何琴玉， 李維謙， Hao Qing, 李　煒， 王銀珍， 曾葆青

(1.華南師范大學(xué)物理與電信工程學(xué)院，量子工程和量子材料實(shí)驗(yàn)室，廣州 510006;2.亞利桑那大學(xué)航空和機(jī)械工程系, 亞利桑那 85721, 美國(guó);3.成都電子科技大學(xué)物理電子學(xué)院，成都 610054)

熱電材料發(fā)電的商業(yè)化進(jìn)展與前景展望

何琴玉1*， 李維謙1， Hao Qing2*, 李煒1， 王銀珍1， 曾葆青3

(1.華南師范大學(xué)物理與電信工程學(xué)院，量子工程和量子材料實(shí)驗(yàn)室，廣州 510006;2.亞利桑那大學(xué)航空和機(jī)械工程系, 亞利桑那 85721, 美國(guó);3.成都電子科技大學(xué)物理電子學(xué)院，成都 610054)

綜述了熱電材料作為綠色環(huán)保能源轉(zhuǎn)換材料的商業(yè)化進(jìn)展與應(yīng)用前景. 闡述了商業(yè)化的熱電材料熱電性能指標(biāo)要求，分析了提高熱電性能的條件和理論上的局限性. 通過(guò)分析近20年提高熱電性能的研究，總結(jié)了一些突破上述局限性的有效手段，提出了將來(lái)提高熱電性能的一些有效方法. 預(yù)測(cè)了熱電性能商業(yè)化的前景.

熱電材料; 商業(yè)化; 績(jī)效因子(ZT)調(diào)制

熱電材料是一種基于溫差電效應(yīng)(Seebeck效應(yīng))的固態(tài)能量轉(zhuǎn)換材料； 涉及材料的熱、電、半導(dǎo)體性能；能將熱能轉(zhuǎn)換成電能，或?qū)崮苻D(zhuǎn)移和制冷[1]. 熱發(fā)電與蒸汽發(fā)電相比，其優(yōu)點(diǎn)在于[2]：(1)無(wú)需運(yùn)動(dòng)部件，因而無(wú)噪音.(2)設(shè)備簡(jiǎn)單. 由一塊連接外部負(fù)載的導(dǎo)線(xiàn)和置于冷熱端的熱電模塊(包括陶瓷導(dǎo)熱片)組成發(fā)電器件. (3)應(yīng)用器件大小各異. 可用于熱發(fā)電工廠、家用太陽(yáng)灶集熱發(fā)電和廢熱發(fā)電;規(guī)模和場(chǎng)所靈活多變. (4)性能穩(wěn)定，設(shè)備輕盈. 美國(guó)的“遨游號(hào)”宇航飛船即采用熱電模塊將同位素衰變時(shí)放出的熱量發(fā)電，以維持電子設(shè)備的正常運(yùn)作60多年[3]. 在太空宇航上需要持久提供電源和重量輕盈的供電器件. 所以，熱電材料的發(fā)電具有蒸汽發(fā)電無(wú)法比擬的優(yōu)點(diǎn)[4]. (5)熱電材料制冷和制熱是同一套熱電系統(tǒng)的反轉(zhuǎn)[3]. 即在熱電材料的一端制冷，那么在同樣的電源提供下該熱電模塊的另一端必然制熱. 即該熱電模塊相當(dāng)于1個(gè)熱量搬運(yùn)裝置——熱泵，將熱量從制冷端搬運(yùn)至制熱端. 這些現(xiàn)象類(lèi)似于現(xiàn)有的以液體作為工質(zhì)運(yùn)行的空調(diào).

熱電材料制冷與目前以氟利昂作為工質(zhì)的制冷方式相比，由于其制冷的運(yùn)行工質(zhì)是載流子，沒(méi)有噪音. 氟利昂排放在大氣中后破壞大氣層中能吸收紫外線(xiàn)的臭氧，導(dǎo)致大氣中紫外線(xiàn)增加. 而過(guò)量的紫外線(xiàn)輻射對(duì)人體有害. 熱電材料制冷對(duì)環(huán)境沒(méi)有破壞，綠色環(huán)保. 運(yùn)行工質(zhì)-電子的運(yùn)行對(duì)熱電模塊沒(méi)有機(jī)械磨損，壽命長(zhǎng).

熱電效應(yīng)發(fā)現(xiàn)了100多年，但熱電材料在發(fā)電、制冷、熱泵等方面的應(yīng)用仍然沒(méi)有商業(yè)化，其主要原因是熱電材料的轉(zhuǎn)換效率與制冷效率較低，無(wú)法和現(xiàn)有商業(yè)化的發(fā)電和制冷方式相競(jìng)爭(zhēng). 本文就熱電材料商業(yè)化的熱電性能指標(biāo)要求、提高熱電性能的條件和理論上的局限性、近20年突破上述局限性的一些有效手段，以及對(duì)熱電性能商業(yè)化的前景進(jìn)行闡述.

1　影響熱電模塊發(fā)電效率的內(nèi)稟因素及其調(diào)制途徑

1.1影響熱電模塊發(fā)電效率的內(nèi)稟因素

決定熱電材料能量轉(zhuǎn)換效率與制冷效率的內(nèi)稟因素主要是溫差電優(yōu)值系數(shù)(ZT):

ZT=S2σT/K,

(1)

式中，S為材料的Seebeck系數(shù)，σ為電導(dǎo)率，K為熱導(dǎo)率，T為絕對(duì)溫度. 為了獲得高ZT，熱電材料的σ與S越大越好，K越小越好. 如果ZT為2.0，熱電發(fā)電才具有商業(yè)價(jià)值；ZT接近3.0，熱電發(fā)電效率才能和現(xiàn)有的蒸汽發(fā)電的效率相當(dāng)[5]. 比如若冷端溫度為300 K，熱端溫度為700 K，則對(duì)于ZT為1.0、2.0、3.0、4.0及5.0的熱電模塊，其熱電轉(zhuǎn)換效率分別為12.84%、19.36%、23.53%、26.51%和28.78%. 但由于S、σ、K之間在物理上相互關(guān)聯(lián)，其中1個(gè)量的優(yōu)化往往使另1個(gè)或2個(gè)量向不利的方向變化[5]，3個(gè)參數(shù)難獨(dú)立調(diào)制. 使得熱電材料的商業(yè)化進(jìn)展緩慢.

Seebeck效應(yīng)由Seebeck于1821年在金屬中首次發(fā)現(xiàn).由于金屬中S較低，即Seebeck效應(yīng)較小，從發(fā)現(xiàn)到隨后的1個(gè)世紀(jì)內(nèi)僅在溫差熱電偶方面有商業(yè)應(yīng)用，尚無(wú)能量轉(zhuǎn)換方面的概念.1957年Ioffe[6]研究半導(dǎo)體材料的Seebeck效應(yīng)后，才有Seebeck效應(yīng)在能量轉(zhuǎn)換方面的探索. 對(duì)于能量轉(zhuǎn)換材料，金屬的S較低；絕緣體材料的σ較低；而對(duì)于半導(dǎo)體材料,S和σ居中，剛好合適. 因而其ZT最高，半導(dǎo)體更適合作為能量轉(zhuǎn)換材料使用[6]. 以下針對(duì)半導(dǎo)體材料進(jìn)行討論.

1.2熱電參數(shù)

1.2.1熱導(dǎo)率K半導(dǎo)體材料的熱導(dǎo)率主要來(lái)自電子熱導(dǎo)率(Ke)和聲子熱導(dǎo)率(KL)，即K=Ke+KL. 半導(dǎo)體塊體材料的Ke基本滿(mǎn)足Wiedemann-Franz定律：

Ke=L0σT，

(2)

與σ成正比，L0≈2.45×10-8V2/K2為Wiedemann-Franz常數(shù). 式(1)可以變?yōu)?/p>

ZT=S2σT/(Ke+KL)=S2σT/(L0σT+KL)=S2T/(L0T+KL/σ).

(3)

從式(3)可知：對(duì)于普通半導(dǎo)體材料，不能通過(guò)過(guò)度減小Ke降低K，因?yàn)闀?huì)直接導(dǎo)致σ下降，達(dá)不到大幅度提升ZT的效果. 關(guān)鍵在于通過(guò)降低KL來(lái)降低K. 于是Slack[7]提出一種理想的半導(dǎo)體熱電材料的模型：在半導(dǎo)體材料中，其電導(dǎo)率與單晶中一樣高，晶格熱導(dǎo)率(KL)與玻璃中一樣低， 即該材料是一種“電子晶體，聲子玻璃”. 在聲子玻璃中，聲子截止，KL=0，此時(shí)

Kmin=Ke=0.25-0.5 W/(m·K)[8].

實(shí)際材料中，即使是晶格振動(dòng)無(wú)序的非晶體，KL仍不可能為0(聲子以愛(ài)因斯坦局域化振動(dòng)模為特征，聲子主要局限在1個(gè)原子范圍內(nèi)，KL≈0.1 W/(m·K). 總之，K一般大于0.25～0.5 W/(m·K). 半導(dǎo)體熱電材料的聲子波長(zhǎng)和平均自由程在很寬的范圍分布，在高溫其能量主要由長(zhǎng)波聲子攜帶[9].

1.2.2電導(dǎo)率σ普通半導(dǎo)體的電導(dǎo)載流子是電子與空穴，由載流子濃度與遷移率μ決定:

σ=neeue+npeup,

(4)

ne、np分別為半導(dǎo)體材料中電子與空穴濃度，e為單個(gè)電子和空穴電量，ue、up分別為熱電材料中電子與空穴的遷移率. 如圖1所示，由于熱電材料所用半導(dǎo)體主要為重?fù)诫s半導(dǎo)體，處于飽和區(qū)或者過(guò)渡區(qū)，因而載流子濃度主要由雜質(zhì)濃度和溫度決定[9]. 削弱μ的主要因素是各種雜質(zhì)、缺陷、聲子、電子的散射.

1.2.3Seebeck系數(shù)S影響Seebeck效應(yīng)的因素有費(fèi)米能級(jí)、載流子速度與能量、電子能帶結(jié)構(gòu)，低溫時(shí)還與聲子速度相關(guān).

圖1　不同濃度和不同溫度下n型與p型半導(dǎo)體的費(fèi)米能級(jí)[9]

Figure 1The Fermil levels of n-type and p-type semiconductor in case of different concentration and different temperature[9]

(1)費(fèi)米能級(jí)的影響

Seebeck效應(yīng)的產(chǎn)生主要是熱端的載流子往冷端擴(kuò)散的結(jié)果. 在開(kāi)路情況下，就在半導(dǎo)體的冷端出現(xiàn)凈的多子，熱端出現(xiàn)凈的少子電荷. 此宏觀積累電荷在半導(dǎo)體內(nèi)部產(chǎn)生電場(chǎng)；當(dāng)擴(kuò)散作用與電場(chǎng)的漂移作用相互抵消時(shí)，即達(dá)到穩(wěn)定狀態(tài)，在半導(dǎo)體的兩端產(chǎn)生了由于溫度梯度所引起的電動(dòng)勢(shì)——溫差電動(dòng)勢(shì). 自然，n型半導(dǎo)體的溫差電動(dòng)勢(shì)的方向是從低溫端指向高溫端(Seebeck系數(shù)為負(fù))；相反，p型半導(dǎo)體的溫差電動(dòng)勢(shì)的方向是高溫端指向低溫端(Seebeck系數(shù)為正). 因此利用溫差電動(dòng)勢(shì)的方向即可判斷半導(dǎo)體的導(dǎo)電類(lèi)型. 物理上，Seebeck系數(shù)是平均電子能量與Fermi能級(jí)差.費(fèi)米能級(jí)隨溫度的變化率越大，Seebeck系數(shù)在該溫度段的值越大[10]. 由圖1 可知，費(fèi)米能級(jí)隨溫度的變化可以通過(guò)摻入雜質(zhì)進(jìn)行調(diào)制. 此時(shí)載流子濃度也發(fā)生了變化. 合適的雜質(zhì)摻入可以獲得較優(yōu)化的Seebeck系數(shù).

(2)載流子速度與能量

因?yàn)闊岫撕屠涠说妮d流子能量不同，因而半導(dǎo)體Fermi能級(jí)在兩端存在著差異. 這種作用也會(huì)增強(qiáng)Seebeck效應(yīng).

(3)電導(dǎo)率σ

S與載流子濃度滿(mǎn)足對(duì)數(shù)反比關(guān)系：

(5)

而如式(4)所示σ與載流子濃度成正比. 故σ越大，S越小. 其實(shí)是以上綜合因素作用的結(jié)果.

(4)聲子速度

因?yàn)闊岫说穆曌訑?shù)多于冷端，則聲子也將要從高溫端向低溫端擴(kuò)散，并在擴(kuò)散過(guò)程中與載流子碰撞時(shí)將能量傳遞給了載流子，故加速了載流子的運(yùn)動(dòng)(聲子牽引)，這種作用同樣會(huì)增加載流子在冷端的積累、增強(qiáng)Seebeck效應(yīng).

近年來(lái)調(diào)制S的方法有[11-13]：

①利用μ對(duì)能量的依賴(lài)性：增強(qiáng)μ對(duì)能量依賴(lài)性，引入強(qiáng)烈依賴(lài)于載流子濃度的散射機(jī)制，且保證遷移率對(duì)能量導(dǎo)數(shù)為正——即載流子能量增加時(shí)，其μ與S均增加. 合適的電離雜質(zhì)滿(mǎn)足這一要求.

②利用載流子濃度對(duì)能量依賴(lài)性：如摻入適當(dāng)?shù)碾s質(zhì)，該雜質(zhì)能與導(dǎo)帶(n型)或者價(jià)帶(p型)發(fā)生共振，引起導(dǎo)帶(n型)或價(jià)帶(p型)附近態(tài)密度明顯變化(如圖2的ER附近的態(tài)密度)，從而提升S[14].

③低能載流子過(guò)濾：利用顆粒納米化界面能產(chǎn)生勢(shì)壘過(guò)濾低能載流子，使S增加但σ受影響小.

由于S和σ有關(guān)，一般情況下σ增加時(shí)S會(huì)減小，故前兩種方法提升ZT幅度不大，而且適合的雜質(zhì)不多，且雜質(zhì)的摻入可能影響σ，因此前兩種方法較難獨(dú)立地調(diào)節(jié)S. 后一種方法是最近發(fā)展起來(lái)，能相對(duì)獨(dú)立地調(diào)節(jié)S而不使σ、K往不利的方向變化.

圖2　EF附近的態(tài)密度ER[14]

以上闡述說(shuō)明，σ的增加會(huì)使K增加、使S減小;這3個(gè)因素之間相互牽制. 但是自從 Glass 提出“電子晶體、聲子玻璃”的理念后，通過(guò)尋找一些特殊的材料、人工摻雜、人工結(jié)構(gòu)等方法獲得了一些準(zhǔn)“電子晶體、聲子玻璃”的材料. 如在方鈷礦中存在天然的較大空洞結(jié)構(gòu)，在其中插入較小的原子. 該原子與晶格的耦合較弱，能很好地散射聲子，但對(duì)電子的輸運(yùn)影響較??；從而實(shí)現(xiàn)降低K及提高ZT. 或者通過(guò)一些特殊的人造微結(jié)構(gòu)來(lái)散射聲子，若該微結(jié)構(gòu)對(duì)電子影響不大，但提高S、降低K，則能獲得ZT的大幅度提高. 這些晶體相比天然的材料在理念上更趨于 “電子晶體、聲子玻璃”的目標(biāo)，故稱(chēng)之為準(zhǔn)“電子晶體、聲子玻璃”.

2　熱電材料績(jī)效因子提高的標(biāo)志性成果

近年來(lái)通過(guò)S、σ、K的人工調(diào)制，熱電材料的ZT獲得了較大的提高. 表1列出了關(guān)于熱電材料及其調(diào)制機(jī)制的標(biāo)志性研究進(jìn)展.

表1　熱電材料的發(fā)展史上標(biāo)志性研究成果及其調(diào)制機(jī)制

近年來(lái)為提高熱電績(jī)效因子ZT所采取的一些有效機(jī)制總結(jié)如下：

2.1在具有天然較大空洞的材料中填入小原子

采用小原子的無(wú)規(guī)振動(dòng)散射聲子[27-30]，即讓小原子與攜熱聲子耦合，降低KL，但對(duì)σ影響不大[15,31-34]；這類(lèi)是典型的準(zhǔn)“電子晶體、聲子玻璃”. 也有另外的觀點(diǎn)，有人認(rèn)為主要是小原子與長(zhǎng)波聲子的諧振引起的KL的降低[35].

2.2能帶工程

通過(guò)摻入雜質(zhì)，該雜質(zhì)能與導(dǎo)帶(n型)或者價(jià)帶(p型)發(fā)生共振，引起導(dǎo)帶(n型)或價(jià)帶(p型)附近態(tài)密度明顯變化,提升S，降低KL[18,36-68];或通過(guò)軌道相互作用產(chǎn)生載流子囊，增加載流子濃度[39-40];或改變能帶，改變電子的有效質(zhì)量[41].

在鈦(Ti)摻雜的p型FeV0.6Nb0.4Sb中發(fā)現(xiàn)8重的能帶簡(jiǎn)并，熱電性能大幅提高[42-44].

2.3熱自旋流

自旋Seebeck效應(yīng)[45-50]：通過(guò)裁剪材料的成分、人工結(jié)構(gòu)等調(diào)制熱電參數(shù),提高熱電性能的方法總受到一些限制. 因?yàn)檩d流子引起的Ke無(wú)法避免. 自旋電子不與晶格產(chǎn)生碰撞，不傳導(dǎo)熱，打破了Ke=LσT的約束，使得對(duì)熱導(dǎo)率的調(diào)制不但可以獨(dú)立調(diào)制晶格部分，也可以獨(dú)立調(diào)制電導(dǎo)部分而不影響其他參數(shù). 同時(shí)熱自旋電子的熱電性能只需要通過(guò)調(diào)節(jié)電壓就可以調(diào)制,調(diào)制不涉及到材料本身，因而不受材料的限制，預(yù)期能有很高的ZT值.

2.4熱電材料層狀結(jié)構(gòu)制造新技術(shù)

由于在層狀結(jié)構(gòu)材料中，大都能被獨(dú)立調(diào)制1個(gè)或者多個(gè)熱電參數(shù),獲得較大的ZT值. 如具有層狀結(jié)構(gòu)的NaxCoO2-δ的ZT800 K=1.2, Bi2Sr2Co2Oy的ZT973 K=1.1，Ca2Co2O5的ZT873 K≈1.2～2.7[51-54]. 又如，最近在一些材料中調(diào)制其微結(jié)構(gòu)制備出微米級(jí)層狀結(jié)構(gòu)，在垂直層結(jié)構(gòu)的方向由于聲子受到散射降低了熱導(dǎo)率，而低能載流子被過(guò)濾又提高了Seebeck系數(shù)，由于層和層之間的晶體結(jié)構(gòu)匹配較好，電導(dǎo)率幾乎不變，只有略微降低；從而ZT大幅度提高至2～3[26-27, 55-56]. 本課題組在Cu1.94Al0.02Se中制備層狀結(jié)構(gòu)實(shí)現(xiàn)ZT從1.7提高到2.62[26]. 制備的Cu1.94Al0.02Se層結(jié)構(gòu)一致，層間晶體結(jié)構(gòu)匹配好，層橫向尺寸在微米級(jí)以上(圖3). 獲得了很低的熱導(dǎo)率和很好的Seebeck系數(shù)，電導(dǎo)率沒(méi)有顯著的降低. ZT在2.62以上，是目前塊體材料中最高的ZT. 但是Cu1.94Al0.02Se的制備條件很苛刻，對(duì)氧特別敏感而降低了熱電性能. 所以，如何將該材料在高溫使用時(shí)隔離氧，或者說(shuō)隔離氧的裝置是否廉價(jià)是該材料商業(yè)化應(yīng)用的關(guān)鍵.在研究中還發(fā)現(xiàn)：Bi0.45Sb1.55Te3.02中無(wú)層結(jié)構(gòu)時(shí)，ZT僅為1.15，形成層微結(jié)構(gòu)后ZT達(dá)到1.4.

圖3　室溫時(shí)Cu1.94Al0.02Se的SEM照片[26]

2.5摻雜重原子、增加納米結(jié)構(gòu)塊體材料中晶粒界面或形成異質(zhì)結(jié)構(gòu)勢(shì)壘

重金屬元素質(zhì)量重，其共振頻率低，與長(zhǎng)波聲子頻率相當(dāng)，能很好地散射半導(dǎo)體中晶格導(dǎo)熱的主要載體——長(zhǎng)波聲子；降低KL[57].

第二相或者異質(zhì)結(jié)形成的勢(shì)壘會(huì)散射聲子，降低熱導(dǎo)率[57-66]. 異質(zhì)結(jié)構(gòu)的勢(shì)壘為內(nèi)建場(chǎng)，內(nèi)建場(chǎng)的特點(diǎn)是對(duì)半導(dǎo)體中少子和多子有相反的作用力，能有效地降低K[59]. 納米尺寸增強(qiáng)了界面散射低能載流子和長(zhǎng)波聲子的效果，提高S，降低KL[19,57,67-72].

2.6超導(dǎo)離子體中“電子晶體、聲子液體”機(jī)制

在一些超導(dǎo)離子體中，超導(dǎo)離子做類(lèi)似于液體的無(wú)規(guī)運(yùn)動(dòng)，散射聲子，具有甚至比非晶還低的超低的K值，類(lèi)似于液體的低導(dǎo)熱能力. 其行為更像“電子晶體、聲子液體”[73]. 本課題組在超導(dǎo)離子體Cu1.94Al0.02Se塊體材料中獲得了2.62的ZT值[26]，雖然單晶的成本高，不適于商業(yè)化，但通過(guò)這些手段將熱電材料的ZT提高到2以上，從某種意義上說(shuō)是提高塊體材料ZT的又一個(gè)里程碑[26],同時(shí)說(shuō)明熱電技術(shù)進(jìn)入可商業(yè)化時(shí)代[27].

2.7調(diào)制微結(jié)構(gòu)至各向異性的織構(gòu)、層結(jié)構(gòu)與其它各向異性的微結(jié)構(gòu)

織構(gòu)[74-75]、層結(jié)構(gòu)[76-77]、各項(xiàng)異性大的材料的熱電性能好[78-81]. 織構(gòu)會(huì)引起應(yīng)力的各向異性，電子有效質(zhì)量的各向異性[82]. 層結(jié)構(gòu)在層的法線(xiàn)方向和層平面內(nèi)對(duì)聲子的散射不同. 如果讓溫差落在垂直層結(jié)構(gòu)方向，則能獲得低KL. 這一點(diǎn)在多層人工膜Bi/Cu里也得到證實(shí)[83]. 天然復(fù)雜結(jié)構(gòu)的材料具有天然低的KL.

2.8空洞散射聲子

有大量文獻(xiàn)報(bào)道在材料中引入孔洞，由于其散射低頻聲子，故能大幅降低K[28， 84-86].在材料InSb-In2Te3中，空洞、質(zhì)量起伏、應(yīng)力三者中空洞對(duì)聲子散射最強(qiáng)[87]. 在石墨烯中存在有序的空洞能大大降低K，提高熱電性能[88]. 有孔洞的鍺膜其K值也大幅降低[89].在具有納米孔洞的氧化鋅(ZnO)中存在類(lèi)似的結(jié)果[90].

還有一些其他的非主流機(jī)制增強(qiáng)熱電材料ZT的方法. 這里不一一列舉.

3　熱電發(fā)電商業(yè)化有待解決的問(wèn)題和可能的研究發(fā)展方向

3.1熱電發(fā)電商業(yè)化有待解決的問(wèn)題

熱電材料的發(fā)電商業(yè)化的瓶頸主要有：一是熱電材料的成本問(wèn)題，ZT高的塊體材料，如Cu2Se體系，Bi-Te體系等均采用多步制備，而且要求原材料是高純單質(zhì)，這些因素使這些塊體材料具有較高的成本[26]. 二是p、n型的匹配. Bi-Te體系中p型的ZT高于n型的ZT;Cu2Se體系只有p型材料，暫時(shí)還沒(méi)有好的n型材料. 對(duì)于熱電發(fā)電模塊，既需要p型的“腳”，也需要n型的“腳”.

3.2熱電發(fā)電商業(yè)化可能的研究方向

3.2.1研究體系高ZT和低成本是今后熱電材料選擇和研發(fā)的總原則. Bi-Te體系、Cu2Se超導(dǎo)離子體系、SnSe體系、β-Fe2Si體系[23]等由于其原材料成本相對(duì)低，目前獲得的體材料的ZT最高在1.5～2.62之間，因而這些體系將是今后幾年商業(yè)化研究的熱點(diǎn). 這些材料都是天然層狀微結(jié)構(gòu)，如果通過(guò)控制制備工藝可以獲得宏觀層狀結(jié)構(gòu)的材料，則可望獲得更高的ZT. 低成本大批量制備熱電材料的工藝和制備設(shè)備是在制備方面的努力方向[91]. 其次高導(dǎo)熱聚合物材料由于其柔性、質(zhì)輕、可批量生產(chǎn)等優(yōu)點(diǎn)可望成為下一個(gè)低成本熱電材料的研究熱點(diǎn)[32，92-94]. 自旋絕緣體也可能是下一個(gè)熱電材料的研究熱點(diǎn)[95-96].

3.2.2熱電系數(shù)的調(diào)制方法大體的思路還是盡量朝著“電子晶體、聲子玻璃”或者“電子晶體、聲子液體”的理念努力，在保持高σ，S的前提下盡可能降低KL. 在文獻(xiàn)[26]中提到了超導(dǎo)離子“液體”輸運(yùn)需要啟動(dòng)能量，使得KL有可能變負(fù)，將K拉得很低. 最近也有相變附近的熱電效應(yīng)研究[97]，也是想通過(guò)相變儲(chǔ)熱來(lái)阻止熱量從高溫端向低溫端輸運(yùn)，從而降低K. 這些新的研究方向有可能帶來(lái)更高的熱電性能.

4　結(jié)語(yǔ)

本文綜述了熱電材料作為綠色環(huán)保能源轉(zhuǎn)換材料的商業(yè)化進(jìn)展與應(yīng)用前景. 闡述了商業(yè)化的熱電材料的電導(dǎo)率、熱導(dǎo)率、Seebeck系數(shù)等熱電性能指標(biāo)要求，分析了提高熱電性能的條件和理論上的局限性. 通過(guò)分析近20年提高熱電性能的研究，總結(jié)了突破上述局限性的有效手段，提出了提高熱電性能的有效方法.

[1]Bozin E S. Malliakas C D, Souvatzis P, et al. Entropically stabilized local dipole formation in lead chalcogenides[J]. Science, 2010, 330(6011): 1660-1663.

[2]He J, Liu Y F. Funahashi R. Oxide thermoelectrics: The challenges, progress, and outlook[J]. Journal of Materials Research, 2011, 26(15): 1762-1772.

[3]Ji H. Caillat T. Thermoelectric materials for space and automotive power generation[J]. Material Research Science Bulletin, 2006, 31(3): 224-229.

[4]Morse J G. Energy for remote areas: Generators fueled with radionuclides are supplying power in small terrestrial and space systems[J]. Science, 1963, 139(3560): 1175-1180.

[5]Terry M T. Recent trends in thermoelectrics materials research, semiconductors and semimetals[M]. New York: Academic Press, 2001.

[6]Ioffe A F. Semiconductor thermoelements and thermoelectric cooling[M]. London: Infosearch Ltd.,1957.

[7]Slack G A. In Solid State Physics[M]. New York: Acedemic Press, 1979.

[8]Sood K C, Roy M K. Longitudinal phonons and high-temperature heat conduction in germanium[J], Journal of Physics: Condensed Matters, 1993, 5(3): 301-312.

[9]劉恩科，朱秉升，羅晉生, 等. 半導(dǎo)體物理學(xué)[M]. 西安:西安交通大學(xué)出版社，2008.

[10]Chen G. Nanoscale energy transport and conversion[M]. Oxford: Oxford University Press, 2005.

[11]Bharat J, Susanne S. Large seebeck coefficients and thermoelectric power factor of La-doped SrTiO3thin films[J]. Applied Physics Letters, 2010,97(4): 042106.

[12]Liu J, Sun Q F, Xie X C. Enhancement of the thermoelectric figure of merit in a quantum dot due to the Coulomb blockade effect[J]. Physical Review B:Condensed Matter，2010,81(24): 245323.

[13]Bell L E. Cooling, heating, generating power, and recovering waste heat with thermoelectric systems[J]. Science, 2008, 321(5895): 1457-1461.

[14]Heremans J P, Jovovic V, Toberer E S, et al. Enhancement of thermoelectric efficiency in PbTe by distortion of the electronic density of states[J]. Science,2008, 321(5888): 554-557.

[15]Sales B C, Mandrus D, Williams R K. Filled skutterudite antimonides: A new class of thermoelectric materials[J]. Science, 1996, 272(5266): 1325-1328.

[16]Duck Y C, Hogan T, Brazis P, et al. CsBi4Te6: A high-performance thermoelectric material for low-temperature applications[J]. Science, 2000, 287(2455): 1024-1027.

[17]Harman T C, Taylor P J, Walsh M P, et al. Quantum dot superlattice thermoelectric materials and devices[J]. Science, 2002, 297(5590): 2229-2232.

[18]Kuei F H, Sim L, Guo F, et al. Cubic AgPbmSbTe2+m: Bulk thermoelectric materials with high figure of merit[J]. Science, 2004, 303(5659):818-821.

[19]Poudel B, Hao Q, Ma Y, et al. High-thermoelectric performance of nanostructured bismuth antimony telluride bulk alloys[J]. Science, 2008, 320(5876): 634-638.

[20]Singh D J. Doping-dependent thermopower of PbTe from Boltzmann transport calculations[J]. Physical Review B: Condensed Matter, 2010,81(19): 195217.

[21]Jong S R, Kyu H L, Lee S M, et al. Peierls distortion as a route to high thermoelectric performance in ln4Se3-delta crystals[J]. Nature, 2009, 459(7249): 965-968.

[22]Gao P, Lu X, Berkun I, et al. Reduced lattice thermal conductivity in Bi-doped Mg2Si0.4Sn0.6[J]. Applied Physics Letters, 2014, 105(20): 202104.

[23]Pei Y Z, Shi X Y, LaLonde A, et al. Convergence of electronic bands for high performance bulk thermoelectric[J]. Nature, 2011, 473(7345): 66-69.

[24]Biswas K, He J Q, Blum I D, et al. High-performance bulk thermoelectrics with all-scale hierarchical architectures[J]. Nature, 2012, 489(7416):414-418.

[25]Zhao L D, Lo S H, Zhang Y S, et al. Ultralow thermal conductivity and high thermoelectric figure of merit in SnSe crystals[J]. Nature, 2014, 508(7496): 373-377.

[26]Zhong B, Zhang Y, Li W Q, et al. High superionic conduction arising from aligned large lamellae and large figure of merit in bulk Cu1.94Al0.02Se[J]. Applied Physics Letters. 2014, 105(12):123902.

[27]Tyagi K, Gahtori B, Bathula S, et al. Band structure and transport studies of copper selenide: An efficient thermoelectric material[J]. Applied Physics Letters, 2014, 105(17):173905.

[28]Hopkins P E, Rakich P T, Olsson R H, et al. Origin of reduction in phonon thermal conductivity of microporous solids[J]. Applied Physics Letters, 2009, 95(16):161902.

[29]Pei Y Z, Yang J, Chen L D, et al. Improving thermoelectric performance of caged compounds through light-element filling[J]. Applied Physics Letters, 2009, 95(4): 042101.

[30]Yu J, Zhao W Y, Wei P, et al. Enhanced thermoelectric performance of (Ba,In) double-filled skutterudites via randomly arranged micropores[J]. Applied Physics Letters, 2014, 104(14):142104.

[31]Prokofiev A, Sidorenko A, Hradil K, et al. Thermopower enhancement by encapsulating cerium in clathrate cages[J].Nature Materials, 2013, 12(12): 1096-1101.

[32]Voneshen D J, Refson K, Borissenko E, et al. Suppression of thermal conductivity by rattling modes in thermoelectric sodium cobaltate[J]. Nature Materials,2013,12(11): 1028-1032.

[33]Guo L, Xu X, Salvador J R, et al. Coupled vibrational modes in multiple-filled skutterudites and the effects on lattice thermal conductivity reduction[J]. Applied Physics Letters, 2013, 102(11): 111905.

[34]Paul A, Klimeck G. Tuning lattice thermal conductance by porosity control in ultrascaled Si and Ge nanowires[J]Applied Physics Letters, 2011, 98(8): 083106.

[35]Koza M M, Johnson M R, Viennois R, et al. Breakdown of phonon glass paradigm in La- and Ce-filled Fe(4)Sb(12)skutteruditesv[J]. Nature Materials, 2008, 7(10):805-810.

[36]Ohta H, Kim S, Mune Y, et al. Giant thermoelectric Seebeck coefficient of two-dimensional electron gas in SrTiO3[J]. Nature Materials, 2007, 6(2): 129-134.

[37]He Q Y, Hu S J,Tang X, et al. The great improvement effect of pores on ZT in Co(1-x)Ni(x)Sb(3)system[J]. Applied Physics Letters, 2008, 93(4):042108.

[38]Wang Q Q, Qin X Y, Li D, et al. Enhancement of thermopower and thermoelectric performance through resonant distortion of electronic density of states of beta-Zn4Sb3doped with Sm[J]. Applied Physics Letters, 2013, 102(15):154101.

[39]Zhu H, Sun W H, Armiento R, et al. Band structure engineering through orbital interaction for enhanced thermoelectric power factor[J]. Applied Physics Letters, 2014, 104(8): 082107.

[40]Liu M, Qin X Y, Liu C S, et al. Enhanced thermoelectric power factor with impurity-induced resonant level[J]. Applied Physics Letters, 2011, 99(6): 062112.

[41]Takagiwa Y, Pei Y, Pomrehn G, et al. Dopants effect on the band structure of PbTe thermoelectric material[J]. Applied Physics Letters, 2012, 101(9): 092102.

[42]Pei Y Z, Shi X Y, LaLonde, et al. Convergence of electronic bands for high performance bulk thermoelectrics[J]. Nature, 2011,473(7345): 66-69.

[43]Liu W, Tan X J, Yin K, et al. Convergence of conduction bands as a means of enhancing thermoelectric performance of n-type Mg2Si1-xSnxsolid solutions[J]. Physical Review Letters, 2012,108(16):166601.

[44]Zaitsev V K, Fedorov M I, Gurieva E A, et al. Highly effective Mg2Si1-xSnxthermoelectrics[J]. Physical Review B, 2006,74(4):045207.

[45]Jaworski C M, Myers R C, Johnston-Halperin E, et al. Giant spin Seebeck effect in a non-magnetic material[J]. Nature, 2012, 487(7406): 210-213.

[46]Uchida K, Takahashi S, Harii K, et al. Observation of the spin Seebeck effect[J]. Nature, 2008, 455(7214):778-781.

[47]Kun R J, Byoung C M, Spiesser A, et al. Voltage tuning of thermal spin current in ferromagnetic tunnel contacts to semiconductors[J]. Nature Materials, 2014, 13(4):360-366.

[48]Kirihara A, Uchida K, Kajiwara Y, et al. Spin-current-driven thermoelectric coating[J]. Nature Materials, 2012, 11(8):686-689.

[49]Bauer G E W, Saitoh E, van Wees B J. Spin caloritronics[J]. Nature Materials, 2012, 11(5):391-399.

[50]Lu H F, Zhu L C, Zu X T, et al. Thermospin effects of a quasi-one-dimensional system in the presence of spin-orbit interaction[J]. Applied Physics Letters, 2010, 96(12):123111.

[51]Fujita K, Mochida T, Nakamura K. High-temperature thermoelectric properties of NaxCoO2-δsingle crystals[C]∥IEEE，20th International Conference on Thermoelectrics. Beijing, 2001: 168-171.

[52]Funahashi R, Shikano M. Bi2Sr2Co2Oywhiskers with high thermoelectric figure of merit[J]. Applied Physics Letters, 2002, 81(8):1459-1461.

[53]Ohkubo I, Mori T. Two-dimensional layered complex nitrides as a new class of thermoelectric materials[J]. Chemistry of Materials, 2014, 26(8): 2532-2536.

[54]Funahashi R, Matsubara I, Ikuta H, et al. An oxide single crystal with high thermoelectric performance in air[J]. Journal of Japanese Applied Physics, 2000, 39 (11B): L1127-L1129.

[55]Coleman J N, Lotya M, O′N(xiāo)eill A, et al. Two-dimensional nanosheets produced by liquid exfoliation of layered materials[J]. Science, 2011, 331(6017): 568-571.

[56]Shan R, Ouardi S, Fecher G H, et al. A p-type Heusler compound: Growth, structure, and properties of epitaxial thin NiYBi films on MgO (100)[J]. Applied Physics Letters, 2012, 101(21):212102.

[57]Skoug E J, Cain J D, Morelli D T. Structural effects on the lattice thermal conductivity of ternary antimony- and bismuth-containing chalcogenide semiconductors[J].Applied Physics Letters,2010, 96(18):181905.

[58]Li H, Tang X F, Zhang Q J, et al. High performance InxCeyCo4Sb12thermoelectric materials with in situ forming nanostructured InSb phase[J]. Applied Physics Letters, 2009, 94(10):102114.

[59]Bahk J H, Shakouri A. Enhancing the thermoelectric figure of merit through the reduction of bipolar thermal conductivity with heterostructure barriers[J]. Applied Physics Letters, 2014, 105(5): 052106.

[60]Zhang B P, Cai C F, Zhu H, et al. Phonon blocking by two dimensional electron gas in polar CdTe/PbTe heterojunctions[J]. Applied Physics Letters, 2014, 104(16):161601.

[61]Li Z Y, Li J F, Zhao W Y, et al. PbTe-based thermoelectric nanocomposites with reduced thermal conductivity by SiC nanodispersion[J]. Applied Physics Letters, 2014, 104(11):113905.

[62]Quintela C X, Rodriguez G B, Rivadulla F. Thermoelectric properties of heavy-element doped CrN[J]. Applied Physics Letters, 2014, 104(2): 022103.

[63]Basu T S, Yang R G, Thiagarajan S J, et al. Remarkable thermal conductivity reduction in metal-semiconductor nanocomposites[J]. Applied Physics Letters, 2013, 103(8): 083115.

[64]Min H L, Jong S R, Vaseem M, et al. Thermoelectric properties of SrTiO3nano-particles dispersed indium selenide bulk composites[J]. Applied Physics Letters, 2013, 102(22): 223901.

[65]Lin J P, Qiao G J, Ma L Z, et al. Heterogeneous in-situ nanostructure contributes to the thermoelectric performance of Zn4Sb3[J]. Applied Physics Letters, 2013, 102(16):163902.

[66]Bahk J H, Bian Z X, Zebarjadi M, et al. Thermoelectric power factor enhancement by ionized nanoparticle scattering[J]. Applied Physics Letters, 2011, 99(7): 072118.

[67]Zebarjadi M, Esfarjani K, Shakouri A, et al. Effect of nanoparticle scattering on thermoelectric power factor[J]. Applied Physics Letters, 2009, 94(20): 202105.

[68]Wang X W, Lee H, Lan Y C, et al. Enhanced thermoelectric figure of merit in nanostructured n-type silicon germanium bulk alloy[J]. Applied Physics Letters, 2008, 93(19):193121.

[69]Ghoshal U, Ghoshal S, McDowell C, et al. Enhanced thermoelectric cooling at cold junction interfaces[J]. Applied Physics Letters, 2002, 80(16): 3006-3008.

[70]Zhao H Z, Pokheral M, Zhu G H, et al. Dramatic thermal conductivity reduction by nanostructures for large increase in thermoelectric figure-of-merit of FeSb2[J]. Applied Physics Letters, 2011, 99(16): 163101.

[71]Li Z Y, Li J F. Fine-grained and nanostructured AgPbmSbTem+2alloys with high thermoelectric figure of merit at medium temperature[J]. Advanced Energy Materials, 2014, 4(2): 1300937.

[72]Chen S, Lukas K C, Liu W S, et al. Effect of Hfconcentration on thermoelectric properties of nanostructured N-Type Half-Heusler materials HfxZr1-xNiSn0.99Sb0.01[J]. Advanced Energy Materials, 2013, 3(9): 1210-1214.

[73]Liu H L, Shi X, Xu F F, et al. Copper ion liquid-like thermoelectric[J]. Nature Materials, 2012, 11(5):422-425.

[74]Lin Y H, Lan J L, Shen Z J, et al. High-temperature electrical transport behaviors in textured Ca3Co4O9-based polycrystalline ceramics[J]. Applied Physics Letters, 2009, 97(7): 072107.

[75]Bhame S D, Pravarthana D, Prellier W, et al. Enhanced thermoelectric performance in spark plasma textured bulk n-type Bi2Te2.7Se0.3and p-type Bi0.5Sb1.5Te3[J]. Applied Physics Letters, 2013, 102(21): 211901.

[76]Hopkins P E, Piekos E S. Lower limit to phonon thermal conductivity of disordered, layered solids[J]. Applied Physics Letters, 2009, 94(18): 181901.

[77]Ishiwata S, Shiomi Y, Lee J S, et al. Extremely high electron mobility in a phonon-glass semimetal[J]. Nature Materials, 2013, 12(6):512-517.

[78]Takahashi K, Sakai A, Kanno T, et al. Tailoring of inclined crystal orientation in layered cobaltite thin films for the development of off-diagonal thermoelectric effect[J]. Applied Physics Letters, 2009, 95(5): 051913.

[79]Rhyee J S, Cho E, Lee K H, et al. Thermoelectric properties and anisotropic electronic band structure on the In4Se3-xcompounds[J]. Applied Physics Letters, 2009, 95(21): 212106.

[80]Carrete J, Mingo N, Curtarolo S. Low thermal conductivity and triaxial phononic anisotropy of SnSe[J]. Applied Physics Letters, 2014, 105(10):101907.

[81]Wan C L, Wang Y F, Norimatsu W, et al. Nanoscale stacking faults induced low thermal conductivity in thermoelectric layered metal sulfides[J]. Applied Physics Letters, 2012, 100(10):101913.

[82]Liao A D, Yao M L, Katmis F, et al. Induced electronic anisotropy in bismuth thin films[J]. Applied Physics Letters, 2014, 105(6): 063114.

[83]Kanno T, Yotsuhashi S, Sakai A, et al. Enhancement of transverse thermoelectric power factor in tilted Bi/Cu multilayer[J]. Applied Physics Letters, 2009, 94(6): 061917.

[84]Lee J H, Grossman J C. Thermoelectric properties of nanoporous Ge[J]. Applied Physics Letters, 2009, 95(1): 013106.

[85]Singh R, Bian Z X, Shakouri A, et al. Direct measurement of thin-film thermoelectric figure of merit[J]. Applied Physics Letters,2009,94 (21): 212508.

[86]Kashiwagi M, Hirata S, Harada K, et al. Enhanced figure of merit of a porous thin film of bismuth antimony telluride[J]. Applied Physics Letters, 2011, 98(2): 023114.

[87]Pei Y Z, Morelli D T. Vacancy phonon scattering in thermoelectric In2Te3-InSb solid solutions[J]. Applied Physics Letters, 2009, 94(12): 122112.

[88]Thiyagarajan P, Oh M W, Yoon J C, et al. Thermoelectric properties of nanoporous three-dimensional graphene networks[J]. Applied Physics Letters, 2014, 105(3): 033905.

[89]Stoib B, Filser S, Petermann N, et al. Thermal conductivity of mesoporous films measured by Raman spectroscopy[J]. Applied Physics Letters, 2014, 104(16): 161907.

[90]Ohtaki M, Hayashi R. Enhanced thermoelectric performance of nanostructured zno: A possibility of selective phonon scattering and carrier energy filtering by nanovoid structure[C]∥25th International Conference on Thermoelectrics, ICT′06. Vienna, Austria, 2006: 276-279.

[91]Snyder G J, Toberer E S. Complex thermoelectric materials[J]. Nature Materials, 2008, 7(2): 105-114.

[92]Chabinyc M. Thermoelectric polymers behind organics’ thermopower[J]. Nature Materials, 2014, 13(2): 119-121.

[93]Bubnova O, Khan Z U, Wang H, et al. Semi-metallic polymers[J]. Nature Materials, 2014, 13(2): 190-194.

[94]Bubnova O, Khan Z U, Malti A, et al. Optimization of the thermoelectric figure of merit in the conducting polymer poly(3,4-ethylenedioxythiophene)[J]. Nature Materials, 2011, 10(6): 429-433.

[95]Uchida K, Xiao J, Adachi H, et al. Spin Seebeck insulator[J]. Nature Materials, 2010, 9(11): 894-897.

[96]Schreier M, Roschewsky N, Dobler E, et al. Current heating induced spin Seebeck effect[J]. Applied Physics Letters, 2013, 103(24): 242404.

[97]Mi W L, Qiu P F, Zhang T S, et al. Thermoelectric transport of Se-rich Ag2Se in normal phases and phase transitions[J]. Applied Physics Letters, 2014, 104(13): 133903.

【中文責(zé)編：譚春林英文責(zé)編：肖菁】

Development in Commercialization and Application Prospect of Thermoelectric Materials

He Qinyu1*, Li Weiqian1, Hao Qing2*, Li Wei1, Wang Yinzhen1, Zeng Baoqing3

(1.Guangdong Key Laboratory of Quantum Engineering and Quantum Material, School of Physics and Telecommunication Engineering,South China Normal University, Guangzhou 510006, China;2.Department of Aerospace and Mechanical Engineering, University of Arizona, Arizona 85721, USA;3.School of Physical Electronics, University of Science and Technology of China, Chengdu 610054, China)

The development in commercialization and the application prospect of thermoelectric materials as green energy material are reviewed in this paper. The parameters requirement of thermoelectric materials in application is discussed. And the upper limit of thermoelectric properties has been analyzed. Some effective routes to break the above-said limit have been advanced after analyzing the last two decades researches in thermoelectric materials. Finally, the application prospect of thermoelectric materials is forecasted.

thermoelectric material; commercialization; figure of meritz (ZT) adjustment

2015-03-06《華南師范大學(xué)學(xué)報(bào)(自然科學(xué)版)》網(wǎng)址：http://journal.scnu.edu.cn/n

國(guó)家自然科學(xué)基金項(xiàng)目(51172078，51372092)；廣州市科技計(jì)劃項(xiàng)目(2013J4100045)

何琴玉，教授，Email: gracylady@163.com; Hao Qing, assistant research professor, Email: qinghao@email.arizona.edu.

TN304

A

1000-5463(2015)05-0009-09