高压热电材料
热电材料(全面的)

热电材料thermoelectric material将不同材料的导体连接起来,并通入电流,在不同导体的接触点——结点,将会吸收(或放出)热量.1834年,法国物理学家佩尔捷(J.C.A.Peltier)发现了上述热电效应.1838年,俄国物理学家楞次(L.Lenz)又做出了更具显示度的实验:用金属铋线和锑线构成结点,当电流沿某一方向流过结点时,结点上的水就会凝固成冰;如果反转电流方向,刚刚在结点上凝成的冰又会立即熔化成水.热电效应本身是可逆的.如果把楞次实验中的直流电源换成灯泡,当我们向结点供给热量,灯泡便会亮起来.尽管当时的科学界对佩尔捷和楞次的发现十分重视,但发现并没有很快转化为应用.这是因为,金属的热电转换效率通常很低.直到20世纪50年代,一些具有优良热电转换性能的半导体材料被发现,热电技术(热电制冷和热电发电)的研究才成为一个热门课题.目前,在室温附近使用的半导体制冷材料以碲化铋(Bi2Te3)合金为基础.通过掺杂制成P 型和N型半导体.如前所述,将一个P型柱和一个N型柱用金属板连接起来,便构成了半导体制冷器的一个基本单元,如果在结点处的电流方向是从N型柱流向P型柱,则结点将成为制冷单元的“冷头”(温度为Tc),而与直流电源连接的两个头将是制冷单元的“热端”(温度为Th). N型半导体的费米能级EF位于禁带的上部,P型的则位于禁带的下部.当二者连接在一起时,它们的费米能级趋于“持平”.于是,当电流从N型流向P型时(也就是空穴从N到P;电子从P到N),载流子的能量便会升高.因此,结点作为冷头就会从Tc端吸热,产生制冷效果.佩尔捷系数,其中是单位时间内在结点处吸收的热量,I是电流强度,Π的物理意义是,单位电荷在越过结点时的能量差.在热电材料研究中,更容易测量的一个相关参数是泽贝克(Seebeck)系数α,,其中T是温度.显然,α描述单位电荷在越过结点时的熵差.对于制冷应用来说,初看起来,电流越大越好,佩尔捷系数(或泽贝克系数)越大越好.不幸的是,实际非本征半导体的性质决定了二者不可兼得:电流大要求电导率σ高,而σ和α都是载流子浓度的函数.随着载流子浓度的增加,σ呈上升趋势,而α则下跌,结果ασ只可能在一个特定的载流子浓度下达到最大(注:由热激活产生的电子-空穴对本征载流子,对提高热电效益不起作用).半导体制冷单元的P型柱和N型柱,都跨接在Tc和Th之间.这就要求它们具有大的热阻.否则,将会加大Tc和Th间的漏热熵增,从而抵消从Tc端吸热同时向Th端放热的制冷效果.最终决定热电材料性能优劣的是组合参数,其中κ是材料的热导率.参数Z和温度T的乘积ZT无量纲,它在评价材料时更常用.目前,性能最佳的热电材料,其ZT值大约是1.0.为要使热电设备与传统的制冷或发电设备竞争,ZT值应该大于2.Glen Slack把上述要求归纳为“电子-晶体和声子-玻璃”.也就是说,好的热电材料应该具有晶体那样的高电导和玻璃那样的低热导.在长程有序的晶体中,电子以布洛赫波的方式运动.刚性离子实点阵不会使传导电子的运动发生偏转.电阻的产生来源于电子同杂质、晶格缺陷以及热声子的碰撞.因此,在完善的晶体中σ可以很大.半导体中的热导包含两方面的贡献:其一由载流子(假定是电子)的定向运动引起的(κe);其二是由于声子平衡分布集团的定向运动(κp).根据维德曼-弗兰兹定律,κe∝σ.人们不可能在要求大σ的同时,还要求小的κe.减小热导的潜力在于减小κp,它与晶格的有序程度密切相关:在长程有序的晶体中,热阻只能来源于三声子倒逆(umklapp)过程和缺陷、边界散射;在非晶态玻璃结构中,晶格无序大大限制了声子的平均自由程,从而添加了对声子的散射机制.因此,“声子-玻璃”的热导率κ可以很低.以无量纲优值系数ZT来衡量热电材料:BiSb系列适用于50—150K温区;Bi2Te3系列适用于250—500K;PbTe系列适用于500—800K;SiGe系列适用于1100—1300K.低温热电器件(T≤220K)主要用于冷却计算机芯片和红外探测器.高温热电设备可将太阳能和核能转化成电能,主要用于航天探测器和海上漂浮无人监测站的供电.最近,氟里昂制冷剂的禁用,为半导体制冷的发展提供了新的契机.1998年秋季在美国波士顿召开的材料研究学会(MRS)学术会议上,热电材料研究再一次成为讨论的热点.Brian Sales等研究了一类新型热电材料,叫作填隙方钴矿锑化物(filled skutterudite antimonides).未填隙时,材料的化学式是CoSb3(或Co4Sb12).晶体中每个Co4Sb12结构单元包含一个尺寸较大的笼形孔洞.如果将稀土原子(例如La)填入笼形孔洞,则化学式变为LaCo4Sb12.由于La原子处于相对宽松的空间内,它的振动幅值也较大.于是,在LaCo4Sb12中,Co4Sb12刚性骨架为材料的高电导提供了基础,而稀土La在笼中的振动加强了对声子的散射——减小了材料的热导.B.Sales 的工作朝着“电子-晶体和声子-玻璃”的方向迈出了第一步.高压(~2GPa)技术已经被用于改进热电材料的性能.如果在高压下观察到了母材料性能的改善,人们将可以通过化学掺杂的办法获得类似的结构,并将它用于常压条件下.ZrNiSn的σ和α都很高,但它的热导率κ并不低.或许可以通过加入第4或第5组元,增强对声子的“质量涨落散射”,达到减小热导的目的.准晶的结构复杂多变,具有“声子-玻璃”的性能.有关研究的重点是改善准晶的导电性能.将纳米金属(Ag)嵌入导电聚合物,当电流流过这种复合材料时,可以产生大的温度梯度.对此,还没有理论上的解释.有两种低维热电材料具有应用前景:CsBi4Te6实际上就是填隙的Bi2Te3;硒(Se)掺杂的HfTe5,在T<220K的温区,其泽贝克系数α远远超出了Bi2Te3.此外,薄膜、人工超晶格、纳米碳管、Bi纳米线和量子阱系统、类猫眼结构等都展现出了在改进热电材料性能方面的潜力.热电效应所谓的热电效应,是当受热物体中的电子(洞),因随着温度梯度由高温区往低温区移动时,所产生电流或电荷堆积的一种现象。
氮化硼导热率

氮化硼导热率
氮化硼是一种高导热率的陶瓷材料,具有优异的导热性能。
其导热率比金属铜高出3倍以上,达到了220-280 W/mK。
这种高导热性能使得氮化硼成为了一种重要的导热材料,在高温、高压等极端环境下应用广泛。
氮化硼的导热性能与其晶体结构有关,其晶体结构为六角晶系,由相互紧密堆积的BN层组成。
这种结构使得氮化硼具有优异的导热性能和高温稳定性,且不易受化学腐蚀。
氮化硼的高导热性能使得它被广泛应用于热导材料、散热材料、电子元器件、热电材料等领域。
例如,氮化硼导热膏被用于CPU散热器和LED散热器中,氮化硼陶瓷管可用于高压电气设备和高速列车的制动系统中。
此外,氮化硼也是一种重要的热电材料,可用于制造热电发电机和热电制冷器等设备。
总之,氮化硼的高导热性能使得它在各种极端环境下都能发挥重要的作用,是一种非常有前途的材料。
- 1 -。
k型热电偶分类

k型热电偶分类K型热电偶分类K型热电偶是一种常用的温度测量设备,广泛应用于工业领域。
根据不同的特性和用途,K型热电偶可以分为以下几个分类:1. 基本分类•标准型(K):采用镉镍合金和铂铑合金作为热电偶材料,适用于0℃至1200℃的温度测量。
•高温型(KP):使用硅化钛作为热电偶材料,可在1200℃至1700℃的高温环境下进行测量。
•超高温型(KU):采用耐高温陶瓷作为保护管,可以在1700℃至2300℃的超高温环境下进行测量。
2. 应用分类•工业型(KI):适用于工业自动化控制系统,如炼油厂、化工厂等。
•航空航天型(KA):经过特殊处理,能适应航空航天领域的高温和高压环境。
•生物医学型(KB):具有防护液体渗入的特性,适用于生物医学实验和医疗设备中的温度测量。
•全素套管式:热电偶连接头和温度点在同一轴线上,适用于一般工况下的温度测量。
•隔离套管式:热电偶被隔离在保护管内,能够防止腐蚀性气体和液体对热电偶的侵蚀。
•可曲式:热电偶的保护管经弯曲处理,便于测量弯曲或曲面表面的温度。
4. 特殊分类•防爆型(KE):符合防爆标准,适用于有爆炸危险的场所。
•防腐蚀型(KC):热电偶材料经过特殊处理,能够抵御腐蚀性气体和液体的腐蚀。
•导电型(KD):具有导电性能,适用于需要同时进行温度测量和电导率测量的场合。
以上是常见的K型热电偶分类,不同的分类适用于不同的场景和环境。
根据具体需求,可以选择合适的K型热电偶进行温度测量。
5. 精密分类•高精度型(KH):采用高精度的热电偶材料和制造工艺,具有更高的测量精度。
•超高精度型(KUH):在高精度型的基础上进一步提高了测量精度,适用于对温度要求极高的场合。
•固定端接头:热电偶连接头和检测点固定在同一位置,适用于固定位置的温度测量。
•可调端接头:热电偶连接头和检测点可以调整位置,适用于需要灵活调节位置的温度测量。
7. 长度分类•标准长度:一般采用标准长度的热电偶,常见长度为1米或2米。
电热材料及热电材料

简述电热材料及热电材料电热材料一、总述◆电热材料:利用电流热效应的材料。
一般应用于电热器。
◆性能要求:高电阻率和低的电阻温度系数,在高温时有良好的抗氧化性,并有长期的稳定性,有足够高的高温强度,易于拉丝。
◆分类:金属型和非金属型。
常用的为Ni-Cr系和Fe-Cr-Al系合金◆用途:主要用作电热器。
二、电热材料分类及其特点2.1金属电热材料●贵金属及其合金:铂、铝铂、铜铂、铂铱合金等,铱易挥发和氧化,能显著地提高铂的耐腐蚀性,具有高硬度、高熔点、高耐蚀能力和低的接触电阻。
●重金属及其合金:钨等,可用于工业炉中。
●镍基合金:铬镍合金、铬镍铁合金等。
这类合金的特点是以氧化铬构成表面保护膜,耐蚀性强,高温强度高,成型加工和焊接性能好。
缺点是价格高。
高电阻电热合金、高温合金、精密合金、耐热合金、特种合金、不锈钢等都是常见和常用的镍铬合金。
●铁基合金:铁铬铝合金、铁铝合金等。
具有高的电阻率和硬度,密度较小(6。
5~7。
2g/mm3),抗振动和抗冲击性能良好。
在450℃和700℃左右分别有脆化区,在高温下长期使用,晶粒容易粗化,因而高温抗蠕变性能和室温韧性较低,但电阻率高,抗氧化性良好,且价格便宜,因而应用广泛。
2.2非金属电热材料●碳化硅:具有优良的常温力学性能,如高的抗弯强度、优良的抗氧化性、良好的耐腐蚀性、高的抗磨损以及低的摩擦系数,而且高温力学性能是已知陶瓷材料中最佳的,其高温强度可一直维持到1600℃。
缺点是断裂韧性较低,即脆性较大。
●碳化硅粉易升华分解,一般碳化硅陶瓷都是用粉末冶金法制备。
●二硅化钼:具有金属与陶瓷的双重特性,是一种性能优异的高温材料。
极好的高温抗氧化性,抗氧化温度高达1600℃以上。
主要应用作发热元件、集成电路、高温抗氧化涂层及高温结构材料。
●石墨:耐高温性好,导电性比一般非金属矿高一百倍。
导热性超过钢、铁、铅等金属材料。
常温下有良好的化学稳定性,能耐酸、耐碱和耐有机溶剂的腐蚀。
高压处理残余应力

3.2.1残余应力模型理论值
两球接触处烧结并撤掉压力后,由于两球完全相等,根据对称原理, 撤压 后的两球之间接触面仍为原圆接触面,并只存在法向接触压力, 两球在接触处 的竖向位移仍应满足相同形式的方程式:
从力学意义上分析,暂不考虑上球,暴露赫兹接触接 触压力,而在下球接 触处叠加上希赖什尔接触压力大 小相等,方向相反的拉力,此时下球接触处的 受力状 态正好为烧结卸压后的状态。此时接触面上点竖直位 移为常数口一口’, 方向指向球心。如图3.7。上球 同理。
由于烧结块体在六个面上分别受到相同大小的压力的作用,颗粒堆积情况 非常复杂,本文模型假设的集中力与对应的远场真实烧结压力的关系也非 常复 杂,所以本文在模型的集中力与烧结压力的对应关系上,也做了一定 的简化: 将烧结颗粒做整齐排列,并只考虑单轴上的压力,从而得到烧结 压力90和集中 力P之间的对应关系:P=4R2吼,R为颗粒的半径。
当两球体以某一力P相压时,在接触点附近将发生局部 形变而出现一个边界为圆形的接触面。由于接触面的边界 半径总是远小于足和足,故可以用关于半空间体的成果来 分析此种局部形变。命M沿zl方向的位移及腹沿z:方向的 位移分别为w和%,并命zI轴上及乞轴上“距O较远处’’ 的两点相互趋近的距离为口,则M和M:之间距离的缩短 为a-(w。+w2)。这里所谓“距D较远处", a-(w +w2) D " 是指该处的形变已经可以略去不计。假定在发生局部形变 以后,M和M,成为接触面上的同一点M,则由几何关系 有
2.烧结的模型化理论
• 对烧结过程模型化,即把烧结过程分解为一系列依次进行 的烧结阶段,尽管在实际烧结过程中这些接管有时互相重 叠。烧结过程一般可以被分解为下述 7个阶段: • 1.颗粒之间形成接触; • 2.烧结颈长大(颈长); • 3.连通孔洞闭合; • 4.孔洞圆化; • 5.孔洞收缩和致密话; • 6.孔洞粗化; • 7.晶粒长大。
简述铁电、压电和热电纳米材料的催化研究

简述铁电、压电和热电纳米材料的催化研究铁电、压电和热电纳米材料近年来在催化研究领域引起了广泛关注。
这些材料具有特殊的结构和性质,对催化反应具有重要作用。
本文将对铁电、压电和热电纳米材料的催化性能进行简要介绍,并分析其应用前景。
铁电材料是一类具有铁电性质的材料,其具有正负两个永久电偶极矩的材料。
研究表明,铁电材料可以用作催化剂,提高催化反应的速率和选择性。
铁电材料的催化性能主要归因于其特殊的电荷分布和表面性质。
例如,铁电材料可以通过调节电荷重排来改变催化活性位点的结合能,从而影响催化反应的速率和选择性。
此外,铁电材料还可以通过外加电场和应力来调控其催化性能。
因此,铁电材料已被广泛应用于氧化还原反应、电催化和光催化等领域。
压电材料是一类具有压电效应的材料,其具有在外力作用下产生电荷分离的特性。
研究表明,压电材料可以用作催化剂,提高催化反应的速率和选择性。
压电材料的催化性能主要归因于其特殊的结构和电荷分布。
例如,压电材料的晶格变形可以改变催化活性位点的结合能,从而影响催化反应的速率和选择性。
此外,压电材料还可以通过外加电压来调控其催化性能。
因此,压电材料已被广泛应用于氧化还原反应、电催化和光催化等领域。
热电材料是一类具有热电效应的材料,其具有在温度梯度下产生电荷分离的特性。
研究表明,热电材料可以用作催化剂,提高催化反应的速率和选择性。
热电材料的催化性能主要归因于其特殊的热导率和电子结构。
例如,热电材料的热导率可以影响催化反应的热量传递和分子扩散,从而调控反应速率。
此外,热电材料的电子结构可以影响催化活性位点的结合能和反应中间体的稳定性,从而影响反应选择性。
因此,热电材料已被广泛应用于热催化和光催化等领域。
目前,铁电、压电和热电纳米材料在催化研究中的应用还处于起步阶段,但已经取得了一些重要的进展。
例如,一些研究发现,通过调控铁电、压电和热电纳米材料的晶格结构和表面性质,可以实现催化活性位点的精确定位和调控。
高分子压电材料及热电材料

热电材料性能表征与评价
热电优值
电导率测试
评价热电材料性能的重要指标,与材料的 电导率、热导率和塞贝克系数有关。
通过测量材料的电阻率或电导率来表征其 导电性能。
Hale Waihona Puke 热导率测试塞贝克系数测试
通过测量材料的热扩散系数、比热容和密 度来计算其热导率。
通过测量材料两端的温差和产生的电动势来 计算其塞贝克系数,表征材料的热电转换效 率。
高分子压电材料性能表征与评价
01
02
03
04
压电常数
衡量高分子压电材料压电性能 的重要参数,包括压电应变常
数和压电电压常数。
介电性能
影响高分子压电材料的极化程 度和电荷存储能力,包括介电
常数和介电损耗。
机械性能
高分子压电材料的机械强度、 刚度和韧性等对其应用性能有
重要影响。
稳定性
高分子压电材料在长期使用过 程中需要保持稳定的压电性能
特点
具有柔韧性好、易于加工成型、 成本低廉、生物相容性好等优点 ,但压电性能相对较弱。
高分子压电材料制备技术
溶胶-凝胶法
通过溶胶-凝胶过程制备 高分子压电薄膜或复合
材料。
电纺丝法
利用电场力制备高分子 压电纤维或纳米线。
压电共混法
其他方法
将压电陶瓷颗粒与高分 子基体共混,制备高分
子压电复合材料。
包括化学气相沉积、物 理气相沉积、喷涂法等。
出了多种具有优良性能的材料,并在一些领域得到了应用。
02 03
存在问题与挑战
然而,高分子压电材料和热电材料的研究仍面临一些问题和挑战,如性 能稳定性、耐温性、耐候性等方面的不足,以及制备工艺和成本等方面 的限制。
PTFE(聚四氟乙烯)的用途

PTFE独特的性能使其在化工、石油、纺织、食品、造纸、医学、电子和机械等工业和海洋作业领域都有着广泛的应用。
1、防腐蚀性能的应用由于橡胶、玻璃、金属合金等材料在耐腐蚀方面存在缺陷,难以满足条件苛刻的温度、压力和化学介质共存的环境,由此造成的损失相当惊人。
而PTFE材料以其卓越的耐腐蚀性能,业已成为石油、化工、纺织等行业的主要耐腐蚀材料。
其具体应用包括:输送腐蚀性气体的输送管、排气管、蒸汽管,轧钢机高压油管,飞机液压系统和冷压系统的高中低压管道,精馏塔、热交换器,釜、塔、槽的衬里,阀门等化工设备。
密封件的性能好坏对整个机器设备的效率与性能都有很大的影响。
PTFE材料具有的耐腐蚀、耐老化、低摩擦系数及不粘性、耐温范围广、弹性好的特性使其非常适合应用于制造耐腐蚀要求高,使用温度高于100℃的密封件。
如机器、热交换器、高压容器、大直径容器、阀门、泵的槽形法兰的密封件,玻璃反应锅、平面法兰、大直径法兰的密封件,轴、活塞杆、阀门杆、蜗轮泵、拉杆的密封件等等。
2、低摩擦性能在载荷方面的应用由于有的设备的摩擦部分不宜加油润滑,比如在润滑油脂会被溶剂溶解而失效的场合或者造纸、制药、食品、纺织等工业领域的产品需要避免润滑油沾污,这就使填充PTFE材料成为机械设备零件无油润滑(直接承受载荷)的最理想材料。
这是因为该材料的摩擦系数是已知固体材料中最低的。
其具体用途包括用于化工设备、造纸机械、农业机械的轴承,用作活塞环、机床导轨、导向环;在土木建筑工程广泛用作桥梁、隧道、钢结构屋架、大型化工管道、贮槽的支承滑块,以及用作桥梁支座和架桥转体等。
3、在电子电气方面的应用PTFE材料固有的低损耗与小介电常数使其可做成漆包线,以用于微型电机、热电偶、控制装置等;PTFE薄膜是制造电容器、无线电绝缘衬垫、绝缘电缆、马达及变压器的理想绝缘材料,也是航空航天等工业电子部件不可缺少的材料之一;利用氟塑料薄膜对氧气透过性大,而对水蒸汽的透过性小的这种选择透过性,可制造氧气传感器;利用氟塑料在高温、高压下发生极向电荷偏离现象的特性,可制造麦克风、扬声器、机器人上的零件等;利用其低折射率的特性,可制造光导纤维。
- 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。
Journal of Alloys and Compounds460(2008)8–12Effects of high-pressure high-temperature treatment on thethermoelectric properties of PbTeMichael A.McGuire a,1,Abds-Sami Malik b,Francis J.DiSalvo c,∗a Department of Physics,Clark Hall,Cornell University,Ithaca,NY14853,USAb Diamond Innovations,Worthington,OH43085,USAc Department of Chemistry and Chemical Biology,Baker Laboratory,Cornell University,Ithaca,NY14853,USAReceived21January2007;accepted22May2007Available online25May2007AbstractChanges in the thermoelectric properties of nominally undoped stoichiometric PbTe caused by high-pressure high-temperature(HPHT)treatment near6.5GPa and900◦C are parison of the electrical resistivity,thermopower,and thermal conductivity suggest that the carrier concentration and lattice defect density are decreased by HPHT treatment.Annealing the treated samples at intermediate temperatures(500–600◦C) reversed the changes.These observations are consistent with Pb vacancies being“squeezed out”by the applied pressure,and subsequently reintroduced by annealing.©2007Elsevier B.V.All rights reserved.Keywords:Semiconductors;Point defects;Electrical transport;Heat conduction1.IntroductionGood thermoelectric(TE)performance requires a specific combination of transport properties[1].In particular,the effi-ciency of a material for TE conversion increases monotonically with the parameter Z=S2/ρκ,where S is the Seebeck coefficient or thermopower,ρthe electrical resistivity,andκis the thermal conductivity.Often Z is quoted as the dimensionless product ZT,where T is the absolute temperature,and ZT values near1 are typical for good TE materials near their optimal operating temperature[2].Lead Telluride(PbTe),which adopts the rock salt crystal structure,is a good TE material when optimally doped with ZT values close to1near400◦C.Although this material has been extensively studied for many years,researchers are still working toward improving the TE performance of PbTe.Recent reviews of the TE properties of PbTe[3]and its use in TE power generation devices[4]have been published.As with other TE materials,considerable work has been devoted to ∗Corresponding author.Tel.:+16072557238;fax:+16072554137.E-mail address:fjd3@(F.J.DiSalvo).1Present address:Department of Chemistry,Frick Laboratory,Princeton Uni-versity,Princeton,NJ08644,USA.understanding the effects of synthetic techniques and materials processing on the properties of PbTe,including single crystals [5],thin-films[6],andfine grained materials[7].Since many TE devices are made from sintered polycrystalline materials, the effects of the hot-pressing process are of particular rele-vance.Recently,large improvements in the TE properties of PbTe synthesized under high-pressure high-temperature(HPHT)con-ditions have been reported[8].In contrast to hot-pressing,in which powdered PbTe is sintered under applied pressure at ele-vated temperatures(but below the melting point of PbTe which is near910◦C),the researchers loaded Pb and Te powders into a pressure cell and reacted them to form PbTe under pressure at high temperature.The reported result is a significant improve-ment in ZT at room temperature and pressure,which changed from about0.04for a cast sample to0.87for a sample made at the highest pressure(5.2GPa).Here we report the effects of post-synthetic HPHT treatment on the TE properties of nominally undoped,stoichiometric PbTe. Significant changes in the measured properties are observed after HPHT treatment,which can be partially reversed by annealing the HPHT treated samples.We propose that the observed effects are related to vacancies in the PbTe struc-ture,which may prove to be important in controlling the carrier0925-8388/$–see front matter©2007Elsevier B.V.All rights reserved. doi:10.1016/j.jallcom.2007.05.072M.A.McGuire et al./Journal of Alloys and Compounds460(2008)8–129concentration in doped PbTe TE materials prepared by hot-pressing.2.Experimental detailsInside an argonfilled glove box,pieces were broken away from a high purity Te ingot(99.9999%,Johnson Matthey)and cut from a high purity Pb rod (99.999%,ASARCO),and loaded into silica tubes in a1:1molar ratio(total sample mass of10g per tube).A conservative estimate of the uncertainty in the loaded compositions,assuming an error of±0.5mg in weighing the starting materials,is about±0.01%.After removal from the glove box the tubes were quickly attached to a vacuum line for sealing,minimizing exposure of the reac-tants to air.The sealed tubes were then heated in an electrical box furnace.The best PbTe ingots,which included the fewest and smallest voids,were made by heating at930◦C for1h and then cooling to850◦C over6h.Powder X-ray diffraction and electron microprobe analysis showed the product to be single phase PbTe.A rectangular parallelepiped sample was cut from the ingot using a dia-mond wire saw,and subsequently used for TE property measurements.HPHT treatment of a coarse powder of this same material was performed at Diamond Innovations using a hydraulic press at∼6.5GPa and∼900◦C.Powder X-ray diffraction under ambient conditions after HPHT treatment confirmed that the sample was still single phase with the NaCl structure type.The sample was cut into several rectangular bars using wire electrical discharge machining(EDM) at Diamond Innovations.The surfaces of the bars were cleaned with SiC paper prior to characterization.Inspection of polished surfaces of the samples by elec-tron microprobe in composition mode showed no significant difference between the cast sample and the HPHT treated material(Fig.1).Annealing of the HPHT treated samples was carried out in evacuated,sealed silica tubes in an electrical box furnace.Before annealing,the surfaces of the pellets were cleaned with a file and SiC paper and rinsed with hexanes to remove any contamination from the application of the electrical and thermal contacts.Preliminary attempts to synthesize PbTe from Pb and Te powders under HPHT conditions following Zhu et al.[8]were not position maps and wavelength dispersive electron microprobe analysis showed the prod-ucts of these reactions to be multiphase,made up of a PbTe matrix incorporating Te rich regions near the composition“PbTe3”and Pb rich regions containing Pb,Te and O(Fig.1).This is likely due to the presence of an oxide layer on the surface of thefine Pb powder starting material.Measurements of TE properties were performed over the temperature range of100–300K using the home built apparatus described elsewhere[9].Copper was plated onto the ends of the samples so that good thermal and electrical contacts could be made using indium metal for the measurement of thermal conductivityκ,thermopower S,and electrical resistivityρ.The voltage contacts for the resistivity measurements were made usingfine gauge copper wire and silver paste(Dupont Conductor Composition4922N).Based on measurements of reference materials with similar values of S andκ,and on errors associated with geometrical measurements,conservative estimates of the uncertainties on these measurements are given as±3%for S,±5%forρ,and±15%forκ. 3.Results and discussionThe TE properties of two bars cut from the HPHT treated material are shown in Fig.2,along with the data from the cast sample before HPHT treatment for comparison.We willfirst address the properties of the cast sample,and then discuss the effects of the HPHT treatment.The positive values of S and the low values ofρwhich increase with temperature indicate that the cast material may be best described as a heavily p-doped semiconductor.Stoichio-metric PbTe is a small band gap semiconductor and is expected to be n-type since the mobility of electrons in the Pb-based conduc-tion band is greater than that of holes in the Te-based valence band.The observed p-type behavior of the cast samplemay Fig.1.Electron microprobe(JEOL8900,15kV,20nA)composition maps for conventionally synthesized PbTe before(a)and after(b)HPHT treatment,and for HPHT synthesized PbTe(c).The light grey regions in all three maps are PbTe. The medium grey regions in(c)are Te rich with compositions near“PbTe3”. The dark grey regions in(c)contain Pb,Te,and O.indicate the presence of Pb vacancies.The congruently melt-ing composition in the Pb–Te phase diagram has been reported as PbTe1.003[5],equivalent to about5×1019Pb vacancies per cubic centimeter.If each Pb vacancy acts as a p-type dopant, degenerate behavior is expected,as observed in the cast sample.10M.A.McGuire et al./Journal of Alloys and Compounds460(2008)8–12Fig.2.The measured electrical resistivityρ,thermopower S,and thermal con-ductivityκof conventionally synthesized PbTe and two samples of the same material after HPHT treatment.Also shown is ZT=S2T/ρκcalculated from the measured data.Fig.2shows that HPHT processing had little effect on themeasured thermal conductivity,but dramatically changed S and ρ.The resistivity increased by over two orders of magnitude, and the temperature dependence changed to display more of anactivated type of behavior.HPHT treatment changed the sign ofthe thermopower,indicating that the conduction is dominatedby electrons in the HPHT processed sample.Due to the largeincrease in resistivity,ZT of the HPHT treated material is signif-icantly diminished,from about0.2for the cast sample to about0.005for the HPHT treated sample at300K.In situ measurements of Pb chalcogenides under pressures up to∼10GPa have revealed changes inρand S of similar magnitude to those shown in Fig.2[10,11].In those investi-gations,samples were subjected to pressure,without heating, and it was found that the changes were reversible upon release of pressure.The observed changes inρand S with increas-ing pressure were attributed to a narrowing of the band gap and subsequent metallization,followed by a structural phase transition near6GPa.At this pressure the cubic lattice of PbTe distorts to orthorhombic symmetry.Much of the earlier literature ascribes the GeS structure-type to this phase;how-ever,recent synchrotron powder diffraction experiments have shown that it is a different orthorhombic modification[12].It is proposed that this structural phase transition is electronically driven,resulting in the opening of a band gap and a transi-tion back into a semiconducting state with a high thermopower [10,11].We believe that the changes we observe after HPHT treatment(Fig.2)arise from a fundamentally different mecha-nism than those observed at high pressure in previous investi-gations.The n-type conduction,larger thermopower,higher resistiv-ity,and activated behavior of the material after HPHT treatment suggest that these samples are closer to intrinsic PbTe than the cast material(Fig.2).We propose that this may be attributed to the“squeezing out”of Pb vacancies from the cast material.To test this hypothesis,a HPHT treated sample was annealed for 12h at500◦C,and then again for12h at600◦C.The TE proper-ties of the sample were measured after each anneal.It is expected that annealing should allow vacancies to diffuse back into the bulk.The samples were not visibly changed by the annealing processes.No sign of grain growth was seen on the sample sur-faces,and no vapor transported material was observed in the silica tube.The results of the properties measurements after annealing are shown in Fig.3.Annealing the samples is seen to have decreased the resistivity to values and behavior intermediate between the cast material and HPHT treated samples.After annealing,the thermopower was again positive,but greater than that in the cast sample.These observations indicate that the density of p-type dopants increased during annealing,but not up to the level present in the cast sample.This is consis-tent with Pb vacancies beginning to diffuse back into the bulk of the material from the sample surface or the grain bounda-ries.The thermal conductivity of the annealed samples is lower than that of the cast sample and the HPHT treated samples (Fig.3).It is important to note that the measured values shown in Fig.3are the total thermal conductivities,which include contributions from the lattice as well as the electronic charge carriers.To understand the behavior of the thermal con-ductivity in this series of measurements,it is instructive to consider separately the lattice thermal conductivityκL,by sub-tracting the electronic contributionκe,which is related to the electrical resistivity by the Wiedemann–Franz lawκe=L0T/ρ, L0=2.45×10−8W /K2[13].The results are shown in Fig.4. HPHT treatment clearly increased the lattice thermal conductiv-ity.This is consistent with phonon scattering defects(vacancies)M.A.McGuire et al./Journal of Alloys and Compounds460(2008)8–1211Fig.3.The measured electrical resistivityρ,thermopower S,and thermal con-ductivityκafter annealing of HPHT treated,conventionally synthesized PbTe. Also shown is ZT=S2T/ρκcalculated from the measured data.being removed by HPHT processing.Fig.3also shows that annealing the HPHT treated samples decreased the lattice ther-mal conductivity,consistent with vacancies diffusing back into the material.A decrease in lattice thermal conductivity is expected in hot-pressedfinegrained materials due to enhanced grain boundary scattering of heat carrying phonons and has been observed exper-imentally[14–16].This phenomenon does not seem to play a significant role in the behavior of the thermal conductivity in the present study.This is consistent with SEM observations,which Fig. 4.The lattice thermal conductivity of PbTe samples as cast,after HPHT treatment,and after subsequent annealing.The electronic contribu-tion was subtracted from the measured total thermal conductivity using the Wiedemann–Franz law[13].did not show afine grained sintered compact,but rather a smooth appearance with no obvious grain structure.It is not surprising that the high pressures(6.5GPa)and high temperatures(900◦C, near the melting point of PbTe)used here did not produce the fine grained structure observed in many hot-pressed refractory materials.Longer annealing times were also briefly explored.After5 days at500◦C small PbTe crystals(approximately0.5mm)had formed on the surface of the pellet and on the tube walls.This likely indicates the presence of some impurities in the sample or tube which acted as vapor transport agents for the crystals. After the5-day anneal,this sample had a negative thermopower and low resistivity(close to that of the cast sample).This is not consistent with the explanation given above for the effects of the shorter anneals,and suggests an increase in the concentra-tion of some n-type dopant species.This is likely due to some unidentified impurity unintentionally introduced into the sample or tube,or to a change in composition of the bulk sample due to the transport of material away from the pellet.Further work is required to understand how these small crystals are transported and how they grow.4.ConclusionsHigh-pressure high-temperature tuning of the TE proper-ties of conventionally synthesized,nominally undoped PbTe was demonstrated.The effects of HPHT processing presented above are satisfactorily explained by the removal of lattice defects,which act like p-type dopants(possibly Pb vacancies) by HPHT treatment,and their reintroduction through annealing. This may prove to be a technologically important observation, since PbTe TE materials are sometimes prepared by hot-pressing,and since it is expected that vacancy concentration may influence the effectiveness of intentionally introduced dopants. These results suggest that further work should be carried out to examine the use of pressure to control the defect density, and therefore the carrier concentration,in optimally doped PbTe.12M.A.McGuire et al./Journal of Alloys and Compounds460(2008)8–12AcknowledgementsThis work was funded by Cornell University.We thank John W.Lucek at Diamond Innovations for helpful discussions regarding the HPHT work,and John Hunt for assistance with the electron microprobe facility,which is part of the Cornell Center for Materials Research(MRSEC Grant DMR-0520404). References[1]R.Heikes,R.Ure Jr.,Thermoelectricity:Science and Engineering,Firsted.,Interscience Publishers Inc.,1961.[2]G.Mahan,Solid State Phys.51(1997)81–157.[3]Z.H.Dughaish,Physica B322(2002)205.[4]J.Yang,T.Caillat,MRS Bull.31(2006)224.[5]I.K.Avetisov,A.Y.Mel’kov,A.Y.Zinov’ev,E.V.Zharikov,Crystallogr.Rep.50(2005)S124.[6]K.Kishimoto,M.Tsukamoto,T.Koyanagi,J.Appl.Phys.92(2002)5331.[7]K.Kishimoto,T.Koyanagi,J.Appl.Phys.92(2002)2544.[8]P.Zhu,X.Jia,H.Chen,L.Chen,W.Guo,D.Mei,B.Liu,H.Ma,G.Ren,G.Zou,Chem.Phys.Lett.359(2002)89.[9]T.Reynolds,M.McGuire,F.DiSalvo,J.Solid State Chem.177(2004)2998.[10]V.Shchennikov,S.Ovsyannikov,A.Derevskov,Phys.Solid State44(2002)1845.[11]S.Ovsyannikov,V.Shchennikov,Phys.Status Solidi B241(2004)3231.[12]G.Rousse,S.Klotz,A.Saitta,J.Rodriguez-Carvajal,M.McMahon,B.Couzinet,M.Mezouar,Phys.Rev.B71(2005)224116.[13]N.W.Ashcroft,N.D.Mermin,Solid State Physics,First ed.,SaundersCollege,1976.[14]J.Parrott,J.Phys.C:Solid State Phys.2(1969)147.[15]N.Savvidest,H.Goldsmid,J.Phys.C:Solid State Phys.13(1980)4671.[16]N.Savvidest,H.Goldsmid,J.Phys.C:Solid State Phys.13(1980)4657.。