Recent Advances in Manganese Oxide Nanocrystals: Fabrication

Mar 30, 2012 - Zhiwen Chen was born in Hefei, Anhui, China, and received his MS degree three years in Inorganic Chemistry from the University of Scien...
3 downloads 15 Views 18MB Size
Review pubs.acs.org/CR

Recent Advances in Manganese Oxide Nanocrystals: Fabrication, Characterization, and Microstructure Zhiwen Chen,*,†,§ Zheng Jiao,*,†,‡ Dengyu Pan,‡ Zhen Li,† Minghong Wu,*,†,‡ Chan-Hung Shek,§ C. M. Lawrence Wu,§ and Joseph K. L. Lai§ †

Shanghai Applied Radiation Institute and ‡Institute of Nanochemistry and Nanobiology, School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, People’s Republic of China § Department of Physics and Materials Science, City University of Hong Kong, Tat Chee Avenue, Kowloon Tong, Hong Kong 4.3.6. Formation Mechanism of Mn3O4 Nanocrystals 5. Summary and Outlook 5.1. Summary 5.2. Outlook 5.3. Concluding Remarks Author Information Corresponding Author Notes Biographies Acknowledgments List of Abbreviations References

CONTENTS 1. Introduction 1.1. General Overview 1.2. Outline 2. Crystal Structures of Manganese Oxides 2.1. Crystal Structure of Mn2O3 2.2. Crystal Structure of Mn3O4 3. Synthesis and Grain Growth Kinetics of Mn2O3 Nanocrystals 3.1. Progress in Mn2O3 Nanomaterials 3.2. Synthesis of Mn2O3 Nanocrystals 3.2.1. Method 3.2.2. Instrumental Analysis 3.3. Grain Growth Kinetics of Mn2O3 Nanocrystals 3.3.1. Kinetic Analysis 3.3.2. Microstructure Analysis 3.4. Spectral Properties of Mn2O3 Nanocrystals 3.4.1. Raman Features of Different Grain Sizes 3.4.2. IR Features of Different Grain Sizes 3.4.3. XPS Analysis of Different Grain Sizes 3.4.4. ESR Analysis of Different Grain Sizes 4. Controllable Synthesis and Microstructure Evolution of Mn3O4 Nanocrystals 4.1. Progress in Mn3O4 Nanomaterials 4.2. Shape-Controlled Synthesis of Mn3O4 Nanocrystals 4.2.1. Synthesis Techniques 4.2.2. Instrumental Analysis 4.3. Microstructure Evolution of Single-Crystal Mn3O4 Nanocrystals 4.3.1. Mn3O4 Nanoparticles 4.3.2. Mn3O4 Nanorods 4.3.3. Mn3O4 Nanofractals 4.3.4. HRTEM Investigation of Mn3O4 Nanocrystals 4.3.5. Spectral Characteristics of Mn3O4 Nanocrystals

© 2012 American Chemical Society

3833 3833 3834 3835 3835 3835

3849 3849 3849 3850 3850 3850 3850 3850 3850 3852 3852 3852

1. INTRODUCTION 3835 3835 3836 3836 3836

1.1. General Overview

Nanomaterials are generally classified into nanostructured materials and nanophase or nanoparticle materials, which form a bridge linking single elements with single crystalline bulk structures.1−5 Nanostructured materials refers to condensed bulk materials made of grains with grain sizes in the nanometer size range. They have a wide range of very interesting new physical properties, as well as potential applications in different fields, such as magnetorecording, permanent magnets, sensors, biomedicine, etc.6−10 Nanophase or nanoparticle materials are usually dispersive nanoparticles. In the past few decades, nanomaterials with controlled size and shape have been investigated in great detail. To distinguish nanomaterials from bulk materials, it is vitally important to demonstrate their unique properties and their prospective impacts in science and technology.11−15 Nanotechnology in the twenty first century facilitates the miniaturization of devices into nanometer sizes with concomitant dramatic enhancement of their ultimate performance. This raises many issues regarding the use of new materials for achieving specific functionality and selectivity. Nanostructured and nanophase or nanoparticle materials, belonging to a new branch of materials research, are attracting a great deal of attention because of their potential applications in areas such as optics,16 electronics,17 catalysis,18 ceramics,19,20 magnetic data storage,21,22 and nanocomposites.23 The unique properties and the improved performances of nanomaterials are determined by their sizes, surface

3836 3836 3838 3839 3839 3840 3841 3843 3844 3844 3845 3845 3845 3845 3845 3846 3846 3847 3847

Received: December 2, 2011 Published: March 30, 2012 3833

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

structures, and interparticle interactions.24,25 Because of finite size effects, some basic magnetic properties of materials, such as the spontaneous magnetization, the Curie temperature, and the anisotropy energy, are strongly influenced by the particle size.26,27 As the particle size decreases, surface to volume ratio increases resulting in an increasing fraction of atoms lying at or near the surface. Surface and interface effects then become more important. Because of the presence of defects, missing bonds, fluctuations in number of atomic neighbors and interatomic distances, the surface is characterized by topological and magnetic disorder. Such disorder propagates from the surface to the particle core, rendering the picture of the particle as a perfectly ordered single domain invalid, with spins rotating in a synchronous way as a large single spin.28−34 The role played by particle size is comparable, in some cases, to that played by the particle chemical composition, adding another flexible parameter for designing and controlling their behavior.35,36 To fully understand the impacts of nanomaterials in nanoscience and nanotechnology and answer the question of why nanomaterials is so special, we need to study the complex magnetic behavior of very small particles and the effect of some surface related phenomena such as surface anisotropy, which can give a dominant contribution to the total particle anisotropy, core−surface exchange interaction (exchange anisotropy, in the case of different magnetic phases) and possible competition between surface and core effects.37,38 This article will review some of the rational fabrication methodologies and the unique properties of manganese oxide nanomaterials, aiming at elucidating their distinct characteristics. Transition metal oxides belong to a class of materials that are vitally important for developing new materials with functionality and smartness.39−45 In particular, because of their unique material properties and potential for desired nanostructures, transition metal oxide nanocrystals, and their assemblies have been widely utilized in various fundamental research and technological applications.46−49 The unique properties of these materials are related to the presence of elements with mixed valences of transition elements. Therefore, it is not surprising that long-term endeavors have focused on the synthesis of monodisperse metal oxide nanocrystals and their use as convenient nanobuilding blocks in constructing ordered superlattice assemblies with advanced functions.50−52 Research on manganese oxides has been a key topic among studies on transition metal oxides. This is due to their potential applications in diverse areas, including rechargeable lithium ion batteries, catalysis, molecular adsorption, gas sensors, energy storage, and magnetics. They are of considerable interest in many technological applications, for example, electrochemical reaction and batteries, due to their outstanding structural diversity combined with novel chemical and physical properties.53−56 Manganese oxide is also an important catalyst for removing carbon monoxide and nitrogen oxide from waste gas and often is used to produce soft magnetic materials, such as manganese zinc ferrite.57−59 Many manganese-oxide-based compounds also exhibit typical colossal magnetoresistance behavior.60 Manganese oxide nanocrystals are expected to exhibit better properties compared with their bulk material counterparts in that a larger surface-to-volume ratio would lead to an improved capacity in absorbing oxygen for oxidation of carbon monoxide and nitrogen.61−63 In early studies of transition metal oxides, many authors have reported a lot of very interesting and meaningful results,64−68 for example, the

oxidation states, the systematics of core−electron exchange splitting, the oxygen 1s X-ray absorption and K near-edge fine structure, and the L3/L2 white-line intensity ratios in manganese oxides. These results provide useful guidance to researchers engaging in more in-depth follow-up studies. Notwithstanding the fact that manganese oxides have applications in many areas, this topic has not been reviewed in detail from the perspective of our latest understanding of its precise technical functions in the vast nanomaterial domain. Herein, we will present a comprehensive review on the recent advances in the synthesis, microstructure, and characterization of manganese oxide materials. This article mainly focuses on the wide-ranging research efforts on the development of preparation methodologies and microstructural assessments of manganese oxides in the past three decades. In particular, the novel and simple approaches to synthesize manganese oxide nanocrystals, including nanoparticles, nanorods, and nanofractals, are discussed in detail. This is an interdisciplinary work that integrates the areas of physics, chemistry, materials science, and nanotechnology. 1.2. Outline

A brief outline of the organization and contents of this review is presented here to facilitate ease of reading this article. A widely applicable synthesis route for various morphologies of manganese oxide nanocrystals including nanoparticles, nanorods, and nanofractals, and their unique microstructural characteristics is summarized in the following. The research work performed to elucidate these fundamental intricate aspects includes: First, we will discuss the synthesis, microstructure, grain growth kinetics and various spectra evaluation of Mn2O3 nanocrystals, particularly in the isothermal grain growth and kinetics models of Mn2O3 nanocrystals as well as the spectral analysis by Raman, infrared (IR) spectra, X-ray photoelectron spectroscopy (XPS), and electron spin resonance (ESR). The grain growth data of Mn2O3 nanocrystals have been analyzed by two different models. The first model, assuming normal grain growth as that in conventional polycrystalline materials, yields large grain growth exponent (n) and extremely low activation energy (Q). Although it can describe the evolution of grain sizes, it fails to give satisfactory physical interpretation of n and Q, with values of both parameters extending beyond theoretical predictions. The second model is based on the structural relaxation of the interface component in nanocrystalline materials. In this case, the ordering of distorted interfaces by structural relaxation proceeds with grain growth. This structure relaxation model not only describes the evolution of grain growth well, but also makes reasonable attribution of the low activation energy to the short-range rearrangement of atoms in the interface region as well. The influence of grain size on the spectral properties of Mn2O3 nanocrystals is especially remarkable for nanoscale materials. It is speculated that the various spectral properties are sensitively dependent upon the grain sizes of Mn2O3 nanocrystals. The 3s−3d electrons exchange interaction of Mn3+ ion and the energy level splitting of Mn3+ ion in octahedral crystal field with tetragonal distorted are presented clearly. Second, we will introduce the shape-controlled synthesis and microstructure evolution of Mn3O4 nanocrystals. A widely applicable chemical reaction route to prepare single-crystal Mn3O4 nanocrystals including nanoparticles, nanorods, and nanofractals is decsribed. The nanoparticles, nanorods, and 3834

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

Mn2O3 possesses the characteristics of the space group Ia3̅ cube structure.

nanofractals of Mn3O4 with tetragonal structure were prepared synchronously by a chemical reaction method. The research results indicate that the dripping speed of NaOH solution plays an important role in the nanostructural evolution of Mn3O4 nanocrystals. The difference in the dripping speed of NaOH solution leads to a large difference in Mn3O4 nanomorphologies (nanoparticles, nanorods, and nanofractals). In particular, we present in detail the electronic structure characteristics of Mn3O4 nanorods analyzed by electron energy loss spectroscopy (EELS) technique. The studies indicate that the hybridization between oxygen 2p and manganese 3d orbits plays an important role when considering the electronic structures of Mn3O4 nanorods. In the last section, we will provide a summary and discussion of future work which could throw new light on this interesting research area.

2.2. Crystal Structure of Mn3O4

Figure 1b demonstrates that the Mn3O4 shows the tetragonal hausmannite crystal structure model with the space group I41/ amd. The oxides of low-valency metals, that is, with cations in oxidation number ≤4, are typically ionic compounds. They are most frequently easily obtained in crystalline forms. For sufficiently small cations (r < 0.8 Å), the oxide structures are obtained by close packing of oxide ions (either ccp or hcp) with cations in tetrahedral or octahedral interstices.70 The bigger cations do not enter such interstices, so that structures with less densely packed oxide ions, e.g. simple cubic structures, must be built and higher coordinations for the cations must occur (seven or eight). In ionic metal oxides, the coordination of the cations (four to eight) is generally higher than the valency (one to four) and this also occurs for the coordination of O2‑ oxide ions (three to six). The bulk basic nature of the ionic metal oxides is associated with the strong polarization of the metal−oxygen bond, to its tendency to be dissociated by water and to the basic nature of the products of their reaction with water, i.e. the metal hydroxides.71−74 Table 1 shows the crystal structures and ion coordinations in simple solid oxides of interest in heterogeneous catalysis. From Table 1, it is known that the oxides of the metal elements in very high oxidation states, for example, Mn7+, Cr6+, Mo6+, W6+, V5+, Nb5+, Ta5+, are also frequently denoted as anhydrides and give rise by reaction with water, to species denoted as acids (sometimes polyoxoacids) that are actually quite acidic. In fact, because of the high oxidation states of the metal elements in these compounds, their electronegativity is very high,75 so that the oxides and acids have nearly the same properties of nonmetal oxides and acids. Therefore, it is expected that these high oxidation state manganese oxides (Mn2O3 and Mn3O4) nanostructures may constitute important building blocks for heterogeneous catalysis industry and offer exciting opportunities for both fundamental research and technological applications.

2. CRYSTAL STRUCTURES OF MANGANESE OXIDES 2.1. Crystal Structure of Mn2O3

Figures 1a and b show the crystal structures of Mn2O3 and Mn3O4, respectively.69 Using the referenced data for Mn2O3, this picture (see Figure 1a) is drawn with Mn atoms on the (24d) sites in order to better delineate the difference in the crystallographic behavior of the sites. It is found that the

3. SYNTHESIS AND GRAIN GROWTH KINETICS OF MN2O3 NANOCRYSTALS 3.1. Progress in Mn2O3 Nanomaterials

Manganese oxides, including MnO, MnO2, Mn2O3, and Mn3O4, are of considerable importance in many applications, such as ion-exchange, molecular adsorption, catalysis, electrochemical reaction, batteries, and magnetism, due to their structural flexibility with novel chemical and physical properties.76−81 Among these manganese oxides, Mn2O3 is wellknown as a cheap and environment-friendly catalyst for removing carbon monoxide and nitrogen oxide from waste gas,82−87 and is also used to produce soft magnetic materials such as manganese zinc ferrite.88 Much attention has been paid to nanostructures due to the fact that the drastic increase in the ratio of surface area to volume causes the appearance of unique physical and chemical properties.89−93 The nanostructured materials have potential technical applications because of their distinctive optical, mechanical, electrical, acoustic, and magnetic properties.94−98 Mn2O3 nanocrystals are expected to exhibit better properties in the following respects: a larger surface-tovolume ratio and an improved capacity to absorb more oxygen for oxidation of carbon monoxide and nitrogen.99−101 Therefore, the investigations of Mn2O3 nanocrystals have become the

Figure 1. Crystal structures of (a) Mn2O3 and (b) Mn2O3. Reprinted with permission from ref 69. Copyright 1991 ASM International. 3835

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

Table 1. Crystal Structures and Ion Coordinations in Simple Solid Oxides of Interest in Heterogeneous Catalysis.a Reprinted with Permission from Ref 70. Copyright 1991 Royal Society of Chemistry

microscopy can be used to examine the microstructure at high resolution. X-ray powder diffraction (XRD) patterns, typically recorded at a scanning rate of 0.05° s−1 in 2θ ranges from 15° to 65° with Cu Kα radiation (1.5418 Å), will generate information on the particle size. The X-ray line-widths provide an estimate of the average particle size through the Scherrer formula Dhkl = (kλ)/(B cos 2θ), where Dhkl is the diameter of the particle in Å, k is a constant which is related to the (hkl) index and crystalline shape, λ is the wavelength of the X-rays, B is the full-width-half-maximum of the diffraction lines, and θ is the Bragg angle. Raman spectroscopy could also yield useful information. For example, Raman spectra could be obtained from a SPEX-1403 laser Raman spectrometer using the 514.5 nm line from an Ar+ laser as the excitation in backscattering configuration and the laser power delivered to the samples at about 30 mW. The measurements of infrared (IR) spectra could be carried out with powder samples for which the KBr is used as a carrier. Typically, the IR spectra taken are in the frequency range from 350 to 4000 cm−1. X-ray photoelectron measurements could be performed on clean sample surfaces obtained by scraping them in situ with a diamond file in ultrahigh vacuum. Typically, X-ray photoelectron spectra (XPS) are taken using a monochromatized X-ray source of Mg Kα radiation (hν = 1253.6 eV) with energy resolution at about 0.9 eV. All the measurements are carried out at room temperature. Electron spin resonance (ESR) meaasurements are also useful. Typically, ESR spectra are recorded with microwave frequency of 9.35 GHz.

new focus in the field of materials science and engineering technology. Much effort is being made to investigate the synthesis and microstructure of Mn2O3 nanomaterials in order to obtain new applications.100−104 Nanometer-size Mn2O3 particles are used in some areas while the surface properties are very important.63,82−84 Many material scientists are exploring the effects of introduction of metal oxides into nanomaterials in order to achieve novel valuable properties,99,105−109 for example, temperature indicator, because of the obvious grain size dependence of physical properties.110 3.2. Synthesis of Mn2O3 Nanocrystals

3.2.1. Method. Mn2O3 nanocrystals are prepared by a chemical liquid homogeneous precipitation (CLHP) method.102 The method employed to synthesize these nanocrystals is described in detail by the following chemical reaction processes using reactants: MnCl2·4H2O, NaOH, and H2O2. Here, the H2O2 aqueous solution is used as an oxidizing agent, and a surfactant, C18H29NaO3S, is added to the reaction mixture to prevent nanoparticles from growing in MnCl2 aqueous solution. First, an appropriate amount of MnCl2·4H2O is dissolved in water. Second, 2.5 M H2O2 solution and C18H29NaO3S surfactant are added to the above solution, and then 2.5 M NaOH solution is dripped into the system very slowly. The aqueous solution becomes a brown suspension after a rapid chemical reaction. Precipitation occurs after allowing the solution to stand for 2 h. The reaction equation is 2MnCl2 + H2O2 + 4NaOH

3.3. Grain Growth Kinetics of Mn2O3 Nanocrystals

→ Mn2O3 + 4NaCl + 3H2O

3.3.1. Kinetic Analysis. Previous work shows that a good understanding of the vibration behavior of bulk and supported manganese oxides could be achieved by spectroscopic techniques.111 The situation is, however, quite different for

3.2.2. Instrumental Analysis. After synthesis of the Mn2O3 nanocrystals, instrumental analysis should be performed to examine the material synthesized. Transmission electron 3836

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

appropriate for the short-range rearrangement of atoms in nanocrystalline SnO2. However, the grain growth kinetic of Mn2O3 nanocrystals is also very important for widespread technological applications. Figure 2 shows the isothermal grain

manganese oxide nanomaterials. In fact, the spectral characteristics of Mn2O3 nanocrystals with different grain sizes are also important information in the understanding of various physical and chemical properties. To our knowledge, there are only a few experimental studies on the synthesis, microstructure, and spectral characteristics of manganese oxide nanomaterials.104,111−116 For example, Praserthdam et al. reported the grain growth behavior of some transition metal oxides and spinel oxides nanocrystals.117 They found that both the initial crystallite size and calcinations temperature affected crystal growth, and this behavior was unambiguously demonstrated by log−linear plotting between (d/d0) and T/(d0)1/2. The behavior of grain growth kinetics and their mechanism for Mn2O3 nanocrystals are essential to the understanding of various properties. The investigation of the thermal stability, microstructure and spectral analysis is therefore important from the technological point of view as well as for scientific interests. We will discuss the grain growth kinetics and spectral evaluation of Mn2O3 nanocrystals. For conventional polycrystalline or coarse grain materials, it is known that the driving force for grain growth results from the decreasing system energy by decreasing the total grain boundary energy. The rate of grain growth is proportional to the curvature radius of the grains, and the equation of grain growth kinetic is118 Dn − D0n = Kt

Figure 2. Grain size as a function of annealing time at various temperatures, while points are the experimental data and the lines are the curves fitted with the eq 1. Reprinted with permission from ref 61. Copyright 2005 Springer.

(1)

growth results of Mn2O3 nanocrystals in the temperature range between 200 and 500 °C for different annealing times. The analysis of grain growth kinetics shows that the grain size increases rapidly at short annealing times and then continues to grow at a very low rate. The grain growth is very fast in the first few minutes, which implies a low activation energy process dominates in this growth stage. The grain size data were fitted to eq 1 using a nonlinear fitting routine. The fitted lines are also shown in Figure 2. The grain growth exponent n and the rate constant K for the four annealing temperatures are shown in Figure 3. The n values (from 5 to 11) exceed the theoretical predicated values (from 2 to 4) and increase with increasing annealing temperature. Since the exponent n is not a constant, the activation energy Q cannot be determined with eq 1. In view of the fact that the average grain sizes seem to settle to

where D is the average grain size after annealing, D0 the initial average grain size, t the annealing time, and K is a temperature (T) dependent rate constant. The constant K can be expressed in an Arrhenius type equation, K ∝ exp(−Q/RT) with Q being the activation energy for isothermal grain growth and R the gas constant. The index n in eq 1 is called the grain growth exponent. According to previous works,119,120 the n value depends on the microstructure and grain growth mechanism. The value of n equals to 2 for normal grain growth in a pure, single-phase system; 3 for grain growth in the presence of solutes; and 4 for the presence of pores. If the rate of grain growth is assumed to be proportional to the difference between the grain boundary curvatures for the instantaneous grain size (1/D) and that corresponds to the limiting grain size (1/Dm), the kinetic equation takes on another form121 ⎛ D − D0 ⎞ D0 − D + ln⎜ m ⎟ = K ′t Dm ⎝ Dm − D ⎠

(2)

where K′ is a temperature dependent rate constant like K in eq 1. The behavior of grain growth for nanocrystalline materials exhibits two distinguishing features. First, the values of n cover a region from 2 to >4, for example, n = 3−11 for nanocrystalline Fe.121 Second, the values of activation energy measured in several nanocrystalline ceramic experiments are very low, for example, about 96 kJ/mol for nanocrystalline TiO2.120 Very large grain growth exponent and extremely low activation energy observed in nanocrystalline ceramics are not commonly observed in conventional materials. It appears that the description of grain growth behavior for nanocrystalline materials by conventional grain growth kinetics is not satisfactory. Zhang et al proposed that the structural relaxation of interfaces might be an alternative mechanism of grain growth for nanocrystalline Si3N4.122 Lai et al. have studied the isothermal grain growth of nanocrystalline SnO2.122,123 They found that the lower activation energy of 31 kJ/mol was

Figure 3. Variations in grain growth exponent n and rate constant K for isothermal grain growth. Reprinted with permission from ref 61. Copyright 2005 Springer. 3837

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

used for calculating activation energy (Q), where 0 ≤ m ≤ 1. The activation energy was found to be 32 kJ/mol. Comparing eq 2 with eq 3, we note that the fitting lines by using eq 3 reveal the important features, namely, the existence of limiting grain sizes (1/Dm) for different annealing temperatures, and the fitting variances of curve fitting using the relaxation eq 3 are much smaller than those of using eq 2. It is thus evident that the fitting results by eq 3 are better than that by eq 2. The higher the annealing temperature, the larger the limiting grain size is and the longer it takes for reaching the limiting grain size. In eq 2, the concept of limiting grain size is used to describe the isothermal grain growth of conventional polycrystalline materials, where the limiting grain size is larger than the initial grain size by several orders of magnitude. However, for nanocrystalline Mn2O3, the limiting grain size is still less than 100 nm and is only 1 order of magnitude larger than the initial grain size. In these experiments, for the longer annealing time, the grains have already attained the limiting size. Equation 2 might not be valid for describing the isothermal grain growth kinetics of nanocrystalline Mn2O3. It is reasonable to expect that another mechanism is involved for eq 3. The low activation energy obtained from eq 3 can be attributed to the large percentage, approximately 65−75% in the as-prepared powders, of highly disordered interfaces in Mn2O3 nanocrystals. As temperature or annealing time increases, structural relaxation of these interfaces takes place and reduces the degree of disorder. The energy barrier of this structural change arising from this local rearrangement of atoms is much lower than the energy barrier for diffusional grain growth. Thus, an alternative model is based on the assumption that the ordering of the interface regions in Mn2O3 nanocrystals occurs simultaneously with grain growth by structural relaxation. This structural relaxation model describes the grain growth kinetics satisfactorily and also yields a low activation energy of 32 kJ/mol appropriate for the rearrangement of atoms. This structure relaxation model not only describes the evolutions of grain growth well, but also makes reasonable attribution of the low activation energy to the short-range rearrangement of atoms in the interface region as well. 3.3.2. Microstructure Analysis. High-resolution transmission electron microscopy (HRTEM) observations indicate that the Mn2O3 nanoparticle sizes are about from 5 to 20 nm. The clear lattice fringes imply that the Mn2O3 nanoparticles are well crystallized.104 When the grain size of Mn2O3 nanocrystals decreases, the internal lattice fringes of the Mn2O3 nanoparticles show very good continuities (the regions labeled A as shown in Figure 5). However, it is worth noting that the lattice fringes of many nanoparticle edges show discontinuity with about 1 nm thickness to this edge (the regions labeled B as shown in Figure 5). This implies that the existence of surface oxygen dangling bonds enabling some middle energy levels to be relaxed after absorbing photons and releasing phonons, which may lead to the reduction in the intensities of green and ultraviolet (UV)-emission spectra with decreasing grain size of Mn2O3 nanocrystals. For example, two emission bands of Mn2O3 nanocrystals with different grain sizes (see Figure 6a, 20 nm; Figure 6b, 13 nm; and Figure 6c, 9 nm) can be observed at 560 (2.21 eV) and 380 nm (3.26 eV). It is found that the intensities of the clear green-emission and UV-emission decrease with decreasing grain sizes. The perplexing variation of this emission intensity is probably due to the fact that the smaller Mn2O3 particles show a lot of surface oxygen dangling bonds and possess the stronger reactive force. The room-

steady values after longer annealing time, we may be able to fit the data using eq 2. The data corresponding to annealing time longer than 2400 s are not included in the analysis because the quality of the fit to the longer times was not significantly better. The activation energy was found to be 44 kJ/mol. This apparent activation energy is very low compared with that of conventional materials. However, it agrees with those observed in other nanocrystalline ceramics, for example, approximately 41 kJ/mol for ZrO2−CeO2 ceramic.124 This low activation energy suggests that some mechanisms than diffusional boundary migration are controlling the grain growth process in nanocrystalline ceramics. In conventional polycrystalline materials, the volume fraction of grain boundaries is negligibly small while nanocrystalline materials can consist of over 50% of boundary regions, depending on the average grain size. The volume fraction (Ct) of interface component can be calculated with the formula, Ct = 3Δ/D,125 where D is the grain size, and Δ, usually taken as about 1 nm, is the average thickness of the interface. In this case, the volume fraction (Ct) remains up to 6% even after annealing at 500 °C for 2 h. The existence of a large number of interfaces and their changes during grain growth is thought to play an important role in grain growth processes of nanocrystalline materials. The nanocrystals size can significantly increase if the interface component gradually changes to crystalline state by ordering. The simplest ordering mechanism is structural relaxation by readjustment of Mn−O bond lengths and O−Mn−O bond angles. The low activation energies of these processes agree well with those found experimentally for grain growth in nanocrystalline materials. Therefore, the relaxation equation can be used to describe the grain growth kinetics126 ⎛ tm ⎞ X = 1 − exp⎜ − ⎟ = 1 − exp( −KTt m) ⎝ τ ⎠

(3)

where X is the relaxation parameter defined as X = (D − D0)/ (Dm − D0) and τ is the relaxation time. KT is a rate constant and can be expressed in an Arrhenius type equation: KT = 1/τ ∝ exp(−Q/RT). The exponent m in eq 3 is the relaxation order and equals to 1 for linear relaxation. Other symbols have the same definitions as those in eqs 1 and 2. Figure 4 shows the fitted lines for isothermal growth data using eq 3 and assuming m = 0.5. In general, this is because an average value of m was

Figure 4. Fitted lines according to the structural relaxation eq 3 for isothermal grain growth, while points are the experimental data and the lines are the fitted curves. Reprinted with permission from ref 61. Copyright 2005 Springer. 3838

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

the different types of defects in silicon oxide, for example, nonbridging oxygen holes center, on the surface of silicon nanocrystals. However, future work should strive to make the models more quantitative so that the origin of the blueemission and UV-emission can be firmly established. In the case of Mn2O3, we can reasonably speculate that the original cause may be due to the relaxation of some middle energy levels, such as the surface, crystal boundary and oxygen holes energy levels, after absorbing photons and releasing phonons. HRTEM observations also proves the existence of the surface oxygen dangling bonds, which demonstrate that the relaxation of some middle energy levels would result in the reduction in intensities of the green-emission and UV-emission spectra with decreasing the grain sizes. This behavior is consistent with the discontinuity of the lattice fringes of many particle edges in Mn2O3 nanocrystals. 3.4. Spectral Properties of Mn2O3 Nanocrystals

Figure 5. HREM image of Mn2O3 nanocrystals. The labeled A regions show the internal lattice fringes of Mn2O3 nanoparticles. The labeled B regions show the lattice fringes of Mn2O3 nanoparticles edges. Reprinted with permission from ref 103. Copyright 2002 Kluwer Academic Publishers.

3.4.1. Raman Features of Different Grain Sizes. Raman spectroscopy is commonly used in many research areas since vibrational information is specific to the chemical bonds and symmetry of molecules. Therefore, it would provide a fingerprint by which the molecules can be identified. Figure 7

Figure 6. Room-temperature green-emission and UV-emission spectra from Mn2O3 nanocrystals with different grain sizes: (a) 20, (b) 13, and (c) 9 nm. Reprinted with permission from ref 103. Copyright 2002 Kluwer Academic Publishers.

temperature UV-excitation spectra of Mn2O3 nanocrystals with different grain sizes also show similar behavior.104 This behavior is completely opposite to the previous studies from the blue-emission and UV-emission in porous silicon and ZnO nanocrystals,127−130 while the intensities of the blue-emission and UV-emission increased with decreasing grain sizes. Three possible models for the blue-emission of the porous silicon have been proposed:131 (i) band-to-band recombination in silicon nanocrystals; (ii) emission from the silicon oxide; and (iii) emission due to surface states. Among these models, the emission originated in the silicon oxide seems to be more acceptable due to the observed correlation between the intensities of the blue-emission and Si−O infrared absorption.131 Chen et al reported that the silicon quantum dots are dispersed orderly in SiO2 columns to the blue-emission sample,107,108 which is suggested to be responsible for the blue-emission. The origin of the UV-band may be linked with

Figure 7. Room-temperature Raman spectra of Mn2O3 nanocrystals with different grain sizes: (a) 34, (b) 21, (c) 20, and (d) 19 nm. Reprinted with permission from ref 47. Copyright 2006 Elsevier B.V.

shows the room-temperature Raman spectra for Mn2O3 nanocrystals with different grain sizes. Three peaks at 651.6, 267.0, and 176.0 cm−1 were observed, in agreement with literature reference values for the bulk Mn2O3.70 The peak at about 651 cm−1 is characteristic of the Raman spectra of Mn2O3, which confirms that the Mn2O3 nanocrystals possess the characteristics of the space group Ia3̅ structure.70,112 It is 3839

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

found that the intensity of the dominant Raman peak (about 650 cm−1) decreased clearly with decreasing average particle sizes of Mn2O3 nanocrystals. Among these vibrations, the symmetry modes are optically inactive but are infrared active. The remaining optic modes are Raman active in first order with the polarizability tensors.132 However, the Raman peak at 267 cm−1 disappears in smaller nanoparticles due to its low intensity with respect to the other modes. In fact, this observation might indicate that the Raman peak at 267 cm−1 is low compared to the one arising from peaks of the bulk material vibrations, consistent with the very small grain size. The experimental results reveal that the phonon Raman spectra are broadened (∼176 and ∼267 cm−1) in the lower frequency regions. However, the dominant Raman peak is only slightly broadened and the positions of these dominant Raman peaks remain unchanged with decreasing grain size. This behavior might be due to some growth defects, such as vacancies of oxygen, vacancy clusters, and local lattice disorder at the interface and surface of Mn2O3 nanocrystals. To investigate the possible reasons for the Raman spectra dependence with the grain size, the conservation of phonon momentum can be considered, which is very important in the study of the vibrational characteristics of various phonons. The existence of minor features in the lower frequency regions might be due to the partial breakdown of Raman selection rules. When the nanoparticle size decreases, the number of surface atoms increase rapidly. For example, surface atoms comprise 3% of the total number of atoms when the size is 100 nm, and 30% when it is 10 nm.133 The surface atoms have a very large number of dangling bonds, and coordination is not complete. With decreasing grain size, the diameter of the spherical particle is equivalent to that of the primitive cell. A large number of crystal boundaries and defects can be confined in space by crystal boundaries or defects resulting in the decay of phonons, and destruction or alteration in the conservation of phonon momentum. If the conservation of phonon momentum is not satisfied, the phonons with q ≠ 0 can contribute to the Raman spectrum. The additional transitions involving phonons with q ≠ 0 can lead to the Raman spectra in the lower frequency regions and the peaks broaden with decreasing grain size. 3.4.2. IR Features of Different Grain Sizes. Infrared (IR) is the subset of spectroscopy that deals with the infrared region of the electromagnetic spectrum. It covers a range of techniques, the most common being a form of absorption spectroscopy. As with all spectroscopic techniques, it can be used to identify compounds and investigate sample composition. IR spectroscopy is widely used in both scientific research and industrial applications as a simple and reliable technique for the quality control and dynamic measurement. It is also used in forensic analysis in both criminal and civil cases, e.g. enabling identification of polymer degradation. Figures 8a and b show the IR spectra for the as-prepared Mn2O3 nanocrystals (grain size: 9 nm) and standard Mn2O3 bulk materials,134 respectively. The spectrum of Figure 8a exhibits remarkable differences in the spectral shape and mean wavenumber from that of Figure 8b. The mean wavenumber of the IR spectrum in Mn2O3 nanocrystals (see Figure 8a) is higher than that of bulk materials (see Figure 8b). This phenomenon may be attributed to the microstructural evolution resulting from the surface disorder of Mn2O3 nanoparticles. As seen in Figure 8a, three peaks at 565.4, 691.5, and 1715.5 cm−1 are observed. It would be tempting to ascribe the 691.5 cm−1 peak to longitudinaloptical (LO) mode, and the 565.4 cm−1 peak to transverse-

Figure 8. Infrared spectra taken in the frequency range of 400−2000 cm−1 for (a) the as-prepared Mn2O3 nanocrystals and the (b) standard Mn2O3 bulk material (ref 133.). Reprinted with permission from ref 47. Copyright 2006 Elsevier B.V.

optical (TO) mode. The absorption peak centered at approximately 1715.5 cm−1 may correspond to CO2 absorbed on the surface of Mn2O3 nanoparticles.135 Figure 9 shows the IR spectra for Mn2O3 nanocrystals at different grain sizes. It is obvious that there is no remarkable difference in the features of the IR spectra among the samples whose grain size ranges from 11 to 19.5 nm. The positions of negative peaks are summarized in Table 2. Fox example, two peaks are observed in Figure 9a: one appears at 602.9 cm−1 and the other at 506.1 cm−1. The assignment of 602.9 cm−1 peak and 506.1 cm−1 peak in Figure 9a to the stretching vibrations of Mn−O units suggests that the 602.9 cm−1 peak might correspond to the asymmetric Mn−O− Mn stretching vibration, and the 506.1 cm−1 peak to the symmetric one. These peaks are assigned to the stretching vibration mode in which the Mn−O bond distance is modulated. Meanwhile, the peak position at 404.0 cm−1 is assigned to the bending vibration in which the Mn−O−Mn bond angle is also modulated. Moreover, the peak of the symmetric stretching mode at 506.1 cm−1 and the peak of the bending mode at 404.0 cm−1 should be ascribed to the surface phonon modes. This is because the surface modes result from 3840

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

electric dipole moment is created in the unit cell. Thus, the doubly degenerated polar modes are split into distinct TO and LO modes. However, one would expect increased TO-LO splitting when the nanocrystals become larger and larger. For the larger nanocrystals, in view of the long-range Coulomb interactions, one would indeed expect that local vibrations would be increasingly affected by dipolar interactions with remote vibrators, hence a larger shift. A realistic evaluation of these effects is complicated, and should probably take into account the geometry of actual nanocrystals. It is noted that, for the samples with small grain size, the intensity from the surface phonon is comparable with that from a bulky phonon. It is probable that the useful range of Coulomb interactions has already been reached for the 9-nm nanocrystals, and that the progressive appearance of the splitting could result from another effect. In general, it can be concluded that the surface phonon contribution increases with decreasing grain size.136 As the grain size continues to decrease, the influence of the surface regime enlarges. The surface phonon couples with the electron in the surface regime leading to the kinetic energy decrease of the tunneling electron.137−139 Comparing Figure 8a with Figure 9d, it is found that the spectrum of Figure 8a (grain size: 9 nm) exhibits remarkable differences from that of Figure 9d (grain size: 11 nm). In order to find some continuity between the series of IR spectra in Figure 9 and Figure 8a, it would be tempting to ascribe the 691.5 cm−1 peak of Figure 8a to the LO mode, and the 565.4 cm−1 peak to the TO mode. This would mean that the 477.0 cm−1 peak in Figure 9d vanishes from the 9-nm nanocrystals, and that the 429.4 cm−1 mode is less intense. Albello et al. proposed that the relaxation of the k = 0 selection rule is progressive when the rate of disorder increases or grain size decreases, and IR modes can become weakly active when the structural changes induced by disorder and grain size effects take place.140 Therefore, the Raman band at 506.1 cm−1 vanished from Figure 7 seems to correspond to IR active LOTO modes.141,142 It is reasonable to assign this mode to IR mode, whose Raman activities are induced by the grain size effect of the smaller nanocrystals. Moreover, when the grain size of Mn2O3 nanocrystals reaches 9 nm, the 506.1 cm−1 mode vanishes also from Figure 8a. The relaxation of the k = 0 selection rule may result from the grain size decrease. To sum up, disorder and nanoparticle size strongly influence the vibrational properties of this material. When the nanoparticle size is small, the vibrational characteristics of Mn2O3 nanocrystals are more complicated. 3.4.3. XPS Analysis of Different Grain Sizes. X-ray photoelectron spectroscopy (XPS) is a quantitative spectroscopic technique that measures the elemental composition, empirical formula, chemical state and electronic state of the elements that exist within a material. XPS spectra are obtained by irradiating a material with a beam of X-rays while simultaneously measuring the kinetic energy and number of electrons that escape from the top 1−10 nm of the material being analyzed. The dependence of electron structure on the grain size in Mn2O3 nanocrystals can be investigated by using this XPS technique.143 Figure 10 shows the O-1s core-level spectra for the samples of Mn2O3 nanocrystals with different average grain size. There is a dominant O-1s maximum at about 530 eV binding energy. It is found that the peak position at about 530 eV does not change significantly with decreasing average grain size within the range (0.9 eV) of energy resolution. That is, the O-1s peak energy for the 30-nm sample was 530.10 eV (see Figure 10a), 530.15 eV for the 20-nm

Figure 9. Infrared spectra taken in the frequency range of 350−800 cm−1 for Mn2O3 nanocrystals with different grain sizes: (a) 19.5, (b) 18, (c) 13, and (d) 11 nm. Reprinted with permission from ref 47. Copyright 2006 Elsevier B.V.

Table 2. Average Grain Sizes (D) of Mn2O3 Nanocrystals.a Reprinted with permission from ref 47. Copyright 2006 Elsevier B.V. Mn2O3 samples

(a)

(b)

(c)

(d)

D (nm) ν1 (cm−1)

19.5 602.9 506.1 404.0

18 597.3 486.3 420.0

13 599.7 482.9 420.0

11 585.7 477.0 429.4

ν2 (cm−1) a

ν1 and ν2 represent the stretching and bending modes, respectively.

the destroyed periodicities in some crystal planes and significant distortion of the lattices. The peak at 602.9 cm−1 should still be associated with the bulk vibrational mode. From Figure 9, we find that all the associated peaks are red-shifted with decreasing particle size, which might be due to different local stresses. With decreasing grain size of Mn2O3 nanocrystals, many defects, such as vacant lattice site, vacancy cluster, or local disorder at the surface and interface of Mn 2 O 3 nanocrystals, may cause lattice distortion and low lattice space symmetry, culminating in different local stresses. The remarkable feature of the IR spectra in Mn2O3 nanocrystals in the band shapes of 570−670 cm−1 is strongly aroused by transverse-optical (TO) and longitudinal-optical (LO) splitting. This splitting appears more and more evident with decreasing grain size. This might be due to the fact that long-range Coulomb field splits the degeneracy of polar modes in which an 3841

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

Figure 10. O-1s core-level XPS spectra of Mn2O3 nanocrystals with different grain sizes: (a) 30, (b) 20, (c) 13, and (d) 9 nm. Reprinted with permission from ref 143. Copyright 2004 Springer.

Figure 11. Mn-2p core-level XPS spectra of Mn2O3 nanocrystals with different grain sizes: (a) 30, (b) 25, and (c) 9 nm. Reprinted with permission from ref 143. Copyright 2004 Springer.

sample (see Figure 10b), 530.24 eV for the 13-nm sample (see Figure 10c), and 529.84 eV for the 9-nm sample (see Figure 10d). It is also found that the peak line shape seems to be asymmetric and a satellite structure exists with decreasing average grain size. It is noted that the O-1s core-level exhibits its major peak at 529.84 eV with a clear shoulder at about 2−3 eV higher binding energy for the sample with a small grain size (see Figure 10d). The half-peak widths were 2.02, 2.07, 2.79, and 2.83 eV for the 30-, 20-, 13-, and 9-nm samples, respectively. This indicates that the peak width increases with decreasing average grain size, and the shoulder peak becomes evident with decreasing average grain size. It is suggested that the appearance of the clear shoulder peak in Figure 10d is due to the surface effects of Mn2O3 nanocrystals.133 Surface atoms are different from the inner atoms with respect to the environment of the crystal field and the binding energy. They have a large number of dangling bonds, their coordination is not complete, and they combine other atoms very easily. Therefore, a clear shoulder peak significantly appears in the 9nm sample with decreaing average grain size. Figure 11 shows the Mn-2p core-level spectra for the samples of Mn2O3 nanocrystals with different average grain size. The doublet peaks of the Mn-2p core-level spectra appears at about 642 and 654 eV, while they are 642.28 and 654.07 eV for the 30-nm sample as shown in Figure 11a, 641.94 and 653.97 eV for the 25-nm sample as shown in Figure 11b, and 641.96 and 653.97 eV for the 9-nm sample as shown in Figure 11c. From this figure, it is seen that the main peak width becomes wide for samples with decreasing average grain size. The half-peak widths are about 3.83, 4.39, and 4.81 eV for the 30-, 25-, and 9nm samples, respectively. This suggests that the peak width increased with decreasing average grain size. Increase of the peak width could arise from the chemical changes around the

Figure 12. Mn-3s core-level XPS spectra of Mn2O3 nanocrystals with different grain sizes: (a) 9, (b) 13, and (c) 20 nm. Reprinted with permission from ref 143. Copyright 2004 Springer. 3842

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

result. The intensity ratio of the two peaks is given by the equation144

Mn sites. In particular, with decreasing average grain size, this leads to a decrease in oxygen coordination and oxygen displacement due to electric charge balance. This could cause a structure change such as oxygen ordering. According to our knowledge on trivalent manganese oxide, this structure originates from the effects of covalency of the manganicoxygen bond. The main peak at around 642 eV is due to the well-screened final state, and the satellite centered at around 654 eV is due to the poorly screened final state. The increase in the main peak width is due to the enhancement of the Mn3+ component induced by the grain size. Figure 12 shows the Mn3s core-level spectra for samples of Mn2O3 nanocrystals with different average grain size. It was found that the Mn-3s corelevel spectra split into two peaks, but the spectral intensity ratio (calculated in the following section) of the two peaks did not change relative to the decreasing grain size. For example, one appears at 84.33 eV and the other at 88.52 eV for the 9-nm sample as shown in Figure 12a. The 3s core-level spectra of the 13- and 20-nm samples also split into two peaks, at 83.45 and 89.56 eV as shown in Figure 12b, and at 83.24 and 88.77 eV as shown in Figure 12c, respectively. It is found that the main peak position is shifted to a higher binding energy with decreasing grain size. The shift in main peak toward high binding energy could have originated from the strong Mn−O bonding. The remarkable aspect of the Mn-3s core-level spectra in Mn2O3 nanocrystals is that the peak is multiply split and has a complex peak structure. It is believed that this is due to the 3s−3d exchange interaction of the Mn3+ ion. It is assumed that the 3s electron of the Mn3+ ion may leave an electron with parallel or antiparallel spin to that of the four 3d electrons as shown in Figure 13. In the ground state, the four 3d electrons are all

( 12 ) = S + 1 1 S I(S − ) 2 I S+

where S is the total spin of the unpaired electrons in the valence levels (4/2 in this case) and I is the intensity of the peak. For the 3s electron in Mn3+ ion, this theoretical ratio is

( 12 ) = S + 1 = 1 S I(S − ) 2 I S+

4 +1 2 = 1.5 4 2

However, the experimental results as shown in Figure 12 suggest that the actual ratio is about 2.0. The result shows the splitting is not proportional to the nominal ionic 3dconfiguration spin. This discrepancy may be attributed to an increase in s−d overlap with increasing oxidation state in the series of manganese oxides.64,144,145 The data for the Mn2O3 nanocrystals shown in Figure 12 illustrate the fact that this model predicts the correct trends of changes in spin states. This is also in accord with the increasing d-electron binding energy and the attendant reduction in the extent of the d-electron wave functions.64 3.4.4. ESR Analysis of Different Grain Sizes. Electron spin resonance (ESR) spectroscopy is a technique for studying chemical species that have one or more unpaired electrons, such as organic and inorganic free radicals or inorganic complexes possessing a transition metal ion. Since ESR is able to detect paramagnetic isolated species and gives information on the coordination of isolated sites, it can be used to detect Mn ion.114,146 Figure 14 shows the ESR patterns

Figure 13. Scheme of the 3s−3d exchange interaction in a Mn3+ ion. The initial state shows that the four 3d electrons are all unpaired, with parallel spins. The final state (1) shows that the spin of a 3s electron is parallel to that of the four 3d electrons, and the final state (2) shows that the spin of another 3s electron is antiparallel to that of the four 3d electrons. Reprinted with permission from ref 143. Copyright 2004 Springer.

Figure 14. ESR spectra of Mn2O3 nanocrystals with different grain sizes: (a) 9, (b) 13, (c) 22, (d) 25, and (e) 50 nm. Reprinted with permission from ref 102. Copyright 1997 Elsevier B.V.

unpaired and have parallel spins. Ionization of a 3s electron in the Mn3+ ion may occur, leaving an electron with parallel or antiparallel spin to that of the four 3d electrons, so that two final ionic states Mn4+ (parallel and antiparallel) will result (see the final states 1 and 2 of Figure 13). Because of the 3s−3d exchange interaction, the parallel state will have the lower binding energy. Thus, the core 3s electron ionization is only concerned with two final ion states. Therefore, two peaks will

of Mn2O3 nanocrystals with different particle sizes at (a) 9, (b) 13, (c) 22, (d) 25, and (e) 50 nm. It can be seen that the intensities of resonance peaks increased with decreasing particle size. Figure 14a shows the hyperfine structure of the ESR line. It is well-known that Mn2O3 possesses tetragonal crystal structure and the spectral term is 5D0. From the view of the crystal structure, the Mn3+ ion is situated in an octahedral crystal field, which is tetragonally distorted.133 The energy level could split into two items (E2g and T2g) in the crystal field as 3843

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

into wires, whiskers, or rods. The synthesis of nanostructures of functional oxides, with a controlled structure and morphology, is critical for scientific and technological applications. In recent years, the synthesis of nanostructures based on metal oxides has been widely demonstrated.175−178 In the past few years, the interface and surface microstructure of nanomaterials have been extensively and intensively investigated. Diversified types of interface−structure models have also been proposed for nanostructural materials,179−183 such as the gas-like model, order and extended order models. It is common that the peculiar properties of nanomaterials are explained in terms of the interface and surface structures, while the effects from the microstructure of the grains are neglected. In fact, for different preparation methods, the microstructure of nanomaterials can be nanocrystallite, nanoamorphous grain, or nanocluster agglomerations with some crystalline features. Since the grains are the basic components of nanomaterials, changes in their microstructure should inevitably change the physical and chemical properties.184,185 Therefore, the study of the microstructure of grains can help us reveal the general structure of nanomaterials and explain the corresponding experimental results. It is expected that the nanostrucural evolution may represent important building blocks for nanodevices and may offer exciting opportunities for both fundamental research and technological applications. Manganese oxides are of considerable interest in many technological applications, for example, electrochemical reaction and batteries due to their outstanding structural diversity combined with novel chemical and physical properties. Manganese oxide is also an important catalyst for removing carbon monoxide and nitrogen oxide from waste gas, and often is used to produce soft magnetic materials, such as manganese zinc ferrite.143 Many manganese-oxide-based compounds also exhibit typical colossal magnetoresistance behavior. Recently, various manganese oxides have aroused attention because of their unique catalysis, ion-exchange, electrochemical, molecular adsorption, and anomalous magnetic properties. These materials have also attracted interest recently as an electrochromic material of anodic coloration since they have a reversible color change from brown (colored state) to yellow (bleached state).116 The nucleation site and mechanism leading to growth of bulk-quantity Mn3O4 nanorods have been investigated.40 In general, Mn3O4 is known to be an active catalyst in several oxidation and reduction reactions, and can be used as a catalyst for the oxidation of methane and carbon monoxide186 or the selective reduction of nitrobenzene.187 More importantly, catalytic applications of different polymorphs of Mn3O4 (hausmannite) have been extended to the combustion of organic compounds at temperatures of the order of 373−773 K.124 These combustion catalytic technologies are of interest in relation to several air-pollution problems, enabling the control of the emission of NOx and volatile organic compounds from waste gases of different origins.125 The discovery of single-crystal Mn3O4 nanocrystals including nanoparticles, nanorods, and nanofractals may provide us with another kind of manganese oxide with different characteristics. The solid−liquid-solution (SLS) growth method has been successfully applied to cubic-structured InAs nanorods and InP nanowires, but there are only a few reports on the shapecontrolled synthesis of Mn3O4 nanocrystals including nanoparticles, nanorods and nanofractals. Usually, Mn3O4 has tetragonal structure with different polymorphs, but at higher

shown in Figure 15. The E2g shows the double degeneracy. The T2g shows the triple degeneracy. Because of the distoration of

Figure 15. Energy level splitting of Mn3+ ion in octahedral crystal field with tetragonal distorted. Reprinted with permission from ref 102. Copyright 1997 Elsevier B.V.

the crystal structure, the E2g level could further split into two items. Thus, the four paired 3d electrons are located on E2g energy level (S = 0), so it should have no ESR signal. When the intensity of the crystal field decreases with decreasing particle size, the splitted energy of the 3d level becomes smaller. When the particle size has decreased to a particular value, the configuration of the electron arrangement changes gradually to Figure 15c. When the splitted energy is smaller than the electron spin binding energy, the 3d electrons become unpaired (S = 4) as shown in Figure 15d and causes the ESR signal. From the shape and peak position of the ESR line, it can be concluded that this hyperfine structure of the ESR line must be the typical signal of the Mn2+ ion. Although the signal of the Mn2+ ion in Figure 15b is discernible, it is very weak. This result indicates that the Mn2O3 nanocrystals show a broad signal exhibiting “wings” on both sides and a poorly resolved hyperfine structure. The signal appears to be isotropic, which suggests an octahedral/tetrahedral symmetry of the paramagnetic Mn ion. Fine structure lines are found at higher and lower fields indicating a change in axial symmetry.

4. CONTROLLABLE SYNTHESIS AND MICROSTRUCTURE EVOLUTION OF MN3O4 NANOCRYSTALS 4.1. Progress in Mn3O4 Nanomaterials

Low-dimensional system such as nanocrystals, thin films, twodimensional heterostructures, clusters, and surface layers, demonstrates a variety of physical, chemical, and functional properties different from those of the bulk materials.29,147−153 Nanoscale one-dimensional microstructures have stimulated much interest because of their peculiar properties and potential applications as a result of their low dimensionality and the quantum confinement effect.102,154−160 Nanostructured materials have many potential technological applications because of their distinctive optical, mechanical, electrical, acoustic and magnetic properties.91,161−165 Considerable efforts have been devoted to research on the bulk-quantity synthesis of nanowires or nanorods using arc discharge, 166 laser ablation, 167 template,168 solution,169 vapor−liquid−solid,170 and other methods.171 The early synthesis techniques of low-dimensional nanomaterials include the use of photolithography and scanning tunneling microscopy.172,173 While these methods have provided nanomaterials for fundamental study, they are obviously not suitable for industrial applications. Screwdislocations174 and vapor−liquid−solid170 are two established mechanisms by which materials can grow one-dimensionally 3844

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

200 kV could be used to determine the low-dimensional distribution of Mn3O4 nanocrystals. High-resolution TEM (HRTEM) images of Mn3O4 nanocrystals could be obtained, for example, using a JEOL-2010 HRTEM with a point-to-point resolution 1.94 Å operating at 200 kV, and with electron energy loss spectroscopy (EELS). XRD could be performed, for example, with a Philips X’pert diffractometer using Cu Kα radiation (1.5418 Å) in reflection geometry, and aproportional counter with an operating voltage of 40 kV and a current of 40 mA. XRD patterns are recorded at a scanning rate of 0.05° s−1 in the 2θ ranges from 15° to 65°. Raman scattering measurements could be obtained by backscattering geometry with a SPEX-1403 laser Raman spectrometer. The excitation source could be an argon-ion laser operated at a wavelength of 514.5 nm in the backscattering configuration and a low incident power to avoid thermal effects. Fourier transform infrared (FTIR) spectrum (samples in a KBr flake) could be recorded with a Nicolet 700 spectrometer (λex = 514.5 nm, 30 mW) in the range 1500−400 cm−1 at room temperature. Electron spin resonance (ESR) spectra could be recorded on a Bruker model ER-200D-SRC instrument with microwave frequency 9.35 GHz.

temperature, the hausmannite crystal structure predominates. This is why the low-dimensional Mn3O4 nanocrystals including nanoparticles, nanorods and nanofractals are difficult to synchronously achieve these morphologies via a widely applicable synthetic route at lower temperature. In this section, we will discuss the probability of achieving this goal by using some experimental approaches developed over the past few years. We will present in detail a novel and simple approach to synchronously synthesize the single-crystal Mn3O4 nanoparticles, nanorods and nanofractals that exploits a one-step chemical reaction to prepare nanometer-diameter particles, rods, fractals under a lower heating temperature. This approach requires neither complex apparatus and sophisticated techniques nor metal catalysts or templates, as are usually needed in other methods. These findings indicate that other nanoparticles, nanorods or nanowires and nanofractals may be manipulated synchronously by using this simple technique, and might provide insight into new opportunities to control material fabrication. 4.2. Shape-Controlled Synthesis of Mn3O4 Nanocrystals

4.2.1. Synthesis Techniques. Single-crystal Mn3O4 nanocrystals including nanoparticles, nanorods, and nanofractals are prepared by a chemical liquid homogeneous precipitation method. The reactants for this chemical reaction process are: MnCl2·4H2O, H2O2, and NaOH. Here, the H2O2 solution is used as an oxidant, and a surfactant, C18H29NaO3S, is added to the reaction mixtures to prevent the agglomeration of nanocrystals. The fabrication procedure is described in detail as follows: an appropriate amount of MnCl2·4H2O is dissolved in distilled water, 2.5 M H2O2 solution, and an appropriate amount of surfactant C18H29NaO3S is added to the MnCl2 solution. Then, 2.5 M NaOH solution is dripped into the above system at three drip rates: rapid (about 120 drops/min), slow (about 60 drops/min), and very slow (about 30 drops/min). The heating-temperature is set at 200 °C for the chemical reaction and the growth of Mn3O4 nanocrystals. After the reaction, the suspension solution is kept for 2 h without disturbance for precipitation. The reaction equation is as follows: For Mn3O4 nanoparticles

4.3. Microstructure Evolution of Single-Crystal Mn3O4 Nanocrystals

4.3.1. Mn3O4 Nanoparticles. Figure 16 displays an XRD pattern of Mn3O4 nanoparticles prepared by the very slow

3MnCl2 + H2O2 + 6NaOH NaOH: ∼ 30 drops/min

Figure 16. Typical XRD pattern of Mn3O4 nanoparticles, which was obtained using Cu Kα radiation. The structure corresponds to a tetragonal hausmannite crystal structure model. Reprinted with permission from ref 51. Copyright 2006 Elsevier Ltd.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ Mn3O4 + 6NaCl + 4H2O

For Mn3O4 nanorods 3MnCl2 + H2O2 + 6NaOH

dripping speed of NaOH solution (about 30 drops/min) under heating-temperature at 200 °C for 2 h. This shows that the Mn3O4 nanoparticles have the same tetragonal structure as the bulk phase (International Center for Diffraction Data (ICDD), PDF File No. 24-0734). The lattice constant is determined to be a = 5.762 Å and c = 9.470 Å, which is consistent with the bulk value. All diffraction peaks can be successfully refined with the tetragonal hausmannite crystal structure model (space group I41/amd) of Mn3O4. No characteristic peaks of impurities, such as other forms of manganese oxides, are detected. The particle sizes are calculated from the XRD line widths B = Δ(2θ), taken as the full-width at half-maximum, using the Scherrer’s formula with K = 0.9. Crystalline domain sizes determined from the line widths of the diffraction peaks are somewhat smaller than those obtained by TEM measure-

NaOH: ∼ 120 drops/min

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ Mn3O4 + 6NaCl + 4H2O

For Mn3O4 nanofractals 3MnCl2 + H2O2 + 6NaOH NaOH: ∼ 60 drops/min

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ Mn3O4 + 6NaCl + 4H2O

4.2.2. Instrumental Analysis. The above precipitates are separated and dried at 100 °C. After grinding, it is put into a muffle furnace for heating at 200 °C for 2 h and the pure Mn3O4 black powders are obtained. The nanocrystals are characterized by low- and high-resolution transmission electron microscopy (TEM) and X-ray diffraction (XRD). For example, a Philips CM 20 TEM operating at an acceleration voltage of 3845

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

distribution, which shows an asymptotic centric distribution. A Gaussian fitting indicates that the maximal probability for crystallite diameter was about 13 nm. This result further demonstrates that the size distribution of the nanoparticles is in good agreement with the XRD results. 4.3.2. Mn3O4 Nanorods. Figure 18a shows a lowermagnification TEM bright-field image of a typical morphology

ments. With the measured full-width at half-maximum of the peaks and Scherrer’s formula, the average particle diameter is estimated to be about 12 nm, which is in good agreement with the roughly estimated value of 10−30 nm from the TEM micrograph (see Figure 17). Here, the individual particle

Figure 17. (a) Typical TEM image of Mn3O4 nanoparticles, (b) SAED pattern of Mn3O4 nanoparticle shown in the white square of panel a, and (c) the particle number distribution with a center size of about 13 nm, which was obtained using a Gaussian fitting. Reprinted with permission from ref 51. Copyright 2006 Elsevier Ltd.

Figure 18. (a) Lower-magnification TEM image of Mn3O4 nanorods, (b) the higher-magnification TEM image of a single Mn3O4 nanorod with 20 nm in width and larger than 550 nm in length, and (c) SAED pattern of a single nanorod shown in panel b. Reprinted with permission from ref 51. Copyright 2006 Elsevier Ltd.

diameter in the TEM micrograph is not clearly visible due to aggregation between Mn3O4 nanoparticles. Strain contributions to the peak broadening and corrections due to particle shape are not considered and might account for some of the discrepancy between the particle sizes determined by XRD as compared to TEM. In addition, Mn3O4 nanorods and nanofractals also have similar XRD patterns as those of the Mn3O4 nanoparticles. Figure 17a shows TEM bright-field image of a typical morphology distribution of Mn3O4 nanoparticles for the sample prepared by the very slow dripping speed of NaOH solution (about 30 drops/min) under heating-temperature at 200 °C for 2 h. It is evident that the sample consists of bulk-quantity nanoparticles. The average grain size found by intercept method ranged from 10 to 30 nm in diameters. From the TEM image, it is seen that the grain size of the Mn3O4 nanoparticles is quite small. Different regions of TEM image contrast of the particles indicates different density, which might be related to the different levels of grain size. It is found that the particle sizes observed in the TEM micrographs are larger than those estimated from XRD data, indicating an appreciable agglomeration of the particles with each other. TEM image reveals that the Mn3O4 nanoparticle is structurally perfect in geometrical shape. The chemical composition of the nanoparticles is determined by energy dispersive X-ray spectroscopy, which corresponds to Mn3O4. Figure 17b shows a typical selected area electron diffraction (SAED) pattern taken from a single nanoparticle (the white square in Figure 17a) with a diameter of about 15 nm. The SAED pattern indexed based on a tetragonal cell with lattice parameters of a = 5.760 Å and c = 9.469 Å, is consistent with the above XRD results. The SAED pattern also confirms that the nanoparticle is a single crystal of tetragonal Mn3O4. Figure 17c depicts the particle size

distribution of Mn3O4 nanorods for the sample prepared by the faster dripping speed of NaOH solution (about 120 drops/ min) under heating-temperature at 200 °C for 2 h. It is clear that the sample consists of bulk quantity of one-dimensional nanorods with lengths from several hundred nanometers to a few micrometers, and diameters from 10 to 30 nm. Figure 18b shows a higher-magnification TEM bright-field image of a single Mn3O4 nanorod with about 20 nm in width and larger than 550 nm in length. It can be seen that the Mn3O4 nanorod is very smooth, straight, and uniform. The geometrical shape is structurally perfect. The profile of the fringes implies that the geometrical shape of this nanostructure is likely to be a nanorod. Figure 18c shows a SAED pattern of a single Mn3O4 nanorod shown in Figure 18b. From SAED pattern, the crystal planes of (202), (020), and (222) can be determined. A moredetailed analysis can be made based on the highly magnified high-resolution TEM image (below). 4.3.3. Mn3O4 Nanofractals. Figure 19a shows TEM brightfield image of a typical Mn3O4 nanofractals for the sample prepared by the slow dripping speed of NaOH solution (about 60 drops/min) under heating-temperature at 200 °C for 2 h. It can be seen that the fractal branches of Mn3O4 nanofractal show a few micrometers in length and several hundred nanometers in width. With the change in the dripping speed of the NaOH solution, a large difference is observed in the TEM images (see Figures 17a, 18a, and 19a). This indicates that the dripping speed of the NaOH solution is a crucial factor in the preparation of Mn3O4 nanocrystals. It determines the chemical reaction rate, which has important effect on the growth of Mn3O4 nanocrystals. The fractal dimension, namely the scaling behavior, of the Mn3O4 nanofractals is measured by using the box-counting method.188 Figure 19b shows that the 3846

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

Figure 20. (a) Higher-magnification TEM image of Mn3O4 nanoparticles, (b) HRTEM image of Mn3O4 nanoparticle, which was shown in the white square of panel a, (c) HRTEM image of Mn3O4 nanorod, and (d) HRTEM image of Mn3O4 nanofractal. Reprinted with permission from ref 51. Copyright 2006 Elsevier Ltd.

Figure 19. (a) Typical TEM image of Mn3O4 nanofractals, (b) the plots of ln(N) versus ln(1/L) of nanofractals, where L is the box size and N represents for the number of boxes occupied by Mn3O4 nanocrystals, (c) SAED pattern of the fractal branch of Mn3O4 nanofractal, and (d) the histogram of branch-lengths of Mn3O4 nanofractals. Reprinted with permission from ref 51. Copyright 2006 Elsevier Ltd.

single Mn3O4 nanoparticle shown in the white square of Figure 20a. In HRTEM observations, these are mainly constituted of single-crystal domain Mn3 O 4 nanoparticles. On closer inspection, recurrent values of separation distance between lattice layers are found (in particular, 0.3089 nm, evidenced in the Figure 20b), which corresponds to lattice parameters of the tetragonal structure of the Mn3O4 hausmannite phase (arising from {112} planes). The above data support the theory that Mn3O4 segregates in a single phase belonging to the tetragonal Mn3O4 hausmannite structure, and that particular nanoparticle is a single-crystal with the {112} planes. Figure 20c represents a typical HRTEM image of a single Mn3O4 nanorod with about 9 nm in width and larger than 200 nm in length. The profile of the fringes implies that the geometrical shape of this nanostructure is likely to be a nanorod, where the smooth and straight shape of the nanorod is evident. The clear lattice fringes show that the nanorod is a single-crystal. The interplanar spacing is about 0.3089 and 0.4924 nm, which corresponds to {112} and {101} planes of tetragonal Mn3O4, and reveals that the growth planes of the nanorods are one of the longitudinal {112} planes and one of the lateral {101} planes. HRTEM observations also indicate that Mn3 O4 nanorods are structurally uniform. No dislocations or other planar defects are detected in the examined area of Mn3O4 nanorods. Figure 20d displays a typical HRTEM image of the top of a branch in Mn3O4 nanofractal. The interplanar spacing is about 0.2768 and 0.4924 nm, which corresponds to {103} and {101} planes of tetragonal Mn3O4, and reveals that the growth planes of the nanofractals are one of the longitudinal {103} planes and one of the lateral {101} planes. 4.3.5. Spectral Characteristics of Mn3O4 Nanocrystals. Transition metal oxides belong to a class of materials that are vitally important for developing new materials with function-

plots of ln(N) versus ln(1/L) of the fractal patterns corresponding Mn3O4 nanofractals, where L is the box size and N represents for the number of boxes occupied by Mn3O4 nanocrystals. It can be seen that all plots have good linearity, which means that the cluster morphologies have scaling invariance in these ranges. So the Mn3O4 nanocrystals in Figure 19a can be regarded as fractals. From the ln(N) versus ln(1/L) curve, the fractal dimension value is estimated to be 1.89. Figure 19c shows a SAED pattern of fractal branch of the Mn3O4 nanofractals shown in Figure 19a. From the SAED pattern, the crystal planes of (101) and (103) is determined. To illustrate the effect of the fractal branch width on the fragmentation propensity of Mn3O4 nanocrystals, Figure 19d shows the histogram of the branch-lengths of the Mn3O4 nanofractals. Treating the fractal as a set of consecutive bars and focusing on the main branches (defined as the longest continuous quasi-linear segments), this histogram depicts the number distribution of bars as a function of the bar length, showing a shift to smaller length as the branch width decreases. This indicates that the grown islands can be characterized by thicker branches, and they are not fragmented. 4.3.4. HRTEM Investigation of Mn3O4 Nanocrystals. Figure 20a shows the higher-magnification TEM image of a few Mn3O4 nanoparticles for the sample prepared by the very slow dripping speed of NaOH solution under heating-temperature at 200 °C for 2 h. It is evident that the geometrical shape of these nanoparticles mainly shows a tetragonal distribution, and a few of them are the spherical distribution. To provide further insight into the nanostructural evolution from these Mn3O4 nanocrystals, Figure 20b shows a typical HRTEM image of a 3847

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

by the previous report.65 To reduce multiple-loss contributions, the original core-loss spectra have been deconvoluted by lowloss spectra recorded on the same area in equivalent conditions. In Figure 21b, the manganese L2,3 spectra from the Mn3O4 nanorods are shown after deconvolution using the Fourier ratio technique and the backgrounds are subtracted using the standard AE−r model.190 Because of the structure of the EELS spectrum near the L2,3 edge due to excitations to bound states, as well as to the continuum, it is necessary to subtract the continuum contribution to determine the bound state contribution accurately. This method has been reported by Pearson et al in EELS spectra taken from transition-metal oxides with varying oxidation state of the metal, indicated by lines under the white lines.191 In general, Mn3O4 contains the mixed valence states of Mn2+ and Mn3+ and can be represented as MnO·Mn2O3. Ardizzone et al found that the external surface of Mn3O4 is 2MnO·MnO2.192 Therefore, the EELS spectrum of Mn3O4 may be the overlap of the spectra from Mn2+, Mn3+, and Mn4+ ions in Mn3O4. Every one of these ions contributes differently to the overall spectrum. Sparrow and Wang et al reported that the L3/L2 ratio of Mn3O4 was higher than that of Mn2O3,66,193 while Paterson reported a little lower L3/L2 ratio of Mn3O4,191 similar to the results of experiments performed by the authors of this paper. The difference may be caused by the different method used for sample preparation or different structures used for EELS analysis. In any case, the combination of the fine structure of the oxygen K-edge and the L3/L2 ratio of manganese may provide a characteristic way to analyze the oxidation state of manganese. Groot et al have systematically analyzed the prepake structure on the oxygen K-edge for the series of transition-metal oxides.67 They have pointed out that the prepake region may consist of one or several peaks which can be interpreted in terms of the ligand-field and exchange splitting. Figure 21c shows the oxygen K-edge spectrum after background subtraction for the Mn3O4 nanorods. The oxygen K-edge is dominated by three peaks and the fine structures can be interpreted in terms of transition processes governed by the dipole selection rule (Δl = ±1). The prepeak (a) around 530 eV has been attributed to the transition from the 1s oxygen core state to the 2p state hybridized with manganese 3d orbits.67,68 The second peak b is related to the projected unoccupied oxygen 2p states mixed with the manganese 4sp band at higher energy above the Fermi level.192 The third region including peaks c and d can be interpreted in terms of the multiple scattering of the excited electron with low kinetic energy. These features are sensitive to the local structure around the oxygen site, which means that the features are similar if crystal structures are identical.194 More precisely, the peaks c and d arising from multiple scattering within oxygen shells of increased size around the excited atom195 and their peak positions can be correlated to the interatomic distances between oxygen atoms through the resonance condition.196 To interpret the prepeak structures observed on the oxygen Kedges, several theoretical models have been proposed. Molecular-orbital theory is very useful for a qualitative interpretation of the features.67,68 Figure 22 shows that the molecular-orbital energy-level diagrams for (A) MnO610− representing the environment around Mn in MnO, (B) MnO69− representing the environment around Mn in Mn2O3, and (C) MnO68− representing the environment around Mn in MnO2. In this figure, only the ligand-field orbitals (3eg and 2t2g states) corresponding to the final states involved in the prepeaks are presented. One notes that the ligand-field splitting

ality and smartness. The unique properties of these materials are related to the presence of elements with mixed valences of transition elements. Electron energy loss spectroscopy (EELS) in the transmission electron microscope is a powerful technique for measuring the valences of some transition metal elements of practical importance. It has the unique advantage of high spatial resolution.189 In most transition-element compounds, atomiclike white-line features are clearly visible on the L2,3 edges, corresponding to transitions from the 2p1/2 and 2p3/2 to empty 3d orbits. EELS spectra of Mn3O4 nanorods have been fully studied and the results show that the position of the oxygen K and manganese L2,3 edges and the intensity ratio of white-lines, L3/L2 are characteristic of the oxidation state of the manganese ions. Figure 21a shows the EELS spectra taken from the Mn3O4 nanorods. It can be seen that the EELS spectrum shape is representative of the manganese oxidation state as pointed out

Figure 21. (a) EELS spectra of the typical Mn3O4 nanorods. (b) Corresponding Mn−L2,3 edge spectra after deconvolution and background subtraction. (c) Corresponding O−K edge spectra after deconvolution and background subtraction. Reprinted with permission from ref 55. Copyright 2006. American Institute of Physics. 3848

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

Mn3O4 nanorods, the formation process of the Mn3O4 nanorods could be divided into two stages from a chemical reaction point of view. In the first stage, the starting materials MnCl2 and H2O2 are dispersed to the surroundings of a suitable surfactant. In the second stage, the MnCl2 is oxidized by H2O2 to Mn3O4 in a suitable alkaline solution. Herein NaOH solution with the appropriate concentration is introduced into the above system. Although the chemical reaction is rather simple, the formation of the Mn3O4 nanorod involves a very complicated process because the nanorods could be formed only under a faster dripping speed of the NaOH solution (about 120 drops/ min). A possible Mn3O4 nanorod formation process is outlined as follows: MnCl2 is carried by the processing surfactants and contacts H2O2 under heating-temperature at 200 °C, where it deposits in the form of a liquid droplet. The liquidized MnCl2 then reacts with H2O2, when the NaOH solution is dripped quickly into the system, and forms Mn3O4, which further serves as some seeds (nucleation sites) for Mn3O4 nanorod growth. The above growth is most likely to be controlled by the SLS mechanism.198 However, interestingly, in the present SLS growth of the Mn3O4 nanorods, no additional transition metals are added as catalysts; it is, therefore, proposed that the formation of the Mn3O4 nanorods undergoes a self-catalyzed SLS growth process.198 The MnCl2 not only acts as the reactant but also provides an energetically favored site for the oxidation of H2O2. The newly formed Mn3O4 functions as a nanorod seed, which further grows to a Mn3O4 nanorod in the presence of MnCl2 and H2O2. The size and, hence, the uniformity of the Mn3O4 nanorods are predefined by the size of the liquid MnCl2 droplets.40 Experimental results indicate that the dripping speed of the NaOH solution plays an important role in the formation of the Mn3O4 nanocrystals. The difference in dripping speed of the NaOH solution leads to a large difference in Mn3O4 morphologies, which have been observed in transmission electron microscopy (Figures 17a, 18a, and 19a). The surfactant, C18H29NaO3S, could aid the formation of fine particles when the NaOH solution was very slowly dripped into the above system (about 30 drops/min) because it forms a “shell” surrounding the particles and prevents the fine particles from aggregating to larger particles. When the NaOH solution is quickly dripped into the above system (about 120 drops/ min), the added NaOH significantly decreases the viscosity of the melt, and thus the formation of the “shell” surrounding the particles in the flux becomes difficult. Hence, the formation of Mn3O4 nanorods becomes easier. As a consequence, the formation of Mn 3 O 4 nanofractals originates from the competition of the Mn3O4 nanoparticles and nanorods formation is governed by the NaOH solution dripping speed. That is why the Mn3O4 nanorods can grow only in the system in which NaOH solution is quickly dripped (about 120 drops/ min).

Figure 22. Molecular-orbital energy-level diagrams for (a) MnO610− represents the environment around Mn in MnO, (b) MnO69− represents the environment around Mn in Mn2O3, and (c) MnO68− represents the environment around Mn in MnO2. Reprinted with permission from ref 55. Copyright 2006. American Institute of Physics.

(the energy separation between 2t2g and 3eg states) increases with the formal oxidation state of the manganese ion, while the exchange splitting (the energy difference between spin-up and spin-down states) decreases. In Figure 22a, there is no splitting in prepeak region although two unoccupied 2t2g↓ and 3eg↓ states are separated. The absence of peak splitting may be related to the relative strength of the transitions to both states, which is mainly governed by the 2p-hole population in each unoccupied state. In Figure 22b, a triply peak structure is expected in the prepeak region on the basis of the accessible unoccupied final states, which is attributed to the transition to 3eg↑, 2t2g↓, and 3eg↓ unoccupied states. In Figure 22c, only two peaks may be observed in the prepeak region. The first peak can be assigned to transition to 2t2g↓ and 3eg↑ states because these two states correspond to similar energy positions. The second shoulder peak can be attributed to the transition to 3eg↓ state. Thus, the relative intensity of each peak seems to be satisfactorily correlated to the number of 2p holes in the unoccupied ligand-field orbitals corresponding to the final state of each peak. This reinforces the importance of hybridization effect between manganese 3d and oxygen 2p orbitals for the Mn2O3 and MnO2 cases. Therefore, the usefulness of molecular-orbital analysis for a semiquantitative interpretation of the oxygen prepeak structures plays an important role when considering electronic structures of the Mn3O4 nanorods. 4.3.6. Formation Mechanism of Mn3O4 Nanocrystals. Further advancement of this approach to nanocrystals synthesis requires a clear understanding of the growth mechanism. Experimental results by the present authors show that only Mn3O4 nanorods are produced when the NaOH solution is quickly dripped into the above system (about 120 drops/min) (Figure 18a). When the NaOH solution is dripped into the above system very slowly or slowly (about 30 drops/min or 60 drops/min, respectively), Mn3O4 nanoparticles (Figure 17a) or the Mn3O4 nanofractals (Figure 19a) are produced respectively. These results indicate that the growth mechanism of the Mn3O4 nanorods in our approach is different from that in the usual “vapor−liquid−solid (VLS)” or “solution−liquid−solid (SLS)” mechanisms. Figures 18a and b show the TEM images of the tips of the as-synthesized Mn3O4 nanorods. There are no spherical droplets, which are known to be good evidence for the VLS or SLS growth mechanisms, at the tips of these nanorods (tips indicated by the white arrows in Figures 18a and b). This observation suggests that the Mn3O4 nanorods did not grow by the VLS or SLS mechanisms proposed for nanorods grown by catalytic technique, in which a liquid metal droplet is located at the growth front of the rods and acts as the catalytic active site.197 From the above studies on the formation of the

5. SUMMARY AND OUTLOOK 5.1. Summary

This review introduces in detail the current state of the synthesis, characterization and microstructure of manganese oxide architectures including Mn2O3 and Mn3O4. Manganese oxide nanocrystals are promising strategic materials in the fabrication of novel nanostructures with controllable size, morphology, and properties. With recent advances in the field of transition metal oxide nanofabrication techniques, smart artificial manganese oxide nanocrystals would attract increasing 3849

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

attention and find many uses in a variety of applications including ion-exchange, molecular adsorption, catalysis, electrochemical reaction, batteries, optics, electronics, magnetism, mechanics, acoustics, and sensing. Some of these properties are tunable based on the adjustment of the related fabrication processes and the structural parameters of the nanostructures. Because of these interesting and tunable properties, some of the fabricated nanostructures are close to being incorporated into micro/nanodevices. Many research efforts have been devoted to developing new strategies for fabricating nanostructures owing to their widespread potential applications. Our focus in this review has been limited to those solution−liquid−solid chemical reaction strategies in which the solution−liquid−solid mechanism has been extensively and intensively studied and the nanocrystals formation processes are generally applicable to many materials. The literature discussed in this review represents part of the challenging ongoing work aimed at elucidating a detailed understanding of the interaction between artificial nanostructures and macroscopic systems. This area of research is still in its infancy, both with respect to the reproducible synthesis of nanostructures and understanding the detailed mechanisms that influence material characterization and advanced functions, so there is still much insight to be gained. It is also a young and quickly growing research field located at the crossroads of physics, chemistry, materials science, biology, and nanotechnology. The approaches described employ solution−liquid−solid chemical reaction with diverse morphologies at multiscales, and yield nanomaterials that inherit the natural morph structures. Therefore, unique and amazing properties can be expected from manganese oxide materials as a result of combining microstructures with synthetic techniques.

reliable synthesis of nanostructures with high uniformity and controlled dimension, microstructure, composition, location, orientation, and surface/interface states remains a challenge, and will inevitably influence the feasibility and reproducibility of the micro/nanodevices. Therefore, the new methodologies are still needed to further improve the synthesis techniques. Third, some of detailed problems should be solved, which include clarification of the microstructures and functional relationship in nanostructural systems and manganese oxide materials, development of an effective method for controllable nanomaterial fabrication on a large scale, exploration of new microstructure features and functions to produce novel materials in manganese oxide architectures. Fourth, the challenging issue for future research in the field of the manganese oxides is the fabrication of the related advanced composite architectures with novel and improved properties. Most surface and interface nanostructures prepared using the chemical reaction method, such as the nanoparticle, nanorods, and nanofractals, can be good functional parts of advanced composite architectures. 5.3. Concluding Remarks

In summary, this review has provided the preparation methodologies and the microstructural characteristics concerning several types of manganese oxides including Mn2O3 and Mn3O4 nanocrystals: nanoparticles, nanorods, and nanofractals prepared by chemical liquid homogeneous precipitation method.

AUTHOR INFORMATION Corresponding Author

*Tel.: +86 21 66137503. Fax: +86 21 66137787. E-mail: [email protected] (Z.W.); [email protected] (Z.J.); mhwu@ staff.shu.edu.cn (M.W.).

5.2. Outlook

Notes

Research on manganese oxide materials will bring improvements to techniques, such as microstructure and morphology associated technique, microstructure and characterization coupling technique, microstructure and advanced function simulation technique, etc. To achieve optimized catalysis performance with the manganese oxide nanocrystals, however, several challenges still remain. Further investigation and additional effort are needed to tackle the following important issues for promoting the practical applications of manganese oxide nanostructures: First, as a new research field, some basic and important characteristics of manganese oxides have not been fully investigated. So far, property tuning of the controllable nanostructural morphologies based on the adjustment of size, spacing, and shape of the nanostructures has not been systematically studied. The adjustable structural parameters of the fabricated nanostructures are among the most attractive advantages in manganese oxide nanocrystals. By changing the size, spacing, and shape of the nanostructures, the properties of nanostructural manganese oxide materials can be adjusted, hence realizing an active property tuning of the nanostructures. Future research of manganese oxide materials should focus on property tuning based on the adjustment of structural parameters of the nanostructures. Second, the properties of nanostructures are determined by their geometries, stoichiometric compositions, microstructures, and surface states. Although various methods have been employed to synthesize manganese oxide nanostructures,

The authors declare no competing financial interest. Biographies

Zhiwen Chen was born in Hefei, Anhui, China, and received his MS degree three years in Inorganic Chemistry from the University of Science and Technology of China (USTC) and a PhD degree in Materials Physics and Chemistry from the City University of Hong Kong (CityUHK). Since 1993, he has worked at USTC as an Assistant Professor, Associate Professor. He joined the Teikyo University in Japan as a Research Fellow in 1999 and the CityUHK as a Research Fellow and Senior Research Fellow in 2001 and 2007, respectively, and is now a Chartered Professor at the Shanghai University, China. His 3850

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

research interests mainly focus on the controllable-synthesis, characterization, and properties of advanced functional materials and oxides nanomaterials. He has published over 130 papers in refereed journals. He served as a Guest Editor, the member of Editorial Board, and Reviewer or Referee for many international journals, and is now Editor-in-Chief of Advances in Materials Physics and Chemistry, Website: http://www.scirp.org/journal/ampc.

Zhen Li was born in 1972 in Hunan, China. She received her MS and PhD degrees form the Shanghai University in 2001 and 2008, respectively. She was appointed as a Professor at the Shanghai University in 2010. Her research interests focus on the semiconductor nanostructured materials. She has published more than 30 papers in international journals. Zheng Jiao was born in 1972 in Peking, China. After he got his PhD degree in Inorganic Chemistry from the University of Science and Technology of China in 2000, he worked at the Osaka University in Japan from 2002 to 2003 and at the Lille University of Technology in France in 2004. Now he worked at the Shanghai University in China as a full professor. His research interests are focused on the functional materials and applications in environmental and biochemistry science.

Minghong Wu earned her PhD from the Chinese Academy of Science in 1999. After some years of postdoctoral research in Japan, she became a Professor in 2002 at the School of Environmental and Chemical Engineering of Shanghai University in China. She was the Distinguished Young Scientist supported by the National Natural Science Foundation of China. Her main research interests include the environmental chemistry, nanostructures of sensors, and biomedical materials. So far she has authored more than 130 publications and patents.

Dengyu Pan is now a Professor of Chemistry in the Institute of Nanochemistry and Nanobiology, Shanghai University, China. He received his PhD degree from the University of Science and Technology of China in 2003 under the supervision of Professor J. G. Hou. From 2003 to 2006, he pursued postdoctoral research in the group of Professor G. H. Wang in Nanjing University, China. His research interests are centered on the synthesis and applications of nanomaterials.

Chan-Hung Shek got his PhD in 1994 from the Department of Mechanical Engineering at the University of Hong Kong. He then joined City University of Hong Kong as an Assistant Professor and became an Associate Professor in 2001. Since 2007, he has been 3851

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

(Project Numbers: 11074161, 11025526 and 41073073), Science and Technology Commission of Shanghai Municipality (Project Numbers: 10JC1405400, 09530501200), and Shanghai Leading Academic Discipline Project (Project Number: S30109). This work was also supported by a strategic research grant (Project Number: 7002554) and a grant from the City University of Hong Kong (Project Number: 7002657).

appointed as Assistant Head of Department of Physics and Materials Science, and in 2011, he was concurrently appointed as Assistant Dean of College of Science and Engineering. He major research interests are the correlation among processing, microstructures and properties of materials, including nanostructured metal oxides, bulk metallic glasses, and conventional engineering alloys.

LIST OF CLHP XRD ICDD SEM TEM HRTEM SAED EDS EELS UV XPS IR ESR SLS VLS D LO TO

C. M. Lawrence Wu received his PhD from the University of Bristol, U.K., in 1986 and has been working in City University of Hong Kong since 1987. He is currently a Professor in the Department of Physics and Materials Science. Apart from working on the synthesis and characterization of nanocrystals, he also works on plasmonic effects in the application of biosensing and solar energy.

ABBREVIATIONS chemical liquid homogeneous precipitation X-ray diffraction international center for diffraction data scanning electron microscopy transmission electron microscopy high resolution transmission electron microscopy selected area electron diffraction energy dispersive X-ray spectroscopy electron energy loss spectroscopy ultraviolet X-ray photoelectron spectroscopy infrared electron spin resonance solution−liquid−solid vapor−liquid−solid dimension longitudinal−optical transverse−optical

REFERENCES (1) Wang, Z. L. Characterization of Nanophase Materials; Wiley-VCH: New York, USA, 2000. (2) Hochbaum, A.; Yang, P. Chem. Rev. 2010, 110, 527. (3) Ying, J. Y. Chem. Eng. Sci. 2006, 61, 1540. (4) Thota, S.; Prasad, B.; Kumar, J. Mater. Sci. Eng., B 2010, 167, 153. (5) Liu, L.; Yang, Z.; Liang, H.; Yang, H.; Yang, Y. Mater. Lett. 2010, 64, 891. (6) Dorman, J. L.; Fiorani, D.; Tronc, E. Adv. Chem. Phys. 1997, 98, 283. (7) Hadjipanayis, G. C.; Prinz, G. A. Science and Technology of Nanostructurated Materials; Plenum: New York, 1991. (8) Chantrell, R. W.; O’Grady, K. Applied Magnetism; Kluwer Academic: Dordrecht, the Netherlands, 1994. (9) Wang, Z. H.; Geng, D. Y.; Hu, W. J.; Ren, W. J.; Zhang, Z. D. J. Appl. Phys. 2009, 105, 07A315. (10) Bayer, V.; Podloucky, R.; Franchini, C. Phys. Rev. B 2007, 76, 165428. (11) Yang, P.; Fardy, M.; Yan, R. Nano Lett. 2010, 10, 1529. (12) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.; Kim, F.; Yan, H. Adv. Mater. 2003, 15, 353. (13) Chen, Z. W.; Pan, D. Y.; Zhao, B.; Ding, G. J.; Jiao, Z.; Wu, M. H.; Shek, C. H.; Wu, C. M. L.; Lai, J. K. L. ACS Nano 2010, 4, 1202. (14) Audi, A. A.; Sherwood, P. M. A. Surf. Interface Anal. 2002, 33, 274. (15) Lin, C. K.; Chuang, K. H.; Lin, C. Y.; Tsay, C. Y.; Chen, C. Y. Surf. Coat. Technol. 2007, 202, 1272. (16) Brus, L. Appl. Phys. A: Mater. Sci. Process. 1991, 53, 465. (17) Kastner, M. A. Phys. Today 1993, 46, 24. (18) Lewis, L. N. Chem. Rev. 1993, 93, 2693. (19) Freer, R. Nanoceramics; Institute of Materials: St. James,London, 1993. (20) Chen, C. Y.; Lin, C. K.; Tsai, M. H.; Tsay, C. Y.; Lee, P. Y.; Chen, G. S. Ceram. Int. 2008, 34, 1661. (21) Awschalom, D. D.; DiVincenzo, D. P. Phys. Today 1995, 48, 43. (22) Smyth, J. F. Science 1992, 258, 414. (23) Yang, H.; Zhuang, Y.; Hu, H.; Du, X.; Zhang, C.; Shi, X.; Wu, H.; Yang, S. Adv. Funct. Mater. 2010, 20, 1733.

Joseph K. L. Lai won an open scholarship to study Physics at Keble College, Oxford University in 1971, and graduated with first class honours in 1974. He joined the Central Electricity Research Laboratories (CERL) in U.K. after graduation and was appointed Project Leader of the Remaining Life Study Group in 1984. While working at CERL, he studied for a PhD at the City University, London, U.K., and was awarded the degree in 1982. He returned to Hong Kong in 1985 and has worked at CityU ever since. He is now Chair Professor of Materials Science in the Department of Physics and Materials Science. Professor Lai is a Fellow of the Hong Kong Institution of Engineers, a Chartered Engineer, and a Fellow of the following UK professional institutions: Institute of Materials, Minerals and Mining, Institute of Physics, and Institution of Mechanical Engineers.

ACKNOWLEDGMENTS The work described in this article was financially supported by the Shanghai Pujiang Program (Project Number: 10PJ1404100), China, Key Innovation Fund of Shanghai Municipal Education Commission (Project Number: 10ZZ64), National Natural Science Foundation of China 3852

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

(24) Li, P.; Nan, C.; Wei, Z.; Lu, J.; Peng, Q.; Li, Y. Chem. Mater. 2010, 22, 4232. (25) Na, C. W.; Han, D. S.; Kim, D. S.; Park, J.; Jeon, Y. T.; Lee, G.; Jung, M. H. Appl. Phys. Lett. 2005, 87, 142504. (26) Winkler, E.; Zysler, R. D.; Fiorani, D. Phys. Rev. B 2004, 70, 174406. (27) Francis, T. M.; Lichty, P. R.; Weimer, A. W. Chem. Eng. Sci. 2010, 65, 3709. (28) Kodama, R. H.; Berkovitz, A. E.; McNiff, E. J.; Foner, S. Jr. Phys. Rev. Lett. 1996, 77, 394. (29) Kodama, R. H.; Makhlouf, S. A.; Berkovitz, A. E. Phys. Rev. Lett. 1997, 79, 1393. (30) Kodama, R. H.; Berkovitz, A. E. Phys. Rev. B 1999, 59, 6321. (31) Kachkachi, H.; Noguès, M.; Tronc, E.; Garanin, D. A. J. Magn. Magn. Mater. 2000, 221, 158. (32) Iglesias, O.; Labarta, A. Phys. Rev. B 2001, 63, 184416. (33) De Biasi, E.; Ramos, C. A.; Zysler, R. D.; Romero, H. Phys. Rev. B 2002, 65, 144416. (34) Zysler, R. D.; Romero, H.; Ramos, C. A.; De Biasi, E.; Fiorani, D. J. Magn. Magn. Mater. 2003, 266, 233. (35) Wang, Z. H.; Geng, D. Y.; Zhang, Y. J.; Zhang, Z. D. J. Cryst. Growth 2008, 310, 4148. (36) Musić, S.; Ristić, M.; Popović, S. J. Mol. Struct. 2009, 924−926, 243. (37) Jiao, F.; Harrison, A.; Hill, A. H.; Bruce, P. G. Adv. Mater. 2007, 19, 4063. (38) Pike, J.; Hanson, J.; Zhang, L.; Chan, S. W. Chem. Mater. 2007, 19, 5609. (39) Zarur, A. J.; Ying, J. Y. Nature 2000, 403, 65. (40) Chen, Z. W.; Lai, J. K. L.; Shek, C. H. Appl. Phys. Lett. 2005, 86, 181911. (41) Zhang, W.; Yang, Z.; Liu, Y.; Tang, S.; Han, X.; Chen, M. J. Cryst. Growth 2004, 263, 394. (42) Ramesh, K.; Chen, L.; Chen, F.; Zhong, Z.; Chin, J.; Mook, H.; Han, Y. F. Catal. Commun. 2007, 8, 1421. (43) Marbán, G.; Valdés-Solís, T.; Fuertes, A. B. J. Catal. 2004, 226, 138. (44) Folch, B.; Larionova, J.; Guari, Y.; Guérin, C.; Reibel, C. J. Solid State Chem. 2005, 178, 2368. (45) Tang, X.; Li, J.; Sun, L.; Hao, J. Appl. Catal., B 2010, 99, 156. (46) Sun, S. H.; Zeng, H. J. Am. Chem. Soc. 2002, 124, 8204. (47) Chen, Z. W.; Lai, J. K. L.; Shek, C. H. J. Non-Cryst. Solids 2006, 352, 3285. (48) Ren, T. Z.; Yuan, Z. Y.; Hu, W.; Zou, X. Microporous Mesoporous Mater. 2008, 112, 467. (49) Shao, C.; Guan, H.; Liu, Y.; Li, X.; Yang, X. J. Solid State Chem. 2004, 177, 2628. (50) Ge, J. P.; Hu, Y. X.; Biasini, M.; Beyermann, W. P.; Yin, Y. D. Angew. Chem., Int. Ed. 2007, 46, 4342. (51) Chen, Z. W.; Lai, J. K. L.; Shek, C. H. Scr. Mater. 2006, 55, 735. (52) Yuan, Z. Y.; Ren, T. Z.; Du, G.; Su, B. L. Chem. Phys. Lett. 2004, 389, 83. (53) Kim, D. K.; Muralidharan, P.; Lee, H. W.; Ruffo, R.; Yang, Y.; Chan, C. K.; Peng, H.; Huggins, R. A.; Cui, Y. Nano Lett. 2008, 8, 3948. (54) Shen, Y. F.; Zerger, R. P.; Deguzman, R. N.; Suib, S. L.; Mccurdy, L.; Potter, D. I.; O’Young, C. L. Science 1993, 260, 511. (55) Chen, Z. W.; Lai, J. K. L.; Shek, C. H. J. Chem. Phys. 2006, 124, 184707. (56) Han, Y. F.; Chen, F.; Zhong, Z. Y.; Ramesh, K.; Widjaja, E.; Chen, L. W. Catal. Commun. 2006, 7, 739. (57) Seo, W. S.; Jo, H. H.; Lee, K.; Kim, B.; Oh, S. J.; Park, J. T. Angew. Chem., Int. Ed. 2004, 43, 1115. (58) Yang, L. X.; Zhu, Y. J.; Tong, H.; Wang, W. W.; Cheng, G. F. J. Solid State Chem. 2006, 179, 1225. (59) Yang, Z.; Zhang, W.; Wang, Q.; Song, X.; Qian, Y. Chem. Phys. Lett. 2006, 418, 46. (60) Shu, Q.; Zhang, J.; Yan, B.; Liu, J. Mater. Res. Bull. 2009, 44, 649.

(61) Chen, Z. W.; Shek, C. H.; Lai, J. K. L. Appl. Phys. A: Mater. Sci. Process. 2005, 80, 703. (62) Hosono, E.; Kudo, T.; Honma, I.; Matsuda, H.; Zhou, H. Nano Lett. 2009, 9, 1045. (63) Isber, S.; Majdalani, E.; Tabbal, M.; Christidis, T.; Zahraman, K.; Nsouli, B. Thin Solid Films 2009, 517, 1592. (64) Wertheim, G. K.; Hüfner, S.; Guggenheim, H. J. Phys. Rev. B 1973, 7, 556. (65) Rask, J. H.; Miner, B. A.; Buseck, P. R. Ultramicroscopy 1987, 21, 321. (66) Sparrow, T. G.; Williams, B. G.; Rao, C. N. R.; Thomas, J. M. Chem. Phys. Lett. 1984, 108, 547. (67) De Groot, F. M. F.; Grioni, M.; Fuggle, J. C.; Ghijsen, J.; Sawatzky, G. A.; Petersen, H. Phys. Rev. B 1989, 40, 5715. (68) Grunes, L. A.; Leapman, R. D.; Wilker, C. N.; Hoffmann, R.; Kunz, A. B. Phys. Rev. B 1982, 25, 7157. (69) Villars, P.; Calvert, L. D. Pearson’s Handbook of Crystallographic Data for Intermetallic Phase; ASM International: Materials Park, OH, 1991. (70) Busca., G. Phys. Chem. Chem. Phys. 1991, 1, 723. (71) West, A. R. Solid State Chemistry and its Applications; Wiley: New York, 1984. (72) Wells, A. F. Structural Inorganic Chemistry; Clarendon Press: Oxford, U.K., 1984. (73) Haber, J. in: Catalysis, Science and Technology; Anderson, J. R.. Boudart, M., Eds.; Springer Verlag: Berlin, 1981; Vol. 2, p 13. (74) Vogel, W. Chemistry of Glass; The American Ceramic Society: Columbus, OH, 1985. (75) Sanderson, R. T. Chemical Bonds and Bond Energy; Academic Press: New York, 1976. (76) Yachandra, V. K.; Sauer, K.; Klein, M. P. Chem. Rev. 1996, 96, 2927. (77) Moore, G. J.; Portal, R.; Salle, A.; Guyomard, D. J. Power Sources 2001, 97, 393. (78) Sato, K.; Yamamoto, A.; Nomura, M. Jpn Patent 9035915,1990. (79) Kapoor, I. P. S; Srivastava, P.; Singh, G. Propellants Explos. Pyrotech. 2009, 34, 351. (80) Huo, Y.; Zhang, Y.; Xu, Z.; Zhu, J.; Li, H. Res. Chem. Intermed. 2009, 35, 791. (81) Mukherjee, S.; Pal, A. K.; Bhattacharya, S.; Raittila, J. Phys. Rev. B 2006, 74, 104413. (82) Marignier, J. L.; Belloni, J.; Delcourt, M. O.; Chevalier, J. P. Nature 1985, 317, 344. (83) Hegedus, L.; Beckman, J. W.; Pan, W. H.; Solor, J. P. European Patent 1989, 345695. (84) Berry, F. J. Adv. Catal. 1982, 30, 97. (85) Giuntini, J. C.; Granier, W.; Zanchetta, J. V.; Taha, A. J. Mater. Sci. Lett. 1990, 9, 1383. (86) Golosovsky, I. V.; Salazar-Alvarez, G.; López-Ortega, A.; González, M. A.; Sort, J.; Estrader, M.; Suriňach, S.; Baró, M. D.; Nogués, J. Phys. Rev. Lett. 2009, 102, 247201. (87) Kim, M.; Chen, X. M.; Joe, Y. I.; Fradkin, E.; Abbamonte, P.; Cooper, S. L. Phys. Rev. Lett. 2010, 104, 136402. (88) Chugai Electric Industrial Co., Ltd. Jpn Patent 82209833, 1982. (89) Thackeray, M. M. Solid State Chem 1997, 25, 1. (90) Zhang, S. Y.; Chen, Z. W.; Tan, S.; Wang, J.; Jin, S. Z. Nanostruct. Mater. 1997, 8, 719. (91) Uyeda, R. Prog. Mater. Sci. 1991, 35, 1. (92) Wang, L. C.; He, L.; Liu, Y. M.; Cao, Y.; He, H. Y.; Fan, K. N.; Zhuang, J. H. J. Catal. 2009, 264, 145. (93) Walanda, D. K.; Lawrance, G. A.; Donne, S. W. J. Solid State Chem. 2009, 182, 1336. (94) Gleiter, H. Prog. Mater. Sci. 1989, 33, 223. (95) Gouda, G. M.; Nagendra, C. L. Sens. Actuat. A 2009, 155, 263. (96) Han, Y. F.; Chen, L.; Ramesh, K.; Widjaja, E.; Chilukoti, S.; Surjami, I. K.; Chen, J. J. Catal. 2008, 253, 261. (97) Yang, L. X.; Zhu, Y. J.; Tong, H.; Wang, W. W. Ultrasonics Sonochem. 2007, 14, 259. 3853

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

(98) Chandra, N.; Bhasin, S.; Sharma, M.; Pal, D. Mater. Lett. 2007, 61, 3728. (99) Liu, Y. P.; Qian, Y. T.; Zhang, Y. H.; Zhang, M. W.; Chen, Z. Y.; Yang, L.; Wang, C. S.; Chen, Z. W. Mater. Lett. 1996, 28, 357. (100) Chen, Z. W.; Zhang, S. Y.; Tan, S.; Wang, J.; Jin, S. Z. Mater. Res. Bull. 1999, 34, 1583. (101) Škapin, S. D.; Kunej, Š.; Suvorov, D. J. Euro. Ceram. Soc. 2008, 28, 3119. (102) Chen, Z. W.; Zhang, S. Y.; Tan, S.; Li, F. Q.; Wang, J.; Jin, S. Z.; Zhang, Y. H. J. Cryst. Growth 1997, 180, 280. (103) Chen, Z. W.; Zhang, S. Y.; Tan, S.; Wang, J.; Jin, S. Z.; Hou, J. G. J. Mater. Sci. Lett. 2002, 21, 411. (104) Salavati-Niasari, M.; Mohandes, F.; Davar, F.; Saberyan, K. Appl. Surf. Sci. 2009, 256, 1476. (105) Wang, Y. Nature 1992, 356, 585. (106) Wang, D. Z.; Zuo, J.; Zhang, Q. J.; Luo, Y.; Ruan, Y. Z.; Wang, Z. J. Appl. Phys. 1997, 81, 1413. (107) Chen, Q. W.; Qian, Y. T.; Chen, Z. Y.; Wu, W. B.; Chen, Z. W.; Zhou, G. E.; Zhang, Y. H. Appl. Phys. Lett. 1995, 66, 1608. (108) Chen, Q. W.; Zhu, J. S.; Li, X. G.; Fan, C. G.; Zhang, Y. H. Phys. Lett. A 1996, 220, 293. (109) Chen, Q. W.; Li, X. J.; Zhou, G. E.; Zhu, J. S.; Zhang, S. Y.; Jia, Y. B.; Zhang, Y. H. J. Appl. Phys. 1997, 81, 7970. (110) Shek, C. H.; Wong, K. W.; Lai, J. K. L. Mater. Sci. Eng. R 1997, 19, 153. (111) Buciuman, F.; Patcas, F.; Craciun, R.; Zahn, D. R. T. Phys. Chem. Chem. Phys. 1999, 1, 185. (112) Chen, Z. W.; Tan, S.; Zhang, S. Y.; Wang, J.; Jin, S. Z.; Zhang, Y. H.; Sekine, H. Jpn. J. Appl. Phys. 2000, 39, 6293. (113) Ruetschi, P.; Giovanoli, R. J. Appl. Electrochem. 1982, 12, 109. (114) Kijlstra, W. S.; Poels, E. K.; Weckhuysen, A. B. M.; Schoonheydt, R. A. J. Phys. Chem. B 1997, 101, 309. (115) Kapteijn, F.; Langeveld, A. D.; Moulijn, J. A.; Andreini, A.; Vuurman, M. A.; Turek, A. M.; Jehng, J. M.; Wachs, I. E. J. Catal. 1994, 150, 94. (116) Bernard, M. C.; Goff, A. H. L.; Thi, V. B.; Torresi, S. C. J. Electrochem. Soc. 1993, 140, 3065. (117) Praserthdam, P.; Silveston, P. L.; Mekasuwandumrong, O.; Pavarajarn, V.; Phungphadung, J.; Somrang, P. Cryst. Growth Des. 2004, 4, 39. (118) Mayo, M. J. Int. Mater. Rev. 1996, 41, 85. (119) Burke, J. E.; Turnbull, D. Prog. Met. Phys 1952, 3, 220. (120) Ardell, A. J. Acta Metall. 1972, 20, 601. (121) Malow, T. R.; Koch, C. C. Acta Meter. 1997, 45, 2177. (122) Shek, C. H.; Lai, J. K. L.; Lin, G. M. Nanostuct. Mater. 1999, 11, 887. (123) Lai, J. K. L.; Shek, C. H.; Lin, G. M. Scr. Mater. 2003, 49, 441. (124) Baldi, M.; Finocchio, E.; Milella, F.; Busca, G. Appl. Catal., B 1998, 16, 43. (125) Zwinkels, M. F. M.; Järås, S. G.; Menon, P. G.; Griffik, T. A. Catal. Rev. Sci. Eng. 1993, 35, 319. (126) Wang, W. Z.; Xu, C. K.; Wang, G. H.; Liu, Y. K.; Zheng, C. L. Adv. Mater. 2002, 14, 837. (127) Yao, B. D.; Shi, H. Z.; Zhang, X. Y.; Zhang, L. D. Appl. Phys. Lett. 2001, 78, 174. (128) Takagi, H.; Ogawa, H.; Yamazaki, Y.; Ishizaki, A.; Nakagiri, T. Appl. Phys. Lett. 1990, 56, 2379. (129) Kong, Y. C.; Yu, D. P.; Zhang, B.; Fang, W.; Fang, S. Q. Appl. Phys. Lett. 2001, 78, 407. (130) Tang, Z. K.; Wong, G. K. L.; Yu, P.; Kawasaki, M.; Ohtomo, A.; Koinuma, H.; Segawa, Y. Appl. Phys. Lett. 1998, 72, 3270. (131) Tsybeskov, L.; Vandyshev, J. V.; Fauchet, P. M. Phys. Rev. B 1994, 49, 7821. (132) Peercy, P. S.; Morosin, B. Phys. Rev. B 1973, 7, 2779. (133) Wang, G. H.; Han, M. Prog. Phys 1990, 10, 248. (134) Sadtler Research Laboratories. Inorganics IR Grating Spectra; Sadtler Research Laboratories, Inc.: Philadelphia, 1972. (135) Yu, S. H.; Qian, Y. T.; Shu, L.; Xie, Y.; Yang, L.; Wang, C. S. Mater. Lett. 1998, 35, 116.

(136) Zheng, L.; Li, K. B.; Zhang, Y. H. Phys. Rev. B 1998, 58, 8613. (137) Fonseca, C. D.; Ozanam, F.; Chazalviel, J. N. Surf. Sci. 1996, 365, 1. (138) Tsu, D. V.; Lucovsky, G.; Davidson, B. N. Phys. Rev. B 1989, 40, 1795. (139) Zhou, S. M.; Zhu, H.; Shi, L.; Zhao, Z. Y.; Zhou, G. E.; Zhang, Y. H. J. Phys.: Condens. Matter. 1999, 11, 6877. (140) Abello, L.; Bochu, B.; Gaskov, A.; Koudryavtseva, S.; Lucazeau, G.; Roumyantseva, M. J. Solid State Chem. 1998, 135, 78. (141) Liu, Y. K.; Zheng, C. L.; Wang, W. Z.; Yin, C. R.; Wang, G. H. Adv. Mater. 2001, 13, 1883. (142) Peng, X. S.; Zhang, L. D.; Meng, G. W.; Tian, Y. T.; Lin, Y.; Geng, B. Y.; Sun, S. H. J. Appl. Phys. 2003, 93, 1760. (143) Chen, Z. W.; Zhang, S. Y.; Tan, S.; Wang, J.; Jin, S. Z. Appl. Phys. A: Mater. Sci. Process. 2004, 78, 581. (144) Fadley, C. S. In Electron Spectroscopy; Shirley, D. A., Ed.; North Holland: Amsterdam, 1972, 781. (145) Kowalczyk, S. P.; Ley, L.; Pollak, R. A.; McFeely, F. R.; Shirley, D. A. Phys. Rev. B 1973, 7, 4009. (146) Vos, D. E. D.; Weckhuysen, B. M.; Bein, T. J. Am. Chem. Soc. 1996, 118, 9615. (147) Lee, G. H.; Huh, S. H.; Jeong, J. W.; Choi, B. J.; Kim, S. H.; Ri, H. C. J. Am. Chem. Soc. 2002, 124, 12094. (148) Nayak, S. K.; Jena, P. J. Am. Chem. Soc. 1999, 121, 644. (149) Ren, Y.; Liu, Q.; Wang, J.; Wang, H.; Xue, D. Mater. Lett. 2009, 63, 661. (150) Chen, W.; Chen, C.; Guo, L. J. Appl. Phys. 2010, 108, 043912. (151) Zhang, H.; Liang, C.; Tian, Z.; Wang, G.; Cai, W. J. Phys. Chem. C 2010, 114, 12524. (152) Tackett, R.; Lawes, G.; Melot, B. C.; Grossman, M.; Toberer, E. S.; Seshadri, R. Phys. Rev. B 2007, 76, 024409. (153) Suzuki, T.; Katsufuji, T. Phys. Rev. B 2008, 77, 220402. (154) Li, J.; Wang, Y. J.; Zou, B. S.; Wu, X. C.; Lin, J. G.; Guo, L.; Li, Q. S. Appl. Phys. Lett. 1997, 70, 3047. (155) Hiruma, K.; Yazawa, M.; Katsuyama, T.; Ogawa, K.; Haraguchi, K.; Kouguchi, M.; Kakibayashi, H. J. Appl. Phys. 1995, 77, 447. (156) Chowdhury, A. N.; Azam, M. S.; Aktaruzzaman, M.; Rahim, A. J. Hazard. Mater. 2009, 172, 1229. (157) Dubal, D. P.; Dhawale, D. S.; Salunkhe, R. R.; Pawar, S. M.; Lokhande, C. D. Appl. Surf. Sci. 2010, 256, 4411. (158) Davar, F.; Salavati-Niasari, M.; Mir, N.; Saberyan, K.; Monemzadeh, M.; Ahmadi, E. Polyhedron 2010, 29, 1747. (159) Park, J. P.; Kim, S. K.; Park, J. Y.; Hwang, C. H.; Choi, M. H.; Kim, J. E.; Ok, K. M.; Kwak, H. Y.; Shim, I. W. Mater. Lett. 2009, 63, 2201. (160) Askarinejad, A.; Bagherzadeh, M.; Morsali, A. Appl. Surf. Sci. 2010, 256, 6678. (161) Dubal, D. P.; Dhawale, D. S.; Salunkhe, R. R.; Fulari, V. J.; Lokhande, C. D. J. Alloys Compd. 2010, 497, 166. (162) Raj, A. M. E.; Victoria, S. G.; Jothy, V. B.; Ravidhas, C.; Wollschläger, J.; Suendorf, M.; Neumann, M.; Jayachandran, M.; Sanjeeviraja, C. Appl. Surf. Sci. 2010, 256, 2920. (163) Salazar-Alvarez, G.; Sort, J.; Suriñach, S.; Baró, M. D.; Nogues, J. J. Am. Chem. Soc. 2007, 129, 9102. (164) Ahmed, K. A. M.; Zeng, Q.; Wu, K.; Huang, K. J. Solid State Chem. 2010, 183, 744. (165) Askarinejad, A.; Morsali, A. Ultrasonics Sonochem. 2009, 16, 124. (166) Iijima, S. Nature 1991, 354, 56. (167) Yu, D.; Lee, C. S.; Bello, I.; Sun, X. S.; Tang, Y. H.; Zhou, G. W.; Bai, Z. G.; Zhang, Z.; Feng, S. Q. Solid State Commun. 1998, 105, 403. (168) Martin, C. R. Science 1994, 266, 1961. (169) Holmes, J. D.; Johnston, K. P.; Doty, R. C.; Korgel, B. A. Science 2000, 287, 1471. (170) Wagner, R. S.; Ellis, W. C. Appl. Phys. Lett. 1964, 4, 89. (171) Morales, A. M.; Lieber, C. M. Science 1998, 279, 208. (172) Liu, H. I.; Maluf, N. I.; Pease, R. F. W. J. Vac. Sci. Technol. B 1992, 10, 2846. 3854

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855

Chemical Reviews

Review

(173) Ono, T.; Saitoh, H.; Esashi, M. Appl. Phys. Lett. 1997, 70, 1852. (174) Frank, F. C. Discovery Faraday 1949, 5, 48. (175) Pan, Z. W.; Dai, Z. R.; Wang, Z. L. Science 2001, 291, 1947. (176) Huang, M. H.; Wu, Y.; Feick, H.; Tran, N.; Weber, E.; Yang, P. Adv. Mater. 2001, 13, 113. (177) Leite, E. R.; Gomes, J. W.; Oliveira, M. M.; Lee, E. J. H.; Longo, E.; Varela, J. A.; Paskocimas, C. A.; Boschi, T. M.; Lanciotti, F. Jr.; Pizani, P. S.; Soares, P. C. Jr. Appl. Sci. Res. 2002, 2, 125. (178) Law, M.; Kind, H.; Messer, B.; Kim, F.; Yang, P. Angew. Chem. 2002, 114, 2511. (179) Wang, D. Z.; Wen, S. L.; Chen, J.; Zhang, S. Y.; Li, F. Q. Phys. Rev. B 1994, 49, 14282. (180) Zhu, X.; Birringer, R.; Herr, U.; Gleiter, H. Phys. Rev. B 1987, 35, 9085. (181) Giovannelli, F.; Chartier, T.; Autret-Lambert, C.; Delorme, F.; Zaghrioui, M.; Seron, A. J. Solid State Chem. 2009, 182, 1021. (182) Gibot, P.; Laffont, L. J. Solid State Chem. 2007, 180, 695. (183) Konstantinov, K.; Tournayre, Y.; Liu, H. K. Mater. Lett. 2007, 61, 3189. (184) Tietz, F.; Sebold, D. Mater. Sci. Eng., B 2008, 150, 135. (185) Chang, Y. Q.; Yu, D. P.; Long, Y.; Xu, J.; Luo, X. H.; Ye, R. C. J. Cryst. Growth 2005, 279, 88. (186) Stobbe, E. R.; Boer, B. A. D.; Geus, J. W. Catal. Today 1999, 47, 161. (187) Grootendorst, E. J.; Verbeek, Y.; Ponce, V. J. Catal. 1995, 157, 706. (188) Chen, Z. W.; Zhang, S. Y.; Tan, S.; Hou, J. G.; Zhang, Y. H. Thin Solid Films 1998, 322, 194. (189) Muller, D. A.; Sorsch, T.; Moccio, S.; Baumann, F. H.; EvansLutterodt, K.; Timp, G. Nature 1999, 399, 758. (190) Du, G. H.; Yuan, Z. Y.; Van Tendeloo, G. Appl. Phys. Lett. 2005, 86, 63113. (191) Paterson, J. H.; Krivanek, O. L. Ultramicroscopy 1990, 32, 319. (192) Ardizzone, S.; Bianchi, C. L.; Tirelli, D. Colloids Surf. A 1998, 134, 305. (193) Wang, Z. L.; Yin, J. S.; Jiang, Y. D. Micron 2000, 31, 571. (194) Davoli, I.; Marcelli, A.; Bianconi, A.; Tomellini, M.; Fanfoni, M. Phys. Rev. B 1986, 33, 2979. (195) Lindner, T.; Sauer, H.; Engel, W.; Kambe, K. Phys. Rev. B 1986, 33, 22. (196) Kurata, H.; Lefèvre, E.; Colliex, C.; Brydson, R. Phys. Rev. B 1993, 47, 13763. (197) Duan, X. F.; Huang, Y.; Cui, Y.; Wang, J. F.; Lieber, C. M. Nature 2001, 409, 66. (198) Trentler, T. J.; Goel, S. C.; Hickman, K. M.; Viano, A. M.; Chiang, M. Y.; Beatty, A. M.; Gibbons, P. C.; Buhro, W. E. J. Am. Chem. Soc. 1997, 119, 2172.

3855

dx.doi.org/10.1021/cr2004508 | Chem. Rev. 2012, 112, 3833−3855