含氧纳米多晶 α-Ti拉伸力学性能与变形机制的分子动力学模拟

图1 计算模型示意图

Fig.1 Schematic diagram of calculation model (The left side grains G1-G5 correspond to the right side grains G1′-G5′, respectively)

表1 纯Ti、O的2NNMEAM + Qeq参数^[15]

Table 1 2NNMEAM + Qeq parameters of pure Ti and O^[15]

Parameter	hcp Ti	Dimer O	Unit
E_c	4.87	2.56	eV
R_e	0.292	0.121	nm
α	4.7195	6.8800	-
A	0.66	1.44	-
t⁽¹⁾	6.80	0.10	-
t⁽²⁾	-2.00	0.11	-
t⁽³⁾	-12.00	0.00	-
β⁽⁰⁾	2.70	5.47	-
β⁽¹⁾	1.00	5.30	-
β⁽²⁾	3.00	5.18	-
β⁽³⁾	1.00	5.57	-
C_min	1.00	2.00	-
C_max	1.44	2.80	-
χ⁰	-1.192	10.11	eV·e^-1
J⁰	8.436	20.5	eV·e^-2
ΔE⁽²⁾	7.97	5.63	eV
ΔE⁽³⁾	10⁵^a	10⁵^a	eV
ζ	8.3	23.9	nm^-1
Z	1.408	0.00	e

Note:^a —an arbitrary large value, 2NNMEAM—second nearest-neighbor modified embedded-atom method, Qeq—charge equilibration, E_c—cohesive energy, R_e—equilibrium nearest neighbor distance, α—exponential decay factor for the universal energy function, A—scaling factor for embedding energy, t⁽¹⁾-t⁽³⁾—weighting factor for atomic density, β⁽⁰⁾-β⁽³⁾—exponential decay factor for atomic densities, C_min and C_max—shielding parameters, χ⁰—electronegativity, J⁰—atomic hardness or the self-Coulomb repulsion, ΔE⁽ⁿ⁾ (n = 2, 3)—energy of stable charge, ζ—valence orbital exponent, Z—charge number

新窗口打开| 下载CSV

表2 二元系统Ti-O的2NNMEAM参数^[15]

Table 2 2NNMEAM parameters for the Ti-O binary systems^[15]

Parameter	Ti-O	Unit
ΔE_c	1.5280	eV
R_e	0.20649	nm
α	7.4455	-
d	0.01	-
C_min(_{i j i}₎	1.00	-
C_min(_{j i j}₎	0.97	-
C_min(_{i i j}₎	1.46	-
C_min(_{i j j}₎	1.46	-
C_max(_{i j i}₎	2.80	-
C_max(_{j i j}₎	2.51	-
C_max(_{i i j}₎	2.80	-
C_max(_{i j j}₎	2.80	-
ρ₀(O) / ρ₀(Ti)	12.0	-

Note:d—adjustable parameter, ρ₀—atomic density scaling, i and j indicate cationic elements Ti and oxygen, respectively

新窗口打开| 下载CSV

E^{T o t a l} = E^{M E A M} (X) + E^{E S} (X, q)

(1)

式中，E^MEAM和E^ES分别为非静电和静电能量，X = {X₁，X₂，…，X_N }和q = {q₁，q₂，…，q_N }分别是原子位置和电荷的可变集合。非静电项和静电项相互独立。Rose通用状态方程(EOS)^[16]很好地描述了金属和共价固体中总能量和原子间距离之间的普遍关系。因此系统总能量使用基于Rose状态方程的2NNMEAM项(式(2))以及新引入的静电项(式(3))表示。非静电能量近似为：

E^{M E A M} = \sum_{i} [F_{i} (\bar{ρ_{i}}) + \frac{1}{2} \sum_{(j \neq i)} ϕ_{i j} (R_{i j})]

(2)

式中，F_i 是嵌入函数， $\bar{ρ_{i}}$ 是位置i处的电子密度， $ϕ$ _ij (R_ij )是原子 $i$ 和原子 $j$ 在距离R_ij 处的对相互作用。

静电能用原子能 $E_{i}^{a t o m}$ 和Coulomb对相互作用 $V_{i j}^{C o u l}$ 之和表示：

E^{E S} = \sum_{i}^{N} E_{i}^{a t o m} (q_{i}) + \sum_{i, j (i \neq j)}^{N} \frac{1}{2} V_{i j}^{C o u l} (q_{i}, q_{j}, R_{i j})

(3)

建立模型过程中，利用最速下降法使模型的能量最小化以消除模型中的不合理结构。然后采用Nose-Hoover热浴法控制体系温度达到目标温度，采用等温等压(NPT)系综进行弛豫10 ps。系统达到平衡后，在X方向施加拉伸应变，模拟的时间步长均为1 fs。为了消除尺度限制带来的边界效应，模型的X、Y、Z方向均采用周期性边界。计算采用LAMMPS开源软件^[17]。使用ATOMSK^[18]构建分子动力学模型，采用OVITO^[19]和ATOMEYE^[20]软件对结果可视化处理，采用共近邻原子分析(CNA)^[21]方法分析变形过程中的结构变化，并采用位错分析(DXA)方法^[22]分析加载过程中的位错变化。

2 计算结果

2.1 力学性能

图2a为不同O含量的多晶α-Ti在温度为300 K、应变速率为5 × 10⁸ s^-1下拉伸的应力-应变曲线。可以明显观察到O含量的增加使多晶α-Ti的屈服应力不同程度的增加。间隙O原子造成周围形成晶格畸变应力场，该应力场与位错应力场产生交互作用，使位错运动受阻，导致屈服应力增大。图2a中，应力上升阶段曲线的重合部分为弹性变形阶段，达到屈服点时应力最大，随后进入塑性变形阶段，含O多晶α-Ti的拉伸应力整体上随应变的增加而降低。图中多晶Ti-0.1O拉伸曲线可观察到明显的加工硬化现象(AB部分)。

图2

图2 不同条件下纳米多晶α-Ti拉伸应力-应变曲线

Fig.2 Tensile stress-strain curves of polycrystalline α-Ti with O content at 300 K and 5.0 × 10⁸ s^-1 (a) and of Ti-0.3O with temperature at 5.0 × 10⁸ s^-1 (b) and strain rate at 300 K (c)

选择力学性能相对优异的多晶Ti-0.3O研究温度对多晶α-Ti力学性能和变形机制的影响。图2b为300、500和700 K下多晶Ti-0.3O的拉伸应力-应变曲线。随着温度升高，含O多晶α-Ti中原子的平均动能增大，原子之间更容易发生相对移动进而造成原子间相互作用力降低，导致宏观性能下降。同时，温度升高会减弱间隙O原子对位错的钉扎阻碍作用，使含O纳米多晶α-Ti的屈服应力降低。

图2c给出了温度为300 K、不同应变速率下多晶Ti-0.3O的拉伸应力-应变曲线。含O多晶α-Ti的屈服应力随着应变速率的增高而增大。而且在高应变速率下，多晶体的塑性变形也更倾向于均匀变形。在单晶Ni的分子动力学模拟中发现，高应变速率时发生非晶化转变且变形均匀化^[23,24]。

2.2 间隙O对变形机制的影响

图3所示为图2a应力-应变曲线中AB部分的原子截图。本文图3、6~9和12中红色原子代表hcp结构，绿色原子代表fcc结构，蓝色原子代表bcc结构，白色原子代表晶界原子以及OVITO无法识别的紊乱原子，用OTHER表示。应变为0.064时，间隙O所在位置出现位错缠结，如图3d黑色虚线圆圈所示，位错缠结导致位错运动受阻，因此造成加工硬化，这种缠结效果并不能激活新的位错。可见α-Ti中添加间隙O原子可以增加其硬度，然而同时也容易导致材料变脆。而在复合材料中，α-Ti中引入层状石墨烯后力学性能显著提高^[25]。在Ti层较薄时，应变诱发位错的传播受到显著限制，从而导致更多的位错诱发滑移，产生独特的塑性变形。图3c中黑色箭头所指的原子半径小的白色原子为O原子。删除理想排列的原子后可以看到晶粒中的位错缠结和堆垛层错。

图3

图3 多晶Ti-0.1O拉伸后位错缠结引起的加工硬化

Fig.3 Work hardening caused by dislocation tangle after tensile of polycrystalline Ti-0.1O (The red atom represents the hcp structure, the green atom the fcc structure, the blue atom the bcc structure, and the white atom the grain boundary atom and the disordered atom being not able to be recognized by OVITO, which are represented by OTHER. Same as in Figs.6-9 and 12) (a, c) atomic diagrams at strains ε = 0 (a) and 0.064 (c), respectively (b) diagram with only fcc atoms and dislocation lines at ε = 0.062 (d) dislocation tangle diagram with only fcc atoms and dislocation lines at ε = 0.064

图4

图4 不同O含量多晶α-Ti总位错密度随应变变化曲线

Fig.4 Total dislocation density with various O contents and strains

图5

图5 不同O含量多晶α-Ti拉伸产生的位错类型

(a) Ti (b) Ti-0.1O (c) Ti-0.3O (d) Ti-0.5O (e) Ti-0.7O (f) Ti-1.0O

Fig.5 Activated dislocation types in polycrystalline α-Ti under tensile loading at different O contents

图6

图6 {10 $\bar{1}$ 1}<10 $\bar{1}$ $\bar{2}$ >孪晶原子图和“孪晶位错”

Fig.6 Atomic structure of twin{10 $\bar{1}$ 1}<10 $\bar{1}$ $\bar{2}$ > (a) and “zonal dislocation” on twin plane (b)

图7

图7 300 K下多晶Ti-0.1O拉伸过程中层错的形成

Fig.7 Formation of stacking faults during tensile of polycrystalline Ti-0.1O at 300 K and ε = 0.191

图8

图8 300 K时多晶Ti-0.3O的Ti_hcp→Ti_bcc→Ti_hcp转变过程

Fig.8 Snapshots of transformation process of Ti_hcp→Ti_bcc→Ti_hcp of polycrystalline Ti-0.3O at 300 K (a-c) Ti_hcp→Ti_bcc (d-f) Ti_bcc→Ti_hcp (g) oxygen atom pinning grain boundary

图9

图9 不同温度下多晶Ti-0.3O拉伸过程中晶界的运动

Fig.9 Grain boundary movement of polycrystalline Ti-0.3O at 300 K (a-c), 500 K (d-f), and 700 K (g-i) with ε = 0 (a, d, g), ε = 0.086 (b, e, h), and ε = 0.166 (c, f, i) (Black dotted circles in Figs.9b, c, e, f, h, and i represent the disordered atoms caused by lattice distortion after adding interstitial O atoms, and dotted yellow circles represent the disordered atoms produced during the strain at different temperatures)

图10

图10 不同应变速率下多晶Ti-0.3O中各类原子含量随应变的变化

Fig.10 Atomic fraction curves of bcc (a), hcp (b), fcc (c), and OTHER (d) atoms at different strain rates

图11

图11 不同应变速率下多晶Ti-0.3O拉伸过程中晶粒的数量

Fig.11 Number of grains in polycrystalline Ti-0.3O under tensile loading at different strain rates

图12

图12 多晶Ti-0.3O拉伸过程中不稳定相变的产生以及相变机理结构示意图

Fig.12 Unstable phase transition during tenison of polycrystalline Ti-0.3O (a-c), and schematic of Ti_hcp→Ti_bccphase transition mechanism (d) (μ—a parameter describing relative slip distance)

不同O含量下多晶α-Ti拉伸时总位错密度随应变的变化如图4所示。总位错密度随O含量的增高而降低。300 K下拉伸不同O含量的多晶α-Ti，在零应变时，所有的纳米晶样品中都存在不同数量的弛豫位错(集中在晶界区域)。这种晶界位错较短而且不稳定，在外加载荷下会被晶界吸收而湮灭，对应图4中总位错密度在应变开始时有降低趋势。随着应变增加，位错源被激活，产生大量位错，位错密度升高。在位错增殖和湮灭的共同作用下，位错密度也呈现出高低起伏的趋势。变形过程中位错类型主要为全位错 $\frac{1}{3}$ <11 $\bar{2}$ 0>、Schockley不全位错 $\frac{1}{3}$ < $\bar{1}$ 100>以及无法识别的位错、缺陷(OTHER)。多晶α-Ti中间隙O的加入使得局部范围内的原子偏离其正常的点阵平衡位置，造成点阵畸变。在金属材料中点阵畸变会引起晶体能量升高，阻碍位错运动使得位错数量减少。

低O含量多晶α-Ti在拉伸加载时，通常首先激活柱面<a>滑移和变形孪晶。随着应变增加，<a>位错会发生分解反应，生成不全位错。图5为不同O含量多晶α-Ti在拉伸应变为0.2时产生的位错与孪晶。图5a和b中观察到平直且平行的红色位错(黑色虚线圆圈所示)定义为“孪晶位错”^[26]，这种“孪晶位错”通常称之为“带状位错”(zonal dislocation)，协调孪晶长大。将图5a中下方黑色虚线圆圈部分的原子图放大如图6所示。图6b为孪晶面和孪晶面上的“孪晶位错”的原子截图。图中红色线条为“孪晶位错”，整齐排列一层的白色原子组成的面是孪晶边界。晶体学分析显示该孪晶类型为{10 $\bar{1}$ 1}<10 $\bar{1}$ $\bar{2}$ >，图6a为孪晶{10 $\bar{1}$ 1}<10 $\bar{1}$ $\bar{2}$ >的晶格结构图。观察到孪晶面两边的原子排布呈镜像对称。绿色原子为堆垛层错。孪晶界(10 $\bar{1}$ 1)由指向孪晶方向 $\frac{1}{2}$ [10 $\bar{1}$ $\bar{2}$ ]的箭头标记。孪晶{10 $\bar{1}$ 1}<10 $\bar{1}$ $\bar{2}$ >由孪晶界上的“孪晶位错”协调长大。孪晶位错Burgers矢量 b = $\frac{1}{2}$ [10 $\bar{1}$ $\bar{2}$ ]，这种孪晶位错能量较高而且不够稳定，通常分解为2个Burgers矢量为 b = $\frac{1}{2} \cdot \frac{1}{2}$ [10 $\bar{1}$ $\bar{2}$ ]的“带状位错”^[27]。在纯Ti的塑性变形中，孪晶{10 $\bar{1}$ 1}<10 $\bar{1}$ $\bar{2}$ >被认为是压缩孪晶，说明含O多晶α-Ti在拉伸加载时，垂直于拉伸方向的压应力使晶粒锥面{10 $\bar{1}$ 1}孪晶滑移系激活，发生孪晶变形。

体系内孪晶的数量随着O含量的增加而减小^[10]，含量增加到0.3%及以上时，几乎没有“孪晶位错”的出现，如图5c~f所示，位错类型受晶粒尺寸限制，主要以排列无序的刃型位错为主。透射电子显微镜研究^[28]表明，对于在室温下拉伸变形的CP-Ti，垂直和平行于柱面的位错主要是螺型位错。增加O含量到0.3%产生的主要是刃型位错，位错几乎没有对齐的趋势^[29]。当O含量超过0.3%时，在多晶α-Ti中观察到了锥面<c+a>位错 $\frac{1}{3}$ <1 $\bar{2}$ 13>和基面位错，如图5c中橘黄色的 $\frac{1}{3}$ <1 $\bar{1}$ 00>位错、图5d中黄色的 $\frac{1}{3}$ <1 $\bar{2}$ 13>位错、图5e中蓝色的<0001>位错和图5f中绿色的 $\frac{1}{3}$ <1 $\bar{2}$ 10>位错。“带状位错”具有较小的Burgers矢量，在OVITO中通常被识别为OTHER，这也是OVITO统计结果中OTHER位错含量较高的一个主要原因，但是实际孪晶位错整齐排列于孪晶界，促进孪晶长大。O含量对柱面滑移临界分切应力(CRSS_prismatic)的影响高于锥面滑移临界分切应力(CRSS_pyramidal)，导致CRSS_pyramidal / CRSS_prismatic比降低和c / a比的增加，从而O含量增加导致柱面密排面效果减弱，锥面滑移在较高O含量多晶体中更容易被激活^[10]。

在纳米晶粒中，位错运动区域有限，全位错更容易分解为不全位错产生堆垛层错。图7为300 K下多晶Ti-0.1O拉伸应变为0.191时产生层错的原子图。图中绿色原子为fcc结构原子，箭头指向的橙色位错为 $\frac{1}{3}$ <1 $\bar{1}$ 00>位错，绿色位错为 $\frac{1}{3}$ [11 $\bar{2}$ 0]位错。可以看到，全位错 $\frac{1}{3}$ [11 $\bar{2}$ 0]分解为头位错 $\frac{1}{3}$ [10 $\bar{1}$ 0]和尾位错 $\frac{1}{3}$ [01 $\bar{1}$ 0]，中间为具有fcc结构的堆垛层错。其位错反应式可以表示为^[30]： $\frac{1}{3}$ [11 $\bar{2}$ 0]→ $\frac{1}{3}$ [10 $\bar{1}$ 0] + $\frac{1}{3}$ [01 $\bar{1}$ 0]。

2.3 温度对变形机制的影响

在纳米多晶Ti-0.3O塑性变形过程中fcc结构原子以层错的形式存在，观察到300 K下堆垛层错比较宽，700 K下层错明显减少或变小。同时观察到具有bcc结构的原子，图8给出了300 K时多晶Ti-0.3O应变过程中bcc相从出现到消失的全过程。首先在结构较混乱的晶界区域出现bcc结构原子，即在晶界处优先形成新相晶核如图8b所示，形成了一个hcp结构和bcc结构共存的新晶粒。从图8a~c中可以看出，随着应变的增加，Ti_hcp→Ti_bcc转变增多，相变不仅伴随晶粒旋转还伴随着晶界的运动。应变大于0.125后bcc结构逐渐变小直至消失，如图8d~f所示。在整个Ti_hcp→Ti_bcc→Ti_hcp转变过程中，生成新晶粒G2。可见新晶粒的生成与晶粒旋转有密切的关系。Li等^[31]用实验展示了单个晶粒旋转和相邻晶粒旋转/生长的完整过程。揭示了诱导晶粒旋转和生长是纳米多晶材料的塑性变形机制之一。

间隙O对晶界运动有阻碍作用。新晶粒G2和晶粒G1之间的晶界受到间隙O原子的阻碍，如图8f黑色虚线圆圈圈出部分。图8g是黑色虚线圆圈部分的放大图，图中原子半径小的红色原子为O原子，白色Ti原子形成的条状代表晶界，黑色粗箭头为晶界的运动方向。O原子阻碍了其所在部分晶界的运动，使存在O原子的部分晶界运动明显慢于同晶界其他部分。

纳米晶的变形机制主要是基于晶界的晶界滑动、晶界旋转^[32]等机制。图9是多晶Ti-0.3O分别在300、500和700 K温度下沿X轴拉伸晶界运动的原子截图。比较图9应变为0.086 (图9b、e和h)和应变为0.166 (图9c、f和i)时晶界(黑色虚线标记)运动后的位置，700 K时相同应变下晶界运动明显快于500和300 K时。研究^[33]发现当温度升高或晶粒尺寸减小时，晶界滑动成为主要机制，并能有效地协调塑性变形。另外标记晶粒G1和晶粒G2中的紊乱原子如图9中的虚线圆圈所示，其中黑色虚线圆圈是加入间隙O原子后引起晶格畸变造成的紊乱原子，黄色虚线圆圈为不同温度下应变过程中产生的紊乱原子。这些紊乱原子中的一部分来自于晶界，随着拉伸进行晶界原子向晶粒内迁移，使得晶界变薄。同样比较图9b、e、h和图9c、f、i中晶粒G1和晶粒G2中的黄色虚线圆圈看出，随着温度升高，晶界周围紊乱原子增多，这是因为温度升高，原子的活性大幅度增加，打破了原子间结合力，使得原子扩散到周围的区域。

2.4 应变速率对变形机制的影响

图10为不同应变速率时多晶Ti-0.3O拉伸过程中bcc、hcp、fcc、OTHER原子含量占比随应变的变化曲线。观察图10b和d，不同应变速率下hcp和OTHER原子均在屈服点0.07左右出现大幅度转变，转变速率随着应变速率的增加而增加。应变速率越高，hcp和OTHER转换的原子数越多。观察图10a中3个应变速率下bcc原子占比曲线，在应变为0.10~0.15之间bcc原子占比出现先增大后减小的浮动，说明晶体内部出现Ti_hcp→Ti_bcc→Ti_hcp转变，这种原子转变伴随着新晶粒的形成。图11统计了1.0 × 10⁸、5.0 × 10⁸和1.0 × 10⁹ s^-1下多晶Ti-0.3O不同应变下的晶粒数量。多晶Ti-0.3O的初始晶粒有8个，随着应变的进行生成不同数量的新晶粒。统计发现生成新晶粒的数量与应变速率存在一定的线性关系，即含O多晶α-Ti应变速率越高生成新晶粒的数量越多。不同应变速率下含O多晶α-Ti在应变区间0.10~0.15时生成新晶粒的数量最多，这与图10a中bcc原子数量占比的起伏相对应，说明生成新晶粒的过程中确实伴随着不稳定的相变。

3 分析讨论

在低温下，纯Ti和大多数Ti合金为hcp结构，当温度高于(882 ± 2)℃时转变为稳定的bcc结构，发生同素异构转变。然而在300 K纳米多晶Ti-0.3O拉伸时观察到了不稳定的bcc结构，如图12a~c中蓝色原子所示。晶界上的Ti原子受到拉伸载荷作用，平面内晶粒取向改变，2个晶粒之间相对旋转导致出现不稳定的bcc晶体结构。比较图12b和c发现，当应变从0.088增加到0.118时G1相对旋转了1.72°，G2相对旋转了2.03°。图12d显示了Ti_hcp→Ti_bcc相变机理的结构示意图。黄色和粉色的原子球分别表示在[00 $\bar{1}$ ]_bcc方向或[1 $\bar{2}$ 10]_hcp方向上，原子位于不同的(00 $\bar{1}$ )_bcc面或(1 $\bar{2}$ 10)_hcp面，虚线表示bcc晶格，细实线表示相变过程中原子的滑移面，箭头表示原子的滑移方向。参数µ用于描述相邻的(110)_bcc面沿[ $\bar{1}$ 10]_bcc方向(或[10 $\bar{1}$ 0]_hcp方向)的相对滑移距离。滑移完成即hcp转变为bcc相。Chen等^[34]在研究垂直于c轴的单晶Ti在单轴压缩下的变形过程中也发现了这种不稳定的相变现象，表明相变是由晶格转换造成的，具体的晶格转换可以描述为(10 $\bar{1}$ 0)_hcp→(110)_bcc。

不同温度下多晶Ti-0.3O拉伸过程bcc原子占比如图13所示，表明bcc原子占比随着应变的增加先增加后减小。多晶Ti-0.3O塑性变形过程中的主要机制是Ti_hcp→Ti_bcc→Ti_hcp。图13中还观察到高温下Ti_hcp→Ti_bcc→Ti_hcp转变的峰值点所对应的应变值小，高温载荷下更利于晶粒旋转而发生相变。

图13

图13 不同温度下多晶Ti-0.3O拉伸过程bcc原子占比与应变关系

Fig.13 Relationship between proportion of bcc atoms and strain during tension of polycrystalline Ti-0.3O at different temperatures

模拟单晶Ti^[34]和多晶Ti^[35]在压缩载荷下的塑性变形过程中都发现Ti_hcp→Ti_bcc不稳定相变。为了定量评估纯Ti的不同晶体结构的能量和晶格转变势垒，了解纳米多晶Ti变形过程中不稳定相变的原因，Zope和Mishin^[36]计算了Ti_hcp和Ti_bcc的平衡晶格能，计算结果显示Ti_hcp在平衡晶格参数(0.295 nm)处具有最低能量(-4.85 eV/atom)。该能量比Ti_bcc低30 meV/atom。Mehl和Papaconstantopoulos^[37]使用密度泛函理论(DFT)计算晶体结构的晶格能量得出Ti_hcp的能量最低，Ti_bcc的能量最高，Ti_hcp和Ti_bcc之间的能量差为67 meV/atom。动态Ti_hcp→Ti_bcc转换的能垒通过原子模拟计算出约为38 meV/atom^[34]。结合原子计算结果和DFT计算结果，可见2种密排结构的Ti_bcc和Ti_hcp之间的能量非常接近。因此，不稳定相变是由外部应变触发的。

4 结论

(1) 通过分子动力学计算了拉伸载荷下不同O含量纳米多晶α-Ti以及多晶Ti-0.3O在不同温度、速率下的力学性能和变形机制。结果表明，含O纳米多晶α-Ti的变形机制主要为位错滑移、晶粒旋转、晶界滑移和晶界迁移。

(2) 间隙O原子造成晶格畸变阻碍位错运动，使含O多晶α-Ti的屈服应力随O含量的增加而增大，位错密度随O含量的增加而减小。O含量小于0.3%时，观察到变形孪晶{10 $\bar{1}$ 0}<10 $\bar{1}$ 2>，该孪晶通过孪晶面上的“带状位错”协调长大。O含量增加到0.3%及以上，孪晶数量减少，位错以排列无序的刃型位错为主。

(3) 多晶Ti-0.3O在不同温度下的拉伸实验表明，随着温度的升高，屈服应力、位错密度降低。拉伸过程中由于晶粒旋转造成Ti_hcp→Ti_bcc→Ti_hcp的不稳定相变，进而产生新晶粒，且高温载荷下更利于此现象产生。高温下晶界运动速率明显加快，晶界上O原子的钉扎作用阻碍钉扎部位及附近晶界迁移，使其明显慢于同一晶界其他部位的迁移。

(4) 多晶Ti-0.3O在拉伸载荷下生成新晶粒的个数随应变速率的升高而增加，新晶粒的产生伴随不稳定相变并且使晶界原子数增多，因此hcp与OTHER原子之间的转换随着应变速率的增大而增多。

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]

Tabie

V M

, Li

, Saifu

, et al.

Mechanical properties of near alpha titanium alloys for high-temperature applications—A review

[J]. Aircr. Eng. Aerosp. Technol., 2020, 92: 521

[2]

Gao

P F

, Fu

M W

, Zhan

, et al.

Deformation behavior and microstructure evolution of titanium alloys with lamellar microstructure in hot working process: A review

[J]. J. Mater. Sci. Technol., 2020, 39: 56

DOI [本文引用: 1]

Titanium alloys have been widely used in many industrial clusters such as automotive, aerospace and biomedical industries due to their excellent comprehensive properties. In order to obtain fine microstructures and favorable properties, a well-designed multi-step thermomechanical processing (TMP) is critically needed in manufacturing of titanium components. In making of titanium components, subtransus processing is a critical step to breakdown lamellar microstructure to fine-structure in hot working process and thus plays a key role in tailoring the final microstructure and properties. To realize this goal, huge efforts have been made to investigate the mechanisms of microstructure evolution and flow behavior during the subtransus processing. This paper reviews the recent experimental and modelling progresses, which aim to provide some guidelines for the process design and microstructure tailoring for titanium alloy research community. The characteristics of the initial lamellar microstructure are presented, followed by the discussion on microstructure evolution during subtransus processing. The globularization of lamellar α is analyzed in detail from three aspects, i.e., globularization mechanism, heterogeneity and kinetics. The typical features of flow behaviors and the explanations of significant flow softening are then summarized. The recent advances in modelling of microstructure evolution and flow behaviors in the subtransus processing are also articulated. The current tantalized issues and challenges in understanding of the microstructure evolution and flow behaviors of the titanium alloys with lamellar microstructure are presented and specified in future exploration of them.

[3]

Dong

Y H

, Chen

I W

Onset criterion for flash sintering

[J]. J. Am. Ceram. Soc., 2015, 98: 3624

[4]

Mahata

, Sikdar

Molecular dynamics simulation of nanometer scale mechanical properties of hexagonal Mg-Li alloy

[J]. J. Magnes. Alloy., 2016, 4: 36

[5]

Suryanarayana

Structure and properties of nanocrystalline materials

[J]. Bull. Mater. Sci., 1994, 17: 307

[6]

Hajizadeh

, Alamdari

S G

, Eghbali

Stored energy and recrystallization kinetics of ultrafine grained titanium processed by severe plastic deformation

[J]. Physica, 2013, 417B: 33

[7]

Huang

Q S

, Zhu

, Chen

Y B

, et al.

Twinning-assisted dynamic adjustment of grain boundary mobility

[J]. Nat. Commun., 2021, 12: 6695

DOI PMID [本文引用: 1]

Grain boundary (GB) plasticity dominates the mechanical behaviours of nanocrystalline materials. Under mechanical loading, GB configuration and its local deformation geometry change dynamically with the deformation; the dynamic variation of GB deformability, however, remains largely elusive, especially regarding its relation with the frequently-observed GB-associated deformation twins in nanocrystalline materials. Attention here is focused on the GB dynamics in metallic nanocrystals, by means of well-designed in situ nanomechanical testing integrated with molecular dynamics simulations. GBs with low mobility are found to dynamically adjust their configurations and local deformation geometries via crystallographic twinning, which instantly changes the GB dynamics and enhances the GB mobility. This self-adjust twin-assisted GB dynamics is found common in a wide range of face-centred cubic nanocrystalline metals under different deformation conditions. These findings enrich our understanding of GB-mediated plasticity, especially the dynamic behaviour of GBs, and bear practical implication for developing high performance nanocrystalline materials through interface engineering.© 2021. The Author(s).

[8]

Liu

, Sun

W K

, Liew

K M

Multiscale modeling of crystal plastic deformation of polycrystalline titanium at high temperatures

[J]. Comput. Methods Appl. Mech. Eng., 2018, 340: 932

[9]

Sotniczuk

, Kuczyńska-Zemła

, Królikowski

, et al.

Enhancement of the corrosion resistance and mechanical properties of nanocrystalline titanium by low-temperature annealing

[J]. Corros. Sci., 2019, 147: 342

DOI [本文引用: 1]

Annealing deformed metals usually leads to reduced strength. In the case of nanocrystalline materials, the opposite effect was observed after low-temperature annealing. The annihilation of mobile dislocations and a relaxation of grain boundaries are probable reasons for the observed phenomenon. Dislocation density and grain boundaries character affect also on the corrosion resistance. Hence, in this work the effect of low-temperature annealing on both the hardness and corrosion resistance of nanotitanium was investigated. The results indicated that low-temperature annealing is a fast, simple and inexpensive way to further enhance nanotitanium's mechanical and corrosion properties, which are crucial for implantable materials.

[10]

Ren

J Q

, Wang

, Lu

X F

, et al.

Effect of oxygen content on active deformation systems in pure titanium polycrystals

[J]. Mater. Sci. Eng., 2018, A731: 530

[本文引用: 3]

[11]

Conrad

, Okasaki

, Gadgil

, et al.

Dislocation structure and the strength of titanium

[A]. Electron Microscopy and Structure of Materials [M]. Berkeley: University of California Press, 1972: 438

[12]

Shechtman

, Brandon

D G

Orientation dependent slip in polycrystalline titanium

[J]. J. Mater. Sci., 1973, 8: 1233

[13]

, Qi

, Tsuru

, et al.

Origin of dramatic oxygen solute strengthening effect in titanium

[J]. Science, 2015, 347: 635

\n The motion of dislocations or defects in a metal influences its strength and toughness. If these defects can be “pinned” by adding alloying elements, it should be possible to create a stronger alloy. It was thought that there shouldn't be much of an interaction between screw dislocations and any alloying elements. However, Yu\n et al.\n show that for α-Ti, the profound hardening effect of oxygen is due to the strong interactions with the core of the dislocations. First-principles calculations reveal that distortion of the interstitial sites at the dislocation core creates a very strong but short-range repulsion for oxygen atoms.\n

[14]

Voronoi

Nouvelles applications des paramètres continus à théorie des formes quadratiques. Deuxième Mémoire. Recherches sur les paralléloèdres primitifs

[J]. J. Reine Angew. Math., 1909, (136): 67

[15]

Lee

, Lee

K R

, Baskes

M I

, et al.

A modified embedded-atom method interatomic potential for ionic systems: 2NNMEAM + Qeq

[J]. Phys. Rev., 2016, 93B: 144110

[本文引用: 6]

[16]

Rose

J H

, Smith

J R

, Guinea

, et al.

Universal features of the equation of state of metals

[J]. Phys. Rev., 1984, 29B: 2963

[17]

Plimpton

Fast parallel algorithms for short-range molecular dynamics

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

[18]

Hirel

Atomsk: A tool for manipulating and converting atomic data files

[J]. Comput. Phys. Commun., 2015, 197: 212

[19]

Stukowski

Visualization and analysis of atomistic simulation data with OVITO—The Open Visualization Tool

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

[20]

AtomEye: An efficient atomistic configuration viewer

[J]. Modell. Simul. Mater. Sci. Eng., 2003, 11: 173

[21]

Sills

R B

, Cai

Solute drag on perfect and extended dislocations

[J]. Philos. Mag., 2016, 96: 895

[22]

Stukowski

, Bulatov

V V

, Arsenlis

Automated identification and indexing of dislocations in crystal interfaces

[J]. Modell. Simul. Mater. Sci. Eng., 2012, 20: 085007

[23]

Ikeda

, Qi

, Çagin

, et al.

Strain rate induced amorphization in metallic nanowires

[J]. Phys. Rev. Lett., 1999, 82: 2900

[24]

Wen

Y H

, Zhu

Z Z

, Zhu

R Z

Molecular dynamics study of the mechanical behavior of nickel nanowire: Strain rate effects

[J]. Comput. Mater. Sci., 2008, 41: 553

[25]

Tang

W H

, Zhang

, Wu

J Y

, et al.

Mechanical properties and enhancement mechanisms of titanium-graphene nanocomposites

[J]. Acta Mech. Sin., 2020, 36: 855

DOI [本文引用: 1]

[26]

, Ma

Zonal dislocations mediating $\{10\bar{1}1\}$ $<10\bar{1}\bar{2}>$ twinning in magnesium

[J]. Acta. Mater., 2009, 57: 1734

[27]

Ren

J Q

, Sun

Q Y

, Xiao

, et al.

Atomistic simulation of tension-compression asymmetry and its mechanism in titanium single-crystal nanopillars oriented along the $[11\bar{2}0]$ direction

[J]. Comput. Mater. Sci., 2018, 147: 272

[28]

Sakai

, Fine

M E

Plastic deformation of Ti-Al single crystals in prismatic slip

[J]. Acta. Metall., 1974, 22: 1359

[29]

Conrad

Effect of interstitial solutes on the strength and ductility of titanium

[J]. Prog. Mater. Sci., 1981, 26: 123

[30]

Wang

Y N

, Chen

S J

, Dong

X C

. Dislocation Theory and Its Application [M]. Beijing: Metallurgical Industry Press, 2007: 101

王亚男, 陈树江, 董希淳. 位错理论及其应用 [M]. 北京: 冶金工业出版社, 2007: 101

[31]

Wang

Y B

, Li

B Q

, Sui

M L

, et al.

Deformation-induced grain rotation and growth in nanocrystalline Ni

[J]. Appl. Phys. Lett., 2008, 92: 011903

[32]

Hasan

M S

, Lee

, Xu

W W

Deformation nanomechanics and dislocation quantification at the atomic scale in nanocrystalline magnesium

[J]. J. Magnes. Alloy., 2020, 8: 1296

[33]

Ghaffarian

, Taheri

A K

, Kang

, et al.

Molecular dynamics simulation study of the effect of temperature and grain size on the deformation behavior of polycrystalline cementite

[J]. Scr. Mater., 2015, 95: 23

[34]

Chen

, Wang

F X

, Li

Transitory phase transformations during $\{10\bar{1}2\}$ twinning in titanium

[J]. Acta Mater., 2019, 171: 65

DOI URL [本文引用: 3]

[35]

Zhang

, Pan

A Q

, Hei

R Y

, et al.

An atomistic simulation on the tensile and compressive deformation mechanisms of nano-polycrystalline Ti

[J]. Appl. Phys., 2021, 127A: 362

[36]

Zope

R R

, Mishin

Interatomic potentials for atomistic simulations of the Ti-Al system

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

[37]

Mehl

M J

, Papaconstantopoulos

D A

Tight-binding study of high-pressure phase transitions in titanium: Alpha to omega and beyond

[J]. Europhys. Lett., 2002, 60: 248