双相不锈钢脉冲激光焊接焊缝金属显微组织的发展 毕业论文外文翻译.doc

英文原文development of weld metal microstructures in pulsed laser welding of duplex stainless steel f. mirakhorli, f. malek ghaini, and m.j. torkamany (submitted october 30, 2011) the microstructure of the weld metal of a duplex stainless steel made with nd:yag pulsed laser is investigated at different travel speeds and pulse frequencies. in terms of the solidication pattern, the weld microstructure is shown to be composed of two distinct zones. the presence of two competing heat transfer channels to the relatively cooler base metal and the relatively hotter previous weld spot is proposed to develop two zones. at high overlapping factors, an array of continuous axial grains at the weld centerline is formed. at low overlapping factors, in the zone of higher cooling rate, a higher percentage of ferrite is transformed to austenite. this is shown to be because with extreme cooling rates involved in pulsed laser welding with low overlapping, the ferrite-to-austenite transformation can be limited only to the grain boundaries.keywords duplex stainless steel, microstructure, pulsed laser welding, solidicationintroductionduplex stainless steels (dss) are widely used in petro- chemical and chemical processings because of the combination of corrosion resistance and advantageous mechanical proper- ties. the wrought alloys microstructure at room temperature is composed of austenite and ferrite phases (ref 1, 2). however, the microstructure resulting from a fusion welding process can be signicantly different because of the cooling rates involved (ref 3-5). figure 1 depicts a typical dss alloy that would solidify completely into ferrite and then, while cooling through solid state transformation, it partially transforms into austenite (ref 1, 2). considering the comparatively higher cooling rates involved in welding processes, the weld metal and the haz microstructure could contain higher amounts of ferrite phase than the base metal. this also can affect the mechanical and corrosion resistance properties of dss welds (ref 2-7).welding dss alloys with continuous power laser has been the subject of previous research studies (ref 8-10). it is shownthat the low heat input and consequently high cooling rates canlead to the formation of higher a/c ratio. on the other hand, pulsed laser can provide further controls on power and heat input. however, there can be questions on how the microstruc- ture of a dss alloy is affected by the rapid pulsating nature of the heat source, since consecutive melting and solidication of weld spots would occur (ref 11-13).in the present study, the focus is on the evaluation of the microstructure in different regions in the weld metal of a dss and also analyzing the effect of variation in weld travel speed and pulse frequency.experimental procedurebead-on-plate laser welding was applied on 2-mm-thick commercial saf 2205 dss plate. the base metal chemical composition is given in table 1. laser welding machine was iql-10, with a pulsed nd:yag laser connected to a computer controlled working table and with a maximum mean laser power of 400 w. the available range for the laser parameters were 1-1000 hz for pulse frequency, 0-40 j for pulse energy, and 0.2-20 ms for pulse duration.during laser welding, argon shielding gas with a coaxial nozzle was used to protect the heated surface from oxidation. work pieces were polished and cleaned with acetone to be prepared for welding. the welded samples were observed in cross sections from three different perpendicular directions (top, transverse, and longitudinal). the etchant was beraha (0.7 k2s2o5 20 ml hcl in 100 ml solution). the wrought base metal consisted of 55% ferrite and 45% austenite as measured by image analysis, with an average hardness of 280 hv as measured by a 500 g load. after establishing the range of parameters to achieve an acceptable weld appearance, the experiments were carried out with varying travel speeds and pulse frequencies, as shown in table 2.overlap factor was calculated by the eq 1 (ref 12, 13). (eq 1)where t is the pulse duration, v is the welding speed, f is the laser frequency, and d refers to the laser spot size on the work piece measured as 0.9 0.1 mm. 3.results and discussionfigure 2 shows the top view of welds at a low and high overlapping. as observed from the gure, the weld spots are clearly distinguishable from each other specially at lower overlapping. when the time (or distance) between two pulses increases, high cooling rates can cause the earlier spots to solidify completely before coincident of the next pulse (ref 11). on the other hand, when the time (or distance) between two pulses decreases, the former spot temperature can still be high enough to the extent that semisolid condition is dominant and the next pulse can raise the temperature to a degree which can almost disappear the fusion line. the solidication pattern of the weld metal was found to vary with the travel speed and/ or frequency because of variations of the overlap factor. with a low overlapping factor, as shown in fig. 2(a), from a solidi- cation pattern point of view, two zones can be identiedzone i is the part of the weld metal which is remelted by the next pulse before becoming cooled thoroughly. in this zone, the grains nucleate on the previous spot epitaxially and grow toward the center.zone ii refers to a single pulse microstructure which is not affected by the next pulse heat and is solidied mainly from the base metal. in this part, the grain boundaries were relatively ner and more jagged.fig. 1 pseudo binary section of fe-cr-ni system at 70% irontable 1 chemical composition (in wt.%)elementcsimnpscrnimofewt.%0.031.01.420.0230.00522.315.483.34bal.between zones i and ii, there exists a very narrow band of material which is affected by the heat of the next pulse welding, i.e., the haz of zone i in zone ii. in fig. 2(a), this region is marked as 3, and from a solidication pattern point of view, it is a part of zone ii. the development of the observed solidica- tion patterns is because in pulse laser welding, when the weld spots are not too close to each other, the previous weld spot is relatively cool when the next pulse strikes, and therefore, effectively two different competing routes exist for the extraction of heat from any point in the molten weld pool. the rst route is directly through the side walls (fusion line with the base metal), and the second route is through the previous weld spot (fusion line between consecutive weld spots). the temperature distribution eld of the weld pool is affected by both of these two heat sinks. proximity of any point in the weld pool to each of these two routes of heat extraction is one of the factors determining the dominant cooling route and solidication orientation. the preferential solidication orien- tation is also affected by the orientation of the grains on which the weld metal grows epitaxially. zone 1 shown in fig. 2(a) is mainly cooled through heat transfer to the previous weld spot (and then to the base metal), but zone 2 is cooled through transferring heat from side walls to the base metal. also, when weld spots overlap each other extensively, zone 2 almost disappears and becomes only limited to a narrow band just next to the two side walls. in such condition (high overlapping), zone 1 solidication pattern dominates most of the weld metal central part, and they can effectively grow on each other epitaxially without being disturbed by zone 1 grains coming in between. here, the grains in the consecutive zone 1 s form a clear preferred orientation, and the axial solidication pattern is formed as shown in fig. 2(b). the authors have experienced pulsed laser welding of various alloys including carbon steel, aluminum alloys, and titanium. however, it was in the case of dss that such progress in understanding of the process of weld microstructures development became possible. it would be interesting to study the weld microstructures in other alloys in the light of the knowledge gained. as stated earlier, solidi- cation in zone 2 is dominated by heat extraction to the side walls, but solidication in zone 1 is dominated by heat extraction to the previous weld spot. the larger grain sizes in zone 1 are due to a higher effective preheat temperature of the material the heat of which escapes to i.e., the metal which itself has been molten just a little earlier. however, the sidewalls are expected to have a lower temperature, as a steeper temperature gradient occurs for solidication of zone 2. thus, the cooling rate in zone 1 is expected to be comparatively lower resulting in a coarser microstructure. now, our attention turns to the postfig. 2 microstructures of the welds with various overlap factors as viewed from top direction at (a) 55% (sampleb2) and (b) 73% (sample b4)fig.3 sem microstructure various parts of the weld metal and the base metal. (a) zone i in fig. 2(a). (b) zone ii in fig. 2(a). (c) the bulk of the weld metal (away from axial grains) in fig. 2(b). (d) the wrought base metalsolidication transformations and the corresponding micro- structural features as shown in fig. 3. from the phase diagram in fig. 1, the microstructure of the initial solidifying weld metal is expected to be fully ferritic and austenite, formed during solid state transformations. one might expect to nd a higher percentage of austenite in zone 1 because of its lower cooling rate. however, the percentage of austenite in zones 1 and 2 at 55% overlap was measured as 1.6 and 4.8%, respectively. a closer look at fig. 3(a) and (b) indicates that formation of austenite is limited to the grain boundaries. it seems that at lower overlaps, the cooling rate is so high that formation of austenite has been mainly limited to higher free energy sites at grain boundaries. in this sense, zone 2 material which has a ner solidication structure has provided relatively more preferential sites to form austenite rather than ferrite during the cooling process. however, on increasing the overlap factor to 73%, the cooling rate has decreased so much to form austenite both at the grain boundaries and within the grains, resulting in an austenite content of 18.5%, see fig. 3(c).the microhardness in these regions was measured using a 500-g load. at 55% overlap, zone 1 was found to have a higher hardness of 385 10 hv, in comparison with zone 2 a hardness of 328 10 hv. this lower hardness can be due to the presence of higher amount of austenite in the microstructureof zone 2. the result of vickers microhardness survey in b4 specimen with a high overlap factor of 73% showed mostly homogenous distribution of microhardness with an average value of 313 hv.fig. 4 microhardness prole along the transverse cross section of the weld in different travel speedsthe microhardness measurement was carried out on a number of transverse cross sections of weld specimens with various overlap factors. the result of hardness surveys is shown in fig. 4. basically, the hardness at the center of the weld pool is higher than that of the spots near the fusion line. higher hardness at the weld center can be due to the higher volume fraction of ferrite phase in this region.conclusionthe results of pulsed nd:yag laser welding dss are summarized as following:in pulsed laser welding, an interesting range of microstruc- tures is formed. dss provided an opportunity to enhance the understanding of how the solidication patterns and the resulting microstructures can develop. when the weld spots do not overlap each other much, two distinct zones of solidication can be identied. one zone is formed under the inuence of heat extraction directly to the base metal, while the second one is formed under the inuence of heat extraction to the previous weld spot. proximity to each of these two heat transfer channels and the presence of any strong preferred growth direction of the grains inherited through epitaxial nucleation determines the solidication pattern at any point. at low overlaps, these two zones appear consecutively in the middle part of the weld. by increasing the overlapping, the solidication pattern governed by heat extraction to the previous weld spot can form continuously without disruptionleading to an array of axial grains in the middle. the cooling rates involved in pulsed laser welding of dss can be so high that the formation of austenite from ferrite can be limited to grain boundaries resulting in an observation that some regions with a higher cooling rate may have slightly higher nal austenite contents.references1.j.c. lippold and d.j. kotecki, welding metallurgy and weldability of stainless steels, wiley & sons inc, new york, 20052.zh.l. jiang, x.y. chen, h. huang, and x.y. liu, grain renement 3.of cr25ni5mo1.5 duplex stainless steel by heat treatment, mater. sci. eng., 2003, a363, p 2632674.h. sieurin and r. sandstro m, austenite reformation in the heat- affected zone of duplex stainless steel 2205, mater. sci. eng., 2006, a418, p 2502565.g. berglund and p. wilhelmsson, fabrication and practical experience of duplex stainless steels, mater. des., 1989, 10, p 23286.p. bala srinivasan, v. muthupandi, v. sivan, p. bala srinivasan, and w. dietzel, microstructure and corrosion behavior of shielded metal arc- welded dissimilar joints comprising duplex stainless steel and low alloy steel, j. mater. eng. perform., 2004, 15(6), p 758764j.d. kordatos, g. fourlaris, and g. papadimitriou, the effect of cooling rate on the mechanical and corrosion properties of saf2205 (uns 31803) duplex stainless steel welds, scripta mater., 2001, 44, p 4014087.j.s. ku, n.j. ho, and s.c. tjong, properties of electron beam welded saf 2205 duplex stainless steel, j. mater. process. technol., 1997, 63(1997), p 7707758.r.s. huang, l. kang, and x. ma, microstructure and phase composition of a low-power yag laser-mag welded stainless steel joint, j. mater. eng. perform., 2005, 17(6), p 9289359.j. pekkarinen and v. kujanpaa, the effects of laser welding parameters on the microstructure of ferritic and duplex stainless steels welds, phys. procedia, 2010, 5, p 51752310.v. amigo, v. bonache, l. teruel, and a. vicente, mechanical properties of duple stainless steel laser joints, weld. int., 2006, 20(5), p 36136611.j. sabbaghzadeh, m.j. hamedi, f. malek ghaini, and m.j. torkamany, effect of process parameters on melting ratio in overlap pulsed laser welding, metall. mater. trans., 2008, 39b, p 34034712.f. malek ghaini, m.j. hamedi, m.j. torkamany, and j. sabbaghzadehb, weld metal microstructural characteristics in pulsed nd:yag laser welding, scripta mater., 2007, 56, p 95595813m.j. torkamany, m.j. hamedi, f. malek, and j. sabbaghzadeh, the effect of process parameters on keyhole welding with a 400 w nd:yag pulsed laser, j. phys., 2006, d39, p 45634567附录2 译文双相不锈钢脉冲激光焊接焊缝金属显微组织的发展 f. mirakhorli, f. malek ghaini, and m.j. torkamany (submitted october 30, 2011) 一种用钕（nd）焊接的双相不锈钢金属的显微组织:，研究在不同的行驶速度和脉冲频率下的yag脉冲激光。在凝固模式方面，焊缝的显微组织示出两个不同的区域组成。两个相互竞争的传热通道的相对较冷的基体金属与以前的焊点相对较热的存在下，提出开发两个区。在高重叠的因素，在焊缝中心线的阵列形成连续的轴向颗粒。区域中的较高的冷却速率，在低重叠的因素，较高的百分比的铁素体转变成奥氏体。这证明是由于涉及脉冲激光焊接具有低重叠，可以仅限于在晶界的铁素体 - 奥氏体相变的极端冷却速率。关键词 双相不锈钢的钢，显微组织，脉冲激光焊接，凝固1.引言双相不锈钢（dss）的广泛使用，由于拥有耐蚀性和良好地机械性能的，所以广泛用于石油化工和化学处理上。在室温下的锻造合金显微组织是由奥氏体和铁素体相互参杂的（参考文献1，2）。然而，产生的融合焊接过程中的微观结构，可以显着不同的，因为所涉及的冷却速率（参考文献3-5）。图1示出了典型的dss合金完全凝固成铁素体，然后，冷却的同时通过固态转化，部分转变为奥氏体（参考文献1，2）。考虑在焊接过程中所涉及的相对较高的冷却速率，在焊缝金属和haz的显微组织比母材中含有较多量的铁素体相。这也可以影响dss焊缝的机械性能和耐腐蚀性（参考2-7）。已经连续功率激光焊接的dss合金与先前研究的主题（参考8-10）。它显示热输入低，因此高的冷却速率可以导致形成较高的a / c比值。另一方面，脉冲激光可以提供进一步的控制，上电和热输入。然而，可以有一个dss合金的微观结构，关于如何由快速脉动特性的热源的影响，由于焊点的连续熔化和凝固，会发生的问题（参考文献11-13）。在本研究中，重点是关于评价的微观结构在不同的区域在焊缝金属中的dss，并分析焊接移动速度和脉冲频率的变化的效果。2.实验步骤钢板激光焊接采用2毫米厚的商业saf2205双相不锈钢板。基体金属的化学成分列于表1。激光焊接机是iql-10，用脉冲nd：yag激光的连接到一个计算机控制的工作表，具有最大平均激光功率为400 w的激光参数的可用范围为1-1000赫兹脉冲频率，0 -40 j的脉冲能量，脉冲持续时间为0.220毫秒。在激光焊接过程中，氩气保护气体与同轴喷嘴用于保护加热的表面不被氧化。工件进行抛光，并用丙酮清洗准备焊接。焊接的样品中观察到的横截面，从三个不同的垂直方向（顶部，横向和纵向）。该蚀刻剂是beraha（0.7 k2s2o5盐酸在20毫升100毫升溶液）。变形基体金属由55的铁素体和45的奥氏体，通过图像分析测量，所测得的500克负荷的平均硬度为280 hv。建立参数的范围内，以达到可接受的焊缝外观后，进行了实验，不同的行驶速度和脉冲频率，如表2中所示。重叠系数计算由式（1）（参考文献12，13）。 (eq 1)其中t是脉冲持续时间中，v是焊接速度，f是激光频率，d是指在工件上测量为0.90.1毫米的激光光斑大小。 3.结果与讨论图2是顶视图，显示出低和高的重叠焊缝区。正如从图中观察时，焊接点清楚地相互区分，特别在较低的重叠。当两个脉冲之间的增加的时间（或距离），高的冷却速率可能会导致早期斑点完全凝固之前的下一个脉冲（参考文献11）相重合。另一方面，两个脉冲之间的时间（或距离）减小时，前点的温度仍可以是足够高的范围内，半固体的状态是显性的，下一个脉冲的温度提高到一定程度，可以几乎消失熔合线。的焊接金属的凝固模式被发现由于重叠系数的变化的行进速度和/或频率变化。具有低重叠因子，如图所示。图2（a），从固相线fication柄的角度来看，两个区域可以被识别：区域i是部分焊缝金属重熔在下一个脉冲，方可彻底冷却。在此区域，外延以前区域出现了类似谷物核的中心生长。区域ii指的是一种单脉冲是不会受到下一个脉冲的热量，主要的基体金属凝固微观结构。在这一部分中，晶界相对更精细，更锯齿状。fig. 1 pseudo binary section of fe-cr-ni system at 70% irontable 1 chemical composition (in wt.%)区域i和ii之间，存在一个很窄的频带的材料，它是由下一个脉冲焊接，即，在区ii区i的haz的热的影响。在图图2（a）中，这个区域被标记为3时，从凝固图案的角度来看，它是第二区域的一部分。所观察到的凝固模式的发展是因为在脉冲激光焊接时，在焊接点都不会太接近对方，以前的焊点时，下一个脉冲的打击，因此，存在两个不同的有效竞争的航线相对凉爽用于提取的热量从焊接熔池中的任何点。第一个路由是直接穿过侧壁（与基体金属的熔合线），而第二路径是通过以前的焊点（连续的焊点之间的熔合线）。熔池的温度分布领域的影响由这两个散热器。这两个航线的热提取熔池中的任何点的接近程度是占主导地位的散热途径，并凝固方向的决定因素之一。的优惠凝固方向塔季翁也受焊缝金属在其上外延生长的晶粒的取向。示于图1区。图2（a）主要是通过热传递到前一个焊点（然后对基体金属）冷却，但冷却区2，通过将热量从侧壁的基体金属。此外，当焊点彼此重叠的广泛，区域2几乎完全消失，变成只限于两个侧壁毗邻的窄带。处于这种状态（重叠），区域1的凝固模式占主导地位的焊接金属中心部分，它们能够有效地相互外延生长在不被干扰的1区晶粒之间。在这里，连续区1形成了明显的择优取向，轴向的凝固模式中的晶粒形成的，如图所示。图2（b）。作者都经历包括碳钢，铝合金，钛，各种合金的脉冲激光焊接。然而，这是决策支持系统的情况下，焊缝微观结构发展的过程中，理解这样的进步成为可能。这将是有趣的研究中获得的知识，其他合金焊缝微观结构。如前所述，在区域2的凝固阳离子主要是通过热提取的侧壁，但在区域1的凝固主要是由以前的焊点的热提取。在区域1中的较大的晶粒尺寸是由于更高的有效的预热温度的热逸出的物质，即，金属本身已经被熔融早一点。然而，预期的侧壁有一个较低的温度，更陡的温度梯度发生凝固区2。因此，区域1中的冷却速度预期为较粗的微观结构，得到的值相对较低。fig. 2 microstructures of the welds with various overlap factors a