Modeling the Drying Process of Porous Catalysts: Impact of the Pore Size Distribution

Abstract

The distribution of catalytically active species in heterogeneous porous catalysts strongly influences their performance and durability in industrial reactors. A drying model for investigating this redistribution was developed and implemented using the finite volume method. This model embeds an analytical approach regarding the permeability and capillary pressure from arbitrary pore size distributions. Subsequently, a set of varying pore size distributions are investigated, and their impact on the species redistribution during drying is quantified. It was found that small amounts of large pores speed up the drying process and reduce internal pressure build up significantly while having a negligible impact on the final distribution of the catalytically active species. By further increasing the amount of large pores, the accumulation of species at the drying surface is facilitated.

Figure 1

Overview of the transport phenomena coupled within the drying model at the pellet (left) and pore length (right) scale: The porous pellet is exposed to a heated gas flow, leading to a vapor flow across the external boundary. Within the pores, gas (white), gas vapor (light blue), liquid (dark blue), aqueous species (yellow), and energy transport (red) are included. Once the species precipitated within the pore, it is considered to be immobile. Within the pores, the three-phase contact line (with contact angle θ) leads to capillary suction and subsequently a flow of liquid, while evaporation leads to the local generation of vapor.

Figure 2

Drying with uniform distributions of varying widths at T_∞ = 100 °C and liquid viscosity with μ_l = 0.001 Pa·s and μ_l = 0.1 Pa·s: (a) permeability over Δr_p, (b) partial permeability for varying Δr_p, (c,d) vapor flux across surface, and (e,f) precipitate load distribution.

Figure 3

Comparison of uniform (U), monomodal (M), and bimodal (B) distribution within the evaluation range of r_p = 10 ± 2.5 nm: (a) normalized distribution, (b) phase permeabilities, and (c) representative radius with gas–liquid–solid contact line.

Figure 4

Drying with uniform (U), monomodal (M), and bimodal (B) distribution of same pore size spread Δr_p at T_∞ = 100 °C with the liquid viscosity values μ_l = 0.001 Pa·s and μ_l = 0.1 Pa·s: (a,b) vapor flux across surface and (c,d) precipitate load distribution.

Figure 5

Water vapor flux J_w over time with varying f_U compared against a purely monomodal distribution (ref).

Figure 6

Derived properties of the widely spread distributions: (a) total permeability, (b) phase permeability, and (c) largest filled radius.

Figure 7

Results for drying with widely spread pore size distributions with varying V_p,2: (a) water vapor flux at the surface, (b) average saturation, (c) gas pressure at center of the sphere, and (d) precipitate load distribution.