基于纳米活性结构的不互溶W-Cu体系直接合金化及其热力学机制

图1 经过两步阳极氧化和H₂退火还原W板的表面和截面形貌SEM像

Fig.1 Surface SEM images and corresponding EDS results (insets) (a, c) and cross-section SEM images of W plate (b, d) with nanoporous structure layer prepared through two-step anodizing process without (a, b) and with (c, d) deoxidation process

为了分析表面纳米化是否能够提高W表面活性，对表面纳米化前后的W板进行了线性伏安曲线(LSV)对比测试，结果如图2所示。可以看出，当电流密度为1 mA/cm²时，具有表面纳米多孔结构的W片的过电位为199 mV，小于未经过处理的的W片的过电位(272 mV)。一般来说，析氢起始电位越低，析氢所需的能耗越少，材料的活性越好^[41]。由此可以得出，具有表面纳米多孔结构的W的析氢能力明显提高，具有更高的活性。

图2

图2 表面纳米多孔化前后W板的线性伏安曲线

Fig.2 Linear sweep voltammetry (LSV) curves of the nanoporous W plate and the pretreated W plate (RHE—reversible hydrogen electrode)

2.2 W/Cu层状复合材料界面显微结构

基于纳米多孔活性结构诱发直接合金化制备的W/Cu层状复合材料的界面HRTEM分析结果如图3所示。其中，图3a是W/Cu界面的高角度环形暗场(HAADF)像；图3b是按照图3a中箭头的位置和方向进行的W/Cu界面EDS线扫描元素分布曲线。图3c和d分别是W/Cu界面处W元素和Cu元素的面扫描结果。可以看出，W和Cu之间确实发生了相互扩散，实现了直接合金化，扩散层的厚度约为27 nm。W表面未纳米多孔化而在近Cu熔点温度与Cu直接合金化所获得的W/Cu扩散层厚度可以达到22 nm^[23]。与之相比，基于W表面纳米多孔化所获得的W/Cu界面上扩散层厚度增加了5 nm，该增加量对于液、固两态均不互溶的W-Cu金属系统而言是比较大的。实际上本团队最新研究表明，如果改进纳米化方法，可以将W-Cu合金化扩散层厚度增加至32 nm，相关结果将另文发表。上述结果表明纳米多孔结构确实促进了W-Cu的直接合金化，在W和Cu之间实现了冶金结合。

图3

图3 W/Cu界面高角环形暗场(HAADF)像及元素分布图

Fig.3 High-angle annular dark field (HAADF) image of W/Cu interface (a), atomic fraction along the red arrow marked in Fig.3a (b), and elemental mappings of W (c) and Cu (d)

Color online

图4是W/Cu界面处的HRTEM像，其中可以观察到2个明暗衬度不同的区域，图中的插图I和II分别为暗、明区域的快速Fourier变换(FFT)谱。对插图I所示FFT谱的标定结果表明，图4中的较暗区域为[001]晶带轴的bcc晶体结构，说明该较暗区域为W。插图II是下半部分较亮区域的FFT谱，其标定结果是[01 $\bar{1}$ ]晶带轴的fcc晶体结构，说明该区域为Cu。在W和Cu的界面中间明显地形成了连续的晶格结构，表明W、Cu之间形成了冶金结合界面，实现了直接合金化，佐证了上文的成分扩散分析结果。同时可以看出，W的(110)晶面间距约为0.219 nm，Cu的(111)晶面间距约为0.212 nm。对照XRD标准卡片，平衡态下，W的(110)晶面间距为0.224 nm，Cu的(111)晶面间距为0.209 nm。由此可知，W/Cu界面处W的晶面间距有所减小，Cu的晶面间距有所增加。上述所得晶面间距变化规律与文献报道是一致的^[22,23]。

图4

图4 W/Cu界面HRTEM像

Fig.4 HRTEM image of W/Cu interface (Inset I shows the fast Fourier transform (FFT) pattern from the darker area and inset II shows the brighter one, d—interplanar spacing)

2.3 W/Cu层状复合材料的界面剪切强度及断口分析

图5显示了W/Cu层状复合材料剪切测试的应力-应变曲线，其中图5a是基于纳米多孔化制备的W/Cu层状复合材料的曲线，图5b是未经表面纳米化制备的W/Cu层状复合材料的曲线。纳米多孔化制备的W/Cu层状复合材料3个试样的剪切强度分别为98、103和102 MPa，平均剪切强度为101 MPa。而未经表面纳米化制备的W/Cu层状复合材料3个试样的剪切强度分别为79、96和86 MPa，平均剪切强度为87 MPa。经过表面纳米化后，W/Cu层状复合材料的剪切强度提升大约16%。结合图3所获结果可知，纳米化提高了W-Cu合金化扩散深度后，也提高了W-Cu合金化界面的结合强度。另外，图5也显示直接合金化获得的W/Cu界面具有一定的塑性。

图5

图5 基于纳米活性结构制备的W/Cu层状复合材料的剪切强度测试应力-应变曲线

Fig.5 Stress-strain curves of shear strength of W/Cu layered composites prepared based on nano active structure (a) and pretreated W block (b)

对W/Cu层状复合材料剪切断裂后W侧和Cu侧的剪切断口分别进行观察。图6a为W/Cu层状复合材料W侧剪切断口的SEM像。图6b为图6a中用矩形框I标记区域的EDS分析结果，可以看出W侧的剪切断口上残留了约30%的Cu。图6c和d分别为图6a用矩形框II和III标记区域的EDS分析结果，可以看到较小的II区和III区也含有Cu的成分，其中III区完全由Cu构成。这些都表明W侧断口有大面积Cu存在，说明W/Cu层状复合材料界面强度较高导致剪切断裂发生在Cu侧。

图6

图6 W/Cu层状复合材料的剪切断口形貌(W侧)的SEM像以及EDS分析结果

Fig.6 SEM image of shear fracture surface (W side) of the W/Cu layered composite (a); EDS analysis results of the region marked with rectangles I (b), II (c), and III (d)

图7a为W/Cu层状复合材料剪切断裂后Cu侧断口的SEM像，图7b显示了图7a中矩形框区域的EDS分析结果，可以看到标记区域内仅有Cu元素，没有W元素存在。基于上述分析结果可以再次推断W/Cu层状复合材料的剪切断裂发生在Cu一侧，也再次表明W/Cu形成了冶金结合，具有较好的界面结合强度。

图7

图7 W/Cu层状复合材料剪切断口形貌(Cu侧)的SEM像以及EDS结果

Fig.7 SEM image of shear fracture surface (Cu side) of the W/Cu layered composite (a), and EDS analysis result of the region marked with the rectangle in Fig.7a (b)

2.4 基于纳米活性结构的W-Cu直接合金化热力学机制

根据式(3)和(4)，对基于纳米活性结构的W-Cu直接合金化的热力学机制进行研究。在表面能的计算中，首先对未纳米多孔化的W表面、纳米多孔化的W表面以及Cu电沉积层进行了XRD测试，分析各个情况的晶面构成，结果如图8所示。采用JADE软件标定图8中各衍射峰对应的晶面指数并分析各个晶面所占的比例。结果表明：未纳米多孔化的W表面中主要有(110)、(200)、(211)和(200) 4个晶面，所占百分比分别为63.82%、9.13%、20.10%和6.96%；表面纳米多孔化后的W表面中主要含有(110)、(200)和(211) 3个晶面，所占百分比分别为1.88%、30.82%和67.30%；Cu镀层中包含(111)、(200)和(220) 3个晶面，所占百分比分别为71.17%、23.70%和5.13%。

图8

图8 纳米多孔化前后W板表面以及Cu电沉积层的XRD谱

Fig.8 XRD spectra of W without nanopores (a), nanoporous W (b), and Cu plating layer (c)

根据前文所述，对W-Cu直接合金化进行了有压力和无压力2种情况下W-Cu直接合金化热力学机制的计算，以探究压力能对W-Cu直接合金化的作用。由于在不互溶金属基层状复合材料制备过程中采用过106 MPa^[22,29]的压力，于是采用式(6)计算了106 MPa下E_p = 179.47 kJ/mol，没有压力下E_p = 0。

根据前文所述新的表面能计算思路，采用第一性原理并结合式(10)对W和Cu的表面能进行了计算。所采用的第一性原理计算软件为Materials Studio材料计算软件中的CASTEP密度泛函模块。计算时首先对W和Cu晶胞进行几何优化，然后在优化后的W和Cu结构上构建W和Cu表面及表面能真空层模型，真空层厚度经反复计算设置为1.5 nm，计算出W、Cu完整晶胞和表面能真空层模型的能量，最后采用式(10)计算出各晶面的表面能。计算中赝势设置为超软赝势，交换关联能项采用广义梯度近似项(GGA)，采用BFGS几何优化方法搜寻最低能量结构。结果见表1。

表1 基于第一性原理和新表面能公式计算的W和Cu表面所含各晶面的表面能(E_surf) (kJ·mol^-1)

Table 1 Surface energy (E_surf) of crystal faces contained on the surface of W and Cu based on the first-principles and new formula of surface energy

Metal	E_surf (110)	E_surf (111)	E_surf (200)	E_surf (211)	E_surf (220)
W	48.16	-	85.41	92.47	100.33
Cu	-	48.16	18.70	-	13.24

新窗口打开| 下载CSV

根据图8和表1所示结果，表面纳米多孔化之后，W中(200)和(211)晶面占比明显增多，也即具有高表面能的晶面含量增多。究其原因，可能是由于W金属的纳米化过程中电解液更容易沿着高表面能的晶面进行刻蚀造成的。

基于式(11)，按照各个晶面所占比例进行加权平均，计算纳米化前后W和Cu电沉积层的表面能，最终得到各材料表面能为E_{surf(plate,W)} = 64.09 kJ/mol，E_surf(nano,W) = 89.46 kJ/mol，E_{surf(coated,Cu)} = 15.36 kJ/mol。从该结果可以看出，表面纳米多孔化之后W的表面能相比于未纳米多孔化W明显增高，可以认为W中具有高表面能的晶面增多是W的表面能提高的主要原因。最后采用式(7)计算W-Cu直接合金化热力学模型(见式(3)和(4))中的表面能E_surf。

参考表2^[25,42]中的其他热力学参数，利用Miedema理论和Alonso计算方法计算W和Cu之间形成晶态相和非晶相的能量(式(3)和(4)中的ΔG_cryst、ΔG_amor和ΔG_int)，最终得到了基于W表面纳米多孔结构的W-Cu体系直接合金化的热力学能量曲线，包括了ΔG_alloying和E_initial等曲线，如图9所示。其中，图9a为没有压力的情况下直接合金化的能量曲线，图9b为有压力的情况下直接合金化的能量曲线。

表2 W-Cu合金化热力学计算所用的参数^[25,42]

Table 2 Parameters used in thermodynamic calculation for the W-Cu direct alloying^[25,42]

Element	V^2/3	$n_{w s}^{- 1 / 3}$	φ	γ	E	K	G	S_f	γ^S,0	ρ	M	T_m
	cm²	(d.u.)^1/3	V		GPa	GPa	GPa	10⁵ m²	mJ·m^-2	g·cm^-3	g·mol^-1	K
Cu	3.70	1.47	4.45	0.343	129.8	137.8	48.3	1.67	1825	8.96	63.55	1356
W	4.50	1.81	4.80	0.280	411.0	160.6	311.0	2.03	3675	19.32	183.84	3680

Note:V—molar volume of metal, n_ws—electron density at the boundary of the crystal cell, φ—electrical chemical potential of metal, γ—Poisson's ratio of metal, E—elastic modulus of metal, K—bulk modulus of metal, G—shear modulus of metal, S_f—surface area of 1 mol interfacial atoms, γ^S,0—specific surface energy of interfacial atoms, ρ—density of metal, M—molar mass of metal, T_m—melting point of metal

新窗口打开| 下载CSV

图9

图9 基于纳米活性结构的W-Cu直接合金化过程中的能量变化曲线

Fig.9 Calculated energy curves of W-Cu direct alloying processes based on nano active structure for the W-Cu systems (Δ $G_{a l l o y i n g}^{c}$ —Gibbs free energy for the formation of W/Cu crystalline phase, Δ $G_{a l l o y i n g}^{a}$ —Gibbs free energy for the formation of W/Cu amorphous phase, E_initial,W—initial energy of W-Cu system without nano-treatment, E_initial,nano—initial energy of W-Cu system with nano-treatment)

(a) without pressure

(b) under 106 MPa pressure

图9还表明，压力能可以大幅度提升初始能，但无论是有压力还是没有压力，合金化体系初始能量E_initial均高于ΔG_alloying，这说明不使用压力、以表面能为主的W-Cu的初始能量也能克服正的合金化能量，实现W-Cu直接合金化，由此本工作在W-Cu直接合金化中未使用压力(见1.1节所述合金化实验方法)，减少了工艺的复杂性。由图9可知，W-Cu直接合金化过程为：随着能量的降低，W-Cu界面上首先形成非晶相，然后非晶相转变成晶态相，这种晶态相应该为一种过饱和固溶体^[22]。W-Cu体系向非晶相或者晶态相转变的过程受储藏能和温度降低速率(即能量降低速率)的约束^[43,44]，储藏能大，温度降速快，非晶相可以更多地保留下来，反之只有晶态相保留下来。据此分析，由于本工作中W的储藏能已经去除，合金化过程又是随炉冷却、冷速较慢，W-Cu合金化后界面上应为晶态相，这一点与图4展示的晶态界面结构是吻合的，证明了所建热力学模型的可靠性。另外，从图9也能看出，经过表面纳米多孔化之后的E_initial曲线始终要高于未经表面纳米多孔化处理的E_initial曲线，这表明W纳米多孔化之后W-Cu金属体系更容易实现直接合金化。未加压力时，由于E_p和E_stor均为0，E_initial完全由表面能E_surf提供，此时经过纳米多孔化后的表面能是W-Cu直接合金化的热力学驱动力。

2.5 纳米结构形状对W-Cu直接合金化热力学的影响

本研究认为，纳米结构形状也有可能会影响W的表面能，进而影响到W-Cu的直接合金化。目前，这方面的研究报道较少，本工作对此进行了探讨。图1显示了W表面纳米多孔活性层表面和截面的形貌特征。可以看出，W纳米多孔层的形状非常复杂。为了定量化探究纳米结构形状对W-Cu直接合金化的影响，设定W纳米多孔层近似等效成圆锥体和圆柱体的组合模型，该模型有2种存在方式，一种是尖端凸出的结构，圆锥体对应截面中的尖端区域，即图10a和b中虚线内的部分，圆柱体对应凸出结构的下端部分；另一种是凹陷的结构，如图10b所示，其中实线圆表示纳米孔的位置。通过对较大纳米孔(见图10c)的观察，发现孔洞内会存在衬度更暗的小孔洞，故也选用圆锥体作为孔洞底端的近似等效结构。图10d和e分别为这2种情况下模型的结构示意图和剖面图。其中，圆柱体高为h₀，底面半径为R₁，圆锥体高为h₁，d_surf为金属中能提供表面能的表面层厚度，此处表面层厚度取2个晶格常数^[24]。采用式(14)和(15)分别计算这2种模型结构与W平板模型(见图10f所示，图中a为平板长度和宽度尺寸，表面层厚度d_surf同样取为2个晶格常数)的表面能并进行比较，以此获得纳米结构形状对W表面能的影响，进而获得纳米结构形状对W-Cu直接合金化热力学的影响。

图10

图10 W纳米多孔结构及W平板的简化模型

Fig.10 Origin shape of W nanoporous with convex structure (a), concave structure (b, c), structural diagram and cross-section of simplified geometric models of Figs.10a and c (d, e), and geometry model for the W plate without nano-treatment (f) (D—nanopore diameter, d_surf—the thickness of the surface layer in a metal that provides surface energy, h₀—the height of a cylinder in the convex or concave structure, h₁—the height of a cone in the convex or concave structure, h₂—obtained from h₁ ± d_surf, R₁—the bottom radius of cylinder and cone in the convex or concave structure, R₂—obtained from R₁ ± d_surf, a—the side length of the model of W plate without nano-treatment)

根据上述模型，对于圆锥体和圆柱体组成的尖端凸出结构，W的表面积S_nano,W和表面层体积Δ $V_{n a n o, W}^{'}$ 为：

S_{n a n o, W} = 2 π R_{1} h_{0} + π R_{1} \sqrt[]{R_{1}^{2} + h_{1}^{2}}

(14)

Δ V_{n a n o, W}^{'} = (π R_{1}^{2} h_{0} + \frac{π}{3} R_{1}^{2} h_{1}) - (π R_{2}^{2} h_{0} + \frac{π}{3} R_{2}^{2} h_{2})

(15)

式中，π $R_{1}^{2}$ h₀和 $\frac{π}{3} R_{1}^{2}$ h₁分别为圆柱和圆锥的体积，R₂ = R₁ - d_surf，h₂ = h₁ - d_surf。根据图2和10，取h₀ = 280 nm，h₁ = 40 nm，R₁ = 15 nm。

对于圆锥体和圆柱体组成的凹陷结构，S_nano,W和Δ $V_{n a n o, W}^{'}$ 为：

S_{n a n o, W} = 2 π R_{1} h_{0} + π R_{1} \sqrt[]{R_{1}^{2} + h_{1}^{2}}

(16)

Δ V_{n a n o, W}^{'} = (π R_{2}^{2} h_{0} + \frac{π}{3} R_{2}^{2} h_{2}) - (π R_{1}^{2} h_{0} + \frac{π}{3} R_{1}^{2} h_{1})

(17)

式中，π $R_{1}^{2}$ h₀和 $\frac{π}{3} R_{1}^{2}$ h₁分别为圆柱和半圆球的体积，R₂ = R₁ + d_surf，h₂ = h₁ + d_surf，取h₀=280 nm，h₁ = 40 nm，R₁ = 30 nm。

未纳米多孔化的平板W的表面积S_plate,W和表面层体积Δ $V_{p l a t, W}^{'}$ 为：

S_{p l a t e, W} = a^{2}

(18)

Δ V_{p l a t e, W}^{'} = S_{p l a t e, W} \cdot d_{s u r f} = a^{2} \cdot d_{s u r f}

(19)

根据式(14)~(19)计算出各情况下的表面积和表面层体积，再代入式(8)计算表面能(计算所用比表面能 $γ_{i}^{S, 0}$ 的参数见表2^[25,42])，得到未纳米多孔化的W板的表面能E_{surf(plate,W)} = 55.24 kJ/mol，纳米多孔化结构的圆锥体和圆柱体组合尖端凸出部分等效模型的表面能E_surf(nano,W) = 57.52 kJ/mol，等效凹陷模型的表面能E_surf(nano,W) = 60.44 kJ/mol。计算结果表明，当W表面纳米多孔活性结构中存在更多的凸尖端出部分时，纳米多孔W的表面能相比于平板W的表面能仅有较小的增加(约4%)，而当W表面纳米多孔活性结构中存在更多的凹陷结构时，纳米多孔W的表面能相比平板W的表面能有较多的增加(约9.5%)。虽然前文提到式(8)由于表面层厚度取值不同会造成表面能结果具有随意性，但可以在采用相同表面层厚度下进行比较。由此可以认为，W表面纳米多孔化之后的结构形状对W的表面能E_surf也有一定的提升作用，进而促进W-Cu直接合金化的进行。但纳米结构形状对W金属表面能影响的计算目前还不能实现各种形状的统一准确定量，所以本工作并未将其加入到式(3)和(4)所示的W-Cu直接合金化热力学模型中来计算合金化过程中的能量变化曲线。

2.6 合金化过程中表面能释放的可能途径

由分子动力学模拟和实验验证结果^[36]可知，W-Cu直接合金化的原子扩散机制为：在一定温度下，Cu表面变得原子松散，出现非晶无序结构，这使得Cu原子可以挣脱Cu基体的束缚，直接挤入W金属中，W金属表面结构也随即变得开始松散混乱，进一步促进Cu原子的扩散进入，最终实现了合金化。模拟结果和实验结果都表明W-Cu直接合金化有一个临界温度范围，这个温度范围模拟结果为(0.67~0.96)T_m,Cu (T_m,Cu为Cu的熔点)^[36]，实验结果为(0.84~0.97)T_m,Cu^[22]。

图11为W和Cu在627、727、827、927和1027℃下直接合金化时W/Cu界面及其附近的原子结构精细密度分布函数。可知，当合金化温度升高到627℃，界面处Cu的ρ(Z)曲线周期性开始逐渐减小，在927~1027℃之间包括界面在内的整体都出现了周期性消失；W的ρ(Z)曲线在Cu周期性减弱后也出现周期性减小，到后期周期性减弱明显，但W的ρ(Z)曲线周期性减弱自始至终仅限于界面处。这一结果表明，随着合金化温度的升高，Cu在界面处及其附近的结构逐渐变得无序化，界面处W结构在Cu无序化后也呈现出了无序化现象。该分析结果与上文所获合金化原子扩散机制相符。由于金属表面由周期结构变得散乱、出现无序结构时，表面能可以得到释放。因此可认为，W-Cu合金化过程中W和Cu的表面能有可能是在Cu和W结构出现混乱时得到释放，而参与到直接合金化中去。Cu表面能首先释放，随后W表面能得到释放。实验也证明了合金化温度在927~1027℃(如980℃^[23])效果最好。

图11

图11 分子动力学模拟所获W和Cu在不同温度下直接合金化时W/Cu界面及其附近的精细密度分布函数

Fig.11 Fine-scale density profiles (ρ(Z)) for W and Cu near and at the W/Cu interface constructed through direct alloying at different temperatures

本研究结果可以为核聚变堆用W/Cu PFCs的制备新方法研究提供参考。同时，不互溶金属体系在包括电池能源等在内的其他领域也有着广泛的应用，可为这些领域涉及的不互溶金属合金化问题研究提供参考。例如Cu-Li是一个比较典型的不互溶体系，由于热力学平衡态下2者之间不发生合金反应，Cu被大规模用于各类锂电池中作为负极集流材料，以保证在电池循环过程集流过程中不会发生电化学锂化形成合金相。但有研究^[45]在原位TEM下观察到在纳米尺寸效应作用下，Cu和Li将发生合金化，即超小Cu纳米颗粒发生电化学锂化和CuLi _x 非晶合金相形成将成为可能。并认为出现这种情况的原因在于Cu纳米颗粒导致的表面能显著增加所带来的反应活性极大增强，但并未给出热力学计算分析。而本工作所建立的基于纳米活性结构的不互溶金属合金化热力学模型可以解释Cu-Li发生合金化的原因，区别在于Cu-Li的合金化是通过电化学驱动原子运动释放表面能，而本工作中W-Cu是通过加热作用驱动原子运动释放表面能。

3 结论

(1) 通过两步阳极氧化和H₂还原退火可以在W金属表面制备出纯W的纳米多孔层，该纳米多孔层具有较高的活性，可以在近熔点温度高温退火时诱发和促进W与Cu之间发生直接合金化。显微结构表征和界面剪切强度测试实验表明，基于纳米化的W/Cu合金化扩散深度和界面剪切强度都要高于未纳米化的W/Cu界面。

(2) 将前期工作建立的W-Cu直接合金化热力学模型应用于基于纳米活性结构的W-Cu直接合金化方法热力学机制的解释，提出了压力能和表面能的新计算方法，获得了单位为kJ/mol的压力能和表面能，从而实现了基于W表面纳米多孔活性结构的W-Cu直接合金化热力学模型计算，证明了基于W表面纳米多孔活性结构实现W-Cu直接合金化并制备出W/Cu层状复合材料在热力学上是可行的。

(3) 表面纳米多孔化后W金属的表面能得到了较大的提升，可以作为W、Cu之间直接合金化的热力学驱动力。W表面纳米多孔活性结构提高表面能的原因主要有2个：一是W金属表面具有高表面能的晶面增多；二是纳米多孔化后的结构形状。

参考文献

原文顺序

文献年度倒序

文中引用次数倒序

被引期刊影响因子

[1]

Tsunematsu

, Nagami

Status of the ITER project

[J]. J. Plasma Fusion Res., 2002, 5: 137

[2]

Philipps

Tungsten as material for plasma-facing components in fusion devices

[J]. J. Nucl. Mater., 2011, 415: S2

[3]

J S

, Zuo

G Z

, Wang

, et al.

Brief review of the interactions between plasma and walls in magnetic controlled fusion devices

[J]. J. Univ. Sci. Technol. China, 2020, 50: 1193

胡建生, 左桂忠, 王亮

等.

磁约束核聚变装置等离子体与壁相互作用研究简述

[J]. 中国科学技术大学学报, 2020, 50: 1193

[4]

Kaufmann

, Neu

Tungsten as first wall material in fusion devices

[J]. Fusion Eng. Des., 2007, 82: 521

DOI URL

[5]

Hirai

, Ezato

, Majerus

ITER relevant high heat flux testing on plasma facing surfaces

[J]. Mater. Trans., 2005, 46: 412

[6]

Bolt

, Barabash

, Krauss

, et al.

Materials for the plasma-facing components of fusion reactors

[J]. J. Nucl. Mater., 2004, 329-333: 66

[7]

Pintsuk

, Smid

, Döring

J E

, et al.

Fabrication and characterization of vacuum plasma sprayed W/Cu-composites for extreme thermal conditions

[J]. J. Mater. Sci., 2007, 42: 30

[8]

Mitteau

, Missiaen

J M

, Brustolin

, et al.

Recent developments toward the use of tungsten as armour material in plasma facing components

[J]. Fusion Eng. Des., 2007, 82: 1700

[9]

Batra

I S

, Kale

G B

, Saha

T K

, et al.

Diffusion bonding of a Cu-Cr-Zr alloy to stainless steel and tungsten using nickel as an interlayer

[J]. Mater. Sci. Eng., 2004, A369: 119

[本文引用: 2]

[10]

Wang

C B

, Shen

, Zhou

Z G

, et al.

Diffusion welding of 93W alloy to OFC and structural control of 93W/OFC joint

[J]. J. Mater. Sci., 2005, 40: 2105

[11]

Tokitani

, Hamaji

, Hiraoka

, et al.

Deformation and fracture behavior of the W/ODS-Cu joint fabricated by the advanced brazing technique

[J]. Fusion Eng. Des., 2019, 146: 1733

DOI [本文引用: 2]

In our previous work, the joint between oxide dispersion strengthened copper alloy (ODS-Cu), GlidCop (R) (Cu-0.3 wt%Al2O3) and tungsten (W) demonstrated superior fracture strength (similar to 200 MPa). This joint was fabricated by the direct brazing method between W and ODS-Cu using BNi-6 (Ni-11%P) filler material without any intermediate layer. This method was named as the improved or the advanced brazing technique. In the present study, deformation and fracture behavior of the joint after the three-point bending test was investigated. At first, it was found that the crack initiation points were dominantly in the W bulk, although it was not clear that the crack initiated from grain boundary or not. Secondly, the crack propagation proceeded mostly in the W bulk but tended to deflect towards the bonding layer. These results interpret the strength of the bonding interface is superior and the present bonding technique is applicable for severe environments such as a high heat flux divertor component on the fusion reactor. Based on the above physical and technological understanding, we successfully fabricated the large scale divertor mock-up which has twenty-eight plates of W with each size of 20 x 20 x 5 mm(3).

[12]

Peng

S X

, Mao

Y W

, Min

, et al.

Joining of tungsten to CuCrZr alloy with Cu-TiH₂-Ni filler and Cu interlayer

[J]. Int. J. Refract. Met. Hard Mater., 2019, 79: 31

[13]

Bang

, Choi

, Kim

H C

, et al.

Manufacturing and testing of flat type W/Cu/CuCrZr mock-ups by HIP process with PVD coating

[J]. Fusion Eng. Des., 2019, 146: 603

[14]

Mou

N Y

, Han

, Yao

D M

, et al.

Manufacturing and high heat flux testing of flat-type W/Cu/CuCrZr mock-up by HIP assisted brazing process

[J]. Fusion Eng. Des., 2021, 169: 112670

[15]

Niu

Y R

, Lu

, Huang

L P

, et al.

Comparison of W-Cu composite coatings fabricated by atmospheric and vacuum plasma spray processes

[J]. Vacuum, 2015, 117: 98

[16]

Zhou

Z J

, Guo

S Q

, Song

S X

, et al.

The development and prospect of fabrication of W based plasma facing component by atmospheric plasma spraying

[J]. Fusion Eng. Des., 2011, 86: 1625

[17]

Song

J P

, Yu

, Zhuang

Z G

, et al.

Preparation of W-Cu functionally graded material coated with CVD-W for plasma-facing components

[J]. J. Nucl. Mater., 2013, 442: S208

[18]

Dai

, Wu

M L

, Shu

S C

, et al.

Thermal CVD growth of graphene on copper particles targeting tungsten-copper composites with superior wear and arc ablation resistance properties

[J]. Diam. Relat. Mater., 2020, 104: 107765

[19]

Ibrahim

, Abdallah

, Mostafa

S F

, et al.

An experimental investigation on the W-Cu composites

[J]. Mater. Des., 2009, 30: 1398

[20]

Perez-Soriano

E M

, Arévalo

, Montealegre-Meléndez

, et al.

Influence of starting powders on the final properties of W-Cu alloys manufactured through rapid sinter pressing technique

[J]. Powder Metall., 2021, 64: 75

[21]

Saito

, Fukaya

, Ishiyama

, et al.

Mechanical properties of HIP bonded W and Cu-alloys joint for plasma facing components

[J]. J. Nucl. Mater., 2002, 307-311: 1542

[22]

Zhang

, Huang

, Wang

Z M

, et al.

Thermodynamic mechanism for direct alloying of immiscible tungsten and copper at a critical temperature range

[J]. J. Alloys Compd., 2019, 774: 939

DOI URL [本文引用: 10]

[23]

Zhang

, Huang

, Liu

Y C

, et al.

Direct diffusion bonding of immiscible tungsten and copper at temperature close to copper's melting point

[J]. Mater. Des., 2018, 137: 473

DOI URL [本文引用: 7]

[24]

J L

, Li

, Wang

Z M

, et al.

Direct alloying of immiscible molybdenum-silver system and its thermodynamic mechanism

[J]. J. Mater. Sci. Technol., 2021, 65: 18

DOI [本文引用: 2]

Direct alloying is difficult to be realized in an immiscible Mo-Ag system with a positive formation heat due to the absence of thermodynamic driving force at equilibrium. In this work, a direct alloying method is developed to realize the direct alloying between Mo and Ag and construct Mo-Ag interface. The direct alloying method was mainly carried out through a direct diffusion bonding for Mo and Ag rods at a temperature close to the melting point of Ag (TmAg). Then the microstructure and phase constitution of the as-constructed Mo-Ag interface are characterized. The results show that Mo-Ag metallurgical bonding interface has been constructed successfully, indicating that a direct alloying in the immiscible Mo-Ag system has been realized. Additionally, mechanical tests are carried out for the Mo-Ag joints prepared through the direct alloying method. The test results show that the average maximum tensile strength of the joints is about 107 MPa. The effect of alloying parameters on the tensile strength is also discussed, which shows that there is an effective temperature range for the direct alloying between Mo and Ag. Lastly, an improved thermodynamic model that considers the formation of Mo-Ag crystalline and amorphous phase is presented to reveal the thermodynamic mechanism of the direct alloying. Combining the calculation and differential scanning calorimetry (DSC) tests results, the Gibbs energy diagram for the direct alloying is obtained. It is confirmed that the co-release of storage energy and surface energy can serve as the thermodynamic driving force to overcome the effect of positive formation heat and lead to direct alloying for Mo-Ag systems.

[25]

Pan

X C

, Zhang

, Huang

, et al.

Construction of metallurgical interface with high strength between immiscible Cu and Nb by direct bonding method

[J]. J. Alloys Compd., 2017, 723: 1053

DOI URL [本文引用: 5]

[26]

Zhang

, Huang

, Wang

Z M

, et al.

Preparation of a nanoporous active tungsten foil by two-step anodizing and deoxidized annealing for hydrogen evolution reaction

[J]. Nanotechnology, 2019, 30: 015603

[27]

, Chen

Y Y

, Chen

, et al.

The improvement of bonding strength of W/Cu joints via nano-treatment of the W surface

[J]. Metals, 2021, 11: 844

W/Cu joining is key for the fabrication of plasma-facing compounds of fusion reactors. In this work, W and Cu are joined through three steps: (1) hydrothermal treatment and reduction annealing (i.e., nano-treatment), (2) Cu plating and annealing in a pure H2 atmosphere, and (3) W/Cu bonding at 980 °C for 3 h. After nano-treatment, nanosheets structure can be found on the W substrate surface. The tensile strength of the W/Cu joint prepared via nano-treatment reaches as high as approximately 93 MPa, which increases by about 60% compared with the one without nano-treatment. The microhardness curves exhibited continuous variations along the W/Cu interface. The TEM images show that the W/Cu interface is compact without any cracks or voids. This work may also be applied for enhancing bonding strength in other immiscible materials.

[28]

Zhao

, Li

, Chen

Y Y

, et al.

Joining of oxygen-free high-conductivity Cu to CuCrZr by direct diffusion bonding without using an interlayer at low temperature

[J]. Fusion Eng. Des., 2020, 151: 111400

[29]

Huang

, Du

J L

, Wang

Z M

Progress in research on the alloying of binary immiscible metals

[J]. Acta Metall. Sin., 2020, 56: 801

DOI [本文引用: 3]

Materials based on binary immiscible metal systems are widely used in aerospace, nuclear fusion engineering, electronic packaging, anti-armor weapons and other fields. However, due to the positive formation heat and the large differences in the properties of the component, the direct alloying of binary immiscible metals and the preparation of the corresponding materials are very difficult. Varieties of methods have been developed for direct alloying of binary immiscible metals at home and abroad, and the thermodynamic and diffusion mechanism of these methods have been studied. In this review, firstly the principle and thermodynamic mechanism of mechanical alloying, physical vapor deposition and ion beam mixing, as well as their applications in binary immiscible metal powder alloys and nano-multilayer films are reviewed. Then the irradiation damage alloying (IDA) and high-temperature structure induced alloying (HTSIA) methods that are proposed and developed by our group are introduced. Besides, the principle, interfacial microstructure, thermodynamic mechanism, diffusion mechanism and application of these two methods were described in detail. Finally, the development trend of the research on alloying of binary immiscible metals is proposed.

黄远, 杜金龙, 王祖敏.

二元互不固溶金属合金化的研究进展

[J]. 金属学报, 2020, 56: 801

[本文引用: 3]

[30]

Y G

. Physical Chemistry[M]. Shanghai: Fudan University Press, 2013: 20

李元高

. 物理化学[M]. 上海: 复旦大学出版社, 2013: 20

[31]

Pan

C H

Development of surface analytical instruments abroad

[J]. Anal. Instrum., 1980, (suppl.1) : 31

潘承璜

国外表面分析仪器的进展

[J]. 分析仪器, 1980, (): 31

[32]

Chen

X Y

, Zhang

P F

, Liu

Y C

, et al.

Nanoconical active structures prepared by anodization and deoxidation of molybdenum foil and their activity origin

[J]. J. Alloys Compd., 2021, 851: 156896

DOI URL [本文引用: 2]

[33]

Fan

. X-Ray Metallography[M]. Beijing: China Machine Press, 1981: 16; 23

[本文引用: 3]

范雄

. X射线金属学[M]. 北京: 机械工业出版社, 1981: 16; 23

[本文引用: 3]

[34]

Mai

Z H

. X-Ray Characterization of Thin Film Structure[M]. 2nd Ed., Beijing: Science Press, 2015: 20

麦振洪

. 薄膜结构X射线表征[M]. 第2版. 北京: 科学出版社, 2015: 20

[35]

Han

, Jeurgens

L P H

, Cancellieri

, et al.

Anomalous texture development induced by grain yielding anisotropy in Ni and Ni-Mo alloys

[J]. Acta Mater., 2020, 200: 857

[36]

Zeng

, Li

, Huang

Construction of an n-body potential for revealing the atomic mechanism for direct alloying of immiscible tungsten and copper

[J]. Materials, 2021, 14: 5988

DOI URL [本文引用: 3]

W-Cu laminated composites are critical materials used to construct nuclear fusion reactors, and it is very important to obtain direct alloying between W and Cu at the W/Cu interfaces of the composites. Our previous experimental studies showed that it is possible to overcome the immiscibility between W and Cu and obtain direct alloying when the alloying temperature is close to the melting point of Cu. Because the W-Cu interatomic potentials published thus far cannot accurately reproduce the alloying behaviors of immiscible W and Cu, an interatomic potential suitable for the W-Cu system has been constructed in the present study. Based on this potential, direct alloying between W and Cu at high temperature has been verified, and the corresponding diffusion mechanism has been studied, through molecular dynamics (MD) simulations. The results indicate that the formation of an amorphous Cu layer at the W/Cu interface plays a critical role in alloying because it allows Cu atoms to diffuse into W. The simulation results for direct alloying between W and Cu can be verified by experimental results and transmission electron microscopy observations. This indicates that the constructed W-Cu potential can correctly model the high-temperature performance of the W-Cu system and the diffusion mechanism of direct alloying between W and Cu.

[37]

Gan

X L

, Xiao

S F

, Deng

H Q

, et al.

Atomistic simulations of the Fe(001)-Li solid-liquid interface

[J]. Fusion Eng. Des., 2014, 89: 2894

[38]

Yang

, Olmsted

D L

, Asta

, et al.

Atomistic characterization of the chemically heterogeneous Al-Pb solid-liquid interface

[J]. Acta Mater., 2012, 60: 4960

[39]

, Meng

X C

, Sun

X G

, et al.

Atomic scale analysis of the corrosion characteristics of Cu-Li solid-liquid interfaces

[J]. J. Alloys Compd., 2018, 763: 1

[40]

Luo

M Z

, Liang

, Lang

, et al.

Molecular dynamics simulations of the characteristics of Mo/Ti interfaces

[J]. Comput. Mater. Sci., 2018, 141: 293

[41]

Laursen

A B

, Kegnæs

, Dahl

, et al.

Molybdenum sulfides-efficient and viable materials for electro- and photoelectrocatalytic hydrogen evolution

[J]. Energy Environ. Sci., 2012, 5: 5577

[42]

Liu

B X

, Lai

W S

, Zhang

Z J

Solid-state crystal-to-amorphous transition in metal-metal multilayers and its thermodynamic and atomistic modelling

[J]. Adv. Phys., 2001, 50: 367

DOI URL [本文引用: 4]

[43]

J L

, Huang

, Xiao

, et al.

Building metallurgical bonding interfaces in an immiscible Mo/Cu system by irradiation damage alloying (IDA)

[J]. J. Mater. Sci. Technol., 2018, 34: 689

DOI [本文引用: 1]

For the immiscible Mo/Cu system with a positive heat of mixing (ΔHm > 0), building metallurgical bonding interfaces directly between immiscible Mo and Cu and preparing Mo/Cu laminar metal matrix composites (LMMCs) are very difficult. To solve the problem, a new alloying method for immiscible systems, which is named as irradiation damage alloying (IDA), is presented in this paper. The IDA primarily consists of three steps. Firstly, Mo is damaged by irradiation with multi-energy (186, 62 keV) Cu ion beams at a dose of 2 × 1017 ions/cm2. Secondly, Cu layers are superimposed on the surfaces of the irradiation-damaged Mo to obtain Mo/Cu laminated specimens. Thirdly, the irradiation damage induces the diffusion alloying between Mo and Cu when the laminated specimens are annealed at 950 °C in a protective atmosphere. Through IDA, Mo/Cu LMMCs are prepared in this paper. The tensile tests carried out for the Mo/Cu LMMCs specimens show that the Mo/Cu interfaces constructed via IDA have high normal and shear strengths. Additionally, the microstructure of the Mo/Cu interface is characterized by High Resolution Transmission Electron Microscopy (HRTEM), X-ray diffraction (XRD) and Energy Dispersive X-ray (EDX) attached in HRTEM. The microscopic characterization results show that the expectant diffusion between Mo and Cu occurs through the irradiation damage during the process of IDA. Thus a Mo/Cu metallurgical bonding interface successfully forms. Moreover, the microscopic test results show that the Mo/Cu metallurgical interface is mainly constituted of crystalline phases with twisted and tangled lattices, and amorphous phase is not observed. Finally, based on the positron annihilation spectroscopy (PAS) and HRTEM results, the diffusion mechanism of IDA is discussed and determined to be vacancy assisted diffusion.

[44]

J L

, Huang

, Liu

J W

, et al.

Irradiation damage alloying for immiscible alloy systems and its thermodynamic origin

[J]. Mater. Des., 2019, 170: 107699

[45]

Sun

M H

, Wei

J K

, Xu

, et al.

Electrochemical solid-state amorphization in the immiscible Cu-Li system

[J]. Sci. Bull., 2018, 63: 1208