PDF《博士学位论文公示材料》

―1― 博士学位论文公示材料 学生姓名 巴要帅 学号

0410054 二级学科 流体机械及工程 导师姓名 巴德纯 论文题目 钛酸基材料的声子热输运调控及其热电性能研究 论文研究方向 热电技术、热电材料 论文关键词 热电材料;

氧化物;

超晶格;

复合材料;

声子输运 论文摘要(中文) 热电材料可以实现热能和电能的直接相互变换,作为清洁新能源之一近来受到越来越多的重视.

热电材料当前以 Bi2Te

3、PbTe 等合金化合物为代表应用较多,但是它们存在着有毒、成本高、热稳定 性差等问题.与此同时,环保、价格低廉、热化学稳定性高的氧化物被发现也具有一定的热电技术上 应用的潜力,迅速引起了研究者的兴趣,被期望在大规模废热回收上发挥重要作用. 材料的热电性能用无因次热电优值 ZT 来衡量, , 其中 为塞贝克系数, 为电导率, 为热导率,T 为绝对温度, 又被称为功率因子) .优异的热电性能要求材料具有高电导率、大塞贝 克系数和低热导率.与当前较为成熟的合金化合物热电材料相比,氧化物的无因次热电优值 ZT 还比较 小,主要原因是过高的声子热输运导致的高热导率.因此设法减小材料的声子热输运、降低材料热导 率成为当前氧化物热电材料的主要工作. 本文以 Nd2/3-xLi3xTiO3(简称 NLTO)和SrTiO3(简称 STO)两类钛酸盐为研究对象,分别采用构 建纳米超晶格结构和构建低热导率复合材料的方法,减小材料中声子热输运,降低材料热导率,改善 材料的电学性能,最终实现材料热电性能的提高. 对于 NLTO 来说,本论文是首个对其声子热输运及热电性能开展研究的工作.本文首先制备了具 有纳米超晶格结构的 NLTO 体晶陶瓷,并对其微观结构、声子热输运(晶格热导率)进行了研究.陶 瓷晶粒中的超晶格结构形成了大量的声子散射界面,有效地减小了声子的热输运,使得 NLTO 晶体陶 瓷表现出玻璃态热导率,且其值仅为~2W/(m?K),远低于常见的其他钛酸盐的热导率. 进一步,通过阳离子空位填充的方法进行电子掺杂,提供电子载流子,提高 NLTO 的电导率,优 化材料的电学性能,使材料从一个电导率很低的锂离子多晶陶瓷转变成为一个以电子传导为主的高电 导率的电子多晶陶瓷.空位填充后的晶体陶瓷维持了~2 W/(m・ K)的玻璃态热导率,在500K 时获得了最 高的无因次热电优值 ZT=0.019. 鉴于空位填充方法得到的晶体陶瓷的热电性能不高, 本文对 NLTO 又进行了 Ti 位Nb 掺杂的研究. 研究表明,Nb 掺杂之后,晶体陶瓷的纳米超晶格结构以及相应的玻璃态热导率同样得到了维持,同时 与阳离子空位填充方法相比,耐氧化温度的提高使其热电性能得到了进一步的改善,最终在 650K 时得 到最高的无因次热电优值 ZT=0.05. 对SrTiO3(简称 STO)的热电性能优化的研究,本文以 Sr(Ti0.85Nb0.15)O3 (简称 Nb-STO)为基体, 通过低热导率材料,包括氧化钇平衡氧化锆(Yttria-stabilized zirconia, 简称 YSZ) 、多孔二氧化硅 (Mesoporous Silica,简称 MS) 、钛酸钾(K2TiO3,简称 KTO)纳米线添加形成复合材料的方法,考察Nb-STO 微观结构、声子热输运、热导率以及热电性能的变化.复合材料的制备采用无压烧结方法 来实现.研究结果表明,低热导率材料的添加可以促进 Nb-STO 晶粒生长,增强陶瓷体的致密度,增 大载流子迁移率,使Nb-STO 的电导率得到了提高;

Nb-STO 晶粒边界处形成了低热导率的热障层,显 著减小了声子热输运, 降低了 Nb-STO 的热导率;

Nb-STO 的载流子浓度不受低热导率材料添加的影响, 使得 Nb-STO 的塞贝克系数近乎不变.最终,YSZ、MS 和KTO 纳米线的添加均大幅提升了 Nb-STO ―2― 的无因次热电优值.其中 KTO 纳米线的添加取得的无因次热电优值 ZT=0.34 在三者之中最高. 本文也对 NLTO 和Nb-STO 的研究结果从耐高温氧化性、热导率、电导率、塞贝克系数、功率因 子和无因次热电优值六个方面进行了比较,指出了两类钛酸基材料各自存在的问题及今后应开展的工 作. 本文的研究确认了具有纳米超晶格结构的 NLTO 作为新型热电材料的潜力,而对于 Nb-STO 复合 材料的研究结果表明,添加低热导率材料构建复合材料的方法可有效提升 Nb-STO 的热电性能.本文 所采用的这两种减小声子热输运、降低材料热导率、提高材料热电性能的手段也为其他氧化物乃至非 氧化材料的热电性能的研究提供了参考. ―3― 论文摘要(英文) Thermoelectric materials can interchange heat and electricity directly, and as one of green energies, have drawn more and more interests. Currently, typical thermoelectric materials include Bi2Te3, PbTe, etc. However, toxicity, high cost and low thermal stability restricts their large-scale applications. Meanwhile, the potential of oxide as thermoelectric material was found, and benefitting from their environment- friendliness, low cost and stable thermochemical properties, oxide has rapidly become one of the hot topics in thermoelectric community, which is expected to play an important role in the large-scale recovery of waste heat. Thermoelectric performance of materials is usually evaluated by dimensionless figure of merit ZT= (S2 σT)/κ, where S, σ, κ and T represent Seebeck coefficient, electrical conductivity, thermal conductivity and absolute temperature, respectively. Moreover, S2 σ is also called as power factor. Thermoelectric materials with high performance require high electrical conductivity, high Seebeck coefficient, and low thermal conductivity. Compared with current alloy thermoelectric materials, oxides'

ZTs are still low, and the main reason lies at high phonon treat transport. As a result, reducing the heat transport by phonons and decreasing the thermal conductivity have become the main work on the oxide thermoelectric materials for thermoelectric researchers. This paper focuses on Nd2/3-xLi3xTiO3 (briefly as NLTO) and SrTiO3 (briefly as STO). Approaches of preparing nano-scale superlattcie structure and constructing composite with low thermal conductivity were applied to the research on NLTO and Nb-STO respectively to decrease phonon heat transport and thermal conductivity, improve the electrical properties and advance the thermoelectric performance. For NLTO, it'

s for the first time to research its thermoelectric properties. Firstly, NLTO ceramics with nano-scale superlattice structure were prepared and their micro-strucuture, phonon heat transport and thermal conductivity were researched. Superlattice structure in the grain produces plenty of interfaces to scatter phonons, effectively reduces phonon heat transport, leading to a glass-like thermal conductivity of ~2W/(m?K) which is much lower than that of most other titanates. Furthermore, electron-doping was achieved by cation-sites vacancy compensation. Electron carriers were produced and the electrical conductivity of NLTO was drastically increased, enabling NLTO polycrystalline ceramics change from lithium ionic conductor with low electrical conductivity into electronic conductor with high electrical conductivity. NLTO with vacancy compensation still holds the glass-like thermal conductivity of ~2 W/(m・ K) and processes the maximum dimensionless thermoelectric figure of merit ZT=0.019 at 500K. To further advance ZT values, the substitution of Nb for Ti in NLTO was also performed. After the substitution, superlattice structure and the corresponding glass-like thermal conductivity was also reserved. Moreover, compared with that of cation-vacancy compensation, thermoelectric performance was improved due to the advanced oxidation-resistant temperature and the maximum ZT=0.05 was achieved at 650K. For the optimization of the thermoelectric performance of SrTiO3 (briefly as STO), taking Sr(Ti0.85Nb0.15)O3 (briefly as Nb-STO) as the matrix, via the addition of second phase, including yttria-stabilized zirconia (briefly as YSZ), meso-porous silica (briefly as MS) and K2TiO3 (briefly as KTO) nanowires, to form composites to investigate the variation of the microstructure, phonon heat transport, thermal conductivity and thermoelectric properties of Nb-STO. Composites were prepared by conventional pressureless sintering. Results show that the addition of second phase with low thermal conductivity can promote the growth of grains, strengthen the relative density, increase the carrier mobility and finally advance the electrical conductivity of Nb-STO. Meanwhile, thermal barrier was formed at grain boundaries of Nb-STO due to the addition, which dramatically reduce the phonon heat transport, lower the thermal conductivity of Nb-STO. Carrier concentration is almost unaffected as the addition, which leads to an almost unchanged Seebeck coefficient. In summary, the addition of any........