Battery cathodes for improved stability

Abstract

A LiO.sub.2 battery and method for fabricating the same are provided herein. The battery cathode comprises a carbon structure filled with a palladium nanoparticle catalyst, including palladium-filled carbon nanotubes (CNTs). The carbon structure provides a barrier between the catalyst and the electrolyte providing an increased stability of the electrolyte during both discharging and charging of a battery.

Claims

1. A method of manufacturing a lithium battery, the method comprising: decapping a carbon structure in a nitric acid solution to form a decapped carbon structure; immersing the decapped carbon structure in a salt solution until a slurry is formed, the salt solution comprising a platinum group metal; providing a cathode by coating the slurry on a carbon cloth gas diffusion layer; assembling the cathode on an anode including a lithium metal such that an electrolyte is disposed between the anode and the cathode to provide the lithium battery comprising the anode, the electrolyte, and the cathode, the cathode comprising the carbon cloth gas diffusion layer and the carbon structure having a catalyst, wherein the catalyst is filled in the carbon structure without a surface coating of the catalyst on the carbon structure, and the catalyst has nanoparticles of a platinum group metal only.

2. The method according to claim 1, the carbon structure including at least one of graphene, fullerenes, amorphous carbons, and carbon nanotubes.

3. The method according to claim 1, further comprising attaching a tube on the cathode and a rod on the anode.

4. The method according to claim 1, further comprising first drying the slurry and calcinating the slurry before coating the slurry on the carbon cloth gas diffusion layer.

5. The method according to claim 4, the first drying being performed under an oxygen gas.

6. The method according to claim 5, further comprising hydrogenating the slurry under a hydrogen gas.

7. The method according to claim 4, further comprising drying the cathode.

8. The method according to claim 1, further comprising storing the cathode in an Ar-filled box.

9. The method according to claim 1, further comprising rinsing the decapped carbon structure in a water and drying the decapped carbon structure before immersing the decapped carbon structure in the salt solution.

10. The method according to claim 1, further comprising providing a separator between the anode and the cathode, the electrolyte being soaked in the separator, and the separator being a polypropylene separator.

11. The method according to claim 1, the platinum group metal being palladium.

12. A method of manufacturing a lithium battery, the method comprising: decapping a carbon structure in a nitric acid solution to form a decapped carbon structure; immersing the decapped carbon structure in a salt solution until a slurry is formed, the salt solution comprising a platinum group metal; providing a cathode by coating the slurry on a carbon cloth gas diffusion layer; assembling the cathode on an anode including a lithium metal such that an electrolyte is disposed between the anode and the cathode to provide the lithium battery comprising the anode, the electrolyte, and the cathode, the cathode comprising the carbon cloth gas diffusion layer and the carbon structure having a catalyst, the carbon structure including at least one of graphene, fullerenes, amorphous carbons, and carbon nanotubes, the method further comprising: rinsing the decapped carbon structure in a water and drying the decapped carbon structure before immersing the decapped carbon structure in the salt solution; hydrogenating the slurry under a hydrogen gas; first drying the slurry under an oxygen gas and calcinating the slurry before coating the slurry on the carbon cloth gas diffusion layer; attaching a tube on the cathode and a rod on the anode; drying the cathode; storing the cathode in an Ar-filled box; and providing a separator between the anode and the cathode, the electrolyte being soaked in the separator, and the separator being a polypropylene separator, the catalyst being a nanoparticle catalyst comprising nanoparticles of a platinum group metal only, the catalyst filled in the carbon structure without a surface coating of the catalyst on the carbon structure, and the platinum group metal being palladium.

Description

BRIEF DESCRIPTION OF DRAWINGS

(1) FIG. 1 shows a LiO.sub.2 battery according to an embodiment of the subject invention.

(2) FIG. 2 shows a plurality of carbon structures of a LiO.sub.2 battery according to an embodiment of the subject invention.

(3) FIG. 3(a) is a transmission electron micrograph of Pd-coated CNTs.

(4) FIG. 3(b) is a transmission electron micrograph of Pd-filled CNTs.

(5) FIG. 3(c) is a plot of a Raman spectrum of Pd-filled and Pd-coated CNTs.

(6) FIG. 4 is a plot of the first discharge/charge capacity of pristine CNT, PD-coated CNTs and Pd-filled CNTs at a constant current density of 250 mAh.

(7) FIG. 5 is a plot of cyclic voltammetry of pristine, Pd-filled, and Pd-coated CNTs in range of 2-4.5 V vs Li/Li.sup.+, wherein all scans rates are 1 mV.Math.s.sup.1.

(8) FIG. 6(a) is a plot of Raman spectrum after discharging and charging the batteries for pristine, Pd-coated, and Pd-filled CNTs.

(9) FIG. 6(b) is a plot of an x-ray diffraction patterns confirming the presence of Li.sub.2O.sub.2 and Li.sub.2CO.sub.3 in Pd-coated and Pd-filled discharged cathodes.

(10) FIG. 7(a) is an image of a scanning electron micrograph of cathodes after discharge for pristine CNTs.

(11) FIG. 7(b) is an image of a scanning electron micrograph of cathodes after discharge for Pd-coated CNTs.

(12) FIG. 7(c) is an image of a scanning electron micrograph of cathodes after discharge for Pd-filled CNTs.

(13) FIG. 8(a) is a plot of a chronopoteniometric test at 250 mA.Math.g.sup.1 from OCV to 4.5 V for pristine, Pd-coated, and Pd-filled CNTs.

(14) FIG. 8(b) is a plot of a linear sweep voltammetry of pre-discharged Pd-coated and Pd-filled CNTs under oxygen between OCV and 4.5 V vs Li/Li.sup.+ at a scan rate of 1 mV.Math.s.sup.1, wherein the scale bars are 1 m.

(15) FIG. 9 is a plot of the cyclability of the LiO.sub.2 batteries for fixed cycle capacities of 500 mAh.Math.g.sup.1 at a current density of 250 mA.Math.g.sup.1 and with voltage cutoffs of 2.0-4.5 V for pristine, Pd-coated, and Pd-filled CNTs. The marker denotes discharge capacity and the marker denotes charge capacity.

DETAILED DESCRIPTION

(16) Embodiments of the subject invention provide Li-oxygen (LiO.sub.2) cathodes using palladium-coated and palladium-filled carbon nanotubes (CNTs). It should be appreciated by one of ordinary skill in the art that the CNTs can be replaced with various catalysts (for example, ruthenium, or platinum-based catalysts) filled carbon structures, (for example fullerenes, buckminsterfullerenes, or graphenes). Empirical data shows that the full discharge of batteries in a 2-4.5 V range shows 6-fold increase in the first discharge cycle of the Pd-filled over the pristine CNTs and 35% increase over their Pd-coated counterparts. The Pd-filled also exhibits improved cyclability with 58 full cycles of 500 mAh.Math.g.sup.1 at current density of 250 mA.Math.g.sup.1 versus 35 and 43 cycles for pristine and Pd-coated CNTs, respectively. The effect of encapsulating the Pd catalysts inside the CNTs leads to increased stability of the electrolyte during both discharging and charging of the battery. Voltammetry, Raman spectroscopy, FTIR, XRD, UV/Vis spectroscopy and visual inspection of the discharge products using scanning electron microscopy can be used to confirm the improved stability of the electrolyte due to this encapsulation and that this approach could lead increasing the LiO.sub.2 battery capacity and cyclability performance.

(17) Multi-walled carbon nanotubes (MWCNTs) can be decapped by nitric acid solution treatment and then 1 mM aqueous solution of PdCl.sub.2 can be used to swell 100 mg of decapped MWCNTs until a slurry is formed. Pd-coated CNTs can also be prepared following the same procedure on untreated capped MWCNTs. Both slurries of Pd-coated and Pd-filled MWCNTs can be dried overnight at room temperature and calcinated in air at 350 C. for 2 hours. Corresponding particles can then be hydrogenated in an oven under hydrogen gas to yield5 wt % Pd nanoparticles. Cathodes can be prepared by coating a slurry of MWCNT (Pristine, Pd-filled and Pd-coated)/PVDF (90/10 wt % in NMP) on a 0.5 diameter carbon cloth gas diffusion layer (CCGDL) followed by drying at 120 C. for 12 hours. The cathodes can then be stored in an Ar-filled glove box to be used later. The typical loading of MWCNT can be 0.50.01 mg. All reported capacities in this application are reported per total mass of active cathode (CNTs and catalyst).

(18) FIG. 1 shows a LiO.sub.2 battery according to an embodiment of the subject invention. Referring to FIG. 1, the LiO.sub.2 battery 100 comprises a cathode 200, an anode 400, and an electrolyte 300 disposed between the cathode 200 and the anode 400. The LiO.sub.2 battery 100 further comprises a tube 250 disposed on the cathode 200 and a rod 450 disposed on the anode 400. The LiO.sub.2 battery 100 can be assembled by using a Swagelok type cell, in which the tube 250 is a stainless steel tube and the rod 450 is a stainless steel rod. The electrolyte 300 can be formed as a separator soaked with an electrolyte, and the separator can be a polypropylene. The anode 400 is formed by a lithium metal disk including a lithium metal.

(19) The cathode 200 comprises a carbon structure with a metal catalyst or metal oxide catalyst, wherein the metal catalyst or metal oxide catalyst includes a platinum group metal. The platinum group metal includes at least one of ruthenium, rhodium, palladium, osmium, iridium, and platinum. In an embodiment of the subject invention, a palladium nanoparticle catalyst is coated on a surface of the carbon structure or filled in the carbon structure. In addition, the cathode 200 further comprises the CCGDL, and the carbon structure having a platinum group metal catalyst is coated on the CCGDL. The cathode 200 includes a porous structure open to an oxygen and the CCGDL has a woven structure.

(20) FIG. 2 shows a plurality of carbon structures of a LiO.sub.2 battery according to an embodiment of the subject invention. Referring to FIG. 2, the carbon structure of LiO.sub.2 battery includes at least one of graphene, fullerenes, amorphous carbons, and carbon nanotubes. The carbon nanotubes can be a single-walled carbon nanotube or a multi-walled carbon nanotube.

(21) Referring to FIGS. 1 and 2, the electrolyte 300 can be prepared by adding 1 mol.Math.kg.sup.1 of LiTFSI salt (i.e., Lithium salt) into tetraethyl glycol di-methyl ether (TEGDME) solvent. The lithium metal disc of the anode 400 is covered by an electrolyte soaked separator, MWCNT-CCGDL and a stainless steel mesh can be used as a current collector. LiO.sub.2 batteries can be rested inside an Ar-filled glove box overnight before electrochemical tests. All electrolyte preparation and cell assembly can be performed inside an Ar-filled glove box (<1 ppm O.sub.2 and <0.1 ppm H.sub.2O).

(22) The LiO.sub.2 batteries can be removed from the argon glove box and placed in the gastight desiccator filled with ultra-high purity oxygen gas (Airgas, purity>99.994%). The batteries can be rested under oxygen for 5 hours before testing.

(23) In certain embodiments of the subject invention, the CNTs can be prepared such that the Pd nanoparticles fill the carbon nanotubes without a Pd surface coating. CNTs can be decapped by introducing the nanotubes to an acid treatment. The decapped CNTs can then be rinsed with water in order to remove any remaining acid treatment. The decapped CNTs can be dried and then immersed into a palladium salt solution and swelled until a slurry is formed. The CNTs can remain in the palladium salt solution until such time that the nanotubes are filled. The CNTs can then be dried, in a drying device, under oxygen to convert the pallidum salt to palladium oxide particles. The CNTs can then be rinsed to remove any debris remaining on the surface of the nanotubes. The CNTs can then be hydrogenated in a furnace to convert the palladium oxide into palladium. The Pd-filled CNTs can then be stored, for example in Argon, until future use.

(24) As materials of the LiO.sub.2 battery 100 according to the present invention, Palladium (II) chloride (PdCl.sub.2, 59% Pd), Bis (trifluoromethane) sulfonamide (LiTFSI, purity>99.95%), tetraethylene glycol dimethyl ether (TEGDME, purity>99.00%), N-Methylpyrrolidine (NMP, purity>97.00%), multi-walled carbon nanotubes (MWCNT, D=5-20 nm, L=5 m, purity>95.00% carbon basis), Titanium (IV) oxysulfate (TiOSO.sub.4) (>29% Ti (as TiO.sub.2) basis), and Lithium Peroxide (Li.sub.2O.sub.2) can be used. In addition, carbon cloth gas diffusion layer (CCGDL, thickness300 m), Lithium foil chips (purity>99.90%), a polypropylene separator (thickness25 m), and Polyvinylidene fluoride (PVDF) can be also be used.

(25) A Solartron 1470 battery tester can be used for galvanostatic discharge/charge tests within a voltage range of 2.0-4.5 V at a current density of 250 mA.Math.g.sup.1. Voltammetry measurements are performed by an electrochemical workstation (Gamry reference 600) at the rate of 1 mV.Math.s.sup.1 in the range of 2.0-4.5 V to investigate the catalytic behavior of oxygen electrodes. All charge/discharge and electrochemical tests are measured in a temperature controlled environment at 25 C. After charge/discharge cycling, the oxygen cathodes are recovered from the batteries in the Ar-filled glove box, rinsed with acetonitrile and dried under vacuum. Cathodes can be investigated by Raman spectroscopy (BaySpec's Nomadic, excitation wavelength of 532 nm), Fourier transform infrared (FTIR) spectroscopy (JASCO FT-IR 4100), and Scanning electron microscopy (SEM) (JEOL 6330F). Bruker GADDS/D8 X-ray powder diffraction (XRD) with MacSci rotating Molybdenum anode (k=0.71073) operated at 50 kV generator and 20 mA current is also used to collect the diffraction patterns. A parallel X-ray beam in size of 100 m diameter is directed on to the samples and diffraction intensities are recorded on large 2D image plate during exposure time. Li.sub.2O.sub.2 is quantified in the cathodes after discharge using a colorimetric method. Briefly, discharged cathodes are first immersed in water then aliquots are taken and added to 2% aqueous solution of TiOSO.sub.4. Instantaneously a color change occurred and the absorbance spectra of the solutions are collected using a UV-Vis spectrophotometer (Gamry UV/Vis Spectro-115E). The peak intensity at 408 nm is calibrated against solutions with known concentrations of Li.sub.2O.sub.2, in the range of 0.1 to 10 mg/ml and linear calibration curve is obtained. Transmission Electron Microscopy (Phillips CM-200 200 kV) is also used to inspect the carbon nanotubes.

(26) The cathodes of the LiO.sub.2 battery can comprise MWCNTs (pristine, Pd-coated and Pd-filled) coated on the woven carbon cloth gas diffusion layer (CCGDL). Homogenous three-dimensional networks of carbon nanotubes over CCGDL yield high surface area with an open structure which improves the electronic contact during charging and discharging processes. FIGS. 3(a) and 3(b) show TEM images of Pd-coated and Pd-filled CNTs, respectively. Referring FIG. 3(b), in Pd-filled cathodes, Pd nanocatalysts are formed in the inner tubular region of decapped CNTs. FIG. 3(c) shows the Raman spectra of the Pd-coated and Pd-filled CNTs. Referring to FIG. 3(c), D and G bands of Pd-filled and Pd-coated CNTs are identical in location and intensity ratios [I(G)/I(D)1.2] indicating that the decapped CNTs do not have high density of defects on their surface. The left peak corresponds to the D band and the right peak corresponds to the G band.

(27) FIG. 4 shows a plot of the first discharge/charge capacity pf pristine CNT, PD-coated CNTs and Pd-filled CNTs at a constant current density of 250 mAh. The first discharge and charge behaviors of the Pd-coated, Pd-filled and pristine CNTs batteries using 1 M LiTFSI in TEGDME electrolyte in the voltage window of 2.0-4.5 vs Li/Li+ at the constant current density of 250 mA.Math.g.sup.1 are shown in FIG. 4. Referring to FIG. 4, the pristine CNTs show a first discharge capacity of 1980 mAh.Math.g.sup.1 compared to 8197 mAh.Math.g.sup.1 and 11,152 mAh.Math.g.sup.1 for the Pd-coated and Pd-filled CNTs, respectively. The 6-fold discharge capacity improvement of Pd-coated CNTs over pristine CNTs resulted from improved ORR and increased surface sites for lithium discharge product deposition due to the presence of Pd nanocatalysts.

(28) FIG. 5 shows a plot of cyclic voltammetry of pristine, Pd-filled, and Pd-coated CNTs in range of 2-4.5 V vs Li/Li.sup.+, wherein all scans rates are 1 mV.Math.s.sup.1. Referring to FIG. 5, Cyclic voltammetry (CV) measurements confirm higher ORR and OER currents for Pd containing CNTs over pristine CNTs. This also confirms retained catalytic activity of Pd when encapsulated inside the CNTs, consistent with previously observations. In addition, the presence of Pd nanocatalysts in both Pd-filled and Pd-coated CNTs shifts the onset of ORR peak of pristine CNTs from 2.8 V to 2.9 V, showing enhanced cathodic activity. The Pd-filled CNTs also demonstrates a 36% increase in first discharge capacity over Pd-coated. In the FIG. 5, an oxidation peak at3.3 V is attributed to the OER and is shown to be more pronounced for the Pd-filled compared to Pd-coated. It is considered that the presence of Pd inside the CNTs strengthens the electron density on the CNT surface yielding homogeneously distributed nucleation sites for Pd-filled CNTs compared to heterogeneously distributed nucleation sites for Pd-coated CNTs. This delocalization of Li.sub.2O.sub.2 seeding sites contributes to intimate contact between Li.sub.2O.sub.2 and CNTs and helps promote the formation of high surface discharge products. At voltage exceeding 3.7 V vs Li/Li.sup.+ Pd-coated CNTs show the highest rate of oxidation of electrolyte and electrolyte decomposition products.

(29) FIG. 6(a) shows a plot of Raman spectrum after discharging and charging the batteries for pristine, Pd-coated, and Pd-filled CNTs. Referring to FIG. 6(a), Raman spectroscopic characterization on the discharged cathodes shows Li.sub.2O.sub.2 formation at 790 cm.sup.1 Raman shift for all discharged cathodes. However, the Pd-coated CNTs cathode shows a pronounced Raman shift peak at 1080 cm.sup.1, which corresponds to the electrolyte decomposition product Li.sub.2CO.sub.3. This peak is absent from the pristine CNT cathodes. This behavior confirms the observation from CV, indicating reduced electrolyte stability due to the presence of Pd on Pd-coated CNTs and enhanced electrolyte stability for Pd-filled compared to Pd-coated CNTs. Following charging, Raman spectroscopic analyses on the cathodes shows efficient removal of the Li.sub.2O.sub.2 peaks from all cathodes, and a decrease in peak intensity of Li.sub.2CO.sub.3 peak for the Pd containing CNTs. The decrease in Li.sub.2CO.sub.3 was previously reported to be enabled in catalyst-containing cathodes. Li.sub.2O.sub.2 content in the cathode is quantified using a colorimetric approach. The amounts of Li.sub.2O.sub.2 in the cathode are determined to be 11.6, 21.4, and 35.4 mols for pristine, Pd-coated, and Pd-filled CNT cathodes, respectively. These amounts correspond to capacity yields of approximately 88%, 23%, and 37% of the experimental capacities recorded by the pristine, Pd-coated, and Pd-filled CNT cathodes, respectively.

(30) In order to determine the molar ratio of Li.sub.2O.sub.2 in the discharged cathodes, the cathodes are analyzed using FTIR. Using peak intensities ratio at 600 cm.sup.1 (Li.sub.2O.sub.2) and 862 cm.sup.1 (Li.sub.2CO.sub.3), Pd-coated and Pd-filled cathodes have 19.3% and 33.2% Li.sub.2O.sub.2 by mole, respectively. By only considering the Li.sub.2O.sub.2 and Li.sub.2CO.sub.3 discharge species, this observation is in agreement with the UV-Vis quantification and further confirms the stabilizing effect of the encapsulation of Pd inside the CNTs compared to coating the CNTs. The CV and Raman data also back up these claims, indicating that the electrolyte undergoes more decomposition in cells with Pd-coated CNTs cathodes.

(31) FIG. 6(b) show a plot of an x-ray diffraction patterns confirming the presence of Li.sub.2O.sub.2 and Li.sub.2CO.sub.3 in Pd-coated and Pd-filled discharged cathodes. Referring to FIG. 6(b), the presence of Li.sub.2O.sub.2 and Li.sub.2CO.sub.3 is additionally confirmed by XRD.

(32) FIGS. 7(a), 7(b), and 7(c) show images of a scanning electron micrograph of cathodes after discharge for pristine CNTs, Pd-coated CNTs, and Pd-filled CNTs, respectively. Referring to FIGS. 7(a), 7(b), and 7(c), the formation of discharge products is visually confirmed using scanning electron microscopy on discharged cathodes. The discharge products (Li.sub.2O.sub.2) of the pristine CNTs are conformal around the cathode in rod-like structures with low porosity. The growth of layers in this morphology often yields to passivation and blockage of oxygen to the cathode and eventually limits the discharge capacity and cycle life of the battery. The Pd-coated CNTs cathode as shown FIG. 7(b) shows some platelet-shaped Li.sub.2O.sub.2 buried by thick conformal layer, suspected to be Li.sub.2CO.sub.3 as shown in Raman and FTIR measurements. In contrast, the Pd-filled CNTs show nano-thin platelets of Li.sub.2O.sub.2 covering the cathode as shown FIG. 7(c). These platelets of Li.sub.2O.sub.2 yield high surface porosity which in turn do not block the access to the CNTs and enhance the performance.

(33) In order to identify the synergy of electrolyte and Pd nanocatalysts, the oxidation stability limit of the electrolyte is determined using a chronopotentiometric stability test and linear sweep voltammetry under oxygen atmosphere. Batteries using Pd-coated, Pd-filled and pristine CNTs are assembled and charged without prior discharging at constant current density of 250 mA.Math.s.sup.1 up to cutoff voltage of 4.5 V.

(34) FIG. 8 shows higher capacity for Pd-coated compared to Pd-filled and pristine. In particular, FIG. 8(a) shows a plot of a chronopoteniometric test at 250 mA.Math.g.sup.1 from OCV to 4.5 V for pristine, Pd-coated, and Pd-filled CNTs, and FIG. 8(b) shows a plot of a linear sweep voltammetry of pre-discharged Pd-coated and Pd-filled CNTs under oxygen between OCV and 4.5 V vs Li/Li.sup.+ at a scan rate of 1 mV.Math.s.sup.1, wherein the scale bars are 1 m. Since no discharge products existed, the capacities obtained are attributed to electrolyte and cathode undesirable reactions. This suggests that Pd-filled CNTs promotes better stability than Pd-coated CNT cathodes. This observation is also confirmed using linear sweep voltammetry following a discharge showing increased OER peak intensity at 3.3-3.4 V and improved electrolyte stability above 3.7 V vs Li/Li.sup.+ for Pd-filled vs Pd-coated CNTs in FIG. 5. These results again support previous observations that encapsulating the Pd nanocatalyst helps improve the electrolyte stability during the operation of the LiO.sub.2 battery.

(35) FIG. 9 shows the cycling stability of LiO.sub.2 batteries based on Pd-filled, Pd-coated, and pristine CNTs. Referring to FIG. 9, Galvanostatic discharge/charge cycling of LiO.sub.2 batteries at a current density of 250 mA.Math.g.sup.1 at a limited capacity of 500 mAh.Math.g.sup.1 in a voltage window of 2.0-4.5 V vs Li/Li.sup.+ are conducted. LiO.sub.2 cells using Pd-filled CNTs show the highest discharge cycling performance of 58 cycles compared to 43 cycles for Pd-coated, and 35 for pristine CNTs. The cycling stability improvement in the case of Pd-coated and Pd-filled CNTs is a result of previously confirmed OER/ORR improvement due to the Pd nanocatalysts. The cycling stability improvement of Pd-filled CNTs over the Pd-coated CNTs is ultimately credited to the decrease in undesirable discharge/charge products formation, e.g. Li.sub.2CO.sub.3, afforded by the encapsulation approach. The inclusion of Pd catalyst in CNTs for use as cathode materials in LiO.sub.2 batteries has been demonstrated. Using two modes of CNT loading, inside (Pd-filled CNTs) and outside (Pd-coated CNTs), shows that both approaches yielded significant improvement in full discharge capacities of the batteries while the Pd-filled cycled for 35% more cycles of 500 mAh.Math.g.sup.1 at current density of 250 mA.Math.g.sup.1. The encapsulation of nanocatalyst inside the CNTs improves the stability of the electrolyte by decreasing the formation of Li.sub.2CO.sub.3 compared to nanocatalyst-coated CNTs. These observations are confirmed by voltammetry, Raman and UV/Vis spectroscopy, FTIR, chronopotentiometry, electron microscopy, and charge/discharge cycling. The Pd-filled CNTs and the Pd-coated CNTs are discussed independently, but the CNTs that are coated with Pd outside and filled with Pd inside at the same time can be used.

(36) It should be understood that the examples and embodiments described herein are for illustrative purposes only and that various modifications or changes in light thereof will be suggested to persons skilled in the art and are to be included within the spirit and purview of this application.

(37) All patents, patent applications, provisional applications, and publications referred to or cited herein are incorporated by reference in their entirety, including all figures and tables, to the extent they are not inconsistent with the explicit teachings of this specification.