Reprinted from SEP/OCT 2013 LNGINDUSTRY

Bill Howe, Geoff Skinner and Tony Maunder, Gasconsult Ltd, UK, discuss the development

of new liquefaction technology.

This article describes the development of Gasconsult’s ZR-LNG™ dual methane expander cycle liquefaction technology and its fit with current trends in LNG project

development.Most LNG production takes place in large scale plants with single

train LNG outputs up to 7.8 million mtpa. All are characterised by a high degree of complexity to maximise energy efficiency and/or co-product value realisation, and carry the knock-on burdens of high capital cost and extended project schedules. Demanding planning approvals and high labour cost inflation have also been a feature of some recent developments.

Over the past decade, interest has developed in so-called mid-scale LNG for exploitation of smaller gas fields with reserves of around 1 trillion ft3, and offshore opportunities unable to support the pipeline capital cost to a land based liquefaction plant. Of necessity, these smaller gas monetisation prospects required lower capacity and lower capital cost plants than current base load schemes. Despite initial interest in the mid-scale sector, little materialised,

LNGINDUSTRY Reprinted from SEP/OCT 2013

particularly in respect of FLNG. Strong interest is only now resurfacing.

A feature of the current committed and planned FLNG facilities has been the increase in envisaged plant capacity. Most oil majors are now talking of facilities producing circa 4 million mtpa, which would likely require single train capacities of a minimum of 2 million mtpa. The move to this larger capacity is driven by superior project returns; it also requires no more internal resources to get a 4 million mtpa project through the major’s capital approval systems than a 1 million mtpa plant.

A further feature is a marked preference by certain operators for elimination of liquid hydrocarbon refrigerants offshore. Higher molecular weight hydrocarbons, particularly propane, are extremely hazardous and represent an explosion/fire risk when accumulating in confined spaces.

For safety reasons, a level of support has thus developed for nitrogen expander processes for FLNG applications.

Power consumption for nitrogen cycles is high compared with mixed refrigerant processes, and the inherent large gas recirculation rates also lead to larger line sizes and heavier plants. These factors put nitrogen cycle schemes at a disadvantage, particularly for the higher plant capacities now under consideration. Even with a low cost energy source there are strong arguments for pursuing high process efficiency. Lower power consumption reduces the size of the compression equipment, the CAPEX of the plant and associated CO2 emissions.

DevelopmentGasconsult’s ZR-LNG process was originally conceived in the mid-2000s. Extensive engineering development was completed on early versions of the technology on a 1 million mtpa modular train for FLNG application. Capital costs and power demands for the latest patented variant are presented in the case studies included later in this article.

Recent developments have seen Gasconsult address further market opportunities, which arise from the advantageous CAPEX and OPEX characteristics of ZR-LNG. This is for a larger single train nominal capacity of 2 million mtpa for greater capital efficiency; whilst simultaneously achieving the low power demand required for larger LNG schemes. This opens up the possibility of deploying expander technology on higher capacity plants based on multiple 2 million mtpa trains. Details of this development are provided later in this article.

ProcessThe need to reduce the power demand for an expander based process while preserving the safety and simplicity of the nitrogen cycle led to the development of the ZR-LNG process. In this process, the refrigerant is methane derived from the feed natural gas. ZR-LNG can achieve a net liquefaction unit drive power of 260 – 320 kWh/tonne of LNG (depending on the feedstock composition, pressure and ambient conditions). This low power demand is achieved without the additional process complexity arising from feed gas pre-cooling. A schematic of the process is shown in Figure 1.

Liquefaction is achieved through the use of two separate expander refrigeration circuits indicated in red and blue. Typically, 35% of the compression power requirement to operate the process is recovered through the gas phase expanders. A further reduction in energy demand is effected by a turbine on the liquid product run down to storage.

With its low energy consumption and low capital cost, ZR-LNG is suitable for both onshore and offshore application up to a capacity of 2 million mtpa per train and can operate on a full range of hydrocarbon gases, including very lean feeds containing insufficient C2+ for production of a hydrocarbon refrigerant.

Design simplicity is encapsulated in ZR-LNG; a typical plant comprising only two compressor packages plus eight major equipment items. The cold box has only three passages (or four when pre-condensation of NGLs is necessary); all passages in the heat exchange cores having vapour phase feeds. As the process has no external cryogenic refrigerant cycle and no liquid or nitrogen top-up system, equipment items are eliminated, together with

Table 1. Basis of designGas composition Mol% CH4 95%, C2H6 4%, C3H8 1%Gas pressure at liquefaction inlet 60 bar gSea water temperature 13 ˚CIndirect cooling – sea water/circ water

3 ˚C approach

Process streams cooled to 20 ˚CHeat leak to cold box 0.5%Minimum cryogenic approach temp. 3 ˚CRecycle gas compressor polytropic η 85%Expander adiabatic η 87%

Table 2. Basic operating parameters

Online factor 345 d/yr

Flow rate 121 mtph

Main recycle compressor power demand 54.7 MWe

Flash gas compressor power demand 3.4 MWe

Total power 58.1 MWe

Expander power recovered to process 21.4 MWe

Net power 36.7 MWe

kWh/t of product 305

Figure 1. ZR-LNG schematic.

Reprinted from SEP/OCT 2013 LNGINDUSTRY

associated bulk materials and their fabrication and construction. The focus on simplicity achieves a significant reduction in capital cost.

Three factors contribute to ZR-LNG’s other key attribute – its significantly lower power requirement relative to nitrogen cycles. The main contributing factor is the higher molar specific heat and lower molar compression power requirement of methane. This yields lower recycle flow rates and attendant lower power demand. A second factor is that liquefaction of part of the feed gas occurs in the liquefying

expander, converting latent heat directly into mechanical work. The third factor is that with methane refrigerant, it is convenient to condense the feed at -120 to -130 ˚C and to flash the resulting condensate to the required product temperature, usually -160 ˚C. The resulting flash gas is recovered through the methane recycle compressor. With typical nitrogen cycles, it is necessary to cool the feed to a much lower temperature to minimise the amount of flashed vapour, as its recovery is not possible through the nitrogen compression system.

Gasconsult has quantified the benefits of the above factors. Several dual expander nitrogen cycle configurations were evaluated on the same basis as ZR-LNG with respect to ambient conditions, machine efficiencies, loop pressure drops, heat exchanger temperature approaches and heat in-leakage. Hysys simulations indicate ZR-LNG has up to 30% lower suction compressor volumes and over 20% lower aggregate machine kW than nitrogen expander schemes.

Case studiesTwo case studies are presented below. Figure 1 provides the basic ZR-LNG flow scheme applicable to these 1 million mtpa FLNG schemes.

1 million mtpa FLNG modular schemeThe basis of the design is recorded in Table 1 and the related power demands are recorded in Table 2. The power consumption of 305 kWh/tonne is achieved by ZR-LNG in its basic form, and with no feed gas pre-cooling.

The cost estimate using pre-fabricated liquefaction modules for FLNG application is provided in Table 3. This estimate, based on vendor quotations against fully detailed equipment specifications, covers an EPIC work scope and is provided on a 2013 instant execution basis. It relates to the liquefaction unit only and excludes the vessel, feed gas purification, NGL fractionation, utilities, LNG/NGL storage, flare and owners costs.

Design excursion – revised process conditionsRecognising that plant performance is impacted by project specific factors, ZR-LNG was modelled using various revised process conditions. Specifically, this design excursion allows for the use of low temperature deep sea water cooling as proposed by some industry professionals for FLNG applications. Power demand was then calculated for various liquefaction operating pressures. Feed gas pre-cooling was not considered because of the wish to avoid increased plant complexity and the growing concern within the industry regarding liquid refrigerant safety issues.

The outcomes are provided in Table 4. Projected power demands are in the range 255 – 270 kWh/tonne without feed gas pre-cooling.

Development of 2 million mtpa trainWith the realisation that some oil majors are showing preference for larger FLNG schemes, and recognising that power demands in the range of 255 – 305 kWh/tonne (see Tables 2 and 4) repositions expander technology for larger capacity plants, Gasconsult developed a conceptual design

Table 3. CAPEX estimate – 2013 (US$ millions)

Equipment supply plus spares 62.8

Bulks supply 14.8

Installation/construction/fabrication 18.9

Transportation 1.9

Plant total 98.4

License fee/insurance/certification 6.0

Project management/engineering/commissioning

28.1

Total engineering plus fees 34.1

Contingency 19.9

Total 152.4

Table 4. Revised designSea water termperature 4 ˚C

Power consumption kWh/t

ZR-LNG 80 bar liquefaction pressure – no pre-cooling 255

ZR-LNG 70 bar liquefaction pressure – no pre-cooling 260

ZR-LNG 60 bar liquefaction pressure – no pre-cooling 270

Table 5. CAPEX estimate – 2013 (US$ millions)

Equipment supply plus spares 105.2

Bulks supply 24.9

Installation/construction/fabrication 31.6

Transportation 3.2

Plant total 164.9

License fee/insurance/certification 11.4

Project management/engineering/commissioning

47.2

Total engineering plus fees 58.6

Contingency 33.5

Total 257

LNGINDUSTRY Reprinted from SEP/OCT 2013

for a nominal 2 million mtpa single train ZR-LNG plant. Technical feedback and pricing from equipment suppliers indicates that it is a feasible and attractive proposition.

A leading supplier has proposed a compact single casing recycle compressor with power in the range of 70 – 80 MW. This compressor can be driven by an industrial-type gas turbine such as Hitachi H-80. Alternatively, and particularly for FLNG applications, two aero-derivative gas turbines, such as GE LM6000 or Rolls-Royce Trent, may be used. Two or three expanders would be required, depending on the suppliers’ designs.

A schematic of a candidate compressor/expander configuration for the 2 million mtpa scheme is shown in Figure 2. This is based on the basic process data provided for the 1 million mtpa FLNG modular scheme and achieves a liquefaction power demand of 300 kWh/tonne. Figure 2

shows the arrangement of the main compressor drives only (excluding supplementary drives).

In addition to the compression equipment, the cold box design has been verified by a leading manufacturer. Based on vendor budget quotations for all major equipment, this larger ZR-LNG module appears competitive for medium-size production; and in multiples as a building block for larger LNG projects both on land and offshore. Table 5 provides the cost estimate using pre-fabricated liquefaction modules for FLNG application for a 2 million mtpa plant. This estimate covers an EPIC work scope and is provided on a 2013 instant execution basis. It relates to the liquefaction unit only and excludes the vessel, feed gas purification, NGL fractionation, utilities, LNG/NGL storage, flare and owners costs.

The expander envelopeGasconsult has modelled the generic dual expander nitrogen and single mixed refrigerant (SMR) cycles against ZR-LNG. Figures 3 and 4 depict the relative combined annual cost of the fuel and capital amortisation for the liquefaction units of a 2 million mtpa LNG facility.

The figures exclude the CAPEX and OPEX of utilities and gas pre-treatment, which are assumed equivalent and assume factors such as maintenance, staffing, insurances, etc. are equal for all the technologies. Figure 3 relates to a typical stranded gas cost of US$ 2/million Btu and Figure 4 to anticipated longer term US gas costs for export projects of US$ 6/million Btu. Fuel consumption data is based on the use of aero-derivative gas turbine drives.

The charts indicate the high influence of capital cost on the overall liquefaction cost for plants operating on low cost feed gas, and the significant increase in influence of energy efficiency on overall costs as the gas cost increases. At the lower stranded gas cost there is little to choose between the SMR and dual expander nitrogen processes at the 2 million mtpa capacity level. For US market conditions, SMR achieves lower overall costs than the nitrogen system. This reflects SMR’s superior energy efficiency. The ZR-LNG combined fuel and capital cost is lower than both SMR and dual nitrogen under all gas pricing and capacity criteria by 20 – 25%. It enables expander technology to be applied in areas where the SMR process previously held a competitive advantage.

ConclusionIn the mid-scale single train capacity range up to 2 million mtpa, the ZR-LNG process is positioned as a simpler lower capital and operating cost process than both nitrogen expander cycles and SMR schemes. The significant reduction in complexity and cost is achieved with a quite limited sacrifice of energy efficiency compared to existing base load plants. ZR-LNG repositions expander technology; widening its application envelope to larger capacity and higher gas cost schemes, whilst securing a CAPEX, OPEX, and operational advantage for the small to mid-scale market. For FLNG applications, ZR-LNG offers efficiency and CAPEX advantages, whilst preserving the operational benefits of nitrogen cycles including safety, tolerance to ships motion, rapid start-up and reduced flaring.

Figure 2. Schematic – drive configuration.

Figure 3. Gas cost US$ 2/million Btu.

Figure 4. Gas cost US$ 6/million Btu.