![]() |
INSTITUTUL NAŢIONAL DE
CERCETARE DEZVOLTARE PENTRU TEHNOLOGII IZOTOPICE SI MOLECULARE Str. Donat 67-103, 400293, Cluj-Napoca, Romānia Tel: +40-264-584037; Fax: +40-264-420042 Email: itim@itim-cj.ro, web: www.itim-cj.ro |
![]() |
Phases Duration
Phase 1 The metal-organic compounds (MOF) are considered as promising for hydrogen storage applications due to extremely high surface area (4000-6000 m2/g Langmuir surface area), almost an infinite number of geometrical and chemical variations for structures and possibilities for post-synthetic modification. However, dispersive forces implied in physisorption (typically < 10 kJ/mol H2) cannot facilitate hydrogen uptake required by DOE target. MOFs reach DOEs demand for hydrogen storage only at 77K. Hydrogen storage on MOFs at ambient temperature is still a challenge.
High quality (high surface area, large pore volume) MOFs (IRMOF-8, MIL-101, HKUST-1) were synthesized using the methods developed by our group. MOFs characterizations were made by PXRD, FT-IR, RAMAN, TGA, nitrogen adsorption/desorption at 77K. Hydrogen adsorption was investigated by volumetric methods on a Sievert home-made device in ITIM which proved to supply accurate data in very good agreement with literature.October 2011 - December 2011 (3 months)
Phase 2 There is a controversy due to contradictory literature results on hydrogen uptake enhancement in some un-bridged and bridged composites MOF-Pt/carbon mixtures. Promising results for room temperature hydrogen adsorption on MOFs were reported, assigned to spillover effect (dissociation of hydrogen on Pt and subsequent migration of atomic hydrogen onto MOF's surface). On the other hand, other results disagree with the significant hydrogen adsorption enhancement due to addition of noble metal catalysts presented above. The question is still open and needs to be solved due to its possible high impact on hydrogen storage.
IRMOF-8, HKUST-1 and MIL-101 bridged and un-bridged composites with Pt on activated carbon were synthesized with the aim to reproduce the considerable enhancement of hydrogen storage at ambient temperature reported in literature. X-ray powder diffraction patterns, surface areas and pore volumes of the prepared composites are comparable with the literature data. However, the hydrogen sorption isotherms at room temperature of any composite do not exhibit any signi?cant enhancement of hydrogen uptake as compared to the starting MOFs materials. In order to avoid such controversial situation it must be done measurements in really independent laboratories on the same sample. In this way would prove without doubt if the high enhancements reported are due to some subtleties in the bridging procedure or to experimental errors in the measurement techniques at high hydrogen pressure.
Composites of prepared MOFs with Pt/AC were prepared by ball-milling, characterized (PXRD, FT-IR, RAMAN, TGA, nitrogen adsorption/desorption at 77K).
The home-made Sievert device was extended (design, execution, functionality tests) with a unit for high pressure hydrogen isotherms (up to 400 bar).January 2012 - December 2012
(12 months)
Phase 3 Composites of MOFs with Pt on different supports (AC, carbon nanotubes, graphene) were prepared by direct synthesis and characterized (PXRD, FT-IR, RAMAN, TGA, nitrogen adsorption/desorption at 77K). Each prepared composite had surface area and volume pore smaller than metal-organic framework synthesized in the same condition. Hydrogen adsorption isotherms were measured at 77 K and room temperature.
There are two types of DOE targets (volumetric and gravimetric) to reach for mobile applications. The volumetric storage capacity, the sum of excess (adsorbed) hydrogen contained within the pores of the material and in the intergranular voids, can be improved by minimizing the intergranular voids. The most logical way to made this is by densification of the storage material. Pellets of different densities (up to crystal density) from MIL-101 were prepared and characterized (PXRD, FT-IR, TGA, nitrogen adsorption/desorption at 77K). Their XRD patterns do not exhibit notable changes with compression. The pellet surface area and pore volume decrease gradual with the increasing density and sharp at density close to the crystal density. The hydrogen excess evolve in the same manner but the total volumetric storage capacity reach values of 40 g/L for certain densities at 77K.January 2013 - December 2013
(12 months)
Phase 4 Ball milling of HKUST-1 and Pt/AC mixtures within of wide range of energies revealed that even low impact energies do not lead to enhancement of hydrogen adsorption capacity. The infrared spectra show that mechanical energy transferred to the sample both by ball milling or compression as monoliths result in local changes within the metal-ligand environment but the crystal lattice does not change as shown by XRD. Increasing ball milling energy leads to the decrease of surface area/ pore volume and consequently to a decrease of H2 adsorption capacity.
Composites Ni@MIL-101, prepared by impregnating MOF with certain precursor followed by its thermally decomposition, were obtained and characterized. Hydrogen adsorption isotherms showed a little lower sorption capacity as compared to pure MIL-101, in agreement with literature data on in Ni@MIL-101.
A compressed monolith with an intermetallic compound does not reduce its adsorption capacity and might contribute to increase the volumetric capacity – an important factor for applications. This promising direction will be continued to find the optimum intermetallic compound.
Significant advances in cryo-adsorption hydrogen storage was achieved with compressed monoliths of various densities – a promising direction for applications to hydrogen powered fuel cell vehicles. MIL-101 was compressed as pellets of different shapes and hydrogen adsorption was studied. The surface area and pore volumes decrease when approaching the crystal density of the metal-organic framework. However, the slow decrease of hydrogen adsorption capacity is compensated by the possibility to fill in a storage container with more the double of the tapped MOFs density. It was investigated the superior hydrogen storage capacity of MIL-101, powder and pellets, at hydrogen boiling point (Max Planck Institute - Stuttgart) revealing one of the highest limit of adsorption capacity reported for MOFs: 13.5wt.% for powder and 10.5 wt. % for pellets of 0.40 g/cm3 density.
Hexagonal pellets were prepared using a compressing device manufactured in INCDTIM. This shape is the best possible to achieve a maximum occupation degree of a hydrogen storage tank and to increase the volumetric capacity. These pellets show a remarkable mechanical integrity even after adsorption measurements. Hydrogen adsorption isotherms were measured in the temperature range 77-160K, of interest for cryo-adsorption storage method. 46.5 g/L volumetric capacity was obtained at 77K and 150 bar – a value satisfying the DOE targets for 2015 and a deliverable capacity of 45 g/L can be obtained by discharging down to 5 bar at 159K. These, together with its high stability towards water, make MIL-101 a promising candidate for applications. The isosteric enthalpy of hydrogen adsorption evaluated with van’t Hoff equation from the experimental isotherms using fugacity is in good agreement with the calorimetric heat of adsorption reported in literature.January 2014 - December 2014
(12 months)
Phase 5
High surface area templated carbon (CT) was synthesized according to literature [Yang et al. Carbon 49 (2011) 1305] by the chemical vapor deposition method from acetylene utilizing zeolite Y as support. After processing and purification, the resulting CT, with surface area 2500 m2g-1 and pore volume 1.2 cm3g-1, was doped with palladium via impregnation with H2PdCl4 and reduction to Pd metal with sodium borohydride solution. The dispersion measured by TEM is rather homogeneous, showing most of Pd as nanoparticles of 3-6 nm and some larger agglomerations. The resulting Pd@CT was characterized and hydrogen adsorption isotherms were measured by the volumetric Sievert method at 77K and ambient temperature. The hydrogen adsorption capacity shows a typical maximum Nex=4.5%H at 77K, higher than the value ~3.9% reported for CT in literature referred to above. The comparison of hydrogen adsorption isotherms at 295K revealed a small decrease of Nex for Pd@CT as compared to the starting CT for doping experiment, contrary to some other studies reported by other authors for higher Pd contents.
The effect of other metals on the possible enhancement of hydrogen adsorption in metal-organic frameworks (MOF) was extended, using a compressed pellet of MIL-101 with a metal hydride (MH) which absorbs hydrogen (a bulk process with hydrogen entering the metal as atoms and occupying the lattice interstices). Contrary to molecular physisorption of hydrogen in MOFs, in metal hydrides, the hydrogen molecule dissociates on the surface and diffuses as atoms in the metal lattice. We supposed that the dissociation on MH could serve as source for the transfer of hydrogen to MOF via the "spillover" effect – the mechanism by which some literature reports explained the enhancement of hydrogen storage capacity in MOF:Pt/activated carbon. In compressed pellets, the close contact MOF/MH might ensure the effect. Pellets of 8-14% Zr0.1Ti1.9Fe1.7V0.3 (an intermetallic with the enthalpy of hydrogen absorption ~14 kJ/mol) mixed with MIL-101 and compressed to densities 0.4-0.46 g.cm-3 have been studied. Up to 170 bar H2 no enhancement effect for hydrogen adsorption by MIL-101 was detected as compared to pure MIL-101 pellet so that even if the spillover effect exists in this case, the sorption capacity remains unchanged.
Another promising direction emerging from our skills in compressed MOFs pellets (monoliths) was the increase of their thermal conductivity – a key parameter for mobile applications of hydrogen storage in sorbents, e.g. fuel cells based electric vehicles. Monoliths with additives which increase the thermal conductivity were prepared, temperature and pressure dependence of the hydrogen adsorption/desorption kinetics has been studied. The adsorption rate was measured and rate constants within the cryo-adsorption temperature range 77-160K have been obtained using different kinetic equations. A composite was obtained with ~95% completion of adsorption in less than 2 min, suitable for applications in cryo-adsorption but also interesting from the fundamental point of view as long as there are almost no data reported in literature on hydrogen sorption kinetics by monoliths, except MOF-5 at 77K and high pressure (DOE Annual Merit Review 2013). A manuscript is in preparation to be sent for publication.January 2015 - December 2015
(12 months)
Phase 6 MIL-101 was synthesized directly on templated carbon impregnated with 4-10 % Pd. The resulting composite Pd/CT@MIL-101 shows a decreased BET surface area of 2269 m2g-1 and a pore volume Vp= 1.14 cm3g-1, much lower values as compared to 3725 m2g-1 and pore volume 1.91 for pure MIL-101 synthesized by the same method in the same conditions. The thermal analysis shows that the composite contains 75% MIL-101 and 25% Pd/CT from which it can be said that the surface area of the composite Pd/CT@MIL-101 does not result from the sum of the components normalized to their ratio. These significantly decreased characteristics of MIL101@Pd/CT reveal some interactions at the contacts between the components as a result of the direct synthesis, which might lead to blocking of some adsorption sites for nitrogen adsorption.
In the powder XRD spectra of MIL101@Pd/CT, the presence of Pd is evidenced by the line at 2θ of ~ 40° and the rather intense line at 2θ of ~ 17° indicates a significant content of terephthalic acid, even if the same procedure was applied for purification as for pure MIL-101, in which this precursor is absent. Though the reaction conditions were the same, it seems that the direct synthesis does not lead to a composite with higher surface area than MIL-101.
After activation at 493K and 2x10-5 mbar for 20 h, the hydrogen adsorption isotherms were measured at 295K and 77K for MIL-101 and Pd/CT@MIL101. Measurements at 295K and low pressure were carefully performed in order to evaluate the hydrogen amount corresponding to PdH0.67 palladium hydride, which has to be found in the composite at hydrogen pressures below 0.03 bar, if the Pd nanoparticles were well activated. From the hydrogen isotherms it was found that there is ~ 2.3% Pd content in the composite in agreement with ~ 2.5 %Pd from the analysis of the composite (75% MIL-101 and 25% Pd/CT). The hydrogen amount adsorbed at low pressure proves that the surface of Pd is activated and corresponds to the formation of PdH0.67 hydride.
A complementary method was also utilized to check the activity of Pd surface in composite: the hydrogen/deuterium isotopic exchange, with the analysis of the components (H2 , HD and D2) measured by mass spectrometry. The analysis of the mass spectra of the H2/D2 mixture extracted from the sample cell shows indeed a high increase of ~ 20 times of the HD/D2 as compared to the initial composition of gas, which proves that Pd is active for the dissociation of the hydrogen molecules and it can be a source of hydrogen atoms.
Hydrogen adsorption isotherms at 295K and 77K in the range 0.03-100 bar are reported for Pd/CT@ MIL-101 and MIL-101. The comparison reveals lower adsorption capacity for the composite at both temperatures. At 295K there is a linear increase with pressure, with a lower slope for the composite. At 77K, the maximum capacity is 5.4% H2 for MIL-101 and 4% H2 for Pd/CT@MIL101. No enhancement is found for the composite even if the Pd nanoparticles were proved to be active for hydrogen dissociation and spillover. It can be concluded that there is no significant contribution of the spillover effect to an enhancement of adsorption capacity for MIL-101 supported on Pd/CT catalyst. The results are in agreement with other experiments reported in literature on MIL-101 doped with Pd.
The practice accumulated in the synthesis of MIL-101 resulted in the development of a new, fast and environment friendly method of purification, suitable for scale extension, registered as patent claim.January 2016- September 2016
(9 months)