Letter pubs.acs.org/NanoLett
Synthesis and Characterization of a Lithium-Doped Fullerane (Lix-C60Hy) for Reversible Hydrogen Storage Joseph A. Teprovich, Jr.,† Matthew S. Wellons,† Robert Lascola,† Son-Jong Hwang,‡ Patrick A. Ward,§ Robert N. Compton,§ and Ragaiy Zidan*,† †
Clean Energy Directorate, Savannah River National Lab, P.O. Box A, Aiken, South Carolina 29808, United States Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States § Department of Chemistry, University of Tennessee, Knoxville, Tennessee 37996, United States ‡
S Supporting Information *
ABSTRACT: Herein, we present a lithium-doped fullerane (Lix-C60-Hy) that is capable of reversibly storing hydrogen through chemisorption at elevated temperatures and pressures. This system is unique in that hydrogen is closely associated with lithium and carbon upon rehydrogenation of the material and that the weight percent of H2 stored in the material is intimately linked to the stoichiometric ratio of Li:C60 in the material. Characterization of the material (IR, Raman, UV−vis, XRD, LDI-TOF-MS, and NMR) indicates that a lithiumdoped fullerane is formed upon rehydrogenation in which the active hydrogen storage material is similar to a hydrogenated fullerene. Under optimized conditions, a lithium-doped fullerane with a Li:C60 mole ratio of 6:1 can reversibly desorb up to 5 wt % H2 with an onset temperature of ∼270 °C, which is significantly less than the desorption temperature of hydrogenated fullerenes (C60Hx) and pure lithium hydride (decomposition temperature 500−600 and 670 °C respectively). However, our Lix-C60-Hy system does not suffer from the same drawbacks as typical hydrogenated fullerenes (high desorption T and release of hydrocarbons) because the fullerene cage remains mostly intact and is only slightly modified during multiple hydrogen desorption/absorption cycles. We also observed a reversible phase transition of C60 in the material from face-centered cubic to body-centered cubic at high levels of hydrogenation. KEYWORDS: Hydrogen storage, metal hydride, fullerene, fullerane, metal-doped C60
I
A major drawback to fulleranes as hydrogen storage materials is the high temperature (>500 °C) required for hydrogen desorption due to the increase in stability of the molecule as the degree of hydrogenation is increased. Concomitant with the desorption of hydrogen from hydrofullerenes is the release of hydrocarbons (i.e., CH4)5,10 or formation of polycyclic aromatic hydrocarbons (PAH) as the integrity of the C60 cage is compromised.5,7,13,14 This indicates the irreversible decomposition of the fullerene nanostructure, resulting in the loss of capacity over multiple dehydrogenation/rehydrogenation cycles. Some work has shown that milder conditions for the hydrogenation/dehydrogenation of fullerenes is possible when dissolved in organic solvent, however, the dehydrogenation was incomplete and/or the system was not shown to be reversible.15−17 Intercalation of C60 with alkali metals has shown that the physical and chemical properties of C60 can be significantly altered and are highly dependent on the type of metal and its molar ratio with C60 (MxC60).18−20 To date, the majority of
nterest in hydrofullerenes/fulleranes (C60Hx) with high hydrogen content (C60H18, 2.4 wt % and C60H36, 4.8 wt %) has increased due to their potential applications as hydrogen storage materials. Highly hydrogenated hydrofullerenes have been synthesized via a variety of methods including Birch reduction,1 hydrogen atom transfer from 9,10-dihydroanthracene,2 and through direct solid phase hydrogenation at elevated temperatures and pressures for extended periods of time (>400 °C and >100 bar H2).3 The degree of hydrogenation (C60:H ratio) of the fullerene molecule has been monitored by the appearance of sharp IR bands in the range of 3000−2700 cm−1 accompanied by a peak shift to lower 2θ in the X-ray diffraction (XRD) pattern.4−12 Through a series of XRD studies, it has been shown that there is a nonlinear relationship between the expansion of the cell parameter and hydrogen content because the rate at which the cell expands increases with an increase of hydrogen content. When the level of hydrogenation reaches ∼5 wt %, the lattice will continue to expand even though the hydrogen content of the fullerene does not increase, due to the poor stability of highly reduced fullerenes (>C60H36). This ultimately results in the fragmentation/collapse and increase in the amorphous nature of the material after extended periods of hydrogen exposure.5,7,13 © 2011 American Chemical Society
Received: September 1, 2011 Revised: December 14, 2011 Published: December 29, 2011 582
dx.doi.org/10.1021/nl203045v | Nano Lett. 2012, 12, 582−589
Nano Letters
Letter
“dead weight” and is not incorporated into the active hydrogen storage material. This results in the lower H2 capacity for the samples synthesized with these stoichiometries. There can also be lithium deficient materials (i.e., 3:1, 2:1), in which there is not enough Li present to fully form the active hydrogen storage material. A direct linear relationship exists between the mole ratio of C60:Li and the number of hydrogen atoms per lithium atom from 1:120 to 1:6 (Figure 1). The slope of this line is 32, which
studies examining the use of alkali fullerides as hydrogen storage materials are theoretical explorations focusing on intercalated or dispersed alkali metal atoms within the carbon matrix.21−24 Early theoretical work also suggests that it is possible for a positively charged lithium atom to associate with more than one hydrogen atom through a charge polarization mechanism.25 Preliminary experimental studies on alkali metal fullerides have demonstrated the ability to absorb hydrogen via chemisorption, however, the dehydrogenation of these alkaliintercalated fullerides was problematic (only 0.62 wt % H2 was desorbed at 200 °C at 100 bar H2).26 The hydrogen storage capabilities of metal intercalated carbon allotropes (i.e., graphite and nanotubes) have also been explored.27−34 Many of these carbon based systems are plagued by diminishing capacity over multiple cycles due to the formation of volatile hydrocarbons (CH4) as well as metal carbides during rehydrogenation/ dehydrogenation. Previous work from our group examined the effect of C60 on the hydrogen storage properties of NaAlH4, LiBH4, and LiAlH4 and determined that these materials are capable of reversibly storing hydrogen.35−39 Upon close examination of the LiAlH4:C60 material, it was determined that Al did not play a role in the material’s ability to reversibly store hydrogen and suggested that the active hydrogen storage material is a lithium intercalated fullerane (Lix-C60-Hy). These results indicated that it may be possible to develop new hydrogen storage materials based on simple metal hydrides (i.e., LiH), which are by themselves typically thought of as impractical for hydrogen storage due to the high temperature required for hydrogen desorption. This impetus led to the current work which examines the hydrogen storage properties of a series of lithiumdoped fulleranes (synthesized from LiH and C60 via solventassisted mixing).
Figure 1. Plot of number of hydrogen atoms per lithium atom vs the mole ratio of C60:Li. All samples were rehydrogenated at 250 °C and 105 bar H2 for 11 h. The wt % H2 was determined by TGA on the third desorption of the material.
is nearly identical to the value of the number of hydrogens (y) in the Lix-C60-Hy compositions that were rehydrogenated at 250 °C and 105 bar H2 (Supporting Information). Although the 1:3 and 1:2 ratios do not follow this trend, it still emphasizes the importance of lithium content in this system since they still contain >5 hydrogen atoms per intercalated lithium atom, which is a value very similar to the optimal 1:6 ratio. This indicates that lithium dictates the hydrogenation sites in the material and is intimately linked to the rehydrogenation/dehydrogenation of the material. The use of graphite in the place of C60 was also examined for comparison purposes (see Supporting Information). The materials prepared with graphite, intercalated with lithium, displayed very little hydrogen capacity under the same desorption/absorption conditions and further emphasize the importance of C60 in the system. Based on the hydrogen desorption/absorption properties of the samples examined, the optimum ratio of Li:C60 is 6:1 (the ‘as prepared’ composite contains 6.2 wt % LiH and 93.8 wt % C60) and will be the primary focus of the characterization. The effect of pressure and temperature on H2 capacity on this ratio was examined for comparison with previous reports on the hydrogenation of pure C60 via H2 over pressure and elevated temperature (Table 1 and Supporting Information). The rehydrogenation of the material was performed at H2 pressures from 25 to 105 bar and at temperatures from 200 to 350 °C. It was determined that for every increase of 25 bar of H2 pressure, an increase of approximately 1 wt % hydrogen content was observed (vs the entire material) when rehydrogenation was performed at 350 °C. The effect of increasing the rehydrogenation temperature had a similar influence on the H2 capacity, but that influence was lower than the pressure effect (when the rehydrogenation was performed at 105 bar
■
RESULTS AND DISCUSSION In order to determine the preferred stoichiometry of the LixC60-Hy system, a series of samples were prepared by solventassisted mixing in which the mol ratio of LiH:C60 was varied from 120:1 to 2:1. The as prepared samples were then subjected to a series of desorption/absorption cycles on a Sieverts apparatus. The as prepared and rehydrided materials were then analyzed by thermogravimetric analysis-residual gas analysis (TGA-RGA) to calculate the wt % of H2 desorbed from the material as well as any other volatile components. The RGA did not detect any volatile hydrocarbons during the dehydrogenation process, indicating that our material is thermally stable and did not decompose as previously reported for other carbon-based materials in which the mechanism of hydrogen storage was chemisorption. In general, the samples showed an unusual first desorption with the release of tetrahydrofuran (THF) as well as H2 upon heating. It is likely that the active hydrogen storage material (LixC60) is formed during this first desorption, which serves as an annealing step. After the samples were rehydrogenated, a much larger quantity of H2 was released at a much lower onset temperature (∼270 °C). The results of the desorption study are summarized in Supporting Information Table S1. Interestingly, the number of hydrogen atoms (y) stored in each stoichiometry (Lix-C60-Hy) is approximately the same value (∼32 H atoms) between mole ratios of 120:1 and 6:1, when the samples were rehydrided at 250 °C under 105 bar H2. This indicates that there is a preferred Li:C60 ratio and that the material will have an excess of lithium (i.e., 120:1 to 8:1) in which the additional LiH acts as 583
dx.doi.org/10.1021/nl203045v | Nano Lett. 2012, 12, 582−589
Nano Letters
Letter
prepared’ sample, is the expected hydrogen content of the material based on the amount of LiH used to make the sample. The release of H2 from the as prepared sample is attributed to a reaction between LiH and C60 (at temperatures >350 °C) to form a lithium-doped fullerene, which releases hydrogen gas in the process. The release of H2 from LiH was unexpected during this initial desorption because pure LiH does not decompose until it is exposed to temperatures above 670 °C. This suggests that the lithium intercalated fulleride produced during this reaction is the active hydrogen storage material in subsequent rehydrogenation/dehydrogenation cycles. After three rehydrogenation cycles (at 350 °C, 105 bar H2, 11 h), the lithium-doped fullerane (6:1 mol ratio) was able to desorb up to 5 wt % H2 with an onset temperature of ∼270 °C with no apparent loss in hydrogen capacity (black line). This indicates that the active hydrogen storage material is similar to Li6-C60-H40 when the sample was rehydrogenated at 350 °C and 105 bar H2. However, the value of 40 for the number of hydrogen atoms (y) is likely an average value because it is known that there is some variability in the degree of hydrogenation of pure C60 when it is performed with H2 overpressure to synthesize highly reduced species (see LDITOF-MS data, Supporting Information). The RGA was set to monitor for volatile hydrocarbons (i.e., methane, ethane), but none were detected. Although it is not readily apparent in the TGA, the RGA and differential thermal analysis (DTA) (see Supporting Information) indicate that there are at least two overlapping H2 desorption events and indicates that multiple active hydrogen storage sites exist in the material. This could be due the proximity of the chemisorbed hydrogen to the intercalated Li. For comparison purposes, a hydrofullerene was prepared by exposing a sample of pure C60 to the same desorption/ absorption conditions used for the lithium intercalated fulleride (i.e., mix with THF, dehydrogenate, and rehydrogenate at 350 °C under 105 bar H2 for 11 h). Although hydrogen desorption is observed in this sample, it occurs at a much higher temperature (∼500 °C) and at a lower weight percent (2.5 wt %) than our rehydrogenated lithium-doped fullerane. Accompanying the release of hydrogen from the sample is the release of methane and indicates that the irreversible decomposition/ fragmentation of the fullerene molecule occurs in this sample. Due to the release of CH4 from the sample, it is likely that actual H2 content of the material is less than 2.5 wt % (red line). The additional weight loss observed from this sample beyond 625 °C is attributed to the sublimation of C60. Figure 3 shows the laser desorption ionization time-of-flight mass spectra (LDI-TOF-MS) for the rehydrogenated and dehydrogenated form of the Li6-C60-Hy material as well as pure C60 that was hydrogenated under the same conditions. Upon rehydrogenation, the lithium intercalated C60 displays a higher degree of hydrogenation compared to the pure C60 hydrogenated (Figure 3a) under the same conditions. It is evident that fulleranes up to C60H48 are present as well as atomic lithium in hydrogenated samples. The presence of an ion corresponding to exohedral lithium on a hydrogenated fullerene has not been confirmed. This is likely due to the labile nature of lithium in the system because individual lithium ions can readily be detected in the LDI-TOF-MS in the rehydrogenated version (Supporting Information). There are many low mass peaks (
