- Email: [email protected]
Decomposition kinetics study of zirconium hydride by interrupted thermal desorption spectroscopy
Journal of Alloys and Compounds xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Decomposition kinetics study of zirconium hydride by interrupted thermal desorption spectroscopy Mingwang Ma, Li Liang, Binghua Tang, Wei Xiang, Yuan Wang, Yanlin Cheng, Xiaohua Tan ⇑ Institute of Electronic Engineering, China Academy of Engineering Physics, Mianyang 621900, China
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Zirconium hydride Thermal desorption spectroscopy Desorption kinetics Phase transforming Phase diagram
a b s t r a c t Thermal desorption kinetics of zirconium hydride powder were studied using thermogravimetry and simultaneous thermal desorption spectroscopy. The activation energies for observed desorption peaks were estimated according to Kissinger relation. The intermediate phase composition was studied using X-ray diffraction by rapid cooling on different stages of heating. The origins of the peaks were described as the equilibrium hydrogen pressure of a number of consecutive phase regions that decomposition reaction passed through. The zirconium monohydride cZrH was observed for extended periods of time at ambient conditions, which has been supposed to be metastable for a long time. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Zirconium hydride is widely employed in nuclear technology (e.g. as the moderators [1–4] and fuel [5–7] in reactors). However, hydrogen would lose from zirconium hydride at the reactor working temperature (e.g. peak temperature of 1123 K for the SNAP reactor and 823 K at the LWR [5]). This may affect the stability and integrity of hydride fuel, which is the major performance constraint for these reactors. Although many studies on the thermal, mechanical and electrical properties of zirconium hydride have been reported [8–11], limited information is available on hydrogen desorption kinetic process. It is required to understand the hydrogen desorption kinetics of zirconium hydride, which is important for understanding the final state of the hydride, such as the amount of hydrogen remaining after thermal processing. Thermal desorption spectroscopy (TDS) is commonly used to identify the rate-controlling step of H2 evolution and obtain the kinetic parameters for many metal hydrides [12–15]. Interrupted TDS, which means a step-by-step heating in thermal desorption regime interrupted by fast sample cooling, is especially useful to test intermediate phase composition and has been applied to study the decomposition kinetics of TiH2 by Borchers et al. [16]. In this work, decomposition kinetics of zirconium hydride powder were studied by conventional TDS and interrupted TDS in detail, with the goal of obtaining desorption kinetic parameters and identifying the origins of desorption peaks, respectively. Based on the combined analysis of interrupted TDS and X-ray diffraction ⇑ Corresponding author. Tel.: +86 816 2489785; fax: +86 816 2487549. E-mail address: [email protected] (X. Tan).
(XRD), the observed desorption peaks can be attributed to different phase transforming steps. Furthermore, the zirconium monohydride cZrH was observed once again for extended periods of time at ambient conditions, which has been supposed to be metastable for a long time [17 and references therein]. 2. Experimental Commercial ZrH2 powder (99% pure, 400 mesh) was used in this study. The decomposition of ZrH2 was studied by thermogravimetry and simultaneous thermal desorption spectroscopy (TG-TDS; IGA-003, Hiden, UK). The samples with the weight of nearly 100 mg were heated with a linear temperature ramp of 2, 5, 10, 12, 15 and 20 K/min up to 1253 K, respectively. The desorption hydrogen gas was measured using quadrupole mass spectrometer (QMS) under pure Ar gas (P99.99%) flow conditions with a flow rate of 50 ml/min. Interrupted TDS experiments were carried out with a heating rate of 20 K/min by heating the samples to the chosen temperature, namely 795, 890, 1013 and 1109 K, and then cooling to room temperature at a rate of rough 40 K/min. The phase structures were identified by XRD (Rigaku D/Max-2400) using Cu-Ka radiation, at a scanning rate of 4°/ min using a generator voltage of 40 kV and a current of 150 mA. All the XRD measurements were carried out at room temperature and the phase composition was determined on base of JCPDS files provided by the International Center for Diffraction Data.
3. Results and discussion Fig. 1 shows the TG and TDS results of the total ZrH2 decomposition for various heating rate. The H/Zr atom ratio was calculated from the weight loss and the signal of H2 evolution rate has been normalized by sample weight. It can be seen that the TDS spectrum consists of one main peak (Peak IV) and three low-temperature (Peak I, II and III) and one high-temperature (Peak V) shoulders. With the increase of heating rate, all the five peaks shift to high
http://dx.doi.org/10.1016/j.jallcom.2015.01.054 0925-8388/Ó 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: M. Ma et al., J. Alloys Comp. (2015), http://dx.doi.org/10.1016/j.jallcom.2015.01.054
2
M. Ma et al. / Journal of Alloys and Compounds xxx (2015) xxx–xxx
temperature. The activation energy for each peak was determined according to Kissinger relation [18], which is presented in inset of Fig. 1(b). The values of activation energies derived from this plot are 120, 147, 167, 188 and 114 kJ/mol for Peak I, II, III, IV and V, respectively. The magnitude is coherent with the value of 160 kJ/ mol for the enthalpy of formation of zirconium hydride through calorimetric experiments [19] and the value of 205 kJ/mol for the activation energy of hydrogen desorption from zirconium hydride through thermogravimetric analysis [7]. Fig. 2 shows a comparison of TDS spectra of the fresh ZrH2 powder and ZrH2 powder heated up to different temperature and then cooled to room temperature. It can be seen that the peaks disappear in turn from Peak I to Peak IV when the sample were quenched from 795 to 1109 K, while the peak positions of the residual peaks do not change for all the samples. To establish a correspondence between the TDS spectra and structural evolution during decomposition the phase compositions of samples were determined by XRD after heated up to the chosen temperature and then cooled to room temperature, which are marked in Fig. 3. Table 1 summarizes corresponding TG-TDS and XRD data. The original zirconium hydride containing nearly 2 H/Ti was identified as e phase (space group I4/mmm) with lattice constants a = 0.499 nm, c = 0.445 nm. The e phase exists as a single phase at hydrogen concentrations roughly above ZrH1.7 at room temperature [17]. It has a distorted-fluorite fct structure, c/a < 1, and the lattice parameters vary with composition [17]. After TDS was interrupted at 795 K, the sample still contains 1.83 H/Zr and has a structure of eZrH1.801. The lattice constants a decreases to 0.495 nm while c increases to 0.451 nm. After TDS was interrupted at 890 K, the sample contains 1.67 H/Zr and the structure mostly transfers to dZrH1.66 (a = 0.478 nm) phase. The d hydride has a fcc structure, CaF2-type, in which hydrogen occupy tetrahedral positions. A weak peak observed at 2h = 36.2° was identified as the residual eZrH1.801, which was probably caused by the temperature difference inside the powder in the experiments. After TDS was interrupted at 1013 K, the sample contains 1.35 H/Zr. The following three phases were detected: dZrH1.66 (a = 0.478 nm), cZrH (a = 0.460 nm, c = 0.496 nm) and aZr (a = 0.326 nm, c = 0.518 nm). The phase cZrH has a fct structure like the e phase from which it differs by having a lattice parameter ratio c/a greater than unity [17]. The H atoms occupy half of the tetrahedral sites in an ordered fashion, leading to a composition of H/Zr = 1. In early works, cZrH is supposed to be metastable, although it has been found under certain conditions (e.g., in the a + d phase field, at temperatures below 528 K) from experimental results [17]. Many questions about c phase still remain unanswered such as the exact composition ranges in which it
Fig. 2. TDS spectra of as received ZrH2 and ZrH2 heated up to different temperature and then cooled to room temperature.
exists, the conditions and mechanism of its precipitation, even its metastability and the possibility that it may be an equilibrium phase [17]. Since the XRD experiments were carried out more than 72 h after the heat treatment of the sample in our work, we can confirm that cZrH can exist for extended periods of time at ambient conditions, formed by quench from high temperature (e.g. 1013 K). After TDS was interrupted at 1109 K, the sample contains 0.63 H/Zr and still has the three phases: dZrH1.66 (a = 0.478 nm), cZrH (a = 0.461 nm, c = 0.495 nm) and aZr (a = 0.325 nm, c = 0.517 nm). Compared with the XRD pattern of sample quenched from 1013 K, it can be seen from the change of peak intensity that the amounts of dZrH1.66 and cZrH reduce and that of aZr augments. The sample contains 0.03 H/Zr after decomposition was completely finished at 1253 K, with the pure structure of aZr (a = 0.324 nm, c = 0.517 nm). The phase diagram of the Zr–H system [20,21] is shown in Fig. 4. (c) is put into brackets because the c phase was supposed to be metastable in early works as discussed above. The initial position, eZrH2, is located at the bottom right corner. The solid line with upward arrows denotes the changing trend of temperature/ hydrogen content obtained from the TG experiment at a heating rate of 20 K/min, which is extended with dash line toward high temperature because the experimental data was not obtained above 1200 K. The stars show the temperature points from which the heating was interrupted and the dash lines with downward arrows point to the corresponding H/Zr ratio in the sample. The phase transformation sequence for ZrH2 decomposition under heating linearly to 1253 K conditions can be schematized as follows:
Fig. 1. TG (a) and TDS (b) spectra of as received ZrH2 at different heating rate. Inset of (b) shows the Kissinger plot for different desorption peaks.
Please cite this article in press as: M. Ma et al., J. Alloys Comp. (2015), http://dx.doi.org/10.1016/j.jallcom.2015.01.054
M. Ma et al. / Journal of Alloys and Compounds xxx (2015) xxx–xxx
3
Fig. 4. Zr–H phase diagram [20,21]. The solid line with upward arrows denotes the changing trend of temperature/hydrogen content obtained from the TG experiment at a heating rate of 20 K/min. The stars show the temperature points from which the heating was interrupted and the dash lines with downward arrows point to the corresponding H/Zr ratio in the sample.
Fig. 3. XRD spectra of as received ZrH2 powder and ZrH2 powder quenched from different temperature.
e ! d þ e ! d ! b þ d ! b: The five peaks of the TDS spectra may be attributed to the five phase regions passed through. The equilibrium hydrogen pressure of each phase may be the origin of the peak and determine the intensity of the peak. Detailed descriptions of relationship between equilibrium hydrogen pressure and temperature/hydrogen content can be found in [6] and references therein. Peak I may be attributed to the equilibrium hydrogen pressure of e phase with a lower magnitude. After the sample was heated up to 795 K, then rapidly cooled and heated again, the decomposition started from d + e phase, leading to the disappearance of Peak I from the TDS spectrum (curve 2 of Fig. 2). Therefore Peak II may be attributed to the equilibrium hydrogen pressure of d + e phase. After the sample was quenched from 890 K, the decomposition started from d phase, which may be the
origin of Peak III (curve 3 of Fig. 2). The equilibrium hydrogen pressure of b + d phase may be the origin of Peak IV. Since the equilibrium hydrogen pressure of b + d and b phase is several orders larger than that of a + d + (c) phase, there is not any peak attributed to a + d + (c) phase (most likely appear at low temperature) that was observed for the sample with 1.35 H/Zr either 0.63 H/Zr even though the decomposition reaction has passed through this phase region (curve 4 and 5 of Fig. 2). Furthermore, since the b + d phase is located at the plateau regions in the pcT curves and the equilibrium hydrogen pressure increases sharply with rising temperature, Peak IV has a sharply increased leading peak and becomes the main peak of TDS spectrum. Peak V may be attributed to the equilibrium hydrogen pressure of b phase, which is the stable form of pure zirconium at high temperature with bcc structure [17] and therefore is not detected by XRD. The last step of decomposition occurs in this phase region at the high temperature. Note that the TDS results showed in this work are only suitable for the thermal desorption system, where it is assumed that the ultra-high vacuum conditions are sufficient to remove gas-phase effects. For fuel application, the reactor system would not achieve this ideal case. Therefore the decomposition process would be affected by hydrogen gas accumulated in the system. Taking into account the effect of gas pressures, the desorption rate can be modified by a factor ln(Peq/P) [22], where Peq is the temperature dependent hydrogen equilibrium pressure and P is the hydrogen partial pressure in the reactor system. According to the equilibrium hydrogen pressure presented in [7], Peq is in the range of 10– 105 Pa for the temperature ranging 773–1073 K. If P is less than but similar in magnitude to Peq, desorption would be depressed
Table 1 XRD parameters. T (K)
Residual H (H/Zr)
Phase composition (JCPDS no.)
Lattice parameters (nm)
293 795 890
2 1.83 1.67
eZrH2 (17-0314) eZrH1.801 (36-1340)
1013
1.35
a = 0.499; a = 0.495; a = 0.478 a = 0.495; a = 0.478 a = 0.460; a = 0.326; a = 0.478 a = 0.461; a = 0.325; a = 0.324;
1109
1253
0.63
0.03
dZrH1.66 (34-0649) eZrH1.801 (36-1340) dZrH1.66 (34-0649) cZrH (34-0690) aZr (05-0665) dZrH1.66 (34-0649) cZrH (34-0690) aZr (05-0665) aZr (05-0665)
c = 0.445 c = 0.451 c = 0.450 c = 0.496 c = 0.518 c = 0.495 c = 0.517 c = 0.517
Please cite this article in press as: M. Ma et al., J. Alloys Comp. (2015), http://dx.doi.org/10.1016/j.jallcom.2015.01.054
4
M. Ma et al. / Journal of Alloys and Compounds xxx (2015) xxx–xxx
at a significantly reduced rate, therefore leading to a lower weight loss rate. 4. Conclusions Thermal desorption kinetics of zirconium hydride powder were studied using conventional TDS and interrupted TDS. Five peaks were observed in the TDS spectra and phase transforming was considered as the rate limiting step. The origins of the peaks were attributed to the equilibrium hydrogen pressure of e, d + e, d, b + d and b phase region, respectively. The corresponding activation energies were determined to be 120, 147, 167, 188 and 114 kJ/mol, respectively. The cZrH phase was observed once again for extended periods of time at ambient conditions, formed by quench from high temperature (e.g. 1013 and 1109 K). Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant Nos. 51406187) and Science and Technology Development Foundation of China Academy of Engineering Physics (Grant Nos. 2014B0401060). References [1] W. Chen, L. Wang, S. Lu, Influence of oxide layer on hydrogen desorption from zirconium hydride, J. Alloys Comp. 469 (2009) 142–145. [2] K. Konashi, T. Ikeshoji, Y. Kawazoe, H. Matsui, A molecular dynamics study of thermal conductivity of zirconium hydride, J. Alloys Comp. 356–357 (2003) 279–282. [3] S. Yamanaka, K. Yamada, K. Kurosaki, M. Uno, K. Takeda, H. Anada, et al., Thermal properties of zirconium hydride, J. Nucl. Mater. 294 (2001) 94–98. [4] T. Hayashi, K. Tobita, Y. Nakamori, S. Orimo, Advanced neutron shielding material using zirconium borohydride and zirconium hydride, J. Nucl. Mater. 386–388 (2009) 119–121. [5] D. Olander, E. Greenspan, H.D. Garkisch, B. Petrovic, Uranium–zirconium hydride fuel properties, Nucl. Eng. Des. 239 (2009) 1406–1424.
[6] D. Wongsawaeng, S. Jaiyen, High-temperature absolute hydrogen desorption kinetics of zirconium hydride under clean and oxidized surface conditions, J. Nucl. Mater. 403 (2010) 19–24. [7] K.A. Terrani, M. Balooch, D. Wongsawaeng, S. Jaiyen, D.R. Olander, The kinetics of hydrogen desorption from and adsorption on zirconium hydride, J. Nucl. Mater. 397 (2010) 61–68. [8] S. Yamanaka, K. Yoshioka, M. Uno, M. Katsura, H. Anada, T. Matsuda, S. Kobayashi, Isotope effects on the physicochemical properties of zirconium hydride, J. Alloys Comp. 293 (1999) 908–914. [9] S. Yamanaka, K. Yoshioka, M. Uno, M. Katsura, H. Anada, T. Matsuda, S. Kobayashi, Thermal and mechanical properties of zirconium hydride, J. Alloys Comp. 293 (1999) 23–29. [10] B. Tsuchiya, M. Teshigawara, K. Konashi, M. Yamawaki, Thermal diffusivity and electrical resistivity of zirconium hydride, J. Alloys Comp. 330–332 (2002) 357–360. [11] M. Uno, K. Yamada, T. Maruyama, H. Muta, S. Yamanaka, Thermophysical properties of zirconium hydride and deuteride, J. Alloys Comp. 366 (2004) 101–106. [12] R. Checchetto, N. Bazzanella, A. Miotello, R.S. Brusa, A. Zecca, A. Mengucci, Deuterium storage in nanocrystalline magnesium thin films, J. Appl. Phys. 65 (2004) 1989. [13] R. Checchetto, L.M. Gratton, A. Miotello, A. Tomasi, P. Scardi, Structural evolution and thermal stability of deuterated titanium thin films, Phys. Rev. B 58 (1998) 4130–4137. [14] W. Lisowski, E.G. Keim, Z. Kaszkur, M.A. Smithers, Decomposition of thin titanium deuteride films; thermal desorption kinetics studies combined with microstructure analysis, Appl. Surf. Sci. 254 (2008) 2629–2637. [15] J.F. Fernández, F. Cuevas, C. Sánchez, Simultaneous differential scanning calorimetry and thermal desorption spectroscopy measurements for the study of the decomposition of metal hydrides, J. Alloys Comp. 298 (2000) 244–253. [16] C. Borchers, T.I. Khomenko, A.V. Leonov, O.S. Morozova, Interrupted thermal desorption of TiH2, Thermochim. Acta 493 (2009) 80–84. [17] F.A. Lewis, A. Aladjem, Hydrogen Metal Systems I, Scitec Publications, Zurich, Switzerland, 1996. [18] H.E. Kissinger, Reactive kinetics in differential thermal analysis, Ann. Chem. 29 (1957) 1702. [19] D.R. Fredrickson, R.L. Nuttall, H.E. Flotow, W.N. Hubbard, The enthalpies of formation of zirconium dihydride and zirconium dideuteride, J. Phys. Chem. 67 (1963) 1506–1509. [20] E. Zuzek, J.P. Abriata, A. San-Martin, F.D. Manchester, The H–Zr (hydrogen– zirconium) system, Bull. Alloy Phase Diagrams 11 (1990) 385–395. [21] D.R. Olander, M. Ng, Hydride fuel behavior in LWRs, J. Nucl. Mater. 346 (2005) 98–108. [22] W. Luo, K.J. Gross, A kinetics model of hydrogen absorption and desorption in Ti-doped NaAlH4, J. Alloys Comp. 385 (2004) 224–231.
Please cite this article in press as: M. Ma et al., J. Alloys Comp. (2015), http://dx.doi.org/10.1016/j.jallcom.2015.01.054
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Decomposition kinetics study of zirconium hydride by interrupted thermal desorption spectroscopy Mingwang Ma, Li Liang, Binghua Tang, Wei Xiang, Yuan Wang, Yanlin Cheng, Xiaohua Tan ⇑ Institute of Electronic Engineering, China Academy of Engineering Physics, Mianyang 621900, China
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Zirconium hydride Thermal desorption spectroscopy Desorption kinetics Phase transforming Phase diagram
a b s t r a c t Thermal desorption kinetics of zirconium hydride powder were studied using thermogravimetry and simultaneous thermal desorption spectroscopy. The activation energies for observed desorption peaks were estimated according to Kissinger relation. The intermediate phase composition was studied using X-ray diffraction by rapid cooling on different stages of heating. The origins of the peaks were described as the equilibrium hydrogen pressure of a number of consecutive phase regions that decomposition reaction passed through. The zirconium monohydride cZrH was observed for extended periods of time at ambient conditions, which has been supposed to be metastable for a long time. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Zirconium hydride is widely employed in nuclear technology (e.g. as the moderators [1–4] and fuel [5–7] in reactors). However, hydrogen would lose from zirconium hydride at the reactor working temperature (e.g. peak temperature of 1123 K for the SNAP reactor and 823 K at the LWR [5]). This may affect the stability and integrity of hydride fuel, which is the major performance constraint for these reactors. Although many studies on the thermal, mechanical and electrical properties of zirconium hydride have been reported [8–11], limited information is available on hydrogen desorption kinetic process. It is required to understand the hydrogen desorption kinetics of zirconium hydride, which is important for understanding the final state of the hydride, such as the amount of hydrogen remaining after thermal processing. Thermal desorption spectroscopy (TDS) is commonly used to identify the rate-controlling step of H2 evolution and obtain the kinetic parameters for many metal hydrides [12–15]. Interrupted TDS, which means a step-by-step heating in thermal desorption regime interrupted by fast sample cooling, is especially useful to test intermediate phase composition and has been applied to study the decomposition kinetics of TiH2 by Borchers et al. [16]. In this work, decomposition kinetics of zirconium hydride powder were studied by conventional TDS and interrupted TDS in detail, with the goal of obtaining desorption kinetic parameters and identifying the origins of desorption peaks, respectively. Based on the combined analysis of interrupted TDS and X-ray diffraction ⇑ Corresponding author. Tel.: +86 816 2489785; fax: +86 816 2487549. E-mail address: [email protected] (X. Tan).
(XRD), the observed desorption peaks can be attributed to different phase transforming steps. Furthermore, the zirconium monohydride cZrH was observed once again for extended periods of time at ambient conditions, which has been supposed to be metastable for a long time [17 and references therein]. 2. Experimental Commercial ZrH2 powder (99% pure, 400 mesh) was used in this study. The decomposition of ZrH2 was studied by thermogravimetry and simultaneous thermal desorption spectroscopy (TG-TDS; IGA-003, Hiden, UK). The samples with the weight of nearly 100 mg were heated with a linear temperature ramp of 2, 5, 10, 12, 15 and 20 K/min up to 1253 K, respectively. The desorption hydrogen gas was measured using quadrupole mass spectrometer (QMS) under pure Ar gas (P99.99%) flow conditions with a flow rate of 50 ml/min. Interrupted TDS experiments were carried out with a heating rate of 20 K/min by heating the samples to the chosen temperature, namely 795, 890, 1013 and 1109 K, and then cooling to room temperature at a rate of rough 40 K/min. The phase structures were identified by XRD (Rigaku D/Max-2400) using Cu-Ka radiation, at a scanning rate of 4°/ min using a generator voltage of 40 kV and a current of 150 mA. All the XRD measurements were carried out at room temperature and the phase composition was determined on base of JCPDS files provided by the International Center for Diffraction Data.
3. Results and discussion Fig. 1 shows the TG and TDS results of the total ZrH2 decomposition for various heating rate. The H/Zr atom ratio was calculated from the weight loss and the signal of H2 evolution rate has been normalized by sample weight. It can be seen that the TDS spectrum consists of one main peak (Peak IV) and three low-temperature (Peak I, II and III) and one high-temperature (Peak V) shoulders. With the increase of heating rate, all the five peaks shift to high
http://dx.doi.org/10.1016/j.jallcom.2015.01.054 0925-8388/Ó 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: M. Ma et al., J. Alloys Comp. (2015), http://dx.doi.org/10.1016/j.jallcom.2015.01.054
2
M. Ma et al. / Journal of Alloys and Compounds xxx (2015) xxx–xxx
temperature. The activation energy for each peak was determined according to Kissinger relation [18], which is presented in inset of Fig. 1(b). The values of activation energies derived from this plot are 120, 147, 167, 188 and 114 kJ/mol for Peak I, II, III, IV and V, respectively. The magnitude is coherent with the value of 160 kJ/ mol for the enthalpy of formation of zirconium hydride through calorimetric experiments [19] and the value of 205 kJ/mol for the activation energy of hydrogen desorption from zirconium hydride through thermogravimetric analysis [7]. Fig. 2 shows a comparison of TDS spectra of the fresh ZrH2 powder and ZrH2 powder heated up to different temperature and then cooled to room temperature. It can be seen that the peaks disappear in turn from Peak I to Peak IV when the sample were quenched from 795 to 1109 K, while the peak positions of the residual peaks do not change for all the samples. To establish a correspondence between the TDS spectra and structural evolution during decomposition the phase compositions of samples were determined by XRD after heated up to the chosen temperature and then cooled to room temperature, which are marked in Fig. 3. Table 1 summarizes corresponding TG-TDS and XRD data. The original zirconium hydride containing nearly 2 H/Ti was identified as e phase (space group I4/mmm) with lattice constants a = 0.499 nm, c = 0.445 nm. The e phase exists as a single phase at hydrogen concentrations roughly above ZrH1.7 at room temperature [17]. It has a distorted-fluorite fct structure, c/a < 1, and the lattice parameters vary with composition [17]. After TDS was interrupted at 795 K, the sample still contains 1.83 H/Zr and has a structure of eZrH1.801. The lattice constants a decreases to 0.495 nm while c increases to 0.451 nm. After TDS was interrupted at 890 K, the sample contains 1.67 H/Zr and the structure mostly transfers to dZrH1.66 (a = 0.478 nm) phase. The d hydride has a fcc structure, CaF2-type, in which hydrogen occupy tetrahedral positions. A weak peak observed at 2h = 36.2° was identified as the residual eZrH1.801, which was probably caused by the temperature difference inside the powder in the experiments. After TDS was interrupted at 1013 K, the sample contains 1.35 H/Zr. The following three phases were detected: dZrH1.66 (a = 0.478 nm), cZrH (a = 0.460 nm, c = 0.496 nm) and aZr (a = 0.326 nm, c = 0.518 nm). The phase cZrH has a fct structure like the e phase from which it differs by having a lattice parameter ratio c/a greater than unity [17]. The H atoms occupy half of the tetrahedral sites in an ordered fashion, leading to a composition of H/Zr = 1. In early works, cZrH is supposed to be metastable, although it has been found under certain conditions (e.g., in the a + d phase field, at temperatures below 528 K) from experimental results [17]. Many questions about c phase still remain unanswered such as the exact composition ranges in which it
Fig. 2. TDS spectra of as received ZrH2 and ZrH2 heated up to different temperature and then cooled to room temperature.
exists, the conditions and mechanism of its precipitation, even its metastability and the possibility that it may be an equilibrium phase [17]. Since the XRD experiments were carried out more than 72 h after the heat treatment of the sample in our work, we can confirm that cZrH can exist for extended periods of time at ambient conditions, formed by quench from high temperature (e.g. 1013 K). After TDS was interrupted at 1109 K, the sample contains 0.63 H/Zr and still has the three phases: dZrH1.66 (a = 0.478 nm), cZrH (a = 0.461 nm, c = 0.495 nm) and aZr (a = 0.325 nm, c = 0.517 nm). Compared with the XRD pattern of sample quenched from 1013 K, it can be seen from the change of peak intensity that the amounts of dZrH1.66 and cZrH reduce and that of aZr augments. The sample contains 0.03 H/Zr after decomposition was completely finished at 1253 K, with the pure structure of aZr (a = 0.324 nm, c = 0.517 nm). The phase diagram of the Zr–H system [20,21] is shown in Fig. 4. (c) is put into brackets because the c phase was supposed to be metastable in early works as discussed above. The initial position, eZrH2, is located at the bottom right corner. The solid line with upward arrows denotes the changing trend of temperature/ hydrogen content obtained from the TG experiment at a heating rate of 20 K/min, which is extended with dash line toward high temperature because the experimental data was not obtained above 1200 K. The stars show the temperature points from which the heating was interrupted and the dash lines with downward arrows point to the corresponding H/Zr ratio in the sample. The phase transformation sequence for ZrH2 decomposition under heating linearly to 1253 K conditions can be schematized as follows:
Fig. 1. TG (a) and TDS (b) spectra of as received ZrH2 at different heating rate. Inset of (b) shows the Kissinger plot for different desorption peaks.
Please cite this article in press as: M. Ma et al., J. Alloys Comp. (2015), http://dx.doi.org/10.1016/j.jallcom.2015.01.054
M. Ma et al. / Journal of Alloys and Compounds xxx (2015) xxx–xxx
3
Fig. 4. Zr–H phase diagram [20,21]. The solid line with upward arrows denotes the changing trend of temperature/hydrogen content obtained from the TG experiment at a heating rate of 20 K/min. The stars show the temperature points from which the heating was interrupted and the dash lines with downward arrows point to the corresponding H/Zr ratio in the sample.
Fig. 3. XRD spectra of as received ZrH2 powder and ZrH2 powder quenched from different temperature.
e ! d þ e ! d ! b þ d ! b: The five peaks of the TDS spectra may be attributed to the five phase regions passed through. The equilibrium hydrogen pressure of each phase may be the origin of the peak and determine the intensity of the peak. Detailed descriptions of relationship between equilibrium hydrogen pressure and temperature/hydrogen content can be found in [6] and references therein. Peak I may be attributed to the equilibrium hydrogen pressure of e phase with a lower magnitude. After the sample was heated up to 795 K, then rapidly cooled and heated again, the decomposition started from d + e phase, leading to the disappearance of Peak I from the TDS spectrum (curve 2 of Fig. 2). Therefore Peak II may be attributed to the equilibrium hydrogen pressure of d + e phase. After the sample was quenched from 890 K, the decomposition started from d phase, which may be the
origin of Peak III (curve 3 of Fig. 2). The equilibrium hydrogen pressure of b + d phase may be the origin of Peak IV. Since the equilibrium hydrogen pressure of b + d and b phase is several orders larger than that of a + d + (c) phase, there is not any peak attributed to a + d + (c) phase (most likely appear at low temperature) that was observed for the sample with 1.35 H/Zr either 0.63 H/Zr even though the decomposition reaction has passed through this phase region (curve 4 and 5 of Fig. 2). Furthermore, since the b + d phase is located at the plateau regions in the pcT curves and the equilibrium hydrogen pressure increases sharply with rising temperature, Peak IV has a sharply increased leading peak and becomes the main peak of TDS spectrum. Peak V may be attributed to the equilibrium hydrogen pressure of b phase, which is the stable form of pure zirconium at high temperature with bcc structure [17] and therefore is not detected by XRD. The last step of decomposition occurs in this phase region at the high temperature. Note that the TDS results showed in this work are only suitable for the thermal desorption system, where it is assumed that the ultra-high vacuum conditions are sufficient to remove gas-phase effects. For fuel application, the reactor system would not achieve this ideal case. Therefore the decomposition process would be affected by hydrogen gas accumulated in the system. Taking into account the effect of gas pressures, the desorption rate can be modified by a factor ln(Peq/P) [22], where Peq is the temperature dependent hydrogen equilibrium pressure and P is the hydrogen partial pressure in the reactor system. According to the equilibrium hydrogen pressure presented in [7], Peq is in the range of 10– 105 Pa for the temperature ranging 773–1073 K. If P is less than but similar in magnitude to Peq, desorption would be depressed
Table 1 XRD parameters. T (K)
Residual H (H/Zr)
Phase composition (JCPDS no.)
Lattice parameters (nm)
293 795 890
2 1.83 1.67
eZrH2 (17-0314) eZrH1.801 (36-1340)
1013
1.35
a = 0.499; a = 0.495; a = 0.478 a = 0.495; a = 0.478 a = 0.460; a = 0.326; a = 0.478 a = 0.461; a = 0.325; a = 0.324;
1109
1253
0.63
0.03
dZrH1.66 (34-0649) eZrH1.801 (36-1340) dZrH1.66 (34-0649) cZrH (34-0690) aZr (05-0665) dZrH1.66 (34-0649) cZrH (34-0690) aZr (05-0665) aZr (05-0665)
c = 0.445 c = 0.451 c = 0.450 c = 0.496 c = 0.518 c = 0.495 c = 0.517 c = 0.517
Please cite this article in press as: M. Ma et al., J. Alloys Comp. (2015), http://dx.doi.org/10.1016/j.jallcom.2015.01.054
4
M. Ma et al. / Journal of Alloys and Compounds xxx (2015) xxx–xxx
at a significantly reduced rate, therefore leading to a lower weight loss rate. 4. Conclusions Thermal desorption kinetics of zirconium hydride powder were studied using conventional TDS and interrupted TDS. Five peaks were observed in the TDS spectra and phase transforming was considered as the rate limiting step. The origins of the peaks were attributed to the equilibrium hydrogen pressure of e, d + e, d, b + d and b phase region, respectively. The corresponding activation energies were determined to be 120, 147, 167, 188 and 114 kJ/mol, respectively. The cZrH phase was observed once again for extended periods of time at ambient conditions, formed by quench from high temperature (e.g. 1013 and 1109 K). Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant Nos. 51406187) and Science and Technology Development Foundation of China Academy of Engineering Physics (Grant Nos. 2014B0401060). References [1] W. Chen, L. Wang, S. Lu, Influence of oxide layer on hydrogen desorption from zirconium hydride, J. Alloys Comp. 469 (2009) 142–145. [2] K. Konashi, T. Ikeshoji, Y. Kawazoe, H. Matsui, A molecular dynamics study of thermal conductivity of zirconium hydride, J. Alloys Comp. 356–357 (2003) 279–282. [3] S. Yamanaka, K. Yamada, K. Kurosaki, M. Uno, K. Takeda, H. Anada, et al., Thermal properties of zirconium hydride, J. Nucl. Mater. 294 (2001) 94–98. [4] T. Hayashi, K. Tobita, Y. Nakamori, S. Orimo, Advanced neutron shielding material using zirconium borohydride and zirconium hydride, J. Nucl. Mater. 386–388 (2009) 119–121. [5] D. Olander, E. Greenspan, H.D. Garkisch, B. Petrovic, Uranium–zirconium hydride fuel properties, Nucl. Eng. Des. 239 (2009) 1406–1424.
[6] D. Wongsawaeng, S. Jaiyen, High-temperature absolute hydrogen desorption kinetics of zirconium hydride under clean and oxidized surface conditions, J. Nucl. Mater. 403 (2010) 19–24. [7] K.A. Terrani, M. Balooch, D. Wongsawaeng, S. Jaiyen, D.R. Olander, The kinetics of hydrogen desorption from and adsorption on zirconium hydride, J. Nucl. Mater. 397 (2010) 61–68. [8] S. Yamanaka, K. Yoshioka, M. Uno, M. Katsura, H. Anada, T. Matsuda, S. Kobayashi, Isotope effects on the physicochemical properties of zirconium hydride, J. Alloys Comp. 293 (1999) 908–914. [9] S. Yamanaka, K. Yoshioka, M. Uno, M. Katsura, H. Anada, T. Matsuda, S. Kobayashi, Thermal and mechanical properties of zirconium hydride, J. Alloys Comp. 293 (1999) 23–29. [10] B. Tsuchiya, M. Teshigawara, K. Konashi, M. Yamawaki, Thermal diffusivity and electrical resistivity of zirconium hydride, J. Alloys Comp. 330–332 (2002) 357–360. [11] M. Uno, K. Yamada, T. Maruyama, H. Muta, S. Yamanaka, Thermophysical properties of zirconium hydride and deuteride, J. Alloys Comp. 366 (2004) 101–106. [12] R. Checchetto, N. Bazzanella, A. Miotello, R.S. Brusa, A. Zecca, A. Mengucci, Deuterium storage in nanocrystalline magnesium thin films, J. Appl. Phys. 65 (2004) 1989. [13] R. Checchetto, L.M. Gratton, A. Miotello, A. Tomasi, P. Scardi, Structural evolution and thermal stability of deuterated titanium thin films, Phys. Rev. B 58 (1998) 4130–4137. [14] W. Lisowski, E.G. Keim, Z. Kaszkur, M.A. Smithers, Decomposition of thin titanium deuteride films; thermal desorption kinetics studies combined with microstructure analysis, Appl. Surf. Sci. 254 (2008) 2629–2637. [15] J.F. Fernández, F. Cuevas, C. Sánchez, Simultaneous differential scanning calorimetry and thermal desorption spectroscopy measurements for the study of the decomposition of metal hydrides, J. Alloys Comp. 298 (2000) 244–253. [16] C. Borchers, T.I. Khomenko, A.V. Leonov, O.S. Morozova, Interrupted thermal desorption of TiH2, Thermochim. Acta 493 (2009) 80–84. [17] F.A. Lewis, A. Aladjem, Hydrogen Metal Systems I, Scitec Publications, Zurich, Switzerland, 1996. [18] H.E. Kissinger, Reactive kinetics in differential thermal analysis, Ann. Chem. 29 (1957) 1702. [19] D.R. Fredrickson, R.L. Nuttall, H.E. Flotow, W.N. Hubbard, The enthalpies of formation of zirconium dihydride and zirconium dideuteride, J. Phys. Chem. 67 (1963) 1506–1509. [20] E. Zuzek, J.P. Abriata, A. San-Martin, F.D. Manchester, The H–Zr (hydrogen– zirconium) system, Bull. Alloy Phase Diagrams 11 (1990) 385–395. [21] D.R. Olander, M. Ng, Hydride fuel behavior in LWRs, J. Nucl. Mater. 346 (2005) 98–108. [22] W. Luo, K.J. Gross, A kinetics model of hydrogen absorption and desorption in Ti-doped NaAlH4, J. Alloys Comp. 385 (2004) 224–231.
Please cite this article in press as: M. Ma et al., J. Alloys Comp. (2015), http://dx.doi.org/10.1016/j.jallcom.2015.01.054