Next Article in Journal
SnO2 Quantum Dot-Decorated g-C3N4 Ultrathin Nanosheets: A Dual-Function Photocatalyst for Pollutant Degradation and Hydrogen Evolution
Previous Article in Journal
Coupling Carbon Dioxide and Cyclohexane Oxide Using Metal-Free Catalyst with Tunable Selectivity of Product Under Mild Conditions
Previous Article in Special Issue
Nano-TiO2-Enhanced Surface Functionalization of Recycled Concrete Aggregates for Improved Degradation Efficiency of Low-Concentration Sulfur Dioxide
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Catalysts
Volume 14
Issue 11
10.3390/catal14110823
catalysts-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Deactivation and Regeneration Studies of Molybdenum-Based Catalysts in the Oxidative Desulfurization of Marine Fuel Oil

by
Teddy Roy
1,*,
Joy Alakari
1,
Christine Lancelot
2,
Pascal Blanchard
2,
Line Poinel
1 and
Carole Lamonier
2
1
SEGULA Technologies, 71 rue Henri Gautier, F-44550 Montoir de Bretagne, France
2
Université de Lille, CNRS, Centrale Lille, ENSCL, Université d’Artois, UMR 8181—UCCS—Unité de Catalyse et Chimie du Solide, F-59000 Lille, France
*
Author to whom correspondence should be addressed.
Catalysts 2024, 14(11), 823; https://doi.org/10.3390/catal14110823
Submission received: 15 October 2024 / Revised: 8 November 2024 / Accepted: 12 November 2024 / Published: 15 November 2024
(This article belongs to the Special Issue Designing Catalytic Desulfurization Processes to Prepare Clean Fuels, 2nd Edition)
Browse Figures
Versions Notes

Abstract

:
The oxidative desulfurization (ODS) of heavy fuel oil (HFO) offers a promising solution for desulfurizing marine fuels under mild conditions, in line with current environmental regulations. While most studies focus on model or light fuels, explaining deactivation through leaching or sulfone adsorption, the deactivation mechanisms of catalysts in HFO remain poorly understood. In this work, Mo-based catalysts supported on alumina were extensively characterized before and after catalytic reactions, and regeneration through air calcination was considered. Techniques such as XRD, Raman spectroscopy, XRF, and TGA, alongside catalytic testing with H2O2 as an oxidant, revealed that Mo surface speciation significantly impacted both activity and deactivation. Contrary to well-dispersed polymolybdates, crystalline MoO3 induced low activity and hindered regeneration. No leaching of the active phase was demonstrated during the reaction. Sulfone adsorption had minimal impact on deactivation, while non-sulphur compounds appeared to be the key contributors. Regeneration outcomes were found to be molybdenum content-dependent: 10Mo/Al recovered its activity, while 20Mo/Al formed inactive phases, like Al2(MoO4)3. Using an organic oxidant (tBHP) during ODS influenced the regeneration, as it prevented Al2(MoO4)3 formation and redispersed crystalline MoO3, enhancing performance. These findings advance understanding of catalyst deactivation and suggest strategies to extend catalyst life in the ODS of HFO.

Graphical Abstract

1. Introduction

To limit the SO2 emissions in marine transport, the authorized sulphur content of marine fuel oils is controlled by strict environmental regulations: It was reduced by the International Maritime Organization from 0.5 wt.% to 0.1 wt.% in Sulphur Emissions Control Areas (SECAs) in 2015, and from 3.5 to 0.5 wt.% outside these areas in 2020 (MARPOL ANNEX VI) [1]. Upgrading fuel quality is usually achieved through the hydrodesulfurization (HDS) process, but the high contents of sulphur and the large amount of refractory compounds in marine fuels require more severe temperature and pressure conditions, inducing important energy consumption and thus cost increase [2]. Among the non-hydrogenation desulfurization processes, oxidative desulfurization (ODS) is considered a promising alternative or complementary method to conventional HDS [3]. This is mainly due to the economic benefits resulting from the mild operating conditions, the absence of expensive hydrogen, and also the high reactivity in ODS of refractory compounds in HDS, like 4,6-dimethyldibenzothiophene (4,6-DMDBT) [4,5,6,7].
In the ODS process, sulphur compounds are oxidized into their corresponding sulfones, which, due to their polar properties, can be further separated from the reaction mixture by extraction or adsorption [8,9,10,11]. The increasing number of papers dealing with ODS demonstrates the growing interest in this process. However, most studies focus on ODS of model fuels or light real feedstocks, such as low-sulphur diesel [12,13,14,15,16,17,18]. In cases addressing sulphur content higher than 0.5 wt.%, use of homogeneous catalysts [19], despite their efficiency, is not realistic for industrial-scale applications due to challenges in catalyst recovery. As a result, heterogeneous catalysts are increasingly researched and utilized for the ODS of heavy fuel oil. Mo-based catalysts are among the most commonly used materials. Tang et al. [20] use molybdenum oxide for the ultrasound-assisted ODS of marine fuel in which total sulphur content is reduced from 3.2 to 2.6 wt.%. On a raw marine fuel IFO500 with an initial sulphur content of 3.3 wt.%, Houda et al. [21] achieve a residual sulphur content of 0.5 wt.% after ODS on a molybdenum catalyst supported on alumina, with ultrasound assistance.
Catalyst deactivation and their potential regeneration are indeed key challenges to be addressed for the further development of the process. One cause given in the literature to explain the observed activity decrease in the ODS reaction is active phase leaching, mainly reported in flow reactors. Chica et al. [16] observe a rapid deactivation of Mo/Al2O3 catalysts in the ODS of a model diesel in a fixed-bed reactor, about 20% of Mo leaches out under the reaction conditions. Vedachalam et al. [22] demonstrate progressive leaching of molybdenum HPA on mesoporous TUD-1 in the ODS of a mildly hydrotreated heavy gas oil when using hydrogen peroxide as an oxidative agent, contrary to the use of cumene hydroperoxide, where no deactivation is noticed. On oxovanadium (IV) polymer-supported catalysts, Ogunlaja et al. [23] show that active phase leaching during the ODS of a model feed in a flow reactor could be limited by modifying the vanadium encapsulation polymer used. To limit the leaching phenomena, modifying the support to increase its interaction with the active phase is considered by Rivoira et al. [24], who limited deactivation by doping SBA-15 with Ga and Al in vanadium-oxide-supported catalysts in the ODS of DBT. Estephane et al. [25] successfully reduce the active phase leaching by the incorporation of W-HPA in SBA-15, thus maintaining the activity in the ODS of an SRGO in a fixed-bed reactor, contrary to its classically impregnated counterpart.
Sulfone adsorption on the catalyst is also reported as an important cause of deactivation, and several strategies have been developed to prevent this retention. On Ti-MCM41 catalysts in the ODS of a model diesel, Chica et al. [16] note a rapid deactivation without any leaching of Ti and relate the observed sulfone retention to the presence of silanol groups on the MCM support. It is confirmed by the much better stability of the silylated Ti-MCM41 catalyst, where no silanol groups are available. Prasad et al. [12] propose that the presence of a Bi promoter in a Mo-based catalyst might hinder sulfone adsorption and thus prevent catalyst deactivation during ODS of a model molecule. Recently, by tuning the support properties with hydrophobic properties, Akopyan et al. [26] minimize sulfone adsorption and increase catalyst lifetime. However, extrapolating deactivation phenomena obtained in the ODS of model molecules to those occurring with real feeds is difficult. Indeed, a better solubility of sulfones, leading to a decrease in adsorbed sulfones on the catalyst surface is reported in real feeds, which are more polar than model feeds [12,27]. On tungsten-supported SBA-15 catalysts, Estephane et al. [27] report a sulfone retention of 98% after the ODS of a model feed, versus only 50% for LGO and 20% for SRGO.
In addition to the retention of sulphur-containing molecules, other species such as nitrogen compounds, hydrocarbon compounds present in the feed, and water or degradation products introduced by the oxidant can also contribute to catalyst deactivation through adsorption on the catalyst surface. Ishihara et al. [8] also observe that among different nitrogen molecules, the presence of indole is more detrimental to ODS activity stability than the presence of quinoline, acridine, and carbazole. Several authors also mention that, besides the competitive adsorption of sulphur and nitrogen compounds, the basic characteristics of these nitrogen compounds can also influence catalyst activity [28]. Mokhtari et al. [29] compare the reuse of molybdenum oxide supported on silica gel under identical conditions in both a real diesel fraction and a model feed composed of DBT in n-octane. No deactivation is observed after seven successive tests with the model feed, while a performance decline of over 30% occurs after just four tests with the real diesel feed. The authors conclude that the primary factor affecting catalyst reuse in the ODS of real diesel is the interaction with various non-sulphur hydrocarbons. Water can also affect the reaction through strong adsorption on the catalyst surface. As reported by Tu et al. [30], an excess of hydrogen peroxide generates more water, which encapsulates the catalyst and decreases its interaction with sulphur compounds. However, these studies are mainly performed on model compounds, which cannot represent the diversity and complexity of the composition of real heavy feeds.
It appears that while catalyst stability is sometimes described in the literature, no study has yet investigated or tried to discriminate the catalyst deactivation phenomena occurring during the ODS of real heavy feed. This justifies our comprehensive study to discern the underlying factors contributing to deactivation, with the final aim of designing efficient and robust catalysts.

2. Results

2.1. Characterization of the Marine Fuel RMG380-DC, Before and After Oxidation

The International Maritime Organization (IMO) classifies marine fuels into categories based on sulphur content and viscosity. High Sulphur Fuel Oils (HSFO) are highly viscous residual fuels, exhibiting viscosities up to 700 cSt (at 50 °C) and sulphur contents reaching 3.5 wt.%. These properties can be adjusted by blending residual fuels with lighter distillate fractions. In our case, RMG380 is a residual marine fuel oil with a sulphur content of 1.3 wt.% and a viscosity of 380 cSt (at 50 °C). The ODS reaction can be performed directly on raw marine fuel oil [31], but dilution in n-dodecane (C12) was performed to facilitate manipulations, mixing conditions, and analysis according to the methodology described elsewhere [21]. The resulting precipitated asphaltenes were further removed by centrifugation and the recovered liquid phase, noted RMG380-DC, was further used for ODS testing.

2.1.1. Elemental Analysis of RMG380 and RMG380-DC

The announced sulphur content of 1.3 wt.% was confirmed by CHNS analysis for the raw fuel RMG380, which is in the range of intermediate marine fuel oils, with a nitrogen mass content of 0.37 wt.%. Similar high N contents have been reported for marine fuels, 0.37 wt.% for an IFO500 with 3.3 wt.% S, 0.39 wt.% for an HSFO700 with 2.9 wt.% S [32], which was significantly higher than in lighter fractions like diesel and SRGO, where total nitrogen content was not higher than a few hundred ppm [17,33]. A H/C ratio of 1.47 was found, which was similar to H/C ratios around 1.5 reported for other heavy fuels, in accordance with their high aromaticity [2,34].
By centrifugation of the diluted sample, 16 wt.% of solid particles were removed; however, the absolute sulphur content of the liquid phase RMG380-DC was found unchanged, indicating that the majority of the sulphur-containing molecules were present in the soluble fraction of the raw fuel in dodecane. In other marine fuels IFO380, IFO500, and HSFO700, a similar mass of solid phase was reported, between 12 and 16 wt.%. For IFO380, characterized by low viscosity and a sulphur content of 0.76 wt.%, 75% of the sulphur present in the fuel was recovered in the liquid phase. In contrast, for fuels with higher viscosity and sulphur content (>2.9 wt.% S), only 43% and 35% of the sulphur were retained in the liquid phases of IFO500-DC and HSFO700-DC, respectively. These results suggest that in these higher viscosity fuels, precipitated asphaltenes and/or solid particles contain a greater proportion of sulphur compared to IFO380 and RMG380.

2.1.2. Analysis of Sulphur-Containing Compounds by GC SCD of RMG380-DC Before and After Oxidation

The chromatogram obtained for RMG380-DC is shown in Figure 1, where alkyl-benzothiophenes (Cx-BTs) elute between 15 and 27 min, and alkyl-dibenzothiophenes (Cx-DBTs) appear between 27 and 40 min. Beyond 40 min, larger molecules elute as unresolved peaks. Some of them can be identified using 2D GC coupled with an SCD detector such as benzonaphthothiophene-type compounds found in various fuel oils [35,36,37]. In this retention time range, a significant baseline rise, attributed to co-eluting compounds, is observed, forming Unresolved Complex Mixtures (UCM). As shown by 2D GC analysis, this UCM likely includes high-molecular-weight sulphides and thiols that may co-elute with alkyl-dibenzothiophenes and naphthobenzothiophenes [38,39].
Quantification of the different families obtained from GC areas shows around 20% of Cx-BTs, 23% of Cx-DBTs, and 55% of UCM and sulphur-containing compounds eluting after 40 min. These values appear similar to those obtained with other marine fuels such as IFO380, LSMFO380 when diluted and centrifuged under the same conditions, indicating that the liquid phases recovered are similar, even when starting from different marine fuels.
After the ODS reaction, the peaks corresponding to sulphur species have clearly decreased, with the simultaneous appearance of peaks attributed to sulfones in the retention time region above 29 min (Figure 1). Beyond 29 min, a mixture of sulphur and sulfone species is present, making it challenging to accurately calculate the conversion of these compounds. Therefore, the conversion of sulphide species can be obtained by comparing the area of the peaks before and after oxidation in the region below 29 min. It is possible to follow the evolution of the quantities of Cx-BTs compounds as no sulfone peaks interfere with the range of their retention times. The peak areas corresponding to the five alkylated BT families are integrated (Figure 2), and the global conversion of Cx-BTs is calculated from the sum of these five regions.
Cx-BT sulfone peaks appear within the Cx-DBT retention region, resulting in overlapping peaks from both families. Only the evolution of the DBT molecule can be distinctly monitored during ODS, as its peak, located at the beginning of the alkyl-DBT retention range (28.7 min, Figure 2), does not overlap with sulfone peaks.

2.2. Effect of Calcination and Mo Loading on xMo/Al2O3 Catalysts

The effects of Mo loading and calcination temperature during catalyst synthesis were studied on the performance of the catalyst in the ODS of RMG380-DC.

2.2.1. Effect of the Calcination Step

Calcination parameters were varied, the corresponding catalysts being denoted 20Mo/Al(x,y,z), with x being the temperature ramp in °C·min−1, y the temperature of the plateau in °C, and z its duration in hours. The first part of the study consisted of increasing the plateau temperature from 350 to 550 °C. An increase in performance is first noted up to 480 °C, which corresponds to the optimum plateau temperature (Table 1). The corresponding Raman mappings of the extrudates’ cross-section show mainly polymolybdates, with some MoO3 crystallites appearing above 350 °C (Figure 3). Increasing the plateau duration from 3 to 5 h led to more MoO3 crystallites, concomitant with a slight decrease in performance. Raising the temperature ramp from 1 to 5 °C·min−1 has the most significant effect: the extrudates are surrounded by a shell of MoO3, which appears to be detrimental to the conversion, falling to 40% for the DBT against 60% observed for the best performance.

2.2.2. Effect of Mo Loading

Characterization of the catalysts
The previously determined optimal calcination conditions (1 °C·min−1, 480 °C, 3 h) were applied to prepare a series of Mo/Al2O3 catalysts with Mo loadings between 5 and 30 wt.% MoO3. The Mo content was confirmed by XRF analysis (Table 2).
The measured BET surface areas and the pore volume were corrected for the weight gain due to molybdenum incorporation (Table 2). No significant modification is observed due to molybdenum impregnation. The catalysts show a very similar specific surface area to that of the bare support, indicating little effect of the impregnation process.
XRD analysis of the xMo/Al catalysts shows only the peaks corresponding to the alumina support (at 38°, 45°, and 67°) for the 5Mo/Al and 10Mo/Al catalysts, while the major peaks at 12°, 23°, 26°, 27.5°, and 33°, attributed to crystalline MoO3, began to appear on 15Mo/Al with increasing intensity up to 30Mo/Al (Figure 4).
Raman spectra confirm this evolution, with only polymolybdate species detected at 948 cm−1 on 5Mo/Al and 10Mo/Al. On 15Mo/Al, together with polymolybdates, some analysed grains reveal the presence of MoO3 with its major peaks at 991, 812, 668, and 285 cm−1, which become predominant at higher Mo loadings (Figure 4).
Thermogravimetric analysis (TGA) was performed on all solids, and the weight loss and derivative curves are presented in Figure 5, while the unzoomed version can be found in the Supporting Information (Figure S1). In all samples, the first weight loss observed in the range of 25–250 °C is primarily due to the removal of physisorbed water held on the surface by hydrogen bonding. For the catalysts and alumina sample, the second weight loss in the range of 300–700 °C corresponds to the loss of water due to dehydroxylation of the support surface. For alumina and 5Mo/Al, small mass losses are observed on the TGA curve, although they are not clearly identifiable on the derivative curve (DTG). This absence of a distinct signal in the DTG curve is attributable to the gradual nature of the mass loss, which occurs diffusely over a broad temperature range. A lower mass loss is observed on 5Mo/Al (~3%) compared to the bare alumina support (~7%) and can be explained by the ion exchange process occurring during the impregnation process. The presence of molybdenum may have reduced the surface hydroxyl concentration on the alumina, meaning less mass is lost during thermal treatment, since fewer hydroxyl volatile species are released.
For the solids with higher Mo content, the DTG analysis shows a peak recorded between 800 and 925 °C, with a maximum measured between 876 and 925 °C for all solids. For comparison, TGA on pure MoO3 is also performed in the same conditions. The corresponding derived weight shows two main weight losses: an intense one at 856 °C corresponding to MoO3 sublimation, occurring after its melting at 795 °C, and a less intense one at 931 °C attributed in the literature to MoO3 in interaction with the alumina crucible walls and impurities [40]. Regarding these results, the observed peak for xMo/Al catalysts in the region around 900 °C is attributed to the sublimation of MoO3 in interaction with the alumina support. The nature of the interaction varies according to the type of support (e.g., ceria or alumina) and may involve the formation of Al-O-Mo bonds, as reported in the literature [41]. Consequently, this interaction induces slight modifications in the thermal stability of MoO3 observed by our TGA analysis.
The weight loss obtained by the mass difference between 800 and 1000 °C evolves linearly with the molybdenum content up to 20 wt.%, with a 5% difference from the theoretical weight loss if all molybdenum on the solid is considered (Figure 5). Thus, it can be hypothesized that 5% of the molybdenum is present as well-dispersed polymolybdates at these loadings and is in strong interaction with the support [42,43]. This is in agreement with the results of Ibrahim et al. [44], who observed that 5% of MoO3 remained in their solid containing initially 12.4 wt.% MoO3. On 30Mo/Al, the weight loss decreases compared to other solids, as it represents only 2/3 of the expected value. Besides molybdate species in strong interaction with Al2O3, some molybdenum could react with the support structure to form an Al2(MoO4)3 phase. Indeed, as observed by Ibrahim et al. [44] when supporting MoO3 on Al(OH)3, the presence of MoO3 on the support led to the formation of Al2(MoO4)3 at a temperature between 500 and 700 °C. In our work, this formation is supported by the endothermic phenomenon occurring at 650 °C observed on the DSC analysis only for a MoO3 content greater than 15 wt.% (Figure S2) ([45,46]).

2.3. Deactivation of xMo/Al Catalysts in the ODS of RMG380-DC

As mentioned earlier, strong adsorption (or no desorption) of ODS products can poison the active sites and lead to catalyst deactivation. Along the same lines, the presence of water and its adsorption as well as the active metal leaching could also be responsible for the catalyst deactivation. Multiple factors can influence catalyst deactivation; yet, it is challenging to discern the specific contribution of each factor. To investigate the potential influence of Mo surface speciation on the deactivation process, two catalysts with distinct surface compositions, 10Mo/Al (polymolybdates) and 20Mo/Al (polymolybdates + MoO3 crystallites), are used in the study of deactivation mechanisms.

2.3.1. Deactivation Tests on 10Mo/Al and 20Mo/Al

Resistance to deactivation of the 10Mo/Al and 20Mo/Al catalysts is evaluated in four consecutive tests, with intermediate filtration and drying at 90 °C overnight for the catalyst between each test (Figure 6). The same behaviour is observed for the three recovered solids, with a continuous decrease in conversion for the first three tests and a stabilization of the values for the fourth one. The conversion for 10Mo/Al evolves from around 60% to 40% for DBT, and from 20% to 12% for the Cx-BTs family. Conversely, the loss of activity appears to be more pronounced on the 20Mo/Al catalyst, with a decrease from 62% to 32% for the conversion of DBT.

2.3.2. Characterization of the Used Catalysts

The catalysts recovered after tests are characterized to identify the possible causes of the decrease in activity.
Attrition of the catalyst extrudates
Attrition of the extrudates can result from mechanical stirring in the reaction medium, leading to the presence of powder that may not be recovered during filtration. Indeed, the mass of recovered powder after filtration and drying decreases from 0.6 g in the initial test to 0.45 g after the fourth test for both 10Mo/Al and 20Mo/Al. To check if this loss of mass is responsible for the decrease in activity observed in the consecutive tests, a reaction was performed with 0.45 g of fresh catalyst for both 10Mo/Al and 20Mo/Al. In these conditions, the conversions of Cx-BTs and DBT are quite similar to those measured in the fourth run on 10Mo/Al, showing that for this catalyst, the loss of activity is due predominantly to the attrition of the extrudates and to the subsequent loss of catalyst mass during the filtration step (Figure 7). However, for 20Mo/Al, conversions of Cx-BTs and DBT are significantly higher with 0.45 g of fresh catalyst than after four runs using the same mass of solid (50% vs. 34% for DBT, and 23% vs 15% for Cx-BTs). To explain this loss of activity, other causes must be investigated when higher molybdenum content is considered.
Leaching of molybdenum
Active-phase leaching in the reaction medium may be at the origin of the loss of activity [16]. Molybdenum loading is thus measured on the catalyst filtered after the reaction, dried overnight at 90 °C, and calcined for 2 h at 550 °C to remove the solvent and any adsorbed compounds. Table 3 shows no molybdenum loss during the test, with similar contents before and after the reaction. This is not surprising, as the leaching observed in the study of Chica et al. [16] is due to a continuous flow process. Using a batch catalytic system, Tian et al. [47] observed a decrease in the leaching proportion during successive batch tests with a model fuel. They reported that catalytic activity is maintained due to low Mo leaching. This finding aligns with our results, which demonstrate minimal or no leaching in a batch process, thereby suggesting that deactivation is not attributable to metal loss.
Adsorption of sulfur-containing compounds on the catalysts
CHNS analysis on 10Mo/Al and 20Mo/Al after one run shows similar sulphur contents around 0.5 wt.%, corresponding to sulphur retention around 4% (Table 4), which is in agreement with Sulf-UV analysis of the liquid phase before and after the test (sulphur retention of 3.5% on 20Mo/Al). This quantity increases after four runs, with 7% retention on 20Mo/Al. For each used catalyst, the atomic H/C ratio is 1.5, indicating that the majority of the retained compounds are unsaturated. As an example, the H/C values for model molecules of DBT and BT sulfones, DBTO2 and BTO2, are 0.67 and 0.75, respectively, while their S/C ratios are 0.08 and 0.12, respectively. The experimental values of these ratios are far from the ones obtained for BTO2 and DBTO2; this suggests that the main components adsorbed do not contain sulphur and are probably polyaromatic. These results align with the lower retention of sulfones observed when a real charge is used [27]. Indeed, non-sulphur compounds present in a real charge can increase the polarity of the liquid phase, favouring the solubility of sulfones and decreasing their interaction with the surface of the catalyst.
The TG and DTG curves of both 10Mo/Al and 20Mo/Al after one test, presented in Figure 8, evidenced a weight loss in the temperature range below 500 °C, possibly corresponding to adsorbed components on the solid during the reaction, which could be responsible for catalyst deactivation. As previously noted, studies [27] have demonstrated that the use of real feedstock can reduce the amount of retained sulfones, with only 4% of introduced sulphur in our experiment. Consequently, the adsorbed compounds are not predominantly sulfone compounds. In another work, Barman et al. [48] evaluate the oil and coke contents on spent catalysts used in a reprocessing unit of heavy fuel oil. Using TGA, they observe two separate ranges of weight loss, from RT to 300 °C attributed to residual oil and from 300 °C to 700 °C attributed to coke and asphaltenes. In their study, coke and asphaltenes are desorbed in the same region as the large loss of weight observed in our experiment, suggesting that these kinds of compounds could be present at the surface.
On the other hand, the DTG curves also indicate an important change in the temperature zone compared to those obtained for fresh catalysts. The peak initially at 908 °C and 920 °C for 10Mo/Al and 20Mo/Al is shifted to 807 °C and 832 °C, respectively. The sublimation temperature of MoO3 is significantly lowered and approaches that of unsupported MoO3. This suggests that the presence of adsorbed compounds decreases the interaction between the molybdenum phase and the catalyst support. On 10Mo/Al, the corresponding weight loss of 5.2% is similar to that of the fresh solid. However, the weight loss of 9.5% observed for 20Mo/Al is significantly lower than that of the fresh solid (15.9%). This could indicate an evolution of the molybdenum-based phase during the test performed with high Mo loadings.
Modification of catalyst surface species
Regeneration temperature to remove coke varies between 400 and 750 °C. From TGA results, most of the carbonaceous deposit can be removed at or above 500 °C. To discriminate whether the adsorbed compounds and/or the evolution of the molybdenum phase impact the performance, both solids are calcined at 550 °C for two hours under air to regenerate the catalysts. TGA analysis, with the plot of the derivative curves, confirms that all adsorbed species are removed by the thermal treatment (Figure 8).
After calcination on 10Mo/Al, the peak attributed to molybdenum is observed back at the same temperature as for the fresh solid, with a similar weight loss (4.3% vs. 4.8% on the fresh solid). This indicates the recoverability of the interaction between the molybdenum phase and the support when adsorbed compounds are removed. The performance of this regenerated catalyst corresponds to that of the fresh one, with Cx-BTs and DBT conversion at 20% and 62.3%, respectively. This demonstrates the efficiency of calcination to recover the initial active phase.
On 20Mo/Al, the peak attributed to molybdenum loss is also back to its initial position, but it shows a shoulder at a slightly lower temperature (866 °C). The weight loss remains the same as that on the used catalyst (9.5%), indicating a loss of molybdenum either in the support or in the formation of a non-volatile phase. The evolution of the molybdenum phase that occurred during the test is thus not entirely recoverable. This is confirmed by the activity of the regenerated solid, which remains similar to that of the used catalyst (Figure 9).
To better understand why the initial activity is not retrieved on the 20Mo/Al, XRD (Figure 10) and Raman spectroscopy (Figure 11) are performed on the regenerated catalysts. For the 10Mo/Al, the comparison analysis of the diffractograms does not indicate any change in the catalyst, as only diffraction peaks of the support are observed for the fresh and regenerated catalyst. On the contrary, for the 20Mo/Al, some changes are observed, with the MoO3 diffraction peaks absent on the regenerated catalyst diffractogram. Instead, small peaks corresponding to a crystallized Al2(MoO4)3 are observed. The presence of this phase is confirmed by Raman spectroscopy (Figure 11) and seems to be detrimental to the catalytic activity. Thus, this suggests that the adsorbed compounds are not the only contributors to the decreased activity; the quantity of molybdenum and its interaction with the support seem to be of strong importance for the catalytic performance.
Oxidant impact on the deactivation process
To gain a deeper understanding of the deactivation observed in the 20Mo/Al catalyst, the study is also conducted using a non-aqueous oxidant, tBHP. Figure 12 shows that by using the catalyst for four successive tests with tBHP, the Cx-BTs and DBT conversions decreased from 30% to 20%, and from 40% to 21%, respectively. As for the catalyst used with H2O2, the characterization of the catalyst using TGA indicates the presence of adsorbed compounds, a modification of the content of MoO3 sublimated (15.9 wt.% to 10.5 wt.%) and a decrease in its sublimation temperature from 918 °C to 812 °C (Figure 12). The change of oxidant does not change the deactivation phenomenon, as the evolution of Cx-BTs and DBT conversions are similar to the one observed with H2O2 tests, only with a lower DBT conversion.
To more thoroughly investigate the deactivation process, the standard regeneration protocol is also applied to a used catalyst. As expected, adsorbed compounds are completely removed from the catalyst surface after calcination, as no weight loss is observed below 500 °C in the TGA analysis. However, unlike the regenerated 20Mo/Al catalyst treated with H2O2, this catalyst exhibits a similar amount of volatile MoO3 compared to the fresh catalyst (15.3 wt.% compared to 15.9 wt.%). More unexpectedly, the initial sublimation temperature of MoO3 is not retrieved. The sublimation temperature of MoO3 after regeneration is observed at 872 °C, which is lower than the 918 °C observed for the fresh catalyst. The impact on catalytic activity is illustrated in Figure 12. After regeneration of the catalyst after the four successive tests, the conversion of DBT increased to 46%, slightly higher than the initial conversion of 42%. Remarkably, the conversion of Cx-BTs significantly improved, reaching 41% after calcination compared to 30% with the fresh catalyst. Attrition is also observed in this case, and the catalytic performance of the regenerated catalyst is achieved with a lower catalyst quantity, demonstrating an even better performance relative to the initial one.
Characterization of the catalyst using Raman spectroscopy reveals the presence of Al2(MoO4)3, albeit to a lesser extent and primarily coexisting with polymolybdate species (Figure S3). The high amount of Al2(MoO4)3 previously observed when using H2O2 as an oxidant could be explained by a formation favoured by the presence of water. The presence of a large amount of water could lead to support hydrolysis leading to an increased availability of Al3+. Moreover, unlike the catalyst used with H2O2, there are no MoO3 crystallites detected on the several grains analysed for the used catalyst with tBHP. These results seem to indicate that the MoO3 crystallites are not necessarily active for the ODS reaction with tBHP. Finally, an intermediate interaction strength, illustrated by an intermediate sublimation temperature, seems to increase conversion performance.

3. Discussion

3.1. Active Phase

The surface speciation of molybdenum is very broad, and multiple species are present in the fresh, used, and regenerated catalysts, depending also on the Mo loadings. With low Mo contents (5 and 10 wt.% MoO3), the surface species identified as highly dispersed polymolybdates are very active in the oxidation reaction. This was previously observed with either molybdenum [49] or tungsten (polytungstate) [50] as active elements. Formation of crystalline phases is often reported to be detrimental, as it results in poorer molybdenum dispersion, reduced interaction with the support, and blockage of active sites. The negative impact of the crystalline phase within the support is demonstrated by the study on calcination parameters. Specifically, the formation of a MoO3 crust around the extrudate appears to hinder catalytic activity. Thereby, it highlights the necessity for accessibility to the active polymolybdates and the need for optimal loading and interaction with the alumina support.
Intriguingly, the behaviour of the 20Mo/Al catalyst is dependent on the type of oxidant used, and its surface speciation changes after its regeneration using two different oxidants. The fresh 20Mo/Al2O3 catalyst exhibits heterogeneous surface speciation with polymolybdates and MoO3 crystallites. The catalyst used with H2O2 and further calcined contains polymolybdates, MoO3 crystallites, and an Al2(MoO4)3 phase. This catalyst is less active than the fresh one and thus demonstrates the negative impact of the presence of an inactive species, Al2(MoO4)3, as observed in other studies [51]. Conversely, on the catalyst used with tBHP and further calcined, polymolybdates and a small amount of Al2(MoO4)3 are identified. For this regenerated catalyst that no longer contains MoO3 crystallites, activity is greater than its fresh equivalent. This observation supports the assertion that MoO3 induces limited or negligible activity in the ODS of Heavy Fuel Oil. Moreover, the difference in activity between the fresh and regenerated catalysts may not only be due to the nature of the active phase but also to its interaction with the support. Indeed, the strength of this interaction, illustrated in our study by the sublimation temperature of MoO3 in TGA analysis, decreased during the regeneration of a catalyst used with tBHP. Catalytic activity can be significantly impacted by modifying the interaction strength between the molybdenum phase and the support, as observed by changing the support [52,53] or by adding a dopant [54,55]. In our case, the use of an organic oxidant and regeneration by calcination allows for the modification of this interaction between the molybdenum phase and the support, thereby ensuring better oxidation, particularly of the more refractory compounds, Cx-BTs.

3.2. Nature of the Deactivation

The conducted study helps distinguish specific factors that can lead to a decrease in catalytic activity from one test to another. In our case, leaching of the active phase is not observed, but attrition is present, caused by mechanical agitation used in the batch reactor for catalytic tests. Regularly observed in batch studies, weight loss is attributed to attrition when contact occurs with the magnetic stirrer and/or the reactor walls. Attrition is a major contributor to the deactivation of the 10Mo/Al catalyst and does not depend on the molybdenum content or the oxidant used, as the same weight proportion is lost with 20Mo/Al used with H2O2 and tBHP. Consequently, to distinguish other deactivation phenomena, the catalysts in different states (fresh, used, and regenerated catalysts) were characterized when the attrition was minimal for the used state, after the first test. According to the literature [16], a major cause of the deactivation of ODS catalysts is the retention of sulfones in the catalyst’s active phase. However, in our study, the quantity of sulphur adsorbed (only 4 wt.% of S) indicates that sulfones are not the main compounds adsorbed onto the surface. This is attributed to the nature of the cut, which is more polar, thereby enhancing the solubility of sulfones in the media. Conversely, the presence of non-sulphur compounds, such as high-molecular-weight polyaromatic hydrocarbons, appears significant, as illustrated by the H/C ratio in CHNS analysis and the volatilization temperature in TGA. This phenomenon is rarely studied in ODS literature, as most studies use model or light cuts with no or low content of heavy polyaromatic compounds. In the case of heavy products, carbonaceous residues can cover the active surface or block the pores by coke deposition and thus lead to diminished activity. Coke deposits can account for up to 15–20% (w/w) of the catalyst [56]. The nature of coke deposits is not well defined and depends on different factors such as the initial hydrocarbons, the reaction conditions, the nature of the products, etc. However, several authors pointed out a direct relationship between the amount of coke deposited and the aromatic and polynuclear aromatic content of the feed [57].
Consequently, the adsorption of heavy compounds on the surface, the interaction of the Mo phase strongly decreases with the support, as shown by the lower sublimation temperature of the molybdenum phase in the TGA analysis. The regeneration step highlights the importance of this interaction, as when it is retrieved, the activity is also recovered for “low” molybdenum content (i.e., 10Mo/Al). However, a too-high content of molybdenum and, overall, the presence of MoO3 phase at the surface prevent the recovery of the initial interaction mainly due to a change in molybdenum phases. Indeed, we have illustrated the presence of the detrimental Al2(MoO4)3 at the surface consequent to the regeneration step. The latter phase is formed by a solid-state reaction between alumina and MoO3 [46], or by the calcination of a previously formed Anderson heteropolyanion [Al(OH)6Mo6O18]3− intermediate [58]. The quantity of Al2(MoO4)3 is more significant when using H2O2 as an oxidant. This could be explained by the presence of water in the reaction medium, which may promote the hydration or dissolution of alumina, thereby making aluminium atoms more readily available or enhancing the formation of the Anderson anion, and thus facilitating the formation of aluminium molybdate during thermal treatment. The presence of water could thus be a factor of deactivation, indicating the importance of designing the catalyst to minimize the adsorption of heavy aromatic compounds, as well as avoiding its surface modification by water. Finally, the study performed with tBHP also shows that the interaction of the active phase with the support is of tremendous importance. An intermediate strength of interaction leads to increased ODS performance. The presence of an organic oxidant seems to redisperse crystallites of MoO3 during the test and to minimize the presence of Al2(MoO4)3 during the regeneration process. This behaviour has already been observed for HDS catalysts [59], where organic additives are used to reduce the interaction strength between the metallic precursors and the support surface, stabilize the surface aluminium atoms, and reduce the formation of refractory phases to sulfidation, such as [Al(OH)6Mo6O18]3−- based crystalline phase to improve the activity. These results need further exploration in order to understand how the active phase can be monitored to achieve major catalytic performance improvements.

4. Materials and Methods

The marine fuel RMG380 used in this study was provided by SEGULA Technologies (Montoir de Bretagne, France). Hydrogen peroxide (27 wt.% in water), tert-butyl hydroperoxide (5.0–6.0 M in decane), ammonium molybdate tetrahydrate (puriss., ≥99%), and n-dodecane (≥99%, Alfa Aesar™, Haverhill, MA, USA) were purchased from Sigma Aldrich (St. Louis, MO, USA) and Fisher Scientific (Hampton, NH, USA), respectively.

4.1. Catalyst Preparation

Molybdenum-based catalysts supported on Al2O3 were synthesized via incipient wetness impregnation of extrudates of industrial alumina (177 m2·g−1) with a solution of ammonium heptamolybdate. The maturation step of 3 h under a humid atmosphere was followed by drying at 90 °C overnight and calcination at 480 °C for 3 h under airflow (0.3 L·min−1) with a temperature rate of 1 °C·min−1 (except as otherwise stated in the calcination parameters study). Molybdenum final loading in the calcined catalyst was set at 10 and 20 wt.% of MoO3. The obtained solids were referred to as xMo/Al, with x being the weight percent in MoO3.

4.2. Catalysts Characterization

Textural properties were investigated based on nitrogen physisorption at −196 °C on an automated gas adsorption analyser, Tristar 3020 (Micromeritics, Norcross, GA, USA). The samples were degassed at 150 °C in a vacuum for 3 h.
X-ray fluorescence analysis (XRF) was performed on an S2 Ranger (Bruker, Billerica, MA, USA) apparatus.
X-ray diffractograms were acquired on a Siemens D5000 diffractometer (Siemens, Munich, Germany) using Cu Kα radiation, in a range of 2θ from 10 to 90° with a 0.02° step size and 2 s step time.
Raman spectra were recorded on a Horiba Xplora apparatus (Horiba, Kyoto, Japan), equipped with a Nd-YAG laser. A laser wavelength of 532 nm with 1% or 10% filters was used. Raman mapping was performed on an 800 × 1000 µm zone of a section of an extrudate with 5 s of accumulation. The data treatment was performed using LabSpec 6 software. Classical Least Squares (CLS) fitting method with reference spectra of MoO3 and polymolybdates was used to attribute a colour to the major species in the mix and helped to visualize the distribution in the extrudates.
Fresh, used, and regenerated catalysts were characterized by TGA on an SDT Q600 apparatus. The samples were analysed under a nitrogen flow up to 1000 °C with a temperature ramp of 10 °C/min.

4.3. Analysis of the Fuels

Elemental C, H, N, and S contents of marine fuels were determined using CHNS analysis on a Thermo Fisher EA1110 Flash instrument (Thermo Fisher Scientific, Waltham, MA, USA). The overall sulphur content was additionally measured by UV fluorescence (Sulf-UV) using an ANTEK 9000S analyser (AlyTech, Juvisy-sur-Orge, France), with samples stirred at 80 °C for 1 h prior to injection. Sulphur-containing compounds were analyzed by gas chromatography on a Varian 3800 equipped with a Sievers SCD 355 (Sulphur Chemiluminescence Detector) (Agilent Technologies, Santa Clara, CA, USA). A high-temperature Phenomenex ZB-5HT column (Agilent Technologies, Santa Clara, CA, USA), coupled with an integrated guard column due to the heavy nature of marine fuel samples, was used. Optimal conditions for analysis included an initial temperature of 50 °C, ramped to 260 °C at a rate of 5 °C/min.

4.4. Dilution and Centrifugation of the Fuel

Due to its high viscosity, the fuel RMG380 was diluted 5 times in dodecane (RMG380-D). It was then centrifuged for 20 min at 8000 rpm to separate precipitated particles. The liquid phase recovered after centrifugation was used for the ODS tests (RMG380-DC).

4.5. Oxidative Desulfurization Protocol

The ODS tests were carried out in a 100 mL glass batch reactor at atmospheric pressure. A total of 30 g of diluted centrifuged fuel RMG380-DC was introduced into the reactor together with the oxidant hydrogen peroxide H2O2 (oxidant to sulphur ratio of 5). Mixing at 1300 rpm was performed prior to the addition of 600 mg of catalyst in extrudate shape. The system was heated to 80 °C under reflux, with careful attention given to the selection and optimization of mixing conditions due to the heavy nature of the feedstock [32]. The ODS test was carried out for one hour. Then, the catalyst was recovered by filtration and dried overnight at 100 °C for reuse in subsequent catalytic tests.
The percentage of sulphur retained on the catalyst was calculated as the difference between the sulphur content in the initial solution and that in the reaction solution post-filtration.

5. Conclusions

The characterization of 10Mo/Al and 20Mo/Al catalysts through BET, XRD, Raman spectroscopy, TGA, and their catalytic evaluation in the ODS of a marine fuel RMG380 provided valuable insights into the active phase involved in the ODS process and contributed to identifying the factors involved in activity loss and catalyst deactivation. The study emphasizes the importance of well-dispersed polymolybdates on the alumina surface and their interaction with the support. Excessive molybdenum loading leads to the formation of poorly dispersed crystalline MoO3, which exhibits low or no activity in ODS and can even hinder catalyst regeneration. Moreover, this study allowed us to distinguish the different phenomena causing activity loss through the characterization of used catalysts and their regeneration. It shows that attrition and subsequent mass loss between two successive tests account for the activity decrease observed on 10Mo/Al but not for 20Mo/Al, underscoring the critical role of molybdenum content in deactivation. Active phase leaching and sulfone adsorption are ruled out as major deactivation mechanisms, while they are typically considered primary deactivation routes in lighter or model feedstocks. Indeed, sulfone solubility is higher in heavy fuel oil than in model or light feeds, thus minimizing their impact on catalyst deactivation. The presence of polyaromatics, on the other hand, likely plays a crucial role, possibly in obstructing catalyst pores and altering molybdenum-support interactions. Finally, the study highlights the challenges in catalyst regeneration. While recovery of initial activity is observed for 10Mo/Al, the 20Mo/Al catalyst formed irreversible phases like Al2(MoO4)3 that prevented regeneration. Interestingly, using an organic oxidant such as tBHP on the 20Mo/Al catalyst resulted in better performance after regeneration, with reduced Al2(MoO4)3 formation and the absence of MoO3 crystallites in the regenerated solid. This underscores the role of water in deactivation and paves the way for future research on catalyst design.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/catal14110823/s1, Figure S1: TG curves of Al2O3, MoO3, and the xMo/Al catalysts obtained by TGA. Figure S2: DSC curves of xMo/Al in the 200–700 °C region; Figure S3: Raman spectra collected from different regions on 20Mo/Al2O3 catalysts, after the test with tBHP + calcination.

Author Contributions

J.A.: investigation and review; T.R.: investigation, supervision, writing—review and editing; L.P.: project administration, funding acquisition, review; C.L. (Christine Lancelot): supervision, writing—review and editing; P.B.: supervision and review; C.L. (Carole Lamonier): supervision and review. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

Acknowledgments

Chevreul Institute (FR 2638), Ministère de l’Enseignement Supérieur et de la Recherche, Région Nord—Pas de Calais, and FEDER are acknowledged for supporting and funding partially this work. We gratefully acknowledge Martine Trentesaux, who conducted Raman cartography analysis, and Olivier Gardoll, who conducted TGA analysis.

Conflicts of Interest

Authors Teddy Roy, Joy Alakari and Line Poinel are employed by the company SEGULA Technologies. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Halff, A.; Younes, L.; Boersma, T. The likely implications of the new IMO standards on the shipping industry. Energy Policy 2019, 126, 277–286. [Google Scholar] [CrossRef]
  2. Rana, M.S.; Sámano, V.; Ancheyta, J.; Diaz, J.A.I. A review of recent advances on process technologies for upgrading of heavy oils and residua. Fuel 2007, 86, 1216–1231. [Google Scholar] [CrossRef]
  3. Stanislaus, A.; Marafi, A.; Rana, M.S. Recent advances in the science and technology of ultra low sulfur diesel (ULSD) production. Catal. Today 2010, 153, 1–68. [Google Scholar] [CrossRef]
  4. Li, M.; Zhang, M.; Wei, A.; Zhu, W.; Xun, S.; Li, Y.; Li, H.; Li, H. Facile synthesis of amphiphilic polyoxometalate-based ionic liquid supported silica induced efficient performance in oxidative desulfurization. J. Mol. Catal. Chem. 2015, 406, 23–30. [Google Scholar] [CrossRef]
  5. Xun, S.; Zhu, W.; Zhu, F.; Chang, Y.; Zheng, D.; Qin, Y.; Zhang, M.; Jiang, W.; Li, H. Design and synthesis of W-containing mesoporous material with excellent catalytic activity for the oxidation of 4,6-DMDBT in fuels. Chem. Eng. J. 2015, 280, 256–264. [Google Scholar] [CrossRef]
  6. Zheng, H.; Sun, Z.; Chen, X.; Zhao, Q.; Wang, X.; Jiang, Z. A micro reaction-controlled phase-transfer catalyst for oxidative desulfurization based on polyoxometalate modified silica. Appl. Catal. Gen. 2013, 467, 26–32. [Google Scholar] [CrossRef]
  7. Fernandes, S.C.; Leo, P.; Santos-Vieira, I.; de Castro, B.; Cunha-Silva, L.; Balula, S.S. Improving Oxidative Catalytic Efficiency for Fuels Desulfurization Using Hybrid Materials Based in MOF-808@SBA-15. ChemCatChem 2024, 16, e202400355. [Google Scholar] [CrossRef]
  8. Ishihara, A.; Wang, D.; Dumeignil, F.; Amano, H.; Qian, E.W.; Kabe, T. Oxidative desulfurization and denitrogenation of a light gas oil using an oxidation/adsorption continuous flow process. Appl. Catal. Gen. 2005, 279, 279–287. [Google Scholar] [CrossRef]
  9. Otsuki, S.; Nonaka, T.; Takashima, N.; Qian, W.; Ishihara, A.; Imai, T.; Kabe, T. Oxidative Desulfurization of Light Gas Oil and Vacuum Gas Oil by Oxidation and Solvent Extraction. Energy Fuels 2000, 14, 1232–1239. [Google Scholar] [CrossRef]
  10. Chen, T.-C.; Shen, Y.-H.; Lee, W.-J.; Lin, C.-C.; Wan, M.-W. An economic analysis of the continuous ultrasound-assisted oxidative desulfurization process applied to oil recovered from waste tires. J. Clean. Prod. 2013, 39, 129–136. [Google Scholar] [CrossRef]
  11. Mirshafiee, F.; Movahedirad, S.; Sobati, M.A.; Alaee, R.; Zarei, S.; Sargazi, H. Current status and future prospects of oxidative desulfurization of naphtha: A review. Process Saf. Environ. Prot. 2023, 170, 54–75. [Google Scholar] [CrossRef]
  12. Prasad, V.V.D.N.; Jeong, K.-E.; Chae, H.-J.; Kim, C.-U.; Jeong, S.-Y. Oxidative desulfurization of 4,6-dimethyl dibenzothiophene and light cycle oil over supported molybdenum oxide catalysts. Catal. Commun. 2008, 9, 1966–1969. [Google Scholar] [CrossRef]
  13. García-Gutiérrez, J.L.; Fuentes, G.A.; Hernández-Terán, M.E.; García, P.; Murrieta-Guevara, F.; Jiménez-Cruz, F. Ultra-deep oxidative desulfurization of diesel fuel by the Mo/Al2O3-H2O2 system: The effect of system parameters on catalytic activity. Appl. Catal. Gen. 2008, 334, 366–373. [Google Scholar] [CrossRef]
  14. Caero, L.C.; Hernández, E.; Pedraza, F.; Murrieta, F. Oxidative desulfurization of synthetic diesel using supported catalysts: Part I. Study of the operation conditions with a vanadium oxide based catalyst. Catal. Today 2005, 107–108, 564–569. [Google Scholar] [CrossRef]
  15. Abdullah, W.N.W.; Ali, R.; Bakar, W.A.W.A. In depth investigation of Fe/MoO3–PO4/Al2O3 catalyst in oxidative desulfurization of Malaysian diesel with TBHP-DMF system. J. Taiwan Inst. Chem. Eng. 2016, 58, 344–350. [Google Scholar] [CrossRef]
  16. Chica, A.; Corma, A.; Dómine, M.E. Catalytic oxidative desulfurization (ODS) of diesel fuel on a continuous fixed-bed reactor. J. Catal. 2006, 242, 299–308. [Google Scholar] [CrossRef]
  17. Cho, K.-S.; Lee, Y.-K. Effects of nitrogen compounds, aromatics, and aprotic solvents on the oxidative desulfurization (ODS) of light cycle oil over Ti-SBA-15 catalyst. Appl. Catal. B Environ. 2014, 147, 35–42. [Google Scholar] [CrossRef]
  18. Khademian, F.; Bazyari, A.; Haghighi, P.; Thompson, L.T. Enhanced oxidative desulfurization of dibenzothiophene in the presence of cyclohexene, indole, quinoline, and p-xylene using surface-silylated titania-silica catalyst. Fuel 2023, 344, 128022. [Google Scholar] [CrossRef]
  19. Farshi, A.; Shiralizadeh, P. Sulfur reduction of heavy fuel oil by oxidative desulfurization (ODS) method. Pet. Coal 2015, 57, 1337–7027. [Google Scholar]
  20. Tang, Q.; Lin, S.; Cheng, Y.; Liu, S.; Xiong, J.-R. Ultrasound-assisted oxidative desulfurization of bunker-C oil using tert-butyl hydroperoxide. Ultrason. Sonochem. 2013, 20, 1168–1175. [Google Scholar] [CrossRef]
  21. Houda, S.; Lancelot, C.; Blanchard, P.; Poinel, L.; Lamonier, C. Ultrasound assisted oxidative desulfurization of marine fuels on MoO3/Al2O3 catalyst. Catal. Today 2021, 377, 221–228. [Google Scholar] [CrossRef]
  22. Vedachalam, S.; Boahene, P.; Dalai, A.K. Oxidative Desulfurization of Heavy Gas Oil over a Ti–TUD-1-Supported Keggin-Type Molybdenum Heteropolyacid. Energy Fuels 2020, 34, 15299–15312. [Google Scholar] [CrossRef]
  23. Ogunlaja, A.S.; Alade, O.S.; Tshentu, Z.R. Vanadium(IV) catalysed oxidation of organosulfur compounds in heavy fuel oil. Comptes Rendus Chim. 2017, 20, 164–168. [Google Scholar] [CrossRef]
  24. Rivoira, L.; Martínez, M.L.; Anunziata, O.; Beltramone, A. Vanadium oxide supported on mesoporous SBA-15 modified with Al and Ga as a highly active catalyst in the ODS of DBT. Microporous Mesoporous Mater. 2017, 254, 96–113. [Google Scholar] [CrossRef]
  25. Estephane, G. Développement de Catalyseurs à Base de Silice Mésostructurée et de Tungstène Pour L’oxydésulfuration de Charges Modèles et de Gazoles. Ph.D. Theses, Université de Lille, Lille, France, 2016. [Google Scholar]
  26. Akopyan, A.V.; Polikarpova, P.D.; Arzyaeva, N.V.; Anisimov, A.V.; Maslova, O.V.; Senko, O.V.; Efremenko, E.N. Model Fuel Oxidation in the Presence of Molybdenum-Containing Catalysts Based on SBA-15 with Hydrophobic Properties. ACS Omega 2021, 6, 26932–26941. [Google Scholar] [CrossRef]
  27. Estephane, G.; Lancelot, C.; Blanchard, P.; Toufaily, J.; Hamiye, T.; Lamonier, C. Sulfur compounds reactivity in the ODS of model and real feeds on W–SBA based catalysts. RSC Adv. 2018, 8, 13714–13721. [Google Scholar] [CrossRef]
  28. Jia, Y.; Li, G.; Ning, G.; Jin, C. The effect of N-containing compounds on oxidative desulphurization of liquid fuel. Catal. Today 2009, 140, 192–196. [Google Scholar] [CrossRef]
  29. Mokhtari, B.; Akbari, A.; Omidkhah, M. Superior Deep Desulfurization of Real Diesel over MoO3/Silica Gel as an Efficient Catalyst for Oxidation of Refractory Compounds. Energy Fuels 2019, 33, 7276–7286. [Google Scholar] [CrossRef]
  30. Tu, Y.; Li, T.; Yu, G.; Wei, L.; Ta, L.; Zhou, Z.; Ren, Z. Study on Modification and Desulfurization Performance of a Molybdenum-Based Catalyst. Energy Fuels 2019, 33, 8503–8510. [Google Scholar] [CrossRef]
  31. Mitra, S.; Racha, S.M.; Shown, B.; Mandal, S.; Das, A.K. New insights on oxidative desulfurization for low sulfur residual oil production. Sustain. Energy Fuels 2023, 7, 270–279. [Google Scholar] [CrossRef]
  32. Houda, S. Oxidative Desulfurization of Marine Fuels: Optimization of Reaction Conditions and Development of Efficient catalysts. Ph.D. Theses, Université de Lille, Lille, France, 2019. [Google Scholar]
  33. Palomeque-Santiago, J.F.; López-Medina, R.; Oviedo-Roa, R.; Navarrete-Bolaños, J.; Mora-Vallejo, R.; Montoya-de la Fuente, J.A.; Martínez-Magadán, J.M. Deep oxidative desulfurization with simultaneous oxidative denitrogenation of diesel fuel and straight run gas oil. Appl. Catal. B Environ. 2018, 236, 326–337. [Google Scholar] [CrossRef]
  34. Garaniya, V.; McWilliam, D.; Goldsworthy, L.; Ghiji, M. Extensive chemical characterization of a heavy fuel oil. Fuel 2018, 227, 67–78. [Google Scholar] [CrossRef]
  35. Hua, R.; Wang, J.; Kong, H.; Liu, J.; Lu, X.; Xu, G. Analysis of sulfur-containing compounds in crude oils by comprehensive two-dimensional gas chromatography with sulfur chemiluminescence detection. J. Sep. Sci. 2004, 27, 691–698. [Google Scholar] [CrossRef] [PubMed]
  36. Mahé, L.; Dutriez, T.; Courtiade, M.; Thiébaut, D.; Dulot, H.; Bertoncini, F. Global approach for the selection of high temperature comprehensive two-dimensional gas chromatography experimental conditions and quantitative analysis in regards to sulfur-containing compounds in heavy petroleum cuts. J. Chromatogr. A 2011, 1218, 534–544. [Google Scholar] [CrossRef] [PubMed]
  37. Lorentz, C.; Laurenti, D.; Zotin, J.L.; Geantet, C. Comprehensive GC×GC chromatography for the characterization of sulfur compound in fuels: A review. Catal. Today 2017, 292, 26–37. [Google Scholar] [CrossRef]
  38. Blomberg, J.; Riemersma, T.; van Zuijlen, M.; Chaabani, H. Comprehensive two-dimensional gas chromatography coupled with fast sulphur-chemiluminescence detection: Implications of detector electronics. J. Chromatogr. A 2004, 1050, 77–84. [Google Scholar] [CrossRef]
  39. Ruiz-Guerrero, R.; Vendeuvre, C.; Thiébaut, D.; Bertoncini, F.; Espinat, D. Comparison of comprehensive two-dimensional gas chromatography coupled with sulfur-chemiluminescence detector to standard methods for speciation of sulfur-containing compounds in middle distillates. J. Chromatogr. Sci. 2006, 44, 566–573. [Google Scholar] [CrossRef]
  40. Qing, Y.; Yang, K.; Chen, Y.; Zhu, J.; Li, Y.; Chen, C.; Li, Q.; Sun, B.; He, J. Thermal Stability, Optical and Electrical Properties of Substoichiometric Molybdenum Oxide. Materials 2023, 16, 2841. [Google Scholar] [CrossRef]
  41. Leung, Y.L.; Flinn, B.J.; Wong, P.C.; Mitchell, K.A.R.; Smith, K.J. A surface spectroscopic study of calcinated molybdenum-aluminum oxide model catalysts. Appl. Surf. Sci. 1998, 125, 293–299. [Google Scholar] [CrossRef]
  42. Drake, T.L.; Stair, P.C. Multiwavelength Raman Spectroscopic Characterization of Alumina-Supported Molybdenum Oxide Prepared by Vapor Deposition. Top. Catal. 2017, 60, 1618–1630. [Google Scholar] [CrossRef]
  43. Tian, H.; Roberts, C.A.; Wachs, I.E. Molecular Structural Determination of Molybdena in Different Environments: Aqueous Solutions, Bulk Mixed Oxides, and Supported MoO3 Catalysts. J. Phys. Chem. C 2010, 114, 14110–14120. [Google Scholar] [CrossRef]
  44. Ibrahim, A.A.; El-Shobaky, G.A. Solid-solid interactions in the MoO3-Al2O3, system. Thermochim. Acta 1989, 147, 175–188. [Google Scholar] [CrossRef]
  45. Cordero, R.L.; Guerra, S.L.; Fierro, J.L.G.; Agudo, A.L. Formation of Al2(MoO4)3 and MoO3 phases induced by phosphate in molybdena-phosphorus catalysts. J. Catal. 1990, 126, 8–12. [Google Scholar] [CrossRef]
  46. Matsumoto, Y.; Shimanouchi, R. Synthesis of Al2(MoO4)3 by Two Distinct Processes, Hydrothermal Reaction and Solid-State Reaction. Procedia Eng. 2016, 148, 158–162. [Google Scholar] [CrossRef]
  47. Tian, Y.; Yao, Y.; Zhi, Y.; Yan, L.; Lu, S. Combined Extraction–Oxidation System for Oxidative Desulfurization (ODS) of a Model Fuel. Energy Fuels 2015, 29, 618–625. [Google Scholar] [CrossRef]
  48. Barman, B.N.; Skarlos, L.; Kushner, D.J. Simultaneous Determination of Oil and Coke Contents in Spent Hydroprocessing Catalyst by Thermogravimetry. Energy Fuels 1997, 11, 593–595. [Google Scholar] [CrossRef]
  49. García-Gutiérrez, J.L.; Fuentes, G.A.; Hernández-Terán, M.E.; Murrieta, F.; Navarrete, J.; Jiménez-Cruz, F. Ultra-deep oxidative desulfurization of diesel fuel with H2O2 catalyzed under mild conditions by polymolybdates supported on Al2O3. Appl. Catal. Gen. 2006, 305, 15–20. [Google Scholar] [CrossRef]
  50. Luna, M.L.; Caero, L.C. Tungsten based catalysts for oxidative desulfurization: Surface species and partially reduced systems as key features to improve the activity. J. Appl. Res. Technol. 2018, 16, 455–465. [Google Scholar] [CrossRef]
  51. Heracleous, E.; Lee, A.F.; Vasalos, I.A.; Lemonidou, A.A. Surface Properties and Reactivity of Al2O3-Supported MoO3 Catalysts in Ethane Oxidative Dehydrogenation. Catal. Lett. 2003, 88, 47–53. [Google Scholar] [CrossRef]
  52. Houda, S.; Lancelot, C.; Blanchard, P.; Poinel, L.; Lamonier, C. Oxidative Desulfurization of Heavy Oils with High Sulfur Content: A Review. Catalysts 2018, 8, 344. [Google Scholar] [CrossRef]
  53. Dominguez Garcia, E.; Chen, J.; Oliviero, E.; Oliviero, L.; Maugé, F. New insight into the support effect on HDS catalysts: Evidence for the role of Mo-support interaction on the MoS2 slab morphology. Appl. Catal. B Environ. 2020, 260, 117975. [Google Scholar] [CrossRef]
  54. Khishgar, M.; Meshkat, S.S.; Safaie, Z. An efficient oxidative desulfurization method of diesel fuel via iron oxide—Molybdenum oxide nanocomposite catalysts on gamma alumina. 2023; preprint. [Google Scholar] [CrossRef]
  55. Alvarez-Amparán, M.A.; Cedeño-Caero, L. MoOx-VOx based catalysts for the oxidative desulfurization of refractory compounds: Influence of MoOx-VOx interaction on the catalytic performance. Catal. Today 2017, 282, 133–139. [Google Scholar] [CrossRef]
  56. Forzatti, P.; Lietti, L. Catalyst deactivation. Catal. Today 1999, 52, 165–181. [Google Scholar] [CrossRef]
  57. Appleby, W.G.; Gibson, J.W.; Good, G.M. Coke Formation in Catalytic Cracking. Ind. Eng. Chem. Process Des. Dev. 1962, 1, 102–110. [Google Scholar] [CrossRef]
  58. Le Bihan, L.; Blanchard, P.; Fournier, M.; Grimblot, J.; Payen, E. Raman spectroscopic evidence for the existence of 6-molybdoaluminate entities on an Mo/Al2O3 oxidic precursor. J. Chem. Soc. Faraday Trans. 1998, 94, 937. [Google Scholar] [CrossRef]
  59. Oliviero, L.; Maugé, F.; Afanasiev, P.; Pedraza-Parra, C.; Geantet, C. Organic additives for hydrotreating catalysts: A review of main families and action mechanisms. Catal. Today 2021, 377, 3–16. [Google Scholar] [CrossRef]
Catalysts 14 00823 g001
Figure 1. Chromatograms of diluted centrifuged samples RMG380-DC and oxidized RMG380-DC after oxidation on a 20Mo/Al catalyst at 80 °C for 60 min with Ox/S = 5. Dashed lines separate the Cx-BTs, Cx-DBTs and UCM zones.
Figure 1. Chromatograms of diluted centrifuged samples RMG380-DC and oxidized RMG380-DC after oxidation on a 20Mo/Al catalyst at 80 °C for 60 min with Ox/S = 5. Dashed lines separate the Cx-BTs, Cx-DBTs and UCM zones.
Catalysts 14 00823 g001
Catalysts 14 00823 g002
Figure 2. Chromatogram of diluted centrifuged samples RMG380-DC in the 15–30 min retention time region with peaks corresponding to Cx-BTs (purple) and DBT (blue).
Figure 2. Chromatogram of diluted centrifuged samples RMG380-DC in the 15–30 min retention time region with peaks corresponding to Cx-BTs (purple) and DBT (blue).
Catalysts 14 00823 g002
Catalysts 14 00823 g003
Figure 3. Raman mapping of extrudate cross-sections for 20Mo/Al catalysts calcined under different conditions: (1) 20Mo/Al (1,350,3), (3) 20Mo/Al (1,480,3), (4) 20Mo/Al (1,550,3), (5) 20Mo/Al (1,480,5), (6) 20Mo/Al (5,480,3). (2) 20Mo/Al (1,400,3) is not shown due to fluorescence.
Figure 3. Raman mapping of extrudate cross-sections for 20Mo/Al catalysts calcined under different conditions: (1) 20Mo/Al (1,350,3), (3) 20Mo/Al (1,480,3), (4) 20Mo/Al (1,550,3), (5) 20Mo/Al (1,480,5), (6) 20Mo/Al (5,480,3). (2) 20Mo/Al (1,400,3) is not shown due to fluorescence.
Catalysts 14 00823 g003
Catalysts 14 00823 g004
Figure 4. (left) XR diffractograms and (right) Raman spectra of the xMo/Al catalysts.
Figure 4. (left) XR diffractograms and (right) Raman spectra of the xMo/Al catalysts.
Catalysts 14 00823 g004
Catalysts 14 00823 g005
Figure 5. (Left) TG and DTG curves (insert) of Al2O3, MoO3 and the xMo/Al catalysts obtained by TGA. (Right) Experimental mass loss versus the theoretical mass loss and comparison with results from Ibrahim and El-Shobaky [39].
Figure 5. (Left) TG and DTG curves (insert) of Al2O3, MoO3 and the xMo/Al catalysts obtained by TGA. (Right) Experimental mass loss versus the theoretical mass loss and comparison with results from Ibrahim and El-Shobaky [39].
Catalysts 14 00823 g005
Catalysts 14 00823 g006
Figure 6. Evolution of the conversion of Cx-BTs and DBT during four successive ODS tests of RMG380_DC with 10Mo/Al and 20Mo/Al catalysts at 80 °C for 60 min with Ox/S = 5.
Figure 6. Evolution of the conversion of Cx-BTs and DBT during four successive ODS tests of RMG380_DC with 10Mo/Al and 20Mo/Al catalysts at 80 °C for 60 min with Ox/S = 5.
Catalysts 14 00823 g006
Catalysts 14 00823 g007
Figure 7. Comparison of Cx-BTs and DBT conversion in Run 4 (fully filled) and in the test with 0.45 g (stripped) of catalyst for 10Mo/Al and 20Mo/Al.
Figure 7. Comparison of Cx-BTs and DBT conversion in Run 4 (fully filled) and in the test with 0.45 g (stripped) of catalyst for 10Mo/Al and 20Mo/Al.
Catalysts 14 00823 g007
Catalysts 14 00823 g008
Figure 8. TG and DTG curves of 10Mo/Al (left) and 20Mo/Al (right) catalysts before testing (_Fresh), after testing (_Used), and after regeneration by calcination at 550 °C (_Calcined). Remark: The water content in the used catalysts explains the TG curves starting below 100%.
Figure 8. TG and DTG curves of 10Mo/Al (left) and 20Mo/Al (right) catalysts before testing (_Fresh), after testing (_Used), and after regeneration by calcination at 550 °C (_Calcined). Remark: The water content in the used catalysts explains the TG curves starting below 100%.
Catalysts 14 00823 g008
Catalysts 14 00823 g009
Figure 9. Comparison of Cx-BTs and DBT conversion for test run 1 (_Fresh), test run 2 (_Used), and after regeneration by calcination at 550 °C (Calcined) for 10Mo/Al and 20Mo/Al catalysts.
Figure 9. Comparison of Cx-BTs and DBT conversion for test run 1 (_Fresh), test run 2 (_Used), and after regeneration by calcination at 550 °C (Calcined) for 10Mo/Al and 20Mo/Al catalysts.
Catalysts 14 00823 g009
Catalysts 14 00823 g010
Figure 10. XR diffractograms of 10Mo/Al and 20Mo/Al catalysts, before and after test + calcination.
Figure 10. XR diffractograms of 10Mo/Al and 20Mo/Al catalysts, before and after test + calcination.
Catalysts 14 00823 g010
Catalysts 14 00823 g011
Figure 11. Raman spectra collected at three different grains on 20Mo/Al2O3 catalysts, after test with H2O2 + calcination.
Figure 11. Raman spectra collected at three different grains on 20Mo/Al2O3 catalysts, after test with H2O2 + calcination.
Catalysts 14 00823 g011
Catalysts 14 00823 g012
Figure 12. (Left) Evolution of the conversion of Cx-BTs and DBT during four successive ODS tests of RMG380_DC using tBHP with the 20Mo/Al catalyst and conversion of Cx-BTs and DBT after its regeneration by calcination at 550 °C. (Right) TG and DTG curves of the 20Mo/Al catalyst used in ODS with tBHP before testing (_Fresh), after testing (_Used), and after regeneration by calcination at 550 °C (_Calcined).
Figure 12. (Left) Evolution of the conversion of Cx-BTs and DBT during four successive ODS tests of RMG380_DC using tBHP with the 20Mo/Al catalyst and conversion of Cx-BTs and DBT after its regeneration by calcination at 550 °C. (Right) TG and DTG curves of the 20Mo/Al catalyst used in ODS with tBHP before testing (_Fresh), after testing (_Used), and after regeneration by calcination at 550 °C (_Calcined).
Catalysts 14 00823 g012
Table 1. Cx-BTs and DBT conversions for 20Mo/Al (x,y,z) catalysts, with x being the temperature ramp in °C·min−1, y the temperature of the plateau in °C, and z its duration in hours. ODS was performed on RMG380_DC at 80 °C for 60 min with Ox/S = 5.
Table 1. Cx-BTs and DBT conversions for 20Mo/Al (x,y,z) catalysts, with x being the temperature ramp in °C·min−1, y the temperature of the plateau in °C, and z its duration in hours. ODS was performed on RMG380_DC at 80 °C for 60 min with Ox/S = 5.
CatalystCxBTs Conv/%DBT Conv/%
20Mo/Al (1,350,3)1446
20Mo/Al (1,400,3)958
20Mo/Al (1,480,3)2360
20Mo/Al (1,550,3)1653
20Mo/Al (1,480,5)1352
20Mo/Al (5,480,3)839
Table 2. Theoretical and measured Mo loadings and textural properties of the series of xMo/Al catalysts.
Table 2. Theoretical and measured Mo loadings and textural properties of the series of xMo/Al catalysts.
Catalyst%theo MoO3%exp MoO3 (XRF)SBET (m²/g)SBET (m²/gsup)Vp (cm3/g)Vp (cm3/gsup)Ømean (nm)
Al2O3//177/0.7/13.6
5Mo/Al57.11671800.70.812.1
10Mo/Al109.11731900.70.811.2
15Mo/Al1515.51541820.60.711.9
20Mo/Al2019.81501870.50.610.6
30Mo/Al3029.51381960.50.711.5
Table 3. Molybdenum content determined by FX on the fresh and used (after four tests) 10Mo/Al and 20Mo/Al.
Table 3. Molybdenum content determined by FX on the fresh and used (after four tests) 10Mo/Al and 20Mo/Al.
Sample NameFresh Catalyst/%MoO3Used Catalyst/%MoO3
10Mo/Al19.818.6
20Mo/Al9.19.4
Table 4. CNHS analysis of 10Mo/Al after 1 test and of 20Mo/Al after one and four tests, with the corresponding sulphur retention percentages.
Table 4. CNHS analysis of 10Mo/Al after 1 test and of 20Mo/Al after one and four tests, with the corresponding sulphur retention percentages.
CatalystC/wt.%H/wt.%N/wt.%S/wt.%S Retention/%H/CS/C
10Mo/AlAfter one run14.341.800.030.524.21.50.01
20Mo/AlAfter one run14.561.800.040.514.31.50.01
After four runs17.512.160.090.836.91.50.02
After one run + calcination0.060.3500.33nd702
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Roy, T.; Alakari, J.; Lancelot, C.; Blanchard, P.; Poinel, L.; Lamonier, C. Deactivation and Regeneration Studies of Molybdenum-Based Catalysts in the Oxidative Desulfurization of Marine Fuel Oil. Catalysts 2024, 14, 823. https://doi.org/10.3390/catal14110823

AMA Style

Roy T, Alakari J, Lancelot C, Blanchard P, Poinel L, Lamonier C. Deactivation and Regeneration Studies of Molybdenum-Based Catalysts in the Oxidative Desulfurization of Marine Fuel Oil. Catalysts. 2024; 14(11):823. https://doi.org/10.3390/catal14110823

Chicago/Turabian Style

Roy, Teddy, Joy Alakari, Christine Lancelot, Pascal Blanchard, Line Poinel, and Carole Lamonier. 2024. "Deactivation and Regeneration Studies of Molybdenum-Based Catalysts in the Oxidative Desulfurization of Marine Fuel Oil" Catalysts 14, no. 11: 823. https://doi.org/10.3390/catal14110823

APA Style

Roy, T., Alakari, J., Lancelot, C., Blanchard, P., Poinel, L., & Lamonier, C. (2024). Deactivation and Regeneration Studies of Molybdenum-Based Catalysts in the Oxidative Desulfurization of Marine Fuel Oil. Catalysts, 14(11), 823. https://doi.org/10.3390/catal14110823

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop