金属学报, 2025, 61(9): 1438-1448 DOI: 10.11900/0412.1961.2023.00451

研究论文

增材制造钛合金等效弹性张量的细观力学建模与实验研究

谭若涵1, 宋永锋1,2, 陈超3, 李丹3, 成庶1, 李雄兵,1

1 中南大学 交通运输工程学院 长沙 410075

2 广东工业大学 省部共建精密电子制造技术与装备国家重点实验室 广州 510006

3 中南大学 粉末冶金研究院 长沙 410083

Mesomechanical Modeling and Experimental Study of Effective Elastic Tensors in Additively Manufactured Titanium Alloys

TAN Ruohan1, SONG Yongfeng1,2, CHEN Chao3, LI Dan3, CHENG Shu1, LI Xiongbing,1

1 School of Traffic & Transportation Engineering, Central South University, Changsha 410075, China

2 State Key Laboratory of Precision Electronic Manufacturing Technology and Equipment, Guangdong University of Technology, Guangzhou 510006, China

3 Powder Metallurgy Research Institute, Central South University, Changsha 410083, China

通讯作者: 李雄兵,lixb213@csu.edu.cn,主要从事无损检测与评价的研究

责任编辑: 肖素红

收稿日期: 2023-11-14   修回日期: 2024-06-04  

基金资助: 湖南省科技创新领军人才项目(2023RC1015)
湖南省自然科学基金项目(2023JJ60159)

Corresponding authors: LI Xiongbing, professor, Tel: 15084761518, E-mail:lixb213@csu.edu.cn

Received: 2023-11-14   Revised: 2024-06-04  

Fund supported: Hunan Provincial Science and Technology Innovation Leading Talent Project(2023RC1015)
Natural Science Foundation of Hunan Province(2023JJ60159)

作者简介 About authors

谭若涵,女,1993年生,博士

摘要

孔隙和织构均为增材制造(AM)多晶体金属的重要特征,但已有的细观力学模型无法研究2者的耦合作用对材料力学性能的影响机理。因此,本工作构建了双相金属材料的改进Mori-Tanaka (modified Mori-Tanaka,MMT)模型,在此基础上预测AM Ti-6Al-4V合金的等效弹性张量,进而探究AM 多晶体材料的微观结构对材料宏观力学性能的影响规律。本模型结合了传统MT模型和微分法,可综合分析各向异性多晶体织构和孔隙2个耦合因素与宏观力学性能的内在联系。通过有损和无损实验,研究了2种孔隙率和织构的AM Ti-6Al-4V试块的相变行为、晶粒取向分布函数、孔隙率和孔隙形貌特征。为验证理论预测的准确性,采用稀疏法与所建立模型进行对比分析;同时进行拉伸实验和超声实验,基于MMT模型得到的Young's模量与拉伸实验结果的平均绝对百分比误差(MAPE)分别为0.87%和2.51%,基于MMT模型得到的等效弹性刚度张量(Ceff)与超声实验结果的MAPE分别为9.47%和4.45%。可见,实验结果从力学和无损检测2个角度验证了MMT模型的有效性,为研究AM 多晶体材料的微观结构对宏观力学性能的作用机理提供了一种有效的细观力学分析方法。

关键词: 等效弹性张量; 增材制造; 钛合金; 各向异性; 改进MT模型; 孔隙; 织构

Abstract

Additively manufactured (AM) Ti-6Al-4V alloys are known for their lightweight and superior mechanical properties and are increasingly being favored in the aerospace, energy, and biomedical industries. Among the available AM technologies, selective laser melting (SLM) stands out for its ability to fabricate complex-shaped Ti-6Al-4V efficiently and economically through near-net-shape manufacturing. Despite their advantages, SLM-produced parts often suffer from mechanical anisotropy and contain microdefects such as pores, cracks, and residual stresses, restricting their wider application. Addressing these issues requires a thorough understanding of the effects of texture and porosity on the mechanical properties of AM materials, which is necessary for developing components that comply with strict industry standards. Previous research has predominantly focused on studying the impact of texture or porosity on material properties, overlooking the interplay between these two critical factors. This work introduces an advanced modified Mori-Tanaka (MMT) model that simultaneously considers both texture and porosity in dual-phase AM Ti-6Al-4V alloys. This model enhances the traditional Mori-Tanaka approach by integrating it with the differential method, facilitating a nuanced analysis of how texture and porosity jointly influence the mechanical behavior of polycrystals. For model development, phase transitions and grain orientations are characterized using EBSD, whereas the 3D Gaussian distribution function describes the texture distribution within the polycrystal. Micro/nano computed tomography (CT) plays a pivotal role in determining the volume fraction and morphology of pores, providing crucial data for the model. These parameters were incorporated into the mesomechanical model to analyze the effective elastic stiffness tensor and Young's modulus, quantifying their collective impact on the alloy's mechanical properties. To validate the model, tensile and ultrasonic tests are conducted for two samples with different porosities and textures and compared their outcomes to evaluate the material's mechanical behavior, which exhibits characteristics of transverse isotropy. The comparison highlights the MMT model's superior precision, especially at higher porosity levels, with mean absolute percentage errors (MAPE) between Young's modulus obtained from the MMT model and the tensile test for two samples were 0.87% and 2.51%, respectively. Similarly, the MAPE between the effective elastic stiffness tensor derived from the MMT model and the ultrasonic test were 9.47% and 4.45%, respectively. These findings underscore the model's effectiveness in predicting material properties and its potential as a robust tool for exploring the interactions between microstructural elements and macroscopic mechanical properties of AM polycrystalline materials.

Keywords: effective elastic tensor; additive manufacturing; titanium alloy; anisotropy; modified MT model; porosity; texture

PDF (2522KB) 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

谭若涵, 宋永锋, 陈超, 李丹, 成庶, 李雄兵. 增材制造钛合金等效弹性张量的细观力学建模与实验研究[J]. 金属学报, 2025, 61(9): 1438-1448 DOI:10.11900/0412.1961.2023.00451

TAN Ruohan, SONG Yongfeng, CHEN Chao, LI Dan, CHENG Shu, LI Xiongbing. Mesomechanical Modeling and Experimental Study of Effective Elastic Tensors in Additively Manufactured Titanium Alloys[J]. Acta Metallurgica Sinica, 2025, 61(9): 1438-1448 DOI:10.11900/0412.1961.2023.00451

增材制造(additively manufactured,AM) Ti-6Al-4V合金密度低且力学性能优异,广泛应用于航空航天、能源和生物医疗等领域,尤其选区激光熔化(selective laser melting,SLM)工艺具有近净形状生产、高材料利用率和低成本等优点,特别适合制造复杂形状的Ti-6Al-4V零件[1,2]。然而,SLM Ti-6Al-4V零件不可避免地会出现织构及微观缺陷(如孔隙、裂缝、残余应力等)。织构会造成力学性能各向异性,微观缺陷降低材料力学性能,从而在一定程度上限制了其应用[3,4]。例如,Liu等[5]发现钛合金中由α晶粒聚集体形成的局部强织构将明显降低钛合金的疲劳性能;任永明等[6]和Cheng等[7]的研究分别表明了微观孔隙对AM Ti-6Al-4V合金疲劳性能和Young's模量的影响显著。为了设计和生产满足工业需求的AM Ti-6Al-4V零件,有必要建立细观力学性能模型,研究织构和孔隙2个耦合因素对材料力学性能的作用机理。

众多学者已单独针对织构或孔隙问题,开展了材料力学性能的细观力学建模研究。一方面,对于仅含织构的多晶体材料,均匀化模型可计算代表性体积单元(representative volume element,RVE)的等效弹性刚度张量( Ceff)。目前,常用的均匀化模型有Voigt模型、Reuss模型和Hill模型等,其中晶粒的取向分布函数(orientation distribution function,ODF)是这些模型实现均匀化的关键所在。Zhao等[8]和Man等[9]在Wigner-D基函数下展开ODF,得到织构系数,计算结果表明以此来表征晶粒取向,可以较为准确地得到 Ceff,但是算法复杂度高、计算耗时长。因此,一些学者采用近似的ODF来表征织构。例如Sha[10]和Li等[11]采用3D Gaussian形分布函数对ODF进行近似,通过与Euler角的半峰全宽(full width half maximum,FWHM)相关的参数来描述织构,结果显示虽损失了部分计算精度,但可显著提升计算效率。

另一方面,对于仅含孔隙的材料,Park等[12]通过微纳CT得到孔隙信息,并以此作为Mori-Tanaka (MT)模型的输入,构建了微观孔隙与 Ceff的细观力学模型,计算结果与有限元方法的计算结果吻合良好。Giraud等[13]研究了岩石的MT模型,通过椭球形夹杂的Eshelby张量推导了RVE的 Ceff,揭示了孔隙对Young's模量的影响规律,且该模型对高孔隙率的情况仍然适用。严寒冰[14]基于MT模型预测了含空心球复合泡沫塑料的有效模量,预测结果与实验结果一致性良好。可见,通过建立MT模型,可有效反映含孔隙材料的微观结构对宏观性能的影响机理。然而,增材制造金属材料属于多晶体,基体的各向异性是由织构决定的,这增加了使用已有MT模型计算 Ceff的难度[15~17]

本文作者前期工作[18]针对含孔隙、织构的多晶体材料建立了改进的Mori-Tanaka (modified Mori-Tanaka,MMT)模型,并通过弹性动力学理论预测了超声波在此类介质中的传播速度。但该模型还没有得到实验验证,且AM Ti-6Al-4V合金为同时含立方结构晶体和六方结构晶体的双相材料,尚不明确与只含有立方结构晶体的单相材料有何差别。

本工作以双相材料AM Ti-6Al-4V合金为研究对象,建立含孔隙和织构的MMT模型,并通过多种实验验证模型的有效性。采用电子背散射衍射(electron backscatter diffraction,EBSD)实验研究晶粒的特征,通过3D Gaussian形分布函数描述多晶体的织构分布,作为均匀化的基础;利用微纳CT实验研究孔隙的分布特征及相关参数,得到MMT模型所需的孔隙参数;将织构和孔隙纳入材料的细观力学模型,分析AM Ti-6Al-4V的 Ceff和Young's模量,定量研究试块中织构和孔隙对材料宏观力学性能的影响。最后,通过拉伸实验和超声实验,分别得到AM Ti-6Al-4V试块的Young's模量和 Ceff,从力学角度和无损检测角度对该模型的有效性进行对比验证。

1 实验方法与模型

1.1 实验方法

本工作技术路线如图1所示。其中所用Ti-6Al-4V合金的主要化学成分(质量分数,%)为:Al 6.08,V 4.04,O 0.0956,N 0.0027,H 0.0008,Fe 0.043,Ti余量。采用如图1所示的SLM技术制备AM Ti-6Al-4V合金试块,其中扫描速率为1000 mm/s、层厚为0.03 mm、扫描间距为0.12 mm,且每层旋转67°。分别在95和230 W 2种功率下进行打印,得到尺寸为40 mm × 40 mm × 40 mm的Ti-6Al-4V试块A和B;再将其放置于真空炉中,600 ℃保温2 h,并在炉中冷却,实现去应力退火。

图1

图1   增材制造(AM) Ti-6Al-4V合金(MMT)模型的流程图

Fig.1   Flowchart of modified Mori-Tanaka (MMT) model in additively manufactured (AM) Ti-6Al-4V alloy (w is any initial plane layer, Ceff is effective elastic stiffness tensor, Cijeff is the component of Ceff, γ is Young's modulus. Samples A and B are manufactured with different printing parameters)


首先从试块A和B的XY平面及XZ平面分别切取尺寸为5 mm × 4 mm × 3 mm的试块,依次采用600、1000、1500、2000号砂纸对试块进行打磨,随后进行电解抛光,使用带有CCD相机的Nordly max3 EBSD仪,以步长0.4 μm对试块中500 μm × 500 μm的区域进行EBSD检测,研究试块的晶粒微观特征。另从试块A和B上分别切取尺寸为1.9 mm × 1.9 mm × 1.9 mm的CT试块,利用Xradia 520 3D X射线显微镜,以1.2 μm的分辨率进行微纳CT实验,观测材料中的孔隙相关信息。然后根据GB/T 228.1—2021,分别在试块A和B垂直于XY平面和XZ平面方向上,通过线切割切取长度为40 mm的哑铃形试块,如图1所示。使用Instron 3369力学试验机以2.0 mm/min的拉伸速率进行室温拉伸,由应力-应变曲线计算Young's模量。最后利用JSR DPR300超声发生接收器,采用V109 RB接触式纵波探头和V155 RB横波探头,无损测量试块的Ceff,对理论模型进行验证。

1.2 MMT模型

针对含孔隙和织构的金属材料,采用3D Gaussian形分布函数对织构进行均匀化处理,同时结合微分法和传统MT法将孔隙分次添加进材料,获得材料的 Ceff。考虑双相时,根据EBSD观察可见每个试块的两相占比不同。此外,不同相的晶粒择优取向并不相同,导致两相的取向分布函数存在差异,且各相的单晶弹性常数也不相同。因此,对两相分别均匀化后,加权得到多晶体的等效弹性刚度张量( CMatrix)。首先不考虑孔隙的存在,借助物理简化模型——Voigt模型,计算多晶体基体体积(ΩMatrix)的 CMatrix。该模型假设多晶体中所有的晶粒具有相同的变形,因此其体积平均应力(T¯)与应变(E¯)的关系为[8]

T¯=1ΩMatrixrΩrcφ, θ, ζE¯dx=rΩrcφ, θ, ζΩMatrixE¯=        SO3cφ, θ, ζFφ, θ, ζdgE¯=CMatrixE¯

式中,dg = sinθdφdθdζ (其中,θφζ均是Euler角,0 ≤ φ ≤ 2π,0 ≤ θ ≤ π,0 ≤ ζ ≤ 2π),SO3为Euler空间中转动张量( R )的集合,Ωr 是参考晶粒r所占据的区域体积, c (φ, θ, ζ)是依赖于晶粒取向Euler角的单晶弹性刚度张量,其分量cijkl (φ, θ, ζ)r的单晶弹性常数(c(r))经转动得到:

cijkl(φ, θ, ζ)=RimRjnRkpRlqcmnpq(r)

式中,RijRi行第j列的分量,cmnpq (r)是c(r)的分量,c(r)取值如表1[19]所示。弹性刚度张量分量的下标通常采用Voigt标记法,即11→1,22→2,33→3,23→4,13→5,12→6,例如c1122c12

表1   Ti-6Al-4V合金的单晶弹性常数(c(r))[19] (GPa)

Table 1  Single crystal elastic constant (c(r)) of Ti-6Al-4V alloy[19]

Phasec11c12c13c33c44
α170927019252
β138108--51

Note:cij is the component of c(r)

新窗口打开| 下载CSV


F(φ, θ, ζ)为3D Gaussian形分布函数,在织构框架中定义如下[20]

Fm(φ, θ, ζ)=F0mexpcosθ2σφm+cosφ2σθm+cosζ2σζm
                      (m=α, β)

式中,右上角标m可由α相和β相分别代入,3个与FWHM相关的织构参数σm = (σφm, σθm, σζm)描述了晶粒的择优取向程度。F0m为满足定义18π202π0π02πFmφ, θ, ζsinθdφdθdζ=1的标准化系数,因此在上式中,

F0m=2Sθmσθmcsch(Sθm)[I0(Sφm)+I0(Sζm)]csch(Sθm)+2σθm

式中,Sφm = 1 / (2σφm)Sθm = 1 / (2σθm)Sζm = 1 / (2σζm)I0x = k=0(x / 2)2k/(k!)2为第一类零阶修正Bessel函数。由 式(1)可得多晶体的 CMatrix[11]

CMatrix=cijkl=
18π2pm02π0π02πcmφ, θ, ζFmφ, θ, ζsinθdφdθdζ

式中,pm 为每个相在多晶体中所占比例,且pα + pβ= 1。假定孔隙的弹性刚度张量为 CPore = 0,含孔隙材料的 Ceff可通过下式迭代得到[21]

Ci=Ci-1I+ρi1-ρiCi-1I-Si-1Ci-1-1
(i=1, 2, , n)

式中, CiρiSi 分别为第i次迭代的等效弹性刚度张量、瞬时孔隙率和Eshelby张量,n代表孔隙分组数(即迭代次数), I 是单位张量。初始输入是多晶体均匀化后的 CMatrix,即 C0 = CMatrix,而迭代完成后所得到的 Cn 为材料的 Ceff,即 Cn = Ceff。孔隙呈现短轴长度为a、长轴长度为b的椭球体形态,其长宽比(rPore = b / a) ≥ 1 (当rPore = 1时即为球形孔隙)。令v0vi 分别为多晶体和不同长宽比孔隙的体积分数,且ν0+1nνi = 1。试块中的总孔隙率ρPore = 1nνi,而瞬时孔隙率为:

ρi=νi/ν0+1iνj

式(7)代表第i次添加到当前基体的孔隙体积分数。该模型不仅综合考虑了孔隙和织构2个耦合因素对宏观弹性性能的联合影响,还考虑了孔隙之间的相互作用,从而提高了预测结果的准确性。

MMT模型所需的关键参数包括了晶粒的织构参数以及各相所占比例,这些参数可借助EBSD实验获得的晶粒数据进行精确计算。同时,孔隙的体积分数和长宽比也是重要参量,可通过微纳CT实验进行有效测定。

2 实验结果

2.1 EBSD实验

AM Ti-6Al-4V试块的EBSD实验结果如表2所示。可见,试块A和B的相解析率(fp)之和分别达到88.47%和83.00%,在已解析的相中,β相较少,其原因可能是:(1) 部分β相未能被正确解析出;(2) 试块在室温下呈现近α相,这种情况也曾被Simonelli等[22]观察到,他们认为这可能是由于在等离子体雾化过程中经历了高冷却速率,大多数β相转变为α相,转变数量与β相区的冷却速率有关。织构指数(J)表明试块A的晶粒取向更为集中,而试块B中β相的晶粒取向分布更为集中,这可能导致试块A和B的力学性能存在差异[23]。取向特性的变化可能与材料的加工工艺及热处理条件紧密相关。

表2   AM Ti-6Al-4V试块的EBSD分析结果

Table 2  EBSD analysis results of AM Ti-6Al-4V samples

SamplePhasefp / %qpJ
Aβ0.86133843.94
α87.6113689183.31
Bβ1.66258922.20
α81.3412709801.26

Note:fp—phase resolution, qp—phase counting, J—texture index

新窗口打开| 下载CSV


图2为AM Ti-6Al-4V试块中晶粒的微观结构及取向分布。由图2a1b1可见,α晶粒形成了局部织构。由于从图2a1b1中无法辨认母相β晶粒的晶界,为了观察AM Ti-6Al-4V合金的柱状晶,需要遵循Burgers晶体关系对母相β晶粒进行重构[24]

图2

图2   AM Ti-6Al-4V试块中晶粒的微观结构及取向分布

Fig.2   Inverse pole figures (IPFs) paralleling to axis Z (a1, b1), IPFs of reconstructed phase β (a2, b2), orientation distribution function (ODF) maps of phase β (a3, b3), and ODF maps of phase α (a4, b4) of AM Ti-6Al-4V sample A (a1-a4) and sample B (b1-b4) (θ—Euler angle)


(11¯1)β(112¯0)α, (110)β(0001)α

根据 式(8),反演出母相β晶粒的晶界。在重构的β相晶体取向微观结构中,β柱状晶通过连续沉积层外延生长,如图2a2b2所示。图2a3a4图2b3b4分别展示了试块A和B的β相和α相的ODF图。观察结果显示试块A的整体各向异性明显强于试块B,而试块B中β相的各向异性强于α相,这表明在试块B中α相的内部结构较β相更不均匀[25]。从图2a3a4图2b3b4中可观察到许多平行的条状图形,这可能是由于特定的晶体学特征关系导致的。ODF图能够揭示多晶材料的三维取向特性,准确地获得材料的织构组成,然而在本工作所提出的细观力学模型中,为了简化计算过程,采用基于Euler角所得织构参数的3D Gaussian形分布函数进行计算。这种简化方法虽然会导致多晶体的对称性降低,但是显著减少了计算量。

使用AZtecCrystal软件对EBSD结果进行分析,得到Euler角的分布结果,如图3所示。根据Euler角的分布,计算得到试块A中α相和β相的σασβ 分别为(0.39, 0.23, 0.03)和(0.79, 0.01, 0.11),而试块B的σασβ 分别为(0.38, 0.24, 0.04)和(1.04, 0.02, 0.09)。

图3

图3   AM Ti-6Al-4V试块Euler角(θ)的概率密度分布图

Fig.3   Probability density distribution map of θ in AM Ti-6Al-4V sample


2.2 微纳CT实验

通过Avizo软件对试块A和B的孔隙进行三维重建和分析,并通过体渲染对不同孔隙进行区分,结果如图4所示。通过CT法测得试块A和B的孔隙率分别为4.51%和1.18%,其中小体积孔隙数量占据大多数。试块A和B的孔隙体积分别主要分布在2000和500 μm3以内。而大体积的孔隙占比少,这可能是由于孔隙合并导致的,因为较大孔隙在材料中的形成和稳定需要更多的能量[26]。根据孔隙形貌的不同,将其分为独立球形孔隙和球形孔隙重叠形成的不规则连通孔隙。在理论模型计算时,使用等体积椭球体对这2种孔隙进行拟合。椭球形孔隙可能有择优取向,这在一定程度上会导致材料表现出宏观力学性能的各向异性。

图4

图4   AM Ti-6Al-4V试块的孔隙重构

Fig.4   Pore reconfigurations of AM Ti-6Al-4V sample A (a) and sample B (b)


在MMT模型中,可以根据孔隙的长宽比以间隔为1将孔隙分为不同组,以便分次添加。此外,孔隙的球度定义为t = 36πVPore2 / SPore33[27],其中VPore为孔隙的体积,SPore为孔隙的表面积,球度越接近1,代表孔隙越接近球体。图5为AM Ti-6Al-4V试块孔隙体积和球度的概率密度分布图。从图5a可以看出,试块A的圆形孔隙占比小于试块B,这是由于试块B的孔隙率较小,孔隙的体积较小,而小孔隙多由于冷却速率过快,气体来不及逸出而形成[26]

图5

图5   AM Ti-6Al-4V试块孔隙体积和球度的概率密度分布图

Fig.5   Probability density distribution maps of pore volume (a) and sphericity (b) in AM Ti-6Al-4V samples (Inset in Fig.5a shows the locally enlarged map)


2.3 力学实验

AM Ti-6Al-4V试块A和B经去应力退火后,其力学性能如表3所示。测量结果显示,试块A和B的最大载荷和Young's模量在不同方向上差异显著,尤其试块A的Young's模量在2个方向上的差异较大,高达14.3%。同时,试块A的最大载荷最大值出现在水平方向上,试块B则出现在垂直方向上;2个试块的Young's模量最大值均出现在垂直方向[28]。这是由于当Ti-6Al-4V合金有较强的晶体织构时,其内部的针状晶会表现出择优取向,这会对Young's模量产生显著的影响[29]。此外,所有的SLM Ti-6Al-4V合金都表现出针状晶粒择优取向形成的弱织构[24]

表3   AM Ti-6Al-4V试块在不同方向的最大载荷和Young's模量

Table 3  Maximum load (Pmax) and γ of AM Ti-6Al-4V samples in different directions

SampleDirectionPmax / kNγ / GPa
AHorizontal14.52109.46
Vertical13.88125.11
BHorizontal15.56115.56
Vertical16.09126.12

新窗口打开| 下载CSV


2.4 超声实验

在试块不同方向进行多次超声实验,采用峰值法得到纵波声速和2个正交方向的横波声速并取平均值[30,31]。试块A和B均呈现横观各向同性。在横观各向同性材料中,沿着主轴所测得的声速与 Ceff之间具有以下关系[32]

C11eff=C22eff=ρV112=ρV222, C33eff=ρV332
C44eff=C55eff=ρV232=ρV132=ρV322=ρV312
C66eff=ρV122=ρV212
C13eff=C31eff=C23eff=C32eff=-C44eff+
(C11eff+C44eff-2ρV452)(C33eff+C44eff-2ρV452)

式中,CijeffCeff的分量;Vij 表示沿i方向传播、沿j方向振动的纵波声速(i = j)和横波声速(ij);V45是在对称轴45°方向测量的声速,对于P波为VqP45,对于S波为VqSV45。根据测量的超声声速,试块A和B的 Ceff表4所示。试块A的各向异性强于试块B,与EBSD的实验结果吻合。试块A的横向模量小于纵向模量,与试块B恰好相反。

表4   超声实验测得AM Ti-6Al-4V试块的等效弹性刚度张量( Ceff) (GPa)

Table 4  Ceff of AM Ti-6Al-4V samples measured by ultrasonic test

SampleC11effC22effC33effC44effC55effC66eff
A142.7 ± 1.0143.4 ± 0.2137.4 ± 0.444.1 ± 0.744.6 ± 0.443.2 ± 0.4
B169.4 ± 0.7168.6 ± 0.7171.6 ± 0.147.1 ± 2.246.8 ± 2.145.2 ± 1.2

新窗口打开| 下载CSV


3 分析与讨论

3.1 MMT理论模型结果

正交各向异性材料的Young's模量与等效弹性柔度张量( Seff)之间的关系可表示为[33]

γ1=1S11eff, γ2=1S22eff, γ3=1S33eff

式中,SijeffSeff的分量, SeffCeff具有自洽关系,Seff:Ceff-1=IIII为四阶单位张量;γ的下标1~3代表不同方向。根据MMT模型迭代20次和迭代0次所得到的C11effC55effC13effC23eff (图6),随着孔隙率的增加,所得结果的差距逐渐增大,并在孔隙率逐渐接近1时又逐渐变小。以C11eff为例,在孔隙率为5%时,2种迭代次数所得结果的绝对误差为0.5,当孔隙率提升至15%时,绝对误差增大至3.3;迭代超过20次时Ceff趋于稳定。需要说明的是,本工作研究的2个试块孔隙率相对较小,因此迭代结果基本无变化。

图6

图6   MMT模型得到的AM Ti-6Al-4V试块的 Ceff

Fig.6   C11eff (a), C55eff (b), C13eff (c), and C23eff (d) of AM Ti-6Al-4V samples obtained by MMT model (Insets show the locally enlarged curves. n—iteration number)


为了验证MMT模型的有效性,使用稀疏法(sparse method,SM)与其对比[34]。稀疏法假定孔隙平均应变近似于孤立孔隙在无限基体中的应变,MMT模型则考虑了孔隙间的相互作用。根据2种方法计算得到 Ceff随孔隙率的变化趋势,以分量C22eff为例,结果如图7所示。可见,2种方法在孔隙率较低时所得结果差异较小。但是,随着孔隙率的增大,预测结果之间的差异显著扩大,特别是当孔隙率接近40%时,稀疏法所计算的C22eff降至零,这与实际物理现象[35]相悖。

图7

图7   MMT模型与稀疏法得到的AM Ti-6Al-4V试块的 Ceff

Fig.7   C22eff of AM Ti-6Al-4V samples obtained by MMT model and sparse method (SM) (Inset shows the locally enlarged curves)


3.2 理论模型与拉伸实验结果对比

试块A和B在水平和垂直方向之间的力学性能存在差异,因此用拉伸Young's模量差值比(tensile modulus difference ratio,TDR)来定量评估其差异性[36]

TDR=|TSh-TSv| / TSv

式中,TSh、TSv分别代表沿着AM Ti-6Al-4V试块水平方向和垂直方向的拉伸Young's模量。

为了准确评估MMT模型理论结果与超声测量结果、拉伸实验结果之间的差异,引入平均绝对百分比误差(mean absolute percentage error,MAPE)和均等系数(equal coefficient,EC) 2个评价指标[37]

MAPE=1s1sxi-x^ixi
EC=1-i=1sxi-x^i2i=1sxi2+i=1sx^i2

式中,x^ixi分别表示理论预测值和实验所得值的第i个分量,s为样本向量的长度。MAPE的值越小,表示理论预测值与实验所得值之间的吻合性越好,模型的预测能力越强,而EC则相反。

试块A和B的TDR分别为12.51%和8.33%,该结果表明试块B的孔隙率较低,其各向异性程度较小,这与试块的超声检测结果相符合。表5列出了基于Voigt (V)、Reuss (R)和Hill (H)模型进行基体均匀化的MMT模型、稀疏法与拉伸实验的结果,以及每种方法的评价指标。基于本工作理论模型(V-MMT)所得试块A和B的Young's模量和拉伸实验结果的MAPE分别为0.87%和2.51%,EC分别为99.51%和98.20%。在试块A的评估中,V-MMT模型预测结果与拉伸实验结果吻合度最高,且MMT模型在预测精度上更具优越性;试块B的MAPE所呈现结论与试块A一致,但EC显示基于Hill模型的稀疏法所得结果最佳,值得注意的是,6种模型所得指标结果的差异性并不显著。

表5   基于MMT模型和稀疏法以及拉伸实验测得的AM Ti-6Al-4V试块不同方向的Young's模量及其平均绝对百分比误差(MAPE)和均等系数(EC)

Table 5  γ in different directions of AM Ti-6Al-4V alloy by MMT model, SM, and tensile test, along with its average absolute percentage error (MAPE) and equalization coefficient (EC)

SampleIndicatorHorizontal γ / GPaVertical γ / GPaMAPE / %EC / %
AV-MMT107.86124.760.8799.51
R-MMT104.16121.663.8098.06
H-MMT106.02123.212.3398.80
V-SM107.41122.611.9399.02
R-SM103.74119.484.8697.53
H-SM105.58121.053.4098.28
Tensile test109.46125.11
BV-MMT115.46132.252.5198.20
R-MMT110.61128.232.9898.42
H-MMT111.55129.002.9398.49
V-SM114.99132.352.7298.19
R-SM110.58127.822.8398.45
H-SM112.80130.102.7798.59
Tensile test115.56126.12

Note: V, R, and H refer to Voigt, Reuss, and Hill, respectively

新窗口打开| 下载CSV


对比结果表明,尽管存在一定差异,但V-MMT预测的Young's模量与拉伸实验结果表现出高度一致性。此外,孔隙率高的试块A的理论结果比试块B吻合度更好,这可能是因为不同的打印参数会导致层与层之间的熔合程度不同,而理论模型并未考虑打印试块的层间结合,从而影响了预测结果[38]

3.3 理论模型与超声实验对比

基于MMT模型所得试块A和B的 Ceff与超声实验所得结果的MAPE分别为9.47%和4.45%,EC分别为94.35%和98.49%,试块A的误差明显大于试块B (表6)。这种现象是由于随着孔隙率的增加,超声声速对孔隙形貌、尺寸等的敏感性增强。即使在相同的孔隙率下,孔隙形貌的随机变化也可能导致纵波波速出现波动;且分布函数以及计算过程中模型的简化也可能会引入部分误差。

表6   基于MMT模型和超声实验获得的AM Ti-6Al-4V试块 Ceff及其MAPE和EC

Table 6  C11eff, C22eff, C33eff, C44eff, C55eff, and C66eff of AM Ti-6Al-4V alloy by MMT method and ultrasonic test, along with its MAPE and EC

SampleIndicator

C11eff

GPa

C22eff

GPa

C33eff

GPa

C44eff

GPa

C55eff

GPa

C66eff

GPa

MAPE

%

EC

%

AV-MMT152.60163.09157.3648.4240.5044.069.4794.35
Ultrasonic143.68143.23136.9244.1144.9843.18
BV-MMT165.65175.67170.3451.2543.3146.974.4598.49
Ultrasonic169.43169.60171.6047.0946.7945.21

新窗口打开| 下载CSV


综上所述,MMT模型在预测AM Ti-6Al-4V合金力学性能方面具有较高的准确性和可靠性。在后续的研究中,为了实现材料微观结构的超声无损检测评估,将建立基于超声声速的模型,用以反演材料孔隙率和各向异性程度,以期能够快速、准确地评价材料的使用性能,这对于材料的研发和生产具有重要的实际指导意义。此外,本工作使用的单晶弹性常数[19]可能与实际值有误差,因此有必要分析该误差在细观力学模型中的误差传播规律。可见,深入探究单晶弹性常数等关键参数如何对MMT模型预测结果产生不确定性影响也是未来的重要工作之一,可为相关领域的研究与应用提供更加坚实的理论基础。

4 结论

(1) 通过对比拉伸实验所得Young's模量和超声实验得到的 Ceff,发现具有较高孔隙率的试块在不同方向上的Young's模量呈现出更加明显的差异,这表明增材制造金属材料的各向异性随孔隙率的增加而增强。

(2) 当孔隙率小于15%时,迭代次数对 Ceff的影响微弱;随孔隙率的增加,迭代次数越多,迭代结果与未迭代结果的差距呈先增大后减小趋势。将孔隙分为球形孔和椭球孔添加时,MMT模型的迭代超过20次时 Ceff趋于稳定。

(3) 对于孔隙率高的AM Ti-6Al-4V试块,基于Voigt均匀化模型的MMT法在预测精度上相较于稀疏法更具优势。

(4) 不同孔隙率和织构的AM Ti-6Al-4V试块,基于MMT模型得到的Young's模量与拉伸实验结果的MAPE分别为0.87%和2.51%,EC分别为99.51%和98.20%;基于MMT模型得到的 Ceff与超声实验结果的MAPE分别为9.47%和4.45%,EC分别为94.35%和98.49%。且孔隙率小的试块超声检测误差更小,这可能是由于孔隙取向、形貌对超声声速造成的影响。

参考文献

Tshephe T S, Akinwamide S O, Olevsky E, et al.

Additive manufacturing of titanium-based alloys—A review of methods, properties, challenges, and prospects

[J]. Heliyon, 2022, 8(3): e09041

[本文引用: 1]

Li S J, Hou W T, Hao Y L, et al.

Research progress on the mechanical properties of the biomedical titanium alloy porous structures fabricated by 3D printing technique

[J]. Acta Metall. Sin., 2023, 59: 478

DOI      [本文引用: 1]

Porous titanium alloys have been used for biomedical implants owing to their low-modulus matching with that of human bones and interconnecting pores with suitable size, which facilitates bone in-growth and satisfies the requirement of a successful implant. Recently, additive manufacturing (3D printing) has emerged as an excellent technology for manufacturing porous implants with accurate designed pore parameters and overcoming processing difficulties caused by high melting temperatures of metals. In this paper, the microstructure and mechanical properties of porous Ti-6Al-4V, commercial pure titanium (CP-Ti), and low-modulus Ti2448 alloys produced by 3D printing, obtained mainly by the authors' group, are reviewed. For Ti-6Al-4V, its fatigue properties are affected by the type of mesh struts and post processing. The better fatigue life of CP-Ti compared to that of Ti-6Al-4V derives from its superior ductility and the strain hardening effect caused by deformation twins. The excellent fatigue life of the low-modulus Ti2448 alloy results from its superelasticity and the high toughness, which increases the crack nucleation life and fatigue crack propagation life, respectively. Future directions of corrosion-fatigue properties of materials in complex physiological environments, surface biological functionalization, and porous material of new metallic alloy systems are discussed.

李述军, 侯文韬, 郝玉琳 .

3D打印医用钛合金多孔材料力学性能研究进展

[J]. 金属学报, 2023, 59: 478

DOI      [本文引用: 1]

钛合金多孔材料具有与人体骨匹配的弹性模量,可有效解决金属植入物与人体骨弹性错配;其内部存在的大量孔隙有利于周围细胞的长入和新骨的生长,从而促进骨组织形成。近年来,增材制造(3D打印)技术被用于钛合金多孔材料制备,该方法可以精确控制孔隙参数,并且克服了因金属高熔点造成的制备困难。本文综述了作者团队在3D打印医用Ti-6Al-4V、纯Ti以及低模量钛合金多孔材料组织及力学性能的研究结果。对于Ti-6Al-4V两相合金,其疲劳性能受多孔结构设计和多种后处理的影响。纯Ti多孔材料较Ti-6Al-4V更优的疲劳寿命源于其更好的塑性和形变孪晶的应变硬化效应。低模量Ti2448合金的优异疲劳寿命则源于其超弹性提高裂纹萌生寿命,高韧性提高裂纹扩展寿命。最后展望了复杂生理环境腐蚀疲劳性能、多孔材料表面生物活化处理和新型医用金属体系多孔材料等发展方向。

Song B, Zhang J L, Zhang Y J, et al.

Research progress of materials design for metal laser additive manufacturing

[J]. Acta Metall. Sin., 2023, 59: 15

[本文引用: 1]

宋 波, 张金良, 章媛洁 .

金属激光增材制造材料设计研究进展

[J]. 金属学报, 2023, 59: 15

[本文引用: 1]

Vilaro T, Colin C, Bartout J D.

As-fabricated and heat-treated microstructures of the Ti-6Al-4V alloy processed by selective laser melting

[J]. Metall. Mater. Trans., 2011, 42A: 3190

[本文引用: 1]

Liu W C, Huang J, Liu J W, et al.

Experimental and crystal plasticity modelling study on the crack initiation in micro-texture regions of Ti-6Al-4V during high cycle fatigue tests

[J]. Int. J. Fatigue, 2021, 148: 106203

[本文引用: 1]

Ren Y M, Lin X, Huang W D.

Research progress of microstructure and fatigue behavior in additive manufacturing Ti-6Al-4V alloy

[J]. Rare Met. Mater. Eng., 2017, 46: 3160

[本文引用: 1]

任永明, 林 鑫, 黄卫东.

增材制造Ti-6Al-4V合金组织及疲劳性能研究进展

[J]. 稀有金属材料与工程, 2017, 46: 3160

[本文引用: 1]

Cheng M, Lu Z G, Wu J, et al.

Effect of thermal induced porosity on high-cycle fatigue and very high-cycle fatigue behaviors of hot isostatic-pressed Ti-6Al-4V powder components

[J]. J. Mater. Sci. Technol., 2022, 98: 177

DOI      [本文引用: 1]

The present work reports the effect of thermal induced porosity (TIP) on the high-cycle fatigue (HCF) and very high-cycle fatigue (VHCF) behaviors of hot-isostatic-pressed (HIPed) Ti-6Al-4V alloy from gas-atomized powder. The results show that the residual pores in the as-HIPed powder compacts present no obvious effect on the HCF life. The regrowth of the residual pores can be observed after solution heat treatment. The pore location ranks the most harmful for the fatigue life compared with the other initiating defects. The maximum stress intensity factors were calculated. The plastic zone size of fine granular area (FGA) is much less than the characteristic size of the microstructure, and the crucial size of the internal pores in this study is about 40 μm. The failure types of fatigue specimens in the VHCF regime were classified, and the competition of different failure types was described based on the modified Poisson distribution.

Zhao T F, Zhang L, Huang M J.

Effective elasticity tensors of fiber-reinforced composite materials with 2D or 3D fiber distribution coefficients

[J]. Acta Mech., 2019, 230: 4175

[本文引用: 2]

Man C S, Zhao D.

Remarks on texture coefficients of polycrystals with improper crystallite symmetry

[J]. J. Elast., 2020, 138: 111

[本文引用: 1]

Sha G F.

Explicit backscattering coefficient for ultrasonic wave propagating in hexagonal polycrystals with fiber texture

[J]. J. Nondestr. Eval., 2018, 37: 51

[本文引用: 1]

Li J, Rokhlin S I.

Elastic wave scattering in random anisotropic solids

[J]. Int. J. Solids Struct., 2016, 78-79: 110

[本文引用: 2]

Park I, Moon J, Bae S, et al.

Application of micro-CT to Mori-Tanaka method for non-randomly oriented pores in air-entrained cement pastes

[J]. Constr. Build. Mater., 2020, 255: 119342

[本文引用: 1]

Giraud A, Huynh Q V, Hoxha D, et al.

Effective poroelastic properties of transversely isotropic rock-like composites with arbitrarily oriented ellipsoidal inclusions

[J]. Mech. Mater., 2007, 39: 1006

[本文引用: 1]

Yan H B.

Determination of the effective modulus of the syntactic foam containing hollow spheres by differential scheme and Mori-Tanaka method

[J]. J. Beijing Univ. Aeronaut. Astronaut., 2000, 26: 688

[本文引用: 1]

严寒冰.

用微分法及Mori-Tanaka法求解复合泡沫塑料的有效模量

[J]. 北京航空航天大学学报, 2000, 26: 688

[本文引用: 1]

Gui J C, Ma T S, Chen P, et al.

Anisotropic damage to hard brittle shale with stress and hydration coupling

[J]. Energies, 2018, 11: 926

[本文引用: 1]

Zhang Q G, Fan X Y, Chen P, et al.

Geomechanical behaviors of shale after water absorption considering the combined effect of anisotropy and hydration

[J]. Eng. Geol., 2020, 269: 105547

Asef M R, Farrokhrouz M.

A semi-empirical relation between static and dynamic elastic modulus

[J]. J. Petrol. Sci. Eng., 2017, 157: 359

[本文引用: 1]

Tan R H, Song Y F, Li X B, et al.

Effective elastic stiffness tensor and ultrasonic velocities for 3D printed polycrystals with pores and texture

[J]. Res. Nondestr. Eval., 2022, 33: 196

[本文引用: 1]

Warwick J L W, Coakley J, Raghunathan S L, et al.

Effect of texture on load partitioning in Ti-6Al-4V

[J]. Acta Mater., 2012, 60: 4117

[本文引用: 4]

Cho J H, Rollett A D, Oh K H.

Determination of volume fractions of texture components with standard distributions in Euler space

[J]. Metall. Mater. Trans., 2004A, 35: 1075

[本文引用: 1]

Gong S, Li Z, Zhao Y Y.

An extended Mori-Tanaka model for the elastic moduli of porous materials of finite size

[J]. Acta Mater., 2011, 59: 6820

[本文引用: 1]

Simonelli M, Tse Y Y, Tuck C.

Effect of the build orientation on the mechanical properties and fracture modes of SLM Ti-6Al-4V

[J]. Mater. Sci. Eng., 2014, A616: 1

[本文引用: 1]

Bozzolo N, Gerspach F, Sawina G, et al.

Accuracy of orientation distribution function determination based on EBSD data—A case study of a recrystallized low alloyed Zr sheet

[J]. J. Microsc., 2007, 227: 275

[本文引用: 1]

Simonelli M, Tse Y Y, Tuck C.

On the texture formation of selective laser melted Ti-6Al-4V

[J]. Metall. Mater. Trans., 2014, 45A: 2863

[本文引用: 2]

Huang S S, Ma Y J, Zhang S L, et al.

Influence of alloying elements partitioning behaviors on the microstructure and mechanical properties in α + β titanium alloy

[J]. Acta Metall. Sin., 2019, 55: 741

[本文引用: 1]

黄森森, 马英杰, 张仕林 .

α + β两相钛合金元素再分配行为及其对显微组织和力学性能的影响

[J]. 金属学报, 2019, 55: 741

DOI      [本文引用: 1]

研究了两相区固溶温度及固溶后冷速对Ti-6Al-4V (TC4)合金元素再分配行为的影响,利用EPMA技术表征了初生α相(α<sub>p</sub>)以及β转变区域(β<sub>t</sub>)的元素浓度,考察了β<sub>t</sub>显微组织尺寸随固溶温度及元素浓度的变化。结果表明:随着固溶温度升高,β<sub>t</sub>区域元素浓度变化显著,表现为Al含量升高、V含量降低,而α<sub>p</sub>晶粒中元素浓度变化较小,导致两区域元素浓度差异减小;同一固溶温度下,以不同冷却方式(水冷、空冷及炉冷)冷却的显微组织及元素分布显示,冷却速率越低,α<sub>p</sub>比例越高,α<sub>p</sub>与β<sub>t</sub>之间元素浓度差异越明显。合金经固溶水冷、空冷后,β<sub>t</sub>分别为淬火马氏体、次生α相(α<sub>s</sub>)+残余β相,2种冷速下β<sub>t</sub>的显微组织尺寸均与高温β相内的元素浓度水平有关,即β<sub>t</sub>内部显微组织尺寸受固溶温度的显著影响。利用纳米压痕技术表征了不同固溶温度下微区域(α<sub>p</sub>、β<sub>t</sub>)的力学特征,结果表明,密排六方(hcp)晶格α<sub>p</sub>本身呈现的力学行为的各向异性对其纳米压痕性能起决定性作用,而β<sub>t</sub>的弹性模量及硬度主要受α<sub>s</sub>片层尺寸的影响。最后讨论了“固溶温度-微区元素浓度-微区显微组织-微区力学性能”之间的关系。

Kasperovich G, Haubrich J, Gussone J, et al.

Correlation between porosity and processing parameters in TiAl6V4 produced by selective laser melting

[J]. Mater. Des., 2016, 105: 160

[本文引用: 2]

Ge J G, Yan X C, Lei Y P, et al.

A detailed analysis on the microstructure and compressive properties of selective laser melted Ti6Al4V lattice structures

[J]. Mater. Des., 2021, 198: 109292

[本文引用: 1]

Liu Z, Liu J R, Zhao Z B, et al.

Microstructure and tensile property of TC4 alloy produced via electron beam rapid manufacturing

[J]. Acta Metall. Sin., 2019, 55: 692

DOI      [本文引用: 1]

<p>Electron beam rapid manufacturing (EBRM) is one of the 3D printing technologies. The main attractions of EBRM technology are its high efficiency and economy in fabricating large, complex near net shape components dielessly and only needing limited machining. In general, the microstructure and texture of titanium alloy can play a significant role in determining its mechanical behaviors. In the present work, the microstructure, texture and tensile property of TC4 alloy produced by electron beam rapid manufacturing (EBRM) are investigated. Results show that the microstructure is comprised of columnar prior <i>β</i> grains that orient parallel to the building direction. The width of the columnar <i>β</i> grains increased rapidly at the initial several build layers, and the subsequent increase rate of the width of the columnar <i>β</i> grains tends to slow down. Fine <i>α</i> lamellae with gradient size are observed inside the columnar prior <i>β</i> grains, which occur because the alloy experiences different complex thermal histories during the EBRM-produced process. The size of <i>α</i> lamellae tends to decrease with the increase of build layers. The XRD result shows that the TC4 alloy has a typical <i>α</i> phase texture, (the <i>c</i>-axes are either concentrated at about 45° or are perpendicular to the building direction). At the same time, the <$10\bar{1}0$> poles are relative to random distribution. For the tensile samples along the electron beam scanning direction, the yield strengths do not show significant change with the increase of build layers, but the tensile strengths increase. The ductility of the alloy also has an upward trend, despite of a slightly decreasing ductility in the top sample. The tensile samples at the bottom of the alloy (10 mm and 20 mm away from the substrate) have similar work hardening exponents, which are lower than the top sample. The top sample shows the highest work hardening exponent. This difference in the tensile properties can be highly attributed to the gradient microstructure. The alloy also presents obvious anisotropy in tensile strength. The tensile sample along the 45° direction has a higher strength than the sample along the <i>X</i> direction, while the tensile sample along the <i>Z</i> direction shows the lowest strength. This anisotropic strength is strongly associated with the <i>α</i> phase texture. When the loading direction is 45° to the building direction, most of the <i>c</i>-axes of <i>α</i> phase are about parallel to the loading direction, showing a "hard" orientation, leading to a higher strength than other oriented samples. Conversely, when the loading direction is along the building direction, most of the <i>α</i> phase present a "soft" orientation, resulting in lower strength compared to the tensile samples along the 45° or the <i>X</i> direction.</p>

刘 征, 刘建荣, 赵子博 .

电子束快速成形制备TC4合金的组织和拉伸性能分析

[J]. 金属学报, 2019, 55: 692

[本文引用: 1]

Banerjee D, Williams J C.

Perspectives on titanium science and technology

[J]. Acta Mater., 2013, 61: 844

[本文引用: 1]

Yuan M D, Dai A B, Ma J T, et al.

Nondestructive measurement of anisotropic elastic constants of selective laser melted 316L based on a tri-mode ultrasonic method

[J]. Meas. Sci. Technol., 2023, 34: 045101

[本文引用: 1]

Dai A B, Yuan M D, Wu J W, et al.

Nondestructive characterization of elastic constants for 316L parts by selective laser melting based on dual-mode ultrasonic transducer

[J]. Chin. J. Sci. Instrum., 2021, 42(6): 95

[本文引用: 1]

戴安帮, 袁懋诞, 吴俊伟 .

基于双模式超声换能器的选择性激光熔化316L制件弹性常数无损表征研究

[J]. 仪器仪表学报, 2021, 42(6): 95

[本文引用: 1]

Prioul R, Bakulin A, Bakulin V.

Nonlinear rock physics model for estimation of 3D subsurface stress in anisotropic formations: Theory and laboratory verification

[J]. Geophysics, 2004, 69: 415

[本文引用: 1]

Mishra N, Das K.

A Mori-Tanaka based micromechanical model for predicting the effective electroelastic properties of orthotropic piezoelectric composites with spherical inclusions

[J]. SN Appl. Sci., 2020, 2: 1206

[本文引用: 1]

Wu J.

Thermal expansion coefficient prediction of asphalt mixture with the Eshelby equivalent inclusion theory

[J]. Appl. Mech. Mater., 2014, 584-586: 1071

[本文引用: 1]

Zhang Y, Han L. Foundation of Mesomechanics [M]. Beijing: Science Press, 2014: 91

[本文引用: 1]

张 研, 韩 林. 细观力学基础 [M]. 北京: 科学出版社, 2014: 91

[本文引用: 1]

Yang J J, Yu H C, Wang Z M, et al.

Effect of crystallographic orientation on mechanical anisotropy of selective laser melted Ti-6Al-4V alloy

[J]. Mater. Charact., 2017, 127: 137

[本文引用: 1]

Yang L, Zhao F, Jiang L, et al.

Development of composition and heat treatment process of 2000 MPa grade spring steels assisted by machine learning

[J]. Acta Metall. Sin., 2023, 59: 1499

DOI      [本文引用: 1]

The rapid development of rail transit has led to the proposition of higher requirements for the mechanical properties of springs and spring steels. Thus, bogies have been identified as the key components for trains to achieve high speed since they are connected with train bodies and wheel sets through springs. Alternatively, since the properties of spring steel materials have an important effect on the safety and comfort of high-speed trains, the development of spring steels with ultra-high strength and good plasticity has attracted the attention of researchers and industrial circles. However, simultaneously improving strength and plasticity has remained an important challenge for the research and development of high-end steels. Notwithstanding, machine learning has recently made substantial progress in designing and predicting various materials, and is expected to become a powerful tool for clarifying the relationship between the composition, process, and properties of complex alloys like steels. Based on the above background, this study reports the realization of rapid chemical composition and heat treatment process-design parameters for new spring steels, using a performance-oriented machine learning design system with high strength and good plasticity (tensile strength (2050 ± 50) MPa, elongation 10.5% ± 1.5%) after collecting literature data on spring steels and other typical quenched + tempered steels. Experimental studies were also carried out to obtain a further optimized heat treatment process (heating at 950oC for 30 min and oil quenching + tempering at 380oC for 90 min and water cooling). Investigations revealed that the tensile strengths of the two new spring steel materials developed were 2183.5 and 2193.0 MPa, their yield strengths were 1923.0 and 2024.5 MPa, their elongations after fracture were 10.5% and 9.7%, and the area reductions were 42.4% and 41.5%, respectively, with grain boundary strengthening and dislocation strengthening being the main strengthening mechanisms of the new spring steels. It was also observed that the fine grain size and appropriate amounts of austenite made the spring steels maintain good plasticity and have ultra-high strength. Moreover, compared with the existing ultra-high strength steels at the same strength grade, the new spring steels had significant technological and cost advantages. Hence, based on the above research, a new method and theory are provided to design chemical composition and heat treatment processes for quenched and tempered steels.

杨 累, 赵 帆, 姜 磊 .

机器学习辅助2000 MPa级弹簧钢成分和热处理工艺开发

[J]. 金属学报, 2023, 59: 1499

[本文引用: 1]

Zhang S Y, Lin X, Chen J, et al.

Influence of processing parameter on the microstructure and forming characterizations of Ti-6Al-4V titanium alloy after laser rapid forming processing

[J]. Rare Met. Mater. Eng., 2007, 36: 1839

[本文引用: 1]

张霜银, 林 鑫, 陈 静 .

工艺参数对激光快速成形TC4钛合金组织及成形质量的影响

[J]. 稀有金属材料与工程, 2007, 36: 1839

[本文引用: 1]

/