Carbon Capture Systems Analysis: Comparing Exergy Efficiency and Cost of Electricity of Existing and Future Technology Options

Given the likelihood that carbon-based fuels will continue to play a major role in electricity generation, carbon capture and storage systems will be a necessary part of future generating capacity. With this research project, we aim to develop a framework for quantitative evaluation of existing and proposed carbon-capture technologies. This framework is based on an exergy-based systems analysis performed at the local (system) level and at the global (life cycle) level, and a techno-economic evaluation of the technology. At the system level, work to date has focused on three tasks, all related to adsorptionbased systems: (1) the derivation of a thermodynamically rigorous definition of the exergy of any adsorbed mixture of species, including non-ideal mixtures that may include water; (2) building a set of property methods for adsorbed phases, so that thermodynamic information can be quickly accessed and used for modeling; and (3) performing an exergy analysis of a pressure swing adsorption system. At the life cycle level, work to date has focused on two tasks: (1) the development of custom-written Matlab code for exergetic life cycle calculations, using data from the Ecoinvent Data v3 life cycle assessment database; and (2) applying this code to models of two fossil-fuel based power plants, with and without an amine absorption carbon capture system, to demonstrate the functionality of the code. Overall then, the work to date has involved building robust and re-usable tools to enable the localand global-scale exergy analysis of carbon capture systems. Introduction With this research project, we seek to answer the set of questions: What technologies or combination of technologies can provide CO2 capture at energy costs much closer to underlying thermodynamic limits? Are the conditions and technology configurations required to achieve such efficiencies possible with available or foreseeable materials? What are the broader (i.e., full system) environmental implications of such technology implementations? What are the likely costs of these technologies? And, given all of the above considerations, which technologies are likely to supply practical, cost-competitive electricity with low life-cycle CO2 emissions? We are approaching these questions by focusing on four types of carbon capture and storage (CCS) systems with different characteristics, ensuring that a representative range of technologies are studied. Each of the four CCS technologies will be evaluated on three bases: (1) an exergetic systems analysis so that the evaluation takes place with respect to absolute, thermodynamic limits; (2) a full life-cycle basis such that all ancillary exergy consumption and destruction—both embodied components and in-process interactions with the environment—are included; and (3) a techno-economic analysis of the technology such that technology options can be evaluated via economic metrics as well. The systems include one baseline option—namely, a MEA/NGCC reference system such as the one studied by NETL [1]—for calibration of methodology, and three advanced technologies. The advanced technologies were chosen in conjunction with GCEP to represent the range of CCS options, and to complement the work of other GCEP research groups. They include one metal-organic framework (MOF) based adsorption system, one biomimetic sorbent, and one ionic liquids absorption system. Our three-part analysis will result in three metrics for each system: work-specific exergy consumption, work-specific carbon emissions to the atmosphere, and first-order estimates of levelized cost of electricity (where the work is the output of the associated power plant, as reduced by the added carbon-capture system). Each of these metrics can be defined on a local (system-only) and global (life cycle) basis. In order to arrive at these metrics, we will be generating exergy distribution plots of each system. These are plots which show the magnitude of exergy destruction in each sub-process in the system, which can both help us understand the inherent inefficiencies in the system and, in the future, to target the most inefficient sub-systems for improvement. The end goal of this project is not only to provide accurate analyses of selected carbon-capture options, but to frame the comparisons of these options in a coherent, compelling way. Background Despite growing concerns about climate change, it is very likely that carbon-based fuels will continue to be used in the coming decades for base-load power generation and for firming of intermittent renewable power sources. Carbon dioxide capture and storage (CCS) is therefore a necessary part of any comprehensive strategy for achieving required reductions in carbon emissions to the atmosphere. Significant studies of proposed carbon capture systems have been performed, including those produced as part of the IPCC process [2], those produced by the U.S. DOE [1], those from the IEA [3], and numerous academic and industrial analyses [4], [5], [6]. These studies arrive at the common conclusion that existing options for post-combustion carbon capture lead to a significant loss of plant output (as much as 30%) and a significant rise in cost of electricity (as much as 80%) when compared to a similar plant without carbon capture. These numbers might be deemed acceptable—perhaps even unavoidable—were it not for the fact that thermodynamic limits suggest that much more efficient processes are possible. Considered in terms of the minimum work required to separate carbon dioxide generated by combustion of natural gas from its flue gas at atmospheric pressure, only a 2.4% loss of plant output is actually required [7]. Note that even after compressing the CO2 to 100 bar (the pressure prescribed by NETL for use in comparing separation options, [1]), the fraction of exergy required has only risen to 3.5%. These figures illustrate a key point: Assuming an exergy efficiency of ~50% for the overall plant, it should be possible to develop a separation methodology that meets NETL specifications while incurring only a 7% loss of plant output. This leads to a need to analyze the performance of carbon capture systems via an exergy analysis, in order to compare their operation to this thermodynamic baseline. The objective of this type of analysis is to understand the distribution of exergy within a system such that the causes of its destruction are exposed and quantified. With the exergy distribution of a plant or process known, not only is its overall quality assessed with respect to a fundamental, absolute yardstick, but the locations and magnitudes of losses are revealed for inspection (and thereby improvement). Exergy analysis provides both an unambiguous assessment and immediately points the way to process improvement. For two reasons, we chose to begin our systems analysis with an adsorption system. The first reason is that two of our advanced carbon capture systems are adsorption-based (the MOF system and the biomimetic sorbent system). The second reason, which became clearer as we advanced in the project, is that very little had been done previously in developing the tools necessary for performing exergy analyses of adsorption-based gas separation systems (whereas absorption-based systems have been studied more extensively). Therefore, our progress on this topic has consisted of two complementary efforts: the development of a generalized and thermodynamically rigorous understanding of the exergy of adsorbed species and mixtures, and the application of this new understanding to an adsorption system model for CO2 capture from flue gas. Our research takes the analysis beyond the local system level by incorporating global effects in the comparison of carbon capture systems. Using results from the exergetic systems analysis, we are conducting a life-cycle assessment (LCA) in order to better understand how a given CCS technology interacts with the environment at broad scales. This assessment includes not only direct energy consumption (i.e., exergy consumption) at a facility, but also the exergy destruction associated with all “embodied” materials and components, as well as the operational interactions with the environment. The goal of this analysis is to understand which technology option results in the smallest disruption to the environment and results in the smallest drawdown of valuable energy and material resources (e.g., exergy stores) across the entire supply chain. This will enable a better understanding of the tradeoffs between using sophisticated material inputs (which often require more embodied energy inputs) and reducing energy impacts on site. A number of LCAs have been performed of CCS technologies [8], [9], [10]. These LCAs serve to situate a CCS technology within the broader industrial ecosystem, tracking all material and energy flows both upstream and downstream of a technology of interest. In our case, this set of material and energy flows is utilized in an exergetic approach to LCA [11], [12], [13]. Several professional LCA datasets exist, such as the GaBi databases or the U.S. LCI database. After an overview of the available options, we chose to purchase the Ecoinvent Data life cycle assessment database, commissioned by the Swiss Centre for Life Cycle Inventories. The selection was based largely on the 1 For example, the exergy efficiency of a modern (F-class) NGCC power plant is ~52%. accessible nature of the raw database files, from which we could construct and manipulate individual matrices and perform exergetic life-cycle assessments without the utilization of a third-party LCA software package. Results Exergetic Systems Analysis: Exergy of Adsorbed Mixtures In order to perform a detailed exergy analysis of an adsorption-based carbon capture system, we needed to be able to define the exergy of adsorbed species. This required a firm grounding in the thermodynamics of adsorbed species and mixtures in general, and of adsorbed mixtures of CO2, N2, and water in particular. To our knowledge, the only previous attempt to characterize the exergy of adsorbed phases was done by Kearns and Webley in 2004 [14]. In their paper, they arrive at an expression for the exergy of a combined system of a binary gas mixture, a given mass of sorbent, and a binary adsorbed mixture in equilibrium with the gas phase. Although they did not do this, the exergy of the adsorbed phase alone could then be isolated by subtracting the exergy of the other parts of the system (the sorbent and the gas phase). The two key underlying assumptions in their derivation are (1) that the adsorbed species are at very low surface coverage, such that there is a linear relationship between gas phase pressure and the amount adsorbed, and (2) that all gases, both in the system and in the environment, behave as ideal gases with constant specific heats. The first assumption is almost never valid for adsorption systems that separate CO2 from N2; in most regions of interest, the CO2 adsorption follows a Langmuir isotherm form, not a linear one. This tool, then, wasn’t useful for evaluating carbon capture systems. The second assumption is often valid, but it isn’t necessary—because exergy is a state property of a substance, we should be able to define it without needing to impose an equation of state on the environment outside the system. In keeping with our mission of providing a set of tools for evaluating any new adsorption-based carbon capture technology, we set out to derive an expression for the exergy of adsorbed phases that could be applied completely generally. First, we solved for the maximum useful work possible if an adsorbed phase were to be equilibrated with the environment—i.e., we found the exergy of this phase—while imposing only that matter and energy must both be conserved, and that transfers of heat, work, and matter should all be allowed between the system and the environment during equilibration. We arrived at an expression that imposed no equation of state on any part of the system or environment, and is therefore completely general: