亞穩(wěn)相圖研究及其在特種陶瓷涂層中的應(yīng)用進(jìn)展

黃燁琰, 徐凱, 吳波, 李朋, 常可可, 黃峰, 黃慶

亞穩(wěn)相圖研究及其在特種陶瓷涂層中的應(yīng)用進(jìn)展

黃燁琰1,2, 徐凱1, 吳波2, 李朋1, 常可可1, 黃峰1, 黃慶1

(1. 中國(guó)科學(xué)院 寧波材料技術(shù)與工程研究所, 先進(jìn)能源材料工程實(shí)驗(yàn)室, 寧波 315201; 2. 福州大學(xué) 材料科學(xué)與工程學(xué)院, 多尺度材料計(jì)算實(shí)驗(yàn)室, 福州 350100)

相圖, 又稱相平衡圖, 是“材料設(shè)計(jì)的索驥圖”, 而涂層的制備過(guò)程中(如物理氣相沉積, Physical Vapor Deposition, 簡(jiǎn)稱PVD), 系統(tǒng)一般遠(yuǎn)離平衡態(tài), 獲得的相為亞穩(wěn)相, 相圖計(jì)算CALPHAD (CALculation of PHAse Diagrams)方法的應(yīng)用遇到了挑戰(zhàn)。本文概述了模擬涂層材料亞穩(wěn)相圖的研究歷程, 重點(diǎn)介紹了近期建立的臨界表面擴(kuò)散亞穩(wěn)相圖模型, 即通過(guò)耦合CALPHAD、第一性原理計(jì)算和高通量磁控濺射鍍膜實(shí)驗(yàn)的方法對(duì)涂層材料的亞穩(wěn)相進(jìn)行表面擴(kuò)散模擬, 相關(guān)計(jì)算僅需要一個(gè)高通量鍍膜實(shí)驗(yàn)作為基礎(chǔ)數(shù)據(jù), 獲得的亞穩(wěn)相圖也得到了實(shí)驗(yàn)驗(yàn)證。由此, 可以建立相關(guān)材料體系的穩(wěn)定和亞穩(wěn)相圖數(shù)據(jù)庫(kù), 通過(guò)組分–制備工藝–組織結(jié)構(gòu)和性能的關(guān)系, 指導(dǎo)陶瓷涂層材料的設(shè)計(jì), 助推研發(fā)時(shí)間和成本“雙減半”目標(biāo)的實(shí)現(xiàn)。

亞穩(wěn)相圖; 模型; 表面擴(kuò)散; 應(yīng)用; 綜述

涂層材料廣泛應(yīng)用于機(jī)械制造、生物醫(yī)學(xué)、光電器件及核電工程等領(lǐng)域。例如, TiAlN涂層由于具有高硬度、低摩擦系數(shù)等性能而被用于切削刀具領(lǐng)域[1-6]; 非晶合金涂層由于具有良好的生物兼容性及優(yōu)越的機(jī)械性能而被用于生物移植材料[7-9]和手術(shù)刀具[10-12]; 透明導(dǎo)電薄膜由于具有優(yōu)良的光學(xué)和電學(xué)性能而被用于太陽(yáng)能電池和顯示器電極[13-22]; FeCrAl涂層由于具有高溫抗氧化以及耐腐蝕的特性而被用作核包殼材料[23-30]。常見(jiàn)的涂層制備方法有化學(xué)氣相沉積(CVD)、物理氣相沉積(PVD)、溶膠– 凝膠、噴涂、電鍍, 其中, PVD方法不涉及化學(xué)反應(yīng), 對(duì)沉積材料和基底材料限制較少, 被廣泛應(yīng)用于制備陶瓷或合金涂層。

TiAlN涂層是一種用PVD方法制備的典型材料,常見(jiàn)的有磁控濺射[31-32]和陰極電弧蒸發(fā)[33]。在磁控濺射鍍膜過(guò)程中, 系統(tǒng)遠(yuǎn)離平衡態(tài), 涂層材料由于快速沉積和急速冷卻, 會(huì)形成亞穩(wěn)態(tài)的相結(jié)構(gòu)。在材料學(xué)領(lǐng)域, 相圖(又稱為相平衡圖)直觀地反映了材料的成分、溫度和相結(jié)構(gòu)的關(guān)系, 被稱為“材料設(shè)計(jì)的索驥圖”。在TiAlN基涂層材料的研發(fā)中, Al在fcc相中的最大固溶度至關(guān)重要, 而穩(wěn)態(tài)相圖(如圖1(a)所示, Al在fcc-TiN中的固溶度幾乎為0, Ti在hcp-AlN中的固溶度也同樣可以忽略不計(jì))不能直接應(yīng)用于這類材料的研究。以TiAlN涂層為例, 如何定量描述非平衡態(tài)的亞穩(wěn)相, 成為相關(guān)領(lǐng)域研究的重點(diǎn)和難點(diǎn)之一。早期的研究主要集中于實(shí)驗(yàn), 如1986年, Münz[34]嘗試在TiN涂層中加入不同含量的Al用以提高TiAlN涂層的抗氧化性, 發(fā)現(xiàn)fcc單相結(jié)構(gòu)涂層的性能隨著Al含量的增加而增強(qiáng), 而第二相hcp結(jié)構(gòu)的出現(xiàn)會(huì)導(dǎo)致涂層性能急劇下降。隨后, 許多研究者[35-41]測(cè)定了Al在fcc結(jié)構(gòu)的Ti1–xAlN中的最大固溶度(max), Hans等[42]近期整理了前人的實(shí)驗(yàn)結(jié)果, 如圖1(b)所示,max的范圍在0.40–0.67。

許多研究者通過(guò)相圖計(jì)算CALPHAD方法, 研究Al在fcc-Ti1–xAlN中的固溶度, 如圖1(b)所示。1990年,Spencer和Stolten等[43-44]率先通過(guò)吉布斯自由能曲線估測(cè)出了max的范圍為0.70~0.72; 1998年, Chen和Sundman[45]在Zeng和Schmid-Fetzer[46]熱力學(xué)計(jì)算結(jié)果的基礎(chǔ)之上, 估測(cè)的max范圍為0.6~ 0.7; 2001年, Spencer等[47]在吉布斯自由能中引入了穩(wěn)態(tài)–亞穩(wěn)態(tài)結(jié)構(gòu)轉(zhuǎn)變能, 評(píng)估的max值為0.71。

基于密度泛函理論(DFT)的第一性原理計(jì)算也被廣泛應(yīng)用于TiAlN涂層的研究, 如圖1(b)所示。2003年, Hugosson等[48]通過(guò)計(jì)算fcc相和hcp相的能量曲線, 估測(cè)max值為0.6; 2006年, Mayrhofer等[49]考慮了Al在fcc晶格上的幾種典型分布情況, 計(jì)算出max的范圍為0.64~0.74; 2010年, Holec等[50]計(jì)算出了壓應(yīng)力下的max的范圍為0.70~0.79; 2015年, Euchner和Mayrhofer[51]考慮了金屬和非金屬亞晶格上的空位, 計(jì)算出max的范圍為0.65~0.72; 2017年, Hans等[42]基于微晶尺寸效應(yīng)計(jì)算出max的范圍為0.50~0.75。

以上的計(jì)算結(jié)果, 無(wú)論是CALPHAD方法還是DFT理論研究, 都只考慮了fcc和hcp相的能量因素, 可以將他們歸納為熱力學(xué)研究的范疇[42](如圖 1(b) 所示“Thermodynamics”)。綜上, CALPHAD獲得的max范圍為0.60~0.72, 占實(shí)驗(yàn)所測(cè)范圍的24%; DFT計(jì)算的max范圍為0.50~0.79, 占實(shí)驗(yàn)所測(cè)范圍的58%。同時(shí), 相關(guān)的理論預(yù)測(cè)始終沒(méi)有包含實(shí)驗(yàn)值的最低邊界[52]。因此, 僅考慮熱力學(xué)的理論計(jì)算無(wú)法覆蓋實(shí)驗(yàn)全部的max的范圍。

近期, Sangiovanni等[53-55]從動(dòng)力學(xué)的角度引入了表面擴(kuò)散吸附能來(lái)模擬穩(wěn)態(tài)TiN的生長(zhǎng), Alling等[56]在研究原子沿著TiAlN(001)面擴(kuò)散時(shí)考慮了構(gòu)型無(wú)序?qū)Ρ砻鏀U(kuò)散的影響。相關(guān)的動(dòng)力學(xué)模擬有助于理解涂層材料的生成過(guò)程。為了全面地描述TiAlN涂層材料制備過(guò)程中亞穩(wěn)相的形成機(jī)理, 需要將動(dòng)力學(xué)和熱力學(xué)有機(jī)結(jié)合起來(lái)。2015~2016年, Chang等[57-58]在Cantor和Cahn[59]的理論基礎(chǔ)之上發(fā)展了用于預(yù)測(cè)亞穩(wěn)相圖的新模型, 這一新模型結(jié)合了熱力學(xué)和動(dòng)力學(xué)計(jì)算, 通過(guò)一個(gè)組合式磁控濺射鍍膜實(shí)驗(yàn)輔以高通量表征, 耦合CALPHAD和第一性原理計(jì)算提供的關(guān)鍵數(shù)據(jù), 對(duì)涂層材料的亞穩(wěn)相進(jìn)行表面擴(kuò)散模擬并通過(guò)實(shí)驗(yàn)驗(yàn)證了該模型的可靠性, 結(jié)果表明, Chang等[57-58]的模型比Saunders和Miodowni[60]的方法更直接有效。這種新的模型結(jié)合了熱力學(xué)和動(dòng)力學(xué)計(jì)算, 通過(guò)一個(gè)組合式磁控濺射鍍膜實(shí)驗(yàn)輔以高通量表征, 耦合CALPHAD和第一性原理計(jì)算提供的關(guān)鍵數(shù)據(jù), 對(duì)涂層材料的亞穩(wěn)相進(jìn)行表面擴(kuò)散模擬, 成功地構(gòu)建了簡(jiǎn)單二元合金體系Cu-W和Cu-V的亞穩(wěn)相圖, 相關(guān)結(jié)果被實(shí)驗(yàn)數(shù)據(jù)證實(shí)。2018年, Liu等[61]借助于Chang等[57-58]提出的模型預(yù)測(cè)出了Al在Ti1–xAlN中的最大溶解度范圍為0.42~0.68, 如圖1(b)所示, 與實(shí)驗(yàn)值[35-41,62-66]十分吻合。本文將圍繞亞穩(wěn)相圖模型的發(fā)展歷程, 重點(diǎn)介紹近期建立的臨界表面擴(kuò)散模型及其在特種涂層領(lǐng)域的應(yīng)用, 并對(duì)模型未來(lái)的發(fā)展方向進(jìn)行展望。

圖1 TiAlN體系的相圖

(a)穩(wěn)態(tài)TiN-AlN偽二元相圖[61], Al在fcc相中的固溶度可以忽略不計(jì); (b)不同計(jì)算方法獲得fcc-Ti1–xAlN涂層中Al的固溶度(max)與實(shí)驗(yàn)值[35-37, 40-41, 49-52, 61-66]的對(duì)比

Fig. 1 The phase diagram of TiAlN

(a) The stable TiN-AlN pseudo binary phase diagram[61], among which the Al solubilities of fcc phase is negligible; (b) The critical Al solubilities (max) in Ti1–xAlN by different calculation methods compared with the experimental data[35-37, 40-41, 49-52, 61-66]

1 亞穩(wěn)相圖模型的構(gòu)建

通過(guò)磁控濺射方法制備薄膜時(shí), 處于氣態(tài)的濺射原子快速到達(dá)基體表面急速冷卻, 同時(shí)受到熱力學(xué)和動(dòng)力學(xué)的控制, 材料的相結(jié)構(gòu)取決于濺射原子(如圖2所示)沉積在基底后的擴(kuò)散行為[67-68]。濺射原子存在表面擴(kuò)散和體擴(kuò)散兩種擴(kuò)散形式, 由于制備溫度一般遠(yuǎn)低于薄膜材料本身的熔點(diǎn)[67-68], 體擴(kuò)散的影響可以忽略。根據(jù)動(dòng)力學(xué)相關(guān)理論可知, 如果濺射原子表面擴(kuò)散的距離足夠長(zhǎng), 就可以形成穩(wěn)定的相, 反之原子的擴(kuò)散距離非常短, 則會(huì)形成亞穩(wěn)相。因此, 表面擴(kuò)散距離對(duì)于涂層的相結(jié)構(gòu)至關(guān)重要。

1.1 Einstein擴(kuò)散模型

Einstein首先提出了原子的表面擴(kuò)散距離公式[69], 如式(1)所示:

其中, 表示擴(kuò)散距離, 表示表面擴(kuò)散系數(shù), 表示擴(kuò)散時(shí)間,為表面原子的振動(dòng)頻率(通?？醋饕粋€(gè)常數(shù), 約為), 表示擴(kuò)散激活能, 為玻爾茲曼常數(shù), 是溫度。

1.2 Cantor和Cahn擴(kuò)散模型

Cantor和Cahn[59]認(rèn)為在磁控濺射鍍膜過(guò)程中, 原子擴(kuò)散時(shí)間受到相鄰原子的限制, 在式(1)的基礎(chǔ)之上, Cantor和Cahn[59]結(jié)合Al-Cu、Al-Ni和Al-Fe的PVD實(shí)驗(yàn)總結(jié)出了式(2):

1.3 Saunders和Miodownik擴(kuò)散模型

Saunders和Miodownik[60]認(rèn)為氣相沉積形成的相結(jié)構(gòu)與沉積在基底最底層的原子的分解擴(kuò)散有關(guān)。他們認(rèn)為, Cantor和Cahn[59]的實(shí)驗(yàn)溫度較低, 所以不用考慮體擴(kuò)散的影響; 當(dāng)溫度上升時(shí), 體擴(kuò)散會(huì)使原子從原先的沉積層遷移到表面再次擴(kuò)散, 此時(shí)擴(kuò)散距離的公式可以在式(1)的基礎(chǔ)上修改, 如式(3)所示:

1.4 臨界表面擴(kuò)散模型

圖3 室溫沉積涂層獲得的相結(jié)構(gòu)與平衡相圖的對(duì)比(a) Al-Cu[59,71]; (b)Al-Ni[59,72]; (c)Al-Fe[59,70]

圖4 Saunders和Miodownik[60]根據(jù)擴(kuò)散公式得到的擴(kuò)散距離與溫度的關(guān)系(a) Cu-11.5at% Sn和(b) Cu-19.5at% Sn及(c) Cu-Sn體系的穩(wěn)態(tài)相圖[73]

結(jié)合理論計(jì)算和關(guān)鍵實(shí)驗(yàn), Chang等[58]總結(jié)出獲得亞穩(wěn)相圖的流程圖, 表示在圖5中, 即通過(guò)耦合CALPHAD、第一性原理計(jì)算和高通量磁控濺射鍍膜實(shí)驗(yàn)的方法對(duì)涂層材料的亞穩(wěn)相進(jìn)行表面擴(kuò)散模擬, 相關(guān)計(jì)算僅需要一個(gè)高通量鍍膜實(shí)驗(yàn)作為基礎(chǔ)數(shù)據(jù)。

2 亞穩(wěn)相圖的應(yīng)用

亞穩(wěn)相圖模型不僅可以用來(lái)預(yù)測(cè)涂層材料的相形成關(guān)系, 還可以從原子擴(kuò)散的角度解釋亞穩(wěn)相的分解及穩(wěn)態(tài)相的形成。將模型應(yīng)用于Cu-W和Cu-V薄膜體系得到了最終的相形成圖; 模型成功地解釋了Pt-Ir合金薄膜不易分解的原因; TiAlN涂層的亞穩(wěn)相圖也通過(guò)模型計(jì)算獲得, 由此預(yù)測(cè)了不同溫度下Al在fcc相中的最大固溶度。

2.1 Cu-W和Cu-V體系的亞穩(wěn)相圖

Chang等[57-58]通過(guò)耦合CALPHAD、第一性原理計(jì)算和高通量磁控濺射鍍膜實(shí)驗(yàn)的方法獲得了Cu-W和Cu-V的亞穩(wěn)相圖, 如圖6(a~b)所示, 其穩(wěn)態(tài)相圖[74-75]如圖6(c~d)所示。不同的能量密度引起沉積速率發(fā)生變化, 因此亞穩(wěn)相圖不是單一的成分溫度關(guān)系圖, 而是一系列成分溫度曲線, 曲線之間的變化趨勢(shì)基本相同, 且在0 ℃時(shí)有相同的最大固溶度。對(duì)于給定的成分, 若溫度低于臨界溫度則原子擴(kuò)散不充分, 會(huì)形成非平衡態(tài)的單相, 當(dāng)溫度逐漸上升時(shí), 原子擴(kuò)散變得相對(duì)充分, 穩(wěn)定的第二相將會(huì)逐漸形成, 預(yù)測(cè)得到的亞穩(wěn)相圖與實(shí)驗(yàn)數(shù)據(jù)完全吻合。

2.2 Pt-Ir涂層亞穩(wěn)相形成機(jī)理

Pt-Ir合金具有優(yōu)異的抗腐蝕性能和優(yōu)良的力學(xué)性能[76-78], 可以應(yīng)用在高溫和腐蝕性強(qiáng)的環(huán)境中。例如在精密玻璃模具制造行業(yè), 常常要求模具中的相在400~700 ℃[79-80]加載時(shí)仍能穩(wěn)定存在不分解, 而Pt-Ir保護(hù)層[81-84]剛好可以滿足這些條件。

為了探究Pt-Ir涂層在高溫下不易分解的特性, Saksena等[85]引入了Pt-Au作為參考體系和Pt-Ir進(jìn)行對(duì)比。Pt-Au和Pt-Ir體系的穩(wěn)態(tài)相圖類似(如圖7(a~b)所示), 都存在fcc相的固溶度間隙。不同溫度下的磁控濺射鍍膜實(shí)驗(yàn)結(jié)果顯示, Pt-Ir涂層即使在高溫(~950 ℃)下仍是以fcc單相存在, 如圖7(c)所示; 而Pt-Au涂層在600 ℃時(shí)已經(jīng)存在相分解的情況, 如圖7(d)所示。根據(jù)式(2)可知擴(kuò)散距離隨著擴(kuò)散激活能的增加成指數(shù)式減小, 而在溫度相同的情況下, Ir原子的表面擴(kuò)散激活能在Pt0.5Ir0.5中是Au原子在Pt0.5Au0.5中的6倍多[85], 因此Pt-Ir體系中的原子表面擴(kuò)散距離很小, 亞穩(wěn)相結(jié)構(gòu)容易保持, 不容易出現(xiàn)穩(wěn)定的第二相。由此, 基于表面擴(kuò)散, 可以很好地理解Pt-Ir涂層材料的穩(wěn)定性能。

圖5 PVD涂層材料的亞穩(wěn)相圖模擬流程圖[58]

圖6 (a)計(jì)算預(yù)測(cè)的Cu-W體系亞穩(wěn)相圖與實(shí)驗(yàn)數(shù)據(jù)的對(duì)比[58]; (b)計(jì)算預(yù)測(cè)的Cu-V體系亞穩(wěn)相圖與實(shí)驗(yàn)數(shù)據(jù)的對(duì)比[58]; (c) Cu-W體系的穩(wěn)態(tài)相圖[74]; (d) Cu-V體系的穩(wěn)態(tài)相圖[75]

圖7 (a) Pt-Ir體系的穩(wěn)態(tài)相圖[86]; (b) Pt-Au體系的穩(wěn)態(tài)相圖[87]; (c) Pt-Ir體系的鍍膜實(shí)驗(yàn)結(jié)果[85]; (d) Pt-Au體系的鍍膜實(shí)驗(yàn)結(jié)果[85]

2.3 TiAlN涂層的亞穩(wěn)相圖

Liu等[61]采用相關(guān)模型預(yù)測(cè)了Ti1–xAlN涂層的亞穩(wěn)相形成圖, 如圖8(a)所示, 圖8(b~d)分別表示預(yù)測(cè)亞穩(wěn)相圖與實(shí)驗(yàn)值的比較。通過(guò)PVD制備的Ti1–xAlN涂層的最大固溶度在0 ℃的計(jì)算值為0.68, 500 ℃的計(jì)算值為0.42, 預(yù)測(cè)的最大固溶度與實(shí)驗(yàn)值吻合很好。預(yù)測(cè)結(jié)果表明, 溫度較低時(shí), 原子擴(kuò)散不充分, 僅存在非平衡的單相區(qū); 而當(dāng)溫度進(jìn)一步升高時(shí), 原子擴(kuò)散更加充分, 因此開(kāi)始形成第二相, 兩相區(qū)出現(xiàn)。同理, 沉積速率的改變會(huì)得到一系列臨界溫度隨成分變化的曲線, 相關(guān)的預(yù)測(cè)結(jié)果都得到了實(shí)驗(yàn)驗(yàn)證[66], 如圖8(b~d)所示。

研究者[1-2]普遍認(rèn)為TiAlN涂層的亞穩(wěn)相形成主要是由擴(kuò)散動(dòng)力學(xué)控制的, 因此僅依賴相圖計(jì)算或者密度泛函理論的熱力學(xué)模擬(圖1(b)所示)所得到的結(jié)果是不可靠[42-51]的。Liu等[61]的工作僅基于一個(gè)關(guān)鍵的組合式磁控濺射鍍膜實(shí)驗(yàn)得到相關(guān)的數(shù)據(jù), 輔以第一性原理計(jì)算和CALPHAD結(jié)果, 對(duì)TiAlN體系進(jìn)行表面擴(kuò)散動(dòng)力學(xué)模擬, 成功預(yù)測(cè)了TiAlN涂層材料的亞穩(wěn)相形成圖。該模型基于熱力學(xué)得到了亞穩(wěn)相圖上的一個(gè)點(diǎn), 通過(guò)動(dòng)力學(xué)計(jì)算獲得了覆蓋寬廣溫度和成分范圍的亞穩(wěn)相圖。尤其是Al在fcc 相中的固溶度極限可以擴(kuò)展到更低的成分, 這是前人[42-51]僅通過(guò)熱力學(xué)計(jì)算無(wú)法解釋的。因此, Liu 等[61]將相關(guān)模型率先成功地運(yùn)用到了氮化物PVD涂層材料亞穩(wěn)相圖的模擬中。

3 總結(jié)與展望

在PVD磁控濺射鍍膜過(guò)程中, 表面擴(kuò)散控制亞穩(wěn)相的形成, 而臨界擴(kuò)散距離決定材料所屬的相區(qū)。研究者近期提出的模型, 將第一性原理、CALPHAD與組合式磁控濺射實(shí)驗(yàn)結(jié)合, 僅基于一次高通量關(guān)鍵實(shí)驗(yàn)數(shù)據(jù)與理論計(jì)算的結(jié)果, 便可以預(yù)測(cè)涂層材料的亞穩(wěn)相圖。相關(guān)工作不僅打破了CALPHAD難以直接計(jì)算PVD涂層材料亞穩(wěn)相圖的局限, 為亞穩(wěn)相圖的預(yù)測(cè)提供了一個(gè)全新的思路, 而且與“材料基因工程”的理念相符, 助推材料研發(fā)時(shí)間和成本“雙減半”目標(biāo)的實(shí)現(xiàn)。同時(shí), PVD涂層材料亞穩(wěn)相圖的模型需要不斷地完善, 從而應(yīng)用到更寬廣的科學(xué)與工程領(lǐng)域。

圖8 Ti1–xAlxN亞穩(wěn)相圖

(a)預(yù)測(cè)的亞穩(wěn)相圖[61], 不同顏色的曲線代表不同的磁控濺射能量密度; (b)預(yù)測(cè)亞穩(wěn)相圖與能量密度為2.3 W?cm–2, 溫度在100~550 ℃的實(shí)驗(yàn)數(shù)據(jù)對(duì)比[61]; (c)預(yù)測(cè)亞穩(wěn)相圖與能量密度為4.6 W?cm–2, 溫度在100~550 ℃的實(shí)驗(yàn)數(shù)據(jù)對(duì)比;(d)預(yù)測(cè)亞穩(wěn)相圖與能量密度為6.8W?cm–2, 溫度在100~550 ℃的實(shí)驗(yàn)數(shù)據(jù)對(duì)比

Fig. 8 Metastable Ti1–xAlN phase formation diagrams

(a) Predicted diagrams at different power densities; (b) The predicted diagram compared with the experimental data with power density of 2.3 W?cm–2at 100–550 ℃; (c) The predicted diagram compared with the experimental data with power density of 4.6 W?cm–2at 100–550 ℃; (d) The predicted diagram compared with the experimental data with power density of 6.8 W?cm–2at 100–550 ℃

在科學(xué)研究中, 模型需要進(jìn)一步拓展。實(shí)際材料體系包含的元素種類眾多, 有必要探索多元體系的亞穩(wěn)相圖。當(dāng)前, 利用亞穩(wěn)相圖模型研究材料的重點(diǎn)為二元、三元體系, 可以借助CALPHAD方法外推的思路, 獲得多元材料體系的亞穩(wěn)相圖數(shù)據(jù)庫(kù), 從而與CALPHAD構(gòu)建的穩(wěn)定相圖數(shù)據(jù)庫(kù)結(jié)合起來(lái)。在此基礎(chǔ)上, 通過(guò)組分–制備工藝–組織結(jié)構(gòu)和性能的關(guān)系, 為合成新型涂層材料提供設(shè)計(jì)思路, 通過(guò)組元和成分的快速篩選, 可以實(shí)現(xiàn)材料的高效開(kāi)發(fā)。

在工程領(lǐng)域, 模型需要做進(jìn)一步的實(shí)際應(yīng)用驗(yàn)證。涂層材料應(yīng)用廣泛、制備工藝種類多, 除了本文提及的PVD鍍膜技術(shù), 化學(xué)氣相沉積(CVD, Chemical Vapor Deposition)也是常見(jiàn)的一種涂層制備技術(shù), 材料的制備過(guò)程也是非平衡態(tài), 從而獲得亞穩(wěn)相。因此, 亞穩(wěn)相圖模型的拓展應(yīng)用需要不同技術(shù)制備樣品的實(shí)驗(yàn)結(jié)果來(lái)驗(yàn)證, 尤其是在關(guān)鍵工藝參數(shù)改變的情況下, 模型的可靠性與完善性需要實(shí)驗(yàn)數(shù)據(jù)的支撐。圍繞模型進(jìn)行更深入的基礎(chǔ)理論和關(guān)鍵實(shí)驗(yàn)研究, 是未來(lái)工作的重點(diǎn)。模型的普適性將極大拓展其在新型結(jié)構(gòu)/功能陶瓷涂層材料研發(fā)中的應(yīng)用。

[1] ENDRINO J C, ?RHAMMAR C, GUTIéRREZ C,Spectral evidence of spinodal decomposition, phase transformation and molecular nitrogen formation in supersaturated TiAlN films upon annealing., 2011, 59(2): 6287–6296.

[2] RACHBAUER R, GENGLER J J, VOEVODIN A A,Temperature driven evolution of thermal, electrical, and optical properties of Ti–Al–N coatings., 2012, 60: 2091–2096.

[3] MAYRHOFER P, RACHBAUER R, HOLEC D,Protective transition metal nitride coatings., 2014, 4: 355–387.

[4] NORRBY N, ROGSTR?M L, JOHANSSON-J?ESAAR M P,X-ray scattering study of the cubic to hexagonal transformation of AlN in Ti1–xAlN., 2014, 73: 205–214.

[5] SCHRAMM I C, J?ESAAR M P J, JENSEN J,Impact of nitrogen vacancies on the high temperature behavior of (Ti1?xAl)Nalloys., 2016, 119: 218–228.

[6] ZOU H K, CHEN L, CHANG K K,Enhanced hardness and age-hardening of TiAlN coatings through Ru addition., 2019, 162: 382–386.

[7] RAJAN S T, KARTHIKA M, BENDAVID A,Apatite layer growth on glassy Zr48Cu36Al8Ag8sputtered titanium for potential biomedical applications., 2016, 369: 501–509.

[8] CAI C N, ZHANG C, SUN Y S,ZrCuFeAlAg thin film metallic glass for potential dental applications., 2017, 86: 80–87.

[9] LEE C M, CHU J P, CHANG W Z,Fatigue property improvements of Ti–6Al–4V by thin film coatings of metallic glass and TiN: a comparison study., 2014, 561: 33–37.

[10] TSAI P H, LIN Y Z, LI J B,Sharpness improvement of surgical blade by means of ZrCuAlAgSi metallic glass and metallic glass thin film coating., 2012, 31: 127–131.

[11] TSAI P H, LI T H, HSU K T,Effect of coating thickness on the cutting sharpness and durability of Zr-based metallic glass thin film coated surgical blades., 2016, 618: 36–41.

[12] CHU J P, JANG J S C, HUANG J C,Thin film metallic glasses: unique properties and potential applications., 2012, 520: 5097–5122.

[13] GORDON R G. Criteria for choosing transparent conductors., 2000, 25: 52–57.

[14] CHENG C H, TING J M. Transparent conducting GZO, Pt/GZO, and GZO/Pt/GZO thin films., 2007, 516: 203–207.

[15] KIM D. Improved electrical and optical properties of GZO films with a thin TiO2buffer layer deposited by RF magnetron sputtering., 2014, 40: 1457–1460.

[16] WANG H L, SUN Y H, FANG L,Growth and characterization of high transmittance GZO films prepared by Sol-Gel method., 2016, 615: 19–24.

[17] LU L F, SHEN H L, FENG J,The enhanced conductivity of AZO thin films on soda lime glass with an ultrathin Al2O3buffer layer., 2010, 405: 3320–3323.

[18] SHEN H L, ZHANG H, LU L F,Preparation and properties of AZO thin films on different substrates., 2010, 20: 44–48.

[19] SUBRAMANYAM T K, GOUTHAM P, PAVAN KUMAR S,Optimization of sputtered AZO thin films for device application., 2018, 5: 10851–10859.

[20] Lü K, ZHU B L, LI K,Effect of hydrogen and Cu interlayer on the optical and electrical properties of GZO thin film., 2014, 29(5): 494–497.

[21] CHEN D, Lü J G, HUANG J Y,Performances of GaN-based LEDs with AZO films as transparent electrodes., 2013, 28(6): 650–652.

[22] QIN X J, HAN S H Z, ZHAO L,Fabrication of transparent conductive Al-doped ZnO thin films by aerosol-assisted chemical vapour deposition., 2011, 26(6): 608–611.

[23] AYDOGAN E, WEAVER J S, MALOY S A,Microstructure and mechanical properties of FeCrAl alloys under heavy ion irradiations., 2018, 503: 250–262.

[24] DRYEPONDT S, UNOCIC K A, HOELZER D T,Development of low-Cr ODS FeCrAl alloys for accident-tolerant fuel cladding., 2018, 501: 59–71.

[25] GUSSEV M N, CAKMAK E, FIELD K G,Impact of neutron irradiation on mechanical performance of FeCrAl alloy laser-beam weldments., 2018, 504: 221–233.

[26] HOGGAN R E, HE L F, HARP J M,Interdiffusion behavior of U3Si2with FeCrAldiffusion couple studies., 2018, 502: 356–369.

[27] JIN D L, NI N, GUO Y,Corrosion of the bonding at FeCrAl/ Zr alloy interfaces in steam., 2018, 508: 411–422.

[28] PARK D, MOUCHE P A, ZHONG W C,TEM/STEM study of zircaloy-2 with protective FeAl(Cr) layers under simulated BWR environment and high-temperature steam exposure., 2018, 502: 95–105.

[29] TANG C C, JIANU A, STEINBRUECK M,Influence of composition and heating schedules on compatibility of FeCrAl alloys with high-temperature steam., 2018, 511: 496–507.

[30] CHANG K K, MENG F P, GE F F,Theory-guided bottom- up design of the FeCrAl alloys as accident tolerant fuel cladding materials., 2019, 516: 63–72.

[31] CAIRNEY J M, HARRIS S G, MUNROE P R,Transmission electron microscopy of TiN and TiAlN thin films using specimens prepared by focused ion beam milling., 2004, 183: 239–246.

[32] RACHBAUER R, STERGAR E, MASSL S,Three-dimen-sional atom probe investigations of Ti–Al–N thin films., 2009, 61: 725–728.

[33] WEBER F R, FONTAINE F, SCHEIB M,Cathodic arc evaporation of (Ti, Al) N coatings and (Ti, Al) N/TiN multilayer-coatings–correlation between lifetime of coated cutting tools, structural and mechanical film properties., 2004, 177–178: 227–232

[34] MüNZ W D. A new alternative to TiN coatings., 1986, 4(6): 2717–2725.

[35] WAHLSTR?M U, HULTMAN L, SUNDGREN J E,Crystal growth and microstructure of polycrystalline Ti1?xAlN alloy films deposited by ultra-high-vacuum dual-target magnetron sputtering., 1993, 235: 62–70.

[36] ZHOU M, MAKINO Y, NOSE M,Phase transition and properties of Ti–Al–N thin films prepared by rf-plasma assisted magnetron sputtering., 1999, 339: 203–208.

[37] H?RLING A, HULTMAN L, ODéN M,Mechanical properties and machining performance of Ti1?xAlN-coated cutting tools., 2005, 191: 384–392.

[38] KUTSCHEJ K, MAYRHOFER P, KATHREIN M,Structure, mechanical and tribological properties of sputtered Ti1–xAlN coatings with 0.5≤≤0.75., 2005, 200: 2358–2365.

[39] RAFAJA D, POKLAD A, SCHREIBER G,On the preferred orientation in Ti1–xAlN and Ti1–x–yAlSiN thin films., 2005, 96: 738–742.

[40] GRECZYNSKI G, LU J, JOHANSSON M,Role of Ti+and Al+ion irradiation (=1, 2) during Ti1–xAlN alloy film growth in a hybrid HIPIMS/magnetron mode., 2012, 206: 4202–4211.

[41] GROSSMANN B, SCHALK N, CZETTL C,Phase composition and thermal stability of arc evaporated Ti1?xAlN hard coatings with 0.4≤≤0.67., 2017, 309: 687–693.

[42] HANS M, MUSIC D, CHEN Y T,Crystallite size-dependent metastable phase formation of TiAlN coatings., 2017, 7(16096): 1–7.

[43] SPENCER P, HOLLECK H. Application of a thermochemical data-bank system to the calculation of metastable phase formation during PVD od carbide, nitride and boride coatings., 1990, 27: 295–309.

[44] STOLTEN H, SPENCER P, NEUSCHüTZ D,Estimation of formation enthalpies for transition metal mono nitrides, carbides and oxides., 1993, 90: 209–219.

[45] CHEN Q, SUNDMAN B. Thermodynamic assessment of the Ti-Al-N system., 1998, 19(20): 146–160.

[46] ZENG K J, SCHMID-FETAER R. Thermodynamic Modeling and Applications of the Ti-Al-N Phase Diagram. Thermodynamics of Alloy Formation, 1997, 275–293.

[47] SPENCER P. Thermodynamic prediction of metastable coating structures in PVD processes., 2001, 92: 1145–1150.

[48] HUGOSSON H W, H?GBERG H, ALGREN M,Theory of the effects of substitutions on the phase stabilities of Ti1?xAlN., 2003, 93(8): 4505–4511.

[49] MAYRHOFER P, MUSIC D, SCHNEIDER J,Influence of the Al distribution on the structure, elastic properties, and phase stability of supersaturated Ti1?xAlN., 2006, 100: 1–5.

[50] HOLEC D, ROVERE F, MAYRHOFER P H,Pressure- dep-endent stability of cubic and wurtzite phases within the TiN–AlN and CrN–AlN systems., 2010, 62: 349–352.

[51] EUCHNER H, MAYRHOFER P. Vacancy-dependent stability of cubic and wurtzite Ti1?xAlN., 2015, 275: 214–218.

[52] SHULUMBA N, HELLMAN O, RAZA Z,Lattice vibrations change the solid solubility of an alloy at high temperatures., 2016, 117(11): 1–6.

[53] SANGIOVANNI D, EDSTR?M D, HULTMAN L,Dynamics of Ti, N, and TiN(= 1–3) admolecule transport on TiN (001) surfaces., 2012, 8: 1–10.

[54] SANGIOVANNI D, EDSTR?M D, HULTMAN L,and classical molecular dynamics simulations of N2desorption from TiN (001) surfaces., 2014, 624: 25–31.

[55] SANGIOVANNI D, ALLING B, STENETEG P,Self- interstitial diffusion, and Frenkel-pair formation/dissociation in B1 TiN studied byand classical molecular dynamics with optimized potentials., 2015, 91: 1–17.

[56] ALLING B, STENETEG P, THOLANDER C,Configurational disorder effects on adatom mobilities on Ti1?xAlN (001) surfaces from first principles., 2012, 85(24): 1–5.

[57] CHANG K K, TO BABEN M, MUSIC D,Estimation of the activation energy for surface diffusion during metastable phase formation., 2015, 98: 135–140.

[58] CHANG K K, MUSIC D, TO BABEN M,Modeling of metastable phase formation diagrams for sputtered thin films., 2016, 17: 210–219.

[59] CANTOR B, CAHN R. Metastable alloy phases by co-sputtering., 1976, 24: 845–852.

[60] SAUNDERS N, MIODOWNIK A P. Phase formation in co-deposited metallic alloy thin flms., 1987, 22: 629–637.

[61] LIU S D, CHANG K K, MRáZ S,Modeling of metastable phase formation for sputtered Ti1–xAlN thin films., 2019, 165: 615–625.

[62] KIM K H, LEE S H. Comparative studies of TiN and Ti1?xAlN by plasma-assisted chemical vapor deposition using a TiCl4/AlCl3/N2/ H2/Ar gas mixture., 1996, 283: 165–170.

[63] LEE S H, KIM B J, KIM H H,Structural analysis of AlN and (Ti1?xAl) N coatings made by plasma enhanced chemical vapor deposition., 1996, 80(3): 1469–1473.

[64] PRANGE R, CREMER R, NEUSCHüTZ D,Plasma- enhanced CVD of (Ti, Al) N films from chloridic precursors in a DC glow discharge., 2000, 133–134: 208–214.

[65] ENDLER I, H?HN M, HERRMANN M,Novel aluminum- rich Ti1?xAlN coatings by LPCVD., 2008, 203: 530?533.

[66] TODT T, ZALESAK J, DANIEL R,Al-rich cubic Al0. 8Ti0. 2N coating with self-organized nano-lamellar microstructure: thermal and mechanical properties., 2016, 291: 89–93.

[67] GROVENOR C R M, HENTZELL H T G, SMITH D A,The development of grain structure during growth of metallic films., 1984, 32(5): 773–781.

[68] PETROV I, BARNA P B, HULTMAN L,Microstructural evolution during film growth., 2003, 21(5): S117–S128.

[69] EINSTEIN A. Elementare teorie der brownschen bewegung., 1908, 14: 371–381.

[70] SUNDMAN B, OHNUMA I, DUPIN N,An assessment of the entire Al-Fe system including D03ordering., 2009, 57: 2896–2908.

[71] LIANG S M, SCHMID-FETZER R. Thermodynamic assessment of the Al–Cu–Zn system, part II: Al–Cu binary system., 2015, 51: 252–260.

[72] ZHOU C Y, CUI J X, GUO C P,Thermodynamic description of the Al–Ge–Ni system over the whole composition and temperature ranges., 2017, 58: 138–150.

[73] LI M, DU Z M, GUO C,Thermodynamic optimization of the Cu–Sn and Cu–Nb–Sn systems., 2009, 477: 104–117.

[74] SUBRAMANINA P R, LAUGHLIN D E. Cu-W (copper tungsten)., 1991, 12: 76–79.

[75] ZHAO J R, DU Y, ZHANG L J,Thermodynamic reassessment of the Cu–V system supported by key experiments., 2008, 32: 252–255.

[76] WEILAND R, LUPTON D F, FISCHER B,High-temperature mechanical properties of the platinum group metals., 2006, 50(4): 158–170.

[77] GANSKE G, SLAVCHEVA E, VAN OOYEN A,. Sputtered platinum–iridium layers as electrode material for functional electrostimulation., 2011, 519: 3965–3970.

[78] TSENG S F, LEE C T, HUANG K C,Mechanical properties of Pt-Ir and Ni-Ir binary alloys for glass-molding dies coating., 2011, 11(10): 8682–8688.

[79] YI A Y, JAIN A. Compression molding of aspherical glass lenses–a combined experimental and numerical analysis., 2005, 88(3): 579–586.

[80] KLOCKE F, DAMBON O, YI A Y,Process Chain for the Replication of Complex Optical Glass Components. Berlin Heidelberg: Springer, 2013: 119–132.

[81] MA K J, CHIEN H H, CHUAN W H,Design of protective coatings for glass lens molding., 2008, 364–366: 655–661.

[82] KLOCKE F, DAMBON O, GEORGIADIS K,Comparison of nitride and noble metal coatings for precision glass molding tools., 2010, 438: 9–6.

[83] BOBZIN K, BAGCIVAN N, EWERING M,Influence of interlayer thickness of a thin noble metal MSIP-PVD coating on compound and system properties for glass lens moulding., 2012, 6: 311–318.

[84] BOBZIN K, BAGCIVAN N, BR?GELMANN T,Development and qualification of a MSIP PVD iridium coating for precision glass moulding., 2013, 44(8): 673–678.

[85] SAKSENA A, CHIEN Y C, CHANG K K,Metastable phase formation of Pt-X (X=?Ir,?Au)?thin?flms., 2018, 8(1): 10198.

[86] YAMABE-MITARAI Y, AOYAGI T, ABE T,An investigation of phase separation in the Ir–Pt binary system., 2009, 484: 327–334.

[87] XU X N, REN Y P, LI C F,Thermodynamic assessment of Au-Pt system., 2012, 22: 1432–1436.

Review on Metastable Phase Diagrams: Application Roles in Specialty Ceramic Coatings

HUANG Ye-Yan1,2, XU Kai1, WU Bo2, LI Peng1, CHANG Ke-Ke1, HUANG Feng1, HUANG Qing1

(1. Engineering Laboratory of Advanced Energy Materials, Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences, Ningbo 315201, China; 2. Multiscale Computational Materials Facility, College of Materials Science and Engineering, Fuzhou University, Fuzhou 350100, China)

Phase diagrams, also known as equilibrium phase diagrams, serve as a road map for materials design. However, preparation process of coatings (such as Physical Vapor Deposition, PVD) is generally far from equilibrium and results in metastable phases. Thus, the CALPHAD (Calculation of Phase Diagrams) approach faces a challenge in calculating the metastable phase diagrams for PVD coating materials. Here we summarized the development of the modeling methodology for the metastable phase diagrams, where the model with critical surface diffusion distance established in recent years were highlighted. The CALPHAD approach, first-principles calculations coupled with high-throughput magnetron sputtering experiments were used to model the atomic surface diffusion, while only one key combinatorial experiment was performed to obtain the basic data for the computation, and the calculated metastable phase diagrams were confirmed by further experiments. Therefore, the database of the stable and metastable phase diagrams can be established, which will be used to guide the design of the ceramic coating materials by the relationship of composition, processing, microstructure, and performance. This model can also help to achieve the goal to shorten the time and reduce the costs of materials research and development.

metastable phase formation diagram; model; surface diffusion; application; review

TQ174

A

1000-324X(2020)01-0019-10

10.15541/jim20190272

2019-06-03;

2019-07-22

國(guó)家自然科學(xué)基金(51701232); 中國(guó)科學(xué)院率先行動(dòng)“百人計(jì)劃”(2017-118)

National Natural Science Foundation of China (51701232); Hundred Talents Program of Taking the Lead of the Chinese Academy of Sciences (2017-118)

黃燁琰(1995–), 女, 博士研究生. E-mail: huangyeyan@nimte.ac.cn

HUANG Ye-Yan(1995–), female, PhD candidate. E-mail: huangyeyan@nimte.ac.cn

?？煽? 研究員. E-mail: changkeke@nimte.ac.cn

CHANG Ke-Ke, professor. E-mail: changkeke@nimte.ac.cn