ANALYSIS SETS OPTIMUM C3 RECOVERY FOR NORTH SEA GAS TERMINAL
David E. Connell
Chevron U.K. Ltd.
London
Paul Horth, R. H. W. Powell
Conoco U.K. Ltd.
London
Analysis for propane recovery in a planned European gas terminal indicated recovery to be most economic at 72%: recovery above that level would yield a lower rate of return and only a small increase in net present value (NPV).
These results led to the decision that the gas terminal for the Britannia project in the U.K. North Sea should proceed with a flow designed for 72% propane recovery.
Britannia is a $3 billion gas development project due on stream in 1997. The field is believed to hold more than 3 tcf of gas in place and 100200 million bbl of recoverable condensate.
Most of the condensate will be separated offshore and exported by pipeline or an offshore tanker-loading system. Gas production is planned to be approximately 500-650 MMscfd sales gas.
Potential customers for the gas are in several North European countries as well as the U.K., and the gas could be landed at several North European locations (Fig. 1).
Gas will be exported from the development in dense phase to prevent liquids forming in the pipeline. This eliminates the need for liquids-handling facilities and avoids the operational problems associated with two-phase pipelines.
Transport of the gas in dense phase requires high pressure. The export gas pressure at the offshore development is 179 barg (2,600 psig) and the cricondenbar is 110 bara (1,595 psig).
Although liquids are not present in the pipeline, the gas contains about 27 bbl/MMscf of C3+ NGLS.
To meet likely sales-gas specifications, the offshore gas must be processed in a gas terminal to recover the majority of these NGLS.
COMMERCIAL CONSTRAINTS
Unlike the U.S., no readily available open market for ethane exists in Europe. Most northern European ethylene is produced from naphtha.
Liquid recovery from landed gas typically is limited to propane, butanes, and heavier condensate, in contrast to the U.S. where a large proportion of ethane in the feed gas would typically be recovered.
The economic attraction of extracting ethane and propane from natural gas is governed by the availability of infrastructure to transport large volumes of these products to market.
There are two types of gas in the North Sea.
In the U.K., most rich natural gas from the northern North Sea which is suitable for processing is landed at St. Fergus. Southern basin North Sea gas is leaner and unsuited for gas processing,
Typical southern basin fields use a simple refrigeration cycle to condition the gas to sales specifications; only small amounts of NGLs are recovered.
The few fields currently being developed in the central North Sea are richer and suitable for processing. This is demonstrated by Enron Corp.'s subsidiary Teesside Gas Processing Ltd. which has built an NGL extraction plant at Teesside for gas it has purchased (OGJ, June 7, p. 37).
This project suggests that propane extraction can be economic if the siting conditions are carefully chosen.
In the U.K., only the Shell/Esso Flags terminal extracts ethane from natural gas (OGJ, Mar. 8, p. 37). The NGL extracted at St. Fergus is transported by pipeline about 130 miles to Mossmorran where it is fractionated and the ethane is used as feedstock to the adjacent Exxon Chemical Ltd. Fife ethylene plant.
Other producers at St. Fergus have chosen to limit NGL extraction in part because of high transportation costs. There are only two NGL transport systems out of St. Fergus. Pipeline and fractionation charges can significantly erode profit margins for third-party users of these systems.
Current charges are 30-40/metric ton (mt). High NGL transport costs at St. Fergus have left producers to find innovative ways to avoid tariff on those NGLs that must be removed from the gas to meet sales-quality specifications.
For example, the Mobil North Sea Ltd. SAGE terminal at St. Fergus negotiated a novel gas specification which allowed 4% CO2. The higher CO2 concentration allowed most of the propane to be left in the gas where it could be sold for thermal value.
The clear challenge in determining prospective gas-terminal locations in Europe is to choose a location which minimizes new NGL infrastructure requirements while maximizing gas value.
PROCESS MARGINS;
FUEL VALUES
Typical sales-gas contracts with major European gas utilities are based on field sales with an agreed daily contract quantity (DCQ) and defined swing factor. The gas price either is fixed or carries an agreed-upon escalation mechanism.
Because gas prices are higher than those prevailing in the U.S., at around $3/MMBTU,14 process margins on NGLs are lower in Europe than in the U.S. NGL liquids are typically sold on the spot market where prices can fluctuate usually in conjunction with world crude prices.
At times of low liquid prices, propane can have more value in the sales gas than as a liquid product.
With a DCQ contract, the annual quantity of gas sold is fixed. Reductions in the amount of gas consumed as fuel, both on land and off-shore, do not therefore lead to increased revenue.
In this situation, there is little economic justification for increased capital expenditure (capex) to bring about improved efficiency because fuel gas that is saved effectively remains in the reservoir until near the end of field life.
Thus the fuel value is the effective value of today's price discounted to near the end of the field life. This arrangement contrasts with NGL processing in the U.S. where fuel must be paid for at the same price as the sales gas.
Another consequence of a supply contract is that increased liquids recovery can improve the economics of the project.
NGL liquids represent hydrocarbon reserves which are sold in addition to the fixed gas production, thus accelerating revenue to the project. This effect is gained as well as the extra value that the hydrocarbons might have as liquids rather than gas.
Equipment and construction costs are typically higher in Europe than in the U.S. Standards and plant designs are more robust, leading to an increase in plant costs.
For illustration, estimated costs escalated to 1992 for two U.K. gas terminals are given in the accompanying box.
BRITANNIA TERMINAL
Because of the high pressures in dense-phase pipelines and 1,000 psig (69 barg) transmission systems, gas plants can be configured to use free energy.
Gas delivered to the terminal will have a minimum pressure of 113 bara. A letdown in pressure of about 40 bar (580 psi) is necessary in the terminal.
This pressure letdown can be used to create low temperatures for liquid recovery, either by Joule-Thomson expansion or in turboexpanders. Some liquid recovery is therefore possible without sales-gas compression or refrigeration, making use of the energy provided by the offshore compression.
Because of the limited market for ethane and the need to limit investment in NGL infrastructure, however, ethane recovery is unattractive. Ethane stripped from the NGL will be returned to the sales gas.
The process design must be relatively simple (compared to the typical U.S. high-efficiency plant) for two reasons:
- Because equipment and installation costs are higher in Europe than in the U.S., revenue generated by sales of liquid can justify fewer items of equipment.
- Increased complexity leads to lower reliability. Failure to meet the nominated sales-gas quantity because of plant upsets can lead to serious financial penalties.
Because fuel is valued at a discounted future price, more costly design features intended to reduce fuel consumption per barrel of liquid produced are generally unjustified.
Poor efficiency, however, which leads to significantly higher capital costs (such as increased compression power) is to be avoided.
DESIGN REQUIREMENTS
The gas-terminal design for the Britannia project was governed by considerations in the following areas, but had also to meet certain other key requirements.
- Economic performance. Capital investment in the terminal facilities will have to meet the same acceptance criteria as laid down for the Britannia project as a whole.
An unavoidable minimum investment is necessary to produce sales gas to specification. Additional investment to increase liquids recovery must produce increased revenue, at a higher internal rate of return (IRR) than the minimum investment.
- Flexibility. The liquid product prices can vary widely on the world market, and the sales gas' calorific value must be controlled within a narrow range (typically 0.5%).
The gas terminal's design, therefore, must be flexible enough to control the liquid recovery so as to optimize product revenues as well as to enable different gas compositions and inlet conditions to be accommodated.
- Reliability. Careful consideration must be given to provision of standby equipment to improve Reliability of the terminal. Downtime is costly both in lost revenue (which cannot be recovered later) and in penalties for failure to meet nomination.
The value of lost revenue for a terminal with capacity of 500 MMscfd is about $1.6 million/day. Considerable capital expenditure can be justified to save even 1 day's downtime per year.
Control, instrumentation, and isolation philosophies must lead to a robust and easily operated plant, but also one in which safe maintenance and changeout of equipment are possible without shutting down production.
PROPANE-RECOVERY
OPTIMIZATION
The next step in the design process was to find the optimum liquid production for the terminal, while meeting the requirements previously described. Liquid production is conveniently measured by portion (%) of propane recovered from the feed gas.
ROLE OF CAPITAL COST
At an earlier stage in the project, detailed equipment sizing and cost estimates had been developed for two process designs: for 55% propane recovery and 95% propane recover.
Each design had a capacity of 1 bcfd feed gas, in two trains. As might be expected, there was a considerable difference in costs.
Fig. 2 shows the installed capital cost of the gas-processing section for these two designs. Other units common to both designs, such as sweetening, dehydration, liquid fractionation and storage, and utility systems, are excluded.
It was clear from preliminary calculations that the increase in capital expenditure in going from 55 to 95% propane recovery could not be justified by the incremental product revenue. It was felt, however, that more economic designs could be developed which would be at intermediate positions on this diagram.
The minimum propane recovery that would enable the sales gas to meet specification was identified as 27%. A design to meet this propane recover defines the minimum investment. Incremental investment and incremental revenue could be calculated from this base.
Attempts were made to improve the 95% recovery design, which led to some cost reductions and began to indicate the shape of the capex/propane-recovery diagram (Fig. 3).
Fig. 3 includes the cost estimates for the 27% recovery design and a design at 80% recovery. It appeared that there might be a sharp increase in capex between 55% and 80% propane recovery.
The higher NGL-recovery designs included two stages of pressure letdown with large gas-turbine-driven sales-gas compression, while the 55% recovery design had a single stage of pressure reduction without an externally driven sales gas compressor (Fig. 4).
To investigate by how much propane recovery could be increased beyond 55% without going to a second stage of pressure reduction and major sales-gas compression, alternative designs were developed, and cost estimates were prepared.
It was found that the propane recover could be increased to as high as 79% with a single stage of expansion still being used without introduction of externally driven sales-gas compression.
The increased propane recover was achieved by an increase in the amount and location of heat-exchange surface and by introduction of a bypass so that some expanded gas could be recompressed by the de-ethanizer overhead compressor.
Results appear in Fig. 5 which shows a sharp, almost stepwise increase in capex associated with the increased complexity and compression required at greater than 80% propane recovery.
The rise in cost between 55 and 79% recover reflects in general the increased heat-exchange surface and the increasing proportion of it which is in stainless steel.
ECONOMIC PERFORMANCE
Some of the design options developed at this stage could be eliminated on grounds of operability or obvious higher cost. five designs were taken forward for a comparison of incremental economic performance above the minimum recovery case.
For compatibility with the basis for the rest of the Britannia project, the terminal capacity was scaled to 715 MMscfd sales gas, and costs for a liquid fractionation unit were included in this evaluation. These designs are listed in Table 1.
Revenue streams were calculated for a 20-year project life based on the offshore production profiles. Typical study assumptions for product pricing, escalation, and operating expenses (opex) are given in the accompanying box.
With these revenue streams, the incremental values of three economic indicators were calculated for each increase in propane recovery across the five designs. The indicators used were internal rate of return (IRR), net present value (NPV), and profitability index (PI).
The values of these indicators for the incremental steps between designs are shown in Figs. 6, 7, and 8.
ECONOMIC RECOVERY
These economic indicators show that, for a European-based gas terminal, sound economic justification exists for recovering NGLs in excess of the minimum. Deep recover using a two-stage process with sales gas compression, however, is not justified.
The cumulative NPV resulting from increased liquid recover above the minimum (Fig. 7) continues to increase up to about 79% propane recovery. But the greatest incremental IRR (Fig. 6) is obtained for the improvement from 55 to 62% propane recover.
Further increases in propane recovery yield lower rates of return on the incremental capex, but the improvement from 62 to 72% propane recover still provides an improved incremental IRR which is greater than the project target.
Increasing from 72 to 79% recovery provides a lower rate of return and only a small increase in NPV.
These results led to the conclusion that the project should proceed with the flowsheet for 72% propane recover.
The flowsheet in Fig. 9 reflects the process chosen. Some significant features of this flowsheet are:
- Single separator, single column feed
- Gas expanded in one stage in a turboexpander
- Gas remaining in dense phase until the expander inlet
- About 20% of separated gas bypassed to the deethanizer overheads, which transfers load from the expander-driven recompressor to the motor-driven compressor
- Liquid recover to be varied by variations in the split of feed between the inlet heat exchangers.
REFERENCES
- World Gas Intelligence, Pet. & Eng. Intelligence Weekly Inc.
- FLAGS newsletter, June 2, 1982.
- County Natwest Woodmac report.
- North Sea Newsletter, Financial Times, London.
- Engineering News Record.
Copyright 1993 Oil & Gas Journal. All Rights Reserved.