Original Title: Development and Application of Controllable Synthesis of Nanoscale Morphology and Structure in Recent Years from Nature & Science Since the advent of carbon nanotubes, there has been a worldwide upsurge in the study of nanomaterials. Nanomaterials usually refer to materials with at least one dimension in the range of 1-100 nm. The reason why nanomaterials have aroused widespread research interest in the field of materials science all over the world in the past decade is not only because of their unique and fascinating properties, but also because these properties change with the size and morphology of nanomaterials. For example, in semiconductor materials, due to electronic transition, different characteristics of bulk materials are produced; gold nanoparticles can show plasmon effect, which is widely used in the field of Raman enhancement; transition metal particles, with the decrease of size, the specific surface area of particles increases, which has great potential in the field of catalysis. Therefore, the synthesis of nanomaterials with different sizes and morphologies has become a hot research topic in the field of nanotechnology. Generally, nanomaterials can be divided into zero-dimensional (nanoparticles, atomic clusters), one-dimensional (nanorods, nanowires, nanotubes), two-dimensional (superlattices, ultrathin films) and three-dimensional (materials composed of the first three materials as basic units) systems. At present, the main effects of nanomaterials include: 1) small size effect; 2) surface effect; 3) quantum size effect; 4) macroscopic quantum tunneling effect. Various properties of materials are closely related to the four effects. It can be said that to some extent, the influence of morphology and structure on the properties of materials determines the research direction of nanomaterials, so researchers have invested a lot of manpower and material resources to study nano-synthesis. Based on this, this paper will review the relevant articles published in Science and Nature in recent years, and show the development and application of controllable synthesis of morphology and structure of nanomaterials in recent years. 1. Zero-dimensional material The usual structures of zero-dimensional nanomaterials are mainly clusters and nanoparticles. Clusters are relatively stable microscopic or submicroscopic aggregates composed of several or even thousands of atoms, molecules or ions through physical or chemical binding forces, and their physical and chemical properties vary with the number of atoms contained. Cluster is a concept of nanomaterials at the material scale. The spatial scale of clusters ranges from a few angstroms to hundreds of angstroms, which is too large to be described by inorganic molecules and too small to be described by small solids. Many properties of clusters are different from those of single atoms and molecules, solids and liquids, and can not be obtained by simple linear extension or interpolation of their properties. Therefore, clusters are regarded as a new level of material structure between atoms, molecules and macroscopic solid matter, and are the transitional state of various substances from atoms and molecules to bulk matter, or represent the initial state of condensed matter. Nanoparticles refer to particles with particle size between 1 and 100 nm (nanoparticles are also called ultrafine particles). It belongs to the category of colloidal particle size. They are located in the transition region between atomic clusters and macroscopic objects, between microscopic systems and macroscopic systems, and are clusters composed of a small number of atoms or molecules, so they are both atypical microscopic systems and atypical macroscopic systems. It can be predicted that nanoparticles should have some novel physical and chemical properties. Nanoparticles differ from macroscopic objects in that they have a large proportion of surface area, while the surface atoms have neither long order nor short order amorphous layers. It can be considered that the state of atoms on the surface of nanoparticles is closer to the gas state, while the atoms inside the particles may be arranged in an orderly manner. Even so, due to the small particle size and the large surface curvature, a high Gilibs pressure is generated internally, which can lead to some deformation of the internal structure. The structural characteristics of nanoparticles make it have the following four effects. (1) Volume effect; (2) surface effect; (3) quantum size effect and (4) macroscopic quantum tunnel effect. Over the past decade, more and more scientists have devoted themselves to the research of nanomaterials, and have achieved fruitful results in preparation, properties and applications. Uch as metal nanoparticles, quantum dots and the like. The colleagues who do the morphology control of metal nanoparticles (structure) must be familiar with Xia Younan's group, right? To some extent, it was the work of Xia Younan and Sun Yugang on the preparation of nano-silver cubes and gold nanocages published in Science in 2002 (Science, 2002, 298, 2176-2179) that formally initiated the era of fine control of the morphology of noble metal nanoparticles. Since then, based on silver cubes and gold nanocages produced by displacement reactions, the team has continuously expanded its territory and published a series of work in high-level journals, leading the progress of this field and influencing many young researchers to devote themselves to this field. After working on silver cubes and gold nanocages, Xia Younan's team prepared PdPt nanodendrites structure by Pd seed growth method in 2009, and its highly active ORR may come from a special branch structure (high index crystal plane). The prominent feature of this paper may be that the seed method controls the growth of crystal faces to obtain uniform branch nanoparticles, while the improvement of catalytic activity is not the focus (Science, 2009, 324, 1302-1305). Expand the full text Figure 1. Morphology and ORR Performance of PdPt Later, Yang Peidong of UC-Berkeley and the Vojislav R. Stamenkovic team of the Argonne National Laboratory synthesized framework-structured nanoparticles in 2014. The open structure and full exposure of the Pt3Ni (111) face greatly increase the ORR activity. This article can be said to be a good continuation of the theoretical model proposed by Vojislav R. Stamenkovic in 2007 and the application of single crystal Pt3Ni (111) with ultra-high activity to practical nanoparticles. Figure 2. Morphology and Properties of Pt3Ni A year later, Huang Yu of UCLA and Tim Mueller of Johns Hopkins University combined Mo-doped PtNi octahedra with ultra-high activity and stability ORR. (Science, 2015, 348, 1230-1234) Figure 3. Morphology and Properties of Mo-Pt3Ni/C In recent years, researchers have found that zero-dimensional nanomaterials have great potential for application and development in catalytic industry. Among them, bimetallic catalysts have attracted wide attention because of their excellent performance. Heterogeneous catalysts comprising bimetallic nanoparticles (NPs) are used in many petrochemical processes, including selective hydrogenation, dehydrogenation, and acetoxylation. The differences in the catalytic performance of bimetallic NPs compared to the parent metal result from their unique geometric and electronic structures and the synergistic effect between the two metals. In particular, the synthetic scheme is crucial to the structural, electronic, and catalytic properties of bimetallic NPs. Conventional impregnation methods often lead to the formation of bimetallic NPs with non-uniform particle size and composition. New strategies for the synthesis of bimetallic NPs include colloidal synthesis, surface organometallic chemistry, atomic layer deposition, coadsorption and coreduction of metal cations, and carbon thermal shock synthesis. Nonetheless, it remains challenging to synthesize ultrasmall (< 3 nm diameter) supported bimetallic NPs with well-defined stoichiometry and compactness between the constituent metals. The synthesis of ultrasmall supported bimetallic nanoparticles (between 1 and 3 nm in diameter) with well-defined stoichiometry and compositional intermetallic compactness remains a major challenge. Kunlun Ding of Louisiana State University in the United States synthesized 10 different supported bimetallic nanoparticles by surface inorganic metal chemistry. By decomposing and reducing the surface adsorbed heterometallic bimetallic complex salts, the salts can easily adsorb target cations and anions continuously on silica substrates. For example, adsorption of tetraamminepalladium (II) [Pd (NH3) 42 +] followed by adsorption of tetrachloroplatinate [PtCl42 ?] is used to form palladium-platinum (Pd-Pt) nanoparticles. These supported bimetallic nanoparticles showed enhanced catalytic performance in the selective hydrogenation of acetylene, which clearly demonstrated the synergistic effect between the constituent metals (Science, 2018, DOI: 10.1126/science.aau4414). Figure 4. Schematic and electron microscopic analysis of supported bimetallic nanoparticles The successful preparation of highly active nanoparticles is not easy, and the subsequent storage is also a problem. Because of the problems such as aggregation and oxidation between nanoparticles with very small size. In fact, particles or a solute phase can be immediately dispersed homogeneously in a solvent to form colloidal solutions. Colloids are very common in biology, chemistry and processing, and their high homogeneity involves different components. In order to make the solute particles exist stably against polymerization and aggregation, it is necessary to increase the repulsive force of the solute surface to overcome the Van der Waals force between them, such as electrostatic stabilization, surface modification to connect the solute to form a three-dimensional space and other methods. Dmitri V. Talapin et al. Of the University of Chicago reported a colloidal system in which solute particles (metals, semiconductors, or magnetic materials) are stable in a variety of molten inorganic salts. However, this stabilization is different from the traditional electrostatic and steric mechanisms. The observation of solute-solvent binding indicates that the colloidal stability is mainly due to the strength of the chemical bonds on the solute-solvent surface. Theoretical and molecular dynamic modeling analysis shows that the surface bonds of the solvent ion layer produce long-range charge density vibrations around the molten salt solute particles, thus preventing their aggregation. The study of colloidal compounding of inorganic particles in inorganic salts also opens a new chapter for the development of colloidal solid-state science and engineering applications. Nature,2017,DOI:10.1038/nature21041) Figure 5. Nanocrystalline colloids in molten inorganic salts 2. One dimensional material 2.1 Synthesis mechanism and method of one-dimensional nanomaterials The nanocrystal growth of one-dimensional materials is essentially a crystallization process, and the transformation from liquid, gas or solid phase to solid involves two basic processes, namely nucleation and growth. In recent years, a series of bottom-up synthesis methods for one-dimensional nanomaterials have been developed. So far, there are several methods to synthesize nanowires, such as VLS, SLS, template method, liquid phase method, self-assembly and so on. Figure 6. Nanowire structures grown by different methods (A) utilize anisotropic orientation growth of that crystal structure; (B) inducing growth of the nanostructure by introduce asymmetry of the crystal nucleus obtained by a liquid-solid interface; (C) utilizing a template confinement method to obtain a one-dimensional confined space; (D) utilize a coating agent-assisted kinetic control method; (E) forming nanowires from nanoparticle self-assembly; (F) reduce that size of the nanowire from the one-dimensional micrometer material. 2.2 Development of one-dimensional nanomaterials Speaking of the synthesis and application of nanowires, it is natural to think of C. Professor of M. Lieber. In the mid-1990s, Lieber's main research direction began to shift to the controlled synthesis of nanomaterials (bottom up). After careful consideration, they chose one-dimensional nanomaterials as a breakthrough to synthesize homogeneous Si nanowires by VLS method for the first time in 1997, and then quickly extended this method to other materials in the periodic table and became a general method (Science, 1997, 279, 208-211). In 2001, they pioneered the successful large-scale assembly of nanowire materials using microfluidics (Science, 2001, 291, 630-633), and Lieber used this concept to develop the first nano-computer. In the same year, Lieber's group first realized the response of Si nanowires to protein molecules (Science, 2001, 293, 1289-1292). This work has attracted the attention of many research groups around the world. In addition to Professor Lieber, another person who has made significant contributions in the field of nanowires is Professor Yang Peidong of UC Berkeley, a student of Professor Lieber. For the first time, Professor Yang Peidong's team observed the formation process of Ge nanowires in situ by electron microscopy, which provided direct and effective evidence for the VLS mechanism. Over the years, Professor Yang Peidong has been committed to using solar energy to convert CO2 into chemical energy. Although biological photosynthesis and pure artificial photosynthesis have been able to achieve low-cost transformation systems through photosynthesis. However, both of them have their own shortcomings. In order to combine their respective advantages, Professor Yang Peidong prepared a semiconductor nanoparticle (cadmium sulfide) on the surface of bacteria (Moorella thermoacetica) without photosynthesis, and obtained a bio-inorganic hybrid system. In this system, CdS can capture light energy and then selectively convert CO2 into a natural "by-product": acetic acid, achieving the goal of fixing atmospheric CO2 and converting it into a useful energy source for human beings. The biological system can ensure the advantages of high selectivity,nickel titanium wire, low cost and self-repair of photosynthesis; and the artificial semiconductor material can also ensure the function of efficiently capturing light energy, and still has a good effect after being circulated for several days under the light-dark condition of simulating day and night! (Science,2016, 251, 74-77)。 Figure 7. Bacteria -Schematic diagram of CdS reaction system In addition, in the Photoelectrochemical system, the average performance of the whole nanowire array can not represent the performance of a single nanowire. In view of this, Professor Yang Peidong's team reported a photoelectrode platform based on a single silicon nanowire, which can reliably detect the I-V characteristics of a single silicon nanowire (Nature Nanotechnology, 11, 609 – 612 (2016)). This model system of single nanowire photoelectrode can be further studied and designed, for example, nanostructures with different morphologies can be introduced to provide reasonable structures for supporting different electrocatalysts. It provides theoretical and technical support for the next generation of nanowire photoelectrode solar-fuel conversion devices. Figure 8. A single silicon nanowire photoelectrode is used for Schematic representation of the PEC (Photoelectrochemical) test In order to make nanowires play a greater advantage, researchers have made unremitting efforts. The joint research team of Huang Yu/Duan Xianfeng of UCLA and California Institute of Technology (Caltech) used alloy and etching methods to change the morphology of platinum nanowires and obtained jagged Pt nanowires. This paper is characterized by showing the great influence of metal surface defects on the catalytic activity, and also proposing a new method for preparing metal nanowires by etching. (Science, 2016, DOI: 10.1126/science.aaf9050) The researchers prepared the nanowires in a two-step process. In the first step, platinum-nickel alloy nanowires are prepared by heat treatment. Then, the nickel atoms in the alloy fibers are selectively removed by electrochemical methods to form nanowires that appear to have fluff and contain only platinum atoms. Through testing, it is found that the zigzag platinum nanowires have very high redox activity, and their performance is nearly 50 times higher than that of commercial catalysts. The amount of that platinum catalyst use in the fuel cell is greatly reduced, and the cost of the fuel cell is greatly reduce. Platinum nanowires with one-dimensional sawtooth surface structure have abundant active sites, which can greatly accelerate the catalytic reaction and the catalyst still has considerable stability after many cycles of testing, so this structure can bring unprecedented breakthroughs in performance. Figure 9. Serrated Structure and Properties of Pt Nanowires In addition to nanowires, another very important one-dimensional material is carbon nanotubes. As a one-dimensional nanomaterial, carbon nanotubes (CNTs) have many unusual mechanical, electrical and chemical properties due to their light weight and perfect connection of hexagonal structure. In recent years, with the in-depth study of carbon nanotubes and nanomaterials, their broad application prospects have been constantly demonstrated. Carbon nanotube (CNT) is a kind of one-dimensional quantum material with a special structure (the radial size is in the nanometer range, the axial size is in the micron range, and both ends of the tube are basically sealed). Carbon nanotubes are coaxial circular tubes composed of several to dozens of layers of carbon atoms arranged in a hexagonal pattern. The distance between the layers is about 0.34 nm, and the diameter is generally 2 ~ 20 nm. According to the different orientation of carbon hexagon along the axial direction, it can be divided into three types: zigzag, armchair and spiral. Among them, helical carbon nanotubes have chirality, while zigzag and armchair carbon nanotubes have no chirality. Carbon nanotubes can be seen as graphene sheets curled, so according to the number of layers of graphene sheets, they can be divided into single-walled carbon nanotubes (or single-walled carbon nanotubes, walled Carbon nanotubes, SWCNTs) and multi-walled carbon nanotubes (or multi-layer carbon nanotubes, SWCNTs). Multi-walled Carbon nanotubes (MWCNTs). At the beginning of the formation of multi-walled tubes, it is easy to become trap centers between layers to capture various defects, so the walls of multi-walled tubes are usually covered with small hole-like defects. Compared with the multi-wall tube, the single-wall tube has smaller diameter distribution range,Titanium welding pipe, fewer defects and higher uniformity. Therefore, in recent years, researchers have been committed to the efficient preparation of single-walled carbon nanotubes. The electronic properties of single-walled carbon nanotubes (SWNTs) depend on their chirality — that is, the way the SWNTs twist along their axis — which is characterized by two indices (n, m). Controlling chirality during nanotube synthesis will enable us to reduce the cost of sorting, develop promising applications, and overcome the limitations of silicon. It is reported that the development of carbon nanotube computers is very rapid, and major breakthroughs have been made. However, selective synthesis still seems to be the weak link, although new studies using solid-state catalysts have reported the growth of chiral specificity in SWNTs. The specific mechanism of this selective growth is still under debate, so it is necessary to clarify the realistic growth mode including the role of catalyst. Existing models focus on kinetics and neglect the role of catalysts, but fail to calculate the chiral distribution from experiments. Atomic computer simulations emphasize chemical accuracy, but need to be supplemented with models in order to provide a comprehensive understanding of the process. Single-walled carbon nanotubes are hollow cylinders that can be grown to the centimeter scale by incorporating carbon at the interface with the catalyst. They have semiconducting or metallic properties depending on the helicity generated during the growth process. In order to support the exploration of selective synthesis, Professor Christophe Bichara's team from the University of Marseille, France, has established a thermodynamic model that correlates the tube-catalyst interface energy, temperature and the resulting tube chirality. The authors show that nanotubes can grow chirally due to the configurational entropy of their nanometer-sized edges, thus explaining the experimentally observed temperature evolution of the chirality distribution. By considering the chemical properties of the catalyst through the interfacial energy, the authors have derived the structure map and phase diagram, which will guide the rational selection of catalyst and growth parameters to achieve better selectivity. (Science, 2018, DOI: 10.1126/science.aat6228) Figure 10. From test to model In the application of single-walled carbon nanotubes, with the development of industrial technology, the transistors of microprocessor chips are gradually replaced by nano-electronic device materials such as single-walled carbon nanotubes (SWNTs). Due to the excellent current output of horizontal nanotubes, they have a great attraction for practical technical applications. Horizontal SWNT arrays with controlled chirality have a wider range of applications and can ensure the uniformity of the fabricated devices, but the direct growth of such SWNT arrays has not yet been achieved. Professor Zhang Jin from Peking University controlled the chirality of horizontal SWNT arrays by controlling the surface symmetry of active catalysts, and obtained horizontal SWNT arrays with controlled chirality on the surface of solid carbide catalysts. The obtained horizontally aligned metal SWNT arrays have an average density of more than 20 tubes/micron, with 90% of the tubes having a chiral index of (12,6). At the same time, SWNT array semiconductors were also obtained with an average density of more than 10 tubes/micron, where 80% of the nanotubes have a chiral index of (8,4). (Nature, 2017, DOI: 10.1038/nature21051) Figure 11. Controlled by ethanol chemical vapor deposition SWNTs chirality In the same year, Cao Qing et al. Of IBM Research Center used a new strategy to manufacture contacts of carbon nanotube transistors, which reduced the size while maintaining a low contact resistance. In order to ensure that enough current is transmitted between the contacts, they set up an array of several parallel semiconductor carbon nanotubes (s-CNTs) between the contacts. The size of the final p-channel transistor is only 40 nm, which is the smallest record reported so far. What's more, electrical tests show that the new s-CNT transistor is faster and more efficient than today's silicon-based transistors. Furthermore, the fabrication of technologically relevant high performance nanotube array devices with identical footprint, resistive end contact points, using a high purity s-CNT source, self-assembled compression of nanotubes into full surface coverage aligned arrays is demonstrated. The s-CNT array transistor exhibits a high saturated on-state current of more than 1.2 mA μm-1 and a conductance of more than 2 mS μm-1, exceeding the current of the best competing silicon device at the benchmark under equivalent gate drive, source-drain biased (VDS). (Science, 2017, DOI: 10.1126/science.aan2476) Figure 12. Single of extremely large size Inset and electron microscope image of the s-CNT transistor In addition to the traditional methods of preparing nanowires, Erik P. A. M. Bakkers from Delft University of Technology and Eindhoven University of Technology in the Netherlands recently applied molecular beam epitaxy technology to the design of nanowire quantum devices. They first introduced a technique for the bottom-up synthesis of single-crystal InSb nanowire networks, with a special focus on nanowire networks with a predefined number of superconducting islands. Structural analysis confirms the high quality crystallinity of the nanowire junction and the existence of an epitaxial superconductor-semiconductor interface. The Aharonov-Bohm effect and the weak-antipositioning effect revealed by quantum transport measurements of nanowire "tags" confirm a phase-coherent system with strong spin-orbit coupling. In addition, a proximally-induced hard superconducting gap is also exhibited in this hybrid superconducting-semiconductor nanowire, indicating that the researchers successfully developed the materials required for the first braiding experiment. This work opens up a new way to fabricate epitaxial three-dimensional quantum structures with great potential for application in key components of quantum devices. (Nature, 2017, DOI: 10.1038/nature23468)。 Figure 13. Deterministic growth of InSb nanowire networks In addition to nanowires and carbon nanotubes, another important one-dimensional material is nanorods. One-dimensional rod-like nanostructured materials have important applications in the study of the dependence of size, morphology and properties, because few materials can grow naturally under such anisotropic conditions (linear pores or surface templates). For the synthesis of one-dimensional materials, it is necessary to limit the reaction space to guide the growth of materials. At present, the commonly used methods include the use of porous matrix materials (such as alumina and polycarbonate films as templates). These linear pores in matrix materials can be filled with soluble precursors. The diameter of the synthesized nanomaterial can be limited due to the existence of these pores, but the additional template matrix material needs to be further removed to obtain the desired nanorod nanomaterial. Recent molecule-based templates can overcome these shortcomings. Professor Lin Zhiqun's research team at the Georgia Institute of Technology in the United States reported a synthesis method that can precisely control the diameter, composition, morphology and structure of the synthesized nanomaterials. The polymer brush used by the researchers to synthesize nanowires is different from that reported in previous articles. Professor Lin Zhiqun of Georgia Institute of Technology used a bottle-scrubbing block copolymer (BBCP), which has a cellulose skeleton with dense grafting functional block copolymers as side chains, including multiple cavities. In addition, the tri-substituted hydroxycellulose functional groups on the cellulose backbone allow dense polymer side chains to be grafted onto the cellulose backbone. The researchers used this advantage to synthesize the required straight BBCP. A series of BBCPs were synthesized by atom transfer radical polymerization. Then the BBCP was dispersed in DMF polar solvent. Due to the difference of polarity between BBCP and the solution, the molecular chains in BBCP were pulled together to form a large reaction chamber. The inorganic precursors dispersed in the solvent are preferentially distributed in the reaction chamber formed by the molecular chains in the BBCP due to the polar effect of the solvent, so the high concentration of aggregation drives the nucleation of inorganic materials and the growth of inorganic nanorods. (Science, 2016, DOI: 10.1126/science.aad8279) Figure 14. Amphoteric molecule use straight cylinder type Illustration of the synthesis mechanism of BBCP as a nanoreactor for the synthesis of one-dimensional nanocrystals A) nanorods synthesized with the assistance of a cellulose template; (B) a nanorod with a core-shell structure which is synthesized by adopting a template-assisted method; (C) nanotubes synthesized by a cellulose template-assisted method; Figure 15. Transmission electron micrographs of different kinds of nanorods synthesized with the assistance of cellulose templates A year later, Andr Andrés Guerrero-Mart Martínez, Luis M. Liz Marz Marzán, and OvidioPe OvidioPe?a-Rodr Rodríguez reported a femtosecond laser pulse annealing process that reshaped gold nanorod sols so that their SPR spectra were as sharp as single gold nanorods. Figure 16. Characterization of gold nanorods after laser treatment Lei's team reported a simple batch preparation technology for oxide nanowires, which directly converts bulk alloy materials into oxide nanowires without catalysts or external conditions. (Science 2017, 355, 267-271.) Figure 17. Morphology of nanowire during synthesis 3. Two-dimensional material Two-dimensional materials refer to materials in which electrons can only move freely (planar motion) in two dimensions of non-nanometer scale (1-100 nm), such as nano-thin films, superlattices and quantum wells. Two-dimensional materials were proposed with the successful isolation of graphene, a single atomic layer of graphite material, by Geim's group at the University of Manchester in 2004. Transition metal disulfides (TMDs) such as MoS2, MoSe2, WS2, etc., their 2D atomic crystals have attracted much attention in recent years due to their excellent electronic properties. In order to exploit their more application potential, we need to find a reliable synthesis method that can precisely control their chemical composition and electronic structure. Professor Duan Xiangfeng of Hunan University (the first unit) and the University of California, Los Angeles, and Duan Xidong of Hunan University reported a universal synthesis method that can control the growth of 2D atomic crystals of multiple heterojunctions, multiple heterojunctions and superlattices. By modified vapor deposition, titanium filler rod ,titanium tubing price, with a reverse gas flow during a continuous vapor deposition process with temperature fluctuations, existing 2D crystals can be cooled to prevent unwanted thermal degradation and uncontrolled homogeneous nucleation, thus enabling highly robust block-by-block epitaxial growth. While not being limited to a single heterojunction, More different 2D heterojunctions (e.g., WS2-WSe2 and WS2-MoSe2), multi-heterojunctions (e.g., WS2-WSe2-MoS2 and WS2-MoSe2-WSe2), and superlattices (e.g. WS2 ? WSe ? WS2 ? WSe ? WS2) were successfully fabricated by precise spatial modulation. The prepared WSe2-WS2 exhibits excellent diode characteristics with a high rectification ratio of 105. (Science,2017,DOI: 10.1126/science.aan6814) Figure 18. By modification CVD, Robust Epitaxial Growth of 2D Single-Layer Heterojunctions, Multi-Heterojunctions, and Superlattices Figure 19. Different Growth of 2D lateral heterojunction Accurate design of high performance semiconductor films with vertical structure at the atomic scale can be used in modern integrated circuits and new materials. One way to obtain such films is to achieve continuous layer-by-layer self-assembly, that is, to use two-dimensional building materials stacked in the vertical direction and connected by van der Waals forces. Two-dimensional materials have been developed vigorously in recent years, but at present, graphene and transition metal disulfides, which are only 1 or 3 atoms thick, are used to realize some heterojunctions which are difficult to prepare earlier, and show excellent physical properties. Moreover, there is no large-scale self-assembly method that can not only maintain the intrinsic characteristics of two-dimensional material construction, but also produce interlayer interface, which limits the transformation of layer-by-layer self-assembly method to a small-scale large-scale preparation. Jiwoong Park et al., Cornell University, reported methods to achieve high levels of spatial uniformity and intrinsic sandwich interfaces to produce semiconductor thin films on a wafer-scale scale. The vertical component of the film is realized by the self-assembly of two-dimensional material modules on the atomic scale in vacuum. At the same time, some large-scale and high-quality heterojunction films and devices have been prepared, including superlattice films, resistance-adjustable tunnel junction arrays produced in batches, band-controlled heterojunction tunnel diodes and millimeter-scale ultra-thin films. The stack of films is removable, interruptible, and compatible with interfaces such as water and plastic, enabling integration with other optical and mechanical systems. (Nature,2017,Doi:10.1038/nature23905) Figure 20. High-quality semiconductor film obtained by layer-by-layer self-assembly Figure 21. Programmed vacuum stack (PVS) process In view of the strong in-plane (intra-layer) stability and the relatively weak out-of-plane (inter-layer) interaction of two-dimensional materials, such materials can be stacked with each other to form a variety of device types with a wide range of functions. To some extent, building two-dimensional material heterostructures is like building Lego bricks. In order to better control the function of two-dimensional heterostructures, it is necessary to prepare two-dimensional material modules at the level of single-layer thin films and control their single-level stacking. However, the current lift-off method for preparing the single-layer sheet has the disadvantages of high cost and difficulty in stably lift-off the two-dimensional crystal structure, so it is urgent to introduce a new preparation method to improve the existing lift-off process. A team led by Jeehwan Kim of the Massachusetts Institute of Technology in the United States has developed a general technology of layer resolution separation (LRS) to produce two-dimensional material monolayers on a wafer scale (5 cm in diameter). This technology first requires the rapid growth of thicker two-dimensional materials on the wafer, then the collection of single stacked layers of these materials, and finally the preparation of single layers through multiple splitting processes. The method can be used for preparing a plurality of single-layer materials including molybdenum disulfide and tungsten sulfide, and the van der Waals heterostructure designed and prepared on the basis has atomic-scale thickness and good performance. (Science, 2018, DOI: 10.1126/science.aat8126) Figure 22. Process for preparing monolayer two-dimensional material by LRS technology Figure 23. Fabrication of Wafer Level Monolayer 2D Materials by LRS Artificial superlattices based on van der Waals heterostructures of two-dimensional atomic crystals (such as graphene or molybdenum disulfide) have solved the problems that existing materials have not yet broken through. Typical strategies for making such artificial superlattices rely on arduous layer-by-layer peeling and re-stacking, with limited yield and reproducibility. Bottom-up methods using chemical vapor deposition can produce high-quality heterostructures, but become increasingly difficult for higher-order superlattices. While the intercalation of two-dimensional atomic crystals with alkali ions provides an alternative to superlattice structures, these generally have poor stability and severely modify the electronic properties. In the same year, Professor Duan Xiangfeng, Professor Huang Yu and Professor Liao Lei of the University of California, Los Angeles, reported an electrochemical molecular intercalation method for a new class of stable superlattices in which single molecular atomic crystals alternate with molecular layers. Using black phosphorus as a model system, the insertion of cetyltrimethylammonium bromide produces a monomolecular phosphorus molecular superlattice in which the interlayer distance is more than twice that in black phosphorus, effectively separating the phosphorus heterocyclic monolayer. A study of transistors fabricated from a single layer of phosphorus molecular superlattices shows on/off current ratios in excess of 107, as well as excellent mobility and superior stability. It has further been shown that several different two-dimensional atomic crystals, such as molybdenum disulfide and tungsten diselenide, can be intercalated with quaternary ammonium molecules of different sizes and symmetries to produce crystals with specific molecular structures, interlayer distances, phase compositions, etc. These studies define a versatile materials platform for fundamental research and potential technology applications. Figure 24. In the process of dynamic intercalation , Evolution of structure and properties from BP to MPMS Figure 25. From TEM characterization of the structural evolution of BP to MPMS Graphene is a two-dimensional carbon nanomaterial with hexagonal honeycomb lattice composed of carbon atoms with sp 2 hybrid orbitals. Graphene is considered to be a revolutionary material in the future because of its excellent optical, electrical and mechanical properties, and its important application prospects in materials science, micro-nano processing, energy, biomedicine and drug delivery. Andre Geim and Konstantin Novoselov, physicists at the University of Manchester, UK, won the 2010 Nobel Prize in Physics for successfully separating graphene from graphite by micromechanical exfoliation. The common production methods of graphene powder are mechanical exfoliation, oxidation-reduction and SiC epitaxial growth, and the film production method is chemical vapor deposition (CVD). However, due to the low yield, sub-micron thickness and poor electrical properties of single-layer graphene, the preparation process of solution exfoliated graphene sheets still faces great challenges. The monolayer graphene oxide with larger lateral size can be obtained by adopting the graphene oxide for exfoliation, and the yield is less than 100%; although a large amount of work is completed in the research, the complete removal of oxygen-containing functional groups still cannot be realized, so the obtained reduced graphene oxide has high disorder, Resulting in inferior physicochemical properties to materials prepared using chemical vapor deposition (CVD). Reduced graphene oxide (Rgo) has been considered to have potential applications in the fields of catalysis and energy, and even the disordered Rgo still has great value. How to reduce graphene oxide efficiently is an urgent problem to be solved. Professor Manish Chhowalla of Rutgers University in the United States reported a microwave method to prepare high-quality graphene in only 1-2 seconds. When the graphene material prepared by the method is used as a channel material in a field effect transistor, the electron mobility can be more than 1000m2/V/s, and the graphene material used as a catalyst carrier material shows excellent oxygen evolution catalytic performance. (Science, 2016, DOI: 10.1126/science.aah3398) The researchers used an improved hummers method to remove graphene oxide and dissolve it into multilayer graphene oxide sheets in aqueous solution. The stable graphene oxide array dispersion solution can be recombined into different forms (e.g., film-like, paper-like, fibrous) in an aqueous solution. The synthesized oxidized graphene precursor is insulating due to the oxygen-containing functional groups which are covalently connected with carbon atoms inside the oxidized graphene precursor. Oxidized graphene with a side thickness of tens of microns was placed in a traditional microwave oven and heated at 1000 W for 1-2 s to achieve the reduction of oxidized graphene. Excitation of graphene oxide under microwave conditions has been previously reported, but the reduction efficiency is still very low. Researchers first put the material under microwave for pre-annealing, which can make the material conductive, so as to absorb the microwave, so as to achieve rapid heating of oxidized graphene, and finally the microwave heating process can achieve the desorption of oxidizing functional groups, and make the surface of the material ordered. Figure 26. With intact single graphene oxide , reduced graphene oxide, and CVD-grown graphene, characterization results of the physical properties of the microwave-assisted reduced graphene oxide Scanning electron micrograph of a monolayer graphene oxide sheet deposited on a silicon substrate. The nanoarray of graphene oxide has a lateral dimension of 50 um. The Raman spectrum is consistent with the Raman spectrum of the graphene grown by CVD, showing a symmetric 2D band and a minimum D band, and the sharp Raman peaks indicate that the microwave-assisted reduction oxidized graphene has high crystallinity and microwave-assisted reducibility. The crystal size and I2D/IG of microwave-assisted reduced graphene oxide are close to those of the samples prepared by other methods, but significantly higher than those of reduced graphene oxide and dispersed graphene materials. In addition to graphene, researchers have developed a new material, boron nitride, in recent years. Known as white graphene, hexagonal boron nitride (hBN) consists of atomically flat layers of alternating hexagonal B and N atoms held together by interlayer van der Waals interactions. Unlike the excellent conductivity of graphene, hNB has extremely excellent insulating properties, which makes it play an important role in various basic scientific and technological fields, such as charge fluctuation, contact resistance, gate dielectric, passivation layer and atomic tunneling layer. Although micron-sized polycrystalline hBN has been realized and used for fundamental research, wafer-scale single-crystal hBN (SC-hBN) thin films are not yet available for practical applications. One way to synthesize sc-hBN films is to start with randomly oriented triangular particles and eventually merge them to form polycrystalline hBN (pc-hBN) films. However, the grain boundaries between randomly oriented hBN grains inevitably produce PC-hBN films, and the large number of grain boundaries in PC-hBN leads to charge scattering and site trapping, which hinders the development of high-performance electronic devices. Therefore, researchers are expected to obtain an alternative to SC-hBN films. Young Hee Le of Sungkyunkwan University, Ki Kang Kim of Dongguk University, and Soo Min Kim of KIST-Research Institute, Korea, jointly reported a method for synthesizing a wafer-scale single-crystal hBN (SC-hBN) monolayer film by chemical vapor deposition. The limited solubility of boron (B) and nitrogen (N) atoms in liquid gold promotes a high degree of diffusion of atoms at the liquid surface at high temperatures, resulting in round hBN particles. These grains further evolve into closely packed grains through the self-alignment of B and N edges formed by the electrostatic interaction between grains, and finally form the SC-hBN film on the wafer scale. This SC-HBn film also enables the synthesis of wafer-scale graphene/hBN structures and single-crystal tungsten disulfide. (Science, 2018, DOI: 10.1126/science.aau2132) Figure 27. Single crystal Synthesis of hBN film In addition to the above new methods of preparing two-dimensional materials, Kourosh Kalantar-zadeh and Torben Daeneke of RMIT University successfully prepared very thin sub-nanoscale HfO2, Al2O3 and Gd2O3 by using liquid metals. Liquid metal refers to an amorphous metal. Liquid metal can be regarded as a mixture of positive ion fluid and free electron gas. Liquid metal is also an amorphous, flowable liquid metal. Room-temperature liquid metals have many interesting surface and volume properties, which make them widely used in various engineering applications, including flexible electronic devices and microfluidics. Gallium-based eutectic alloys such as EGaIn (containing gallium and indium), gallium-indium-tin alloys are liquid at room temperature, non-toxic, and are held together by metallic bonds. Unlike molecular and ionic liquids, liquid metals are rarely used as reaction solvents. The team pointed out that although two-dimensional (2D) oxides have a wide range of applications in electronics and other technologies, many oxides are not easy to synthesize 2D materials by conventional methods. The team used a non-toxic eutectic gallium-based alloy as the reaction solvent and added the alloyed desired metal to the melt. Thermodynamically, the composition of self-limiting interfacial oxides is predicted. At the same time, the surface oxides are separated as 2D layers both on the substrate and in the suspension, and it is found that very thin sub-nanoscale HfO2, Al2O3 and Gd2O3 can be produced. The reaction route based on liquid metal can be used to produce 2D materials which can not be obtained by conventional methods before. Using liquid metal at room temperature as the reaction environment for the synthesis of low-dimensional oxide nanomaterials is another powerful tool for the method of obtaining 2D materials. (Science,2017,DOI:10.1126/science.aao4249) Figure 28. Basic principle and synthesis method Figure 29. Characterization of materials obtained by the gas injection method In addition to the above large-scale two-dimensional nanomaterials, there is another kind of two-dimensional nanosheet morphology with smaller size. In 2016, Professor Huang Xiaoqing's team prepared multi-level Pt-Co nanosheets by wet chemical reduction method, and combined with a series of advanced characterization methods and theoretical simulations, characterized in detail the ordered intermetallic structure, high-index crystal surface and platinum-rich nanostructure on the surface of multi-level Pt-Co nanosheets. The effect of this unique structure on the catalytic performance of MOR, EOR, and ORR was studied in depth. Among them, the ORR performance of Pt3Co nanosheets is the best among the reported ORR catalysts in the Pt-Co system, even comparable to that of many Pt-Ni-based catalysts. This study has important guiding value and significance for the fine control of the surface structure of one-dimensional platinum-based nanomaterials, the design of high-index crystal planes and ordered intermetallic nanostructures, and the control of the catalytic performance of nanomaterials. (Science, 2016, 354, 6318, 1410-1414) Figure 30. Structure and DFT Simulation of Pt-Co Nanosheet 4. Other Noble metal catalyst is a kind of efficient and popular catalyst at present, and the research on noble metal catalyst has been continuously carried out in recent years. Atomic-scale noble metal catalysts have the highest atom utilization efficiency, and their catalytic activity is many times higher than that of nanoparticles, clusters and bulk noble metals. Over time, catalysis by individual transition metal atoms (not just noble metals) has become increasingly popular. Figure 31. Development History of Material Morphology The process of preparing a single transition metal atom catalyst is generally to reduce the ratio of the amount of metal precursor to the amount of substrate, then to use a solvent or other means (usually using vacancies or defects in the substrate) to increase the interaction between the metal precursor and the substrate, and finally to reduce the metal precursor to a single metal atom. Of course, in this process, there is no mention of how to disperse the monoatomic metal precursor uniformly on the substrate, which is a common challenge in the study of single transition metal atom catalysts. Following the evolution in fig. 29, some of the same rules and challenges can be obtained. 1) the amount of the transition metal precursor: generally, the amount of the transition metal precursor is reduced to ensure that individual transition metal atoms are not agglomerated into clusters, nanoparticles, etc. According to some work reported in the latest literature, the ratio of the amount of metal precursor to that of the substrate is generally less than 0.5%, however, If the amount of transition metal precursor can be increased while achieving uniform dispersion of individual metal atoms, it will be beneficial to the industrialization of atomic-scale catalysts. 2) The study of catalytic mechanism: It is generally discussed whether a single transition metal atom activates its surrounding atoms to form a new active site, or whether a single transition metal atom still plays a role in reducing the activation energy of the reaction. At the same time, whether the path of a single transition metal atom participating in the catalytic reaction is consistent with that of the traditional catalytic reaction. 3) Stability: whether a single noble metal atom can stably exist on the substrate. For example, a single noble metal atom is dispersed on graphene, but due to the weak bonding ability between carbon atoms and noble metal atoms, a single metal atom will agglomerate in the catalytic reaction, so whether Pd atoms in this paper can stably exist on the titanium dioxide substrate is worth further study. In response to these challenges, Professor Zheng Nanfeng of Xiamen University and his team reported a photochemical method for preparing atomically dispersed Pd1/TiO2 catalysts with relatively high content and high stability at room temperature in Science. The first author of this paper, Dr. Liu Pengxin et al., successfully realized the stable dispersion of single-atom Pd on TiO2 nanosheets protected by ethylene glycol by photodeposition technology, and the loading of Pd atoms reached 1.5%. The Pd1/TiO2 catalyst exhibited a very high catalytic activity for the hydrogenation of carbon-carbon double bonds, and its performance was still 9 times that of the commercial Pd catalyst after 20 cycles! More importantly, the Pd1/TiO2 catalyst-eg system can catalyze the hydrogenation of aldehydes by splitting hydrogen molecules, which is 55 times higher than that of commercial Pd catalyst! Figure 32. Pd/TiO2 catalyst preparation method A certain amount of H2PdCl4 acid was added to a beaker in which titanium dioxide was uniformly dispersed, and the chloropalladate acid was adsorbed by titanium dioxide, then irradiated by low-density UV light of a xenon lamp for 10 min, and then washed with deionized water. Finally, a light gray dispersed product is obtained. Similarly, noble metal catalysts have excellent catalytic performance and have good application prospects in many aspects, such as catalytic conversion, fuel cells and so on. However, due to the high price of precious metals, the production cost has been seriously increased, which greatly hinders the actual development of precious metal catalysis. This requires us to reduce the loading of noble metals as much as possible while maintaining the high activity and stability of noble metals, such as the preparation of atomic-scale catalysts, the synthesis of hollow or core-shell catalysts. However, these methods can not accurately control the composition and size of the core and shell, and in the catalytic process, the catalyst is easy to recombine, form alloys and lose its original activity. In view of the above problems, Sean T. Hunt et al. Of MIT used a template method to prepare a single-atomic-layer noble metal-supported core-shell catalyst with a transition metal carbide surface (Science, 2016, DOI: 10.1126/science.aad8471). The noble metal salt and that transition metal oxide are wrap in a SiO2 template at first, then the SiO2 template is carbonized step by step, and finally the SiO2 template is remove, so that the transition metal carbide surface loaded single-layer noble metal core-shell structure catalyst is prepared, and the catalyst shows ultrahigh electrocatalytic activity and stability. The following is a detailed explanation of the work. Figure 33. The preparation method Figures A-E show the whole evolution process of material preparation. First, the silica-coated noble metal salt and oxide nanoparticles (A) are calcined in CH4/H2 mixed gas at 200 (B), 600 (C) and 900 (D), respectively. Finally, the SiO2 template is removed to prepare the single-layer Pt/WC core-shell structure catalyst. The lower part is the STEM diagram of each stage. The element distribution and linear scan of E diagram well illustrated the successful preparation of Pt/WC core-shell structure catalyst. In daily life, the endless flow of vehicles on the road will emit harmful gases such as CO, NO and various hydrocarbons, which will pollute the environment and reduce air quality. Therefore, in the automobile manufacturing industry, diesel oxidation catalyst materials (DOCs) are widely used in automobile exhaust emissions, in which Pt metal can play an efficient catalytic role in this oxidation reaction process (the application of noble metal catalysts based on single atoms has attracted wide attention due to their high catalytic activity and high catalytic selectivity for catalytic substrates). However, under high temperature oxidation conditions, Pt nanoparticles sintered on oxide supports tend to form larger particles (single-atom-based catalysts have strong mobility and aggregation properties, which make such materials prone to agglomeration during heating), thus reducing the catalytic effective area and thus reducing the catalytic reaction efficiency. How to keep the efficiency of the catalyst unchanged in the process of using the catalyst at high temperature for a long time has been a key research topic for the actual commercialization of the catalyst. The research team of Abhaya K. Datye (corresponding author) from the University of New Mexico in the United States used Ce oxide powder with the same surface area and different exposed crystal faces, and mixed CeO2 with Pt/Al oxide catalyst for heat treatment at 800 C in air. Researchers have ingeniously used the high-temperature mobility of metal atoms to prepare a high-temperature resistant single-atom Pt/CeO2 catalyst. The experimental results show that the Pt particles will move to the surface of CeO2 and be trapped during the high temperature heat treatment. The characterization results show that polyhedral and nanorod CeO2 are more efficient in immobilizing Pt atoms than cubic CeO2. Meanwhile, the high-temperature synthesis condition ensures that the binding sites of Pt atoms and CeO2 are in the most stable binding state, so the catalyst material with atomic-level dispersion and high temperature resistance can be prepared. (Science 353.6295 (2016): 150-154. DOI: 10.1126/science.aaf8800) Figure 34. Synthesis mechanism and properties In the field of catalysis, in addition to the current hot monatomic systems, porous materials have also attracted considerable attention due to their wide range of industrial applications, from gas separation to catalysis. The topological characteristics (in particular the size of the pores in the individual pores and the uniformity of these pores) are key factors in determining their excellent performance in a particular application. The construction of hierarchical porous structures that maintain their overall crystalline sequence is theoretically desirable because highly ordered structures can, in turn, significantly improve performance. At present, there are some studies related to the above theory, such as mesoporous TiO2 single crystals with significantly improved conductivity and electron mobility compared with nanocrystalline TiO2, and mesoporous crystalline zeolite molecular sieves have stronger framework acidity and stability compared with amorphous molecular sieves. Professor Li Yingwei of South China University of Technology and Professor Chen Banglin of the University of Texas at San Antonio have constructed highly oriented and ordered micropores in metal-organic framework (MOF) single crystals, opening up a new era of three-dimensional ordered macro-microporous materials (I. E. Materials containing both macro-and micro-pores) in the form of single crystals. This strategy benefits from the powerful modeling effect of the polystyrene nanosphere template and the dual solvent-induced heterogeneous nucleation method. This process synergistically enables the in situ growth of MOFs within the ordered voids, resulting in single crystals with oriented and ordered macroscopic microporous structures. Compared with conventional polycrystalline hollow and disordered pore ZIF-8, the improved diffusion properties of this layered framework and its stable single-crystal properties enable it to have excellent catalytic activity and recyclability for macromolecular reactions. (Science,2017,DOI:10.1126/science.aao3403) Figure 35. In-situ Synthesis of SOM-ZIF-8 Nanoparticles and Its Structural Confirmation In addition, DNA is widely used in the assembly of nanoparticles to construct highly ordered materials. Through the interaction between specific DNA sequences, nanoparticles can be controlled to form a variety of structures, including more than 30 different lattice symmetrical structures, and the distance between nanoparticles can be controlled from 3 nm to 130 nm. Compared with a variety of assembly structures of nanoparticles in solution, DNA-mediated surface assembly of nanoparticles has only a very limited number of structures. Moreover, at present, the performance of nanoparticle assembly technology, including DNA regulation, on nanoparticle surface assembly is also unsatisfactory. The size, shape, and composition of the nanoparticles involved in the formation of independent nanostructures cannot be clearly explained in the formation of two-dimensional and three-dimensional extended lattices. At present, there is still a lack of nanoparticle assembly technology that can quickly, accurately and controllably control the assembly of nanoparticles on a large surface to form the expected two-dimensional or three-dimensional structure, and can clearly explain its size, shape and composition. Three professors, Vinayak P. Dravid, Koray Aydin and Chad A. Mirkin, from Northwestern University in the United States, worked together to glue 300 nm thick PMMA on the surface of a gold-plated silicon wafer, and then use electron beam lithography (EBL) to form an orderly hole array on the PMMA. The exposed gold surface at the bottom of the well is densely decorated with DNA sequences with sticky ends. The surface of gold nanoparticles was also modified with DNA sequences with sticky ends. These DNA-modified gold nanoparticles are assembled layer by layer on the gold surface at the bottom of the PMMA pore through complementary base pairing at the sticky end. The nano-particle assembly technology which uses PMMA as a template and is regulated by DNA can control the arrangement, the interval and the sequence of the nano-particles in each assembly structure so as to realize controllable broadband absorption. In addition, this assembly technology can form responsive plasmonic nanostructures that cannot be formed by other assembly technologies. (Science,2018,DOI:10.1126/science.aaq0591) Figure 36. With Schematic Diagram of Reconfigurable Nanoparticle Assembly Regulated by DNA with PMMA as Template Figure 37. With DNA-mediated monolayer assembly of nanoparticles using PMMA as templates Figure 38. Two floors Three-layer nano particle assembly structures 5. Sum up In a word, through the unremitting efforts of researchers worldwide, the synthesis and application of nanomaterials have made gratifying progress in the past 20 years, providing a series of high-quality nanostructures with precisely controllable structures and properties at the molecular or atomic level for subsequent basic and applied research. As a functional structural unit, high-quality nano-morphology structure can also be widely used in self-assembly research. Although gratifying progress has been made in the synthesis and application of nanomaterials, its development still has a long way to go, which requires the efforts of interdisciplinary and interdisciplinary researchers to carry forward the past and forge ahead into the future. Due to the limited ability of Xiaobian, if there is any omission, please add it! This article is contributed by Z, Chen,titanium bar gr5, the academic group of the editorial department of Material Man, and edited by Material Niu. Cailiaorenvip Back to Sohu, see more Responsible Editor:. yunchtitanium.com