- Email: [email protected]
Effect of thermal history on the terminal solid solubility of hydrogen in Zircaloy-4
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Effect of thermal history on the terminal solid solubility of hydrogen in Zircaloy-4 Ju-Seong Kim a,*, Yong-Soo Kim b a b
Korea Atomic Energy Research Institute, 989-111 Daedeokdaero, Yuseong-gu, Daejeon, 305-353, Republic of Korea Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul, 133-791, Republic of Korea
article info
abstract
Article history:
The terminal solid solubility (TSS) of hydrogen in zirconium alloys has a hysteresis. The
Received 16 June 2014
TSS of hydrogen in Zircaloy-4 during cooling and heating were studied using differential
Received in revised form
scanning calorimetry (DSC) with a hydrogen content of 40e731 wppm. A significant hys-
1 August 2014
teresis gap was observed between the TSS for dissolution (TSSD) and precipitation (TSSP). It
Accepted 10 August 2014
was confirmed that the hydrogen dissolution temperature was unaffected by the previous
Available online 2 September 2014
thermal history in comparison with the hydride precipitation temperature. The TSSP temperature increased with a decrease in the maximum temperature, but a significant
Keywords:
temperature gap remained even when the maximum temperature was equal to the TSSD
Terminal solid solubility
temperature. The terminal solid solubility of hydrogen in Zircaloy-4 can be represented by
Hydrogen
the following equations.
Zircaloy-4 Thermal history
TSSD: C ¼ 2.255 105 exp (39101/RT). TSSP: C ¼ 4.722 104 exp (26843/RT). TSSP2: C ¼ 8.612 105 exp (30583/RT).
Based on the experimental results hydrogen solubility path depending on the previous thermal history was proposed. Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Introduction Zirconium alloys have been used as reactor component materials owing to a low neutron absorption cross section, high corrosion resistance, and proper mechanical properties at high temperature. Hydrogen pick-up in zirconium
alloy is one of the most important issues in reactor operation [1]. Typical nuclear cladding discharged from the pressurized water reactor has 200e600 wppm of hydrogen depending on the burn-up [2]. Most of the hydrogen can dissolve in Zr-matrix at high temperatures. However, the brittle hydrides can be precipitated during the cool-down which can reduce the ductility of the cladding [3], and
* Corresponding author. Tel.: þ82 42 868 8689; fax: þ82 42 863 0565. E-mail addresses: [email protected], [email protected] (J.-S. Kim). http://dx.doi.org/10.1016/j.ijhydene.2014.08.018 0360-3199/Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
threaten the integrity of cladding such as hydride reorientation [4] or delayed hydride cracking (DHC) [5,6]. Interestingly the TSS of hydrogen in zirconium alloys for dissolution (TSSD) and precipitation (TSSP) have a hysteresis. This hysteresis can affect the DHC behavior and hydride re-orientation. Consequently, the terminal solid solubility (TSS) of hydrogen in zirconium and its alloys has been measured using a variety of techniques: diffusion equilibrium [7e10], dilatometry [11e15], internal friction and dynamic elastic modulus [16e19], small-angle neutron diffraction [20], differential scanning calorimetry (DSC) [15,16,21e25], and synchrotron X-ray diffraction [26]. Kearns [8] reviewed the previous investigations [7,11,27e30] and performed experiment with employing diffusion method to determine the TSS of hydrogen in zirconium alloys. His best fit of the data often used as a standard TSSD for zirconium and Zircaloy-2 and Zircaloy-4. The other methods such as internal friction, DSC and dilatometry are dynamic methods to measure the hydrogen thermal behaviors during the temperature transient. However, most of the works on the TSS of hydrogen in zirconium alloys are about the TSSD rather than the TSSP. In addition, there are no available data on TSSP2 of hydrogen in Zircaloy-4. The TSSD of hydrogen in zirconium alloy seems to be less sensitive to the previous thermal history, but the TSSP is strongly dependent on the thermal history. The extent of the hysteresis depends on the temperature history such as peak temperature, a holding time at the peak temperature and a cooling rate; therefore, the published data on TSSP is considerable scatter. The objectives of this study are to confirm the TSS of hydrogen in Zircaloy-4 and evaluate the maximum temperature effect on the TSSP. The DSC method was employed to measure the TSS of hydrogen in Zircaloy-4 owing to simplicity and reproducibility.
Experimental Materials
16443
Differential scanning calorimetry Basically, DSC measures the changes in the difference in the heat exchange between the sample and the reference. The change in the heat flow signal indicates an exothermic or endothermic reaction when a phase transformation occurs. Fig. 1 shows a typical DSC heat flow and its derivative heat flow curves of hydrided Zircaloy-4 during heating. As the temperature increases, the hydrogen begins to dissolve and the heat flow curve decreases steadily indicating that the hydrogen dissolution process is an endothermic reaction. In this work, the TSS of hydrogen in Zircaloy-4 was measured using a heat flux DSC (Netzsch 200 F3). The calibrations were conducted prior to the experiment using metal standard samples (In, Sn and Zn). DSC measurements were carried out in purified N2 at a flow rate of 50 cm3/min. Generally, DSC is more sensitive at higher heating/cooling rates but such rates tend to slightly increase the TSSD temperature and decrease the TSSP temperature. It is known that the effect of heating rate on the dissolution of hydrogen is negligible [21,22], whereas the effect of cooling rate on the precipitation is more susceptible. Considering these effects, TSS measurements were conducted at cooling and heating rate of 20 C/min. Most samples were heated to peak temperature of 550 C. The dwell time at the peak temperature was 5 min and then cooled to 40 C. Some high hydrogen containing specimens were heated to 580 C. Each specimen was experienced four identical thermal cycles. All the data were used to calculate the mean TSS values excluding that of first heating cycle. Additional tests were conducted to investigate the hydrogen behavior during thermal cycling and effect of the maximum temperature.
Determination of the TSSD and TSSP temperatures There are three important temperatures for determining the TSS: peak temperature, completion temperature, and maximum slope temperature. The peak temperature (PT) is the maximum or minimum point in the heat flow curve and the maximum slope temperature (MST) is the point of maximum deviation of the heat flow. The completion
Cold worked stress-relieved (CWSR) Zircaloy-4 finally annealed at 470 C for 5 h was used for the present TSS measurements. Table 1 shows the chemical composition of Zircaloy-4. Hydrogen charging was conducted at 400 C by a Sieverts-type apparatus. After hydrogen charging, the heat treatment was performed at 400 C for 5e10 h to ensure a uniform distribution of hydride. As a result, 40e1121 wppm of hydrided specimens were obtained. All of the hydrogen concentrations were analyzed by a hydrogen determinator (LECO RH-404), which uses an inert gas fusion method. The error of the determinator is less than 5 wppm.
Table 1 e Chemical composition of Zircaloy-4 (unit: weight%). Element
Sn
Fe
Cr
C
O
Si
Zr
Weight
1.32
0.21
0.11
0.013
0.13
0.0092
Balance
Fig. 1 e DSC curve and its time derivative of Zircaloy-4 specimen with a hydrogen content of 288 wppm during heat-up.
16444
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
temperature (CT) or onset temperature (OT) is the point of intersection of the two tangent lines, which is the interpolated baseline and the dissolution or precipitation discontinuity line. However, there are no rules to determine the TSS temperature. In the present study, the TSSD temperature was determined by MST as a detectable transition temperature for both heating and cooling. The typical DSC heating curve shows a broad peak, for example, the temperature difference between PT and MST is around 30 C as shown in Fig. 1. On the other hand, the typical DSC curve during cooling has a sharp peak with respect to heating curves (Fig. 2) but the cooling curve also has a similar broad peak when the maximum temperature is reduced. The solid solubility limit of hydrogen in a-Zr is known to be around 700e713 wppm [31,32] and the temperature of the eutectoid reaction of a-Zr þ hydride to (aþb)-Zr is 550 ± 3 C [32]. Fig. 3 shows a DSC curve of this phase transition. The reaction of a-Zr þ hydride to (aþb)-Zr is a strong endothermic reaction, which is different from the reaction of aZr þ hydride to a-Zr (weak endothermic). In this case, the onset point can be available as a transition temperature. The peak height of 1121 wppm is greater than that of 940 wppm, but the peak position is almost identical to that of 940 wppm indicating that these reactions are a eutectic reaction. The phase transition reaction of 731 wppm was found in the aZr þ hydride to a-Zr reaction and the OT of the aZr þ hydride to (aþb)-Zr reaction was around 553 C, moreover, there is also a hysteresis gap between phase transition in a-Zr þ hydride # (aþb)-Zr. However, the determination of OT is not straightforward and is rather subjective especially when the peak has a broad and low intensity. Some investigators used the PT [22,33e35] and the others utilized MST [16,18,21,23,25,36]. Khatamian et al. [34,37] proposed that the PT is the most appropriate temperature to determine phase transition because it agrees with the reference equilibrium data and neutron diffraction pattern. On the other hand, Pan and Puls [18] claimed that MST is the end point of dissolution because MST is in good agreement with their internal friction method, moreover, MST is recognized as a representable temperature because even very small hydrides are present at MST, the hydrogen value is under the error estimation. Namely, the TSSD is the end point
Fig. 2 e DSC curves and its time derivative of Zircaloy-4 with a hydrogen content of 288 wppm during cool-down.
Fig. 3 e DSC curves of phase transition in Zircaloy-4 during heat-up and cool-down.
of the hydrogen dissolution reaction and the TSSP is the initial point of the hydride precipitation. Consequently, in the present study, the TSS temperatures were determined by MST as a detectable transition temperature for both heating and cooling, which also agrees with the reference data.
Results and discussion Terminal solid solubility of hydrogen in Zircaloy-4 for dissolution and precipitation To investigate the wide range of TSS of hydrogen in Zircaloy-4, 40e731 wppm of hydrided specimen was used. Previous studies on the TSS of hydrogen in Zircaloy-4 were conducted using various methods [8,10,12,21,22,25,35]. The present data of TSSD and TSSP are plotted with references in Figs. 4 and 5, respectively. The curves from our experiment are.
Fig. 4 e TSSD of hydrogen in Zircaloy-4 with reference data.
16445
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
partly due to the different thermal history and measurement methods. Nevertheless, the present data are consistent with other TSSP data of 25e34 kJ/mol. The reference data are summarized in Table 2.
Effects of the maximum temperature on TSSP
Fig. 5 e TSSP of hydrogen in Zircaloy-4 with reference data.
TSSD of Zircaloy 4 : C ¼ 2:255 105 expð39101=RTÞ; TSSP of Zircaloy 4 : C ¼ 4:722 104 expð26843=RTÞ; where C is the hydrogen content (wppm), R is the gas constant and T is the absolute temperature. The activation energies of TSSD and TSSP are 39.101 and 26.843 kJ/mol, respectively. The present activation energy for dissolution is 39.101 kJ/mol in good agreement with the other TSSD data of 32e39 kJ/mol in Fig. 4. The measured data at 286 C is slightly lower than that of the other references. On the other hand, the presented TSSP in Fig. 5 is in good agreement with the results of the three references [10,35], however, our data have a slightly lower slope than that of Slattery [12] and McMinn et al. [21] and measured hydrogen content reported herein is greater than that of Slattery [12] and McMinn et al. [21] below 250 C. The peak temperature during the solubility measurement in McMinn et al. [21] is 400 C; therefore, this relative low temperature may affect the TSSP. The TSSP data of references are significantly scattered as shown in Fig. 5, which might be
Previous studies showed that increasing the maximum temperature and hold time at the maximum temperature have little effect on the TSSD temperature, whereas the TSSP temperature can be reduced [16,17,21], in other words, lowering the maximum temperature can increase the TSSP temperature. Fig. 6 shows the effect of the maximum temperature on TSSP temperature of 288 wppm specimen. The change in TSSP temperature is negligible when the maximum temperatures are 500 C and 550 C. However, when the maximum temperature is close to TSSD temperature of 434.56 C, TSSD temperature increases and simultaneously the peak height gradually lowers and the peak width broadens but the hatched area is similar to that of an integration of the truncated peak in Fig. 6. Pan et al. [17] defined this increased temperature as the TSSP2 temperature. They regarded the broad TSSP2 curve as a hydride growth process and the truncated TSSP peak as a nucleation stage of hydride formation in the cooling curve. The hatched area in Fig. 6 is closely related to the amount of hydrogen dissolution implying that hydride nucleation is faster than hydride growth. To determine the TSSP2 temperature, additional tests were conducted by lowering the maximum temperature to TSSD temperature. The temperature difference between the two temperatures is less than 4 C. The TSSP2 of hydrogen in Zircaloy-4 yields the following relationship. TSSP2 of Zircaloy 4 : C ¼ 8:612 105 exp 30583=RT
Fig. 7 shows the TSS of hydrogen in Zircaloy-4. There is a significant temperature gap between the TSSD and TSSP. The temperature gap between TSSD and TSSP tends to decrease with increasing hydrogen content. The gap between TSSD and TSSP is 41e85 C. The gap between TSSD and TSSP2 is around 44e58 C; however, there is no correlation with hydrogen content. On the other hand, the gap between TSSP and TSSP2
Table 2 e Reference data on terminal solid solubility of hydrogen in Zircaloy-4. Data
Method
Temperature range (K)
Aa
Q (kJ/mol)
Reference
TSSD
Diffusion Diffusion Dilatometry DSC DSC DSC DSC DSC Diffusion Dilatometry DSC DSC DSC DSC
533e798 533e755 506e714 423e598 493e717 533e783 533e783 560e818 533e700 461e645 423e543 443e643 533e733 480e777
9.9 104 6.6 104 1.076 105 1.06 105 5.25 104 2.63 105 1.72 105 2.255 105 3.1 104 1.066 105 1.387 105 4.01 104 5.72 104 4.722 104
34.541 32.196 35.515 35.990 32.117 39.310 36.308 39.101 25.279 32.088 34.469 27.336 27.287 26.843
Kearns [8] Kammenzind [10] Slattery [12] McMinn [21] Tang [25] Vizcaı´no [22] Vizcaı´no [35] This work Kammenzind [10] Slattery [12] McMinn [21] Tang [25] Vizcaı´no [35] This work
TSSP
a
A is the pre-exponential coefficient.
16446
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
that eases hydride precipitation, which is normally called the memory effect [41]. The hypothetical theory of TSS of hydrogen in zirconium alloys was developed by Puls [42e44]. An equilibrium solvus, Ceq, often called stress-free solvus is the state when no accommodation energy is required and can be expressed as followings: Ceq ¼ A expðQ=RTÞ
Fig. 6 e DSC curves with 288 wppm at the different maximum temperatures.
decreases with increasing hydrogen content. To determine the magnitude of the thermal hysteresis of hydrogen solubility, (TTSSDTTSSP)/TTSSP was proposed and reported that the ratio decreases with increasing temperature [36]. The ratio is dependent on the temperature and between 1.29 and 0.35 and the ratio between TSSD and TSSP2 is 0.08e0.17. The ratio clearly showed that the extent of hysteresis decreases in increasing hydrogen content. It is believed that when the maximum temperature reaches the TSSD temperature, the temperature gap between the TSSD and TSSP is reduced due to the presence of dislocations. It has been observed that precipitated hydrides make dislocations [38,39]. The elimination of dislocations at high temperature makes it difficult to precipitate the hydrides [14,17,21]. These dislocations can be removed at high temperatures; however, some dislocations still remain in the matrix at a medium temperature [40]. Consequently, it is believed that even if all of the hydrogen is completely dissolved, this dislocation can act as a pre-existing hydride site
where A is the pre-exponential constant and Q is the enthalpy for solution hydrogen. The disparity in TSS of hydrogen in Zircaloy-4 accounts for the misfit strain [45] energy loss due to the plastic deformation during cool-down. A hydride precipitation induces a positive misfit strain due to its larger volume than matrix. This misfit stain accompanies the elastic and plastic deformation. Strain energy stored during accommodation of hydride in the matrix is termed accommodation energy. During hydride nucleation, TSS solvus is governed by the elastic accommodation energy. Thus the hydride nucleation process, TSSP, is expressed as below: CTSSP ¼ Ceq exp wtel RT where wtel is the total elastic strain energy of matrix and hydride precipitation per mole H. On the other hand, the hydride growth solvus, TSSP2, is dependent on elasticplastic accommodation energy. The growth solvus is given by CTSSP2 ¼ Ceq exp wrel þ wp
RT
where wrel is the remaining elastic accommodation energy and wp is the plastic component of accommodation energy. A TSS determined on heat-up is balanced between wrel and wp. The dissolution solvus, TSSD, can be presented as follows: r wel wp CTSSD ¼ Ceq exp RT thus, the present activation energies for dissolution and growth will differ by 2wp . Therefore, dissipated plastic work determined experimentally corresponds to 4.259 kJ/mol.
Discussion on the solid solubility during thermal cycling
Fig. 7 e Terminal solid solubility of hydrogen in Zircaloy-4.
Recently, it was suggested that there is no hysteresis gap between the TSSD and TSSP if the hydrides are present at the maximum temperature [46], in other words, precipitation can occur immediately during subsequent cool-down. On the contrary, Pan et al. [17] showed that the TSSP temperature can be detected even if the maximum temperature is below the TSSD temperature. Fig. 8 shows the DSC cooling curve when the maximum temperature is equal to or less than the TSSD temperature. As shown in the figure, there is still a hysteresis gap when the maximum temperature is equal to the TSSD temperature. However, when the maximum temperature is less than the TSSD temperature, in other words, some hydrides in matrix are not dissolved, the cooling curve does not show a clear peak. In this case, it is ambiguous to define the point that represents the onset of hydride precipitation. Moreover, it is difficult to determine the peak position when the maximum temperature is less than 300 C due to the low sensitivity at low hydrogen solubility.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
Fig. 8 e DSC cooling curves of a 288 wppm specimen when the maximum temperature is less than the TSSD temperature.
Nevertheless, it is clear that the cooling DSC curve follows the intrinsic precipitation line regardless of the maximum temperature, which indicates that the hydride precipitation curve follows the TSSP2 line even if hydrides are present in the matrix. Another experiment was conducted to confirm the existence of a hysteresis gap when pre-existing hydrides were present during reheating. The specimen was cooled from 550 to 300, 350 and 400 C with a hold time of 5 min at the transition temperatures, and the specimens were then re-heated to 550 C. As shown in Fig. 9 there was a temperature gap between the reheating point and initiation point of hydrogen dissolution; however, the temperature gap was narrower than the typical temperature gap of hysteresis. The reheated DSC curve is also a weak endothermic reaction indicating that hydrogen dissolves slowly with an increasing temperature. Therefore, it might be concluded that when hydrides exist during cooling, a sudden temperature increase can induce a quasi-immediate dissolution of hydride and the previous thermal history has little effect on the TSSD temperature implying that hydrogen solubility on heating eventually follows the TSSD line.
Fig. 9 e DSC curves of a 442 wppm specimen during reheating and cooling.
16447
Therefore, it was proposed that there is a hysteresis gap on re-heating similar to a cool-down, and hydrides may not dissolve during the hysteresis gap (FeA in Fig. 10) [47]. However, based on the experimental results, the hypothetical hydrogen solubility path can be inferred as shown in Fig. 10. For example, when hydrogen content of a 210 wppm specimen is heated to over the dissolution temperature, the hydrogen solubility follows the TSSD until all hydrides are completely dissolved (OeAeBeC in Fig. 10). When the temperature begins to decrease, there is a temperature gap owing to plastic and elastic accommodation energy until the precipitation solvus reaches point D (CeBeD in Fig. 10), and the solubility then follows the TSSP2 (DeE in Fig. 10) during subsequent cool-down, but the temperature gap can be reduced (BeD1 in Fig. 10) depending on the peak temperature and holding time at the peak temperature. On the other hand, when the specimen is subjected to re-heating, the dissolution solvus can follow FeG instead of FeA due to the recovery of elastic deformation [14]. The proposed re-heating path is similar to synchrotron X-ray results when the specimen is thermally cycled [26]. Similarly, hydrides can be precipitated without a temperature gap on cooling if the hydrides are present in the matrix but the amount of precipitation would be negligible and the precipitation solvus would follow the TSSP2 line (HeIeF in Fig. 10).
Conclusions The TSS of hydrogen in Zircaloy-4 was studied by the DSC method with 40e731 wppm of hydrided specimen. The main findings were as follows: The solubility limit in alpha-Zircaloy-4 was near 731 wppm and the temperature of eutectoid reaction was around 552.5 C. The activation energy for dissolution and precipitation were 39.101 kJ/mol and 26.843 kJ/mol, respectively. The temperature gap between TSSD and TSSP lines was around 41e85 C. The TSSD temperature was unaffected by the previous thermal history but the TSSP temperature was influenced by
Fig. 10 e Schematic diagram of hydrogen solubility path in Zircaloy-4 during thermal cycling.
16448
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
the previous maximum temperature. The measured activation energy of the TSSP2 was 30.583 kJ/mol and the temperature gap between the TSSD and TSSP2 was 44e58 C. Hydrogen solubility during heating and cooling tries to follow an intrinsic dissolution and precipitation solvi, in other words, solubility path is dependent on the direction of heat flow.
Acknowledgments This work has been carried out under the Nuclear R&D Program supported by the Ministry of Science, ICT&Future Planning (NRF-2012M2A8A5025823).
references
ski A, Sobieszczyk S. Hydrogen-enhanced degradation [1] Zielin and oxide effects in zirconium alloys for nuclear applications. Int J Hydrogen Energy 2011;36:8619e29. [2] Kim JS, Yoon HK, Kook DH, Kim YS. A study on the initial characteristics of domestic spent nuclear fuels for long term dry storage. Nucl Eng Technol 2013;45:337e84. [3] Coleman CE, Hardie D. The hydrogen embrittlement of azirconiumda review. J Less Common Metals 1966;11:168e85. [4] Aomi M, Baba T, Miyashita T, Kamimura K, Yasuda T, Shinohara Y, et al. Evaluation of hydride reorientation behavior and mechanical properties for high-burnup fuelcladding tubes in interim dry storage. ASTM Spec Tech Publ 2009;1505:651e73. [5] Ivanova SV. Hydrogen effected defects evolution in zirconium items of light-water reactors. Int J Hydrogen Energy 2006;31:295e300. [6] Zhao C, Song X, Yang Y, Zhang B. Hydrogen absorption cracking of zirconium alloy in the application of nuclear industry. Int J Hydrogen Energy 2013;38:10903e11. [7] Sawatzky A. The diffusion and solubility of hydrogen in the alpha phase of Zircaloy-2. J Nucl Mater 1960;2:62e8. [8] Kearns JJ. Terminal solubility and partitioning of hydrogen in the alpha phase of zirconium, Zircaloy-2 and Zircaloy-4. J Nucl Mater 1967;22:292e303. [9] Sawatzky A, Wilkins BJS. Hydrogen solubility in zirconium alloys determined by thermal diffusion. J Nucl Mater 1967;22:304e10. [10] Kammenzind BG, Franklin DG, Peters HR, Duffin WJ. Hydrogen pickup and redistribution in alpha-annealed Zircaloy-4. ASTM Spec Tech Publ 1996;1295:338e69. [11] Erickson WH, Hardie D. The influence of alloying elements on the terminal solubility of hydrogen in a-zirconium. J Nucl Mater 1964;13:254e62. [12] Slattery GF. The terminal solubility of hydrogen in zirconium alloys between 30 and 400 C. J Inst Metals 1967;95:43e7. [13] Slattery GF. The terminal solubility of hydrogen in the zirconium/2 at% chromium/0.16 at% iron alloy. J Nucl Mater 1969;32:30e8. [14] Singh RN, Mukherjee S, Gupta A, Banerjee S. Terminal solid solubility of hydrogen in Zr-alloy pressure tube materials. J Alloys Compd 2005;389:102e12. [15] Giroldi JP, Vizcaı´no P, Flores AV, Banchik AD. Hydrogen terminal solid solubility determinations in Zr-2.5 Nb pressure tube microstructure in an extended concentration range. J Alloys Compd 2009;474:140e6.
[16] Khatamian D, Pan ZL, Puls MP, Cann CD. Hydrogen solubility limits in excel, an experimental zirconiumbased alloy. J Alloys Compd 1995;231:488e93. [17] Pan ZL, Ritchie IG, Puls MP. The terminal solid solubility of hydrogen and deuterium in Zr-2.5 Nb alloys. J Nucl Mater 1996;228:227e37. [18] Pan ZL, Puls MP. Precipitation and dissolution peaks of hydride in Zr-2.5 Nb during quasistatic thermal cycles. J Alloys Compd 2000;310:214e8. [19] Pan ZL, Puls MP, Ritchie IG. Measurement of hydrogen solubility during isothermal charging in a Zr alloy using an internal friction technique. J Alloys Compd 1994;211e212:245e8. [20] Root JH, Fong RWL. Neutron diffraction study of the precipitation and dissolution of hydrides in Zr-2.5Nb pressure tube material. J Nucl Mater 1996;232:75e85. [21] McMinn A, Darby EC, Schofield JS. Terminal solid solubility of hydrogen in zirconium alloys. ASTM Spec Tech Publ 2000;1354:173e95. [22] Vizcaıno P, Banchik AD, Abriata JP. Solubility of hydrogen in Zircaloy-4: irradiation induced increase and thermal recovery. J Nucl Mater 2002;304:96e106. [23] Une K, Ishimoto S. Terminal solid solubility of hydrogen in unalloyed zirconium by differential scanning calorimetry. J Nucl Sci Technol 2004;41:949e52. [24] Vizcaı´no P, Banchik A, Abriata J. Hydrogen in Zircaloy-4: effects of the neutron irradiation on the hydride formation. J Mater Sci 2007;42:6633e7. [25] Tang R, Yang X. Dissolution and precipitation behaviors of hydrides in N18, Zry-4 and M5 alloys. Int J Hydrogen Energy 2009;34:7269e74. [26] Zanellato O, Preuss M, Buffiere JY, Ribeiro F, Steuwer A, Desquines J, et al. Synchrotron diffraction study of dissolution and precipitation kinetics of hydrides in Zircaloy4. J Nucl Mater 2012;420:537e47. [27] Mallett MW, Albrecht WM. Low-pressure solubility and diffusion of hydrogen in zirconium. J Electrochem Soc 1957;104:142e6. [28] Douglas TB. The zirconiumehydrogen system: some thermodynamic properties from a heat content Study1. J Am Chem Soc 1958;80:5040e6. € [29] Ostberg G. Determination of hydride solubility in alpha phase zirconium, zircaloy-2 and zircaloy 4. J Nucl Mater 1962;5:208e15. [30] Westerman R. Charging Zircaloy-2 with hydrogen beyond the solubility limit. J Nucl Mater 1966;18:31e8. [31] Gulbransen EA, Andrew KF. Solubility and decomposition pressures of hydrogen in alpha-zirconium. J Metal Trans AIME 1955;203:136e44. [32] Zuzek E, Abriata JP, San-Martin A, Manchester FD. The HeZr (hydrogenezirconium) system. Bull Alloy Phase Diagr 1990;11:385e95. [33] Khatamian D, Ling VC. Hydrogen solubility limits in a- and bzirconium. J Alloys Compd 1997;253e254:162e6. [34] Khatamian D. DSC “peak temperature” vs “maximum slope temperature” in determining TSSD temperature. J Nucl Mater 2010;405:171e6. [35] Vizcaı´no P, Flores AV, Bozzano PB, Banchik AD, Versaci RA, Rios RO. Hydrogen solubility and microstructural changes in Zircaloy-4 due to neutron irradiation. ASTM Spec Tech Publ 2011;1529:754e83. [36] Une K, Ishimoto S. Dissolution and precipitation behavior of hydrides in Zircaloy-2 and high Fe Zircaloy. J Nucl Mater 2003;322:66e72. [37] Khatamian D, Root JH. Comparison of TSSD results obtained by differential scanning calorimetry and neutron diffraction. J Nucl Mater 2008;372:106e13.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
[38] Bailey JE. Electron microscope observations on the precipitation of zirconium hydride in zirconium. Acta Metall 1963;11:267e80. [39] Carpenter GJC, Watters JF, Gilbert RW. Dislocations generated by zirconium hydride precipitates in zirconium and some of its alloys. J Nucl Mater 1973;48:267e76. [40] Carpenter GJC, Watters JF. An in-situ study of the dissolution of g-zirconium hydride in zirconium. J Nucl Mater 1978;73:190e7. [41] Cameron DJ, Duncan RG. On the existence of a memory effect in hydride precipitation in cold-worked Zr-2.5% Nb. J Nucl Mater 1977;68:340e4. [42] Puls MP. The effects of misfit and external stresses on terminal solid solubility in hydride-forming metals. Acta Metall 1981;29:1961e8.
16449
[43] Puls MP. Elastic and plastic accommodation effects on metalehydride solubility. Acta Metall 1984;32:1259e69. [44] Puls MP. On the consequences of hydrogen supersaturation effects in Zr alloys to hydrogen ingress and delayed hydride cracking. J Nucl Mater 1989;165:128e41. [45] Carpenter GJC. The dilatational misfit of zirconium hydrides precipitated in zirconium. J Nucl Mater 1973;48:264e6. [46] Billone MC, Burtseva TA, Einziger RE. Ductile-to-brittle transition temperature for high-burnup cladding alloys exposed to simulated drying-storage conditions. J Nucl Mater 2013;433:431e48. [47] IAEA. Delayed hydride cracking in zirconium alloys in pressure tube nuclear reactors. IAEA-TECDOC-1410; 2004.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Effect of thermal history on the terminal solid solubility of hydrogen in Zircaloy-4 Ju-Seong Kim a,*, Yong-Soo Kim b a b
Korea Atomic Energy Research Institute, 989-111 Daedeokdaero, Yuseong-gu, Daejeon, 305-353, Republic of Korea Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul, 133-791, Republic of Korea
article info
abstract
Article history:
The terminal solid solubility (TSS) of hydrogen in zirconium alloys has a hysteresis. The
Received 16 June 2014
TSS of hydrogen in Zircaloy-4 during cooling and heating were studied using differential
Received in revised form
scanning calorimetry (DSC) with a hydrogen content of 40e731 wppm. A significant hys-
1 August 2014
teresis gap was observed between the TSS for dissolution (TSSD) and precipitation (TSSP). It
Accepted 10 August 2014
was confirmed that the hydrogen dissolution temperature was unaffected by the previous
Available online 2 September 2014
thermal history in comparison with the hydride precipitation temperature. The TSSP temperature increased with a decrease in the maximum temperature, but a significant
Keywords:
temperature gap remained even when the maximum temperature was equal to the TSSD
Terminal solid solubility
temperature. The terminal solid solubility of hydrogen in Zircaloy-4 can be represented by
Hydrogen
the following equations.
Zircaloy-4 Thermal history
TSSD: C ¼ 2.255 105 exp (39101/RT). TSSP: C ¼ 4.722 104 exp (26843/RT). TSSP2: C ¼ 8.612 105 exp (30583/RT).
Based on the experimental results hydrogen solubility path depending on the previous thermal history was proposed. Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Introduction Zirconium alloys have been used as reactor component materials owing to a low neutron absorption cross section, high corrosion resistance, and proper mechanical properties at high temperature. Hydrogen pick-up in zirconium
alloy is one of the most important issues in reactor operation [1]. Typical nuclear cladding discharged from the pressurized water reactor has 200e600 wppm of hydrogen depending on the burn-up [2]. Most of the hydrogen can dissolve in Zr-matrix at high temperatures. However, the brittle hydrides can be precipitated during the cool-down which can reduce the ductility of the cladding [3], and
* Corresponding author. Tel.: þ82 42 868 8689; fax: þ82 42 863 0565. E-mail addresses: [email protected], [email protected] (J.-S. Kim). http://dx.doi.org/10.1016/j.ijhydene.2014.08.018 0360-3199/Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
threaten the integrity of cladding such as hydride reorientation [4] or delayed hydride cracking (DHC) [5,6]. Interestingly the TSS of hydrogen in zirconium alloys for dissolution (TSSD) and precipitation (TSSP) have a hysteresis. This hysteresis can affect the DHC behavior and hydride re-orientation. Consequently, the terminal solid solubility (TSS) of hydrogen in zirconium and its alloys has been measured using a variety of techniques: diffusion equilibrium [7e10], dilatometry [11e15], internal friction and dynamic elastic modulus [16e19], small-angle neutron diffraction [20], differential scanning calorimetry (DSC) [15,16,21e25], and synchrotron X-ray diffraction [26]. Kearns [8] reviewed the previous investigations [7,11,27e30] and performed experiment with employing diffusion method to determine the TSS of hydrogen in zirconium alloys. His best fit of the data often used as a standard TSSD for zirconium and Zircaloy-2 and Zircaloy-4. The other methods such as internal friction, DSC and dilatometry are dynamic methods to measure the hydrogen thermal behaviors during the temperature transient. However, most of the works on the TSS of hydrogen in zirconium alloys are about the TSSD rather than the TSSP. In addition, there are no available data on TSSP2 of hydrogen in Zircaloy-4. The TSSD of hydrogen in zirconium alloy seems to be less sensitive to the previous thermal history, but the TSSP is strongly dependent on the thermal history. The extent of the hysteresis depends on the temperature history such as peak temperature, a holding time at the peak temperature and a cooling rate; therefore, the published data on TSSP is considerable scatter. The objectives of this study are to confirm the TSS of hydrogen in Zircaloy-4 and evaluate the maximum temperature effect on the TSSP. The DSC method was employed to measure the TSS of hydrogen in Zircaloy-4 owing to simplicity and reproducibility.
Experimental Materials
16443
Differential scanning calorimetry Basically, DSC measures the changes in the difference in the heat exchange between the sample and the reference. The change in the heat flow signal indicates an exothermic or endothermic reaction when a phase transformation occurs. Fig. 1 shows a typical DSC heat flow and its derivative heat flow curves of hydrided Zircaloy-4 during heating. As the temperature increases, the hydrogen begins to dissolve and the heat flow curve decreases steadily indicating that the hydrogen dissolution process is an endothermic reaction. In this work, the TSS of hydrogen in Zircaloy-4 was measured using a heat flux DSC (Netzsch 200 F3). The calibrations were conducted prior to the experiment using metal standard samples (In, Sn and Zn). DSC measurements were carried out in purified N2 at a flow rate of 50 cm3/min. Generally, DSC is more sensitive at higher heating/cooling rates but such rates tend to slightly increase the TSSD temperature and decrease the TSSP temperature. It is known that the effect of heating rate on the dissolution of hydrogen is negligible [21,22], whereas the effect of cooling rate on the precipitation is more susceptible. Considering these effects, TSS measurements were conducted at cooling and heating rate of 20 C/min. Most samples were heated to peak temperature of 550 C. The dwell time at the peak temperature was 5 min and then cooled to 40 C. Some high hydrogen containing specimens were heated to 580 C. Each specimen was experienced four identical thermal cycles. All the data were used to calculate the mean TSS values excluding that of first heating cycle. Additional tests were conducted to investigate the hydrogen behavior during thermal cycling and effect of the maximum temperature.
Determination of the TSSD and TSSP temperatures There are three important temperatures for determining the TSS: peak temperature, completion temperature, and maximum slope temperature. The peak temperature (PT) is the maximum or minimum point in the heat flow curve and the maximum slope temperature (MST) is the point of maximum deviation of the heat flow. The completion
Cold worked stress-relieved (CWSR) Zircaloy-4 finally annealed at 470 C for 5 h was used for the present TSS measurements. Table 1 shows the chemical composition of Zircaloy-4. Hydrogen charging was conducted at 400 C by a Sieverts-type apparatus. After hydrogen charging, the heat treatment was performed at 400 C for 5e10 h to ensure a uniform distribution of hydride. As a result, 40e1121 wppm of hydrided specimens were obtained. All of the hydrogen concentrations were analyzed by a hydrogen determinator (LECO RH-404), which uses an inert gas fusion method. The error of the determinator is less than 5 wppm.
Table 1 e Chemical composition of Zircaloy-4 (unit: weight%). Element
Sn
Fe
Cr
C
O
Si
Zr
Weight
1.32
0.21
0.11
0.013
0.13
0.0092
Balance
Fig. 1 e DSC curve and its time derivative of Zircaloy-4 specimen with a hydrogen content of 288 wppm during heat-up.
16444
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
temperature (CT) or onset temperature (OT) is the point of intersection of the two tangent lines, which is the interpolated baseline and the dissolution or precipitation discontinuity line. However, there are no rules to determine the TSS temperature. In the present study, the TSSD temperature was determined by MST as a detectable transition temperature for both heating and cooling. The typical DSC heating curve shows a broad peak, for example, the temperature difference between PT and MST is around 30 C as shown in Fig. 1. On the other hand, the typical DSC curve during cooling has a sharp peak with respect to heating curves (Fig. 2) but the cooling curve also has a similar broad peak when the maximum temperature is reduced. The solid solubility limit of hydrogen in a-Zr is known to be around 700e713 wppm [31,32] and the temperature of the eutectoid reaction of a-Zr þ hydride to (aþb)-Zr is 550 ± 3 C [32]. Fig. 3 shows a DSC curve of this phase transition. The reaction of a-Zr þ hydride to (aþb)-Zr is a strong endothermic reaction, which is different from the reaction of aZr þ hydride to a-Zr (weak endothermic). In this case, the onset point can be available as a transition temperature. The peak height of 1121 wppm is greater than that of 940 wppm, but the peak position is almost identical to that of 940 wppm indicating that these reactions are a eutectic reaction. The phase transition reaction of 731 wppm was found in the aZr þ hydride to a-Zr reaction and the OT of the aZr þ hydride to (aþb)-Zr reaction was around 553 C, moreover, there is also a hysteresis gap between phase transition in a-Zr þ hydride # (aþb)-Zr. However, the determination of OT is not straightforward and is rather subjective especially when the peak has a broad and low intensity. Some investigators used the PT [22,33e35] and the others utilized MST [16,18,21,23,25,36]. Khatamian et al. [34,37] proposed that the PT is the most appropriate temperature to determine phase transition because it agrees with the reference equilibrium data and neutron diffraction pattern. On the other hand, Pan and Puls [18] claimed that MST is the end point of dissolution because MST is in good agreement with their internal friction method, moreover, MST is recognized as a representable temperature because even very small hydrides are present at MST, the hydrogen value is under the error estimation. Namely, the TSSD is the end point
Fig. 2 e DSC curves and its time derivative of Zircaloy-4 with a hydrogen content of 288 wppm during cool-down.
Fig. 3 e DSC curves of phase transition in Zircaloy-4 during heat-up and cool-down.
of the hydrogen dissolution reaction and the TSSP is the initial point of the hydride precipitation. Consequently, in the present study, the TSS temperatures were determined by MST as a detectable transition temperature for both heating and cooling, which also agrees with the reference data.
Results and discussion Terminal solid solubility of hydrogen in Zircaloy-4 for dissolution and precipitation To investigate the wide range of TSS of hydrogen in Zircaloy-4, 40e731 wppm of hydrided specimen was used. Previous studies on the TSS of hydrogen in Zircaloy-4 were conducted using various methods [8,10,12,21,22,25,35]. The present data of TSSD and TSSP are plotted with references in Figs. 4 and 5, respectively. The curves from our experiment are.
Fig. 4 e TSSD of hydrogen in Zircaloy-4 with reference data.
16445
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
partly due to the different thermal history and measurement methods. Nevertheless, the present data are consistent with other TSSP data of 25e34 kJ/mol. The reference data are summarized in Table 2.
Effects of the maximum temperature on TSSP
Fig. 5 e TSSP of hydrogen in Zircaloy-4 with reference data.
TSSD of Zircaloy 4 : C ¼ 2:255 105 expð39101=RTÞ; TSSP of Zircaloy 4 : C ¼ 4:722 104 expð26843=RTÞ; where C is the hydrogen content (wppm), R is the gas constant and T is the absolute temperature. The activation energies of TSSD and TSSP are 39.101 and 26.843 kJ/mol, respectively. The present activation energy for dissolution is 39.101 kJ/mol in good agreement with the other TSSD data of 32e39 kJ/mol in Fig. 4. The measured data at 286 C is slightly lower than that of the other references. On the other hand, the presented TSSP in Fig. 5 is in good agreement with the results of the three references [10,35], however, our data have a slightly lower slope than that of Slattery [12] and McMinn et al. [21] and measured hydrogen content reported herein is greater than that of Slattery [12] and McMinn et al. [21] below 250 C. The peak temperature during the solubility measurement in McMinn et al. [21] is 400 C; therefore, this relative low temperature may affect the TSSP. The TSSP data of references are significantly scattered as shown in Fig. 5, which might be
Previous studies showed that increasing the maximum temperature and hold time at the maximum temperature have little effect on the TSSD temperature, whereas the TSSP temperature can be reduced [16,17,21], in other words, lowering the maximum temperature can increase the TSSP temperature. Fig. 6 shows the effect of the maximum temperature on TSSP temperature of 288 wppm specimen. The change in TSSP temperature is negligible when the maximum temperatures are 500 C and 550 C. However, when the maximum temperature is close to TSSD temperature of 434.56 C, TSSD temperature increases and simultaneously the peak height gradually lowers and the peak width broadens but the hatched area is similar to that of an integration of the truncated peak in Fig. 6. Pan et al. [17] defined this increased temperature as the TSSP2 temperature. They regarded the broad TSSP2 curve as a hydride growth process and the truncated TSSP peak as a nucleation stage of hydride formation in the cooling curve. The hatched area in Fig. 6 is closely related to the amount of hydrogen dissolution implying that hydride nucleation is faster than hydride growth. To determine the TSSP2 temperature, additional tests were conducted by lowering the maximum temperature to TSSD temperature. The temperature difference between the two temperatures is less than 4 C. The TSSP2 of hydrogen in Zircaloy-4 yields the following relationship. TSSP2 of Zircaloy 4 : C ¼ 8:612 105 exp 30583=RT
Fig. 7 shows the TSS of hydrogen in Zircaloy-4. There is a significant temperature gap between the TSSD and TSSP. The temperature gap between TSSD and TSSP tends to decrease with increasing hydrogen content. The gap between TSSD and TSSP is 41e85 C. The gap between TSSD and TSSP2 is around 44e58 C; however, there is no correlation with hydrogen content. On the other hand, the gap between TSSP and TSSP2
Table 2 e Reference data on terminal solid solubility of hydrogen in Zircaloy-4. Data
Method
Temperature range (K)
Aa
Q (kJ/mol)
Reference
TSSD
Diffusion Diffusion Dilatometry DSC DSC DSC DSC DSC Diffusion Dilatometry DSC DSC DSC DSC
533e798 533e755 506e714 423e598 493e717 533e783 533e783 560e818 533e700 461e645 423e543 443e643 533e733 480e777
9.9 104 6.6 104 1.076 105 1.06 105 5.25 104 2.63 105 1.72 105 2.255 105 3.1 104 1.066 105 1.387 105 4.01 104 5.72 104 4.722 104
34.541 32.196 35.515 35.990 32.117 39.310 36.308 39.101 25.279 32.088 34.469 27.336 27.287 26.843
Kearns [8] Kammenzind [10] Slattery [12] McMinn [21] Tang [25] Vizcaı´no [22] Vizcaı´no [35] This work Kammenzind [10] Slattery [12] McMinn [21] Tang [25] Vizcaı´no [35] This work
TSSP
a
A is the pre-exponential coefficient.
16446
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
that eases hydride precipitation, which is normally called the memory effect [41]. The hypothetical theory of TSS of hydrogen in zirconium alloys was developed by Puls [42e44]. An equilibrium solvus, Ceq, often called stress-free solvus is the state when no accommodation energy is required and can be expressed as followings: Ceq ¼ A expðQ=RTÞ
Fig. 6 e DSC curves with 288 wppm at the different maximum temperatures.
decreases with increasing hydrogen content. To determine the magnitude of the thermal hysteresis of hydrogen solubility, (TTSSDTTSSP)/TTSSP was proposed and reported that the ratio decreases with increasing temperature [36]. The ratio is dependent on the temperature and between 1.29 and 0.35 and the ratio between TSSD and TSSP2 is 0.08e0.17. The ratio clearly showed that the extent of hysteresis decreases in increasing hydrogen content. It is believed that when the maximum temperature reaches the TSSD temperature, the temperature gap between the TSSD and TSSP is reduced due to the presence of dislocations. It has been observed that precipitated hydrides make dislocations [38,39]. The elimination of dislocations at high temperature makes it difficult to precipitate the hydrides [14,17,21]. These dislocations can be removed at high temperatures; however, some dislocations still remain in the matrix at a medium temperature [40]. Consequently, it is believed that even if all of the hydrogen is completely dissolved, this dislocation can act as a pre-existing hydride site
where A is the pre-exponential constant and Q is the enthalpy for solution hydrogen. The disparity in TSS of hydrogen in Zircaloy-4 accounts for the misfit strain [45] energy loss due to the plastic deformation during cool-down. A hydride precipitation induces a positive misfit strain due to its larger volume than matrix. This misfit stain accompanies the elastic and plastic deformation. Strain energy stored during accommodation of hydride in the matrix is termed accommodation energy. During hydride nucleation, TSS solvus is governed by the elastic accommodation energy. Thus the hydride nucleation process, TSSP, is expressed as below: CTSSP ¼ Ceq exp wtel RT where wtel is the total elastic strain energy of matrix and hydride precipitation per mole H. On the other hand, the hydride growth solvus, TSSP2, is dependent on elasticplastic accommodation energy. The growth solvus is given by CTSSP2 ¼ Ceq exp wrel þ wp
RT
where wrel is the remaining elastic accommodation energy and wp is the plastic component of accommodation energy. A TSS determined on heat-up is balanced between wrel and wp. The dissolution solvus, TSSD, can be presented as follows: r wel wp CTSSD ¼ Ceq exp RT thus, the present activation energies for dissolution and growth will differ by 2wp . Therefore, dissipated plastic work determined experimentally corresponds to 4.259 kJ/mol.
Discussion on the solid solubility during thermal cycling
Fig. 7 e Terminal solid solubility of hydrogen in Zircaloy-4.
Recently, it was suggested that there is no hysteresis gap between the TSSD and TSSP if the hydrides are present at the maximum temperature [46], in other words, precipitation can occur immediately during subsequent cool-down. On the contrary, Pan et al. [17] showed that the TSSP temperature can be detected even if the maximum temperature is below the TSSD temperature. Fig. 8 shows the DSC cooling curve when the maximum temperature is equal to or less than the TSSD temperature. As shown in the figure, there is still a hysteresis gap when the maximum temperature is equal to the TSSD temperature. However, when the maximum temperature is less than the TSSD temperature, in other words, some hydrides in matrix are not dissolved, the cooling curve does not show a clear peak. In this case, it is ambiguous to define the point that represents the onset of hydride precipitation. Moreover, it is difficult to determine the peak position when the maximum temperature is less than 300 C due to the low sensitivity at low hydrogen solubility.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
Fig. 8 e DSC cooling curves of a 288 wppm specimen when the maximum temperature is less than the TSSD temperature.
Nevertheless, it is clear that the cooling DSC curve follows the intrinsic precipitation line regardless of the maximum temperature, which indicates that the hydride precipitation curve follows the TSSP2 line even if hydrides are present in the matrix. Another experiment was conducted to confirm the existence of a hysteresis gap when pre-existing hydrides were present during reheating. The specimen was cooled from 550 to 300, 350 and 400 C with a hold time of 5 min at the transition temperatures, and the specimens were then re-heated to 550 C. As shown in Fig. 9 there was a temperature gap between the reheating point and initiation point of hydrogen dissolution; however, the temperature gap was narrower than the typical temperature gap of hysteresis. The reheated DSC curve is also a weak endothermic reaction indicating that hydrogen dissolves slowly with an increasing temperature. Therefore, it might be concluded that when hydrides exist during cooling, a sudden temperature increase can induce a quasi-immediate dissolution of hydride and the previous thermal history has little effect on the TSSD temperature implying that hydrogen solubility on heating eventually follows the TSSD line.
Fig. 9 e DSC curves of a 442 wppm specimen during reheating and cooling.
16447
Therefore, it was proposed that there is a hysteresis gap on re-heating similar to a cool-down, and hydrides may not dissolve during the hysteresis gap (FeA in Fig. 10) [47]. However, based on the experimental results, the hypothetical hydrogen solubility path can be inferred as shown in Fig. 10. For example, when hydrogen content of a 210 wppm specimen is heated to over the dissolution temperature, the hydrogen solubility follows the TSSD until all hydrides are completely dissolved (OeAeBeC in Fig. 10). When the temperature begins to decrease, there is a temperature gap owing to plastic and elastic accommodation energy until the precipitation solvus reaches point D (CeBeD in Fig. 10), and the solubility then follows the TSSP2 (DeE in Fig. 10) during subsequent cool-down, but the temperature gap can be reduced (BeD1 in Fig. 10) depending on the peak temperature and holding time at the peak temperature. On the other hand, when the specimen is subjected to re-heating, the dissolution solvus can follow FeG instead of FeA due to the recovery of elastic deformation [14]. The proposed re-heating path is similar to synchrotron X-ray results when the specimen is thermally cycled [26]. Similarly, hydrides can be precipitated without a temperature gap on cooling if the hydrides are present in the matrix but the amount of precipitation would be negligible and the precipitation solvus would follow the TSSP2 line (HeIeF in Fig. 10).
Conclusions The TSS of hydrogen in Zircaloy-4 was studied by the DSC method with 40e731 wppm of hydrided specimen. The main findings were as follows: The solubility limit in alpha-Zircaloy-4 was near 731 wppm and the temperature of eutectoid reaction was around 552.5 C. The activation energy for dissolution and precipitation were 39.101 kJ/mol and 26.843 kJ/mol, respectively. The temperature gap between TSSD and TSSP lines was around 41e85 C. The TSSD temperature was unaffected by the previous thermal history but the TSSP temperature was influenced by
Fig. 10 e Schematic diagram of hydrogen solubility path in Zircaloy-4 during thermal cycling.
16448
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
the previous maximum temperature. The measured activation energy of the TSSP2 was 30.583 kJ/mol and the temperature gap between the TSSD and TSSP2 was 44e58 C. Hydrogen solubility during heating and cooling tries to follow an intrinsic dissolution and precipitation solvi, in other words, solubility path is dependent on the direction of heat flow.
Acknowledgments This work has been carried out under the Nuclear R&D Program supported by the Ministry of Science, ICT&Future Planning (NRF-2012M2A8A5025823).
references
ski A, Sobieszczyk S. Hydrogen-enhanced degradation [1] Zielin and oxide effects in zirconium alloys for nuclear applications. Int J Hydrogen Energy 2011;36:8619e29. [2] Kim JS, Yoon HK, Kook DH, Kim YS. A study on the initial characteristics of domestic spent nuclear fuels for long term dry storage. Nucl Eng Technol 2013;45:337e84. [3] Coleman CE, Hardie D. The hydrogen embrittlement of azirconiumda review. J Less Common Metals 1966;11:168e85. [4] Aomi M, Baba T, Miyashita T, Kamimura K, Yasuda T, Shinohara Y, et al. Evaluation of hydride reorientation behavior and mechanical properties for high-burnup fuelcladding tubes in interim dry storage. ASTM Spec Tech Publ 2009;1505:651e73. [5] Ivanova SV. Hydrogen effected defects evolution in zirconium items of light-water reactors. Int J Hydrogen Energy 2006;31:295e300. [6] Zhao C, Song X, Yang Y, Zhang B. Hydrogen absorption cracking of zirconium alloy in the application of nuclear industry. Int J Hydrogen Energy 2013;38:10903e11. [7] Sawatzky A. The diffusion and solubility of hydrogen in the alpha phase of Zircaloy-2. J Nucl Mater 1960;2:62e8. [8] Kearns JJ. Terminal solubility and partitioning of hydrogen in the alpha phase of zirconium, Zircaloy-2 and Zircaloy-4. J Nucl Mater 1967;22:292e303. [9] Sawatzky A, Wilkins BJS. Hydrogen solubility in zirconium alloys determined by thermal diffusion. J Nucl Mater 1967;22:304e10. [10] Kammenzind BG, Franklin DG, Peters HR, Duffin WJ. Hydrogen pickup and redistribution in alpha-annealed Zircaloy-4. ASTM Spec Tech Publ 1996;1295:338e69. [11] Erickson WH, Hardie D. The influence of alloying elements on the terminal solubility of hydrogen in a-zirconium. J Nucl Mater 1964;13:254e62. [12] Slattery GF. The terminal solubility of hydrogen in zirconium alloys between 30 and 400 C. J Inst Metals 1967;95:43e7. [13] Slattery GF. The terminal solubility of hydrogen in the zirconium/2 at% chromium/0.16 at% iron alloy. J Nucl Mater 1969;32:30e8. [14] Singh RN, Mukherjee S, Gupta A, Banerjee S. Terminal solid solubility of hydrogen in Zr-alloy pressure tube materials. J Alloys Compd 2005;389:102e12. [15] Giroldi JP, Vizcaı´no P, Flores AV, Banchik AD. Hydrogen terminal solid solubility determinations in Zr-2.5 Nb pressure tube microstructure in an extended concentration range. J Alloys Compd 2009;474:140e6.
[16] Khatamian D, Pan ZL, Puls MP, Cann CD. Hydrogen solubility limits in excel, an experimental zirconiumbased alloy. J Alloys Compd 1995;231:488e93. [17] Pan ZL, Ritchie IG, Puls MP. The terminal solid solubility of hydrogen and deuterium in Zr-2.5 Nb alloys. J Nucl Mater 1996;228:227e37. [18] Pan ZL, Puls MP. Precipitation and dissolution peaks of hydride in Zr-2.5 Nb during quasistatic thermal cycles. J Alloys Compd 2000;310:214e8. [19] Pan ZL, Puls MP, Ritchie IG. Measurement of hydrogen solubility during isothermal charging in a Zr alloy using an internal friction technique. J Alloys Compd 1994;211e212:245e8. [20] Root JH, Fong RWL. Neutron diffraction study of the precipitation and dissolution of hydrides in Zr-2.5Nb pressure tube material. J Nucl Mater 1996;232:75e85. [21] McMinn A, Darby EC, Schofield JS. Terminal solid solubility of hydrogen in zirconium alloys. ASTM Spec Tech Publ 2000;1354:173e95. [22] Vizcaıno P, Banchik AD, Abriata JP. Solubility of hydrogen in Zircaloy-4: irradiation induced increase and thermal recovery. J Nucl Mater 2002;304:96e106. [23] Une K, Ishimoto S. Terminal solid solubility of hydrogen in unalloyed zirconium by differential scanning calorimetry. J Nucl Sci Technol 2004;41:949e52. [24] Vizcaı´no P, Banchik A, Abriata J. Hydrogen in Zircaloy-4: effects of the neutron irradiation on the hydride formation. J Mater Sci 2007;42:6633e7. [25] Tang R, Yang X. Dissolution and precipitation behaviors of hydrides in N18, Zry-4 and M5 alloys. Int J Hydrogen Energy 2009;34:7269e74. [26] Zanellato O, Preuss M, Buffiere JY, Ribeiro F, Steuwer A, Desquines J, et al. Synchrotron diffraction study of dissolution and precipitation kinetics of hydrides in Zircaloy4. J Nucl Mater 2012;420:537e47. [27] Mallett MW, Albrecht WM. Low-pressure solubility and diffusion of hydrogen in zirconium. J Electrochem Soc 1957;104:142e6. [28] Douglas TB. The zirconiumehydrogen system: some thermodynamic properties from a heat content Study1. J Am Chem Soc 1958;80:5040e6. € [29] Ostberg G. Determination of hydride solubility in alpha phase zirconium, zircaloy-2 and zircaloy 4. J Nucl Mater 1962;5:208e15. [30] Westerman R. Charging Zircaloy-2 with hydrogen beyond the solubility limit. J Nucl Mater 1966;18:31e8. [31] Gulbransen EA, Andrew KF. Solubility and decomposition pressures of hydrogen in alpha-zirconium. J Metal Trans AIME 1955;203:136e44. [32] Zuzek E, Abriata JP, San-Martin A, Manchester FD. The HeZr (hydrogenezirconium) system. Bull Alloy Phase Diagr 1990;11:385e95. [33] Khatamian D, Ling VC. Hydrogen solubility limits in a- and bzirconium. J Alloys Compd 1997;253e254:162e6. [34] Khatamian D. DSC “peak temperature” vs “maximum slope temperature” in determining TSSD temperature. J Nucl Mater 2010;405:171e6. [35] Vizcaı´no P, Flores AV, Bozzano PB, Banchik AD, Versaci RA, Rios RO. Hydrogen solubility and microstructural changes in Zircaloy-4 due to neutron irradiation. ASTM Spec Tech Publ 2011;1529:754e83. [36] Une K, Ishimoto S. Dissolution and precipitation behavior of hydrides in Zircaloy-2 and high Fe Zircaloy. J Nucl Mater 2003;322:66e72. [37] Khatamian D, Root JH. Comparison of TSSD results obtained by differential scanning calorimetry and neutron diffraction. J Nucl Mater 2008;372:106e13.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 4 4 2 e1 6 4 4 9
[38] Bailey JE. Electron microscope observations on the precipitation of zirconium hydride in zirconium. Acta Metall 1963;11:267e80. [39] Carpenter GJC, Watters JF, Gilbert RW. Dislocations generated by zirconium hydride precipitates in zirconium and some of its alloys. J Nucl Mater 1973;48:267e76. [40] Carpenter GJC, Watters JF. An in-situ study of the dissolution of g-zirconium hydride in zirconium. J Nucl Mater 1978;73:190e7. [41] Cameron DJ, Duncan RG. On the existence of a memory effect in hydride precipitation in cold-worked Zr-2.5% Nb. J Nucl Mater 1977;68:340e4. [42] Puls MP. The effects of misfit and external stresses on terminal solid solubility in hydride-forming metals. Acta Metall 1981;29:1961e8.
16449
[43] Puls MP. Elastic and plastic accommodation effects on metalehydride solubility. Acta Metall 1984;32:1259e69. [44] Puls MP. On the consequences of hydrogen supersaturation effects in Zr alloys to hydrogen ingress and delayed hydride cracking. J Nucl Mater 1989;165:128e41. [45] Carpenter GJC. The dilatational misfit of zirconium hydrides precipitated in zirconium. J Nucl Mater 1973;48:264e6. [46] Billone MC, Burtseva TA, Einziger RE. Ductile-to-brittle transition temperature for high-burnup cladding alloys exposed to simulated drying-storage conditions. J Nucl Mater 2013;433:431e48. [47] IAEA. Delayed hydride cracking in zirconium alloys in pressure tube nuclear reactors. IAEA-TECDOC-1410; 2004.