ReviewCO2 transport: Data and models – A review
Introduction
In the two-degree scenario (2DS) of the International Energy Agency [1], which is one possible way of reaching the two-degree goal, CO2 capture and storage (CCS) contributes to reducing the global CO2 emissions by about six billion tonnes per year in 2050. To achieve this scenario, CO2 must be transported from the points of capture to the storage sites. A large fraction of the captured CO2 is likely to be transported in pipeline networks. Pipeline transport of CO2 is different from that of natural gas in a number of ways. First, the CO2 will normally be in a liquid or dense liquid state [2], [3], whereas the natural gas most often is in a dense gaseous state, see e.g. Aursand et al. [4]. Second, depending on the capture technology, the CO2 will contain various impurities [5], [6], [7], which may, even in small quantities, significantly affect the thermophysical properties [8], [9]. The thermophysical properties, in their turn, influence the depressurization and flow behaviour [10]. Transport of CO2 by pipeline is in operation for the purpose of enhanced oil recovery (EOR), mainly in the USA [11], [12]. The CCS case is likely to be different, due to different impurities, and the proximity to densely populated areas. Thus, there is a need to develop modelling tools which can aid in the safe and economical design and operation of CO2-transport pipelines.
Flow models for CO2 transport should be able to take a number of phenomena into account. As already alluded to, multiple chemical components need to be catered for. Already in a single-phase case, CO2 mixtures from different capture technologies will give different dynamic behaviour in pipeline transport [13]. Depending on the conditions, hydrates [14], or other solids, may form. Two-phase liquid–vapour flow may also occur, even if the pipeline is designed to be operated in the single-phase region. This may be due to varying CO2 supply [15], or during transient events, such as start-up, shut-in or depressurization [16], [17], [18]. During these events, among other things, it is important to be able to estimate the temperatures, since the construction materials may have a minimum temperature below which they begin to lose their toughness, e.g. the ductile–brittle transition temperature of steel. Furthermore, it is of great importance to be able to calculate the single-phase and two-phase (mixture) speed of sound. This is because a given pipeline filled with CO2 is more susceptible to running-ductile fracture than if filled with natural gas [2], [19], [20], and the running fracture is governed by a ‘race’ between the fracture velocity and the speed of sound.
The dispersion of CO2 resulting from a leakage [21], [22], [23], [24] is an input to risk assessments. To obtain realistic input boundary conditions to dispersion models, it is necessary to have good depressurization models for pipes and vessels.
Koornneef et al. [25] pointed out various knowledge gaps which affect the uncertainties of quantitative risk assessments for CO2 pipelines. However, the fact that CO2 pipelines have different challenges when compared to natural gas pipelines does not mean that CO2-pipeline networks will be associated with high risks. The study by Duncan and Wang [26] suggested a very small likelihood of having potentially lethal releases for CO2 from pipelines, assuming, among other things, that fracture propensity can be successfully mitigated.
Due to the large investments associated with offshore pipelines, transportation by ship may be a viable alternative due to its flexibility, especially in a start-up phase with relatively low CO2 volumes. Among the issues needing further attention, is the design of the offloading system, which also has to be compatible with the restrictions imposed by the storage site. Such considerations require modelling tools accurately representing the thermophysical properties of CO2 and CO2-rich mixtures, including the vapour–liquid phase boundary and the precipitation of solids. Further, the emptying or depressurization of vessels have similarities with the depressurization of pipelines.
Regarding the content of other substances (‘impurities’) in the CO2 to be transported, there appears to be at least two views. The first is that one should arrive at a ‘transport specification’ listing the maximum allowable content of impurities. The second is to perform knowledge-based optimization for each case. We believe that the latter approach may lead to a more efficient CCS system, preventing e.g. the oversizing of capture and conditioning plants. In the case of ship transport in particular, the liquefaction process should be optimized together with the capture process.
In view of the above, we want to review the state of the art with respect to data and models for transient two- and multiphase flow of CO2 and CO2-rich mixtures in CO2-transport systems. Emphasis is put on developments having taken place after the reviews by Aursand et al. [4], Li et al. [8], [9], Gernert and Span [27], or on relevant subjects not covered therein. We put our boundary conditions around the transport system itself, focusing on thermo- and fluid dynamics in, and out of, pipes and vessels.
Although it would lead too far to enter into details in this paper, it should not be forgotten that the accuracy of a simulation not only depends on the accuracy of the physical model, but also on the employed numerical method. Numerical methods for multiphase flow models is a subject where there are still challenges with respect to robustness, accuracy and efficiency. For instance, numerical diffusion can smear out the resolution of a depressurization wave in a pipeline [28], [29]. The numerical methods employed to solve for the thermophysical properties also need to be highly consistent, robust and efficient. This is particularly true in conjunction with CFD methods, where the thermophysical properties are needed in each computational cell at each time step.
The remainder of this article is organized as follows. In Section 2 we review the data situation for phase equilibria, density, speed of sound, viscosity and thermal conductivity of CCS-relevant CO2-rich mixtures. Section 3 deals with property models and briefly discusses implementation in fluid-dynamic models and flow through restrictions, which is relevant for decompression calculations. In Section 4, we review published data and models for transient multiphase flow of CO2-rich mixtures in pipes. In particular, we include pipe-depressurization case study. Section 5 considers ship transport of CO2. The study is concluded in Section 6.
Section snippets
Thermophysical property data
Knowledge of the relevant thermophysical properties of the relevant fluids is needed to optimize CO2 transport with respect to economy, operability and safety. A few examples will be provided in the following. We will then review the situation regarding thermophysical property data for CO2-rich CCS-relevant mixtures.
In pipeline transport it is usually desirable to have the fluids in dense phase, and hence knowledge of the vapour–liquid phase behaviour is essential [30], [31]. Accurate knowledge
Thermophysical property models
In this section, we briefly review models and methods for calculating the thermophysical properties of CO2 and CO2-rich mixtures. Highly relevant topics, such as implementation in fluid-dynamic models, and methods for calculating flow through restrictions, are also covered.
Pipeline transport of CO2
Some challenges related to CO2 transport in pipes, such as enlarged two-phase region and free water, increase with an increasing amount of impurities in the CO2. On the other hand, the removal of impurities at the capture plant entails increased investment and operational costs. This constitutes a techno-economic optimization problem which has to be considered for each specific project. Some of the data and models needed to perform detailed studies are yet lacking.
For CO2-transport pipelines
Ship transport of CO2
CO2 transport with ship is interesting in scenarios involving CO2 sources close to the coast and offshore storage. Since ships are more flexible than pipelines, ship transport may be preferable for early CCS deployment, where the CO2 quantities are small.
Conclusions
Although CO2 is transported in various ways today, the amount required for full-scale CCS implementation motivates the search for solutions being as safe and reliable as required and as efficient as possible. To do this, simulation tools handling the transient flow of single- and multiphase CO2 and CO2-rich mixtures inside, and out of, pipes and vessels are needed to perform calculations relevant for design, operation and safety. Today’s models are in need of improvement with respect to both
Acknowledgements
This publication has been produced with support from the BIGCCS Centre, performed under the Norwegian research programme Centres for Environment-friendly Energy Research (FME), and from the CO2Mix project in the CLIMIT research programme. The authors acknowledge the following partners for their contributions: Engie, Gassco, Shell, Statoil, TOTAL, and the Research Council of Norway (193816 and 200005).
The authors would like to thank Dr. Johannes Gernert and Professor Roland Span for sharing the
References (386)
- et al.
Optimization of pipeline transport for CO2 sequestration
Energy Convers Manage
(2006) - et al.
Pipeline transport of CO2 mixtures: models for transient simulation
Int J Greenh Gas Contr
(2013) - et al.
Dynamis CO2 quality recommendations
Int J Greenh Gas Contr
(2008) - et al.
PVTxy properties of CO2 mixtures relevant for CO2 capture, transport and storage: review of available experimental data and theoretical models
Appl Energy
(2011) - et al.
Viscosity, thermal conductivity and diffusion coefficients of CO2 mixtures: review of experimental data and theoretical models
Int J Greenh Gas Contr
(2011) - et al.
Thermo- and fluid-dynamical modelling of two-phase multi-component carbon dioxide mixtures
Int J Greenh Gas Contr
(2010) - et al.
Dynamic simulation of pipelines containing dense phase/supercritical CO2-rich mixtures for carbon capture and storage
Int J Greenh Gas Contr
(2012) Evaluating the risk of hydrate formation in CO2 pipelines under transient operation
Int J Greenh Gas Contr
(2013)- et al.
Modelling the impact of stream impurities on ductile fractures in CO2 pipelines
Chem Eng Sci
(2012) - et al.
Assessing the risk for CO2 transportation within CCS projects, CFD modelling
Int J Greenh Gas Contr
(2011)