The possibility to expose as-cast alloys to the air before first hydrogenation could greatly reduce the synthesis complexity and cost. Therefore, we investigated the effect of air exposure on our as-cast alloys by preparing two identical samples of composition TiFe+ 12 wt.% (Zr+2Mn). One sample was processed in air, the other one in an argon-filled glove box. The sample labelled Ar was crushed into a glove box and never exposed to the air. Sample labelled Air was crushed and handled in air. The first hydrogenation was measured under an initial hydrogen pressure of 2 MPa at room temperature. Activation curves are shown in Fig 3.2.9. Remarkably, the air exposed sample has the same activation kinetics as the sample handled under argon. These results indicate that the TiFe+ 12 wt.% (Zr+2Mn) alloy could be handled in air without detrimental effects in the activation behaviour.
Improving first hydrogenation kinetics by co Id rolling
The first hydrogenation kinetics shown in Fig 3.2.7 indicates that an incubation time is present for x=2 and x=4. In a previous investigation, it was shown that cold rolling is very effective to enhance the first hydrogenation of LaNis alloy . This led to the use of cold rolling on the as-cast material to see if there is any first hydrogenation kinetic improvement. Cold rolling was done in air for one and five rolling passes. Fig 3.2.11 shows the first hydrogenation of TiFe+x wt.% (Zr+2Mn) (x=2(a), 4(b), 8(c), 12(d)) alloys in as cast and after cold rolling. For x=8 and x= 12 the curves levelled off after two hours. In order to have a better view of the kinetics we only plotted the first two hours. It is seen clearly that cold rolling has a clear beneficial effect only for the x=4 where rolling reduces the incubation time. In the case of x=8 the effect was more limited. Thus cold rolling could reduce incubation time but the main factor to enhance first hydrogenation is by adding (Zr+2Mn) alloy.
Microstructure evolution, phase transformation and hydrogenation performance of TiFe alloy doped with x wt.% (Zr+2V)
It was recently reported that the first hydrogenation of TiFe alloy could be improved by adding Zr [49, 71] and V  . This indicates that zirconium and vanadium play an important role on improving the first hydrogenation properties of TiFe alloy. The positive effect of zirconium and vanadium leads to the investigation of (Zr+2V) as an additive to TiFe alloy. In this chapter the effects of different (Zr+2V) content on the microstructure, hydrogen storage properties and air resistance of TiFe alloy are reported.
Results and discussion
Fig 3.3.1 shows backscattered electron micrographs of TiFe+x wt.% (Zr+2V) (x=2, 4, 8, 12) alloys prepared by arc melting. They show different microstructure which includes bright and dark phases. For small amount of (Zr+2V) (x=2), it can be seen that small bright areas increase with incipient dendritic structure. For x=4, the bright areas are larger and the dendritic morphology more apparent. For x=8 the bright areas increase and the dendrites are weil developed. When x=12 the bright surface is qui te important. Using image J software, the proportion of bright and dark phase areas is measured. These values are plotted in Fig 3.3.2.
From Fig 3.3.2 it is seen clearly that the bright phase area increases linearly with x.
The fact that the ordinate at the origin is not zero and the slope is not 1 means that the bright phase composition is not pure (Zr+2V) but instead contains Ti and Fe elements.
In order to confirm this, the chemical composition of each phase was analyzed by EDX.
Fig 3.3.3 shows a higher magnification backscattered electron micrographs of TiFe+x wt.% (Zr+2V) alloy. A quantitative analysis of the dark phase (point 1) and bright phase (point 2) was performed and the proportions of elements are reported in Table 3.3.2 and 3.3.3, respectively. Analysis of the dark phase (Table 3.3.2) indicates that for x=2, 4, and 8, zirconium is totally absent, the iron abundance is constant while titanium abundance slightly decreases. The decreasing of titanium abundance is due to substitution by vanadium. From the elemental abundances, we could conclude that the dark phase has a composition TiyFeV1_y. However, for x=12 the situation is more complicated. Here, iron abundance is more than 50 atomic % and zirconium is present at a level of 3 atomic %.
Inspection of Table 3.3.3 shows that the titanium abundance is constant over the whole range of x values. However, contrary to the dark phase, it is the iron that is ?substituted by vanadium and zircon ium wh en x increases. In fact the ratio of ZrN is almost 1/2 which is the ratio of (Zr+2V). It is clear that the chemical composition in the dark and bright phases does not change in the same fashion wh en x increases.
The XRD patterns of the TiFe+x wt.% (Zr+2V) (x=2, 4, 8, 12) alloys in as-cast state are presented in Fig 3.3.4. It can clearly be seen that ail alloys prepared have TiFe phase. However the secondary phase on ly can be seen for x=8 and 12. Due to the small intensities of these peaks, indexation is difficul t. By inspection, these small peaks can match the Fe2 Ti type crystal structure.
In order to further identify the secondary phase, the synchrotron pattern is shown in Fig 3.3.5. From this figure, it can be seen that TiFe phase is very obvious and shows (100) and (110) lattice plane. In addition, sorne rings and bright points also can be found at both sides of (110) lattice plane. Actually this suggests the presence of Fe2 Ti-like phase.
Combining XRD and synchrotron, the secondary phase can be speculated Fe2Ti-like phase. Using Rietveld refinement method, these small peaks could be fitted with a Fe2Ti phase (Space group: P63/mmc, structure type MgZn2) . This phase was also seen in other Ti-Fe-Zr compounds . Rietveld refinement was performed on all patterns. For the bright phase structure, the chemical composition was fixed as the one determined from EDX measurements. Fig 3.3.6 shows the Rietveld refinement for x= 12 alloy. The refinement is good except for a possible small peak at around 41.5°.
Despite the smallness of the peaks, the Fe2 Ti-like phase abundance was determined to be 57 wt.% which is very close to the value determined from SEM micrographs.Likewise, the Fe2 Ti-like phase abundance in the x=8 pattern was given by Rietveld refinement to be 39 wt.% which again, is close to the SEM value.
Fig 3.3.7 shows the first hydrogenation of TiFe + x wt.% (Zr+2V) (x=2, 4, 8, 12) alloys measured under an initial hydrogen pressure of 2 MPa at room temperature. It can be seen that x=2 can ‘t absorb any hydrogen ev en after 25 h. There is around 4.5 h of incubation time for x=4. For x=8 and 12, the alloys show no incubation time and directly absorb hydrogen with fast kinetics. Also, the first hydrogen absorption capacity increases with x and the maximum capacity is around 1.65 wt.% (x=12).
As seen from the SEM investigation and powder diffraction patterns, amount ofbright phase increases with x. Therefore, the conclusion is that enhancement of first hydrogenation kinetic is proportional to the amount of bright phase. Furthermore, taking into account that the hydrogen capacity of pure Ti Fe is 1.86 wt.% and that, for x=12, the amount of TiFe phase is around 50% as determined by SEM and Rietveld refinement, this phase could only contribute to less than 1 wt.% to the hydrogen capacity. This means that the secondary phase Fe2 Ti-like also absorbs hydrogen with a capacity similar to the TiFe phase. Moreover, there is not any change of slope or kink in the absorption curve for x=8 and 12. This means that the dark and bright phases are acting as a single phase during hydrogenation. A possible mechanism could be that the bright phase acts as a gateway for the hydrogen. Such mechanism was also proposed for a closely related system (Ti-Fe-Zr) [49,70,71 , 76].
In order to know the mechanism of the first hydrogenation kinetics, the experimental data of different first hydrogenation curves was analyzed by using the rate limiting step model equation (Nucleation-growth-impingement model (JMA) [16- 18], Contracting volume model (CV) [19-21] and Ginstling-Brounshtein model (GB) [22, 23]) are reported in Table 1.3. The left side of the equation is a function of the reaction ‘s completion ratio (a). In these equations, a is the reaction rate (a = %Hab/%Hmax) , t is reaction time; k is kinetic rate constant. In order to find the correct rate Iimiting step, the left side of these equations as a function of time is plotted. For this type of plot, the correct rate limiting step will be a linear curve fitted by Origin software. The model with the best fit to the data represents the reaction ‘ s rate limiting step.
Fig 3.3.9 shows the plot for the first hydrogenation kinetics of representative x=4 alloy. Linear regressions were performed on each mode\. As shown in the literature, in order to get a wider range of the reaction fitted and the regressions from 10% to 90% of the reaction ‘s completion (a = 0.1 ~0.9) are chosen . It can be seen that CV3D, GB2D and GB3D ail show a reasonable fit for x=4 alloy. In order to compare this three models clearly, the adjusted R2 value for each alloy was shown in Table 3.3.6. It is clear that the adjusted R2 value ofGB3D model shows the maximum value which is closest to 1. So GB3D is the best model for x=4 alloy. From the description of the GB3D model in Table 1.3, it is 3D growth and the growth interface velocity of diffusion decreases with time . For x=8 and 12 alloys GB3D model also gives the best fit (figures are not plotted here), it can be concluded that the first hydrogenation kinetics ofthree alloys are in accordance with GB3D mode\.
Table des matières
LIST OF FIGURES
LIST OF TABLES
1.1.1. Hydrogen energy
1. 1.2. Hydrogen storage
1.1.3. Metal hydrides
l .l.3.1. AB5 intermetallic compounds
l.l.3.2. AB2 intermetallic compounds
l.l.3.3. A28 intermetall ic compounds
l.l.3.4. AB intermetallic compounds
1. 104. Hydriding process
l .l.4.I . Absorption mechanism
l .l.4.2. Kinetic models
l. l .4.3. Thermodynamics
1.2. DOPING AND MECHANICAL DEFORMATION OF TtFE ALLOY
1.3. RESEARCH GOALS
l A. THESIS STRUCTURE
2. EXPERIMENTAL DETAILS
2. 1. SYNTHESIS OF MA TERlALS
2.1.1. Arc melting
2. 1.2. Bali mi ll ing
2. 1.3. Cold rolling
2. 104. Forging
2.2. CHARACTERIZA TION OF MA TERLALS
2.2.1 . Sieverts-type apparatus
2.2.2. X-ray diffraction
2.204. Scanning electron microscopy
3. EFFECT OF DIFFERENT ADDITIVES ON THE MICROSTRUCTURE AND HYDROGENATION PROPERTIES OF TIFE ALLOY
3. 1. HYDROGEN STORAGE PROPERnES OF TlO.95FEZRO.05, TtFEo.95ZRo.05 AND TtFEZRo.05 ALLOYS
3.1.1. Results and discussion
3. 1.1.1. Morphology
3. 1.1.2. Structural characterization
3.1. 1.3. First hydrogenation
3. 1.1.4. Pressure-composition isotherms
3.2. MICROSTRUCTURE AND HYDROGENATION PROPERTIES OF TIFE DOPED WITH (ZR+ 2MN) AS ADDITIVE
3.2.1. Results and discussion
3.2. 1.1. Morphology
3.2. 1.2. Crystal structural characterization
3.2. 1.3. First hydrogenation
3.2. 1.4. Pressure-composition isotherms
3.2. 1.5 . Air exposure
184.108.40.206. Improving tirst hydrogenation kinetics by cold rolling
3.3 . MICROSTRUCTURE EVOLUTION, PHASE TRANSFORMATION AND HYDROGENATION PERFORMANCE OF TIFE ALLOY DOPED WITH X WT.% (ZR+2V)
3.3.1. Results and discussion
3.3. 1. 1. Morphology
3.3. 1.2. Structure characteri zation
3.3. 1.3. Fi rst hydrogenation
3.3. 1.4. Phase transformation
3.3 .1.5 . Pressure-composition isotherms
3.3. 1.6. Cycling properties
3.3. 1.7. Airexposure
4. EFFECT OF MECHANICAL DEFORMATION ON THE MICROSTRUCTURE AND FIRST HYDROGENA TION KlNETICS OF TIFE-BASED ALLOY
4.1. EFFECTOF BALL MILLING ON THE FIRST HYDROGENATION PROPERTIES OF TIFE ALLOY DOPED WITH 4 WT.% ZR AS ADDITiVE
4.1.1 . Results and discussion
4. 1.1. 1. Chemical composition
4. 1.1.2. Morphology
220.127.116.11. Crystal structure characterization
4.1 . 1.3. First hydrogenation
4.1. 1.4. Kinetic models of tirst hydrogenation
4.2. EFFECT OF COLD ROLLING AND FORGING ON THE FIRST HYDROGENATION PROPERTIES OF TIFE ALLOY DOPED WITH 4 WT. % ZR AS ADDITIVE
4.2. 1. Results and discussion
4.2. 1.1. Morphology
18.104.22.168. Crystal structure characteri zation
22.214.171.124 . First hydrogenation
126.96.36.199. Kinetic models of first hydrogenation
4.3. EFFECT OF COLO ROLLING AND FORGlNG ON THE FIRST HYDROGENATION PERFORMANCE OF TIFE ALLOY DOPED WITH 4 WT.% (ZR+2MN) AS ADDITTVE
4.3 .1. Results and discussion
4.3 .1. 1. Morphology
188.8.131.52 . Crystal structure
184.108.40.206 . First hydrogenation
220.127.116.11. Rate limiting step of tirst hydrogenation
5. CONCLUSION AND FUTURE WORK
5.2. FUTURE WORK
5.2.1 . Study the dark and bright phases ofTiFe alloy doped with Zr+2Mn/Zr+2V