CoMoNi Catalyst Texture and Surface Properties in Heavy Oil Processing. Part II: Macroporous Sepiolite-Like Mineral Victoria S. Semeykina,* ,,,§ Ekaterina V. Parkhomchuk, ,,§ Alexander V. Polukhin, , Pavel D. Parunin, ,,§ Anton I. Lysikov, , Artem B. Ayupov, Svetlana V. Cherepanova, , Vasily V. Kaichev, and Tatyana S. Glazneva , Boreskov Institute of Catalysis SB RAS, Lavrentieva Avenue 5, Novosibirsk 630090, Russia Novosibirsk State University, 2 Pirogova Street, Novosibirsk 630090, Russia § Research and Education Center, Novosibirsk State University, 2 Pirogova Street, Novosibirsk 630090, Russia * S Supporting Information ABSTRACT: A set of novel CoMoNi hydrotreating catalysts supported on sepiolite-like mineral and modied by H 3 PO 4 have been prepared and studied in hydrodesulfurization (HDS) and hydrodemetallization (HDM) of heavy Tatar oil with extremely high viscosity and sulfur content. Catalysts had a multiphase composition, represented by calcium/magnesium oxides, silicates, or phosphates, and were found to be of great interest for studying the role of support surface properties in heavy oil hydrotreating. For monitoring the catalyst properties, all the samples have been investigated by X-ray diraction (XRD), X-ray uorescence spectroscopy (XFS), X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM), CO- and CDCl 3 -Fourier transform infrared (FTIR), mercury porosimetry, and N 2 adsorption methods. The catalyst with a small phosphate content showed higher initial HDS conversion due to the more developed specic surface area, increased Lewis acidity, and better active component distribution; however, the sulfur removal substantially reduced during 240 h on stream. The sample with a high amount of phosphates demonstrated better stability, higher HDM, and hydrocracking/ hydroisomerization activity despite lower acidity and poor active phase dispersion that may be accounted for by the higher fraction of macropores. 1. INTRODUCTION A typical catalyst employed for crude oil hydrotreatment conventionally includes shaped alumina-supported Mo sulde promoted by Co or Ni. According to the widely known theory of Co(Ni)MoS active species, proposed by Topsoe et al., 1 for the best catalytic performance structure of active component should be represented by slabs of MoS 2 with edges decorated by promoter atoms. Two types of active sites have been put forward in this hypothesis: the edge sites, responsible for direct desulfurization/denitrogenation of reactant molecules, and brim sites, exhibiting activity both in HDS/HDN and hydrogenation. The remote control theory, proposed by Delmon et al. 2,3 turns particular attention to the eect of hydrogen spillover: migration of hydrogen forming on Co(Ni)S x phase (donor) to MoS 2 species (acceptor). The theory emphasizes that support properties such as isoelectric point, acidity and continuity as well as a distance between donor and acceptor species directly aect the hydrogen spillover rate. It should be stressed that accordingly to P. Baeza et al. 4 sepiolite was shown to be the most eective support from this point of view. Various admixtures such as alkaline and earth metals, boron, phosphorus, and inorganic oxides are proved to inuence the catalytic performance of alumina-supported Mo sulde promoted by Co or Ni. Apart from shifting acid-base properties of the support, they result in changing the active component characteristics such as dispersion and an electronic state. Despite some contradictory data reported on similar catalytic systems, most researchers accept the proposal that uoride additives 5 and oxide additives such as SiO 2 , 6,7 B 2 O 3 , 7,8 and P 2 O 5 9,10 increase Lewis/Bronsted acidity of the support and enhance hydrocracking/isomerization activity with hydro- desulfurization (HDS) being slightly increased, decreased, or unchanged that suggested the presence of two types of catalytic sites, the sites on the molybdenum suldes responsible for hydrogenation and C-S(N) bond cleavage and the support acidic groups providing hydrocracking and isomerization. However, the eect of admixtures on active component dispersion was also taken into consideration: it was shown that phosphorus, silica, and boron additives diminished active Received: March 29, 2016 Revised: July 29, 2016 Accepted: August 1, 2016 Article pubs.acs.org/IECR © XXXX American Chemical Society A DOI: 10.1021/acs.iecr.6b01208 Ind. Eng. Chem. Res. XXXX, XXX, XXX-XXX