Bih-Show Lou1, Pitchaimani Veerakumar2, Shen-Ming Chen3, Vediyappan Veeramani3, Rajesh Madhu3, Shang-Bin Liu2,4. 1. Chemistry Division, Center for General Education, Chang Gung University, Taoyuan 33302, Taiwan. 2. Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 10617, Taiwan. 3. Department of Chemical Engineering and Biotechnology, National Taipei University of Technology, Taipei 10608, Taiwan. 4. Department of Chemistry, National Taiwan Normal University, Taipei 11677, Taiwan.
Abstract
The synthesis of highly dispersed and stable ruthenium nanoparticles (RuNPs; ca. 2-3 nm) on porous activated carbons derived from Moringa Oleifera fruit shells (MOC) is reported and were exploited for supercapacitor applications. The Ru/MOC composites so fabricated using the biowaste carbon source and ruthenium acetylacetonate as the co-feeding metal precursors were activated at elevated temperatures (600-900 (o)C) in the presence of ZnCl2 as the pore generating and chemical activating agent. The as-prepared MOC carbonized at 900 (o)C was found to possess a high specific surface area (2522 m(2) g(-1)) and co-existing micro- and mesoporosities. Upon incorporating RuNPs, the Ru/MOC nanocomposites loaded with modest amount of metallic Ru (1.0-1.5 wt%) exhibit remarkable electrochemical and capacitive properties, achiving a maximum capacitance of 291 F g(-1) at a current density of 1 A g(-1) in 1.0 M H2SO4 electrolyte. These highly stable and durable Ru/MOC electrodes, which can be facily fabricated by the eco-friendly and cost-effective route, should have great potentials for practical applications in energy storage, biosensing, and catalysis.
The synthesis of highly dispersed and stable ruthenium nanoparticles (RuNPs; ca. 2-3 nm) on porous activated carbons derived from Moringa Oleifera fruit shells (MOC) is reported and were exploited for supercapacitor applications. The Ru/MOCcomposites so fabricated using the biowaste carbon source and ruthenium acetylacetonate as the co-feeding metal precursors were activated at elevated temperatures (600-900 (o)C) in the presence of ZnCl2 as the pore generating and chemical activating agent. The as-prepared MOCcarbonized at 900 (o)C was found to possess a high specific surface area (2522 m(2) g(-1)) and co-existing micro- and mesoporosities. Upon incorporating RuNPs, the Ru/MOC nanocomposites loaded with modest amount of metallicRu (1.0-1.5 wt%) exhibit remarkable electrochemical and capacitive properties, achiving a maximum capacitance of 291 F g(-1) at a current density of 1 A g(-1) in 1.0 M H2SO4 electrolyte. These highly stable and durable Ru/MOC electrodes, which can be facily fabricated by the eco-friendly and cost-effective route, should have great potentials for practical applications in energy storage, biosensing, and catalysis.
Porous activated carbon (PAC)-related materials offer great advantages for practical applications12, such as adsorbents, energy storage, catalyst supports, and electrodes for fuel cells3. Moreover, owing to the intrinsic properties, such as high surface area, diversified morphology, good electrical conductivity, and tailorable porosity, PACs are also favorable materials as supports for metal nanoparticles (NPs) or metal oxides, and have been widely applied as biomolecule sensors131415, biomedical engineering materials16, toxic molecules/heavy metal detectors1718, and supercapacitors18192021222324252627282930313233. In terms of the latter, aside from the widely studied carbon materials such as graphene3435363738, carbon nanotubes37383940, and ordered mesoporous carbons41424344, PACs are easy to prepare, eco-friendly, and cost-effective; they may be prepared from renewable biomass precursors through facile carbonization and activation routes, for examples, from tree barks or leaves182021222324, plants2526, fruits2728, grain or seed shells2930313245464748495051, lignin5253, food derivatives54, marine products33, and so on. Typically, the synthesis of PACs from biomass feedstock invokes a chemical activation method in which activating agents such as ZnCl2, KOH, NaOH, or H3PO4 are commonly introduced along with the biomass precursor214152032. Upon completion of a subsequent carbonization treatment, the substrate was then washed with concentrated HCl to obtain the PAC as the final product5354.We report herein the synthesis of stable, highly dispersed ruthenium nanoparticles (RuNPs) on PACs derived from Moringa Oleifera fruit shells. Moringa Oleifera is a fast-growing, deciduous tree also known as the “drumstick tree”, mostly cultivated in Asian countries such as India, Philippines, and is also commonly seen in Africa, South America, the Caribbean and other Oceania countries5556. Compare to other biomass feedstock, Moringa Oleifera, which belongs to the family of Moringaceae, is not only abundant in nature but also known to produe highly nutritious fruit5758 that have medicinal596061, and other applications62. Moreover, the Moringa Oleifera also contains cellulose, hemicellulose, and lignin that are desirable as precursors for fabrication of PACs with wormhole-like microstructures626364656667 preferable as supports for electochemical and energy storage applicaions. By graphitizing the biowaste carbon precursor, namely Moringa Oleifera fruit shells along with ZnCl2 as activating agent under N2 atmosphere at elevated temperatures, the PAC (hereafter termed as Moringa Oleiferacarbons; MOC-Tc) so fabricated at different carbonization temperatures (Tc = 600–900 oC) was further washed with HCl to remove the Zn2+ species. Subsequently, the as-synthesized MOC-900 was mixed with the metal precursor (i.e., ruthenium(III) acetylacetonate; Ru(acac)3) in ethanol solution followed by a thermal reduction treatment to disperse RuNPs onto the carbon support. It is noteworthy that, here, the Ru(acac)3 serves as the metal precursor as well as a secondary carbon source, which warrants not only simultaneous reduction of RuIII to Ru0 but also a high dispersion of RuNPs on the MOC789101112. As illustrated in Fig. 1, after a thermal reduction treatment at 900 oC, the Ru/MOC-900 nanocomposites so fabricated with vaired Ru loadings (x = 1.0 and 1.5 wt%) were found to exhibit excellent electrochemical properties desirable for electrochemical applications. To the best of our knowledge, this represents the first report to exploit Moringa oleifera fruit shells as the primary carbon source to fabricate PAC materials for such applications. As will be shown latter that the Ru/MOC nanocomposites prepared by this innovated facile and cost-effective route are indeed suitable as high-performance electrode materials for high-power supercapacitors.
Figure 1
Illustration of the synthesis route for Ru/MOC nanocomposites.
(a) Moringa Oleifera fruit shells; (b) thermal irradiation by microwave (MW); (c) activation and carbonization treatments; (d) addition of Ru(acac)3 followed by MW irradiation and thermal reduction at 900 °C; (e) Ru/MOC nanocomposite electrode for supercapacitor application.
Results and Discussion
Figure 2a shows the XRD patterns of the as-prepared MOC and Ru/MOC samples. For the MOC-Tc samples prepared at different temperatures (Tc = 600–900 oC), two broad peaks at 2θ = 22.2 and 43.4° respectively corresponding to the (002) and (100) diffractions of amorphous graphiticcarbon68 were observed. On the other hand, for the Ru/MOC-900composites reduced at 900 oC with different metal loading (x = 1.0 and 1.5 wt%), additional sharp diffraction peaks at 2θ = 38.3, 42.2, 44.0, 58.3, 69.4, and 78.4o were evident, which may be assigned to the (100), (002), (101), (102), (110), and (103) diffraction planes of hexagonal close-packed (hcp) Ru metal (JCPDS-ICDD card No. 06-0663)69. The Raman spectra of the MOC and Ru/OMC samples exhibited two main peaks located at 1363 and 1585 cm−1 (Fig. 2b), which may be attributed to vibration bands of carbons in disorderedgraphite (D band) and the E2g mode of the graphite (G band), which is related to vibrations of sp2 carbon structure in two-dimensional (2D) hexagonal lattice70. Moreover, for MOCcarbonized at elevated temperatures (Tc ≥ 800 oC), an additional vibrational peak corresponding to the overtone of the D band (i.e., the 2D band) at 2712 cm−1 was also observed71. Accordingly, the G to D band intensity ratio (IG/ID) is normally used to assess the crystalline structure of the graphiticcarbons, as depicted in Table 1. That an IG/ID ratio of 1.21 was observed for the MOC-900 and Ru/MOC-900, indicating a high degree of graphitization for the PAC supports.
Figure 2
Physicochemical properties of various MOC-Tc and Ru/MOC-Tc materials.
Textural properties of various as-prepared MOC and Ru/MOC samples.
Sample
Tca
Ru loadingb
Mpc
Surface aread
Pore volumee
Dpf
IG/IDg
Stot
Smicro
Vtot
Vmicro
Vmeso
MOC-600
600
—
—
718
286
0.98
0.20
0.78
3.7
0.97
MOC-700
700
—
—
1384
456
1.02
0.30
0.72
3.8
0.98
MOC-800
800
—
—
1924
480
1.33
0.39
0.94
3.9
0.99
MOC-900
900
—
—
2522
576
1.78
0.46
1.33
4.3
1.21
Ru1.0/MOC-900
900
1.0
2.8
2473
506
1.69
0.37
1.32
4.1
1.20
Ru1.5/MOC-900
900
1.5
3.1
2015
439
1.48
0.29
1.19
3.9
1.21
(a) Carbonization temperature in oC. (b) Ru loading in wt% deduced from EDX and TGA results. (c) RuNP size in nm determined by FE-TEM analysis. (d) Brunauer-Emmet-Teller (BET) surface areas in unit of m2 g−1; Stot and Smicro denotes total and microporous surface area, respectively; Smicro determined by t-plot analysis. (e) Total pore volume in cm3 g−1 calculated at P/P0 = 0.99 of the N2 adsorption/desorption isotherm; Vtot, Vmicro, and Vmeso represents total, microporous, and mesoporous pore volume, respectively, Vmeso = Vtot–Vmicro. (f) Average pore size determined by non-local DFT calculations. (g) G to D band intensity ratio obtained from Raman data.
Results obtained from N2 adsorption/desorption isotherms (77 K) showed that all MOC-based samples exhibit H1-type isotherms (see Fig. 2c; cf. IUPACclassification). The presence of a weak hysteresis loop at P/P0 of ca. 0.4 together with the notable increase in N2 uptade at low relative pressures reveal the coexistence of micro- and mesoporosities in these carbon substrates72. Accordingly, the specific surface areas and pore volumes responsible for the micro- and mesoporosities may be derived, as depicted in Table 1. Further calculations by density functional theory (DFT) indicate that these MOC materials have an average mesopore size of ca. 4.0 nm (see Supporting Information, Fig. S1). Based on the above results, it is clear that the MOC-900, which possesses the highest BET surface area (Stot = 2522 m2 g−1) and total pore volume (Vtot = 1.78 cm3 g−1), exhibits superior textural properties and IG/ID value. Compared to its counterparts carbonized at relatively lower temperautres, it is indicative that the MOC-900 substrate is more suitable for application as electrode material owing to the anticipated higher electrical conductivity and porosity, which are favorable for electron transport and ion diffusion. Upon loading RuNPs onto the MOC-900 support, consistent decreases in both microporous and mesoporous surface area and pore volume with increasing Ru loading were observed (Table 1), indicating the successiful dispersion of metal NPs in both types of pores. This is also supported by the pore size distribution profiles (Supporting Information, Fig. S1), which showed notable decrease in micropores along with narrowing of mesopore distribution.The role of activating agent, namely ZnCl2, during activation of MOC is worthy for further exploration. During the process, the impregnated ZnCl2 tends to promote dehydration of the carbon substrate, leading to charring and aromatization along with the creation of porosities. It is anticipated that mobile liquid ZnCl2 (m.p. ∼ 283 oC) should be formed during the initial stage of the activation. Further increasing the activation temperature beyond 750 oC (b.p. of ZnCl2ca. 730 oC), strong interactions between carbon atoms and Zn species, which result in considerable collapses between the carbon interlayers to create pores in the matrix, as illustrated in Fig. 3 73. It has been shown that the generation of micro- and mesoporosities (Table 1) is provoked by the elimination of hydrogen and oxygen atoms from the carbon substrate by ZnCl2, leading to the formation of water rather than oxygenated organic species74. The amount of activating agent, activation temperature, and subsequent washing by HCl were all found to have profound effect on the evolution of porosity within the MOC. At a fixed activation temperature, a gradual increase in BET surface area (Stot) of the as-synthesized MOC with increasing dosage of ZnCl2 was observed (Supporting Information, Fig. S2). For examples, at a low carbonization temperature (600 oC), the MOC prepared in absence of ZnCl2 exhibited a rather low Stot = 50.6 m2 g−1. On the other hand, the surface area of MOC increased from ca. 210 to 718 m2 g−1 when prepared in the presence of 0.5 and 2.0 g of the activating agent (Supporting Information, Fig. S3). A more pronounced effect was found for activation temperature, for example, upon increasing the temperature from 600 to 900 oC while in presence of a fixed amount of ZnCl2 (2.0 g), the surface area of the resulting MOC increased drastically from 718 to 2522 m2 g−1. These results are consistent with earlier literature reports7576.
Figure 3
Schematic of the pore formation during activation of MOC at 900 oC in the presence of ZnCl2 as activating agent.
The TGA profiles of various MOC and Ru/MOC nanocomposites are displayed in Fig. 2d, their corresponding DTAcurves are shown in Fig. S4 of the Supporting Information. Aside from the slight weight-loss below 150 oC due to desorption of physisorbed water, all samples also showed a strong weight-loss at ca. 620 oC, which should be associated with combustion of the MOC material77. The nearly complete weight-loss observed for the as-syntheized MOC-Tc (Tc = 600–900 oC), indicating a complete oxidation of MOC by combustion and that nearly no trace of other ingredients (e.g., Zn species) were present in the PAC materials78. By comparison, while the Ru-loaded MOCs also exhibited two distinct weight-loss peaks at 50–150 and 400–620 oC, respectively, a residual weight-loss of ca. 9.8 and 11.6 wt% was observed for the Ru1.0/MOC-900 and Ru1.5/MOC-900, respectively, indicating the anticipated presence of remanent rutheniumn oxides.The morphology and structural properties of the as-syntheiszed and Ru-loaded MOCs are examined by using FE-SEM/TEM, as shown in Figs 4 and 5. The as-prepared MOCs clearly possess abundant porosities, which tend to increase with increasing carbonization temperature, as illustrated in the magnified SEM and TEM images (see Supporting Information, Figs. S5 and S6)79. This can be seen by the TEM images of MOC-900 sample taken at different magnifications (Fig. 4), which clearly indicate the presences of interconnected micro- and mesopores with a curl-like morphology. For the MOC-900, a rather board distribution of mesopore sizes in the range of ca. 5−20 nm on the surfaces of the curl-like MOCs may be inferred.
Figure 4
FE-TEM image of the MOC-900 at different magnifications.
(a–d) FE-TEM images and (e) EDX profiles of Ru-loaded MOCs. (a) Ru1.5/MOC-900 and (b–d) Ru1.0/MOC-900 nanocomposites; the white circiles identify the presence of RuNPs. Inset in (c) shows the corresponding SAED pattern of RuNPs. (e) EDX profile of the Ru1.0/MOC-900 sample.
Moreover, the dispersion of RuNPs on the surfaces of MOC-900 may be clearly observed for both Ru-loaded samples, as shown in Fig. 5. Further analysis indicate that the RuNPs has an average particle size of ca. 3 nm for both Ru1.5/MOC-900 and Ru1.0/MOC-900 nanocomposites, as illustrated in Fig. S7 of the Supproting Information. For the latter, analysis based on selected area electron diffraction (SAED) pattern (Inset, Fig. 5c) revealed the presences of (100), (002), (101), (102), (110), and (103) reflections corresponding to crystalline hcp structure of Ru metal. Based on the FE-TEM image of a single Ru metal NP (Fig. 5d), a lattice spacing of 0.230 nm was determined, in excellent agreement with the value derived from the XRD data of the (002) lattice plane (Fig. 2a). Moreover, analysis of the energy-dispersive X-ray (EDX) result (Fig. 5e) indicates the existences of various signals corresponding to C Kα (0.2 keV), Cu Lα,β (0.9 keV), Cu Kα (8.0 keV), Cu Kβ (8.9 keV), Ru Lα,β (2.6 keV), Ru Lγ (3.2 keV), and Ru Kα (19.2 keV), respectively. The Cu signals arise from diffuse scattering of the Cu grid support80. The above results confirm a complete thermal reduction of Ru(acac)3 to RuNPs, which are uniformly dispersed over the structural framework of the MOC support.The surface properties of the MOC-900 sample and Ru/MOCcomposites were further examined by X-ray photoelectron spectroscopy (XPS), as shown in Fig. 6. The XPS survey spectrum of the as-prepared MOC-900 (Fig. 6a) exhibits the anticipated C 1s (282–290 eV) and O1s (530–535 eV) signals. The spectrum near the C 1s and O 1s regions, which are displayed in Fig. 6d,e, respectively, may further be deconvoluted to identify C–C (284.8 eV), C–O (286.0 eV), and C = O (289.2 eV) functional groups as well as the corresponding oxygen states of C = O (521.8 eV) and C–O (533.1 eV). For the Ru1.0/MOC-900 and Ru1.5/MOC-900 nanocomposites, additional overlapping peaks at binding energy of 284.3 and 280.7 eV may be ascribed to the charistic peak of Ru 3d3/2 and Ru 3d5/2 81, further confirming the formation of metallicRu0 by the chemical reduction.
Figure 6
XPS spectra of various samples:
(a) as-prepared MOC-900, (b) Ru1.0/MOC-900, and (c) Ru1.5/MOC-900 and their corresponding (d) C1s and (e) O1s spectrum.
Electrochemical Behavior of Ru/MOC-900 Electrodes
The electrochemical properties of the MOC-900, Ru1.0/MOC-900, and Ru1.5/MOC-900 electrode materials were assessed by cyclic voltammetry (CV) and galvanostaticcharge-discharge (GCD) method. The CV curves observed for various Ru/MOC-based electrodes in 1.0 M H2SO4 aqueous electrolyte solution are depicted in Fig. 7a, which all exhibited the typical rectangular shape electri double-layer capacitor (EDLC) behavior. Nonetheless, compared to the Ru/MOC electrode, the metal-free MOC-900 electrode showed only weak capacitive behavior. Among the three MOC-based electrodes examined, the Ru1.0/MOC-900 electrode was found to exhibit an excellent pseudocapacitive redox property in the potential ranges of 0.4–0.6 V. As shown earlier, the RuNPs embedded in the Ru/OMC-900 are highly dispersed in the carbon support, leading to a notable decrease in total surface area and pore volume with increasing metal loading (Table 1). As such, the inferior capacitive performance observed for the Ru1.5/OMC-900 electrode compared to its counterpart with lower Ru loading (1.0 wt%) may be attributed to hindrance in electron transport and ion diffusion by the embedded metal. Overall, a maximum capacitance of 291 F g−1 was observed for the Ru1.0/MOC-900 electrode comparing to that of its MOC-900 and Ru1.5/MOC-900counterparts, as shown in Fig. 7b.
Figure 7
Electrochemical performances of assorted MOC-based electrodes.
(a) CV curves recorded in 1.0 M H2SO4 aqueous electrolyte at a scan rate of 10 mV s−1. (b) Corrsponding specific capacitances observed for various electrodes. (c) CV curves recorded at different scan rates (10–500 mV s−1). (d) GCD curves at different current densities (1–20 A g−1). (e) Variations of specific capacitance with current density. (f) Cyclic stability test at a constant current density of 4 A g−1.
To further assess the characteristics of the Ru1.0/MOC-900 nanocomposite for its practical application as electrode material for supercapacitors, we conducted further electrochemical studies. As can be seen from the CV curves recorded at vaired scan rates from 10 to 500 mV s−1 shown in Fig. 7c, all curves retained the rectangular-shape even at high scan rates, which indicates that the Ru1.0/MOC-900 electrode indeed exhibits excellent capacitive property with good electrical conductivity82. Moreover, based on the GCD curves measured for the Ru1.0/MOC-900 electrode at different current densities from 1 to 20 A g−1 (Fig. 7d), a gradual decrease in the corresponding specificcapacitance with current densitiy from 291 to 94 F g−1 was observed (Fig. 7e). The durability of the Ru/MOC electrode was also evaluated. As shown in Fig. 7f, the Ru1.0/MOC-900 electrode retained over 90% of its initial capacitance after more than 2000 charge-discharge cycles when recorded at a constant current density of 4 A g−1, revealing an extraordinary stability and durability. For comparision, the textural and capacitive properties of other activated carbon (AC)-electrodes derived from various biomass feedstock are depicted in Table 2
2324284546474849505152. It is indicative that the Ru/MOC electrode exhibits comparable capacitive performance even with a modest Ru loading of only 1 wt%. The MOC material derived from Moringa Oleifera fruit shells clearly has the advantage of achieving a high surface area (2473 m2 g−1), this together with the use of a activating agent (ZnCl2) and thermal reduction procedures help to facilitate dispersion of the metal (Ru) NPs and formation of micro- and mesoporosities favorable for electron transport and ion diffusion in the MOC matrix, hence, the superior electrochemical performances and excellent stability and durability well-suited for application of high-performance supercapactiors.
Table 2
Comparisons of textual and capacitive properties of various biomass-derived activated carbons.
Biomass feedstock
SBETa
Precursor/ZnCl2b
Cc
Electrolyte
Reference
Sugar cane bagasses
1000
1:1.75
300
1 M H2SO4
23
Banana fibers
1097
1:5
296
1 M Na2SO4
28
Coffee beans
1019
1:1
368
1 M H2SO4
45
Coffee shells
842
1:20
150
6 M KOH
46
Camellia Oleifera fruit shells
1935
1:4
374
1 M H2SO4
47
Walnut shells
1073
2 M ZnCl2
117
6 M KOH
48
Fermented rice
2106
—
219
0.1 M KOH
49
Potato wastes
1052
1:2
255
2 M KOH
50
Coconut shells
2440
1:3
246
0.5 M H2SO4
51
Waste filter paper
2170
1:4
302
6 M KOH
52
Moringa Oleifera fruit shells
2473
1:1
291
1 M H2SO4
This work
(a) BET surface area in m2 g−1. (b) Ratio of carbon precursor to ZnCl2. (c) Capacitance in F g−1.
To enhance the energy densities, the fabricated symmetriccell supercapacitor was operated at various potential ranges (Fig. 8a) and scan rates (Fig. 8b). Accordingly, the corresponding calculated specificcapacitance of 36. 2 to 11. 5 F g−1 was obtained using the known total mass of the electrode. The increase in capacitive current with increasing scan rate may be ascribed due to intercalation and deintercalation of ions24. Likewise, the GCD curves recorded at different potential ranges and current densities are displayed in Fig. 8c,d, respectively, which corresponds to a maximum specificcapacitance and current density of 31.6 F g−1 and 0.25 A g−1, respectively. The specificcapacitance so obtained is comparable to earlier literature reports83.
Figure 8
Performance assessments of the symmetric cell supercapacitor fabricated based on the Ru1.0/MOC-900 composite.
CV curves recorded with varied (a) potential ranges and (b) scan rates. Corresponding GCD curves measured at varied (c) potential ranges and (d) current densities. (e) Cyclic stability test at a constant current density of 0.75 A g−1; insets: photographs of the as-prepared electrode during charge and discharge process. (f) Ragone plot of the solid-state device.
The long-term stability of the symmetriccell supercapacitor was also tested upto 1000 charge-discharge cycles at a current density of 0.75 A g−1. As a result, ca. 97% of its initial capacitance was retained after 1000 cycles (Fig. 8e). Finally, the correlation between power and energy densities of such solid-state device was also investigated; a maximum energy density of 6.3 Wh kg−1 at a power density of 250 W kg−1 was achieved (Fig. 8e).In summary, a series of novel porous activated carbon materials derived from Moringa Oleifera fruit shells via a facile annealing and chemical activation process using ZnCl2 as activating agent have been developed and exploited as electrode support for electrochemical energy storage. When incorporated with RuNPs by thermal reduction using ruthenium(III) acetylacetonate as the metal precursor, the Ru/MOC materials so fabricated exhibits not only superior textural properties but also excellent capacitive properties with extraordinarly stability and durability. A maximum specificcapacitance of 291 F g−1 was achieved for the Ru1.0/MOC-900 oC electrode (Ru loading 1.0 wt%; activation temperature 900 oC) at a current density of 1 A g−1. The results obtained from a cycliccharge-discharge test showed that the same electrode retained more than 90% of it original capacitance after 2000 consecutive test cycles at 4 A g−1. We believe that these MOC materials, which possess desirable textural properties, co-exsisting micro- and mesoporosities, and good electrical conductivities should render prospective applications in high-performance energy storage devices, biosensors, and catalysis, especially when combining with different metal oxides or conductive polymers as porous composites.
Experimental
Materials
Research grade ruthenium(III) acetylacetonate (Ru(acac)3, 96%; Acros), zinc chloride (ZnCl2; Sigma-Aldrich), potassium hydroxide (KOH; Sigma-Aldrich)) and poly vinyl alcohol (PVA; Shimakyu) were purchased commercially and used without further purification. Drumstick fruit shells (Moringa Oleifera; Family: Moringaceae) were collected from Theni district, Tamil Nadu, India. All solutions were prepared using doubly distilled water.
Characterization Methods
All powdered x-ray diffraction (XRD) experiments were recorded on a PANalytical (X’Pert PRO) diffractometer using CuKα radiation (λ = 0.1541 nm). Nitrogen adsorption/desorption isotherm measurements were carried out on a Quantachrome Autosorb-1 volumetric adsorption analyzer at −196 oC. Prior to measurement, the sample was purged with flowing N2 at 150 oC for at least 12 h. The pore size distributions were derived from density functional theory (DFT) calculations. The morphology of the sample was studied by field-emission transmission electron microscopy (FE-TEM) at room temperature (25 oC) using an electron microscope (JEOL JEM-2100F) operating at 200 kV. Elemental compositions of various samples were carried out with an energy-dispersive x-ray (EDX) analyser; an accessory of the FE-TEM facility. X-ray photoelectron spectroscopy (XPS) measurements were performed using a ULVAC-PHI PHI 5000 VersaProb apparatus. Thermogravimetric analyses (TGA) were done on a Netzsch TG-209 instrument under air atmosphere. All Raman spectra were recorded on a Jobin Yvon T64000 spectrometer equipped with a charge coupled device (CCD) detector cooled with liquid nitrogen. The backscattering signal was collected with a microscope using an Ar+ laser centered at 488 nm as the excitation source.
Synthesis of Activated Carbon (MOC)
The Moringa Oleifera fruit shells were first air-dried and cut into pieces, then subjected to microwave irradiation (typically for 1–2 h) to obtain the pyrolytic MO fruit shells (denoted as MOFS). Typically, the activation of the carbon precursor was carried out by impregnating 1.0 g of MOFS in aqueous solution of ZnCl2 (2.0 g) under sonication, followed by an evaporation treatment at 80 oC. The dried MOFS/ZnCl2 mixture was then heated at different temperatures (600, 700, 800, and 900 oC) for 5 h under N2 atmosphere at a heating rate of 5 oC min−1. Finally, the sample was washed with 2M HCl solution and washed with hot deionized water until reaching a pH = 7. The sample was then dried at 80 oC in an air oven. The final as-synthesized carbon products so obtained are named as activated Moringa Oleiferacarbons (MOCs) and the samples prepared at different cabonization temperature (Tc; in oC) are denoted as MOC-Tc.
Preparation of Ru/MOC nanocomposite
Typically, a known amount of activated carbon (MOC) was first impregnated in nitric acid (HNO3) solution for 3 h at 80 oC in an ultrasonic bath. The hydrophilic mixture was then filtrated and dried at room temperature. Then, 0.1 g of the treated powdered MOC-900 sample was impregnated into an ethanol solution (3 mL) containing desirable amount of Ru(acac)3 under continuous ultrasonication for 1 h. Subsequently, the sample vacuum dried to remove the solvent, then subjected to microwave irradiation (at 200 oC for 2 h), followed by pyrolysis at high temperature (900 oC) under N2 atmosphere for 5 h. The nanocomposites so prepared are denoted as Ru/MOC-900, where x represents the Ru loading (x = 1.0 or 1.5 wt%).
Fabrication of Rux/MOC Electrodes
For supercapacitor applications, the Ru/MOC-electrodes were prepared by mixing Ru/MOC (85wt%) and graphite (15 wt%) with 0.4 mL of N-methylpyrrolidone to form a homogeneous slurry. Then, ca. 15 μL of the above slurry was coated on a stainless steel electrode with a dimension of about 1 × 1 cm2 by means of the solution-casting method, followed by drying overnight at 60 oC. The mass loading of Ru/MOC on the substrate was estimated to be ca. 1.5 mg cm−2. For comparison purpose, separate MOC-electrodes (in absence of Ru) were also prepared following the above procedures. For fabricating the symmetric supercapacitor cell, two Ru/MOC electrodes were attached face to face by using PVA/H2SO4 gel electrolyte as a separator.
Electrochemical Studies
Electrochemical properties of various MOC-900 and Ru1.0/MOC-900 electrodes were assessed by using a three-electrode system (in 1.0 M KOH aqueous electrolyte solution) consisting of the fabricated working electrode, Ag/AgCl as the reference electrode, and a platinum (Pt) wire as the counter electrode. All the cyclic votammetry (CV) and galvanostaticcharge-discharge (GCD) experiments were performed on an electrochemical work station (CH Instrument; CHI627). Prior to each measurement, the cell was soaked in the aqueous electrolyte solution (1 M H2SO4) for a few hours. Typically, the CV profiles of the electrodes were recorded in the potential range of −1.0 to 0.0 V. The specificcapacitance of various electrodes were calculated according to equation reported elsewhere2084.
Additional Information
How to cite this article: Lou, B.-S. et al. Ruthenium nanoparticles decorated curl-like porous carbons for high performance supercapacitors. Sci. Rep.
6, 19949; doi: 10.1038/srep19949 (2016).
Authors: Jiabin Wang; Han Zhang; Michael R C Hunt; Alasdair Charles; Jie Tang; Oana Bretcanu; David Walker; Khalil T Hassan; Yige Sun; Lidija Šiller Journal: ChemSusChem Date: 2017-01-18 Impact factor: 8.928