pubs.acs.org/Langmuir © 2010 American Chemical Society
Accelerated Formation of THF-H2 Clathrate Hydrate in Porous Media Dipendu Saha and Shuguang Deng* Chemical Engineering Department, New Mexico State University, Las Cruces, New Mexico 88003 Received December 23, 2009. Revised Manuscript Received January 28, 2010 Porous media were used to control the hydrogen clathrate particle size in order to accelerate its formation kinetics. Stoichiometric tetrahydrofuran-hydrogen binary clathrate hydrates with ∼1 wt % hydrogen loading formed in the mesopores of four porous media with median pore diameters of 49, 65, 100, and 226 A˚ at 270 K and hydrogen pressure of 65 bar. The minimum formation time for the tetrahydrofuran-hydrogen binary clathrate hydrates was 27 min in a porous medium with a median pore diameter of 49 A˚, which is 6-22 times faster than the tetrahydrofuran-hydrogen binary clathrate hydrates formed in the bulk ice. The clathrate formation time was found to increase with pore size of the porous media. A modified shrinking core kinetic model was used to calculate the diffusivity of hydrogen in the tetrahydrofuranhydrogen binary clathrate hydrates. Hydrogen diffusivities in the tetrahydrofuran-hydrogen binary clathrate hydrates were found to be on the order of 10-18-10-19 m2/s and decrease with increasing pore size or clathrate particle size.
1. Introduction Clathrates are crystalline solid compounds that consist of a host and one or more guest molecules are usually stable at elevated pressures and low temperatures. The well-known clathrate hydrates are typically generated with water (ice) as the host and small gaseous molecules including methane, ethane, propane, argon, xenon, ammonia or chlorine as the guest. Several years ago, researchers tended to believe that hydrogen clathrate hydrates do not exist because hydrogen molecules are too small (kinetic diameter of 2.72 A˚) to be stabilized inside the cavities of a host. In 1994, Udachin et al.1 stabilized tetrahydrofuran(THF-) water clathrate hydrates with hydrogen molecules, but did not report detailed structural information on this clathrate and hydrogen storage data. Mao et al.,2 for the first time, synthesized a pure hydrogen clathrate hydrate at an extreme condition of 200 MPa and 280 K, and reported a hydrogen storage amount of 5.3 wt %. However, the extreme synthesis condition hindered its real-world applicability as a potential hydrogen storage material. Later, Fluresse et al.3 improved Mao’s work by synthesizing a stoichiometric THF/H2O (5.56 mol % THF) binary clathrate hydrate at a less extreme condition of 7 MPa and 280 K, and reported a hydrogen uptake of 1 wt %. The THF/H2 binary clathrate hydrate crystallizes with a structure-II (CS-II) morphology that has 16 small 512 and eight large 51264 cavities within the crystal.4 This binary H2 clathrate has attracted quite a few researchers worldwide to investigate its synthesis and properties since Fluresse et al. published their research work.3 Smaller cage occupancy relating to the weight percent capacity was the most controversial result in this binary *Corresponding author. Telephone: 575-646-4346. Fax: 575-646-7706. E-mail:
[email protected]. (1) Udachin, K.; Lipkowski, J.; Tzacz, M. Supramol. Chem. 1994, 3, 181. (2) Mao, W. L.; Mao, H.; Goncharov, A. F.; Struzhkin, V. V.; Guo, Q.; Hu, J.; Shu, J.; Hemley, R. J. Science 2002, 297, 2247. (3) Florusse, L. J.; Peters, C. J.; Schoonman, J.; Hester, S. C.; Koh, C. A.; Dec, S. F.; Marsh, K. N.; Sloan, E. D. Science 2004, 306, 469. (4) Sloan, E. D. Clathrate Hydrates of Natural Gases, 2nd ed.; Marcel Dekker: New York, 1998. (5) Lee, H.; Lee, J.; Kim, D. Y.; Park, J.; Seo, Y.-T.; Zeng, H.; Moudrakovski, I. L.; Ratcliffe, C. I.; Ripmeester, J. A. Nature 2005, 434, 743.
8414 DOI: 10.1021/la904857e
clathrate. Lee et al.5 reported a 4 wt % hydrogen storage as a result of multiple small cage occupancies, but this claim could not be verified by other researchers. The results of all the later publications were consistently around 1 wt % of hydrogen storage capacity, which was later verified by Hester et al.6 by a neutron diffraction study and Fourier mapping. In an effort to enhance the hydrogen uptake, the THF concentration was lowered to 0.5 mol % in a nonstoichiometric hydrate,7 but this approach did not increase the hydrogen storage capacity. Very recently, Sughara et al.8 lowered the THF concentration below the eutectic composition of THF and H2O binary mixture (THF mole fraction e 0.0106) which raised the H2 uptake up to 3.4 wt %. Anderson et al.9 experimentally determined the phase relations of the binary clathrates at temperatures between 260 and 290 K and 45 MPa. Strobel et al.10 and Hashimoto et al.11 performed thermodynamic studies on binary clathrate hydrates. Raman spectroscopic investigations of the binary clathrates were made in details by Ogata et al.12 and Strobel et al.13 Several other research works on this THF/H2 binary clathrate hydrate were also reported by Mulder et al.,14 Su et al.15 or Talyzin.16 In a different approach, Saha and Deng17 increased the hydrogen uptake in an ordered mesoporous carbon by clathrate mediated adsorption within its pores. A few other types of hydrogen clathrates including TBAB semiclathrates,18,19 (6) Hester, K. C.; Strobel, T. A.; Sloan, E. D.; Koh, C. A.; Huq, A.; Schutz, A. J. J. Phys. Chem. B 2006, 110, 14024. (7) Strobel, T. A.; Taylor, C. J.; Hester, K. C.; Dec., S. F.; Koh, K. C.; Miller, K. T.; Sloan, E. D. J. Phys. Chem. B 2006, 110, 17121. (8) Sughara, T.; Haag, J. C.; Prasad, P. S. R.; Warntjes, A. A.; Sloan, E. D.; Sum, A. K.; Koh, C. A. J. Am. Chem. Soc. 2009, 131, 14616. (9) Anderson, R.; Chapoy, A.; Tohidi, B. Langmuir 2007, 23, 3440. (10) Strobel, T. A.; Koh, C. A.; Sloan, E. D. Fluid Phase Equilib. 2009, 280, 61. (11) Hashimoto, S.; Muryama, S.; Sughara, T.; Sato, H.; Ohgaki, K. Chem. Eng. Sci. 2006, 61, 7884. (12) Ogata, K.; Hashimoto, S.; Sughara, T.; Moritoki, M.; Sato, H.; Ohgaki, K. Chem. Eng. Sci. 2008, 63, 5714. (13) Strobel, T. A.; Koh, C. A.; Sloan, E. D. Fluid Phase Equilib. 2007, 261, 382. (14) Mulder, F. M.; Wagemaker, M.; Van Eijck, L.; Kearley, G. J. ChemPhysChem. 2008, 9, 1331. (15) Su, B.; Christopher, L. B.; Cooper, A. L. Adv. Mater. 2008, 20, 2663. (16) Talyzin, A. Intl. J. Hydrogen Energy 2008, 33, 111. (17) Saha, D.; Deng, S. Intl. J. Hydrogen Energy 2009, 34, 8583. (18) Chapoy, A.; Anderson, R.; Tohidi, B. J. Am. Soc. 2007, 129, 746. (19) Strobel, T. A.; Koh, C. A.; Sloan, E. D. Fluid Phase Equilib. 2007, 361, 282.
Published on Web 02/11/2010
Langmuir 2010, 26(11), 8414–8418
Saha and Deng
Article Table 1. Identity of Different Pore Sized Materials
pore size (A˚)
materials
vendor or lab synthesis
49 65 100 226
activated alumina synthesized by sol-gel technique 30 -DMT-da(Bz)-Suc-CPG silica gel packing materials of a HPLC column separated from the binder
synthesized in our lab by a sol-gel technique23 Biosearch Technology Sigma-Aldritch Sigma-Aldritch
hydroquinone20 or cyclohexanone clathrates21 were also reported recently. Despite extensive experimental studies on the synthesis, characterization and unique properties of the THF-H2 binary clathrate hydrates were reported, no effort was attempted to lower the clathrate formation time to make it a practical and efficient hydrogen storage technique. The reported clathrate formation kinetics of binary clathrates and pure clathrates were quite slow, and the clathrate formation period lasted for several days in the bulk ice particles. In order to lower the formation time, big ice particles were crashed into fine particles before they were exposed to pressured hydrogen, cutting the formation time to 3-10 h,12,16 which is still unsuitable for practical hydrogen storage application. Another disadvantage of introducing crushed ice particles is that the ice particles will regain its bulk shape after melting, making it unsuitable for future cycles of enclathration irrespective of its initial size range. Lokshin et al.22 minimized the formation of pure hydrogen clathrate hydrate to an extremely low value of 2 min by introducing hexagonal ice-Ih structure to the system. However, the increasing trend of formation time from 2 to 30 min with the lowering of pressure from 2000 to 500 bar suggests that the formation time will further increase if the formation pressure is lowered to less than 100 bar, which is the sufficient condition for synthesizing binary clathrate hydrates. The aim of this present study is to develop a technique to lower the synthesis time of binary clathrates. Despite the fact that the hydrogen storage amount itself is said to be the main drawback of THF/H2 binary clathrate hydrates to be used in reality, the sluggish kinetics is also a contributing problem of its usage and this work provides an approach to improve its kinetics of formation. It is clear from the literature that lowering the particle size lowered the formation time of clathrates, so lowering the particle size even more could decrease the formation time to a desirable range. To apply this principle, we choose porous materials with four different pore sizes, 49, 65, 100, and 226 A˚ to form clathrates within the pores. Another advantage of introducing porous structure is that the ice particle size will remain intact even after melting thereby keeping the formation time unaltered in any number of enclathration cycles. We also examine the pore size/ice particle size effects on the binary clathrate formation. A macroscopic model was used to estimate the diffusivity of hydrogen in the binary THF/H2 clathrates.
2. Experimental Methods 2.1. Materials. The four porous materials with different pore sizes and their sources were summarized Table 1. The median pore size of all the materials were measured and monitored by N2 adsorption-desorption study at 77 K repeatedly before and after each set of experiments to ensure that pore size remains intact. The original pore size of as received [30 -DMT-da(Bz)-Suc-CPG] was (20) Strobel, T. A.; Kim, Y.; Gary, S.; Jack, R, A. J. Am. Chem. Soc. 2008, 130, 14975. (21) Strobel, T. A.; Hester, K. C.; Sloan, E. D.; Koh, C. A. J. Am. Chem. Soc. 2007, 129, 9544. (22) Lokshin, K. A.; Zhao, Y. Appl. Phys. Lett. 2006, 88, 131909.
Langmuir 2010, 26(11), 8414–8418
Figure 1. Kinetic plots of clathrate formation in porous media. 1000 A˚, but it was shrunk to 65 A˚ after the first addition of H2O/ THF mixture by some chemical or physical or combined alterations which were unknown to us. But for the remaining experiments, the pore size remained unchanged which were confirmed by repeated N2 adsorption-desorption study. We also rule out the possibility of chemical reaction of these materials with THF/ H2O mixture as the pore size did not alter after the experiments. For each type of material, around 1 g of the sample has been employed. 2.2. Clathrate Formation. The high pressure experiment was performed gravimetrically in a Rubotherm magnetic suspension balance at slightly below the freezing point of water (270 K) and hydrogen pressures up to 50-100 bar. The process flow diagram for the Rubotherm balance was shown in our previous publications.24,25 Like all other gravimetric devices, this balance was also pre-examined with the blank run of empty balance and volume run sample loaded balance in order to measure the weight and volume of empty sample holder and sample itself before introducing the particular gas of interest, in this case, hydrogen. The detailed operation procedures were described in our previous publications.24,25 The four porous materials were initially checked for their capability of adsorbing hydrogen in the dry state at identical clathrate formation conditions (270 K and up to 100 bar) to confirm that the pure materials do not adsorb hydrogen at all in that experimental conditions. After the evacuating and volume measurement runs, the adsorbent sample was removed from the balance sample holder, loaded with H2O/THF mixture (in 1:0.056 mol ratios, which is the stoichiometric ratio for the CS-II structure), and put on a shaker for 10 min to ensure that H2O/ THF mixture was completely dispersed into the pores. The H2O/ THF loaded samples were then put back to the balance sample holder and cooled the system with liquid nitrogen before connecting to a vacuum line to ensure that THF was not desorbed during the evacuation step. After the system was cooled to the desired temperature, helium gas was introduced to the system for the volume run. When volume run was complete, the system was evacuated to remove helium keeping the freezing temperature (23) Deng, S.; Lin, Y. S. AIChE J. 1997, 43, 505. (24) Saha, D.; Wei, Z.; Deng, S. Intl. J. Hydrogen Energy 2008, 33, 7479. (25) Saha, D.; Deng, S. Langmuir 2009, 25, 12550.
DOI: 10.1021/la904857e
8415
Article
Saha and Deng
Figure 2. Variation of clathrate formation time with median pore diameter. intact. Then, 99.999% pure hydrogen gas was introduced to the system before raising the system temperature to 270 K and maintained at this value for the entire experiment time. Hydrogen flow rate maintained in a moderate range of 200 mL/min in the entire set of experiment and pressure was increased periodically to a maximum of 100 bar or when the system reached equilibrium. The kinetic data of hydrogen adsorption in the clathrate formation run was also recorded automatically. The hydrogen loaded THF-H2 binary clathrate hydrates were then undergone a desorption step by reducing the pressure to 1 bar and increasing the system temperature to 298 K. The hydrogen adsorption/ desorption cycle was repeated several times to ensure reproducibility of the experimental data and reversibility of the adsorption-desorption processes.
3. Results and Discussion 3.1. Kinetic Study. The hydrogen loadings in the THF-H2 binary clathrate hydrates formed in all four porous media were approximately 1.0 wt %, which is consistent with the published data on THF-H2 binary clathrate hydrates.6-11 Although it was claimed that the THF-H2 clathrates would form in bulk ice at 50 bar and 273 K, it was observed in our experiments that the pressure required for reaching the equilibrium was 65 bar. The excess pressure requirement could be attributed to the capillary force generated in the mesopores, which reduces the activity of water in the pores resulting in a freezing point depression.26 Handa et al.27 observed a similar phenomenon that 20-70% excess pressure was required to form methane and propane hydrates in a silica gel with a median pore width of 70 A˚. The typical kinetic plots (fractional hydrogen uptake versus time) for THF-H2 binary clathrate hydrate formation in four porous media are shown in Figure 1, and the effect of pore size on the average formation time is plotted in Figure 2. It can be observed that the average clathrate formation time in the porous medium with a median pore diameter of 49 A˚ pore size was about 27 min, which is the shortest formation time for THF-H2 binary clathrate hydrates ever reported. As expected the formation time increased with the pore size of the porous media, which can be attributed to larger clathrate particle size generated in the meso(26) Clarke, M. A.; Pooladi-Darvish, M.; Bishnoy, P. R. Ind. Eng. Chem. Res. 1999, 38, 2485. (27) Handa, Y. P.; Stupin, D. J. Phys. Chem. 1992, 96, 8599.
8416 DOI: 10.1021/la904857e
Figure 3. Schematic representation of the modified Jander’s kinetic model.
pores. The average time of clathrate formation in the porous media with median pore size of 65, 100, and 226 A˚ were 50, 154, and 304 min, respectively. Although it could be argued that the formation time can be further shortened if a microporous material is used, the large size of the lattice in the CS-II structure (17.3 A˚) would prohibit formation clathrates in micropores. 3.2. Estimation of Intraparticle Diffusivity. To estimate the diffusivity of hydrogen in the clathrates, we used a modified shrinking core kinetic model originally developed by Jander28 and further modified by several other researchers for modeling argon,29 carbon dioxide30 and methane31 hydrate formation kinetics. The model assumes a sphere of radius r of material A reacting with a mobile phase (gas) of material B as illustrated in Figure 3. The product thickness l is inversely proportional to the rate of reaction, dl/dt. dl k ¼ dt l
ð1Þ
(28) Jander, W. Z. Anorg. Allg. Chem. 1927, 163, 1. (29) Halpern, Y.; Thieu, V.; Henning, R. W.; Wang, X.; Schultz, A. J. J. Am. Chem. Soc. 2001, 123, 128126. (30) Henning, R. W.; Schultz, A. J.; Thieu, V.; Halpern, Y. J. Phys. Chem. A 2000, 104, 5566. (31) Wang, X.; Schultz, A. J.; Halpern, Y. J. Phys. Chem. A 2002, 106, 7304.
Langmuir 2010, 26(11), 8414–8418
Saha and Deng
Article
Figure 4. Linear plot of (t - t*) versus (1 - R)1/3.
Integrating the above equation gives l 2 ¼ 2kt
ð2Þ
where k (m2/s) is a proportionality constant representing the intraparticle diffusivity of mobile phase (B) in the solid (A). The volume of the unreacted sphere is given by 4π ðr - 1Þ3 3
ð3Þ
4π 3 r ð1 - RÞ 3
ð4Þ
V ¼
V ¼
where R is the fraction of the original sphere that has reacted. Equating right side of eqs 3 and 4 gives 1 ¼ rð1 - ð1 - RÞ1=3 Þ
ð5Þ
Substituting eq 5 into eq 2 and rearranging yields t ¼
r2 ð1 - ð1 - RÞ1=3 Þ2 2k
ð6Þ
This model is based on the assumption that an initial product layer is formed rapidly after the solid phase A is exposed to the mobile phase B, after which the reaction becomes diffusion controlled and the reaction kinetic can be described by eq 6. Fuji and Kondo32 modified the above model by introducing a definite period of time t* after which the reaction becomes diffusion controlled at a conversion R*. The modified model equation becomes ð1 - RÞ1=3 ¼
-ð2kÞ1=2 ðt - t Þ1=2 þ ð1 - RÞ1=3 r
ð7Þ
The value of t* and R* can be determined by trial and error technique so the linear fit of (t - t*) versus (1 - R)1/3 plot becomes the best fit. It was reported that R* lies within 10 to 20% of the total conversion.29-31 In our study, a trial and error technique also gives a R* value of 20%. (32) Fujii, K.; Kondo, W. J. Am. Ceram. Soc. 1974, 57, 492.
Langmuir 2010, 26(11), 8414–8418
Figure 5. Effect of pore size on hydrogen diffusivity in clathrates.
Figure 4 is the linear plots of (t - t*) versus (1 - R)1/3 for all the kinetic shown Figure 2. It can be observed that the modified kinetic model described in eq 7 fits the kinetic data well with a relatively high regression coefficient (R2 >0.97), suggesting the modified kinetic model can be used to describe the formation kinetics of THF-H2 binary clathrate hydrates in porous media. The diffusivity of the hydrogen in the clathrates formed in the porous media were calculated from the slope of the linear plot -ð2kÞ1=2 r shown in Figure 4. The radius of the clathrates is assumed to be the same as the radius of the pores. The effect of median pore diameter on diffusivity was illustrated in Figure 5. The order of magnitude of the hydrogen diffusivity in the tetrahydrofuranhydrogen clathrate hydrates is 10-18-10-19 m2/s. It is observed that the diffusivity decreases with pore size or clathrate particle size, which is consistent with what was stated by Halpern et al.32 and Henning et al.30 that gas diffusion constants in clathrates depend on the clathrate particle size.
4. Conclusion Stoichiometric tetrahydrofuran-hydrogen binary clathrate hydrates with ∼1 wt % hydrogen loading formed in the mesopores of four porous media with a median pore diameter of 49, 65, 100, and 225 A˚ at 270 K and hydrogen pressure of 65 bar. DOI: 10.1021/la904857e
8417
Article
The clathrate formation kinetics were greatly accelerated by using the porous media to control the clathrate particle size. The formation time for the THF-H2 binary clathrate hydrates was cut from 3 to 10 h in bulk ice to 27 min in a porous medium with a median pore diameter of 49 A˚. It was observed that the clathrate formation time
8418 DOI: 10.1021/la904857e
Saha and Deng
increases with pore size of the porous media. A modified shrinking core kinetic model was employed to estimate the diffusivity of hydrogen in the THF-H2 binary clathrate hydrates. The hydrogen diffusivities in the THF-H2 binary clathrate hydrates decrease with increasing pore size or clathrate particle size.
Langmuir 2010, 26(11), 8414–8418
