Tuning Cage Dimension in Clathrate Hydrates for Hydrogen Multiple

Dec 30, 2013 - ... A. Moultos , George E. Romanos , Athanassios K. Stubos , Ioannis G. Economou ... Ioannis N. Tsimpanogiannis , Ioannis G. Economou...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/JPCC

Tuning Cage Dimension in Clathrate Hydrates for Hydrogen Multiple Occupancy Dong-Yeun Koh,† Hyery Kang,† Jiwon Jeon,‡ Yun-Ho Ahn,† Youngjune Park,†,§ Hyungjun Kim,*,‡ and Huen Lee*,†,‡ †

Department of Chemical and Biomolecular Enginieering and ‡Graduate School of EEWS, Korea Advanced Institute of Science and Technology, 373-1 Guseong-dong, Yuseong-gu, Daejeon 305-701, South Korea S Supporting Information *

ABSTRACT: As hydrogen molecules enter the clathrate hydrate body, the ubiquitous dodecahedral cavity (512) is too small to allow anything but single occupancy thermodynamically. The possibility that H2 double occupancy can occur in the dodecahedral cavity has been suggested and is still under debate. Here we uncover the unique feature of multiple occupancy of the hydrogen molecule in a dodecahedral cavity as induced by tuning the cage dimensions. The guest promoter population in the hydrate matrix spontaneously controls the degree of molecular hydrogen storage by tuning the cage dimensions. Our analysis combined with computational study reveals that only ∼1% expansion (∼3% in volume) of the cage dimensions is sufficient to provide thermodynamically stable room for double occupancy in the dodecahedral cavity. The findings in this research provide a strategy for doubling the hydrogen population in dodecahedral cavities in structure II hydrates.



INTRODUCTION Clathrate hydrate with high levels of chemical engineering versatility is a new avenue to create remarkably environmental sound materials. Deliberate introduction of large quantities of hydrogen molecules in an icy water framework is a key factor in establishing a more efficient and environmentally friendly hydrogen-based economy. Numerous studies on hydrogen storage in clathrate hydrate materials have evolved since pure hydrogen hydrate formation under extremely high pressure conditions (∼2 kbar) was identified.1−4 Binary gas hydrates containing both H2 and another competing guest molecule (i.e., a promoter) provide a simple way to store H2 using much milder conditions. The most well-known promoter, tetrahydrofuran (THF), easily forms the sII hydrate itself and significantly lowers the hydrogen hydrate formation pressure.5 Because of the small size of the dodecahedral cavities (512), it has been widely thought that H2 molecules singly occupy most 512 cavities, while the promoter molecule THF fills all of the large cavities (51264). To maximize hydrogen storage capacity, it will certainly be beneficial to achieve multiple hydrogen occupancy per cage, and multiple occupancy of the small cage is more promising because the liquid guest former strongly interacts with the surrounding host water framework. However, the tuning mechanism driven by the relative ratio of guest to host molecules can cause the effective exclusion of liquid guest molecules while at the same time the spontaneous inclusion of gaseous molecules.6 On a molecular level, hydrogen hydrates highlight the unposed question: how can the concept of creating possible routes for multiple hydrogen molecules in a small dodecahedral (512) cage be explored. This is becoming crucial for assessing © 2013 American Chemical Society

whether clathrate hydrate materials can be used as hydrogen storage media in both pure H2 and binary (or semiclathrate) H2 hydrates. Many reports have suggested that no more than one hydrogen molecule can be entrapped in the small cavity in either pure-H2 hydrate or binary-H2 hydrates. This issue has been under debate because the research by Mao et al.1 suggested the possibility of hydrogen double occupancy in the 512 cavity of a pure-H2 hydrate, and several other studies on the H2+icy matrix have explored this possibility as well.5,7−17 Thus, probing the possibility of doubly occupancy for H2 in the small cavity of the sII (sII-S) hydrate along with the development of an appropriate tuning method to enable double occupancy will determine the hydrogen storage capacity of the clathrate hydrate as about 1 to 2 wt %, which is an essential factor in determining the feasibility of clathrate hydrates as future hydrogen storage media. Furthermore, the use of a gaseous sII promoter can be beneficial in weakening the guest−host interaction, thereby inducing the facile escape of guest molecules from the confined cage space, followed by a direct mutual chemical exchange between the promoter and hydrogen.18 We note a recent interesting approach by Lu et al.18 in which a H2 multiply loaded hydrate (in both cavity types in an sII hydrate) is built using the N2 hydrate under moderate hydrogen pressure (15 MPa). Spectroscopic evidence of double H2 occupancy in sII-S was in good agreement with previous MD simulation results. Received: October 28, 2013 Revised: December 22, 2013 Published: December 30, 2013 3324

dx.doi.org/10.1021/jp410632q | J. Phys. Chem. C 2014, 118, 3324−3330

The Journal of Physical Chemistry C

Article

Figure 1. (a) Snapshot of doubly occupied hydrogen molecules from the molecular dynamics simulation. The cage geometry, symmetry and kinetic diameter of H2 are presented. (b) Structure of hollow sII hydrate and large guest promoters used in this study. (c), (d) Structure of hollow sI hydrate and sH hydrate. sI and sH hydrate both have 512 cavity as a main building block. Dodecahedral cavity (D) is represented in red letter.

(∼0.8 MPa) or isobutane (∼0.2 MPa) at 243 K. During hydrate formation, ice particles are isolated from liquid-phase propane or isobutane for better quality of samples. After a week of formation period, propane or isobutane hydrate particles were collected and finely ground (∼200 μm) in the LN2. Ground solid particles of propane and isobutane hydrates were again pressurized in the high-pressure reactor with H2 up to 25−130 bar at 77 K. Then, the reactors were placed in a temperaturecontrolled bath at 240 K. The pressure of each system was stabilized at about 10, 20, 30, 40, and 50 MPa within 6 h of reaction. Under this temperature and pressure condition, pure hydrogen hydrate could not be formed without any promoters. The reactor was kept at 243 K bath for 3days. Reactor was quenched in LN2 and simultaneously the pressure was released to atmospheric pressure for further spectroscopic analysis. For the THF hydrates, identical experimental procedure was adopted. For pure-H2 hydrate, finely ground ice particles were reacted with 100 MPa of H2 pressure at 243 K in the same high-pressure reactor. Preparation of Nonstoichiometric Samples. Nonstoichiometric samples were prepared by controlling the propane concentration mixed with finely ground ice particles. Exact amount of liquefied propane was transferred (by precooled pipet) to the ice particle (∼200 μm) and well mixed in the reactor at 77 K. The reactors were pressurized with H2 of 120−125 bar at 77 K, and the reactors were moved to a temperature-controlled bath at 243 K. The pressure of the reactors was stabilized at ∼50 MPa within 6 h of reaction, and they were kept at 243 K bath for 3 days. Reactors were quenched in LN2 and simultaneously the pressure was released to atmospheric pressure for further spectroscopic analysis. Characterization. For Raman measurements, the Horiba Jobin Yvon LabRAM HR UV/vis/NIR high-resolution dispersive Raman microscope was used in which a CCD detector is equipped and cooled by liquid nitrogen. The excitation source was an Ar-ion laser emitting a 514.53 nm line. The laser intensity was typically 30 mW. Samples were kept at 77 K during Raman measurements. For variable temperature

At this point, we recognize that the hydrogen bonded hydrate cage lattice is quite flexible, either contracted or stretched, compared with the rigid zeolite or MOF matrix. A sensitive polyhedral water cage made with a fixed number of lattices can be tuned through external stimulation methods, such as temperature and pressure changes, and more significantly, by altering the molecular details of a specific guest promoter, thus enabling multiple H2 molecules to occupy a cage. Eventually, the most stable cage structure is achieved, leading to the maximum hydrogen storage capacity. Henceforth, we mainly focus on the effect of a small change of the cage dimensions on the number of enclosed H2 molecules. Such a critical issue needs to be clearly addressed owing to its direct link to gas storage capacity. The 512 cavity (average radius ∼3.9 Å) is known to be the most common for clathrate hydrate (Figure 1a); it can easily accommodate light gaseous molecules mainly in the single occupancy form. Three wellknown types of clathrate hydrate structures (sI, sII, and sH) have a 512 cavity as a basic building block (Figure 1b,c).19 We need to concentrate on the ubiquity of 512 cavities in the clathrate hydrate system to determine if the 512 cavity could provide a key solution for increasing hydrogen quantity in an icy matrix. If the 512 cavity size can be tuned via a promoter exchange mechanism, the newly formed 512 cavity forms will differ considerably. Here we attempted to control the cage size by simply changing the guest promoter from THF to propane and isobutane promoters (structure II hydrate) to induce plastic-like lattice elongation, and the possible occurrence of multiple occupancies as the most plausible clathrate hydrate structure for accommodating hydrogen is shown to be structure II as proved by structure transition (sH to sII) study.22



MATERIALS AND METHODS Preparation of (Stoichiometric) Clathrate Hydrate Samples. Deionized water with ultrahigh purity was supplied from Merck (Germany). Propane, isobutane, and hydrogen gas with a purity of 99.9995 mol % were supplied by Special Gas (Korea). Powdered ice (∼200 μm) was pressurized by propane 3325

dx.doi.org/10.1021/jp410632q | J. Phys. Chem. C 2014, 118, 3324−3330

The Journal of Physical Chemistry C

Article

Figure 2. Raman spectra of the products of the reaction of propane (a) and isobutane (b) hydrate with H2 gas at 243 K under various amounts of pressure. Peaks at 4120 [Q1(1)] and 4125 cm−1 [Q1(0)] for a single H2 molecule in the 512 cage (1S). Peaks at 4153 and 4159 cm−1 for two H2 molecules in a single 512 cage (2S). (c) Raman spectra of the binary THF-H2 hydrate (at various amounts of hydrogen pressure) and the pure-H2 (formation condition: 243 K, 100 MPa) hydrate. Peaks at 4136, 4143, and 4150 cm−1 for up to four H2 molecules in a 51264 cage (2L-4L). 2S peaks are not detected in both cases. (d) Schematic illustration of cage expansion and appearance of H2 doubly occupied cages.

experiment, LINKAM unit was used to control the sample temperature. The PXRD patterns were obtained using Rigaku D/max-IIIC diffractometer with Cu Kα as a light source (λ = 1.5406 Å) at a generator voltage of 40 kV and a generator current of 300 mA. Low-temperature stage attached to XRD kept the working temperature to 90K and step scan mode of 0.01°/3s was applied. The HRPD patterns were collected using PAL (Pohang Accelerator Laboratory) Synchrotron. During the measurements, the θ/2θ scan mode with a fixed time of 2 s, and a step size of 0.005° for 2θ = 0−120° and the beamline with a wavelength of 1.5472 Å were used for each samples. The loading of samples was performed at 77 K to minimize the possible sample damage. Computational Methods. To predict the hydrogen storage capacity of sII-S depending on pressure, we used the

grand canonical Monte Carlo (GCMC) simulations as implemented in the sorption module of the Cerius2.29 We prepared three different sII hydrate structure by varying the lattice parameters. The partial charge distribution of guest molecules (THF, propane, and isobutane) are defined by using QEq model,30 and each guest molecule is located at the center of large cage of sII hydrate. To describe the interaction of H2 with H2O, we used force field (FF) designed by Pascal et al.26 that is carefully fitted to reproduce the quantum mechanical energetics. Water molecules are described using TIP4/ice water model,31 and H2 molecule contains pseudoatom at the H2 bond midpoint. GCMC simulations were carried out for 107 trials with three different guest molecules (THF, propane, and isobutane) as a function of pressure at 243 K. Among these trajectories, we analyzed the last 8 000 000 trajectories to obtain the ensemble averaged values. 3326

dx.doi.org/10.1021/jp410632q | J. Phys. Chem. C 2014, 118, 3324−3330

The Journal of Physical Chemistry C

Article

Figure 3. Schematic illustration and corresponding Raman spectra of the reaction product from the stoichiometric sample (a) and the nonstoichiometric sample (b). When propane has already built the solid hydrate networks with a larger lattice parameter, hydrogen molecules easily form hydrogen double clusters in the empty dodecahedral cavities. In the propane-diluted sample, however, a guest tuning effect takes place and the hydrogen double cluster is not detected.



RESULTS AND DISCUSSION When solid propane (or isobutane) gas hydrate particles (