Effect of blending on hydrotreating reactivities of atmospheric residues: Synergistic effects. (1st June 2021)

• Record Type: Journal Article
• Authors: Cui, Qingyan; Ma, Xiaoliang; Nakano, Koji; Nakabayashi, Koji; Miyawaki, Jin; Al-Mutairi, Adel; Marafi, Abdulazeem M.J.; Al-Otaibi, Ahmed M.; Yoon, Seong-Ho; Mochida, Isao
• Is Part Of: Fuel. Volume 293(2021)
• Publication Date: 2021-06-01
• Subjects: Atmospheric residue -- Hydrotreatment -- Blending -- Synergy; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2021.120429
• Languages: English
• ISSNs: 0016-2361

Abstract

Highlights: Different synergistic effects were found in hydrotreating the blends of two ARs. The synergistic effects were attributed to the change of SARA fractions in blends. A synergistic effect index was proposed. Composition of product from two AR blends can be estimated on the developed models. Abstract: In order to clarify the effect of blending heavy atmospheric residues (AR) and light AR on reactivities for hydrodesulfurization (HDS), hydrodenitrogenation (HDN), hydrodevanadium (HDV), hydrodenickel (HDNi), hydrodeasphaltenes (HDAsp) and hydrodeconradson-carbon-residue (HDCCR) in hydrotreating process, hydrotreating of the blends from two ARs at different ratios was conducted in a fixed-bed unit at 370 °C under a H2 pressure of 13.5 MPa over a set of commercial HDM and HDS catalysts. The compositions and properties of the two ARs and their blends as well as their hydrotreated products were analyzed and compared in detail. A positive synergistic effect of the blending was found for HDS, HDN and HDCCR, while a negative synergistic effect of the blending was observed for HDAsp, HDV and HDNi. A synergistic effect index ( ηi, s ) was proposed to describe quantitatively the extent of the blending synergistic effect. The observed synergistic effects can be attributed to change of the aromatics-to-saturates ratio in the blends. In addition, composition and property of the hydrotreated products from the blending feedstocks with different blending ratios were well estimated on the basis of the developed models with an average relative deviation less than 1.5 %.