Catalysis 2023 Issue

15 10 5 20 25 30 40 35 50 45 65 55 60

SRGO + 15% LCO

SRGO + 30% LCO SRGO + 45% LCO

SRGO + 15% LCO

SRGO feed

Sulphur target = 8 ppmw

100

150

200

250

300

350

400

450

500

550

Elapsed run time , h

3 4 5 6 7 8

CoMo CoMo/NiMo/CoMo NiMo NiMo/CoMo

Figure 6 Sulphur product trend along both runs

As expected, we can see in Figure 5 that the highest hydrogen consumption during each condition is obtained using the NiMo catalyst, while the CoMo catalyst has the lowest hydrogen consumption. Hydrogen consumption for the NiMo catalyst is almost twice as much as for the CoMo catalyst. This result could be explained by the fact that NiMo catalysts better promote the indirect sulphur removal mechanism, as briefly pointed out in Figure 1. It is a known fact, that NiMo catalysts are inferior to CoMo ones when it comes to hydrodesulphurisation (HDS) activity, therefore shifting the reaction pathway to the pre-aromatic satura- tion route for better sulphur accessibility. Additionally, the hydrodearomatisation (HDA) activity of NiMo catalysts is stronger than that of CoMo catalysts, hence there is more aromatics conversion in parallel. Lastly, the extent of nitrogen conversion, which requires pre- aromatic saturation, is also higher for NiMo catalysts from their superior hydrodenitrogenation (HDN) activity. These combined effects substantially increased total hydrogen consumption when 100% NiMo catalysts were used. Similarly, hydrogen consumption obtained for the mixed catalytic beds (NiMo/CoMo) averaged between the NiMo and CoMo catalysts, with slightly higher consumption (around 5%) obtained in the case of the ‘sandwich’ con - figuration CoMo/NiMo/CoMo. For the NiMo/CoMo configuration, part of the NiMo cata - lyst was used for direct sulphur removal of easier sulphur species, such as sulphides, thiophenes, and benzothio- phenes. These easy sulphur species do not really require pre-aromatics saturation prior to sulphur abstraction (see Figure 1 for mechanism details). As such, this part of the NiMo catalysts consumed approximately the same hydro- gen as when using CoMo catalysts for the same purpose. Consequently, there was less NiMo catalyst volume available for HDN, indirect HDS, and HDA in the subse- quent reactor section, thus lowering the total hydrogen consumption (fewer NiMo active sites to promote aromatic saturation). On top of that, more CoMo catalyst volume was available in the last section of the reactor for converting

more difficult sulphur species (alkyl DBT). As such, the direct sulphur abstraction pathway (fewer molecules of hydrogen consumed, also see Figure 1) dominated in this last reactor zone. In contrast the CoMo/NiMo/CoMo scheme initially uti - lised CoMo catalyst in the frontal section for the treatment of easy sulphur compounds, leaving more NiMo catalyst volume in the subsequent reactor zone to promote aro- matic saturation and effectively convert organic nitro- gen compounds (ULSD inhibitor – requires pre-aromatic saturation). The HDS rate is relatively slow in this reactor section as the remaining sulphur species became more sterically hindered (below 500 ppmw level – pre-aromatic saturation route dominates) with relatively high levels of inhibiting organic nitrogen. The use of CoMo catalyst near the reactor outlet opti- mised hydrogen consumption in the low nitrogen zone, for example <80 ppmw,3 by promoting direct sulphur abstrac- tion (less hydrogen consumed pathway). The HDS rate for this last zone became much faster than the previous zone as there was less organic nitrogen to inhibit sulphur con- version. This catalyst positioning strategy explains why total hydrogen consumption was slightly higher than NiMo/ CoMo stacking, as the NiMo catalyst did more hydrogena - tion by employing this sandwich design. Hydrodesulphurisation The main objective of the ULSD process replicated during this experimental program is to reduce the total sulphur content in the feed below 10 ppmw. For this study, a target product sulphur of 8 ppmw was selected, so the SOR operating tem- perature was estimated for reaching such conversion. As already mentioned, the initial temperature provided was overestimated, causing overtreating of the feed (very low sulphur product), so this was immediately adjusted to reach the desired product sulphur. Figure 6 shows the profile of the sulphur product obtained during the test. A slightly rising trend in product sulphur was observed during most of the operating conditions,

Catalysis 2023