SLM制备Ti6Al4V时缺陷同拉伸性能的关系

来源：：长三角G60激光联盟

据悉，美国Lawrence Livemore国家实验室的学者研究了SLM制备Ti6Al4V时缺陷同拉伸性能的关系，其成果发表在《Materials & Design》上。

成果的Graphical abstract
来自美国Lawrence Livemore国家实验室的学者在《Materials & Design》上发表了论文《Defects -dictated tensile properties of Selective laser melted Ti6Al4V》，为我们报道了如下研究成果：

一个正常的能量图用来识别用于高能量激光制备Ti6Al4V（简写为Ti64）样品的工艺图。

图 Lawrence Livemore国家实验室

屈服强度和均匀的拉伸伸长率通过高质量的样品的显微组织和缺陷的生成来判断。

小比例的空穴（≤ 1%）可以在拉伸时生长和聚集，从而会影响到总的应变到断裂的过程。

图 沉积态的Ti64样品的相组成和显微组织：

a) Synchrotron XRD patterns collected for six different samples. Small peaks arising from β-Ti are pointed with arrows in V8 and H8. b) 3D reconstruction of the microstructure using optical micrographs (OMs). c) EBSD mapping on the BD/LD surface of an as-printed Ti64 sample (IPF image). d) Pole figures (0001 and 11–20 reflections) of the area highlighted by the black dashed-line in c) and stereographic projection of the parent β grain reconstructed based on these two pole figures. e) IPF of the high-temperature β microstructure based on c). f) and g), Bright field TEM pictures showing the defect structures in as-built materials, consisting of compression twins, tension twins, and 〈c + a〉 dislocations.

微塑形的早起发生是由于制造的缺陷所造成的。边缘的扫描道是缺陷产生的重要来源，从而影响拉伸性能。

成果简介：
采用SLM（又叫LPBF）来沉积制备金属，经常会存在缺陷，如位错、孪生、元素偏析、杂质和气孔等，从而正面的或负面的影响到产品的机械性能。在这里来自美国Lawrence Livemore国家实验室的研究人员系统的研究了室温条件下准静态时SLM制备的Ti64合金的拉伸行为，并采用当前的原位同步X射线衍射（SXRD）和SXCT扫描技术进行了在拉伸屈服强度和均匀伸长时，主要由沉积态的显微组织来决定，而应变-失效则同气孔的关系非常敏感，甚至是高密度样品时（大于99.5%）时也是如此，原位SXRD揭示了在沉积态时微塑性的萌生起源于应力水平，显著的显示为早起晶格应变偏离行为。SXCT揭示了气孔的生长机制是在拉伸轴垂直于制造方向的时候，而在沿着制造方向时却观察不到这一现象。这一不均匀的气孔生长机制导致了SLM时的应变-失效的巨大差异。研究人员所发现的熔池动力学模拟和实验结果揭示了早先并不清楚的气孔源的生成机制，即扫描道边缘的气孔。同时研究人员还为大家展示了一种正常的能量图来识别获得高质量样品的优化的工艺区间。

图  沉积态Ti64钛合金的样品的机械性能

a) Stress/strain curves obtained during tensile testing at 10−3/s. b) Yield strength versus strain-to-failure taken from the data in a) and compared to the literature. c) Normalized work hardening as a function of strain for four samples, each representative of a group presented in b). Dashed lines highlight a normalized work hardening of 1 which represents the beginning of the necking. d) Yield strength versus uniform elongation taken from the data presented in a).

Ti64由于具有高强度、低密度、高耐蚀性和生物相容性，而成为应用最为广泛的钛合金，从而广泛的应用到航空航天、海洋、医疗和动力工程。然而，Ti64合金难以铸造和机加工，这是因为该合金溶体具有较高的反应活性和较低的热传导性。增材制造技术吸引了人们的广泛注意。但在实际应用过程中，一个主要的问题在于SLM制备的Ti64存在较差的拉伸韧性，这是因为在SLM制备过程中存在脆性的马氏体相α’，这是在采用SLM制备钛合金时的一个非常严重的问题，需要克服。早起的研究，大多聚焦在如何将α’相分解成更为韧性的α+β相上来，采取的策略是改变工艺参数和/或后续热处理。改变β相使其成为强且韧的相的研究，目前还很少看到有报道。作为比较，很少有研究用于揭示SLM制备Ti64时的拉伸性能的影响因素，这一性能不仅受到内在的显微组织和缺陷的影响，同时还对气孔等非常敏感，这是增材制备时常见的问题。

图 在拉伸测试时原位 SXRD 在P1垂直方向（ (a, c, e)）和P1平行方向 (b, d, f)时的测试结果

Lattice strain and FWHM as a function of the macroscopic engineering stress for the six planes: . a) and b) Engineering strain/stress curves for the two samples. The crosses mark the end of the uniform elongation. Note that SXRD data collected after necking are not shown in this figure. c) and d) Evolution of the lattice strain along both tensile and compressive directions in the two samples. Note that the {0002} peak in both samples could not be accurately fitted along the compressive direction due to weak intensity. All lattice strains are normalized by the initial d-spacing measured in the as-printed state, just before loading. e) and f) FWHM for the same set of planes in the two samples along the tensile direction. Note that FWHM measurement accuracy is in the range of ±0.002 1/nm to ±0.005 1/nm, depending on the reflection.

早期的研究工作曾经指出，气孔的形状、尺寸、分布和体积分数、强烈的影响着拉伸强度和拉伸韧性。同时研究已经指出，气孔同时还会显著的影响到制品的疲劳性能。对相对来说比较脆性的材料，少量的气孔会严重的影响到拉伸韧性。当然，也有人认为，空穴也许会有积极的效果，甚至会由于位错-空穴的相互作用而可以促进拉伸韧性性能的提高。由于Ti64是一种低韧性的合金，相似的负面效应也许会获得。众所周知，SLM时，激光参数（功率、速度）、层厚、扫描策略、支撑结构等均会影响最终的相和晶粒尺寸的，同时也会影响到气孔的含量。无论如何，由于气孔在高密度样品中的作用的不确定性，气孔对拉伸性能的影响还是值得详细的研究的。

图 在原位SXRD拉伸测试时织构的变化

a) and b) {0002} peak distribution over the entire diffracted ring as a function of the macroscopic strain for both V8 and H2 samples. Note that the y-axis does not follow a linear scale since SXRD data acquisition was controlled by the time and not the strain. Stars of the same color are 65° apart, corresponding to a set of matrix/compression twins. Circles of the same color are 85° apart, corresponding to a set of matrix/tension twins. Azimuthal angles of 0 and 180 correspond to the loading direction (marked with solid white lines). In b), the blue arrows highlight compression twin activity with one peak intensity decreasing when the other one increases. c) and d) Engineering tensile curves for the same samples V8 and H2, respectively. Red crosses correspond to the locations of the blue curves in a) and b).

在这里，研究人员采用原位同步辐射SXRD技术和扫描技术SXCT来研究缺陷，即预先存在的位错和孪生，以及气孔对SLM制造的Ti64拉伸性能的影响。

图 同步X-CT技术在拉伸测试前和拉伸测试后得到的实验结果

Synchrotron X-ray computed tomography conducted before and after tensile testing in several dog-samples with a 1 × 1 mm2 gauge section. Red regions correspond to pores. a) and b) As-printed samples cut from plates P1 and P7, respectively. c) Yield strength versus strain-to-failure of the samples concerned by this SXCT study. d) Optical images of dog-bone samples cut vertically and horizontally from the plate P1. e) Same samples as in d) analyzed by XCT after failure.

图 气孔的分布和体积分数

a) and b), respectively in the samples P1 (Horizontal and Vertical) before (as-printed) and after (postmortem) tensile test, measured by SXCT. UR and NR are the uniform and necking regions of the dog-bone sample after test, respectively.

图 上图中（a）中的2D横截面和3D重构的结果：

a) Locations of the two cross-sections displayed in b) and c). b) Cross-section corresponding to the green plane in a). This plane cut through the laser turn-around induced pores. c) Cross-section corresponding to the blue plane in a). This plane follows the pores aligned along the laser track. The dimensions correspond to the average distance between pores calculated from the projected intensity profiles in red and blue.

图 本实验后得到的正常的能量图用于工艺判断 （见红色的部分）

a previous study conducted with Ti64 pillars printed in the same conditions [37], and the literature. a) Normalized heat input diagram (log10 scale) where the dash lines are the results of the product of 1/h⁎ and E⁎ (see text). b) Sample density as a function of the normalized heat input (semi-log10 scale).

图：采用 ALE3D代码在不同的激光参数下进行扫描时得到的气孔形成的模拟图

The first column (a, b, c) corresponds to the set P1 (laser power of 100 W, and speed of 600 mm/s) while the second column (d, e, f) corresponds to the set P7 (laser power of 250 W, and speed of 1300 mm/s). The interface between the liquefied or solidified Ti64 and the ambient is represented in yellow color. The red envelope shows the solid/liquid interface. a) and d) show the melt pool from below the surface. b) and e) the side views at a given time during printing, while c) and f) are side views after the laser was turned off.



本研究的主要结论如下：

L-PBF制备 Ti64样品得到的显微组织为为马氏体，其强度可以达到 (>1110 MPa),该强度的提供得益于复杂的显微组织的组成，主要是细小的α′板条、元素偏析和高密度的缺陷，缺陷主要是位错、和孪生。

1.构建了一个正常的能量图，该图可以使我们直接比较其他SLM制备和EBM制备Ti64钛合金的样品时的工艺窗口。该工艺图可以进一步的识别本工作中的高密度样品。

2.我们对高密度样品进行的拉伸测试，沿着不同的制备方向和在不同的测量几何形状的条件下，显示出增材制造的Ti 64样品的屈服强度和均匀的伸长率主要由材料内在的显微组织所确定。而应变-失效则强烈的依赖于气孔的存在，即使是其体积率 <1%。均匀的应变-失效可以推荐为在增材制造材料时拉伸韧性的有效评估手段。

3.TEM和XRD分析结果进一步的证明了高密度位错和孪生在沉积态显微组织中的存在，其主要来源于马氏体的转变和在打印过程中的塑性变形。材料的强度的增加得益于位错和孪生的自由移动或相互作用。小比例的β相在SLM制备的样品中也存在。

4.原位同步X射线衍射显示，在沉积态的Ti64合金中，微塑性的萌生在应力水平比较低的时候，只有宏观屈服强度的一半。这表明变形行为的比较小的差异，甚至是在不同的参数下加工的样品或在不同的方向上加载也是如此。

5.X-CT结果揭示了气孔的聚集和生长长大，从而造成的失效机制，这是当拉伸轴垂直于制造方向的时候，导致了更加有限的应变-失效，而这一机制并没有在平行于制造方向的时候观察到。

6.多物理场模拟结果揭示了我们并没有知道的在SLM制备钛合金时的气孔源的生成机制：扫描道边缘的生成。他们造成了空穴在粉末颗粒之间的存在，从而看起来对拉伸性能的影响非常大。

文章来源：Defects-dictated tensile properties of selective laser melted Ti-6Al-4V,Materials & Design,Volume 158, 15 November 2018, Pages 113-126,https://doi.org/10.1016/j.matdes.2018.08.004