Effect of Mo, V and Zr on the microstructure and mechanical proper es of Ti2AlNb alloys

Small bu�on ingots of Ti2AlNb alloys with different contents of Mo, V and Zr were melted by vacuum nonconsumable arc furnace. Due to the rapid cooling rate during mel ng process, only β grains without precipita on were observed in most of the bu�on ingots and no regular phenomenon was found. However, when the samples were heated to β phase region and then furnace cooled to room temperate, different morphologies and quan es of primary α phase and second O phase formed from the β grains of different samples. It is suggested that the morphology of α phase was changed from lamellar to quadrilateral with increasing V and the lath O increased with increasing Zr. Besides, the residual β/B2 phase increased with increasing Mo and V. The EDS results showed that Al and Zr were enriched in α phase whereas Nb, Mo and V were enriched in β/B2 phase. The micro-hardness of these samples before and a�er heat treatment was detected and the micro-hardness increased with increasing Zr and decreasing Mo and V.


Introduc on
Ti 2 AlNb-based alloys have been extensively studied due to their excellent specific strength and good oxida on resistance at 650~750 °C [1][2][3]. These advantages made them prospec ve material to replace high-temperature tanium alloy and superalloy in the field of engineering applica on. However, the weakness in fracture toughness and creep resistance significantly limited the development and applica on of Ti 2 AlNb-based alloys at present. Although great effort has been performed by op mizing the hot working and heat treatment process, only limited improvement was achieved over the past years. Therefore, alloying is considered another poten al method to promote the combina on property of Ti 2 AlNb alloys.
In the study performed by Tang et. al [4,5], Mo and W were used to replace the Nb in Ti-22Al-27Nb alloy. The results showed that V was effec ve in improving room temperature duc lity, and W was effec ve in enhancing high-temperature tensile strength and creep resistance. Emura et. al [6] studied the effect of Mo and Fe on the room-temperature duc lity of Ti 2 AlNb alloy, and a higher tensile elonga on at room temperature was obtained. The effect of Zr on the microstructure and proper es was inves gated by Germann [7], and it is suggested that the oxida on resistance and creep property were obviously increased when the content of Zr was up to 2 at.%. Besides, the effect of Ta [8], Si [7], B [9] on the phase transforma on and mechanical proper es was also studied. Notably, most of these works have been done by adding one or two elements into the Ti 2 AlNb based alloys, few studies focused on the simultaneous and commuta ve effect of three or more elements on the phase transforma on, microstructure and the relevant proper es.
In fact, according to the development and applica on of superalloy, the genera on and the opera ng temperature is typically classified by the kind and content of the elements in the material. Similarly, different contents of elements, such as Mo, Nb, Zr, Si and so on, were usually added into tanium alloys simultaneously to improve the long me performance at elevated temperature. For this case, it is believed that adding mul ple alloying elements is a prospec ve method to improve the proper es of Ti 2 AlNb alloy. In this paper, 9 kinds of Ti 2 AlNb alloys with different content of Mo, V and Zr are melted. The effect of these elements on the microstructure and micro-hardness are detected and discussed. Besides, the microsegrega on in different phases was also analyzed.

Experiment
According to the Mo equilibrium and orthogonal method, 9 kinds of ingots with different composi ons were designed based on Ti-22Al-25Nb alloy (at.%). The detailed nominal composi ons of the ingots are listed in Table 1. Small bu�on ingots with approximate160g were melted by vacuum non-consumable arc furnace. To assure the homogeneity, each ingot was melted 5 mes at least. The purity of raw materials used in this study were pure Ti, pure Al, pure Zr, AlMo, AlV and NbTi master alloy, respec vely. Leica DM4000M op cal microscope and JSM-6460 were used to observe the microstructures of these samples. The chemical composi on was analyzed by an energy dispersive X-ray spectroscopy (EDS) detector. Micro-hardness was detected by a Vickers hardness tester. For the measuring accuracy, each sample was detected 5 mes at least.

Microstructures of different ingots
The typical bu�on ingots prepared by vacuum non-consumable arc furnace are shown in Fig. 1. The ingot has no obvious oxida on and the surface of the ingot is glossy. Columnar grains, which grown direc onally from the edge to the inner of the ingot, can be obviously observed. Fig. 2 shows the microstructures of the as-cast ingots. The grain boundary is clearly observed in these ingots. With the increase of Mo and V, the grain boundary became wider and the precipita ons within the grains increased, as shown in 1#, 2# and 3# samples. When the content of Zr is increased up to 2%, it is found that rod-like phase formed within the grains, as shown in 8# and 9# samples. However, for the 4# and 6# samples, although the Mo and V are up to 2%, few precipita ons can be observed within most of the grains. Besides, no more regularity was found among these 9 samples before heat treatment and it is confusing that the effect of composi on on the microstructure.
According to phase diagram, although the phase transforma on is complicated in the range of 960~1060 °C, the solidifica on path is simple phase transforma on from liquid phase to β phase. Thus, only β grains were formed during the solidifica on. A�er subsequent cooling procedure, although complicated phase transforma on path is inclined undergone, most of the precipita on was restricted due to the high cooling rate and then only a small quan ty of precipita ons could be observed within those β grains. Besides, because the cooling rate is uncontrollable during the mel ng process, the cooling rate is inconsistent and thus no obvious regularity with regard to the volume ra o and morphology could be found in these ingots.

Microstructures a�er heat treatment
In order to inves gate the effect of elements on the microstructure and microsegrega on of Ti 2 AlNb alloy, all the samples were heated to single β phase region, held on 1 house, and then cooled to room temperature in furnace. Due to the cooling rate is sufficient low, the α 2 phase and O phase have enough me to form from the β/B2 grains. Fig. 3 shows the microstructures a�er heat treatment. Different from those microstructures shown in Fig. 2, various morphologies and volumes of equiaxed α 2 /O phase, coarse α 2 /O lath, fine O lamellae and residual β/B2 phase were observed a�er heat treatment. Within these pictures, the black phase is α 2 phase, the white phase is β/B2 phase and the gray phase is O phase.
When the content of Mo is increased from 0% (1#, 5# and 9#) to 2% (3#, 4# and 8#), the volume of β/B2 phase obviously increases and the fine O lamellae par cipated from the β/B2 decrease. Besides the primary coarse α 2 lath, a great deal of α 2 phase with tetragon morphology is also observed, which suggests that the orienta on rela onship between α 2 and β/B2 phase may change with vary of Mo. When the content of V is increased from 0% (1#, 4# and 7#) to 2% (3#, 6# and 9#), the volume of β/B2 phase also increases and the morphology of α 2 phase transforms from coarse lath to tetragon largely. Moreover, it is found that more β/B2 phase is reserved with increasing Mo whereas the amount of tetragon α 2 increases with increasing V. Different from the effects of Mo and V, the volume of β/B2 and α 2 phase decreases obviously when the content of Zr is increased from 0% (1#, 2#, 3#) to 2% (7#, 8#, 9#). Besides, it is also found that the O phase between the β/B2 and α 2 phase increases correspondingly, which suggested that the increasing of Zr can improve the phase transforma on of β/B2 +α 2 → O. Fig. 3(a-c) show the microstructures that Mo and V increase from 0 to 2% simultaneously, the volume of β/B2 phase is increased obviously. Besides, the length of coarse α 2 /O lath and the volume of O phase decrease obviously. When Mo and V is increased up to 2%, most of O phase disappears and the microstructure only consists of β/B2 and α 2 phases, as shown in Fig. 3(c). Besides, numerous primary tetragon α 2 phase is formed from the original β grains. Fig. 3(a, e, i) show the microstructures that V and Zr increase from 0 to 2% simultaneously, the volume of β/B2 and O phase increases whereas the α 2 phase decreases obviously. Simultaneously, it is also found that volume and length of coarse lath also increase. By contrast, with the decrease of Mo and increase of Zr, the β/B2 decreases and the volume and length of lath O phase increase, as shown in Fig. 3 c, f and i.
As discussed above, it is believed that Mo and V are beneficial to the reserve of β/B2 phase and the precipita on of fine O phase is restricted largely correspondingly. Besides, the orienta on rela onship between the β/B2 and α 2 phase is also changed with the increase of Mo and V. By contrast, with the increase of Zr, both the primary and second α 2 lath becomes more coarse and longer, and the volume of O phase around the α 2 lath increases largely. For this case, only li�le β/B2 phase can be reserved a�er the heat treatment.

Composi on of different phases
The composi on of different phases within the samples is analyzed by EDS. Fig. 4 shows the results of lineal analyze within sample 3 and 5. It is found that the Ti and Al are enriched in α 2 phase and poor in β/B2 phase. By contrast, Nb, Mo and V are enriched in β/B2 phase and poor in α 2 phase. The rim O phase is formed between β/B2 and α 2 phase and then has the transi on content between β/B2 and α 2 phase. As is well-known, increasing Al will improve the forma on of α 2 phase, whereas Nb, Mo, V is beneficial to increase the ra o of residual β/B2 phase.
The content in different phases of the typical samples is listed in Table 2. The microsegrega on in different phases is different with the change of alloy composi on. While in 1 # sample, the content of Nb in β/B2 phase is 46% higher than α 2 phase. While in 3# sample, the content of Nb in β/B2 phase is only 28% higher than α phase. Similarly, the content of Mo and V of β/B2 phase is also higher than that of α 2 phase, and it is found that the microsegrega on between β/B2 and α 2 phase increases with the increase of residual β/B2 phase, such as 4#, 6# and 8# sample. Note that, when the nominal composi on of Zr is lower than 1%, it is difficult to detect the content of Zr in different phases due to the detec on precision of EDS equipment. Nevertheless, the microsegrega on of Zr in different phases can s ll be analyzed qualita ve. While in 4# and 6# samples, the α 2 phase has more Zr than β/B2 phase. By contrast, while in 8# and 9# samples, the β/B2 phase has more Zr than α 2 phase. What's more, the Zr in O phase is higher than both of α 2 and β/B2 phase, which is somewhat different the other elements.

Micro-hardness
The images in Fig. 6 depict the micro-hardness of different samples before and a�er heat treatment. Obvious regularity can be observed although li�le par cipa on is found within most of the samples. With the increase of Zr from 0% to 2%, the mean micro-hardness of the samples increases from 335 to 355. Similarly, the mean micro-hardness slightly increases from 346 to 348 with the increase of Mo. This suggests that the solid solu on strengthening of Zr within β/B2 phase is higher than Mo. Different from Mo and Zr, however, the mean micro-hardness decreases obviously with the increase of V. For this case, it is believed that the micro-hardness is depended on the solid solu on strengthening. Thus, the micro-hardness of Ti 2 AlNb alloy increases with the increase of Mo and Zr, whereas decreases with the increase of V.
A�er heat treatment, the micro-hardness of all these samples decreases obviously. As shown in Fig. 2 and Fig. 3, various morphologies and volumes of equiaxed α 2 /O, lath α 2 /O and lamellar O phase were formed a�er the heat treatment. For this case, the effect of solid solu on strength on the micro-hardness decreases whereas the effect of precipita on strength becomes more significant. Therefore, it is suggested that the solid solu on strength has more effect on the micro-hardness of Ti 2 AlNb alloy than the solid solu on strength.
Nevertheless, it is s ll found that the effect of element content on the micro-hardness is same to that before heat treatment. With the increase of Mo and Zr and decrease of V, the micro-hardness increase correspondingly, and vice versa. Somewhat differently, the trend of decrease or increase becomes more pronounced a�er heat treatment, which suggests that the precipita on of different phases enhances the effect of alloy on the micro-hardness of Ti 2 AlNb alloy.

Conclusions
1) The microstructure and composi on of different phases of Ti 2 AlNb samples a�er heat treatment are different and are largely dependent on the alloy composi ons. The morphology of α 2 phase changed from lamellar to quadrilateral with increasing V and the lath O increased with increasing Zr. Besides, the residual β/B2 phase increased with increasing Mo and V. Besides, it is suggested that that Al and Zr were enriched in α phase whereas Nb, Mo and V were enriched in β/B2 phase.
2) The micro-hardness of these samples before and a�er heat treatment increased with increasing Zr and decreasing Mo and V. And it is suggested that the micro-hardness a�er heat treatment is lower than that before heat treatment.  Figures   Fig. 1 Button ingot obtained by non-consumable arc furnace.