Exothermically Efficient Exfoliation of Biomass Cellulose to Value

Article Cite This: Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

pubs.acs.org/IECR

Exothermically Efficient Exfoliation of Biomass Cellulose to ValueAdded N‑Doped Hierarchical Porous Carbon for Oxygen Reduction Electrocatalyst Cheong Kim,† Chunyu Zhu,*,†,‡ Yoshitaka Aoki,†,‡ and Hiroki Habazaki†,‡ †

Division of Applied Chemistry & Frontier Chemistry Center, Faculty of Engineering, and ‡Graduate School of Chemical Sciences and Engineering, Hokkaido University, Sapporo, Hokkaido 060-8628, Japan

Ind. Eng. Chem. Res. Downloaded from pubs.acs.org by UNIV OF TEXAS AT DALLAS on 02/20/19. For personal use only.

S Supporting Information *

ABSTRACT: In this work, high-value-added N-doped hierarchical porous carbon (NHPC) for oxygen reduction reaction (ORR) electrocatalysis was prepared using biomass cellulose as the raw material. The pyrolysis of cellulose is accelerated by a redox combustion reaction of magnesium nitrate−carbohydrates (urea and cellulose) as absorbed in cellulose fibers, which endows doping with nitrogen, exfoliates the cellulose to highly porous particles, and creates numerous pores simultaneously. After being further carbonized at high temperature and washed with acid, NHPCs were produced that have hierarchical porous structure and large specific surface area. These features are beneficial to the ORR. The influence of four preparation parameters, including species of magnesium salt, carbonization temperature, urea amount, and magnesium salt amount, on the porous characteristics and ORR performance is comprehensively investigated. properties. Charge delocalization in the sp2-hybridized carbon frameworks may be induced due to the doping of electronegative N atoms, which can promote the adsorption of the oxygen (O2) molecule and enrich the densities of free charge carriers that act as active sites for the ORR.3,5,8−11 In addition, the ORR performance of carbon materials is also influenced by factors such as the specific surface area and porosity. Those factors can strongly impact the number of accessible active catalytic sites and the efficient mass transfer. Generally, a 3D hierarchical porous carbon, which has a large specific surface area and an open multiporous structure (containing micro/ meso/macropores), is considered to be efficient. These characteristics can enhance the mass transport of ORRrelevant species (H+, O2, H2O, OH−, etc.) and improve their easy accessibility to the active sites.12−15 Here, a large number of micro/mesopores and the high surface area can increase the number of active sites. In addition, the macropores could

1. INTRODUCTION The oxygen reduction reaction (ORR) is the most challenging step in clean electrochemical energy conversion devices, such as metal−air batteries and fuel cells. Therefore, the exploration of efficient electrocatalysts for the ORR are significantly required for their practical wide applications. So far, platinum (Pt) and other precious metal-based compounds are the most effective ORR catalysts. However, the scarcity, high cost, and poor tolerance of precious-metal-based compounds to methanol and carbon monoxide have hampered their widespread commercialization. Therefore, it is necessary to explore alternative materials that are inexpensive and have high electrocatalytic activity and long-term stability. So far, much effort has been made to explore inexpensive precious-metalfree catalysts, such as oxide1 and heteroatom-doped carbonaceous nanomaterials2−4 and so on. In the past 10 years, researchers have found that carbon nanomaterials with Ndoping are the most promising candidates. This is because of their advantages of low-cost, desirable catalytic activity, good resistance to carbon dioxide poisoning, and low susceptibility to methanol crossover effect.3,5−7 The N-doping strategy is effective because of the unique electronic and structural © XXXX American Chemical Society

Received: Revised: Accepted: Published: A

December 26, 2018 February 4, 2019 February 14, 2019 February 14, 2019 DOI: 10.1021/acs.iecr.8b06410 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

pestle, after which the samples were carbonized at high temperatures ranging from 700 to 1100 °C in Ar. It is noted that the prepyrolysis and carbonization processes can also be combined into one step by heating the precursors directly to high temperatures. We separated the treatment into two processes due to the following reasons: (1) to observe and characterize the samples after pyrolysis and (2) to avoid the pollution of our high-temperature furnace during pyrolysis with a large amount of gas emission. Finally, the carbonized samples were washed with 0.5 M hydrochloric acid (HCl) aqueous solution, washed with ultrapure water and ethanol several times, and dried for further use. The samples were named as MgxUryCot1g-temperature, based on the magnesium salt amount, urea amount, and calcination temperature, for example, Mg10Ur0Cot1g-1000, Mg10Ur20Cot1g-1000, and so on. When magnesium acetate tetrahydrate was used, the samples were referred to as MgAc10Ur0Cot1g-1000 and MgAc10Ur20Cot1g-1000. In total, four experimental parameters, including magnesium salt species (magnesium nitrate and magnesium acetate, 10 mmol), calcination temperature (from 700 to 1100 °C), urea amount (0, 20, 30, and 40 mmol), and magnesium nitrate amount (10 and 20 mmol), were investigated to find an optimized production condition for NHPC possessing the highest electrocatalytic ORR properties. Generally, the formation of MgO@C composite precursor by the exothermic pyrolysis of magnesium nitrate−urea− cellulose can be represented by the following reaction formula:

promote the mass transport of ORR-relevant species to the active sites. Until now, N-doped carbons with good ORR properties have been obtained by thermal annealing of carbon materials with ammonia (NH3)14 and the direct pyrolysis and calcination of N-containing precursors such as melamine, dopamine, porphyrin, cyanamide, and aniline.13,16−19 However, the high-cost raw materials and low output of these synthetic methods have limited their application. In addition, the direct preparation of porous carbon with designed hierarchical porous structure is still difficult. Therefore, the development of an ecofriendly and inexpensive strategy to prepare N-doped hierarchical porous carbon (NHPC) for superior ORR electrocatalysis is still of great interest. Biomass, as a renewable source, has received tremendous attention in the preparation of carbon materials because of its low cost, abundance, and environmental friendliness.20−22 Carbon materials for ORR electrocatalyst have been explored from diverse biomass resources, such as from grass,9 pomelo peel,23 cocoon silk,24 sewage sludge,25 Lycium barbarum L.,26 pine needles,27 cellulose,8 gelatin,15 chitosan,28 and starch.29 Among various biomass resources, cellulose is one of the most sustainable and abundant biomass materials in nature. Cellulose is an important structural component of the primary cell wall of green plants such as cotton and wood. Therefore, cellulose is a promising raw material for carbon.30 To obtain a carbon product using cellulose raw material such as cotton, an endothermic pyrolysis process is conventionally required under an Ar/N2 atmosphere.31,32 In order to obtain a product with high specific surface area, an activation process is normally needed, for example, using KOH,31 ammonium phosphate [(NH4)3PO4],33 urea,8 and zinc chloride (ZnCl2).32 However, the electrocatalytic properties still need to be modified. Especially, the efficient and clean preparation of NHPC with high specific surface area and 3D hierarchical porous structure is still challenging. Herein, by considering all the above problems, we report a facile and efficient strategy to prepare value-added NHPC particles using biomass cellulose as raw material. The pyrolysis of cellulose is accelerated by an exothermic reaction using magnesium nitrate−carbohydrates (urea and cellulose), which are absorbed in the cellulose fibers. The incorporation of magnesium nitrate−urea endows the doping with nitrogen, exfoliates the fibers to fine 3D porous powders, and creates a large quantity of pores at various scales simultaneously. The NHPC exhibits good electrochemical performance as an ORR electrocatalyst, and the preparation of NHPC is optimized in terms of species of magnesium salt precursors, calcination temperature, urea amount, and magnesium salt amount.

Mg(NO3)2 + CH4N2O + (C6H10O5)n → MgO + C (containing dopant N, H, O) + gases (NOx , CO, CO2 , H 2O, et al.)

(1)

In this work, cotton cellulose was selected as the raw material for the initial assessment of this new exothermic reaction. In fact, various kinds of low-cost cellulose raw materials, such as pulp cellulose and recycled cellulose, can also be employed, which could further reduce the production cost. 2.2. Material Characterization. The samples were characterized by X-ray diffraction (XRD; Rigaku Miniflex with Cu Kα radiation), transmission electron microscopy (TEM; 200 kV, JEOL, JEM-2010F), and scanning electron microscopy (SEM; JEOL, JSM-7400F). An X-ray photoelectron spectroscopy (XPS; JEOL, JPS-9200) system using a Mg Kα X-ray source (hν = 1253.6 eV) was used to characterize the surface functional groups of the samples. Raman spectra of the samples were obtained by a RENISHAW Raman spectrometer (inVia Reflex). The pyrolysis behavior for the nitrate, urea, cotton, and their mixtures was evaluated by a thermogravimetric (TG) analyzer combined with a differential scanning calorimetric (DSC) analyzer (Mettler Toledo). The carbon contents of the calcined MgO@C composites were also determined by their air combustion during TG analysis. The specific surface area (SSA), pore size distribution, and pore volume of the samples were characterized by N2 adsorption−desorption using a Microtrac-BEL (BELSORPmini) surface area analyzer, which is cooled by liquid nitrogen. The samples were outgassed at 250 °C for 6 h under vacuum pumping prior to the gas adsorption measurements. The total pore volumes (V0.95) were determined from the N2 absorption amount at the relative pressure P/P0 of 0.95. The SSA were

2. EXPERIMENTAL SECTION 2.1. Preparation of N-Doped Porous Carbon. In the experiment, a certain amount of magnesium nitrate hexahydrate [Mg(NO3)2·6H2O; 10 or 20 mmol] or magnesium acetate tetrahydrate [Mg(CH3COO)2·4H2O; 10 mmol] and urea [CO(NH2)2; 0, 20, 30, or 40 mmol] were dissolved in 5 mL of ultrapure Milli-Q water to form a homogeneous solution. Subsequently, 1 g of commercially available medical absorbent cotton cellulose was put into the above solution. After that, the wet cotton was dried. The dried samples were heated to 500 °C for prepyrolysis at a heating rate of 10 °C min−1 under Ar flow in a vertical tubular reactor. Subsequently, the pyrolyzed samples were pulverized using a mortar and B

DOI: 10.1021/acs.iecr.8b06410 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article


Figure 1. Schematic diagram of the preparation of N-doped porous carbon particles with hierarchical pore structure.

Figure 2. TG−DSC curves of the pyrolysis behavior of different precursors: (a) magnesium nitrate hexahydrate, (b) urea, (c) cotton, (d) Mg10Ur20, (e) Mg10Ur0Cot1g, (f) Mg10Ur20Cot1g, (g) MgAc10Ur0Cot1g, and (h) MgAc10Ur20Cot1g.

presents the schematic diagram for the preparation of NHPC from cotton cellulose raw material, which is innovated by an exothermic reaction accelerated pyrolysis process to exfoliate the micron-sized bulky cellulose fibers to highly porous carbon. The raw materials of cotton fibers are first impregnated with magnesium nitrate−urea aqueous solution and subsequently dried. The dried fiber precursors are further heated to 500 °C at 10 °C min−1 for their prepyrolysis, where the intensive exothermic reactions occur at around 200 °C. The prepyrolyzed precursors are further pulverized and heat treated at high temperatures ranging from 700 to 1000 °C to increase the carbonization of the final products. Finally, after acid washing, the porous carbon products are obtained. From the SEM observation shown in Figure S1 (Supporting Information, SI), it is confirmed that the raw cotton fibers have diameters of several tens of micrometers and are hollow. These bulky fibers can retain their original morphology after the conventional direct pyrolysis process under Ar atmosphere. Figure S2 (SI) presents the SEM images of the cotton fibers after the impregnation and drying of magnesium nitrate or magnesium nitrate−urea aqueous solutions. The cellulose

determined by both the Brunauer−Emmett−Teller (BET) and t-plot methods. The total surface area (atotal), external surface area (aex), and micropore volume (Vmicro) were obtained by tplot analysis. The micropore surface area (amicro) was calculated as atotal − aex. The mesopore volume (Vmeso) was determined as V0.95 − Vmicro. Here, we used V0.95 as the sum of meso- and micropores to avoid the overestimation of mesopore volume. The pore size distributions were evaluated by both the Barrett−Joyner−Halenda (BJH) and nonlocalized density functional theory (NLDFT) methods by using the absorption data. The slit-type pore model is used for the NLDFT simulation. 2.3. Electrochemical Measurement. The electrochemical characterization was performed in a conventional threeelectrode system using a rotating disk electrode (RDE). The details are shown in the Supporting Information.

3. RESULTS AND DISCUSSION 3.1. Materials Preparation and Characterization. 3.1.1. Exothermic Reaction Accelerated Pyrolysis of Nitrate−Urea−Cellulose to Porous Powders. Figure 1 C


Article


Figure 3. Morphology observation results. Typical SEM images of the samples after acid washing: (a, b) MgAc10Ur0Cot1g-1000, (c, d) MgAc10Ur20Cot1g-1000, (e, f) Mg10Ur0Cot1g-1000, and (g, h) Mg10Ur20Cot1g-1000. TEM observation results for the highly porous MgO@C precursor (i, j, k, l) and the corresponding carbon sample (m, o, p) for Mg10Ur20Cot1g-1000: (i, m) typical TEM images at low magnifications, (j, k) typical TEM images at high magnifications, (l, o, p) the high-resolution TEM images at high magnifications, and (n) a typical SAED pattern.

fibers can retain their fiber shape after the absorption of nitrate or nitrate−urea, indicating the impregnation of nitrate or nitrate−urea into the fibers. The pyrolysis of metal nitrate and urea is a well-known solution combustion synthesis (SCS) process to produce metal oxides.34,35 SCS is an exothermic redox reaction process. The reaction often occurs at around 150−300 °C, which also emits a large amount of gases in a short period. The key for the SCS process is the employment of nitrates as the oxidizer and organic carbohydrate compounds as the reductant fuels, such as urea, glycine, and critic acid, to form an intensive exothermic redox reaction.36−38 The simultaneous absorption of magnesium nitrate and urea in cotton fibers can accelerate the pyrolysis of cellulose. It is noted that cellulose itself can also be a reductant fuel when mixed with nitrates to form an exothermic reaction like SCS, although the fibers are in a solid form. To evaluate the pyrolysis behavior, different raw materials, including magnesium nitrate, urea, cotton, and their mixtures, are heated in a TG−DSC measurement setup under Ar flow, as shown in Figure 2. The decompositions of magnesium nitrate, urea, and cotton are all endothermic reactions. The complete decomposition of magnesium nitrate needs a temperature of

higher than around 500 °C, while the decomposition of urea finishes at around 450 °C. The pyrolysis of cotton cellulose starts at around 310 °C and finishes at around 390 °C. The pyrolysis of magnesium nitrate−urea shows a weak exothermic reaction at around 320 °C under the Ar atmosphere. It is very interesting that when cotton cellulose is mixed with magnesium nitrate or magnesium nitrate−urea, high exothermic reactions occur at 150 and 200 °C, respectively. Here, magnesium nitrate acts the oxidizer, while urea and/or cotton cellulose are employed as the reductant fuel to form the redox exothermic reaction. It is observed in the experiment that the addition of urea can make a homogeneous distribution of magnesium nitrate and urea as adsorbed in cotton fibers as compared to that of only magnesium nitrate, which is due to the chelating of urea molecule with nitrate and cellulose. As shown in Figure S2 (SI), some particles referring to magnesium nitrate are dispersed on the surface of fibers for the nitrate−cotton precursor, while the surface of the fibers are smooth when magnesium nitrate and urea are absorbed. Furthermore, the addition of urea together with magnesium nitrate can create a good exfoliation of cotton fibers to give a highly porous architecture and also offer nitrogen-doping to the final carbon product. It is observed that, during the D


Article


Figure 4. N2 absorption/desorption isotherms (a, d), BJH pore size distribution (b, e), and NLDFT pore size distribution (c, f) for the samples obtained after calcination treatment (a, b, c) and after acid washing (c, d, f). Here, the samples obtained with magnesium nitrate and magnesium acetate precursors are compared.

pyrolysis process under Ar flow in a vertical tubular reactor for the magnesium nitrate or nitrate−urea absorbed cotton fibers, a large amount of gases were emitted in a short period and the pyrolysis process finished in several minutes; however, the pyrolysis for pristine cotton required much longer time and higher temperature. In order to further demonstrate the importance of using nitrate as the oxidizer in the redox exothermic reaction, magnesium acetate was also used to substitute for the nitrate. As evaluated by TG−DSC, shown in Figure 2, a typical exothermic reaction is not observed for the pyrolysis of the magnesium acetate-added cotton fibers (samples MgAC10Ur0Cot1g and MgAC10Ur20Cot1g). The nitrate-induced exothermic reaction can accelerate the pyrolysis of cotton fibers (pyrolysis temperature decreases from around 400 °C to less than around 200 °C) and exfoliate the fibers to 3D highly porous architecture, while the acetate-induced samples retain almost the original fiber shape, as shown by the comparison of SEM images of the prepyrolyzed samples in Figure S3 (SI). A further comparison of these differences for various production conditions will be shown for their high-temperature carbonized samples and acid-washed samples. 3.1.2. High-Temperature Carbonization To Obtain WellCarbonized Samples (XRD, SEM, and TEM Characterization). Figure S4 (SI) displays the XRD patterns for the samples after high-temperature carbonization and after acid washing, respectively. The samples are classified in terms of four experimental parameters, including magnesium salt species, calcination temperature, urea amount, and magnesium nitrate amount. After calcination, all samples present XRD peaks that can be indexed to the MgO phase (JCPDS No. 00004-0829). The peaks for carbon are not detected due to the

low crystallinity of the annealed carbon and the low amount of carbon as compared with the MgO phase. After acid washing, from the XRD patterns, it is confirmed that the peaks belonging to MgO phase disappear for all samples and broadened peaks corresponding to carbon 002 and 101 peaks are observed, indicating the successful removing of MgO by acid leaching. The micromorphology of the calcined MgO@C composites and acid-washed samples was observed by SEM. Figure 3a−h shows the typical SEM images for the samples after acid washing to compare effect of the addition of magnesium nitrate and magnesium acetate in the raw materials. The acetateinduced samples (Figure 3a−d; MgAC10Ur0Cot1g-1000 and MgAC10Ur20Cot1g-1000) present the bulky pulverized micron fibers, which are not porous from the SEM morphology. However, the nitrate-induced samples (Figure 3e−h; Mg10Ur0Cot1g-1000 and Mg10Ur20Cot1g-1000) show highly 3D porous architectures, which contain pores in a wide range from several tens of nanometers to several hundreds of nanometers. Sample Mg10Ur0Cot1g-1000 contains some porous fibers indicative of its raw material of cotton fibers, while sample Mg10Ur20Cot1g-1000 consisted of highly porous powders, indicating the complete exfoliation of cotton fibers by a nitrate−urea-induced exothermic reaction as absorbed in the raw cellulose fibers. More SEM images for the samples prepared at different experimental parameters are shown in Figuress S5 and S6 (SI). Figure 3i−p shows the typical TEM observation results for the highly porous samples of Mg10Ur0Cot1g-1000 obtained after carbonization and after acid washing. For the MgO@C precursor obtained after carbonization, the TEM image at a low magnification (Figure 3i) indicates that the sample is E


Article

Industrial & Engineering Chemistry Research Table 1. Summary of the Porous Characteristics for Porous Carbon Obtained after Acid Washing Mg10Ur20Cot1g-700 Mg10Ur20Cot1g-800 Mg10Ur20Cot1g-900 Mg10Ur20Cot1g-1000 Mg10Ur20Cot1g-1100 MgAc10Ur0Cot1g-1000 MgAc10Ur20Cot1g-1000 Mg10Ur0Cot1g-1000 Mg10Ur30Cot1g-1000 Mg10Ur40Cot1g-1000h Mg20Ur40Cot1g-1000

BET SSAa (m2 g‑1)

atotalb (m2 g‑1)

aexc (m2 g‑1)

amicrod (m2 g‑1)

V0.95e (cm3 g‑1)

Vmicrof (cm3 g‑1)

Vmesog (cm3 g‑1)

994 1191 1114 1173 1163 720 252 969 1021 400 887

1020 1190 1079 1158 1167 705 251 1038 1027 − 862

360 446 460 495 549 54 52 129 550 − 514

660 744 619 663 618 696 199 909 477 − 348

1.05 1.33 1.35 1.51 1.59 0.68 0.33 0.69 1.43 0.73 1.29

0.39 0.53 0.51 0.61 0.61 0.59 0.25 0.46 0.45 0.04 0.35

0.66 0.80 0.84 0.90 0.98 0.09 0.08 0.23 0.98 0.69 0.94

a Specific surface area as calculated from the adsorption data by the BET method. bThe total surface area as determined by the t method. cThe external surface area as determined by the t method. dThe micropore surface area as determined by atotal − aex. eTotal pore volume at P/P0 = 0.95, corresponding to the pores with diameters up to around 40 nm. fThe micropore volume as analyzed by the t method. gThe mesopore volume as determined by V0.95 − Vmicro. hSample Mg10Ur40Cot1g-1000 is not applicable to t-plot analysis due to the lack of micropores for this sample.

show the same conclusions for the meso/macropore distribution as the BJH results. It is further confirmed that large mesopores and macropores are absent for the acetatederived samples. These observations about the difference in pore structure of the MgO@C composites obtained with different magnesium salt precursors demonstrate the importance of the nitrate-assisted exothermic reaction in creating numerous meso/macropores by exfoliating the micron-sized cotton fibers to a highly 3D porous architecture, which was also confirmed by previous morphology observation. The BET SSA, pore volume information, for the MgO@C composites are summarized in Table S1 (SI). After acid washing of the MgO@C composites, the obtained carbon samples show greatly increased volume for N2 adsorption, as shown in Figure 4d, especially for the magnesium nitrate-derived samples. The N2 adsorption isotherms for the acid-treated samples (excluding sample MgAC10Ur20Cot1g-1000) show strong N2 adsorption at a relative pressure of less than 0.1, which is caused by the presence of numerous micropores. The increased hysteresis loops between the adsorption and desorption branches at ∼0.5−0.8 P/P0 for the magnesium nitrate-derived samples indicate the further creation of abundant mesopores by acid removing of the MgO template. Additionally, the steep adsorption at a relative pressure of ∼0.9−1.0 demonstrates the presence of macropores for sample Mg10Ur20Cot1g-1000. The BJH pore size distribution shown in Figure 4e further confirms the increased pore volumes in both meso- and macropores for sample Mg10Ur20Cot1g-1000 as compared with sample Mg10Ur0Cot1g-1000. However, the acetatederived samples (MgAC10Ur0Cot1g-1000 and MgAC10Ur20Cot1g-1000) show a very limited number of pores with size larger than around 10 nm as compared with the nitrate-derived samples. In addition, the acid-washed samples present especially increased pore volumes in mesopores, with the size ranging from around 2 to 10 nm as compared with their MgO@C precursors, which are created by removing the MgO template. The pore size distributions of the final carbon samples were also confirmed by the NLDFT model, as shown in Figure 4f. The analysis using the NLDFT model can show the same conclusions as the BJH analysis in the meso- and macropore size distribution for these samples. Furthermore, the NLDFT model can further give the detailed information for the micropores. These acid-washed samples show greatly

highly porous, containing a large number of interconnected mesopores in the range of several tens of nanometers (50 nm) to several hundreds of nanometers. The TEM observation at larger magnifications (Figure 3j−l) illustrates that the porous carbon matrix is incorporated with plenty of MgO nanoparticles with size ranging from several nanometers to dozens of nanometers. After acid washing to remove the MgO nanoparticles, it is confirmed that a highly porous carbon with numerous interconnected meso- and macropores is produced (Figure 3m). The porous carbon is constructed of very thin layers of porous nanosheets. The high-resolution images in Figure 3o,p indicate that the sample is highly defective. More importantly, the sample contains abundant micropores and mesopores of several nanometers. The selected area electron diffraction (SAED) pattern is shown in Figure 3n, representing a typical amorphous carbon structure. 3.1.3. Characterization of the Porous Carbon by N2 Adsorption Measurement. The porous characteristics, including SSA, pore size distribution, and pore volume, of the MgO@C composites and final carbon samples were further evaluated by a N2 adsorption experiment. Figure 4 shows the N2 absorption−desorption isotherms (a, d), BJH pore size distributions (b, e), and NLDFT pore size distribution (c, f) of the MgO@C composites (a−c) and final carbon samples (d− f) obtained with magnesium nitrate and magnesium acetate precursors. As shown in Figure 4a, the MgO@C composites illustrate the hysteresis-type isotherms (absorption−desorption hysteresis with relative pressure P/P0 ranging from ∼0.5 to 0.8), indicating the existence of mesopores. It is noted that the magnesium nitrate-derived composites present larger hysteresis as compared to that of the magnesium acetate-derived samples. The isotherms for these composites show limited adsorption at P/P0 less than 0.1, indicating the shortage of micropores for these samples. The magnesium nitrate-derived composites show a sharp increase of adsorption at a relative pressure of ∼0.9−1.0, demonstrating the presence of macropores. However, such a sharp increase is not observed for magnesium acetate-derived samples. The BJH pore size distributions for these composite samples are shown in Figure 4b, confirming the existence of numerous mesopores for these four samples; however, the macropores for the acetate-derived samples are almost neglectable as compared with nitrate-derived samples. NLDFT pore size distributions are shown in Figure 4c, which F


Article


Figure 5. N2 absorption/desorption isotherms (a, d, g), BJH pore size distribution (b, e, h), and NLDFT pore size distribution (c, f, i) for the samples obtained after acid washing. The samples obtained with different calcination temperature (a−c), different urea amount (d−f), and different MgO template amount (g−i) are compared.

volumes for the obtained porous carbons are summarized in Table 1. The porous characteristics as analyzed by N2 adsorption measurement are also compared for the samples obtained with three other experimental parameters, including calcination temperature, urea amount, and magnesium nitrate amount, which are shown in Figure S7 (SI) for the MgO@C composites and Figure 5 for acid-washed carbon samples. The details for the comparison of SSA and pore volumes for these samples are also summarized in Tables 1 and S1 (SI). For the calcined MgO@C composites, the similar hysteresistype isotherms were obtained with the adsorption features (limited adsorption below 0.1 and steep adsorption higher than 0.9), representing the existence of meso- and macropores, as presented in Figure S7 (SI). However, for the acid-treated carbon products shown in Figure 5, hysteresis-type isotherms with both steep adsorption at low and high P/P0 sides (P/P0 < 0.1 and > 0.9) are confirmed (excluding sample Mg10Ur40-

increased micropore volumes as compared with their MgO@C precursors, indicating the importance of the MgO template in also creating micropores. It is noted that the MgO nanoparticles as dispersed in the carbon matrix are not smooth spheres but have a coarse and rough surface, which can induce many micropore sized cracks and slits, as can be confirmed by the TEM observation in Figure 3l. To summarize, a highly multiporous carbon coexisting with micro-, meso-, and macropores was designed by a magnesium nitrate−urea-assisted pyrolysis of cotton cellulose. The vigorous exothermic reaction accelerated pyrolysis with a large amount of gas emissions exfoliates the micron-size cotton fibers to porous MgO@C powders containing numerous mesoand macropores, and at the same time, MgO nanoparticles are introduced to be dispersed in the carbon matrix, which can create abundant micro- and mesopores after acid washing. The whole process for the formation of NHPC is illustrated in Figure 1. The details for the comparison of SSA and pore G


Article


Figure 6. XPS analysis of the carbon samples: (a) the wide scan of the samples obtained at different calcination temperatures (the wide scan for other samples are shown in Figure S9, SI), (b, c) the N 1s narrow scan, and (d, e) the summary of percentage of different N species.

Cot1g-1000), indicating the coexisting of micro-, meso-, and macropores. As for the carbonization temperature effect (Figure 5a−c), the SSA of the carbon samples is increased from around 1000 m2 g−1 at 700 °C to larger than 1100 m2 g−1 when the temperature is higher than 800 °C. The values of total SSA for the carbon samples calcined from 800 to 1100 °C are similar; however, the external surface area and pore volume show an increasing tendency with temperature. As confirmed by pore size distributions in Figure 5b,c, the samples present pore volume shifts to higher values in the pore size range of around 2−20 nm as the calcination temperature rises, which is caused by the crystal growth of MgO template. As for the urea amount effect in Figure 5d−f, when too much urea (40 mmol) is added, the obtained carbon product shows greatly decreased SSA from larger than 1000 m2 g−1 to around 400 m2 g−1. This is because urea is also a carbon source, and in this case, too much urea makes a greatly weakened exothermic reaction and a greatly decreased ratio of MgO template in the MgO@C composite. Thereby, a decreased number of pores are created when an excess amount of urea is added. It is found that the optimal urea amount is 20 mmol, making the sample Mg10Ur20Cot1g-1000 have the highest SSA and a well-developed hierarchical pore structure, hence indicating the highest ORR catalytic properties, which will be discussed in detail in the following text. The effect of magnesium nitrate precursor amount is presented in Figure 5g−i. It is concluded that too much magnesium precursor will make an excess ratio of MgO template in MgO@C composites; thereby, the original porous structure may collapse when removing the template. Therefore,

the SSA and pore volume are greatly decreased when the MgO template amount is doubled. 3.1.4. TG Air Combustion Analysis To Determine the Yield. The carbon ratios in the MgO@C composites were determined by the combustion in oxygen flow by using a TG analyzer. The TG curves are shown in Figure S8 (SI). The mass percentages of carbon in the MgO@C were calculated by the weight loss from 100 to 600 °C, since carbon is combusted to CO2 gas in temperatures ranging from around 300 to 500 °C, creating the weight loss, while magnesium oxide is retained as the residue. Note that the carbon ratio here refers to the amorphous carbon with doping elements such as N, O, and H, which were also combusted to gases during the measurement. On the basis of the carbon ratios in the MgO@C composites, we can calculate the theoretical yields in the unit of grams of carbon/grams of cotton for these samples, correspondingly. The results are summarized in Table S2 (SI). 3.1.5. X-ray Photoelectron Spectroscopy (XPS) and Raman Analysis of the Porous Carbon. The characterization of the bonding features of the carbon samples was performed by XPS and Raman spectroscopy. Figures 6 and S9 (SI) show the XPS measurement results. Figure 6a presents the XPS survey scan curves for the Mg10Ur20Cot1g carbon samples obtained under different carbonization temperatures, which show three peaks at ∼285.0, 400.0, and 533.0 eV, referring to the C 1s, N 1s, and O 1s electrons, respectively.14,39 Oxygen is normally contained in the amorphous carbon. Nitrogen atom is also doped in carbon, since urea was added in the raw materials. The atomic percentages of N in (N + C) are determined to be 9.98, 9.28, 3.36, 2.02, and 1.03% for the samples calcined from 700 to 1100 °C, respectively. The ratio of doped nitrogen is decreased with carbonization temperature, H


Article


Figure 7. ORR activities. (a) CV curves in both Ar- and O2-saturated electrolyte and (b) LSV curves at different rotation speeds for sample Mg10Ur20Cot1g-1000. The comparison of the ORR activity using LSV curves at 1600 rpm for the samples prepared under different parameters: (c) the calcination temperature effect, (d) magnesium salt precursor effect, (e) urea amount effect, and (f) MgO template amount effect. The comparison with commercial Pt/C is also shown in part f. More information about CV, LSV, K−L plots and electron transfer number are shown in Figures S11, S12, S13, and S14 (SI), respectively.

especially when the temperature is higher than 800 °C. The XPS survey scan curves for other samples obtained with different magnesium salt precursors, different urea amount, and different magnesium precursor amount are also presented in Figure S9 (SI), respectively. At the same carbonization temperature of 1000 °C and when urea is also added in the raw materials, the carbon samples present similar atomic percentages of N in (N + C), which are around 2.01−2.45%. When urea is not added in the raw materials, the obtained samples (MgAcUr0Cot1g-1000 and Mg10Ur0Cot1g-1000) present very limited N atomic percentages of lower than 1%, which is also properly derived from the measurement error. The N atomic percentages for these samples are summarized in Table S3 (SI). The N 1s spectra for the typical samples with N-doping were carefully scanned, as presented in Figure 6b,c. The N 1s spectra can be divided into three typical peaks, including the quaternary N (graphitic N, ∼401.3 eV), pyrrolic or pyridonic N (∼400.1 eV), and pyridinic N (∼398.6 eV).4,9 The ratios for the above three peaks are greatly influenced by the carbonization temperature. Graphitic N is increasing with the temperature increase, while pyridinic and pyrrolic N are decreasing with the temperature increase. The results are summarized in Figure 6d. Graphitic and pyridinic N are usually assumed to be the active sites for the ORR.3,14 With the increase of carbonization temperature, the graphitization degree is enhanced, making a stable and highly conductive carbon; however, the doping amount of N is also decreased, although the ratio for the sum of graphitic and pyridinic N is increased. Therefore, it is significantly important to find the

optimized carbonization temperature to prepare a carbon sample with high ORR activities. Raman spectroscopy has been widely used to characterize carbon materials and is sensitive to the slight changes of C−C bonds. Figure S10 (SI) shows the typical Raman spectra for the obtained carbon samples. The typical D-band and G-band of carbon located at approximately 1350 and 1600 cm−1, respectively, are observed for all samples. D-band is associated with a disordered carbon structure. G-band is referred to the ordered sp2 bonded graphitic carbon. The intensity ratio of ID/ IG represents the disordering degree of carbon materials. The ID/IG ratios for samples is higher than 1.0, representing the great disordering and defective characteristic for these carbon samples, which is attributed to the doping of N element. 3.2. Electrochemical Test. 3.2.1. ORR Activity Measurement by RDE. The electrocatalytic activities of the as-prepared NHPCs were investigated using a RDE setup in 0.1 M KOH electrolyte. CV measurements on the porous carbon electrode were conducted in both Ar- and O2-saturated electrolyte, as shown in Figures 7a and S11 (SI). In Ar-saturated solution, no redox features were observed for any electrode. However, in the O2-saturated electrolyte, significant cathodic ORR peaks between 0.6 to 0.9 V vs RHE are observed, implying the electrocatalytic activity of the NHPCs for the ORR. Moreover, sample Mg10Ur20Cot1g-1000 exhibits the most well-defined cathodic curve and the most positive cathodic peak potential (0.76 V), indicating the good ORR performance for such an electrode. To further investigate the ORR reaction kinetics, RDE measurements were conducted at different rotating speeds. Figure 7b shows the typical LSV curves for sample I


Article


Figure 8. Electrochemical measurement using the chronoamperometric method. (a) Chronoamperometric responses of NHPC and Pt/C on glass carbon electrodes kept at 0.4 V vs Ag/AgCl in O2-saturated 0.1 M KOH electrolyte and (b) chronoamperometric responses of NHPC and Pt/C electrodes with the addition of methanol at 300 s.

nitrate is used in the raw material, which has been discussed in detail in the Material Characterization section. It is also observed that when urea is not added in the raw material, the corresponding carbon products illustrate poorer activity, representing the importance of urea for the doping of N element to carbon. The amount of urea added in the raw material is also optimized, as shown in Figure 7e. Urea20 is an optimal amount, and when excess amount of urea is added, the corresponding carbon shows decreased specific surface area; thereby, a decreased ORR activity is obtained. MgO template amount is also very important, and when an excess amount of MgO template is introduced in the precursor, the corresponding carbon also shows decreased specific surface area; hence, a poorer ORR activity is obtained, as shown in Figure 7f. In summary, sample Mg10Ur20Cot1g-1000 presents the best ORR activities, which are even similar to that of commercial Pt/C, as shown in Figure 7f. The summary of the ORR properties, including onset potential, half-wave potential, and limiting current density are shown in Table S4 (SI). 3.2.2. Chronoamperometric Measurement. Besides the good electrocatalytic activity, the NHPC also exhibits a high durability superior to that of Pt/C catalyst as evaluated by chronoamperometric examination, which is shown in Figure 8a. The measurement was conducted at −0.4 V vs Ag/AgCl in O2-saturated KOH solution at a rotation speed of 1600 rpm. The NHPC electrode (Mg10Ur20Cot1g-1000) can retain ∼88% of its original current density after holding for 10 h. However, commercial Pt/C loses ∼17% of its original current density under the same condition. Moreover, it is well-known that the possible crossover of methanol from anode to cathode in fuel cells causes the poisoning of the Pt catalyst in the cathode. Therefore, the durability of catalysts against possible methanol crossover is another important issue for their practical application. Methanol tolerance was evaluated by the chronoamperometric measurement at −0.4 V vs Ag/AgCl in O2-saturated KOH solution, as shown in Figure 8b, in which 2 vol % methanol was added to the electrolyte. The ORR current does not change obviously after methanol addition for the NHPC electrode, while a significantly decreased current shift is observed immediately after methanol addition for the Pt/C electrode. The results indicate that NHPC has a superior methanol tolerance property, making it a promising cathodic catalyst for alkaline direct methanol fuel cells. The superior electrochemical ORR performance of the NHPC can be explained by its novel hierarchical porous

Mg10Ur20Cot1g-1000 at rotation speeds from 400 to 1600 rpm. The LSV curves for other samples are summarized in Figure S12 (SI). The corresponding K-L plots reveal good linearity, as shown in Figure S13 (SI), suggesting the first-order reaction kinetics of the ORR as a function of dissolved oxygen levels. The estimated numbers of the transferred electrons (n) per oxygen molecule in the potential range of 0.3−0.6 V are plotted in Figure S14 (SI). It is observed that for the hierarchical porous carbon with N-doping and high specific surface area, including Mg10Ur20Cot1g-800, Mg10Ur20Cot1g-900, Mg10Ur20Cot1g-1000, Mg10Ur20Cot1g-1100, Mg10Ur30Cot1g-1000, and Mg20Ur40Cot1g-1000, the corresponding electrodes present n values varying from 3.5 to ∼4.0, which is approaching the ideal four electron transfer pathway. Sample Mg10Ur20Cot1g-700 presents transfer numbers slightly higher than 4, which might be caused by the unexpected side reactions, since the carbon is not wellgraphitized at a low calcination temperature. In contrast, samples Mg10Ur0Cot1g-1000, Mg10Ur40Cot1g-1000, MgAc10Ur0Cot1g-1000, and MgAc10Ur20Cot1g-1000 present low values of transfer number between ∼2.5 and 3. The low selectivity toward the direct four-electron transfer reaction for these four samples is due to their lack of N-doping or nonhierarchical porous structure with low specific surface area. The ORR performance for all electrodes are compared in their LSV curves at a rotating speed of 1600 rpm, as shown in Figure 7c−f. The ORR activities are discussed in terms of experimental parameters, including carbonization temperature (c), magnesium salt species (d), urea amount (e), and magnesium salt amount (f). For the samples prepared with different carbonization temperatures, as shown in Figure 7c, it is observed that sample Mg10Ur20Cot1g-1000 illustrates the best ORR activities, including highest onset potential (0.94 V) and half-wave potential (0.83 V). The carbonization temperature is a very important parameter for preparing N-doped carbon with high ORR performance, which could greatly influence the doping amount of N, the doping species of N, and the graphitization degree of carbon. The results here indicate that an optimized carbonization temperature is 1000 °C in our experimental system, which is also frequently used by other research groups.8,19,23,27 Figure 7d shows the comparison of the samples using different magnesium salt precursors. It is obvious that the nitrate-derived samples present greatly enhanced ORR activity compared with that of the acetate-derived samples. This is because of the formation of a highly hierarchical porous structure and high specific surface area when magnesium J


Article


template amount). Benefiting from the characteristics of hierarchical multiporous, high specific surface area, and Ndoping, the optimized NHPC sample shows superior electrocatalytic ORR activities, comparable to those of the commercial Pt/C catalyst. The good performance combined with the ecofriendly and sustainable method for synthesis reinforces cellulose-derived porous carbon, which is also useful in supercapacitors, Li ion batteries, and pollutant absorbents, as a potential candidate for advanced carbon-based electrocatalysts.

structure with N-doping and high specific surface area. First, the abundant graphitic and pyridinic N-doped carbon offers a large amount of N−C active sites. At the same time, the high specific surface area and hierarchical open pore structure can enhance the penetrability of electrolytes into the interior surfaces, which can increase the number of exposed active sites, provide efficient electrolyte/mass contact, and facilitate fast ion transportation. These characteristics of NHPC greatly contribute to the enhancement of the electrocatalytic activities. Finally, the electrocatalytic performance of NHPC is also comparable to the best reported carbon-based ORR catalysts,2,12,19,29 making it a promising substitute for commercial Pt/C. 3.3. Summary of the Four Parameters for the Formation of NHPC Powders and Their Electrochemical Properties. In summary, the influence of four preparation parameters, including species of magnesium salt precursors, carbonization temperature, added urea amount, and magnesium salt amount, on the porous properties and electrochemical performance of NHPC were comprehensively investigated. The following conclusions can be obtained for each parameter. (1) For the comparison of the effect of magnesium salt precursors, magnesium nitrate and magnesium acetate were used. It was confirmed that the use of nitrate introduced exothermic reactions, which could promote the pyrolysis at a low temperature and quickly exfoliate the cellulose fibers to a highly porous architecture; however, the magnesium acetate-induced reaction was endothermic and the corresponding pyrolyzed samples still had the bulky fiber shape, which was less porous. (2) The carbonization temperature was optimized, which could greatly influence the doping amount of N, the doping species of N, and the graphitization degree of carbon, and it was concluded the NHPC prepared at 1000 °C showed the best electrochemical performance. (3) The importance of urea addition in the raw materials was obvious, since it offered the doping source for N element and promoted the complete exfoliation of cellulose fibers; however, excess addition of urea in the raw materials would decrease the specific surface area of the final carbon. (4) MgO template was important for the creation of micro- and mesopores; therefore, the added MgO amount was optimized and it was found that an excess amount of MgO template could decrease the specific surface area of the final carbon, making a low-activity ORR catalyst. After the discussion of the above four parameters, the optimized NHPC with the highest ORR performance is Mg10Ur20Cot1g-1000.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.iecr.8b06410. Details of electrochemical measurement; tables summarizing the porous characteristics for MgO@C composites, the yields, the nitrogen contents by XPS analysis, and ORR activities; and figures showing the following: SEM images for the original cotton fiber, carbonized cotton fiber, cotton fibers absorbed with nitrate and urea, and prepyrolyzed samples; XRD patterns; SEM images of calcined samples; SEM images of the final carbon samples; N2 absorption/desorption isotherms and pore size distribution; TG curves for MgO/C composites under air flow; XPS wide scans; Raman spectra; CV curves under both Ar- and O2-saturated electrolytes; LSV curves; and K−L plots and calculated transfer number (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Tel: +81-011-706-6736. E-mail: [email protected]. ac.jp. ORCID

Chunyu Zhu: 0000-0002-5975-9308 Yoshitaka Aoki: 0000-0001-5614-1636 Hiroki Habazaki: 0000-0002-7172-8811 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by JSPS KAKENHI Grant Number 18K14047. The authors also thank the “Nanotechnology platform” Program of the Ministry of Education, Culture, Sports, Science and Technology (MEXT) for the use of analysis equipment. This work was also supported partially by 2018 Feasibility Study Program of the Frontier Chemistry Center, Faculty of Engineering, Hokkaido University. The authors also thank Prof. Shin R. Mukai and Prof. Shinichiroh Iwamura for the discussion on nitrogen absorption analysis results.

4. CONCLUSIONS In this study, we have developed a facile and efficient process for the fabrication of NHPC powders from biomass raw material of cellulose for electrochemical ORR catalysts. The pyrolysis of cellulose is accelerated by a redox combustion reaction of magnesium nitrate−carbohydrates (urea and cellulose) as absorbed in the cellulose fibers, which endows the doping with nitrogen, exfoliates the fibers to highly porous particles, and creates a large quantity of pores at multiscales simultaneously. After the exothermic prepyrolysis, the samples were further carbonized at high temperatures and finally washed with acid solution to obtain NHPC. The influence of four preparation parameters on the porous properties and electrochemical performance were comprehensively investigated, including magnesium salt species, carbonization temperature, urea amount, and magnesium salt amount (MgO

■

REFERENCES

(1) Qaseem, A.; Chen, F.; Qiu, C.; Mahmoudi, A.; Wu, X.; Wang, X.; Johnston, R. L. Reduced Graphene Oxide decorated with Manganese Cobalt Oxide as Multifunctional Material for Mechanically Rechargeable and Hybrid Zinc−Air Batteries. Part. Part. Syst. Charact. 2017, 34, 1700097. (2) Hu, C.; Dai, L. Multifunctional Carbon-Based Metal-Free Electrocatalysts for Simultaneous Oxygen Reduction, Oxygen Evolution, and Hydrogen Evolution. Adv. Mater. 2017, 29, 1604942.

K


Article

Industrial & Engineering Chemistry Research (3) Guo, D.; Shibuya, R.; Akiba, C.; Saji, S.; Kondo, T.; Nakamura, J. Active sites of nitrogen-doped carbon materials for oxygen reduction reaction clarified using model catalysts. Science 2016, 351, 361−365. (4) Chung, D. Y.; Lee, K. J.; Yu, S. H.; Kim, M.; Lee, S. Y.; Kim, O. H.; Park, H. J.; Sung, Y. E. Alveoli-Inspired Facile Transport Structure of N-Doped Porous Carbon for Electrochemical Energy Applications. Adv. Energy Mater. 2015, 5, 1401309. (5) Xu, Z.; Zhuang, X.; Yang, C.; Cao, J.; Yao, Z.; Tang, Y.; Jiang, J.; Wu, D.; Feng, X. Nitrogen-Doped Porous Carbon Superstructures Derived from Hierarchical Assembly of Polyimide Nanosheets. Adv. Mater. 2016, 28, 1981−1987. (6) Wang, Q.; Ji, Y.; Lei, Y.; Wang, Y.; Wang, Y.; Li, Y.; Wang, S. Pyridinic-N-Dominated Doped Defective Graphene as a Superior Oxygen Electrocatalyst for Ultrahigh-Energy-Density Zn−Air Batteries. ACS Energy Lett. 2018, 3, 1183−1191. (7) Wang, S.; Ji, X.; Ao, Y.; Yu, J. Vertically Aligned N-Doped Diamond/Graphite Hybrid Nanosheets Epitaxially Grown on BDoped Diamond Films as Electrocatalysts for Oxygen Reduction Reaction in an Alkaline Medium. ACS Appl. Mater. Interfaces 2018, 10, 29866−29875. (8) Wu, X.; Yu, X.; Lin, Z.; Huang, J.; Cao, L.; Zhang, B.; Zhan, Y.; Meng, H.; Zhu, Y.; Zhang, Y. Nitrogen doped graphitic carbon ribbons from cellulose as non noble metal catalyst for oxygen reduction reaction. Int. J. Hydrogen Energy 2016, 41, 14111−14122. (9) Zhang, H.; Chen, J.; Li, Y.; Liu, P.; Wang, Y.; An, T.; Zhao, H. Nitrogen-Doped Carbon Nanodots@Nanospheres as An Efficient Electrocatalyst for Oxygen Reduction Reaction. Electrochim. Acta 2015, 165, 7−13. (10) Yu, D.; Zhou, L.; Tang, J.; Li, J.; Hu, J.; Peng, C.; Liu, H. Nitrogen-Doped Porous Carbon Nanosheets Derived from Coal Tar Pitch as an Efficient Oxygen-Reduction Catalyst. Ind. Eng. Chem. Res. 2017, 56, 8880−8887. (11) Wang, Q.; Lei, Y.; Chen, Z.; Wu, N.; Wang, Y.; Wang, B.; Wang, Y. Fe/Fe3C@C nanoparticles encapsulated in N-doped graphene−CNTs framework as an efficient bifunctional oxygen electrocatalyst for robust rechargeable Zn−air batteries. J. Mater. Chem. A 2018, 6, 516−526. (12) Wan, K.; Tan, A.-d.; Yu, Z.-p.; Liang, Z.-x.; Piao, J.-h.; Tsiakaras, P. 2D nitrogen-doped hierarchically porous carbon: Key role of low dimensional structure in favoring electrocatalysis and mass transfer for oxygen reduction reaction. Appl. Catal., B 2017, 209, 447−454. (13) Chung, H. T.; Cullen, D. A.; Higgins, D.; Sneed, B. T.; Holby, E. F.; More, K. L.; Zelenay, P. Direct atomic-level insight into the active sites of a high-performance PGM-free ORR catalyst. Science 2017, 357, 479−484. (14) Wang, Y.; Liu, H.; Wang, K.; Song, S.; Tsiakaras, P. 3D interconnected hierarchically porous N-doped carbon with NH3 activation for efficient oxygen reduction reaction. Appl. Catal., B 2017, 210, 57−66. (15) Fan, H.; Shen, W. Gelatin-Based Microporous Carbon Nanosheets as High Performance Supercapacitor Electrodes. ACS Sustainable Chem. Eng. 2016, 4, 1328−1337. (16) Rafti, M.; Marmisollé, W. A.; Azzaroni, O. Metal-Organic Frameworks Help Conducting Polymers Optimize the Efficiency of the Oxygen Reduction Reaction in Neutral Solutions. Adv. Mater. Interfaces 2016, 3, 1600047. (17) Sonkar, P. K.; Prakash, K.; Yadav, M.; Ganesan, V.; Sankar, M.; Gupta, R.; Yadav, D. K. Co(ii)-porphyrin-decorated carbon nanotubes as catalysts for oxygen reduction reactions: an approach for fuel cell improvement. J. Mater. Chem. A 2017, 5, 6263−6276. (18) Liu, Y.; Chen, F.; Ye, W.; Zeng, M.; Han, N.; Zhao, F.; Wang, X.; Li, Y. High-Performance Oxygen Reduction Electrocatalyst Derived from Polydopamine and Cobalt Supported on Carbon Nanotubes for Metal−Air Batteries. Adv. Funct. Mater. 2017, 27, 1606034. (19) Zhang, J.; Zhao, Z.; Xia, Z.; Dai, L. A metal-free bifunctional electrocatalyst for oxygen reduction and oxygen evolution reactions. Nat. Nanotechnol. 2015, 10, 444.

(20) Deng, J.; li, m.; Wang, Y. Biomass-derived Carbon: Synthesis and Application on Energy Storage and Conversion. Green Chem. 2016, 18, 4824−4854. (21) Basta, A. H.; Lotfy, V. F.; Hasanin, M. S.; Trens, P.; El-Saied, H. Efficient treatment of rice byproducts for preparing high-performance activated carbons. J. Cleaner Prod. 2019, 207, 284−295. (22) Jiang, W.; Xing, X.; Li, S.; Zhang, X.; Wang, W. Synthesis, characterization and machine learning based performance prediction of straw activated carbon. J. Cleaner Prod. 2019, 212, 1210−1223. (23) Yuan, W.; Feng, Y.; Xie, A.; Zhang, X.; Huang, F.; Li, S.; Zhang, X.; Shen, Y. Nitrogen-doped nanoporous carbon derived from waste pomelo peel as a metal-free electrocatalyst for the oxygen reduction reaction. Nanoscale 2016, 8, 8704−8711. (24) Wang, Y.; Lei, Y.; Wang, H. Astridia velutina-like S, N-codoped hierarchical porous carbon from silk cocoon for superior oxygen reduction reaction. RSC Adv. 2016, 6, 73560−73565. (25) Yuan, S.-J.; Dai, X.-H. An efficient sewage sludge-derived bifunctional electrocatalyst for oxygen reduction and evolution reaction. Green Chem. 2016, 18, 4004−4011. (26) Zuo, L.-X.; Wang, W.-J.; Song, R.-B.; Lv, J.-J.; Jiang, L.-P.; Zhu, J.-J. NaCl Crystal Tuning Nitrogen Self-Doped Porous Graphitic Carbon Nanosheets for Efficient Oxygen Reduction. ACS Sustainable Chem. Eng. 2017, 5, 10275−10282. (27) Wang, M.; Lei, X.; Hu, L.; Zhang, P.; Hu, H.; Fang, J. Highperformance Waste Biomass-derived Microporous Carbon Electrocatalyst with a Towel-like Surface for Alkaline Metal/air batteries. Electrochim. Acta 2017, 250, 384−392. (28) Rybarczyk, M. K.; Lieder, M.; Jablonska, M. N-doped mesoporous carbon nanosheets obtained by pyrolysis of a chitosanmelamine mixture for the oxygen reduction reaction in alkaline media. RSC Adv. 2015, 5, 44969−44977. (29) Men, B.; Sun, Y.; Liu, J.; Tang, Y.; Chen, Y.; Wan, P.; Pan, J. Synergistically Enhanced Electrocatalytic Activity of Sandwich-like NDoped Graphene/Carbon Nanosheets Decorated by Fe and S for Oxygen Reduction Reaction. ACS Appl. Mater. Interfaces 2016, 8, 19533−19541. (30) Perez-Madrigal, M. M.; Edo, M. G.; Aleman, C. Powering the future: application of cellulose-based materials for supercapacitors. Green Chem. 2016, 18, 5930−5956. (31) Shen, W.; Hu, T.; Wang, P.; Sun, H.; Fan, W. Hollow Porous Carbon Fiber from Cotton with Nitrogen Doping. ChemPlusChem 2014, 79, 284−289. (32) Ma, G.; Guo, D.; Sun, K.; Peng, H.; Yang, Q.; Zhou, X.; Zhao, X.; Lei, Z. Cotton-based porous activated carbon with a large specific surface area as an electrode material for high-performance supercapacitors. RSC Adv. 2015, 5, 64704−64710. (33) Liu, Q.; Zhou, Y.; Chen, S.; Wang, Z.; Hou, H.; Zhao, F. Cellulose-derived nitrogen and phosphorus dual-doped carbon as high performance oxygen reduction catalyst in microbial fuel cell. J. Power Sources 2015, 273, 1189−1193. (34) Zhu, C.; Han, C.-g.; Akiyama, T. Controlled synthesis of LiNi0.5Mn1.5O4 cathode materials with superior electrochemical performance through urea-based solution combustion synthesis. RSC Adv. 2015, 5, 49831−49837. (35) Prakasha, K. R.; Prakash, A. S. A time and energy conserving solution combustion synthesis of nano Li1.2Ni0.13Mn0.54Co0.13O2 cathode material and its performance in Li-ion batteries. RSC Adv. 2015, 5, 94411−94417. (36) Ail, S. S.; Benedetti, V.; Baratieri, M.; Dasappa, S. Fuel-Rich Combustion Synthesized Co/Al2O3 Catalysts for Wax and Liquid Fuel Production via Fischer−Tropsch Reaction. Ind. Eng. Chem. Res. 2018, 57, 3833−3843. (37) Adams, R. A.; Pol, V. G.; Varma, A. Tailored Solution Combustion Synthesis of High Performance ZnCo2O4 Anode Materials for Lithium-Ion Batteries. Ind. Eng. Chem. Res. 2017, 56, 7173−7183. (38) Cross, A.; Kumar, A.; Wolf, E. E.; Mukasyan, A. S. Combustion Synthesis of a Nickel Supported Catalyst: Effect of Metal Distribution L


Article

Industrial & Engineering Chemistry Research on the Activity during Ethanol Decomposition. Ind. Eng. Chem. Res. 2012, 51, 12004−12008. (39) Chen, Y.; Ma, R.; Zhou, Z.; Liu, G.; Zhou, Y.; Liu, Q.; Kaskel, S.; Wang, J. An In Situ Source-Template-Interface Reaction Route to 3D Nitrogen-Doped Hierarchical Porous Carbon as Oxygen Reduction Electrocatalyst. Adv. Mater. Interfaces 2015, 2, 1500199.

M


Exothermically Efficient Exfoliation of Biomass Cellulose to Value

Recommend Documents