Inspired by the Seeking Knowledge Strategy for Analects of Confucius to Screen Highly Electrochemically Active Porous Carbon: The Critical Source of Electroactive Sites

In this work, the forestry wastes are converted into a series of porous carbons using H3PO4 activation. These porous carbons feature a large specific surface area (1045.20 m2 g−1) and porosity that combines micro‐, meso‐, and macropores in various amounts depending on the fuel properties recorded for precursors used. Importantly, the C content recorded for forestry waste is one of the crucial factors in defining the specific surface area of the derived porous carbons. In addition, the total capacitance of the pine‐sawdust‐based porous carbon (PS‐C) sample is the highest, such as 220.55 F g−1 upon 5 mV s−1. Notably, the electrical double‐layer capacitance recorded for the samples remains essentially constant with increasing scan rates, such as ≈91.50 F g−1 for the olive‐shell‐based porous carbon, ≈123.70 F g−1 for PS‐C, and ≈105.66 F g−1 for the pine‐needle‐based porous carbon. Encouragingly, the pore‐associated sp3 site holds significant roles in the electrochemical application of the porous carbons. More importantly, the O/C value recorded for the precursor can be employed as a universal predictor of electrochemically active sites produced in porous carbons. In the findings, crucial insights are exhibited into the optimized fabrication of porous carbon with target electrochemically active sites for other applications such as catalysis.


Introduction
[3][4][5][6][7] Given the irreplaceable role of porous carbon, the demand for porous carbon is expected to be extremely high in the forthcoming years.10] For using porous carbon in energy storage systems, it is crucial to understand the structure-performance relationship.
As is known, porous carbon requires a high specific surface area and proper pore size distribution for ion transport and storage electrolytes when used as electrodes for supercapacitors. [11,12]To yield porous carbon characterized by these specific properties, various synthetic strategies have been developed, [13][14][15] e.g., Deng and coworkers obtained the capacitive porous carbon (S BET = 1238.81m 2 g À1 ) by UZnCl 2 -deep eutectic solvent (DES)-assisted synthesis strategy and suggested that high specific surface area was favorable to increase the specific capacitance. [16]Not only that Luo and coworkers prepared a series of chitosan-based hydrogel-beadsderived porous carbons by a rapid one-step method, and the results showed that the porous carbons characterized by a great specific surface area possessed high specific capacitances. [17]ifferently, Liao et al. indicated that the specific surface area is nonlinearly related to the specific capacitance. [18]Similarly, Wang et al. found that an increase in specific surface area on top of the ultrahigh specific surface area decreased the specific capacitance of the samples [19] ; from this, we can infer that there is a non-totally absolute linear relationship between specific capacitance and specific surface area.Despite the significant advances achieved in the mentioned studies, the quite different viewpoints reveal that the relationship between structures and capacitive properties of porous carbon has remained elusive.
[22][23] For instance, Chen and coworkers successfully prepared rice-husk-based porous carbons and tried to identify the inherent connection between pore DOI: 10.1002/sstr.202300329 In this work, the forestry wastes are converted into a series of porous carbons using H 3 PO 4 activation.These porous carbons feature a large specific surface area (1045.20 m 2 g À1 ) and porosity that combines micro-, meso-, and macropores in various amounts depending on the fuel properties recorded for precursors used.Importantly, the C content recorded for forestry waste is one of the crucial factors in defining the specific surface area of the derived porous carbons.In addition, the total capacitance of the pine-sawdust-based porous carbon (PS-C) sample is the highest, such as 220.55 F g À1 upon 5 mV s À1 .Notably, the electrical double-layer capacitance recorded for the samples remains essentially constant with increasing scan rates, such as ≈91.50 F g À1 for the olive-shell-based porous carbon, ≈123.70F g À1 for PS-C, and ≈105.66F g À1 for the pine-needle-based porous carbon.Encouragingly, the pore-associated sp 3 site holds significant roles in the electrochemical application of the porous carbons.More importantly, the O/C value recorded for the precursor can be employed as a universal predictor of electrochemically active sites produced in porous carbons.In the findings, crucial insights are exhibited into the optimized fabrication of porous carbon with target electrochemically active sites for other applications such as catalysis.
structures and capacitive performances. [24]Similarly, Zhang and coworkers fabricated hickory-shell-based porous carbons and established the volcanic-eruptive-type relationships via drawing curves of specific capacitance versus the ratio of micro-and mesopores. [25]Although previous efforts have exploited biomassderived porous carbons concerning synthesis, structures, and performance, to the knowledge of these authors, thus far, there have been very scarce in-depth studies of their relationship in the literature, which primarily is attributed to the fact that both the capacitive performance and the specific surface area are dynamically variable for a range of samples and that there are numerous factors that influence their capacitive performances (e.g., specific surface area, surface functional group, and current density).
The seeking knowledge strategy (i.e., query at both ends for understanding) for the Analects of Confucius-Zihan refers to considering both ends of the question so that all is clear on this question.To address the mentioned bottlenecks, inspired by the seeking knowledge strategy for the Analects of Confucius-Zihan, we have considered the following questions: 1) for biomassderived porous carbon, the obtained specific capacitances from the constant current charge-discharge (GCD) curves contain some pseudocapacitance, and typically the specific capacitance decreases gradually with increasing current density.Accordingly, how to ensure that the specific capacitance value remains unchanged despite changes in testing conditions within a range is one of the crucial objectives in solving the mentioned bottlenecks; 2) based on the well-established critical role played by specific surface area in affecting specific capacitance, the only way to eliminate any ambiguities regarding the specific surface area and to determine the effective area is through theoretical calculations to investigate how the nature of the carbon structure affects the capacitive properties when the effect recorded for the specific surface area is excluded; and 3) the presence of heteroatoms generates pseudocapacitance by the Faraday reaction for enhancing the total capacitance; the sole approach to eliminate any ambiguity in terms of heteroatoms is to exclude the pseudocapacitance caused by heteroatoms.Herein, using six forestry wastes with very different fuel properties as starting materials, we report on the effect of the fuel properties of forestry wastes on the texture properties and capacitive properties of the resulting forestry-wastes-derived porous carbon based on H 3 PO 4 activation.Over a range, the C content of forestry wastes is one of the crucial factors in defining the specific surface areas for their derived porous carbons.In addition, the specific capacitance recorded for pine-sawdust-based porous carbon (PS-C) achieved 156.36 F g À1 at 0.2 A g À1 in 6 M KOH aqueous electrolyte.
Encouragingly, pores associated with sp 3 bonds serve as the predominant origin of the electrocatalytic sites.Within a range, the formation of porous carbon characterized by numerous electrochemically active sites can be predicted from the O/C ratio reported for the carbonaceous matter.In summary, the motivation of this work is to produce cheap and sustainable porous carbons and identify the critical source of electroactive sites, which would assist in understanding the relationship between structure and performance and facilitating the applications for porous carbons in different fields.

Fuel Properties
As is known, porous carbons produced from various forestry wastes exhibit different structures and properties. [26]Based on this, the olive shell (OS), pine sawdust (PS), pine needle (PN), rice straw (RS), cotton stem (CS), and corn cob (CC) samples as the precursors for preparing porous carbons, and the fuel properties are shown in Table 1.There are remarkable differences in the fuel properties reported for these forestry wastes.Elemental C levels were highest in the PS sample, followed by OS, PN, CS, CC, and RS.Contrastingly, elemental O levels were highest in the CS sample, followed by OS, PN, CC, RS, and PS.Furthermore, the elemental H content in the precursors was observed to be ranging from 5.05 to 6.50 wt%.Non-negligibly, N content is one of the most crucial parameters for evaluating the cleanliness of feedstocks.Encouragingly, OS, PS, PN, RS, CS, and CC exhibited low N content, especially PS, which possessed as low as 0.17 wt% N. Consequently, the PS samples are clean, renewable, and promising energy materials, the preparation of which into porous carbon is a simultaneous strategy for promoting sustainable development, contributing to the "carbon peak and carbon neutrality" and material valorization. [27]urthermore, Zhang et al. indicated that rice husk had high ash content (15.84 wt%). [28]Notably, RS had up to 13.73 wt% ash, which implies that the ash content of rice husk was slightly higher than that of RS.Interestingly, the elemental C content of the feedstocks was ≈2-3 times their fixed carbon contents, and similar laws were recorded in the reported palm kernel shell, reed, and apricot shell. [26,29]In addition, analytical results indicated varying O/C ratios for these forestry wastes (e.g., 1.05 for OS, 0.90 for PS, 1.03 for PN, 1.17 for RS, 1.14 for CS, and 1.11 for CC). Figure 1 illustrates the variation of H/C and works. [30,31]Not to be ignored, in the case of OS, PS, PN, RS, CS, and CC, their H/C and O/C values were essentially in the vicinity of the H/C and O/C ratios recorded for biomass, but their values were significantly different, which is a crucial feature as this feature enables us to differentiate the fuel properties of the forestry wastes.-C) exhibit hierarchical micro/meso/macropore properties. [32,33]In particular, the N 2 sorption profile of the samples can be divided into three stages: P/P 0 < 0.4, 0.4 < P/P 0 < 0.8, and P/P 0 < 0.8.Of these, the first stage is ascribed to the filling of micropores with N 2 . [34]The hysteresis loop in the second stage signals that the samples possess  mesopore structures. [33]In addition, the tiny hysteresis loop in the third stage indicates that the samples have meso-and macropores structures. [35]It was expected that all porous carbon samples would display profile characteristics resembling those of OS-C since they were prepared at an analogous condition.Of note, not all N 2 adsorption curves indicate the same microporosity, as seen in Figure 2b, and some of the N 2 adsorption curves are significantly dissimilar to those for OS-C.In particular, OS-C featured meso-and macropore properties.Accordingly, the fuel properties recorded for precursors strongly dominate the texture properties recorded for porous carbons.In contrast, the amount of N 2 adsorbed is an indicator of the level of porosity (i.e., surface area and total pore volume) generated by porous carbons.As a result, PS-C had a high specific surface area, followed by PN-C and OS-C.Interestingly, the ranking in O/C values of forestry wastes was identified: PS < PN < OS.Based on this, we interpret the high specific surface area to mean that PS was characterized by a low activation resistance.Such an explanation for the lower activation resistance may be related to the O/C value.Actually, the semilog plot recorded for the N 2 adsorption isotherm (Figure 2c) shows the adsorption of biomimetic-like pores of porous carbon.Given the different fuel properties of OS, PS, PN, RS, CS, and CC, various profiles of pore size distribution curves can be predicted for their derived porous carbon samples even though they were obtained under similar conditions.Encouragingly, the pore size distribution curves for OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C were distinct, as shown in Figure 2d.Nevertheless, the micropore size recorded for the as-prepared porous carbon concentrated at 0.7-2 nm, with numerous mesopores at 2-5 nm.One noteworthy feature of the OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C samples shows multi-peaked pore size distributions featuring abundant mesopores clustered at ≈2.5 nm.Further, PN-C exhibited a higher pore size distribution of around 0.7 nm, suggesting a tendency toward micropores.Such findings can equally be verified from microporosity (Table 2).Summarily, changes in pore size distribution also support that the fuel properties recorded for forestry waste have a strong influence on the pore size in the resulting porous carbon.Table 2 summarizes the texture properties recorded for porous carbons.Overall, OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C exhibited texture property parameters in agreement with their N 2 adsorption isotherms and pore size distribution curves.As an example, PS-C demonstrated the greatest specific surface area, followed by PN-C, OS-C, CS-C, CC-C, and RS-C.Encouragingly, the enhancement of porosity was achieved through higher C element content and was particularly evident at a C element content of 46.72 wt%.As such, the specific surface area for PS-C showed 1045.20 versus 484.70 m 2 g À1 for RS-C, an increase of 115.64%.Generally, all porous carbon samples presented low or moderate specific surface areas in the range of 484.70-1045.20m 2 g À1 , with total pore volumes between 0.33 and 0.83 cm 3 g À1 .Noteworthy, the remarkable difference in specific surface area is one of the critical aims of the present work, as it facilitates determining the mechanism of the electroactivity source.Encouragingly, the highest specific surface area, i.e., 1045.20 m 2 g À1 for PS-C, was significantly greater compared to those of the porous carbons prepared by two-step methods, e.g., carbonization after ambient drying (600 m 2 g À1 ), [36] carbonization after freeze drying (1013 m 2 g À1 ), [37] and carbonization after supercritical drying (587 m 2 g À1 ). [38]As previously reported, lignin-derived porous carbon prepared using H 3 PO 4 activation provided medium-specific surface areas (≈1000 m 2 g À1 ). [39]vidently, PS-C had a specific surface area comparable to the lignin-derived porous carbon.In contrast, PS-C featured the thinnest pore-wall thickness (8.70 nm), facilitating the interchannel transfer of electrolyte ions in porous carbon.Encouragingly, the yields (27.06%-30.02%) of the porous carbon samples were higher owing to the lower activation temperature (Table 2).

Nature of the Samples
The properties and purity of the porous carbon samples were investigated using powder X-ray diffraction (XRD).From Figure 3a, it can be seen that the XRD patterns of all samples possess wider (002) diffraction peaks at ≈23°and fainter diffraction peaks at ≈43°(corresponding to the (100)/(101) diffractions caused by graphene stacks, which are typical of graphitized structures and main amorphous phase characteristics). [40,41]Based on this, we can infer the presence of multiple graphene-like stacks in the as-prepared porous carbon samples, in which the changes occurring in the stacks correspond to the destructive behavior during the activation process.Additionally, the intensity of the samples was noticeable in the low-angle regions, demonstrating highly developed nanopore structures.According to the literature, electrolyte ions stored in nanopores formed by irregular graphene layers in amorphous carbon indicate that this porous carbon with low crystallinity and nanopore structure is favorable for charge storage. [42]The R value (the A intensity divided by the B background, Figure S1, Supporting Information), proposed by Liu et al. for quantifying defects inside the porous carbons, was estimated from the XRD patterns. [43]As shown in Figure S1, Supporting Information, OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C had R values of 1.76, 1.44, 1.75, 1.85, 1.80, and 1.84, respectively.As is known, a lower R value reflects a higher defect density. [44]Significantly, PS-C presented such poorest R value, followed by PN-C, OS-C, CS-C, CC-C, and RS-C.Consequently, PS-C featured such greatest defect density, followed by PN-C, OS-C, CS-C, CC-C, and RS-C.More importantly, the R value correlates inversely with the specific surface area, from which it can be inferred that high specific surface areas favor the production of defective structures.Conclusively, Figure 3a demonstrates the generated samples to be typical of chemically activated carbons and amorphous.
In addition, the defect structures of the resultant porous carbon were further investigated by Raman spectra.For all porous materials, the D band is at ≈1340 cm À1 and the G band is at ≈1585 cm À1 . [45]Typically, the D band is assigned to defect sites or stretching vibrations of disordered carbon (e.g., carbon atoms near the edges of a graphene layer).In addition, the G band contributes directly to the bond extension of the paired sp 2 carbons, typically suggesting that graphite is crystalline. [46]Crucially, the intensity ratio of the D and G peaks, i.e., I D /I G , is usually utilized for identifying the defect or disorder levels of porous materials. [47]The As is known, many small organic molecules transformed into conjugated aromatic rings that form edge structures with numerous defects along the boundaries of biochar. [48]The presence of edge vacancies can improve the performance of porous carbon.Further, the as-yielded porous carbons achieved higher I D /I G values than those of the biomass-derived porous carbons manufactured through the two-step process (I D /I G value: 0.88 [49] and 0.96 [50] ); from this, we can infer that as-obtained porous carbon, especially PS-C had numerous defects.As is known, the defective structures in porous carbon favor the transport of electrolyte ions and reinforce the energy storage sites of porous carbon. [51]ccordingly, PS-C presents great promise as an electrode material used in capacitors.Generally, the Raman spectra results showed agreement with the XRD analysis.
To observe the surface elements recorded for the samples, PS-C and RS-C were used as case studies for X-ray photoelectron spectroscopy (XPS) analysis.Figure 3c demonstrates the survey spectra of porous carbon with prominent peaks at ≈285.5, ≈532.5, ≈400.1, ≈200, and ≈134.3 eV for C1s, O1s, N1s, P2s, and P2p, respectively. [52,53]As is known, the O/P/N-rich structures favor the wettability and pseudocapacitance recorded for porous carbons.Specifically, PS-C and RS-C possessed high O content values (21.18-26.02at%), superior to previously reported data for porous materials (7.30% [54] and 2.80% [55] ).Alternatively, PS-C and RS-C had P data ranging from 3.96 to 5.18 at%, indicating some doping effect using phosphoric acid.Notably, the C1s content recorded for PS-C was 73.3d-f and S2a-c, Supporting Information).Nevertheless, the content of the corresponding functional groups was significantly different, e.g., 6.84% C-O for PS-C and 10.30% C-O for RS-C.[58][59] As such, the FTIR spectra were used to analyze the surface functional groups of samples, and a range of typical functional groups can be recognized by their spectral intensity.Additionally, the various characteristic peaks corresponding to the wavelengths are summarized in Table S1, Supporting Information.Notably, the FTIR spectra of PS-C and RS-C were similar, indicating that PS-C and RS-C exhibited similar surface functional groups (Figure S2d, Supporting Information). [60]Nevertheless, the FTIR spectral intensity of PS-C and RS-C was slightly different, suggesting a difference in the surface functional group content of PS-C and RS-C (Figure S2d, Supporting Information). [60]More encouragingly, the spectrum of PS-C and RS-C exhibits a peak at ≈668 cm À1 , which belongs to the C-P stretching. [52]This result, as with XPS, suggests that phosphorus species were introduced into the porous carbons during the H 3 PO 4 activation process.We then investigated the morphologies of OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C using scanning electron microscopy and high-resolution transmission electron microscopy.Forestry wastes have extremely few pore structures except for PS (Figure 4a,b,j,m,p,s).Contrarily, porous carbon exhibited numerous pore structures after activation by H 3 PO 4 (Figure 4b,c,e-i,k,l,n,o,q,r, and t,u).Interestingly, the PS samples featured natural hierarchical porous structures in the transverse and longitudinal directions (Figure 4d), and these natural hierarchical porous structures facilitate the entry of the activator.Not only that the hierarchical porous structures recorded for PS were preserved to some extent after activation (Figure 4e-i).Encouragingly, the pore wall of PS possessed numerous pore structures (Figure 4f ), and this interpenetrating pore structure not only provided a path for the high-speed transport of electrolyte ions but also promoted the transfer of electrolytes across the channel.Also, forestry-waste-based porous carbons (i.e., OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C) featured plenty of irregular nanopores (Figure S3, Supporting Information), suggesting abundant sites for absorbing electrolyte ions. [61]Further, the presence of nanopores reveals that the forestry-waste-based porous carbons (i.e., OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C) showed amorphous states, agreeing with the XRD and Raman spectra results.Notably, the presence of numerous bulge-like structures on the surface of the RS-C samples (SiO 2 is present in the bulge) is consistent with the fact that they were high ash. [62]In conclusion, the significant differences in the morphology and number of pore structures further indicate that the fuel properties and structures of forestry wastes are crucial factors dominating the pore structures recorded for their derived porous carbons.

Electrochemical Properties
In general tests, the electrochemical properties in terms of cyclic voltammetry (CV), GCD, and electrochemical impedance spectrum (EIS) of all samples were evaluated in an aqueous 6 M KOH electrolyte using a three-electrode system.As shown in Figure 5a, the CV curves of all samples at a scan rate of 5 mV s À1 exhibit a nearly rectangular shape.Consequently, the electrochemical properties recorded for the as-obtained porous carbons were dominated by the electrical double-layer capacitive (EDLC) behaviors.Additionally, these humps in the CV curves are typically attributed to pseudocapacitance produced by the active functional groups. [63]Nevertheless, the PS-C sample exhibited the largest CV integral area among these porous carbons, suggesting that it offers the best charge-storage capacity, which matches its great specific surface area and high micropore content.Moreover, the CV curves recorded for the as-prepared porous carbons (i.e., OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C) upon various scanning rates are illustrated in Figure S4a-f The symmetric GCD curves reveal reversible ion adsorption/desorption procedures and efficient ion transfer at the electrode/electrolyte interface.In addition, the neartriangular GCD curves illustrate the remarkable EDLC properties of the as-obtained porous carbons. [64]Consequently, samples OS-C, PS-C, PN-C, CS-C, and CC-C possessed outstanding electrical double-layer capacitance (EDLC).Despite this, the sample PS-C provided the best performance regarding EDLC, followed by PN-C, OS-C, and CS-C.From this, we can infer that the fuel properties recorded for the precursors remarkably affect the electrochemical performances recorded for the carbon electrodes.Additionally, the voltage drop (IR drop) recorded for the PS-C electrode is negligible at elevated current densities compared to the electrode RS-C.In contrast, the specific capacitance at various current densities can be determined based on the discharge curves, as shown in Table 3 and Figure 5c.The specific capacitance of forestry-waste-based porous carbon decreases with increasing current density, which agrees with the previously reported trend. [65]Significantly, the PS sample possessed the greatest specific capacitance (e.g., 156.36 F g À1 at 0.2 A g À1 ), and its capacitance retention was as high as 66.51% at 10 A g À1 (Figure 5d).Encouragingly, the highest result (i.e., 66.51%) showed that the pore structure and high specific surface area of PS-C significantly improved the capacitance retention compared to previous findings (10 A g À1 : 53% [66] and 64% [67] ).Moreover, ion diffusion and transport kinetics recorded for samples were estimated in depth by EIS tests.All Nyquist profiles exhibit similar trends, both one vertical line as well as one semicircle, as illustrated in Figure 5f.Generally recognized, the vertical lines in the low-frequency region reveal the primary features of the EDLC. [68]As such, the as-obtained porous carbons, especially PS-C, had desirable EDLC performances.Previously several efforts have provided equivalent circuit diagrams in Nyquist plots. [58,59,69]Similarly, the present work equally offered equivalent circuit diagrams in the Nyquist plots (as given in Figure 5f ).In the enlarged Nyquist plot (shown in Figure 5f ), the diagonal line with a slope of ≈45°in the mid-frequency region correlates to the Warburg resistance (W o ), demonstrating the diffusion of electrolyte ions across the porous electrodes.Thus, electrolyte ions can be quickly transferred in sample PS-C.In addition, the equivalent series resistance (R s ) is defined by the intercept of the high-frequency region concerning the real axis.The fitted R s , R ct (charge-transfer resistance), and W o values of samples are listed in Table S2, Supporting Information.Encouragingly, the low R s value of 0.78 Ω for the PS-C was lower compared to the previously published value (0.93 Ω). [70] In other words, the sample PS-C had favorable electrical conductivity.In reviewing the literature, Le et al. demonstrated that the R s of the porous carbon (CKKF700-3) was the lowest (0.35 Ω), attributing to its loose porous structures [71] ; from this, it can be inferred that PS-C in this work endowed with a lower R s value may also be due  Table 3. Variation of the specific capacitance in the three-electrode system.to its loose porous structures.As is known, the cycle life is also an important indicator for evaluating the properties of porous carbon electrodes.As shown in Figure 5e, the capacitance retention of PS-C at 1 A g À1 was as high as 88.33% in 5000 cycles, which indicates its excellent cycling stability.Based on the CV curves, the kinetics of the porous carbon electrodes can be analyzed using the formula below (I = kv b ; log(i) = blog(v) þ log(k); where i and v are the current density and scan rate, respectively). [72,73]he category of reaction kinetics of the electrodes is defined by the value of b. b equal to 0.5 indicates the electrode reaction kinetics of Faraday redox behaviors, and b equal to 1 indicates the fast kinetic process of the EDLC. [74,75]As shown in where k 1 v denotes the quick kinetics procedure followed to EDLC and k 2 v represents the slow-kinetic procedures for pseudocapacitance).Evidently, the capacitive behaviors of the samples, except RS-C, were the dominant contributors to the total capacitance, while the diffusive contributions primarily occurred at lower scan rates.In addition, the capacitive behaviors of PS-C were the most prominent, followed by OS-C, CS-C, and PN-C.Notably, an extremely small percentage of pseudocapacitance was recorded at high scan rates owing to insufficient redox reaction time.

Sources of Electrochemically Active
As is known, the relationship between specific surface area and capacitance performance is not a simple linear one, which results from many factors affecting capacitance performance, such as specific surface area, current density, and change in total capacitance. [76]As a result, the exclusion of specific surface area, current density, and changing total capacitance assists in understanding the structure-capacitance performance relationship.
Based on this, we calculated the EDLC and pseudocapacitance of the prepared samples from the CV curves, and the equations are listed in Supporting Information.As shown in Table S3, Supporting Information, the total capacitance of the porous carbon samples decreases with increasing scan rates, regardless of the structural nature of the porous carbon samples.Interestingly, this trend is consistent with the tendency of the specific capacitance to change with increasing current density.Remarkably, PS-C possessed the highest total capacitance, followed by PN-C and CS-C, which is generally consistent with the ranking of the specific capacitance calculated using GCD curves.Encouragingly, the EDLC of the samples (i.From Table S3 and S4, Supporting Information, we realize that the decrease in pseudocapacitance was responsible for the decrease in total capacitance as the scan rate increased.Capacitance that does not vary with changes in current density or scan rate is one of the critical goals of this work, as essentially constant capacitance facilitates the construction of structure-capacitance performance relationships.As aforementioned, the specific surface area is not simply linearly related to the capacitance performance, so further eliminating the effect of the specific surface area is equally an important objective of the present study.Interestingly, the normalization method can eliminate the effect of specific surface area.Encouragingly, the specific surface-area-normalized EDLC of OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C were 123.8a).Notably, the specific surface-area-normalized EDLC of CS-C was significantly higher than that of previous works (Table S5, Supporting Information).Evidently, the specific surface-area-normalized EDLC represents the effective electrochemically active sites.To further identify the mechanism of the electrochemically active source, we performed the electron energy loss spectroscopy (EELS) test.The sp 3 content was significantly different in the samples (Figure 8b).Specifically, the sp 3 content was 22% for OS-C, 21% for PS-C, 20% for PN-C, 17% for RS-C, 24% for CS-C, and 23% for CC-C.In other words, the sp 3 content of CS-C was the highest, followed by CC-C, OS-C, PS-C, PN-C, and RS-C.In contrast, linear fitting is commonly used to assess the correlation between two variables, and a correlation coefficient higher than 0.7 indicates a strong linear relationship. [77,78]According to the literature, Wang et al. evaluated the relationship between naproxen degradation rate on a logarithmic scale and porous carbon structure by linear fitting. [79]The results indicated that there was a positive linear correlation between the naproxen degradation rate on a logarithmic scale and the I D /I G value.Zhou and co-workers constructed the relationship between N content and CO 2 adsorption rate by linear fitting. [80]Kim and co-workers established the relationship between carbon structure and CO 2 adsorption rate employing linear fitting. [81]Inspired by this, linear fitting was equally used in this work to evaluate the relationship between the sp 3 content and specific surface-area-normalized EDLC, as shown in Figure 8c.Encouragingly, the correlation coefficient (Pearson's r) between the sp 3 content and specific surface-area-normalized EDLC was as high as 0.98, suggesting that the pore-associated sp 3 site holds significant roles in the electrochemical application of the porous carbons.In addition, the O/C ratios of forestry wastes showed a linear relationship with the specific surfacearea-normalized EDLC recorded for their derived porous carbons (Figure 8d); from this, we can infer that the sp 3 site becomes an electrocatalytic site due to the localized charge density and facilitates ion insertion by widening the sp 2 layer spacing (Figure 8e).More importantly, according to the O/C ratio of carbonaceous matter, it is possible to predictably produce porous carbons characterized by an abundance of electrochemically active sites.Non-negligibly, the elemental C content of forestry wastes is one of the critical factors in determining the specific surface area of their derived porous carbons (Figure S6, Supporting Information).In summary, we have fabricated a series of porous carbons using H 3 PO 4 activation and shown that the poreassociated sp 3 site is the dominant source recorded for electrochemically active, which provides a theoretical basis for further understanding of electrochemistry or catalysis.It is necessary to mention that the mass loading of the porous carbon in the electrode was ≈3.5 mg cm À2 , which is lower than the high mass loadings of commercial electrodes. [56,69]Of note is that the present work has examined only the electrochemical properties of porous carbon with low mass loading (i.e., ≈3.5 mg), which does not indicate that the electrochemical properties of porous carbon with high mass loading (>10 mg cm À2 ) are not crucial.Notwithstanding its limitations, this study does provide a lot of important data for understanding the critical source of electroactive sites.

Conclusions
Herein, a series of porous carbons with various texture properties have been synthesized using H 3 PO 4 activation.The analytical results revealed that the fuel properties of forestry wastes significantly influence the texture properties recorded for their derived porous carbons.Specifically, the C content recorded for forestry waste is one of the crucial factors in defining the specific surface areas recorded for the derived porous carbons.

Experimental Section
In the present work, the OS, PS, PN, RS, CS, and CC samples were crushed into uniform particles (40-60 mesh), washed using deionized water, and dried out in an oven at 105 °C for 12 h.Subsequently, the OS, PS, PN, RS, CS, and CC feedstocks were fabricated into porous carbons through an H 3 PO 4 -assisted procedure.Specifically, a mixture of H 3 PO 4 solution (58.10 wt%) and feedstocks in a weight ratio of 3:1 was placed in a ceramic ark, which was then subjected to heating to 550 °C using a tube furnace at 10 °C min À1 for 60 min.Additionally, the activation temperature (i.e., 550 °C) refers to the activation temperature of H 3 PO 4 in the literature. [82]Notably, the feedstock was a variable in the whole process, where the feedstock could be OS, PS, PN, RS, CS, and CC.The as-obtained porous carbons, OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C, were washed with deionized water to neutral (pH ≈ 7) and then dried in an oven at 105 °C for 12 h.Moreover, material characterization and electrochemical test methods are available in Supporting Information.

Figure 2a presents a
Figure 2a presents a classic preparation procedure for porous carbon electrodes.For more information, see Experimental Section.In addition, the N 2 sorption curves are displayed in Figure 2b.Meanwhile, the pore size distribution curves are displayed in Figure 2d.Irrespective of the fuel properties recorded for feedstocks, the N 2 sorption curves of the samples match the International Union of Pure and Applied Chemistry (IUPAC)-IV type, demonstrating that OS-based porous carbon (OS-C), PS-based porous carbon (PS-C), PN-based porous carbon (PN-C), RS-based porous carbon (RS-C), CS-based porous carbon (CS-C), and CC-based porous carbon (CC-C) exhibit hierarchical micro/meso/macropore properties.[32,33]In particular, the N 2 sorption profile of the samples can be divided into three stages: P/P 0 < 0.4, 0.4 < P/P 0 < 0.8, and P/P 0 < 0.8.Of these, the first stage is ascribed to the filling of micropores with N 2 .[34]The hysteresis loop in the second stage signals that the samples possess

Figure 2 .
Figure 2. a) Fabrication of the porous carbon electrodes.b) N 2 sorption isotherms of samples.c) Semilog plots of N 2 sorption isotherms.d) Pore size distribution of samples.
I D /I G values of OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C were detected as 1.34, 1.43, 1.36, 1.25, 1.30, and 1.29.From this, the defect degree of the porous carbon samples was determined: PS-C > PN-C > OS-C > CS-C > CC-C > RS-C.
61 versus 67.72 at% for RS-C, an increase of 8.69%.In addition, the types of C-(C-C/ C-H, COOH/COOR, C-O, and O-C=O), O-(C-O-C, C-O/ P-OH, O=C-O, and C=O/P=O), and P-(C-P-O, C-O-P, C 3 -P=O, and P-O) containing functional groups were identical in PS-C and RS-C (Figure , Supporting Information, respectively.The closed CV curves recorded for the PS-C sample exhibited excellent quasirectangular shapes even at elevated scanning rates compared to the RS-C sample, indicating the optimal capacitance properties and rate capability of PS owing to the rational pore structures.Significantly, the weak and broad peaks presented in the CV curves demonstrate the redox reaction occurring, which supports the production of pseudocapacitance.The GCD curves recorded for OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C at 0.2 A g À1 are shown in Figure 5b.Additionally, Figure S4g-l, Supporting Information, shows the GCD curves for OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C at current densities ranging from 0.2 to 20 A g À1 .

Figure 5 .
Figure 5. Electrochemical properties of the three-electrode system, a) CV curves of samples at 5 mV s À1 ; b) GCD curves of samples at 0.2 A g À1 ; c) plot of specific capacitance upon various current densities; d) plot of capacitance retention upon different current densities; e) cycling stability at a current density of 1 A g À1 for samples; and f ) Nyquist plots for samples in 6.0 M KOH, were recorded at a frequency range from 0.01 Hz to 100 kHz.
Figure 6a and S5a-k, Supporting Information, the b values of charge and discharge states for all samples are in the range of 0.50-1, which indicates the electrochemical behaviors dominated by the EDLC and accompanied by some pseudocapacitance.As is known, the total capacitance of biomass-based porous carbon generally includes the EDLC and Faraday pseudocapacitance.Moreover, the relationships between capacitive and diffusion contributions for OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C are shown in Figure 6b and 7a-f.Notably, these relationships were determined using the Dunn method e., OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C) remained essentially constant with increasing current density, e.g., the EDLC recorded for PS maintained at 123.70 F g À1 (Table S4, Supporting Information).Nonetheless, the order for the EDLC of the samples was established: PS-C (≈123.70F g À1 ) > CS-C (≈109.78F g À1 ) > PN-C (≈106.11F g À1 ) > CC-C (≈95.80 F g À1 ) > OS-C (≈91.50F g À1 ) > RS-C (≈10.56F g À1 ).

Figure 6 .
Figure 6.a) The b values of charge state for the PS-C electrode at different voltages; b) capacitance contribution at 5 mV s À1 for the PS-C electrode.

Figure 7 .
Figure 7.The columnar plot of capacitance contributes at different scan rates for a) OS-C, b) PS-C, c) PN-C, d) RS-C, e) CS-C, and f ) CC-C.

Figure 8 .
Figure 8. a) The normalized capacitance of samples, numbers 1-6 stand for OS-C, PS-C, PN-C, RS-C, CS-C, and CC-C, respectively.b) EELS spectra for samples.c) Correlation between sp 3 content and specific surface-area-normalized electrical double-layer capacitance.d) Correlation between O/C and specific surface-area-normalized electrical double-layer capacitance.e) Molecular dynamics (MD) simulation of sp 3 carbon layer evolution as O/C increases.

Table 1 .
Fuel properties of forestry waste.
O/C values for forestry wastes.Notably, the forestry wastes possessed higher amounts of elemental H and O than the H/C and O/C values for anthracite, coal, and peat recorded in previous

Table 2 .
Pore structure parameters of forestry-waste-based porous carbon.
3articularly, PS-C was characterized by a large specific surface area (1045.20 m 2 g À1 ), oxygen content (21.18 at%), nitrogen content (1.31 at%), and phosphorus content (3.96 at%).Benefiting these advantages, the specific capacitance recorded for PS-C achieved 156.36 F g À1 at 0.2 A g À1 in 6 M KOH aqueous electrolyte.Notably, the EDLCs recorded for the samples remained essentially constant as the scan rate increased, with ≈91.50 F g À1 for OS-C, ≈123.70 F g À1 for PS-C, ≈105.66FgÀ1 for PN-C, ≈10.56 F g À1 for RS-C, ≈109.53FgÀ1 for CS-C, and ≈95.82F g À1 for CC-C, respectively.Encouragingly, pore-related sp3bonds are the predominant source of electrocatalytic sites.Over a range, the generation of porous carbon with numerous electrochemically active sites can be predicted based on the O/C value recorded for the carbonaceous matter.Such findings recorded in the present work provide a new clue for understanding electrochemically active sites.