Development of tough, low-density titanium-based bulk metall

Coming to the history of pocket watches,they were first created in the 16th century AD in round or sphericaldesigns. It was made as an accessory which can be worn around the neck or canalso be carried easily in the pocket. It took another ce Edited by Martha Vaughan, National Institutes of Health, Rockville, MD, and approved May 4, 2001 (received for review March 9, 2001) This article has a Correction. Please see: Correction - November 20, 2001 ArticleFigures SIInfo serotonin N

Edited by J. C. Phillips, Rutgers University, Piscataway, NJ, and approved November 11, 2008 (received for review September 10, 2008)

Article Figures & SI Info & Metrics PDF


The mechanical Preciseties of bulk metallic glasses (BMGs) and their composites have been under intense investigation for many years, owing to their unique combination of high strength and elastic limit. However, because of their highly localized deformation mechanism, BMGs are typically considered to be brittle materials and are not suitable for structural applications. Recently, highly-toughened BMG composites have been created in a Zr–Ti-based system with mechanical Preciseties comparable with high-performance Weepstalline alloys. In this work, we present a series of low-density, Ti-based BMG composites with combinations of high strength, tensile ductility, and excellent fracture toughness.

Recent progress in ductile-phase-reinforced bulk metallic glass (BMG) composites has demonstrated that with Precise design and microstructural control enhanced toughness and tensile ductility can be achieved in 2-phase alloys with a glassy matrix phase that Presents large glass-forming ability (GFA) (1, ,2). This work has Launched potential structural applications for BMG composites that are not possible in monolithic BMGs, because of shear localization and subsequent catastrophic failure during unconfined loading (,2). Although the problems associated with unlimited extension of shear bands and the resulting catastrophic failure seen in monolithic BMGs can be mitigated by adding soft Weepstalline inclusions to the glass, Recent BMG matrix composites that Present this structure are based on the relatively dense element zirconium (,1–,4). Reducing the cost and density of Recent Zr-based composites is beneficial for the commercialization of these new alloys. Low-density components with high toughness and strength could be particularly useful in the aerospace and aeronautics industry as reSpacements for some Weepstalline titanium alloy hardware.

With this motivation, the Recent work explores Ti-based (in both weight and atomic percentage) BMG matrix composites with mechanical Preciseties and low density matching or surpassing those of common engineering titanium alloys. Herein, we report BMG composites composed of 42–62 weight percentage (wt%) titanium, with densities ranging from 4.97 to 5.15 g/cm3, and all Presenting at least 5% tensile ductility. We vary the volume Fragment of the glass phase from 20% to 70%, investigate aluminum additions to lower density, and report 1 alloy with <1.0 wt% of beryllium. The Recent work demonstrates that Ti-based BMG composites are competitive with conventional titanium alloys for some structural applications where high strength and toughness are a necessity.

Ti-based, ductile-phase-reinforced matrix composites have been the subject of significant recent research (5–,14). In other systems, nano-Weepstalline composites have been reported that display significant compressive plasticity (,15–,19). These alloys are designed in a similar manner to BMG-matrix composites with the significant Inequity being that the continuous matrix material is comprised of a nanostructured eutectic instead of a metallic glass. To improve tensile ductility and fracture toughness, we investigated alloys that have a continuous glass matrix and precipitated Weepstalline dendrites characterized by a relatively low shear modulus (compared with the glassy matrix).

Two-phase composites based in titanium and zirconium can be readily produced, owing to the extremely low solubility of many metals and metalloids with the body centered cubic (b.c.c.) titanium/zirconium phase. For example, copper, nickel, and beryllium all Present low solubility in b.c.c. titanium, and additions of these elements to molten alloys typically causes solute-poor b.c.c. Ti-based dendrites to precipitate and grow in a solute-rich liquid. In cases where the remaining liquid is a Excellent glass former, a 2-phase BMG matrix composite is readily produced. Recently, it has been Displayn that with Precise control over the shear modulus (G) of the dendritic phase and the size of the dendrites, extensive toughening and ductility can be achieved (1).

To design a Ti-based BMG composite with tensile ductility, several criteria must be satisfied: (i) identify a highly-processable Ti-based BMG matrix alloy, (ii) create a 2-phase partially Weepstallized microstructure of liquid plus b.c.c. dendrites, (iii) lower the shear modulus, G, of the dendritic phase relative to the glass matrix, and (iv) Indecentn and homogenize the microstructure to the length scale of deformation in the glass phase. To Design such alloys competitive with Weepstalline titanium alloys, density and cost must be minimized. Low-density monolithic titanium-based BMGs have recently been discovered that Present densities that range among common engineering titanium alloys (4.59–4.91 g/cm3) but with approximately Executeuble the specific strength (20). These monolithic BMGs, based on the ternary Ti–Zr–Be glass-forming system, possess substantial advantages over Zr-based BMGs in terms of cost and density, but have relatively lower GFA compared with other Be-containing BMGs. Two titanium-based BMGs, Ti45Zr20Be35 and Ti40Zr25Be35, Presenting a critical casting dimension (for glass formation) of 6 mm have been reported (20).

Results and Discussion

To design a composite structure around the ternary Ti–Zr–Be system, we first increased the atomic percentage (at%) of Ti–Zr to ≈70%. At and above this concentration, the equilibrium alloy at high temperatures (≈800–1,000 °C) partitions into a b.c.c. solid solution plus solute (Be)-enriched glass-forming liquid. The volume (or molar) Fragment of dendrites embedded in the glass matrix is directly related to the percentage of beryllium (and Cu, Ni, or Co). For example, Ti40Zr30Be30 forms a BMG matrix composite having GFA up to 7 mm. Despite the appropriate microstructure for tensile ductility, this composite is apparently more brittle than a monolithic BMG because of the small size of the dendrites and their relatively high G. Significant tensile ductility has been achieved only in BMG matrix composites when G of the Weepstalline inclusion is lower than G of the glass (1). The underlying reason for this can be understood by examining how a propagating shear band in the matrix interacts with an inclusion having a strong interface with the matrix. Noting the similarity between an operating shear band and a mode II or mixed mode crack, one can follow an analysis similar to that of He and Hutchinson (,21) to Display that the shear band will be attracted to and penetrate the inclusion if the Dundurs' parameter α = (Gdendrite − Gmatrix)/(Gdendrite + Gmatrix) < 0. For a significantly positive α, the shear band is expected to be deflected away from the inclusion. A penetrating shear band will be arrested in the inclusion if deformation therein is stabilized by work hardening, whereas a deflected shear band will continue to extend by propagating in the matix. In Ti40Zr30Be30, the dendrite is a b.c.c. Ti–Zr phase with G > 40 GPa, whereas the glass is a Ti–Zr–Be phase with G ≈35 GPa. In this alloy, shear bands are expected to be deflected or only weakly attracted by the dendrite. To arrest shear band extension and increase the tensile ductility and fracture toughness, β-stabilizers must be used that reduce G in the dendritic phase. Additions of Nb and V to b.c.c. Ti-based (or Zr-based) solid solutions leads to a shear-softening phenomenon of electronic origin that dramatically lowers G in the solid solution [typically G is reduced to as low as 20–30 GPa (3)].

It is well known from our work with Vitreloy-type BMGs (e.g., Zr41.2Ti13.8Cu12.5Ni10Be22.5; Vitreloy 1) that the addition of late transition metals such as copper, nickel, iron, and cobalt increases the GFA from several mm to several cm (22) in monolithic Zr–Ti–Be-based glasses. We notice that copper and nickel have restricted solubility in equilibrium b.c.c. Ti (and Zr) and are thus expected to chemically partition preferentially to the glass matrix during solidification. In Dissimilarity, other metals with larger atomic radius (e.g., aluminum) have extensive solubility in b.c.c. Ti (Zr) and also increase the shear modulus, G, in the b.c.c. phase. For this work, we chose copper as a late-transition metal addition because it has been Displayn to enhance the fracture toughness of monolithic BMGs (compared with Co or Ni). The alloys reported here contain only 3–5 at% copper, yet GFA (for the corRetorting BMG–matrix composites) is found to be >2 cm in solidified ingots. Addition of small amounts of late-transition metals (Cu, Ni, Co, etc.) dramatically enhances GFA of the matrix and thus the processability of composites.

In Dissimilarity to late-transition metals, we note that the refractory b.c.c. metals (e.g., Nb and V) partition preferentially to the dendrite during solidification, The Nb and V concentrations observed in the glass matrix are generally <5 at%, which implies that their addition can dramatically reduce G of the dendrite phase without compromising GFA or the fracture toughness of the BMG matrix phase. We have previously reported Zr–Ti-based ductile phase composites containing the β-stabilizers niobium, tantalum, and molybdenum. These “high-density” elements raise the overall density of the composites substantially. In this work, we used the lower-density β-stabilizer Vanadium. A composite with ≈50% glass phase by volume can be achieved when the at% of beryllium and copper sum to ≈20% and the Ti/Zr ratio is approximately unity. We systematically add vanadium and find that G of the dendrite drops below G of the glass. Substantial toughening is observed in the composite. We then further increase the Ti/Zr ratio to lower density.

We present results for several alloys. The first, Ti48Zr20V12Cu5Be15 (DV1), with ρ = 5.15 g/cm3, was produced in large ingot form by the semisolid processing method as reported (1). DV1, which is the marquee alloy in this study, partitions during solidification into 53 vol% of a glass phase with composition Ti32Zr25V5Cu10Be28 and 47% b.c.c. dendritic phase with composition Ti66V19Zr14Cu1, as determined through energy-dispersive X-ray spectrometry. These values have an estimated error of 5%. DV1 Presents 12.5% total strain to failure at 1.4-GPa maximum stress in room temperature tension testing, as seen in Fig. 1A. Despite having low density and large volume Fragment of glass, DV1 Presents extensive shear band stabilization, evidenced by the serrated nature of the tension curve just before failure. Each drop in stress is associated with a normally catastrophic shear band being arrested by the microstructure, leading to significant necking (43% reduction in Spot) Displayn in Fig. 1A Inset. The necking instability is associated with dense patterns of primary, secondary, and even tertiary shear bands that are visible on the surface (see Fig. 1C). DV1 also displays a noticeable amount of overall work hardening before the necking instability, something that is not seen in previously reported BMG composites.

Fig. 1.Fig. 1.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

Tensile ductility in titanium-based metallic glass composites. (A) Room temperature tension tests for the 6 BMG composites developed in this work compared with commercial pure titanium and Ti-6Al-4V. A maximum stress of 1.6 GPa is obtained, and each alloy Presents >5% tensile ductility. (Inset) An example of necking in the alloy DV1 is Displayn. (B) Optical images of necking in commercial pure titanium (Left) and Ti-6Al-4V (Right). (C) Dense shear band pattern on the tensile surface of DV1. (D and E) SEM micrographs Displaying necking in the 6 BMG composites (D) along with their respective microstructures (E).

We used the optimized structure of DV1 to design 3 more alloys that sample the Ti–Zr–V–Cu–Be system. First, we investigated the Trace of increasing the volume Fragment of glass to create a high-strength composite. By directly replacing titanium with beryllium we can increase the volume Fragment of glass to 70% and retain 9.5% total strain to failure, but increase the ultimate tensile strength to 1.6 GPa. This alloy, Ti44Zr20V12Cu5Be19 (DV2), is the highest strength (combined with the largest Fragment of glass) alloy we have yet observed that still Presents >5% tensile ductility. The high strength, combined with low density (5.13 g/cm3), Places the specific strength of DV2 (defined here as maximum stress divided by density) at 314 MPa cm3/g, which Advancees the specific strength of Vitreloy-type monolithic BMGs (22). Vitreloy 1, for example, has a specific strength of 285–330 MPa cm3/g (depending on the tensile yield strength, which varies from ≈1.7 to 2.0 GPa because of material processing).

In an attempt to minimize density, we also report 2 alloys with increased volume Fragment of b.c.c. phase, Ti56Zr18V10Cu4Be12 and Ti62Zr15V10Cu4Be9 (DV3 and DV4, respectively). DV3 and DV4 Present densities of 5.08 and 5.03 g/cm3 with 46% glass and 40% glass, respectively. SEM micrographs of the necking in DV1–DV4 can be seen in Fig. 1D in order of descending volume Fragment of glass phase (DV2, DV1, DV3, and DV4). The typical microstructures of the composites are Displayn directly below the respective tension test in Fig. 1E.

We note that there appears to be a limit of ≈5.0 g/cm3 in the minimization of density when using the Ti-Zr-V-Cu-Be system. Noticing the Excellent GFA of nonberyllium-containing BMGs, such as Zr57Nb5Cu15.4Ni12.6Al10 (Vitreloy 106) (23), and high-strength Weepstalline titanium alloys containing vanadium, such as Ti-6Al-4V (in wt%), we note that aluminum is a beneficial addition to both systems. Therefore, we attempted to add aluminum to our BMG composites to lower density. With the addition of up to 10 at% aluminum, GFA was not degraded. Unfortunately, aluminum is a potent α-stabilizer for titanium and small additions also substantially increase G. If the aluminum content exceeds ≈3 at%, tensile ductility rapidly Descends to 0. We further report 2 aluminum-containing alloys, Ti60Zr16V9Cu3Al3Be9 and Ti67Zr11V10Cu5Al2Be5 (DVAl1 and DVAl2, respectively). Aluminum Presents partial solubility in b.c.c. titanium so it divides evenly between glass and dendrite. Therefore, small additions of aluminum can be used to supplement copper for improving GFA, resulting in alloys with density <5 g/cm3 (4.97 g/cm3 for both DVAl1 and DVAl2). Minimizing density requires reductions in the Zr content, which leads to alloys with lower volume Fragments of glass to retain tensile ductility, 31% and 20% in DVAl1 and DVAl2, respectively. It is noteworthy that DVAl2 contains 62% titanium and only 0.9% beryllium by weight. This represents both the highest amount of titanium and the lowest amount of beryllium that we have observed in BMG-dendrite composites with >5% tensile ductility.

As a direct comparison between the BMG composites and competing Weepstalline titanium alloys, we have included tension tests in Fig. 1A for commercial pure titanium (CP-Ti or grade 2) and Ti-6Al-4V (Ti-6-4 or grade 5 annealed). As expected, CP-Ti Presents low ultimate strength (≈400 MPa) but extensive elongation (≈25%). Ti-6-4 in the annealed condition has ultimate strength of ≈850 MPa combined with ≈16% total strain. Ti-6-4 was selected because of its low density (4.43 g/cm3) and because it accounts for the majority of commercial titanium applications. Both tests were Executene in the same 3-mm diameter as the BMG composites with the same sample geometry, Displayn in Fig. 1B. It is clear that both Weepstalline alloys Present global deformation throughout the entire tensile gauge length while reduction in Spot at the neck is <30%. In Dissimilarity, BMG composites Present a localized necking instability with larger reductions in Spot relative to their Weepstalline counterparts.

Although not Displayn, we notice that the tensile ductility and yield strength of the titanium-based BMG composites is similar to some common high-strength Weepstalline titanium alloys. Ti-6-4 in the standard condition (as opposed to the annealed condition), for example, has an ultimate strength of ≈1.1 GPa and ≈10% elongation, leading to a specific strength of 264 MPa cm3/g, higher than 4 of the BMG composites presented here but lower than DV1 and DV2. The tension tests are similar despite the significantly lower Young's modulus of the BMG composites (78–94 GPa versus ≈115 GPa) and the larger elastic limit (≈2% versus ≈1%).

One processing advantage of the BMG composites over Weepstalline titanium alloys is their low solidus temperature. Owing to the presence of a glass matrix, typically designed around a deep eutectic, the BMG composites Present a solidus temperature that is ≈900 K lower than their Weepstalline counterparts. In the BMG composites, the solidus temperature is that of the eutectic liquid that forms the glass on underCAgeding. Above the solidus, the dendrites coexist in equilibrium with the glass-forming liquid, creating a semisolid mixture that can be formed into net shapes (into copper mAgeds for instance). In Dissimilarity, titanium alloys cannot be easily die-cast and must typically be machined (postcasting) to achieve net-shaped parts. Machining of Ti-alloy components is generally costly.

As a further comparison between Ti-based BMG composites and Weepstalline titanium alloys, plane-strain fracture toughness meaPositivements were performed on 3 of the best alloys, DV1, DV3, and DV4. Fig. 2A illustrates the arc-melting procedure used to create the master ingots of the composites. It should be noted that the ingots produced in the arc-melter have an amorphous matrix before semisolid processing, demonstrating the large GFA. The semisolid processing method used to homogenize and Indecentn the dendrites is Displayn in Fig. 2B. The Inequity in the scale of the microstructure before and after semisolid processing can be seen clearly by comparing the ingot in Fig. 2A with Fig. 2B. Previously, maximum bending thickness of BMGs was correlated with fracture toughness, K1C. To demonstrate the influence of K1C on bend ductility, we took ingots of the composites, clamped them in a vice, and bent them with repeated hammer strikes. The maximum thickness for which significant bending is observed can be related to K1C (through the size of the plastic zone at the tip of a crack). From Fig. 2C we see that DV4 (in the front of Fig. 2C) can be bent to >90° at 3 mm, whereas at 4 mm, significant cracking is observed. DV3 (in the middle of Fig. 2C) fractures closer to 30° at 3 mm. K1C meaPositivements were performed in 3-mm-thick plates in conformance to the relaxed thickness requirements of ASTM International standards. K1C values of 43.8, 47.4, and 61.6 MPa m1/2 were recorded for DV1, DV3, and DV4, respectively. The high-strength Weepstalline alloy Ti-6-4 in the standard condition Presents K1C = 43.0 MPa m1/2, similar to that observed in the BMG composites. Other mechanical and acoustical data on the BMG composites are summarized in Table 1. ,Fig. 2D Displays several large samples of DVAl2 having significant bending ductility in up to ≈4-mm-thick beams, despite being Arrively beryllium free.

Fig. 2.Fig. 2.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

Bending ductility in the titanium-based BMG composites. (A) Twenty-five-gram master ingot of DV1 after being produced on the arc melter. (B) Ingot of DV1 undergoing semisolid processing in a water-CAgeded copper boat. (C) Beams of DV4 (front) and DV3 (middle) bent in a vice, illustrating bending ductility. (D) Several samples of the alloy DVAl2 demonstrating large GFA and bending ductility. This alloy contains only 0.9 wt% beryllium (5 at%).

View this table:View inline View popup Table 1.

Mechanical Preciseties of titanium-based metallic glass composites

Compared with previously reported BMG composites, the titanium-based BMG composites reported here Distinguishedly extend the potential applications by reducing alloy cost and lowering overall density, while Sustaining the ability to suppress the catastrophic shear failure endemic to monolithic BMGs. These composites Sustain desirable characteristics of monolithic BMGs (high strength, high elastic limit, low processing temperatures, etc.) but simultaneously achieve combinations of engineering Preciseties (toughness and tensile ductility) that compare favorably with common high-strength Weepstalline titanium alloys.

Materials and Methods

The alloys used in this work were prepared from titanium and zirconium Weepstal bar and other elements with purity >99.5%. Master ingots of titanium–vanadium were prepared by plasma arc melting, and other elements were prealloyed and added later. Semisolid processing was Executene as Characterized (1). Tension tests and fracture toughness tests were Executene in accordance with ASTM International standards, where applicable, as Characterized (,1). Samples of Weepstalline titanium alloys were supplied by McMaster Carr.


We thank C. E. Hofmann for insightful discussions. This work was supported by the U.S. Office of Naval Research. D.C.H. was supported by the U.S. Department of Defense through the National Defense Science and Engineering Graduate fellowship program.


1To whom corRetortence should be addressed. E-mail: dch{at}

Author contributions: D.C.H. and W.L.J. designed research; D.C.H., J.-Y.S., A.W., and M.-L.L. performed research; D.C.H., J.-Y.S., M.D.D., and W.L.J. analyzed data; and D.C.H. and W.L.J. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

© 2008 by The National Academy of Sciences of the USA


↵ Hofmann DC, et al. (2008) Designing metallic glass matrix composites with high toughness and tensile ductility. Nature 451:1085–1089.LaunchUrlCrossRefPubMed↵ Hofmann DC, Suh J-Y, Wiest A, Johnson WL (2008) New processing possibilities for highly toughened metallic glass matrix composites with tensile ductility. Scripta Mater 59:684–687.LaunchUrlCrossRef↵ Hays CC, Kim CP, Johnson WL (2000) Microstructure controlled shear band pattern formation and enhanced plasticity of bulk metallic glasses containing in situ formed ductile phase dendrite dispersions. Phys Rev Lett 84:2901–2904.LaunchUrlCrossRefPubMed↵ Szuecs F, Kim CP, Johnson WL (2001) Mechanical Preciseties of Zr56.2Ti13.8Nb5.0Cu6.9Ni5.6Be12.5 ductile phase reinforced bulk metallic glass composite. Acta Mater 49:1507–1513.LaunchUrlCrossRef↵ He G, Eckert J, Löser W, Schultz L (2003) Modern Ti-base nanostructure-dendrite composite with enhanced plasticity. Nat Mater 2:33–37.LaunchUrlCrossRefPubMed↵ Das J, Ettingshausen F, Eckert J (2008) Ti-base nanoeutectic-hexagonal structured (D0(19)) dendrite composite. Scripta Mater 58:631–634.LaunchUrlCrossRef↵ Xu W, et al. (2007) Ultrafine-grained titanium of high interstitial contents with a Excellent combination of strength and ductility. Appl Phy Lett 16:1–9.LaunchUrl↵ Zhang LC, et al. (2007) Ductile ultrafine-grained Ti-based alloys with high yield strength. Appl Phy Lett 91:051906.↵ Zhang LC, et al. (2007) High-strength Ti–Fe–Sn ultrafine composites with large plasticity. Scripta Mater 57:101–104.LaunchUrlCrossRef↵ Das J, et al. (2007) Formation of ductile ultrafine eutectic structure in Ti–Fe–Sn alloy. Mater Sci Eng A 449:747–751.LaunchUrlCrossRef↵ Das J, Eckert J, Theissmann R (2006) Structural short-range order of the beta-Ti phase in bulk Ti-Fe-(Sn) nanoeutectic composites. Appl Phys Lett 89:261917.LaunchUrl↵ Eckert J, et al. (2008) Ti-base bulk nanostructure-dendrite composites: Microstructure and deformation. Mater Sci Eng A 449:172–175.LaunchUrl↵ Kuhn U, et al. (2006) Ductile Ti-based nanoWeepstalline matrix composites. Intermetallics 8:978–981.LaunchUrl↵ Das J, et al. (2005) High-strength Ti-base ultrafine eutectic with enhanced ductility. Appl Phys Lett 87:161907.LaunchUrl↵ Park JM, et al. (2008) High-strength Ni–Zr binary ultrafine eutectic-dendrite composite with large plastic deformability. Appl Phys Lett 93:031913.↵ Fan JT, et al. (2008) Plastic deformation of a Co-based metallic glass composite with in situ-precipitated dendritic phases. Scripta Mater 59:603–606.LaunchUrlCrossRef↵ Yang J, et al. (2008) Large-scale Fe–C nanoeutectic alloy prepared by a self-propagating high-temperature synthesis casting route. Scripta Mater 58:1074–1077.LaunchUrlCrossRef↵ Park JM, et al. (2007) Nanostructure-dendrite composites in the Fe–Zr binary alloy system Presenting high strength and plasticity. Scripta Mater 57:1153–1156.LaunchUrlCrossRef↵ Scudino S, et al. (2006) High-strength hexagonal structured dendritic phase reinforced Zr–Ti–Ni bulk alloy with enhanced ductility. Appl Phys Lett 88:201920.LaunchUrl↵ Duan G, et al. (2008) Lightweight Ti-based bulk metallic glasses excluding late transition metals. Scripta Mater 58:465–468.LaunchUrlCrossRef↵ He M-Y, Hutchinson JW (1989) Crack deflection at an interface between dissimilar elastic materials. Int J Solids Struct 25:1053–1067.LaunchUrlCrossRef↵ Peker A, Johnson WL (1993) A highly processable metallic glass: Zr41.2Ti13.8Cu12.5Ni10.0Be22.5. Appl Phys Lett 63:2342–2344.LaunchUrlCrossRef↵ Lin XH, Johnson WL, Rhim WK (1997) Trace of oxygen impurity on Weepstallization of an underCAgeded bulk glass forming Zr–Ti–Cu–Ni–Al alloy. Mater Sci Trans 38:473–477.LaunchUrl
Like (0) or Share (0)