Energy Storage

Carbon nanotubes are being considered for energy production and storage. Graphite, carbonaceous materials and carbon fiber electrodes have been used for decades in fuel cells, battery and several other electrochemical applications [50]. Nanotubes are special because they have small dimensions, a smooth surface topology, and perfect surface specificity, since only the basal graphite planes are exposed in their structure. The rate of electron transfer at carbon electrodes ultimately determines the efficiency of fuel cells and this depends on various factors, such as the structure and morphology of the carbon material used in the electrodes. Several experiments have pointed out that compared to conventional carbon electrodes, the electron transfer kinetics take place fastest on nanotubes, following ideal Nernstian behavior [51]. Nanotube microelectrodes have been constructed using a binder and

aAS^A

• Nanotubes a Commercial GDT (manufacturer I > ° Commercial GDT (manufacturer 2)

have been successfully used in bioelectrochemical reactions (e.g., oxidation of dopamine). Their performance has been found to be superior to other carbon electrodes in terms of reaction rates and reversibility [52]. Pure MWNTs and MWNTs deposited with metal catalysts (Pd, Pt, Ag) have been used to electro-catalyze an oxygen reduction reaction, which is important for fuel cells [53,54,55]. It is seen from several studies that nanotubes could be excellent replacements for conventional carbon-based electrodes. Similarly, the improved selectivity of nanotube-based catalysts have been demonstrated in heterogeneous catalysis. Ru-supported nanotubes were found to be superior to the same metal on graphite and on other carbons in the liquid phase hydrogenation reaction of cinnamaldehyde [55]. The properties of catalytically grown carbon nanofibers (which are basically defective nanotubes) have been found to be desirable for high power electrochemical capacitors [56].

2.1 Electrochemical Intercalation of Carbon Nanotubes with Lithium

The basic working mechanism of rechargeable lithium batteries is electrochemical intercalation and de-intercalation of lithium between two working electrodes. Current state-of-art lithium batteries use transition metal oxides (i.e., LiKCoO2 or LiKMn2O4) as the cathodes and carbon materials (graphite or disordered carbon) as the anodes [57]. It is desirable to have batteries with a high energy capacity, fast charging time and long cycle time. The energy capacity is determined by the saturation lithium concentration of the electrode materials. For graphite, the thermodynamic equilibrium saturation concentration is LiC6 which is equivalent to 372mAh/g. Higher Li concentrations have been reported in disordered carbons (hard and soft carbon) [58,59] and metastable compounds formed under pressure [60].

It has been speculated that a higher Li capacity may be obtained in carbon nanotubes if all the interstitial sites (inter-shell van der Waals spaces, inter-tube channels, and inner cores) are accessible for Li intercalation. Electrochemical intercalation of MWNTs [61,62] and SWNTs [63,64] has been investigated by several groups. Figure 8 (top) shows representative electrochemical intercalation data collected from an arc-discharge-grown MWNT sample using an electrochemical cell with a carbon nanotube film and a lithium foil as the two working electrodes [64]. A reversible capacity (Crev) of 100-640 mA h/g has been reported, depending on the sample processing and annealing conditions [61,62,64]. In general, well-graphitized MWNTs such as those synthesized by the arc-discharge method have a lower Crev than those prepared by the CVD method. Structural studies [65,66] have shown that alkali metals can be intercalated into the inter-shell spaces within the individual MWNTs through defect sites.

Single-walled nanotubes are shown to have both high reversible and irreversible capacities [63,64]. Two separate groups reported 400-650 mAh/g reversible and ^1000mAh/g irreversible capacities in SWNTs produced by the

Capacity(mAh/g)

0.0 500.0 1000.0 1500.0 Capacity(mAh/g)

Fig. 8. Top: Electrochemical intercalation of MWNTs with lithium. Data were collected using 50mA/h current. The electrolyte was LiClO4 in ethylene carbonate/dimethyl carbonate. Bottom: Charge-discharge data of purified and processed SWNTs. The reversible capacity of this material is 1000mAh/g. (Figures are from Gao et al. in [64])

0.0 500.0 1000.0 1500.0 Capacity(mAh/g)

Fig. 8. Top: Electrochemical intercalation of MWNTs with lithium. Data were collected using 50mA/h current. The electrolyte was LiClO4 in ethylene carbonate/dimethyl carbonate. Bottom: Charge-discharge data of purified and processed SWNTs. The reversible capacity of this material is 1000mAh/g. (Figures are from Gao et al. in [64])

laser ablation method. The exact locations of the Li ions in the intercalated SWNTs are still unknown. Intercalation and in-situ TEM and EELS measurements on individual SWNT bundles suggested that the intercalants reside in the interstitial sites between the SWNTs [67]. It is shown that the Li/C ratio can be further increased by ball-milling which fractures the SWNTs [68]. A reversible capacity of 1000mAh/g [64] was reported in processed SWNTs. The large irreversible capacity is related to the large surface area of the SWNT films (~300m2/g by BET characterization) and the formation of a solid-electrolyte-interface. The SWNTs are also found to perform well under high current rates. For example, 60% of the full capacity can be retained when the charge-discharge rate is increased from 50mA/h to 500mA/h [63].

The high capacity and high-rate performance warrant further studies on the potential of utilizing carbon nanotubes as battery electrodes. The large observed voltage hysteresis (Fig. 8) is undesirable for battery application. It is at least partially related to the kinetics of the intercalation reaction and can potentially be reduced/eliminated by processing, i.e., cutting the nanotubes to short segments.

2.2 Hydrogen Storage

The area of hydrogen storage in carbon nanotubes remains active and controversial. Extraordinarily high and reversible hydrogen adsorption in SWNT-containing materials [69,70,71,72] and graphite nanofibers (GNFs) [73] has been reported and has attracted considerable interest in both academia and industry. Table 2 summarizes the gravimetric hydrogen storage capacity reported by various groups [74]. However, many of these reports have not been independently verified. There is also a lack of understanding of the basic mechanism(s) of hydrogen storage in these materials.

Table 2. Summary of reported gravimetric storage of H2 in various carbon materials (adapted from [74])

Material

Max. wt% H2

T(K)

P (MPa)

SWNTs(low purity)

5-10

133

0.040

SWNTs(high purity)

300

0.040

GNFs(tubular)

11.26

298

11.35

GNFs (herringbone)

67.55

298

11.35

GNS(platelet)

53.68

298

11.35

Graphite

4.52

298

11.35

GNFs

0.4

298-773

0.101

Li-GNFs

20

473-673

0.101

Li-Graphites

14

473-674

0.101

K-GNFs

14

<313

0.101

K-Graphite

5.0

<313

0.101

SWNTs(high purity)

8.25

80

7.18

SWNTs(~50% pure)

4.2

300

10.1

Materials with high hydrogen storage capacities are desirable for energy storage applications. Metal hydrides and cryo-adsorption are the two commonly used means to store hydrogen, typically at high pressure and/or low temperature. In metal hydrides, hydrogen is reversibly stored in the interstitial sites of the host lattice. The electrical energy is produced by direct electrochemical conversion. Hydrogen can also be stored in the gas phase in the metal hydrides. The relatively low gravimetric energy density has limited the application of metal hydride batteries. Because of their cylindrical and hollow geometry, and nanometer-scale diameters, it has been predicted that the carbon nanotubes can store liquid and gas in the inner cores through a capillary effect [76]. A Temperature-Programmed Desorption (TPD) study on SWNT-containing material (0.1-0.2 wt% SWNT) estimates a gravimetric storage density of 5-10 wt% SWNT when H2 exposures were carried out at 300torr for 10 min at 277 K followed by 3 min at 133 K [69]. If all the hydrogen molecules are assumed to be inside the nanotubes, the reported density would imply a much higher packing density of H2 inside the tubes than expected from the normal H2-H2 distance. The same group recently performed experiments on purified SWNTs and found essentially no H2 absorption at 300 K [75]. Upon cutting (opening) the nanotubes by an oxidation process, the amount of absorbed H2 molecules increased to 4-5 wt%. A separate study on higher purity materials reports ^8 wt% of H2 adsorption at 80 K, but using a much higher pressure of 100 atm [77], suggesting that nanotubes have the highest hydrogen storage capacity of any carbon material. It is believed that hydrogen is first adsorbed on the outer surface of the crystalline ropes.

An even higher hydrogen uptake, up to 14-20 wt%, at 20-400°C under ambient pressure was reported [70] in alkali-metal intercalated carbon nanotubes. It is believed that in the intercalated systems, the alkali metal ions act as a catalytic center for H2 dissociative adsorption. FTIR measurements show strong alkali-H and C-H stretching modes. An electrochemical absorption and desorption of hydrogen experiment performed on SWNT-containing materials (MER Co, containing a few percent of SWNTs) reported a capacity of 110mAh/g at low discharge currents [72]. The experiment was done in a half-cell configuration in 6 M KOH electrolyte and using a nickel counter electrode. Experiments have also been performed on SWNTs synthesized by a hydrogen arc-discharge method [71]. Measurements performed on relatively large amount materials (^50% purity, 500 mg) showed a hydrogen storage capacity of 4.2 wt% when the samples were exposed to 10MPa hydrogen at room temperature. About 80% of the absorbed H2 could be released at room temperature [71].

The potential of achieving/exceeding the benchmark of 6.5 wt% H2 to system weight ratio set by the Department of Energy has generated considerable research activities in universities, major automobile companies and national laboratories. At this point it is still not clear whether carbon nanotubes will have real technological applications in the hydrogen storage applications area. The values reported in the literature will need to be verified on well-characterized materials under controlled conditions. What is also lacking is a detailed understanding on the storage mechanism and the effect of materials processing on hydrogen storage. Perhaps the ongoing neutron scattering and proton nuclear magnetic resonance measurements will shed some light in this direction.

In addition to hydrogen, carbon nanotubes readily absorb other gaseous species under ambient conditions which often leads to drastic changes in their electronic properties[78,79,112]. This environmental sensitivity is a double-edged sword. From the technological point of view, it can potentially be utilized for gas detection[112]. On the other hand, it makes very difficult to deduce the intrinsic properties of the nanotubes, as demonstrated by the recent transport[78] and nuclear magnetic resonance[79] measurements. Care must be taken to remove the adsorbed species which typically requires annealing the nanotubes at elevated temperatures under at least 10~6 torr dynamic vacuum.

0 0

Post a comment