Elucidating Local Structure and Positional Effect of Dopants in Colloidal Transition Metal Dichalcogenide Nanosheets for Catalytic Hydrogenolysis

Tailoring nanoscale catalysts to targeted applications is a vital component in reducing the carbon footprint of industrial processes; however, understanding and controlling the nanostructure influence on catalysts is challenging. Molybdenum disulfide (MoS2), a transition metal dichalcogenide (TMD) material, is a popular example of a nonplatinum-group-metal catalyst with tunable nanoscale properties. Doping with transition metal atoms, such as cobalt, is one method of enhancing its catalytic properties. However, the location and influence of dopant atoms on catalyst behavior are poorly understood. To investigate this knowledge gap, we studied the influence of Co dopants in MoS2 nanosheets on catalytic hydrodesulfurization (HDS) through a well-controlled, ligand-directed, tunable colloidal doping approach. X-ray absorption spectroscopy and density functional theory calculations revealed the nonmonotonous relationship between dopant concentration, location, and activity in HDS. Catalyst activity peaked at 21% Co:Mo as Co saturates the edge sites and begins basal plane doping. While Co prefers to dope the edges over basal sites, basal Co atoms are demonstrably more catalytically active than edge Co. These findings provide insight into the hydrogenolysis behavior of doped TMDs and can be extended to other TMD materials.


X-ray Diffraction (XRD)
Samples were characterized with XRD using a Bruker AXS D8 Discover GADDS Microdiffractometer at New York University's Shared Instrument Facility.All samples prepared by drop-casting from cyclohexane and drying on clean 1-cm 2 glass slides, and were measured with a Cu Kα source.

Fourier Transform Infrared Spectroscopy (FT-IR)
FT-IR was performed using a Thermo Scientific Nicolet 6700 FT-IR.Samples were prepared by drop-casting from cyclohexane and drying on clean CaF2 crystals.

Atomic Force Microscopy (AFM)
AFM measurements were performed using a Bruker MultiMode 8 AFM operating in tapping mode.Flattening and z-scale adjustment of AFM images was performed with Gwyddion software. 1ermogravimetric Analysis (TGA) TGA was performed using a TA instruments TGA550.The sample was heated in ultrapure nitrogen from 23°C to 100°C at 10°C/min followed by a 15-minute isothermal hold.The sample was then ramped to 400°C at 20°C/min and held at this temperature for 70 minutes.

High Resolution Transmission Electron Microscopy (HRTEM)
HRTEM was performed using a 200 kV FEI Titan Themis Scanning TEM.Samples were prepared by drop-casting dilute samples of nanosheets in cyclohexane (fresh samples) or dimethylformamide (post-HDS) onto carbon coated copper grids and drying at 80°C.

Scanning Electron Microscopy-Energy Dispersive X-ray Spectroscopy (SEM-EDX)
SEM-EDX was performed using a Zeiss Gemini Ultra-55 Analytical Field Emission Scanning Electron Microscope.Samples were prepared by drop-casting on 1 cm 2 silicon substrates.EDX was performed targeting Co, Mo, and S compositions.

X-ray Photoelectron Spectroscopy (XPS)
XPS for examining Co, Mo, and S was performed using a Physical Electronics Versaprobe II XPS.Spectra were collected using an Al Kα source set to 49.4 W and 14.87 keV (250 meV resolution) with a 200 μm beam diameter.The survey pass energy was set to 117.40 eV, while the elemental pass energies were set to 29.35 eV.Samples were prepared by drop-casting nanosheet suspensions onto glass substrates and drying at 70°C for 10 min.Prior to analysis, spectra were corrected by shifting the C1s peak to 284.8 eV.

X-ray Absorption Spectroscopy (XAS)
Samples were measured at Brookhaven National Laboratory using the National Synchrotron Light Source-II (NSLS-II) facility's Quick X-ray Absorption and Scattering (QAS, 7-BM) and Tender Energy X-ray Absorption Spectroscopy (TES, 8-BM) beamlines.For hard Xray energy measurements at QAS, samples were smeared onto clear adhesive tape, folded up to 8 times, and measured at the Co and Mo K-edges in fluorescence and transmission modes, respectively.At the TES beamline, the samples were smeared onto single-layer, non-adhesive 1cm 2 Kapton films and measured at the S K-edge.The collected XAS data were then analyzed using the Demeter software package; XANES and EXAFS were analyzed in ATHENA, while modeling of the EXAFS data for Co and Mo K-edges was performed in ARTEMIS. 2 Parameters used in the fits include an amplitude factor of 0.76 (determined by fitting Co foil) and a k-weighting of 2. The R-range was 1.0-2.3Å, and the k-range was 2.0-12.0Å -1 (dk = 2).At the Co edge, Co-S and Co-O were modeled iteratively and checked for stability.Wavelet transforms were performed using a modification of the code originally published by Muñoz et al. 3 Cauchy order was set to 200, and the R-space distance was set from 0.2 to 6.0 Å (no. of intervals = 200).

Gas Chromatography-Mass Spectroscopy (GC-MS)
Samples taken from the reaction mixture were analyzed using a GC-2030 gas chromatograph and GCMS-QP2020 NX gas chromatograph-mass spectrometer from Shimadzu.A 0.2-μL volume of analyte was injected at a column temperature of 40°C using a Shimadzu AOC-20i autosampler, then after one minute the column temperature increased to 250°C over 10 min.
Prior to analysis, a calibration curve of the area ratio between thiophene and the n-decane reference was prepared using solutions of known concentrations.The concentrations of thiophene in the mixture before and after reaction were measured using the area under the identified peaks in the chromatograph, and the percent conversion of thiophene was calculated for each reaction run as follows:

Turnover Frequency Computation
Per our geometric structure computation (Table S4

Ligand Removal Procedure on Nanoscale MoS2
The procedure for the removal of ligands from our nanosheets is adapted from the work of Meeree Kim et al. 4 As the ligand-removed particles were only used for measurement in X-ray photoelectron spectroscopy (XPS), small quantities were used.
In a typical ligand removal sequence, 20 mg of as-synthesized MoS2 nanosheets are added to a vial inside a nitrogen-filled glovebox, along with 40 mg of nitrosyl tetrafluoroborate (NOBF4), 4 mL of chloroform, and a stir bar.The vial is sealed, sonicated for 10 min, then placed on a stir plate and allowed to stir at 700 RPM for 60 min at ambient temperature.
Inside the glovebox, the ligand-free particles are cleaned twice by addition of 10 mL of hexane and 10 mL toluene and centrifuging at 9500 RPM for 10 min to remove any remaining long-chain organic ligands and excess chloroform.The particles are then cleaned a further two times by the addition of 5 mL dimethylformamide (DMF) and 15 mL toluene to remove excess NOBF4, again centrifuging as above.The particles are then dried under vacuum overnight.After the ligand removal, particles no longer dispersed in hexane but easily dispersed in DMF.

Triangular
An equilateral triangle morphology is assumed with 13 unit cells on each side.The sides are all assumed to be Mo-edge.An equilateral triangle represents the structure with the maximum ratio of edge sites to total Mo atoms.
Table S4.Calculation of atom and site counts for considered geometries of MoS2.

Figure S11 .
Figure S11.Ex situ Mo K-edge XANES of 21% Co:Mo samples fresh, after heating to 300°C in

Figure S12 .
Figure S12.Ex situ Mo K-edge EXAFS of 21% Co:Mo samples fresh, after heating to 300°C in

Figure S16 .
Figure S16.Co K-edge Artemis fitting of 10% Co:Mo post-HDS catalyst EXAFS showing both

Figure S17 .
Figure S17.Co K-edge Artemis fitting of 21% Co:Mo post-HDS catalyst EXAFS showing both

Figure S18 .
Figure S18.Co K-edge Artemis fitting of 39% Co:Mo post-HDS catalyst EXAFS showing both

Figure S20 .Figure S21 .Figure S22 .
Figure S20.Artemis fitting in k-space of the Co K-edge on the 10% Co:Mo post-HDS catalyst.

Figure 23 .
Figure 23.Artemis fitting in k-space of the Co K-edge on the 39% Co:Mo post-HDS catalyst.

Figure S25 .
Figure S25.Co K-edge WT of the 25% Co:Mo catalyst fresh (a) and post-HDS (b).The stronger

Figure S27 .
Figure S27.S 2p orbital XPS of fresh catalyst (ligands removed) compared with bulk MoS2

Figure S33 .
Figure S33.HRTEM of post-HDS 21% Co:Mo MoS2 nanosheets.Population size = 75 in the SI), we estimate an average of 36 edge site atoms for truncated triangular sheets.Combining this with our DFT and XAS observations of, and assuming decorated Co cannot sit next to another decorated Co (per XAS results and steric considerations), we compute that a Co:Mo ratio of 16.2% is the saturation point for the nanosheet edges.Assuming that Co covers an Mo site on the edge but creates a new site on the basal plane, we therefore can compute the number of sites per nanosheet, as used in Figure1f,   is the number of moles of Co present in the Co-doped MoS2 catalyst. is the time of the reaction, in hours.
as follows: If Co:Mo ≤ 16.2%:   =    , If Co:Mo > 16.2%:   =    +    * ( : − 0.162) The quantity Nedge Mo refers to the number of Mo atoms along a single nanosheet's perimeter.Ntotal Mo refers to the total number of Mo atoms present in a single nanosheet.This is based on the assumption that a Co atom on the edge blocks one Mo atom, thus the number of edge sites does not change.However, any Co in excess of the saturation point (16.2%) would likely adsorb to the inert basal plane, meaning it creates an additional site.The turnover frequency per unit Co in Figure1gof the main text is calculated as follows:

Table S1 .
Results and best fit parameters of EXAFS modeling at the Mo K-edge for the different Co:Mo catalysts post-HDS.

Table S2 .
Results and best fit parameters of EXAFS modeling at the Co K-edge for the different Co:Mo fresh catalysts.

Table S3 .
Results and best fit parameters of EXAFS modeling at the Mo K-edge for the different Co:Mo catalysts post-HDS.