Synthesis of Oxygenated Hydrocarbons from Ethanol over Sulfided KCoMo-Based Catalysts: Influence of Novel Fiber- and Powder-Activated Carbon Supports

Abstract

Ethanol has become a viable feedstock for basic organic synthesis. The catalytic conversion of ethanol provides access to such chemicals as diethyl ether, ethyl acetate, and acetaldehyde. Carbonaceous materials are extensively studied as supports for heterogeneous catalysts due to their chemical and thermal stability, high surface area, and tunable texture. In this paper, ethanol conversion over K₁₀Co_3.7Mo₁₂S-catalysts supported on novel activated carbon (AC) materials (i.e., novel powder-AC (DAS and YPK-1), fiber non-woven AC material (AHM), and fabric active sorption (TCA)) was investigated. The catalysts were prepared by the incipient wetness co-impregnation method followed by sulfidation. The catalysts were characterized by employing N₂ adsorption–desorption measurements, TEM, SEM/EDX, UV–Vis spectroscopy, and XRF. Catalytic performance was assessed in a fixed-bed down-flow reactor operating at 320 °C, 2.5 MPa, and with continuous ethanol feeding in an He atmosphere. Activity is highly dependent on the support type and catalyst’s textural properties. The activity of the fiber-supported catalysts was found to be greater than the powder-supported catalysts. Ethanol conversion at T = 320 °C, P = 2.5 MPa, and GHSV = 760 L h⁻¹ kg_cat⁻¹ increased as follows: (38.7%) KCoMoS₂/YPK-1 < (49.5%) KCoMoS₂/DAS < (58.2%) KCoMoS₂/TCA < (67.1%) KCoMoS₂/AHM. Catalysts supported by powder-AC enhanced the formation of MoS₂-crystallites, whereas the high acidity of fiber-AC seemed to inhibit the formation of MoS₂-crystallites. Simultaneously, a high surface area and a microporous catalytic structure enhance the formation of oxygenates from hydrocarbons. The dehydration and dehydrogenation reactions, which led to the creation of ethene and acetaldehyde, were shown to require a highly acidic catalyst, while the synthesis of ethyl acetate and higher alcohols required a less acidic catalyst.

1. Introduction

Nowadays, the global energy dependency on fossil fuels produces significant instability in the global market since global supplies of fossil fuels are depleting [1,2], resulting in relative price volatility. Currently, ethanol is seen as a raw material for synthesis of petrochemical products over fossil fuels, and it is considered one of the best biofuels for transportation [3]. It may, in fact, be burnt directly or combined with gasoline to enhance fuel combustion in automobiles, resulting in lower CO₂ emissions and lowering greenhouse gas emissions in the environment [3,4]. Furthermore, ethanol is not only regarded as a superior fuel but also as a very versatile chemical product. It is a vital raw material for both food processing and the synthesis of chemical products [5]. Its applications are growing by the day and have been shown to be critical in terms of global research [6,7,8,9,10].

A wide range of catalytic systems have been studied in methanol and ethanol processes, resulting in catalysts based on transition metals [11], metal oxides (Al₂O₃, MgO, TiO₂, CeO₂, and SiO₂) [12,13], and metals or metal oxides supported on activated carbon [14,15]. In [16], Au nanoparticles supported on a range of metal oxides were examined for gas phase oxidation, and the product distribution was demonstrated to be dependent on the kind of supports. Iwasa et al. [17] investigated SiO₂, ZrO₂, Al₂O₃, MgO, and ZnO as supports for Cu-based catalysts and discovered that the selectivity toward acetaldehyde or ethyl acetate depended on the support type. Another study [15] discovered that the synthesis of undesired higher hydrocarbons was accelerated not only with basic oxide supports (ZnO and MgO) but also with Al₂O₃, and they ascribed this production to base-catalyzed aldol condensation. Moreover, mesoporous SiO₂-supported Cu-catalysts work extremely well at 260 °C, converting ethanol with a higher product selectivity toward acetaldehyde [7]. It is proposed that the presence of the Si-OH group may promote side reactions such as C-O bond breaking, which leads to ethanol dehydration [17]. As a consequence, selectivity to acetaldehyde was greatly reduced.

Applications of transition sulfides (TMS) supported on different carriers as catalysts for the conversion of syngas into alcohols look very attractive because of the low sensibility to sulfur poisons present even in very small amounts in synthesis gas [18]. For decades, these systems were the focus of our studies as catalysts for the hydrodesulfurization (HDS) of oil crudes and fractions [19,20,21,22,23,24]. As a result of tracer isotopic studies (S-35 and H-3), it was found that two different active sites (ASs) responsible for the hydrogenolysis of the C-S bond (type I “slow” and II “rapid”) differ by their productivity in HDS and by their structure. The sites of type I consist of a non-promoted MoS₂ single cluster with lower activity in hydrogenolysis. However, these sites are active in hydrogenation (HYD) reactions. The AS of type II consists of the combination of two single MoS-clusters, one of which contains a Co atom [23] and exhibits higher catalytic activity. These results developed the “Rim-edge” model suggested by M. Daage and R. Chianelli [25] by the introduction of two different sites responsible for desulfurization and by the quantitative evaluation of their activity. The logical continuation of this was the suggested concept of interlayer dynamics and the related “Dynamic mechanism” [22,23]. We applied these results for the similar TMS systems modified by potassium as catalysts for syngas conversion [26].

According to the rim–edge model, a hydrogenation reaction predominantly proceeds on the MoS₂ crystallite rims [25] and most of the slow ASs are located on the rims, which follows from interlayer dynamics [23]. The particle length increase determines the increase in the number of active sites located on the rims of the MoS₂ slabs. The presence of potassium is favorable for increasing both the stacking number and the length of the slabs. As the layers grow in number, the number of ASs located on the edges increases. Alcohols are formed on the low-coordinative unsaturation ASs positioned predominantly on the crystallite edges. So, the increase stacking number and linear size of the crystallite promote formation of the active sites and are favorable for alcohol formation. As the percentage of multilayer MoS₂ crystallites grows, the yield of alcohols increases, whereas the yield of HC decreases. This fact suggests that the alcohols were formed on the AS modified with potassium located on the MoS₂ crystallite edges. Hence, it is possible to control the alcohols/HC ratio by varying the K concentration. According to quantum chemical calculations, using the density functional theory approach, potassium donates electronic density onto the Co atoms of the CoMoS active sites [26].

According to [27], support morphology on supported KCoMoS catalysts is critical for the formation of the active phase. K-modified, Co(Ni)-promoted and supported MoS₂-based catalysts have attracted interest owing to their resistance to sulfur poisoning and less severe coke deposition [18,28,29,30]. There has been extensive research on the support effect on MoS₂-based catalysts in hydrotreatment [31,32] and synthesis gas conversion [29,33,34] because of the crucial role they play in the development of the catalyst, the rate at which it deactivates, and the overall cost of the process. Conventionally, γ-Al₂O₃ is used as a support for hydrodesulfurization (HDS), hydrodenitrogenation (HDN), hydrocracking, and CO/H₂ conversion due to its thermal properties, stability, and low cost [35,36]. Mo (W)-based catalysts supported on Al₂O₃ and promoted by Co or Ni are among the most active and stable. However, on Al₂O₃, it is difficult to avoid the formation of undesirable species such as Co(Ni)MoO₄ or Co(Ni)Al₂O₄ in the preparation of oxide precursors. These species may result in confinement sulfides of the Co₉S₈ or Ni₂S₃ type, which are almost inactive in hydrotreatments or in poorly sulfided Mo oxysulphide species [37]. Moreover, TiO₂ and ZrO₂ have shown a high hydrotreatment reaction potential. However, their restricted particular surface area has long prevented their widespread use. Recently developed supports with specific surface areas as high as 300 m²/g show promise as potential Al₂O₃ competitors [28,29].

Carbon-type supports would be perfect due to their high specific surface area and regulated porosity [34,38,39,40], but in practice, they fall short due to their inadequate mechanical qualities. Hydrotreatment conditions may cause the active phase to sinter, and the restricted ability of these supports to be regenerated after deactivation severely limits their use. However, the properties of carbon make it an “inert” and dispersive support, which has been exploited in numerous fundamental studies aimed at understanding the effects of interactions between the support and the active phase, comparing the catalytic activity of carbon-supported sulfide catalysts to those supported on Al₂O₃ and SiO₂ [41,42]. Despite the fact that these catalytic systems have been extensively studied for higher alcohol synthesis from syngas conversion, no studies for direct ethanol conversion over transition metal sulfides catalysts have been highly reported. This study is novel and exciting since studying ethanol reactions on MoS₂-based catalysts may lead to a new set of intriguing ethanol transformations, such as dehydrogenation and dehydrogenative coupling.

The goal of the study is to investigate the support effect of novel fiber- and powder-activated carbon materials on the activity of supported, modified transition metal sulfide catalysts in ethanol conversion by establishing the relationship between the catalytic behavior and the nature, morphology, and acidity of the used carriers.

2. Results

2.1. Catalyst Characteristics

Two novel powder-AC and two fiber-AC were chosen as supports for the preparation of K-promoted CoMo catalysts (i.e., 10 wt.% K, 3.7 wt.% Co, and 12 wt.% Mo). The textural properties of the DAS, KCoMoS₂/DAS, YPK-1, and KCoMoS₂/YPK-1 catalysts are listed in

Table 1. Surface area and pore characteristics for AC-supported KCoMoS₂ catalysts ¹.

Catalysts	DAS	YPK-1	KCoMoS2/DAS	KCoMoS2/YPK-1
SBET (m2/g)	724	771	250	178
Smicro (m2/g)	662	660	243	142
Smeso (m2/g)	62	51	7	35
Vtotal (cm3/g)	0.34	0.32	0.11	0.09
Vmicro (cm3/g)	0.28	0.27	0.10	0.06
Vmeso (cm3/g)	0.06	0.05	0.01	0.03
Dp (nm)	3.2	3.9	3.3	3.2
NSBET	–	–	0.46	0.31
BE	–	–	0.54	0.69
%ME	0.18	0.16	0.11	0.33

¹ S_meso = S_BET − S_micro; micropore volume calculated by t-plot method; mesopore volume calculated by BJH method; NS_BET—normalized BET surface area; BE—blocking extent of the pores of the support due to metal loading calculated by BE = 1 – NS_BET; %ME—mesoporosity calculated via the ratio of mesopore volume to total pore volume.

. When compared to the parent DAS (723.58 m²g⁻¹) and YPK-1 (771.29 m²g⁻¹), the BET surface areas of KCoMoS₂/DAS and KCoMoS₂/YPK-1 dramatically decreased to 249.45 m²g⁻¹ and 177.76 m²g⁻¹, respectively. Additionally, the microporous structure of the KCoMoS₂/DAS catalyst (243 m²g⁻¹) was significantly higher than that of the KCoMoS₂/YPK-1 catalyst (142 m²g⁻¹), although the mesoporous structures exhibited the opposite tendency, at 7 and 35 m²g⁻¹, respectively. The observed reduction in surface area was strongly attributed to the collapse and blockage of porous structures during the deposition of metals in a sulfide state on the support. The large surface area may indicate fine but relatively small pores (about < 2 nm), which may become blocked during catalyst synthesis, particularly if significant loadings are desired. In addition, compared to DAS and YPK-1 supports, KCoMoS₂/DAS and KCoMoS₂/YPK-1 exhibited a decrease in pore volume (from 0.34 to 0.11 cm³g⁻¹ and 0.32 to 0.09 cm³g⁻¹, respectively). Meanwhile, an increase in D_p from 3.169 to 3.285 nm was observed on KCoMoS₂/DAS after the metal-loading process. Surisetty et al. [43] hypothesized that adding metal would likely block the micropores in the support.

The normalized NS_BET values are also shown in Table 1. As an approximation of the degree of pore blockage encountered by the catalyst support during the metal-loading process, the values of NS_BET were calculated [44]. The NS_BET values of the KCoMoS₂/DAS and KCoMoS₂/YPK-1 catalysts were found to be 0.46 and 0.31, respectively. The higher NS_BET value indicates significant pore blocking. The blocking extent (BE) of the pores of the support as a result of metal loading was estimated using the equation BE = 1 − NS_BET. The BE is found to be high on the YPK-1 (i.e., 0.69) alkali-modified trimetallic catalyst compared to that of the catalyst supported on DAS (i.e., 0.54). The results demonstrate that the nature of the support significantly affected the textural characteristics and performance of the catalysts.

The elemental compositions of the stabilized catalysts are shown in

Table 2. Composition of prepared catalysts based on X-ray fluorescence spectroscopy data.

Catalysts	Targeted Compositions (wt.%)			Measured Compositions (wt.%)
	K	Co	Mo	K	Co	Mo
KCoMoS2/DAS	10	3.7	12	8.5	3.7	11.9
KCoMoS2/YPK-1	10	3.7	12	8.4	3.9	15.8
KCoMoS2/TCA	10	3.7	12	11.3	4.2	13.9
KCoMoS2/AHM	10	3.7	12	11.1	4.6	15.1

. In light of these findings, it seems that the vast majority of precursors were deposited on the support surface during incipient wetness impregnation. Notably, the measured K composition on both supports, fiber and powder, seems to be dependent on the support type. The K loading was seen to be between 8–11%, which is a variance of around 30%. On the fiber-supported catalysts, we presume that a large quantity of K was deposited on the actual surface of the support, while on the powder-supported catalysts, K was deposited inside the pores.

The acidity of the supports and catalysts was investigated using pyridine-TPD adsorption experiments, and the results are shown in

Table 3. Total acidity of AC-supported MoS₂ catalysts.

Catalysts	Acidity (µmol·g−1)
DAS	65.81
YPK-1	39.31
AHM	79.67
TCA	47.37
KCoMoS2/DAS	21.19
KCoMoS2/YPK-1	24.05
KCoMoS2/AHM	156.43
KCoMoS2/TCA	165.81

. It should also be noted that pyridine is adsorbed primarily on the strongly acidic sites [11,45]. The data show that the acidity for the powder and fiber supports is similar in values. Thus, the acidity of the DAS powder material is close to the acidity of the AHM fabric material, and the acidity of YPK-1 is close to the acidity of TCA, i.e., the acidity does not depend on whether the material is a powder or a fiber; instead, it is determined by its composition and physical-chemical properties like acidity, porosity, etc. At the same time, the catalysts deposited on the powder support demonstrate significantly lower acidity compared to the support itself, and the catalysts deposited on the fiber support show significantly higher acidity than the original corresponding fiber support.

In Figure 1a, the N₂ sorption isotherms for the KCoMoS₂/DAS and KCoMoS₂/YPK-1 catalysts are reported, while the corresponding pore size distributions are depicted in Figure 1b. According to IUPAC classification [46], both the AC-supported catalysts showed a typical Type I isotherm, which is a typical characteristic of N₂ adsorption in microporous samples. This may be observed because of the presence of narrow micropores. Microporous solids with small external surfaces provide Type I isotherms (e.g., activated carbons, molecular sieve zeolites, and certain porous oxides) [46]. Additionally, both isotherms for the DAS- and YPK-1-supported catalysts show an H4 hysteresis loop. Type H4 loops are frequently associated with narrow slit-like pores [47,48]. In addition, micropores are filled at a low relative vapor pressure before capillary condensation occurs, whereas mesopores are filled at a higher relative pressure after capillary condensation commences. The DAS- and YPK-1-based catalysts exhibited a narrow pore size distribution profile. According to [49], this effect could be caused by the appearance of pores in different size ranges. The pores in the smaller size range could be caused by a blockage in the support’s pore channels, while the pores in the mid-size range could be caused by the collapse of smaller pores in the smaller size range. Moreover, the distinctive broad pore size distribution may result in pore blockage, thus affecting the catalyst’s diffusional features during the synthesis of desired products. Notably, we were unable to quantify the textural properties of the fiber-based catalysts owing to their very low or negligible surface area; hence, the isotherms for the fiber-AC-supported catalysts are not provided. Fibers have a very low surface area, less than 1 m²/g, and it has been shown that the N₂ sorption approach is ineffective at low surface area ranges, below and about 1 m²/g, and that using nitrogen is inadequate [50].

2.2. SEM and TEM

The materials were subjected to SEM/EDX to examine the morphologies of the catalysts (Figure 2). The powder-AC-supported catalysts, i.e., KCoMoS₂/DAS and KCoMoS₂/YPK-1, demonstrated a porous structure with a high degree of uneven porosity. For both catalysts, it is possible that the non-uniform porosity structure is driven by a number of small-particle and large-particle aggregations. In contrast, KCoMoS₂/AHM and KCoMoS₂/TCA exhibit a thin, flexible, threadlike morphology with a strip-axial pattern and a few longitudinal gaps, as well as a huge proportion of irregular particles distributed mostly on the fiber-AC surfaces of the respective materials. On the other hand, it seems that active metals are deposited on the support surface more unevenly with the fiber-AHM than with the fiber-TCA.

Energy dispersive X-ray spectroscopy (EDX) was used to investigate the distributions of K, Co, Mo, and S (Figure S1). The SEM/EDX spectrum mapping confirms that MoS₂ species are generated during the sulfidation process, and K/Co species are homogeneously distributed on the MoS₂ phase, which further forms a single phase and a uniform phase. The elements’ mapping also provides indirect evidence for the formation of MoS_xO_y oxysulfide species and a Co-Mo-S phase. It is also worth mentioning that an EDX spectrum revealed that the KCoMoS₂ catalyst supported on DAS showed impurities on its catalytic surface, such as Fe, Ca, Si, P, Cr, and S. No traces of impurities were observed on the other samples. The impurities seen on DAS support might be attributed to the source used for preparation.

Figure 3 displays the results of a TEM analysis of the microstructure and morphology of MoS₂-like species of sulfided catalysts [41]. The TEM micrographs show the typical layered structure of MoS₂, with crystallites that are randomly oriented. As can be seen, the powder-activated-carbon-based catalysts exhibited much larger and longer layered structures of the MoS₂ crystallite phase than their counterparts. The difference in crystallite morphology might be ascribed to the interaction between the active phase and the support, which may result in type-I or type-II multilayered phase formation. It has been reported that the surface morphology of MoS₂-based catalysts is a consequence of the decorating of K on the edges of MoS₂-slabs, resulting in a K-decorated MoS₂-phase [51,52].

The TEM surface study reveals structural differences in the MoS₂ crystallites, which may be due to K-intercalation. K-intercalation increases the interplanar distance between MoS₂ phases [51,53]. K-intercalation between MoS₂ planes may raise (002) planes’ d-spacing from 6.2 to 8.4 nm. The authors show that intercalation changes throughout time. Dorokhov et al. [54] proposed that K-intercalation in MoS₂ interlayer space creates additional active centers that generate alcohols and alter catalytic activity. In addition, a K-promoter may change the shape and microstructure of active phases, while the support may change the location of active MoS₂-like species (K-decorated MoS₂ and K-intercalated MoS₂).

2.3. Catalytic Tests

The direct conversion of ethanol to various oxygenates was carried out to study the catalytic properties of the supported KCoMoS₂ catalysts. The choice of an adequate support for the catalysts was critical in the process of enhancing their catalytic performance and properties. The experimental results are shown in Figure 4, i.e., the ethanol conversion (%), total liquid product (LP_total), hydrocarbons (HC_total), and LP_total/HC_total ratio. The highest activity for the ethanol reaction was observed on the fiber-AC-based catalysts and the conversion increased in the following manner: (38.7%) KCoMoS₂/YPK-1 < (49.5%) KCoMoS₂/DAS < (58.2%) KCoMoS₂/TCA < (67.1%) KCoMoS₂/AHM. The observed variance in catalytic activity could be a result of the surface morphologies of the catalysts, the extent to which the support interacts with the active phase, and the nature and textural properties of the support utilized. It should be noted that even though fiber-AC-based catalysts show the highest activity, the conversion rate was much greater owing to the increased synthesis of HC_total, and the opposite trend was seen over powder-AC-supported catalysts, with the following increase in LP_total: KCoMoS₂/TCA < KCoMoS₂/AHM < KCoMoS₂/YPK-1 < KCoMoS₂/DAS. Additionally, the increased overall activity of the fiber-based supported catalyst may be attributed to mass-transfer limitations in comparison to the microporous AC.

In contrast to the two powder-Ac supported catalysts, the catalyst with the highest proportion of micropores, KCoMoS₂/DAS (49.5%), was found to be more active than KCoMoS₂/YKP-1 (38.6%). According to Osman et al. [55,56], the quantity of the microporous surface area in a catalyst increases its activity towards CO hydrogenation. This seems to be the case on ethanol conversion. The similar effect of microporous on catalytic activity was seen. Catalysts with a greater microporous content were shown to be more active. The results show that a higher surface area improved the catalytic activity of the MoS₂-based catalysts. It is well-established [29,57] that supporting the Mo active phase on a high-surface area material, such as ACs, may be advantageous because it increases the number of active sites by improving the dispersion and distribution of the metal particles across the support surface. Despite the comparable difference in pore volume, KCoMoS₂/DAS has a slightly improved V_total and V_micro, resulting in enhanced catalytic performance. This corroborates the results of [18]; however, in CO conversion, the catalyst with a larger pore volume was found to facilitate higher conversion and increase the yield of liquid products. In addition, a larger catalyst pore diameter enhanced catalytic activity. Narrow pores are inactive owing to diffusion restrictions, but very wide pores are more prone to coking [58]. This further shows the significance of catalysts’ textural properties in enhancing their performance in targeted products.

Figure 5 illustrates the product distribution achieved and the GC analysis of the reaction products identified as aldehyde, esters, higher alcohols, hydrocarbons, and CO/CO₂. Ethanol was transformed into acetaldehyde (AcH), n-propanol (n-PrOH), i-propanol (i-PrOH), ethyl acetate (EtOAc), n-butanol (BuOH), butyl acetate (BuOAc), ethyl acetoacetate (EAA), CO₂, and light hydrocarbons. It is widely known that ethanol dehydrogenates to directly produce acetaldehyde and dehydrates to form C₂H₄ and diethyl ether through the parallel reaction network, whilst other products are synthesized as secondary reaction products formed via aldol condensation and coupling processes. It is worth noting that only trace amounts of diethyl ether and CO were reported on the KCoMoS₂/TCA catalyst. This could be due to the fact that ether is generally unstable and rapidly decomposes into the equivalent olefin. The synthesis of AcH and C₂H₄ demonstrates that this catalytic system is capable of catalyzing both dehydrogenation and dehydration processes. The product distribution results further provide evidence of the increased synthesis of HC over fiber-based rather than powder-based catalysts, especially for C₂ hydrocarbons. n-PrOH and BuOH, were synthesized in higher quantities over powder-based catalysts than their counterparts. Interestingly, it appears as though the yield of the aldol condensation product, BuOH, was directly proportional to the amount of AcH produced. It seems as if a reduction in AcH promotes aldol-type condensation. While AcH appears to be the second most-synthesized liquid product, no clear correlation was seen between its synthesis for the both fiber- and powder-based catalysts. It was produced at a greater yield than KCoMoS₂/AHM, about two to three times higher than the yield obtained with the other catalysts examined. A considerable amount of EAA was also detected, suggesting that it could be generated through the C–C bond formation process that occurs when two esters or one ester reacts with another carbonyl molecule in the presence of a strong base via the Claisen condensation process. In our prior work [40], a reaction pathway for ethanol over K- and Co-promoted MoS₂-supported catalysts was proposed.

EtOAc was synthesized in significant quantities among all the liquid products under both catalysts, and the yields increased as follows: KCoMoS₂/TCA < KCoMoS₂/AHM < KCoMoS₂/YPK-1 << KCoMoS₂/DAS. It has been shown that [59,60] the dehydrogenation of ethanol employing transition metal-containing catalysts yields a considerable amount of ethyl acetate. Szymanski et al. [61] revealed that the synthesis of ethyl acetate during ethanol decomposition on heterogeneous transition metal catalysts occurs as a consequence of the interaction of the first generated AcH with the surface ethoxy group or ethanol molecules. The adsorbed ethoxy species is dehydrogenated to generate an acetyl species, and the ethoxy and acetyl species react to form adsorbed EtOAc, which eventually desorbs. According to previous studies [62,63], the synthesis of ethyl acetate is the result of two successive reactions, (1) and (2), with the synthesis of acetaldehyde serving as an intermediate step. In order to achieve high selectivity to ethyl acetate, it is also important to lower the partial pressure of acetaldehyde in the system. A fast reaction rate (1) decreases the acetaldehyde concentration, which in turn increases the selectivity to ethyl acetate [63].

CH₃CH₂OH → CH₃CHO + H₂

(1)

CH₃CH₂OH + CH₃CHO → CH₃COOCH₂CH₃ + H₂

(2)

3. Discussion

From Figure 6 it follows that powder carriers (DAS and YPK-1) contain crystallites of the active phase, characterized by a large number of layers and a much longer length, compared with crystallites deposited on fabric materials. It can be assumed that the latter are located in “axial spots” on the surface of the fibers, occupying significantly less area than the crystallites located on the surface of DAS and YPK-1. Therefore, the acid sites of the powder materials are blocked by crystallites of the active phase, while the acid sites of the fiber materials are more open and affect the acidity of the catalyst.

A question remains, namely, why does the acidity of these fiber catalysts become greater than that of the initial supports? It may be due to the size of the formed particles. Most of the active phase particles deposited on the fiber materials are 1–2-layered crystallites with an average size from 2 to 6 nm. The number of more extended particles in these catalysts is insignificant.

The active phase deposited on powder supports contains more particles with 6–8 stacks and a size greater than 10 nm. This indicates that the active phases of the AHM and TCA catalysts, consisting of small, low-layered crystallites, contain more coordination-unsaturated sites (CUS) than the phases of powder catalysts, consisting of larger particles. It is known that CUS are characterized by high acidity. Moreover, hydrogen spillover can play some role the increase of the acidity of active sites on fiber supports. The larger the fraction of free surface, the higher the hydrogen spillover.

Thus, the decrease in the acidity of the catalysts carried on the powder materials and the increase in the acidity of the catalysts deposited on fiber materials can be explained by three interrelated factors: (i) the difference in the active phase particle sizes, (ii) the difference in the number of CUS located on large and small particles, and (iii) hydrogen spillover on the surface free of particles of the active phase.

These considerations correspond to the data on the conversion and yield of the liquid reaction products on powder and fiber catalysts (Figure 4). Catalysts deposited on fiber supports show a higher degree of conversion compared to catalysts on powder supports. In addition, the yield of gaseous products on them is greater than on powder catalysts. At the same time, the yield of liquid products (mainly alcohols) is higher on powder catalysts. Indeed, if hydrogenation and chain-breaking (cracking) reactions occur at CUS, which are located, according to the “Rim-edge” model [23,25,64,65], on rims and solid angles, whose numbers are higher on smaller particles, the formation of alcohols occurs on the edges. Therefore, the longer the edge (more layers and a higher stacking number) of crystallite and the longer its linear size, the greater the yield of alcohols.

Figure 7 depicts catalyst acidity as a function of (a) ethanol conversion and (b) product yields, respectively. The increased surface acidity of the catalyst enhances HC_total synthesis and inhibits LP_total synthesis, while conversion (Figure 7a) increases with increased surface acidity. The most active catalysts (KCoMoS₂/TCA and KCoMoS₂/AHM) are those that have the greatest values for pyridine adsorption (strong acid sites), whilst the least active catalysts (KCoMoS₂/YPK-1 and KCoMoS₂/DAS) have the fewest acidic sites. The increased acidity and activity may be attributed to the reduced ash content of the fiber-AC supports compared to the powder-AC supports.

Figure 7b displays the relationship between the product yields and catalyst acidity. There is a definite correlation, and more evidence indicates that the acidity of the catalyst surface has a significant influence on the synthesis of the desired products. The higher ethylene output observed with the increasing catalyst acidity indicates that the dehydration process needs the presence of strong acid sites in order to produce ethylene products. According to [66], dehydration and the production of C₂H₄ are catalyzed by acidic sites. Additionally, excessive acidity enhanced the synthesis of HC, which was at its maximum under the most acidic catalyst, namely, KCoMoS₂/TCA. Increased acidity promotes alkane production by facilitating C-O cleavage through dehydration [67]. Reduced acidity may help to decrease deoxygenation and boost the synthesis of oxygenates. It also seems as if reduced acidity favors an aldol-type condensation reaction, and this led to an increased yield of higher alcohol, namely, BuOH. According to León et al. [68], catalysts with a greater concentration and strength of their basic sites enhance the formation of C₄-products (i.e., butanol-1), while the presence of acid sites promotes ethanol dehydration, resulting in lower condensation efficiency. Researchers believe that basic sites are critical to the condensation of ethanol towards butanol-1 [8,29]. The slightly higher yields of butyl acetate in the less acidic samples, KCoMoS₂/YPK-1 and KCoMoS₂/DAS, are an indication of an aldol-type process.

Consequently, the synthesis of acetate decreases as the acidity of the catalyst increases [61]. EtOAc was found to decrease as the acidity of the catalyst increased, and it increased in the following manner: KCoMoS₂/TCA < KCoMoS₂/AHM < KCoMoS₂/YPK-1 < KCoMoS₂/DAS. It has been shown that basic sites are associated with the formation of ethoxide species [69]. As a result, the data suggest that the formation of ethoxide is a critical step in the synthesis of ethyl acetate, and that it may even be the rate-limiting step under these conditions. The EAA yield seems to be a byproduct of ethyl acetate synthesis since it appears to be exactly proportional to the EtOAc yield. AcH, the primary reaction product of dehydrogenation, was formed at over twice the rate of its closest counterpart on one of the acidic KCoMoS₂/AHM catalysts, with other catalysts exhibiting a nearly identical performance in their syntheses. However, it was synthesized at a lower yield than EtOAc. It is possible that secondary condensation products contribute to the reduced AcH levels. BuOH-1 seems to rise somewhat in proportion to a drop in AcH yield. Carrasco-Marin et al. [59] observed that the dehydrogenation process that yields AcH occurs on either Lewis acid or basic surface sites located on both the external and internal surfaces. However, this might explain, to some degree, why no clear correlation was not found in the synthesis of AcH in relation to the catalysts’ textural properties. The activity for catalyzing the dehydrogenation of ethanol decreases in the following order: KCoMoS₂/AHM > KCoMoS₂/DAS ≥ KCoMoS₂/TCA > KCoMoS₂/YPK-1.

Additionally, when the acidity increased, a slight drop in pore volume was observed, particularly amongst the powder-AC-supported catalysts. According to Anashkin et al. [70], the total acidity of MoS₂-based catalysts was reduced as the pore volume increased, and the synthesized MoS₂ catalysts exhibited an increase in liquid products. This is consistent with our results when the DAS- and YPK-1-supported catalysts are compared. KCoMoS₂/DAS had a slightly larger pore volume, lower acidity, and higher activity. As a result, MoS₂ with a lower surface acidity showed more activity for LP synthesis (as seen in Figure 4) than MoS₂ with a higher acidity and a smaller pore volume. Importantly, the results demonstrate that the catalyst’s activity and the yields of products are substantially dependent on the support properties. This impact is consistent with our previous research [40], in which it was determined that the type of support plays a crucial role in catalytic activity.

Industrial Relevance of Ethyl Acetate and Other Oxygenates

The experimental results indicated that ethyl acetate (as seen in

Table 4. Product yields from ethanol over KCoMoS₂-supported catalysts.

Catalysts	KCoMoS2/DAS	KCoMoS2/YPK-1	KCoMoS2/AHM	KCoMoS2/TCA
Acetaldehyde	4.6	2.5	11.9	4.2
Ethyl acetate	29.2	19	18.1	13.7
Ethyl acetoacetate	3.2	3.1	2.1	2.7
BuOH	2.5	3.2	0	0.8
n-PrOH	3.5	2.2	0.7	1.1
i-PrOH	0	0.1	0	0
BuOAc	0.6	0.5	0.1	0.3
Dethyl ether	0	0	0	0.3
CO	0	0	0	0.1
CO2	1.4	1.6	8.6	2.9
LPtotal	43.6	30.6	29.8	23.1
HCtotal	4.5	6.4	24.6	32.1
LPtotal/HCtotal	9.7	4.8	1.3	0.7

) was synthesized in greater quantities than the other synthesized products. In contrast, the use of ethyl acetate as a sustainable oxygenated fuel has sparked attention [71]. It has been proven that the fuel properties of ethyl acetate suggest that it has the potential to be employed as an oxygenated addition to gasoline. The fundamental benefit of using ethyl acetate as an oxygenated gasoline is that it increases octane without raising Reid vapor pressure (RVP). Higher alcohols, such as butanol-1, have comparable qualities to gasoline and may be used as a fuel additive, making them a viable replacement. Compared to 1-butanol, ethanol is more corrosive, water-soluble, and has less energy. Other oxygenated compounds obtained from ethanol could be used as an intermediate in the synthesis of a wide range of organic chemical compounds.

4. Materials and Methods

4.1. Carbonaceous Supports

DAS was produced from anthracite (hard coal) by the procedures of dough preparation, granulation, carbonization, and gas-vapor activation. Activated carbon YPK-1 was produced by utilizing a carbonaceous composition generated by gas-vapor activation at 850–900 °C. TCA is an elastic sorbent developed by heating technical fabric previously impregnated with chemical substances. The canvases were 20 m long, 0.55 m wide, and 0.6 mm thick. Heat treatment of a nonwoven, needle-punched material based on viscose and Milton fibers created a nonwoven activated carbon material (AHM), which had Air resistance value of 10 Pa, surface density of 120 g/m², and was 1.0–3.5 mm thick [34].

4.2. Synthesis of K- and Co-Promoted MoS₂ Catalysts

KCoMoS₂ catalysts were prepared using salt solutions of 5 mmol (NH₄)₆Mo₇O₂₄ (obtained from Alfa Aesar, Haverhill, MA, USA, chemically pure 99%) and KOH (analytical grade, 98%, 10 mmol), and C₄H₆CoO₄.4H₂O (Alfa Aesar, tetrahydrate, chemically pure 98%; 2.5 mmol) by the incipient wetness co-impregnation method. The impregnated catalysts were dried at 60–70 °C and 100–110 °C for 2 and 6 h, respectively. Prior to catalytic testing, all catalytic samples were sulfidized using element sulfur (catalyst: sulfur weight ratio = 4:1) in an autoclave at 360 °C under H₂ for 1 h at 6.0 MPa.

4.3. Physical Characteristics of Catalyst

The measurements were carried out using Hitachi SU8000 field-emission scanning electron microscope (FE-SEM). The analytic measures were optimized using a target-oriented methodology. The samples were taped to a 25 mm metal specimen stub before measurement. Magnetron sputtering was used to cover the metal with a thin layer (10 nm, Au/Pd, and 60/40). A working distance of 8–10 mm was used to obtain images in secondary electron mode. The morphology of the samples was examined in relation to the surface metal coating [72].

A Shimadzu EDX-7000 X-ray fluorescence spectrometer was utilized to study the sulfided catalysts’ composition. All samples were crushed first. The study employed a 15–50 kV voltage, an 8–200 mA tube current, and Rh anode. The X-ray fluorescence spectroscopy (XRF) method error was 1 wt.%. The spectra were processed using basic parameters.

The samples’ morphologies were examined using a Hitachi transmission electron microscope (TEM). Analytical measures were optimized via the use of a goal-oriented strategy. Prior to measurements, samples were placed on a 3 mm copper grid and kept in a grid holder. The images were acquired in bright-field TEM mode at a 100 kV accelerating voltage [73] in order to quantify the MoS₂-crystallited. For each catalyst, the lengths of at least 500 slabs were measured using the slab sizes observed in the TEM micrographs. The number of slabs per stack was determined to obtain the average stacking number ($\overline{N}$):

$$\overline{N}=\frac{{{\displaystyle \sum }}_{i=1\dots t}{n}_i{N}_i}{{{\displaystyle \sum }}_{i=1\dots t}{t}^n}$$

(3)

where t is the total number of slabs in the TEM micrographs and n_i is the number of stacks in N_i layers.

N₂ adsorption–desorption isotherms were measured at 77 K using a Quantachrome Nova 1200e (USA). After 3 h of Ar flow, the supports and sulfide catalysts were degassed for 4 h at 110 °C for the supports and 250 °C for the sulfide catalysts. In this case, the BET equation was utilized to compute the surface area (SSA). V_total was calculated at P/P_o = 0.99. The desorption branch of the isotherms was utilized to compute the size distributions of mesopores. The desorption branch of Barrett, Joyner, and Halenda cumulative pore volume was employed as the mesopore volume (considering the adsorption film thickness on the mesopore surface). The micropore volume of the samples was obtained by comparing the total and mesopore volumes. Before using any sample cells for gas sorption analysis, they were properly calibrated. Each sample was obtained at 0.1 g for analysis. The Barrett, Joyner, and Halenda technique was designed to acquire at least 25 adsorption points (10 for BET and 6 for t-plot/-S) and 45 desorption points [74].

To establish the presence of dispersed metal species within the pores of the DAS- and YPK-1-based catalysts, the normalized BET surface area (NS_BET) values of the catalysts were calculated using an approach given by Vradman et al. [44].

$$N{S}_{BET}=\frac{{(S_{BET})}_{catalyst}}{\left(1-y\right).{(S_{BET})}_{support}}$$

(4)

where NS_BET is the normalized BET surface area and y is the weight fraction of the phases.

The acid–base properties of the supported catalysts were determined using UV spectroscopy of pyridine adsorption. An SF-103 single-beam scanning spectrophotometer was used to measure concentrations in liquids. The pyridine adsorption spectra in the ultraviolet region of a blank pyridine solution in octane and solutions of adsorption systems with supports and catalyst were recorded at room temperature for 60 min. There were no changes in absorption maxima at 253 nm (analytical absorption band) with pyridine concentration. This was performed using a calibration curve plotted against the optical density of solution D and pyridine concentration. Equation (3) was used to compute Gibbsian adsorption (G, mol/g):

$$G=\frac{\left(C_0-C_t\right)\times V}{m}=\frac{\left(D_0-D_t\right)\times V}{m\times \epsilon \times l}$$

(5)

where V (solution volume) is 10 mL, m (sample mass) is 0.1 g, D₀ and D_t are equal to the optical density of pyridine at maximum absorption prior to and during the adsorption process. l (cuvette length) is 1 cm; ε is the molar absorption coefficient (extinction, ε of pyridine = 2∙10⁶ L mol⁻¹∙cm⁻¹).

4.4. Catalytic Studies

A stainless-steel packed-bed reactor was utilized to test supported sulfided catalyst performance for converting ethanol to various oxygenates. The catalyst loading was 3 g, with particle sizes ranging from 0.25 to 0.50 mm. The catalyst samples were diluted to 5 mL with quartz granules (1–2 mm) before loading them into the reactor. P = 2.5 MPa, T = 320 °C, GWSV = 760 L h⁻¹ kg_cat⁻¹, and ethanol space velocity 0.3 mL min⁻¹ under He atm.

The ethanol reaction pathways were investigated in a series of studies using KCoMoS₂-based catalysts supported on various support materials. A high-pressure pump injected 30 mL of rectified ethanol into the reactor. Following the ethanol-feeding procedure, the catalyst was left in gas flow for 1 h under the same reaction conditions.

Gas products analysis was performed using the LHM-80 instrument equipped with TCD detector and two 1 m length columns packed with CaA molecular sieves and Porapak Q. For liquid products’ analysis, we utilized Crystall-2000M GS chromatograph with FID detector equipped with 50 m length HP-FFAP capillary column.

The ethanol conversion is defined as:

$${\mathrm{x}}_{\mathrm{EtOH}}=\frac{{\mathrm{mole}\mathrm{EtOH}}_{\mathrm{reacted}}}{{\mathrm{mole}\mathrm{EtOH}}_{\mathrm{feed}}}$$

(6)

while the selectivity S was determined on the basis of a carbon balance for each component:

$$\mathrm{S}=\frac{{\mathrm{mole}\mathrm{products}}_{\mathrm{I}}\mathrm{formed}}{{\mathrm{n}}_{\mathrm{EtOH}}\mathrm{reacted}}\frac{{\mathrm{nC}}_{\mathrm{i}}}{{\mathrm{nC}}_{\mathrm{EtOH}}}=\frac{{\mathrm{AC}}_{\mathrm{i}}}{{\mathrm{Ac}}_{\mathrm{EtOH}}}\frac{{\mathrm{nC}}_{\mathrm{i}}}{{\mathrm{nC}}_{\mathrm{EtOH}}}\left(\frac{1-{\mathrm{x}}_{\mathrm{EtOH}}}{{\mathrm{x}}_{\mathrm{EtOH}}}\right)$$

(7)

where nC_i and nC_EtOH represent the numbers of carbon atoms in the component i and in the ethanol fed, respectively, while Ac_i and Ac_EtOH are the normalized chromatographic peaks’ areas.

5. Conclusions

An attempt was made to develop an extremely efficient K-modified and Co-promoted MoS₂ catalyst based on activated carbon materials. The study offers insight into the catalyst composition necessary for maximum performance during synthesis of oxygenates from ethanol. The novel catalysts showed ethanol reaction activity. The impact of various carbon materials on the MoS₂ ethanol dehydrogenation catalysts was investigated. From the results, it can be concluded that the activity of the catalyst was highly dependent on the support type and catalyst’s textural properties, especially for catalysts with greater surface area, pore volume, and microporous structures. A high surface area and a microporous texture enhanced catalysts’ activity towards the formation of liquid products from hydrocarbons. Powder-activated carbons possess a multiply stacked MoS₂-crystallite structure that leads to a higher quantity of liquid products, while fewer stacked layers on fiber-based catalysts leads to an increasing yield of hydrocarbons. The acidity of the catalyst had a substantial impact on conversion and product distribution. Conversion, dehydration reaction, and dehydrogenation all increased with the increasing catalyst acidity, whereas the aldol-type condensation reaction decreased with the increasing catalyst acidity. By varying the support, it is possible to use KCoMoS₂-based catalysts to control the targeted products, such as ethyl acetate and higher alcohols. Although these insights have been uncovered, there is still space to use them in order to develop catalysts that can provide high yields for certain targeted products.