Feb-2005

ULSD production at low pressure

A large number of existing diesel hydrotreaters are operated at pressures below 30 bar. The catalyst used for production of ULSD in these hydrotreaters differs from catalyst formulations that perform well at higher pressures

Lars K Skyum, Haldor Topsøe

Article Summary

Technology is available for producing catalysts that show superior activity, even at very low hydrogen pressure. This is important for refineries with hydrotreaters operating at below 30 bar. Recently improved CoMo-based catalysts are creating value in hydrotreaters operating across the entire pressure range, from low- to high-pressure applications, where NiMo catalysts were previously the preferred choice.

Figures 1a and 1b show a comparison of two catalysts in a pilot plant test conducted on straight-run gas oil (SRGO) at a constant temperature and LHSV, but at two pressures. The observed HDS and HDN activity is highest with the proprietary TK-576 BRIM at both pressures. Nitrogen is known to inhibit desulphurisation, and if it is removed to near-zero levels a catalyst with high hydrogenation activity will be the best choice. At the lowest pressure (20 bar hydrogen), the nitrogen is reduced to around 30wtppm with TK-576 BRIM and to 40wtppm with TK-574 (Figure 1a).
With these levels of nitrogen in the product, the removal of sterically hindered sulphur compounds will proceed primarily via the direct desulphurisation route, but also to some extent via the hydrogenation route. When the pressure is increased from 20–30 bar hydrogen, the nitrogen is removed to less than 10wtppm with both catalysts (Figure 1b). At this relatively low nitrogen level, the hydrogenation route will predominate, but some of the refractive sulphur compounds are still removed via direct desulphurisation. Also, in these conditions, TK-576 BRIM provides higher desulphurisation activity than TK-574. When this activity improvement is translated into a difference in temperature, it is concluded that TK-576 BRIM performs 7–8°C better than TK-574.

Figure 2 shows a comparison between both the catalysts on a feed mixture consisting of 50% light cycle oil (LCO) and 50% SRGO at a pressure of 30 bar at the same temperature and LHSV. The LCO contains a high concentration of inhibiting nitrogen compounds and, under the operating conditions chosen for the test shown in Figure 2, the two catalysts do not remove the inhibitors to the level where the hydrogenation pathway is used to remove sulphur. This means the direct desulphurisation route is preferred. Also, in these conditions, TK-576 BRIM is superior to TK-574 and shows an improvement of 5°C.

The conclusion drawn from the tests presented in Figures 1a and 1b and Figure 2 is that TK-576 BRIM has a higher activity than TK-574 for the two important reaction routes in the removal of sulphur; namely, the direct route and the pre-hydrogenation route.

In addition to high start-of-run (SOR) activity, TK-576 BRIM exhibits a high stability in low-pressure operation. Figure 3 illustrates its stability in ULSD conditions in very low-pressure operation. After initial stabilisation, this catalyst was operated in ULSD mode (Step 1). In the second step, which lasted one month, the catalyst was exposed to severe operation at a high temperature, resulting in near-zero product sulphur levels (<1wtppm). In the third and final step, the operation was changed back to the conditions used in Step 1. As seen, the observed activity loss with the catalyst was negligible, despite the rough treatment during the period with accelerated deactivation.

Technology objective
The test results presented in Figures 1a and 1b and Figure 2 confirm that the main objectives in the development of TK-576 BRIM have been achieved: to make a CoMo catalyst with improved activity for direct desulphurisation (more Type II reaction sites), as well as enhanced hydrogenation activity (more BRIM reaction sites).

Basically, direct desulphurisation is important in low-pressure operation, but as the pressure increases the hydrogenation activity of the catalyst gains importance. The preferred catalyst in low- to medium-pressure ULSD hydrotreaters, therefore, has a high activity for desulphurisation via the direct route, as well as via the pre-hydrogenation route. There exists promoters, which are known to enhance the hydrogenation activity on CoMo catalysts, but many of these have the adverse effect on the activity for direct desulphurisation.

Testing results
At the end of 2004, more than 4000 tonnes of BRIM catalysts had been sent to refineries, of which a quarter was TK-576 BRIM for ULSD production. One of the TK-576 BRIM charges was delivered to a Western-European refinery that produces ULSD with either 50 or 10wtppm sulphur at a total reactor inlet pressure below 30 bar. The decision to choose this catalyst was made after  pilot plant testing, which test showed it was possible to improve performance by 6°C, relative to the present operation. Table 1 lists the design feedstock characteristics and operating conditions for the reference unit. As with any ULSD production, the co-processing of HGO makes it difficult to reach the desired sulphur specification due to a relatively large concentration of refractive sulphur compounds and inhibiting nitrogen species in this fraction.

By the end of 2004, TK-576 BRIM was loaded and started up using an SRGO spiked with DMDS. Figures 4a and 4b show the normalised temperature and the product sulphur for the first two weeks of operation. SOR performance was as expected and, as such, the refiner was able to operate at a lower temperature than in previous cycles.