PTQ Q1 2026 Issue

Key differences between hydroprocessing units at different scales

Industrial

Large pilot plant

Bench-scale pilot plant

High-throughput (microreactors)

Catalyst volume (L) Diameter reactor (cm) Length reactor (cm)

100,000

5 4

0.2

0.002

250

0.5 10

2,000

400 3.98 0.35

Volumetric flux (m3/m2/h) at LHSV = 1 h -1 Linear liquid velocity at LHSV=1 h -1

20.4 0.88

0.64

0.10

0.021

0.004

Table 1

between refiners and catalyst vendors, with refiners com - monly specifying the Request for Proposals (RFPs). These tests are conducted at refineries’ corporate research labs, third-party facilities, or vendor testing facilities (for example, Topsoe R&D labs). Independent testing enables a side-by- side comparison of different catalyst brands under identical feedstocks and operating conditions. Additionally, pilot plants provide a controlled environment to simulate operational changes, such as feedstock variations or process conditions, and evaluate their impact on catalyst performance. This allows refiners to identify optimal operat - ing strategies without jeopardising the commercial full-scale unit performance. Key disparities between hydroprocessing units at different scales Table 1 highlights key differences in reactor dimensions when operating at a constant LHSV of 1 h - ¹. Due to their significantly smaller reactor size, the equivalent linear liquid velocity in high-throughput micro-reactors can be up to 220 times lower than in industrial units. These geometric dis - parities among TBRs from commercial-scale to pilot-scale create scalability constraints. As a result, matching both liquid hourly space velocity (LHSV) and superficial velocities between pilot and industrial reactors is not feasible. For the same LHSV, the mass flux in pilot-scale units can be approxi - mately 30 times lower, and even more so in high-throughput systems, leading to potential mass transfer limitations and deviations in observed catalyst performance. In short reactors with low linear velocity, there is a forma - tion of a thick liquid film and poor mass transfer between the treat gas and the oil (see Figure 2 ). This phenomenon results in 80-95% of the hydrogen consumption occurring within just 10-20% of the reactor height. In the reactor’s upper

section (see Figure 3 ), hydrogen availability inside the cata - lyst pellets can become limited once the dissolved hydrogen in the feed is depleted, often manifesting as an increased catalyst deactivation rate. Pilot reactors operate at lower superficial mass velocities, so pilot plant design, choice of reactor, and testing conditions should focus on eliminating heat and mass transfer effects. Not eliminating such effects will lead to incomplete catalyst wetting, high axial dispersion, and inter- and intraparticle gradients, resulting in deviations from the true kinetics of the catalytic reactions and reduced conversion. These become more significant in high-throughput micro-scale reactors and will, in most cases, result in inaccurate catalyst comparison. Overcoming scalability constraints: Optimising pilot plant testing Pilot plant testing has become a cornerstone of catalyst qualification; however, the reliability of data depends heavily on the way tests are designed and executed, as well as the design of the pilot plant’s test unit. Variations in activation conditions, temperature profiling, and hydrogen distribution can lead to large discrepancies between pilot and commer - cial performance. To ensure that pilot results provide mean - ingful differentiation between catalysts, test programmes should emphasise gradual temperature step-ups, consistent sulphiding protocols, and sufficient feed sulphur and nitro - gen levels to maintain realistic hydrogen demand. These measures enhance comparability and prevent early satura - tion of catalyst activity. To mitigate scale-induced deviations in pilot-scale TBRs, it remains essential to minimise axial dispersion, ensure uni - form catalyst wetting, and limit wall flow. Table 2 summarises the key parameters that must be controlled during pilot plant testing to ensure reliable catalyst evaluation and meaningful

Gas phase Film diusion rate

< Consumption rate

Liquid phase

Catalyst surface

0.1 0.2

0.3

0.4 0.5

0.6

0.7

0.8 0.9

No hydrogen => coking

Dimensionless bed length

Figure 3 Hydrogen consumption in small reactor

Figure 2 Diffusion limitation in small-scale reactors

PTQ Q1 2026