Jul-2002

Best practices for producing ultra low sulphur diesel

This article presents a methodology for approaching low sulphur diesel design with low pressure reactor solutions. It reviews commercial cases showing proven best practices can be applied to existing hydroprocessing assets to produce ULSD

Sal Torrisi, Dave DiCamillo and Richard Street, Criterion Catalysts & Technologies
Tom Remans, Shell International Chemicals Research
Jahn Svendsen, Shell Global Solutions

Article Summary

As refiners begin to contemplate options to meet ultra low sulphur diesel (ULSD) fuel production (<15wppm sulphur) by 2006, many are concerned that their existing equipment cannot be modified to deliver the required performance. Although utilising existing assets would be the best financial option, many refiners are hearing that high-pressure solutions requiring sizable grassroots investment are the only option. But experience has shown that ULSD can be produced at moderate pressures via unit revamps or grassroots investment for a more cost effective solution.

By understanding the ULSD chemistry, the latest generation catalysts can be applied (in the right combinations), with current process/hardware know-how and proper hydrocarbon management, to achieve <15wppm sulphur at today’s typical diesel hydrotreater pressures. This knowledge plus proper planning can confirm that current refinery capabilities are able to meet the future requirements. Not only does this approach maximise investment returns for ULSD, it allows the refiner to take advantage of improved unit capabilities now to realise immediate financial benefits.
Once the problem is understood, it then becomes an exercise in applying the different tools available, which can improve unit performance to determine whether a revamp option is feasible. Determining whether a given refinery can utilise existing units or equipment to make the jump from current sulphur levels down to <15wppm sulphur can seem an overwhelming endeavour at first. Getting acquainted with the chemistry required to achieve ULSD is a good place to start.

The importance of partial hydrogenation of multi-substituted dibenzothiophenes as an HDS prerequisite when product sulphur is pushed below 50 wppm has been well reviewed in the industry. It is equally important to understand the reactivity of these refractory sulphur species and nature of a given feed. Feedstock properties like crude source, sulphur distribution, prior processing history (catalytically cracked, thermally cracked, hydrotreated) and cut point, all play a role in determining the feedstock reactivity and subsequent processing conditions necessary to produce ULSD.

Applying new generation catalysts and different catalyst combinations can make significant progress towards  lowering product sulphur, but in most cases this is not a total solution. Because ULSD production can become a complex issue, the evaluation effort should also consider utilising best available reactor internals, which can provide as big a boost to performance as the next generation catalyst, especially considering that reactor hardware has improved substantially even over the last five years.

Lastly, hydrocarbon management is critical. Unit designs can be greatly affected by crude selection, feed balancing between units, cut point adjustment to exclude refractory sulphur, and product blending of jet, diesel and heating oil pools.

It is imperative to organise the evaluation effort and weigh options systematically. A single technology or standard approach will not solve all problems. It is essential that vendors work closely with customers to understand refinery capabilities, identify bottlenecks and limitations, and work together to provide a solution that meets ULSD requirements, unit flexibility and investment criteria within the desired time frame.

Ability to meet 15ppm sulphur
Is the facility’s current configuration able to competitively produce diesel with less than 15ppm sulphur? The answer is usually framed in terms of increased reactor temperature required (Treq) or improved catalyst activity. Note that catalyst activity and reactor volume are somewhat interchangeable variables to put the adjustment in perspective. The answer can vary significantly depending on the nature of the feedstock and other operating conditions. Due to the refractory nature of the last-removed sulphur species, the additional Treq can actually be larger for the last 20wppm as it was for the first as 270wppm.

In order to assess an existing diesel hydrotreater (DHT) operation and begin to develop ULSD solutions, it is helpful to examine a specific case and evaluate incremental improvements to see if they are sufficient to make the transition to ultra-low sulphur levels. Assuming that a conventional CoMo catalyst is being used to make 350–500wppm product sulphur today, two solutions could be to either drop in a catalyst with three to  four times the current activity level or raise the start-of-run (SOR) reactor temperature by ~100°F to achieve <15wppm product sulphur with no other unit or feedstock modifications.

While either option can be part of the ultimate solution, other modifications are necessary to supplement better catalysts or reactor temperature changes in order to make ULSD with a reasonable unit cycle length.

Disposition management
Good hydrocarbon management is critical to ensure that refinery economics are not compromised. This element is especially important in refineries that currently operate or will manage multiple units to produce ULSD in the future. Matching the right feeds with the existing unit capabilities will maximise utilisation of existing capital investment and may minimise grassroots investment.

In general, the more difficult feeds containing LCO, coker distillates and even heavy SRGOs should be matched with existing higher-pressure units within a given site, although this is not always the case, depending on the technology solutions employed. Feed undercutting has been commonly employed to remove the heavier-boiling refractory sulphur species and enable a much easier feed to be treated in the existing units. Significant progress towards  lowering the product sulphur with the existing reactor capabilities can be made by excluding most of the refractory sulphur species that boil above 640°F. A unit today processing a feedstock with a T90 of 680°F and making 500wppm product sulphur can cut levels to 50–100wppm by reducing the T90 to 645°F as shown in Figure 1, because almost all of the sterically hindered multi-substituted dibenzothiophenes are excluded.

This undercutting can effectively increase HDS activity by 60–70 per  cent or reduce the SOR Treq by 25–30°F, because the remaining sulphur species are much easier to process. While in principle this modification is easy to implement, the implication of downgrading a non-trivial amount (10–20 per cent) of middle distillate to a lower value product disposition must be considered, as must whether reduction of capital investment is justified against overall refinery economics.