Impact of Nanostructuring on the Phase Behavior ... - ACS Publications

Subscriber access provided by UNIV OF PITTSBURGH

Article

The Impact of Nanostructuring on the Phase Behavior of Insertion Materials: The Hydrogenation Kinetics of a Magnesium Thin Film Lars Johannes Bannenberg, Herman Schreuders, Lambert van Eijck, Jouke R. Heringa, NinaJuliane Steinke, Robert M Dalgliesh, Bernard Dam, Fokko M. Mulder, and Ad A. van Well J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.6b02302 • Publication Date (Web): 03 May 2016 Downloaded from http://pubs.acs.org on May 11, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

The Impact of Nanostructuring on the Phase Behavior of Insertion Materials: The Hydrogenation Kinetics of a Magnesium Thin Film Lars J. Bannenberg,*,† Herman Schreuders,‡ Lambert van Eijck,† Jouke R. Heringa,† Nina-Juliane Steinke,§ Robert Dalgliesh,§ Bernard Dam,‡ Fokko M. Mulder,‡ Ad A. van Well† † Department of Radiation Science & Technology, Delft University of Technology, Mekelweg 15, Delft, 2629 JB, The Netherlands ‡ Department of Chemical Engineering, Delft University of Technology, Julianalaan 134, Delft, 2628 BL, The Netherlands § ISIS, Rutherford Appleton Laboratory, OX11 0QX, Didcot, Oxfordshire, United Kingdom ABSTRACT: Nanostructuring is widely applied in both battery and hydrogen materials to improve the performance of these materials as energy carriers. Nanostructuring changes the diffusion length as well as the thermodynamics of materials. We studied the impact of nanostructuring on the hydrogenation in a model system consisting of a thin film of magnesium sandwiched between two titanium layers and capped with palladium. While we verified optically the coexistence of the metallic alpha MgDx and the insulating beta MgD2-y phase, neutron reflectometry shows significant deviations from the thermodynamic solubility limits in bulk magnesium during the phase transformation. This suggests that the kinetics of the phase transformation in nanostructured battery and hydrogen storage systems is enhanced not only as a result of the reduced length scale but also due to the increased solubility in the parent phases.

INTRODUCTION Nanostructuring is widely applied in lithium-ion battery and hydrogen-storage materials to improve the performance of these materials as energy carriers.1-5 Due to a shortening of the diffusion length and a change in the thermodynamics, the lithium or hydrogen loading and unloading (kinetics) are enhanced.6,7 Several studies on different types of battery electrode materials discuss the effect of decreasing the grain size from several micrometers to tens of nanometers on its structure and kinetics.8,9 An unexpected finding in these studies is that the lithium mole fraction in both the lithium poor and rich phase appear to be particle-size dependent, i.e. the solubility limits increase considerably when decreasing the grain size from ~100 to ~10 nm having an influence on the kinetics. Indications of a similar, non-stoichiometric behavior is observed in diffraction studies on two-phase nanograin refined magnesium hydride.10-12 In this experimental study, we investigate the influence of nanostructuring on the phase behavior, solubility limits and kinetics of the (un)loading of hydrogen in a magnesium thin film. The difference with the studies mentioned above is fourfold: (i) In a 2D system only one dimension is reduced to the nanometer scale. Besides the influence of the surface and interface free energy the strain-energy contributions may play an important role since volume changes upon (un)loading are accommodated in one dimension only. (ii) This 2D system gives access to a larger range of experimental techniques that are potentially more accurate. (iii) Moreover, it can be tailored more precisely than the 3D nano-grain system. (iv) Additionally, 2D films are relevant for applications such as for example hydrogen storage or sensing devices, switchable mirrors and energy saving windows.13-18 In particular, this clamped system can act as model system for melt infiltrated nano-materials that are currently developed for bulk storage of hydrogen.19 As our model system, we select a 10-nm-thick Mg film that we hydrogenate. We include a top layer of 10 nm Pd to prevent oxidation and to catalyze the hydrogen dissociation reaction and sandwich the Mg layer between two 10-nm-thick Tilayers to prevent the mixing of Pd and Mg and reactions with the substrate.20-22 When Mg hydrogenates, it changes its crystal structure from low solubility HCP (metallic α-MgHx) to a hydrogenated tetragonal MgH2 phase (insulating β-MgH2-y), resulting in a volume increase of approximately 30% in bulk. This metal-toinsulator transition of Mg results in a large change in optical contrast. Mooij and Dam utilized this change in optical contrast and studied the same system by optical transmission measurements (hydrogenography) in conjunction with a John

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 21

son-Mehl-Avrami- Kolmogorov (JMAK) analysis for phase nucleation and growth.23-25 They observed the coexistence of the twophases (α/β-Mg) during hydrogenation of their thin film sample.23,24 From the JMAK analysis it was concluded that on dehydrogenation no such 2D behavior takes place.

Figure 1. (a) Top view 2 x 2 mm images of the sample in transmission at various moments in time during hydrogenation at PD2 = 100 Pa for t < 450 min and PD2 = 115 Pa for t > 450 min. The β-MgD2-y domains transmit more light and are therefore brighter than α-MgDx. Both nucleation and growth occur. (b) Number of nucleation sites as a function of time. After the increase in pressure at t = 450 min sites with a higher barrier for nucleation become available for nucleation. (c) Fraction of β-MgD2-y as a function of time. The increase in pressure resulted in an accelerated conversion of α-MgDx to βMgD2-y, which is both due to an increased number of nuclei and an increase in radial growth velocity. Here, we study the magnesium-based model system to investigate the phase behavior of magnesium during the (de)hydrogenation. We study the ab- and desorption kinetics of deuterium in the thin film system by simultaneous optical transmission and (off-)specular neutron reflectometry (NR) measurements. The former technique gives information of the presence and evolution of the reflecting metallic alpha and transmitting non-metallic beta phases, whilst the latter yields structural information about the thickness and deuterium content of the alpha and beta phases. Additionally, Density Functional Theory (DFT) computations are performed to aid the interpretation. We find during (de)hydrogenation of the film an increased solubility range of both α-MgDx and β-MgD2-y, suggesting that the kinetics of the phase transformation in confined battery and hydrogen storage systems can be enhanced not only as a result of the reduced length scale but also due to the enhanced solubility in the parent phases.

RESULTS AND DISCUSSION Absorption of deuterium. In our analysis, we will restrict ourselves to the second cycle of (un)loading of the thin film system as the second cycle is more representative than the first cycle that differs significantly from subsequent cycles.5,23 Deuterium gas is used to enhance contrast in the NR measurements and all the experiments are conducted at a constant temperature of T = 80°C. After the first cycle, the Mg layer is fully unloaded and regains the thickness of the virgin Mg layer. However, the titanium layers remain partly deuterated up to TiD0.6 and expanded by 20% as compared to its virgin value. Subsequently, we exposed the layered system to two different deuterium pressures during loading: first to PD2 = 100 Pa for t < 450 min and subsequently to PD2 = 115 Pa for t > 450 min. Both the neutron reflectometry data and optical transmission measurements indicate that the deuteration of the film saturates after 1200 minutes. In the first 10 minutes, the titanium layers load to TiD1.7 and have an expansion of about 30% as compared with the virgin state. After the pressure increase, a further increase of the titanium loading was observed to TiD2. Top view images of the sample in transmission at various instances in time during the loading of the sample are depicted in Figure 1a. During hydrogenation, the optical transmission images show that there are only regions with a low optical transmission (0.55), corresponding to the metallic α-MgDx phase and domains with a high optical transmission (1.00), corresponding to the β-MgD2-y phase. These nuclei of β-MgD2-y are formed over time and grow radially outward with a constant radial speed in a nucleation-and-growth process. From the images we determine the number of nuclei and the


Page 3 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fraction of β-MgD2-y as a function of time (Figure 1b-c). Note that the change in optical transmission as a result of a different deuterium content of the Ti deuteride layers is negligible. The graph of the fraction of β-Mg2-y nuclei is S-shaped: from t = 100 min onwards, a substantial number of visible nuclei are formed that grow radially outward, accelerating the conversion of α-MgDx. Gradually, fewer nuclei are formed and impingement of nucleation sites slows down the growth of the β-MgD2-y fraction considerably. The increase in pressure from PD2 = 100 Pa to PD2 = 115 Pa at t = 450 min results in an accelerated conversion of α-MgDx to β-MgD2-y. This is due to both an increase in the radial velocity from 0.12 µm min-1 to 0.20 µm min-1 and the formation of new nucleation sites. After the increase in pressure, sites with a higher barrier for nucleation become available for nucleation.24,26

Figure 2. (a) Contour plots of the specular and off-specular reflectivity at various moments in time, z is the perpendicular and x the in-plane coordinate. The amount of off-specular scattering (scattering away from Qx = 0 nm-1) is a measure of the presence of in-plane structures smaller than the in-plane coherence length of the neutron. (b) Specular NR curves, which are a cut at Qx = 0 nm-1 in the images of Figure 2a. Large changes over time occur due to differences in both deuterium content and thickness of the layered system that can be revealed by modelling. (c) Deuterium content as a function of time during hydrogenation. A large increase in deuterium content is seen between t = 0 min and t = 30 min, which is attributed to the loading of the α-MgDx. From t = 500 min, the β-MgD2-y domains exceed the in-plane coherence length of the neutron, allowing to separate the contribution of the α-MgDx and β-MgD2-y to the reflectogram. The grey dashed line indicates the weighted average of the α-MgDx and β-MgD2-y deuterium content. (d) Layer thickness of the magnesium as a function of time. Neutron reflectometry measurements obtained during the absorption of deuterium are depicted in Figure 2a and Figure 2b. Figure 2c-d represent the evolution of the deuterium content and expansion of the Mg layer as a function of time as obtained by fitting the specular NR data (see Supporting Information). At t = 30 min, a deuterium loading of the Mg layer to MgD0.3 is observed. Looking at the transmission results, presented in Figure 1, only a negligible fraction of the available α-MgDx was converted to β-MgD2-y at t = 30 min, leading to the conclusion that the metallic α-MgDx is loaded up to MgD0.3. For t > 30 min the average amount of deuterium in the two-phase system increases to MgD0.6 at t = 500 min. From this time onwards the β-MgD2-y domains have radii of more than 30 µm which is larger than the in-plane coherence length of the neutron. Hence, it is possible to observe the evolution of the deuterium content and layer thicknesses of both phases separately. The results as presented in Figure 2c and 2d, show that the amount of deuterium in α-MgDx remains fairly constant around MgD0.35. The deuterium content of the β−MgD2-y domains increases as a function of time: from roughly MgD1.5 at t = 500 to MgD2.0 at t = 900 min. Additionally, the layer expands approximately 35% with respect to the virgin state. At t = 900 min, the fraction and domain size of the α-MgDx is small, with the implication that the system can only be modelled as being composed of one, average, phase. An increase of the overall deuterium content from MgD1.9 to MgD2.0 between t = 900 min and t = 1200 min is observed.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 21

Figure 3. Deuterium absorption model that is a refinement of the model proposed by Mooij and Dam.23 (a) Starting situation of the second loading after the first loading-unloading cycle. (b) The Ti layers hydrogenate to TiD1.7 within the first 10 minutes of the loading. Subsequently, the α-MgDx loads to MgD0.3, which is much higher than the bulk solubility of D2 in Mg. (c) A nucleation-and-growth process governs the transition from metallic α-MgDx to insulating β-MgD2-y. The deuterium content of β-MgD2-y increases as a function of time from MgD1.5 to MgD2 (d) Fully loaded system. Based on our experimental observations, we formulate a new deuterium absorption model as shown in Figure 3 which is a refinement of the model proposed by Mooij and Dam.23 Starting off with the unloaded situation after an (initial) loading/unloading cycle, first, the titanium layers are loaded within 10 minutes to TiD1.7. Consistent with the more negative enthalpy of formation for titanium hydride (-130 kJ mol-1 H2) as compared with magnesium hydride (-75 kJ mol-1 H2). Next, the α-MgDx is loaded to around MgD0.3. A nucleation and growth process governs the phase transition to substoichiometric magnesium hydride, where the degree of deuterium increases as a function of time. After all the available α-MgDx has been converted to β-MgD2-y, the magnesium layer is fully loaded to MgD2. As shown both by NR and optical transmission measurements, the hydride phase domains extends throughout the entire layer. This is different from the system studied by Kalisvaart et al. where a larger driving force (105 Pa) and a thicker (50 nm) magnesium layer was used.27 In their study, they found that the conversion of α-Mg to β-MgD2 commenced at the top Mg-interface. The transition from α-MgDx to β-MgD2-y is accompanied by a substantial increase in volume. Due to the geometric nature of the thin film, this volumetric expansion manifests itself as an expansion of the thickness and is measured to be around 35%. As a consequence, the Mg, upper Ti and Pd layer need to deform plastically near the α-MgDx to β-MgD2-y domain boundaries to accommodate this large volume expansion of Mg on hydrogenation. This results in an extremely large ‘effective’ edge boundary energy as determined in the context of a classical nucleation and growth theory.14 Another factor is the interface energy between α-MgDx and TiD2-y that might stabilize the α-MgDx phase because of the close lattice match.12 In combination with the implied stress needed to generate the plastic deformation, the interface energy could explain the extended solubility limit of Mg as compared with the bulk solubility limit of 10-4.28 In bulk, the magnesium can expand freely without the necessity to deform. In the system under investigation, the balance of varying internal stresses, interface and edge energies determines when α−MgDx converts to β−MgD2-y during hydrogenation. Similar to the case of lithium insertion materials, this makes it thermodynamically more favourable to load a substantial amount of deuterium in α−MgDx before the conversion to β−MgD2-y commences.8 The enlarged solubility limits have important consequences for the kinetics of the loading of insertion materials. The increased solubility in the α−MgDx and the lower occupancy in the β−MgD2-y enhances the diffusion. Additionally, the enhanced solubility limits causes a lowering of the edge boundary energy involved in the plastic deformation at the interface between α−MgDx and β−MgD2-y which will improve the kinetics of the phase transformation.5

Desorption of deuterium. To unload the multilayer system, we decrease the pressure in two steps: first to PD2 = 10 Pa and then to PD2 = 4 Pa. Top-view images of the sample in transmission at six instances in time during unloading at PD2 = 10 Pa are shown in Figure 4a. Unlike during loading, we do not see the formation of clear nuclei that grow radial outward, but merely an overall decrease in optical transmission. However, some fluctuations in transmission are clearly visible: the transmission tends to be higher near the nucleation sites of the β-MgD2-y during loading, suggesting that desorption is slower in the regions where β-MgD2-y has been formed earlier during deuterium absorption. Next, we average the transmission over the entire image and plot this transmission as a function of time (Figure 4b). Three different regimes can clearly be distinguished: In the first regime, for t < 15 min, the average transmission remains relatively constant. In the second regime, the transmission starts to drop significantly and almost linearly between t = 15 min and t = 70 min. In the third regime, t > 70 min, the optical transmission is back to the transmission of the virgin sample, and all the β-MgD2-y is converted to α-MgDx. Large changes in specular and off-specular intensities are seen during unloading (Figure 5a and 5b). Since the optical transmission images do not show large domains that exceed the estimated in-plane coherence length of the neutron, we fit the model using a uniform (magnesium) layer. After unloading, the fitted titanium layer expansion was 25 % and the


Page 5 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


deuterium content dropped slightly to TiD1.6. The resulting fitted deuterium content and layer thickness of the Mg layer are shown in Figure 5c and 5d.

Figure 4. (a) Top view 2 x 2 mm images of the sample in transmission at various moments in time during dehydrogenation at PD2 = 10. Unlike during loading of the sample, no large domains with either a high (1) or low (0.55) transmission are observed, but more an overall decrease in optical transmission. (b) Average optical transmission of the film during dehydrogenation. In the first 15 minutes of the dehydrogenation, no changes in optical transmission are observed.

Figure 5. (a) Contour plots of the specular and off-specular reflectivity at various moments in time during unloading at PD2 = 10 Pa. The increase in off-specular intensity (outside Qx = 0 nm-1) is a result of the formation of in-plane structures on a length scale smaller than the neutron’s in-plane coherence length. (b) Specular NR curves, which are a cut at Qx = 0 nm-1 in the off-specular images of Figure 5a and change as a function of time due to difference in layer thickness and deuterium content of the different layers. (c) Deuterium content of the magnesium layer as a function of time that are ob-



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 21

tained by fitting the specular NR curves. A linear decrease of the deuterium content is observed for t < 100 min. For t < 15 min, no changes in optical contrast where observed. (d) Magnesium layer thickness as a function of time. From t = 100 min onwards, both the deuterium content and layer thickness of the magnesium layer remained stable until t = 400 min where we stopped the measurement (not shown in the Figure).

Figure 6. Deuterium desorption model at PD2 = 10 Pa. (a) Starting situation after loading the multilayer system. (b) In the first 15 minutes of the unloading, the β-MgD2-y dehydrates to MgD1.7. (c) Metallic α-MgDx domains are formed that are significantly smaller than the resolution of the camera (2.2 µm). (d) Situation after unloading at PD2 = 10 Pa which finished at t = 100 min. A further decrease in pressure to PD2 = 4 Pa will result in a return of the deuterium content and thickness of the layers to their respective starting values of the loading procedure as shown in Figure 3a. The decrease in deuterium content of the Mg layer is almost linear for t < 70 min. However, the optical transmission measurements show a slower decrease in transmission in the first 15 minutes, and no decrease in the first 10 minutes. Hence, almost all magnesium will thus be β-MgD2-y. We therefore conclude that the decrease of deuterium content from MgD1.95 to MgD1.8 in the first 15 minutes is ascribed to the partial unloading of the β-MgD2-y. Between 15 and 70 minutes all the insulating β-MgD2-y is converted to metallic α-MgDx, and the average deuterium content drops from MgD1.6 to MgD0.55. It drops further to MgD0.4 at around t = 100 min, and stays stable afterwards for at least three hours. Note that although the deuterium content of the α-MgDx at this point in time is similar to the deuterium content of the α-MgDx phase during loading, the Mg layer is thicker than during loading (10.4 nm vs 9.7 nm). This can be explained by assuming that the magnesium is slightly, but homogenously, porous, which has been observed earlier with neutron reflectometry for unloaded magnesium films capped with palladium.29 This is at variance with the previous hypothesis of Mooij and Dam that the porosity is concentrated at the Mg-Ti interface, however, its confirms the model which needs a porous film in order to explain the fast rehydrogenation kinetics observed in this system.23 The unloading step to PD2 = 4 Pa resulted in a complete dehydrogenation of the Mg layer and a decrease in layer thickness of the magnesium and titanium layers to the starting values of the loading procedure in a time span of 3 hours. We hypothesize a desorption model as shown in Figure 6. In the first 15 minutes, a drop of the deuterium content to MgD1.8 is observed. For t > 15 min, we see a large increase in off-specular intensity, much larger than during loading. This increased off-specular intensity results from in-plane inhomogeneities on a length scale smaller than the neutron’s inplane coherence length (≈30 µm), with a different deuterium content and/or layer thickness. This overall decrease in transmission is likely to be caused by the formation of many small metallic domains that are significantly smaller than the resolution of the camera (2.2 µm). Hence, the top view image represent an average of the α-MgDz and β-MgD2-y domains. After the formation of α-MgDx, the α-MgDx layer dehydrogenates further to MgD0.3. A decrease in pressure from PD2 = 10 Pa to PD2 = 4 Pa is required to fully dehydrogenate the Mg layer. As for the extended solubility limits during loading, the enlarged limits upon dehydrogenation are likely a result of the stabilizing effect of the titanium interfaces.

Density Functional Theory. To acquire a better understanding of the optical properties of α-MgDx and β-MgD2-y with different amount of hydrogen concentrations, we perform Density Functional Theory (DFT) computations for two different scenarios: 1) we add hydrogen atoms to the HCP α-Mg crystal structure at the 2a and 4f sites; 2) we remove hydrogen atoms from β-MgD2 at various quantities (see Supporting Information for more details).30 We allow the unit cell to relax freely. When hydrogen atoms are inserted into α-MgHx, the unit cell expands with about 5% at MgH0.4. Band gap and density-ofstates computations show that the magnesium remains metallic with the insertion of hydrogen atoms and hence has a low optical transmission for hydrogen concentrations smaller than MgH0.6. The removal of an increasing amount of hydrogen atoms from MgH2 in the β-MgH2-y results in a monotonous decrease of the volume up to 7% for MgH1.5. While the band gap and density of states computations show that the band gap decreases


Page 7 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


monotonously when hydrogen atoms are removed, a band gap of at least 1.5 eV remains visible until MgH1.6. However, due to the used method, the calculated band gap is likely to be an underestimation.31 Conclusions and Implications The (de)hydrogenation of thin film Mg is studied with in situ (off)-specular neutron reflectometry and optical transmission measurements. For the first time the (de)hydrogenation from nano up to the macroscopic scale is observed of the two-phase system. The deuterium content of both phases deviate from the stoichiometric values in the bulk system, i.e. the solubility limits are increased. These findings are supported by density-functional-theory calculations. We conjecture that the increase of solubility limits is a general consequence of nanostructuring and abundant interfacial interactions, both in three dimensions as previously found in lithium-insertion materials, as well as in one dimension, as shown here for a confined hydrogen insertion material. This has important consequences for the design of nano-structured devices, batteries and other storage materials as kinetics can be enhanced not only because of shorter diffusion lengths, but also due to changing stochiometries of the parent phases that enhance diffusion and lower phase transition barriers which is especially important in confined systems.

ASSOCIATED CONTENT Supporting Information. Details concerning the Neutron Reflectometry measurements and data processing, the Hydrogenography setup and measurements and information concerning the DFT computations. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]

Tel.: (+31) 15 2789753 Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENT We thank the ISIS facility at the Rutherford Appleton Laboratory, Oxfordshire, for the provision of beam time and technical support (RB1510185) and wish to thank L.P.A. Mooij, B.N. de Graaff, M. Wagemaker, M.J. Van Setten, and C. Boelsma for fruitful th discussions. This project has received funding from the European Union's 7 Framework Programme for research, technological development and demonstration under the NMI3-II Grant number 283883.

ABBREVIATIONS DFT, Denstity Functional Theory; NR, Neutron Reflectometry.

REFERENCES (1) Berube, V.; Chen, G.; Dresselhaus, M.S. Impact of Nanostructuring on the Enthalpy of Formation of Metal Hydrides. Int. J. Hydrogen Energ. 2008, 33, 4122-4131. (2) Fichtner, M. Properties of Nanoscale Metal Hydrides Nanotechnology 2009, 20, 204009. (3) Kim, K.C.; Dai, B.; Johnson, J.K.; Sholl, D.S. Assessing Nanoparticle Size Effects on Metal Hydride Thermodynamics Using the Wulff Construction. Nanotechnology 2009, 20, 204001. (4) Arico, A.S.; Bruce, P.; Scrosati, B.; Tarascone, J.M.; Van Schalkwijk, W. Nanostructured Materials for Advanced Energy Conversion and Storage Devices. Nature Mater. 2005, 5 366-377. (5) Mooij, L.P.A.; Baldi, A.; Boelsma, C.; Shen, K.; Wagemaker, M.; Pivak, Y.; Schreuders, H.; Griessen, R.; Dam, B. Interface Energy Controlled Thermodynamics of Nanoscale Metal Hydrides. Adv. Energy Mater. 2011, 1, 754-758. (6) Gao, J.; Adelhelm, P; Verkuijlen, M.H.W.; Rongeat, C.; Herrich, M.; van Bentum, P.J.M.; Gutfleisch, O.; Kentgens, A.P.M.; de Jong, K.P.; de Jongh, P.E. Confinement of NaAlH4 in Nanoporous Carbon: Impact on H2 Release, Reversibility, and Thermodynamics. J. Phys. Chem. C 2010, 114, 4675-4682. (7) Jeon, K.J.; Moon, H.R.; Ruminski, A.M.; Jiang, B.; Kisielowski, C.; Bardhan, R.; Urban, J.J. Air-Stable Magnesium Nanocomposites Provide Rapid and High-Capacity Hydrogen Storage without Using Heavy-Metal Catalysts. Nature Mater. 2011, 10, 286-290. (8) Wagemaker, M.; Mulder, F.M.; Van der Ven, A. The Role of Surface and Interface Energy on Phase Stability of Nanosized Insertion Compounds. Adv. Mater. 2009, 21, 2703-2709. (9) Malik, R.; Zhou, F.; Ceder, G. Kinetics of Non-Equilibrium Lithium Incorporation in LiFePO4. Nature Mater. 2011, 10, 587-590. (10) Schimmel, H.G.; Huot, J.; Chapon, L.C.; Tichelaar, F.D.; Mulder, F.M. Hydrogen Cycling of Niobium and Vanadium Catalyzed Nanostructured Magnesium. J. Am. Chem. Soc. 2005, 127, 14348-14354. (11) Flacau, R.; Tan, X.; Danaie, M.; Fritzsche, H.; Mitlin, D. In-situ Neutron Powder Diffraction on TiF3 Catalysed Magnesium for Hydrogen Storage Applications. Can. Metall. Quart. 2015, 54, 47-50. (12) Mulder, F.M.; Singh, S.; Bolhuis, S.; Eijt, S.W.H. Extended Solubility Limits and Nanograin Refinement in Ti/Zr FluorideCatalyzed MgH2. J. Phys. Chem. C 2012, 116, 2001-2012.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 21

(13) Schlapbach, L.; Zuttel , A. Hydrogen-Storage Materials for Mobile Applications, Nature 2001 , 414 , 353-358. (14) Baldi, A.; Dam, B. Thin Film Metal Hydrides for Hydrogen Storage Applications. J. Phys. Chem. C 2011, 21, 4021-4026. (15) Slaman, M.; Dam, B.; Pasturel, M.; Borsa, D.M.; Schreuders, H.; Rector, J.H.; Griessen, R. Fiber Optic Hydrogen Detectors Containing Mg-Based Metal Hydrides. Sens Actuators B Chem. 2007, 123, 538-545. (16) Huiberts, J.N.; Griessen, R.; Rector, J.H.; Wijngaarden, R.J.; Dekker, J.R.; De Groot, D.G.; Koeman, N.J. Yttrium and Lanthanum Hydride Films with Switchable Optical Properties. Nature 1996, 380, 231-234. (17) Bao, S.; Tajima, K.; Yamada, Y.; Okada, M.; Yoshimura, K. . Color-Neutral Switchable Mirrors Based on Magnesium-Titanium Thin Films. Appl. Phys. A 2007, 87, 621-624. (18) Jain, I.; Lal, C.; Jain, A. Hydrogen Storage in Mg: a Most Promising Material. Int. J. Hydrogen Energ. 2010, 25, 5133-5144. (19) De Jongh, P.E.; Eggenhuisen, T. Melt Infiltration: an Emerging Technique for the Preparation of Novel Functional Nanostructured Materials. Adv. Mater. 2013, 25, 6672-6690. (20) Tan, X.; Harrower, C.T.; Amirkhiz, B.S.; Mitlin, D. Nano-scale Bi-Layer Pd/Ta, Pd/Nb, Pd/Ti and Pd/Fe Catalysts for Hydrogen Sorption in Magnesium Thin Films. Int. J. Hydrogen Energ. 2009, 34, 77417748. (21) Baldi, A.; Pálsson, G.K.; Gonzalez-Silveira, M.; Schreuders, H.; Slaman, M.; Rector, J.H.; Krishnan, G.; Kooi, B.J.; Walker, G.S.; Fay, M.J. et al. Mg/Ti Multilayers: Structural and Hydrogen Absorption Properties, Phys. Rev. B 2010, 81, 224203. (22) Jung, H.; Yuh, J.; Cho, S.; Lee, W. Effects of Ti Interlayers on Microstructures and Hydrogen Storage Capacity in Mg/Pd Multilayer Thin Films, J. Alloy. Compd. 2014, 601, 63-66. (23) Mooij, L.P.A.; Dam, B. Hysteresis and the Role of Nucleation and Growth in the Hydrogenation of Mg Nanolayers. PCCP 2013, 15, 2782-2792. (24) Mooij, L.P.A.; Dam, B. Nucleation and Growth Mechanisms of Nano Magnesium Hydride from the Hydrogen Sorption Kinetics. PCCP 2013, 15, 11501-11510. (25) Gremaud, R.; Broedersz, C.P.; Borsa, D.M.; Borgschulte, A.; Mauron, P.; Schreuders, H.; Rector, J.H.; Dam, B.; Griessen, R. Hydrogenography: An Optical Combinatorial Method to Find New Light-Weight Hydrogen-Storage Materials, Adv. Mater. 2007, 19, 28132817. (26) De Yoreo, J.J.; Vekilov, P.G. Principles of Crystal Nucleation and Growth, Rev. Mineral. Geochem. 2003, 54, 57-93. (27) Kalisvaart, W.P.; Luber, E.J.; Poirier, E.; Harrower, C.T.; Teichert, A.; Wallacher, D.; Grimm, N.; Steitz, R.; Fritzsche, H.; Mitlin, D. Probing the Room Temperature Deuterium Absorption Kinetics in Nanoscale Magnesium Based Hydrogen Storage Multilayers Using Neutron Reflectometry, X-ray Diffraction, and Atomic Force Microscopy. J. Phys. Chem. C 2012, 116, 5868-5880. (28) Popovic, Z.D.; Piercy, G.R. Measurement of the Solubility of Hydrogen in Solid Magnesium. Metall. Trans. A 1975, 6, 1915-1917. (29) Dura, J.A.; Kelly, S.T.; Kienzle, P.A.; Her, J.-H.; Udovic, T.J.; Majkrzak C.F.; Chung, C.-J.; Clemens, B.M. Porous Mg Formation upon Dehydrogenation of MgH2 Thin Films. J. Appl. Phys. 2011, 109, 093501. (30) Schimmel, H.G.; Kearley, G.J.; Huot, J.; Mulder, F.M. Hydrogen Diffusion in Magnesium Metal (α phase) Studied by ab initio Computer Simulations J. Alloy. Compd. 2005, 404, 235-237. (31) Van Setten, M.J.; Popa, V.A.; De Wijs, G.A.; Brocks, G. Electronic Structure and Optical Properties of Lightweight Metal Hydrides Phys. Rev. B 2007, 75, 035204.


Page 9 of 21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

190x142mm (300 x 300 DPI)


Page 10 of 21

Page 11 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


190x142mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

190x142mm (300 x 300 DPI)


Page 12 of 21

Page 13 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


190x142mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

190x142mm (300 x 300 DPI)


Page 14 of 21

Page 15 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


190x142mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

190x142mm (300 x 300 DPI)


Page 16 of 21

Page 17 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


190x142mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

194x123mm (300 x 300 DPI)


Page 18 of 21

Page 19 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


190x142mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

190x142mm (300 x 300 DPI)


Page 20 of 21

Impact of Nanostructuring on the Phase Behavior ... - ACS Publications

Recommend Documents