Room-temperature generation and transformation of graphitic CQDs into mixed sp²–sp³ hybridized carbon allotropes from liquid oxygenates

Abstract

The synthesis of nanocrystalline sp²–sp³ hybrid structures holds immense promise for advancing materials science, but conventional synthetic methods, which rely on graphite as a precursor, are economically and energetically impractical. In this study, we report a room-temperature strategy for generating nanocrystalline sp²–sp³ hybridized carbon allotropes from liquid oxygenates via graphitic CQDs intermediates. A range of liquid oxygenates including acetic acid, formic acid, acetone, acetaldehyde, ethanol, glycolic acid, acrylic acid, glycine, and glucose were employed as carbon sources. Under controlled electrochemical conditions that are conducive to an abundance of excess electrons on negatively charged metal nanoparticles (NPs), C₂ radicals are generated from these liquid oxygenates. Experimental results provide evidence of the formation of linear acetylenic carbon intermediates and graphitic CQDs in the electrolyte and on the cathode, alongside the transformation of CQDs into nanocrystalline sp²–sp³ hybrid carbon on the electrode. Complementary DFT studies support a mechanism wherein C₂ radicals align and bond into linear acetylenic carbon chain through intermolecular interactions, rather than strong adsorption to the surface. Meanwhile, the sp²-to-sp³ transformation is facilitated by the negatively charged metal nanoparticles, with the degree of transformation being dependent on the crystallographic structure of the metal (e.g. Ag > Bi). This approach provides not only a simple and energy-efficient synthesis route, but also a platform with potential to revolutionize the fabrication of advanced carbon materials by optimizing each step of the formation process through the interaction between negatively charged metal nanoparticles and liquid oxygenates.

---

Introduction

A carbon configuration having two or more hybridization states is desirable because it can form carbon materials with extraordinary properties and novel function combination that can revolutionize creation of new advanced and functional materials. Unlike the well-known diamond-like carbon (DLC), which is sp²–sp³ hybrid amorphous structure, mixed nanocrystalline sp²- and sp³-hybridized carbons are advantageous as they combine exceptional mechanical strength with excellent electrical properties.^1–5 For example, Li et al.⁶ synthesized a composite composed of nanodiamond incoherently embedded in disordered multilayer graphene, which exhibits a combination of ultrahigh hardness and strength and excellent electrical conductivity. These properties can be varied by modifying the ratio of sp² to sp³ in the carbon structures. To demonstrate, by decreasing such the ratio, the hardness of sp²–sp³ hybrid carbon structures significantly enhances.⁵ Moreover, their electronic properties can be manipulated by a tunable band gap of carbon phases relating to the sp² to sp³ ratio.^3,4 With the phase engineering to transform one phase to another phase, sp²–sp³ hybrid carbon structures with broad range of properties can be achieved.⁷

In general, to synthesize crystalline sp²–sp³ hybrid carbon structures, it requires sp² carbon forms such as graphitic carbon as starting raw material. However, because large mechanical strain is needed in order to initiate and sustain crystal rearrangement, it typically requires high pressures and temperatures to activate the sp²-to-sp³ transformation especially in three dimensional bulk systems, such as graphite,^7–11 leading to the difficulty in synthesizing sp²–sp³ hybrid carbon structures. In literature, the transformation of graphite into hexagonal diamond was naturally formed in meteor impacts under extremely high pressure and elevated temperature.^8,12 Furthermore, formation of sp³ nodes in sp²-hybridized glassy carbon via local buckling of graphene sheets could be induced only at high temperatures and pressures.¹³ More recently, a novel material, Gradia, has been synthesized using sp² hybrid carbon precursor under high temperatures and pressures. It contains both graphite-like sp² and diamond-like sp³ structural units.¹⁴

In contrast to graphitic carbon, liquid oxygenates not only exist abundantly in nature but are also readily manufactured on an industrial scale, rendering them high cost-effective and ideal raw materials for producing nanocrystalline sp²–sp³ hybridized carbon. Although liquid oxygenates have been utilized as carbon sources for generating various carbon allotropes such as graphene, diamond, and carbon nanotubes,^15–19 there is a scarcity of publications detailing the direct synthesis of nanocrystalline sp²–sp³ hybridized carbon from these sources. Moreover, existing procedures often entail substantial energy consumption. For instance, Yolshina et al.²⁰ demonstrated the synthesis of a hierarchically ordered graphene-nanodiamond film by reacting D-glucose with a molten metallic zinc catalyst in an environment of molten chlorides of alkali metals in air, requiring a high temperature exceeding 700 °C. Hence, there is a pressing need for a low energy and facile synthetic process utilizing liquid oxygenates as carbon sources for the generation of nanocrystalline sp²–sp³ hybridized carbon.

In our recent investigations,^21–23 nanocrystalline solid carbons were synthesized under ambient conditions and low applied potentials (−1.1 to −1.6 V vs. Ag/AgCl) by the reduction of CO₂ using various types of electrochemically induced negatively charged metal nanoparticles (NPs) including Ag, Bi, Co, and Zn in a non-reducible N⁺-containing stabilizing medium. When a negative potential was employed in our system, nanoclustering occurred within the self-limiting ultrathin metal oxide layers of metal particles on the highly conductive substrate. Surplus electrons essential for reducing the ultrathin metal oxide layer could result in the creation of negatively charged metal NPs, effectively stabilized by the ternary electrolyte system comprising [BMIm]⁺[BF₄]⁻, propylene carbonate, and water. Interestingly, when negatively charged Ag NPs were in situ formed and utilized as the catalyst, nanocrystalline sp²–sp³ hybrid carbon structures were formed on the cathode.²⁴ Furthermore, the effectiveness of electrochemically activated negatively charged metal NPs has recently been proven across various reactions, such as the oxygen reduction reaction (ORR),²⁵ catalytic sulfenylation of indole,²⁶ the suppression of oxidized surface formation on Cu NPs.²⁷ For instance, studies have shown that electrochemically activated negatively charged Pt NPs exhibit specific and mass activities 89 and 31 times higher, respectively, than those of commercial Pt/C catalysts in ORR. Therefore, it is reasonable to expect that electrochemically activated negatively charged Ag NPs would demonstrate similar efficacy in reducing CO₂ to crystalline solid carbon. However, the precise pathway for the bottom-up synthesis of sp²–sp³ hybrid carbon structures from the interaction between electrochemically activated negatively charged Ag NPs and CO₂ remains unclear. Liquid oxygenates, abundant in nature and addressing mass transfer limitation associated with gaseous CO₂,²⁸ emerge as promising carbon sources. They not only enhance the suitability of the synthetic process but also facilitate investigating the formation mechanism of nanocrystalline sp²–sp³ hybrid carbon structures through the reduction reaction of various carbon sources by negatively charged Ag NPs. Comparing to previous, state-of-the art processes on the synthesis methods of diamond-like carbon (DLC) and nanocrystalline sp²–sp³ hybrid carbon, it can be seen that, for processes conducted at room-temperature and relatively low potential, only amorphous form of sp²–sp³ hybrid carbon or DLC was obtained. The synthesis of a mixed graphite-diamond structure, starting from glassy carbon or graphite, typically requires extremely high pressure. However, there has been a developing trend in the formation of metastable diamond at normal pressure, utilizing molten Zn,²⁰ liquid metal,²⁹ and monoatomic Ta.⁹

This study introduces an innovative single-step synthesis method for creating nanocrystalline sp²–sp³ hybrid carbon structures from various liquid organic oxygenates, including acetic acid, formic acid, acetone, acetaldehyde, ethanol, glycolic acid, acrylic acid, glycine, and glucose. The approach leverages electrochemically activated negatively charged Ag NPs under ambient conditions. Experimental observations with cooperative DFT calculations suggest a four-step synthesis process. Initially, the reaction between negatively charged Ag NPs and liquid oxygenate generates C₂ radicals. These radicals undergo coupling to form linear acetylenic carbon structures, which subsequently cross-link to produce graphitic CQDs. Remarkably, CQDs act as intermediates facilitating the formation of nanocrystalline sp²–sp³ hybrid carbon structures. CQDs are detected both on the cathode and in the electrolyte, underscoring their pivotal role. Ultimately, the CQDs intermediates transform into nanocrystalline sp²–sp³ hybrid carbon structures on the negatively charged Ag NPs. Notably, this transformation is influenced by the crystallographic planes of negatively charged metal NPs. For instance, the (111) plane of FCC Ag promotes the transition from sp² to sp³ hybridization, while employing a post-transition metal like Bi results in the formation of polycrystalline graphene structures from interconnected CQDs. The intricate synthesis mechanism behind sp²–sp³ hybrid carbon structure formation is described, providing valuable insights into the role of electrochemically activated negatively charges metal NPs in driving this transformative process.

Results and discussion

In this study, various types of negatively charged metal nanoparticles (Ag, Bi) with excess electrons were in situ synthesized via electrochemical activation under ambient conditions at relatively low applied potentials (−1.05 to −1.6 V vs. Ag/AgCl). To achieve the negatively charged state with excess electrons on the metal surface, a minimal presence of metal oxides (route 1) or metal cations (route 2) on a highly conductive substrate under a well-stabilized system is imperative. Under these circumstances, liquid oxygenates undergo conversion into nanocrystalline sp²–sp³ hybridized crystalline carbon under ambient conditions. The conditions conducive to an abundance of excess electrons on negatively charged metal NPs to facilitate the formation of nanocrystalline sp²–sp³ hybrid carbon structures under ambient conditions are dependent on the electrodeposition and pre-treatment conditions of the cathode, electrolyte conditions, applied potential, and the carbon source, which leverage the exceptional reductive power of in situ generated negatively charged metal NPs as summarized in Table 1. A strong relationship between experimental conditions leading to an excess of electrons on negatively charged metal NPs and the formation of sp²–sp³ hybrid carbon under ambient conditions is clearly established. The overall concept is shown in Fig. 1.

Fig. 1 The overall concept for the generation and formation of sp²–sp³ hybridized carbon allotropes from liquid oxygenates.

Table 1 Relationship between the experimental conditions conducive to excess electrons on negatively charged metal NPs and the formation of sp²–sp³ hybrid carbon under ambient conditions

Entry no.	Cathode	Electrodep. conditions	Pre-treatment of cathode	Electrolyte	Carbon Source	Applied potential^a (vs. Ag/AgCl)	Excess electron on metal clusters	Nanocrystalline sp^2–sp^3 carbon formation^b	Reasons
a Reaction time 30 min. b Based on Raman results.
1	Electrodep·Ag on Cu foil	(Route 1) 0.01 M AgNO3 in 0.6 M (NH4)2SO4, −1.1 V, 20 s	Exposed to air 12 h	0.4 M [BMIM]^+[BF4]^−	3.5 M Acetic acid	−1.05 V	Yes	Yes	Reduction of ultra-thin Ag oxide layers leading to negatively charged Ag NPs in stabilized system
2	Electrodep·Ag on Cu foil	(Route 1) 0.01 M AgNO3 in 0.6 M (NH4)2SO4, −1.1 V, 20 s	Immediate use	0.4 M [BMIM]^+[BF4]^−	3.5 M Acetic acid	−1.05 V	Minor	No	Less oxide formation, less amount of negatively charged Ag NPs
3	Electrodep·Ag on Cu foil	(Route 1) 0.01 M AgNO3 in 0.6 M (NH4)2SO4, −1.1 V, 200 s	Exposed to air 12 h	0.4 M [BMIM]^+[BF4]^−	3.5 M Acetic acid	−1.05 V	No	No	Too many Ag atoms to induce excess electron on Ag NPs
4	Electrodep·Ag on Cu foil	(Route 1) 0.01 M AgNO3 in 0.6 M (NH4)2SO4, −1.1 V, 20 s	Exposed to air 12 h	0.4 M [BMIM]^+[BF4]^−	3.5 M Acetic acid	−2.1 V	Minor	No	Negative bias induced coalescence of small Ag clusters, less amount of negatively charged Ag NPs
5	Electrodep·Ag on Cu foil	(Route 1) 0.01 M AgNO3 in 0.6 M (NH4)2SO4, −1.1 V, 20 s	Exposed to air 12 h	0.4 M (NH 4 ) 2 SO 4	3.5 M Acetic acid	−1.05 V	Yes	Yes	No salt effect on the reaction with negatively charged Ag
6	Electrodep·Ag on Cu foil	(Route 1) 0.01 M AgNO3 in 0.6 M (NH4)2SO4, −1.1 V, 20 s	Exposed to air 12 h	0.4 M [BMIM]^+[BF4]^−	3.5 M Acetate	−1.05 V	Yes	No	Strong electrostatic repulsion between the negatively charged Ag and acetate anions
7	0.001 M AgNO 3 and Cu foil	(Route 2)	-	0.4 M [BMIM]^+[BF4]^−	3.5 M Acetic acid	−1.05 V	Yes	Yes	Reduction of small amount of Ag^+ on Cu can also lead to negatively charged Ag NPs
8	Electrodep·Bi on Sn foil	(Route 1) 0.1 M Bi(NO3)3 in 1 M HNO3, −0.7 V, 60 s	Exposed to air 1 h	0.4 M [BMIM]^+[BF4]^−	3.5 M Acetic acid	−1.25 V	Yes	Yes	Reduction of thin Bi oxide layers leading to negatively charged Bi NPs in stabilized system
9	Electrodep·Bi on Sn foil	(Route 1) 0.1 M Bi(NO3)3 in 1 M HNO3, −0.7 V, 60 s	Exposed to air 12 h	0.4 M [BMIM]^+[BF4]^−	3.5 M Acetic acid	−1.25 V	No	No	Too many Bi^3+ atoms to induce excess electron on Bi NPS

---

For entry 1, acetic acid was chosen as the liquid oxygenate to demonstrate the negatively charged Ag NPs system via route 1. The creation of negatively charged Ag NPs takes place at the cathode under a well-stabilizing system by N⁺ containing salt. Small dendritic particles were prepared via electrodeposition at −1.1 V vs. Pt rod for 20 s. Subsequently, they were left in air for 12 h to naturally develop ultrathin oxide layers, approximately 2 nm for Ag₂O,²¹ serving as a precursor for the formation of negatively charged Ag NPs within the system. Upon applying a negative potential, the Ag–O bonds are attacked by electrons, leading to the formation of Ag atoms, which are proceeded to nanoclustering and coalescence into Ag NPs.³⁰ The surplus electrons surpassing the requirements for complete reduction of native Ag oxide, are efficiently transferred to Ag NPs due to their excellent electrical conductivity. Within a robust stabilizing system, excess electrons accumulate on the Ag NPs surface, and are not bound to Ag NPs, according to Gauss's law,²⁵ resulting in the formation of negatively charged Ag NPs.

The synthetic process was achieved by applying −1.05 V vs. Ag/AgCl, which was selected from cyclic voltammetry (CVs) (ESI Fig. S1†), to the cathode for 30 min. The characterization of carbon products on the cathode after the reaction on Ag NPs via route 1 and route 2 are shown in Fig. 2. For both cases, the Raman spectra reveal several peaks corresponding to nanocrystalline carbon structures exhibiting both sp² and sp³ hybridization states.^6,31,32 The D and G bands are observed at 1390 and 1540 cm⁻¹, respectively. These bands correspond to the breathing mode of the sp² carbon atoms in graphitic rings arising from the sp² carbon atoms bonded to the defect in the structure such as heteroatoms or sp³ carbon atoms and the bond stretching of sp² carbon atoms, respectively.^33–37 The presence of D and G bands confirms the existence of sp² carbon domains. The presence of sp³ carbon domains is indicated by the T band and the hexagonal-diamond related bands. The T band, observed at 1106 cm⁻¹, corresponds to sp³ bonded carbon features.^6,38–40 Additionally, the L₂, L₁ and L₃ bands, corresponding to the E_2g, A_1g and E_1g vibration modes of hexagonal diamond are detected at wavenumbers 1230, 1274, and 1331 cm⁻¹, respectively.^38,41–43 The broadening of these bands is likely attributed to the small grain size and stacking disorder.^38,42 Moreover, a peak at 1169 cm⁻¹ can be attributed to trans-polyacetylene [trans-(CH)_n] segments (t-PA), which are commonly formed in both natural and synthetic diamonds.^31,44 The XPS results (spectra acquired on Ag tape) also indicate that the crystalline carbon products on the cathode comprise both sp² and sp³ carbon domains. The C 1s spectra reveal the presence of sp² C–C, sp³ C–C, and C=O at binding energies 284.9, 285.6, and 288.5 eV, respectively. The percentage of sp³ carbon from XPS analyses was calculated to be 38%. Furthermore, the sp²–sp³ hybrid structure of synthesized carbon product on the cathode was confirmed by TEM-SAED results. The diffraction patterns correspond to several crystallographic planes of progressive intermediates for the transformation from graphite to diamond, or Diamond-C (JCPDS 89-8494),²⁴ suggesting that sp²–sp³ hybrid structure of the synthesized carbon products originate from the transformation of a graphitic carbon intermediate prior generated. Moreover, many diffraction patterns matching graphite and hexagonal diamond were observed (ESI Fig. S2†). The HR-TEM images show the presence of graphitic nanodomains with interplanar spacing 0.235 nm corresponding to (1120) plane of graphene.⁴⁵ The stacking of these graphitic nanodomains into larger nanocrystalline carbon structures is clearly evident from the parallel-line-like moiré fringes observed in the graphitic domain. In our system, these moiré fringes appear to result from stacking mismatches between the as-formed graphitic nanodomains. Moiré fringes have frequently been observed in natural-impact diamond formed from graphite.^46,47 Concerning the transformation of bilayer graphene into diamond, the nucleation site for sp³ bonds are the stacked areas in the moiré pattern.⁴⁸ The characterization results on the cathode not only confirm the formation of nanocrystalline sp²–sp³ hybrid carbon material from liquid oxygenates, but also validate that this carbon structure is created via nano-graphitic intermediates, graphitic CQDs as they were also detected in the electrolyte.

Fig. 2 Characterization of carbon products on the cathode after the reaction with acetic acid via route 1 [entry 1] and route 2 [entry 7].

For route 2, instead of using electrodeposited Ag electrocatalysts, the electrochemical reduction of acetic acid was conducted in a modified-electrochemical system wherein an ultra-small amount of AgNO₃ (i.e. 0.001 M AgNO₃) was added to the catholyte in the presence of a bare Cu foil substrate [entry 2]. Similar characteristic of the carbon products obtained from negatively charged Ag NPs generated from Ag oxides (route 1) and through the reduction of free Ag cations under excess electrons environment (route 2) verified the concept of this study. The Raman spectra from entry 2 exhibit similar characteristics of the carbon products observed in entry 1. In addition to the D and G bands, peaks relating to hexagonal diamond features, including the L₁ band at 1306 cm⁻¹, and the peak assigned to out-of-plane defects at 1509 cm⁻¹, which demonstrates sp³ hybridization of carbon,⁴⁹ were observed. The peak at 1414 cm⁻¹ also suggests the transformation of a carbon phase.⁵⁰ The SAED results are in good agreement with Diamond-C, graphite, and hexagonal diamond structures (ESI Fig. S3†). Meanwhile, the C 1s XPS spectrum indicates the presence of sp² C–C, sp³ C–C, and C=O at the same binding energies as those of entry 1 with a similar percentage of sp³ at 32%. In the HR-TEM images, the superposition between sp² carbon domains showing the (004) plane of graphite at lattice spacing of 0.168 nm and sp³ carbon domains formed from them, evidenced through the (002) plane of Diamond-C at the lattice spacing of 0.242 nm and the (200) plane of hexagonal diamond⁵¹ at the lattice spacing of 0.206 nm are apparent. Interestingly, we could also detect the (002) plane of graphite at the lattice spacing of 0.339 nm in the stacked area, which suggests transformation from graphite to Diamond-C. In agreement to the previous study,⁴⁸ such results suggest that the nucleation site for sp³ bonds in the system is the stacked area between graphitic clusters forming in moiré pattern. Moreover, the utilization of in situ generated Ag NPs from this route to synthesize metastable nanocrystalline sp²–sp³ hybrid carbon materials from various liquid oxygenates and substrates has also been proven (ESI Fig. S4†). The formation of sp²–sp³ hybrid carbon material via the transformation of as-formed graphitic carbon intermediate under these conditions, which follows the same route as entry 1, not only confirms an alternative pathway for producing negatively charged Ag NPs from Ag cations, but also eliminates the possibility of catalytic effects originating from Ag oxides, such as oxygen vacancies,^52,53 in synthesizing sp²–sp³ hybrid carbon materials.

The gaseous and liquid products obtained after the reduction of acetic acid at −1.05 V vs. Ag/Cl on the negatively charged Ag NPs were analyzed by GC and NMR (ESI Fig. S5†). Under the conditions used, there were no liquid products observed, with only trace amounts of CO and H₂ gaseous products (0.004–0.008 mmol min⁻¹, %FE_CO+H2 < 1) detected during the 30 min reaction time. The presence of graphitic CQDs in the used electrolyte solution was revealed by TEM-EDX-SAED, XRD, and Raman spectroscopy as shown in Fig. 3. The TEM micrographs display crystalline solid carbons with a sheet-like structure, resulting in diffraction peaks corresponding to the (100), (110), (201) graphite planes. Similar diffraction results have been reported for highly crystalline graphitized carbon dots synthesized by ZnCl₂-enabled thermochemical process.⁵⁴ The very weak (002) diffraction line of graphite is possible when the c axis of graphite is preferentially aligned parallel to the incident X-ray beam, while the (100) and (110) diffractions are strong.⁵⁵ A distinguishable (002) peak of graphite at 2θ = 26.5° is evident in the XRD results. Additionally, the aqueous solution containing these carbon products emitted strong blue fluorescence under UV light irradiation. The PL results reveal a maximum emission peak around 435 nm with an excitation wavelength of 360 nm, which is typical for blue CQDs⁵⁶ (ESI Fig. S6†). The carbon products obtained in the electrolyte are suggested to be in the form of CQD assemblies. It has been suggested that self-assembly of CQD end groups spontaneously occurs upon drying, with hydrogen bonding mainly responsible for nanosheet formation.⁵⁷ The PL intensity increases with increasing concentration, however, excessive high concentrations result in agglomeration due to the presence of a high amount of polar functionality.⁵⁸ Additionally, the Raman spectra exhibit two characteristic Raman peaks positioned at ∼1353 and ∼1580 cm⁻¹, corresponding well to the D and G bands in the CQDs structure, respectively. The conversion of acetic acid to CQDs based on GC and NMR analyses was determined to be approximately 35.2% after 30 min of reaction time. The majority of this conversion is attributed to the formation of CQDs dispersed in the electrolyte phase. In addition, the average yield of isolated nanocrystalline sp²–sp³ hybridized carbon material was estimated to be in the range of ∼50–100 mg cm⁻² of electrode surface area.

Fig. 3 Characterization of graphitic carbon products in the electrolyte.

Comparing various experimental conditions to induce an abundance of excess electrons on negatively charged metal NPs to facilitate the formation of nanocrystalline sp²–sp³ hybrid carbon under ambient conditions, it is worth noting that both insufficient Ag oxides [entry 2] and excessive amount of Ag oxides during nanoclustering [entry 3] are unlikely to create conditions conducive to excess electrons. Similarly, excessive negative bias can lead to the coalescence of metal clusters, hindering the formation of negatively charged Ag NPs [entry 4]. Under the specified conditions, the formation of solid carbon nanomaterials is either much less pronounced or absent as confirmed by the Raman results of the used electrodes (Fig. 4A–C). The formation of crystalline solid carbon products correlates strongly with the presence of excess electrons on small metal NPs. Stabilization of the active state of negatively charged Ag NPs via the electrostatic interaction between negatively charged Ag NPs and a positively charged nitrogen atom (N⁺) of the ionic salt in this study was confirmed by our density functional theory (DFT) calculations.^21,59 The anionic Ag clusters are significantly more stable than the neutral ones under [BMIM]⁺[BF₄]⁻ (ESI Table S1†). Moreover, the types of salt used can be varied from [BMIM]⁺[BF₄]⁻ to a simpler molecule (NH₄)₂SO₄, as they function equally well [entry 5] (ESI Fig. S4†). The negative charge state of negatively charged Ag NPs was further confirmed using acetate, which is a deprotonated form of acetic acid, as the carbon source instead of acetic acid [entry 6]. There were no solid carbon products after a 30 min reaction under the same conditions. The results suggest a strong electrostatic repulsion between the negatively charged Ag NPs and acetate anions, which interferes with the catalytic process.

Fig. 4 Raman results of the cathodes obtained from different experimental setups with Ag NPs [entries 1, 3, 4 and 6] (A–C) and TEM and Raman results obtained on the Bi cathodes after reaction [entries 8 and 9] (D).

The influence of crystallographic planes of negatively charged metal NPs on the crystalline structures of the nanostructured carbon formed is elucidated through the utilization of negatively charged metal NPs of post-transition metal (Bi). The TEM-EDX-SAED and Raman results obtained on the Bi cathode [entry 8] are depicted in Fig. 4D and ESI Fig. S5.† A suitable applied potential was determined via CV (ESI Fig. S6†). The cathode for the in situ formation of negatively charged Bi NPs was fabricated by depositing Bi onto a Sn substrate at −0.6 V vs. Pt rod for 60 s. The optimal ambient air exposure time to generate a thin oxide layer for the as-prepared electrodeposited Bi is 1 h as Bi is considerably more prone to oxidation in air, forming a naturally occurring native oxide, compared to Ag. Extending the air exposure time to 12 h leads to the formation of additional Bi oxide species, potentially hindering the creation of excess electrons conducive to the conditions required for negatively charged Bi NPs [entry 9]. Consequently, prolonged exposure resulted in no discernible formation of solid crystalline carbon during the electrochemical reduction reaction.

The TEM-EDX-SAED findings reveal that the crystallographic planes of the carbon products predominantly correspond to graphite planes, while the Raman spectra exhibit only the D and G bands at 1375 and 1573 cm⁻¹, respectively, without the peak associated with sp²–sp³ hybrid carbon materials. There appears to be less transformation of GQDs into nanocrystalline sp²–sp³ hybrid carbon materials on Bi surface. Since Ag is a face-centered cubic transition metal, the hybridization state of GQDs undergoes a remarkable transformation from sp² to sp³ on the negatively charged Ag NPs, whereas on Bi, being a post-transition metal with a hexagonal crystallographic structure, the produced GQDs seem to yield a polycrystalline graphene structure instead. These results align well with DFT calculations, indicating that the presence of (111) planes of a face-centered cubic transition metal can induce preferential buckling in one of the graphene sublattices, facilitating the transformation of sp² to sp³ hybridized carbon in this study.⁶⁰ Such findings suggest the potential to finely tune sp³/sp² ratios in the synthesized nanocrystalline sp²–sp³ hybrid carbon structures by varying the types of negatively charged metal NPs to control the degrees of transformation.

The aforementioned findings suggest the formation of graphitic CQDs from acetic acid and their immediate transformation into a nanocrystalline sp²–sp³ hybrid carbon structure in the presence of negatively charged Ag NPs. It should be noted that direct conversion of acetic acid into CQDs on the surface of negatively charged Ag NPs is unlikely due to several reasons. Firstly, CQDs would cover the surface of negatively charged Ag NPs, deactivating their catalytic activity. Additionally, CQDs would stack and form a hybrid carbon material on the negatively charged Ag NPs, resulting in an undetectable amount of CQDs in the electrolyte. Despite this, as mentioned earlier, CQDs were observed both in the electrolyte and on the cathode, while nanocrystalline sp²–sp³ hybrid carbon materials were only observed on the cathode. These observations suggest the presence of an active intermediate generated from the reaction between acetic acid and negatively charged Ag NPs. These intermediates can spontaneously form CQDs on the cathode and in the electrolyte.

In a recent study, plasmonic Ag NPs were utilized as catalysts to synthesize graphene oxide from liquid oxygenates under ambient conditions,⁶¹ with the formation of graphene oxide proceeding through an intermediate stage involving C₂ radicals.⁶¹ According to the literature,⁶²in situ synthesized C₂ radicals can transform into various graphitic carbon allotropes without the need for catalysts under ambient conditions. Diatomic carbon has a singlet biradical character with a quadruple bond and it serves as a molecular element in the bottom-up chemical synthesis of nanocarbons such as graphite, carbon nanotubes, and C₆₀.⁶² The construction of graphitic carbon allotropes from C₂ radicals is initiated by the coupling of C₂ radicals into linear acetylenic carbon, which is consecutively proceeded through intermolecular cross-linking to from graphitic carbon.⁶³ In our system, C₂ radicals appear to be a plausible active intermediate for the conversion of acetic acid into GQDs. The Raman shift around 2150 cm⁻¹ was observed on the cathode in the case of entry 1 (Fig. 4A), where sp²–sp³ hybrid carbon was formed on the cathode, suggesting the presence of linear acetylenic carbon,^64,65 which could be generated from radical coupling of C₂ radicals in this system. Meanwhile, for the cases marked as minor possibility to form negatively charged Ag NPs [entries 2–4], although they do not exhibit peaks related to graphitic carbon materials, Raman spectra show a prominent peak corresponding to linear acetylenic carbon at around 2150 cm⁻¹ (Fig. 4C). This indicates the potential formation of negatively charged Ag NPs catalyzing the conversion of acetic acid into C₂ radicals, which further proceeds through radical coupling to form linear acetylenic carbon. However, due to the insufficient amount of negatively charged Ag NPs [entry 2] and the instability of negatively charged Ag NPs [entry 4], there is not enough of these linear acetylenic carbon to undergo intermolecular cross-linking to from graphitic carbon.

It should be noted that in the case where the formation of negatively charged Ag NPs is unlikely [entry 4] due to the presence of excessive amount of Ag atoms, there are neither graphitic carbon materials nor peaks related to linear acetylenic carbon observed in Raman spectra. This validates that C₂ radicals are not created from the interaction between the Ag electrocatalyst and acetic acid, but they can be synthesized through the catalysis of negatively charged Ag NPs. From the experimental results, it can be concluded that the interaction of acetic acid and negatively charged Ag NPs leads to the coupling of C₂ radicals to form linear acetylenic carbon. Based on the observation of graphitic carbon intermediate on the cathode and in the electrolyte, as well as previous study,⁶³ it is likely that this linear acetylenic carbon undergoes cross-linking to form graphitic carbon. The phase transformation from linear acetylenic carbon to graphitic carbon and vice versa has been validated by a number of studies.^64,66 Moreover, on the negatively charged Ag NPs, the graphitic carbon is transformed into nanocrystalline sp²–sp³ hybrid carbon materials.

The role of negatively charged Ag NPs on the growth of carbon nanomaterials from C₂ radicals was further investigated under a similar environment as that of entry 1 without acetic acid. C₂ radicals are generated via an electro-oxidation of CaC₂. The removal of two electrons from acetylide anions results in the formation of C₂ radicals, which eventually transform into graphitic carbon (as depicted in Fig. 5A). Conversion of acetylide anions into graphitic carbon by means of electro-oxidation is well-established.^67,68 Upon applying the positive potential (+1.2 V vs. Ag/AgCl), the electrolyte solution near the working electrode darkened and diffused throughout the electrolyte after 1 h reaction time, with carbonaceous material deposited on the working electrode. Raman spectra confirmed these carbonaceous materials on the electrode and in the electrolyte to be graphitic carbon, as indicated by the presence of D and G peaks at 1385 and 1545 cm⁻¹ for the former and 1398 and 1554 cm⁻¹ for the latter. In addition, on the electrode, Raman peak at 1934 cm⁻¹ and 2135 cm⁻¹ attributing to C≡C stretching acetylene⁶¹ and linear acetylenic carbon species^64,65 (–[C≡C]_n–) was also observed, suggesting that the graphitic carbon spontaneously formed from in situ generated C₂ radicals through a linear acetylenic carbon intermediate under the conditions used. This result not only demonstrates the formation of graphitic carbon from in situ generated C₂ radicals on the electrode and in the electrolyte, aligning with the formation of graphitic CQDs intermediate on the cathode and in the electrolyte in the negatively charged Ag NPs system, but also underscores that the formation of nanocrystalline sp²–sp³ hybrid carbon materials required negatively charged Ag NPs surface.

Fig. 5 The growth of carbon nanomaterials from C₂ radicals (generated from acetylide anions from CaC₂) on the negatively charged Ag NPs without other carbon sources: the experimental setup and characterization of the carbon on the cathode and electrolyte after reaction with the negatively charged Ag NPs (A) snapshots from the CP-MD simulation showing the interaction of CC molecules with the Ag(111) surface over the simulation time of 1.8 ps. Initially positioned 2.40 Å above the Ag surface, the CC molecules approach and align, forming a linear acetylenic carbon –[C≡C]_n– chain (B).

The interaction and transformation of linear acetylenic carbon from C₂ radical molecules on an Ag (111) surface were investigated using ab initio Density Functional Theory (DFT) and Car-Parrinello Molecular Dynamics (CP-MD) simulations within the NVE ensemble. The nature of chemical bonding during the CP-MD simulation between 16 isolated C₂ molecules and an Ag(111) surface is illustrated in Fig. 5B. To ensure numerical stability, a time step of 0.12 fs was employed, and the total simulation duration was 1.8 ps. The Ag(111) surface was modeled as a three-layer silver slab, with each layer containing 16 hexagonally arranged Ag atoms, resulting in a total of 48 atoms. The Ag–Ag bond length was calculated to be 2.87 Å, with an interlayer spacing (d-spacing) of 2.45 Å between the Ag layers. The bottom 16 Ag atoms were frozen to replicate the bulk-like behavior. To simulate the excess charge commonly observed on Ag surfaces, 5 additional electrons were added to the neutral simulation system (686e⁻ in cell), resulting in a total charge of −5e⁻ per simulation cell (691e⁻ in cell). The C₂ molecules were initialized with a bond length of 1.33 Å and positioned approximately 2.40 Å above the top Ag layer to study their interaction with the metallic substrate.

As the reaction progresses, the C₂ molecules move closer to the Ag surface, reducing its distance to 2.10 Å above the Ag surface. This weak interaction facilitates molecular alignment and eventually polymerization into a linear acetylenic carbon chain from the C₂ molecules, denoted as, –[C≡C]_n–. Interestingly, no significant chemical bonding was observed between the C₂ molecules and Ag atoms, suggesting that the process was driven by intermolecular interactions rather than strong adsorption to the surface. A total energy reduction was calculated to be approximately ΔE = −1.54 eV per cell by the final state of the simulation. This total energy decrease indicates that the system achieved a more stable configuration with the formation of the –[C≡C]_n– chain, which subsequently detached from the Ag surface. The result highlights the role of weak surface interactions in promoting molecular rearrangement and coupling.

The innovative approach for catalytically synthesizing nanocrystalline carbon from liquid organic oxygenates under ambient conditions has been extended to various carbon sources including ethanol, acetone, acetaldehyde, and formic acid, employing applied potentials ranging from −1.05 to −1.6 V vs. Ag/AgCl as determined from CVs (ESI Fig. S7†). Following a 30 min reaction time, the formation of crystalline solid carbon from these diverse oxygenic compounds was identified on the cathodes by Raman spectroscopy and TEM-SAED analysis, as illustrated in Fig. 6. The Raman spectra of 3D-nanostructured carbon allotropes on the negatively charged Ag NPs subsequent to reactions with different carbon sources reveal sp²–sp³ hybrid carbon structures. Additional peaks observed at 1405 to 1420 cm⁻¹ indicate a transformation of one carbon phase into another,⁵⁰ while defect-related graphene structures yield the double-resonant D′ band at 1620 cm⁻¹.^6,49 These Raman bands have been previously documented for graphite subjected to high shock treatments and synthetic diamond materials.⁵⁰ Similar to the utilization of acetic acid, the lattice spacings of the carbon products synthesized from formic acid, acetone, acetaldehyde, and ethanol align well with those of Diamond-C, hexagonal diamond, and graphite. Stacking of GQDs and moiré fringes are clearly observed in the TEM images.

Fig. 6 Characterization results of the cathode after reaction on negatively charged Ag NPs with various liquid oxygenic carbon sources (*for ethanol, 0.3 M H₂O₂ was added to the reaction system).

We have additionally discovered that alternative feedstocks such as glycolic acid, acrylic acid, glycine, and glucose are viable carbon sources in this method, yielding carbon products on the cathodes with similar characteristics, indicating a comparable formation mechanism (ESI Fig. S8†). Remarkably, the as-prepared carbon products obtained from nitrogenous oxygenates contain substantial amounts of nitrogen atoms within the nanocrystalline carbon structure. This presents an intriguing opportunity to develop a strategy for synthesizing N-hybrid nanocrystalline carbon materials using nitrogenous oxygenates as raw materials. In the images where Ag is not presented, SAED clearly indicates the presence of the (002) plane of graphite in the carbon products synthesized using formic acid and glycolic acid as the carbon source. The formation of C₂ radicals from the liquid oxygenates in the system may occur by breaking all bonds between carbon and other atoms to form triple bonds within the molecules or between single carbon atoms. Compared to C–O bonds in oxygenic functional groups, C–H bonds are more resistant to breaking, as hydrogen is not a favorable leaving group. Therefore, in the case of ethanol, which contains a significantly higher ratio of C–H to C–O bonds compared to the other sources, a small amount of hydrogen peroxide (0.3 M H₂O₂) needs to be added to the system to generate hydroxyl radicals. These radicals then abstract hydrogen atoms from the molecules to create C₂ radicals^61,69 (refer to ESI Fig. S9† for details). A summary of the TEM-EDX-SAED results of the nanocrystalline carbon synthesized from various carbon sources via negatively charged Ag NPs are provided in ESI Fig. S10.†

To demonstrate the role of negatively charged metal NPs in the transformation of CQDs into nanocrystalline sp²–sp³ hybrid carbon, commercial blue-fluorescent graphene quantum dots (GQDs) (Sigma-Aldrich) are exposed to the same reaction conditions in the presence of negatively charged Ag NPs as illustrated in Fig. 7. These GQDs are used as the sole carbon source, devoid of any additional carbon source. Prior to their application in the reaction, the commercial GQDs displayed characteristic Raman D and G peaks at 1365 cm⁻¹ and 1552 cm⁻¹, respectively. Notable, the Raman features of the resultant carbon products on the cathodes subsequent to the reaction with GQDs exhibited distinct L₁, L₂, and L₃ bands at 1323, 1231, and 1356 cm⁻¹, respectively, indicative of the transformation of GQDs into sp³-hybridized structures (referred to route 1). These findings are congruent with the SAED data, which depicted diffraction patterns corresponding to Diamond-C and hexagonal diamond. Furthermore, the presence of the T band at 1098 cm⁻¹ was also noted. Investigating into the growth of GQDs on negatively charged Ag NPs commencing from 0.001 M AgNO₃ was also conducted (referred to route 2). The resulting products manifested Raman spectral characteristics including an L₁ band at 1288 cm⁻¹, L₂ band at 1206 cm⁻¹, T band at 1096 cm⁻¹, D band at 1373 cm⁻¹, G band at 1562 cm⁻¹ and t-PA at 1151 and 1444 cm⁻¹. Additionally, SAED analysis revealed various planes of Diamond-C and graphite, aligning with the Raman findings. ESI Fig. S11† furnishes detailed TEM-EDX-SAED outcomes regarding the conversion of commercial GQDs into sp²–sp³ hybridized carbon. Compared to other metastable carbon phases, the transition from graphite to hexagonal diamond occurs with the least energy barrier.⁷⁰ Previous studies have documented that this transformation typically involves the sliding of graphite base planes succeeded by buckling and plucking,³⁹ necessitating exceedingly high pressure and temperature conditions.^8–12

Fig. 7 Characterization results of the commercial blue-fluorescent GQDs before (A and B) and after applied on the negatively charged Ag NPs (C).

Here, the synergy between chemically induced phase transformation and the surface properties of negatively charged Ag NPs facilitates the ambient transformation of sp² to sp³ hybridization in carbon. This transformation is evident from the moiré fringes observed in TEM images of several samples, indicating the stacking of graphitic CGQDs on the surface of negatively charged Ag NPs during the initiation stages. These stacked regions areas within the moiré patterns serve as nucleation sites for the formation of sp³ hybridized carbon, originating from bilayer graphene.⁴⁸ Recent research demonstrates the potential for nanocrystalline sp²-to-sp³ carbon transformation with the assistance of monoatomic tantalum.^9,10 First-principles calculation suggests that tantalum supplies electrons to adjacent graphene layers, prompting a shift from sp² to sp³ electronic configuration, with an energy barrier of −5.38 eV per unit cell. In our setup, the surplus electrons on the surface of the negatively charged Ag NPs can drive the sp² to sp³ transformation within the stacked regions between graphitic clusters. The negatively charged Ag NPs not only provide excess electrons to adjacent graphene layers but also stabilize sp³ dangling bonds, as the energy barrier is significantly diminished through robust hybridization between the sp³ orbitals of dangling bonds and the metal's d_z² surface orbitals.¹¹ Furthermore, in contrast to three dimensional bulk systems, surface chemistry plays a pivotal role in substantially reducing the activation barrier for sp²-to-sp³ transformation⁷ when carbon structures are reduced to a few atomic layers. Chemical functionalization and chemisorption of active species such as H*, OH*, and F* onto the surface of multilayer graphene promote the transition towards the sp³ hybridization state, a concept known as chemically induced phase transformation.^60,71–73 Thus, the activation barrier for the transformation process of sp² into sp³ hybridized carbon in our system is significantly reduced by chemical functionalization on the CQDs surface, such as hydroxyl group incorporation.

Conclusions

In this contribution, we achieve the single step synthesis of nanocrystalline sp²–sp³ hybrid carbon structures from various liquid organic oxygenates at room-temperature and normal pressure. This process utilizes the super reductive power of the electrochemically activated in situ generated negatively charged Ag NPs to catalyze the conversion of liquid oxygenates following a distinct pathway (i) liquid oxygenates react with negatively charged Ag NPs, yielding C₂ radicals (ii) radical coupling produce linear acetylenic carbon, evidenced by the appearance of carbyne species in Raman spectra, (iii) conversion of linear acetylenic carbon to graphitic CQDs by intermolecular cross-linking, observed both on the cathode and in the electrolyte, as demonstrated by the electro-oxidation of acetylide anion experiment, (iv) transformation of CQDs into nanocrystalline sp²–sp³ hybrid carbon structures, facilitating by negatively charged Ag NPs stabilizing sp³ dangling bonds and supplying excess electrons to graphene layers. Chemical functionalization of CQDs further induces a phase transition towards sp³ hybridization. Notably, the crystallographic planes of transition metals influence the reaction, with the (111) plane of fcc Ag favoring sp² to sp³ transformation, while post-transition metals like Bi result in polycrystalline graphene structures. This approach offers a facile and energy-efficient synthesis route for nanocrystalline sp²–sp³ hybrid carbon structures from liquid oxygenates under ambient conditions. Additionally, insights gained from the synthetic process and formation mechanism could pave the way for advancements in materials science, engineering, and catalysis, particularly regarding the utilization of negatively charged metal NPs in carbon materials synthesis from liquid oxygenates.

Experimental procedure

Preparation of electrocatalysts

In this study, the electrocatalysts for the electrochemical reduction were simply fabricated by means of simple electrodeposition process. The surface preparation of the metallic substrate (Alfa Aesar, thickness 0.1 mm, purity 99.9999%) was performed by scratching with the abrasive paper in order to remove the native metal oxide distracting the electrodeposition process. Then, the electrodeposition process was conducted using a metal precursor stock, which was prepared by dissolving the metal precursor in de-ionized water under a vigorous stirring. The electrodeposition parameters are provided in Table 2. After electrodeposition, the electrocatalysts deposited on a metal substrate were exposed to ambient air for 24 h to create the native oxide layers of the electrocatalysts, which is the key to create negatively charged metal NPs. All the chemicals used in this process were purchased from Sigma-Aldrich and were used without further purification.

Table 2 Electrodeposition conditions for each metal electrocatalysts

The cathode		Metal precursor	Electrodeposition conditions
Electrocatalysts	Substrate		Applied potential (vs. Pt wire)	Electrodeposition time (s)
Ag	Cu foil	0.01 M AgNO3 in 0.6 M (NH4)2SO4	−1.1 V	20
Bi	Sn foil	0.1 M Bi(NO3)3 in 1 M HNO3	−0.7 V	60

---

Electrochemical reaction

A potentiostat machine (M204, Metrohm) was used to supply the current to a three-electrode system H-cell and collect the electrochemical data. To elaborate, the three-electrode system H-cell was separated into the cathodic region comprising the as-prepared electrocatalyst on the metal substrate as the working electrode along with 3 M Ag/AgCl as a reference electrode submerged in the catholyte containing the aqueous solution of liquid oxygenates and [BMIM][BF₄] as ionic salt and the anodic region having Pt foil as a counter electrode submerged in 0.1 M KHCO₃ as the anolyte. All potentials were applied against Ag/AgCl reference electrode. The cyclic voltammetry measurement has been performed using the scan rate of 50 mV s⁻¹ from 0 to −3 V against Ag/AgCl reference electrode in order to obtain the suitable applied potential. The gaseous products were analyzed by gas chromatography (Shimadzu GC-2014). The liquid state ¹H-NMR was determined by Fourier transform nuclear magnetic resonance spectrometer (Bruker AV400 ultra shield) 400 MHz with d₆-DMSO as the solvent. After being used in the electrochemical reduction, the working electrodes were washed with DI water in order to clean the surface before further characterization.

Characterization techniques

The Raman measurements were carried out using a PERKIN ELMER Spectrum GX equipped using the UV line at 532 nm and a TE-cooled CCD detector. TEM-EDX-SAED samples were prepared by dispersing the samples in ethanol with a few drops of the resulting suspension deposited on the TEM grid. After that the samples were observed with a TECNAI G2 Spirit Twin from FEI. The crystallite orientation of thin films was probed by GI-XRD at an incident angle of 3° at 12 keV (wavelength of 0.103 nm) with respect to the substrate surface. The results were calculated based on Cu-K_α (wavelength of 0.154 nm). The measurement was performed at beamlines BL1.1W at synchrotron light research and institute (Public organization) at the synchrotron Thailand central lab. Photoluminescence (PL) spectra were measured at ambient conditions by a spectrofluorophotometer (Fluromax, Horiba) using a Xe lamp as the light source and excitation wavelength 360 nm. X-ray photoelectron spectra (XPS) were acquired on a Ag tape using an Amicus spectrometer with Mg K_α X-ray radiation operated at 10 kV and 20 mA. The liquid state ¹³C-NMR and ¹H-NMR were measured by Fourier transform nuclear magnetic resonance spectrometer (Bruker AV400 ultra shield, topspin 2.1 software) 400 MHz and d₆-DMSO was used as solvent.

Statistical analysis of Raman and XPS spectrum

The baseline correction and the statistical analysis for the whole spectrum of Raman and XPS was attained using Fityk software. The deconvolution of the whole spectrum into single-band profiles was proceeded using a Gaussian function so that the integration of intensity and width of single bands is fitted to the whole spectrum.

Computational details

The interaction and transformation of linear acetylenic carbon from C₂ radical molecules on an Ag (111) surface were investigated using ab initio Density Functional Theory (DFT) and Car-Parrinello Molecular Dynamics (CP-MD) simulations within the NVE ensemble. These calculations were implemented using the Quantum ESPRESSO software suite. To ensure numerical stability, a time step of 0.12 fs was employed, and the total simulation duration was 1.8 ps. The Ag(111) surface was modeled as a three-layer silver slab, with each layer containing 16 hexagonally arranged Ag atoms, resulting in a total of 48 atoms. The Ag–Ag bond length was calculated to be 2.87 Å, with an interlayer spacing (d-spacing) of 2.45 Å between the Ag layers. The bottom 16 Ag atoms were frozen to replicate the bulk-like behavior. A vacuum region of 25 Å was introduced between periodic boundary cells to ensure no interaction between surfaces. To simulate the excess charge commonly observed on Ag surfaces, 5 additional electrons were added to the neutral simulation system (686e⁻ in cell), resulting in a total charge of −5e⁻ per simulation cell (691e⁻ in cell).

Author contributions

Conceptualization: RN. Methodology: RN, JP. Investigation: RN (major), KP, PS, PP, YL, SX, AS, UP. Funding acquisition: JP. Resources: JP, SP, UP. Project administration: JP, KCW. Supervision: JP. Writing – original draft: RN. Writing – review & editing: JP, CP, PAS, YL, KCW.

Conflicts of interest

Authors declare that they have no competing interests.

Data availability

All relevant data are available from the authors on reasonable request and/or are included within the article and the ESI.†

Acknowledgements

Financial supports from Chulalongkorn University C2F program, NSRF via the Program Management Unit for Human Resources & Institutional Development, Research and Innovation [grant number B16F650003], National Research Council of Thailand (Research Team Promotion), and the Royal Society-Newton Mobility Grant are gratefully acknowledged.

References

1. J. Vejpravová, Nanomaterials, 2021, 11, 2469.
2. S. F. Matar and V. L. Solozhenko, Diamond Relat. Mater., 2023, 133, 109747.
3. Y. Y. Zhang, S. Chen, H. Xiang and X. G. Gong, Carbon, 2016, 109, 246–252.
4. M. Hu, M. Ma, Z. Zhao, D. Yu and J. He, AIP Adv., 2016, 6, 055020.
5. M. Kapsiz, S. Geffroy, A. Holzer and K. Schmitz, Chem. Eng. Technol., 2023, 46, 118–127.
6. Z. Li, Y. Wang, M. Ma, H. Ma, W. Hu, X. Zhang, Z. Zhuge, S. Zhang, K. Luo, Y. Gao, L. Sun, A. V. Soldatov, Y. Wu, B. Liu, B. Li, P. Ying, Y. Zhang, B. Xu, J. He, D. Yu, Z. Liu, Z. Zhao, Y. Yue, Y. Tian and X. Li, Nat. Mater., 2023, 22, 42–49.
7. F. Lavini, M. Rejhon and E. Riedo, Nat. Rev. Mater., 2022, 7, 814–832.
8. D. Kraus, A. Ravasio, M. Gauthier, D. O. Gericke, J. Vorberger, S. Frydrych, J. Helfrich, L. B. Fletcher, G. Schaumann, B. Nagler, B. Barbrel, B. Bachmann, E. J. Gamboa, S. Göde, E. Granados, G. Gregori, H. J. Lee, P. Neumayer, W. Schumaker, T. Döppner, R. W. Falcone, S. H. Glenzer and M. Roth, Nat. Commun., 2016, 7, 10970.
9. C. Chen, D. Fan, H. Xu, M. Jiang, X. Li, S. Lu, C. Ke and X. Hu, Carbon, 2022, 196, 466–473.
10. Z. Zhu, C. Jiang, C. Chen, S. Lu, M. Jiang, X. Li and X. Hu, Carbon, 2023, 211, 118098.
11. D. Odkhuu, D. Shin, R. S. Ruoff and N. Park, Sci. Rep., 2013, 3, 3276.
12. M. Murri, R. L. Smith, K. McColl, M. Hart, M. Alvaro, A. P. Jones, P. Németh, C. G. Salzmann, F. Corà, M. C. Domeneghetti, F. Nestola, N. V. Sobolev, S. A. Vishnevsky, A. M. Logvinova and P. F. McMillan, Sci. Rep., 2019, 9, 10334.
13. M. Hu, J. He, Z. Zhao, T. A. Strobel, W. Hu, D. Yu, H. Sun, L. Liu, Z. Li, M. Ma, Y. Kono, J. Shu, H. K. Mao, Y. Fei, G. Shen, Y. Wang, S. J. Juhl, J. Y. Huang, Z. Liu, B. Xu and Y. Tian, Sci. Adv., 2017, 3, e1603213.
14. K. Luo, B. Liu, W. Hu, X. Dong, Y. Wang, Q. Huang, Y. Gao, L. Sun, Z. Zhao, Y. Wu, Y. Zhang, M. Ma, X. F. Zhou, J. He, D. Yu, Z. Liu, B. Xu and Y. Tian, Nature, 2022, 607, 486–491.
15. A. Gnisci, G. Faggio, G. Messina, J. Kwon, J. Y. Lee, G. H. Lee, T. Dikonimos, N. Lisi and A. Capasso, J. Phys. Chem. C, 2018, 122, 28830–28838.
16. V. Baranauskas, H. J. Ceragioli, A. C. Peterlevitz and S. F. Durrant, Diamond Relat. Mater., 2001, 10, 927–930.
17. H. J. Basheer, K. Baba and N. Bahlawane, ACS Omega, 2019, 4, 10405–10410.
18. A. Shawky, S. Yasuda and K. Murakoshi, Carbon, 2012, 50, 4184–4191.
19. A. Kumar, P. Ann Lin, A. Xue, B. Hao, Y. Khin Yap and R. M. Sankaran, Nat. Commun., 2013, 4, 2618.
20. V. A. Yolshina, L. A. Yolshina, V. A. Elterman, E. G. Vovkotrub, A. A. Shatunova, V. I. Pryakhina, N. A. Khlebnikov and N. V. Tarakina, Mater. Des., 2017, 135, 343–352.
21. R. Nganglumpoon, S. Watmanee, T. Teerawatananond, P. Pinthong, K. Poolboon, N. Hongrutai, D. N. Tungasmita, S. Tungasmita, Y. Boonyongmaneerat, N. Jantaping, S. Wannapaiboon, P. Praserthdam, Y. Morikawa, J. G. Goodwin and J. Panpranot, Carbon, 2022, 187, 241–255.
22. S. Watmanee, R. Nganglumpoon, N. Hongrutai, P. Pinthong, P. Praserthdam, S. Wannapaiboon, P. I. Szilágyi, Y. Morikawa and J. Panpranot, Nanoscale Adv., 2022, 4, 2255–2267.
23. P. Pinthong, S. Phupaichitkun, S. Watmanee, R. Nganglumpoon, D. N. Tungasmita, S. Tungasmita, Y. Boonyongmaneerat, N. Promphet, N. Rodthongkum and J. Panpranot, Nanomaterials, 2022, 12(19), 338.
24. J. Fayos, J. Solid State Chem., 1999, 148, 278–285.
25. E. Hua, S. Choi, S. Ren, S. Kim, G. Ali, S. J. Kim, W. S. Jang, S. Joo, J. Zhang, S. Ji, Y. S. Cho, J. Kang, T. Song, S. Hong, H. Choi, Y. M. Kim, H. Han and S. W. Kim, Energy Environ. Sci., 2023, 16, 4464–4473.
26. S. S. Han, A. Thacharon, J. Kim, K. Chung, X. Liu, W. S. Jang, A. Jetybayeva, S. Hong, K. H. Lee, Y. M. Kim, E. J. Cho and S. W. Kim, Adv. Sci., 2023, 10, 2204248.
27. K. Chung, J. Bang, A. Thacharon, H. Y. Song, S. H. Kang, W. S. Jang, N. Dhull, D. Thapa, C. M. Ajmal, B. Song, S. G. Lee, Z. Wang, A. Jetybayeva, S. Hong, K. H. Lee, E. J. Cho, S. Baik, S. H. Oh, Y. M. Kim, Y. H. Lee, S. G. Kim and S. W. Kim, Nat. Nanotechnol., 2022, 17, 285–291.
28. E. Pérez-Gallent, C. Vankani, C. Sánchez-Martínez, A. Anastasopol and E. Goetheer, Ind. Eng. Chem. Res., 2021, 60, 4269–4278.
29. Y. Gong, D. Luo, M. Choe, Y. Kim, B. Ram, M. Zafari, W. K. Seong, P. Bakharev, M. Wang, I. K. Park, S. Lee, T. J. Shin, Z. Lee, G. Lee and R. S. Ruoff, Nature, 2024, 629, 348–354.
30. X. Wu, Y. Guo, Z. Sun, F. Xie, D. Guan, J. Dai, F. Yu, Z. Hu, Y.-C. Huang, C.-W. Pao, J.-L. Chen, W. Zhou and Z. Shao, Nat. Commun., 2021, 12, 660.
31. M. L. Frezzotti, Nat. Commun., 2019, 10, 4952.
32. P. Németh, K. McColl, L. A. J. Garvie, C. G. Salzmann, C. J. Pickard, F. Corà, R. L. Smith, M. Mezouar, C. A. Howard and P. F. McMillan, Diamond Relat. Mater., 2021, 119, 109573.
33. H. Pang, X. Wang, G. Zhang, H. Chen, G. Lv and S. Yang, Appl. Surf. Sci., 2010, 256, 6403–6407.
34. Y. Liu, Y. Hu and J. Zhang, J. Phys. Chem. C, 2014, 118, 8993–8998.
35. Q. Xu, X. Lin, L. Gan, G. Owens and Z. Chen, J. Colloid Interface Sci., 2022, 605, 881–887.
36. J. Hong, M. K. Park, E. J. Lee, D. Lee, D. S. Hwang and S. Ryu, Sci. Rep., 2013, 3, 2700.
37. A. C. Ferrari, Solid State Commun., 2007, 143, 47–57.
38. V. Thapliyal, M. E. Alabdulkarim, D. R. Whelan, B. Mainali and J. L. Maxwell, Diamond Relat. Mater., 2022, 127, 109180.
39. F. Piazza, K. Gough, M. Monthioux, P. Puech, I. Gerber, R. Wiens, G. Paredes and C. Ozoria, Carbon, 2019, 145, 10–22.
40. S. M. Huang, Z. Sun, Y. F. Lu and M. H. Hong, Appl. Phys. A: Mater. Sci. Process., 2002, 74, 519–523.
41. S. V. Goryainov, A. Y. Likhacheva, S. V. Rashchenko, A. S. Shubin, V. P. Afanas'Ev and N. P. Pokhilenko, J. Raman Spectrosc., 2014, 45, 305–313.
42. Y. El Mendili, B. Orberger, D. Chateigner, J. F. Bardeau, S. Gascoin and S. Petit, Chem. Phys., 2022, 559, 111451.
43. S. Ugapeva, V. Afanasiev, A. Pavlushin and A. Eliseev, IOP Conf. Ser.: Earth Environ. Sci., 2019, 362, 012031.
44. F. Klauser, D. Steinmüller-Nethl, R. Kaindl, E. Bertel and N. Memmel, Chem. Vap. Deposition, 2010, 16, 127–135.
45. A. Abbas, T. A. Tabish, S. J. Bull, T. M. Lim and A. N. Phan, Sci. Rep., 2020, 10(11), 21262.
46. H. Ohfuji, T. Irifune, K. D. Litasov, T. Yamashita, F. Isobe, V. P. Afanasiev and N. P. Pokhilenko, Sci. Rep., 2015, 5, 14702.
47. H. Ohfuji, M. Nakaya, A. P. Yelisseyev, V. P. Afanasiev and K. D. Litasov, J. Mineral. Petrol. Sci., 2017, 112, 46–51.
48. D. Luo, D. Hui, B. Wen, R. Li, J. Yang, X. Shen, A. H. Reid, S. Weathersby, M. E. Kozina, S. Park, Y. Ren, T. D. Loeffler, S. K. R. S. Sankaranarayanan, M. K. Y. Chan, X. Wang, J. Tian, I. Arslan, X. Wang, T. Rajh and J. Wen, Phys. Rev. B, 2020, 102, 155431.
49. A. A. Khomich, V. Kononenko, O. Kudryavtsev, E. Zavedeev and A. V. Khomich, Nanomaterials, 2023, 13(1), 162.
50. A. Greshake, T. Kenkmann and R. Schmitt, Meteorit. Planet. Sci. Suppl., 2000, 35, 5.
51. H. Y. Kim, D. S. Kim, K. S. Kim and N. M. Hwang, Nanomaterials, 2020, 10, 1–12.
52. Y. Liu, P. Deng, R. Wu, X. Zhang, C. Sun and H. Li, J. Mater. Chem. A, 2021, 9, 6694–6709.
53. X. Ren, Y. Gao, L. Zheng, Z. Wang, P. Wang, Z. Zheng, Y. Liu, H. Cheng, Y. Dai and B. Huang, Surf. Interfaces, 2021, 23, 100923.
54. S. M. Wei, K. Feng, C. Li, N. Xie, Y. Wang, X. L. Yang, B. Chen, C. H. Tung and L. Z. Wu, Matter, 2020, 2, 495–506.
55. T. Yagi, W. Utsumi, M.-A. Yamakata, T. Kikegawa and O. Shimomura, Phys. Rev. B: Condens. Matter Mater. Phys., 1992, 46, 6031–6039.
56. M. McDowell, A. Metzger, C. Wang, J. Bao, J. M. Tour and A. A. Martí, Chem. Commun., 2018, 54, 4325–4328.
57. B. Fresco-Cala, M. L. Soriano, A. Sciortino, M. Cannas, F. Messina and S. Cardenas, RSC Adv., 2018, 8, 29939–29946.
58. A. Tyagi, K. M. Tripathi, N. Singh, S. Choudhary and R. K. Gupta, RSC Adv., 2016, 6, 72423–72432.
59. S. Watmanee, R. Nganglumpoon, N. Hongrutai, P. Pinthong, P. Praserthdam, S. Wannapaiboon, P. Á. Szilágyi, Y. Morikawa and J. Panpranot, Nanoscale Adv., 2022, 4, 2255–2267.
60. Y. Horbatenko, M. Yousaf, J. Lee, T. H. Choi, R. S. Ruoff and N. Park, Carbon, 2016, 106, 158–163.
61. J. J. Koo and Z. H. Kim, J. Phys. Chem. Lett., 2022, 13, 3740–3747.
62. K. Miyamoto, S. Narita, Y. Masumoto, T. Hashishin, T. Osawa, M. Kimura, M. Ochiai and M. Uchiyama, Nat. Commun., 2020, 11, 2134.
63. D. Kumar, R. Tiwari, R. K. Patel, P. Adhikary and S. Krishnamoorthi, J. Nanopart. Res., 2021, 23, 162.
64. E. O'Bannon, G. Xia, F. Shi, R. Wirth, A. King and L. Dobrzhinetskaya, Diamond Relat. Mater., 2020, 108, 107876.
65. B. Pan, J. Xiao, J. Li, P. Liu, C. Wang and G. Yang, Sci. Adv., 2015, 1, e1500857.
66. S. K. Shin and S. M. Park, Bull. Korean Chem. Soc., 2012, 33, 597–600.
67. A. H. C. Sirk and D. R. Sadoway, J. Electrochem. Soc., 2008, 155, E49–E55.
68. Y. Chen, Q. Xu, Q. Song, H. Li, Z. Ning, X. Lu and D. J. Fray, Electrochem. Commun., 2016, 64, 1–4.
69. G. Feng, P. Cheng, W. Yan, M. Borona, X. Li, J. H. Su, J. Wang, Y. Li, A. Corma, R. Xu and J. Yu, Science, 2016, 351, 1188–1191.
70. S. Srinivasan, R. Batra, D. Luo, T. Loeffler, S. Manna, H. Chan, L. Yang, W. Yang, J. Wen, P. Darancet and S. K. R. S. Sankaranarayanan, Nat. Commun., 2022, 13, 3251.
71. S. V. Erohin, Q. Ruan, P. B. Sorokin and B. I. Yakobson, Small, 2020, 16, 2004782.
72. A. P. M. Barboza, M. H. D. Guimaraes, D. V. P. Massote, L. C. Campos, N. M. Barbosa Neto, L. G. Cancado, R. G. Lacerda, H. Chacham, M. S. C. Mazzoni and B. R. A. Neves, Adv. Mater., 2011, 23, 3014–3017.
73. E. V. Emelin, H. D. Cho, V. I. Korepanov, L. A. Varlamova, S. V. Erohin, D. Y. Kim, P. B. Sorokin and G. N. Panin, Nanomaterials, 2022, 12, 4408.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5nr01388g

---