[0016] 以下針對本發明形態之一例詳細說明，但本發明不限定於此。又，本說明書中只要未特別記載，則表示數值範圍之「A~B」意指「A以上B以下」。 [0017] [1. 與n型導電材料之性能有關之指標] 首先，針對與n型導電材料之性能有關之指標加以說明。作為該指標舉例為輸出因子(功率因子(power factor))。輸出因子係藉由以下之式(1)求出。 [0018] PF=α2
σ (1) 式(1)中，PF表示輸出因子，α表示賽貝克係數，σ表示導電率。n型導電材料中，例如輸出因子於310K較好為100μW/mK2
以上，更好為200μW/mK2
以上，特佳為400μW/mK2
以上。輸出因子於310K若為100μW/mK2
以上，則由於為與以往型之p型導電材料同等或高於其之值故而較佳。為了獲得如此之高輸出之n型導電材料，而思考提高賽貝克係數或導電率之任一者，獲得其兩者。 [0019] 所謂賽貝克係數係指顯示賽貝克效應之電路之高溫接合點與低溫接合點之間之對於溫度差之開放電路電壓之比(取自「McGraw-Hill科學技術用語大辭典第3版」)。賽貝克係數例如可使用後述實施例所用之賽貝克效應測定裝置(MMR Technologies公司製)等測定。賽貝克係數之絕對值越大，表示熱起電力越大。 [0020] 又，賽貝克係數可成為用以判別碳奈米管等之電子材料之極性的指標。具體而言，例如賽貝克係數顯示正值之電子材料可謂具有p型導電性。相對於此，賽貝克係數顯示負值之電子材料可謂具有n型導電性。 [0021] n型導電材料中，賽貝克係數較好為-20μV/K以下，更好為-30μV/K以下，又更好為-40μV/K以下。但，於使用低溫熱源等之微小能量進行發電時，隨著熱起電力增大導電率亦增大，而亦有需要抑制對升壓電路所要求之阻抗之情況。該情況下賽貝克係數更好為-40~-20μV/K。 [0022] 導電率例如可藉由使用電阻率計(三菱化學分析公司製，LORESTA GP)之4探針法測定。 [0023] n型導電材料中，導電率較好為1000S/cm以上，更好為1500s/cm以上，又更好為2000S/cm以上。導電率若為1000S/cm以上，則由於n型導電材料為高輸出故而較佳。 [0024] 又，作為n型導電材料之性能有關之另一指標舉例為無次元性能指數ZT。ZT係藉由以下之式(2)求得。 [0025] ZT=PF．T/κ (2) 式(2)中，PF為輸出因子(=α2
σ)，T表示溫度，κ表示熱傳導率。ZT越大，表示越優異之n型導電材料。由式(2)可知為了獲得較大之ZT，輸出因子亦即賽貝克係數之絕對值及導電率較大較好。 [0026] 又，由式(2)可知為了獲得較大之ZT，熱傳導率越小越好。此係與熱電轉變材料(亦即n型導電材料及p型導電材料)係利用溫度差者對應。熱電轉換材料之熱傳導率較大時，會使物質中之溫度容易變均一，不易產生溫度差。因此，具備熱傳導率較大之熱電轉變材料之熱電轉變裝置有難以有效率地發電之傾向。 [0027] [2. n型導電材料] 本發明之一實施形態之n型導電材料(以下亦稱為本n型導電材料)之特徵為包含內包金屬錯合物之n型碳奈米管。本說明書中，所謂「內包金屬錯合物」意指金屬錯合物位於碳奈米管之空洞內部。本n型導電材料中，金屬錯合物是否內包於碳奈米管可藉由例如後述之實施例所示般以透過型電子顯微鏡觀察而調查。 [0028] 又，金屬錯合物可全部內包於碳奈米管，亦可僅一部份內包於碳奈米管。 [0029] 上述n型碳奈米管成為負電荷非局部化之狀態，成為軟性鹼(soft base)。另一方面，金屬錯合物成為正電荷非局部化之軟性酸(soft acid)。藉由軟性酸對於軟性鹼作為而可安定化。因此，本n型導電材料藉由金屬錯合物對n型碳奈米管作用，而顯示安定之n型導電性。又，軟性酸及鹼之定義係基於HSAB理論(R.G. Pearson, J. Am. Chem. Soc. 85(22), 3533-3539, 1963)。 [0030] 再者，本n型導電材料由於金屬錯合物被內包於n型碳奈米管，故具有優異之熱電特性及化學安定性。此推測係於金屬錯合物被內包於n型碳奈米管時，與金屬錯合物僅附著於n型碳奈米管之情況相比，金屬錯合物難以脫離之故。 [0031] 本n型導電材料可根據需要含有n型奈米管及金屬錯合物以外之物質。作為此等物質，若為不阻礙金屬錯合物之上述效果則未特別限定。 [0032] ＜2-1. n型碳奈米管＞ 本n型導電材料含有n型奈米管。n型奈米管只要具有n型導電性即可，可為藉由任何方法n型化者。又，本說明書中，碳奈米管亦有稱為「CNT」之情況。 [0033] 上述碳奈米管亦可具有單層或多層(兩層、三層、四層或比其多之多層)之構造。例如上述碳奈米管可為單層碳奈米管(single-wall carbon nanotube：SWNT)或多層碳奈米管(multi-wall carbon nanotube：MWNT)。 [0034] 本n型導電材料作為熱電轉換裝置等而認為有各種應用及用途。此處，熱電轉換裝置若具有柔軟性，則可密著於人體或配管等之複雜三次元表面，可有效地利用於體溫或廢熱等故而較佳。為了增加熱電轉換裝置之柔軟性，本n型導電材料中，基於賦予優異機械特性(拉身強度、楊氏模數及彈性率等)之觀點，上述碳奈米管較好為單層碳奈米管。 [0035] 上述n型碳奈米管之平均內徑並未特別限定，例如較好為0.6~1000nm，更好為0.6~100nm，又更好為0.8~20nm，特佳為1~5nm，最好為1.2~3nm。平均內徑若為0.6nm以上，則可於n型碳奈米管內部充分納入金屬錯合物。再者，平均內徑若為1.2nm以上，則由於金屬錯合物更易進入n型碳奈米管內部故而較佳。且，平均內徑若為1000nm以下，則抑制金屬錯合物於n型碳奈米管內部擴散，可獲得常壓常溫下之化學安定性故而較佳。再者，平均內徑若為3nm以下，則金屬錯合物難以自n型碳奈米管內部脫離，或某局面中可獲得優異之熱電轉換特性故而更佳。 [0036] n型碳奈米管之平均內徑可藉由例如透過型電子顯微鏡或掃描型電子顯微鏡觀察n型碳奈米管而測定。例如藉由透過型電子顯微鏡或掃描型電子顯微鏡觀察n型導電材料之任意5處。對每1處隨機選擇10根n型碳奈米管，測量各內徑，將所測量之50根n型碳奈米管之內徑平均值設為n型碳奈米管之平均內徑。 [0037] 上述碳奈米管可成形為期望形狀。例如本n型導電材料可含有將碳奈米管積集而成之薄膜。此處，上述「薄膜」亦可改稱為薄片或膜。薄膜亦可為例如1μm~1000μm之厚度。薄膜之碳奈米管之密度並未特別限制，但可為0.05~1.0g/cm3
，亦可為0.1~0.5g/cm3
。上述薄膜係碳奈米管彼此相互絡合形成不織布狀之構造。因此，上述薄膜為輕量且具有柔軟性。 [0038] ＜2-2. 金屬錯合物＞ 本n型導電材料含有金屬錯合物。本說明書中之金屬錯合物意指使金屬陽離子與配位子配位鍵結而成者。 [0039] 作為金屬陽離子舉例為典型金屬離子(鹼金屬離子及鹼土金屬離子)及過渡金屬離子等。作為上述金屬離子舉例為鋰離子、鈉離子、鉀離子、銣離子、銫離子、鍅離子、鈹離子、鎂離子、鈣離子、鍶離子、鋇離子、鐳離子及鈧離子等。 [0040] 作為上述配位子若為與金屬陽離子形成配位鍵結之化合物則未特別限制。換言之，上述配位子係具有對金屬陽離子之配位單元之化合物。作為上述配位子舉例為例如有機配位子等。有機配位子可為單齒配位子(亦即具有1個配位單元之化合物)，亦可為多齒有機配位子(亦即具有2個以上配位單元之化合物)。基於可更有效率地納入金屬陽離子之觀點，有機配位子較好為多齒有機配位子。作為多齒有機配位子舉例為環糊精、冠狀醚及其衍生物(例如苯并冠狀醚及二苯并冠狀醚)、杯芳烴及該等之衍生物等。其中，有機溶劑中較好使用冠狀醚及其衍生物，更好為冠狀醚衍生物。又，冠狀醚可通過氧上之非共用電子隊與陽離子溶劑合。 [0041] 作為冠狀醚舉例為例如下述通式(I)表示之冠狀醚。 [0042][0043] 式(I)中，n為1以上之整數。 [0044] 又，藉由上述通式(I)表示之冠狀醚納入金屬陽離子而形成之錯合物係以下述式(II)表示。 [0045][0046] 式(II)中，n為1以上之整數。Z為上述金屬陽離子。 [0047] 作為冠狀醚之具體例舉例為例如下述式(a)~(c)表示之冠狀醚。 [0048][0049] 上述式(a)為12-冠狀-4-醚。上述式(b)係15-冠狀-5-醚。上述式(c)係18-冠狀-6-醚。 [0050] 作為冠狀醚衍生物，舉例為例如具有1個以上芳基環者。 [0051] 冠狀醚衍生物與冠狀醚比較，隨著芳基環部位之擴張使電荷更非局部化。因此，使用具有1個以上芳基環之冠狀醚衍生物時，與使用不具有芳基環之冠狀醚時相比，顯示更安定之n型導電性並且顯示高的導電率及化學安定性。 [0052] 基於正電荷之非局部化之觀點，上述1個以上之芳基環較好形成縮合環。亦即，上述冠狀醚衍生物較好為具有縮合環之冠狀醚衍生物。作為上述縮合環，可為縮合2個、3個、4個、5個或其以上之芳基環者。作為上述芳基環或縮合環舉例為例如苯環、萘還、蒽環、四并苯環、五并苯環、六并苯環、七并苯環、八并苯環、菲環、芘環、䓛(chrysene)環、苯并芘環、聯三苯環或苯并呋喃環。 [0053] 作為具有苯環之冠狀醚衍生物舉例為例如下述式(d)表示之苯并-18-冠狀醚-6。 [0054][0055] 上述冠狀醚及其衍生物只要配合成為納入對象之金屬陽離子之尺寸選擇即可。 [0056] [3. n型導電材料之製造方法] 本發明一實施形態之n型導電材料之製造方法(以下亦稱為本製造方法)係包含將與金屬錯合物接觸之n型碳奈米管在真空條件下於100℃以上加熱之真空加熱步驟。依據本製造方法，金屬錯合物係內包於n行碳奈米管中。因此，可獲得具有優異電熱特性及化學安定性之n型導電材料。 [0057] 又，關於[2. n型導電材料]中已說明之事項，於以下說明中適當省略，並援用上述記載。 [0058] ＜3-1. 真空加熱步驟＞ 本步驟係將與金屬錯合物接觸之n型碳奈米管在真空條件下於100℃以上加熱之步驟。藉由進行本步驟，使金屬錯合物內包於n型碳奈米管。 [0059] 作為碳奈米管與金屬錯合物接觸之方法，舉例為例如使含有金屬錯合物之溶液與碳奈米管接觸之方法。 [0060] 只要可使碳奈米管與溶液接觸，則其方法並未特別限制。基於使碳奈米管與溶液充分接觸之觀點，較好藉由將碳奈米管含浸於溶液中，或藉由於溶液中使碳奈米管剪切分散，使碳奈米管與溶液接觸。 [0061] 作為於碳奈米管含浸溶液之方法舉例為將成形為如後述之期望形狀之碳奈米管(例如薄膜)浸漬於溶液中之方法。又，作為於溶液中使碳奈米管剪切分散之方法，舉例為使用均質化裝置將碳奈米管分散於溶液中之方法。 [0062] 上述溶液中之溶劑可為水亦可為有機溶劑。該溶劑較好為有機溶劑，更好為甲醇、乙醇、丙醇、丁醇、乙腈、N,N-二甲基甲醯胺、二甲基亞碸或N-甲基吡咯啶。作為丙醇舉例為1-丙醇及2-丙醇。作為丁醇舉例為1-丁醇及2-丁醇等。 [0063] 溶液中金屬錯合物之濃度可為任意濃度，例如較好為0.001~1mol/L，更好為0.01~0.1mol/L。 [0064] 作為上述均質化裝置只要為可於溶液中均質分散碳奈米管之裝置則未特別限制，但可使用例如均質機或超音波均質機等之習知手段。又本說明書中僅表示為「均質機」時意指「攪拌均質機」。 [0065] 作為均質化裝置之運轉條件只要為可於溶液中均質分散碳奈米管之條件則未特別限制。例如作為均質化裝置使用均質機時，將添加有碳奈米管之溶液，以均質機之攪拌速度(轉數)20000rpm於室溫(23℃)處理10分鐘，可將碳奈米管分散於溶液中。 [0066] 又，將已成形之碳奈米管浸漬於溶液中之方法時，浸漬時間並未特別限制，但較好為10~600分鐘，更好為100~600分鐘，又更好為200~600分鐘。 [0067] 又，本步驟前，亦可包含使碳奈米管n型化之步驟。n型化之方法並未特別限制，舉例為自電極將電子導入碳奈米管之方法及以特定陰離子作用於碳奈米管之方法。 [0068] 又，使碳奈米管n型化之步驟可與本步驟同時進行。該情況下，例如可將溶解有溶解於溶劑時產生與陰離子包接為金屬錯合物之金屬陽離子之金屬鹽及配位子之溶液與碳奈米管接觸，進行本步驟。基於有效率地形成金屬錯合物之觀點，上述溶液較好以其莫耳比為1:1之方式含有金屬陽離子與配位子。 [0069] 上述陰離子使碳奈米管之載子從電洞變化為電子。藉此，使碳奈米管之賽貝克係數產生變化同時使碳奈米管帶負電。 [0070] 作為陰離子之例舉例為羥基離子(OH-
)、烷氧基離子(CH3
O-
、CH3
CH2
O-
、i-PrO-
及t-BuO-
等)、硫離子(SH-
及CH3
S-
及C2
H5
S-
等之烷硫基離子等)、氰脲基離子(CN-
)、I-
、Br-
、Cl-
、BH4 -
、羧基離子(CH3
COO-
等)、NO3 -
、BF4 -
、ClO4 -
、TfO-
及Tos-
等。其中陰離子較好為選自由OH-
、CH3
O-
、CH3
CH3
O-
、i-PrO-
、t-BuO-
、SH-
、CH3
S-
、C2
H5
S-
、CN-
、I-
、Br-
、Cl-
、BH4 -
及CH3
COO-
所成之群中之至少一者，更好為OH-
及CH3
O-
之至少一者。若為上述陰離子，則可效率良好地使碳奈米管之賽貝克係數變化。 [0071] 陰離子作為使碳奈米管n型化之摻雜劑發揮作用之理由之一認為係具有非共用電子對。推測陰離子基於其非共用電子對，與成為摻雜對象之碳奈米管相互作用，或誘發化學反應。且，關於摻雜效率，認為摻雜劑之路易士鹼性、分子間力及解離性至為重要。 [0072] 本說明書中，所謂「路易士鹼性」意指供給對子對之性質。認為路易士鹼性較強之摻雜劑對於賽貝克係數之變化造成更大影響。 [0073] 又，認為分子間力亦與摻雜劑對碳奈米管之吸附性有關聯。作為摻雜劑之分子間力舉例有氫鍵、CH-π相互作用、π-π相互作用等。上述陰離子中，較好為賦予弱氫鍵之陰離子。作為賦予弱氫鍵之陰離子舉例為例如OH-
、CH3
O-
、CH3
CH2
O-
、i-PrO-
、t-BuO-
。且，陰離子較好為賦予π-π相互作用之陰離子。作為賦予π-π相互作用之陰離子舉例為CH3
COO-
。 [0074] 所謂真空條件係100Pa以下。較好為0.01Pa~50Pa，更好為0.1Pa~20Pa，又更好為1Pa~10Pa。 [0075] 進行上述真空加熱之溫度只要考慮金屬錯合物之昇華點或熔點適當決定即可，較好為100℃以上，更好為120℃~250℃，又更好為180℃~250℃。 [0076] 於真空條件下進行加熱之時間並未特別限定，較好為2小時以上，更好為3小時~72小時，又更好為10小時~20小時。 [0077] 壓力、溫度及時間之至少一者若為上述範圍，則由於可更有效率地將金屬錯合物內包於n型碳奈米管故而較佳。 [0078] ＜3-2. 成形步驟＞ 本製造方法於上述真空加熱步驟之前或之後亦可包含成形步驟。亦即，本步驟可為在上述真空加熱步驟之前將碳奈米管成形為期望形狀(例如薄膜)之步驟，亦可為將由上述真空加熱步驟所得之n型導電材料成形為期望形狀之步驟。 [0079] 較好，本製造方法於上述真空加熱步驟之前，包含將碳奈米管積集並成形薄膜之成形步驟。該情況下，上述真空加熱步驟中，較好將上述薄膜浸漬於上述溶液中。 [0080] 作為成形薄膜之方法並未特別限定，舉例為例如於溶劑中分散碳奈米管，所得分散液於過濾器上過濾而成形薄膜之方法。過濾舉例為使用膜過濾器之方法。具體而言，將碳奈米管之分散液使用0.1~2μm孔之膜過濾器進行抽吸過濾，將殘留於膜過濾器上之膜於50~150℃減壓乾燥1~24小時，而可成形薄膜。 [0081] 分散碳奈米管之溶劑可為水亦可為有機溶劑。該溶劑較好為有機溶劑，更好為鄰-二氯苯、溴苯、1-氯萘、2-氯萘或環己酮。若為該等溶劑，則可有效率地分散碳奈米管。 [0082] 作為分散碳奈米管之方法可使用與上述之＜3-1. 真空加熱步驟＞中使用均質化裝置將碳奈米管分散於溶液中之方法同樣之方法。 [0083] 本發明不限定於上述各實施形態，可於申請專利範圍所示範圍內進行各種變更，關於適當組合不同實施形態分別揭示之技術手段所得之實施形態亦包含於本發明之技術範圍。 [實施例] [0084] 以下，基於實施例及比較例更詳細說明本發明，但本發明並非限定於以下實施例者。 [0085] [實施例1] 將5mg之CNT(平均內徑2nm，名城奈米碳公司製，製品名：EC2.0)於10mL之鄰-二氯苯中，使用攪拌均質機(IKA公司製，ULTRA TURRAX)以20000rpm處理10分鐘。隨後，於膜過濾器(0.2mm孔徑，直徑25mm)上抽吸過濾，於120℃、減壓下乾燥12小時，藉此獲得CNT薄膜。 [0086] 所得CNT薄膜於在乙醇中溶解有0.01mol/L KOH(和光純藥工業公司製，試藥等級)及0.01mol/L之苯并-18-冠狀醚(SIGMA ALDRICH公司製)之溶液中浸漬4小時。 [0087] 隨後，自溶液中拉起CNT薄膜，於200℃進行3小時真空加熱，獲得實施例1之n型導電材料。 [0088] [比較例1] 除了自溶液中拉起CNT薄膜後，於80℃進行1小時真空加熱以外，與實施例1同樣，獲得比較例1之n型導電材料。 [0089] [實施例2] 除了使用18-冠狀醚(SIGMA ALDRICH公司製)代替苯并-18-冠狀醚以外，與實施例1同樣，獲得實施例2之n型導電材料。 [0090] [比較例2] 除了自溶液中拉起CNT薄膜後，於80℃進行1小時真空加熱以外，與實施例2同樣，獲得比較例2之n型導電材料。 [0091] [比較例3] 與實施例1同樣獲得之CNT薄膜未浸漬於溶液中而作為比較例3。 [0092] [透過型電子顯微鏡之觀察] 實施例1及比較例1之n型導電材料以透過型電子顯微鏡(JEOL公司製，製品名：JEM-3100FEF)進行觀察。 [0093] 圖2之(a)係顯示比較例1之碳奈米管之透過型電子顯微鏡像。圖2之(b)係顯示實施例1之碳奈米管之透過型電子顯微鏡像。可知相對於圖2之(a)之金屬錯合物位於碳奈米管外部，圖2之(b)係金屬錯合物位於碳奈米管內部。 [0094] 又，圖2之(c)及(d)分別為藉由元素分析觀察實施例1之碳奈米管內部有無碳及鉀之結果的圖。元素分析係使用電子能量損失分光法。由圖2之(c)及(d)可知碳奈米管內部存在碳及鉀。因此，由圖2之(c)及(d)亦可知實施例1中金屬錯合物內包於碳奈米管。 [0095] [熱電特性] 針對實施例及比較例所得之n型導電材料，使用熱電特性評價裝置(ADVANCE理工股份有限公司製，製品名：ZEM-3)，測定37℃~200℃之導電率σ及賽貝克係數α。導電率係藉由四端子法測定，賽貝克係數係藉由二端子法測定。且，使用所得導電率及賽貝克係數，由上述式(1)算出輸出因子PF。 [0096] 圖3之(a)~(c)分別顯示37℃~200℃之實施例1及比較例1之導電率之測定值、賽貝克係數之測定值及輸出因子之算出值的圖。橫軸表示測定溫度。圖3之(a)~(c)中，實施例1之測定值以黑圓圈表示，比較例1之測定值以白圓圈表示。 [0097] 由圖3之(a)可知，37℃~200℃之比較例1之n型導電材料之導電率依存於溫度而變化，另一方面，實施例1之n型導電材料並未依存於溫度顯示安定之導電率。 [0098] 由圖3之(b)可知，實施例1於任一溫度下，賽貝克係數之絕對值均大於比較例1，故於任一溫度下，熱起電力均大。 [0099] 由圖3之(c)可知，實施例1之n型導電材料與比較例1之n型導電材料比較，於37℃~200℃之任一溫度下，均顯示高的輸出因子。由此，可知藉由於碳奈米管內包金屬錯合物，可增大輸出因子。 [0100] 圖4之(a)~(c)分別顯示37℃~200℃之實施例2及比較例2之導電率之測定值、賽貝克係數之測定值及輸出因子之算出值的圖。橫軸表示測定溫度。圖4之(a)~(c)中，實施例2之測定值以黑圓圈表示，比較例2之測定值以白圓圈表示。 [0101] 由圖4之(a)可知，37℃~200℃之比較例2之n型導電材料之導電率依存於溫度而變化，另一方面，實施例2之n型導電材料並未依存於溫度顯示安定之導電率。 [0102] 由圖4之(b)可知，實施例2於任一溫度下，賽貝克係數之絕對值均大於比較例2，故於任一溫度下，熱起電力均大。 [0103] 由圖4之(c)可知，實施例2之n型導電材料與比較例2之n型導電材料比較，於37℃~200℃之任一溫度下，均顯示高的輸出因子。 [0104] 由此，由實施例2及比較例2之比較，可知藉由於碳奈米管內包金屬錯合物，可增大輸出因子。且若比較圖3之(c)及圖4之(c)，則可知配位子具有苯環者，顯示更安定之n型導電性並且顯示高的輸出因子。 [0105] 又，針對實施例1、比較例1及比較例3所得之CNT薄膜，測定熱擴散率、定壓比熱、密度及熱傳導率。作為熱擴散率係使用快速分析儀(NETCH公司製，製品名：LFA 467 HyperFlash)，測定面內方向之熱擴散率。定壓比熱係使用示差掃描熱量計(SII NanoTechnology公司製，製品名：DSC6200)測定。熱傳導率係由所得之熱擴散率、定壓比熱及密度之乘積算出。 [0106] 測定結果示於表1。 [0107] [表1]
[0108] 如表1所示，關於實施例1、比較例1及比較例3，熱傳導率無太大差異。 [0109] 又，針對實施例1及比較例1之ZT進行檢討。實施例1及比較例1之熱傳導率均為38W/m．K。因此，如上述之實施例1輸出因子相較於比較例1更增大。因此，可知實施例1中以輸出因子之增大量使ZT增大。 [0110] [化學安定性] 針對實施例1及比較例1所得之n型導電材料，除了未洗淨(0秒後)，以99%乙醇溶液中洗淨100秒、200秒、300秒、400秒、500秒、600秒、700秒、800秒、900秒及1000秒，於室溫乾燥1小時後，測定37℃下之賽貝克係數。賽貝克係數係使用賽貝克效應測定裝置(NMR technologies公司製，SB-200)測定。 [0111] 圖5顯示隨著乙醇洗淨，實施例1及比較例1於37℃之賽貝克係數變化之圖。橫軸表示乙醇洗淨之處理時間。圖5中實施例1之測定以黑圓圈表示，比較例1之測定值以白圓圈表示。 [0112] 比較例1之n型導電材料於600~700秒之乙醇洗淨期間，賽貝克係數自負值變化為正值。由此可知比較例1因乙醇洗淨而喪失n型導電性。比較例1由於金屬錯合物並未內包於碳奈米管，故認為藉由乙醇洗淨，吸附於碳奈米管表面之金屬錯合物脫離。 [0113] 另一方面，實施例1即使以乙醇洗淨1000秒，賽貝克係數仍顯示負值。由此可知乙醇洗淨後亦安定而具有n型導電性。實施例1由於金屬錯合物內包於碳奈米管，故認為即使藉由乙醇洗淨，金屬錯合物亦不脫離而內包於碳奈米管。亦即，認為藉由將金屬錯合物內包於碳奈米管，可改善化學安定性。 [產業上之可利用性] [0114] 本發明可利用於熱電發電系統、醫療用電源、保全用電源、航空宇宙用途等之各種廣泛產業。[0016] Hereinafter, an example of the aspect of the present invention will be described in detail, but the present invention is not limited thereto. In addition, as long as it is not specifically mentioned in this specification, "A to B" which shows a numerical range means "A to B". [0017] [1. Indexes Related to Performance of n-Type Conductive Material] First, indexes related to performance of n-type conductive material will be described. An example of this index is an output factor (power factor). The output factor is obtained by the following formula (1). [0018] PF = α 2 σ (1) In the formula (1), PF represents an output factor, α represents a Seebeck coefficient, and σ represents a conductivity. In the n-type conductive material, for example, the output factor at 310K is preferably 100 μW / mK 2 or more, more preferably 200 μW / mK 2 or more, and particularly preferably 400 μW / mK 2 or more. If the output factor is at least 100 μW / mK 2 at 310K, it is preferable that the output factor is equal to or higher than that of a conventional p-type conductive material. In order to obtain such a high-output n-type conductive material, it is considered to improve either the Seebeck coefficient or the conductivity, and obtain both. [0019] The Seebeck coefficient refers to the ratio of the open circuit voltage to the temperature difference between the high-temperature junction and the low-temperature junction of a circuit showing the Seebeck effect. "). The Seebeck coefficient can be measured using, for example, a Seebeck effect measuring device (manufactured by MMR Technologies) used in the examples described later. The larger the absolute value of the Seebeck coefficient, the greater the thermal starting power. [0020] Moreover, the Seebeck coefficient can be used as an index for judging the polarity of electronic materials such as carbon nanotubes. Specifically, for example, an electronic material showing a positive Seebeck coefficient can be said to have p-type conductivity. On the other hand, electronic materials with negative Seebeck coefficients can be said to have n-type conductivity. [0021] In the n-type conductive material, the Seebeck coefficient is preferably -20 μV / K or less, more preferably -30 μV / K or less, and even more preferably -40 μV / K or less. However, when power is generated using a small amount of energy such as a low-temperature heat source, the electrical conductivity also increases with the increase of the thermal starting power, and it is necessary to suppress the impedance required for the booster circuit. In this case, the Seebeck coefficient is more preferably -40 to -20 μV / K. [0022] The electrical conductivity can be measured, for example, by a 4-probe method using a resistivity meter (manufactured by Mitsubishi Chemical Analysis Co., Ltd., LORESTA GP). [0023] In the n-type conductive material, the conductivity is preferably 1,000 S / cm or more, more preferably 1500 s / cm or more, and even more preferably 2000 S / cm or more. When the conductivity is 1000 S / cm or more, it is preferable because the n-type conductive material has a high output. [0024] Another index related to the performance of the n-type conductive material is, for example, the dimensionless performance index ZT. ZT is obtained by the following formula (2). [0025] ZT = PF. T / κ (2) In formula (2), PF is the output factor (= α 2 σ), T is the temperature, and κ is the thermal conductivity. A larger ZT indicates a more excellent n-type conductive material. It can be known from formula (2) that in order to obtain a larger ZT, the output factor, ie, the absolute value of the Seebeck coefficient and the conductivity are relatively large and good. [0026] It is also known from the formula (2) that in order to obtain a larger ZT, the smaller the thermal conductivity, the better. This system corresponds to a thermoelectric conversion material (that is, an n-type conductive material and a p-type conductive material) that uses a temperature difference. When the thermal conductivity of the thermoelectric conversion material is large, the temperature in the material will be easily uniformed, and a temperature difference will not be easily generated. Therefore, a thermoelectric conversion device provided with a thermoelectric conversion material having a large thermal conductivity tends to be difficult to efficiently generate electricity. [2. n-type conductive material] An n-type conductive material according to an embodiment of the present invention (hereinafter also referred to as the n-type conductive material) is characterized by an n-type carbon nanotube including a metal complex . In the present specification, the “inclusion metal complex” means that the metal complex is located inside the cavity of the carbon nanotube. In this n-type conductive material, whether or not a metal complex is contained in a carbon nanotube can be investigated by observation with a transmission electron microscope as shown in the examples described later. [0028] Moreover, the metal complex may be entirely enclosed in the carbon nanotube, or only a part thereof may be enclosed in the carbon nanotube. [0029] The n-type carbon nanotube is in a state where the negative charge is not localized and becomes a soft base. Metal complexes, on the other hand, become soft acids with non-localized positive charges. It is stabilized by the action of a soft acid to a soft base. Therefore, the present n-type conductive material exhibits stable n-type conductivity through the action of a metal complex on the n-type carbon nanotube. The definitions of soft acids and bases are based on the HSAB theory (RG Pearson, J. Am. Chem. Soc. 85 (22), 3533-3539, 1963). [0030] In addition, the n-type conductive material has excellent thermoelectric characteristics and chemical stability because the metal complex is enclosed in the n-type carbon nanotube. This is presumably because when the metal complex is contained in an n-type carbon nanotube, the metal complex is harder to detach than when the metal complex is only attached to the n-type carbon nanotube. [0031] The n-type conductive material may contain substances other than the n-type nano tube and the metal complex as required. These substances are not particularly limited as long as they do not hinder the aforementioned effects of the metal complex. [0032] <2-1. N-type carbon nano tube> The n-type conductive material contains an n-type nano tube. As long as the n-type nano tube has n-type conductivity, it may be an n-type by any method. In this specification, a carbon nanotube may be referred to as "CNT". [0033] The carbon nanotube may have a single-layer or multi-layer structure (two, three, four, or more layers). For example, the carbon nanotube may be a single-wall carbon nanotube (SWNT) or a multi-wall carbon nanotube (MWNT). [0034] The present n-type conductive material is considered to have various applications and uses as a thermoelectric conversion device and the like. Here, if the thermoelectric conversion device has flexibility, it can adhere to the complicated three-dimensional surface of the human body or piping, and can be effectively used for body temperature or waste heat. In order to increase the flexibility of the thermoelectric conversion device, in the n-type conductive material, from the viewpoint of imparting excellent mechanical properties (tensile strength, Young's modulus, elastic modulus, etc.), the carbon nanotube is preferably a single-layer carbon nanotube Meter tube. [0035] The average inner diameter of the n-type carbon nanotube is not particularly limited. For example, it is preferably 0.6 to 1000 nm, more preferably 0.6 to 100 nm, still more preferably 0.8 to 20 nm, particularly preferably 1 to 5 nm. It is preferably 1.2 ~ 3nm. If the average inner diameter is 0.6 nm or more, a metal complex can be sufficiently incorporated into the n-type carbon nanotube. In addition, if the average inner diameter is 1.2 nm or more, it is preferable because the metal complex is more likely to enter the inside of the n-type carbon nanotube. In addition, if the average inner diameter is 1000 nm or less, it is preferable that the metal complex is prevented from diffusing inside the n-type carbon nanotube, and chemical stability at normal pressure and temperature can be obtained. In addition, if the average inner diameter is 3 nm or less, it is more preferable that the metal complex is difficult to escape from the inside of the n-type carbon nanotube, or excellent thermoelectric conversion characteristics can be obtained in a certain situation. [0036] The average inner diameter of the n-type carbon nanotube can be measured by observing the n-type carbon nanotube with a transmission electron microscope or a scanning electron microscope, for example. For example, any five points of the n-type conductive material are observed with a transmission electron microscope or a scanning electron microscope. Ten n-type carbon nanotubes are randomly selected at each place, and the inner diameters are measured. The average value of the measured inner diameters of the 50 n-type carbon nanotubes is set to the average inner diameter of the n-type carbon nanotubes. [0037] The above carbon nanotube can be formed into a desired shape. For example, the n-type conductive material may include a thin film formed by collecting carbon nanotubes. Here, the "thin film" may be renamed a sheet or a film. The film may have a thickness of, for example, 1 μm to 1000 μm. The density of the carbon nanotube of the thin film is not particularly limited, but may be 0.05 to 1.0 g / cm 3 , or 0.1 to 0.5 g / cm 3 . The thin-film carbon nanotubes are entangled with each other to form a non-woven structure. Therefore, the film is lightweight and flexible. [0038] <2-2. Metal Complex> The n-type conductive material contains a metal complex. The metal complex in this specification means a metal cation and a ligand coordinately bonded. [0039] Examples of the metal cation include typical metal ions (alkali metal ions and alkaline earth metal ions) and transition metal ions. Examples of the metal ion include lithium ion, sodium ion, potassium ion, rubidium ion, cesium ion, rubidium ion, beryllium ion, magnesium ion, calcium ion, strontium ion, barium ion, radium ion, rubidium ion, and the like. [0040] The ligand is not particularly limited as long as it is a compound that forms a coordination bond with a metal cation. In other words, the above-mentioned ligand is a compound having a coordination unit for a metal cation. Examples of the above-mentioned ligands include organic ligands and the like. The organic ligand may be a monodentate ligand (that is, a compound having one coordination unit), or a multidentate organic ligand (that is, a compound having two or more coordination units). From the viewpoint that metal cations can be more efficiently incorporated, the organic ligand is preferably a multidentate organic ligand. Examples of polydentate organic ligands include cyclodextrin, crown ethers and derivatives thereof (for example, benzocrown ether and dibenzocrown ether), calixarene, and derivatives thereof. Among them, crown ethers and their derivatives are preferably used in organic solvents, and crown ether derivatives are more preferred. In addition, the crown ether can be combined with a cationic solvent through a non-shared electron group on oxygen. [0041] Examples of the crown ether include a crown ether represented by the following general formula (I). [0042] [0043] In the formula (I), n is an integer of 1 or more. [0044] In addition, the complex formed by incorporating the crown ether represented by the general formula (I) into a metal cation is represented by the following formula (II). [0045] [0046] In Formula (II), n is an integer of 1 or more. Z is the aforementioned metal cation. [0047] Specific examples of the crown ether include, for example, a crown ether represented by the following formulae (a) to (c). [0048] [0049] The above formula (a) is 12-crown-4-ether. The formula (b) is a 15-crown-5-ether. The formula (c) is an 18-crown-6-ether. [0050] Examples of the crown ether derivative include those having one or more aryl rings. [0051] Compared with the crown ether derivative, the charge becomes more non-localized as the aryl ring site expands. Therefore, when a crown ether derivative having one or more aryl rings is used, compared with a crown ether having no aryl rings, it exhibits more stable n-type conductivity and exhibits higher conductivity and chemical stability. [0052] From the viewpoint of non-localization of the positive charge, it is preferred that the one or more aryl rings described above form a condensed ring. That is, the above-mentioned crown ether derivative is preferably a crown ether derivative having a condensed ring. As the above-mentioned condensed ring, two, three, four, five or more aryl rings may be condensed. Examples of the aryl ring or the condensed ring include, for example, a benzene ring, a naphthalene ring, an anthracene ring, a tetraacene ring, a pentacene ring, a hexaacene ring, a heptacene ring, an octaphenyl ring, a phenanthrene ring, and a pyrene ring , Chrysene ring, benzofluorene ring, bitriphenyl ring or benzofuran ring. [0053] Examples of the crown ether derivative having a benzene ring include benzo-18-crown ether-6 represented by the following formula (d). [0054] [0055] The crown ether and its derivative may be selected by blending the size of the metal cation to be incorporated. [3. Manufacturing method of n-type conductive material] A manufacturing method of an n-type conductive material according to an embodiment of the present invention (hereinafter also referred to as the manufacturing method) includes an n-type carbon nano which will be brought into contact with a metal complex. The vacuum heating step of heating the rice tube under the vacuum condition above 100 ° C. According to this manufacturing method, the metal complex is contained in n rows of carbon nanotubes. Therefore, an n-type conductive material having excellent electrothermal characteristics and chemical stability can be obtained. [0057] In addition, the matters described in [2. n-type conductive material] are appropriately omitted in the following description, and the above description is referred to. [0058] <3-1. Vacuum heating step> This step is a step of heating an n-type carbon nanotube in contact with a metal complex under a vacuum condition at 100 ° C or higher. By performing this step, the metal complex is encapsulated in the n-type carbon nanotube. [0059] As a method of contacting the carbon nanotube with a metal complex, for example, a method of contacting a solution containing a metal complex with a carbon nanotube. [0060] As long as the carbon nanotube can be brought into contact with the solution, the method is not particularly limited. From the viewpoint of bringing the carbon nanotubes into full contact with the solution, it is preferable to contact the carbon nanotubes with the solution by immersing the carbon nanotubes in the solution or by shear-dispersing the carbon nanotubes in the solution. [0061] An example of a method for impregnating a solution with a carbon nanotube is a method of immersing a carbon nanotube (for example, a film) formed into a desired shape as described later in the solution. In addition, as a method of shearing and dispersing carbon nanotubes in a solution, a method of dispersing carbon nanotubes in a solution using a homogenizing device is exemplified. [0062] The solvent in the solution may be water or an organic solvent. The solvent is preferably an organic solvent, more preferably methanol, ethanol, propanol, butanol, acetonitrile, N, N-dimethylformamide, dimethylsulfinium, or N-methylpyrrolidine. Examples of propanol include 1-propanol and 2-propanol. Examples of butanol include 1-butanol and 2-butanol. [0063] The concentration of the metal complex in the solution may be any concentration, for example, it is preferably 0.001 to 1 mol / L, and more preferably 0.01 to 0.1 mol / L. [0064] The homogenizing device is not particularly limited as long as it is a device capable of homogeneously dispersing carbon nanotubes in a solution, but conventional means such as a homogenizer or an ultrasonic homogenizer can be used. In the present specification, the expression "homogeneous machine" means "agitating homogenizer". [0065] The operating conditions of the homogenizing device are not particularly limited as long as the conditions can homogeneously disperse the carbon nanotubes in the solution. For example, when using a homogenizer as a homogenizing device, the carbon nano tube can be dispersed in a solution with a carbon nano tube added at a stirring speed (revolution) of 20,000 rpm at room temperature (23 ° C) for 10 minutes. In solution. [0066] In the method of immersing a formed carbon nanotube in a solution, the immersion time is not particularly limited, but it is preferably 10 to 600 minutes, more preferably 100 to 600 minutes, and even more preferably 200. ~ 600 minutes. [0067] Before this step, a step of n-type carbon nanotubes may be included. The method of n-type formation is not particularly limited, and examples thereof include a method of introducing electrons into a carbon nanotube from an electrode and a method of acting on a carbon nanotube with a specific anion. [0068] The step of n-type carbon nanotubes can be performed simultaneously with this step. In this case, for example, the carbon nano tube can be contacted with a solution in which a metal salt and a ligand that generate a metal cation that is anion-encapsulated with a metal complex when dissolved in a solvent is dissolved in a carbon nanotube, and this step is performed. From the viewpoint of efficiently forming a metal complex, the solution preferably contains a metal cation and a ligand such that the molar ratio thereof is 1: 1. [0069] The anion changes the carrier of the carbon nanotube from an electric hole to an electron. As a result, the Seebeck coefficient of the carbon nanotube is changed and the carbon nanotube is negatively charged. [0070] The anion is exemplified as Example hydroxyl ions (OH -), alkoxy ion (CH 3 O -, CH 3 CH 2 O -, i-PrO - and t-BuO -, etc.), sulfide (SH - and CH 3 S - and C 2 H 5 S - ion alkylthio, etc.), ureido cyanide ion (CN -), I -, Br -, Cl -, BH 4 -, carboxyl ion (CH 3 COO - etc.), NO 3 -, BF 4 -, ClO 4 -, TfO - and Tos - and the like. Wherein the anion is preferably selected from the group consisting of OH -, CH 3 O -, CH 3 CH 3 O -, i-PrO -, t-BuO -, SH -, CH 3 S -, C 2 H 5 S -, CN -, I -, Br -, Cl - , BH 4 - and CH 3 COO - to the group of the at least one, more preferably OH - and CH 3 O - is at least one. If it is the said anion, the Seebeck coefficient of a carbon nanotube can be changed efficiently. [0071] One of the reasons why the anion functions as a dopant for n-type carbon nanotubes is considered to have a non-shared electron pair. It is presumed that the anion is based on its non-shared electron pair, and interacts with the carbon nanotube that becomes the doping object, or induces a chemical reaction. In addition, regarding the doping efficiency, the Lewis basicity, intermolecular force, and dissociation of the dopant are considered to be extremely important. [0072] In the present specification, the "Lewis basicity" means a property of a pair of pairs. It is believed that the more basic dopants of Louis have more influence on the change of Seebeck coefficient. [0073] It is also believed that the intermolecular force is also related to the dopant's adsorption to carbon nanotubes. Examples of intermolecular forces of dopants include hydrogen bonding, CH-π interaction, π-π interaction, and the like. Among the above-mentioned anions, anions that impart weak hydrogen bonds are preferred. As the example given an anion of weak hydrogen bonds, for example, OH -, CH 3 O -, CH 3 CH 2 O -, i-PrO -, t-BuO -. The anion is preferably an anion that imparts a π-π interaction. As imparting the anionic π-π interaction is exemplified by CH 3 COO -. [0074] The vacuum condition is 100 Pa or less. It is preferably 0.01 Pa to 50 Pa, more preferably 0.1 Pa to 20 Pa, and even more preferably 1 Pa to 10 Pa. [0075] The temperature for performing the above-mentioned vacuum heating may be appropriately determined in consideration of the sublimation point or melting point of the metal complex, preferably 100 ° C or higher, more preferably 120 ° C to 250 ° C, and still more preferably 180 ° C to 250 ° C. . [0076] The time for heating under vacuum is not particularly limited, but it is preferably 2 hours or more, more preferably 3 hours to 72 hours, and still more preferably 10 hours to 20 hours. [0077] If at least one of the pressure, temperature, and time is within the above-mentioned range, it is preferable because the metal complex can be more efficiently enclosed in the n-type carbon nanotube. [0078] <3-2. Forming Step> This manufacturing method may include a forming step before or after the above-mentioned vacuum heating step. That is, this step may be a step of forming the carbon nanotube into a desired shape (for example, a thin film) before the vacuum heating step, or a step of forming the n-type conductive material obtained in the vacuum heating step into a desired shape. [0079] Preferably, the manufacturing method includes a forming step of collecting carbon nanotubes and forming a thin film before the vacuum heating step. In this case, in the vacuum heating step, the film is preferably immersed in the solution. [0080] The method of forming the film is not particularly limited, and examples include a method of dispersing carbon nanotubes in a solvent, and filtering the obtained dispersion on a filter to form a film. An example of filtration is a method using a membrane filter. Specifically, the carbon nano tube dispersion is suction-filtered using a membrane filter with a hole of 0.1 to 2 μm, and the membrane remaining on the membrane filter is dried under reduced pressure at 50 to 150 ° C. for 1 to 24 hours. Shaped film. [0081] The solvent of the dispersing carbon nanotube may be water or an organic solvent. The solvent is preferably an organic solvent, more preferably o-dichlorobenzene, bromobenzene, 1-chloronaphthalene, 2-chloronaphthalene, or cyclohexanone. With these solvents, carbon nanotubes can be efficiently dispersed. [0082] As the method of dispersing the carbon nanotubes, the same method as the method of dispersing the carbon nanotubes in a solution using a homogenizing device in the above-mentioned <3-1. Vacuum heating step> can be used. [0083] The present invention is not limited to the above-mentioned embodiments, and various changes can be made within the scope shown in the scope of the patent application. Embodiments obtained by appropriately combining the technical means disclosed by different embodiments are also included in the technical scope of the present invention. [Examples] Hereinafter, the present invention will be described in more detail based on examples and comparative examples, but the present invention is not limited to the following examples. [Example 1] 5 mg of CNT (average inner diameter: 2 nm, manufactured by Mingcheng Nano Carbon Co., Ltd., product name: EC2.0) was placed in 10 mL of o-dichlorobenzene using a stirring homogenizer (manufactured by IKA Corporation) , ULTRA TURRAX) for 10 minutes at 20000 rpm. Subsequently, the membrane filter (0.2 mm pore diameter, 25 mm diameter) was suction-filtered, and dried at 120 ° C. under reduced pressure for 12 hours, thereby obtaining a CNT film. [0086] The obtained CNT film was dissolved in ethanol with a solution of 0.01 mol / L KOH (manufactured by Wako Pure Chemical Industries, Ltd.) and 0.01 mol / L of benzo-18-crown ether (manufactured by SIGMA ALDRICH). Medium immersion for 4 hours. [0087] Subsequently, the CNT film was pulled up from the solution and vacuum heated at 200 ° C for 3 hours to obtain the n-type conductive material of Example 1. [Comparative Example 1] An n-type conductive material of Comparative Example 1 was obtained in the same manner as in Example 1 except that the CNT film was pulled up from the solution and then subjected to vacuum heating at 80 ° C for 1 hour. [Example 2] An n-type conductive material of Example 2 was obtained in the same manner as in Example 1 except that 18-crown ether (manufactured by SIGMA ALDRICH) was used instead of benzo-18-crown ether. [Comparative Example 2] An n-type conductive material of Comparative Example 2 was obtained in the same manner as in Example 2 except that the CNT film was pulled up from the solution and then subjected to vacuum heating at 80 ° C for 1 hour. [Comparative Example 3] A CNT film obtained in the same manner as in Example 1 was used as Comparative Example 3 without being immersed in the solution. [Observation by Transmission Electron Microscope] The n-type conductive materials of Example 1 and Comparative Example 1 were observed with a transmission electron microscope (manufactured by JEOL, product name: JEM-3100FEF). 2 (a) is a transmission electron microscope image of a carbon nanotube of Comparative Example 1. FIG. FIG. 2 (b) is a transmission electron microscope image of the carbon nanotube of Example 1. FIG. It can be seen that the metal complex in FIG. 2 (a) is located outside the carbon nanotube, and the metal complex in FIG. 2 (b) is located inside the carbon nanotube. 2 (c) and (d) are the results of observing the presence or absence of carbon and potassium in the carbon nanotube of Example 1 by elemental analysis, respectively. Elemental analysis uses electron energy loss spectroscopy. It can be seen from (c) and (d) of FIG. 2 that carbon and potassium exist in the carbon nanotube. Therefore, it can also be seen from (c) and (d) of FIG. 2 that the metal complex in Example 1 is enclosed in a carbon nanotube. [Pyroelectric characteristics] For the n-type conductive materials obtained in the examples and comparative examples, a thermoelectric characteristic evaluation device (manufactured by ADVANCE Technology Co., Ltd., product name: ZEM-3) was used to measure the conductivity from 37 ° C to 200 ° C σ and Seebeck coefficient α. Electrical conductivity is measured by the four-terminal method, and Seebeck coefficient is measured by the two-terminal method. Then, using the obtained conductivity and Seebeck coefficient, the output factor PF is calculated from the above formula (1). 3 (a) to (c) are diagrams showing measured values of electrical conductivity, measured values of Seebeck coefficient, and calculated values of output factors of Example 1 and Comparative Example 1, respectively, at 37 ° C to 200 ° C. The horizontal axis indicates the measurement temperature. In (a) to (c) of FIG. 3, the measurement value of Example 1 is shown by a black circle, and the measurement value of Comparative Example 1 is shown by a white circle. [0097] As can be seen from FIG. 3 (a), the conductivity of the n-type conductive material of Comparative Example 1 at 37 ° C. to 200 ° C. varies depending on the temperature. On the other hand, the n-type conductive material of Example 1 does not depend on it. Shows stable conductivity at temperature. [0098] As can be seen from FIG. 3 (b), the absolute value of the Seebeck coefficient of Example 1 is greater than that of Comparative Example 1 at any temperature, so the thermal starting power is large at any temperature. [0099] As can be seen from FIG. 3 (c), the n-type conductive material of Example 1 and the n-type conductive material of Comparative Example 1 show a high output factor at any temperature between 37 ° C and 200 ° C. From this, it can be seen that the output factor can be increased by the inclusion of metal complexes in carbon nanotubes. 4 (a) to (c) are diagrams showing measured values of electrical conductivity, measured values of Seebeck coefficient, and calculated values of output factors of Example 2 and Comparative Example 2 at 37 ° C to 200 ° C, respectively. The horizontal axis indicates the measurement temperature. In (a) to (c) of FIG. 4, the measurement value of Example 2 is indicated by a black circle, and the measurement value of Comparative Example 2 is indicated by a white circle. [0101] It can be seen from FIG. 4A that the conductivity of the n-type conductive material of Comparative Example 2 at 37 ° C. to 200 ° C. varies depending on the temperature. On the other hand, the n-type conductive material of Example 2 does not depend on it. Shows stable conductivity at temperature. [0102] It can be seen from FIG. 4 (b) that the absolute value of the Seebeck coefficient of Example 2 is greater than that of Comparative Example 2 at any temperature, so the thermal starting power is large at any temperature. [0103] As can be seen from FIG. 4 (c), the n-type conductive material of Example 2 and the n-type conductive material of Comparative Example 2 showed a high output factor at any temperature between 37 ° C and 200 ° C. [0104] Thus, from the comparison between Example 2 and Comparative Example 2, it can be seen that the output factor can be increased by the inclusion of the metal complex in the carbon nano tube. Moreover, if (c) of FIG. 3 and (c) of FIG. 4 are compared, it can be seen that the ligand having a benzene ring exhibits a more stable n-type conductivity and a high output factor. [0105] Regarding the CNT thin films obtained in Example 1, Comparative Example 1, and Comparative Example 3, the thermal diffusivity, constant pressure specific heat, density, and thermal conductivity were measured. As the thermal diffusivity, a rapid analyzer (manufactured by NETCH, product name: LFA 467 HyperFlash) was used, and the thermal diffusivity in the in-plane direction was measured. Constant pressure specific heat was measured using a differential scanning calorimeter (manufactured by SII NanoTechnology, product name: DSC6200). Thermal conductivity is calculated from the product of the obtained thermal diffusivity, specific heat at constant pressure, and density. [0106] The measurement results are shown in Table 1. [Table 1] [0108] As shown in Table 1, regarding Example 1, Comparative Example 1, and Comparative Example 3, there was not much difference in thermal conductivity. [0109] The ZT of Example 1 and Comparative Example 1 were reviewed. Both the thermal conductivity of Example 1 and Comparative Example 1 were 38 W / m. K. Therefore, the output factor of Example 1 as described above is larger than that of Comparative Example 1. Therefore, it can be seen that ZT is increased by the increase amount of the output factor in the first embodiment. [Chemical Stability] The n-type conductive materials obtained in Example 1 and Comparative Example 1 were cleaned in a 99% ethanol solution for 100 seconds, 200 seconds, 300 seconds, except that they were not washed (after 0 seconds). 400 seconds, 500 seconds, 600 seconds, 700 seconds, 800 seconds, 900 seconds, and 1000 seconds, and after drying at room temperature for 1 hour, the Seebeck coefficient at 37 ° C was measured. The Seebeck coefficient is measured using a Seebeck effect measuring device (manufactured by NMR technologies, SB-200). 5 is a graph showing changes in the Seebeck coefficient of Example 1 and Comparative Example 1 at 37 ° C. with ethanol washing. The horizontal axis represents the processing time for ethanol washing. The measurement of Example 1 in FIG. 5 is indicated by a black circle, and the measurement value of Comparative Example 1 is indicated by a white circle. [0112] The n-type conductive material of Comparative Example 1 changed from a negative value to a positive value during the ethanol washing period of 600 to 700 seconds. From this, it can be seen that Comparative Example 1 lost n-type conductivity by washing with ethanol. In Comparative Example 1, since the metal complex was not contained in the carbon nanotube, the metal complex adsorbed on the surface of the carbon nanotube was removed by washing with ethanol. [0113] On the other hand, in Example 1, the Seebeck coefficient showed a negative value even after being washed with ethanol for 1,000 seconds. From this, it can be seen that after washing with ethanol, it is stable and has n-type conductivity. In Example 1, since the metal complex was contained in the carbon nanotube, it was considered that the metal complex was contained in the carbon nanotube without detaching it even after washing with ethanol. That is, it is thought that chemical stability can be improved by inclusion of a metal complex in a carbon nanotube. [Industrial availability] [0114] The present invention is applicable to a wide variety of industries such as thermoelectric power generation systems, medical power supplies, security power supplies, and aerospace applications.