German refiner debottlenecks diesel hydrotreaters for new sulfur specs

Sept. 10, 2001
Bayernoil Raffinerie- gesellschaft mbH successfully revamped two of its three diesel hydrotreaters in preparation for Germany's introduction of ultra low sulfur diesel (ULSD), that is, diesel with a sulfur content of less than 50 ppm (wt), in November 2001.

Bayernoil Raffineriegesellschaft mbH successfully revamped two of its three diesel hydrotreaters in preparation for Germany's introduction of ultra low sulfur diesel (ULSD), that is, diesel with a sulfur content of less than 50 ppm (wt), in November 2001.

Bayernoil, a joint venture of BP plc, Agip SPA, Veba Oil Refining & Petrochemicals GmbH, and Petróleos de Venezuela SA (PDVSA), operates three refineries in Bavaria, Germany. Based on the success of these two revamp projects, Bayernoil is evaluating options to revamp its third hydrotreater to produce 10 ppm(wt) diesel.

Test run data show that the revamps were economically and functionally successful. The revamps improved the desulfurization capabilities of these two diesel hydrotreaters.

The European Commission has mandated a reduction of diesel-fuel sulfur level from 500 ppm(wt) to 350 ppm(wt) in 2000, with a further reduction to 50 ppm(wt) in 2005.

In the UK and several northern European countries, governments are encouraging 50 ppm (wt) diesel through tax incentives. Germany is planning an early introduction of 50 ppm (wt) diesel for Nov. 1, 2001 and is considering tax incentives to encourage refiners to further reduce their diesel sulfur content to 10 ppm(wt) by Jan. 1, 2003.

In the US, the Environmental Protection Agency recently mandated 15 ppm(wt) highway diesel, starting in mid-2006.

Revamp options

Since every refiner processes a different crude slate and has different processing equipment, the choice of technology options to produce ULSD is site specific. In many cases, a combination of options may meet each refiner's needs at the lowest capital expenditure.

On a limited basis, refiners have implemented and commercially demonstrated many diesel hydrotreater revamp options in recent years to produce ULSD.1-5

The industry used many of the same approaches when diesel sulfur specifications were reduced in the US to 500 ppm (wt) in 1993 and in the EU to 350 ppm(wt) in 2000. Meeting the 50 ppm(wt) cap is significantly more difficult than meeting 350 or 500 ppm(wt) rules, however.

In the ascending order of capital cost, these revamp options include:

  • Use of improved catalyst. Recent advances in hydrotreating catalyst technologies have significantly improved the sulfur removal capability of catalysts.4-7 For example, new catalysts developed by Akzo Nobel Catalysts BV, Amersfoot, the Netherlands (KF757, KF848); Haldor Tops e AS, Lyngby, Denmark (TK573, TK574); and Criterion Catalysts & Technologies, Houston (Century and Centinel lines), reportedly increase desulfurization activity by 25-75% more than the catalysts used in the mid to late 1990s.

The use of improved catalyst alone, however, only reduces the sulfur level from the current 340 ppm(wt) average sulfur in US highway diesel to 100-250 ppm(wt). This is still a long way from meeting the 50 ppm(wt) or lower specifications. Additional changes will be required to produce significant amounts of ULSD.

  • Adjustment of feed end point and feed composition. One of the easiest ways to produce ULSD is to use "easier feeds" such as straight-run (SR) kerosine and light gas oil (LGO). By cutting out most of the refractory sulfur compounds (such as sterically hindered dibenzothiophenes) contained in high end point materials and in highly aromatic feedstocks (such as light cycle oil (LCO) and coker LGO), many of the existing hydrotreaters can produce ULSD if combined with the use of high activity catalyst.

Removing these heavier fractions, however, severely reduces the amount of ULSD that can be produced. Moreover, refiners have to find a home for the high end point materials (for example, heating oil).

  • Increase of the reactor temperature. Raising the reactor temperature increases the desulfurization capability of the existing hydrotreater. The effectiveness of this option is quite limited, however, because it incurs additional catalyst and downtime costs as a result of a shorter catalyst life.
  • Improvement of reactor efficiency. Improved vapor-liquid contacting in the hydrotreating reactors can significantly decrease the temperature requirement to achieve the same level of desulfuriza- tion.2 3 8 9 The extent of improvement depends on the type of existing vapor-liquid distributor being used, the feedstock quality, and the degree of desulfurization desired.
  • Additional reactor volume. Adding more catalyst is another simple and moderate cost way to increase desulfurization capability of the existing unit.

The amount of additional catalyst volume depends on how difficult the feed diet is. Very substantial increases of catalyst volume (up to 5-10 times of the existing size) may be required if the feed contains large amounts of LCO.

  • Increase of hydrogen-to-oil ratio. Increasing the treat gas rate (hydrogen-to-oil ratio) can enhance the desulfurization activity of the catalyst by reducing the inhibition effect of hydrogen sulfide (H2S) and ammonia (NH3).

The effect is rather small compared to the needs for achieving 50 ppm(wt) or lower, however. For example, increasing the hydrogen-to-oil ratio by 50% may gain only 4-8° C.

  • Removal of H2S from the treat gas. The catalyst desulfurization activity can also be improved by removal of H2S in the treat gas (recycle gas plus makeup gas). Although the effectiveness of this option is greater than increasing the hydrogen-to-oil ratio, the cost of implementing this option (such as with a high pressure amine scrubber) should be evaluated against the use of other options.
  • Increase of hydrogen partial pressure. Increasing hydrogen partial pressure is another option that will improve sulfur removal and prolong catalyst life. Refiners should evaluate the cost associated with this option (such as hydrogen purification or imported high purity hydrogen) against other options, however.

Case 1: diesel hydrofiner revamp

In Case 1, Bayernoil revamped its 16,600 b/d diesel hydrofiner. By replacing two small reactors with a larger reactor, the unit can now produce ULSD with less than 10 ppm sulfur with a typical feedstock mix of SR gasoil and up to 20% FCC LCO (Fig. 1).
Click here to enlarge image

Bayernoil revamped one of its three diesel hydrotreaters, a diesel hydrofiner, to make 50 ppm sulfur diesel (Fig. 1).

The hydrofiner is a high-pressure unit (70.3 kg/sq cm or 1,000 psig) with two small reactors running in parallel. At a nominal throughput of 110 cu m/hr (16,600 b/d), the liquid hourly space velocity is in excess of 4 hr-1, and the maximum treat gas hydrogen-to-oil ratio is 168 normal cu m hydrogen/cu m (1,000 scf hydrogen/bbl).

There is no H2S scrubber in the recycle gas loop.

The unit cannot produce 50 ppm(wt) automotive diesel without significantly derating the unit or lightening the feed slate. A typical feedstock consists of a mixture of the 40-50% SR LGO, 20-40% SR HGO, 0-15% fluid catalytic cracker (FCC) naphtha (or SR kerosine), and 0-20% FCC LCO.

Bayernoil evaluated several options for both technical feasibility and capital costs. These options included: adding catalyst volume by adding a new reactor, increasing the hydrogen-to-oil ratio by adding a new recycle-gas compressor, removing H2S from the recycle gas by installing a scrubber, or increasing hydrogen partial pressure by installing a hydrogen-purification system.

Based on these evaluations, Bayernoil decided to replace the existing two small reactors with a new larger reactor capable of producing 50 ppm(wt) or lower sulfur. To minimize the size of the new reactor, the company needed an effective, active catalyst.

Bayernoil solicited several process/catalyst vendors and engineering contractors to provide the reactor design. It eventually awarded the contract to Mobil Technology Co. (now ExxonMobil Research & Engineering).9

The new reactor came on stream in October 1999. Table 1 summarizes test-run results to determine the effectiveness of the new reactor.

Comparison of test-run data is complicated because different catalysts were used before (KF757) and after (regenerated KF752) the installation of the new reactor. Bayernoil uses regenerated KF752 because the unit is not required to produce 50 ppm(wt) in 1999-2001; ULSD is not required until November 2001.

Besides the catalyst, the feed diet also varied before the revamp, after the revamp, and during the cycle.

Click here to enlarge image

Nevertheless, the performance of the new reactor exceeded that of the old reactors (Table 1). The improvement is a result of a combination of additional catalyst volume and the use of state-of-the-art reactor design technology.

The unit can currently produce ULSD with less than 50 ppm(wt) sulfur with its regular feed diet (up to 20% LCO) and still achieve 18-24 months cycle length if high activity catalyst such as KF757 or equivalent is used.

Click here to enlarge image

Reactor radial temperature profile at the inlet and outlet of the catalyst beds is an important indicator of how well the reactor internals perform. Small deviations in radial delta temperatures are indicative of well designed reactor internals.10 11 Fig. 2 shows the new reactor can achieve less than 2.5° C. in radial delta temperatures both at the inlet and outlet of each bed.

Although the reactor is equipped with a quench zone, the reactor uses no quench gas during normal operations. Thus, the Bed 2 inlet temperature is nearly the same as the Bed 1 outlet.

Case 2: diesel HDS-II revamp

Bayernoil undertook the second project, a revamp of its diesel HDS II unit, to also improve its desulfurization capability. As well as reducing the sulfur content of the diesel fuel, the revamp increased unit throughput from 15,000 b/d to 18,000 b/d.

Diesel HDS-II is a low-pressure unit (32 kg/sq cm or 450 psig) with a single reactor operated at a nominal throughput of 120 cu m/hr (18,000 b/d).

The existing feed distributor tray is a simple v-notched chimney tray. Observations and test runs suggested that channeling and poor distribution of vapor and liquid might be responsible for the poor reactor performance.

Click here to enlarge image

For example, Bayernoil observed that increasing the unit throughput (thus mass flux) improves the reactor performance as measured by relative catalyst activity. Fig. 3 shows this effect. One published postulate suggests that the combination of low mass flux and out-of-date tray design causes poor reactor performance.

Before the revamp, HDS-II mainly produced heating oil within 0.2% sulfur content specifications and occasionally produced automotive diesel with less than 350 ppm(wt) sulfur. With the proposed introduction of 50 ppm(wt) diesel in Germany in November 2001 and the projected decline in demand of heating oil in Europe, this unit must be capable of producing more ULSD in the future.

As in the diesel hydrofiner revamp case, Bayernoil evaluated several options. The limited flexibilities of this unit (low pressure and low hydrogen-to-oil ratio) did not permit it to be upgraded to a full-time producer of ULSD without significant capital investment.

Bayernoil therefore decided to replace the existing feed distributor tray in the HDS II reactor with a new one. It designed the new unit to produce heating oil on a part-time basis and ULSD on a part-time basis.

Additionally, Bayernoil made several low cost equipment upgrades to increase unit throughput by 20%, including:

  • Replacing the electric motor in one of the feed pumps.
  • Redesigning feed furnace burners to increase the fired duty from 2.1 Mw/burner to 2.6 Mw/burner.
  • Resizing some control valves.
  • Conducting a safety system review and replacing pressure relief valves where necessary.

In the design of the reactor internals, the following criteria are especially important to achieve the maximum reactor efficiency:2 9-11

  • Maximum dispersion of gas and liquid and complete coverage of the catalyst bed below.
  • Low sensitivity to tray out-of-levelness.
  • Low pressure drop.
  • Large flexibility to handle broad range of vapor/liquid ratios.
  • Large turn-down ratio (unit throughput).
  • Low vulnerability to fouling.

The refinery solicited several process and catalyst vendors and engineering contractors to provide the reactor internal design. Again, Bayernoil awarded the contract to Mobil Technology Co. (now ExxonMobil Research & Engineering).9

The new design installed the new tray at a higher level than the existing tray to gain 10% extra catalyst volume. Also, implementation of a new bed grading strategy reduced pressure drop buildup due to fouling. Bayernoil brought the unit on stream in May 2000.

With the installation of the new distributor tray and a change in the feed diet (that is, replacing LCO by SR petroleum), the unit can produce 50 ppm(wt) diesel. Bayernoil plans to continue to use this unit mostly to produce heating oil with less than 0.2 wt % sulfur; the unit will produce ULSD on a limited basis. Bayernoil has two other hydrotreaters that will produce 50-ppm sulfur diesel on a full-time basis.

Click here to enlarge image

Monitoring data of the reactor performance for cycles with the old tray vs. the new tray shows the effectiveness of the new reactor. Fig. 4 plots the desulfurization against the weighted average bed temperature (WABT).

Although the two catalysts are slightly different in activity (old tray with regenerated KF752 vs. new tray with regenerated KF752/KF756 mixture), the majority of improvement in hydrodesulfurization can be attributed to the state-of-the-art tray design of the new tray.

With the new tray, HDS-II mostly produces automotive diesel at the current sulfur specifications of less than 350 ppm(wt); it occasionally produces heating oil with 0.2% sulfur or less. The feed diet is more than 50% cracked stocks (FCC LCO and Combi-Cracker LGO) with a rather high end point (T90 is 360-370° C.).

Production of ULSD will require lightening up the feed and/or reducing the amount of cracked stocks.

References

  1. Lee, C.K., and McGovern, S.J., "Clean Diesel: Overview and Comparison of Clean Diesel Production Technologies," AIChE Spring Meeting, Houston, Apr. 22-26, 2001.
  2. Yeary, D.L., Wrisberg, J., and Moyse, B.M., "Revamp for Low Sulfur Diesel: A Case Study," NPRA 1997 Annual Meeting, AM-97-14.
  3. Ouwerkerk, C.E., Brandland, E.S., Hagan, A.P., Kirkert, J.P., and Zonnevylle, M.C., "Performance Optimization of Fixed Bed Processes," European Refining Technology Conference (ERTC), Nov. 16-18, 1998, Berlin.
  4. Tippett, T., Knudsen, K.G., and Cooper, B.H., "Ultra Low Sulfur Diesel: Catalyst and Process Options," NPRA 1999 Annual Meeting, AM-99-06.
  5. Desai, P.H., Gerritsen, L.A., and Inoue, Y., "Low Cost Production of Clean Fuels with STARS Catalyst Technology," NPRA 1999 Annual Meeting, AM-99-40.
  6. Peries, J-P, Jeanlouis, P-E, Schmidt, M., and Vince, P.W., "Combining NiMo and CoMo Catalysts for Diesel Hydrotreaters," NPRA 1999 Annual Meeting, AM-99-51.
  7. Criterion Catalysts Company, "Centinel Hydroprocessing Catalysts: A New Generation of Catalysts for High-Quality Fuels," October 2000.
  8. Mayo, S.W., "Maximizing Cycle Length with Akzo Nobel's Guard Bed and Liquid Distribution Technology," Akzo Nobel Catalysts Symposium 1998, Noordwijk, The Netherlands.
  9. ExxonMobil Research & Engineering Co., Public Announcement on Reactor Internals for Improving Hydrotreater Performance.
  10. Sarli, M.S., McGovern, S.J., Lewis, D.W., and Snyder, P.W., "Improved Hydrocracker Temperature Control: Mobil Quench Zone Technology," NPRA 1993 Annual Meeting, AM-93-73.
  11. Hunter, M.G., Pappal, D.A., and Pesek, C.L., "Moderate Pressure Hydrocracking: A profitable Conversion Alternative," NPRA 1994 Annual Meeting, AM-94-21.

Based on a presentation at the AIChE Spring National Meeting 2001, Houston, Apr. 22-26, 2001.

The authors

Click here to enlarge image

Daping Li is an engineer at Basell Polyolefins, Wesseling, Germany. At the time of these projects, he was a senior process engineer at Bayernoil. His responsibilities included the planning and execution of several Bayernoil clean fuel projects related to the European Auto Oil Program 2000 and 2005. Previously, he had been responsible for the operations of hydrotreating, catalytic reforming, and isomerization units. Li holds a BS in chemical engineering from Dalian University of Technology, Dalian, China, and a PhD in chemical engineering from Karlsruhe University, Karlsruhe, Germany.

Click here to enlarge image

Dirk Znidarcic is a process engineer at Bayernoil. He is responsible for the operations of FCC, CCR-reformer, isomerization, and hydrotreating units. Previously, he worked as shift supervisor in the refinery. Znidarcic holds a diploma in chemical engineering from Georg-Simon-Ohm University in Nuremberg, Germany.

Click here to enlarge image

Christof Thiel is a process engineer at Bayernoil. Since joining Bayernoil in 1998, he has been responsible for the operations of two crude units, three hydrotreating units, LPG recovery, the amine system, and the Claus plants. Thiel holds a diploma in chemical engineering from Karlsruhe University, Karlsruhe, Germany.

Click here to enlarge image

C.K. Lee is a principal of PetroTech Consultants. His areas of expertise are hydroprocessing, catalytic reforming, and clean fuel technologies. He has provided consulting services to the refining and catalyst industries since April 2000. Previously, he worked for Mobil Technology Co. (now ExxonMobil Research & Engineering) for more than 20 years. Lee holds a BS in chemical engineering from Cheng Kung University, Taiwan, and a PhD in chemical engineering from the University of Houston.