An experimental and modeling study of vacuum residue upgrading in supercritical water

Abstract

Arabian Heavy crude oil was fractionated into distillate and vacuum residue fractions. The vacuum residue fraction was treated with supercritical water (SCW) at 450 °C in a batch reactor for 15 to 90 minutes. The main products were gas, coke, and upgraded vacuum residue; the upgraded residue consisted of gasoline, diesel, and vacuum gas oil range components. The molecular composition of gas and upgraded vacuum residue was analyzed using gas chromatography (GC, GC×GC). SCW treatment converted higher carbon number aliphatics (≥ C21) and long chain (≥ C5) alkyl aromatic compounds into C1-C20 aliphatics, C1-C10 alkylaromatics and multi-ringed species. The concentrations of gasoline and diesel range compounds were greater in the upgraded product, compared to the feed. A first-order, five lump reaction network was developed to fit the yields of gas, coke, diesel and gasoline range components obtained from SCW upgrading of vacuum residue. Distillation of crude oil followed by SCW treatment of the heavy fraction approximately doubled the yield of chemicals, gasoline, and diesel, while forming significantly less coke than conventional upgrading methods. ** Many of the results in this manuscript were presented by S. Gudiyella at the 2016 AIChE Annual Meeting, identified as the Best Presentation in the session "Reaction Engineering of Biomass and Hydrocarbons in Supercritical Water" by the Session Chair K. Choi.

Chemicals from heavy oils by ZSM‐5 catalysis in supercritical water: Model compound and reaction engineering

“…Accomplishing this requires understanding the effect of water on reaction rates and mechanisms, which goes beyond the observations of conversion and yield previously made by Zaker et al 42 Because of the complexity of the reaction mixture, we adopt a group‐type modeling approach to capture the effects of water on broad kinetic phenomena. Gudiyella et al 2 used a similar approach to quantify the role of SCW on the rates of atmospheric residue upgrading, motivating its application to catalytic systems.…”

“…Figure S5 provides empirical data on >50 species, and it shows that addition of SCW influences product distributions obtained from ZSM‐5‐catalyzed dodecane cracking. To develop a reaction engineering model for quantitative determination of the effect of water on dodecane cracking rates, we must first reduce independent species from >50 to a manageable number, as is common practice in group‐type models 2,43‐49 . The specific product categories used in group‐type models place implicit limits on the insight that they can provide, so groups must be judiciously selected based on mechanistic insights and experimental observations.…”

“…Consistent with typical practice, 2,47,64,65 all reactions were assumed to be first order with respect to mass yield and all rate constants are reported with dimensions of inverse time, that is, hr −1 . With this assumption, the mathematical model describing the reaction network shown in Figure 3 becomes the following set of ordinary differential equations: where D refers to mass yield of the dodecane feed, Ar is aromatic, Al is aliphatic, C is coke, and G is gas.…”

“…The reaction network theory is then used for comparison of quantitative and qualitative features of ZSM‐5‐catalyzed dodecane cracking in the presence and absence of SCW. This work provides new insight into the role of water in ZSM‐5‐catalyzed hydrocarbon cracking reations under supercritical conditions, the understanding of which may be useful for many petroleum and renewable oil applications 2,11,18 …”

“…For heavy oil, SCW upgrading can eliminate some of the drawbacks of conventional upgrading techniques, such as the demand for excess hydrogen, which is environmentally 15 and economically disadvantageous 1,16 . In addition, some studies have shown that SCW modifies the reaction network and minimizes carbon rejection as low‐value coke 2,17 . The mechanism of coke suppression is an open question, with the most plausible explanations invoking solvation and dispersion of reactive intermediates to reduce repolymerization reaction rates 7,18‐20 .…”