A Promising Approach to Enhanced Thermoelectric Properties Using Carbon Nanotube Networks
A Promising Approach to Enhanced Thermoelectric Properties Using Carbon Nanotube Networks
复制标题
利用碳纳米管网络增强热电性能的一种有前途的方法
DOI：
10.1002/adma.200902221
复制
发表时间：
2010-01-26
影响因子：
29.4
通讯作者：
Fan, Shoushan
中科院分区：
文献类型：
--
作者：
Meng, Chuizhou;Liu, Changhong;Fan, Shoushan
热电材料在热泵和发电机等应用中具有很大的意义，它们可以在不移动机械部件或危险工作流体的情况下实现热能和电能之间的转换。特别是由于化石燃料的燃烧已经引起了非常令人担忧的环境问题，通过热电装置将废热转化为电能变得更加迫切。热电材料的性能通过品质因数量化，由ZT 1/4 SsT/k给出，其中S、s、T和k分别是塞贝克系数、电导率、绝对温度和热导率。为了在室温下具有高ZT值，需要高S、高s和低k。已经进行了各种尝试来提高热电材料的效率。然而，根据Wiedemann-Franz定律，这三个参数之间存在很强的相关性，这使得它成为一项非常具有挑战性的任务。因此，在过去，取得了非常缓慢的改善，直到最近报道了一些有前途的方法，包括量子阱结构，具有复杂电子结构的晶体，薄膜和多层膜以及所谓的声子玻璃/电子晶体复合材料。合成复合材料被认为是一种有效的策略，以实现改善材料的性能，通过结合每个组件的优点。在介电常数和热导率等研究领域中发现了协同增强效应。然而，对于热电材料，人们认为不可能通过复合材料来增强热电性能，因为早期的理论数值模拟表明，复合材料的塞贝克系数和品质因数不可能高于其组分之一的最大值。近年来，当我们尝试用碳纳米管（CNTs）来改善聚苯胺（PANI，一种典型的导电聚合物）在复合材料中的输运性能时，发现复合材料的热电性能与其本体母体样品相比有显著的提高，这与早期关于复合材料的理论结论不一致。形态学表征表明，聚苯胺组分均匀地涂覆在每个单独的CNT和每个CNT束的表面。它是一种具有低维网络结构的复合材料。在这里，我们探讨了这些意见和survivors的原因，这种方法可以扩展到一个简单的和一般的策略，合成纳米复合材料具有增强的热电性能。研究中使用的CNT/PANI纳米复合材料是通过一个简单的两步方法制备的，即，形成一个厚的CNT网络和聚合的PANI组分。首先，通过在真空的帮助下通过微孔膜过滤均匀的CNT悬浮液，获得由随机缠结的单个CNT和CNT束制成的独立式CNT网络。其次，采用原位化学聚合法在碳纳米管网络上制备了均匀的聚苯胺涂层。该过程的细节在实验部分中给出。宏观上，原始CNT片材表面看起来非常光滑和有光泽。从扫描电子显微镜（SEM）观察，很明显，单个CNT和它们的束随机交织在一起，形成了良好的CNT网络（图1A）。原始CNT直径在约15-30 nm的范围内。在PANI聚合（0.2M苯胺）后，直径增加至约60-200 nm。图1B表明PANI在CNT的表面上形成了完全均匀的涂层。值得注意的是，在液相中的PANI涂覆过程之后，网络的框架被保留。这确实是一个形成良好的随机分布的纳米结构系统。宏观上，其表面变得有点粗糙，然而，其仍然保持柔性（图1B的插图）。它可以很容易地卷起、弯曲或扭曲，甚至折叠起来也不会开裂。这种机械性质比传统的易碎热电膜的机械性质优越上级。图1C和1D中的透射电子显微镜（TEM）图像进一步证明了涂覆有PANI的单个CNT和CNT束的纳米结构，其中一些PANI涂层在样品制备期间通过强烈的超声处理而被破坏。在CNT的外壁周围可以看到清晰的界面（由白色箭头指示）。PANI涂层的厚度为约50-90 nm。这些观察结果提供了强有力的证据表明，聚苯胺链的碳纳米管的外壁上生长。其形成机理可以简单地理解如下。在混合溶液中，苯胺盐酸盐被过硫酸铵（APS）氧化生成不溶性聚苯胺低聚物和聚合物。碳纳米管的p-键合表面与聚苯胺的共轭结构强烈相互作用，这有利于聚合物链在碳纳米管表面的沉积，形成管状涂层。在这里，单个CNT和CNT网络分别用作核心和模板。
Thermoelectric materials are of great interest for applications as heat pumps and power generators, which can realize conversion between thermal and electrical energy without moving mechanical components or hazardous working fluids. Especially since the combustion of fossil fuel has caused very alarming environmental problems, the conversion of waste heat to electric power by means of thermoelectric devices has become more urgent. The performance of thermoelectric materials is quantified by a figure of merit, given by ZT1⁄4 SsT/k, where S, s, T, and k are the Seebeck coefficient, electrical conductivity, absolute temperature, and thermal conductivity, respectively. To have a high ZT value at room temperature, high S, high s, and low k are required. Various attempts have been made to enhance the efficiency of thermoelectric materials. However, there is a strong correlation of these three parameters according to theWiedemann–Franz law, which makes it a very challenging task. Therefore, very slow improvement was achieved in the past, until some promising approaches were reported recently, which involve quantum-well structures, crystals with complex electronic structures, thin and multilayer films and so-called phonon-glass/electroncrystal compound materials. Synthesizing composites was considered an effective strategy to achieve improved material performances by combining the advantages of each component. Synergistic enhancement effects have been found in some research fields, such as dielectric permittivity and thermal conductivity. However, for thermoelectric materials, it was believed impossible to enhance thermoelectric properties through composites, because early theoretical numerical simulations indicated that the Seebeck coefficient and the figure of merit of the composites could not be higher than the maximum of one of its components. Recently, when we tried to use carbon nanotubes (CNTs) to improve the transport properties of polyaniline (PANI, which is a typical kind of conductive polymer) through composites, we found that the thermoelectric performance of the composites could be remarkably enhanced compared with both of their bulk parent samples, which is not consistent with the early theoretical conclusions about composites. Morphology characterization showed that the PANI component uniformly coated the surface of each individual CNTand of each CNT bundle. It is a composite with a low-dimensional network structure. Here, we explore the reason for these observations and surmise that this method may be extended to be a facile and general strategy to synthesize nanocomposites with enhanced thermoelectric properties. The CNT/PANI nanocomposites used in the study were fabricated by a simple two-step method, that is, the formation of a thick CNT network and the polymerization of the PANI component. First, a freestanding CNTnetwork made of randomly entangled individual CNTs and CNT bundles was obtained by filtering a uniform CNT suspension through a microporous membrane with the aid of vacuum. Second, the in situ chemical polymerization approach was used to synthesize a PANI layer uniformly coated on the prepared CNT network. Details of the process are given in the Experimental section. Macroscopically, the pristine CNT sheet surface looks very smooth and shiny. From scanning electron microscopy (SEM) observations, it was apparent that individual CNTs and their bundles randomly intertwined together to form a good CNT network (Fig. 1A). The original CNTdiameters were in the range of about 15–30 nm. After PANI polymerization (0.2 M aniline), the diameters increased to about 60–200 nm. Figure 1B indicates that PANI formed a wholly uniform coating layer on the surface of the CNTs. It is notable that the framework of the network was retained after the PANI coating process in the liquid phase. This is indeed a well-formed randomly distributed nanostructural system. Macroscopically, its surface became a little rough, however, it still remained flexible (inset of Fig. 1B). It can be rolled up, bent, or twisted easily, and even folded without cracking. This mechanical nature is superior to that of conventional fragile thermoelectric films. The transmission electron microscopy (TEM) images in Figure 1C and 1D further demonstrate the nanostructure of individual CNTs and CNT bundles coated with PANI, where some PANI coating was destroyed through intensive ultrasonication during the specimen preparation. Clear interfaces (indicated by white arrows) can be seen around the outer walls of the CNTs. The thickness of the PANI coating layer is about 50–90 nm. These observations provide strong evidence that the PANI chains grow on the outer walls of the CNTs. The formation mechanism can be understood briefly as follows. In themixed solution, aniline hydrochlorides are oxidized by ammonium peroxidisulfate (APS) to form insoluble oligomers and polymers of polyaniline. The p-bonded surface of the CNTs interacts strongly with the conjugated structure of polyaniline, which facilitates the deposition of polymer chains at the surface of the CNTs, forming a tubular coating layer. Here the individual CNTs and the CNT network serve as the core and the template, respectively.