View PDF Version

Open Access Article
This Open Access Article is licensed under a
Creative Commons Attribution 3.0 Unported Licence
DOI: 10.1039/D5SE01500F (Paper) Sustainable Energy Fuels, 2026, Advance Article

Hydrothermal oxidative desulfurization of thiophene to sulfate: the effect of MoOx, WOx and carbon supports

Cheng Changab, Frédéric Vogelac, Oliver Kröcherab and David Baudouin*a
aPSI Center for Energy and Environmental Sciences, Paul Scherrer Institute, Forschungsstrasse 111, 5232 Villigen PSI, Switzerland
bInstitute of Chemical Sciences and Engineering, École Polytechnique Fédérale de Lausanne, (EPFL), 1015 Lausanne, Switzerland
cUniversity of Applied Sciences Northwestern Switzerland (FHNW), Institute for Biomass and Resource Efficiency, 5210 Windisch, Switzerland

Received 12th November 2025 , Accepted 20th March 2026

First published on 30th April 2026

Abstract

Among various forms of sulfur, some organosulfur compounds (particularly alkyl thiophenes) in biomass are rather refractory under hydrothermal conditions, posing a threat to the catalysts used in catalytic hydrothermal gasification (cHTG). In petrochemistry, alkyl thiophenes are usually treated by oxidative desulfurization (ODS) under mild conditions and removed in the form of sulfones, generating a sulfur-free product stream. ODS could be used to oxidize organosulfur compounds to sulfate, allowing efficient separation by exploiting the low salt solubility in supercritical water. To assess the viability of ODS in a cHTG process, we explored the effect of temperature and oxidant concentration (O/S ratio) on sulfate production from the ODS of thiophene. More importantly, the impact of Mo- and W-based carbon materials on the conversion of thiophene to sulfate was investigated. Our results showed a sulfate yield below 5% at temperatures ranging from 50 °C to as high as 400 °C in pressurized water. Experiments varying the oxidant-to-sulfur (O/S) ratio revealed that lower ratios (≤12) enhanced both sulfate yield and oxygen selectivity, whereas higher ratios (58 and 116) led to decreased selectivity due to excess oxidant consumption by organic matter. Carbon nanofibers (CNFs) alone increased the sulfate yield threefold (to 2.3%) at 400 °C, an effect attributed to oxygen-containing surface groups. Acid treatment of CNFs further boosted this yield to 7%. A clear correlation between surface functionalities and catalytic activity was established using FTIR and Boehm titration. Among metal oxides, Mo(IV), in the form of MoO2, was identified as an active phase for oxidative desulfurization (ODS), achieving a sulfate yield of 12%, while MoO3 and WO3 showed no such activity. However, metal oxide loading altered the CNF surface properties, potentially diminishing their promotional effect. These findings provide a basis for further development of MoO2 catalysts supported on surface-modified carbon materials, with the goal of preserving beneficial carbon surface characteristics.

Introduction

In the catalytic hydrothermal gasification (cHTG) process, organic matter present in a wet stream, such as sewage sludge, biogenic organic waste, or manure, can be completely converted to renewable gas, typically rich in methane. The use of supercritical water as a reaction medium allows fast kinetics, and eliminates the need to dry the feedstock prior to its valorization.1–3 Additionally, due to the drastic drop in dielectric constant and the derived salt solubility, SCW enables the separation of salts from the main reaction stream, facilitating their concentration into a brine phase.4–8 Sulfur is a prominent issue for such a process, as it is for all catalytic biomass conversion processes involving heterogeneous metal catalysts, which readily deactivate through sulfidation.9–11 In cHTG, hydrocarbons from biomass can transform various sulfur sources into H2S, which preferentially blocks the surface active sites of the catalyst.9,12,13 With the development of the cHTG process, inorganic sulfur species such as sulfate salts and hydrogen sulfide can be removed by salt separation4–8 and by chemisorption over sulfur scavengers,11,14 respectively. Sulfur scavengers, such as Zn-based11,15 or more recently Cu- and CeO2-based materials,16,17 can be used for the decomposition of a broad variety of organosulfur compounds (OSCs) such as alkyl sulfides, alkyl disulfides or thiols, and the subsequent absorption of the H2S formed. However, alkyl thiophenes are more recalcitrant, and their formation during cHTG of various biomass feedstocks is favored.16,18,19 Therefore, studying how alkyl thiophenes are formed, how to prevent their formation and how to eliminate them from the supercritical stream is crucial for lengthening the lifetime of sulfur scavengers and for protecting the catalyst, bringing economical value to the development of cHTG.

Oxidative desulfurization is a common method applied in petrochemistry to remove alkyl thiophenes from crude oil. In this process, thiophenes can be oxidized to sulfones over fixed-bed catalysts together with an oxidant.20–23 Subsequent separation of water-soluble sulfones by extraction or absorption allows the generation of low-sulfur oil products. Many studies on ODS were carried out with petroleum as the feedstock and sulfone or sulfate as the oxidized sulfur product.24–31 A wide range of catalytic materials can effectively drive oxidative conversions of thiophenic substrates under mild conditions. For H2O2, these include transition-metal oxides and titanosilicates (zeolites), e.g., Ti-MWW32 and Ti-Beta,33 porous ionic liquids,34 and Ti-, V-, Mo-, W- and Zr-based catalysts.35,36 For O2 activation in aerobic ODS, Fe3N,37 FeWC and W oxo-carbide catalysts have shown good performance,38,39 while ozone-assisted oxidative desulfurization can be promoted by transition metal complexes.40 However, these studies had ODS performed under mild conditions (T ≪ 350 °C, P < 3 MPa) and sometimes without water, which is not compatible with cHTG applications. In addition, in a cHTG context, it is desired to oxidize the sulfur in alkyl thiophenes all the way to sulfate, which can then be efficiently removed by salt separation thanks to the low salt solubility under supercritical water conditions.15,41–43 Because cHTG requires pumping an aqueous organic feed at high pressure, employing a liquid oxidizer is ideal, as it ensures efficient mass transfer during preheating, whereas a gaseous oxidizer would remain as a separate phase until the water reaches its critical point, creating localized high oxidizer concentrations and an associated explosion hazard.

To facilitate ODS with H2O2 in the cHTG process, Mo- and W-based catalysts can be expected to be the most active transition metal catalysts for ODS.44–48 Carbon materials, such as activated carbon (AC) and carbon nanofibers (CNFs), are the supports with the best compromise between stability, performance, specific surface area, and cost in supercritical water.49–52 A few studies have focused on Mo- and W-based carbon material for conventional ODS,53–56 but none under cHTG conditions.

This paper aims to bridge the gap in ODS studies of organosulfur compounds between petrochemical research and catalytic hydrothermal gasification (cHTG), thereby exploring a potential pathway for desulfurizing cHTG product streams upstream of a salt separator. Thiophene was chosen as a recalcitrant model organosulfur compound,16,19 together with H2O2 as the oxidant and glycerol as an organic model compound for biomass. This paper investigates the oxidative desulfurization of thiophene to sulfate under hydrothermal conditions, with a focus on near- and supercritical water environments typically encountered in continuous hydrothermal gasification (cHTG). The focus was not on the oxidation of the hydrocarbon framework, but on the selective transformation of sulfur. Particular attention was given to the role of carbon-supported molybdenum and tungsten oxides in influencing both sulfate yield and sulfur oxidation selectivity.

Experimental

Materials and methods

Glycerol (>99%, Carl Roth GmbH & Co. KG) was diluted in deionized water (DI H2O) to 20 wt% for the batch reactor experiments. Thiophene (>99%) was purchased from Fluka and diluted to 20 mmol L−1 in the glycerol–water solution before reaction, representing a typical concentration of sulfur (0.064%) in wet biomass.57 H2O2 (30%, Merck) was diluted shortly before the test in the model solution based on the O/S ratio for each experiment. After the batch reactor experiments, isopropanol (>99.9%, Fisher Scientific AG) was used for rinsing the reactor.

The Mo- and W-based catalysts were prepared using the incipient wetness impregnation (IWI) method on two supports, activated carbon AC-CGRAN (AC) from Cabot and carbon nanofibers NC7000 (CNF) from Nanocyl, crushed and sieved to 250–500 µm. Mo and W precursor solutions were prepared by dissolving ammonium molybdenum tetrahydrate ((NH4)6Mo7O24·4H2O) (>99%, Fisher Scientific AG) and ammonium metatungstate hydrate ((NH4)6H2W12O40) (≥99.9%, Fisher Scientific AG), respectively, in DI H2O. The metal concentration was then determined as the desired metal loading multiplied by the mass of the catalyst, divided by the total pore volume. The support was impregnated by drop-wise addition of the Mo or W precursor solution under vigorous stirring. The impregnated support was then dried overnight in air at 120 °C before calcination for 5 h in a quartz tubular furnace heated to 500 °C (5 °C min−1) under argon (60 mL min−1). According to the literature,58 the corresponding ammonium salt decomposes at high temperature into metal(VI) oxides, along with ammonia and water evacuated by the flowing argon. Due to different reports58–64 on the oxidation state of Mo after calcination, the catalysts prepared in this way are referred to as MoOx/AC, MoOx/CNF, WOx/AC, and WOx/CNF (Table 1). To prepare HNO3-treated CNF, as received CNF was immersed in 65% HNO3 in a round bottom flask equipped with a reflux condenser and heated in a water bath at 95 °C. After 2 h, the acid treated CNF was carefully filtered and washed with DI water until the pH of the filtrate became neutral. Afterwards, the acid treated CNF was dried overnight.

Table 1 Metal loading and starting materials used for the catalyst synthesis
Catalysts Metal loading (wt%) Ammonium salt Support
MoOx/AC 13.5 (NH4)6Mo7O24·4H2O AC
MoOx/CNF 13.5 (NH4)6Mo7O24·4H2O CNF
WOx/AC 21.8 (NH4)6H2W12O40 AC
WOx/CNF 21.8 (NH4)6H2W12O40 CNF

The hydrothermal experiments (Table 2) were conducted in a stainless steel batch reactor system developed in-house.57 The detailed experimental procedure is described in a previous study.65 In this study, the reaction temperature ranged from 50 °C to 400 °C. O/S ratios were mainly maintained at 0, 6 and 12, to study the effect of the O/S ratio under conditions similar to conventional ODS conditions. The O/S ratio is kept in this range to prevent overoxidation of S-free organics. The effect of the relatively high O/S ratios of 58 and 116 on the oxidation selectivity was also studied in supercritical water, for which O/C ratios were around 1/6 and 1/3, respectively. Note that such high O/C ratios will ultimately not only lead to carbon oxidation and hence CO2 formation, but also favor H2 production over CH4,66 both of which are undesirable in cHTG, where methane is typically targeted.

Table 2 List of experiments performed and conditions applied. For all tests, thiophene concentration and residence time were fixed at 20 mmol L−1 and 30 min, respectively
Experiment no. Potential catalysts Temperature (°C) Glycerol (wt%) H2O2 (mmol L−1) Pressure (MPa) O/S/C ratioa Metal/S ratiob
a In the O/S/C ratio, O takes only H2O2 into account and excludes the oxygen from glycerol.b Mo or W.c The mass of pure carbon materials added was kept the same as that of metal-loaded materials.
1 — 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
2 — 350 20 60 18 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
3 — 300 20 60 11 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
4 — 250 20 60 7 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
5 — 200 20 60 5 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
6 — 150 20 60 4 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
7 — 100 20 60 3 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
8 — 50 20 60 2.5 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
9 — 400 0 0 25 0 —
10 — 350 0 0 18 0 —
11 — 400 20 0 25 0 —
12 — 400 20 120 25 12[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
13 — 400 20 580 25 58[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
14 — 400 20 1160 25 116[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 —
15 — 400 0 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]4 —
16 — 350 0 60 18 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]4 —
17 AC 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 0c
18 CNF 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 0c
19 MoOx/AC 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 3[thin space (1/6-em)]:[thin space (1/6-em)]10
20 MoOx/CNF 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 3[thin space (1/6-em)]:[thin space (1/6-em)]10
21 WOx/AC 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 3[thin space (1/6-em)]:[thin space (1/6-em)]10
22 WOx/CNF 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1:330 3[thin space (1/6-em)]:[thin space (1/6-em)]10
23 HNO3 treated CNF 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 0
24 MoO2 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 3[thin space (1/6-em)]:[thin space (1/6-em)]10
25 MoO3 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 3[thin space (1/6-em)]:[thin space (1/6-em)]10
26 WO3 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 3[thin space (1/6-em)]:[thin space (1/6-em)]10
27 MoOx/CNF-3h 400 20 60 25 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330 3[thin space (1/6-em)]:[thin space (1/6-em)]10

Each experiment was performed twice. After each experiment, the gas phase was bubbled through an alkaline trap and collected for microGC analysis (see details in ref. 65). The aqueous phase was poured out directly, after which 4 mL isopropanol (IPA) was added into the reactor to wash out the liquid residue. This organic phase is referred to as the IPA phase. The detailed experimental protocol is described in the SI. Ion chromatography (IC) was used for the determination of sulfate. UV-Vis was used for the determination of hydrogen sulfide. Gas chromatography with a sulfur chemiluminescence detector (GC-SCD) was used for the identification of volatile organosulfur compounds and to determine thiophene conversion. For the catalysts, N2 physisorption was used to measure the specific surface area and total pore volume. X-ray diffraction (XRD) was used to determine the crystal phase of the metal catalysts. Transmission electron microscopy (TEM) was used to assess the presence of particles on the catalyst support. Fourier transform infrared (FTIR) spectroscopy was used to qualitatively identify the surface functional groups of the carbon material, while Boehm titration was used for quantification of the oxygen-containing acidic sites on the surface.67,68 Details of these analytical methods are described in the SI.

Results & discussion

Among the various organosulfur compounds that can be formed during supercritical water treatment of biomass, thiols, dialkyl disulfides and various thiophenes are the most frequent ones.57 Among these compounds, thiophenes are the most stable in supercritical water,69 due to the low electron density of their S atom. A preliminary series of ODS tests under cHTG conditions, at 400 °C and 25 MPa, was performed with various organosulfur compounds typically found in effluents produced from the hydrothermal treatment of biomass in the presence of H2O2 and glycerol (same reaction conditions as exp. 1, Table 2). The results showed that alkylthiols were the most sensitive to oxidation and thiophene the least, with the following sulfate yield order: 1-pentanethiol (29%) > benzothiophene (14%) > DMDS (9%) > thiophene (0.6%).70 Because of its stability towards oxidation, thiophene was selected as a model organosulfur compound to probe the efficiency of potential catalytic material in the ODS of organosulfur compounds in hydrothermal and supercritical water.

Influence of temperature on the hydrothermal oxidative desulfurization of thiophene

The influence of temperature on thiophene conversion and sulfate yield in hydrothermal ODS was studied between 50 and 400 °C. Thiophene conversion was found to be already 87% at 50 °C. The thiophene conversion slowly increased from ca. 87% to 94% as the temperature rose from 50 °C to 200 °C, as shown in Fig. 1. Within the temperature range of 250 °C to 400 °C, the thiophene conversion fluctuated around 90%. The highest conversion was around 95% at 300 °C, and the lowest conversion was around 87% at 50 °C. The respective standard deviation is relatively small compared to the corresponding mean value. The results show that the effect of temperature on sulfate formation or thiophene conversion was insignificant under the reaction conditions in this study. In terms of sulfate production, the change in sulfate yield over the temperature range of 50 °C to 400 °C was found to be limited, with the yield remaining below 5%. A slightly increased yield between 200 °C and 300 °C was observed, with a maximum of around 5% in this range. The selectivity of sulfate production was very low compared to thiophene conversion.
image file: d5se01500f-f1.tif
Fig. 1 Influence of temperature on thiophene conversion and sulfate yield in ODS (results from experiments 1 to 8, with 20 wt% glycerol, 20 mmol L−1 thiophene, 60 mmol L−1 H2O2 for 30 min, O/S/C = 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]330, see Table 2).

The high conversion of thiophene, coupled with low selectivity toward sulfate, suggests that thiophene rapidly reacts to form another/other product(s) during the temperature ramp-up. These products may still contain sulfur (organosulfur compounds) or may result in sulfur being released in forms other than sulfate. This indicates that either the kinetics toward complete oxidation are slow, or the selectivity for sulfur oxidation is inherently low under the tested conditions (reductive conditions).

To assess the role of water alone on thiophene conversion and sulfate yield (blank experiment), experiments were carried out with only thiophene in water as the feedstock. Two different temperatures were chosen, 400 °C and 350 °C, for comparison between supercritical water conditions and subcritical water conditions (exp. 9 and exp. 10 in Table 2). Fig. S1 shows that thiophene conversions were around 99% and 98% at 400 °C and 350 °C, respectively, and the corresponding sulfate yields were around 0.5% and 3%, respectively. A small amount of sulfate was formed, presumably due to the oxidation of thiophene by the residual oxygen dissolved in the water of the feedstock. In this blank experiment, the concentration of hydrogen sulfide collected in the liquid trap was below the detection limit within the temperature range studied, indicating negligible H2S production. This result is in agreement with the data from Lu's study,71 which showed that thiophene in supercritical water at (450 °C) hardly decomposes to produce H2S (with only 0.0012% yield).

The volatile organosulfur compound (VOSC) products were below the detection limit of the GC-SCD for all experiments (0.02 ppm of S72), as shown in the chromatogram in Fig. S2. This result shows that no new VOSCs were produced from the reaction between thiophene, H2O2 and glycerol during all the tests performed. The fact that the conversion of thiophene was comparable from 400 °C down to 50 °C and that no VOSCs were formed under any of these conditions indicates that thiophene does not decompose under hydrothermal conditions below 400 °C, but rather forms polar or larger non-volatile compounds that still contain sulfur. The GC-SCD result in Fig. S3 shows that there were no VOSC products formed when exposed to pure water at 350 °C and 400 °C either, which is in line with the literature.71 This indicates that thiophene or its decomposition products are converted to non-volatile S-compounds, possibly via dimerisation/oligomerisation.

The fact that the experiment in the absence of H2O2 leads to high thiophene conversion cannot be explained by the presence of small amounts of oxygen in the system (O2_dissolved/S is ca. 0.013[thin space (1/6-em)]:[thin space (1/6-em)]1). Hydrolysis can be excluded because volatile decomposition products would have been formed and observed by GC-SCD.73,74 The conversion of thiophene to non-volatile thienylthiophene (dimer) or to di/tetrahydrohydroxythiophenes, catalysed by first-row transition-metal cations, was reported in pure water at 240 °C and 3.4 MPa,75 and might partially explain the conversion of thiophene at high temperature, assuming leaching of Ni from the stainless steel. Finally, thiophene S-alkylation or electrophilic reactions with the reactive carbonyl intermediates derived from glycerol dehydration, can lead to thiophene-based alkylated or condensation products.76 The conversion of thiophene observed even at low temperature likely originates from a reaction with glycerol, but its origin could not be clearly evidenced.

Influence of the O/S ratio on oxidative desulfurization of thiophene

The effect of the O/S ratio during ODS of thiophene under cHTG conditions was studied by examining sulfate production and oxygen selectivity at different O/S ratios. From Fig. 2, it can be seen that increasing the O/S ratio from 0 to 12 caused an increase in the sulfate yield from 0.3% to 1.6%. The yield remained stable when increasing the O/S ratio to 58 (within standard deviation) and then further rose to 3.1% at an O/S ratio of 116. To evaluate the overall selectivity of H2O2 toward sulfate formation, the oxygen selectivity was analyzed, defined as the ratio of oxygen incorporated into sulfate to the total oxygen amount supplied by H2O2 (see eqn (S3)). The oxygen selectivity behaved differently from the sulfate yield. Initially, the increase in oxygen selectivity was insignificant when the O/S ratio increased from 6 to 12. Then, the oxygen selectivity decreased to 0.09% as the O/S ratio increased to 58. At this point, adding more H2O2 to increase the O/S ratio did not significantly affect the oxygen selectivity, which reached 0.11% when the O/S ratio was 112.
image file: d5se01500f-f2.tif
Fig. 2 Influence of the O/S ratio on the sulfate yield and oxygen selectivity for the ODS of thiophene (results from experiments 1 and 11–14: 20 wt% glycerol, 20 mmol L−1 thiophene, 400 °C and 25 MPa for 30 min, S/C = 1[thin space (1/6-em)]:[thin space (1/6-em)]330, see Table 2).

These results reveal that increasing the O/S ratio in the region ≤ 12 is beneficial for sulfate production and oxygen selectivity, but further increasing the O/S ratio to 58 has an adverse impact on oxygen selectivity. Doubling the O/S ratio to 116 would double the sulfate yield, but the oxygen selectivity remained the same. The low selectivity might be a result of the highly reactive conditions met in (near) supercritical water, which involve the formation of small reductants such as H2, short alcohols or aldehydes,77 prone to oxidation. Another possibility is that ODS took place during the temperature ramp-up and the thiophene oxidation product was reduced back once the SCW conditions in the presence of glycerol were reached. It is unclear how fast H2O2 reacts with thiophene and glycerol (and their decomposition products) in the experimental setting of this study. When the reaction system was overloaded with the oxidant (O/S ≥ 12), sulfate production continued to increase; however, excess oxidant – including unselective ˙OH radicals from H2O2 – was likely consumed by organic matter in the feedstock or might even reduce the already formed sulfate. This leads to a decrease in oxygen selectivity, suggesting that while thiophene conversion was rapid, the reaction pathway favors alternative products over selective sulfur oxidation. Within the 30 min residence time, further increasing the H2O2 concentration of the feed did not increase the oxygen selectivity for sulfate production from the ODS of thiophene.

Additionally, Fig. S4 shows that increasing the H2O2 concentration can lead to the measurable production of volatile organosulfur compounds, which are considered as byproducts in the process of sulfate production. Based on these results, an O/S ratio of 6 (H2O2 concentration at 60 mmol L−1) was considered optimal and chosen as the reaction conditions for subsequent experiments. Different feed compositions showed that the addition of 20 mmol L−1 thiophene and 60 mmol L−1 H2O2 did not affect the gas produced from the non-catalytic gasification of glycerol (Fig. S5). However, glycerol itself could affect the selectivity of the ODS of thiophene to sulfate, as its decomposition products in SCW include reducing gases such as CO and H2. Since glycerol was fixed at 20 wt%, 60 mmol L−1 H2O2 resulted in a low O/C ratio of 1/55, which might have hindered the oxidative desulfurization (ODS) of thiophene to sulfate.

To probe whether the sulfate production from thiophene oxidation was hindered by the low O/C ratio caused by glycerol, experiments were carried out without the addition of glycerol (with a high O/C ratio of 1.5) in the feed for comparison. Two different temperatures were chosen, 400 °C and 350 °C, for the comparison between supercritical water conditions and subcritical water conditions (exp. 15 and exp. 16 in Table 2).

Fig. S6 shows that thiophene conversion in the experiments without glycerol was around 99% at both 400 °C and 350 °C, slightly higher than in the cases with glycerol, which was around 92%. This can be explained by the addition of glycerol in the feedstock, which competes with thiophene in reacting with H2O2, thus leading to a lower thiophene conversion. This explanation is supported by the sulfate yield results in Fig. S7, in which the sulfate yields from the experiments without glycerol were around 15% and 33% at 400 °C and 350 °C, respectively, much higher than in the cases with glycerol, which were around 0.7% and 2%, respectively. Note that at the O/S/C ratio of 6[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]4 used, a maximum of 46% of the full oxidation of thiophene to CO2, H2SO4 and H2O can be reached. According to literature, the formation of sulfoxide (C4H4SO) or sulfones (C4H4SO2) during ODS is preferred. Thiophene sulfone is the most stable and common product before ring cleavage, indicating that the conditions applied favor the formation of sulfate. The higher sulfate yield observed at 350 °C compared to 400 °C may be attributed to increased reduction of intermediate organosulfur compounds by thiophene decomposition products, e.g. CO, at the higher temperature, which competes with sulfate formation and limits its accumulation. It should also be noted that VOSC products were below the detection limit without glycerol, meaning that decomposition of thiophene to volatile organosulfur side products did not occur under these conditions, or these compounds were readily oxidized to non-volatile compounds. This would be explained by the faster oxidation kinetics of, e.g., thiols or disulfides that can form upon thiophene decomposition,70 resulting in thiophene being the dominating organosulfur compound observed. These results underline the low selectivity of H2O2 oxidation of thiophene under these conditions.

The low sulfate selectivity further indicates that several competing pathways operate under these reducing conditions. Because thermochemical sulfate reduction is extremely slow78 and the S(+VI) → S(+IV) step is rate-limiting,79 sulfate formed in this system is unlikely to be further reduced. Thus, the limited sulfate yield may arise from partially oxidized sulfur intermediates being diverted into non-volatile organosulfur compounds, and/or from unselective H2O2 reactions with other organic species present.

The strong suppression of sulfate formation by glycerol can be attributed to intense competition between thiophene and the large excess of sulfur-free organic species (glycerol and its decomposition products), present at an S/Cglycerol molar ratio of 1[thin space (1/6-em)]:[thin space (1/6-em)]326. Under the applied conditions—low viscosity, complete miscibility of reactants, and high temperature and pressure—H2O2 is expected to react rapidly and unselectively with any available organic compound. This large excess of reductants relative to H2O2 promotes highly unselective oxidation pathways, thereby limiting sulfur oxidation to sulfate. Achieving high sulfate selectivity therefore requires a catalyst capable of steering the reaction toward selective sulfur oxidation and enhancing the corresponding reaction kinetics.

Influence of Mo- and W-based carbon materials on ODS of thiophene

As described in the former section, the use of catalysts to accelerate thiophene oxidation towards sulfate is necessary. To prevent excessive oxidation of organics, which is undesired in the catalytic hydrothermal gasification to methane, an O/C ratio of 1/55 was kept for the subsequent experiments.

Fig. 3 shows the influence of various Mo- and W-based carbon supported materials, as well as the carbon supports themselves, on the sulfate yield and oxygen selectivity in the ODS of thiophene. First, it can be observed that AC and CNF supports increased the sulfate yield by a factor of 2.4 and 3.4 when compared to the test without the addition of any material. The oxygen selectivity followed the same trend, with approximately 1.4% over AC and 1.8% over CNF, which is a 3 to 4-fold increase compared to the material-free test. The positive effect of the carbon materials might result from the oxygen-containing surface functional groups on the supports.


image file: d5se01500f-f3.tif
Fig. 3 Influence of various Mo- and W-based carbon materials on the sulfate yield and oxygen selectivity from the ODS of thiophene (results from experiments 1 and 17–22: 20 wt% glycerol, 20 mmol L−1 thiophene, 60 mmol L−1 H2O2, 25 MPa pressure and 30 min residence time, see Table 2).

After loading these two carbon supports with WOx, the sulfate yield over WOx/AC and WOx/CNF both dropped to around 0.7%, which is nearly the same as in the case without material addition. Their oxygen selectivity followed the same trend. However, loading MoOx on the carbon supports exhibited a different result. The sulfate yield over MoOx/CNF increased to 2.8%, which is 20% higher than that over pure CNF, while the sulfate yield over MoOx/AC was the same as over pure AC (1.7%). There are a few hypotheses that could explain these results: 1. WOx might favor competing reactions, preventing sulfate formation; 2. WOx might favor the reduction of sulfate or ODS intermediates 3. WOx altered the surface functionalities of the supports favoring thiophene oxidation to sulfate. In parallel, MoOx seems to participate in sulfate formation, or in changing the nature of the surface of the carbon support, e.g. by the formation of more carbon surface functionalities through C surface oxidation. It can, however, not be excluded at this stage that MoOx decreases the surface functional groups of the carbon support active in sulfate formation, while MoOx contributes to the ODS reaction. The difference between MoOx/AC and MoOx/CNF could be explained by the presence of different forms or phases of MoOx, or the much higher mesoporous surface area of CNF (AC being mostly microporous, see Table S3 and Fig. S13) and hence the intensified possible effect of MoOx on surface carbon active sites.

Note that experiments were conducted to quantify the sulfate adsorption capacity of AC, CNF, MoOx/AC and MoOx/CNF (blank adsorption tests, see the SI). These materials were found to adsorb sulfate at a level corresponding to 0.19%, 0.02%, 0.12%, and 0.015% sulfate yield (see Table S1), which are negligible compared to the values obtained during the hydrothermal ODS tests.

To better understand the effect of the carbon surface on the sulfate yield, FTIR spectra of the fresh catalysts were measured. The FTIR spectra of CNF, MoOx/CNF and WOx/CNF are presented in Fig. S8. Due to the strong absorption of carbon, it had to be highly diluted in KBr to allow exploitable IR transmission, at the expense of the signal to noise ratio. Yet, absorption bands at around 1260 cm−1, 1540 cm−1, and 1585 cm−1 were observed on CNFs, which can be assigned to the C–O bond vibration in lactone,80 the asymmetric stretching vibration of the carboxylate group (–CO2−),81 and the C[double bond, length as m-dash]C stretching mode associated with nanofiber surface defects,82 respectively. After loading MoOx and WOx, the absorption at 1260 cm−1 (C–O vibration in lactone) clearly decreased in the order CNF > MoOx/CNF > WOx/CNF while that at the other two bands varied slightly. This supports the previous assumption that WOx and MoOx altered the surface properties of the CNF.

Boehm titration was used to quantify the acidic sites on the potential catalysts, which allowed a more direct analysis of the oxygen-containing surface functional groups. With bases of various strengths (NaHCO3, Na2CO3, and NaOH), different acidic oxygen-containing groups are neutralized distinctively. Fig. 4 reveals that after the loading of MoOx, the total amount of acidic sites decreased, with only lactonic groups as the main contributor, represented by the C–O bond vibration in lactone at 1260 cm−1 on the FTIR spectrum of MoOx/CNF. The decrease in oxygen-containing surface functional groups on MoOx/CNF further supports the previous assumption that the loading of MoOx could alter the surface properties and cover active sites, which is in agreement with the FTIR results.


image file: d5se01500f-f4.tif
Fig. 4 The type and concentration of acidic functional groups on different materials: as received (untreated) CNF, on HNO3-treated CNF and on MoOx supported on (untreated) CNF.

XRD characterization was carried out on fresh AC, CNF, MoOx/AC, MoOx/CNF, WOx/AC, and WOx/CNF to evaluate the crystalline phases, and indirectly the oxidation state of Mo and W present in the supported metal oxides. The XRD patterns are shown in Fig. 5. The diffraction reflections (“diffraction peaks”) on the XRD pattern of MoOx/CNF correspond to the characteristic peaks of MoO2. No signals of MoO3 were found for MoOx/CNF. The diffraction peaks observed for WOx/CNF are characteristic of WO3 and no other type of tungsten phase was found. These results indicate that the oxidation state of Mo was reduced from +6 (ammonium salt used for the synthesis) to +4 after the thermal treatment performed in an inert atmosphere, while that of W remained the same at +6. The only explanation is that the CNF support acted as a reductant for MoO3 during the thermal treatment, reducing it to MoO2. In the same group of the periodic table, W has a higher atomic number (74) than Mo (42), resulting in a greater effective nuclear charge. The standard reduction potentials of MoO3/MoO2 and WO3/WO2 indicate a higher reducibility of molybdenum oxide (0.25 V vs. 0.036 V, respectively). Note that oxycarbides of W or Mo normally form only under high-temperature, non-aqueous, carbon-rich conditions (typically ≥600–900 °C),83,84 while thermodynamics show that neither metal should be reduced below +IV between 150–550 °C in pure or reductive supercritical water.85 Therefore, forming W or Mo oxycarbides under the conditions used here is extremely unlikely.


image file: d5se01500f-f5.tif
Fig. 5 XRD patterns of fresh Mo- and W-based catalysts and the neat supports AC and CNF.

In the case of MoOx/AC and WOx/AC, no diffraction peaks could be observed. This suggests that the metal oxides on AC are amorphous or smaller than ca. 5 nm. TEM was used to further assess the presence of particles, but hardly any particles could be observed (see Fig. S9 and S10).

Apart from the effect of surface functional groups, it has also been reported that the formation of Mo species with an oxidation state lower than +6 from MoO3 leads to high activity for the ODS of organosulfur compounds with Mo(V) as the most active species.86 The combination of XRD results and measured sulfate yields suggest that loading MoOx onto CNF promoted the sulfate yield from the ODS of thiophene due to the presence of Mo(IV). The difference between AC and CNF supported MoOx can be explained either by highly dispersed MoOx, as evidenced by XRD and TEM, which altered the carbon surface functionalities, compensating the promoting effect of the Mo(IV) or by MoOx, which is only present in the micropores of the AC, resulting in very low accessibility of the metal.

Note that Mo- and W-based phases present in the fresh catalyst are prone to evolve during the tests under reductive supercritical water.

The effect of oxygen-containing surface functional groups

The potentially promoting effect of oxygen-containing surface functional groups on the sulfate yield in the ODS of thiophene was studied by treating CNF with HNO3 (reflux for 2 h at 95 °C), which is known to oxidize the carbon surface.87–89 FTIR spectroscopy shows an increase in the absorption bands at around 1260 cm−1, 1540 cm−1, and 1585 cm−1 after the HNO3 treatment, which correspond to the C–O vibration in lactone, the asymmetric stretching vibration of the carboxylate group, and the C[double bond, length as m-dash]C stretching vibration of CNF surface defects, respectively (see Fig. 6). Additionally, two new bands appeared in the spectrum of the HNO3-treated CNF at around 1685 cm−1 and 1730 cm−1, which can be assigned to the C[double bond, length as m-dash]O bond vibration in lactonic groups and carboxylic groups, respectively.90–92 This confirms that the HNO3 treatment of CNF produced more oxygen-containing surface functional groups on CNF.
image file: d5se01500f-f6.tif
Fig. 6 FTIR results of the modified CNF materials (diluted 500-fold with KBr; the CNF used to prepare MoOx/CNF was not treated with HNO3).

Boehm titration (Fig. 4) of the HNO3-treated sample shows that the oxidation treatment of CNF led to a considerable increase in the concentration of acidic sites, which is around 8 times higher than that of the untreated CNF.

The same reaction conditions were applied in the batch reactor experiment (exp. 23 in Table 2) using HNO3-treated CNF as potential catalyst for the ODS of thiophene. Fig. 7 shows that the sulfate yield increased significantly to almost 8% over HNO3-treated CNF compared to around 2.3% over the original CNF, and the oxygen selectivity increased from below 0.5% to above 4%. Note that the quantity of acidic sites on the surface of CNF and HNO3-treated CNF represents approximately 3 and 9% of the sulfate formed, considering conservatively that four acidic sites are needed to oxidize one thiophene to sulfate, and they therefore cannot account for a stoichiometric (non-catalytic) oxidation of thiophene to sulfate. It can be inferred that the addition of oxygen-containing surface functional groups on the CNF promoted the sulfate yield from the ODS of thiophene. This result supports our previous hypothesis regarding the catalytic effect of oxygen-containing surface functional groups on selective thiophene oxidation towards sulfate.

It has been reported that organosulfur compounds (e.g. thiophene or dibenzothiophene) can adsorb onto surface sites93 of the carbon catalyst, and our results, together with prior studies, suggest that activated carbon likely enhances ODS by generating surface-bound percarboxylic acid species formed through the reaction of carboxylic acids with H2O2,94–96 as depicted in Fig. 8. Thiophene is expected to have π–π interactions with the CNF surface, favouring its reaction with percarboxylic acid. This proposed percarboxylic-acid-type active site aligns with established reaction pathways for liquid-phase oxidation of thiophenic sulfur compounds, where percarboxylic acids produced from carboxylic acids (typically formic or acetic acid) and hydrogen peroxide drive the oxidation process.21,97,98 Moreover, ex situ formation of surface-anchored percarboxylic acids on mesoporous silica has been demonstrated to effectively oxidize BT and DBT model compounds.99


image file: d5se01500f-f7.tif
Fig. 7 Influence of HNO3 treatment of CNF on sulfate yield and oxygen selectivity (results from experiments 1, 18, 20 and 23).

Carbon nanofiber and activated carbon (AC) has long been recognized as one of the more hydrothermally stable support materials for heterogeneous catalysts (at 23–30 MPa, 380–450 °C), particularly in comparison to silica, alumina, and zeolites.52,100 However, the stability of activated carbon is not unconditional, and its surface functional groups are subject to transformation under supercritical water conditions. However, to the best of our knowledge, the stability of such functionalities at the surface of supercritical water stable carbon materials has not been systematically studied, making the long-term stability of functionalised CNF difficult to assess.

Supporting MoOx on CNF led to a decrease in the total acidic sites, likely as a result of the adsorption of Mo on these sites. It is noteworthy that despite MoOx/CNF containing four times fewer oxygen-containing surface groups than CNF, it still produces a slightly higher sulfate yield. It is assumed that the addition of Mo(IV) can also promote the sulfate yield from the ODS of thiophene. The promoting effect of Mo(IV) addition might be weaker than the adverse effect of surface functional group removal on MoOx/CNF; thus, the sulfate yield over MoOx/CNF was still higher than that over pure CNF.

The effect of Mo(IV)

Batch reactor experiments (exp. 24 to exp. 26 in Table 2) were carried out using pure MoO2, MoO3 and WO3 as potential catalysts to further study the assumed promoting effect of Mo(IV) on the sulfate yield in the ODS of thiophene. To allow for a comparison, the same molar quantity of active metal as in the supported metal oxide materials was used. Note that the specific surface areas of the carbon-supported materials are two orders of magnitude larger than those of the pure oxides (see Table S2), but the surface areas of MoO3 and MoO2 are similar. The surface area of the oxide (active metal) is not known in MoOx/CNF, but the dispersion (the proportion of metal at the surface, and hence the proportion of metal that is active) is expected to be much higher than in MoO3 and MoO2 considering the TEM observations made. Fig. 9 shows that the sulfate yield over MoO2 increased to around 12%, while that over MoO3 and WO3 and without catalysts was around 0.6%. The oxygen selectivity followed the same trend and reached approximately 8%. This result is in line with the assumption that Mo(IV) promotes sulfate production, while Mo(VI) and W(VI) do not have any impact.
image file: d5se01500f-f8.tif
Fig. 8 Reaction pathway involving the formation of a percarboxylicacid-type active site leading to the oxidation of thiophene to, here, its sulfoxide.

Because the metal/S ratio (3[thin space (1/6-em)]:[thin space (1/6-em)]10) was kept constant and because the dispersion on MoOx/CNF is much higher than in MoO2, this result points out that the selectivity towards sulfate is much higher with MoO2 than with MoOx/CNF. Sintering or coking alone can therefore not explain the low performance of highly dispersed MoOx/CNF compared to poorly dispersed MoO2. Oxidation of Mo to Mo(VI) or rapid leaching of highly dispersed Mo species from MoOx/CNF may account for the reduced performance. Although these effects could not be assessed due to the small catalyst mass and the presence of coke, recent work showed substantial leaching of highly dispersed NiMo catalysts during heating from room temperature to 400 °C in pure water.56 While the conditions here differ and are less prone to leaching, Mo leaching during the batch experiments cannot be excluded, particularly on the highly dispersed MoOx/CNF (Fig. 9).


image file: d5se01500f-f9.tif
Fig. 9 The effect of pure MoO2, MoO3, and WO3 on sulfate yield and oxygen selectivity (results from experiments 1, 18, 20, and 24 to 26).

Then, the effect of calcination temperature of the catalysts on the sulfate yield was investigated. It has been reported that MoO3 can be reduced to MoO2 on the surface of carbon materials at high temperatures due to the reductive nature of carbon.101 At 500 °C, an increasing amount of MoO2 was formed with the increase in residence time during the reaction between MoO3 and carbon. The original calcination time used for the preparation of MoOx/CNF was 5 h. By shortening the calcination time to 3 h (exp. 27 in Table 2), less MoO2 formation was expected. The XRD patterns (Fig. S11) of both catalysts show little difference between both, except for smaller and broader diffraction peaks, indicative of lower MoO2 formation, which is in agreement with Zhang's observations.101

Fig. S12 shows that the sulfate yield over MoOx/CNF-3h is around 1.5%, which is lower than that over MoOx/CNF and CNF alone, indicating that the promoting effect of Mo(IV) on the sulfate yield is weaker with shorter calcination time, and inadequate loading of Mo(IV) (or calcination of CNF) might not be able to compensate for the drop in catalytic activity due to the alteration of surface functional groups. This result also supports that Mo(IV) is a key factor ins enhancing the sulfate yield.

Molybdenum is well known for its redox capability in oxidative processes such as ODS;102 however, under the predominantly reductive conditions applied here, thermodynamic modeling indicates that MoO2 is the dominant stable phase,85 and any redox cycling would more likely occur between Mo(0) and Mo(IV) sites. The reduction of MoO3 to MoO2 in pressurized pure water has been experimentally observed at temperatures as low as 320 °C,103 consistent with thermodynamic calculations,85 when accounting for the measured103 oxygen fugacity in pure water. However, the very low activity of MoO3 suggests that this reduction proceeds slowly. Given that reaction conditions and medium composition evolve continuously during testing, and therefore oxygen fugacity also changes, obtaining direct experimental evidence for molybdenum-based redox mechanisms is particularly challenging and lies beyond the scope of this work.

Conclusion

In the ODS of thiophene under cHTG-relevant conditions, temperature has little effect on conversion and does not significantly promote sulfate formation. In contrast, the O/S ratio strongly influences both sulfate yield and oxygen selectivity. A moderate increase in the O/S ratio enhances sulfate yield, but beyond a certain threshold, selectivity toward sulfate declines. This is likely due to competing oxidation reactions involving hydrocarbons in the feedstock, which consume the excess oxidant. Integrating the ODS of thiophene into a SCWG process in order to convert organosulfur compounds to sulfate would therefore negatively impact methane production.

Selectivity towards sulfate formation from thiophene during ODS reaches a maximum around 350 °C before dropping above the critical point of water. This is likely due to competitive reduction reactions in SCW caused by the rapid decomposition of organics and higher concentrations of small reductive compounds (H2, CO, light alcohols, and alkenes). However, the sulfate yield without the use of a catalyst remains marginal, i.e., below 5%.

Catalyst screening revealed that carbon nanofiber enhances sulfate production, whereas activated carbon does not, which suggests that micropores are less relevant, probably because they become filled with coke. A clear correlation was observed between oxygen-containing surface groups and sulfate yield. By increasing CNF surface functionalities or by using pure MoO2, a significant increase in thiophene conversion to sulfate can be achieved.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: experimental procedures, analytical methods, data analysis equations, additional characterization results (XRD, TEM, FTIR, physisorption, Boehm titration), chromatograms, control experiments, and supporting figures and tables. See DOI: https://doi.org/10.1039/d5se01500f.

Acknowledgements

This work was funded by the Swiss National Science Foundation (SNF grant 183663). Part of this work was performed within the Energy System Integration (ESI) platform at PSI. The authors acknowledge Dr Reinhard Kissner from ETHZ for support with IC analysis, Julian Indlekofer from PSI for help with GC-SCD analysis, Norbert Schmid from PSI for technical help, and Dr Luca Maggiulli and Dr Davide Ferri for their help with N2 physisorption and FTIR. The authors acknowledge Dr Yukihiko Matsumura from Hiroshima University (Japan) for valuable discussions.

References

  1. C. Jesslyn, A. V. Conradie and E. Lester, Chem. Eng. J., 2021, 128837 Search PubMed.
  2. J. A. Okolie, R. Rana, S. Nanda, A. K. Dalai and J. A. Kozinski, Sustain. Energy Fuels, 2019, 3, 578–598 RSC.
  3. F. Vogel, Hydrothermal production of SNG from wet biomass, in Synthetic Natural Gas from Coal, Dry Biomass, and Power-To-Gas Applications, ed. T. J. Schildhauer and S. M. A. Biollaz, John Wiley & Sons, Hoboken, 2016, ch. 10, pp. 249–278,  DOI:10.1002/9781119191339.ch10.
  4. M. Schubert, J. B. Müller and F. Vogel, J. Supercrit. Fluids, 2014, 95, 364–372 CrossRef CAS.
  5. D. Baudouin, D. Salionov, F. Vogel and S. Bjelić, ACS Eng. Au, 2021, 1, 134–147 CrossRef CAS.
  6. J. Reimer, G. Peng, S. Viereck, E. De Boni, J. Breinl and F. Vogel, J. Supercrit. Fluids, 2016, 117, 113–121 CrossRef CAS.
  7. M. Schubert, J. Aubert, J. B. Müller and F. Vogel, J. Supercrit. Fluids, 2012, 61, 44 CrossRef CAS.
  8. M. Schubert, J. Supercrit. Fluids, 2012, 61, 44 CrossRef CAS.
  9. M. Dreher, J. Catal., 2013, 301, 38 CrossRef CAS.
  10. M. Dreher, M. Steib, M. Nachtegaal, J. Wambach and F. Vogel, ChemCatChem, 2014, 6, 626–633 CrossRef CAS.
  11. G. Peng, C. Ludwig and F. Vogel, Appl. Catal., B, 2017, 202, 262–268 CrossRef CAS.
  12. O. Yakaboylu, Biomass Bioenergy, 2013, 59, 253 CrossRef CAS.
  13. M. Osada, Energy Fuels, 2007, 21, 1400 CrossRef CAS.
  14. R. C. Brown, in Thermochemical Processing of Biomass, 2011, p. 1 Search PubMed.
  15. G. Peng, Methane Production from Microalgae via Continuous Catalytic Supercritical Water Gasification: Development of Catalysts and Sulfur Removal Techniques, PhD thesis, École Polytechinque Fédérale de Lausanne, 2015.
  16. D. Baudouin, H. Xiang and F. Vogel, Biomass Bioenergy, 2022, 164, 106529 CrossRef CAS.
  17. H. Xiang, D. Baudouin and F. Vogel, Sci. Rep., 2023, 13, 9987 CrossRef CAS.
  18. A. R. Katritzky, M. Balasubramanian and M. Siskin, Energy Fuels, 1992, 6, 431–438 CrossRef CAS.
  19. A. R. Katritzky, R. A. Barcock, M. Balasubramanian, J. V. Greenhill, M. Siskin and W. N. Olmstead, Energy Fuels, 1994, 8, 498–506 CrossRef CAS.
  20. C. Cai, Chem. Geol., 2003, 202, 39 CrossRef CAS.
  21. S. Otsuki, Energy Fuels, 2000, 14, 1232 CrossRef CAS.
  22. J. Xiao, Appl. Energy, 2014, 113, 78 CrossRef CAS.
  23. W. Jin, RSC Adv., 2017, 7, 48208 RSC.
  24. T. Hirai, K. Ogawa and I. Komasawa, Ind. Eng. Chem. Res., 1996, 35, 586 CrossRef CAS.
  25. S. Pysh’yev, Chem. Technol., 2012, 229 Search PubMed.
  26. S. Pyshyev, B. Korchak, D. Miroshnichenko and N. Vytrykush, ACS Omega, 2022, 7, 26495–26503 CrossRef CAS PubMed.
  27. Z. Long, Fuel, 2014, 130, 19 CrossRef CAS.
  28. Y. Zhang, J. Chem., 2018, 6495826 Search PubMed.
  29. D. K. Bal and J. B. Bhasarkar, Asia-Pac. J. Chem. Eng., 2019, 14, 2271 CrossRef.
  30. B. Khodaei, Chem. Eng. Process. Process Intensif., 2018, 132, 75 CrossRef CAS.
  31. P. Wu, S. Ma and W. Zhu, Green Chem., 2026, 28, 37–95 RSC.
  32. H. Wang, M. Yu and Z. Lin, J. Mol. Model., 2019, 25, 106 CrossRef PubMed.
  33. H. Wang, Y. Deng and R. Zhou, Theor. Chem. Acc., 2018, 137, 66 Search PubMed.
  34. P. Wu, B. Wang, L. Chen, J. Zhu, N. Yang, L. Zhu, C. Deng, M. Hua, W. Zhu and C. Xu, Adv. Sci., 2024, 11, 2401996 Search PubMed.
  35. V. Hulea, E. Dumitriu and F. Fajula, Catalysts, 2021, 11, 867 Search PubMed.
  36. Y. Chen, Q. Tian, Y. Tian, J. Cui and G. Wang, Appl. Sci., 2021, 11, 2018 Search PubMed.
  37. P. Wu, S. Ma, S. Zhou, Y. Sun, L. Chen, J. Liu, W. Zhu and C. Xu, Chem. Eng. Sci., 2025, 305, 121136 CrossRef CAS.
  38. A. V. Akopyan, E. A. Eseva, R. A. Mnatsakanyan, D. A. Davtyan, M. O. Lukashov, I. S. Levin, A. A. Sadovnikov, A. V. Anisimov, A. M. Terzyan, A. M. Agoyan and E. Karakhanov, Chem. Eng. J., 2023, 464, 142641 CrossRef CAS.
  39. A. V. Akopyan, R. A. Mnatsakanyan, E. A. Eseva, D. A. Davtyan, P. D. Polikarpova, M. O. Lukashov, I. S. Levin, K. A. Cherednichenko, A. V. Anisimov, A. M. Terzyan, A. M. Agoyan and E. A. Karakhanov, ACS Omega, 2022, 7, 11788–11798 CrossRef CAS PubMed.
  40. A. V. Akopyan, D. A. Grigoriev, P. L. Polikarpova, E. A. Eseva, V. V. Litvinova and A. V. Anisimov, Pept. Chem., 2017, 57, 904–907 CrossRef CAS.
  41. M. H. Waldner, Catalytic Hydrothermal Gasification of Biomass for the Production of Synthetic Natural Gas, ETH Zurich, 2007 Search PubMed.
  42. M. Schubert, J. W. Regler and F. Vogel, J. Supercrit. Fluids, 2010, 52, 113–124 CrossRef CAS.
  43. M. Schubert, J. W. Regler and F. Vogel, J. Supercrit. Fluids, 2010, 52, 99–112 CrossRef CAS.
  44. D. Wang, Appl. Catal., A, 2003, 253, 91 CrossRef CAS.
  45. W. A. W. A. Bakar, R. Ali, A. A. A. Kadir and W. N. A. W. Mokhtar, Fuel Process. Technol., 2012, 101, 78–84 CrossRef CAS.
  46. R. A. Sheldon, J. Mol. Catal., 1980, 7, 107 CrossRef CAS.
  47. N. A. Khan, B. N. Bhadra, S. W. Park, Y. Han and S. H. Jhung, Small, 2019, 16, 1901564 CrossRef PubMed.
  48. G. Rodriguez-Gattorno, A. Galano and E. Torres-García, Appl. Catal., B, 2009, 92, 1 CrossRef CAS.
  49. A. A. Peterson, Energy Environ. Sci., 2008, 1, 32 RSC.
  50. P. Azadi and R. Farnood, Fuel Energy Abstr., 2011, 36, 9529 CAS.
  51. D. C. Elliott, Biofuel Bioprod. Biorefining, 2008, 2, 254–265 CrossRef CAS.
  52. G. Peng, C. Ludwig and F. Vogel, ChemCatChem, 2016, 8, 139–141 CrossRef CAS.
  53. X. An, Chem. Eng. J., 2024, 480, 147879 CrossRef CAS.
  54. M. A. Astle, Adv. Funct. Mater., 2019, 29, 1808092 CrossRef.
  55. J. Zou, Adv. Funct. Mater., 2021, 31, 2100442 CrossRef CAS.
  56. Y. Liu, Fuel, 2022, 324, 124534 CrossRef CAS.
  57. D. Baudouin, H. Xiang and F. Vogel, Biomass Bioenergy, 2022, 164, 106529 CrossRef CAS.
  58. C. Petit and T. J. Bandosz, Microporous Mesoporous Mater., 2009, 118, 61 CrossRef CAS.
  59. D. G. B. Dionizio, L. Forrer, G. Berhault, P. M. de Souza and C. A. Henriques, Mol. Catal., 2023, 536, 112882 CAS.
  60. U. C. Abubakar, J. Clean. Prod., 2019, 211, 1567 CrossRef CAS.
  61. C. Petit and T. J. Bandosz, Environmental Science & Technology, 2008, 42, 3033 Search PubMed.
  62. L. Feng, J. Catal., 2000, 190, 1 CrossRef CAS.
  63. M. Simnad and A. Spilners, JOM, 1955, 7, 1011–1016 CrossRef CAS.
  64. T. N. Kovács, J. Therm. Anal. Calorim., 2016, 124, 1013 CrossRef.
  65. C. Chang, Chem. Eng. Res. Des., 2024, 205, 459 CrossRef CAS.
  66. Y. Alshammari and K. Hellgardt, Clean Energy via Hydrothermal Gasification of Hydrocarbon Resources, 22nd World Energy Congress paper, Daegu, Korea, 2013 Search PubMed.
  67. S. L. Goertzen, K. D. Thériault, A. M. Oickle, A. C. Tarasuk and H. A. Andreas, Carbon, 2010, 48, 1252–1261 CrossRef CAS.
  68. H. Ren, Nanoscale Adv., 2019, 1, 1432 RSC.
  69. A. R. Katritzky, D. A. Nichols, M. Siskin, R. Murugan and M. Balasubramanian, Chem. Rev., 2001, 101, 837–892 CrossRef CAS PubMed.
  70. C. Chang, Behavior of sulfur under hydrothermal conditions: removal strategies for catalytic processes, EPFL, 2024 Search PubMed.
  71. L. Lu, Fuel, 2023, 353, 129251 CrossRef CAS.
  72. A. S. Calbry-Muzyka, Energy Fuels, 2019, 33, 9859 CrossRef CAS.
  73. P. D. Clark and J. B. Hyne, Fuel, 1984, 63, 1649–1654 CrossRef CAS.
  74. P. D. Clark, J. B. Hyne and J. David Tyrer, Fuel, 1984, 63, 125–128 CrossRef CAS.
  75. P. D. Clark, N. I. Dowling, J. B. Hyne and K. L. Lesage, Fuel, 1987, 66, 1353–1357 CrossRef CAS.
  76. M. Schiel, C. Domini, A. Chopa and G. Silbestri, Synthesis, 2018, 50, 4846–4854 CrossRef CAS.
  77. D. Salionov, C. Hunston, F. Vogel, D. Baudouin and S. Bjelić, J. Catal., 2023, 426, 257–269 CrossRef CAS.
  78. C. Chang, F. Vogel, O. Kröcher and D. Baudouin, Chem. Eng. Res. Des., 2024, 205, 459–466 CrossRef CAS.
  79. A. Meshoulam, G. S. Ellis, W. Said Ahmad, A. Deev, A. L. Sessions, Y. Tang, J. F. Adkins, J. Liu, W. P. Gilhooly, Z. Aizenshtat and A. Amrani, Geochim. Cosmochim. Acta, 2016, 188, 73–92 CrossRef CAS.
  80. J. L. Figueiredo, et al, Carbon, 1999, 37, 1379 CrossRef CAS.
  81. B. C. Smith, Spectroscopy, 2018, 33(5), 20–23 Search PubMed.
  82. A. Peyvandi, P. Soroushian, N. Abdol and A. M. Balachandra, Carbon, 2013, 63, 175–186 CrossRef CAS.
  83. C. Johnson, H. Sellinschegg and D. C. Johnson, Chem. Mater., 2001, 13, 3876–3881 CrossRef CAS.
  84. J. Ouyang, Y. Sun, Y. Zhang, J. Liu, X. Bo and Z. Wang, Molecules, 2025, 30, 84 CrossRef CAS PubMed.
  85. J. N. Jocz, P. E. Savage and L. T. Thompson, Ind. Eng. Chem. Res., 2018, 57, 8655–8663 CrossRef CAS.
  86. Y. Tu, Energy Fuels, 2019, 33, 8503 CrossRef CAS.
  87. S. Sainio, D. Nordlund, R. Gandhiraman, H. Jiang, J. Koehne, J. Koskinen, M. Meyyappan and T. Laurila, J. Phys. Chem. C, 2016, 120, 22655–22662 CrossRef CAS.
  88. D. Sebastián, M. J. Lázaro, R. Moliner, I. Suelves, A. S. Aricò and V. Baglio, Int. J. Hydrogen Energy, 2014, 39, 5414–5423 CrossRef.
  89. Y. Takahashi, H. Fujita and A. Sakoda, Adsorption, 2013, 19, 143–159 CrossRef CAS.
  90. J.-H. Zhou, et al., Carbon, 2007, 45, 785–796 CrossRef CAS.
  91. J. M. O'Reilly and R. A. Mosher, Carbon, 1983, 21, 47 CrossRef.
  92. S. Shin, J. Jang, S.-H. Yoon and I. Mochida, Carbon, 1997, 35, 1739–1743 CrossRef CAS.
  93. A. Zhou, X. Ma and C. Song, Effects of oxidative modification of carbon surface on the adsorption of sulfur compounds in diesel fuel, Appl. Catal., B, 2009, 87(3), 190 CrossRef CAS.
  94. Z. C. Kampouraki, Catalytic oxidative desulfurization of a 4,6-DMDBT containing model fuel by metal-free activated carbons: the key role of surface chemistry, Green Chem., 2019, 21, 6685–6698 RSC.
  95. M. T. Timko, J. A. Wang, J. Burgess, P. Kracke, L. Gonzalez, C. Jaye and D. A. Fischer, Fuel, 2016, 163, 223–231 CrossRef CAS.
  96. F. F. Roman, Catalysts, 2021, 11, 1239 CrossRef CAS.
  97. U. Arellano, J. A. Wang, M. T. Timko, L. F. Chen, S. P. Paredes Carrera, M. Asomoza, O. A. González Vargas and M. E. Llanos, Fuel, 2014, 126, 16–25 CrossRef CAS.
  98. M. Te, C. Fairbridge and Z. Ring, Appl. Catal., A, 2001, 219, 267–280 CrossRef CAS.
  99. P. De Filippis and M. Scarsella, Ind. Eng. Chem. Res., 2008, 47, 973–975 CrossRef CAS.
  100. C. Hunston, D. Baudouin, L. Koning, A. Agarwal, O. Kröcher and F. Vogel, Appl. Catal., B, 2023, 320, 121956 CrossRef CAS.
  101. G.-H. Zhang, Metall. Res. Technol., 2018, 115, 416 CrossRef CAS.
  102. C. Wang, M. Ma, Y. Zhang, Y. Zhang, J. Chen, J. Li, Y. Lu, X. Yao and M. Zhang, Catalysts, 2025, 15, 1146 CrossRef CAS.
  103. G. Solferino and A. J. Anderson, Chem. Geol., 2012, 322–323, 215–222 CrossRef CAS.
This journal is © The Royal Society of Chemistry 2026
Click here to see how this site uses Cookies. View our privacy policy here.