Micro-alloying of yttrium in Zr-based bulk metallic glasses

Micro-alloying of yttrium in Zr-based bulk metallic glasses Wei-jie PENG, Yong ZHANG State Key Laboratory for Advanced Metals and Materials, University of Science and Technology Beijing, Beijing 100083, China Received 18 October 2010; accepted 23 December 2010 Abstract: The effect of yttrium addition on the glass-forming ability (GFA) and mechanical properties of the Zr-based (Zr0.525Al0.1Ti0.05Cu0.179Ni0.146)100−xYx and (Zr0.55Al0.15Ni0.1Cu0.2)100−xYx (x=0, 0.2, 0.4 0.6, 1, 2) alloys was studied. Micro-alloying of 0.6% yttrium enhances the room temperature ductility as well as the GFA of the Zr-based alloys. The mechanism of enhancing the GFA and room temperature ductility was analyzed. It is indicated that proper yttrium addition stabilizes the undercooled liquid by means of forming lots of ordered clusters, thus improving the GFA and the room temperature ductility. Key words: micro-alloying; glass-forming ability; Zr-based alloys; yttrium

1 Introduction Bulk metallic glasses (BMGs) have the potential to be the structural materials as well as the military materials, e.g. to substitute of the depleted uranium[1]. Thanks to a variety of techniques, the BMG samples with distinct shapes have been created[2−3]. It was reported that oxygen impurities have a detrimental effect on Zr-based BMGs and can embrittle them[4−5]. Zirconium oxide may lead to nucleation of crystalline Zr2Ni in the undercooled liquid. This is the main reason why the Zr-based glass-forming alloy is so sensitive to the oxygen content[6]. Highly purified raw materials and strong oxygen affinity elements are both selected so that the detrimental effect of oxygen contamination can be eliminated[7]. It was widely recognized that yttrium can scavenge the oxygen[8−10]. According to thermodynamic principle, yttrium has a stronger affinity with oxygen atom than zirconium, because yttrium has much higher formation enthalpy with oxygen (1 905.0 kJ/mol) than zirconium (1100.8 kJ/mol). Yttrium can react with the oxide impurities during processing of sample preparation and form particles. And the particles are benign to heterogeneous nucleation[8, 11]. Thus, the glass forming ability (GFA) of Zr-based alloys is improved by scavenging oxygen. It was reported that the micro-alloying of yttrium can enhance GFA by three mechanisms[4, 8, 11−13], namely, scavenging oxygen, suppressing the competitive

crystal phase and decreasing liquidus temperature. In this work, the effect of yttrium addition in the Zr52.5Al10Ti5Cu17.9Ni14.6 (ZAT) and Zr55Al15Ni10Cu20 (ZA) alloys is studied by focusing on whether scavenging the oxygen is the main reason by micro-alloying of yttrium to enhance GFA of Zr-based alloys and whether the Y2O3 phases have a lower potency as heterogeneous nucleation sites than other oxides.

2 Experimental Alloy ingots with compositions of (ZAT)100−xYx (x=0, 0.2, 0.4, 0.6, 1, 2 ) and (ZA)100−xYx (x=0, 0.2, 0.4, 0.6, 1) were prepared by vacuum arc melting a mixture of pure Zr (99.8%) (mass fraction), Cu (99.9%), Ni (99.9%), Al (99.9%) and Ti (99.5%) in a purified argon atmosphere. The ingots were remelted several times to ensure its compositional homogeneity, and then the melt was cast into Cu mold to form rod-shaped samples with 1 mm and 5 mm in diameter and wedge shape samples. Rapidly solidified ribbons were prepared in an argon atmosphere with a single roller melt-spinning technique. The wheel speed was maintained at 20 m/s. The composition and microstructure of the alloys were studied by X-ray diffraction (XRD), differential scanning calorimetry (DSC), transmission electron microscopy (TEM) and scanning electron microscopy (SEM). Y2O3 chips were prepared by spark plasma sintering (SPS) at temperature of 1 300 °C and pressure of 40 MPa. In order to study the wetting properties of the Zr-based

Corresponding author: Yong ZHANG; E-mail: [email protected]; [email protected]

Wei-jie PENG, et al/Progress in Natural Science: Materials International 21(2011) 46−52

alloys with the Y2O3 chips, the ZAT and ZA alloys were heated to liquidus temperature of Zr-based alloys in a vacuum furnace.

47

However, the changes of onset melting temperature were not apparent.

3 Results 3.1 XRD results Figure 1 shows the XRD patterns of (ZAT)100−xYx (x=0, 0.2, 0.4, 1, 2) alloys. Increasing Y addition from 0.2% to 0.4%, the peaks of crystalline phase of Zr2Ni became fewer and weaker, indicating that the GFA of Zr-based alloys was improved by micro-alloying with Y. The GFA of (ZAT)100−xYx (x=0, 0.2, 0.4, 1, 2) alloys was greatly enhanced with 0.4% yttrium addition. It exhibited mainly amorphous features but a few crystal peaks were still visible. When yttrium addition reached 1%, more crystalline peaks appeared, indicating an increase of the volume fraction of crystalline phases and a Zr5Al3 phase formed. The result confirms that a proper yttrium addition greatly improves the GFA of the ZAT alloy by suppressing the precipitation of the Zr2Ni phase. Further addition leads to the precipitation of Zr5Al3 phase.

Fig.2 XRD patterns of (ZA)100−xYx (x=0, 0.2, 0.4, 0.6, 1) alloy rod samples

Fig.3 DSC traces obtained from (ZAT)100−xYx (x=0, 0.2, 0.4, 0.8, 1, 2) alloy rod samples

Fig.1 XRD patterns of (ZAT)100−xYx (x=0, 0.2, 0.4 ,1, 2 ) alloy rod samples

Figure 2 shows the XRD patterns of the (ZA)100−xYx (x=0, 0.2, 0.4, 0.6, 1) alloys. It is obvious that the crystalline Zr2Ni peaks became fewer and weaker with yttrium addition, and that the best GFA was obtained with 0.6% yttrium addition. 3.2 Thermal analysis Figures 3 and 4 present the DSC curves of (ZAT)100−xYx (x=0, 0.2, 0.4, 0.8, 1, 2) and (ZA)100−xYx (x=0, 0.2, 0.4) alloys, respectively. No exothermic peak was observed for the alloy without yttrium addition, indicating that no amorphous phase formed in the as-cast ZAT and ZA alloys. With proper yttrium addition, an exothermic peak was observed, confirming the formation of amorphous phase induced by micro-yttrium addition.

Fig.4 DSC traces obtained from (ZA)100−xYx (x=0, 0.2, 0.4) alloy rod samples

3.3 SEM studies The backscattered electron image of ZAT alloys is given in Fig.5(a). The main crystalline phase was Zr2Ni

48

Wei-jie PENG, et al/Progress in Natural Science: Materials International 21(2011) 46−52

Fig.5 SEM images of (ZAT)100−xYx alloy rod samples: (a) ZAT; (b) x=0.2; (c) x=0.4; (d) x=1

phase. It is consistent with the results of XRD analysis. The microstructure changed dramatically with yttrium addition. Fig.5(b) shows Y2O3 phase appearance with 0.2% yttrium addition. When yttrium addition reached 0.4%, a few Y2O3 particles distributed on amorphous matrix (Fig.5(c)). The results indicated that the GFA of ZAT alloys was increased. But, the further increase of yttrium addition could lead to more crystalline phases (Fig.5(d)). Figure 6 shows SEM images of wedge cast sample of (ZA)100−xYx (x=1) alloy. It is found that yttrium addition could lead to the precipitation of Y2O3 phase. But Y2O3 phase led to the precipitation of a new phase. EDS analysis indicated that the new phase was Zr5Al3 phase and grew with decreasing cooling rate. There was more and bigger Zr5Al3 phase in the tail of wedge alloy compared with the tip of wedge alloy. 3.4 TEM studies Figure 7 shows TEM images of (ZAT)100−xYx (x=0, 0.6) ribbon samples. No nano-crystals were observed by TEM technique in the ZAT sample (Fig.7(a)). Figure 7(b) shows the TEM image of ZAT ribbon sample with 0.6% yttrium addition. Crystalline phase was observed. The analysis of the electron diffraction patterns showed that the crystal was Y2O3 phase with a BCC structure and lattice parameter a=10.60 Å. Based on these results, it is concluded that the Y2O3 phase is very likely to precipitate. This was also confirmed by the fact that yttrium has a great affinity with oxygen atom[14].

Fig.6 SEM images of (ZA)100−xYx (x=1) wedge casting sample: (a) Tip of wedge casting sample; (b) Middle part of wedge casting sample; (c) Tail of wedge casting sample

Wei-jie PENG, et al/Progress in Natural Science: Materials International 21(2011) 46−52

Fig.7 TEM images and SAED patterns of ZAT (a) and (ZAT)99.4Y0.6 (b) ribbon samples

Zr2Ni phase (FCC structure and lattice parameter a=12.27 Å) in the ZAT rod sample alloy was observed, as shown in Fig.8. The crystal structure and lattice parameter are very different from Y2O3 phase. Thus, the precipitation of Y2O3 phase affects the nucleation of Zr2Ni phase[15].

Fig.8 TEM image and SAED pattern of ZAT alloy rod sample

In the ZAT alloy rod sample with 0.6% yttrium addition, some changes of the microstructure occurred. Firstly, some new phases precipitated, such as Zr5Al3 and YZrO3 phases, as shown in Fig.9. Secondly, a lot of ordered clusters occurred, as shown in Fig.9(a). The observation of Zr5Al3 phase is consistent with the result of XRD analysis. The analysis of the electron diffraction patterns showed that the Zr5Al3 phase has a hexagonal structure and lattice parameters a=8.184 Å and c=5.702 Å, and the YZrO3 phase has a BCC structure and lattice parameter a=10.51 Å. The crystal structure and lattice

49

Fig.9 TEM images and SAED patterns of (ZAT)100−xYx (x=0.6) alloy rod sample: (a) hexagonal structure Zr5Al3phase; (b) BCC structure YZrO3 phase

parameter of YZrO3 phase are very similar to those of Y2O3 phase. Figure 10 shows the HRTEM images of YZrO3 phase. It is surprising that the zone 1 of YZrO3 phase has no obvious boundary with amorphous matrix (Fig.10(b)). It seems that the YZrO3 phase has dissolved in the matrix. The HRTEM images of zone 2 of YZrO3 phase shows an ordered crystal phase. As for the zone 3, it looks like a collection of many ordered clusters. It is presumed that the YZrO3 phase has formed by ordered clusters. 3.5 Wetting test It is found that yttrium oxide can lead to the precipitation of Zr5Al3 phase in wedge casting sample of (ZA)100−xYx (x=1) alloy. Figure 11 shows the morphology of Y2O3 chips. The interface between Y2O3 chip and ZAT alloy is given in Fig.12(a). There is no obvious reaction between Y2O3 chip and ZAT alloy. However, for the ZA alloy, a white phase formed in the interface (Fig.12(b)). EDS analysis indicated the composition of the white phase is 8.47% Al, 11.16% Cu, 26.27% Zr, 9.32% Y, 39.82% O and 4.96% Ni. 3.6 Mechanical properties Figure 13 shows the compressive stress — strain curves of (ZA)100−xYx (x=0.6, 1) rod samples with diameter of 1 mm. The curve of (ZA)100−xYx (x=0.6,) rod sample presents high strength as well as high ductility. The fracture strength can reach up to 2 700 MPa. Figure 14 shows the fracture surface of (ZA)100−xYx

50

Wei-jie PENG, et al/Progress in Natural Science: Materials International 21(2011) 46−52

Fig.10 TEM images of ZAT alloy: (a) BCC YZrO3 phase; (b) HRTEM image of zone 1 in Fig.10(a); (c) HRTEM image of zone 2 in Fig.10(a); (d) HRTEM images of zone 3 in Fig.10(a)

Fig.11 Morphology of Y2O3 chips prepared by SPS technique Fig.13 Compressive stress—strain curves of (ZA)100−xYx (x=0.6, 1) rod samples with diameter of 1 mm

Fig.14 Fracture surface of (ZA)100−xYx (x=0.6) rod sample

Fig.12 Interface between Y2O3 chip and ZAT (a) and ZA (b) alloy

(x=0.6,) rod sample. Vein patterns and crack were observed. The mechanical property was improved because the matrix was scavenged with 0.6% yttrium addition. However, further yttrium addition lead to decline of mechanical properties.

Wei-jie PENG, et al/Progress in Natural Science: Materials International 21(2011) 46−52

4 Discussion 4.1 Purification of alloy by yttrium micro-alloying Yttrium has a great affinity with oxygen atom[14]. Besides, it can react with the impurity oxygen during processing to form particles. SEM and TEM studies indicate the presence of Y2O3 phase. The formation of crystalline Zr2Ni can be triggered by zirconium oxide nuclei. This is the main reason why the Zr-based glass forming alloy is so sensitive to oxygen content. Yttrium addition can suppress Zr2Ni phase by scavenging oxygen. Moreover, primary phase changes to yttrium oxide (Y2O3 phase), which tends to precipitate even in ribbon samples. In yttrium free alloy oxygen-induced nuclei are Zr4Ni2O phase. It has similar crystal structure and lattice parameter to Zr2Ni phase. Y2O3 phase is formed with yttrium addition. And crystal structure and lattice parameter differ enormously with Zr2Ni phase. 4.2 Mechanism of enhancing GFA by yttrium micro-alloying in ZAT alloy The beneficial effect of Y on glass formation was reported by the following reasons[16−22]: 1) Y improves the manufacturability of these alloys by scavenging the oxygen impurity from it, via the formation of innocuous yttrium oxides; 2) Y adjusts the compositions closer to the eutectic and thus lowers their liquidus temperature; 3) there are appropriate atomic-size mismatch and large negative heat of mixing between Y and existing constitutive elements. DSC results indicate that the liquidus temperature does not change, so yttrium addition can not adjust the composition closer to the eutectic for the ZAT alloy. Based on these results, we can conclude that, yttrium micro-addition enhances the GFA only in case of atomic-size mismatch and large negative heat of mixing between Y and existing constitutive elements for the ZAT alloy. Yttrium has the largest atomic radius of 1.80 Å among all constituent elements and a large negative heat of mixing with constituent elements such as Cu(−22 kJ/mol), Al(−38 kJ/mol). The addition of Y causes the more sequential change in the atomic sizes, as well as the generation of new atomic pairs with large negative heat of mixing[22]. Consequently, the topological and chemical short range orders increase with yttrium addition, thus increasing the packing density of the undercooled liquid with low atomic diffusivity. Then, the liquidus structure is changed, and the ordered clusters precipitate easily. Fig.9(a) shows the presence of many ordered clusters. The ordered clusters may stabilize the liquidus structure by decreasing the Gibbs free energy of the undercooled liquid. However, further addition changes the structure to excess and does not obtain a

51

high GFA composition. The composition modification leads to the precipitation of Zr5Al3 phase. 4.3 Mechanism of yttrium micro-alloying in ZA alloy For the ZA alloy, Y2O3 phase can act as a new heterogeneous nucleant for Zr5Al3 phase. Wetting test also shows a reaction interface between ZA alloy and Y2O3 phase. Although Y2O3 phase can act as a new heterogeneous nucleation site for Zr5Al3 phase in the ZA alloy, the matrix is scavenged. Thus, the mechanical property is improved. 4.4 Action of Ti element Yttrium addition has different effect on the ZAT and ZA alloys. The ZA alloy has an obvious reaction with Y2O3 chip (Fig.12(b)), and Y2O3 particles can act as heterogeneous nucleants for Zr5Al3 phase in the ZA alloy. The composition difference between ZAT and ZA alloy mainly lies in Ti content. Ti is less likely to react with Y, and the solid solubility of Ti in Y or Y in Ti is very small (Ti-Y phase diagram). Therefore, the ZAT alloy with Ti is less likely to react with Y2O3 phase as well. But in Ti free alloy (ZA), the other constituent elements have large negative heat of mixing with Y. Therefore, yttrium is not a proper addition element to enhancing the GFA in the ZA alloy.

5 Conclusions 1) Proper yttrium micro-addition is a viable way to improve the GFA of ZAT BMGs. Yttrium changes the liquidus structure and forms ordered clusters. The ordered structure stabilizes the liquidus. 2) The addition of yttrium leads to the formation of Zr5Al3 phase in the ZA alloys. Yttrium oxides may act as the heterogeneous nucleants for Zr5Al3 phase. Since the ZA alloy reacts with Y2O3, yttrium addition does not apply for all alloys in the case of enhancing GFA. 3) (ZA)100−xYx (x=0.6) rod sample presents high room temperature ductility as well as high strength. The fracture strength reaches up to 2 700 MPa.

References [1]

[2]

[3]

[4]

ZHANG Y, LIN J P, CHEN G L, et al. Mechanical properties of interface between U-0.75Ti alloy and TiC [J]. Journal of University of Science Technology Beijing, 1998, 5: 41−43. WANG S, ZHANG Y. Shear-band spacing controlled by Bridgman solidification in dendrite/BMG composites [J]. Science in China Series G: Physics, Mechanics & Astronomy, 2009, 52: 1632−1636. ZHANG X C, ZHANG Y, CHENG X H, et al. Bulk metallic glass rings prepared by a modified water quenching method [J]. International Journal of Minerals, Metallurgy and Materials, 2009, 16: 108−111. ZHANG Y, PAN M X, WANG R J, et al. Formation of Zr-based bulk metallic glasses from low purity of materials by yttrium addition

52 [5]

[6]

[7]

[8]

[9]

[10]

[11] [12]

[13] [14]

Wei-jie PENG, et al/Progress in Natural Science: Materials International 21(2011) 46−52 [J]. Mater Trans JIM, 2000, 41: 1410−1414. LIU C T, CHISHOLM M F, MILLER M K. Oxygen impurity and microalloying effect in a Zr-based bulk metallic glass alloy [J]. Intermetallics, 2002, 10: 1105−1112. ALTOUNIAN Z, BATALLA E, STROM-OLSEN J O. The influence of oxygen and other impurities on the crystallization of NiZr2 and related metallic glasses [J]. J Appl Phys, 1987, 61: 149−155. YAN M, SHEN J, ZHANG T, et al. Enhanced glass-forming ability of a Zr-based bulk metallic glass with yttrium doping [J]. J Non-Cryst Solids, 2006, 352: 3109−3112. LU Z P, LIU C T, PORTER W D. Role of yttrium in glass formation of Fe-based bulk metallic glasses [J]. Appl Phys Lett, 2003, 83: 2581−2583. WANG H J, SHIFLET G J, POON S J, et al. The role of Y/lanthanides on the glass forming ability of amorphous steel [J]. Appl Phys Lett, 2007, 91: 141910−1−3. LIU C T, LU Z P. Effect of minor alloying additions on glass formation in bulk metallic glasses [J]. Intermetallics, 2005, 13: 415−418. LU Z P, LIU C T. Role of minor alloying additions in formation of bulk metallic glasses: A review [J]. Mater Sci, 2004, 39: 3965−3974. CHEN J, ZHANG Y, HE J P, et al. Metallographic analysis of Cu-Zr-Al bulk amorphous alloys with yttrium addition [J]. Scripta Mater, 2006, 54: 1351−1355. KUBASCHEWSKI O, ALCOCK C B. Metallurgical Thermochemistry [M]. Oxford: Pergamon, 1979. JIANG F, WANG Z J, ZHANG Z B, et al. Formation of Zr-based

[15]

[16]

[17]

[18]

[19]

[20]

[21] [22]

bulk metallic glasses from low purity materials by scandium addition [J]. Scripta Mater, 2005, 53: 487. KUNDIG A A, LEPORI D, PERRY A J. Influence of low oxygen contents and alloy refinement on the glass forming ability of Zr52.5Cu17.9Ni14.6Al10Ti5 [J]. Materials Transactions, 2002, 43: 3206−3210. XU D H, DUAN G, JOHNSON W L. Unusual glass-forming ability of bulk amorphous alloys based on ordinary metal copper [J]. Phys Rev Lett, 2004, 92: 245504−1−4. INOUE A, ZHANG T, TAKEUCHI A. Bulk amorphous alloys with high mechanical strength and good soft magnetic properties in Fe-TM-B (TM=IV-VIII group transition metal) system [J]. Appl Phys Lett, 1997, 71: 464−466. SHEN T D, SCHWARZ R B. Bulk ferromagnetic glasses prepared by flux melting and water quenching [J]. Appl Phys Lett, 1999, 75: 49−51. PONNAMBALAM V, POON S J, SHIFLET G J. Fe-based bulk metallic glasses with diameter thickness larger than one centimeter [J]. Mater Res, 2004, 19: 1320−1323. PONNAMBALAM V, POON S J, SHIFLET G J. Fe-Mn-CrMo-(Y,Ln)-C-B (Ln=Lanthanides) bulk metallic glasses as formable amorphous steel alloys [J]. Mater Res, 2004, 19: 3046−3052. LU Z P, LIU C T, THOMPSON J R, et al. Structure amorphous steels [J]. Phys Rev Lett, 2004, 92: 245503−1−4. INOUE A, ZHANG W, ZHANG T, et al. Cu-based bulk glassy alloys with high tensile strength of over 2000 MPa [J]. J Non-Cryst Solids, 2002, 304: 200−209.