PTQ Q2 2023 Issue

Kinetic model for TGU hydrogenation reactors: Part 2 Catalyst model validation

A rigorous high-fidelity kinetic model can help designers and operators forecast the life expectancy of reactor catalyst beds

Michael A Huffmaster Independent Consultant Prashanth Chandran, Nathan A Hatcher, Daryl R Jensen and Ralph H Weiland Optimized Gas Treating, Inc.

P art 1 of this article, published in PTQ ’s Q1 2023 issue, involved the development of a model for the reactions occurring in the hydrogenation reactor of a sulphur recovery unit (SRU). The model addressed 11 different reactions and alluded to the catalyst ageing and poisoning that inevitably occurs over the life of the catalyst. Part 2 takes a deeper look at catalyst deactivation from a modelling standpoint and quantifies how deactivation can be included in the reaction kinetics model. Part 3 will detail catalyst deactivation and poisoning mechanisms. The model is validated with a case study. Catalyst ageing and poisoning model Ultimately, catalyst activity sets the sulphur recovery per- formance of the tail gas unit (TGU). Deactivation of het- erogeneous catalysts, such as the ones used in the TGU hydrogenation reactor, occurs by ageing and poisoning; this is a ubiquitous problem that causes gradual loss of catalytic rate. For a comprehensive TGU design, catalyst deactivation over the life of the catalyst charges and its effect on meeting sulphur emission requirements must be addressed. When fresh catalyst is loaded into the TGU reactor and activated, it has maximum surface area and activity. On start-up, the catalyst is immediately exposed to several possible deactivation stresses, most causing irreversible damage. Mechanisms that alter catalyst activity do so by affecting the dispersed, active, metal-sulphide phases of cobalt and molybdenum and the high surface area alumina support. Alumina (and titania) are often used in the pro- cess industry as supports for many heterogeneous cata- lysts, as well as for the Claus process, so one can draw on this larger body of knowledge and on sulphur recovery industry experience. The activity of these catalysts is strongly related to the γ -alumina (or mixed phase alumina) surface area of the base, alumina crystallites and their microporous structure that facilitates accessibility to the reactants. The alumina matrix has hydroxyl ions on the catalyst surface that serve as weak Brønsted-Lowry acid sites, promoting hydrolysis and Claus reaction. Their extensive surface area both sup - ports and interacts with the active cobalt and molybde - num metals. Activity declines as a function of time, and exposure to

250 300 350

Waterton, area m/g COS, ppm

200

150

100

120

144

Months in service

Figure 1 Puget Sound Refinery, sulphur-plant samples from SCOT reactor, lab data 1

normal process conditions is treated as ageing and related to loss of surface area and active sites. The remaining frac- tion active surface area can be represented by an ageing factor, AF. Ageing tends to occur uniformly throughout the catalyst bed, with catalytic activity or conversion of reac - tive species declining rapidly at first and then slowly over the catalyst’s life. Spent catalyst activity approaches about 50% of fresh activity, and the model is fitted to this operat - ing data for ageing, as observed in the measurements of the sulphur-plant data shown in Figure 1 . 1 For a comprehensive TGU design, catalyst deactivation over the life of the catalyst charges and its effect on meeting sulphur emission requirements must be addressed Assays of used TGU catalysts report surface area, crush strength, carbon, and sulphate. 2 Typically, when the surface area reaches 120 m 2 /g, they are considered spent. Other ‘spent’ criteria are carbon-on-catalyst as seen at levels approaching 1% and crush strength declining to half the fresh value. Sulphate is not always observed, but about 1% is not unusual (although spent catalyst may have substantial

PTQ Q2 2023