SHELL PLANS $2.2-BILLION RENOVATION OF DUTCH REFINERY

April 26, 1993
Peter Ladeur Shell Internationale Petroleum Maatschappij B. V. The Hague Harry Bijwaard Shell Internationale Petroleum Maatschappij B.V. Pernis Royal Dutch/Shell Group recently approved a $2.2 billion rejuvenation of its Pernis refinery, near Rotterdam. This upgrade will enable the refinery to meet product volume and quality demands well into the next century, while reducing environmental emissions (OGJ, Jan. 25, p. 40).
Peter Ladeur
Shell Internationale Petroleum Maatschappij B. V.
The Hague
Harry Bijwaard
Shell Internationale Petroleum Maatschappij B.V.
Pernis

Royal Dutch/Shell Group recently approved a $2.2 billion rejuvenation of its Pernis refinery, near Rotterdam. This upgrade will enable the refinery to meet product volume and quality demands well into the next century, while reducing environmental emissions (OGJ, Jan. 25, p. 40).

Shell Pernis is a complex, 400,000 b/d refinery with two vacuum gas oil (VGO) catalytic crackers, two visbreakers, and a residue conversion unit.1 A simplified block diagram of the refinery and its processing units is shown in Fig. 1.

Cornerstones of the $1.7-billion main revamp project are a single-train, 8,000 metric tons/sd (mt/sd), or about 56,000 b/sd, hydrocracking unit and a three-train 1,650 mt/sd residue-gasification unit for production of hydrogen and sulfur-free fuel gas. Fuel gas will be used in a new 115-mw electricity cogeneration plant.

In addition, new amine treating, sulfur recovery, and tail gas units will be installed. These will come onstream in 1997.

Another $0.5 billion will be spent over the next 10 years, mainly on environmental improvements of existing facilities. The main objective of the project is to reduce emissions levels and enable the refinery to meet the demand for lighter and environmentally cleaner transportation fuels.

Shell Pernis satisfies local demand and acts as supply center for other Shell companies in and outside Europe. Flexibility in both feedstocks and products is of utmost economic importance.

The refinery was designed to process mainly high-sulfur crudes. Some 50 different crudes are processed every year. In years to come, availability of low-sulfur crudes will decline. At present, low-sulfur crudes represent about 25% of the refinery's crude diet.

On the product side, total output is anticipated to remain constant over the next decade. Forecasts indicate a shift to middle distillates from both the motor gasoline and residual fuel sides.

PROCESS SELECTION

Crude supply, product quality, and demand trends played an important role in the selection of the new processing facilities. The new hydrocracking unit (HCU) will replace a fluid catalytic cracking (FCC) unit that will be 45 years old by 1997.

The HCU was selected in preference to a major upgrade of the existing unit or a new catalytic cracker because of its better middle distillate selectivity and superior product quality (very low sulfur, high cetane number). By this choice, investments in cat-feed hydrotreaters or FCC-product desulfurizers are obviated.

Hydrogen required for the HCU, approximately 255 mt/sd or 3.2 wt % on feed, will be produced from syngas originating from gasification of heavy, vacuum-flashed, visbroken residue.

Steam methane reforming (SMR) was considered as an alternative to residue gasification. Unlike gasification, SMR would not have reduced fuel oil make. This in turn would have led to reduced crude supply flexibility, both in sulfur and API gravity, and thus to a more expensive crude diet. In addition, expensive natural gas would have been required as SMR feed.

HYDROCRACKING UNIT

The new HCU will be a high-conversion, once-through, single-reactor, series-flow hydrocracker. The unit will be fed with heavy straight-run VGO (370-550 C.), flashed distillates originating from the residue hydroconversion (Hycon, Shell's proprietary process) and visbreaking units, and lube oil extracts. The nitrogen content of the combined feed will typically vary between 1,000 and 1,700 ppm.

About 80% of the feed will be converted to high-quality 370 C.-minus products (mainly tops, naphtha, kerosine, and gas oil). 2 The other 20% is low-sulfur, low-cokeforming tendency feed for the remaining FCC unit. A simplified flow diagram for the HCU is shown Fig. 2.

Hot feed from upstream units or cold feed from storage is heat-exchanged with reactor effluent, combined with hot gas from the only furnace of the unit, and sent to the high-pressure (almost 200 bar), high-temperature (up to 450 C.) reactor.

The very large reactor vessel has a height of some 34 m and weighs around 1,200 metric tons. The reactor contains some 400 cu m of two different catalysts in multi plc, stacked-bed configuration. In the top, a denitrogenation/hydrogenation catalyst is used, while the bottom beds contain a proprietary cracking catalyst.

Combined reactor effluent is heat-exchanged with reactor and vacuum-column feed before entering the separator system.

The unit was designed for minimal wash water consumption. The bulk of the wash water will be reused via internal recycle loops; only a small bleed stream from the dedicated sour water stripper (SWS) will be sent to the water treatment plant. Hydrogen sulfide scrubbing of the recycle gas is not applied.

In the atmospheric distillation column, a separation is made between naphtha minus and heavier products. Steam stripping is applied to minimize light component carry-over to the mild vacuum column (MVC). In the MVC, heavy naphtha, kerosine, and gas oil are produced as top product and side cuts, while the unconverted hydrowax ( 370 C. cat-feed) is recovered as the bottom product.

Advantages of the two-column recovery section compared with the more common single-column concept are higher recovery of gas oil products under mild vacuum conditions and significantly lower energy consumption.

The high degree of heat integration present means that design duty of the gas furnace is determined more by start-up conditions than by normal operation. Simulation studies of total HCU system dynamics revealed that in normal operation, heat input from the furnace is so small that control of reactor inlet temperature by varying furnace outlet temperature would be too slow. Therefore, reactor inlet temperature will be directly controlled by bypassing feed/effluent exchangers.

The unit will be equipped with advanced process control for reactor severity and temperature profile control for heat integration aspects (heat recovery vs. heat requirement of Distillation section) and for product quality control in the distillation section.

GASIFICATION UNIT

The three-train gasifier will have a capacity of 1,650 mt/sd (about 11,000 b/sd) and will be fed with heavy, vacuum-flashed, visbroken residue. Approximately 1,600 mt/sd of oxygen will come from future third-party sources nearby. The gasification reactors will be larger than standard sizes constructed so far, but will remain well within the limits of upscaling.

The main reason three gasification trains were selected is that, in case of outage of one gasifier, hydrogen requirement for the HCU can be delivered from two gasifier trains.

Under normal three-train operation, the syngas in excess of the requirement for hydrogen production will be used as gas turbine fuel. Gasifiers in the Shell gasification process typically operate between 50 and 75 bar pressure and 1,200-1,500 C.3

The Pernis gasifiers will be equipped with the recently developed proprietary coannular burner, through which the hydrocarbon fuel and oxygen are introduced separately into the reaction chamber. Steam shielding is applied to ensure long burner life. Advantages over previous designs are longer burner life, improved operating range, and start-up via the main burner.

Each gasifier will have its own waste-heat recovery train and soot/ash quench scrubber. Solids-free syngas from the three reactor chains will be treated in a single-train, H2S-removal unit. About two thirds of the treated syngas will be sent to the hydrogen plant integrated with the gasification unit.

The combined process is called the Shell gasification hydrogen process, or SGHP (Fig. 3). The hydrogen plant consists of a two-stage CO shift (high temperature/low temperature), carbon dioxide removal, and methanation. Lurgi's Rectisol process was selected for H2S removal from the syngas, as well as for CO2 removal downstream of the low-temperature CO shift. Both Rectisol treating steps are highly integrated.

The soot and ash-containing scrubber water from the three trains is filtered and returned to the scrubbers; excess water is exported to water treatment facilities. The filter cake will be worked up.

UTILITIES

Many key equipment choices were governed by the refinery's existing utility infrastructure. Two new gas turbines form the heart of a new 115-mw cogeneration plant.

Saturated steam from the waste-heat exchangers of the SGHP will be superheated in the fired waste-heat boilers of the gas turbines. Part of the high-pressure steam will be used as process steam in the gasifiers and the hydrogen plant; the remainder will be sent to steam turbines for generation of lower-pressure steam and electricity. A large share of the electricity generated in the new utilities complex will be exported to the public grid.

In addition to the Rectisol process in the SGHP, new amine and sulfur-recovery facilities will be installed. Existing Claus units will be equipped with feed preheaters for the treatment of H2S containing a relatively high proportion of inerts (mainly CO2 and NH3).

CONSTRUCTION

A major challenge of this project, besides its scale, will be to construct the units in a fully operational refinery. About 1,000 tie-ins with existing process, utilities, and movements facilities will have to be made. Construction is expected to start in the second quarter of 1994. All new facilities will be operated from a single control room and maximum use will be made of advanced process control.

The project will be coordinated through an engineering, procurement, and construction contractor supported by specialized engineering contractors.

ENVIRONMENTAL ASPECTS

An overview of the environmental impact of this project is given in Table 1, where current refinery operation is compared with post-investment operation.

Crude oil diet is projected to shift toward a higher proportion of Middle East crude, with a consequent increase in sulfur intake from 280,000 mt/y to 410,000 mt/y. In spite of the heavier diet, investments in HCU and SGHP capacity led to a 10% higher white-product make and a 40% reduction in fuel oil production.

Sulfur leaving the refinery in the form of hydrocarbon products, and eventually becoming SO2 emission by end users, is reduced from about 155,000 mt/y to 80,000 mt/y, while sulfur recovery increases almost three-fold, from slightly more than 100,000 mt/y to about 300,000 mt/y.

Refinery emissions of SO2 and NOx are reduced 35-45%, mainly from reduced residual fuel firing. Dust emissions from the refinery will be more than halved as a result of retirement of one of the catalytic crackers and reduced liquid fuel firing.

The only substantial increase in refinery emissions is that of CO2, because of the new hydrogen plant. However, because of the higher hydrogen content of the products, global CO2 emissions are on a par.

This major refinery rejuvenation project will enable Shell Pernis to meet current and foreseeable future product volume and quality demands. In spite of an increased proportion of high-sulfur crudes in the refinery diet, the environmental impact of the refinery and its products is reduced. The project will therefore carry Shell Pernis well into the next century.

REFERENCES

  1. Deelen, W.J., and van Tilburg, J.P.J., "The residue hydroconversion process: commercial experience and further developments for integrated refinery configurations," Topic 18, Paper 6, 13th World Petroleum Congress, Buenos Aires, 1991.

  2. Hoek, A., et a]., "New catalyst improves heavy feedstock hydrocracking," OGJ, Apr. 22, 1991, pp. 77-82.

  3. Higman, C.A.A., and Gruenfelder, G., "Clean power generation from heavy residues," Gasification, I. Mech. E., 1990, C410/071, pp. 101-11.

Copyright 1993 Oil & Gas Journal. All Rights Reserved.