Abstract
Nitrogen-doped graphene holds immense promise for electrocatalysis, where surpassing the traditional 2–5 at% nitrogenation ceiling requires strategic control of thermodynamic and kinetic factors. Here, we systematically investigate the dominant thermodynamic and kinetic parameters to reach higher nitrogenation. It is demonstrated that, within the investigated 140–200 °C hydrothermal regime, neither kinetic nor mass-transport limitations matter, in sharp contrast to the prevailing literature. Instead, nitrogenation levels are fundamentally governed by the population of basal-plane epoxide groups, meaning that the starting graphene oxide oxidation degree alone limits ultimate doping efficiency. Optimizing precursor oxidation successfully drives the final nitrogen content to 17.3 at% (N/C = 0.24). Subsequent thermal treatment efficiently converts this grafted nitrogen into lattice-doped configurations dominated by pyridinic N (57%). This work establishes a framework showing that maximizing reactive oxygen sites is the definitive pathway to advanced nitrogen-doped frameworks.


Nitrogen-doped graphene has been widely investigated for electrocatalysis, energy storage, adsorption, and sensing because nitrogen functionalities can modulate the electronic structure and surface chemistry of the carbon framework. − Higher nitrogen contents are often associated with improved activity or performance, motivating continued efforts to increase the density of accessible nitrogen configurations at the surface. , Despite diverse synthetic strategies, including chemical vapor deposition, thermal annealing, plasma treatment, and hydrothermal or solvothermal processing with ammonia, hydrazine, or urea, the nitrogen contents of nitrogen-doped graphene commonly remain within approximately 2–6 at% and only occasionally reach 7–9 at%. − Even solvent-enhanced nitrogenation in our previous graphene composite reached only approximately 7 at%. The persistence of this nitrogenation ceiling across fundamentally different approaches suggests that the final nitrogen level may not be governed solely by conventional kinetic parameters such as reaction temperature, duration, and nitrogen-source concentration. Mass transport and active-site accessibility have also been proposed as limiting factors because severe restacking of two-dimensional graphene oxide (GO) reduces accessible porosity and restricts the diffusion of nitrogen precursors into interlayer reactive regions. This consideration has motivated the development of expanded or crumpled GO architectures with more open pore structures, although whether enhanced accessibility alone can increase nitrogen incorporation has not been systematically established. ,
A further complication is that the oxidation degree and oxygen speciation of GO precursors are rarely controlled independently of kinetic and structural variables. It therefore remains unclear whether ammonia-based nitrogenation is limited primarily by reaction kinetics, mass transport, or the finite population and intrinsic reactivity of oxygen-containing sites. Because ammonia reacts with oxygen functionalities such as hydroxyl, epoxide, and carboxyl groups, the abundance and chemical identity of these sites may impose an intrinsic upper limit on nitrogen incorporation. , Here, we employ a vapor-phase hydrothermal platform to independently examine the effects of reaction temperature, duration, ammonia dosage, precursor porosity, and oxidation degree on GO nitrogenation. GO precursors with controlled oxidation degrees were prepared using a preformed acidic oxidizing medium method, while three-dimensional crumpled graphene oxide generated through a cryogenic process provided a highly open, nonrestacked architecture for evaluating mass-transport and active-site accessibility effects under otherwise matched chemical conditions. , The initially introduced surface-associated nitrogen functionalities and their subsequent thermal conversion into lattice-associated pyridinic, pyrrolic, and graphitic configurations were evaluated separately, because the lattice-doped nitrogen contents commonly reported for nitrogen-doped graphene remain relatively modest. ,− This design enables kinetic, transport, and oxygen-site effects to be distinguished and provides a framework for identifying the principal factors governing high-level nitrogen incorporation in GO.
To implement this comparison while preventing solvent-induced collapse of CGO, we adopted a vial-in-liner vapor-phase hydrothermal configuration (Figure a). Because highly oxidized CGO readily collapses upon direct contact with polar solvents such as water, the GO or CGO powder was placed in a glass vial, while the aqueous ammonia solution was placed outside the vial within a Teflon liner. Nitrogenation therefore proceeded through vapor-phase species diffusing into the vial, eliminating direct liquid contact with the precursor and allowing two-dimensional GO and three-dimensional CGO to be compared without solvent-induced structural collapse.
1.

Vapor-phase hydrothermal nitrogenation of crumpled graphene oxide and structural evolution after nitrogen incorporation. (a) Schematic illustration of the vapor-phase hydrothermal nitrogenation process. CGO was placed in a vial inside a Teflon liner to avoid direct contact with the surrounding ammonia solution. During heating at 180 °C for 12 h, NH3/H2O vapor diffused into the vial and reacted with CGO to form nitrogenated crumpled graphene oxide (NrCGO). (b) Photographs of rCGO and NrCGO with the same sample mass, showing the apparent volume contraction after nitrogenation. (c–e) SEM images of (c) CGO, (d) rCGO, and (e) NrCGO, confirming that the crumpled morphology is retained after vapor-phase treatment and nitrogenation. Insets show the corresponding high-magnification images. (f) XRD patterns of GO, rGO, rCGO, and NrCGO, showing suppressed long-range restacking in the crumpled samples. (g) FTIR spectra of GO, rCGO, and NrCGO, where the emergence of the C–N vibration at approximately 1420 cm–1 supports nitrogen incorporation. (h) Raman spectra of rCGO and NrCGO, showing the characteristic D and G bands of graphitic carbon frameworks. (i) TGA profiles of rCGO and NrCGO under N2 atmosphere, revealing the different thermal decomposition behavior after nitrogenation. (j) N2 physisorption isotherms with the corresponding BET surface areas and (k) BJH pore-size distribution curves of rCGO and NrCGO, indicating that porous features are retained despite partial densification after nitrogenation.
Before assessing these variables experimentally, we established an idealized nitrogenation capacity by treating reactive oxygen functionalities in GO as the available conversion sites. Representative reaction pathways associated with hydroxyl, epoxide, and carboxyl functionalities are schematically illustrated in Scheme . Highly oxidized 3 × GO, with an O/C atomic ratio of approximately 0.55 determined by XPS (Table ), was used as the benchmark for estimating the idealized nitrogenation capacity. Based on literature analyses, the oxygen species in GO are mainly distributed among hydroxyl (∼12%), epoxide/ether (∼34%), and carboxyl (∼27%) species, but hydroxyl groups are not expected to contribute effectively to nitrogen incorporation. The thermodynamic favorability in terms of free energy and mechanistic plausibility of these representative pathways was evaluated based on the calculated ΔH°, ΔS°, and ΔG° values summarized in Table S1, while the associated functional-group reactivity is further discussed in Supporting Information Section S2. Ketones found on the edge of 3 × GO may also react with NH3(g) to form imines through a carbinolamine intermediate. However, the difficulty of the dehydration step required to convert the unstable carbinolamine intermediate into the corresponding imine under basic conditions renders ketones unlikely to contribute greatly to nitrogen incorporation. The free energies indicate a clear hierarchy in thermodynamic driving force: nitrogenation at carboxyl- and hydroxyl-derived sites is only modestly favorable (ΔG° in the range of −21.94 to −19.04 kJ mol–1), whereas epoxide ring opening is strongly spontaneous (ΔG° ≈ −374.40 kJ mol–1). Direct substitution of a C-OH group with either NH3(g) or aqueous NH4OH is highly unlikely because hydroxide OH– is a poor leaving group that holds onto carbon too tightly to be displaced, while the excess of water shifts the chemical equilibrium backward, and the formation of unreactive ammonium ions significantly inhibits the reactivity. (see Supporting Information Section S2) Accordingly, epoxide functionalities are expected to constitute the most thermodynamically reactive oxygen sites toward ammonia-based nitrogenation. Based on the measured oxygen content of 3 × GO (O/C = 0.55), we estimated the idealized nitrogenation capacity by assuming complete conversion of all reactive oxygen functionalities according to the stoichiometries proposed in Scheme . Hydroxyl groups were excluded from the calculation because they are not expected to contribute significantly to ammonia-based nitrogenation. In contrast, epoxide groups were assumed to undergo a 1O-to-1N conversion through nucleophilic ring-opening by NH3, whereas carboxyl groups were treated as a 2O-to-1N conversion due to ammonium carboxylate formation. Using the reported oxygen-species distribution in GO and the measured O/C ratio of 3 × GO, the idealized nitrogenation capacity was estimated to be N/C = 0.26, without considering additional carbon loss during hydrothermal treatment. In comparison, applying the same estimation framework to the lower-oxidation commercial GO precursor (O/C = 0.41) yields a lower idealized nitrogenation capacity of approximately N/C = 0.20. Apparently, despite extensive reports on hydrothermal/solvothermal nitrogenation of GO, nitrogen contents approaching this idealized nitrogenation capacity have not been achieved, recognized as the widely observed “nitrogenation ceiling”. ,−
1. Possible Reaction Pathways and Oxygen-to-Nitrogen Conversion Assumptions for Ammonia-Based Nitrogenation of Oxygen Functionalities in Graphene Oxide .

a (a) Possible substitution of basal-plane carbon hydroxyl groups by NH3 to form C–NH2 motifs with H2O elimination; this pathway is unlikely because −OH is a poor leaving group and NH4 +/water suppresses effective amination. (b) Nucleophilic ring opening of epoxide groups by NH3, generating amino alcohol functionalities and corresponding to a 1O-to-1N conversion. (c) Acid–base neutralization between edge carboxyl groups and NH3, yielding ammonium carboxylate species and corresponding to an effective 2O-to-1N conversion. These stoichiometries were used with the measured oxygen content of 3 × GO to estimate an idealized nitrogenation capacity of N/C = 0.26.
1. Effects of Synthetic Parameters on Nitrogen Incorporation.
| Reaction temp. (°C) | N/C ratio | N (at%) | Reaction time (h) | N/C ratio | N (at%) | NH3 (aq) volume(mL) | N/C ratio | N (at%) | Oxidation degree of GO (O/C) | N/C ratio | N (at%) | GO structure | N/C ratio | N (at%) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 140 | 0.13 ± 0.01 | 10.9 ± 0.2 | 6 | 0.13 ± 0.03 | 10.7 ± 1.4 | 2.5 | 0.15 ± 0.01 | 12.5 ± 0.6 | 0.41 (com-mercial GO) | 0.16 ± 0.02 | 12.8 ± 0.9 | GO (33.9 m2 g–1) | 0.16 ± 0.01 | 12.7 ± 0.5 |
| 160 | 0.14 ± 0.01 | 11.5 ± 0.2 | 12 | 0.16 ± 0.01 | 12.8 ± 0.9 | 5 | 0.16 ± 0.03 | 13.2 ± 2.4 | 0.49 (1 × GO) | 0.15 ± 0.01 | 12.3 ± 0.2 | CGO (330.7 m2 g–1) | 0.16 ± 0.02 | 12.8 ± 0.9 |
| 180 | 0.16 ± 0.01 | 12.8 ± 2.7 | 18 | 0.15 ± 0.03 | 11.9 ± 1.8 | 10 | 0.16 ± 0.02 | 12.8 ± 0.9 | 0.55 (3 × GO) | 0.24 ± 0.02 | 17.3 ± 1.3 | |||
| 190 | 0.15 ± 0.02 | 12.5 ± 0.5 | 24 | 0.15 ± 0.01 | 12.2 ± 0.5 | 20 | 0.14 ± 0.02 | 11.1 ± 1.8 | ||||||
| 200 | 0.16 ± 0.01 | 13.1 ± 1.0 |
Fixed reaction conditions: reaction time = 12 h, NH3(aq) volume = 20 mL, graphene precursor O/C = 0.41, using CGO as the graphene precursor.
Fixed reaction conditions: reaction temperature = 180 °C, NH3(aq) volume = 20 mL, graphene precursor O/C = 0.41, using CGO as the graphene precursor.
Fixed reaction conditions: reaction temperature = 180 °C, reaction time = 12 h, graphene precursor O/C = 0.41, using CGO as the graphene precursor.
Fixed reaction conditions: reaction temperature = 180 °C, reaction time = 12 h, NH3(aq) volume = 20 mL, using CGO as the graphene precursor.
Fixed reaction conditions: reaction temperature = 180 °C, reaction time = 12 h, NH3(aq) volume = 20 mL, and graphene precursor O/C = 0.41.
With this idealized nitrogenation capacity in mind, we carried out a one-factor-at-a-time study to quantify the gap between the idealized capacity and experimentally achieved nitrogen incorporation. Specifically, we independently varied kinetic-dependent factors of reaction temperature, reaction time, NH3(aq) dosage, precursor oxidation degree (O/C), and precursor surface area (accessibility/mass transport), while holding all other parameters constant. Unless otherwise specified, all parameter studies were conducted using CGO prepared from commercial GO. Only the oxidation-degree experiments employed different GO precursors with systematically varied oxidation levels. The resulting N(at%) and N/C ratios are summarized in Table . In this work, N(at%) was calculated from the EA-derived C/N/O contents by converting the weight fractions into molar amounts and normalizing only over C, N, and O, with hydrogen excluded from the calculation. This definition follows the same normalization concept commonly used for XPS-derived atomic percentages, thereby facilitating comparison with reported N(at%) values in the literature, while the values reported here represent bulk elemental compositions derived from EA rather than surface-sensitive XPS compositions. The detailed calculation method is provided in the Supporting Information. The corresponding elemental compositions (C, O, and N atomic percentages) are summarized in detail in Tables S2 and S3. It should also be noted that elemental analysis of highly hydrophilic GO-based samples may include contributions from residual physisorbed water, which can mainly affect the apparent H and O contents and, in turn, influence the normalized elemental atomic percentages. Because all EA-derived compositions in this work were obtained from samples prepared and analyzed under comparable conditions, this possible contribution does not alter the comparative trends discussed below. TPD–MS provides a useful complementary approach in this regard, because physisorbed species can be removed during thermal desorption, allowing thermally stable and chemically bound nitrogen species to be evaluated more selectively. Further quantitative TPD–MS analysis may therefore provide a more detailed understanding of chemically bound nitrogen contents and nitrogen-species distributions in highly hydrophilic GO-based materials.
We first examined the effect of reaction temperature by scanning the full practical hydrothermal window used in this work (i.e., 140, 160, 180, 190, and 200 °C), with all other reaction parameters kept constant as summarized in Table . Because 200 °C approaches the thermal-stability limit of the Teflon liner, no tests were conducted at higher temperatures. The resulting nitrogen contents remain narrowly confined to N/C = 0.13–0.16: 0.13 ± 0.01 (140 °C), 0.14 ± 0.01 (160 °C), 0.16 ± 0.01 (180 °C), 0.15 ± 0.02 (190 °C), and 0.16 ± 0.01 (200 °C) (Table ). Thus, increasing the hydrothermal temperature from 140 to 200 °C does not further increase the final nitrogen incorporation level. Because epoxide–amine reactions are known to proceed even at substantially lower temperatures, this reaction pathway is expected to be sufficiently activated within the investigated hydrothermal regime. The nearly constant N/C ratio therefore likely reflects saturation of the finite population of reactive oxygen sites rather than insufficient thermal activation. The nearly unchanged N/C ratio over 140–200 °C is therefore interpreted as a saturation behavior arising from the collective outcome of multiple nitrogenation and deoxygenation steps, rather than as evidence that all elementary nitrogenation processes are temperature-independent. Because the same nitrogenation/deoxygenation pathways are expected to operate throughout this range, the absence of a systematic change in N/C suggests that the final nitrogenation level is largely insensitive to preparation kinetics within the investigated hydrothermal regime. Deoxygenation may also occur concurrently at these temperatures, as thermal removal of oxygen functionalities from GO and amine-functionalized GO has been well-established. Importantly, oxygen removal alone, such as C–O bond cleavage from amino-alcohol intermediates, does not alter the N/C atomic ratio unless accompanied by carbon loss, because this process does not change the relative numbers of nitrogen and carbon atoms. Based on these results, 180 °C was selected as the standard temperature for subsequent comparisons. This temperature-insensitive behavior in nitrogenation contrasts with several prior reports that attribute higher nitrogen levels to elevated synthesis temperatures. ,
We next evaluated the role of reaction duration at 180 °C by varying the time from 6 to 24 h, yielding N/C = 0.13 ± 0.03 (6 h), 0.16 ± 0.01 (12 h), 0.15 ± 0.03 (18 h), and 0.15 ± 0.01 (24 h) (Table ). No systematic increase is observed with extended treatment, indicating that the final nitrogen content is likewise time-independent over this range. The resulting N/C values remain within the same narrow plateau range observed in the temperature-dependent study. This conclusion opposes the literature, suggesting that longer reaction times lead to higher nitrogen incorporation in GO. , We therefore adopted 12 h as the standard reaction duration.
The effect of nitrogen-source dosage on the nitrogenation level was evaluated by varying the amount of NH3 from 2.5 to 20 mL. The resulting nitrogen contents (Table ) were N/C = 0.15 ± 0.01 (2.5 mL), 0.16 ± 0.03 (5 mL), 0.16 ± 0.02 (10 mL), and 0.14 ± 0.02 (20 mL). Therefore, even an ∼ 8-fold increase in NH3 dosage does not increase nitrogen incorporation, and the N/C ratios remain nearly constant and similar to those obtained in the temperature- and time-dependent studies above. We fixed the NH3(aq) volume at 20 mL as an overdose condition in the subsequent experiments to ensure that ammonia availability is not a limiting variable.
By adopting a series of CGO precursors made by GO with varied oxidation degrees, such as O/C = 0.41 of commercial GO, 0.49 of 1 × GO, and 0.55 of 3 × GO, the resulting nitrogen contents are N/C = 0.16 ± 0.02, 0.15 ± 0.01, and 0.24 ± 0.02 (Table ), respectively. Importantly, only the highly oxidized precursor (O/C = 0.55) produces a clear step-change to N/C = 0.24 ± 0.02, far above the N/C ≈ 0.13–0.16 range observed across all temperature/time/NH3(aq)-dosage conditions. In addition, commercial GO (O/C = 0.41) and 1 × GO (O/C = 0.49) yield indistinguishable N/C values within error; it appears that only a significant difference in the GO oxidation degrees can lead to a meaningful shift in the N/C values. These results directly identify precursor oxidation degree as the primary factor governing the final nitrogenation level in this ammonia system.
Finally, we directly tested a widely cited hypothesis that higher surface area and more open porosity enhance accessibility/mass transport and thereby increase nitrogen incorporation. , Two precursors with the same O/C = 0.41 but drastically different surface areas, namely GO (33.9 m2 g–1) and CGO (330.7 m2 g–1), yield essentially identical nitrogen contents: N/C = 0.16 ± 0.01 (33.9 m2 g–1) vs 0.16 ± 0.02 (330.7 m2 g–1) (Table ). This quantitative control experiment demonstrates that improved accessibility/mass transport alone does not raise nitrogen incorporation when oxidation degree is matched, directly challenging the common assumption that surface area is the key lever for surpassing the nitrogenation ceiling.
Having established the distinct roles of precursor oxidation and accessibility, we next examined the structural consequences of nitrogenation using CGO prepared from commercial GO as a representative model system. CGO was selected because its distinct three-dimensional morphology enables structural retention, densification, and nitrogen distribution to be directly evaluated after vapor-phase treatment. Figure b compares nitrogenated crumpled graphene oxide (NrCGO) with the blank control treated under identical conditions in the absence of a nitrogen source, denoted as reduced crumpled graphene oxide (rCGO). NrCGO undergoes appreciable volume contraction, accompanied by an increase in tap density from 1.5 mg mL–1 to 100 mg mL–1. A small amount of condensate is observed inside the vial, indicating that NH3/H2O vapor likely condensed during the cooling stage of the preparation process; this condensation is plausibly associated with the observed densification, although the underlying mechanism remains to be clarified. SEM (Figure c–e), however, confirms that the wrinkled, crumpled framework is retained after treatment. No obvious large-scale pore formation or fragmentation characteristic of pronounced steam etching is observed at this scale. Overall, rCGO largely preserves the precursor morphology, whereas NrCGO exhibits reduced particle dimensions while retaining a wrinkled and crumpled framework rather than undergoing structural collapse.
Multiple complementary characterizations further support framework preservation and nitrogen incorporation. XRD (Figure f) shows that GO exhibits a (001) reflection at 2θ of 9.8°, while rGO and NrGO display broad (002) features near 26°, whereas both rCGO and NrCGO show no obvious layered diffraction, consistent with suppressed long-range intersheet stacking and minimal structural change upon nitrogenation. , FTIR (Figure g) reveals attenuated oxygen-related bands after treatment and the appearance of a C–N vibration at approximately 1420 cm–1 in NrCGO, providing spectroscopic evidence for nitrogen incorporation. , Raman spectra of rCGO and NrCGO (Figure h) display the characteristic D and G bands of graphitic carbon, indicating that the overall carbon framework is maintained. TGA (Figure i) provides insight into both the thermal stability and the binding nature of the incorporated nitrogen. Relative to rCGO, NrCGO exhibits a noticeably larger mass loss of 5.76 wt% in the 100–300 °C region, consistent with the decomposition of thermally unstable surface-grafted N functionalities. At higher temperatures, NrCGO continues to show a larger and more sustained weight loss than rCGO, consistent with the gradual decomposition of more stable, lattice-incorporated nitrogen species: the additional loss in 300–1000 °C is commonly associated with lattice-doped N. Overall, the larger total weight loss and lower residue of NrCGO compared with rCGO indicate that nitrogenation increases the population of reactive functionalities/defects and improves chemical reactivity, but it also reduces thermal stability to some extent.
N2 sorption/BJH analyses (Figure j,k) indicate that porosity is retained after nitrogenation, although the BET surface area decreases from 345.7 to 206.0 m2 g–1 with reduced pore volume, consistent with densification. XPS further confirms nitrogen incorporation (Figure a,b): an N 1s signal appears at approximately 400 eV, and the N/C atomic ratio reaches about 0.24. Deconvolution of the N 1s spectrum resolves components assignable to pyridinic N, pyrrolic/amine-like N, ammonium N, and oxidized N. − HAADF-STEM/EDS mapping (Figure c–f) shows homogeneous distributions of C, O, and N throughout the crumpled framework. The combination of retained porosity and abundant, homogeneously distributed nitrogen functionalities also suggests a promising potential in, for example, adsorption and gas capture functionality. In particular, amine-like nitrogen sites in our products should be highly capable of achieving enhanced CO2 affinity, , whereas protonated/cationic nitrogen sites may facilitate interactions with negatively charged species through electrostatic attraction. Additional characterization details, spectral analysis, and peak assignments are provided in Section S3 of the Supporting Information.
2.

Surface nitrogen speciation and elemental distribution of nitrogenated crumpled graphene oxide. (a) XPS survey spectra of rCGO and NrCGO, showing the emergence of the N 1s signal after vapor-phase nitrogenation. (b) High-resolution N 1s spectrum of NrCGO, deconvoluted into pyridinic, pyrrolic, ammonium, and oxidized nitrogen species. (c) HAADF-STEM image of an individual NrCGO particle, showing the retained wrinkled/crumpled morphology. (d–f) EDS elemental maps collected from the same particle, confirming the homogeneous distribution of (d) C, (e) O, and (f) N throughout the crumpled framework.
With the structural characteristics and parameter-dependent trends established, we next examined the chemical origin of the elevated nitrogen incorporation. When the results in Table are considered collectively, a clear pattern emerges: as long as the precursor oxidation degree remains unchanged, N/C stays near 0.15 despite substantial variations in reaction temperature, duration, ammonia dosage, and precursor accessibility. Compared with the idealized nitrogenation capacity (N/C = 0.20), this plateau suggests that only a subset of oxygen sites can be converted to C–N under our conditions, while certain oxygenated groups remain unreacted even under the most intensive conditions we tested. This finding is distinctive relative to most prior studies, which typically tune one kinetic parameter at a time or adopt higher-porosity architectures without independently matching precursor oxidation degree and oxygen speciation. , As a result, the unique N/C plateau can be easily masked. The increase from 1 × GO to 3 × GO is therefore particularly informative because it changes the oxygen-site population and allows us to identify which specific oxygen sites are responsible for pushing nitrogenation beyond the 0.15 plateau.
We first rule out edge carboxyl groups as the dominant contributor: their abundance is inherently constrained by the finite number of edge carbons and therefore cannot increase indefinitely even in highly oxidized 3 × GO. , The key evidence comes from nitrogenation with different nitrogen source molecules (Table and Table S4). As shown in Table , under identical conditions and using 3 × GO, the nitrogenation degree follows NH3 (N/C = 0.24) > CH3NH2 (N/C = 0.14) > ethylenediamine (N/C = 0.1 after normalization by the two nitrogen atoms in each ethylenediamine molecule), opposite to the basicity trend of ethylenediamine > CH3NH2 > NH3. If the additional sites in 3 × GO were mainly carboxyl groups, stronger bases should either increase nitrogen uptake (via acid–base reaction, which is generally much faster than nucleophilic substitution reactions) or, at a minimum, no decrease trend should be observed. Alternatively, if all these nitrogen sources were sufficiently basic to fully react with carboxyl groups, the final nitrogenation levels would be expected to converge (EDA ≈ CH3NH2 ≈ NH3). This nonconvergence may indicate that nitrogenation on these sites should alternatively be controlled by steric hindrance (ethylenediamine > CH3NH2 > NH3), rather than basicity. Such a proposal is most consistent with the sterically sensitive nature of epoxide groups, where small NH3 can access and react more effectively than other bulkier amines, following the order of NH3 > CH3NH2 > ethylenediamine. Therefore, the enhanced nitrogenation observed for highly oxidized GO is attributed to the increased abundance of basal-plane epoxide sites in 3 × GO relative to 1 × GO. This interpretation is further supported by high-resolution C 1s XPS analysis of the three GO precursors (Figure S1 and Table S5). The C–O–C component increases from 28.9% in commercial GO and 30.8% in 1 × GO to 38.9% in 3 × GO, whereas the OC–O component remains nearly unchanged (3.6–4.4%) and neither C–OH nor CO shows a corresponding monotonic increase.
2. Effect of Nitrogen Source and Post-Annealing Treatment Temperature on the Nitrogenated Crumpled Graphene Oxide.
| N-source | N/C ratio | N (at%) | Annealing temp. (°C) | N/C ratio | N (at%) |
|---|---|---|---|---|---|
| NH3 | 0.24 ± 0.02 | 17.3 ± 1.3 | 300 | 0.20 | 14.0 |
| CH3NH2 | 0.14 ± 0.01 | 11.4 ± 0.8 | 600 | 0.11 | 8.54 |
| EDA | 0.20 ± 0.01 | 15.6 ± 0.4 | 900 | 0.07 | 5.72 |
This interpretation is further supported by the comparison between the theoretical and experimentally achieved nitrogenation levels. While lower-oxidation GO (O/C = 0.41) exhibits a calculated idealized nitrogenation capacity of only N/C ≈ 0.20 and experimentally remains near the commonly observed N/C ≈ 0.15 plateau, highly oxidized 3 × GO shows both a higher idealized capacity (N/C ≈ 0.26) and a substantially higher experimentally achieved nitrogenation level (N/C ≈ 0.24). Previous studies on highly oxidized GO systems showed that additional oxidation preferentially increases basal-plane epoxide populations after edge carboxyl groups approach saturation. Such behavior is therefore consistent with the proposal that epoxide groups are the dominant reactive sites responsible for pushing nitrogenation beyond the conventional N/C ≈ 0.15 plateau. Comparison with reported nitrogenated graphene systems (Table S6) further shows that conventional ammonia-based hydrothermal nitrogenation of GO typically results in relatively modest nitrogen incorporation levels. Although alternative nitrogen sources and harsher nitrogenation methodologies can also achieve high nitrogen incorporation, the present results highlight precursor oxidation degree as a critical yet previously underemphasized parameter for increasing nitrogen incorporation. Such oxygen-site-guided considerations may also provide useful insight for further optimization of other graphene nitrogenation methodologies. The resulting oxygen-site-dependent nitrogenation mechanism is summarized in Scheme .
2. Schematic Summary of Oxygen-Site-Dependent Nitrogenation of Crumpled Graphene Oxide (CGO) .

a Nitrogen incorporation is weakly affected by reaction temperature, time, and ammonia dosage but is strongly governed by the precursor oxidation degree. Increasing the O/C ratio, particularly the abundance of epoxide functionalities, provides additional reactive sites and enables the formation of highly N-doped NrCGO.
Having identified epoxide-rich oxygen chemistry as the origin of enhanced nitrogen incorporation, we next examined how the initially introduced nitrogen functionalities evolved during thermal treatment. Elemental analysis of NrCGO annealed at 300, 600, and 900 °C shows a progressive decrease in nitrogen content with increasing temperature (Table and Table S4), indicating that the retained nitrogen level can be tuned through the annealing conditions. The corresponding FTIR spectra of NrCGO after thermal treatment at 300, 600, and 900 °C are provided in Figure S2. This trend indicates the progressive loss of thermally unstable nitrogen species with increasing annealing temperature, while the nature of the retained nitrogen configurations was further examined by TPD–MS and XPS.
We therefore employed TPD–MS as a bulk-sensitive speciation probe and interpreted the NH3/HCN/N2/NO evolution (Figure a-d) using the mechanistic assignment reported by Yoshii et al. Peak deconvolution (Figure e) shows that NrCGO-600 °C is dominated by pyridinic N (57%), followed by pyrrolic N (32%) and a smaller fraction of graphitic N (10%), with oxidized nitrogen negligible (approximately 1%). These results were also confirmed by XPS quantification (Figure e). This pyridinic/pyrrolic bias indicates that annealing at elevated temperatures preferentially stabilizes defect-associated sp2 nitrogen (commonly assigned to vacancy/hole boundaries or nonhexagonal reconstructions), whereas graphitic N substitution within a more intact basal-plane hexagonal network remains a minority under these conditions. Notably, the pyridinic-N fraction observed here is comparatively high relative to many reported nitrogen-doped graphene/carbon systems (Table S7), which may be advantageous for future catalytic applications. ,,
3.

Bulk-sensitive quantification of nitrogen configurations in thermally treated NrCGO. (a–d) TPD profiles of desorbed nitrogen-containing species from NrCGO-600 °C, including (a) NH3, (b) HCN, (c) N2, and (d) NO. The deconvoluted signals are assigned to pyrrolic, pyridinic, graphitic, and oxidized nitrogen species based on their characteristic desorption behavior. (e) Quantitative distribution of nitrogen configurations derived from TPD analysis, compared with the surface-sensitive XPS results.
Although pyridinic and pyrrolic nitrogen are frequently discussed as “edge-type” motifs, , this assignment does not necessarily imply only the outer perimeter of a graphene sheet. In our system, the major nitrogenation pathway originates from basal-plane epoxide mediated by NH3-driven ring opening to introduce a local O–C–C–N motif. During annealing and deoxygenation, effective removal of oxygen in every O–C–C–N motif may result in a large quantity of local defects and thus lead to a reconstruction into nonhexagonal rings and vacancy-like defects, which effectively yield edge-like boundaries within the sheet. In addition, hydrothermal steam etching may also provide a secondary pathway for generating defect-associated sites by preferentially removing defective sp3-rich domains and creating nanoscale pores and internal edges. Such steam-induced defects may therefore contribute to the subsequent formation of edge-like nitrogen configurations. Consequently, the pyridinic-rich signature observed here can be rationalized as arising from both basal plane functional groups and defect-associated reconstruction, not only from macroscopic outer edges.
As a preliminary assessment of functional activity, NrCGO-600 °C was evaluated for the oxygen reduction reaction. Relative to rCGO, NrCGO-600 °C exhibits a 0.08 V positive shift in ORR onset potential (Figure S3), indicating that the nitrogen configurations generated through nitrogenation and subsequent annealing are electrocatalytically active. This result provides functional validation of the incorporated nitrogen species. It is worth noting that ORR performance is governed not only by the total nitrogen content but also by the resulting nitrogen configuration and electrical conductivity. ,, Increasing nitrogen incorporation can raise the population of catalytically relevant sites, particularly pyridinic-N-associated defect sites, , while excessive defect formation may disrupt the conjugated carbon network and impede electron transport. , In addition, different nitrogen configurations influence the electronic structure of graphene differently. Graphitic N is generally more compatible with preservation of the conjugated carbon framework and efficient charge transport, whereas pyridinic and pyrrolic N are commonly found at edge- and defect-associated sites that can be catalytically relevant. Therefore, achieving high ORR activity requires a balance among nitrogen content, nitrogen configuration, and electrical conductivity. Further optimization of this balance and selective control of nitrogen configurations using the present nitrogenation strategy remain subjects for future study.
In summary, we have systematically decoupled the chemical and physical factors governing the ammonia-based hydrothermal nitrogenation of graphene oxide, revealing that the widely observed ″nitrogenation ceiling″ in this commonly used system is neither a kinetic nor a mass-transport limitation, but is fundamentally an oxygen-site-limited phenomenon. Through rigorous control experiments using a vapor-phase hydrothermal platform and 3D crumpled architectures, we demonstrated that conventional parametersincluding reaction temperature (up to 200 °C), duration, ammonia dosage, and even active-site accessibilityexert negligible influence on final nitrogen levels when the precursor oxidation degree is held constant. Our findings establish that the population of basal-plane epoxide groups serves as the primary gateway for nitrogen incorporation. By modulating the oxidation degree of the graphene precursor, we achieved a nitrogen content of 17.3 at% (N/C = 0.24), among the highest values reported for ammonia-based hydrothermal nitrogenation of GO. Comprehensive characterization via TPD–MS and XPS confirms that these surface-grafted nitrogen species can be effectively converted into stable lattice-doped motifspredominantly pyridinic nitrogen (57%). Possible additional carbon loss and framework reconstruction during hydrothermal treatment may also affect the final N/C ratio, and their quantitative contributions remain to be clarified in future studies. It should be noted that this conclusion does not exclude the possibility that higher-temperature treatments, different nitrogen sources, or other nitrogenation methodologies rather than hydrothermal-based treatment may lead to different kinetics of nitrogen incorporation. Rather, this work provides a fundamental guideline in the design of highly heteroatom-doped graphene materials via hydrothermal systems, shifting the focus from kinetic optimization to the precise engineering of precursor oxygen speciation. The oxygen-guided nitrogenation strategy presented here not only offers practical knowledge to pursue even higher nitrogen doping levels, but also may open further opportunities for developing high-performance graphene-based materials for energy storage and catalysis.
Supplementary Material
Acknowledgments
We acknowledge the financial support from the National Science and Technology Council, Taiwan (Grant number: NSTC 112-2628-M-110-001-MY4), JST PRESTO (Grant number: JPMJPR23QA), and JSPS-KAKENHI (Grant number: 26KJ0572). The authors also appreciate the partial financial support from the Green Hydrogen Research Center (GHRC) at National Sun Yat-sen University, Taiwan. We also thank Prof. Chien-Liang Lee for his assistance with the ORR electrochemical measurements.
The data supporting the findings of this study are available in the main manuscript and the Supporting Information.
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jpclett.6c02507.
Experimental procedures for the synthesis of GO, CGO, rCGO, and NrCGO, including vapor-phase hydrothermal nitrogenation and ORR electrode preparation (Section S1); thermodynamic and mechanistic considerations for representative nitrogenation pathways on graphene oxide, including model reactions and calculated thermodynamic parameters (Section S2 and Table S1); supplementary structural and surface characterizations of rCGO and NrCGO by XRD, FTIR, Raman, TGA, BET/BJH, and XPS analyses (Section S3); elemental compositions of nitrogenated crumpled graphene synthesized under different hydrothermal reaction conditions determined by elemental analysis (Table S2); elemental compositions of NrCGO synthesized using graphene oxide precursors with different oxidation degrees and structures determined by elemental analysis (Table S3); elemental compositions of NrCGO synthesized using different nitrogen sources and after annealing at different temperatures determined by elemental analysis (Table S4); abundance percentages of different functional groups on the surfaces of commercial GO, 1 × GO, and 3 × GO based on high-resolution C 1s XPS peak deconvolution (Table S5); comparison of nitrogen incorporation levels among reported nitrogenated graphene-based materials (Table S6); quantification of nitrogen configurations by TPD–MS and comparison with reported nitrogen-doped graphene/carbon materials (Table S7); high-resolution C 1s XPS spectra and peak deconvolution of 3 × GO, 1 × GO, and commercial GO (Figure S1); comparative FTIR spectra of NrCGO and thermally reduced NrCGO samples treated at 300, 600, and 900 °C (Figure S2); LSV curves of rCGO and NrCGO-600 °C in O2-saturated 0.1 M KOH (Figure S3) (DOCX)
Yen-Ting Liu, Guan-Ting Chen, Hong-Han Huang, and Hong-Yin Chen ensured that the original data and materials were used, performed the experiments, and analyzed the data. Guan-Ting Chen conducted the majority of the experiments and wrote the first draft. Shunsuke Shimizu, Takeharu Yoshii, and Hirotomo Nishihara performed the TPD–MS measurements and contributed to the interpretation of the TPD–MS results. Virgilio D. Ebajo, Jr. contributed to the mechanistic interpretation and provided critical revisions to the mechanism discussion. Chun-Hu Chen conceived and designed the research, supervised the project, secured funding, and provided critical revisions to the manuscript. All authors discussed the results, contributed to the whole process of the manuscript writing, and approved its submission.
The authors used OpenAI’s ChatGPT to assist in improving the English grammar and writing clarity of the manuscript. The authors reviewed and verified all content for accuracy and take full responsibility for the final version of the text.
The authors declare no competing financial interest.
References
- Usachov D., Vilkov O., Grüneis A., Haberer D., Fedorov A., Adamchuk V. K., Preobrajenski A. B., Dudin P., Barinov A., Oehzelt M.. et al. Nitrogen-Doped Graphene: Efficient Growth, Structure, and Electronic Properties. Nano Lett. 2011;11:5401–5407. doi: 10.1021/nl2031037. [DOI] [PubMed] [Google Scholar]
- Sforzini J., Hapala P., Franke M., van Straaten G., Stöhr A., Link S., Soubatch S., Jelínek P., Lee T. L., Starke U.. et al. Structural and Electronic Properties of Nitrogen-Doped Graphene. Phys. Rev. Lett. 2016;116:126805. doi: 10.1103/PhysRevLett.116.126805. [DOI] [PubMed] [Google Scholar]
- Panchakarla L. S., Subrahmanyam K. S., Saha S. K., Govindaraj A., Krishnamurthy H. R., Waghmare U. V., Rao C. N. R.. Synthesis, Structure, and Properties of Boron- and Nitrogen-Doped Graphene. Adv. Mater. 2009;21:4726–4730. doi: 10.1002/adma.200901285. [DOI] [Google Scholar]
- Wang H., Xie M., Thia L., Fisher A., Wang X.. Strategies on the Design of Nitrogen-Doped Graphene. J. Phys. Chem. Lett. 2014;5:119–125. doi: 10.1021/jz402416a. [DOI] [PubMed] [Google Scholar]
- Croxall M. P., Lawrence R. T., Ghosh Biswas R., Soong R., Simpson A. J., Goh M. C.. Improved Photocatalytic Performance of TiO2-Nitrogen-Doped Graphene Quantum Dot Composites Mediated by Heterogeneous Interactions. J. Phys. Chem. Lett. 2024;15:3653–3657. doi: 10.1021/acs.jpclett.4c00335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Perazzolo V., Durante C., Pilot R., Paduano A., Zheng J., Rizzi G. A., Martucci A., Granozzi G., Gennaro A.. Nitrogen and sulfur doped mesoporous carbon as metal-free electrocatalysts for the in situ production of hydrogen peroxide. Carbon. 2015;95:949–963. doi: 10.1016/j.carbon.2015.09.002. [DOI] [Google Scholar]
- Wang N., Lu B., Li L., Niu W., Tang Z., Kang X., Chen S.. Graphitic Nitrogen Is Responsible for Oxygen Electroreduction on Nitrogen-Doped Carbons in Alkaline Electrolytes: Insights from Activity Attenuation Studies and Theoretical Calculations. ACS Catal. 2018;8:6827–6836. doi: 10.1021/acscatal.8b00338. [DOI] [Google Scholar]
- Tang R., Aziz A., Yu W., Pan Z.-Z., Nishikawa G., Yoshii T., Nomura K., Taylor E. E., Stadie N. P., Inoue K.. et al. Prominent Structural Dependence of Quantum Capacitance Unraveled by Nitrogen-Doped Graphene Mesosponge. Small. 2024;20:2308066. doi: 10.1002/smll.202308066. [DOI] [PubMed] [Google Scholar]
- Zhai Z., Shen H., Chen J., Li X., Jiang Y.. Direct growth of nitrogen-doped graphene films on glass by plasma-assisted hot filament CVD for enhanced electricity generation. J. Mater. Chem. A. 2019;7:12038–12049. doi: 10.1039/C9TA01768B. [DOI] [Google Scholar]
- Li X., Wang H., Robinson J. T., Sanchez H., Diankov G., Dai H.. Simultaneous Nitrogen Doping and Reduction of Graphene Oxide. J. Am. Chem. Soc. 2009;131:15939–15944. doi: 10.1021/ja907098f. [DOI] [PubMed] [Google Scholar]
- Du D., Li P., Ouyang J.. Nitrogen-Doped Reduced Graphene Oxide Prepared by Simultaneous Thermal Reduction and Nitrogen Doping of Graphene Oxide in Air and Its Application as an Electrocatalyst. ACS Appl. Mater. Interfaces. 2015;7:26952–26958. doi: 10.1021/acsami.5b07757. [DOI] [PubMed] [Google Scholar]
- Zhou Y., Nishina Y.. Bottom-up synthesis of nitrogen-doped nanocarbons by a combination of metal catalysis and a solution plasma process. Nanoscale Adv. 2020;2:4417–4420. doi: 10.1039/D0NA00327A. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wen Y., Rufford T. E., Hulicova-Jurcakova D., Zhu X., Wang L.. Structure Control of Nitrogen-Rich Graphene Nanosheets Using Hydrothermal Treatment and Formaldehyde Polymerization for Supercapacitors. ACS Appl. Mater. Interfaces. 2016;8:18051–18059. doi: 10.1021/acsami.6b04572. [DOI] [PubMed] [Google Scholar]
- Wang M., Wang J., Hou Y., Shi D., Wexler D., Poynton S. D., Slade R. C. T., Zhang W., Liu H., Chen J.. N-Doped Crumpled Graphene Derived from Vapor Phase Deposition of PPy on Graphene Aerogel as an Efficient Oxygen Reduction Reaction Electrocatalyst. ACS Appl. Mater. Interfaces. 2015;7:7066–7072. doi: 10.1021/acsami.5b01025. [DOI] [PubMed] [Google Scholar]
- Kao W.-Y., Chen W.-Q., Chiu Y.-H., Ho Y.-H., Chen C.-H.. General Solvent-dependent Strategy toward Enhanced Oxygen Reduction Reaction in Graphene/Metal Oxide Nanohybrids: Effects of Nitrogen-containing Solvent. Sci. Rep. 2016;6:37174. doi: 10.1038/srep37174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wen Z., Wang X., Mao S., Bo Z., Kim H., Cui S., Lu G., Feng X., Chen J.. Crumpled Nitrogen-Doped Graphene Nanosheets with Ultrahigh Pore Volume for High-Performance Supercapacitor. Adv. Mater. 2012;24:5610–5616. doi: 10.1002/adma.201201920. [DOI] [PubMed] [Google Scholar]
- Zou Y., Kinloch I. A., Dryfe R. A. W.. Nitrogen-doped and crumpled graphene sheets with improved supercapacitance. J. Mater. Chem. A. 2014;2:19495–19499. doi: 10.1039/C4TA04076G. [DOI] [Google Scholar]
- Yeh T.-F., Chen S.-J., Yeh C.-S., Teng H.. Tuning the Electronic Structure of Graphite Oxide through Ammonia Treatment for Photocatalytic Generation of H2 and O2 from Water Splitting. J. Phys. Chem. C. 2013;117:6516–6524. doi: 10.1021/jp312613r. [DOI] [Google Scholar]
- Chen C.-H., Hu S., Shih J.-F., Yang C.-Y., Luo Y.-W., Jhang R.-H., Chiang C.-M., Hung Y. Jr.. Effective Synthesis of Highly Oxidized Graphene Oxide That Enables Wafer-scale Nanopatterning: Preformed Acidic Oxidizing Medium Approach. Sci. Rep. 2017;7:3908. doi: 10.1038/s41598-017-04139-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Liu Y.-P., Tang W.-R., Liu Y.-T., Devi Y., Chiou K., Chen C.-H.. Cryogenic synthesis of crumpled graphene Spheres: Advancing scalability, stability and preserved chemical versatility. Carbon. 2025;242:120435. doi: 10.1016/j.carbon.2025.120435. [DOI] [Google Scholar]
- Guo H.-L., Su P., Kang X., Ning S.-K.. Synthesis and characterization of nitrogen-doped graphene hydrogels by hydrothermal route with urea as reducing-doping agents. J. Mater. Chem. A. 2013;1:2248–2255. doi: 10.1039/C2TA00887D. [DOI] [Google Scholar]
- Long D., Li W., Ling L., Miyawaki J., Mochida I., Yoon S.-H.. Preparation of Nitrogen-Doped Graphene Sheets by a Combined Chemical and Hydrothermal Reduction of Graphene Oxide. Langmuir. 2010;26:16096–16102. doi: 10.1021/la102425a. [DOI] [PubMed] [Google Scholar]
- Lu Y.-F., Lo S.-T., Lin J.-C., Zhang W., Lu J.-Y., Liu F.-H., Tseng C.-M., Lee Y.-H., Liang C.-T., Li L.-J.. Nitrogen-Doped Graphene Sheets Grown by Chemical Vapor Deposition: Synthesis and Influence of Nitrogen Impurities on Carrier Transport. ACS Nano. 2013;7:6522–6532. doi: 10.1021/nn402102y. [DOI] [PubMed] [Google Scholar]
- Saidi W. A.. Oxygen Reduction Electrocatalysis Using N-Doped Graphene Quantum-Dots. J. Phys. Chem. Lett. 2013;4:4160–4165. doi: 10.1021/jz402090d. [DOI] [Google Scholar]
- Ebajo V. D., Santos C. R. L., Alea G. V., Lin Y. A., Chen C.-H.. Regenerable Acidity of Graphene Oxide in Promoting Multicomponent Organic Synthesis. Sci. Rep. 2019;9:15579. doi: 10.1038/s41598-019-51833-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ding Y.-q., Cui Y.-z., Li T.-d.. New Views on the Reaction of Primary Amine and Aldehyde from DFT Study. J. Phys. Chem. A. 2015;119:4252–4260. doi: 10.1021/acs.jpca.5b02186. [DOI] [PubMed] [Google Scholar]
- Compton O. C., Dikin D. A., Putz K. W., Brinson L. C., Nguyen S. T.. Electrically Conductive “Alkylated” Graphene Paper via Chemical Reduction of Amine-Functionalized Graphene Oxide Paper. Adv. Mater. 2010;22:892–896. doi: 10.1002/adma.200902069. [DOI] [PubMed] [Google Scholar]
- Ferré-Pujol P., Obata S., Raya J., Bianco A., Katayama H., Kato T., Nishina Y.. Reversible chemical modifications of graphene oxide for enhanced viral capture and release in water. Carbon. 2025;234:120015. doi: 10.1016/j.carbon.2025.120015. [DOI] [Google Scholar]
- Wu J., Zhang D., Wang Y., Hou B.. Electrocatalytic activity of nitrogen-doped graphene synthesized via a one-pot hydrothermal process towards oxygen reduction reaction. J. Power Sources. 2013;227:185–190. doi: 10.1016/j.jpowsour.2012.11.074. [DOI] [Google Scholar]
- Hassan F. M., Chabot V., Li J., Kim B. K., Ricardez-Sandoval L., Yu A.. Pyrrolic-structure enriched nitrogen doped graphene for highly efficient next generation supercapacitors. J. Mater. Chem. A. 2013;1:2904–2912. doi: 10.1039/c2ta01064j. [DOI] [Google Scholar]
- Li J., Li X., Zhao P., Lei D. Y., Li W., Bai J., Ren Z., Xu X.. Searching for magnetism in pyrrolic N-doped graphene synthesized via hydrothermal reaction. Carbon. 2015;84:460–468. doi: 10.1016/j.carbon.2014.12.024. [DOI] [Google Scholar]
- Song L., Luo Q., Zhao F., Li Y., Lin H., Qu L. T., Zhang Z. P.. Dually functional, N-doped porous graphene foams as counter electrodes for dye-sensitized solar cells. Phys. Chem. Chem. Phys. 2014;16:21820–21826. doi: 10.1039/C4CP03443K. [DOI] [PubMed] [Google Scholar]
- Wang T., Wang L. X., Wu D. L., Xia W., Zhao H. Y., Jia D. Z.. Hydrothermal synthesis of nitrogen-doped graphene hydrogels using amino acids with different acidities as doping agents. J. Mater. Chem. A. 2014;2:8352–8361. doi: 10.1039/C4TA00170B. [DOI] [Google Scholar]
- Vacchi I. A., Spinato C., Raya J., Bianco A., Ménard-Moyon C.. Chemical reactivity of graphene oxide towards amines elucidated by solid-state NMR. Nanoscale. 2016;8:13714–13721. doi: 10.1039/C6NR03846H. [DOI] [PubMed] [Google Scholar]
- Ortiz-Anaya I., Obata S., Nishina Y.. Tunable interlayer distance in graphene oxide through alkylamine surface coverage and chain length. J. Colloid Interface Sci. 2025;695:137727. doi: 10.1016/j.jcis.2025.137727. [DOI] [PubMed] [Google Scholar]
- Sandoval S., Kumar N., Sundaresan A., Rao C. N. R., Fuertes A., Tobias G.. Enhanced Thermal Oxidation Stability of Reduced Graphene Oxide by Nitrogen Doping. Chem. Eur. J. 2014;20:11999–12003. doi: 10.1002/chem.201403833. [DOI] [PubMed] [Google Scholar]
- Zhang H., Kuila T., Kim N. H., Yu D. S., Lee J. H.. Simultaneous reduction, exfoliation, and nitrogen doping of graphene oxide via a hydrothermal reaction for energy storage electrode materials. Carbon. 2014;69:66–78. doi: 10.1016/j.carbon.2013.11.059. [DOI] [Google Scholar]
- Li X., Tang T., Li M., He X.. Photochemical doping of graphene oxide thin films with nitrogen for electrical conductivity improvement. J. Mater. Sci.: Mater. Electron. 2015;26:1770–1775. doi: 10.1007/s10854-014-2606-7. [DOI] [Google Scholar]
- Ajravat K., Rajput S., Brar L. K.. Microwave assisted hydrothermal synthesis of N doped graphene for supercapacitor applications. Diam. Relat. Mater. 2022;129:109373. doi: 10.1016/j.diamond.2022.109373. [DOI] [Google Scholar]
- Yeh C.-N., Huang H., Lim A. T. O., Jhang R.-H., Chen C.-H., Huang J.. Binder-free graphene oxide doughs. Nat. Commun. 2019;10:422. doi: 10.1038/s41467-019-08389-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Luo J., Jang H. D., Sun T., Xiao L., He Z., Katsoulidis A. P., Kanatzidis M. G., Gibson J. M., Huang J.. Compression and Aggregation-Resistant Particles of Crumpled Soft Sheets. ACS Nano. 2011;5:8943–8949. doi: 10.1021/nn203115u. [DOI] [PubMed] [Google Scholar]
- Bosch-Navarro C., Coronado E., Martí-Gastaldo C.. Controllable coverage of chemically modified graphene sheets with gold nanoparticles by thermal treatment of graphite oxide with N,N-dimethylformamide. Carbon. 2013;54:201–207. doi: 10.1016/j.carbon.2012.11.027. [DOI] [Google Scholar]
- Brzhezinskaya M., Kapitanova O. O., Kononenko O. V., Koveshnikov S., Korepanov V., Roshchupkin D.. Large-scalable graphene oxide films with resistive switching for non-volatile memory applications. J. Alloys Compd. 2020;849:156699. doi: 10.1016/j.jallcom.2020.156699. [DOI] [Google Scholar]
- Das N. K., Santra S., Naik P. K., Vasa M. S., Raj R., Bose S., Banerjee T.. Evaluation of Thermophysical Properties and Thermal Performance of Amine-Functionalized Graphene Oxide/Deep Eutectic Solvent Nanofluids as Heat-Transfer Media for Desalination Systems. ACS Sustainable Chem. Eng. 2023;11:5376–5389. doi: 10.1021/acssuschemeng.2c06325. [DOI] [Google Scholar]
- Hsiao M.-C., Liao S.-H., Yen M.-Y., Liu P.-I., Pu N.-W., Wang C.-A., Ma C.-C. M.. Preparation of Covalently Functionalized Graphene Using Residual Oxygen-Containing Functional Groups. ACS Appl. Mater. Interfaces. 2010;2:3092–3099. doi: 10.1021/am100597d. [DOI] [PubMed] [Google Scholar]
- Yoshii T., Nishikawa G., Prasad V. K., Shimizu S., Kawaguchi R., Tang R., Chida K., Sato N., Sakamoto R., Takatani K.. et al. Quantitative and qualitative analysis of nitrogen species in carbon at the ppm level. Chem. 2024;10:2450–2463. doi: 10.1016/j.chempr.2024.03.029. [DOI] [Google Scholar]
- Guo D., Shibuya R., Akiba C., Saji S., Kondo T., Nakamura J.. Active sites of nitrogen-doped carbon materials for oxygen reduction reaction clarified using model catalysts. Science. 2016;351:361–365. doi: 10.1126/science.aad0832. [DOI] [PubMed] [Google Scholar]
- Pels J. R., Kapteijn F., Moulijn J. A., Zhu Q., Thomas K. M.. Evolution of nitrogen functionalities in carbonaceous materials during pyrolysis. Carbon. 1995;33:1641–1653. doi: 10.1016/0008-6223(95)00154-6. [DOI] [Google Scholar]
- Song C., Liu M., Ye W., Liu Y., Zhang H., Lu R., Zhang S.. Nitrogen-Containing Porous Carbon for Highly Selective and Efficient CO2 Capture. Energy Fuels. 2019;33:12601–12609. doi: 10.1021/acs.energyfuels.9b03203. [DOI] [Google Scholar]
- Li P. Z., Zhao Y. L.. Nitrogen-Rich Porous Adsorbents for CO2 Capture and Storage. Chem. Asian J. 2013;8:1680–1691. doi: 10.1002/asia.201300121. [DOI] [PubMed] [Google Scholar]
- Kimura R., Ferré-Pujol P., Nishina Y.. Grafting-through functionalization of graphene oxide with cationic polymers for enhanced adsorption of anionic dyes and viruses. Carbon. 2025;238:120296. doi: 10.1016/j.carbon.2025.120296. [DOI] [Google Scholar]
- Lerf A., He H., Forster M., Klinowski J.. Structure of Graphite Oxide Revisited. J. Phys. Chem. B. 1998;102:4477–4482. doi: 10.1021/jp9731821. [DOI] [Google Scholar]
- Yuan J., Yang X.-B., Wang H.-F., Cao Y., Wang H., Yang G., Yu H.. Pyridinic-N Dominates ORR Performance: Decoding Configuration-Sensitive Electrocatalysis of N-Doped Graphene Quantum Dots. ACS Catal. 2025;15:18112–18122. doi: 10.1021/acscatal.5c04899. [DOI] [Google Scholar]
- Wu Y., Jiang Z., Yu Z., Zhang S., Xiang W., Zheng G., Shu C., Li Z., Zhou T., Zhu B.. et al. Interfused Graphene Fiber Membranes Enable Volumetric Electron-Transfer Advanced Oxidation via Interlayer Site Activation. Angew. Chem., Int. Ed. 2026;65:e6024780. doi: 10.1002/anie.6024780. [DOI] [PubMed] [Google Scholar]
- Lai L., Potts J. R., Zhan D., Wang L., Poh C. K., Tang C., Gong H., Shen Z., Lin J., Ruoff R. S.. Exploration of the active center structure of nitrogen-doped graphene-based catalysts for oxygen reduction reaction. Energy Environ. Sci. 2012;5:7936–7942. doi: 10.1039/c2ee21802j. [DOI] [Google Scholar]
- Yokwana K., Ntsendwana B., Nxumalo E. N., Mhlanga S. D.. Recent advances in nitrogen-doped graphene oxide nanomaterials: Synthesis and applications in energy storage, sensor electrochemical applications and water treatment. J. Mater. Res. 2023;38:3239–3263. doi: 10.1557/s43578-023-01070-1. [DOI] [Google Scholar]
- Han T. H., Huang Y.-K., Tan A. T. L., Dravid V. P., Huang J.. Steam Etched Porous Graphene Oxide Network for Chemical Sensing. J. Am. Chem. Soc. 2011;133:15264–15267. doi: 10.1021/ja205693t. [DOI] [PubMed] [Google Scholar]
- Kim H. W., Bukas V. J., Park H., Park S., Diederichsen K. M., Lim J., Cho Y. H., Kim J., Kim W., Han T. H.. et al. Mechanisms of Two-Electron and Four-Electron Electrochemical Oxygen Reduction Reactions at Nitrogen-Doped Reduced Graphene Oxide. ACS Catal. 2020;10:852–863. doi: 10.1021/acscatal.9b04106. [DOI] [Google Scholar]
- Zhang Y., Ge J., Wang L., Wang D., Ding F., Tao X., Chen W.. Manageable N-doped Graphene for High Performance Oxygen Reduction Reaction. Sci. Rep. 2013;3:2771. doi: 10.1038/srep02771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jiang L., van Dijk B., Wu L., Maheu C. m., Hofmann J. P., Tudor V., Koper M. T. M., Hetterscheid D. G. H., Schneider G. g. F.. Predoped Oxygenated Defects Activate Nitrogen-Doped Graphene for the Oxygen Reduction Reaction. ACS Catal. 2022;12:173–182. doi: 10.1021/acscatal.1c03662. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wang B., Tsetseris L., Pantelides S. T.. Introduction of nitrogen with controllable configuration into graphene via vacancies and edges. J. Mater. Chem. A. 2013;1:14927–14934. doi: 10.1039/c3ta13610h. [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting the findings of this study are available in the main manuscript and the Supporting Information.
