金属学报, 2025, 61(4): 643-652 DOI: 10.11900/0412.1961.2023.00175

研究论文

He泡液态金属铝的动态拉伸断裂机制与损伤模型

周婷婷,, 赵福祺, 周洪强, 张凤国, 殷建伟

北京应用物理与计算数学研究所 北京 100094

Mechanism and Damage Model for the Dynamic Tensile Fracture of Liquid Aluminum Containing He Bubbles

ZHOU Tingting,, ZHAO Fuqi, ZHOU Hongqiang, ZHANG Fengguo, YIN Jianwei

Institute of Applied Physics and Computational Mathematics, Beijing 100094, China

通讯作者: 周婷婷,zhou_tingting@iapcm.ac.cn,主要从事材料动态力学行为研究

责任编辑: 李海兰

收稿日期: 2023-04-20   修回日期: 2023-05-14  

基金资助: 国家自然科学基金项目(12172063)

Corresponding authors: ZHOU Tingting, associate professor, Tel:(010)59872646, E-mail:zhou_tingting@iapcm.ac.cn

Received: 2023-04-20   Revised: 2023-05-14  

Fund supported: National Natural Science Foundation of China(12172063)

作者简介 About authors

周婷婷,女,1986年生,副研究员,博士

摘要

为揭示辐照缺陷(如He泡)对液态金属动态断裂的影响,并发展相应的损伤理论模型,本工作采用分子动力学模拟和连续介质力学理论计算研究了宽应变率范围内含He泡液态金属铝的动态拉伸断裂行为。模拟结果表明,在3.0 × 106~3.0 × 109 s-1应变率范围内,He泡增长是导致液态金属断裂的物理机制;在极高应变率下(如3.0 × 1010 s-1),除He泡增长,也观察到孔洞成核、增长,但He泡增长仍占据主导地位。He泡增长过程可分为快速增长和较慢增长2个阶段,该阶段性增长特征不受应变率影响,但He泡增长速率随着应变率的升高而显著增加。与液态纯Al相比,含He泡液态金属铝的动态拉伸强度显著降低,该差异在极高应变率下有所减小。进一步发展了含He泡液态金属的损伤断裂模型,采用该模型的理论计算结果与微观模拟结果在宽应变率范围内相符。

关键词: 动态断裂; He泡; 液态金属; 分子动力学; 损伤模型

Abstract

The dynamic fracture of metals in liquid state has become a subject of considerable interest in current times because of its observation in various physical and technological processes such as inertial confinement fusion and high-power laser-driven surface micromachining. In addition, it has been found that the fractures at elevated temperature are highly correlated with the microstructure of materials. He bubbles are frequently observed in many metals exposed to irradiation environments as a result of radioactive or self-irradiation. Both experimental and theoretical studies have indicated that He bubbles can substantially affect the mechanical properties of irradiated metals, resulting in hardening, swelling, and embrittlement. In recent years, attention has been drawn to understand the effects of He bubbles on the dynamic properties of materials, including shock compression, dynamic fracture, and surface ejection. This study examines the dynamic tensile fracture behavior of liquid aluminum containing He bubbles across a wide range of strain rates by utilizing molecular dynamics (MD) simulations and continuum modeling. The physical mechanism leading to the dynamic fracture is revealed to be predominated by the growth of He bubble. Under strain rates ranging from 3.0 × 106 s-1 to 3.0 × 109 s-1, tension primarily induces bubble growth. At higher strain rates, such as 3.0 × 1010 s-1, both bubble growth and void nucleation-growth are observed, although bubble growth remains the dominant factor. The growth of He bubbles unfolds in two distinct phases: rapid growth followed by slower growth. These staged evolutionary characteristics appear to be consistent across strain rates, but the growth rate of helium bubbles markedly increases with increasing strain rates. Furthermore, the dynamic tensile strength at varying strain rates indicates a significant reduction for the metal containing He bubbles compared to the pure metal. However, this discrepancy decreases at extremely high strain rates, such as 3.0 × 1010 s-1. In addition, a continuum damage model is constructed based on the insights obtained from MD simulations to describe the dynamic tensile fracture of liquid metal containing He bubbles. This model accounts for external tensile stress, internal pressure of He bubbles, inertia, viscosity, and surface tension. Theoretical calculations using the damage model and the binomial equation of state, which depict the pressure-volume relationship of the metal substrate, exhibit excellent agreement with MD data over a wide range of strain rates. This includes the evolution of the tensile stress and He bubble radius. The self-consistent MD-continuum model proposed in this study has the potential to be applied in macroscopic hydrodynamic simulations, to depict the dynamic tensile fracture behavior of liquid metal with He bubbles.

Keywords: dynamic fracture; He bubble; liquid metal; molecular dynamics; damage model

PDF (1861KB) 元数据 多维度评价 相关文章 导出 EndNote| Ris| Bibtex  收藏本文

本文引用格式

周婷婷, 赵福祺, 周洪强, 张凤国, 殷建伟. He泡液态金属铝的动态拉伸断裂机制与损伤模型[J]. 金属学报, 2025, 61(4): 643-652 DOI:10.11900/0412.1961.2023.00175

ZHOU Tingting, ZHAO Fuqi, ZHOU Hongqiang, ZHANG Fengguo, YIN Jianwei. Mechanism and Damage Model for the Dynamic Tensile Fracture of Liquid Aluminum Containing He Bubbles[J]. Acta Metallurgica Sinica, 2025, 61(4): 643-652 DOI:10.11900/0412.1961.2023.00175

极端条件下材料的动态力学行为(如塑性变形、冲击相变、动态断裂)是冲击动力学、固体力学、材料科学等学科研究的重要内容,也是高速撞击、内爆压缩等动高压工程领域关注的科学问题。当冲击波在金属自由面或金属/气体界面反射时,金属在入射稀疏波与自由面反射稀疏波的拉伸作用下会发生动态断裂,也称为层裂[1]。该现象广泛存在于航空航天、国防及民用领域,因此自发现以来就受到了极大关注。经过一个世纪的实验与理论研究,人们对固态金属动态断裂行为已获得了深入的认识,发展了多种损伤理论模型以描述损伤成核至最终断裂的过程[1~3]

近年来,对液态金属动态断裂的研究逐渐得到关注。在较强的冲击载荷下,金属会发生卸载熔化或冲击熔化,导致动态断裂发生在液态[4,5]。在现代工业装置(如面向未来聚变能源的惯性约束聚变装置)和激光驱动的表面微加工中,也观察到熔化后金属的动态断裂现象[6~10]。对低熔点金属(如Pb和Sn)的层裂实验发现,与固态金属不同,液态金属的自由面速率曲线没有明显的回跳信号,且表面会形成大量的物质喷射[11~16]。对高熔点金属,科研人员通过对样品加热再施加冲击载荷,研究了液态金属Cu、Al等的层裂行为[17~22]。Zaretsky[17]发现Sn和Zn在温度仅低于熔点10 K的情况下仍然具有较高的层裂强度,但是当金属完全熔化后其强度比固态低了一个量级,而液态金属Pb的层裂强度只比固态低3倍。对金属Al的层裂实验则发现高温下的层裂强度仍然与金属的微结构密切相关,商用铝合金AD1的层裂强度在初始温度接近熔点时降低至接近零,但高纯度的多晶铝A999和单晶Al在熔化后的层裂强度仍然很高[20~22]。分子动力学(MD)模拟表明,固态和液态金属的层裂机制都是孔洞成核、增长、贯通,2者的主要差异表现在孔洞数量、体积分布、损伤深度以及层裂强度上[5,23,24]。Mayer等[25]采用MD模拟研究了动态拉伸下金属Pb和Sn的断裂行为,发现纯金属的层裂强度显著高于实验结果,但当考虑金属中存在杂质时,该差异被消除。他们结合MD模拟与连续介质力学理论计算,发展了液态金属的损伤理论模型[25~27]

在辐照环境中,金属材料内部会产生大量的缺陷,如空位、He原子、位错等[28~32]。He原子因其在金属中的低溶解性会聚集成大量纳米尺度的He泡,并具有很高的内压[29,32,33]。大量实验和理论研究[28,29,32,34~37]表明,He泡会对金属的力学性能产生显著影响,包括局部硬化、延展性降低、高裂纹率及低断裂时间等。最近的拉伸实验结果发现,He泡能够同时增强百纳米直径的单晶Cu和单晶Al-Cu合金的强度和延展性[38,39]。此外,近年来He泡对金属冲击压缩与动态拉伸断裂等动态力学性能的影响开始得到关注。对含He泡掺B铝层裂行为的平板撞击实验结果发现,He泡会降低金属的层裂强度[40~42],并且通过样品回收观察到He泡的长大与汇合现象[40,41]。高能激光实验结果发现,He泡和B均会降低金属的层裂强度,且B对层裂强度的影响比He泡更加明显[43]。在理论模型方面,祁美兰等[44]在孔洞增长模型基础上建立了恒定拉应力下He泡的生长方程,讨论了He泡内压、黏性、惯性及温度对He泡生长的影响。张凤国等[45]基于耦合材料初始损伤、孔洞尺寸及惯性效应的损伤模型,将He泡内压作为外力,研究了含He泡金属铝的层裂行为。

由于He泡涉及微观尺度,当前的实验手段难以获得对He泡动态演化的直接观测,而通过回收检测受冲击样品也很难获得对其演化过程和微观机理的认识。此外,实验样品中存在的其他缺陷使得确定He泡单一因素的影响非常困难。另一方面,已有的理论模型只是将He泡内压作为外力耦合进描述纯金属的损伤模型中,它们能否正确描述纳米级He泡的动态演化规律和损伤发展还有待考察。MD模拟为研究动载荷下含He泡金属的动态力学行为提供了另一种可行路径,可为He泡动态演化规律及其对金属动态响应的影响提供原子层面的重要信息。Kubota等[46]采用MD方法研究了含He泡金属铝在冲击作用下的动力学响应,获得了Hugoniot曲线和临界剪切强度。Wang等[47]对比研究了冲击载荷下含He泡或孔洞金属铝的塑性变形,发现He泡会更加显著地促进周围位错环的发射。Shao等[48]研究了衰减冲击波加载下He泡对金属铝层裂破坏的影响,结果表明,当冲击压力较低时He泡显著降低了金属的层裂强度。本课题组研究发现,He泡增长、汇合是导致延性金属断裂的主要机制,He泡会显著抑制孔洞的成核、增长[49,50];采用动态拉伸MD模拟,获得了宽应变率范围内含He泡固态金属铝的损伤演化规律,并结合连续介质力学理论计算发展了能描述固态金属中He泡增长的损伤模型[51]

本工作针对液态金属动态断裂行为及辐照缺陷的影响,采用MD模拟与连续介质力学理论计算,研究宽应变率范围内含He泡液态金属铝的动态拉伸断裂机制、断裂强度及损伤演化规律。

1 模拟方法与计算细节

1.1 分子动力学模拟

含He泡液态金属铝(下文简称含He泡铝)样品的初始结构如图1a所示,在[100]、[010]、[001] 3个方向各有40个晶体单胞。Al单胞为fcc结构,晶格常数为0.405 nm,故样品在3个方向的尺寸均为16.2 nm。为了在样品中构建He泡,首先通过删除金属原子在样品中心构建了一个半径R0 = 1.5 nm的球形孔洞,再往孔洞里添加He原子。He原子的数量由He原子/金属原子空位的数量比(δ)确定,本工作设置为δ = 0.5。根据He原子与金属原子的数量,计算出He原子的浓度为0.188%。根据He泡体积与样品尺寸,计算出He泡的初始体积分数D0 = 0.33%。He泡的初始状态如R0δD0均基于已有研究结果[29,32]

图1

图1   含He泡液态金属铝(简称含He泡铝)样品的初始结构与分子动力学(MD) (1100 K)优化后的结构

Fig.1   Initial configuration of the liquid Al sample with He Bubbles (Al-He sample) (a) and the optimized configuration (by molecular dynamics (MD) simulation) at 1100 K (b) (The sample is cut through the center to show the central He bubble)


MD模拟的核心是描述原子间相互作用的势函数。本工作采用Zope和Mishin[52]发展的嵌入原子方法(embedded atom method,EAM)势函数来描述金属Al原子间的相互作用。He原子间、Al-He原子间相互作用分别采用 式(1)[53]式(2)[47]的对势来描述,势函数参数来自文献[47]。相关势函数已广泛用于金属Al和Al-He体系力学性质的研究[47~51]

ϕ(r)=ε{6a-6exp[a(1-rr*)]-aa-6(r*r)6}
ϕ(r)=-α[1+β(rra-1)]exp[-β(rra-1)]

式中,ϕ为原子间相互作用的势能;r为原子之间的距离;aαβ为拟合系数,a = 13.1,α = 0.024 eV;β = 7.046;ε / k = 12.4,为拟合系数(其中,ε为原子间相互作用势能的最低值,k为Boltzmann常数);r* = 0.325 nm,为He原子间相互作用的截断半径(cut-off radius);ra = 0.567 nm,为Al-He原子间相互作用的截断半径。MD模拟采用国际开源程序Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS)[54]。首先采用共轭梯度法对样品进行能量最小化,再采用等温等压(NPT)系综在300 K和零压下对样品进行20 ps的结构优化。随后将样品升温至熔化后的目标温度(T = 1100 K),再采用NPT系综进行该温度和零压下的结构优化,优化时间为50 ps以确保样品完全熔化。优化后的样品结构如图1b所示,样品密度为2.372 g/cm3,He泡初始半径R0 = 1.868 nm,He泡初始内压pg0 = 0.632 GPa。

以优化后的结构作为施加动态拉伸载荷的初始结构,采用恒定应变率的三轴拉伸,拉伸过程中样品内原子坐标变化通过LAMMPS自带的“deform”命令来实现。动态拉伸过程采用等温等容(canonical,NVT)系综,温度保持在目标温度1100 K。拉伸的体积应变率(ε˙,下文的应变率均指体积应变率)范围为3.0 × 106~3.0 × 1010 s-1,以研究宽应变率范围内样品的动态拉伸断裂行为。MD模拟中采用三维周期性边界条件,时间步长为1 fs。动态拉伸过程中He泡与孔洞等微结构的演化通过Open Visualization Tool (OVITO)软件中的“Construct surface mesh”方法[55]进行可视化分析。在该方法中,当孔洞半径大于最近邻原子半径时被识别为孔洞。孔洞体积采用本课题组最近发展的统计方法和程序来计算[51]。具体做法为:将样品在三维空间划分为若干个立方体单元,单元尺寸与金属的晶格常数相当,将样品内所有原子根据其坐标划分到这些立方体中;若在3个方向连续2个单元内均没有原子,则识别为孔洞;将相互连接的多个空单元合并为一个孔洞,该孔洞的体积为这些空单元的体积之和。与“Construct surface mesh”方法相比,该孔洞统计方法能有效排除点缺陷。He泡作为一种含气孔洞,其统计方法与孔洞相同,只需先删除He原子。

1.2 损伤模型与理论计算

MD模拟结果(见2.1节)表明基体材料熔化后He泡增长是导致材料拉伸断裂的主导物理机制,传统的孔洞成核、增长机制被显著抑制,与基体材料为固态时一致[51]。因此,在建立损伤模型时本工作忽略孔洞成核、增长项,仅考虑He泡增长项。模型以孔洞增长模型[1~3]为基本框架,考虑He泡内压的影响,具体表达式如 式(3)所示,包括外加应力、He泡内压、惯性、黏性及表面张力对He泡增长的作用。

P+pg0R0Rn=ρsRR¨+1.5R˙2+4ηR˙R+2γR

式中,左边第一项P为外界施加的拉伸应力;左边第二项为He泡内压[56]R指He泡半径,R0pg0均指He泡初始参数(即初始半径和初始内压),n表示He泡气压随半径的衰减系数;右边第一项为惯性,ρs为基体材料的密度;R˙表示R随时间(t)的一阶导数,即He泡半径变化的速度,R¨表示Rt的二阶导数,即He泡半径变化的加速度;右边第二项为材料黏性,η为黏性系数;右边第三项为表面张力,γ为表面能。MD模拟可给出He泡内压随He泡半径的变化规律,从而可以确定He泡内压项的指数n (见补充材料图S1)。He泡的表面能可由MD模拟给出的He泡初始半径和初始内压依据Laplace-Young方程(pg0 = 2γ / R0)[57]确定。方程(3)中仅有一个待定参数η。与固态的损伤模型不同[51],在描述液相的损伤发展时,本工作不再考虑基体材料的强度。

与MD模拟类似,在进行连续介质力学理论计算时,首先构建一个周期性的代表体积单元(representative volume element,RVE)。在三轴均匀拉伸作用下,RVE的体积(V)随时间的变化如 式(4)所示:

V=V0expε˙t

式中,V0为样品初始体积。基体材料的体积(Vs)随时间的变化如 式(5)所示:

Vs=V-43πR3

基体材料的状态方程采用二阶多项式表示[26]

P=-K1VsV0-1+K2VsV0-12

式中,K1为体模量;K2为模量的二阶修正。K1K2可通过拟合损伤发展前MD模拟得到的基体材料P-V关系来确定(见补充材料图S2)。

通过式(3)~(6),可对含He泡金属在动态拉伸下的损伤演化过程进行计算,得到He泡半径和拉伸应力随时间的演化。

2 结果与讨论

2.1 动态拉伸断裂机制与损伤演化规律

以应变率为3.0 × 108 s-1为例对动态拉伸过程进行详细讨论。图2a是该应变率下含He泡铝样品中He泡的微结构演化过程,拉伸应力、He泡半径以及He泡内压随时间的变化见图3。由图2a可以看出,在拉伸应力作用下,He泡很快发生膨胀,说明基体材料熔化后,He泡增长的临界应力很低(约为0.068 GPa),显著低于固态。固态时,He泡膨胀需要克服材料的屈服强度,当基体材料发生塑性变形后He泡才会发生明显增长,因此临界应力较高[51]。在膨胀过程中,液态时He泡基本保持为球形,而固态时为八面体结构[51]。此外,在整个拉伸过程中,没有观察到孔洞成核、增长,表明He泡增长是导致液态金属拉伸断裂的主导物理机制,这与固态时的断裂机制[51]相同。

图2

图2   应变率为3.0 × 108 s-1时纯Al和含He泡铝样品在拉伸过程中的He泡与孔洞演化

Fig.2   Evolutions of He bubble/voids during the dynamic tension process at the strain rate (ε˙) of 3.0 × 108 s-1

(a) Al-He sample (b) pure Al


图3

图3   应变率为3.0 × 108 s-1时含He泡铝样品在拉伸过程中的拉伸应力(P)、He泡半径(R)以及He泡内压(Pg)随时间的变化

Fig.3   Time (t) evolutions of tensile stress (P) (a), He bubble radius (R) (b), and the internal pressure of He bubble (Pg) (c) of Al-He sample during the dynamic tension process at ε˙ = 3.0 × 108 s-1 (In Fig.3b, the black curve represents the fast growth of He bubble and the blue curve shows the slower growth of He bubble; the same in Fig.6)


拉伸应力随时间的变化如图3a所示,其先快速增加到最大值再缓慢降低至趋近收敛值。样品所经历的最大拉伸应力即动态拉伸强度为0.381 GPa (t = 62 ps),显著高于He泡增长的临界应力,说明He泡膨胀不会立即导致应力弛豫,这与纯金属不同。He泡增长过程大致可分为早期快速增长与晚期较慢增长2个阶段,如图3b所示。通过对He泡半径随时间的演化进行拟合,发现He泡的早期增长满足指数规律(图中黑色曲线所示),而晚期增长满足幂律(图中蓝色曲线所示),转变点发生在约100 ps。He泡增长使得He泡内压快速降低,尤其在早期约50 ps内几乎呈线性下降趋势,如图3c所示。

相同应变率下液态纯Al中孔洞的演化过程如图2b所示。与含He泡铝不同,纯Al的动态拉伸断裂机制为孔洞成核、增长。由OVITO软件确定的孔洞成核最早发生在约120 ps,但由孔洞的演化过程可知,早期形成的孔洞是随机分布的且不能稳定存在,这可能是由于热涨落而形成的一些点缺陷。直到拉伸应力增大到3.061 GPa ( t = 242 ps),孔洞的成核点才逐渐稳定下来并开始长大(如图2b中最大的孔洞),其余孔洞在表面张力的作用下逐渐塌缩。最终,在模拟样品中仅有一个孔洞长大,这与Mayer等[26,27]的研究结果一致。

图4对比了纯Al和含He泡铝样品的拉伸应力与孔洞体积随时间的变化。由图4a可知,纯Al经历的应力演化与含He泡铝明显不同。纯Al的拉伸应力随时间增大到最大值后迅速降低至零甚至变为正压,并出现一定幅值的振荡。纯Al样品所经历的最大拉伸应力为3.111 GPa (t = 246 ps),仅略微高于孔洞的临界成核应力,说明孔洞增长会迅速导致应力弛豫。此外,纯Al中孔洞成核的临界应力与最大拉伸应力均显著高于含He泡铝,表明含He泡铝的损伤更容易发展且动态拉伸强度更低。纯Al样品中孔洞的增长过程可分为3个阶段,如图4b所示。孔洞成核后在短时间内迅速增长,增长速率显著高于He泡的增长速率;之后因应力振荡导致孔洞出现周期性的收缩与膨胀;当拉伸应力弛豫到接近收敛值时,孔洞开始缓慢增长,其增长速率与He泡增长速率相当。最终,孔洞体积比He泡体积略小。

图4

图4   应变率为3.0 × 108 s-1时纯Al样品与含He泡铝样品在拉伸过程中的拉伸应力及He泡/孔洞体积随时间的变化

Fig.4   Time evolutions of P (a) and volume of voids/He bubble (b) during the dynamic tension process at ε˙ = 3.0 × 108 s-1 for pure Al sample and Al-He sample


2.2 应变率效应

在本工作研究的应变率范围内,含He泡铝样品的动态拉伸断裂机制始终由He泡增长所主导,孔洞成核、增长被明显抑制,如图5所示。图6给出了不同应变率下含He泡铝样品在拉伸过程中的拉伸应力、He泡半径以及He泡内压随时间的变化。结合图3可以看出,不同应变率下He泡的阶段性演化过程是相似的,均可分为快速增长与较慢增长2个阶段。通过数据拟合发现,早期增长均满足指数规律,晚期增长均满足幂律。但是,随着应变率的降低,He泡在2个阶段的增长速率相差越来越小,使得阶段性的演化特征不再明显。这主要是由于随着应变率的降低,快速增长阶段的速率下降比缓慢增长阶段更加显著。不同应变率下的拉伸应力随时间的演化过程基本相同,均是快速增加至最大值再缓慢减小至趋近收敛值,但变化速率随着应变率的升高而增加。此外,在较高应变率下(ε˙ = 3.0 × 109 s-1),拉伸应力减小至接近收敛值后会出现振荡现象。

图5

图5   应变率为3.0 × 109和3.0 × 107 s-1时含He泡铝样品在拉伸过程中的He泡演化

Fig.5   Evolutions of He bubble during the dynamic tension process at ε˙ = 3.0 × 109 s-1 (a) and ε˙ = 3.0 × 107 s-1 (b) for Al-He sample


6 不同应变率下含He泡铝样品在拉伸过程中的拉伸应力、He泡半径及He泡内压随时间的变化

Fig.6 Time evolutions of P (a1-c1), R (a2-c2), and Pg (a3-c3) of He bubble during the dynamic tension process at ε˙ = 3.0 × 109 s-1 (a1-a3), ε˙ = 3.0 × 107 s-1 (b1-b3), and ε˙ = 3.0 × 106 s-1 (c1-c3) for Al-He sample

样品的动态拉伸强度随着应变率的升高而增大,尤其是当应变率高于3.0 × 108 s-1时呈现出明显的应变率强化效应,如图7所示。当应变率为3.0 × 106、3.0 × 107、3.0 × 108和3.0 × 109 s-1时,含He泡铝样品的动态拉伸强度分别为0.332、0.337、0.381和0.743 GPa,显著低于纯Al的强度(2.774、2.920、3.111和3.533 GPa),表明金属熔化后He泡仍然会降低金属的强度,这与Glam等[41]开展的加热层裂实验结果一致。初始温度为600 ℃的纯Al和含He泡铝平板撞击实验结果表明,He泡显著降低了金属的层裂强度(1.37 GPa vs 0.4 GPa)[41]。前期对冲击载荷下含He泡铝层裂的MD模拟发现,熔化后He泡对层裂强度的影响很小[49,50],这与本工作的研究结果不同,其主要原因是应变率不同。在冲击层裂模拟中,样品所经历的应变率接近1011 s-1,高于本工作动态拉伸模拟的应变率。为此,本工作补充了应变率为3.0 × 1010 s-1的动态拉伸模拟,得到含He泡铝和纯Al样品的强度分别为3.355和4.63 GPa。这表明随着应变率进一步增加,含He泡铝的强度显著增加,使得He泡对动态拉伸强度的影响有所减弱,趋向于冲击层裂的MD模拟结果。对于纯Al,飞秒激光实验结果得到的温度为2500 K时的层裂强度高达2.5 GPa[58],比本工作在1100 K的计算结果略低。

图7

图7   含He泡铝样品和纯Al样品的动态拉伸强度随应变率的变化

Fig.7   Variations of dynamic tensile strength with ε˙ for pure Al sample and Al-He sample


大量的实验与理论研究已表明,金属的拉伸强度会随着应变率的升高而增加[25,59~62],而应变率对拉伸强度的影响机制与不同应变率下的损伤成核机制有关[25,62]。对于完美晶体,高/低应变率下的损伤都是均匀成核的,所需的成核应力较高。因此,层裂强度也较高,且层裂强度随应变率的降低而缓慢下降,如图7中的金属纯Al。对于含缺陷金属,应变率较低时,损伤是非均匀成核的,以缺陷处的成核增长为主,所需的成核应力较低;应变率较高时,非均匀成核不足以将应力弛豫,均匀成核也会被激发。因此,含缺陷金属的层裂强度会随着应变率的降低而显著降低,且低于完美晶体,如图7中的含He泡金属。对应变率为3.0 × 1010 s-1时含He泡铝的微结构分析证明,高应变率下除了He泡增长,还出现了新的孔洞成核、增长(如图8所示),这与更低应变率下的损伤仅由He泡增长控制有所不同。

图8

图8   应变率为3.0 × 1010 s-1时含He泡铝样品在拉伸过程中的微结构演化

Fig.8   Microstructure evolutions for the sample with He bubble during the dynamic tension process at ε˙ = 3.0 × 1010 s-1 (The central large pore is He bubble, and the other small pores are new nucleated voids)


2.3 损伤模型计算结果与参数变化规律

采用 式(3)的损伤模型和 式(6)的状态方程,对含He泡铝在不同应变率下的动态拉伸过程进行了理论计算,模型参数见表1。计算得到的拉伸应力与He泡半径随时间的演化如图9a和b所示。通过对比可以看出,模型给出的结果与MD模拟结果整体上符合很好,尤其是当应变率在3.0 × 106~3.0 × 108 s-1范围内。在更高应变率下(3.0 × 109 s-1),2者之间略有差别,主要体现在模型给出的拉伸应力后期的振荡比MD模拟结果略微明显。图9c是不同应变率下含He泡铝的动态拉伸强度。可以看出,模型计算结果与MD模拟结果基本相同。上述结果表明,本工作采用的损伤模型与参数能够很好地描述宽应变率范围内含He泡液态金属铝的动态拉伸断裂行为,包括He泡增长与应力演化过程。

表1   损伤理论模型中的参数

Table 1  Parameters in the present damage model

ε˙ / s-1R0 / nmpg0 / GPaγ / (J·m-2)nK1 / GPaK2 / GPaη / (Pa·s)
3.0 × 1061.8680.6320.594.6511300.03
3.0 × 1071.8680.6320.594.6511300.006
3.0 × 1081.8680.6320.594.6511300.0025
3.0 × 1091.8680.6320.594.6511300.0012

Note:R0initial radius of He bubble, pg0—initial pressure of He bubble, γ—surface energy of He bubble, n—coefficient in the relationship between the pressure and radius of He bubble, K1 and K2—bulk modulus and its quadratic correction, η—viscosity coefficient

新窗口打开| 下载CSV


图9

图9   模型计算与MD模拟得到的不同应变率下含He泡铝样品的拉伸应力和He泡半径随时间的变化,及动态拉伸强度

Fig.9   Time evolutions of P (a) and R (b), and dynamic tensile strength (c) at various strain rates obtained from MD simulations and the present continuum model


损伤模型中唯一的待定参数η通过拟合MD模拟得到的拉伸应力与He泡半径随时间的变化确定,其随应变率的变化如图10所示。黏性系数随着温度的升高会显著降低,因此液体的黏性系数会明显低于固体材料[51]。此外,黏性系数随着应变率的升高而降低,这与已有研究结果[63,64]一致。将黏性系数随应变率的变化按照 式(7)进行拟合,得到拟合系数c1 = 1.208 × 107c2 =10.606。

η=c1(lgε˙)-c2

图10

图10   不同应变率下损伤模型中的黏性系数

Fig.10   Viscosity coefficients at different ε˙ used in the present continuum model (dots) and the fitting curve using Eq.(7)


本工作模型可进一步用于宏观流体动力学模拟,为解决工程实际问题奠定了理论基础。

3 结论

(1) 在恒定应变率的动态拉伸作用下,He泡增长是导致液态金属铝发生断裂的主导物理机制。在3.0 × 106~3.0 × 109 s-1的应变率范围内,金属断裂仅由He泡增长控制;在极高应变率下(如3.0 × 1010 s-1),He泡增长和孔洞成核-增长共同导致了金属断裂,但前者仍占据主导地位。He泡增长过程可分为快速增长和较慢增长2个阶段,分别满足指数规律和幂律。He泡的阶段性增长特征不受应变率影响,但He泡增长速率随着应变率的升高而显著增加。

(2) 获得了宽应变率范围内含He泡液态金属铝的动态拉伸强度,其随应变率的升高而增加,尤其是当应变率高于3.0 × 108 s-1时出现明显的应变率强化效应。与液态纯Al相比,含He泡铝的强度显著降低,但在极高应变率下(3.0 × 1010 s-1) 2者的差异减小,这与极高应变率下的损伤机制有关。

(3) 发展了描述液态金属中He泡增长的损伤理论模型,采用该模型的理论计算结果与微观模拟结果在宽应变率范围内相符,包括He泡增长与应力演化过程,并给出了模型参数随应变率的变化规律。

文中补充材料可通过以下网址查看:https://www.ams.org.cn/CN/10.11900/0412.1961.2023.00175

参考文献

Antoun T, Curran D R, Razorenov S V, et al. Spall Fracture [M]. New York: Springer, 2003: 1

[本文引用: 3]

Curran D R, Seaman L, Shockey D A.

Dynamic failure of solids

[J]. Phys. Rep., 1987, 147: 253

Johnson J N.

Dynamic fracture and spallation in ductile solids

[J]. J. Appl. Phys., 1981, 52: 2812

[本文引用: 2]

Andriot P, Chapron P, Lambert V, et al.

Influence of melting on shocked free surface behavior using Doppler laser interferometry and X ray densitometry

[A]. Shock Waves in Condensed Matter 1983 [M]. Amsterdam: Elsevier, 1984: 277

[本文引用: 1]

Luo S N, An Q, Germann T C, et al.

Shock-induced spall in solid and liquid Cu at extreme strain rates

[J]. J. Appl. Phys., 2009, 106: 013502

[本文引用: 2]

Smalyuk V A, Weber S V, Casey D T, et al.

Hydrodynamic instability experiments with three-dimensional modulations at the national ignition facility

[J]. High Power Laser Sci. Eng., 2015, 3: 1

[本文引用: 1]

Orth C D.

Spallation as a dominant source of pusher-fuel and hot-spot mix in inertial confinement fusion capsules

[J]. Phys. Plasmas, 2016, 23: 343

Tsakiris N, Anoop K K, Ausanio G, et al.

Ultrashort laser ablation of bulk copper targets: Dynamics and size distribution of the generated nanoparticles

[J]. J. Appl. Phys., 2014, 115: 243301

Oboňa J V, Ocelík V, Rao J C, et al.

Modification of cu surface with picosecond laser pulses

[J]. Appl. Surf. Sci., 2014, 303: 118

Shugaev M V, Shih C Y, Karim E T, et al.

Generation of nanocrystalline surface layer in short pulse laser processing of metal targets under conditions of spatial confinement by solid or liquid overlayer

[J]. Appl. Surf. Sci., 2017, 417: 54

[本文引用: 1]

Kanel G I, Savinykh A S, Garkushin G V, et al.

Dynamic strength of tin and lead melts

[J]. JETP Lett., 2015, 102: 548

[本文引用: 1]

de Rességuier T, Signor L, Dragon A, et al.

Dynamic fragmentation of laser shock-melted tin: Experiment and modelling

[J]. Int. J. Fract., 2010, 163: 109

de Rességuier T, Signor L, Dragon A, et al.

Experimental investigation of liquid spall in laser shock-loaded tin

[J]. J. Appl. Phys., 2007, 101: 013506

Signor L, de Rességuier T, Dragon A, et al.

Investigation of fragments size resulting from dynamic fragmentation in melted state of laser shock-loaded tin

[J]. Int. J. Impact Eng., 2010, 37: 887

Chen Y T, Hong R K, Chen H Y, et al.

An improved Asay window technique for investigating the micro-spall of an explosively-driven tin

[J]. Rev. Sci. Instrum., 2017, 88: 013904

Vogan W S, Anderson W W, Grover M, et al.

Piezoelectric characterization of ejecta from shocked tin surfaces

[J]. J. Appl. Phys., 2005, 98: 113508

[本文引用: 1]

Zaretsky E B.

Experimental determination of the dynamic tensile strength of liquid Sn, Pb, and Zn

[J]. J. Appl. Phys., 2016, 120: 025902

[本文引用: 2]

Zaretsky E B, Kanel G I.

Response of copper to shock-wave loading at temperatures up to the melting point

[J]. J. Appl. Phys., 2013, 114: 083511

Razorenov S V, Savinykh A S, Zaretsky E B.

Elastic-plastic deformation and fracture of shock-compressed single-crystal and polycrystalline copper near melting

[J]. Tech. Phys., 2013, 58: 1437

Kanel G I, Razorenov S V, Bogatch A, et al.

Spall fracture properties of aluminum and magnesium at high temperatures

[J]. J. Appl. Phys., 1996, 79: 8310

[本文引用: 1]

Garkushin G V, Kanel G I, Savinykh A S, et al.

Influence of impurities on the resistance to spall fracture of aluminum near the melting temperature

[J]. Int. J. Fract., 2016, 197: 185

Kanel G I, Razorenov S V, Baumung K, et al.

Dynamic yield and tensile strength of aluminum single crystals at temperatures up to the melting point

[J]. J. Appl. Phys., 2001, 90: 136

[本文引用: 2]

Xiang M Z, Hu H B, Chen J, et al.

Molecular dynamics simulations of micro-spallation of single crystal lead

[J]. Modell. Simul. Mater. Sci. Eng., 2013, 21: 055005

[本文引用: 1]

Shao J L, Wang P, He A M, et al.

Spall strength of aluminium single crystals under high strain rates: molecular dynamics study

[J]. J. Appl. Phys., 2013, 114: 173501

[本文引用: 1]

Mayer A E, Mayer P N.

Strain rate dependence of spall strength for solid and molten lead and tin

[J]. Int. J. Fract., 2020, 222: 171

[本文引用: 4]

Mayer A E, Mayer P N.

Evolution of pore ensemble in solid and molten aluminum under dynamic tensile fracture: Molecular dynamics simulations and mechanical models

[J]. Int. J. Mech. Sci., 2019, 157-158: 816

[本文引用: 2]

Mayer P N, Mayer A E.

Evolution of foamed aluminum melt at high rate tension: A mechanical model based on atomistic simulations

[J]. J. Appl. Phys., 2018, 124: 035901

[本文引用: 2]

Ullmaier H.

Radiation damage in metallic materials

[J]. MRS Bull., 1997, 22: 14

[本文引用: 2]

Trinkaus H, Singh B N.

Helium accumulation in metals during irradiation—Where do we stand?

[J]. J. Nucl. Mater., 2003, 323: 229

[本文引用: 3]

Hou P F, Wang X H, Liu Y X, et al.

A neutron irradiation-induced displacement damage of indium vacancies in α-In2Se3 nanoflakes

[J]. Phys. Chem. Chem. Phys., 2020, 22: 15799

Xie H X, Gao N, Xu K, et al.

A new loop-punching mechanism for helium bubble growth in tungsten

[J]. Acta Mater., 2017, 141: 10

Schwartz A J, Wall M A, Zocco T G, et al.

Characterization and modelling of helium bubbles in self-irradiated plutonium alloys

[J]. Philos. Mag., 2005, 85: 479

[本文引用: 4]

Li Q, Parish C M, Powers K A, et al.

Helium solubility and bubble formation in a nanostructured ferritic alloy

[J]. J. Nucl. Mater., 2014, 445: 165

[本文引用: 1]

Chung B W, Thompson S R, Lema K E, et al.

Evolving density and static mechanical properties in plutonium from self-irradiation

[J]. J. Nucl. Mater., 2009, 385: 91

[本文引用: 1]

Schäublin R, Chiu Y L.

Effect of helium on irradiation-induced hardening of iron: A simulation point of view

[J]. J. Nucl. Mater., 2007, 362: 152

Osetsky Y N, Stoller R E.

Atomic-scale mechanisms of helium bubble hardening in iron

[J]. J. Nucl. Mater., 2015, 465: 448

Ding M S, Tian L, Han W Z, et al.

Nanobubble fragmentation and bubble-free-channel shear localization in helium-irradiated submicron-sized copper

[J]. Phys. Rev. Lett., 2016, 117: 215501

[本文引用: 1]

Ding M S, Du J P, Wan L, et al.

Radiation-induced helium nanobubbles enhance ductility in submicron-sized single-crystalline copper

[J]. Nano Lett., 2016, 16: 4118

[本文引用: 1]

Li S H, Zhang J, Han W Z.

Helium bubbles enhance strength and ductility in small-volume Al-4Cu alloys

[J]. Scr. Mater., 2019, 165: 112

[本文引用: 1]

Glam B, Eliezer S, Moreno D, et al.

Dynamic fracture and spall in aluminum with helium bubbles

[J]. Int. J. Fract., 2010, 163: 217

[本文引用: 2]

Glam B, Strauss M, Eliezer S, et al.

Shock compression and spall formation in aluminum containing helium bubbles at room temperature and near the melting temperature: Experiments and simulations

[J]. Int. J. Impact Eng., 2014, 65: 1

[本文引用: 3]

Li Y H, Chang J Z, Zhang L, et al.

Experimental investigation of spall damage in pure aluminum with helium bubbles

[J]. Chin. J. High Press. Phys., 2021, 35: 054101

[本文引用: 1]

李英华, 常敬臻, 张 林 .

氦泡铝的层裂特性实验研究

[J]. 高压物理学报, 2021, 35: 054101

[本文引用: 1]

Xiao D W, He L F, Zhou P, et al.

Spall in aluminium with helium bubbles under laser shock loading

[J]. Chin. Phys. Lett., 2017, 34: 056201

[本文引用: 1]

Qi M L, He H L, Wang Y G, et al.

Dynamic analysis of helium bubble growth in the pure al under high strain-rate loading

[J]. Chin. J. High Press. Phys., 2007, 21: 145

[本文引用: 1]

祁美兰, 贺红亮, 王永刚 .

高应变率拉伸下纯铝中氦泡长大的动力学研究

[J]. 高压物理学报, 2007, 21: 145

[本文引用: 1]

Zhang F G, Hu X M, Wang P, et al.

Numerical analysis of spall response in aluminum with helium bubbles

[J]. Expl. Shock Waves, 2017, 37: 699

[本文引用: 1]

张凤国, 胡晓棉, 王 裴 .

含氦泡金属铝层裂响应的数值分析

[J]. 爆炸与冲击, 2017, 37: 699

[本文引用: 1]

Kubota A, Reisman D B, Wolfer W G.

Dynamic strength of metals in shock deformation

[J]. Appl. Phys. Lett., 2006, 88: 241924

[本文引用: 1]

Wang H Y, Li X S, Zhu W J, et al.

Atomistic modelling of the plastic deformation of helium bubbles and voids in aluminium under shock compression

[J]. Radiat. Eff. Defects Solids, 2014, 169: 109

[本文引用: 4]

Shao J L, Wang P, He A M, et al.

Influence of voids or He bubbles on the spall damage in single crystal Al

[J]. Modell. Simul. Mater. Sci. Eng., 2014, 22: 025012

[本文引用: 1]

Zhou T T, He A M, Wang P, et al.

Spall damage in single crystal Al with helium bubbles under decaying shock loading via molecular dynamics study

[J]. Comput. Mater. Sci., 2019, 162: 255

[本文引用: 2]

Zhou T T, He A M, Wang P.

Dynamic evolution of He bubble and its effects on void nucleation-growth and thermomechanical properties in the spallation of aluminum

[J]. J. Nucl. Mater., 2020, 542: 152496

[本文引用: 2]

Zhou T T, Zhao F Q, Zhou H Q, et al.

Atomistic simulation and continuum modeling of the dynamic tensile fracture and damage evolution of solid single crystalline al with He bubble

[J]. Int. J. Mech. Sci., 2022, 234: 107681

[本文引用: 9]

Zope R R, Mishin Y.

Interatomic potentials for atomistic simulations of the Ti-Al system

[J]. Phys. Rev., 2003, 68B: 024102

[本文引用: 1]

Young D A, McMahan A K, Ross M.

Equation of state and melting curve of helium to very high pressure

[J]. Phys. Rev., 1981, 24B: 5119

[本文引用: 1]

Plimpton S.

Fast parallel algorithms for short-range molecular dynamics

[J]. J. Comput. Phys., 1995, 117: 1

[本文引用: 1]

Stukowski A.

Visualization and analysis of atomistic simulation data with OVITO—The open visualization tool

[J]. Modell. Simul. Mater. Sci. Eng., 2010, 18: 015012

[本文引用: 1]

Reif F. Fundamentals of Statistical and Thermal Physics [M]. New York: McGraw-Hill, 1965: 1

[本文引用: 1]

Ibach H. Physics of Surfaces and Interfaces [M]. Berlin: Springer, 2006: 1

[本文引用: 1]

Agranat M B, Anisimov S I, Ashitkov S I, et al.

Strength properties of an aluminum melt at extremely high tension rates under the action of femtosecond laser pulses

[J]. JETP Lett., 2010, 91: 471

[本文引用: 1]

Kuksin A, Norman G, Stegailov V, et al.

Dynamic fracture kinetics, influence of temperature and microstructure in the atomistic model of aluminum

[J]. Int. J. Fract., 2010, 162: 127

[本文引用: 1]

Ashitkov S I, Agranat M B, Kanel G I, et al.

Behavior of aluminum near an ultimate theoretical strength in experiments with femtosecond laser pulses

[J]. JETP Lett., 2010, 92: 516

Ashitkov S I, Komarov P S, Ovchinnikov A V, et al.

Strength of liquid tin at extremely high strain rates under a femtosecond laser action

[J]. JETP Lett., 2016, 103: 544

Kanel G I.

Spall fracture: Methodological aspects, mechanisms and governing factors

[J]. Int. J. Fract., 2010, 163: 173

[本文引用: 2]

Grady D E.

Strain-rate dependence of the effective viscosity under steady-wave shock compression

[J]. Appl. Phys. Lett., 1981, 38: 825

[本文引用: 1]

Chivapornthip P, Bohez E L J.

Dependence of bulk viscosity of polypropylene on strain, strain rate, and melt temperature

[J]. Polym. Eng. Sci., 2017, 57: 830

[本文引用: 1]

/