Atmospheric Dispersion of CO2 Following Full-Scale Burst Tests

12 Pages Posted: 16 Apr 2019 Last revised: 27 Oct 2020

See all articles by Ajit Godbole

Ajit Godbole

University of Wollongong - Energy Pipelinec CRC Ltd.

Xiong Liu

University of Wollongong - Energy Pipelinec CRC Ltd.

Guillaume Michal

University of Wollongong - Energy Pipelinec CRC Ltd.

Bradley Davis

University of Wollongong - Energy Pipelinec CRC Ltd.

Cheng Lu

University of Wollongong - Energy Pipelinec CRC Ltd.

Keith Armstrong

DNV GL Testing and Research Centre

Clara Huescar Medina

DNV GL Testing and Research Centre

Abstract

The CO2SafeArrest Joint Industry Project (JIP) was set up in May 2016 to study (1) the fracture propagation and arrest characteristics of steel pipelines carrying anthropogenic Carbon Dioxide (CO2), and (2) the atmospheric dispersion of CO2 following release from a fractured pipeline. The participants in the JIP are the Energy Pipelines Cooperative Research Centre (EPCRC), Australia, and DNV GL (UK/Norway). The project is funded by the Commonwealth Government of Australia through the Carbon Capture and Storage Research Development and Demonstration (CCSRDD) fund, and by CLIMIT (Norway).

Two full-scale burst tests of steel pipelines filled with a highly compressed mixture of CO2 and N2 were carried out at the DNV GL Testing and Research Centre site at Spadeadam, UK, on 30 Sept 2017 and 24 March 2018 respectively.

This paper describes the experimental and analytical/numerical investigation of the dispersion of CO2 in the atmosphere following its release in the two burst tests. In the first test, the entire length of the pipe test section was buried under a 1 m deep soil cover. In the second test, only half the length of the pipe test section was buried. In both tests, an explosive charge placed at the top dead center (TDC) at half-length was detonated to initiate a fracture in the pipe wall. The resulting crack propagated in both directions as the pipe wall was torn open sideways. This allowed the high-pressure contents of the pipe to be released into the atmosphere. The resulting CO2 cloud billowed up initially to a considerable height, before beginning to sink to the ground as it was simultaneously swept away by the prevailing wind. A number of sensors arranged in a ‘fan’ pattern over the terrain recorded the concentration of CO2 as the dispersing cloud passed over them. For each test, Computational Fluid Dynamics (CFD) simulations carried out for a number of likely scenarios before the actual event were used to finalize the sensor layout. The CO2 concentration histories obtained in the experiments were compared against predictions of CFD simulations of the dispersion.

The dispersion is influenced predominantly by four factors: (1) nature of the dispersing substance, (2) the nature of the terrain, (3) the prevailing weather and wind, and (4) ‘source strength’. In the CFD simulations, features of the surrounding landscape and wind/weather were incorporated in the problem definition. The source strength was specified as a mathematical function, approximately representing the highly transient, explosive release of CO2 into the atmosphere. It was found that despite the simplifying assumptions used in the CFD models, there was reasonably good agreement between the predicted and measured CO2 concentration histories at the nominated sensor locations. These studies will facilitate a realistic assessment of the risk associated with a sudden large-scale release of CO2 into the atmosphere, particularly in relation to an estimation of the ‘consequence distance’. It is proposed that the concept of consequence distance be associated with time, due to the highly transient nature of the event.

Keywords: Safety and dispersion, GHGT-14

Suggested Citation

Godbole, Ajit and Liu, Xiong and Michal, Guillaume and Davis, Bradley and Lu, Cheng and Armstrong, Keith and Huescar Medina, Clara, Atmospheric Dispersion of CO2 Following Full-Scale Burst Tests. 14th Greenhouse Gas Control Technologies Conference Melbourne 21-26 October 2018 (GHGT-14) , Available at SSRN: https://ssrn.com/abstract=3365779 or http://dx.doi.org/10.2139/ssrn.3365779

Ajit Godbole (Contact Author)

University of Wollongong - Energy Pipelinec CRC Ltd. ( email )

Australia

Xiong Liu

University of Wollongong - Energy Pipelinec CRC Ltd.

Australia

Guillaume Michal

University of Wollongong - Energy Pipelinec CRC Ltd. ( email )

Australia

Bradley Davis

University of Wollongong - Energy Pipelinec CRC Ltd.

Australia

Cheng Lu

University of Wollongong - Energy Pipelinec CRC Ltd.

Australia

Keith Armstrong

DNV GL Testing and Research Centre

Spadeadam
Cumbria
United Kingdom

Clara Huescar Medina

DNV GL Testing and Research Centre

Spadeadam
Cumbria
United Kingdom

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
119
Abstract Views
552
Rank
422,437
PlumX Metrics