Extensive expansion at Kårstø gas plant under way

Statoil's Kårstø gas plant, on the west coast of Norway, is undergoing major expansion to accommodate increased volumes from development of the offshore Aasgard fields.

Prompting this expansion is the landing at Kårstø in 2000 of a 42-in., rich-gas pipeline from Haltenbanken, offshore mid-Norway, and installation of the 42-in. Europipe II dry-gas pipeline from Kårstø to Germany (Fig. 1 [137,236 bytes]).

In the same period, several "spin-off" projects adding value to the overall concept may be constructed. These could include a 350-mw power plant and ethane-shipment facilities.

Total investment at Kårstø in the next 3-4 years will reach approximately $1.1 billion (U.S.). Civil work began in June 1997; the detail engineering contract was awarded in August 1997.

Beginnings

Aasgard is a unitization of three fields: Midgard, Smorbukk, and Smorbukk Sor, all discovered between 1981 and 1985. Estimated combined field reserves are 780 million bbl of liquids and 7.5 tcf of natural gas (Fig. 2 [99,980 bytes]).

Oil production from the fields in Haltenbanken will commence near yearend 1998 from Aasgard A, a 200,000-b/d floating production storage and offloading unit, while Aasgard B, a semisubmersible platform, will be installed in 2000 for gas and condensate processing prior to rich-gas transport through the 450-mile, 42-in. Aasgard transport system to Kårstø.

Gas production from Aasgard will peak at 1.4 bscfd of rich gas and around 100,000 b/d of condensate.

The plan for development and operation (PDO) for the Aasgard field was approved by the Stortinget in June 1996. This also formed the basis for commencing the 400-mile, 42-in. Europipe II sales-gas pipeline at Kårstø .

This pipeline, starting up in 1999, will connect to the continental Europe gas grid from its landing point at Dornum, Germany; the PDO was approved in December 1996.

The Kårstø gas plant, originating as part of the Statpipe system, started up in 1985 processing rich associated gas from Statfjord and later Gullfaks and other fields in the North Sea and exporting gas to Europe via Ekofisk field center and Norpipe transportation system to Emden, Germany.

Statpipe is owned by den norske stats oljeselskap AS (Statoil, operator), Elf , Norsk Hydro, Mobil, Esso, Shell, Conoco, Saga Petroleum, and Total. The Aasgard field and Aasgard transport system are owned by Statoil (operator), Agip, Total, Mobil, Neste Petroleum, Saga Petroleum, and Norsk Hydro.

Initial processing capacity of the Kårstø plant was 600 MMscfd of rich gas from which liquid propane, iso-butane, normal-butane, and naphtha were extracted. Gas-NGL separation takes place in two turboexpander-driven highly heat integrated NGL extraction units after being dehydrated in molecular sieves. The NGL-mix extracted is fractionated into the components in two fractionation trains.

Performance tests and debottlenecking studies performed soon after start-up identified a potential for increased throughput. At the same time, new oil-producing fields were continuously coming onstream in the North Sea prompting a need for new gas-processing capacity through the Statpipe system.

During planned maintenance shutdowns in the summers of 1990 and 1991, the plant throughput capacity was increased by 30% to 775 MMscfd at an investment of less than $7 million.

Increases in NGL fractionation and shipment capacity to around 2 million mty were also obtained. The main bottleneck removed was the flare system through reduction of the flare load by installation of a high integrity pressure protection system (hipps).

This overpressure-protection concept developed by Statoil was an early installation, and Statoil believes it to be the first worldwide installed in a hydrocarbon-processing plant with such complexity. Further work consisted of retraying the demethanizers and some other minor modifications mainly based on making use of design margins identified.

Demand for gas-throughput capacity was increasing steadily and by 1993 the full rich-gas pipeline capacity of 885 MMscfd was called for. As additional volumes were rather small and no low-cost revamp option of the NGL-extraction units was foreseen, a different approach was used.

Through laboratory tests of rich-gas and sales-gas mixtures of actual gas passing through Kårstø, a maximum rich-gas injection rate of around 25 wt % into the sales gas was determined. A rich-gas bypass was started up in 1993 giving yet another boost in gas-throughput capacity (14%) at a cost of less than $3 million.

Increased downstream pipeline pressure prompted addition at Kårstø in 1993 of a second compressor stage, consisting of three electric variable-speed-drive compressor units.

An unstabilized-condensate pipeline from the Sleipner field was landed at Kårstø in 1993, and the Kårstø plant was expanded with two condensate stabilizers (130,000 b/d each) and a third fractionation train including an increase in steam-generating capacity (40%) and cooling-water pump capacity (33%).

This project was conducted in parallel with the compressor station and rich-gas bypass installation. In autumn 1997, an expansion of the unstabilized-condensate capacity to 165,000 b/d was completed, and currently flow rates of around 180,000 b/d can be achieved under certain flow conditions (Fig. 3 [57,695 bytes]).

Today, the Kårstø gas plant is Europe's largest NGL and condensate-processing facility, with a capacity of approximately 4 million mty of stabilized condensate and a further capacity of close to 4 million mty of propane, iso- and normal butane, and naphtha. Dry gas export is approximately 750-800 MMscfd, of which up to 100 MMscfd is rich gas bypassing the NGL-extraction process (Fig. 4 [155,512 bytes]).

Compared to the original plant design, an increase in gas processing of 50% and a five-fold increase in liquid product processing and handling capacity have been achieved through very modest increases in utility and support systems (Fig. 5 [62,106 bytes]). Still, the plant is achieving an on-stream factor close to 100% with regard to nominated gas sales.

Landing Aasgard gas

The large gas reserves to be developed in an area remote from other significant gas infrastructure required cost-efficient transportation solutions to be able to compete with other Norwegian gas sources in the more mature North Sea.

At the same time, the Aasgard development is regarded as the key to opening up the Haltenbanken for further development with a gas export solution in place.

Five main gas-export scenarios were identified for Aasgard:

• Dry-gas export from field to Emden, Germany (reference case).
• Dry-gas export from field to Draupner pipeline riser platform in the North Sea.
• Rich-gas export from field to the giant Troll field onshore gas-conditioning plant at Kollsnes, north of Bergen. Dew point control process.
• Rich-gas export from field to the Troll field onshore gas-conditioning plant at Kollsnes, north of Bergen. NGL recovery process.
• Rich-gas export from field to Statpipe's onshore Kårstø gas plant, north of Stavanger.

The dew-point processing scheme by expanding current process at Kollsnes with another process train yielded the lower investment cost. Norwegian gas reserves are abundant, however, compared with current sales contracts and volumes expected to be absorbed by the European market.

Therefore, the high NGL recovery yielded superior overall economics. This added value results not only from products extracted along the gas-value chain penetrating another and higher priced market, but also because it is possible to produce at a higher overall field hydrocarbon rate as more but slightly leaner sales gas is landed in Europe.

Design studies of the NGL recovery schemes for Kårstø and Kollsnes showed that the existing and more developed infrastructure for utilities and offsite systems at Kårstø, already being a world scale NGL producer, yielded lower investment and operating costs.

Total pipeline length required (gas import/export) differs some 10 km out of a total pipeline route of almost 1,400 km to Europe. Thus, the increased net present value of approximately $150 million by integrating into the existing Kårstø plant rather than constructing a grassroots NGL plant at Kollsnes turned the edge with respect to being the most attractive landing point for the Aasgard transport rich-gas pipeline.

Further strengthening the Kårstø alternative is the flexibility to extract as much as 600,000 mty ethane from gas routed through the existing facilities with Aasgard being the field with reserves large enough to guarantee realization of the liquefaction, storage, and shipment part of such liquid ethane production.

Also the possible integration of a 350-mw natural-gas-fired power plant may add value to the total project portfolio.

The Aasgard transport pipeline is designed as a 42-in. pipeline, being the maximum qualified pipeline size for pressure levels and water depths being discussed for the pipeline. The diameter chosen gives a rich-gas capacity of around 2.3 bscfd, well more than the 1.4 bscfd requirement of Aasgard.

Given the low difference in installed cost per mile for the pipeline diameters in question, however, and the expected impact on area field development acceleration, a larger trunkline option was chosen. It is expected that two or three existing and new oil-producing fields in the Haltenbanken area will also deliver gas into the system from the start-up date in October 2000.

As a consequence of landing Aasgard gas at Kårstø with annual export volumes of 320-390 billion cu ft in addition to the 210-250 bscf/year currently exported from Kårstø , a new export line from Kårstø was required.

Thus, the fifth Norwegian trunkline for gas export to continental Europe, Europipe II with planned landing in Dornum, Germany, will commence at Kårstø. It is a shorter, lower-investment route than going via the originally planned starting point at Draupner.

The problem of the Europipe II pipeline starting up in 1999, a year ahead of the Aasgard gas deliveries, is overcome by converting the existing Statpipe export leg to a bi-directional gas pipeline providing required volumes for Europipe II.

Kårstø expansion

The ownership and investor structure to expand the Kårstø gas plant is rather complicated due not only to differences in the current Statpipe joint venture owners group and the companies represented in the Aasgard joint venture, but also to the fact that other ownership groups were foreseen for several of the projects which might come out of the development.

A common optimization project for all companies and ownership groups was formed. In this project, called "Kårstø 2000," joint work groups were initiated to establish such important parameters as final plant capacity, infrastructure expansions, and development of projects for further refining the products available or possible to be made available at Kårstø, including the simultaneous establishment of sound commercial principles for the realization of the projects.

Processing, NGL recovery

The initial NGL-recovery scheme forming the basis for the PDO was designed in-house as a modified version of the existing Kårstø process but with the deletion of the demethanizer column, thereby losing the flexibility to produce liquid ethane.

The scheme was laid out as a single train with refrigeration obtained through pressure let-down in a Joule-Thomson (J-T) valve and a turboexpander in series. Propane recovery was expected to be less than 80% compared to the 90%+ process which is currently operated at Kårstø.

Export compression requirements ranging from 600 psig to Europipe II's maximum operating pressure of 2,750 psig were proposed either in a 3 x 33% dedicated compressor station or a configuration fully integrated with the existing Kårstø compressor station with a suction pressure of around 600 psig.

The initial part of the conceptual studies concentrated on optimizing the extraction process for NGL recovery, energy efficiency, regularity, and life-cycle cost. Of the various NGL-recovery schemes studied, nine schemes were presented by the three contractors, ranging from simple J-T schemes via heavy-end wash-column concepts to various turboexpander-driven refrigeration one-column and two-column separation schemes.

The schemes' comparison (Table 1 [60,315 bytes]) showed that high NGL recoveries and high compressor suction pressure gave the highest value addition to the project even though this meant that two-column concepts were required to achieve bulk-gas removal in a high-pressure wash column to minimize energy consumption, while high propane recovery is obtained in a low-pressure de-ethanizer column.

Thus, two schemes yielding 95%+ propane recovery and compressor suction pressures at least 290 psi greater than currently employed were selected for further conceptual development.

The increased revenue from high NGL recovery and higher compressor suction pressure more than offset the increased investment of a two-column process and the possibility of a reduced number of compressor units in a station.

The principal difference between the two schemes was the use of two turboexpanders in series vs. a single unit with a broader operating window.

The final concept for pre-engineering and detail engineering combines features from the two schemes, but basically the dual turboexpander scheme was chosen with its higher NGL-recovery but also the uncertainty regarding qualifying turboexpander size and operating from the dense phase through the gas-mixture critical point.

A study of total gas-delivery regularity from Aasgard to the delivery point in Emden prompted the decision to construct two 50%-extraction trains. This decision was also supported by the fact that with the flow rates in question, several equipment units would have to be doubled because of limited construction ability.

Also, the compressor-station design was increased to 3 x 50% from 3 x 33% to increase regularity and because frame size would not be altered for gas-turbine drivers.

Further detailed regularity studies have concluded that initially only a 2 x 50+% configuration is required as commercial arrangements (including the use of downstream gas storage and of any existing spare capacity day-to-day) will resolve any regularity issues in the early years.

This is due to the Europipe II pipeline operating at well below design-export pipeline pressure until further volume increase beyond Kårstø process capacity.

NGL fractionation

Also, some of the premium naphtha from Aasgard condensate would be lost as condensate when NGL would be transferred into the condensate stabilization and fractionation units.

The revamp studies revealed that although feasible, the revamp work would lead to extensive replacement of equipment, valves, and piping in addition to retraying all fractionation columns and even possibly adding one new debutanizer. A 3-week shutdown period followed by a period with only 50% throughput of at least 3 months was foreseen to allow the field work to be performed.

Considering the length of time and the construction risk in virtually reconstructing the existing fractionation units along with the development constraints on both Assgard and existing fields being processed at Kårstø, it was decided to build a new, fourth fractionation train at Kårstø.

This also allows Aasgard to bring forward production of an NGL-rich gas from the Smorbukk area at an earlier stage, increasing project revenue. A single 65,000 b/d fractionation train will be constructed, but with the flexibility to transfer and receive NGL for fractionation to and from the exiting units.

Spur-gas processing

The possibility further to process this gas was reviewed as a means for adding more value to the gas transfer across the Kårstø gas plant.

Final definition of optimal plant size was studied extensively because it could be argued that by utilizing the forecast profiles through Statpipe and the planned build-up of plateau levels for gas export from Aasgard, a plant expansion of 700-1,050 MMscfd would be an adequate solution.

Also, larger volumes from Aasgard's peak plateau of 1.4 bscfd or expansion to 1.75 bscfd up to the Europipe II capacity 0.75 tcf/yr or to the total Kårstø annual export-pipeline capacity of around 1 tcf were evaluated.

It was shown that using a gas mixture dominated by high gross calorific value (GCV) gas from the Sleipner area and extending the reverse flow to tie-in to the front-end of the NGL extraction units would show profits for expansions up to the Europipe II export capacity based on a day-to-day processing capacity.

Higher capacities would require firm commitments from additional fields exporting rich gas.

Expectations for higher export rates through Europipe II prompted the decision to expand the plant equal to the highest Aasgard plateau as the basis for design of new NGL extraction and fractionation facilities.

Thus, total processing capacity at Kårstø will correspond to a dry-gas export level of around 0.64 tcf/year which currently is expected to be fully utilized from 2001 without the processing of spur gases in reverse flow.

Ethane liquefaction

The Norwegian petrochemical industry, in Rafnes and Bamble operated by Norsk Hydro and Borealis, has been operating on a feedstock mix also including ethane. The ethane has come from the declining Ekofisk field's Teesside (U.K.) processing plant.

As well as seeking replacement feedstock for the Bamble plants, Borealis was also evaluating alternative feedstocks for construction of new cracker capacity at its Stenungsund plant in Sweden.

Suddenly a situation occurred in which replacement ethane feedstock and new capacity were being planned in northwest Europe, thus possibly forming the basis for an ethane-liquefaction unit with economy of scale.

At the same time, Aasgard was being planned for landing at Kårstø, a plant whose facilities to extract ethane already are in place, thereby reducing the required investment cost.

Although Aasgard will not necessarily supply ethane physically, it is a rich gas source with reserves large enough to enable the long-term NGL feedstock guarantee required from both buyer and seller of a product with no established market.

Agreements have been concluded to construct a 600,000+ mty ethane-liquefaction unit based on the existing extraction units' ethane capacity.

Flexibility to feed Aasgard gas into the existing units to deliver ethane for shipment will be built, but the new NGL-extraction units constructed for processing Aasgard have no ethane recovery.

The ethane-liquefaction unit will be designed with a dedicated propane-cooling circuit, utilizing auto-refrigeration and the cold from tank boil-off to minimize energy consumption and investments.

It is also generally recognized in the design that rerouting the ethane back into the sales gas might be used as a back-up solution as long as gas-export volumes are adequate to absorb quality variations before investing in costly spare equipment.

Natural-gas-fired power

At Kårstø, studies were carried out in cooperation between the involved parties to establish mutual integration synergies. Nine different heat-integration schemes were studied.

Basic differences were the level of heat integration (high or low-pressure steam) and choice and configuration of Aasgard pipeline compressor drivers, in which electrical, steam turbine, and gas turbine drives all were viable alternatives.

The studies concluded that heat transfer at the low-pressure-steam level would be optimum, yielding a highly energy-efficient plant with the flexibility to fulfill the business objectives of both a power producer and a gas-processing plant providing a service in a transport chain.

Thus, the pipeline compressors will be gas-turbine driven with waste-heat-recovery steam generators with auxiliary firing capacity to cover the standalone operation of the gas-processing capacity expansion.

The combined-cycle power plant, however, will be installed with back pressure and condensing steam turbines transferring low-pressure steam into the gas plant-steam grid. In this way, minimum auxiliary firing and back-up boiler capacity are required at the same time as maximum electric power quantities are made available to the Nordic energy market.

Also, the on-stream dependency of the two plants is decoupled as well as operating windows broadened when performing a more moderate energy integration compared with integration at the high-pressure-steam level.

Other main synergies in the construction phase are common cooling-water tunnel construction and other civil works. Generally being within the Kårstø gas plant perimeters will enable the various plants to back each other up with individually developed infrastructure and services.

Storage, jetties

With another 50% increase in LPG production at Kårstø from 2000, however, it is evident that increased storage is required. Studies to establish cost for aboveground storage were aimed at securing the same relative storage as today (that is, the same number of days to fill tanks as at present).

At the same time, the possibility of excavating a large-scale rock cavern for storage of propane at atmospheric pressure was evaluated and, with its low investment cost/cubic feet, found to be competitive.

With a large increase in LPG exported from Kårstø, the risk of disrupting the schedule of gas deliveries with limited storage will be increasing.

At the same time, increased competition from Algerian and Nigerian volumes in the northwest European market mainly served by Norwegian and U.K. volumes will affect logistics at Kårstø.

Substantially increased storage will not only increase gas-delivery regularity, but will add value by the ability to penetrate other markets overseas as well as generally serving larger ships and thereby yielding more flexibility for customers. A rock cavern for propane storage was decided upon.

A propane cavern size of 885 million cu ft laid out in two cavities was settled on, based on elements such as expected future production rates, shipment sizes, and operational procedures. The new storage size more than doubles current LPG storage.

The size of this cavern also enables conversion of existing propane-storage tanks to normal and iso-butane service giving substantially larger storage volumes for these products at a total investment well below the cost of adding marginal storage for the three LPG products. The cavern excavation work is now nearly complete.

Studies were also conducted to evaluate if one of the existing propane tanks could be converted from propane storage (-40° F.) to liquid-ethane storage (-130° F.). From a material point of view, this was confirmed, but it was concluded that the foundation works were inadequate for such service.

The liquid ethane is now planned to be stored in a single aboveground storage tank at atmospheric pressure, reflecting the needs of the basic long-term contracts for ethane. Any tank overfill can be "spilled" back into the gas-export pipeline without violating specifications.

Shipments

Although increased production of all products at Kårstø is foreseen towards the turn of the century, the need for a fully flexible export jetty was examined. The cost of such a jetty was compared with converting the existing module ro-ro jetty to a combined module and loading jetty dedicated for ethane shipment and possible small size iso-butane and naphtha carriers.

This action, combined with increasing the allowable ship size at the two existing jetties to allow for changes in shipping pattern and serving increased storage, was estimated at approximately half the cost of a new fully flexible jetty, still satisfying the requirements of operational jetty occupancy.

The first step in this jetty system expansion was completed early summer 1997 when dredging and excavation to allow increased berth depth was completed along with two new mooring dolphins for one of the two existing jetties.

Condensate carriers up to 80,000 dwt and LPG carriers up to 40,000 dwt can now be handled compared to around 40,000 tons condensate and 30,000-ton LPG carriers before expansion.

The Author

Sigbjorn Svenes is senior process advisor within the technology and services division of Statoil's exploration and production technology unit. From late 1993 into 1997, he was process-engineering manager for the gas-transportation business unit of Statoil and later with the gas development and marketing business unit. Svenes served as assistant start-up manager for the commissioning and start-up phase of the condensate processing facilities at Karst in 1993. He joined Statoil at its Karst site in 1988, holding various process engineering positions in the period until 1993. In 1986-1988, he worked for Kvaerner Engineering, Stavanger, as a process engineer. Svenes graduated (1985) as a chemical engineer from the Norwegian Institute of Technology, University of Trondheim.

Copyright 1998 Oil & Gas Journal. All Rights Reserved.