张念先，Megumi Kawasaki，黄 毅，Terence G. Langdon,3(.南安普顿大学工程与环境学院,英国南安普顿SO7 BJ；2.汉阳大学材料与工程学院, 韩国首尔33-79；3.南加州大学航空航天,机械工程与材料科学学院, 美国洛杉矶CA90089-453)

晶粒尺寸对大塑性变形的两相合金超塑性的影响

张念先1，Megumi Kawasaki2,3，黄 毅1，Terence G. Langdon1,3
(1.南安普顿大学工程与环境学院,英国南安普顿SO17 1BJ；2.汉阳大学材料与工程学院, 韩国首尔133-791；3.南加州大学航空航天,机械工程与材料科学学院, 美国洛杉矶CA90089-1453)

在室温下对共晶铅锡合金(Pb-62% Sn)进行了高压扭转变形(HPT).采用不同时间的自退火制备不同晶粒尺寸的样品.最长的自退火时间为12天.通过研究这些样品在室温下不同的拉伸行为，从而获得了关于超塑性性能的结果.之后，本文通过将这些结果与等通道挤压(ECAP)获得的样品进行比较，不仅证明了所有样品都具有良好的超塑性，而且具有小晶粒尺寸的样品更容易在大应变速率的条件下获得超塑性.

等通道转角挤压；高压扭转；铅锡合金；自退火；超塑性

Superplasticity is defined as the ability of polycrystalline materials to pull out uniformly in tension to extremely large elongations prior to failure. These elongations, often exceeding 1000%, are evidence of the occurrence of superplasticity[1]. In practice, superplasticity requires an elongation of at least 400% and a measured strain rate sensitivity (SRS), m, close to 0.5[2]. Additionally, since superplastic flow is controlled by diffusion processes, it requires a testing temperature of 0.5Tmor greater whereTmis the absolute melting temperature of the material. A grain size less than 10 μm is necessary to facilitate grain boundary sliding (GBS) which is the dominant deformation mechanism for superplastic flow[3]. However, this introduces difficulties because grain growth usually occurs relatively easily at elevated temperatures. Thus, two-phase alloys are often used to obtain superplastic properties because the presence a second phase restricts grain growth at high temperatures.

The Pb-62% Sn eutectic alloy is a traditional and highly studied model material for investigations of superplasticity. An early investigation showed the possibility of producing an elongation of up to >4,000% in this alloy[4]and later the Pb-Sn alloy was used to produce a record superplastic elongation of 7,550%[5]. Nevertheless, these results were achieved using conventional processing techniques to reduce the grain size, such as rolling or extrusion, and therefore it was not possible to achieve exceptional grain refinement. With the recent development of several new processing techniques, it is now feasible to produce the bulk Pb-Sn alloy with a very small grain size and therefore it is important to investigate the superplastic properties under these conditions.

The application of severe plastic deformation (SPD) to bulk solids represents a relatively new metal processing technique in which materials are heavily strained through the introduction of a high density of dislocations but with limited change in the overall dimensions of the samples[6]. Recently, much attention has focused on the two SPD techniques of equal-channel angular pressing (ECAP)[7]and high-pressure torsion (HPT)[8]where the procedures of ECAP and HPT are both capable of producing exceptionally refined grains. However, it should be noted that metals processed by ECAP generally have ultrafine grain sizes within the submicrometer range whereas processing by HPT has the potential of producing materials where the grain size may lie within the true nanometer range which is less than 100 nm[9]. A very recent investigation showed that it is possible to produce a homogenous microstructure with an average grain size of 0.8 μm in the Pb-Sn alloy processed by HPT[10]and in the following sections examples are presented for superplastic flow in the Pb-Sn alloy processed by either HPT or ECAP.

Superplasticity is a diffusion-controlled process in which the steady-state strain rate,ε, may be described by a relationship of the form[11]:

(1)

whereAis a dimensionless constant,Dis the appropriate diffusion constant [=Doexp(-Q/RT) whereDois a frequency factor,Qis the activation energy,Ris the gas constant andTis the absolute temperature],Gis the shear modulus, b is the Burgers vector,kis Boltzmann’s constant,dis the grain size,σis the flow stress andnandpare the stress and inverse grain size exponents, respectively. The flow behaviour in superplasticity is generally expressed by Eq. (1) withn=2,p=2 andQ=QgbwhereQgbis the activation energy for grain boundary diffusion[12]. Thus, it follows that a decreasing grain size will lead to the occurrence of superplasticity at faster strain rates and this prediction was subsequently confirmed in experiments on two commercial aluminium alloys[13]where superplastic elongations were achieved at fast strain rates above 10-2·s-1which is within the region of high strain rate superplasticity[14].

The present experiments were initiated specifically to examine the superplastic behaviour of the Pb-Sn alloy with different grain sizes and to compare the results with published data for the same alloy using different processing techniques.

1 Experimental material and procedures

The experiments were conducted using a commercial Pb-62% Sn eutectic alloy which was supplied as a cast billet with dimensions of 34 mm×20 mm×15 mm. Discs were machined from this billet having diameters of 10 mm and thicknesses of 1.2～1.5 mm and then each side of the discs was carefully polished using abrasive papers to give a series of HPT disc samples having thicknesses of 0.80±0.01 mm.

The HPT processing was conducted using a conventional facility operating under quasi-constrained conditions in which there is a small outflow of material around the periphery of the disc during the pressing operation[15-16]. A compressive pressure was applied using a load of 235 kN corresponding to an imposed pressuresP, of 3.0 GPa. The disc samples were processed by HPT forN=1 turn at a fixed rotational speed of 1 r/min. This alloy readily self-anneals at room temperature (RT) and therefore the processed discs were stored at RT for different periods of up to 12 days in order to produce samples having a series of increasing grain sizes. An earlier report described the self-annealing of this alloy after processing by HPT[17].

Following the SPD processing, all samples were subjected to careful grinding and polishing with abrasive papers and diamond paste. In order to reveal the grain structures, each polished disc was etched for 30～40 s in a solution of 25 ml H2O, 5 ml HCl with a concentration of 37% and 5 g of NH4NO3. The nature of the microstructures was systematically investigated using a scanning electron microscope (SEM) equipped with a high-brightness field emission gun (JEOL JSM-6500F). Spatial grain sizes,d(d=1.74×L, whereLis the value of the linear intercept grain size), were determined from the SEM images using separate measurements of at least 300 individual grains.

To evaluate the superplastic properties, samples were cut from the processed material and then tested in tension. For the HPT discs, the tensile samples had gauge lengths of 1.5 mm and cross-sections of 1.0 mm× 0.6 mm and these samples were tested under conditions of constant cross-head displacement with initial strain rates from 1.0 × 10-4to 1.0 × 10-1·s-1. The results gave flow stresses for the different testing conditions and provided information on the elongations to failure and the values of the SRS. All of the HPT samples were tested at RT where this is reasonable because the alloy has a very low melting temperature ofTm=140 ℃. This means that deformation at RT is at least within the regime of warm working[18]so that superplastic flow is possible.

In order to examine the effect of different grain sizes, data were also taken from an earlier report which described the processing of the same Pb-Sn alloy by ECAP[19]. In this earlier report, an experimental alloy was prepared by casting 38% Pb and 62% Sn (mass fraction) of commercial purity and ECAP billets were then machined from the cast bars having square cross-sections of 14 mm×14 mm and lengths of 70 mm. The processing by ECAP was also conducted at RT and used an ECAP die with an internal channel angle ofΦ=120(°) and an arc of curvature at the intersection of the channels ofΨ=0(°). These angles lead to an imposed strain of 0.6 on each passage of the billet through the die[20]. The ECAP billets were pressed repetitively from 1 to 5 passes using processing route A in which the billets are not rotated between each pass[21]. For the ECAP samples, tensile specimens were cut with tensile axes lying parallel to the pressing direction and with gauge lengths of 4 mm and cross-sectional areas of 3 mm×2 mm and the specimens were then pulled to failure at an elevated temperature of 150 ℃.

2 Experimental results

2.1 Superplastic properties of samples processed by HPT

Figure 1 shows SEM images of representative microstructures of the Pb-Sn alloy at the edge of a disc after processing by HPT for 1 turn and then storing at RT for (a) 1 day, (b) 4 days and (c) 12 days, where the dark grey phase represents the Sn-rich phase and the light grey phase represents the Pb-rich phase. Microstructures at the edge regions are shown in Fig.1 because the subsequent tensile tests were performed using material from this region of each disc. In Fig.1 it is readily apparent that HPT processing introduces a banded structure in which agglomerates of both phases delineate the direction of torsional flow which is from the top to the bottom in the images in Fig.1. Similar microstructural evolution was reported in the Zn-22% Al eutectoid alloy after processing by HPT[22]. It is evident that with a short storage time of 1 day the width of the banded phases is significantly smaller in Fig.1(a) than after the longer storage times of 4 and 12 days in Fig.1(b) and (c), respectively. This change is due to a regrouping of neighboring phases of the same type and therefore through the formation of larger phases during self-annealing.

The occurrence of grain growth is also obvious in Fig.1 and this is further confirmed by the grain size measurements presented in Table 1. It should be noted that the average grain sizes were measured using the separate Sn grains but with the individual Pb-rich phase measured as a single grain regardless of whether measurements by electron backscatter diffraction (EBSD) indicated that the Pb-rich phases contain several grains. This procedure is consistent with the preferred interfaces for plastic deformation by GBS in the Pb-Sn alloy, since it is well established that most of the sliding occurs between Sn-Sn and Pb-Sn boundaries[23]and the objective of this research was to focus on investigating the superplastic behaviour of the Pb-Sn alloy.

In Fig.2, the tensile results for the HPT samples are summarized in a plot of elongation to failure (upper) and flow stress (lower) against the initial strain rate for samples stored for 1, 4 and 12 days. It is apparent from the upper plot that the HPT samples exhibit typical superplastic elongations of at least 400 % at a strain rate of 1.0×10-4·s-1and the maximum elongation is 630%. Thus, the alloy after HPT demonstrates excellent superplastic ductility at the slower strain rates. The lower plot shows there is a high value ofm≈0.45 for the HPT-processed samples stored at RT for 1 and 4 days at strain rates slower than 1.0×10-2·s-1, thereby suggesting the occurrence of significant GBS at these slower strain rates. It should be noted that the small elongations (<400%) obtained atmvalues close to 0.45 are due to the relatively low testing temperature of RT.

Careful inspection of the upper plot in Fig.2 shows that the sample stored for only 1 day with a finer measured grain size, as documented in Table 1, achieved higher elongations at faster strain rates than in the other samples. Thus, an elongation of 270% was recorded after storage for 1 day at a strain rate of 1.0×10-2·s-1and this contrasts with the measured elongations of 190% and 110% at the same strain rate for samples stored for 4 and 12 days, respectively. It is also apparent that, as the strain rate decreases, higher elongations are then achieved with samples having coarser grain sizes at these slower strain rates.

In Fig.2 the highest elongation of 630% was recorded in a sample stored for 4 days and then tested at a strain rate of 1.0 × 10-4·s-1. Figure 3 shows this specimen after pulling to failure together with an initial untested sample. The sample showing this highest elongation pulled out uniformly with no evidence for the formation of any necking within the gauge length. It is well established that neck-free flow and uniform elongation is a significant characteristic of superplasticity[24].

2.2 Superplastic properties of samples processed by ECAP

Based on an earlier report[19], Table 2 gives the average grain sizes of samples processed by ECAP for 1 to 5 passes. These measurements show that processing by ECAP reduces the grain size and there is only a minor decrease in grain size with increasing numbers of ECAP passes.

The elongation to failure (upper) and the flow stress (lower) are plotted against the initial strain rate for samples processed by ECAP through 1 to 5 passes in Fig.4[19]. The values of the SRS measured in the lower part of Fig.4 are within the range of 0.3 to 0.5 which suggests the occurrence of superplastic flow. This is confirmed by the elongations shown in the upper part of Fig.4 where, excluding the results forN=1 pass, almost all of the specimens show elongations to failure of more than 1000% at 10-4-10-2·s-1Similar to the results shown in Fig.2, higher elongations occur also at the faster strain rates in the samples processed through 5 passes where the grains are the smallest. The maximum elongations occurred in samples processed through 2 and 4 passes at a slower strain rate of 1.0×10-3·s-1. It is also interesting to note that, comparing Figs 2 and 4, the peak in elongation is displaced to faster strain rates when the testing temperature is increased. Thus, in Fig.2 the maximum elongation occurs at a strain rate of 1.0×10-4·s-1or possibly even at a slower strain rate since this was the lowest rate recorded experimentally whereas in Fig.4 the peak elongation at 150 ℃ occurs at a strain rate of 1.0×10-3·s-1. Similar results for the effect of temperature on superplastic elongation were also reported in other studies[25].

3 Discussion

It is demonstrated in Figs. 2 and 4 that fine grains favour superplasticity occurring at faster strain rates compared with coarser grains. This trend is in excellent agreement with the prediction of Eq. (1) and the fundamental model for superplastic flow by GBS[12]. Nevertheless, the results shown in Figs. 2 and 4 are limited by the relatively small numbers of samples.

In order to provide a more comprehensive overview of the superplastic behaviour of the Pb-Sn alloy, additional results were collated and these are summarized in Table 3[5, 19, 25-30]. The results in Table 3 are based on several different processing routes including cold rolling[5,29]and hot extrusion and annealing[30]without SPD processing, conventional ECAP[19,25,27], ECAP and cold rolling[28]and ECAP using a T-shaped die[26]: Table 3 summarizes the relevant spatial grain sizes, the testing temperatures, the strain rates and the measured elongations to failure defined as ΔL/Lo% where ΔLis the increase in length andLois the initial gauge length, respectively. The excellent superplastic properties of the Pb-Sn alloy are readily apparent from inspection of the reported elongations in Table 3. Generally, most of the larger elongations are achieved at relatively slow strain rates with samples having slightly coarser grain sizes. For example, the record-breaking elongation of 7,550% was achieved at a strain rate of 2.1×10-4·s-1in a sample by cold rolling and having an initial grain size of 11.6 μm[5].

In order to more clearly examine the significance of grain size on superplasticity in the Pb-Sn alloy, the elongations obtained at RT in Table 3 were plotted against the initial strain rate in Fig.5 together with the results obtained in the present investigation. All of the datum points in Fig.5 divide reasonably into two groups with grain sizes from 2.6 to 5.7 μm denoted by open symbols and grain sizes from 6.1 to 14.8 μm denoted by solid symbols. It is evident that samples with the finer grain sizes show a rapid increase in the elongation as the strain rate is decreased compared with samples having relatively coarser grain sizes and this.

Tendency is depicted by the two separate solid curves shown in Fig.5. It is obvious from Fig.5 that samples with the finer grains achieve superplastic elongations at relatively faster strain rates and samples with coarser grains achieve superplastic elongations at relatively slower strain rates, thereby demonstrating the displacement of the superplastic region towards faster strain rates in response to the reduction in grain size.

* The elongation of 7,550% represents a world record for tensile superplasticity.

Additionally, all samples show similar and small elongations at strain rates from 1.0×10-2to 1.0×10-1·s-1regardless of the different initial grain sizes, but samples with finer grains reach maximum elongations at faster strain rates but with smaller elongations whereas samples with coarser grains reach maximum elongations at slower strain rates with larger elongations. An example is given by the peak elongation of 1610% at a strain rate of 1.3×10-4·s-1for a sample having a grain size in the range from 2.6 to 5.7 μm which contrasts with the elongation of 2 230% at a strain rate of 6.6×10-6·s-1for a sample having a grain size in the range from 6.1 to 14.8 μm.

4 Summary and conclusions

(1)The Pb-62% Sn eutectic alloy was processed by HPT and then subjected to self-annealing to introduce a range of grain sizes.

(2)After HPT, samples were pulled to failure in tension and the results demonstrate that superplasticity is achieved in testing at room temperature. The results are compared with published data for the same alloy processed by ECAP where superplasticity was achieved at a testing temperature of 150 ℃. The largest elongation for a sample processed by HPT was 630% at an initial strain rate of 1.0×10-4·s-1.

(3)It is demonstrated that fine grains favour the occurrence of superplastic flow at relatively faster strain rates, where this is consistent with the conventional theoretical mechanism for superplasticity.

[1]Chokshi A H, Mukherjee A K, Langdon T G. Materials Science and Engineering R Reports, 1993, 10(93):237-274.

[2]Langdon T G. Seventy-five years of superplasticity: historic developments and new opportunities[J]. Journal of Materials Science, 2009, 44(22):5998-6010.

[3]Langdon T G. An evaluation of the strain contributed by grain boundary sliding in superplasticity[J]. Materials Science and Engineering A, 1994, 174(2):225-230.

[4]Ahmed M M I, Langdon T G. Exceptional ductility in the superplastic Pb-62Pct eutectic[J]. Metallurgical Transactions A, 1977, 8(11):1832-1833.

[5]Ma Y, Langdon T G. The characteristics of cavitation in superplastic metals and ceramics[J]. Metallurgical and Materials Transactions A, 1996, 27(4):873-878.

[6]Valiev R Z, Islamgaliev R K, Alexandrov I V. Bulk nanostructured materials from severe plastic deformation[J]. Progress in Materials Science, 2000, 45(2):103-189.

[7]Valiev R Z, Langdon T G. Principles of equal-channel angular pressing as a processing tool for grain refinement[J]. Progress in Materials Science, 2006, 51(7):881-981.

[8]Zhilyaev A P, Langdon T G. Using high-pressure torsion for metal processing: fundamentals and applications[J]. Progress in Materials Science, 2008, 53(6):893-979.

[9]Bazarnik P, Huang Y, Lewandowska M,etal. Structural impact on the Hall-Petch relationship in an Al-5Mg alloy processed by high-pressure torsion[J]. Materials Science and Engineering A, 2015, 626:9-15.

[10]Zhang N X, Kawasaki M, Huang Y,etal. Microstructural evolution in two-phase alloys processed by high-pressure torsion[J]. Journal of Materials Science, 2013, 48(13):4582-4591.

[11]Langdon T G. The mechanical properties of superplastic materials[J]. Metallurgical Transactions A, 1982, 13(5):689-701.

[12]Langdon T G. A unified approach to grain boundary sliding in creep and superplasticity[J]. Acta Metallurgica Et Materialia, 1994, 42(94):2437-2443.

[13]Valiev R Z, Salimonenko D A, Tsenev N K,etal. Observations of high strain rate superplasticity in commercial aluminum alloys with ultrafine grain sizes[J]. Scripta Materialia, 1997, 37(12):1945-1950.

[14]Higashi K, Mabuchi M, Langdon T G. High-strain-rate superplasticity in metallic materials and the potential for ceramic materials[J]. ISIJ International, 1996, 36(12):1423-1438.

[15]Figueiredo R B, Cetlin P R, Langdon T G. Using finite element modeling to examine the flow processes in quasi-constrained high-pressure torsion[J]. Materials Science and Engineering A, 2011, 528(28):8198-8204.

[16]Figueiredo R B, Pereira P H R, Aguilar M T P,etal. Using finite element modelling to examine the temperature distribution in quasi-constrained high-pressure torsion[J]. Acta Materialia, 2012, 60(6-7):3190-3198.

[17]Zhang N X, Kawasaki M, Huang Y,etal. The significance of self-annealing in two-phase alloys processed by high-pressure torsion[J]. IOP Conference Series-Materials Science and Engineering, 2014:012126.

[18]Doherty R D, Hughes D A, Humphreys F J,etal. Current issues in recrystallization: a review[J]. Materials Science and Engineering A, 1997, 238(97):219-274.[19]Kawasaki M, Mendes A D A, Sordi V L,etal. Achieving superplastic properties in a Pb-Sn eutectic alloy processed by equal-channel angular pressing[J]. Journal of Materials Science, 2011, 46(1):155-160.

[20]Iwahashi Y, Wang J, Horita Z,etal. Principle of equal-channel angular pressing for the processing of ultra-fine grained materials [J]. Scripta Materialia, 1996, 35(2):143-146.

[21]Furukawa M, Iwahashi Y, Horita Z,etal. The shearing characteristics associated with equal-channel angular pressing [J]. Materials Science and Engineering A, 1998, 257(2):328-332.

[22]Kawasaki M, Ahn B, Langdon T G. Microstructural evolution in a two-phase alloy processed by high-pressure torsion[J]. Acta Materialia, 2010, 58(3):919-930.

[23]Vastava R B, Langdon T G. An investigation of intercrystalline and interphase boundary sliding in the superplastic Pb-62% Sn eutectic[J]. Acta Metallurgica, 1979, 27(2):251-257.

[24]Langdon T G. Fracture processes in superplastic flow[J]. Materials Science, 1982, 16(4):175-183.

[25]El-Danaf E A, Khalil K A, Soliman M S. Effect of equal-channel angular pressing on superplastic behavior of eutectic Pb-Sn alloy[J]. Materials and Design, 2012, 34:235-241.

[26]Rao V S, Kashyap B P, Prabhu N,etal. T-shaped equi-channel angular pressing of Pb-Sn eutectic and its tensile properties[J]. Materials Science and Engineering A, 2008, 486(1):341-349.

[27]Kawasaki M, Lee S, Langdon T G. Constructing a deformation mechanism map for a superplastic Pb-Sn alloy processed by equal-channel angular pressing[J]. Scripta Materialia, 2009, 61(10):963-966.

[28]Lugon L P, Figueiredo R B, Cetlin P R. Tensile behavior of an eutectic Pb-Sn alloy processed by ECAP and rolling[J]. Journal of Materials Research and Technology, 2014, 3(4):327-330.

[29]Ahmed M M I, Langdon T G. The effect of grain size on ductility in the superplastic Pb-Sn eutectic[J]. Journal of Materials Science Letters, 1983, 2(7):337-340.

[30]Soliman M S. Effect of strain rate and grain size on the ductility of superplastic Pb 62% Sn eutectic at room temperature[J]. Scripta Metallurgica Et Materialia, 1995, 33(6):919-924.

Influence of grain size on superplastic properties of a two-phase Pb-Sn alloy processed by severe plastic deformation

Nian Xian Zhang1, Megumi Kawasaki2,3, Yi Huang1, Terence G. Langdon1,3
(1.Materials Research Group, Faculty of Engineering and the Environment,University of Southampton, Southampton SO17 1BJ, U.K;2.Division of Materials Science and Engineering, Hanyang University, Seoul 133-791, South Korea; 3.Departments of Aerospace & Mechanical Engineering and Materials Science, University of Southern California, Los Angeles, CA 90089-1453, U.S.A.)

A Pb-62% Sn eutectic alloy was processed by high-pressure torsion (HPT) at room temperature (RT) and then different grain sizes were introduced by subjecting samples to self-annealing for times of up to 12 days. Tensile specimens were cut from the HPT discs and then pulled to failure in order to investigate the development of superplastic properties. The results are compared with similar data reported for this alloy after processing by ECAP. It is shown that samples processed by ECAP or HPT exhibit excellent superplastic elongations at 150 ℃ and RT, respectively, and it is demonstrated also that samples with fine grains show improved ductility at faster strain rates.

equal-channel angular pressing; high-pressure torsion; Pb-Sn alloy; self-annealing; superplasticity

10.14186/j.cnki.1671-6620.2015.04.003

TG 146.2

A

1671-6620(2015)04-0255-08