N-Doped Porous Carbon from Sargassum spp . as Efficient Metal-Free Electrocatalysts for O 2 Reduction in Alkaline Fuel Cells

This work reports the synthesis of N-doped porous carbon (NPC) with a high surface area from Sargassum spp. as a low-cost alternative for electrocatalyst production for the oxygen reduction reaction (ORR). Sargassum spp. was activated with potassium hydroxide at different temperatures (700, 750, and 800 ◦C) and then doped with pyridine (N700, N750, and N800). As a result of the activation process, the 800 ◦C sample showed a high surface area (2765 m2 g−1) and good onset potential (0.870 V) and current density (4.87 mA cm−2). The ORR performance of the electrocatalysts in terms of their current density was N800 > N750 > N700 > 750 > 800 > 700, while the onset potential decreased in the following order: N800 > 800 > 750 > 700 > N700 > N750. The fuel cell performance of the membrane electrode assembly (MEA) prepared with electrocatalyst synthesized at 750 ◦C and doped with pyridine was 12.72 mW cm−2, which was close to that from Pt/C MEA on both the anode and cathode (14.42 mW cm−2). These results indicate that NPCs are an alternative to the problem of Sargassum spp. accumulation in the Caribbean due to their high efficiency as electrocatalysts for ORR.


Introduction
Fuel cells are considered an environment-benign technology providing solutions to a range of environmental challenges, such as harmful levels of local pollutants, in addition to providing economic benefits due to their high efficiency [1], the main challenges for fuel cells are to demonstrate reliability and reduce capital cost [2].Fuel cells (FCs) use a continuous stream of gaseous and/or liquid fuels and oxidizers to electrochemically generate electrical output power [3].Alkaline fuel cells (AFCs) are excellent, efficient performers with excellent reliability and length of life [4]; their overall reaction is 2H 2 + O 2 → 2H 2 O + electric energy + heat.In an AFC anode, the hydrogen is catalytically oxidized, 2H 2 + 4OH − → 4H 2 O + 4e − , while, at the cathode, the oxygen is reduced, O 2 + 2H 2 O + 4e − → 4OH − [5].The oxygen reduction reaction (ORR) is the primary target for overpotential reduction in AFCs [6,7].No single polycrystalline metal catalyzes the ORR as effectively as Pt [8], but its high cost presents a barrier for AFC commercialization.Due to their earth abundance and almost zero cost, the use of biomass-derived materials may be a promising alternative for the development of high-performance and multifunctional electrocatalysts on a large scale to promote AFC commercialization.
In previous years, vegetal and animal origin biomasses such as moss [9], lotus stem [10], oil seed [11], luffa sponge fibers [12], amaranthus [13], pomelo peel [14], brewer's yeast [15], okara [16], fish [17], chicken feather [18], porcine blood [19,20], olive oil [21] and waste leather scraps [22,23] have been reported as carbon and heteroatom precursors for the synthesis of electrocatalysts, as they have been found to enhance ORR performance.Additionally, through chemical treatments, such as activation and doping, physicochemical and electrochemical properties can be tuned.Specifically, KOH activation agents that promote microporosity [24] have been used to synthesize electrocatalysts towards ORR [12,22,23].On the other hand, doping with heteroatoms is one of the most effective methods to tailor the electronic properties of carbon nanomaterials [25], when doping the carbonaceous framework with heteroatoms, electronegativities and atom sizes could lead to alteration of the electron densities, resulting in the formation of active centers for ORR and contributing to an improvement in the electrochemical stability [26].Among heteroatoms, nitrogen has been the most studied (through doping agents such as melamine [27], hidrazine [28], dopamine [10]) because its atomic radius is similar to that of C with one more electron in its valence shell, so n-type electronic modification is created within the carbon lattice, leading to enhanced electron-donor interactions or the basic properties of N-doped carbon materials [29].There are many nitrogen species associated with ORR performance improvement.First, pyridinic species (N1), which refers to N atoms at the edges of graphene planes, where each N is bonded to two carbon atoms and donates one p-electron to the aromatic π system, then pyrrolic ones (N2) where N atoms are integrated into five membered heterocyclic rings, which are bonded to two carbon atoms and contribute two p-electrons to the π system, finally quaternary (N3) N atoms are those substituting carbon atoms within the graphene layer [8].
Sargassum spp. is a brown alga (class Cyclosporeae; order Fucales; family Fucaceae; genus Sargassum) whose proximate composition includes 14.33% moisture, 6.55% protein, 1.90% lipid, 18.50% ash, 58.72% carbohydrate, and 17.00% fiber [30].The Sargasso Sea is one of the main distribution areas of the so-called gulf-weed; here, it forms a holopelagic population and maintains itself by vegetative growth only [31].During June 2011, pelagic Sargassum began washing ashore along the Caribbean, Gulf of Mexico, West African, and Brazilian coastlines in unprecedented quantities [32].Studies inferred that dissolved organic nutrients from this weed may have been taken up and contributed to the development and persistence of blooms during the macroalgal bloom period in 2012 [33].However, green tides along the coasts of the Yellow Sea also reached their record by 2013, when a major event reportedly occurred [34].Later, the total biomass dry weight collected from Atalaia beach was 53.9 tons for the 2014 events and 301.26 tons for 2015.This confirmed the intensity of the algal bloom along the Northern Brazilian coast in 2015, where around 3243 tons of Sargassum spp.wet weight were stranded [35].Later, in 2016, after a severe algal bloom, Chilean aquaculture experienced extreme mortalities of 40 million farmed salmon, which caused a financial loss of US $800 million [36].This year, Sargassum has already arrived at the Mexican Caribbean coasts; however, a new, unexplored opportunity has arisen for this sustainable and Earth-abundant source to be used as an opportunity for fuel cell commercialization, rather than simply being an economical threat.
In 2016, Liu et al. extracted sodium alginate from seaweeds, mixed it with graphene and FeCl 3 solution and calcinated it at 700 • C in NH 3 to prepare an aerogel with physicochemical properties of I D /I G = 0.95 and 465 m 2 g −1 surface area as well as an ORR performance by the four electron-transfer pathways of −4.5 mA cm −2 [37].Further, our metal-free electrocatalyst approach appeared in the study performed by Song et al., where pyrolyzed seaweeds (Undaria pinnatifida) were exposed to a N 2 flow for 5 h at 1000 • C, followed by 3 M HCl acid treatment for 24 h.When the nitrogen content was 1.8 wt % in the bulk, this process resulted in a surface area of 1218 m 2 g −1 ; however, it is noteworthy that quantities of 27.0%, 21.0% and 43.5% of N1, N2 and N3 were found, respectively.Additionally, the ORR performance was 0.972 V vs. Ag/AgCl (onset potential) and 4.5 mA cm −2 (current density) at Energies 2019, 12, 346 3 of 15 1600 rpm in KOH [38].In this work, we proposed a feasible synthesis route using Sargassum spp. as a sustainable and Earth-abundant carbon source for the synthesis of NPC as metal-free electrocatalysts for alkaline fuel cell applications.

Metal-Free Electrocatalyst Synthesis
Electrocatalysts were prepared according to the method reported previously by Perez-Salcedo K. Y. et al. [39], but pyrolysis was performed in an argon atmosphere at three different temperatures, meaning that the activated samples were denoted as 700, 750, and 800.After activation, doping treatment was performed with pyridine (C 5 H 5 N, Sigma-Aldrich, 99.8%, Saint Louis, MO, USA) as a nitrogen source.N-doped samples were labeled as N700, N750, and N800.The synthesis process and nomenclature is summarized in Table 1.

Physical Characterization
Morphological characterization was carried out by scanning electron microscopy (SEM) in a JSM-6360LV, JEOL, Mundelein, IL, USA, equipment.Samples were dried and coated with a gold film in a sputter coater instrument.Additionally, electrocatalysts were analyzed by energy dispersive spectroscopy (EDS).The Brunauer-Emmett-Teller (BET) surface area was characterized in BET equipment (Bel MicrotracBel, Belsorp-max, Osaka, Japan) by degassing the samples at 200 • C for 6 h.The elemental analysis was carried out using an Organic Elemental Analyzer (Thermo Scientific, 2000, Milan, Italy); the measurements were repeated in duplicate with the mean value reported.Structural properties were characterized by Raman spectroscopy in an equipment (DXR xi, Thermo Scientific, Madison, WI, USA) with a He-Ne gas laser (λ = 633 nm).X-ray diffraction (XRD) patterns were obtained with a Bruker D2 Phaser apparatus (D2 Phaser, Bruker, Karlsruhe, Germany) with a step size of 0.0101 2θ degrees and 4.75 step per s, in a range from 9.99 to 99.99 2θ degrees.Finally, surface elemental analysis was conducted by X-ray photoelectron spectroscopy (XPS) using a spectrophometer (K-Alpha, Thermo Scientific, East Grinstead, UK) equipped with an Al-Kα X-ray source at 12 kV; spectra calibration was done using C1s at 284.5 eV as a reference.

Electrochemical Measurements
Electrochemical experiments were carried out in a conventional three-electrode test cell at room temperature using a PAR Bistat Potentiostat (Princeton Applied Research, Oak Ridge, TN, USA) connected to a rotating disk electrode set-up.A mixture of 15 µL Nafion (5 wt %) solution, 500 µL 2-propanol, and 10 mg electrocatalysts was sonicated in an ultrasonic bath for 40 min, and commercial 10% platinum was also prepared for comparison purposes.Thirty microliters of catalytic ink dispersion was dropped onto the polished glassy carbon substrate (5 mm diameter).Cyclic voltammetry (CV) was performed from 1.2 V to 0.0 V vs. the reversible hydrogen electrode (RHE) potential window at a 20 mV s −1 scan rate for 3 cycles in N 2 -saturated 0.5 M KOH electrolyte; an activation treatment was performed at a 50 mV s −1 scan rate for 40 cycles before all measurements.ORR performance was studied using the rotary disk electrode (RDE) technique in O 2 saturated 0.5 M KOH, and all background current was corrected by subtracting the data obtained in N 2 -saturated electrolyte under the same measurement conditions.Linear sweep voltammetry (LSV) experiments at 200, 400, 800, 1200, 1600, and 2000 rpm rotation rates were conducted from 1.2 V to 0.0 V vs. the RHE potential window at a 5 mV s −1 scan rate using a coiled platinum wire as the counter electrode and Ag/AgCl (KCl saturated) as the reference electrode.

Fuel Cell Studies
Membrane electrode assembly (MEA) performance was evaluated using a Greenlight Test Station (G40 Fuel cell system, Hydrogenics, Vancouver, BC, USA).The MEA, gas diffusion layer (GDL), and gasket were assembled in the single cell, with each bolt under uniform torque (4.5 N m).Single cells were tested at 30 • C with humidified hydrogen and oxygen at the anode and cathode, respectively, at a 0.5 L min −1 flow rate, under atmospheric pressure, and in 100% relative humidity.The GDL and MEA were fabricated as described below.
For GDL, a 15 × 15 cm carbon paper substrate (0.40 mm thickness non-woven, GD07508G, Hollingsworth & Vose Company, Walpole, MA, USA) was coated in equipment (Easycoater equipment, EC26) with a homogeneous suspension made of 843.75 mg pure carbon black (nano-chain Pureblack carbon, grade 205-110, Superior Graphite Co.), 281.25 mg vapor-grown carbon fiber (VGCF, Showa Denka, Toyama, Japan) and 270 mg sodium dodecyl sulfate (SDS, Fisher Scientific, Pittsburgh, PA, USA) dispersed in 13.5 mL of water and mixed by sonicating for 30 min and stirring for 1 h at 60 rpm.Then, 30 wt % of polytetrafluoroethylene (PTFE, TE-3859 from DuPont Fluoroproducts, Wilmington, DE, USA) was added to the mixture.Bubbles were avoided during the coating process.The coated substrate was dried at room temperature overnight and sintered at 350 • C for 30 min.Then, it was washed with distilled water for 30 min.
MEA was obtained by coating both sides of the anion exchange membrane (FAA-3-50, Fumatech BWT, Bietigheim-Bissingen, Germany), and this was kept in 1 M KOH solution overnight to replace Br − with OH − .To prepare the anode catalyst layer, a catalyst ink of 50 mg 46 wt % Pt/C (TKK, Tanaka Kikinzoku Group, Tokyo, Japan), 0.2 mL water, 1.2 MeOH, 0.8 mL tetrahydrofuran (THF), and 0.1 mL FAA-3 was made.The mixture was sonicated in an ultrasonic water bath for 10 min and stirred for 10 s (repeated three times).The ink was spray-coated onto the 5 cm 2 active area membrane and dried at room temperature.For the cathode catalyst layer, ink was prepared by mixing 50 mg electrocatalyst powder in 1.2 MeOH, 0.8 mL THF, and 0.1 mL FAA-3.The mixture was treated ultrasonically for 10 min and stirred for 10 s (repeated three times), and then the membrane was coated.Commercial Pt/C was also spray-coated to allow a fuel cell performance comparison.

Physicochemical Characterization
Elemental analyses of C, H, N, and S weight percentages were performed to obtain a better understanding of the electrocatalysts' compositions.The results including the standard deviations from the triple measurements for each electrocatalyst are summarized in Table 2.It can be observed that after activation treatment at 700, 750, and 800 • C, nitrogen disappeared.This was because macroalgae have a low nitrogen ratio, which can result in nitrogen loss during pyrolysis treatment as NH 3 , N 2 O, or even HCN, as reported by composting algae studies [40].After C 5 H 5 N doping, a low amount of nitrogen was incorporated to the material, which, according to previous studies, can be tuned by manipulating the solvothermal treatment temperature.On the other hand, seaweeds contain sodium alginate [41] and high amounts of calcium carbonate (CaCO 3 ) [39], therefore, the samples pyrolyzed at the highest temperature (800 and N800) showed the highest post-pyrolysis carbon amount (68.86 and 80.95 wt %, respectively) among all samples.Finally, the sulfur content was low for the 700 and Energies 2019, 12, 346 5 of 15 750 samples, lower than the detection limit for 800, and dropped to zero after doping treatment for all three sample types.It is noteworthy that these carbon and sulfur contents are in agreement with the amounts found from the EDS results discussed below.The electrocatalysts' N 2 adsorption-desorption isotherms are presented in Figure 1a.They exhibit a type I shape associated with microporous solids.In particular, N750 shows type H3 hysteresis, which is typical in solids consisting of aggregates or particle agglomerates forming slit-shaped pores with a nonuniform size and/or shape [42].The MP plot shown in Figure 1b confirms the electrocatalysts' microporosity with the majority of micropores being between 0.5 and 1.0 nm.The isotherms' textural properties are summarized in Table 3.The synthesized electrocatalysts had high BET surface areas, which suggests a greater availability of active sites for the ORR to occur.As expected, the BET surface area increased with each heat treatment; N800 showed the highest surface area (2765 m 2 g −1 ) because the acid treatment removed any alkaline or inorganic residue when cleaning its structure.To the best of our knowledge, this is the highest surface area reported for electrocatalysts from biomass wastes used for ORR performance studies.The electrocatalysts' N2 adsorption-desorption isotherms are presented in Figure 1a.They exhibit a type I shape associated with microporous solids.In particular, N750 shows type H3 hysteresis, which is typical in solids consisting of aggregates or particle agglomerates forming slit-shaped pores with a nonuniform size and/or shape [42].The MP plot shown in Figure 1b confirms the electrocatalysts' microporosity with the majority of micropores being between 0.5 and 1.0 nm.The isotherms' textural properties are summarized in Table 3.The synthesized electrocatalysts had high BET surface areas, which suggests a greater availability of active sites for the ORR to occur.As expected, the BET surface area increased with each heat treatment; N800 showed the highest surface area (2765 m 2 g −1 ) because the acid treatment removed any alkaline or inorganic residue when cleaning its structure.To the best of our knowledge, this is the highest surface area reported for electrocatalysts from biomass wastes used for ORR performance studies.The electrocatalysts mesoporous structures are given in Figure 2a-f.In accordance with the surface area results, the micrographs show similar highly porous structures.It can be seen that after doping treatment, such pore structures remained, but Figure 2e,f clearly display some agglomerates and fragments that could have been formed due to the doping treatment.EDS spectra confirmed the presence of inorganic compounds (2.29 wt % calcium, 1.29 wt % chloride, 0.96 wt % sodium, 0.86 wt % potassium and 0.66 wt % magnesium) in the raw Sargassum spp.which were completely removed after activation and doping treatment.Moreover, the carbon content increased after chemical treatments from 53.67 wt % in raw seaweed to 83.85, 85.42, 85.89, 83.20, 87.29, and 88.12 wt % for 700, 750, 800, Energies 2019, 12, 346 6 of 15 N700, N750, and N800, respectively.The sulfur content decreased after doping treatment, confirming the CHNS elemental analysis results.On the other hand, 0.06, 0.03, 0.04, 0.13, 0.17, 0.16 wt % chloride was found in 700, 750, 800, N700, N750, and N800, respectively.The electrocatalysts mesoporous structures are given in Figure 2a-f.In accordance with the surface area results, the micrographs show similar highly porous structures.It can be seen that after doping treatment, such pore structures remained, but Figures 2e,f clearly display some agglomerates and fragments that could have been formed due to the doping treatment.EDS spectra confirmed the presence of inorganic compounds (2.29 wt % calcium, 1.29 wt % chloride, 0.96 wt % sodium, 0.86 wt % potassium and 0.66 wt % magnesium) in the raw Sargassum spp.which were completely removed after activation and doping treatment.Moreover, the carbon content increased after chemical treatments from 53.67 wt % in raw seaweed to 83.85, 85.42, 85.89, 83.20, 87.29, and 88.12 wt % for 700, 750, 800, N700, N750, and N800, respectively.The sulfur content decreased after doping treatment, confirming the CHNS elemental analysis results.On the other hand, 0.06, 0.03, 0.04, 0.13, 0.17, 0.16 wt % chloride was found in 700, 750, 800, N700, N750, and N800, respectively.Furthermore, the Raman spectra were studied to understand the structure and quality of the carbon materials.D-and G-band existence was confirmed for the 700, 750, 800, N700, N750, and N800 samples.As presented in Figure 3a, the D-band (1354 cm −1 ) corresponds to a defective sp 3 -phase and the G-band (1588 cm −1 ) to an ordered sp 2 -phase [26].The D-band was present more often than the G-band for all electrocatalysts and structural properties, and defective sites were determined through the D-band and G-band relationship (ID/IG), providing information regarding the surface defects.In particular, sample 700 presented a lower index (ID/IG = 1.35 ± 0.03), indicating that it had fewer sp 3 carbon defects and more sp 2 carbon.Sample 750 showed a slightly superior index (ID/IG = 1.44 ± 0.03), suggesting that the number of defects increased as the chemical treatments were applied.Finally, sample 800 was the most disordered sample (ID/IG = 1.47 ± 0.03), providing evidence that the disorder degree was affected by the temperature.Likewise, N-doped electrocatalysts showed an enhancement in D-band intensity, leading to higher ID/IG indices of 1.40 ± 0.01, 1.45 ± 0.01, and 1.44 ± 0.02 for N700, N750, and N800, respectively.Finally, the amorphous nature of the electrocatalysts was confirmed in Figure 3b by the amorphous carbon diffraction patterns.Furthermore, the Raman spectra were studied to understand the structure and quality of the carbon materials.D-and G-band existence was confirmed for the 700, 750, 800, N700, N750, and N800 samples.As presented in Figure 3a, the D-band (1354 cm −1 ) corresponds to a defective sp 3 -phase and the G-band (1588 cm −1 ) to an ordered sp 2 -phase [26].The D-band was present more often than the G-band for all electrocatalysts and structural properties, and defective sites were determined through the D-band and G-band relationship (I D /I G ), providing information regarding the surface defects.In particular, sample 700 presented a lower index (I D /I G = 1.35 ± 0.03), indicating that it had fewer sp 3 carbon defects and more sp 2 carbon.Sample 750 showed a slightly superior index (I D /I G = 1.44 ± 0.03), suggesting that the number of defects increased as the chemical treatments were applied.Finally, sample 800 was the most disordered sample (I D /I G = 1.47 ± 0.03), providing evidence that the disorder degree was affected by the temperature.Likewise, N-doped electrocatalysts showed an enhancement in D-band intensity, leading to higher I D /I G indices of 1.40 ± 0.01, 1.45 ± 0.01, and 1.44 ± 0.02 for N700, N750, and N800, respectively.Finally, the amorphous nature of the electrocatalysts was confirmed in Figure 3b by the amorphous carbon diffraction patterns.XPS studies were carried out to investigate the surface composition and chemical bonding of the electrocatalysts.The atomic concentrations of carbon, oxygen, nitrogen and sulphur were obtained from the survey and are summarized in Table 4. High-resolution spectra were obtained for C1s, N1s, and O1s and are shown in Figures 4 and 5.The fitting was done at 284.5 eV using C1s as a reference.Carbon was the most abundant element with 90.16, 89.63, 87.32, 89.08, 90.94, and 93.06 wt % for 700, 750, 800, N700, N750, and N800, respectively.C1s peaks located at 284.5 ± 0.2 eV corresponded to C=C/C-C [43], while peaks at 285.9 ± 0.1 and 287.2 ± 0.1 eV were assigned to amorphous carbon and sp 2 carbon atoms bonded to oxygen, respectively [44].The peaks located at 288.6 and 290.0 eV were assigned to C-O bonds (O-C-O/C=O and O-C=O, respectively), and the peak at 291.5 ± 0.1 was related to the π-π* shake-up satellite, which is common in high-temperature treated carbons [44].High-resolution spectra for N1s showed peaks which were assigned as follows: N1 from 398.1 to 399.3 eV; N2 from 399.8 to 401.2 eV, and N3 from 401.1 to 402.7 eV [23].Pyridine treatment eliminated the sulphur from doped samples and changed the nitrogen-bond type.Originally, all samples presented N1, but, after C5H5N treatment, the nitrogen shifted to N2 and N3, while the N1 type disappeared.XPS studies were carried out to investigate the surface composition and chemical bonding of the electrocatalysts.The atomic concentrations of carbon, oxygen, nitrogen and sulphur were obtained from the survey and are summarized in Table 4. High-resolution spectra were obtained for C1s, N1s, and O1s and are shown in Figures 4 and 5.The fitting was done at 284.5 eV using C1s as a reference.Carbon was the most abundant element with 90.16, 89.63, 87.32, 89.08, 90.94, and 93.06 wt % for 700, 750, 800, N700, N750, and N800, respectively.C1s peaks located at 284.5 ± 0.2 eV corresponded to C=C/C-C [43], while peaks at 285.9 ± 0.1 and 287.2 ± 0.1 eV were assigned to amorphous carbon and sp 2 carbon atoms bonded to oxygen, respectively [44].The peaks located at 288.6 and 290.0 eV were assigned to C-O bonds (O-C-O/C=O and O-C=O, respectively), and the peak at 291.5 ± 0.1 was related to the π-π* shake-up satellite, which is common in high-temperature treated carbons [44].High-resolution spectra for N1s showed peaks which were assigned as follows: N1 from 398.1 to 399.3 eV; N2 from 399.8 to 401.2 eV, and N3 from 401.1 to 402.7 eV [23].Pyridine treatment eliminated the sulphur from doped samples and changed the nitrogen-bond type.Originally, all samples presented N1, but, after C 5 H 5 N treatment, the nitrogen shifted to N2 and N3, while the N1 type disappeared.

Electrochemical Characterization
The electrocatalytic activity of the as-prepared NPC was studied using the rotating disk electrode (RDE) technique with linear sweep voltammetry (LSV) in a three-electrode system.Figure 6a shows the ORR polarization curves catalyzed by 700, 750, 800, N700, N750, N800, and 10% commercial Pt/C in O2-saturated alkaline electrolyte (0.5 M KOH) at a 5 mV s −1 scan rate and 2000

Electrochemical Characterization
The electrocatalytic activity of the as-prepared NPC was studied using the rotating disk electrode (RDE) technique with linear sweep voltammetry (LSV) in a three-electrode system.Figure 6a shows the ORR polarization curves catalyzed by 700, 750, 800, N700, N750, N800, and 10% commercial Pt/C in O 2 -saturated alkaline electrolyte (0.5 M KOH) at a 5 mV s −1 scan rate and 2000 rpm.All measurements were performed at room temperature under the same conditions and catalyst load, and the background currents were subtracted.Current density values were determined at 0.0 V and the onset potential, as reported in literature [45]; these are summarized in Table 5.In the LSV data, it is noteworthy that the current density increased in parallel with the rotation rate, but none of the samples reached the plateau of the diffusion limiting current spectroscopy, while its onset potential remained constant (0.828 V).In the same way, doping of the 750 sample improved its current density to 4.32 mA cm −2 , but its N750 onset potential shifted to a more negative value (0.811 V).Finally, at the highest temperature, more positive onset potentials were shown by the 800 and N800 samples.In particular, the N800 sample showed a well-defined curve and the best ORR activity, possibly as a result of having the highest surface area among all samples.
In general, the ORR performance of the electrocatalysts was attributed to their structures, namely, the increase in active sites, which facilitate oxygen diffusion and electron transport.The results of the electrocatalyst cyclic voltammetry (CV) assessment in N2-saturated alkaline electrolyte (0.5 M KOH) at a 20 mV s −1 scan rate under room temperature conditions are presented in Figure 6b.All samples exhibited a pseudo-rectangular shape, which is associated with good charge propagation throughout the porous carbon framework, as well as good electronic conductivity and an accessible porous structure [46].
The K-L plots shown in Figure 7 were obtained from RDE polarization curve data recorded at various rotation rates (200, 400, 800, 1200, 1600, 2000 rpm) applying  =  +   / , where j is the absolute measured current density (mA cm −2 ), jk is the kinetic current, ω is the electrode rotation rate (rpm), and B is determined from the K-L graph slopes.The ORR polarization plots show that the pyrolysis temperature played an important role in the ORR activity.Both undoped and N-doped electrocatalysts exhibited good performance for the ORR.As expected, slower ORR kinetics were shown at the lowest temperature for the 700 sample, whose cathode reduction peak appeared at 0.827 V leading to a maximum current of 2.93 mA cm −2 .In addition, the doping of sample 700 improved its performance, as revealed by the N700 current density value (4.15 mA cm −2 ); however, it exhibited a more negative potential (0.812 V), which could be associated with physicochemical properties, such as its low nitrogen content.
In addition, when the catalysts were synthesized at 750 • C, the current density improved as compared with samples at 700 • C. Specifically, it increased by 33% (3.89 mA cm −2 ), probably due to its larger surface area and nitrogen content, and increase in disorder, as studied by Raman spectroscopy, while its onset potential remained constant (0.828 V).In the same way, doping of the 750 sample improved its current density to 4.32 mA cm −2 , but its N750 onset potential shifted to a more negative value (0.811 V).
Finally, at the highest temperature, more positive onset potentials were shown by the 800 and N800 samples.In particular, the N800 sample showed a well-defined curve and the best ORR activity, possibly as a result of having the highest surface area among all samples.
In general, the ORR performance of the electrocatalysts was attributed to their structures, namely, the increase in active sites, which facilitate oxygen diffusion and electron transport.
The results of the electrocatalyst cyclic voltammetry (CV) assessment in N 2 -saturated alkaline electrolyte (0.5 M KOH) at a 20 mV s −1 scan rate under room temperature conditions are presented in Figure 6b.All samples exhibited a pseudo-rectangular shape, which is associated with good charge propagation throughout the porous carbon framework, as well as good electronic conductivity and an accessible porous structure [46].
The K-L plots shown in Figure 7 were obtained from RDE polarization curve data recorded at various rotation rates (200, 400, 800, 1200, 1600, 2000 rpm) applying j , where j is the absolute measured current density (mA cm −2 ), j k is the kinetic current, ω is the electrode rotation rate (rpm), and B is determined from the K-L graph slopes.Based on the K-L plot slopes and the intercepts obtained from 0.2 to 0.6 V vs. RHE (V vs. RHE), Figure 8   Based on the K-L plot slopes and the intercepts obtained from 0.2 to 0.6 V vs. RHE (V vs. RHE), Figure 8 shows the electron-transfer numbers (n) calculated from the B = 0.2nFD 2/3 0 v −1/6 C 0 equation, where n is the total number of transferred electrons per O 2 molecule, F is the Faraday constant (96,485 C mol −1 ), D 0 is the O 2 diffusion coefficient (1.9 × 10 −5 cm 2 s −1 ), C 0 is the O 2 global concentration (1.0 × 10 −6 mol cm −3 ) [47], and υ is the electrolyte kinetic viscosity (0.01 cm 2 s −1 ) [10].
Based on the K-L plot slopes and the intercepts obtained from 0.2 to 0.6 V vs. RHE (V vs. RHE), Figure 8 shows the electron-transfer numbers (n) calculated from the  = 0.2 /  /  equation, where n is the total number of transferred electrons per O2 molecule, F is the Faraday constant (96,485 C mol −1 ), D0 is the O2 diffusion coefficient (1.9 × 10 −5 cm 2 s −1 ), C0 is the O2 global concentration (1.0 × 10 −6 mol cm −3 ) [47], and υ is the electrolyte kinetic viscosity (0.01 cm 2 s −1 ) [10].According to the slopes of the K-L plots, the averaged ORR electron transfer numbers were 3.4, 3.1, 3.5, 1.2, 2.2, and 3.7 for 700, 750, 800, N700, N750, and N800, respectively.There was a first-order dependence between 700, 750, 800, and N800 ORR kinetics and potentials from 0.2 to 0.5 V vs. RHE that proceeded mainly through a four-electron reduction pathway (2H2O + O2 + 4e − → 4OH − ).For these electrocatalysts, as the potential increased, the number of transferred electrons decreased to around 3.5; hence, both electron pathways coexisted for these materials.In contrast, the N700 and N750 kinetics reactions followed a mechanism closer to a two-electron transfer at potentials lower than 0.55 V.According to the slopes of the K-L plots, the averaged ORR electron transfer numbers were 3.4, 3.1, 3.5, 1.2, 2.2, and 3.7 for 700, 750, 800, N700, N750, and N800, respectively.There was a first-order dependence between 700, 750, 800, and N800 ORR kinetics and potentials from 0.2 to 0.5 V vs. RHE that proceeded mainly through a four-electron reduction pathway (2H 2 O + O 2 + 4e − → 4OH − ).For these electrocatalysts, as the potential increased, the number of transferred electrons decreased to around 3.5; hence, both electron pathways coexisted for these materials.In contrast, the N700 and N750 kinetics reactions followed a mechanism closer to a two-electron transfer at potentials lower than 0.55 V.
A brief comparison of the yield vs. other doped biomasses is provided in Table 6 for more insights.

Fuel Cell Studies
Figure 9 presents the fuel cell polarization and power density curves from MEA with NPC at 30 • C and 100% relative humidity using H 2 and O 2 gases prepared as described above.A correlation parameter brief for each cathode is provided in Table 7.The anode loadings were 0.4 mg Pt cm −2 (34 wt % Pt, 40 wt % C, and 25% ionomer), while, at the cathode, they were 4, 1, 4, 2, 2, and 2 mg cm −2 (75 wt % electrocatalysts and 25% ionomer) for 700, 750, 800, N700, N750, and N800, respectively.All samples outperformed the 700 sample's onset potential, indicating higher activity because of the N-doping effect and the pyrolysis temperature increase.The doping of the 700 sample influenced its polarization curve shape, as N700 demonstrated a higher activity (peak power 9.427 m 2 g −1 and 0.793 V) than the 700 sample.Both 750 and N750 showed similar OCVs, but different peak power densities.At 0.8 V, N750 performed slightly better than 750, but in the mass transport area, 750 had the best performance.Although 800 and N800 had the highest temperatures, and N800 had the highest BET surface area (2675.4mA cm −2 ), they exhibited a mass transfer limitation at a low potential (starting at 0.55 V), whereas samples at 700 and 750 • C did not.It is noteworthy that low loading cathodes (750 = 1.14 mg cm −2 and N750 = 1.59 mg cm −2 ) exhibited the best performance, so further electrode optimization must be performed, as both anode and cathode catalyst loading must be constant in order to get comparable results.

Conclusions
Electrocatalysts were synthesized by KOH activation and C5H5N doping of Sargassum spp.The surface area and pore volume decreased in the following order: 800 > N800 > 750 > N750 > 700 > N700.Chemical treatments performed with N700 allowed the surface area to increase from 60 m 2 g −1 to 2513, 2594, and 2675 m 2 g −1 for N700, N750, and N800, respectively.
The ORR performance of the electrocatalysts in terms of their current density was N800 > N750 > N700 > 750 > 800 > 700, while the onset potential decreased in the following order: N800 > 800 > 750 > 700 > N700 > N750.N800 presented the best ORR performance with the highest onset potential (0.870 V) and current density at 0.0 V vs. RHE (4.87 mA cm −2 ); this current density value is higher than commercial 10% Pt/C (4.0 mA cm −2 ).The fuel cell performance of the MEA prepared with an electrocatalyst synthesized at 750° C and doped with C5H5N was 12.72 mW cm −2 , which was close to that from Pt/C MEA on both the anode and cathode (14.42 mW cm −2 ).The performance was attributed to the high surface area, nitrogen content, defective structure, and graphitic nitrogen species found in Sargassum spp., which indicates that NPCs are an alternative to the problem of Sargassum spp.accumulation in the Caribbean due to their high efficiency as electrocatalysts for ORR.

Figure 3 .
Figure 3. (a) Raman spectra and (b) XRD patterns of electrocatalysts synthesized from Sargassum spp. at different temperatures.

Figure 3 .
Figure 3. (a) Raman spectra and (b) XRD patterns of electrocatalysts synthesized from Sargassum spp. at different temperatures.

Figure 6 .
Figure 6.(a) Linear sweep voltammetry (LSV) assessment of metal-free electrocatalysts at 2000 rpm and a 5 mV s −1 scan rate in O 2 -saturated 0.5 M KOH solution and (b) cyclic voltammetry (CV) in N 2 -saturated alkaline electrolyte (0.5 M KOH) at a 20 mV s −1 scan rate.

Figure 8 .
Figure 8. Electron-transfer numbers obtained from K-L plots for each synthesized electrocatalyst.

Figure 8 .
Figure 8. Electron-transfer numbers obtained from K-L plots for each synthesized electrocatalyst.

Figure 9 .
Figure 9. (a) Fuel cell polarization and (b) power density curves at 30 °C and 100% relative humidity using H2 and O2 gases.

Table 1 .
Nomenclature to identify the synthesized electrocatalysts.
• C and 2 M HCl acid washing at 95 • C 750 Grounded and washed raw sample (RS) impregnated at 1:2 RS/KOH ratio for 1 h, pyrolyzed in Ar@750 • C and 2 M HCl acid washing at 95 • C 800 Grounded and washed raw sample (RS) impregnated at 1

Table 2 .
Analysis of elemental CHNS composition (wt %) of the electrocatalyst bulk.

Table 2 .
Analysis of elemental CHNS composition (wt %) of the electrocatalyst bulk.

Table 3 .
Textural properties of the synthetized samples from Sargassum spp.

Table 3 .
Textural properties of the synthetized samples from Sargassum spp.

Table 5 .
Electrocatalytical parameters of the oxygen reduction reaction (ORR) performance.

Table 5 .
Electrocatalytical parameters of the oxygen reduction reaction (ORR) performance.

Table 6 .
Comparison among doped biomass used for electrocatalysts synthesis for ORR in alkaline media.

Table 7 .
Fuel cell parameters for electrocatalysts tested at 30 °C and 100% relative humidity using H2 and O2 gases.