Holistic Assessment of Carbon Capture and Utilization Value Chains

Abstract

Carbon capture and utilization (CCU) is recognized by the European Union, along with carbon, capture and storage (CCS), as one of the main tools towards global warming mitigation. It has, thus, been extensively studied by various researchers around the world. The majority of the papers published so far focus on the individual stages of a CCU value chain (carbon capture, separation, purification, transportation, and transformation/utilization). However, a holistic approach, taking into account the matching and the interaction between these stages, is also necessary in order to optimize and develop technically and economically feasible CCU value chains. The objective of this contribution is to present the most important studies that are related to the individual stages of CCU and to perform a critical review of the major existing methods, algorithms and tools that focus on the simulation or optimization of CCU value chains. The key research gaps will be identified and examined in order to lay the foundation for the development of a methodology towards the holistic assessment of CCU value chains.

1. Introduction

Effects such as the melting of glaciers, rise in sea level, forests dying, and climate change are indications of global warming, which is caused by the release of heat trapping gases (greenhouse gases) to the atmosphere, with carbon dioxide (CO₂) being by far the major contributor. The CO₂ atmospheric levels have been continuously increasing between 1950 and 2017, have now reached at approximately 400 ppm the highest CO₂ level ever recorded [1], and they are expected to continue increasing if people and businesses keep burning fossil fuels at the same rates as now.

In order to respond to these environmental pressures, in 2011 the European Union (EU) announced that the main target set in the Roadmap for 2050 is to gradually reduce the overall levels of the Greenhouse Gas Emissions (GHG) by 80% below the 1990 values (40% by 2030 and 60% by 2040) [2]. Towards that end, the European Commission’s 2011 Energy Roadmap has identified the three main pillars for supporting the transition to a low carbon economy: (a) wider implementation of renewable energy sources, (b) promotion of low carbon energy supply options (e.g., nuclear power or conventional fossil fuels supported by carbon capture), and (c) implementation of energy saving measures [3].

The major mechanisms put in place by EU to ensure that the targets will be met are (a) the EU Emissions Trading System (EU-ETS) which enables the trading of emission rights and covers 45% of EU’s greenhouse gas emissions coming from large-scale facilities in the power and industry sectors and (b) the national emissions reduction targets, which cover 55% of EU’s greenhouse gas emissions coming from houses, agriculture, waste, and transport. Renewable energy targets are promoted through the Renewable Energy Directive [4], designed to reflect each country’s starting point. Similarly, energy efficiency targets can be reached by following the Energy Efficiency Plan and Energy Efficiency Directive [5,6].

Achieving the milestones set means that (a) EU’s energy security will improve by the reduced dependence on imported energy sources making Europe more competitive; (b) advanced green growth will be effectively promoted; and, (c) more jobs will be created. The governance process will be improved as targets are met and the Energy Union is promoted as a cohesive system [7].

1.1. Carbon Capture as a Mitigation Option

As previously mentioned, carbon capture, either in the form of carbon capture and storage (CCS) or carbon capture and utilization (CCU), is one of the alternative instruments to achieve the emissions reduction targets, since energy production still relies heavily on fossil fuels. In addition, there are other industrial sources of CO₂ emissions other than power generation, which cannot be mitigated by using more renewable energy sources or improving efficiencies. This means that CCS and CCU are able to decrease CO₂ emissions in areas where the other measures have minor or no application.

CCS is the process of capturing and storing anthropogenic CO₂ emissions underground as a way of reducing the amount of CO₂ that is released in the atmosphere. CO₂ can be transported via pipelines and stored in deep saline formations, coalbeds, mature and depleted oil or gas fields. Over the last 40 years, during enhanced oil recovery operation, CO₂ was being stored underground as a consequence [8]. Although such operations have traditionally focused on optimizing oil production and not the storage per se, technologies that are used in EOR have facilitated the development and the wider implementation of CCS, especially in terms of CO₂ transport and injection. Similarly, carbon capture and utilization shares the same initial steps, but instead of storing CO₂ underground, CO₂ is converted into commercial products, so additional research is required to render the whole chain technically and economically feasible [9].

1.2. Carbon, Capture and Utilization

A CCU value chain can be described by six major steps (Figure 1). The first step is the carbon source characterization, including data, such as its location, the CO₂ output flowrate, the CO₂ purity, and the type of output stream. The second step is the CO₂ capture process, where CO₂ is separated from the output stream using an appropriate technology based on the type of stream. Following the separation, purification, and compression take place based on the type of transportation and the required purity of the receiver. The fifth step is the transportation of CO₂ to the receiver. The type of transportation is determined by the stream characteristics (i.e., flowrate, purity), the distance between source and sink, and other regional characteristics. The final step is utilization; where CO₂ is delivered to the receiver at the required purity and flowrate and converted to the final product.

The technical characteristics of all six steps have been extensively studied in the literature. However, in order to be able to identify and eventually develop a viable CCU value chain, the economic characteristics of each component should be also analyzed. Moreover, the value chain should be assessed as a whole, by adopting a systemic approach and taking into account the interdependencies among its different components. The objective of this paper is to present the most important and relevant literature for these three aspects of developing a CCU value chain (technical characteristics, economic characteristics, value chain modelling), identify the existing research gaps, and lay the foundation for the development of a novel methodological framework for the holistic assessment of CCU value chains.

2. Technical Characteristics of a CCU Value Chain

2.1. Carbon Source Characterization

An initial attempt to characterize the carbon sources was performed by IPPC [10] through presenting the profile (purity and available quantity) of the major worldwide large CO₂ stationary sources. Patricio et al. [11] presented an extensive list of CO₂ sources and linked them to the type of industrial activity and the corresponding NACE code that these emissions might be produced by. They also listed the potential impurities in each stream as well as the typical magnitude of flow. Zakkour and Cook [12] emphasized on the high purity sources (>80% CO₂), and briefly summarized the level of industrial activity in each sector. Fennel et al. [13] have also presented a detailed list of industrial and power generation sectors, focusing on the CO₂ partial pressure and the cost thatis associated with its removal from the gas stream.

Regarding sources classification based on purity, there are no nationally or internationally agreed boundaries. Jin et al. [14] have classified CO₂ sources in four different categories (High > 90%, Secondary High 50–90%, Moderate 20–50%, and Low < 20%). The Centre for Low Carbon Futures [15] has also classified CO₂ sources in four categories (High > 90%, Secondary Highest: 50–90%, Moderate Level: 20–50%, Low < 20%) based on the impact of the CO₂ concentration on the energy requirements and the corresponding cost for separating CO₂ from the gas stream, whereas Angelis-Dimakis and Castillo [16] have proposed three classification bins (High > 90%, Moderate 20–90%, and Low < 20%). The overlapping of CO₂ in the proposed categories and the inconsistencies among researchers makes the purpose of the categorization of sources and using the categories to optimize CCU value chains unclear and not very useful.

This short literature account showed that sources have distinctive characteristics in the way that CO₂ is produced and it can be further categorized into three new categories that facilitate the matching with carbon capture technologies [17]. The three proposed categories are (a) high purity CO₂ streams that come from sources that directly produce an output stream of 96–100% CO₂ purity, (b) CO₂ from syngas production, and (c) CO₂ produced from the combustion of fuels.

2.2. Carbon Capture

Carbon capture is the most extensively studied component of the CCU value chain, especially since it is also a stage common to CCS. The objective of this paper is not to perform a detailed literature review, as there are many available either with a more chemical engineering approach [9,10,18,19] or from an industrial/commercial perspective [15,20].

There are also several research papers that provide a list of all possible CO₂ capture options along with detailed process descriptions of each one [9,11,21,22,23,24]. Our focus will be to identify the most widely used technologies and to emphasize the characteristics that can be related to the carbon sources towards the implementation of a CCU value chain.

Carbon capture technologies are classified in terms of CO₂ separation from the stream in the following three categories [9,22,25,26,27,28,29]:

• Post-combustion carbon capture, when separation takes place after combustion. This process involves CO₂ capture from the exhaust gases after the combustion of fossil fuels. The most common separation technique in this category is separation by chemical absorption.
• Pre-combustion carbon capture, when separation takes place before combustion. This process involves CO₂ capture from streams containing H₂ and CO₂ that usually result from natural gas reforming and the produced syngas reacts with H₂ through a water gas shift reaction to produce the mixture of H₂ and CO₂. The most common separation technique in this category would be separation by physical sorbents.
• Oxy-fuel combustion carbon capture, when the combustion is performed while using pure oxygen. This is essentially the same with post combustion carbon capture, but in this case, a lot of impurities can be avoided due to oxygen being pure (21% of air is oxygen), therefore the product would be a high purity CO₂ stream.

Spigarelli and Kawatra [21] also classified the carbon capture technologies in high, medium, and low pressure technologies, in terms of CO₂ partial pressure, as expressed by the share of the carbon dioxide in the flue gas stream (30–70%, 35%, and 3–20%, respectively). However, these classification schemes do not facilitate the identification, formulation, and optimization of the value chains, since the objective is to be able to match the CO₂ sources with carbon capture technologies that are compatible and process the output stream of the source efficiently. The categorization based on the CO₂ partial pressure makes matching difficult, since both the categorization of sources based on purity and carbon capture technologies are unclear. For that purpose, three new categories are proposed in terms of the type of stream that a technology can process and the commercially applicable technologies are matched to the carbon sources categories presented in Section 2.2, based on already published work. These three categories are: (a) general gas conditioning, including the removal of common impurities, such as NO_x, SO_x, dust, and water; (b) absorption with chemical solvents or combustion using pure oxygen; and, (c) absorption with physical solvents. The results are presented in

Table 1. Matching carbon sources and carbon separation technologies (with information from [9,11,12,30,31,32]).

	Source CO2 Purity
Technique	High CO2 %	Medium CO2 %	Low CO2 %
General Gas Conditioning	Fermentation processes
	Beer (100%)
	Wine (100%)
	Bioethanol (100%)
	Biogas production (99%)
	Refineries gas sweetening (96–99%)
Absorption with Chemical Solvents or Combustion using pure oxygen		Iron and steel Industries	Pulp industry
			Recovery boiler (13%)
		Corex (30%)	Paper industry
			Energy production (13%)
		TGRBF (22–38%)	Power generation
			Oil fired boilers (11–13%)
		Blast furnace (22%)	Coal fired boilers (11%)
			Natural gas fired boilers (7–10%)
		Lime production	Incineration of waste (10%)
			Glass industries (10%)
		Combustion of fuels in lime kilns (24–32%)	Aluminium production (3–10%)
			Polyethylene production
		Cement industries	Cracker furnaces (5%)
			Gas turbines (3–4%)
		Cement kilns (22%)	Carbon black manufacturing (2–5%)
		Oil refineries (8–24%)	Brick dryers and kilns (1.5–4%)
Absorption with Physical Solvents	Hydrogen production (100%)		Integrated gasification combined cycles (12–14%)
	Ammonia production
	Haber-Bosch process (30–99%)
	Ethylene oxide production (30–100%)
	Natural gas processing (2–70%)

, where each source is sorted according to purity and is matched to a separation technology.

2.3. Carbon Purification

Purification in the context of carbon capture and utilization includes the separation of CO₂ during carbon capture, followed by dehydration and compression that prepares the gas stream, for transportation and utilization [33,34,35,36]. It is an important step, as it can affect the downstream components of the value chain (its technical characteristics, design parameters, and, by extension, the cost). Wetenhall [37] and Ivan [38] identified and analyzed the effect of impurities on equipment during transportation such as corrosion, altering of physical properties, phase behavior, density and viscosity, which, in extent, affect transportation conditions and design parameters. Their work showed that the addition of impurities to the mixture increases the required pressure (critical pressure) to keep the mixture in the supercritical phase (required phase for transportation), the addition of H₂S, SO₂, and NO₂ increases viscosity and decreases density compromising pipeline capacity and the presence of H₂S, SO₃, and NO₂, but mainly water causes corrosion.

Since some sources have relatively high CO₂ purity, a dilemma arises as to if it is necessary to purify the CO₂ stream. In some occasions, where receivers do not require high CO₂ purity, it might be acceptable to avoid the purification step; however, although capital costs would decrease because of less equipment required, in the long term, the running costs will increase because of damage that is caused to equipment from corrosion and higher required compression due to changes in phase behavior. On the other hand, including purification in all occasions would significantly increase capital costs, but running costs would be much lower. In addition, all streams would have high CO₂ purity, thus satisfying all receivers, and facilitating the creation of a CCU hub by mixing all streams and centrally distributing CO₂. Thus, purification is recommended as a mandatory step of all CCU value chains, independent of the purity specifications of the source and the receiver.

The level of purification that needs to be achieved mainly depends on the technical limitation of the equipment that is involved in conjunction with health and safety considerations. The “Dynamis” project, presented recommendations for the composition limits of gas streams containing mainly CO₂ for CCS, primarily from a transport point of view and to a certain extent, for storage [39]. The study examined oxyfuel, pre- and post-combustion electricity production processes utilising different types of coal and Natural Gas Combined Cycle Gas Turbines to cover a wide range of possible impurities [32]. After considering the safety and toxicity of impurities, avoidance of free water formation, avoidance of hydrate formation, avoidance of corrosion, reduction of CO₂ volume, a list of recommendations for the concentrations of various impurities was compiled [39]. Based on that report, the maximum concentrations of some of the key impurities are H₂O 500 ppm, H₂S 200 ppm, SO_x 100 ppm, NO_x 100 ppm, and the minimum concentration of CO₂ is 95.5%.

2.4. Carbon Transportation

Transportation of captured CO₂ can be achieved either by using pipelines, truck tankers, or railroad tankers (inland transportation), or by using ships (sea transportation). Each type has its own advantages and disadvantages, and if used effectively, it can contribute to the optimization of CCU value chains. Inland transportation using pipelines is the most favored type of transportation, because it can handle large flowrates effectively, but it becomes an economically challenging method for low flowrates. On the contrary, truck tankers and railroad tankers are more useful for handling small quantities and they have lower capital costs.

Pipeline transportation takes place in the liquid or supercritical state at pressures above 50 and 80 MPa respectively and temperatures around 0 °C [38]. When CO₂ is transported by ship, it is usually in the liquid phase (on the saturation line, with pressure a few bars higher than the atmospheric and temperature a few degrees below the ambient). Aspelund et al. [40] have concluded that the most efficient way to transport CO₂ is as a cryogenic liquid and the optimum conditions are 6.5 bar and −51.2 °C. When CO₂ is transported via truck tankers, it should be in liquid state at 1.7 MPa pressure and −30 °C [41]. Although the volumes are very small when compared to what could be required, cryogenic vessels of the size of 2 to 30 tonnes are available to suit customer requirements. CO₂ transportation by railroad tankers takes place in liquid state at 2.6 MPa pressure in specially developed tanks [41].

2.5. Carbon Utilization

The final step of a CCU value chain is carbon utilization. At this stage, captured CO₂ that has been purified to the desired level arrives at the receiver. The main characteristics that need to be specified for each receiver in the context of developing, implementing and optimizing a CCU value chain are two: (a) the minimum CO₂ purity and (b) the magnitude of flow that is required by each receiver. Additional information such as upper limits in the composition of certain impurities found in captured CO₂ that are not compatible with the process of each receiver and modifications that might be necessary for the receiver to accept captured CO_2, can affect the optimization decisions.

An extensive list of current and future receivers was presented by the Global CCS Institute [42]. The end-uses can be classified in four different categories: (a) direct use, (b) mineral carbonation, (c) fuels production, and (d) chemicals production. Patricio et al. [11] have compiled a very detailed table with the required magnitude of flow of all the existing and emerging CO₂ uses, as well as further quantitative characteristics and the potential collaborators from an industrial symbiosis point of view. In this paper, the previously published table is further enhanced with a few more carbon receivers, as well as qualitative values for the purity requirements of each receiver (

Table 2. Quality requirements of existing and emerging carbon dioxide (CO₂) uses.

CO2 Receiver	Required Purity	TRL [11]	Reference
Direct Use
Beverage carbonation	High > 99.0%	9	[43]
Wine making	High > 99.0%	9	[43]
Food processing, preservation and packaging	High > 99.0%	9	[43]
Coffee decaffeination	High > 99.0%	9	[43]
Supercritical CO2 as a solvent	High > 99.0%	9	[43]
Steel manufacture	High	9	[53]
Metal working	High	9	[44]
Welding	High > 99.8%		[46]
Pulp and paper processing	High		[47]
Water treatment	High > 99%		[43]
Electronics	Supercritical phase		[48]
Power generation (as power cycle working fluid)			[42]
Pneumatics		9	[45]
Fire suppression technology	High > 99.5%	9
Refrigerant gas	High > 95.5%	9	[43]
Horticulture	High	9	[45]
Pharmaceutical processes	High > 99.5%		[43]
Mineral Carbonation
Calcium carbonate and magnesium carbonate	Medium-Low	7–8	[42]
CO2 concrete curing	Medium-Low	7–8	[42]
Baking soda (sodium bicarbonate)	Medium-Low		[42]
Bauxite residue treatment (“red mud”)	High	7–9	[42]
Fuels Production
Renewable methanol	High	7–8	[42]
Formic acid	High		[42]
Algae cultivation	Medium-Low	4–7	[42]
CO2 injection to conventional methanol synthesis	High		[42]
Genetically engineered micro-organisms	High > 99.5%		[45]
Enhanced oil recovery (EOR)	High 95.0%	9	[45]
Enhanced coal bed methane recovery (ECBM)	High-Medium-low	7	[42]
Chemicals Production
Urea yield boosting	High	9	[42]
Polymer processing	High	7	[42]
Chemical synthesis (excludes polymers and fuels)	High	6–8	[54]

TRL: Technology Readiness Level.

). The following sections give a few more details.

2.5.1. Direct Use

Carbon dioxide can be used in various processes in the food industry, including cooling while grinding powders, as an inert to prevent food spoilage and in packaging applications, when used both in modified and controlled atmosphere packaging. The specifications in these applications are such that purity must be greater than 99%, moisture content less than 52 vppm, CO less than 10 vppm, total hydrocarbons less than 50 vppm, oil less than 10 ppmw, and pass a test for acidity and red substances [43]. CO₂ is also used in beverage carbonation [44], in coffee decaffeination as an extraction solvent (supercritical CO₂) [45], and wine making, as a seal gas to prevent oxidation of the wine during maturation [42].

Another important consumer of CO₂ is metal industry, where CO₂ can be used for chilling parts for shrink fitting, hardening of sand cores, and moulds, as a shrouding gas to prevent oxidation in a welding process, or in a minority of basic oxygen furnaces as a bottom stirring agent or for dust suppression [46]. Other industrial direct uses of CO₂, include pulp and paper processing (where it is used to reduce pH during pulp washing operations) [47], water treatment (where it is used for the re-mineralization of water following reverse osmosis and for pH control) [43], printed circuit board manufacture (as a cleaning fluid) [48], and power generation (in the supercritical form as a power cycle working fluid) [42]. It can be also used in pneumatic applications, in fire extinguishers and in industrial fire protection systems [46], and in refrigeration units (especially larger industrial air conditioning and refrigeration systems) [49] as the working fluid, replacing other more toxic refrigerant gases [50]. Finally, carbon dioxide is considered as a key instrument in commercial greenhouses towards increasing production. A usual practice is to utilize CO₂ in the form of exhaust gases from the heating system/combined heat and power generator of the greenhouse or to use pure liquid CO₂ [51].

2.5.2. Mineral Carbonation

Mildly concentrated CO₂ is contacted with a mineral-loaded alkaline waste stream (either solid or liquid) and it precipitates out as mineral carbonate (limestone/dolomite equivalent precipitates). The resulting product can be further processed to form an aggregate equivalent product for the construction industry. There are numerous sources of industrial waste that can be used in the mineral carbonation process, such as: metallurgic slags, incineration ashes, mining tailings, asbestos containing materials, and oil shale processing materials [52]. There are several variants of mineralization, such as baking soda production (where CO₂ is contacted with sodium rich brine), concrete curing (where a flue gas stream is used in the cement curing process in order to sequester CO₂ in manufactured concrete products), and bauxite residue (“red mud”) treatment (where CO₂ is added to the highly alkaline bauxite residue slurry) [42].

2.5.3. Fuels Production

A variety of fuels can be produced while using a captured CO₂ stream as feedstock. CO₂ methanol can be produced using a concentrated CO₂ source and H₂ as feedstock. The electrolysis of water produces H₂, which can be then mixed with CO₂ and compressed [44]. Similarly, formic acid can be produced through electrochemical reduction of CO₂, by combining a concentrated stream of captured CO₂ and water [55]. Moreover, power-to-gas technology involves the transformation of electricity to synthetic natural gas, using the surplus electricity from renewable power sources in order to produce methane. A completely different application is the use of CO₂ as a nutrient/external carbon source in algae cultivation, which can then be used for a wide variety of products, including biofuels and other chemicals [56].

2.5.4. Chemicals Production

Carbon and oxygen are both key elements in organic chemistry. Consequently, there is a wide range of chemicals that can at least theoretically use CO₂ as a feedstock for production, including organic acids, alcohols, esters, and sugars. The practicality of CO₂ as a feedstock will vary significantly based on the current production routes. The dominant potential demand, which is based on current markets, could come from acetic acid, which has a current global market of ~6 Mtpa and it can be produced by direct catalysis of CO₂ and methane [42]. Urea can be also produced when carbon dioxide reacts with ammonia. The reaction is catalyst free and there are no requirements for CO₂ quality [57]. Urea production using CO₂ is a mature, but, at the same time, saturated market.

Finally, a more recent synthesis that has been examined is the combination of captured CO₂ with traditional feedstocks in order to produce polymers [42]. More specifically, the use of CO₂ as feedstock for production of polycarbonates, such as polypropylene carbonate (PPC) and polyethylene carbonate (PEC), or as a feedstock in the polymerization of urethanes, to produce polyurethanes, has been examined and piloted by several companies.

3. Economic Characteristics of a CCU Value Chain

The economic characteristics of the individual components of a CCU value chain have not been extensively studied. As it can be seen from the analysis that follows, most of the studies have been performed in the context of CCS (rather than CCU), and thus the majority of them focus on carbon capture (primarily) and carbon transportation. However, there are no studies estimating the potential costs that are linked with carbon utilization (due to either retrofitting/modifying an existing plant in order to receive recycled CO₂ or installing a new plant).

3.1. Carbon Capture

The carbon separation costs are usually presented in literature either per type of source or per type of technology used (i.e., pre-combustion, post-combustion, and oxy-fuel combustion). The separation costs are further divided to fuel, operating and maintenance, and capital costs, and are usually expressed in terms of levelized cost of capture and cost of CO₂ avoidance. In the case of power plants, the costs are also usually presented in terms of levelized cost of electricity.

Kuramochi et al. [58] presented a review of economic studies and comparison of carbon capture technologies for the major industrial sectors. According to their analysis, it is not easy to identify which technology will be widely implemented in the future and suggested that a combination of industrial plants and power plants would lead more easily to economically viable carbon separation projects. The reported costs for the iron and steel sector vary between 40–65 €/tCO₂ (depending on the process and the capture technology). For the cement sector, the selected technology was post-combustion capture with reported costs above 65 €/tCO₂ avoided. For the petroleum industry, the selected technology was oxyfuel capture with reported costs at 50–60 €/tCO₂.

Leeson et al. [49] have performed a similar and more recent review of the main industrial sectors, as well as high-purity sources of CO₂, but with an emphasis on CCS, and the outputs vary slightly when compared to [58]. The reported costs for the iron and steel industry range from $50 to 90/tCO₂ avoided (using post combustion chemical absorption), for the petroleum refineries have an average value of $98/tCO₂ avoided, and for the cement industry, a mean cost of $39.4/tCO₂ avoided using oxy-combustion with calcium looping, which showed better results when compared to post-combustion amine scrubbing. For the high purity sources, the mean values varied from $10 to $41.6/tCO₂ avoided based on the actual process examined. Although the analysis was focusing on CCS deployment, the main conclusions can be also applied in CCU.

Kuramochi et al. [50] have also reviewed and compared the techno-economic performance of carbon capture technologies that are applied in alternative distributed energy systems (including all of the stages involved, such as capture, compression, distributed transport using a pipeline) and took into account all the different parameters (fuel type, energy plant type, capacity factor, energy prices, and interest rate). The reported costs ranged from 30 to 150 €/tCO₂ avoided depending on the capacity of the power plant (with an estimate that these values might drop between 4–9% in the mid and long term). Finally, Element Energy [59] presented a detailed techno economic study for retrofit deployment of different CO₂ capture technologies in the same four sectors, with a UK focus, reporting an average cost of 60 €/tCO₂, 75 €/tCO₂, 65 €/tCO₂, and 10 €/tCO₂ for the iron and steel, cement, petroleum industries, and high purity sources, respectively.

Apart from the above-mentioned review papers and reports, there have been a few more published papers that emphasize on one type of carbon source. For the cement industry, Namzul Hassan et al. [60] performed a techno-economic study for the installation of a post-combustion carbon capture in an existing cement plant, estimating the average cost at $51/tCO₂ using chemical absorption. Barker et al. [61] reviewed all the possible carbon capture technologies that could be applied in a cement industry and estimated the relevant costs at 40 €/tCO₂ avoided by using oxyfuel technology, and substantially higher (107 €/tCO₂) while using post combustion capture for a flow of 1 MtCO₂ per year. An overview of the estimated collected capture cost for the cement industry and the corresponding flow rate of flue gas stream are presented in Figure 2. A similar figure (Figure 3) can be also drawn for the high purity sources, using data from [62,63], although it should be mentioned that the process is not the same but only the output is a high purity (>95%) CO₂ stream in all nine cases.

For the other important industrial CO₂ sources, there were not enough published papers that would allow the authors to perform a similar analysis. For the power plants, the situation is slightly different, since there are various types of plants and not enough data for each one of them. In addition, the range of possible CO₂ flow rate is much bigger, as it can exceed 10 MtCO₂/year. Moreover, the reported costs are often presented in terms of levelized cost of electricity (per MWh of electricity produced), making the comparison more difficult. However, a few indicative papers are mentioned to showcase the range in the cost of CO₂ capture.

Kolstad and Young [64] presented a report dedicated on assessing costs of various carbon capture technologies used for emissions in a brown coal fired power plant in Australia, in the context of the three technologies proposed by literature, (i.e., post-combustion, pre-combustion and oxy-fuel combustion). Various alternative configurations are proposed, data for retrofitting costs, pipeline transport, and storage are included in their analysis and the estimated capture cost varies from $30 to $90/tCO₂ avoided. Borgert and Rubin [65] presented a model for oxyfuel coal power plants and calculated the CO₂ avoidance cost (ranging from $45 to $75 per tCO₂ avoided) and the effects on transport costs with varying outlet CO₂ purity and O₂ purity.

Davison et al. [63] studied power plants, as well as hydrogen co-production plants, and estimated the total plant cost, total capital requirement, levelized cost of electricity and CO₂ avoidance costs. The reported costs vary between 60–65 €/tCO₂ for pulverized coal plants and between 95–100 €/tCO₂ for IGCC plants. Rubin et al. [66] summarized various published studies regarding the cost of post, pre, and oxy-combustion carbon capture technologies applied on different types of power plants. The estimated capture cost ranges from $58–121/tCO₂ for post combustion in NGCC power plants, $28–41/tCO₂ for pre-combustion in IGCC power plants using bituminous coal and $42–87/tCO₂ for pre-combustion in IGCC power plants using supercritical pulverized coal.

Based on our review, a major gap in the literature review is a comparative analysis of all the existing studies in order to estimate the capture cost as a function of the industrial source, stream purity, and the magnitude of flow.

3.2. Carbon Transportation

IPCC [10] has initially attempted to estimate the cost of various transportation options in in the context of captured carbon dioxide. Onshore and offshore pipelines, as well as ship transportation, are included in the analysis, which is complemented by cost information regarding the design of CO₂ ship carriers, the required liquefaction facilities and an overall comparison of all three options. It has been estimated that the transportation cost may vary between $1–5/tCO₂ for every 250 km, depending on the flow rate of the pipeline. Kuramochi et al. [58] have reported slightly higher values. The total cost, including compression and transportation, has been estimated to approximately 10 €/tCO₂ transported for short distances and large quantities (<500 tCO₂/day) and may increase to more than 50 €/tCO₂ transported for low quantities. Borgert and Rubin [64] have estimated the transportation cost for an oxy-fuel power plant at approximately $2.5/tCO₂/150 km. Similarly, Morgan and Grant [67] have reported a cost for onshore pipelined of about $1–3/tCO₂/160 km.

Zero Emissions Platform [31] has produced two different studies regarding the cost of transporting CO₂. In the first one, performed as part of CCS demonstration projects, the transportation cost of captured CO₂ from coal or natural gas-fired power plants while using onshore pipeline, offshore pipeline, and ship for varying transportation has been estimated. The second one is focused entirely on carbon transportation and several likely alternative real-life transport networks are described. It has been determined that pipeline costs are mainly determined by CAPEX and are proportional to the distance covered, thus benefiting from economies of scale and use in full capacity. On contrary, ship and truck transport have relatively lower CAPEX and they are less dependent on distance and on scale of transport. The combination of pipelines and ship transportation towards the creation of offshore networks could be in most cases the most cost-effective solution, especially in the early development of clusters. It has been estimated that the cost for onshore and offshore pipeline transportation ranges from 1.5–5.4 €/tCO₂/180 km and 6–20.4 €/tCO₂/500 km, respectively.

Knoope et al. [68] reviewed and evaluated several cost models that are available in literature, regarding CO₂ transportation using pipelines (including the required booster station), but only reported and compared the overall investment cost and not the cost per tCO₂ transported. The models reviewed take into account many factors, such as carbon steel grade, inlet pressure, region of implementation, terrain, corridor factor, mass flowrate, and distance of transportation for both onshore and offshore applications. Mallon et al. [69] estimated the overall investment cost for different carbon sources in order to compare the reported costs at varying pipeline diameters. They also present the CAPEX costs from two different sources for varying pipeline diameters and lengths.

Aspelund et al. [35] presented the first study with real life data in order to identify the best transportation option for an approximate distance of 1500 km, including all of the steps involved in the transportation of CO₂ using ships (liquefaction, gas conditioning, loading system, offshore unloading system, ship design). The total costs were estimated in the range of $20–30/tCO₂ for volumes larger than 2 MtCO₂ per year. Brownsort [70] reviewed and compared eight different studies in terms of the total lifetime costs of the ship transportation system (expressed as specific cost per unit of CO₂ transported), including compression and liquefaction. The reported transportation cost ranges from $12–58/tCO₂, depending on the distance and the quantity transported. Kjarstad et al. [71] investigated alternative CO₂ transport options in the Nordic region by calculating transportation costs for both ships (13–22 €/tCO₂) and pipelines (4–25 €/tCO₂) at varying flowrates and distances. Weihs et al. [72] used the Integrated Carbon Capture and Storage Economics Model and estimated the CO₂ transportation costs for shipping pure CO₂ and for transporting CO₂ using ship, onshore pipelines, shallow offshore pipelines, and deep offshore pipelines in terms of CAPEX and OPEX, as well as the levelized costs of CO₂ transportation under various flowrates. Pipelines prove to be more economical ($10–15/tCO₂/1000 km) in shorter distances (<500 km) when compared to ships ($20–25/tCO₂/1000 km). However, this is reversed in longer distances (>1500 km), with the two values being $50–70/tCO₂/1000 km and $40/tCO₂/1000 km for pipelines and ships, respectively.

Table 3. Type of transportation examined in previous published studies.

Reference	Pipelines Offshore	Pipelines Onshore	Ships	Truck Tankers	Railroad Tankers
[66]	Yes	Yes	No	No	No
[10]	Yes	Yes	Yes	No	No
[35]	No	No	Yes	No	No
[31]	Yes	Yes	Yes	Yes	No
[68]	Yes	Yes	No	No	No
[69]	Yes	Yes	Yes	No	No
[72]	Yes	Yes	No	No	No
[70]	No	No	Yes	No	No
[71]	No	Yes	No	No	No

summarizes the main characteristics of the reviewed papers. It is apparent that pipelines and ships are the most common and widely applied methods (with pipelines being preferable in shorter distances and ships in longer distance—where applicable), whereas road and rail transportation are rarely considered in real cases and/or in longer distances.

4. Existing Optimization Models and Algorithms

The identification and implementation of novel value chain based on CCU requires both the study of each stage of the value chain separately and the assessment of the entire system as a whole in order to assess the interrelationships among the different components and the way that these affect the technical and economic feasibility of the system.

Several attempts have been made to either optimize each stage of the value chain or model/optimize the entire system, focusing on CCS. These have been recently reviewed by Tapia et al. [73] and they have been classified in three different categories (mathematical programming, pinch analysis, and miscellaneous tools). Our objective is to summarize the papers on the integrated assessment of CCU value chains exclusively and identify the major research gaps.

Hasan et al. [74] developed a Mixed Integer Linear Programming (MILP) model for a CCU network, capturing anthropogenic CO₂ emissions for Enhanced Oil Recovery (EOR) in the United States. They also presented an overview (including the challenges) of CCS and CCU, a multiscale framework for CCS and CCU and proposals for CCS and CCU supply chain networks in the United States (US). Hassiba et al. [75] studied the integration of carbon sources, with different storage and utilization options in an industrial context in order to achieve emissions reduction at a minimum cost, using heat and carbon integration techniques. Agrali et al. [76] developed a MILP model for economic optimization that assesses whether CCS, CCU, a combination of them, or no carbon capture at all is a more viable option for fossil-fired power plants that operate under a cap and trade system. The model considers pipelines and a limited range of sources and receivers. It was applied in Turkey considering two power plants, one utilization point, and two storage sites.

Pérez-Fortes et al. [77] performed a techno economic analysis on a European level to assess two alternative carbon utilization options (methanol synthesis and fly ash aqueous carbonation), with CO₂ captured from a conventional power plant and by using process flow modelling for the estimation of all the relevant technical and economic values. Duraccio et al. [78] attempted a techno economic analysis on a smaller scale (an Italian industrial district) and assessed alternative capture technologies in order to develop a CCU value chain between a power plant (which acted as the source) and a sugar factory (which acted as the sink). Similarly, Reiter and Lindorfer [79] assessed the economic performance of various different carbon sources (taking into account capture cost and CO₂ penalty) in order to identify the best option for the development of power to gas (P2G) industry in Austria. Their analysis concluded that there is surplus of CO₂, capable of providing the necessary input, and the best sources available are CO₂ bioethanol and biogas upgrading facilities.

Karjunen et al. [80] also studied the development of power-to-gas value chains in Finland using biogenic carbon sources. They used a node-based material balance tool, which monitors the carbon flow across the energy systems and estimated all the CCU related costs. Patricio et al. [81] developed a methodological approach to identify European regions with potential in developing CCU value chain networks, while using material flow analysis. The approach is broken down in three major steps; (a) assessment of the regional utilization potential for nine selected technologies, selected based on the Technology Readiness Level; (b) assessment of regional CO₂ availability; and, (c) matching based on geographical criteria on a national and regional level. The same approach has been applied with more detail in one of the previously identified regions (in Sweden), and a number of technically feasible value chains was proposed [11]. von der Assen et al. [82] have mapped the available medium and high flow CO₂ sources in Europe and identified the best regions in Europe for the development of CCU projects that are based on the minimization of the environmental impacts of CO₂ supply, by using the environmental-merit-order curves.

Table 4. Main characteristics of the published CCU modeling approaches.

Reference	CCU/CCS	Type of Receivers	Approach	Application	Economic Assessment	Environmental Impact Assessment	Social Impact Assessment
[74]	Both	EOR only	Optimization (MILP)	USA	Yes	No	No
[75]	Both	Three Selected	Optimization (Pinch)	Qatar	Yes	No	No
[76]	Both	EOR only	Optimization (MILP)	Turkey	Yes	No	No
[77]	CCU	Two Selected	Techno economic Analysis	Europe	Yes	No	No
[78]	CCU	Sugar Production	Techno economic Analysis	Industrial District	Yes	No	No
[79]	CCU	P2G	Techno economic Analysis	Austria	Yes	No	No
[80]	CCU	P2G	Simulation (Node Modelling)	Finland	Yes	No	No
[11]	CCU	Five Selected	Simulation (MFA)	Regional	Yes	No
[81]	CCU	Nine Selected	Simulation (MFA)	Europe	Yes	No	No
[82]	CCU	-	Environmental Merit-Order	Europe	No	Yes	No

CCU: carbon capture and utilization. CCS: carbon capture and storage. MILP: Mixed Integer Linear Programming. MFA: Material Flow Analysis

summarizes the key characteristics of the abovementioned CCU modeling tools and algorithms.

5. Conclusions and Suggestions for Future Work

It is apparent from the literature review that the individual components of a CCU value chain have been extensively studied. The technical characteristics of all five stages (source, capture, purification, transportation, and utilization) have been modeled and optimized and there are clear classification schemes, where appropriate. Regarding the economic characteristics, a detailed and extended analysis have been performed and published only for the capture and transportation stages. This is mainly due to the fact that the research was mostly driven by CCS. Thus, it is necessary to estimate the incurred cost for the major utilization options, since this will play a significant role in the assessment of the entire value chain. This cost might be either due to retrofitting/modifying an existing carbon receiver in order to receive recycled CO₂ or represent the installation cost of a new plant. Moreover, a critical review of the capture and transportation cost related research papers would be useful, in order to assess the impact of economies of scale in these processes and to estimate how the cost has been reduced over the course of the last ten years due to technological progress.

Regarding the optimization of the entire value chain, it can be observed that only a few relevant journal articles have been published, and each one approaches the problem from a different perspective. Most of these attempts focus on a limited number of the potential CO₂ sources and usually a very limited number of receivers. Several research papers have been developed in the context of CCS, and thus the developed algorithms cannot be applied in a straightforward way to a CCU value chain. The majority of the published studies focuses only on the economic assessment of the carbon reuse scheme, based on either the total cost of the project or the economic performance of the carbon source. However, the success of such business models depends on multiple stakeholders with conflicting interests and the economic performance of each one of them and the system as whole should be assessed.

Finally, in almost all cases, the assessment of the environmental and the social impacts of the CCU value chains are neglected. There have been a few studies assessing the environmental impact of carbon capture technologies individually (using life cycle assessment) and one study that examines the environmental performance of the system as a whole. Social impact assessment has been completely ignored, something common for all the industrial symbiosis schemes. However, in order to pass from the first steps of the development of a CCU value chain (opportunity identification and opportunity assessment) to the latter ones (barrier removal and commercialization), apart from the maturity of the technologies involved, a holistic view of the system should be adopted that will be able to examine the performance of all the stakeholders that are involved and all the components of sustainability.