In Situ Thermal-Stage Fitted-STEM Characterization of Spherical-Shaped Co/MoS2 Nanoparticles for Conversion of Heavy Crude Oils

We report the thermal stability of spherically shaped cobalt-promoted molybdenum disulfide (Co/MoS2) nano-catalysts from in-situ heating under electron irradiation in the scanning transmission electron microscope (STEM) from room temperature to 550 ◦C ± 50 ◦C with aid of Fusion® holder (Protochip©, Inc.). The catalytic nanoparticles were synthesized via a hydrothermal method using sodium molybdate (Na2MoO4·2H2O) with thioacetamide (CH3CSNH2) and cobalt chloride (CoCl2) as promoter agent. The results indicate that the layered molybdenum disulfide structure with interplanar distance of ~0.62 nm remains stable even at temperatures of 550 ◦C, as observed in STEM mode. Subsequently, the samples were subjected to catalytic tests in a Robinson Mahoney Reactor using 30 g of Heavy Crude Oil (AGT-72) from the golden lane (Mexico’s east coast) at 50 atm using (ultrahigh purity) UHP hydrogen under 1000 rpm stirring at 350 ◦C for 8 h. It was found that there is no damage on the laminar stacking of Co/MoS2 with temperature, with interlayer spacing remaining at 0.62 nm; these sulfided catalytic materials led to aromatics rise of 22.65% and diminution of asphaltenes and resins by 15.87 and 3.53%, respectively.


Introduction
Layered metal sulfides such as molybdenum disulfide (MoS 2 ) have been extensively used as catalytic materials in oil refineries for removal of noxious impurities including sulfur, nitrogen compounds and metals like Ni and V, mainly by means of hydrodesulfurization (HDS) and hydroprocessing (HDP) [1]. Previous studies indicate that the catalytic activity of MoS 2 is enhanced by certain metal promoters like cobalt and nickel, as described by Chianelli et al. [2] to create the so-called CoMoS or NiMoS phases as coined by Topsøe et al. [3]. The crystallographic structure of MoS 2 is characterized by a sandwich-like stacking of layers with van der Waals-type chemical bonds between S-Mo-S along the c-direction, with an interlayer distance of 0.62 nm, which makes this particular material ideal for high temperature and tribology engineering applications. More recently, it has been considered as a semiconductor electron transport material for electronic devices [4,5]. The location of metals that are promoters responsible for enhancing metallic "brim" or "edge" sites have been discussed in the literature [6,7] and some studies were made by X-ray absorption near edge structure (XANES) [8], scanning tunneling microscopy (STM) [9] as well as high-resolution transmission electron microscopy (HRTEM) [10]. A series of observations to locate the position of cobalt promoters have been described by Deepak et al., who used Z-contrast in STEM mode for determining a bond distance of 2.24 Å between W-Co-W-Co-W, atoms at the laminar edge of WS 2 (which is a material with a structure alike to MoS 2 ) thus demonstrating the location of these promoters [11]. Moreover, the "in-situ" operando HRTEM techniques made it possible to achieve experimental observations that relate it to crystal growth as described by Hansen et al.; these authors directly imaged laminar MoS 2 phase formation from MoO 3 in an atmosphere of H 2 S:H 2 = 1:9 (0.8 mbar) at 690 • C, with the use of an environmental TEM holder [12]. Ramos et al. performed a study at the carbon grid level on "in-situ" TEM carburization effects on fresh Co/MoS 2 unsupported catalyst at the temperatures range of 350 to 450 • C, thus determining bending curvatures of the MoS 2 layers that are due to carbon deposits and replacement of sulfur atoms at the reactive edges of the MoS 2 platelets. This resulted in a stabilized MoS 2 −C X phase that presents texturally stabilized sulfide particles, thus keeping a crystallite size smaller and less stacked, which is in agreement with data reported elsewhere [13,14]. Recently, an electron tomography study determined significant fractal aspects that are related to porosity in cobalt-promoted MoS 2 in spherically shaped nanoparticles [15], which is coincident with previous work reported by Yin et al. for MoS 2 -type catalysts [16]. Herein, we present an "in-situ" heating STEM study whose purpose was to determine the structural stability of Co/MoS 2 catalyst along with "ex-situ" catalytic performance for upgrading heavy crude oils from the golden lane (Mexico east coast, well AGT-72). These crude oils have high viscosities of about 25·10 3 cP at room temperature and a gravity of less than 13 o API, with asphaltene content higher than 20%. Their mean asphaltene structure was reported recently for the first time [17], thus, the purpose of this work was to characterize the structural features of the active-proven layered MoS 2 catalysts by STEM fitted with a thermal stage. For this, off-line catalytic reaction was performed in parallel to prove their activity for the hydroprocessing of heavy feedstocks [18,19].

In-Situ Scanning Transmission Electron Microscopy (STEM)
Albiter et al. [15] show SEM micrographs (geometry nanorods) for MoS 2 samples before and after 5 h of reaction at 350 • C during HDS reaction.By comparing between micrographs, Albiter et al. show important textural changes for MoS 2 samples after 5 h of reaction, as well as a decrease in specific surface area of 56% for the MoS 2 samples synthetized at 400 • C and 74% for those synthetized at 700 • C. In addition, a decrease in the catalytic activity (24%) and selectivity (39%) was observed. Taking these results into account, STEM images were performed and used as references in order to assess the textural effects over the sample synthetized by the method described above. The sample was subjected to in-situ heating from room temperature up to 550 • C. The images from in-situ heating STEM experiments indicate that the sample was texturally stable up to 550 • C, with no diffusion or evaporation observed. Figure 1 presents two STEM images taken at room temperature together with the heating profile used during the experiments. At room temperature, it was possible to observe some porosity typical of the Co/MoS 2 spheres as reported by Ramos et al. [16] and laminar dispersed structures (needle-like) corresponding to MoS 2 as are indicated by red arrows in Figure 1, as described in detail by Vollath et al. [17] for MoS 2 nanocrystals ( Figure S1). This result allows us to suggest that if the hydroprocessing reactions are carried out above 350 • C, and there are changes in the activity and selectivity of the reaction, these would be a consequence of non-textural factors. During heating stages at a ramp rate of 150 °C/min, as indicated in Figure 2, it was possible to determine the structural stability at 300 °C, 350 °C, 400 °C, 450 °C, 500 °C and 550 °C. It was found that there is no damage on the laminar stacking of Co/MoS2 with temperature, which is of interest for catalyst design for in-situ crude oil upgrading, as described by Pereira-Almao et al. [20]. Furthermore, it was possible to observe the structure stability of cobalt-promoted MoS2 layers, as indicated by the red arrows over atomic fringes that are characteristic of the ~0.62-nm interlayer distances typical of van der Waals solids, as described by Chianelli et al. [21]. During heating stages at a ramp rate of 150 • C/min, as indicated in Figure 2, it was possible to determine the structural stability at 300 • C, 350 • C, 400 • C, 450 • C, 500 • C and 550 • C. It was found that there is no damage on the laminar stacking of Co/MoS 2 with temperature, which is of interest for catalyst design for in-situ crude oil upgrading, as described by Pereira-Almao et al. [20]. Furthermore, it was possible to observe the structure stability of cobalt-promoted MoS 2 layers, as indicated by the red arrows over atomic fringes that are characteristic of the~0.62-nm interlayer distances typical of van der Waals solids, as described by Chianelli et al. [21].

Figure 2.
A series of STEM images acquired during in-situ heating at 300 °C, 350 °C, 400 °C, 450 °C, 500 °C and 550 °C as labeled. One can observe c-stacking of S-Mo-S two-dimensional layers, even at of 550 °C, with interplanar distances of ~0.62 nm; also the spherical shape remains as indicated by red-circle on image corresponding to stage at 550 °C.

In-situ Hydrodesulfurization Activity
The HDS process is most commonly used to obtain low sulfur fuels, i.e., ultra-low-sulfurgasoline (sulfur content < 15 ppm). Thus, the proof of concept for the "in-situ"-catalytic test was performed using a heavy crude oil (AGT-72) from "The golden Lane" using a Robinson Mahoneytype reactor with a loading of 30.0 g crude oil AGT-72 and 0.3 g of Co/MoS2 catalyst. The reaction was carried out with ultra-high purity hydrogen at 50 atm and 350 °C for 8 h. HDS results show that the catalyst reduces the sulfur content of crude oil from 5.1% to 4.8%, despite the refractory species present in the Heavy Crude Oil-AGT-72. The catalytic nanoparticles in our study are aimed for applications in oil recovery into oil reservoirs [20][21][22][23]; however but some challenges persist such as:

In-Situ Hydrodesulfurization Activity
The HDS process is most commonly used to obtain low sulfur fuels, i.e., ultra-low-sulfur-gasoline (sulfur content < 15 ppm). Thus, the proof of concept for the "in-situ"-catalytic test was performed using a heavy crude oil (AGT-72) from "The golden Lane" using a Robinson Mahoney-type reactor with a loading of 30.0 g crude oil AGT-72 and 0.3 g of Co/MoS 2 catalyst. The reaction was carried out with ultra-high purity hydrogen at 50 atm and 350 • C for 8 h. HDS results show that the catalyst reduces the sulfur content of crude oil from 5.1% to 4.8%, despite the refractory species present in the Heavy Crude Oil-AGT-72. The catalytic nanoparticles in our study are aimed for applications in oil recovery into oil reservoirs [20][21][22][23]; however but some challenges persist such as: (i) a better crude oil-catalyst mixing, (ii) a close effective H 2 /hydrocarbon ratio and (iii) the influence or role of impurities and sulfidation as described by Afanasiev [24]. Mainly, for an optimal catalytic performance for cobalt-or nickel-promoted MoS 2 catalyst.

Saturates, Asphaltenes, Resins and Aromatics (SARA) Analysis
Results from the SARA analysis are presented in Figure 3, which is used as the metrics to assess the conversion and selectivity as well as to establish the quality of the reaction products. The results after reaction showed a significant decrease in the original amount of asphaltenes, as displayed in Table 1, along with a slight decrease in percentage of resins, while a substantial increase in aromatics content is verified, which hold the asphaltenes micelles in suspension [25]. As reported before, resins have a strong tendency to associate with asphaltenes due to their high polarity, thus, acting as a protective shield of asphaltenes [26]. For the present case a stable and less viscous crude oil remains after the hydrotreatment with CoMoS 2 ; the instability index (CII) developed by Yet et al. [27] was used by means of the Equation (1), for measuring the asphaltenes deposition potential of hydrotreated AGT-72. Our results give a CII value around 0.7 after CoMoS 2 in situ treatment, which confirms a stable product, whereas CII results with values of CII higher than 0.9 are considered as unstable. These results show the increase in stability when the AGT-72 crude oil undergoes hydrotreatment with Co/MoS 2 catalysts.
On the other hand, the increased amount of aromatics content (wt-%) from 25.0 to 32.0 may contribute to a higher-octane number, together with isoparaffins [28]. It is clear from our data that aromatics, resins, asphaltenes and saturates proportions, i.e., SARA, are modified after interaction of AGT-72 with Co/MoS 2 catalyst, which is a basic proof of the concept for the potential use of cobalt-promoted MoS 2 spherically shaped nanoparticles for in-situ heavy oil upgrading [16,22] at a microscopic level.
Catalysts 2020, 10, x FOR PEER REVIEW 5 of 10 (i) a better crude oil-catalyst mixing, (ii) a close effective H2/hydrocarbon ratio and (iii) the influence or role of impurities and sulfidation as described by Afanasiev [24]. Mainly, for an optimal catalytic performance for cobalt-or nickel-promoted MoS2 catalyst.

Saturates, Asphaltenes, Resins and Aromatics (SARA) Analysis
Results from the SARA analysis are presented in Figure 3, which is used as the metrics to assess the conversion and selectivity as well as to establish the quality of the reaction products. The results after reaction showed a significant decrease in the original amount of asphaltenes, as displayed in Table 1, along with a slight decrease in percentage of resins, while a substantial increase in aromatics content is verified, which hold the asphaltenes micelles in suspension [25]. As reported before, resins have a strong tendency to associate with asphaltenes due to their high polarity, thus, acting as a protective shield of asphaltenes [26]. For the present case a stable and less viscous crude oil remains after the hydrotreatment with CoMoS2; the instability index (CII) developed by Yet et al. [27] was used by means of the Equation (1), for measuring the asphaltenes deposition potential of hydrotreated AGT-72. Our results give a CII value around 0.7 after CoMoS2 in situ treatment, which confirms a stable product, whereas CII results with values of CII higher than 0.9 are considered as unstable. These results show the increase in stability when the AGT-72 crude oil undergoes hydrotreatment with Co/MoS2 catalysts.

CII = Saturates + Asphaltenes Aromatics + Resins
On the other hand, the increased amount of aromatics content (wt-%) from 25.0 to 32.0 may contribute to a higher-octane number, together with isoparaffins [28]. It is clear from our data that aromatics, resins, asphaltenes and saturates proportions, i.e., SARA, are modified after interaction of AGT-72 with Co/MoS2 catalyst, which is a basic proof of the concept for the potential use of cobaltpromoted MoS2 spherically shaped nanoparticles for in-situ heavy oil upgrading [16,22] at a microscopic level.

X-Ray Photoelectron Spectroscopy (XPS) Results
Cobalt promoters were not identified by STEM as reported earlier by Deepak et al. [11], partially due to their low concentration and dispersion located along the crystallite edges, as indicated by Zhu et al. [7]. Thus, in this work, the fresh catalyst was focused on the analysis by X-ray photoelectron spectroscopy (XPS) of the chemical environment around the atoms present on the top surface layers. Figure 4 indicates the presence of molybdenum 3d 3/2 (232eV), sulfur 2p 3/2 (168eV) and cobalt 2p 3/2 (782eV) that was verified, in agreement with Joe et al. [21].

X-ray Photoelectron Spectroscopy (XPS) Results
Cobalt promoters were not identified by STEM as reported earlier by Deepak et al. [11], partially due to their low concentration and dispersion located along the crystallite edges, as indicated by Zhu et al. [7]. Thus, in this work, the fresh catalyst was focused on the analysis by X-ray photoelectron spectroscopy (XPS) of the chemical environment around the atoms present on the top surface layers. Figure 4 indicates the presence of molybdenum 3d 3/2 (232eV), sulfur 2p 3/2 (168eV) and cobalt 2p 3/2 (782eV) that was verified, in agreement with Joe et al. [21].

Co/MoS 2 Catalyst Preparation
The catalyst was synthesized by mixing 3 mmol of sodium molybdate (Na 2 MoO 4 ·2H 2 O) (Sigma Aldrich, Inc. 2033 Westport Center Dr, St. Louis, MO 63146, USA) and 9 mmol of thioacetamide (CH 3 CSNH 2 ) (Sigma Aldrich, Inc. 2033 Westport Center Dr, St. Louis, MO 63146, USA), followed by the addition of 30 mL of deionized water and 0.5 g of sodium silicate (Na 2 SiO 3 ·9H 2 O); the resulting solution was kept under vigorous stirring for 30 min. The pH of the solution was adjusted to 6.0 using a 12 M hydrochloric acid (HCl) solution, then 0.50 g of cobalt chloride (CoCl 2 ) was added to the solution, which turns into a purple color, then it was placed into a Parr© reactor and set at a temperature of 220 • C for 4 h, followed by slow cool down. The catalyst was washed using Sodium Hydroxide (NaOH) 1 M to remove possible residues and was dried at 200 • C in autoclave, as described by Ramos et al. [15].
The stoichiometry of the reaction is the following:

Catalytic Activity in the Conversion of Golden Lane Crude Oil
The activity test was carried out in a Robinson Mahoney Reactor (Parker, Inc. 8325 Hessinger Dr., Erie, Pennsylvania 16509 USA) using 30 g of Heavy Crude Oil, i.e., labelled as AGT-72, and 0.3 g Co/MoS 2 catalyst that are placed in a stainless-steel basket to obtain an oil/catalyst ratio of 10/1 wt/wt. The reaction system was adjusted to 50 atm using ultra-high purity hydrogen (UHP), the stirring rate and temperature were 1000 rpm and 350 • C during reaction time of 8 h, respectively. The pressure was controlled at 100 kg/cm 2 during the reaction. The hydrotreated crude oil was analyzed to determine the composition of the lumps by means of SARA analysis, i.e., saturates, aromatics, resins and asphaltenes.

SARA Analysis
The asphaltenes were precipitated with n-heptane following the American Society for Testing and Materials (ASTM) D3279 method and the remaining fraction was separated following the ASTM 2700 method, thus obtaining the concentration of resins, aromatics and saturates of both crude oil and the hydrotreated (i.e., after conversion) crude oil. The identification of some individual compounds and gaseous products was made by gas chromatography coupled to mass spectrometry model PerkinElmer© Claurus 590 TM (GC-MS, PerkinElmer, Norwalk, CT, USA).

In-Situ STEM
Approximately 0.05 g of the (catalyst) black precipitate was dispersed in ethanol to make an almost transparent solution, from which one drop was placed into the Fusion ® sample holder chip from Protochip© (3800 Gateway Centre Blvd #306, Morrisville, NC 27560, USA) and it was dried at room temperature. Observations were made on a FEI ® model Titan© STEM (Hillsboro, OR, USA) unit, fitted with a charge coupled device (CCD) camera, and it was operated at 300 keV in STEM-mode, with an extraction voltage of 4500 V, Gun Lens 6, Camera Length 0.195 m, Spot size 7, data size 1024 pixels × 1024 pixels. The heating rate was set at 150 • C/min starting from room temperature; the first step was at 300 • C, at different magnifications upwards and downwards; continuing to the next temperature step was followed (350 • C, 400 • C, 450 • C, 500 • C, 550 • C, respectively, keeping the ramp at 150 • C/min). The long acquisition time needed for single frames was in the range of 5-10 s and in general, if the sample drift was not too high then a drifting correction was possible, as well as the stability of the Fusion ® holder as presented in ( Figure S2).

X-Ray Photoelectron Spectroscopy (XPS)
XPS was performed in a Thermo-VG Scientific model Escalab-250 equipment (ThermoFisher®Scientific, Waltham, MA, USA) Company, City, State Abbr. and Country.) at ultra-high vacuum (10 −9 mbar) conditions. The data sets were acquired with monochromatic Al Kα radiation source (1486.6 eV), binding energies (BE) were corrected using C 1s (284.6 eV). All data were analyzed with XPS Peak 4.1©, applying a Shirley background subtraction and Gaussian-Lorentzian deconvolution parameters.

Conclusions
A combined study about thermal performance by using in-situ heating-stage-STEM, XPS and a micro-reaction characterization off-line to determined that the catalytic material is composed by Co/MoS2 spherically shaped aggregates catalytic nanoparticles (see supplementary information Figure S1A), with an average size of 2.8 microns (S1) were structural stable even at temperatures of 550 • C in accordance with imaging, turbostratic or layers bending were not observed at those elevated temperatures, as diffraction patterns indicate d(001) = 0.62 nm corresponds to c-axis stacking remains as presented in Figure S3. Furthermore, catalytic experiments using Mexican crude oil (AGT-72) showed partial oil upgrading when a mixture of catalytic material was used; these transition metal sulfides materials led to aromatics rise of 22.65% and the diminution of asphaltenes and resins by 15.87 and 3.53%, respectively.  Figure S2: Collage of pictures of sample holder Fusion© used to perform heating in-situ STEM to determine thermal structural and crystallographic stability of Co/MoS 2 catalytic samples., Figure S3: Diffraction patterns obtained during in-situ heating of spherical shape Co/MoS 2 particles, one can observed that d (002) = 0.62 nm remains at any moment, indicating that stacking of layers remain and not much crystallographic variation was observed even at elevated temperatures of 550 • C.