The prospective and ongoing oil and gas activity in the northwest of Western Australia is of a truly global scale, and its continued realisation presents significant challenges and opportunities. The activity started in 1984 through the establishment of the North West Shelf Venture (NWSV) facilities that, representing an investment of $A27 billion, constitute Australia’s largest oil and gas resource development and currently account for more than 40 per cent of Australia’s oil and gas production.

Beyond the NWSV, several major gas developments in the region – including Pluto, Gorgon and Prelude – are already postfinancial investment decision with several other major projects fast approaching this point. The investment in the development of these offshore gas fields over the next decade will be at least $A100 billion, and since the primary markets for this gas are China, Japan and Korea, the efficient production of LNG is central to the monetisation of these vast gas resources.

However, the natural gas in many of the fields to be developed contains contaminants such as nitrogen gas (N2) and CO2 at levels that were once considered prohibitive. Innovative technical solutions for the separation and treatment of these contaminants will provide engineers the flexibility needed to minimise the energy required to produce the LNG and therefore the environmental footprint of these major developments.

Cryogenic pressure swing adsorption technology

Cryogenic pressure swing adsorption (PSA) is a technology being developed to address two of the most challenging and important gas separation applications: the removal of N2 from methane (CH4) dominant streams (nitrogen rejection), and the removal of CH4 from N2-dominant streams (nitrogen purification). Increasingly, cryogenic distillation towers known as nitrogen rejection units (NRU) are incorporated into the final stages of LNG trains to ensure the N2 content of the LNG is within specification (less than approximately 1 per cent). These NRUs are expensive to construct and operate, but they are needed because the natural gas being fed to the plant often contains N2 levels greater than 4 per cent.

Although it is inert and thus relatively benign, N2 is parasitic because energy is wasted cooling it, even if an NRU is present after the main cryogenic heat exchanger to remove it from the LNG. Furthermore, the overhead vapour from the NRU which contains the N2 also contains significant amounts of CH4, and this cryogenic gas mixture must be processed further to recover the valuable methane and allow disposal of the nitrogen. Further cryogenic distillation is often needed to achieve this, resulting in a significant additional expense.

The Centre for Energy are developing cryogenic PSA technology to more efficiently address this important challenge in gas processing and LNG production. Adsorption-based separation process applications Adsorption-based separation processes are used widely in several industries including natural gas processing for the following applications: dehydration, CO2 removal, and hydrocarbon dew point control. Gas dehydration usually involves temperature swing adsorption (energyintensive) and is effective because traditional adsorbents such as zeolites have a strong affinity for H2O, a polar molecule, and because even saturated natural gas does not contain a large amount of water.

Other current applications of adsorptionbased separation in gas processing involve combinations of these same factors – for example, CO2 is moderately polar – and are usually only applicable to small gas flows. In contrast, N2 and CH4 are very difficult to separate because they are such similar species with cryogenic normal boiling points -196 degrees Celsius and -161 degrees Celsius – very similar molecular sizes – 0.36 nanometres (nm) and 0.38 nm – and are both non-polar molecules.

Furthermore, the scale of the flows associated with gas field developments in Western Australia usually precludes adsorption-based gas separations because of the bed sizes that would be required if operated in a conventional temperature-swing, slow-cycle mode; these typically use regeneration times longer than one hour.

There are three important innovations associated with the development of cryogenic PSA which make it a potentially suitable technology for this demanding application. First, because adsorption increases exponentially as temperature decreases, the cryogenic conditions readily available in LNG plants can be exploited to significantly increase the storage capacity and dynamic selectivity of a given adsorbent. Second, novel adsorbents with large surface areas (SA) and controlled pore-size distributions (PSD) are continuously being developed by us and other researchers to maximise N2/CH4 selectivity and capacity. Third, the use of rapid-cycle PSA, in which beds are regenerated every few minutes by depressurisation rather than once or twice a day by heating, provides an avenue for minimising the adsorbent tower size and the energy required for regeneration.

The search for the optimal adsorbents

The University of Western Australia (UWA) has developed novel experimental systems to find the optimal adsorbents needed for cryogenic PSA processes and to acquire the data necessary to design them. Three apparatus, with very different operating conditions and principles, are used to screen and characterise novel adsorbents.

A commercial ASAP2020 system, which operates with pure fluids over the pressure range 10-6–1 bar, is used to determine the SA and PSD of new adsorbents and determine their pure gas capacities, kinetics and heats of adsorption over the temperature range –30 degrees Celsius. The behaviour of selected adsorbents with gas mixtures is then studied in a custom-designed, static volumetric system which measures the adsorbent’s equilibrium capacity and N2/ CH4 selectivity over the pressure range 0.1–70 bar and temperature range minus 160–150 degrees Celsius.

Finally, a dynamic breakthrough column is used to determine the kinetic selectivity of adsorbents and to extract the parameters needed to design a rapid-cycle PSA process: gas mixtures at pressures between 1 and 10 bar are flowed at rates between 50 and 200 standard cubic cm min-1 through a 1-inch column maintained at temperatures between -90 degrees Celsius and 100 degrees Celsius.

The effluent mass flow and composition are measured, as is the adsorbent’s temperature using three sensors embedded along the bed.

Importantly, all three systems can operate at a common set of conditions and validate the measurements made by one another.

Key adsorbent parameters such as capacity, kinetics, selectivity and heats of adsorption can be extracted from each system. These parameters are then used

in a design tool, which determines the number and sizes of adsorbent towers required for a given application and cycle time. A dynamic model is then used to test various PSA cycles to investigate the feasibility of the design in terms of meeting processing (product purity) and operational specifications (energy cost).

These modeling tools establish whether the adsorbent can serve as the basis of a cryogenic PSA gas separation application, or whether further refinement of the adsorbent material is required.

The UWA has already identified novel adsorbents with potential for N2 rejection and N2 purification applications in LNG plants. Further experiments are underway to more extensively characterise these materials, while we also continue to screen and modify other adsorbents for even better performance. In the near future we plan to demonstrate the most effective of these rapid cryogenic PSA cycles using a prototype multi-column apparatus to validate the model designs.

Acknowledgements: This research is funded by Chevron Corporation, the Australian Research Council and the Western Australian Energy Research Alliance. We thank Tom Rufford, Guillaume Watson, Paul Hofman, Nathan Jensen, Craig Grimm and Ida Chan for their significant contributions to this research.