室温和高应变率下的超塑性纳米晶陶瓷外文翻译资料
2022-12-11 20:12:11
Superplastic nanocrystalline ceramics at room temperature and high strain rates
Tensile-loading molecular dynamics simulations show that nanocrystalline SiC not only becomes ductile, but can be superplas-tically deformed at room temperature when grain sizes are reduced to d ~ 2 nm. The calculated strain rate sensitivity, 0.67, implies a superplastic ceramic able to attain strains of up to 1000% at room temperature and typical strain rates (~10-2s-1). The origin of the superplasticity is linked to an unusually steep rise in creep rate to 10 6s s-1 for d = 2 nm. The results explain recent observations in SiC nanowires and suggest novel opportunities for structural ceramics.
An ideal structural material should be strong,ductile and formable. High strength is desirable in order for the material to support high loads, while ductility and formability make it tolerant to structural defects,resistant to catastrophic failure, and allow the design of complex shape applications. Unfortunately, materials with very high strength, such as ceramics, are usually brittle, failing without significant plastic deformation.The task of designing ceramics that can undergo large plastic deformation at room temperature while main-taining high strength has been a major challenge to material scientists for the last few decades. A potential way to improve the deformability of ceramics has been to reduce the scale of their microstructure into the nanometer regime. While earlier studies report enhancements , the key challenge of achieving sufficient ductility at industrially attractive strain rates and temperatures remains unmet.
We focus on silicon carbide (SiC), a ceramic with applications ranging from gas turbines [7] to brake disks and armor. Earlier reports suggest that the mechanical response of nanocrystalline SiC (nc-SiC) depends critically upon grain size. Our tensile-loading molecular dynamics (MD) simulations, performed at room temperature and high strain rates, show a drastic change in the mechanical response of nc-SiC, leading to superplasticity, upon grain size reduction from 6 to 2 nm. We simulate fully dense SiC with d = 2, 4, and 6 nm, with randomly oriented grains and grain boundaries (GB), see Supplementary Data. Illustrations of the nanostructures are shown in Fig. 1. To ensure the results are not dependent on the interatomic potential chosen, we perform simulations using different SiC potentials: Tersoff-1, Tersoff-2, Tersoff-3, and the Vashishta potential fitted for SiC.
The stress–strain curves for nc-SiC with different grain sizes simulated and with all the interatomic potentials are shown in Figure 1. As a reference, the results for amorphous SiC (a-SiC) and lt;100gt;-oriented crystalline SiC (C-SiC) are also shown. All stress–strain curves show a decay in the ultimate strength upon grain size reduction from 6 to 2 nm from 5% (Vashishta potential) to 27% (Tersff-3). The strain to failure increases with decreasing grain size for all potentials. For d = 2 nm,the strain to failure is gt;250% for Tersoff-1 and Tersoff-2, and 70% for the Tersoff-3 and Vashishta potentials. For all potentials, the simulations show a sharp several-fold rise in the strain to failure, indicative of a brittle to ductile transition, in the 6–2 nm range. Ford = 2 nm, the transverse (contractile) strain is a large fraction of the tensile strain (up to 40%); this is characteristic of superplastic materials.
To understand the failure behavior change shown in Fig. 1, we analyzed the free volume evolution using the Connolly method [16]. Figure 2 shows the atomic structures for d = 2, 4 and 6 nm at different strains between the yield and failure points. The simulation with d = 6 nm show the formation of many voids that rapidly coalesce to form a macrocrack. Immediately prior to failure a classical (nearly) smooth crack forms with few voids elsewhere in the system (Fig. 2(c)). On the other hand, the simulations for d = 2 nm show continual nucleation of new cavities. Eventually, some of these cavities coalesce into a failure-producing defect that spans the system. This difference in behavior leads to very large strains to failure (~50%, see Fig. 2(i)) in the 2 nm sample as compared with the relatively small strains to failure in the 4 nm (~18%, see Fig. 2(f)) and 6 nm (13%, see Fig. 2(c)) samples. While all samples eventually fail via a void coalescence process, the difference in failure mode between the 4 nm/6 nm samples and the 2 nm sample appears as brittle vs. ductile failure behavior.
The change observed in mechanical behavior can be understood considering the deformation and failure in terms of traditional superplasticity studies, i.e. by characterizing the plastic behavior under constant stress or constant strain rate . This is commonly done by determining the strain rate as a function of applied stress and fitting the data to the functional form εprop;sigma;n, where ε is the strain rate, sigma; is the constant applied stress and n is an exponent characteristic of the material. Alternatively, one can deform a material at constant strain rate and measure the flow stress as a function of time and determine the relationship sigma;prop; em, where m is the strainrate sensitivity. The approaches are equivalent, since m = 1/n. Classically, a material is characterized as being superplastic if m≧0.5 .
To evaluate n, we perform a series of room temperature simulations of tensile deformation of nc-SiC(d = 2 nm) at constant stresses from 1 to 5 GPa using the Vashishta potentials [15]. As shown in Figure 1(g),the stresses are all below the yield stress, which is ~ 6 GPa. Figure 3(a) shows the strain as a function of simulation time for each value of applied tensile stress.The creep rates are determined from the
剩余内容已隐藏,支付完成后下载完整资料
室温和高应变率下的超塑性纳米晶陶瓷
拉伸分子动力学模拟的结果表明,纳米SiC不仅变得有韧性,而且可以在室温下进行变形过程分析,当晶粒尺寸减小到2纳米时。计算出的应变率敏感性,0.67,意味着一种超塑性陶瓷应变率能够达到室温和典型的应变率的1000%。超塑性的起源与在d=2nm的时候蠕变率异常陡峭上升到106 s- 有关.这个结果解释最近对SiC纳米线的观测并且对于结构陶瓷提出了新的机会。
理想的结构材料应坚固、韧性和可成型。高强度是可取的,以支持高负荷的材料,而延展性和成形性,使其容忍结构缺陷,对灾难性故障有抵抗力,并允许复杂形状的应用程序的设计。不幸的是,高强度的材料,如陶瓷,通常是脆性的,没有明显的塑性变形。设计可以在室温下经受较大的塑性变形的陶瓷的任务,在过去的几十年里,维护高强度已经成了材料科学家面临的主要挑战任务。一个潜在的方法来改善陶瓷材料的变形能力是减少他们微观结构的规模达到纳米级。通过对早期的研究报告的改进,这个挑战的关键是在工业上有吸引力的应变率取得足够的延展性和温度任然未得到满足。
我们专注于碳化硅(SiC),一种应用范围包括燃气轮机制动盘和装甲的陶瓷。此前的报告表明,纳米SiC(NC SiC)的力学响应关键取决于晶粒尺寸。我们的拉伸分子动力学模拟(MD),在室温和高应变率下进行,纳米SiC显示出急剧变化的力学响应,出现超塑性,在从6到2纳米的晶粒尺寸减少。我们模拟完全致密的SiC与D = 2,4,和6纳米,与随机取向的晶粒和晶界(晶界),见补充数据。纳米结构的插图如图1所示。为了确保结果不依赖于所选择的原子间相互作用势,我们进行模拟,使用不同SiC电位:tersoff- 1,Tersoff- 2,Tersoff- 3和Vashishta用于SiC的电位。
不同晶粒尺寸的模拟SiC应力-应变曲线和所有的原子间相互作用势,如图1所示。作为参考,结果非晶碳化硅(SiC)和lt;1 0 0gt;取向结晶碳化硅(SiC)都显示。所有的应力应变曲线显示在晶粒尺寸从6到2 nm减少时极限强度的衰减从5%(Vashishta势)到27%(tersoff- 3)。破坏应变的增加导致晶粒尺寸的所有电势减少。D = 2 nm,破坏应变gt;250% 为tersoff- 1和 tersoff- 2,为tersoff- 3和Vashishta电位70%。对于所有的电位,模拟显示一个急剧的几倍上升的应变失效,表示脆性-韧性过渡,在6 - 2纳米范围内。d= 2纳米,横向(收缩)应变是拉伸应变(高达40%)的一大部分,这是超塑性材料的特性。
为了理解图1所示的失效行为变化,我们用康纳利方法分析了自由体积的演化。图2显示了原子结构的D = 2,4和6 nm的不同应变之间的屈服点和故障点。D = 6 nm的模拟显示有许多空隙,迅速联合起来形成宏观裂纹。在失效之前,一个经典的(几乎)光滑的裂缝在系统的其他地方形成空洞(图2(c))。另一方面,D = 2 nm的模拟显示连续成核的新的空腔。最终,这些空洞合并失败产生缺陷,跨越系统。这种行为上的差异导致非常大的应变失效,应变50%的2 nm的样品应变失效相当于的18%的4纳米和13%的6纳米样品。虽然所有样品最终失效经过一个空洞的聚结过程中,但4 nm / 6 nm的样品和在2 nm的样品之间的故障模式的差异体现了脆性与韧性性失效行为。
机械行为的变化是可以被考虑理解为在传统的超塑性研究方面的变形和失效,即恒定应力和恒定应变速率下表征塑性行为。这通常是确定应变速率作为施加应力的函数并将数据拟合到函数形式εprop;sigma;n功能通常所做的那样,在ε是应变率,sigma;是不恒定应力和n是物质指数特性。另外,你可以在恒应变率和变形材料的流动应力测量作为时间的函数关系,确定sigma;prop; em,其中m是应变速率敏感性。该方法是等效的,因为M = 1 / N.一般来说,如果M 大于等于0.5,材料的特点是具有超塑性。
为了估计n的值,我们进行了一系列的数控拉伸变形室温模拟,对SiC(D = 2 nm)在恒定的应力从1至5 GPA使用Vashishta电位。如图1所示(G)、应力均小于屈服应力,这是6美元的GPA。图3(a)说明作为施加拉伸应力的每个值的模拟时间的函数的应变,蠕变率由长期数据的直线斜率决定.。所有的应变率超过105s-1。通过拟合蠕变应变速率作为施加应力的函数(图3(b)),我们发现当n = 1.5,或应变速率敏感性M = 0.67,NC-SiC D = 2纳米被划分作为一个真正的超塑性材料。推算下来到10-2s-1这个工业可实现应变率表明,应变超过1000%应该是很容易实现的。图3(c)显示压力4 GPA 时纳米SiC晶粒的蠕变速率对晶粒尺寸的依赖性。这一数据表明εprop;d-3.2。这从D = 6纳米到D = 2纳米的急剧增加的蠕变速率与晶粒尺寸减小解释了机械行为中观察到的戏剧性的变化。
图3显示的结果清楚地表明,SiC在D = 2 nm在室温下表现为一种持续的蠕变变形的弹塑性材料。而蠕变已经通过MD多晶材料中观察到,它只发生在通过晶界扩散高温下(Coble蠕变)。相反,这里描述的蠕变变形机制通过三协调机制:晶界滑动、在晶界S和三联点扩散,和在晶界S的纳米孔成核、生长和迁移。通常,多晶材料的稳态蠕变速率ε由经验关系ε=ADGb/kT(b/d)p (sigma;/G)n描述,KT是热能,P和N反粒度和应力指数,D(格,晶界,或三结)扩散系数,G剪切模量,b是伯格斯矢量的大小和A是有无量纲常数的大小。不同的蠕变机制取决于不同的材料和条件。当晶内机制取决于位错攀移,即P = 0。这显然不是这里的情况,因为没有观察到位错和ε具有较强的晶粒尺寸依赖性。当蠕变是由晶内扩散控制,P = 2,n = 1;这也不是这里的情况,因为没有观察到这种扩散。然而,当蠕变发生通过晶界滑动(Rachinger蠕变),P = 2,n = 2;沿晶界扩散(Coble蠕变),P = 3,n = 1;扩散沿三联点,P = 4,n = 1。图3 d = 2 nm的结果表明,所有这些蠕变机制在共同行动 (P = 3.2,N = 1.5),即晶界滑动以及沿晶界和三联路口扩散。
变形的分析,如图4所示(一)–(C),证实了晶界滑动容纳扩散在晶界和三联点是主要的变形机制。随着晶粒尺寸的减小晶界扩散变得越来越重要。特别是,晶界三联点变得非常重要,一旦晶粒尺寸减小到纳米级的区域。d<3 nm的纳米结构理论计算表明,在三联点的原子数大于在晶界的,表明这些材料是从他们的结构尺寸有本质区别。这些最近的理论与我们的模拟中观察到的超塑性行为是一致的。
在颗粒材料的体积分数进行了详细的分析,模拟晶粒边界和三联点证实了界面越来越重要的想法,尤其是三和更高的连接,随着晶粒尺寸的减小(图4(d)–(f))。计算非晶晶界厚度(图4(d))是~ 0.75 nm(基于环的数据)。它在晶界,特别是,晶界三联点纳米腔倾向于核,如图4(E)。仿真结果表明(见图2),纳米腔在D = 2 nm的样品比D = 4 nm和D = 6 nm的样品在较大的应变生长开始成核更缓慢。我们认为,这种行为与D = 2 nm增加样品的有效扩散率有关系结合更大的晶界和三联点的体积分数。在三联点的扩散系数可以比晶界和晶粒内高出几个数量级。图4(f)量化在晶粒内部的原子的实际分数,晶界和三联点。(后者包括四倍晶界和更高的连接)减少晶粒尺寸D = 4 nm为D = 2 nm,在折痕明显三连接原子的分数,从9.9%到33%。这个分数相近的晶界S(37%)和大于晶粒内(29.5%)。这表明,三联点及其增强扩散系数的急剧增加对在蠕变和超塑性在晶粒尺寸减少这个比例上升的影响。
我们的研究结果可以用来理解对SiC纳米线在弯曲和拉伸试验中显示令人惊讶的超塑性,ε=10-4-10-5s-1的实验报道应变下纳米线断裂,超过100%的弯曲和拉伸变形下超过200%。用MD模拟了解导致所观察到的这种大变形的原子机制预测广泛性塑性应变时,薄的非晶薄膜的晶粒之间的是相对于纳米线拉伸轴对齐剪大解决。这是与我们的观测结果,当提供的纳米SiC的粒度很小超塑性可以被实现,取得了一致的。我们怀疑我们的纳米SiC的模拟中观察到蠕变机制(即晶界滑动,沿晶界扩散的三结和纳米空穴)也可以在含有高密度非晶薄膜的其它纳米SiC结构中工作。
这里给出的结果提供了陶瓷的微观结构设计的重要指导(或具有优异延展性陶瓷)。然而,我们注意到,虽然目前的模拟集中在完全致密,无杂质和夹杂物的无缺陷的陶瓷,但大多数实验系统是不太理想的。此外,碳化硅的晶粒尺寸为几纳米的烧结(例如,通过化学气相合成获得3纳米颗粒的粉末)是出了名的困难。幸运的是,负担得起的烧结(如火花等离子烧结)允许生产近100%致密的SiC的晶粒尺寸低于10纳米。有减少缺陷的策略,例如通过胶体控制烧结,或使陶瓷不敏感也已成功应用。这里所提出的打开了实现新的易于成型,强度高、超塑性陶瓷型的可能性。结果如图1所示非晶SiC应该甚至比D = 2 nm的纳米SiC有更好的超塑性。然而,大块非晶碳化硅只能由中子/离子照射,这严重限制了其应用。
我们的研究结果表明,NC SiC可以在室温下维持超塑性变形,同时保持显着的高强度。通过一系列的晶粒尺寸创建材料,我们可以设想一个仅由SiC晶粒尺寸不同实现不同强度和韧性之间的平衡组合。纳米晶和单晶SiC复合超层将呈现非常高的强度相对于典型的单片SiC,而且支持大变形。这种方法也可以用于装甲涂层设计其中非常强大的抗裂材料是特别可取的。
本选题的目的是专注于对超塑性纳米陶瓷碳化硅的研究,纳米陶瓷碳化硅可以在室温下维持超塑性变形,同时保持显着的高强度。通过一系列的晶粒尺寸创建材料,我们可以设想一个仅由碳化硅晶粒尺寸不同实现不同强度和韧性之间的平衡组合。
通常情况下,理想的结构材料应具有坚固、韧性和可成型的特性。这是因为高强度是可取的,以支持高负荷的材料,而延展性和成形性,使其容忍结构缺陷,对灾难性故障有抵抗力,并允许复杂形状的应用程序的设计。但不幸的是,高强度的材料,如陶瓷,通常是脆性的,没有明显的塑性变形。在过去的几十年里,维护高强度已经成了材料科学家面临的主要挑战任务。所以,对于纳米陶瓷碳化硅这种在保持高强的的同时又可以在室温下维持超塑性变形的材料的研究的显得尤为重要。
剩余内容已隐藏,支付完成后下载完整资料
资料编号:[27405],资料为PDF文档或Word文档,PDF文档可免费转换为Word