Method for producing graphitized nanoporous carbon, carbon particles obtained thereby, and their use as highly stable support materials for electrochemical processes - Patents.com
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- シュトゥディエンゲゼルシャフト·コーレ·ゲマインニュッツィゲ·ゲゼルシャフト·ミト·ベシュレンクテル·ハフツング
- Filing Date
- 2023-06-20
- Publication Date
- 2026-06-24
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[Technical Field]
[0001] The present invention relates to a method for producing graphitized nanoporous carbon, the carbon particles obtained thereby, and their use as highly stable supports for electrochemical processes. [Background technology]
[0002] Carbon materials are widely used catalyst support materials for electrochemical processes due to their high conductivity, stability, and ready availability. For polymer exchange membrane fuel cells (PEMFCs), platinum and its alloys supported on carbon black are the state-of-the-art catalysts. Hollow graphite spheres (HGS), disclosed in WO2013 / 117725A1 (Patent Document 1), combine the advantages of carbon black with their uniform shape and well-defined porosity in the low mesopore range. By confining metal nanoparticles (NPs) in the pores, HGS can be used to improve the degradation process against the usual degradation pathway of nanoparticle aggregation and desorption.
[0003] [ka] ionomers, ultimately resulting in catalysts superior to commercially available alternatives. In addition, it has recently been disclosed that small mesopores aid in the gas transport properties of PEMFC catalysts and resistance to poisoning by ionomers, making HGS an even more promising catalyst support material. The synthesis of HGS involves chemical vapor deposition of ferrocene into the pores of a rigid template, followed by annealing and then leaching the rigid template with hydrofluoric acid. While this procedure is highly controlled, it is economical and difficult to scale.
[0004] Methods for producing resorcinol-formaldehyde structures are known in the prior art. For example, Gaikwad Mayur M. et al., in Materials Today Communications, Vol. 20, September 1, 2019, p. 100569, disclose "Enhanced catalytic graphitization of resorcinol formaldehyde-derived carbon xerogel to improve its anodic performance for lithium ion batteries."
[0005] Similarly, Chen Ling et al. disclose "Hierarchical porous graphitized carbon xerogel for high performance supercapacitor" in DIAMOND AND RELATED MATERIALS, ELSEVIER SCIENCE PUBLISHERS, AMSTERDAM, NL, Vol. 121, 20 December 2021, p. 2021.108781 (Non-Patent Document 2).
[0006] However, a common feature of the obtained structures is that although they have the expected structure, they do not meet the high level material requirements for electrochemical purposes which are still to be improved. [Prior art documents] [Patent documents]
[0007] [Patent Document 1] WO2013 / 117725A1 [Non-patent literature]
[0008] [Non-Patent Document 1] Gaikwad Mayur M.et al.,Materials Today Communications,Vol.20,1 September 2019,p.100569 [Non-patent document 2] CHEN LINGet al.,DIAMOND AND RELATED MATERIALS,ELSEVIER SCIENCE PUBLISHERS,AMSTERDAM,NL,Vol.121,20 December 2021,p.2021.108781 Summary of the Invention [Problem to be solved by the invention]
[0009] It is therefore an object of the present invention to provide a material that can be produced more economically while retaining the beneficial properties of HGS, namely their high degree of graphitization and well-defined porosity in the low mesopore range. [Means for solving the problem]
[0010] The above-mentioned object of the present invention is achieved by providing a method for producing graphitized nanoporous carbon, in particular graphitized nanoporous carbon derived from resorcinol-formaldehyde (RF) gel, which allows for a considerably simplified production procedure, the carbon obtained thereby, and its use as a highly stable support material for electrochemical processes.
[0011] A key inventive aspect of the present synthetic method is the simultaneous formation of graphitic and mesoporous carbon structures using a scalable and simple reaction sequence. More specifically, the mesopores are in a size range (1-10 nm) that allows for the pore confinement of metal nanoparticles, and are located within the carbon matrix (intracerebrospinal pores) rather than between individual particles. This well-defined porosity allows for the pore confinement of the active metal, which leads to the demonstrated superior stability of the catalyst against electrochemical degradation. Additionally, the method does not rely on laborious steps such as supercritical drying, freeze-drying, or solvent exchange, as used in the prior art, and is therefore economically favorable compared to other previously reported methods.
[0012] More particularly, the present invention provides a method for producing graphitized nanoporous carbon, comprising the steps of: a) a mixture of the following: i), ii), iii), optionally and iv), namely i) a nucleophilic compound selected from melamine, melam, melem, ammeline, 4,6-amino-dihydroxy-1,3,5-triazine, aminophenol, diaminobenzene, dihydroxybenzene, trihydroxybenzene, and combinations thereof; ii) formaldehyde or its oligomers in stoichiometric excess relative to compound i) for complete polymerization; iii) metallic graphitization catalysts, wherein the molar ratio of the metallic graphitization catalyst to compound i) is in the range of 1-30:1-3, optionally in the range of 1-10:1-3, and is selected from Fe, Co or Ni salts of organic acids or acidic organic compounds, in particular from Fe formate, Fe oxalate, Ni oxalate, Ni formate, Co oxalate, Co formate, Fe(II)(acac)2, Fe(III)(acac)3, Ni(II)(acac)2, Co(II)(acac)2, Co(III)(acac)3 or combinations thereof, arbitrarily and iv) a catalytic amount of a promoter compound, preferably ammonia; in a solvent comprising water and a C1-C3 aliphatic alcohol in a volume ratio of 0% to 100% by volume based on the total volume of the solvent, with the remainder being C1-C3 aliphatic alcohol, preferably an aqueous solution having a C1-C3 aliphatic alcohol content of 10% to 40% by volume; b) heating the mixture obtained in step a) to a temperature range between 30°C and 100°C and maintaining the mixture at said temperature range for at least 6 hours, preferably from 6 to 60 hours, more preferably from 6 to 50 hours; c) separating the reaction product obtained in step b) from the aqueous solvent, optionally grinding the separated reaction product, and optionally washing the separated reaction product with fresh aqueous solvent; d) subjecting the separated reaction product obtained in step c) to a high temperature graphitization process at a temperature ranging from 600°C to 1000°C to provide a graphitic framework; e) subjecting the graphitized product obtained in step d) to a process for leaching the metals of the metallic graphitization catalyst, preferably by treatment with an inorganic acid; and f) washing the reaction product obtained in step e) with water and C1-C3 aliphatic alcohol, and finally drying the washed product, optionally followed by a grinding step; The present invention relates to the method comprising the steps of:
[0013] In the method of the present invention, a catalytic amount of a promoter compound can be used to accelerate or initiate the condensation process. The promoter compound can be a weak acid or basic compound, such as sodium carbonate or sodium bicarbonate. Preferably, ammonia is used because it evaporates completely during the carbonization process.
[0014] The solvent for condensation contains water and a C1-C3 aliphatic alcohol in a volume ratio of 0 to 100% by volume based on the total volume of the solvent, with the remainder being C1-C3 aliphatic alcohol, and the solvent is preferably an aqueous solution containing 10 to 40% by volume of the C1-C3 aliphatic alcohol, preferably ethanol.
[0015] In the method of the present invention, a nucleophilic compound selected from melamine, melam, melem, ammeline, 4,6-amino-dihydroxy-1,3,5-triazine, aminophenol, diaminobenzene, dihydroxybenzene, trihydroxybenzene, and combinations thereof is reacted with a stoichiometric excess of formaldehyde or an oligomer thereof relative to compound i) to substantially completely polymerize / condense compound i) with formaldehyde to form a polycondensed network.
[0016] Yet another elaborate method for producing graphitized nanoporous carbon comprises: g) impregnating the graphitized nanoporous product obtained in step f) with a catalytically active metal in the form of a solution of a metal salt or a mixture thereof selected from Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Al, Mo, Se, Sn, Pt, Ru, Pd, W, Ir, Os, Rh, Nb, Ta, Pb, Bi, Au, Ag, Sc, Y (preferably Pt, Ru, Pd, Au, Ag, Ni, Co, Ir) or a salt or mixture thereof, preferably by the incipient wetness method, h) subjecting the graphitized nanoporous product obtained in step g) to a hydrogenation process to form catalytically active metal sites; i) subjecting the graphitized nanoporous product obtained in step h) to a further temperature treatment in an atmosphere inert to the metal particles and the graphitized nanoporous product at a high temperature ranging from 600°C to 1400°C, preferably from 600°C to 1000°C, in order to modify the microstructure and confine catalytically active metal sites in the mesopores, improving the chemical and electrochemical properties; Additional steps are included, including:
[0017] In one embodiment of the process of the present invention, compound i) is selected from resorcinol, melamine and its derivatives, in particular resorcinol.
[0018] In another embodiment of the method of the present invention, the metallic graphitization catalyst is selected from Fe(II)(acac)2, Fe(III)(acac)3, Ni(II)(acac)2, Co(II)(acac)2, Co(III)(acac)3, or combinations thereof.
[0019] In yet another embodiment of the process of the present invention, the concentration of compound i) in the solvent ranges from 0.01 mol / l to 0.5 mol / l when stirring the mixture in step b).
[0020] In yet another embodiment of the process of the present invention, the concentration of compound i) in the solvent ranges from 0.5 mol / L to 2.0 mol / L if the mixture is not stirred in step b), and the resulting product often needs to be crushed due to its blocky structure.
[0021] The present invention also relates to graphitized nanoporous carbon particles obtainable according to any one of the methods described above, as well as to their use as catalysts, in particular for the cathode-side oxygen reduction reaction (ORR) in PEM fuel cells.
[0022] Step i), in which the metal-loaded graphite particles obtained in step h) are further treated / heat-treated at elevated temperatures ranging from 600 to 1400°C, preferably from 600 to 1000°C, for a time period ranging from 1 to 10 hours, more preferably from 1 to 5 hours, preferably under an inert atmosphere such as argon, with the aim of modifying the microstructure of the particles and promoting the confinement of the metal particles in the pore structure, is particularly important for improving the electrochemical stability of the nanoparticles under adverse potential cycling conditions.
[0023] After the leaching step, mesoporous graphite particles are obtained, which can be further impregnated in an impregnation step with a solution of a salt of a catalytically active metal such as Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Al, Mo, Se, Sn, Pt, Ru, Pd, W, Ir, Os, Rh, Nb, Ta, Pb, Bi, Au, Ag, Sc, or Y, or a combination thereof, preferably Pt, Ru, Pd, Au, Ag, Ni, Co, or Ir, where the volume of the solution of the metal salt, preferably a chloride, or a combination of salts is completely absorbed into the mesopores of the graphite particles, and optionally dried. The impregnation can be carried out with an alcoholic solution of the metal salt in an amount such that the loading of the catalytically active metal on the particles is 10 to 40 wt. %, calculated based on the total weight of the final dried particles.
[0024] The catalytically active metal that can be used in the graphite mesoporous particles according to the present invention is not limited and may be one of Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Al, Mo, Se, Sn, Pt, Ru, Pd, W, Ir, Os, Rh, Nb, Ta, Pb, Bi, Au, Ag, Sc, Y, or a combination thereof. Suitable metals will vary depending on the reaction to which the gold-loaded graphite particles according to the present invention are to be subjected. In addition, the metal may be a single metal or an alloy of two or more metals.
[0025] The resulting graphite particles, loaded with a metal precursor or, in the case of alloys, a mixture of metal precursors, are subjected to a hydrogenation process, preferably carried out as a gas-phase hydrogenation. Such gas-phase reduction can be applied at elevated temperatures ranging from 100 to 500°C under normal pressures. This hydrogenation results in the formation of metal clusters within the mesopore structure of the graphite particles. A subsequent heat treatment step at elevated temperatures defines the final microstructure of the metal or alloy nanoparticles and promotes pore confinement for improved chemical and electrochemical properties.
[0026] To complete this invention, the inventors analyzed options for producing suitable carbon support materials. For example, resorcinol-formaldehyde gels are polymers produced via the polycondensation of resorcinol and formaldehyde. Their synthesis is diverse, and variations in pH, temperature, solvent, drying conditions, and added salts, bases, or acids result in different polymer morphologies. Due to their high surface area and tailorable porosity, carbonized RF gels have been investigated in the state of the art for application in electrochemical devices, such as supercapacitors or batteries. For PEMFCs, RF gels have been considered as support materials, but their success has been limited, likely due to their low degree of graphitization and high micropore volume. Existing literature describing the synthesis of RF gels is extensive, and indeed describes both mesoporous and graphitized carbon materials derived from RF gels, as well as combinations thereof. To analyze why none of these materials have been competitive with carbon black for use as support materials for PEMFCs, the properties of RF systems are described in more detail below.
[0027] The production of RF polymers can follow a sol-gel process. First, hydroxymethylresorcinol is formed through the reaction of resorcinol with formaldehyde. These polyalcohols then condense with the elimination of water to form dispersed nanometer-sized clusters, also known as sols. This sol can be aged, often at elevated temperatures in a closed container, to prepare a gel. This gel is a three-dimensional, highly porous, fractal structure of individual primary particles, the striking shape of which must often be preserved. For this purpose, supercritical drying (using, for example, CO2) is applied to form aerogels that retain up to 90% of the gel's volume. In some cases, this drying can be followed with equal success by freeze-drying, which is referred to as a cryogel. If such elaborate drying techniques are to be avoided, subcritical drying can yield xerogels, i.e., polymers that are collapsed but still have a highly porous structure. Alternatively, if the synthesis is carried out under high dilution and agitation, individual spheres can be produced similar to the Stöber process for preparing nanodispersed silica spheres.
[0028] For use in electrochemical devices, RF gels must be conductive and can therefore be carbonized and further activated to tailor the properties of the RF carbon. The molecular composition of the gel is dominated by the benzene rings of resorcinol, thus retaining a high proportion of its mass and pore structure upon carbonization. Without the aggregate structure of primary particles, carbonized RF spheres possess porosity in the micropore range. These pores form because reaction products such as water, CO, and CO2 escape from the bulk polymer during the conversion to carbon. For the production of mesoporous RF carbon, RF gels are the precursor of choice. Because the size and connectivity of the primary particles that form the gel can be tailored to a great extent, pores can be generated at the junctions of these clusters or as pockets within the gel structure. Through modifications of the synthetic procedure, RF carbons with pore sizes spanning the entire mesopore range have already been produced in the prior art. However, it is important to note that the pore size distribution is often broad and, moreover, the pores are not located within the particle but rather at the edges and corners of small, fractal primary particles, making them different from the typical pores of hollow graphite spheres. Therefore, it is highly questionable whether the confinement of metals in these mesopores can provide the same stability enhancement known from HGS, where the mesoporosity is intraparticle-based. Similar concerns are raised about the improved transport properties and protection against poisoning provided by ionomers that are enabled by mesoporous supports. RF with intraparticle porosity has been reported using surfactants or hard templates, but although very well-defined textures can be obtained, the synthetic sequence is complicated and shares the drawbacks of the HGS production described above.
[0029] The improvement of RF carbon conductivity through graphitization, as demonstrated here, offers a promising route to improving graphitic carbon's suitability for electrochemical processes where high stability and conductivity are beneficial. In the prior art, the conductivity of different RF xerogels carbonized at 800°C was compared with that of carbon black, and the latter was found to be 10 times higher, regardless of the xerogel's composition. The addition of a graphitization catalyst before carbonization and annealing at high temperatures is a means to enhance the graphitization of RF carbon. Without the use of a catalyst, temperatures well above 2000°C are generally required to form graphitic carbon from the polymer. Therefore, transition metal catalysts, such as iron or nickel, are typically incorporated into the RF polymer before carbonization, either by addition to the synthesis mixture or by post-impregnation of the polymer. While graphitic carbon can be produced from RF polymers via these methods, the flexibility of the synthesis is significantly reduced. When metal salts are incorporated into the synthesis mixture, the complex reaction mixture requires further balancing, for example, to counteract the effect of acidic metal salts on pH. Post-impregnation treatment can serve as an alternative, but it adds another step to the synthesis and probably has a less pronounced effect on graphitization because the catalyst is deposited on the carbon surface rather than in the structure itself.
[0030] In summary, the preparation of RF carbon is highly tailorable, and the relevant literature is abundant. However, in the prior art, the synthesis of mesoporous graphitic RF carbon relies on the use of hard templates, surfactants, or large amounts of metal salts. Furthermore, additional steps such as thorough drying, hydrothermal treatment, post-impregnation treatment, or activation complicate what is essentially a simple reaction. Therefore, the search for a simple synthetic route to graphitic mesoporous RF carbon is of great importance. If a simple and scalable preparation sequence, such as that provided in this invention, can be found, carbon will become a good candidate as a support material for electrochemical processes, especially for PEMFC catalysts.
[0031] As described herein, we synthesize graphitic RF spheres by introducing transition metal salts, such as iron salts, into the RF synthesis mixture. We demonstrate how various salts affect the pH of the mixture and the morphology of the polymer. Finally, by using iron acetylacetone as a hybrid porogen-graphitizing agent, we provide a scalable synthesis method for mesoporous and highly graphitic carbon spheres (GRFs). We do not require hydrothermal treatment and overnight drying in air, thus avoiding solvent exchange, supercritical drying, or post-activation treatments. The GRFs are comprehensively characterized by XRD, TG-MS, N2-physisorption, and TEM. We then load metal nanoparticles, such as platinum nanoparticles, onto the GRFs (Pt / GRFs) and evaluate this material as a catalyst for the oxygen reduction reaction (ORR). Pt / GRF offers comparable activity to commercially established Pt supported on carbon black, while being significantly more stable in accelerated aging tests (ADT). Based on the simple and scalable synthetic procedure and the beneficial shape of GRF, which resembles highly stable HGS, we believe that GRF is a potential candidate for industrial applications.
[0032] The invention is further illustrated by the accompanying figures and examples. [Brief explanation of the drawings]
[0033] [Figure 1] a) Simplified reaction scheme for the polycondensation of resorcinol and formaldehyde. b) Overview of different synthetic routes to RF spheres or gels. [Figure 2] Overview of the pH of the reaction mixture (pHRct-Sol) compared to the pH of a 1.0 M solution of the salts in water (pHWater). Below the table, an exemplary digital photograph of the polymer is shown. Due to the low solubility of iron acetylacetonate in water, these salts did not dissolve completely. Therefore, the measured pH values represent those of saturated solutions of these salts. [Figure 3]Comparison of N2 physisorption isotherms (left) and XRD patterns (right) of pristine RF polymers prepared with Fe(acac)2 after carbonization and after leaching. The intensities of the XRD patterns are normalized based on the graphite (002) reflection (approximately 26°) for the "after leaching" and "after carbonization" states, whereas the patterns derived from polymer analysis are plotted as absolute intensities. After graphitization, a significant increase in porosity is observed. Additionally, the (002) reflection (approximately 26°) in the XRD patterns indicates the formation of graphitic carbon. [Figure 4] TEM micrographs at various magnifications of CRF prepared with iron acetylacetonate. In contrast to the reference material prepared without iron ions (CRF-Ref, left), the formation of a porous, highly graphitic structure is observed. [Figure 5] Summary of characterization of the prepared CRFs by a) XRD, b) Raman spectroscopy, and c) N2 physisorption. In addition, important features such as external surface area (Ext. SA), micropore volume (VP,micro), and total pore volume (VP,total) (determined at 0.95 p / p) are compared in Table d). In contrast to the reference materials, high graphitization of CRF-Fe(acac)2 and CRF-Fe(acac)3 is observed in both XRD and Raman spectroscopy analyses. A large difference in porosity is also observed, with the samples prepared with iron being predominantly mesoporous, while CRF-Ref is almost exclusively microporous. [Figure 6] Overview of TG-MS measurements of the polymer RF-Fe(acac)2 in argon. Mass loss is shown at the top, and iron counts for selected fragments are shown below. In contrast to the reference material without iron ions (right), a sharp mass loss is observed at 600 °C. This suggests a pore-forming effect of the acetylacetonate ligands upon graphitization and explains the observed mesoporosity of the carbon material. [Figure 7]Characterization of GRFs prepared with Co and Ni acetylacetonates by N2 physisorption. Isotherms are shown. Similar to the material prepared with iron acetylacetonate, the porosity of the carbon is primarily in the mesopore range desired for ORR catalyst supports, but the use of Ni(acac)2 as the iron salt results in a material with less overall porosity. [Figure 8] Comparison of XRD patterns of Pt nanoparticles (NPs) deposited on GRF (left). The carbon GRF was produced via graphitization of RF-Fe(acac)2 and subsequent leaching with HCl. Pt NPs were produced by incipient wetness impregnation of the GRF with an aqueous solution of HPtCl6, followed by annealing at the indicated temperature for 2.5 hours. Additionally, a TEM micrograph and a histogram of the Pt particle size distribution of a sample annealed at 750 °C are shown. Pt NP size can be tuned by varying the annealing temperature. The NPs are deposited in the pore system of the GRF and have an average size of 2.8 ± 1.0 nm when annealed at 750 °C, which is within the optimal range for Pt-based ORR catalysts. [Figure 9]Overview of the activity (left) and stability (right) of the Pt / GRF catalyst for ORR catalysis, as determined by evaluation in a rotating disk electrode (RDE) half-cell setup. Mass activity (MA) was calculated from the current at 0.9 V (RHE) at a rotation speed of 1600 rpm. ECSA was determined by CO stripping. Specific activity (SA) was calculated from MA and ECSA. For stability verification, the catalyst was subjected to an accelerated aging protocol consisting of 20,000 square-wave cycles from 0.6 to 1.0 V (RHE) with a 1-second hold time. ECSA was determined by CO stripping after the indicated number of cycles. For comparison, a catalyst consisting of PtNPs supported on Vulcan XC72R was subjected to the same treatment. More details are described in the experimental section. The Pt / GRF catalyst exhibits high ORR activity and a large specific surface area of approximately 100 m2 g-1 Pt. The stability of Pt / GRF under potential cycling is superior to that of PtNPs supported on commercial carbon black, with a smaller ECSA loss of approximately 15% compared to a loss of >30% for Pt / V. [Figure 10] Comparison of XRD data (top left) and N2 physisorption data (bottom left) for GRF and GRFG, as well as micrographs of GRFG at various magnifications (right). GRFG was produced by increasing the reactant concentrations in the water / ethanol mixture by 25-fold, while maintaining the same resorcinol:formaldehyde:Fe(acac)2 ratio and ammonia concentration, compared to the production of GRF. Even though the spherical shape of the carbon is lost, GRFG still possesses the highly graphitized and mesoporous texture of GRF, confirming the excellent scalability of this reaction.
[0034] As illustrated in Figure 12, a simplified reaction scheme for the polycondensation of resorcinol and formaldehyde is shown along with an overview of various synthetic routes to RF spheres or RF gels.
[0035] Because RF polymerization is highly sensitive to pH, the latter was measured 15 minutes after the reactants had completely dissolved. Figure 2 compares the polymer textures for various salts, referencing the pH of the salt and the respective synthesis mixtures. Due to the low solubility of iron acetylacetonate in water, the mixtures are not completely transparent; therefore, the measured pH represents the value of a saturated solution of the salt. As expected, the effect of acidic iron salts on the pH of the reaction mixture is significant. For both FeCl3 and Fe(NO3)3, the pH shifts from approximately 11 in the control case without iron salt to 2, completely changing the behavior of the reaction. Macroscopically, this is evident in the dense polymer, which cannot be dispersed, thus precluding the successful formation of uniform spherical particles. For all other salts, the pH remains in the alkaline region, and an orange powder is obtained after filtration and drying. The significantly darker color of this powder suggests the incorporation of iron into the polymer. The reaction using iron acetylacetonate is particularly promising, as the salt has little effect on the pH of the mixture.
[0036] Furthermore, Figure 3 shows a comparison of the N physisorption isotherm (left) and XRD pattern (right) of the RF polymer prepared with Fe(acac) after carbonization and leaching in the initial state. The intensity of the XRD pattern is normalized based on the graphite (002) reflection (approximately 26°) for the "after leaching" and "after carbonization" states, while the pattern derived from the polymer analysis is plotted as absolute intensity. After graphitization, a significant increase in porosity is observed. Additionally, the (002) reflection (approximately 26°) in the XRD pattern indicates the formation of graphitic carbon.
[0037] In the next part, CRF-S-Fe(acac) xThe synthesis of CRF-Fe(acac)2 and CRF-Fe(acac)3 is examined in more detail. First, we scaled up the synthesis tenfold to a 1-liter flask, producing approximately 6 g of iron-containing polymer per batch. In addition, we significantly reduced the complexity by replacing hydrothermal treatment (24 hours at 100 °C in a Teflon-lined autoclave) with stirring at 70 °C for 24 hours, especially for larger-scale production. As shown in the TEM micrographs in Figure 4, CRF-Fe(acac)2 and CRF-Fe(acac)3 are mesoporous and highly graphitic. In fact, compared to materials produced using a solvothermal step, the synthesis at lower temperatures, but with stirring, allows for the production of carbon spheres with narrower size distributions, e.g., 367 ± 32 nm for CRF-Fe(acac)3.
[0038] Characterization using N2 physisorption, XRD and Raman spectroscopy is summarized in Figure 5.
[0039] To verify the formation of the mesoporous system in the two samples prepared with iron acetylacetonate, TG-MS measurements in argon were carried out starting from the non-porous polymer, as shown in Figure 6.
[0040] N2 physisorption characterization of GRFs prepared with Co and Ni acetylacetonates is shown, as illustrated in Figure 7. Similar isotherms are shown to the material prepared with iron acetylacetonate, and the porosity of the carbon is primarily in the mesopore range, which is desirable for ORR catalysts, although the use of Ni(acac)2 as the salt results in a material with lower overall porosity.
[0041] The optimized synthesis of CRF-Fe(acac)2 and CRF-Fe(acac)3 affords mesoporous carbon spheres in a highly graphitic system. The preparation sequence is a one-pot reaction, followed by filtration, drying in air at 80 °C, and carbonization under an inert atmosphere, followed by subsequent HCl leaching. Compared to literature, this procedure is remarkably simple, taking advantage of the use of iron acetylacetonate as a composite porogen-graphitizing agent. To demonstrate the utility of CRF-Fe(acac)2, hereafter referred to as GRF, as a support material for PEMFC catalysts, we load it with PtNPs via incipient wetness impregnation (IWI) using aqueous HPtCl6, followed by reduction and annealing. The annealing temperature was varied to determine the conditions suitable for producing platinum particles with a diameter of 3 nm. The XRD pattern and TEM micrograph of the final selected Pt / GRF annealed at 750 °C are summarized in Figure 8. Samples annealed at different temperatures have the expected reflections of Pt in the XRD patterns. As expected, the reflections tend to become sharper as the temperature increases, with the (111) reflection being prominent. Correspondingly, TEM micrographs show small particles of approximately 3 nm dispersed on the carbon.
[0042] As shown in Figure 9, Pt / GRF was evaluated as a catalyst for ORR in a rotating disk electrode setup. For details of the electrochemical characterization, see the Experimental Section. The specific activity (SA), mass activity (MA), and electrochemical surface area (ECSA) are summarized in Figure 9. The activity and ECSA of Pt / GRF are in good agreement with values expected from the literature and are essentially determined by the size of the Pt NPs. However, NPs of approximately 3 nm offer a good compromise between high mass activity and stability, since NPs smaller than 2 nm tend to dissolve more easily. The stability of Pt / GRF was verified by an accelerated aging protocol in the typical voltage range of PEMFCs, between 0.6 and 1.0 V (RHE). Compared to a catalyst consisting of platinum NPs supported on Vulcan (Pt / V), Pt / GRF retained a significantly higher proportion of its ECSA after extended cycling (Figure 9). This indicates the successful confinement of PtNPs within the mesopores of GRF, which prevents the aggregation and exfoliation of the active metal.
[0043] Figure 10 shows a comparison of the XRD data (top left) and N2 physisorption data (bottom left) of GRF and GRFG, as well as various magnifications of photomicrographs of GRFG (right). GRFG was produced by increasing the reactant concentrations in the water / ethanol mixture by 25-fold compared to the production of GRF, while maintaining the same resorcinol:formaldehyde:Fe(acac)2 ratio and ammonia concentration. Although the spherical shape of the carbon is lost, GRFG still possesses the highly graphitized and mesoporous texture of GRF, confirming the excellent scalability of this reaction. Like GRF, GRFG is highly graphitic and has nearly the same porosity. [Example]
[0044] Show me how High-resolution transmission electron microscopy (HR-TEM) images were obtained on an HF-2000 microscope equipped with a cold field emitter (CFE) and operated at a maximum accelerating voltage of 200 kV. Typically, samples were placed on a Lacey carbon film supported by a copper grid. Solid samples were deposited on the Lacey carbon film without prior dissolution.
[0045] Room temperature XRD patterns were collected using a Bragg-Brentano diffractometer (STOE THETA / THETA) with a secondary graphite monochromator (CuKα 1,2 The instrument was equipped with a 1000 W (radiation) and proportional gas detector. The divergence slit was set at 0.8°, the receiving slit was set at 0.8 mm, and the horizontal mask width was 6 mm. The sample was prepared on a background-free single-crystal quartz sample holder.
[0046] Thermal analysis (thermogravimetry, TG) coupled with mass spectrometry (MS) was performed on a Netsch STA449F3 Jupiter thermal analyzer connected to a Netsch QMS403D Aeoros mass spectrometer. Samples (approximately 10 mg) were heated from 45 to 1000 °C (heating rate, 10 °C / min) in a continuous flow of Ar (40 mL / min).
[0047] Nitrogen adsorption isotherms were recorded using a Micromeritics 3Flex instrument. The apparent specific surface area was determined using the Brunauer-Emmett-Teller (BET) algorithm. The t-plot method was used to determine the pore volume and external surface area. Carbon black STSA was used as the thickness curve. The total pore volume was determined using the Güllwitsch rule at 0.95 p / p0.
[0048] Inductively coupled plasma optical emission spectroscopy (ICP-OES) analysis was performed with a Spectrogreen FMX46 using UVPlus Optic ORCA (optimized Rowland circle alignment) optics. To prepare the liquid sample for measurement, approximately 10 mg of catalyst was leached in a mixture of 7.5 mL of HCl, 2.5 mL of HNO3, and 0.8 mL of H2O2. The suspension was heated at 180 °C for 5 minutes in an Anton Paar Multiwave 5000 microwave oven, after which it was allowed to cool to ambient temperature. The suspension was then diluted to 50 mL with mQ water and stirred overnight before analysis and metal concentration evaluation. Sample weights were measured throughout the procedure using a Mettler Toledo XA205 Dual Range and a Sartorius Entris II.
[0049] Electrochemical measurements The specific activity (SA) and mass activity (MA) for ORR, as well as the electrochemical surface area (ECSA), were evaluated by electrochemical half-cell measurements in a three-electrode compartment Teflon cell using a rotating disk electrode (RDE) setup. A graphite rod was used as the counter electrode, and a double-junction Ag / AgCl (3 M KCl, Metrohm) electrode was used as the reference electrode. The reference electrode's cell compartment was separated using a Nafion membrane to avoid chloride ion leaching during activity and stability measurements. All potentials are measured against the reversible hydrogen electrode (RHE). To ensure accurate correction of the Ag / AgCl electrode potential, the potential was previously measured with a Pt electrode in H2-saturated electrolyte. This procedure was repeated for all activity and stability measurements using the tested Pt-based ORR catalyst as the RHE.
[0050] The catalyst film was prepared by first dispersing the catalyst powder in ultrapure water (VWR) using an ultrasonic vial processor. Then, after 30 minutes of initial treatment, a 20 μL ink droplet was applied to a freshly polished glassy carbon disk electrode (0.196 cm) with a diameter of 5 mm. 2The catalyst was pipetted onto a surface of 1000 .mu.m (geometric surface area of 1000 .mu.m), embedded in a Teflon shell, and finally dried in air. The amount of catalyst on the electrode was 10 μg cm.sup.2 as a platinum loading relative to the carbon support and dispersion. -2 For both activity and stability measurements, 0.1 M HClO (prepared by diluting 70% HClO (Ultrapure, Carl Roth) with ultrapure water) was used as the electrolyte. The ohmic drop was corrected via positive feedback to have a residual uncompensated resistance of less than 3 Ω.
[0051] Before each activity and stability measurement, the catalyst was subjected to 200 to 300 cleaning cycles in N2 saturated electrolyte until a stable cyclic voltammogram (CV) was recorded. The ORR activity was measured at 50 mVs. -1 The specific activity (SA) was calculated from the anodic scan of the CV recorded in O2-saturated electrolyte using a scan rate of 0.9 V / RHE and a rotation speed of 1600 rpm according to Couterie-Levich analysis. The CVs were corrected for capacitive processes by subtracting a CV recorded in nitrogen within the same potential window to isolate the current for oxygen reduction. The specific activity (SA) is expressed as the kinetic current density normalized to the actual surface area, derived from the electrochemical CO oxidation charge. For this, the peak of the CO stripping CV was integrated after subtracting the CV recorded in N2-saturated electrolyte to correct for non-Faraday contributions. The electrochemical surface area (ECSA) was calculated as 380 μC cm -2 Pt The mass activity (MA) was calculated assuming a surface charge density of 0.4 V and 1.0 V / RHE. The mass activity (MA) was derived from the SA and ECSA. Stability measurements consisted of an accelerated aging protocol based on voltage cycling of 10,800 CV at 1 V / s between 0.4 V and 1.0 V / RHE. The aging tests were performed in N2-saturated electrolyte without rotation, and the ECSA was determined via CO stripping.
[0052] Synthesis of RF carbon spheres The synthesis of RF carbon spheres is based on a Stöber-type process aimed at uniform RF spheres. Typically, 40 mL of water, 16 mL of ethanol, and 0.20 mL of ammonia solution (25 wt % in water) were mixed for 5 minutes. Then, 0.40 g of resorcinol and an amount of iron salt to give a 1:3 molar ratio of iron to resorcinol were dissolved in the mixture with stirring. The pH of the reaction mixture was measured at this stage. Then, 0.56 mL of formaldehyde (37 wt % in water with 10–15% methanol) was added to the solution, which was further stirred at 30°C for 24 hours. After this step, for the sample designated CRF-S-SALT, the reaction mixture was transferred to an autoclave equipped with a Teflon inlet and subjected to solvothermal treatment at 100°C for 24 hours without stirring. For the sample designated CRF-SALT, instead of solvothermal treatment, the reaction mixture was stirred at 70 °C for 24 h in a round-bottom flask equipped with a reflux condenser. After both procedures, the solid was filtered, washed with water and ethanol, and dried overnight in air at 80 °C. Unless otherwise noted, carbonization was carried out in a tube furnace at 1000 °C under a 100 mL / min argon flow for 4 h. After cooling to room temperature, the iron residues in the black solid were leached twice in semi-concentrated HCl at 60 °C overnight. Finally, the powder was washed with copious amounts of water and ethanol and dried in air at 80 °C. Using the above reactant amounts, a yield of approximately 600 mg of polymer and a final yield of roughly 200 mg of carbon were obtained. For the preparation of the CRF-SALT sample, the reaction was scaled up 10-fold.
[0053] Exemplary synthesis of RF carbon gel To prepare the RF gel, 13.75 mL of water, 5.52 mL of ethanol, and 0.063 mL of ammonia solution (25 wt % in water) were mixed in a 50 mL PE bottle for 5 minutes. Then, 3.00 g of resorcinol and 2.29 g of Fe(acac)2, giving a 1:3 iron:resorcinol molar ratio, were dissolved under stirring. After complete dissolution, 4.06 mL of formaldehyde solution (37 wt % in water containing 10–15% methanol) was added. The sealed bottle was then stirred at 50°C for 2 hours and then at 70°C for 24 hours. The mixture solidified after approximately 6 hours, and stirring was discontinued. The bottles were then opened, and the orange gel was dried overnight in air at 80°C. After drying, the polymer monolith was crushed and ground to an orange powder, which was then carbonized and leached according to the procedure described above for the RF carbon spheres.
[0054] Preparation of Pt / GRF GRFs were loaded with PtNPs according to our recent publication on the synthesis of Pt / HGS. 100 mg of GRFs was dried overnight under vacuum. The powder was then impregnated with a solution of HPtCl in water, in an amount equal to 90% of the total pore volume of the carbon. Impregnation was performed in three equal-volume steps under vigorous stirring and shaking. To ensure complete pore filling, the powder was sonicated for 30 min. The powder was dried, reduced, and annealed in a tube furnace. The sample was dried at 120 °C (3 K / min) under flowing Ar (200 mL / min) for 60 min and then heated at 220 °C (2 K / min) for 90 min in a mixture of H and Ar (40 / 160 mL / min). Finally, the powder was annealed at various temperatures for 2.5 h (5 K / min) under an Ar atmosphere (100 mL / min).
[0055] General As explained earlier, CRF-Fe(acac) x The carbon materials produced using the methods of the present invention, including Fe(acac), provide spherical, mesoporous, and highly graphitic carbons and are therefore candidates for further research and ultimately for use as catalyst supports for electrochemical devices. xWe believe that the metal salts, such as iron acetylacetonate, are incorporated into the carbon structure due to the high affinity of the acetylacetonate anion for the metal cation and behave as non-dissociating ligands in solution. This behavior is evident from the pH effect and the low water solubility of the salts described above. Since both iron acetylacetonate and the polymer are expected to have similar hydrophilic properties, the latter may also facilitate incorporation into the polymer.
[0056] The optimized synthesis of CRF transition metal salts, such as Fe(acac)2, produces highly graphitic mesoporous carbon spheres. The preparation sequence is a one-pot reaction, followed by filtration, drying in air at 80 °C, and carbonization under an inert atmosphere, followed by subsequent HCl leaching. Compared to literature, this procedure is remarkably simple, taking advantage of the use of iron acetylacetonate as a hybrid porogen-graphitizing agent. To demonstrate the utility of CRF-Fe(acac)2, hereafter referred to as GRF, as a support for PEMFC catalysts, we loaded it with PtNPs via incipient wetness impregnation (IWI) using aqueous HPtCl6, followed by reduction and annealing. The annealing temperature was varied to determine conditions suitable for producing platinum particles with a diameter of 3 nm. The resulting catalyst, Pt / GRF, was evaluated for ORR and showed high activity and stability, resulting directly from the beneficial structure of the carbon support.
[0057] While the synthesis of GRF is simplified compared to conventional literature, the high control over shape, i.e., the production of uniform carbon spheres, comes at the expense of relatively high dilution of the synthesis mixture. Using the current approach, approximately 2 g of carbon can be produced per liter. For industrial purposes, increasing the volumetric yield of production is desirable. Therefore, we increase the resorcinol and formaldehyde concentrations by 25-fold compared to the synthesis of GRF, while maintaining the NH3 concentration, ethanol / water ratio, and Fe / resorcinol ratio. Polycondensation is carried out at 70°C for 24 hours with stirring, but stirring is stopped after 6-8 hours when the gel solidifies, producing a polymer gel, which is then dried in air at 80°C. After carbonization and leaching, we obtain a highly graphitic carbon gel (GRFG) with physical characteristics very similar to spherical GRF. Similar to GRF, GRFG is highly graphitic and has almost the same porosity. However, iron particles trapped in graphite shells are present throughout the structure: ICP-OES measurements performed after disintegrating the carbon by heating in a mixture of aqua regia and H2O2 under elevated pressure (see Experimental Section for details) confirmed an iron content of less than 1 wt% in both GRF and GRFG.
[0058] Considering similar properties in terms of graphitization and porosity, GRFG offers a significantly more promising catalyst support than GRF. The improvement in volumetric yield to approximately 70 g of carbon per liter removes the last obstacle preventing the successful use of GRFG as a PEMFC support with potential for industrial applications.
[0059] As previously shown, we developed a synthetic route to graphitic mesoporous carbon starting from the established resorcinol-formaldehyde chemistry. By using iron salts, specifically acetylacetonate, as the graphitization catalyst and porogen, we produced uniform carbon spheres that exceed most carbon blacks in terms of graphitic carbon content. Furthermore, the carbon possesses intraparticle mesoporosity, a property rarely found in RF materials. By exploiting its advantageous pore structure, we loaded PtNPs onto the carbon, thereby producing a highly active electrocatalyst for the oxygen reduction reaction (ORR). Most importantly, this has enhanced stability compared to industrial catalysts. A clear advantage of the procedure is its simplicity. Additionally, because no extensive drying, templating, or activation is required, the method appears promising for scaling up to even larger product amounts. In fact, the inventors have shown that the method can be significantly optimized when the reaction concentration is increased by 25-fold and still obtains graphitic mesoporous carbon with a commercially viable volumetric yield of 70 g per liter of synthesis mixture.