Greening up LNG [NGW Magazine]
To maintain public acceptability of natural gas, help Asian and European buyers meet their Paris emission targets, and make it easier for LNG producing countries to meet their own Nationally Declared Commitments – assuming they have them – developers are increasingly working towards cutting the wellhead to waterline greenhouse gas (GHG) emissions of their projects.
The improved GHG performance has the added benefit of increasing the likelihood of these long-life projects surviving their natural lives in an increasingly carbon-constrained world.
There are two main ways developers are decreasing the carbon footprint associated with their LNG projects: carbon capture and storage (CCS); and the electrification of plant and upstream facilities.
Natural gas is regularly billed as the cleanest of the fossil fuels, with its full-cycle GHG emissions intensity half that of coal and 30% less than crude oil. But the actual performance of natural gas depends on numerous factors, including distance from wellhead to final market; whether it is transported primarily by pipeline or as LNG; and numerous project-specific factors, both upstream and downstream, including the quality and source of the gas and the amount of methane emissions.
The Clean Energy Association of British Columbia – also known as Clean Energy BC – has reported global average emissions of 0.58 metric tons (mt) of CO2 equivalent (CO2e) mt of LNG produced on a well-to-water basis. But a 2018 study by Delphi Group found the GHG intensity performance of 19 LNG liquefaction plants from around the world ranged from 0.15 mt - 0.70mt CO2e/mt LNG.
How old a plant is, and where it is, influence the emissions. For example those in arctic conditions such as Yamal LNG tend to be more efficient. The source of power for the liquefaction process and auxiliary units, and other various technologies employed in the plant, also affect the emissions.
For example, a conventional LNG plant uses so-called direct drive natural gas power to chill the gas to -162 °C to liquefy it, accounting for up to 90% of a facility’s relatively high GHG emissions. In contrast, a modern electric drive (e-drive) plant, powered by combined-cycle gas power, or better yet renewable sources of power, can drastically slash the carbon intensity.
According to the Delphi Group study, upstream GHG emissions from producing, processing and transporting the gas are even more variable than for LNG plants and their volume depends on a number of factors unique to each project. These include the type of gas production – whether onshore or offshore, hydraulically fractured or conventional – and the percentage of CO2 found in the gas, as well as the distance from the gas formation to the coast.
Currently the two lowest GHG emitting LNG projects well-to-water are Norway’s Snoehvit and Australia’s Gorgon with emission intensities of around 0.35 tCO2e/mt LNG. These best-in-class projects benefit from the use of offshore gas with sub-sea gathering systems, which emit low levels of methane into the atmosphere; high-efficiency electricity plants for their cooling and other power needs; and CCS to reduce the amount of CO2 vented from their gas processing plants.
Carbon Capture and Storage
To date, CCS has not been used on either a pre-combustion basis for a LNG liquefaction plant, or on a retrofitted post-combustion basis for liquefaction for a simple reason: It is prohibitively expensive to add CCS to the front end of a plant, and there are better options in countries with access to low-carbon electric power (see below).
No estimates are readily available for capturing carbon from a liquefaction plant, but a large-scale CCS retrofit project to capture 1mn mt/year of CO2 from a coal-fired power plant presently costs roughly $1bn.
A 2015 study by the US National Energy Technology Laboratory (NETL) found retrofitted CO2 capture cost an average of $56/mt for a pulverized-coal (PC) plant and $71/mt for a natural gas combined-cycle (NGCC) plant, given the more dilute CO2 stream of a natural gas-fired plant. The same would go for a LNG cooling plant powered by gas.
On the other hand, CCS is relatively cost effective for gas processing plants, including those associated with liquefaction. Processing plants must reduce CO2 to no more than a 2% share of gas when transporting it by pipelines, while an LNG processing plant must reduce it to a mere 50 parts/mn by volume (ppmv) before liquefaction.
The Global CCS Institute (GCCSI) estimates the cost of capturing CO2 from large-scale industrial emitters to range between $20/mt for a natural gas processing plant and about $200/mt for a cement plant.
As a result, it should be no great surprise that the two LNG plants to use CCS mentioned above, Snoehvit LNG since 2007 and Gorgon LNG since last August – which was roughly three years behind schedule – are sequestering CO2 that their gas processing plants are capturing anyway.
There are presently 24 large-scale CCS facilities – at least 0.4mn mt/yr – in operation, under construction and completed but no longer operating around the world, according to the GCCSI. Eleven of those capture CO2 from gas processing plants, often for the purpose of enhanced oil recovery (EOR).
In early October, Qatar – the world’s largest LNG producer and exporter for years, but presently neck-and-neck with Australia – announced it had commissioned a 5mn mt/yr CCS project in conjunction with its planned 40% increase in LNG liquefaction capacity by 2025. The captured CO2 is to ultimately support EOR in the country’s oilfields.
The Qatari government has yet to provide details about its CCS project, but it should be noted natural gas from its massive North Field tends to have relatively low CO2 concentrations – 2.2%, compared with the global average of 2% and a high of around 14%.
Qatar has no GHG emissions target under the Paris agreement, hence the purpose of its CCS project appears to be to make its LNG more appealing to potential buyers with targets and plans to retire higher emitting coal-fired power plants. And the European Union might go ahead and tax the carbon footprint of imports.
For the sake of comparison, Norway’s Snoehvit stores 0.7mn mt/yr of CO2e, while the Gorgon CCS project is projected to store between 3.4mn to 4mn mt/yr once it is fully operational, making it the world’s largest. Natural gas from Gorgon has a CO2 concentration of 14% before processing, and from Snoehvit it is between 5% and 8%, depending on which field the gas is coming from.
In contrast to CSS, high-efficiency electrification of a LNG liquefaction project’s cooling and ancillary power needs provide a relatively cost effective option for reducing GHG emissions, especially if low-cost renewable energy sources are readily available.
As previously mentioned, the emission profiles of Snoehvit LNG and Gorgon LNG both benefit from the use of high-efficiency, gas-fired electric plants for both cooling and other power needs. These e-drive plants, with energy efficiency levels up to 70%, compress the gas – rather than less efficient direct drive systems.
The next iteration for improving GHG emissions from LNG liquefaction plants is to bypass gas-fired electricity plants altogether, and instead use renewable energy for all of their power needs. This is the route most LNG projects in Canada are at least considering, and some in the US as well.
All six of the most advanced LNG projects in Canada are considering using the country’s abundant, low-cost renewable energy – especially hydropower – for their secondary power needs, and all but LNG Canada project plan to use it for compression.
The first phase of the LNG Canada project, sanctioned in October 2018, is too far along in the process to swap renewable energy for gas-fired power plants for its two 7mn mt/yr liquefaction plants.
Despite this, the emissions intensity of the first phase of LNG Canada’s project is forecast to be a mere 0.15 mt CO2e/mt LNG, which is half the global average and about a fifth less than the best current facility. The project’s emission reductions come from a combination of low emission aeroderivative engines to liquefy the gas, relatively low temperatures on the British Columbia (BC) coast, and hydropower for its auxiliary power needs.
On the other hand, the 18mn mt/yr Kitimat LNG project, currently led by Chevron, and the 2.1mn mt/yr Woodfibre LNG, both also on the BC coast, as well as the 11mn mt/yr Energie Saguenay LNG project in Quebec, are all planning to be all-electric from the outset and powered by low-cost hydropower from each of these provinces. It has been estimated this could push down plant emission intensities to as low as 0.065mt CO2e/mt of LNG -- and even further in the case of Energie Saguenay, which is aiming for just 0.04/mt CO2/mt LNG.
The two LNG projects on the east coast of Canada in the province of Nova Scotia – Pieridae Energy’s 5-10mn mt/yr Goldboro LNG and the up to 8mn mt/yr Bear Head LNG project – had planned to use gas-fired power generation for compression, but are now exploring the possibility of reconfiguring their designs to source hydropower from Newfoundland and Labrador instead, depending on the reliability and cost of the supply.
In Texas, Freeport LNG sources power from the grid to run liquefaction, utility compressors, and all pumps and fans at its plants. The improvement in emissions intensity is presently not substantial, sitting around 0.24 mt CO2e/mt LNG – not much better than the global average of 0.30 mt CO2e/mt LNG – given the Texas grid is fairly carbon intensive. But the state’s grid is decarbonising quickly, with rapid development of wind and solar power projects, which will bring down the carbon intensity of Freeport LNG over time.
The largest potential GHG reductions in the upstream are by switching compressor and generator power from natural gas and diesel to clean grid electricity – hydro, wind, solar and thermal. This not only reduces emissions from source fuel, but the replacement of gas activated pumps and compressors with electric equipment also eliminate or reduce methane venting and fugitive emissions.
Methane is a powerful, albeit relatively short-lived GHG when released directly into the atmosphere. Its estimated global warming potential is 86 times greater than CO2 over a 20-year period, declining to 34 times over a 100-year period.
A prime example of an upstream project incorporating measures to significantly reduce its carbon footprint is Shell Canada’s Groundbirch asset in northeast BC – presently encompassing some 500 wells and four gas processing plants – which will feed Shell’s share of gas for the LNG Canada project upon completion in 2024.
Since 2016, Shell has reduced its GHG emissions by a quarter at Groundbirch through a number of different measures. The biggest gains have come from electrifying its gas plants and sourcing hydropower from the grid, according Shell Canada. Emissions from these gas plants have been slashed by 90%, or an estimated 150,000 mt CO2e/yr.
In addition, the installation of electric actuator valves, instead of pneumatic valves that release methane into the atmosphere every time they are activated, have helped Shell slash methane emissions at Groundbirch to a mere 0.1% of gas production. The US Energy Information Administration estimates global average methane emissions from gas production to be 1.7% and 1.3% for US gas production.
Finally, Shell has gone to great lengths to decrease GHG emissions associated with drilling new wells at Groundbirch. Besides using electric actuator valves, the company has switched from diesel to gas-powered rigs and uses solar panels and thermoelectric generators to meet the power needs of new well pads. Shell is looking at adopting further measures in the future, including the use of zero-emissions methanol fuel cells.
Depending on the LNG project, upstream emissions from producing, processing and transporting natural gas – including so-called fugitive methane emissions – can be higher than GHG emissions associated with the liquefaction process.