(2) vapor-solid (VS),and (3) aqueous solution growth (ASG). The growthmechanisms will be thoroughly discussed in the nextsection.The most successful synthesis of ZnO nanostruc-tures were carried out by Wang0s group via the VSand VLS mechanisms[40–50]. They have systemati-cally synthesized almost the biggest family of knownZnO nanostructures, including nanowires (nanorods),nanobelts, nanoplatelets, nanotubes, nanocombs,nanorings, nanosprings, nanotetrapods, nanohelixes,nanocastles, and so on. They proposed and theo-retically modeled the growth of unconventional ZnOnanostructures, such as nanorings, nanospring, andnanohelixes. In their model, the polarity of ZnOcrystal lattice plays a key role[51–53]. The unconven-tional structures form as a consequence of energy min-imization by neutralizing positive and negative sur-face charges. Regarding to the formation of other con-ventional ZnO nanostructures, the growth processeswill be discussed in the following sections.2. Growth MechanismsAs we mentioned in the Introduction Section, thegrowth mechanisms for ZnO nanomaterials mainly in-clude VLS, VS, and ASG processes. These mecha-nisms have their own distinct characteristics; at thesame time, they also have similarities. Even the vaporphase growth mechanisms (VLS and VS) have manyaspects in common to the solution growth mechanism(ASG). However, the ASG growth mechanism is muchmore complicated than the vapor phase growth mech-anisms. In the following sections, we will thoroughlydiscuss the growth mechanisms, and compare theirsimilarities and differences.2.1 VLS Growth mechanismVapor-liquid-solid mechanism, with the shortname of VLS mechanism, has clear indication ofphases involved during the growth, namely that, avapor phase, a liquid phase, and a solid phase areinvolved in sequence. Precursors generally impinge asubstrate in a vapor phase. The grown materials formalloys with metal particle catalysts in a liquid phase.And the grown materials precipitate from the alloydroplets in a solid phase. There are four steps forVLS mechanism: (1) diffusion of precursor moleculesin the reaction chamber; (2) adsorption and crack-ing of the precursor followed by incorporation of theatomic species into the alloy droplets; (3) diffusion ofthe atoms through the droplets; and (4) precipitationof the grown atoms at the liquid-solid interface andincorporation into the grown 1D nanomaterials[54].Many experimental parameters are closely related tothe first step, which will be discussed in the follow-ing sections, however, the importance of the first stepwas underestimated in the past and sometimes it waseven omitted from the four steps wrongfully. It was acontroversial topic regarding which of the four steps isthe rate determining one. It was assumed that the dif-fusion step could not be rate determining, since thediffusion through the liquid alloy droplet is simplytoo fast[55], however, this assumption might not betrue, as we will explain later. Concerning the othertwo steps, Bootsma and Gassen favored step (2)[56],whereas Givargizov believed that the incorporation ofthe grown atoms into the growing 1D nanomaterialsat the liquid-solid interface is the rate determiningstep[55].
It is clear that under a steady state growthcondition, the flow of precursor materials through theVLS sequence has to be continuous. Therefore, somekinds of interactions between the neighboring stepsare necessary, and none of the four steps can be dealtas an independent process. Variation of any of thefour steps will have an influence on other steps, and ifthe response cannot be justified, the growth will de-viate from the steady growth regime. Either a newsteady state is obtained or the growth terminates.For VLS mechanism, it is well-known that a metalalloy droplet is needed to guide the growth of the1D nanomaterial, and the function of which is sim-ilar to a catalyst for a chemical reaction. Recently,metalloid particles have also been reported to guidethe growth of ZnO nanowires[57]. Both the size andthe shape of the grown material can be controlled bythe alloy droplet. Sometimes, even the orientation ofthe material also follows the crystallographic interfacematching between the grown material and the alloydroplet when it is in a solid state[58]. In general, thealloy droplet surface is assumed to be ideally rough,such that all the impinging vapor molecules are cap-tured. The word ‘capture’ has richer meaning thana simple sticking. According to Parker[59], actually,the sticking efficiency of any surface, no matter solidor liquid surfaces, is approximately 1 for the elements other than H2 and He. For a perfect flat solid surface,the sticking atoms either re-evaporate into the vaporphase or relocate to a favorable site with steps, kinks,or defects. In contrary, the sticking atoms on a dropletsurface will be immobilized and incorporated into thealloy droplets. Taking 1D nanomaterial growth viathe VLS mechanism into account, it becomes imme-diately clear that the radius r of the 1D nanomateri-als is directly related to the radius R of the droplet.Considering the situation where a droplet in thermalequilibrium is sitting on top of a cylindrical rod, addi-tionally assuming a flat interface, the radius r of therod is determined by the droplet radius R as[60]r = Rr1 ¡³¾ls¾l´2(1)where ¾ls and ¾l denote the liquid-solid interface ten-sion and the liquid surface tension of the droplet, re-spectively. From Eq.(1), at lease two significant im-plications can be drawn out. Firstly, before the ter-mination of 1D nanomaterial growth, the radius ofthe cylindrical rod is smaller than the radius of thedroplet. The larger the difference between the surfacetension for liquid-solid interface and liquid surface,the smaller the difference between the radius of therod and the droplet. When ¾ls <<¾l, r approachesthe value of R. Secondly, since a solid surface witha specified index possesses a certain interface energywith a liquid, the value of ¾ls is a variant instead of afixed number. The consequence is that the 1D nano-materials will grow with a different radius if the crys-tallographic axial direction is changed, even all otherexperimental parameters are the same.Two terms, alloy droplet and alloy particle, can beused interchangeably here. Basically, these two termswere used to indicate the same meaning: the catalyst.As we will discuss later, the catalyst may neither bea pure liquid phase nor be a pure solid phase. Andthe VLS mechanism might, under certain situations,be a vapor-solid-solid one. We describe the more gen-eral phenomena using both of these two terms. As thename indicated, VLS growth, the alloy particle shouldbe in a liquid alloy state with the grown material;however, controversies do exist over the state of thealloy particle[61–63]. Both liquid and solid states havebeen observed for the alloy particle[64]. A recent arti-cle resolved the conflict and demonstrated that bothliquid and solid states of the alloy particle can guidethe growth of 1D nanomaterials, and these two statescan even coexist in a synthesis run[64]. Whatever thestate the alloy particle is in, it acts as a favorablecapturing site for impinging atoms with 100% 氧化锌纳米材料的合成英文文献和中文翻译(2):http://www.youerw.com/fanyi/lunwen_52568.html