- Email: [email protected]
Structural phase transitions in the Ti45Zr38Ni17−xFex nano-alloys and their deuterides
Journal of Alloys and Compounds 646 (2015) 90e95
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Structural phase transitions in the Ti45Zr38Ni17xFex nano-alloys and their deuterides D. Rusinek a, J. Czub a, J. Niewolski a, Ł. Gondek a, *, M. Gajewska b, A. Takasaki c, A. Hoser d, b _ A. Zywczak w, Poland AGH University of Science and Technology, Faculty of Physics and Applied Computer Science, Mickiewicza 30, 30-059 Krako w, Poland AGH University of Science and Technology, Academic Centre for Materials and Nanotechnology, Mickiewicza 30, 30-059 Krako c Department of Engineering Science and Mechanics, Shibaura Institute of Technology, Toyosu, Kotoku, Tokyo 135-8548, Japan d Helmholtz Zentrum Berlin, Hahn-Meitner-Platz 1, 14109 Berlin, Germany a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 February 2015 Received in revised form 1 June 2015 Accepted 2 June 2015 Available online 10 June 2015
The mechanically alloyed TieZreNi materials are extensively studied due to their promising properties concerning biomedical, electronic or hydrogen related applications (for example the gaseous hydrogen storage and the MNiH batteries). In this paper we address the very crucial issue of the structural properties and transformations of the amorphous and quasicrystalline Ti45Zr38Ni17xFex (x ¼ 0, 4, 8) and their hydrides. According to the neutron diffraction results, the transformation of the amorphous Ti45Zr38Ni17 phase into the icosahedral quasicrystalline state (the i-phase) is quasi-continuous and starts at the relatively low temperature of 300 C. At 500 C the i-phase is well-developed. At higher temperatures the i-phase transforms into the approximant w-phase and eventually into the cubic phase (the c-phase). Interestingly, the deuterided i-phase exhibits completely different thermal evolution. Namely, this phase decomposes into the simple intermetallic compounds above 625 C. What is worthmentioning is that the release of deuterium is strictly related to that structural decomposition. The possibility of hydrogenation of the amorphous Ti45Zr38Ni17xFex phases with maintaining the amorphous nature of the alloys is the other extremely important field of our interest. We established a processing route to meet our goal. Finally, we show that introducing deuterium triggers an exciting phase transition from the deuterided amorphous phase into the unknown before, partially disordered, quasicrystallinelike phase (the glassy quasicrystal) without releasing of deuterium. © 2015 Elsevier B.V. All rights reserved.
Keywords: Hydrides Quasicrystals Amorphous alloys Neutron diffraction X-ray diffraction
1. Introduction The TieZreNi alloys have several interesting properties that make them very promising in certain fields of applications. Namely, they might be used for high temperature protective coatings [1], biomedical applications [2], metal-hydride based batteries [3], and memory shape alloys [4]. It seems that the hydrogen sorption properties, e.g. the concentration of hydrogen exceeding significantly 2 wt.%, indicate their potential in the hydrogen related applications [5e8]. On the other hand, the basic properties of such alloys and their hydrides are still not fully understood. The TieZreNi compounds can be synthesized in the crystalline, amorphous and quasicrystalline phases. The quasicrystalline phase
* Corresponding author. E-mail address: [email protected] (Ł. Gondek). http://dx.doi.org/10.1016/j.jallcom.2015.06.023 0925-8388/© 2015 Elsevier B.V. All rights reserved.
is particularly interesting for the hydrogen storage applications. The quasicrystalline TieZreNi alloys (the i-phase) may be obtained by the melt spinning technique, but the material may react with a crucible [7]. The mechanical alloying (MA) is the more advantageous method of producing the quasicrystalline TieZreNi compounds. However, a subsequent thermal treatment is necessary [5,9,10]. In this process, the amorphous alloy is obtained by the MA synthesis, while the further annealing transforms the alloy into the stable quasicrystalline phase. This is in agreement with the ternary phase diagrams that confirm that the TieZreNi quasicrystals can be thermodynamically stable [11,12]. The icosahedral structure of the TieZreNi compounds was evidenced by X-ray and neutron diffraction, while the local atomic structure of the quasicrystals and approximants were derived from the EXAFS measurements [13e17]. For the Ti40Zr40Ni20 the transformation from the amorphous into the quasicrystalline phases occurs mainly between 500 and 550 C [16].
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95
While hydrogenation of the quasicrystalline TieZreNi alloys could be easily performed, hydrogenation of the amorphous Ti45Zr38Ni17xFex conducted at elevated temperatures led to decomposition of the specimen [18]. Particularly, formation of the simple hydride phases such as the TiH2, ZrH2 and (ZrTi)H2 was observed. In the case of the quasicrystal 45:38:17-related phases, the hydrides of the quasicrystal and the simple hydride phases were found after hydrogenation [5,6,19]. The aim of the present work was to track the Ti45Zr38Ni17 transformation from the amorphous to the quasicrystalline state as the nature of such a process is not entirely understood. For the first time, the in-situ neutron diffraction technique was used for this purpose and unveiled the quasi-continuous character of this transformation. On the other hand, the issue of the hydrogen influence on the structural properties of the investigated alloys was addressed. Namely, we succeeded in obtaining the amorphous Ti45Zr38Ni17xFex (x ¼ 4, 8) alloys with the full capacity of hydrogen and then we evidenced an extraordinary phase transition during increasing of temperature. 2. Experimental The amorphous Ti45Zr38Ni17 material and the Fe-doped Ti45Zr38Ni17, that were the base for the further processing, were synthesized by MA. Then, the product was tested by means of X-ray powder diffraction (XRD) in order to check its quality. The resulting diffraction pattern revealed no reflections related to the crystalline phases confirming the fully amorphous state of the powder. The quasicrystalline phase was obtained by thermal treatment of the amorphous powder. Afterwards, the quasicrystalline powder was deuterided. All the above syntheses were performed following the procedures reported in details in Ref. [17]. The XRD measurements were done by means of the Panalytical Empyrean powder diffractometer equipped with the Anton Paar HTK1200 high-temperature chamber. The deuteriding of the alloys was processed using the Setaram PCT-PRO Sievert's apparatus. Introducing deuterium instead of hydrogen created the possibility to perform the neutron diffraction (ND) experiments. Deuterium is a good neutron scatterer, while hydrogen exhibits a very large incoherent scattering cross section which makes it not suitable for ND. The procedure for the amorphous specimens was as follows. Firstly, the samples were placed in a controlled-atmosphere glove-box, grinded in a mortar and sieved (20 mm mesh) in order to avoid agglomeration of the powder grains. Then each sample was weighted and placed in a reaction chamber. Then the closed reaction chamber was transferred outside the glove-box and connected to the Sievert's system. The reaction chamber was then evacuated and flushed several times by helium (purity 6N). Next, a leak check was done and the volume of the filled chamber was measured (at 30 C). The sample was evacuated again after that procedure and the temperature ramp (1 deg/min.) up to 180 C was applied. Evacuation at that temperature lasted for at least 18 h in order to desorb the spurious gases. The stability of the amorphous specimen was checked prior to the thermal treatment by the in-situ XRD measurements. Then the sample was cooled down to the temperature of 30 C that was maintained. Next, deuterium gas (99.8% D þ 0.2% H, of 6N purity) was applied to the calibrated volume and expanded to the reaction chamber under the pressure of 40 bar. The uptake reaction was observed just a few minutes after applying the pressure so there was no need to increase the temperature to trigger the reaction. The uptake lasted up to 100 min resulting in 2.35 wt.% of the deuterium concentration. Eventually, the pressure was lowered to 1 bar (absolute) and the sample was checked for possible desorption by next 64 h. As a matter of fact, the observed desorption was negligibly small and the
91
total concentration of 2.34 wt.% was achieved (x ¼ 0). Introducing the Fe-doping lowered the overall deuterium uptake: 2.31 wt.% (x ¼ 4) and 2.26 wt.% (x ¼ 8), however the reaction proceeded faster. Then the samples were checked by XRD in order to get a preliminary confirmation of their amorphous nature. The XRD measurements were repeated and the specimens were prepared shortly before the ND experiments. The ND measurements were performed by means of the E6 focusing powder diffractometer (BENSC-Berlin) and an incident neutron wavelength of 2.45 Å. The in-situ measurements were performed up to 800 C using the high-temperature HTF1 furnace (AS Scientific Products). Each powdered sample was enclosed in a quartz tube under a flow of the high-purity argon gas (6N). The background originating from the furnace and the quartz tube was measured prior to the experiment in the full temperature range. Therefore, it was possible to extract the background from the collected data. It is worth mentioning that ND has a huge advantage over XRD as the structural features related to deuterium can be easily evidenced. The transmission electron microscopy (TEM) investigations were performed using the FEI Tecnai TF20 X-TWIN (FEG) microscope at an accelerating voltage of 200 kV. The bright field (BF) imaging and the selected area electron diffraction (SAED) technique were applied.
3. Results and discussion 3.1. The Ti45Zr38Ni17 amorphous phase transformations As it is apparent from the results of the DSC measurements, only one well-defined anomaly at about 530e560 C can be noticed (see Fig. 1). However, a broad feature existing only for the first heating cycle is visible at the left side of that maximum. This feature ranging from 300 up to 500 C has been commonly attributed to releasing of the energy gained during MA. However, our neutron diffraction studies unveiled the true origin of such behavior. The insitu neutron diffraction measurements in the 250e600 C temperature range are presented in Fig. 2. The data below 250 C are not shown as there were no significant changes noticed and therefore the fully amorphous state of the sample between 30 and 250 C was confirmed. The temperature of the sample was stabilized for 15 min before collection of each diffraction pattern. After
Fig. 1. The DSC curve collected for the Ti45Zr38Ni17 amorphous phase (the 1st cycle) and the repeated measurement (the 2nd cycle) for the same specimen.
92
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95
Fig. 2. The neutron diffraction patterns of the initially amorphous Ti45Zr38Ni17 at elevated temperatures.
that, the sample was maintained at particular temperature for further 60 min. Afterwards, an additional scan was performed to check the time evolution of the diffraction pattern. Then, the temperature was changed and the above sequence was performed again. We noticed just a slight time-dependence of the diffraction patterns at the investigated temperatures. According to Fig. 2, the neutron diffraction patterns exhibit apparent evolution starting from 300 C. Namely, several well defined reflections appeared at about 38.1, 62.8, 94.9 and 115.1 of the 2Q angle. Those reflections can be attributed to the i-phase. At about 500 C, the reflections of the i-phase were well developed. Hence, the quasi-continuous character of the transition from the amorphous into quasicrystalline phase was evidenced. Above 525 C, several small additional reflections appeared. Consequently, at temperature of 600 C the apparent lines at 56.1, 59.1, 92.0 and 102.0 of the 2Q angle were observed. Moreover, the main reflections became significantly narrowed. All of those indicated the crystalline character of the structure stabilized above 550 C. Indeed, this structure can be indexed as the approximant of the i-phase (the w-phase). The transition at about 550 C was reversible, therefore the w-phase transformed back into the i-phase during cooling. This is the reason for the existence of the DSC maxima in the first and the next cycles. The diffraction pattern collected at 550 C can be indexed as the cubic I m-3 space group with the lattice constant a ¼ 14.344 (1) Å as was expected for the wphase. Tracking the thermal evolution of the w-phase of the Ti45Zr38Ni17 at high temperatures was the next step. Therefore, the in-situ heating was performed again, but at that time temperature of 700 C was reached (see Fig. 3). Apart from the expected transition from the i-phase into the w-phase, an additional transition at about 675 C from the w-phase into the new cubic phase was noticed. The new phase seemed to be not very different from the wphase as a part of reflections were in similar positions. However, several reflections disappeared (see eg. at 50 and 56.1 of the 2Q) while the additional appeared (at 45 of the 2Q). Nevertheless, the structure can be indexed in the cubic system (the c-phase) similarly to the w-phase. The new phase was stable and remained unchanged during cooling down to room temperature. To confirm the sequence of the phase transitions, the in-situ high-temperature XRD studies were conducted. The obtained results entirely confirmed that the w-phase transforms to the c-phase above 675 C. Only the changes related to the lattice contraction were observed
Fig. 3. The thermal evolution of the w-phase of the Ti45Zr38Ni17 as seen by ND.
during cooling down to room temperature. Therefore, it must be concluded that the newly discovered c-phase was stable. To sum up this paragraph, we evidenced the quasi-continuous transition from the amorphous into the quasicrystalline i-phase, what is in perfect agreement with the DCS data. At temperatures of 550 C the approximant w-phase appeared, what accords with the DSC data as well. Further heating of the Ti45Zr38Ni17 specimen above 675 C led to the occurrence of another transition into an unknown cubic phase. The latter transition was not reversible, as the phase was maintained during cooling down to room temperature. 3.2. Deuterium desorption from the quasicrystalline Ti45Zr38Ni17 iphase The icosahedral i-phase can be deuterided up to 2.3 wt% [17]. After deuterization its quasicrystalline structure is maintained. The thermal desorption studies (TDS) presented in Ref. [17] show that the maximum of the hydrogen release occurred at the temperature of about 430 C. However, the sample was continuously evacuated for the TDS measurements, while in our studies, the sample was kept under 1 bar of argon (absolute pressure). That pressure was maintained during the whole experimental cycle, therefore it should be expected that desorption of deuterium occurred at higher temperatures. In Fig. 4 the neutron diffraction results for the deuterided iphase are presented. In the RT-350 C temperature range the diffraction patterns did not exhibit significant changes. As can be seen for the deuteride, an additional strong reflection that was not found for the pure i-phase was visible at about 97 of the 2Q. Appearing of such an additional reflection/reflections is typical for deuterides as deuterium atoms, in contradiction to hydrogen, are good neutron scatterers. This reflection can be attributed to the metal-deuterium mean distance and in our case the related dspacing is about 1.7 Å. Therefore, t existence of this reflection is a good probe for the deuterium concentration in the sample. As can be inferred from the Fig. 4, the significant changes were triggered at the temperature of 475 C at which splitting of the above mentioned reflection occurred. However, lowering of its intensity was noticed. That tendency was more pronounced when temperature was increased, while above 525 C the reflections at 46 and 53 of the 2Q increased their intensities what suggests that some kind of rearrangements of the i-phase occurred. Eventually, at about 625 C the deuterium related reflection disappeared. In fact, at that
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95
93
should be underlined that the release of deuterium was strictly connected with the above discussed structural decomposition of the i-phase. 3.3. Deuterization of the Ti45Zr38Ni17xFex (x ¼ 4,8) amorphous phases
Fig. 4. The in-situ ND patterns of the deuterided i-phase of the Ti45Zr38Ni17.
temperature formation of the new phase that was apparently different from the i- and w-phases was observed. Moreover, that phase was different from the c-phase evidenced for non-deuterided i-phase. At 725 C additional rearrangements of the reflections occurred suggesting further decomposition of the i-phase. No traces of the i-phase were noticed after cooling down. The sample consisted of a mixture of some crystalline phases and only the Ti2Ni phase was unambiguously recognized among them. That is in line with the findings reported in Ref. [20], where an occurrence of the Ti2Ni accompanying hydrogen desorption from the i-phase was shown. The products formed during the hydrogen release from the Ti45Zr38Ni17 at elevated temperatures were extremely dependent on the thermodynamic conditions (pressure, a heating ramp) and the other factors, e.g. the morphology of the sample and the hardly avoidable oxygen contamination of the surface during long time experiments at elevated temperatures [20]. We suggest a possibility of decomposition of the deuterided iphase into the simple intermetallic compounds just above 625 C. This is an important difference as compared to the i-phase for which the transition into the approximant phase was evidenced. It
As it is mentioned in the introduction and the experimental part, we found the route of maintaining the amorphous nature of the specimens after introducing deuterium. Such a possibility is of the utmost interest as it addresses the crucial issue of the behavior of hydrogen/deuterium in the amorphous alloys. It is commonly believed that hydrogen in an amorphous alloy is bounded to its particular constituents leading to the formation of the nanocrystalline simple hydrides. The XRD measurements suggested the diffraction patterns for both the base and deuterided specimens are typical for the amorphous materials. Moreover, the broad main maxima of the deuterided specimens were shifted towards the lower angles with respect to the base alloys (for x ¼ 4 see Fig. 5), what was an evident sign of enhancing of the mean metalemetal distances. SAED confirmed the amorphous nature of the deuterided sample (an inset to Fig. 5). All the results indicate that deuterium is not bounded to the particular constituent. The morphology of the raw and deuterided powders was similar as can be inferred from the bright field TEM experiment (Fig. 5). Presence of the submicrometer agglomerates of primary 300 nm particles was observed in both types of specimens. Thus, the deuterization process did not lead to further decomposition of the material. As deuterium cannot be tracked by the X-rays or electron diffraction, the neutron diffraction experiment was performed in order to obtain the final confirmation. Indeed, the collected diffraction patterns exhibited clearly that apart from the broad maxima related to the distribution of the metalemetal distances there was another broad maxima that can be attributed to the distribution of the metal-deuterium distances (the Fig. 5). The width of the latter is a direct confirmation that deuterium is not bounded to any specific element in a well-defined manner but it is just an equivalent constituent of the alloy. To strengthen this conclusion, the thermal evolution of the amorphous deuterided alloys was observed in-situ by means of ND as depicted in Fig. 6. First of all, at the temperature of about 250 C
Fig. 5. The diffraction patterns and TEM images of the amorphous Ti45Zr38Ni13Fe4 before (black) and after (red) deuterization. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
94
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95
Fig. 6. Thermal evolution of the deuterided amorphous Ti45Zr38Ni13xFex (x ¼ 4 and x ¼ 8) specimens.
two sharp, well-defined reflections appeared for the deuterided amorphous Ti45Zr38Ni17xFex (x ¼ 4, 8) alloys. The extraordinary nature of such diffraction patterns must be underlined as there were no other reflections in the entire investigated 2Q range. A careful inspection of the patterns revealed that those strong reflections appeared at the same angles as two main reflections visible in the case of the deuterided Ti45Zr38Ni17 quasicrystal (please compare Fig. 4). Although the smaller reflections were missing, it seems that the newly evidenced structure was somewhat similar to the deuterided quasicrystalline phase that might be considered as the partially disordered quasicrystalline i-phase (the PDi-phase). The further thermal evolution of that phase was different from that for the deuterided Ti45Zr38Ni17 quasicrystal (there is no intermediate phase in the 625e700 C temperature range), but the final product was similar to the deuterided quasicrystal related mainly to the Ti2Ni, TiFe and Zr2Ni phases. To conclude our observations, the presence of deuterium in the amorphous alloys Ti45Zr38Ni17xFex (x ¼ 4, 8) triggered the phase transition from the amorphous into the PDi-phase in the relatively narrow temperature range (250e275 C). The latter is in a contradiction to the transition from the amorphous to the i-phase for the pure Ti45Zr38Ni17 that occurred in the much broader temperature range (see Fig. 2). However, it supports the conclusion that the PDiphase was formed. In the higher temperatures the PDi-phase decomposed directly into the simple intermetallic compounds with the release of deuterium. 4. Concluding remarks The mechanically-alloyed, initially amorphous, Ti45Zr38Ni17 material undergoes the structural transformation into the quasicrystalline state (the i-phase) in a quasi-continuous manner. The transformation occurs in a broad temperature range between 300 and 500 C, while at 550 C the approximant w-phase stabilizes. Our studies show that at higher temperatures the approximant wphase disappears and the cubic phase (the c-phase) is formed. The latter is stable during cooling down to the room temperature. On the other hand, the deuterided i-phase exhibits completely different thermal evolution. It decomposes into the simple intermetallic compounds above 625 C. What is worth-mentioning is
that the release of deuterium is mostly related to that structural decomposition. Another interesting issue is that decomposition of the icosahedral phase is not preceded by establishing of the approximant w-phase. These results show that the deuterided icosahedral phase is more stable against the transformation into its approximant, however its decomposition is irreversible. The extremely interesting results were obtained by investigating the deuterided, but still amorphous Ti45Zr38Ni17xFex (x ¼ 4, 8) alloys. The inspection of the X-ray, electron and neutron diffraction patterns indicates that deuterium is an equivalent constituent of the alloy and influences significantly the thermal evolution. Namely, it triggers the phase transition from the amorphous to the partially disordered i-phase phase that have not been evidenced yet (the glassy quasicrystal). It seems that occurrence of such a glassy quasicrystal phase might be an intrinsic property of the thermal evolution of the other amorphous, hydrided materials. The possibility of obtaining the amorphous, but hydrided alloys with the hydrogen capacity exceeding 2.4 wt.% is extremely interesting in the context of the hydrogen storage and will be an object of further studies. Acknowledgments This research project was supported by the Polish-Norwegian Research Programme through the project 'Nanomaterials for hydrogen storage’ number 210733, the European Commission under the 7th Framework Programme through the 'Research Infrastructure' action of the 'Capacities' Programme, NMI3-II Grant number 283883 and by the Polish Ministry of Science and Higher Education (MNiSW). References [1] Y. Guan, J.-G. Liu, C.-W. Yan, Int. J. Electrochem. Sci. 6 (2011) 4853. [2] H. Lefaix, F. Prima, S. Zanna, P. Vermaut, P. Dubot, P. Marcus, D. Janickovic, P. Svec, Mater. Trans. 48 (2007) 278. camps, E. Leroy, A. Percheron-Gue gan, Electro[3] B. Guiose, F. Cuevas, B. De chimica Acta 54 (2009) 278. [4] S. Inoue, N. Sawada, T. Namazu, Vacuum 83 (2009) 664. [5] YuV. Zhernovenkova, L.A. Andreev, S.D. Kaloshkin, T.A. Sviridova, V.V. Tcherdyntsev, I.A. Tomilin, J. Alloys Compd. 434e435 (2007) 747. [6] A. Takasaki, K.F. Kelton, J. Alloys Compd. 347 (2002) 295.
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95 [7] R.R. Shahi, T.P. Yadav, M.A. Shaz, O.N. Srivastava, S. van Smaalen, Int. J. Hyd. Energy 36 (2011) 592. [8] A. Kocjan, P.J. McGuiness, S. Kobe, Int. J. Hyd. Energy 35 (2010) 259. [9] YuV. Zhernovenkova, T.A. Sviridova, V.V. Tcherdyntsev, S.D. Kaloshkin, A.L. Pomadchik, J. Non-Crystalline Solids 353 (2007) 3429. [10] A. Matsumoto, K. Kobayashi, T. Nishio, K. Ozaki, J. Alloys Compd. 434e435 (2007) 315. [11] J.P. Davis, E.H. Majzoub, J.M. Simmons, K.F. Kelton, Mater. Sci. Eng. 294e296 (2000) 104. [12] R.G. Hennig, A.E. Carlsson, K.F. Kelton, C.L. Henley, Phys. Rev. B 71 (2005) 144103. [13] R.G. Hennig, K.F. Kelton, A.E. Carlsson, C.L. Henley, Phys. Rev. B 67 (2003) 134202.
95
[14] A. Sadoc, E.H. Majzoubc, V.T. Huettd, K.F. Kelton, J. Alloys Compd. 356e357 (2003) 96. [15] A. Sadoc, J.Y. Kimc, K.F. Kelton, Mater. Sci. Eng. 294e296 (2000) 348. [16] A. Kocjan, P.J. McGuiness, M. Raji c Linari c, S. Kobe, J. Alloys Compd. 457 (2008) 144. [17] A. Takasaki, C.H. Han, Y. Furuya, K.F. Kelton, Phil. Mag. Lett. 82 (2002) 353. _ [18] A. Zywczak, D. Shinya, Ł. Gondek, A. Takasaki, H. Figiel, Sol. State Commun. 150 (2010) 1. _ [19] A. Takasaki, A. Zywczak, Ł. Gondek, H. Figiel, J. Alloys Compd. 580 (2013) S216. [20] R.M. Stroud, A.M. Viano, P.C. Gibbons, K.F. Kelton, S.T. Misture, Appl. Phys. Lett. 69 (1996) 2998.
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Structural phase transitions in the Ti45Zr38Ni17xFex nano-alloys and their deuterides D. Rusinek a, J. Czub a, J. Niewolski a, Ł. Gondek a, *, M. Gajewska b, A. Takasaki c, A. Hoser d, b _ A. Zywczak w, Poland AGH University of Science and Technology, Faculty of Physics and Applied Computer Science, Mickiewicza 30, 30-059 Krako w, Poland AGH University of Science and Technology, Academic Centre for Materials and Nanotechnology, Mickiewicza 30, 30-059 Krako c Department of Engineering Science and Mechanics, Shibaura Institute of Technology, Toyosu, Kotoku, Tokyo 135-8548, Japan d Helmholtz Zentrum Berlin, Hahn-Meitner-Platz 1, 14109 Berlin, Germany a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 February 2015 Received in revised form 1 June 2015 Accepted 2 June 2015 Available online 10 June 2015
The mechanically alloyed TieZreNi materials are extensively studied due to their promising properties concerning biomedical, electronic or hydrogen related applications (for example the gaseous hydrogen storage and the MNiH batteries). In this paper we address the very crucial issue of the structural properties and transformations of the amorphous and quasicrystalline Ti45Zr38Ni17xFex (x ¼ 0, 4, 8) and their hydrides. According to the neutron diffraction results, the transformation of the amorphous Ti45Zr38Ni17 phase into the icosahedral quasicrystalline state (the i-phase) is quasi-continuous and starts at the relatively low temperature of 300 C. At 500 C the i-phase is well-developed. At higher temperatures the i-phase transforms into the approximant w-phase and eventually into the cubic phase (the c-phase). Interestingly, the deuterided i-phase exhibits completely different thermal evolution. Namely, this phase decomposes into the simple intermetallic compounds above 625 C. What is worthmentioning is that the release of deuterium is strictly related to that structural decomposition. The possibility of hydrogenation of the amorphous Ti45Zr38Ni17xFex phases with maintaining the amorphous nature of the alloys is the other extremely important field of our interest. We established a processing route to meet our goal. Finally, we show that introducing deuterium triggers an exciting phase transition from the deuterided amorphous phase into the unknown before, partially disordered, quasicrystallinelike phase (the glassy quasicrystal) without releasing of deuterium. © 2015 Elsevier B.V. All rights reserved.
Keywords: Hydrides Quasicrystals Amorphous alloys Neutron diffraction X-ray diffraction
1. Introduction The TieZreNi alloys have several interesting properties that make them very promising in certain fields of applications. Namely, they might be used for high temperature protective coatings [1], biomedical applications [2], metal-hydride based batteries [3], and memory shape alloys [4]. It seems that the hydrogen sorption properties, e.g. the concentration of hydrogen exceeding significantly 2 wt.%, indicate their potential in the hydrogen related applications [5e8]. On the other hand, the basic properties of such alloys and their hydrides are still not fully understood. The TieZreNi compounds can be synthesized in the crystalline, amorphous and quasicrystalline phases. The quasicrystalline phase
* Corresponding author. E-mail address: [email protected] (Ł. Gondek). http://dx.doi.org/10.1016/j.jallcom.2015.06.023 0925-8388/© 2015 Elsevier B.V. All rights reserved.
is particularly interesting for the hydrogen storage applications. The quasicrystalline TieZreNi alloys (the i-phase) may be obtained by the melt spinning technique, but the material may react with a crucible [7]. The mechanical alloying (MA) is the more advantageous method of producing the quasicrystalline TieZreNi compounds. However, a subsequent thermal treatment is necessary [5,9,10]. In this process, the amorphous alloy is obtained by the MA synthesis, while the further annealing transforms the alloy into the stable quasicrystalline phase. This is in agreement with the ternary phase diagrams that confirm that the TieZreNi quasicrystals can be thermodynamically stable [11,12]. The icosahedral structure of the TieZreNi compounds was evidenced by X-ray and neutron diffraction, while the local atomic structure of the quasicrystals and approximants were derived from the EXAFS measurements [13e17]. For the Ti40Zr40Ni20 the transformation from the amorphous into the quasicrystalline phases occurs mainly between 500 and 550 C [16].
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95
While hydrogenation of the quasicrystalline TieZreNi alloys could be easily performed, hydrogenation of the amorphous Ti45Zr38Ni17xFex conducted at elevated temperatures led to decomposition of the specimen [18]. Particularly, formation of the simple hydride phases such as the TiH2, ZrH2 and (ZrTi)H2 was observed. In the case of the quasicrystal 45:38:17-related phases, the hydrides of the quasicrystal and the simple hydride phases were found after hydrogenation [5,6,19]. The aim of the present work was to track the Ti45Zr38Ni17 transformation from the amorphous to the quasicrystalline state as the nature of such a process is not entirely understood. For the first time, the in-situ neutron diffraction technique was used for this purpose and unveiled the quasi-continuous character of this transformation. On the other hand, the issue of the hydrogen influence on the structural properties of the investigated alloys was addressed. Namely, we succeeded in obtaining the amorphous Ti45Zr38Ni17xFex (x ¼ 4, 8) alloys with the full capacity of hydrogen and then we evidenced an extraordinary phase transition during increasing of temperature. 2. Experimental The amorphous Ti45Zr38Ni17 material and the Fe-doped Ti45Zr38Ni17, that were the base for the further processing, were synthesized by MA. Then, the product was tested by means of X-ray powder diffraction (XRD) in order to check its quality. The resulting diffraction pattern revealed no reflections related to the crystalline phases confirming the fully amorphous state of the powder. The quasicrystalline phase was obtained by thermal treatment of the amorphous powder. Afterwards, the quasicrystalline powder was deuterided. All the above syntheses were performed following the procedures reported in details in Ref. [17]. The XRD measurements were done by means of the Panalytical Empyrean powder diffractometer equipped with the Anton Paar HTK1200 high-temperature chamber. The deuteriding of the alloys was processed using the Setaram PCT-PRO Sievert's apparatus. Introducing deuterium instead of hydrogen created the possibility to perform the neutron diffraction (ND) experiments. Deuterium is a good neutron scatterer, while hydrogen exhibits a very large incoherent scattering cross section which makes it not suitable for ND. The procedure for the amorphous specimens was as follows. Firstly, the samples were placed in a controlled-atmosphere glove-box, grinded in a mortar and sieved (20 mm mesh) in order to avoid agglomeration of the powder grains. Then each sample was weighted and placed in a reaction chamber. Then the closed reaction chamber was transferred outside the glove-box and connected to the Sievert's system. The reaction chamber was then evacuated and flushed several times by helium (purity 6N). Next, a leak check was done and the volume of the filled chamber was measured (at 30 C). The sample was evacuated again after that procedure and the temperature ramp (1 deg/min.) up to 180 C was applied. Evacuation at that temperature lasted for at least 18 h in order to desorb the spurious gases. The stability of the amorphous specimen was checked prior to the thermal treatment by the in-situ XRD measurements. Then the sample was cooled down to the temperature of 30 C that was maintained. Next, deuterium gas (99.8% D þ 0.2% H, of 6N purity) was applied to the calibrated volume and expanded to the reaction chamber under the pressure of 40 bar. The uptake reaction was observed just a few minutes after applying the pressure so there was no need to increase the temperature to trigger the reaction. The uptake lasted up to 100 min resulting in 2.35 wt.% of the deuterium concentration. Eventually, the pressure was lowered to 1 bar (absolute) and the sample was checked for possible desorption by next 64 h. As a matter of fact, the observed desorption was negligibly small and the
91
total concentration of 2.34 wt.% was achieved (x ¼ 0). Introducing the Fe-doping lowered the overall deuterium uptake: 2.31 wt.% (x ¼ 4) and 2.26 wt.% (x ¼ 8), however the reaction proceeded faster. Then the samples were checked by XRD in order to get a preliminary confirmation of their amorphous nature. The XRD measurements were repeated and the specimens were prepared shortly before the ND experiments. The ND measurements were performed by means of the E6 focusing powder diffractometer (BENSC-Berlin) and an incident neutron wavelength of 2.45 Å. The in-situ measurements were performed up to 800 C using the high-temperature HTF1 furnace (AS Scientific Products). Each powdered sample was enclosed in a quartz tube under a flow of the high-purity argon gas (6N). The background originating from the furnace and the quartz tube was measured prior to the experiment in the full temperature range. Therefore, it was possible to extract the background from the collected data. It is worth mentioning that ND has a huge advantage over XRD as the structural features related to deuterium can be easily evidenced. The transmission electron microscopy (TEM) investigations were performed using the FEI Tecnai TF20 X-TWIN (FEG) microscope at an accelerating voltage of 200 kV. The bright field (BF) imaging and the selected area electron diffraction (SAED) technique were applied.
3. Results and discussion 3.1. The Ti45Zr38Ni17 amorphous phase transformations As it is apparent from the results of the DSC measurements, only one well-defined anomaly at about 530e560 C can be noticed (see Fig. 1). However, a broad feature existing only for the first heating cycle is visible at the left side of that maximum. This feature ranging from 300 up to 500 C has been commonly attributed to releasing of the energy gained during MA. However, our neutron diffraction studies unveiled the true origin of such behavior. The insitu neutron diffraction measurements in the 250e600 C temperature range are presented in Fig. 2. The data below 250 C are not shown as there were no significant changes noticed and therefore the fully amorphous state of the sample between 30 and 250 C was confirmed. The temperature of the sample was stabilized for 15 min before collection of each diffraction pattern. After
Fig. 1. The DSC curve collected for the Ti45Zr38Ni17 amorphous phase (the 1st cycle) and the repeated measurement (the 2nd cycle) for the same specimen.
92
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95
Fig. 2. The neutron diffraction patterns of the initially amorphous Ti45Zr38Ni17 at elevated temperatures.
that, the sample was maintained at particular temperature for further 60 min. Afterwards, an additional scan was performed to check the time evolution of the diffraction pattern. Then, the temperature was changed and the above sequence was performed again. We noticed just a slight time-dependence of the diffraction patterns at the investigated temperatures. According to Fig. 2, the neutron diffraction patterns exhibit apparent evolution starting from 300 C. Namely, several well defined reflections appeared at about 38.1, 62.8, 94.9 and 115.1 of the 2Q angle. Those reflections can be attributed to the i-phase. At about 500 C, the reflections of the i-phase were well developed. Hence, the quasi-continuous character of the transition from the amorphous into quasicrystalline phase was evidenced. Above 525 C, several small additional reflections appeared. Consequently, at temperature of 600 C the apparent lines at 56.1, 59.1, 92.0 and 102.0 of the 2Q angle were observed. Moreover, the main reflections became significantly narrowed. All of those indicated the crystalline character of the structure stabilized above 550 C. Indeed, this structure can be indexed as the approximant of the i-phase (the w-phase). The transition at about 550 C was reversible, therefore the w-phase transformed back into the i-phase during cooling. This is the reason for the existence of the DSC maxima in the first and the next cycles. The diffraction pattern collected at 550 C can be indexed as the cubic I m-3 space group with the lattice constant a ¼ 14.344 (1) Å as was expected for the wphase. Tracking the thermal evolution of the w-phase of the Ti45Zr38Ni17 at high temperatures was the next step. Therefore, the in-situ heating was performed again, but at that time temperature of 700 C was reached (see Fig. 3). Apart from the expected transition from the i-phase into the w-phase, an additional transition at about 675 C from the w-phase into the new cubic phase was noticed. The new phase seemed to be not very different from the wphase as a part of reflections were in similar positions. However, several reflections disappeared (see eg. at 50 and 56.1 of the 2Q) while the additional appeared (at 45 of the 2Q). Nevertheless, the structure can be indexed in the cubic system (the c-phase) similarly to the w-phase. The new phase was stable and remained unchanged during cooling down to room temperature. To confirm the sequence of the phase transitions, the in-situ high-temperature XRD studies were conducted. The obtained results entirely confirmed that the w-phase transforms to the c-phase above 675 C. Only the changes related to the lattice contraction were observed
Fig. 3. The thermal evolution of the w-phase of the Ti45Zr38Ni17 as seen by ND.
during cooling down to room temperature. Therefore, it must be concluded that the newly discovered c-phase was stable. To sum up this paragraph, we evidenced the quasi-continuous transition from the amorphous into the quasicrystalline i-phase, what is in perfect agreement with the DCS data. At temperatures of 550 C the approximant w-phase appeared, what accords with the DSC data as well. Further heating of the Ti45Zr38Ni17 specimen above 675 C led to the occurrence of another transition into an unknown cubic phase. The latter transition was not reversible, as the phase was maintained during cooling down to room temperature. 3.2. Deuterium desorption from the quasicrystalline Ti45Zr38Ni17 iphase The icosahedral i-phase can be deuterided up to 2.3 wt% [17]. After deuterization its quasicrystalline structure is maintained. The thermal desorption studies (TDS) presented in Ref. [17] show that the maximum of the hydrogen release occurred at the temperature of about 430 C. However, the sample was continuously evacuated for the TDS measurements, while in our studies, the sample was kept under 1 bar of argon (absolute pressure). That pressure was maintained during the whole experimental cycle, therefore it should be expected that desorption of deuterium occurred at higher temperatures. In Fig. 4 the neutron diffraction results for the deuterided iphase are presented. In the RT-350 C temperature range the diffraction patterns did not exhibit significant changes. As can be seen for the deuteride, an additional strong reflection that was not found for the pure i-phase was visible at about 97 of the 2Q. Appearing of such an additional reflection/reflections is typical for deuterides as deuterium atoms, in contradiction to hydrogen, are good neutron scatterers. This reflection can be attributed to the metal-deuterium mean distance and in our case the related dspacing is about 1.7 Å. Therefore, t existence of this reflection is a good probe for the deuterium concentration in the sample. As can be inferred from the Fig. 4, the significant changes were triggered at the temperature of 475 C at which splitting of the above mentioned reflection occurred. However, lowering of its intensity was noticed. That tendency was more pronounced when temperature was increased, while above 525 C the reflections at 46 and 53 of the 2Q increased their intensities what suggests that some kind of rearrangements of the i-phase occurred. Eventually, at about 625 C the deuterium related reflection disappeared. In fact, at that
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95
93
should be underlined that the release of deuterium was strictly connected with the above discussed structural decomposition of the i-phase. 3.3. Deuterization of the Ti45Zr38Ni17xFex (x ¼ 4,8) amorphous phases
Fig. 4. The in-situ ND patterns of the deuterided i-phase of the Ti45Zr38Ni17.
temperature formation of the new phase that was apparently different from the i- and w-phases was observed. Moreover, that phase was different from the c-phase evidenced for non-deuterided i-phase. At 725 C additional rearrangements of the reflections occurred suggesting further decomposition of the i-phase. No traces of the i-phase were noticed after cooling down. The sample consisted of a mixture of some crystalline phases and only the Ti2Ni phase was unambiguously recognized among them. That is in line with the findings reported in Ref. [20], where an occurrence of the Ti2Ni accompanying hydrogen desorption from the i-phase was shown. The products formed during the hydrogen release from the Ti45Zr38Ni17 at elevated temperatures were extremely dependent on the thermodynamic conditions (pressure, a heating ramp) and the other factors, e.g. the morphology of the sample and the hardly avoidable oxygen contamination of the surface during long time experiments at elevated temperatures [20]. We suggest a possibility of decomposition of the deuterided iphase into the simple intermetallic compounds just above 625 C. This is an important difference as compared to the i-phase for which the transition into the approximant phase was evidenced. It
As it is mentioned in the introduction and the experimental part, we found the route of maintaining the amorphous nature of the specimens after introducing deuterium. Such a possibility is of the utmost interest as it addresses the crucial issue of the behavior of hydrogen/deuterium in the amorphous alloys. It is commonly believed that hydrogen in an amorphous alloy is bounded to its particular constituents leading to the formation of the nanocrystalline simple hydrides. The XRD measurements suggested the diffraction patterns for both the base and deuterided specimens are typical for the amorphous materials. Moreover, the broad main maxima of the deuterided specimens were shifted towards the lower angles with respect to the base alloys (for x ¼ 4 see Fig. 5), what was an evident sign of enhancing of the mean metalemetal distances. SAED confirmed the amorphous nature of the deuterided sample (an inset to Fig. 5). All the results indicate that deuterium is not bounded to the particular constituent. The morphology of the raw and deuterided powders was similar as can be inferred from the bright field TEM experiment (Fig. 5). Presence of the submicrometer agglomerates of primary 300 nm particles was observed in both types of specimens. Thus, the deuterization process did not lead to further decomposition of the material. As deuterium cannot be tracked by the X-rays or electron diffraction, the neutron diffraction experiment was performed in order to obtain the final confirmation. Indeed, the collected diffraction patterns exhibited clearly that apart from the broad maxima related to the distribution of the metalemetal distances there was another broad maxima that can be attributed to the distribution of the metal-deuterium distances (the Fig. 5). The width of the latter is a direct confirmation that deuterium is not bounded to any specific element in a well-defined manner but it is just an equivalent constituent of the alloy. To strengthen this conclusion, the thermal evolution of the amorphous deuterided alloys was observed in-situ by means of ND as depicted in Fig. 6. First of all, at the temperature of about 250 C
Fig. 5. The diffraction patterns and TEM images of the amorphous Ti45Zr38Ni13Fe4 before (black) and after (red) deuterization. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
94
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95
Fig. 6. Thermal evolution of the deuterided amorphous Ti45Zr38Ni13xFex (x ¼ 4 and x ¼ 8) specimens.
two sharp, well-defined reflections appeared for the deuterided amorphous Ti45Zr38Ni17xFex (x ¼ 4, 8) alloys. The extraordinary nature of such diffraction patterns must be underlined as there were no other reflections in the entire investigated 2Q range. A careful inspection of the patterns revealed that those strong reflections appeared at the same angles as two main reflections visible in the case of the deuterided Ti45Zr38Ni17 quasicrystal (please compare Fig. 4). Although the smaller reflections were missing, it seems that the newly evidenced structure was somewhat similar to the deuterided quasicrystalline phase that might be considered as the partially disordered quasicrystalline i-phase (the PDi-phase). The further thermal evolution of that phase was different from that for the deuterided Ti45Zr38Ni17 quasicrystal (there is no intermediate phase in the 625e700 C temperature range), but the final product was similar to the deuterided quasicrystal related mainly to the Ti2Ni, TiFe and Zr2Ni phases. To conclude our observations, the presence of deuterium in the amorphous alloys Ti45Zr38Ni17xFex (x ¼ 4, 8) triggered the phase transition from the amorphous into the PDi-phase in the relatively narrow temperature range (250e275 C). The latter is in a contradiction to the transition from the amorphous to the i-phase for the pure Ti45Zr38Ni17 that occurred in the much broader temperature range (see Fig. 2). However, it supports the conclusion that the PDiphase was formed. In the higher temperatures the PDi-phase decomposed directly into the simple intermetallic compounds with the release of deuterium. 4. Concluding remarks The mechanically-alloyed, initially amorphous, Ti45Zr38Ni17 material undergoes the structural transformation into the quasicrystalline state (the i-phase) in a quasi-continuous manner. The transformation occurs in a broad temperature range between 300 and 500 C, while at 550 C the approximant w-phase stabilizes. Our studies show that at higher temperatures the approximant wphase disappears and the cubic phase (the c-phase) is formed. The latter is stable during cooling down to the room temperature. On the other hand, the deuterided i-phase exhibits completely different thermal evolution. It decomposes into the simple intermetallic compounds above 625 C. What is worth-mentioning is
that the release of deuterium is mostly related to that structural decomposition. Another interesting issue is that decomposition of the icosahedral phase is not preceded by establishing of the approximant w-phase. These results show that the deuterided icosahedral phase is more stable against the transformation into its approximant, however its decomposition is irreversible. The extremely interesting results were obtained by investigating the deuterided, but still amorphous Ti45Zr38Ni17xFex (x ¼ 4, 8) alloys. The inspection of the X-ray, electron and neutron diffraction patterns indicates that deuterium is an equivalent constituent of the alloy and influences significantly the thermal evolution. Namely, it triggers the phase transition from the amorphous to the partially disordered i-phase phase that have not been evidenced yet (the glassy quasicrystal). It seems that occurrence of such a glassy quasicrystal phase might be an intrinsic property of the thermal evolution of the other amorphous, hydrided materials. The possibility of obtaining the amorphous, but hydrided alloys with the hydrogen capacity exceeding 2.4 wt.% is extremely interesting in the context of the hydrogen storage and will be an object of further studies. Acknowledgments This research project was supported by the Polish-Norwegian Research Programme through the project 'Nanomaterials for hydrogen storage’ number 210733, the European Commission under the 7th Framework Programme through the 'Research Infrastructure' action of the 'Capacities' Programme, NMI3-II Grant number 283883 and by the Polish Ministry of Science and Higher Education (MNiSW). References [1] Y. Guan, J.-G. Liu, C.-W. Yan, Int. J. Electrochem. Sci. 6 (2011) 4853. [2] H. Lefaix, F. Prima, S. Zanna, P. Vermaut, P. Dubot, P. Marcus, D. Janickovic, P. Svec, Mater. Trans. 48 (2007) 278. camps, E. Leroy, A. Percheron-Gue gan, Electro[3] B. Guiose, F. Cuevas, B. De chimica Acta 54 (2009) 278. [4] S. Inoue, N. Sawada, T. Namazu, Vacuum 83 (2009) 664. [5] YuV. Zhernovenkova, L.A. Andreev, S.D. Kaloshkin, T.A. Sviridova, V.V. Tcherdyntsev, I.A. Tomilin, J. Alloys Compd. 434e435 (2007) 747. [6] A. Takasaki, K.F. Kelton, J. Alloys Compd. 347 (2002) 295.
D. Rusinek et al. / Journal of Alloys and Compounds 646 (2015) 90e95 [7] R.R. Shahi, T.P. Yadav, M.A. Shaz, O.N. Srivastava, S. van Smaalen, Int. J. Hyd. Energy 36 (2011) 592. [8] A. Kocjan, P.J. McGuiness, S. Kobe, Int. J. Hyd. Energy 35 (2010) 259. [9] YuV. Zhernovenkova, T.A. Sviridova, V.V. Tcherdyntsev, S.D. Kaloshkin, A.L. Pomadchik, J. Non-Crystalline Solids 353 (2007) 3429. [10] A. Matsumoto, K. Kobayashi, T. Nishio, K. Ozaki, J. Alloys Compd. 434e435 (2007) 315. [11] J.P. Davis, E.H. Majzoub, J.M. Simmons, K.F. Kelton, Mater. Sci. Eng. 294e296 (2000) 104. [12] R.G. Hennig, A.E. Carlsson, K.F. Kelton, C.L. Henley, Phys. Rev. B 71 (2005) 144103. [13] R.G. Hennig, K.F. Kelton, A.E. Carlsson, C.L. Henley, Phys. Rev. B 67 (2003) 134202.
95
[14] A. Sadoc, E.H. Majzoubc, V.T. Huettd, K.F. Kelton, J. Alloys Compd. 356e357 (2003) 96. [15] A. Sadoc, J.Y. Kimc, K.F. Kelton, Mater. Sci. Eng. 294e296 (2000) 348. [16] A. Kocjan, P.J. McGuiness, M. Raji c Linari c, S. Kobe, J. Alloys Compd. 457 (2008) 144. [17] A. Takasaki, C.H. Han, Y. Furuya, K.F. Kelton, Phil. Mag. Lett. 82 (2002) 353. _ [18] A. Zywczak, D. Shinya, Ł. Gondek, A. Takasaki, H. Figiel, Sol. State Commun. 150 (2010) 1. _ [19] A. Takasaki, A. Zywczak, Ł. Gondek, H. Figiel, J. Alloys Compd. 580 (2013) S216. [20] R.M. Stroud, A.M. Viano, P.C. Gibbons, K.F. Kelton, S.T. Misture, Appl. Phys. Lett. 69 (1996) 2998.