formation of molybdenum oxide nanostructures

Chinese Journal of Polymer Science Vol. 27, No. 1, (2009), 11−22

Chinese Journal of Polymer Science ©2009 World Scientific

Feature Article

FORMATION OF MOLYBDENUM OXIDE NANOSTRUCTURES CONTROLLED BY POLY(ETHYLENE OXIDE) Chirakkal V. Krishnan, Rafael Muñoz-Espí, Qi Li, Christian Burger and Benjamin Chu* Department of Chemistry, Stony Brook University, NY 11794-3400, USA Abstract Polymeric systems have played an important role as structure-directing agents and in the control of nucleation and growth of crystals. This article reviews the work of our research group in the field of the polymer-assisted crystallization of inorganic materials, mainly focused on the formation of highly ordered, porous molybdenum oxide nanostructures. Different experimental parameters including the influence of poly(ethylene oxide)-containing polymers on the morphology and structure of the products obtained from peroxomolybdate solutions are examined. Our electrochemical investigations on molybdate species are also briefly described. Finally, the importance of the precursor species in the formation of the final product is discussed. Keywords: Polymer-assisted crystallization; Molybdenum oxide; Peroxomolybdate; Polyoxomolybdate.

INTRODUCTION The catalytic, electrochromic, and photochromic properties of MoO3 make the preparation and characterization of structures, and especially nanostructures, of this transition metal oxide and its derivatives challenging and interesting[1]. A variety of oxidation and hydration states of molybdenum oxide, such as oxides, sub-oxides, peroxides, hydroxides, and crystalline hydrates are known. Three different crystalline polymorphs of anhydrous MoO3 have been reported: the thermodynamically stable orthorhombic phase α-MoO3, the metastable β-MoO3, and the metastable high-pressure phase MoO3-II[2]. Furthermore, several hydrates can be found, including a monoclinic dihydrate (MoO3·2H2O), a yellow monoclinic monohydrate (β-MoO3·H2O), a white triclinic monohydrate (α-MoO3·H2O), a monoclinic hemihydrate (MoO3·1/2H2O), and an orthorhombic MoO3·1/3H2O[2−5]. With less and less hydration water in the crystal, MoO3·2H2O →α-MoO3·H2O → MoO3·1/3H2O → MoO3, there is an increasing tendency toward the formation of linear Mo―O―Mo chains in the crystal structure[6]. Nanostructures from molybdenum oxide and its derivatives have been synthesized by a variety of methods. Starting from a structure-directing organic template, α-MoO3·H2O with fibrous structures was prepared using a lamellar molybdenum oxide-amine (with different alkyl chains) composite and a hydrothermal reaction followed by nitric acid treatment[7]. One-dimensional α-MoO3 nanoribbons and nanorods were also synthesized via acidification and aging (about one month) of (NH4)6Mo7O24·4H2O solutions under hydrothermal conditions. The as-grown α-MoO3 crystals were further used as metal oxide precursors for the preparation of hexagonal MoS2 nanorods under H2S/H2 reduction[8]. Molybdenum trioxide nanobelts and prism-like particles were also prepared from acidic solutions in the presence of nitrate salts and a hydrothermal procedure[6]. Very recently, a flexible one-step solvothermal procedure for the synthesis of MoO3 nanorods using MoO3·2H2O as the starting material in both neutral and acidic media has been reported[9, 10]. The MoO3·2H2O precursor for this synthesis was *

Corresponding author: Benjamin Chu, E-mail: [email protected] Invited lecture presented at the Session in Memory of the 90th Birthday of Professor Renyuan Qian of the International Symposium on Polymer Physics, 2008, Xiamen, China Received August 25, 2008; Accepted September 12, 2008

12

C.V. Krishnan et al.

prepared by dissolving Na2MoO4 in HClO4[11]. Multi-lamellar mesostructured molybdenum oxide nanofibers and nanobelts have been prepared from MoO3 by a hydrothermal process using the surfactant cetyltrimethylammonium bromide as a template[12]. Nanobelts were also prepared in the absence of any template using Na2MoO4 and HClO4 under hydrothermal conditions[13]. However, all these hydrothermal or solvothermal routes for the formation of MoO3 nanostructures required high pressures with the use of stainless steel autoclaves, high temperatures (120 to 200°C) and long reaction times (20 h to 7 days). Helical nanosheets, cross-like nanoflowers and nanobelts of molybdenum oxides have also been prepared hydrothermally from a blue slurry obtained by dissolving molybdenum metal in a restricted amount of hydrogen peroxide[14]. A peroxo species, MoO2.67(O2)0.33·0.75H2O, was also reported by hydrothermal process using peroxomolybdic acid[15]. Platelets of molybdenum oxide with maximum basal planes (010) could be obtained by oxidizing molybdenum metal at high temperatures[16]. Among the many possible synthetic pathways, we have been particularly interested in how polymeric systems can affect nanofabrication[17]. Macromolecules may specifically control nucleation and growth processes or may act as templates or structure-directing agents, as well as participate in chemical reactions[18−20]. In some cases, polymers can stabilize phases that are otherwise metastable. The organic component can, at times, be incorporated into the inorganic matrix and then be easily removed by dissolution or calcination. POLYMER-ASSISTED GROWTH OF MOLYBDENUM OXIDE NANOSTRUCTURES Nanostructured polymers are widely used as templates for inorganic materials[21−23]. However, rather than following the traditional template route, we are interested on the effects that polymers (nanostructured or not) can have on the growth of inorganic crystals and in their ability to lead to highly ordered, highly symmetric and comparatively large inorganic nanostructures. Therefore, we will refer to “polymer-assisted” or “polymercontrolled” growth, rather than “templating,” which should, in fact, be considered as representing a different approach. Some years ago, we reported the formation of highly uniform hollow molybdenum oxide nanospheres, in the presence of micelles of a PEO–PBO–PEO [PEO, poly(ethylene oxide); PBO, poly(butylene oxide)] triblock copolymer. Peroxomolybdate at pH ca. 1.5, was obtained from the dissolution of molybdenum metal in hydrogen peroxide[24]. Highly ordered, porous polyoxomolybdate structures with a very unusual primitive cubic structure were obtained from the same precursor in the presence of high concentrations of PEO–PPO–PEO [PEO, poly(ethylene oxide); PPO, poly(propylene oxide)] triblock copolymers[25]. The originally expected templating mechanism of the polymer could be excluded based on the facts that the resulting inorganic nanostructure had a different symmetry (primitive cubic) than the block copolymer (body-centered cubic) and that analogous cubic structures to those obtained with the triblock copolymers could also be obtained with simple PEO homopolymers. This primitive cubic structure, depicted in Fig. 1 together with a transmission electron microscopy (TEM) image, resembles that of a zeolite, albeit on a much larger length scale (lattice constant of ca. 5 nm vs. the ca. 1 nm of a typical zeolite). The dark blue color of the products indicates the presence of Mo(V) and/or Mo(IV). The presence of polyoxyethylene-containing polymers seems to play two different roles: first, they act as structure-directing agents, most likely by coordinating with the molybdenum ions in a crown ether-like fashion; and second, they act as weak reducing agents to partially reduce Mo(VI) to Mo(V) and/or Mo(IV). The coexistence of molybdenum of lower oxidation states seems to be crucial in the final assembly of the structures. Müller et al. synthesized a smaller icosahedral polyoxomolybdate cluster (known as “Keplerate”) with a diameter of 2.5 nm and having 132 Mo atoms (60 MoV and 72 MoVI) from solutions of ammonium molybdate, ammonium acetate, acetic acid and hydrazine sulfate[26]. These studies clearly demonstrated the dominant role of the mixed valence of molybdenum in forming nanostructures and/or supramolecular assemblies.

Formation of Molybdenum Oxide Nanostructures Controlled by Poly(ethylene oxide)

13

Fig. 1 Primitive cubic nanoporous molybdenum oxide with a cubic lattice constant of about 5.0 nm Left: experimental TEM image; Right: schematic sketch[25]

The formation of the highly ordered primitive cubic structures described above required long crystallization times (more than two weeks at room temperature), which could be reduced to only few hours by using a more recently reported sonochemical process at a temperature of ca. 70°C[27]. By controlling the metal-to-polymer ratio, we have succeeded in synthesizing nanoribbons and whiskers of mixed-valent molybdenum oxides with well-defined shapes[28]. The nature of the molybdenum species formed in the presence of H2O2 has been studied by Raman spectroscopy, and it is generally agreed that the polymerization processes are far less complex in the presence of peroxide[29]. Stock solutions are often prepared by dissolution of the metal in an excess peroxide and then removing the excess peroxide either by evaporation[30, 31] or by using a platinum gauze[32]. Raman spectra of freshly prepared solutions with and without heating to remove excess peroxide, the effect of aging, and the effect of added peroxide on aged solution are shown in Figs. 2 and 3. For a fresh solution, heating to remove the excess peroxide did not change the integrity of the species, as shown in Fig. 2. The stretching modes associated with the side-bonded peroxo-ligand [νsym(O―O), νsym[MoO2], and νasym[MoO2] are generally found near 880, 600, and 530 cm−1 respectively. Terminal Mo＝O stretches are expected near 950 cm−1 with the corresponding deformation near 300 cm−1. Our data are consistent with this observation and addition of excess peroxide to an aged solution restores the νsym(O―O) mode at 880 cm−1, as shown in Fig. 3.

Fig. 2 Raman spectra: (a) 1 mol/L peroxomolybdate with A) excess peroxide, 1) 545, 2) 584, 3) 661, 4) 881, 5) 978 cm–1; B) after heating to remove excess peroxide, 1) 543, 2) 573, 3) 659, 4) 880, 5) 976 cm−1; (b) 1 mol/L peroxomolybdate after 43 days 1) 373, 2) 564, 3) 918, 4) 963 cm−1

14


Fig. 3 Raman spectra: (a) 0.1 mol/L peroxomolybdate A) after 43 days, B) with added excess peroxide; (b) 0.5 mol/L peroxomolybdate A) after 43 days, B) with added excess peroxide

The structural and morphological features of the molybdenum oxide products obtained by the sonochemical pathway depend on different experimental parameters. The influence of the nature of polymer, pH, added peroxide, sonication time, and presence of counter ions were investigated and the results are briefly described in the following paragraphs. Effect of Polymer We have previously reported the results of the influence of different polyoxyethylene-containing polymers and of different molybdenum/polymer ratios on the morphology of molybdenum oxide[27, 28]. As an example for the effect of the polymer, Fig. 4 compares the SEM micrographs of a sample prepared from a peroxomolybdate solution at pH ca. 1.5 without any added polymer with those of samples prepared in the presence of PEO-100k and PEO-1000k. Results from small angle X-ray scattering (SAXS) are more illustrative than micrographs. Figure 5 presents the SAXS patterns obtained without polymer and in the presence of PEO-100k, together with the radially averaged scattering curves. One can unambiguously observe the absence of any SAXS peak in the samples without any polymer, whereas the presence of polymer leads to peaks ascribable to the primitive cubic structure. This observation clearly indicated an increase in the long-range order when the samples were synthesized with the polymer. It should be noted that when we added a small amount of ethylene glycol (acting as a mild reducing agent) to the system containing PEO homopolymers, the change from yellow to a blue color was accelerated.

Fig. 4 Products obtained from a 0.6 mol/L peroxomolybdate solution after a sonication time of 12 h a) In the absence of any polymer; b) In the presence of PEO-100k (Mo:EO = 1:5); c) In the presence of PEO-1000k (Mo:EO = 1:5)


15

Fig. 5 SAXS patterns obtained (a) in the absence of any polymer and (b) in the presence of ca. 17 wt% of PEO-100k; (c) scattering curves corresponding to the patterns

Effect of pH There are major changes in both isopolyoxomolybdate and peroxomolybdate species depending on the pH, as described in more detail in a later section. We have established that, at a pH ca. 1.5, irrespective of the peroxomolybdate species, the product formed has the same crystal structure. The role of the peroxide in the formation of molybdenum oxide crystals has not been clearly established. Also, there is no direct evidence for

Fig. 6 Products obtained at different pH values from a 0.6 mol/L peroxomolybdate solution in the presence of PEG-4600 (Mo:EO = 1:5) after a sonication time of 12 h The pH was raised by adding NaOH.

16


the involvement of peroxide. The peroxide is also involved in the interaction with the polymer. The products formed from this reaction with the polymer interacted either with the peroxo species or the isopolyoxomolybdate to produce the final mixed-valent molybdenum oxides. Figure 6 shows how the morphology of the obtained materials ranged from balls at pH 1.5 to prismatic-shaped crystals at pH > 6, going through more poorly defined shapes at intermediate pH values. In these experiments, the pH of the peroxomolybdate solution, containing a fixed molybdenum/polymer ratio, was increased by using sodium hydroxide. Effect of Added Hydrogen Peroxide The lack of information on the extent of free peroxide present in the peroxomolybdate obtained by dissolution of the metal in H2O2 has prompted us to investigate the influence of excess peroxide in the starting precursor on the morphology of the products. The extent of decomposition of peroxide on sonication is also not known. Figure 7 presents the SEM micrographs of products obtained at different excess concentrations of H2O2. The results, when compared with those of materials obtained with no excess peroxide added (first panel of Fig. 6), indicate that the morphology is not greatly influenced by the excess peroxide.

Fig. 7 Products obtained from a 0.6 mol/L peroxomolybdate solution in the presence of PEG-4600 (Mo:EO = 1:5) at different concentrations of H2O2 (sonication time of 12 h): (a) [H2O2] = 0.44 mol/L, (b) [H2O2] = 1.32 mol/L, and (c) [H2O2] = 3.52 mol/L

Fig. 8 Products obtained in the presence of polyoxyethylene homopolymers (Mo:EO = 1:5) from a 1.0 mol/L peroxomolybdate solution: (a) PEG-4600, sonication for 12 h, (b) PEG-4600, sonication for 48 h, (c) PEO-100k, sonication for 12 h and (d) PEO-100k, sonication for 48 h


17

Effect of Sonication Time The color of a 0.6 mol/L solution of peroxomolybdate in the presence of a polymer changed from yellow to dark blue following the sequence: yellow → light orange → orange → darker orange → light greenish orange → light green → blue → dark blue (0, 2, 4, 8, 12, 13, 13.5, and 14 h of sonication, respectively). This observation suggests that more and more mixed-valent oxides are formed progressively when increasing the sonication time. The micrographs of Fig. 8 for materials obtained in the presence of PEG-4600 and PEO-100k show that an increase in the sonication time from 12 h to 48 h does not have any significant influence on the final morphology. However, the micrograph shown in Fig. 9 indicates that well-defined and uniform cubes were obtained by waiting for 5 days after an initial sonication period of 12 h. During this time the color also changed from yellow to deep blue indicating further reduction of Mo(VI) to lower oxidation states in the final product.

Fig. 9 Product obtained after standing of the solution (0.4 mol/L peroxomolybdate, PEG 4600 (Mo:EO = 1:5), pH ca. 1.5 sonication for 12 h) for 5 days

Effect of Simple Counter Ions Since we had introduced sodium ions for changing the pH and since salts, such as NaCl and NaClO4 or the corresponding acids, were used in many hydrothermal procedures with isopolyoxomolybdates, we also

Fig. 10 Products obtained in the presence of PAA-g-PEO (partial sodium salt) from 0.4 mol/L peroxomolybdate solutions: (a) and (b) SEM micrographs of the materials prepared at concentrations of the polymer of 0.1 wt% and 9.1 wt%, respectively; (c) XRD patterns


18

examined the influence of the nature of the counter ions in the present studies with peroxomolybdate. As an example of the effect that counter ions can have in the crystallization process, Fig. 10 presents the SEM micrographs and the X-ray diffraction (XRD) patterns of materials crystallized in the presence of a partial sodium salt of a poly(acrylic acid)-graft-PEO (NaPAA-g-PEO) copolymer. The XRD peaks for the sample prepared with 0.1 wt% NaPAA-g-PEO could all be assigned to monoclinic MoO3·1/2H2O, which is the common phase obtained in our experiments at low concentrations of polymer without counter ions. However, the micrographs of the sample obtained at a higher concentration (9.1 wt%) show the presence of a hexagonal phase, which turned out to be sodium molybdenum oxide from the XRD patterns. Small amounts of counter ions do not seem to significantly affect either the morphology or the crystal structure, but higher amounts can lead to the crystallization of other crystal phases, as it is the case of hexagonal sodium molybdate crystals when sodium is present. ELECTROCHEMICAL STUDIES Typically, MoO3 layers have been deposited onto substrates using various techniques, such as chemical vapor deposition, vacuum evaporation, radiofrequency sputtering, electron beam evaporation, sol-gel and spin coating[1]. There was less hydration in the oxides prepared by non-electrochemical methods and, consequently, less proton transport and poorer charge transfer kinetics. The quantity of metal oxide grown by anodic oxidation using a metal substrate is difficult to control. Cathodic electrodeposition using different substrates from peroxomolybdates to pure molybdates has been reported[32]. Low temperature and soft processing capabilities make electrodeposition attractive. The products obtained at various potentials[33, 34], viz., Mo2O5·3H2O at +0.18 V to +0.02 V, 2[Mo3O8·H2O] at +0.02 V to −0.01 V, Mo3O8·H2O at −0.01 V to −0.4 V, and Mo3O8−x(OH)x·xH2O at −0.4 V to −0.6 V, indicated the richness of the electrodeposition method. It could offer the possibility to control the stoichiometry, water content, and oxidation state by controlling the applied potentials. We were focusing our attention on the frequency response analysis or the impedance technique to characterize the substrate, peroxomolybdate and isopolyoxomolybdate[35, 36]. We also utilized this technique to gain information polyoxomolybdate cluster, on the self assembly process of the icosahedral (NH4)42[MoVI72MoV60O372(CH3COO)30(H2O)72]·ca. 300H2O[37]. The information gained from electrochemical studies should also help us to fine-tune the synthesis of these large clusters using electrochemical techniques or by judicious choice of reducing agents. In the past, the reason for the selection of a particular reducing agent has not been explained in the literature, even though the investigators may be aware of it. For example, molybdate can be reduced from +6 to +5 oxidation state by reducing agents, such as sodium dithionate, hydrazine, cysteine, sodium thiosulfate, and ascorbic acid under different experimental conditions for synthesizing different isopolyoxomolybdate clusters. The number of electrons involved in the redox process for many of these reducing agents depends on the pH. However, the actual number of electrons utilized in the synthesis of the clusters has not been clearly explored. Further reduction of molybdenum to +4 and +3 oxidation states are possible with some of these reducing agents. If that is possible, then there may be disproportionation reactions. This aspect of the synthetic process has also not been studied. It is noted that some of these complications can be avoided using electrochemical reduction at controlled potentials. For example, electrochemical techniques have been employed for making clusters of actinyl peroxide nanospheres[38]. Other ingenious schemes, such as electrochemical step-edge decoration on a highly oriented pyrolytic graphite surface have been employed to prepare nanowires and nanoribbons of MoOx: MoO42− + (4 − x )H2O + (6 − 2x)e− → MoOx + (8 − 2x)OH− [39]

They can be subsequently converted to MoS2

(1)

.

COMPLEXITY OF SUBSTRATES Isopolyoxomolybdates In alkaline solutions, the molybdate species exists only as MoO42−. At pH values below 7.0, heptamolybdates are formed by the condensation reaction


19

7MoO42− + 8H+ → Mo7O246− + 4H2O

(2)

At lower pH values, protonated species, such as HMo7O245− and H2Mo7O244−, are formed. Our molybdenum oxide materials were typically synthesized from solutions at pH 4.5 and below. At lower pH, octamolybdates are formed by the reaction: 8MoO42− + 12H+ → Mo8O264− + 6H2O

(3) 3−

Further lowering the pH can produce protonated species of the octamolybdates, such as HMo8O26 and H2Mo8O262−. Octamolybdates are also known to exist as α-, β-, and γ-type, where the γ-form was an intermediate in the α↔β interconversion. The formation of some of these octamolybdates is counter ion-dependent. The existence of significant amounts of HMoO4− and H2MoO4 before the heptamolybdates were formed has also been postulated[40, 41]. Confusing ways of writing the formula for the same species can complicate the search and interpretation of data in the literature. For example, H2MoO4·H2O has been written as Mo(OH)6 or MoO2(H2O)2(OH)2 or MoO3(H2O)3[42] and HMoO4− has been written as MoO3(OH)−. At very high acidities, cations, such as MoO22+[43], and large anions, such as Mo36O1128−[44], have also been reported to be formed. The counter ion dependence of the various species formed became obvious from the fact that clusters, such as (NH4)42[MoVI72MoV60O372(CH3COO)30(H2O)72]·ca. 300H2O. ca. 10CH3COONH4 were formed around pH 4.2 in the presence of an excess of acetate and of small amounts of a reducing agent, instead of a polymer[26, 37]. The complicated nature of the self-assembly process was evident from the fact that other compounds, such as 36H2O·3CH3COOH, (NH4)12Na12[Mo40O128]·ca. 70H2O, Na10[H4Mo18O56(CH3COO)2]·ca. 64H2O, Na32[H4Mo54O168(CH3COO)4]·ca. 128H2O, (NH4)12Na20[H4Mo54O168(CH3COO)4]·ca. 98H2O, [Mo48O128]28−, [Mo116O331](CH3COO)30(H2O)56]46− and Na32[H4Mo54O168(CH3COO)4]·ca. (NH4)32[MoVI110 MoV28O416H6(H2O)58(CH3COO)6]·xH2O (x ca. 250) had been synthesized and characterized from aqueous molybdate solutions containing acetate at a pH of 3.5−4 using different reducing agents as well as different amounts of the same reducing agent to control the ratio of MoVI and MoV[45]. Peroxomolybdates The dissolution of molybdenum in hydrogen peroxide has been reported to produce solid MoO2(OH)(OOH)[30, 31]. Later work has reported the species in solution, H2[Mo2O3(O―O)4(H2O)2], obtained by the reaction of molybdenum and hydrogen peroxide[46]. This is in agreement with the data on the oxidation of molybdenum at the anode to produce the species [Mo2O3(O―O)4]2−. It is known that molybdate interacts with hydrogen peroxide to form a variety of peroxo complexes with 1 to 6 attached peroxide groups[47−49]. In our experiments, when the molybdenum metal was dissolved in 30% hydrogen peroxide to make a 1.0 mol/L solution, the resulting pH was ca. 1.5. The most probable dissolution reaction was suggested as[35] 2Mo + 10H2O2 → H2Mo2O3(O2)4(H2O)2 + 7H2O

(4)

[46]

This is in agreement with the electrochemical oxidation of molybdenum

2Mo + 4H2O2 → [Mo2O2(O2)4]2− + 14H+ + 12e−

(5)

When the yellow solution obtained by dissolution of the molybdenum metal in 30% H2O2 was evaporated to dryness, the formation of MoO2(OH)(OOH) suggested the reaction[35] Mo + 4H2O2 → MoO2(OH)(OOH) + 3H2O

(6)

[47−50]

and therefore changes in the concentration of Molybdenum can form a variety of peroxo complexes the precursor could shift the equilibrium. For example, it has been suggested that in the presence of an excess amount of peroxide, the tetraperoxodimolybdate formed, H2Mo2O3(O2)4, might split into H2MoO2(O2)2 upon dilution[47]. H2Mo2O3(O2)4 + H2O ⇄ 2 H2MoO2(O2)2

(7)


20

With limited peroxide, the hydrolysis reaction upon dilution could be H2Mo4O9(O2)4 + H2O ⇄ 2 H2Mo2O5(O2)2

(8)

Whether the species in aqueous solution was H2Mo2O3(O2)4, H2MoO2(O2)2, or MoO2(OH)(OOH), we noted that they all disproportionated to give the final product MoO3 by a catalytic process[28]. It should also be noted that the oxygen produced in this process was 100% singlet oxygen, 1O2(1∆g), a reactive oxygen species, that may interact with the polymer. H2Mo2O3(O2)4 ⇄ 2MoO3 + H2O + 2O2

(9)

2H2MoO2(O2)2 ⇄ 2MoO3 + 2H2O + O2

(10)

2MoO2(OH)(OOH) ⇄ 2MoO3 + H2O + O2

(11)

In the synthesis of molybdenum oxides from molybdates under high acidities, the major species in the substrate is either heptamolybdate, octamolybdate, or their protonated forms. Peroxomolybdates have also been used under similar high acid conditions for the synthesis of molybdenum oxides in the presence of polymers. At this point we attempt to illustrate a possible connection between the two. In the idealized structure of heptamolybdate, shown in Fig. 11, the peroxidation starts with the most negatively charged (−1.5) octahedra V and VII[47]: Mo7O246− + 2H2O2 → Mo7O22(O2)26− + 2H2O

(12)

This is followed by octahedra (charge −1.17) I−IV. The structure can be maintained reasonably well in all these steps. Further addition of peroxide starts to disrupt the heptamolybdate structure, which would be completely destroyed at a peroxide/molybdenum ratio of 14:1, with the end products being H2[Mo2O5(O2)2] and H[MoO2(O2)(OH)][47]. What could have happened when the peroxomolybdate obtained by dissolving the metal in 30% excess H2O2 was allowed to decompose in the presence of polymers at a pH of ca. 1.5? Did the structure progressively revert back to the hepta- and octamolybdates and their protonated forms? Were the true substrates in the formation of molybdenum oxides really hepta- and octamolybdates and their protonated forms or was there a role for the peroxo species? If the peroxo species were not involved in the process of molybdenum oxide formation, the role of the peroxide could be to partially react with the polymers to make reactive species that would reduce some Mo(VI) to Mo(V) or Mo(IV) to produce mixed-valent oxides of different morphologies, depending on the concentration of the polymer. We plan to study this aspect by investigating the morphology of the molybdenum oxide at two different pH values of about 1.5 and 4 where hepta- and octamolybdates are dominant and using different Mo/peroxide ratios.

Fig. 11 Idealized structure of heptamolybdate, Mo7O246−[47] Out of the 7 MoO6 octahedra , I–IV are placed on their edges, I and II as well as III and IV share their edges, I and III as well as II and IV share corners. Octahedra V–VII share their edges with all their neighbors. 12 oxygens share 1 molybdenum each (6 of these form acidic OH groups and can be replaced by a peroxide ligand while maintaining the heptamolybdate structure), 8 oxygens share 2 molybdenums each, 2 oxygens share 3 molybdenums each and one oxygen shares 4 molybdenums.


21

Finally, it can be concluded that the selection of the substrate, whether it is isopolyoxomolybdate or peroxomolybdate, has to be done with extreme caution because of the complexity of the different species involved depending on the pH, concentration, and the presence of other counter ions. The choice of the substrate, the polymer, and the technique will control the morphology and size. CONCLUSIONS AND OUTLOOK Our polymer-assisted approach for the formation of molybdenum oxide nanostructures has led to a range of surprisingly highly ordered symmetric and porous structures. We have synthesized mixed-valent porous molybdenum oxides from aqueous solutions of peroxomolybdates in the presence of different polyoxyethylenecontaining polymers. The highly ordered primitive cubic structure obtained at high polymer concentrations, which is formed with a certain independence of the exact polymer structure, resembles that of zeolites, although on a much larger scale. The control of pH appears to be very critical in controlling the morphology, and this is probably due to the great variation in the substrate species at different pH values for both isopolyoxomolybdates and peroxomolybdates. By using an ultrasound irradiation at 70°C, we have been able to reduce the time required for crystallization from weeks to hours. Nanoribbons and whiskers with very homogeneous shape could be obtained by this method, but also the unusual primitive cubic structure when high polymer concentrations were used. The morphology and the crystal structure have been shown to depend on the polymer concentration and on the polyoxyethylene chain length. Sonication, probably assisted by peroxide decomposition, did seem to break down some of the polymers, because the solution became less viscous. At this point it remains an open question whether or not there are free radicals and the influence that they may have on the crystal formation. In addition to the high surface area and porosity, the presence of mixed valence states in the materials prepared strongly suggest that our materials can form improved molybdenum oxide-based catalysts. Preliminary results on catalytic epoxidation of olefins seem to confirm this potential application. ACKNOWLEDGEMENT Financial support by the Basic Energy Sciences, Department of Energy (DEFG0286ER45237), is gratefully acknowledged.

REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Kumari, L., Ma, Y.R., Tsai, C.C., Lin, Y.W., Wu, S.Y., Cheng, K.W. and Liou, Y., Nanotechnology, 2007, 18: 115717 Murugan, R., Ghule, A., Bhongale, C. and Chang, H., J. Mater. Chem., 2000, 10: 2157 Seguin, L., Figlarz, M., Cavagnat, R. and Lassegues, J.C., Spectroc. Acta Pt. A-Molec. Biomolec. Spectr., 1995, 51: 1323 Seguin, L., Gerand, B., Chevrier, G. and Touboul, M., Mater. Sci. Forum, 1996, 228-231: 695 Bénard, P., Seguin, L., Louer, D. and Figlarz, M., J. Solid State Chem., 1994, 108: 170 Xia, T.A., Li, Q., Liu, X.D., Meng, J.A. and Cao, X.Q., J. Phys. Chem. B, 2006, 110: 2006 Niederberger, M., Krumeich, F., Muhr, H.J., Müller, M. and Nesper, R., J. Mater. Chem., 2001, 11: 1941 Lou, X.W. and Zeng, H.C., Chem. Mat., 2002, 14: 4781 Michailovski, A., Krumeich, F. and Patzke, G.R., Helv. Chim. Acta, 2004, 87: 1029 Patzke, G.R., Michailovski, A., Krumeich, F., Nesper, R., Grunwaldt, J.D. and Baiker, A., Chem. Mat., 2004, 16: 1126 Cruywagen, J.J. and Heyns, J.B.B., S. Afr. J. Chem., 1981, 34: 118 Song, R.Q., Xu, A.W., Deng, B. and Fang, Y.P., J. Phys. Chem. B, 2005, 109: 22758 Li, X.L., Liu, J.F. and Li, Y.D., Appl. Phys. Lett., 2002, 81: 4832 Li, G.C., Jiang, L., Pang, S.P., Peng, H.R. and Zhang, Z.K., J. Phys. Chem. B, 2006, 110: 24472 Fang, L., Shu, Y.Y., Wang, A.Q. and Zhang, T., J. Phys. Chem. C, 2007, 111: 2401 Baiker, A. and Gasser, D., Z. Physik. Chem., 1986, 149: 119

22


17 18 19 20

Liu, T.B., Burger, C. and Chu, B., Prog. Polym. Sci., 2003, 28: 5 Xu, A.W., Ma, Y.R. and Colfen, H., J. Mater. Chem., 2007, 17: 415 Gorna, K., Muñoz-Espí, R., Gröhn, F. and Wegner, G., Macromol. Biosci., 2007, 7: 163 Wegner, G., Demir, M.M., Faatz, M., Gorna, K., Muñoz-Espí, R., Guillemet, B. and Gröhn, F., Macromol. Res., 2007, 15: 95 Förster, S. and Plantenberg, T., Angew. Chem.-Int. Edit., 2002, 41: 689 van Bommel, K.J.C., Ariannna, F. and Shinkai, S., Angew. Chem.-Int. Edit., 2003, 42: 980 Rozenberg, B.A. and Tenne, R., Prog. Polym. Sci., 2008, 33: 40 Liu, T.B., Xie, Y. and Chu, B., Langmuir, 2000, 16: 9015 Liu, T.B., Wan, Q., Xie, Y., Burger, C., Liu, L.Z. and Chu, B., J. Am. Chem. Soc., 2001, 123: 10966 Müller, A., Krickemeyer, E., Bögge, H., Schmidtmann, M. and Peters, F., Angew. Chem.-Int. Edit., 1998, 37: 3360 Chen, J.L., Burger, C., Krishnan, C.V. and Chu, B., J. Am. Chem. Soc., 2005, 127: 14140 Krishnan, C.V., Chen, J.L., Burger, C. and Chu, B., J. Phys. Chem. B, 2006, 110: 20182 Campbell, N.J., Dengel, A.C., Edwards, C.J. and Griffith, W.P., J. Chem. Soc.-Dalton Trans., 1989, 1203 Segawa, K., Ooga, K. and Kurusu, Y., Bull. Chem. Soc. Jpn., 1984, 57: 2721 Kurusu, Y., Bull. Chem. Soc. Jpn., 1981, 54: 293 Kondrachova, L., Hahn, B.P., Vijayaraghavan, G., Williams, R.D. and Stevenson, K.J., Langmuir, 2006, 22: 10490 McEvoy, T.M. and Stevenson, K.J., Anal. Chim. Acta, 2003, 496: 39 McEvoy, T.M. and Stevenson, K.J., J. Mater. Res., 2004, 19: 429 Krishnan, C.V., Garnett, M. and Chu, B., Int. J. Electrochem. Sci., 2007, 2: 444 Krishnan, C.V., Garnett, M., Hsiao, B. and Chu, B., Int. J. Electrochem. Sci., 2007, 2: 29 Krishnan, C.V., Li, Q. and Chu, B., Electrochim. Acta, 2007, 53: 975 Burns, P.C., Kubatko, K.A., Sigmon, G., Fryer, B.J., Gagnon, J.E., Antonio, M.R. and Soderholm, L., Angew. Chem.Int. Edit., 2005, 44: 2135 Li, Q., Walter, E.C., van der Veer, W.E., Murray, B.J., Newberg, J.T., Bohannan, E.W., Switzer, J.A., Hemminger, J.C. and Penner, R.M., J. Phys. Chem. B, 2005, 109: 3169 Sasaki, Y., Lindqvist, I. and Sillen, L.G., J. Inorg. Nucl. Chem., 1959, 9: 93 Schwarzenbach, G. and Meier, J., J. Inorg. Nucl. Chem., 1958, 8: 302 Cruywagen, J.J. and Heyns, J.B.B., Inorg. Chem., 1987, 26: 2569 Mitchell, P.C.H., Q. Rev. Chem. Soc., 1966, 20: 103 Tytko, K.H., Schonfel.B, Buss, B. and Glemser, O., Angew. Chem.-Int. Edit. Engl., 1973, 12: 330 Müller, A., Kuhlmann, C., Bögge, H., Schmidtmann, M., Baumann, M. and Krickemeyer, E., Eur. J. Inorg. Chem., 2001, 2271 Chakravorti, M.C., Ganguly, S. and Bhattacharjee, M., Polyhedron, 1993, 12: 55 Csanyi, L.J., Transit. Met. Chem., 1989, 14: 298 Dickman, M.H. and Pope, M.T., Chem. Rev., 1994, 94: 569 Tytko, K.H., Gras, D., Katscher, H. and Schröder, F., "Gmelin Handbook of Inorganic Chemistry", 8th ed., Mo Suppl., Springer Verlag, New York, 1989, Vol. B(3b) Nardello, V., Marko, J., Vermeersch, G. and Aubry, J.M., Inorg. Chem., 1995, 34: 4950

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50