Negative CO₂ emissions in the lime production using an indirectly heated carbonate looping process

Abstract

Lime is an essential raw material for iron and steel production, in construction and agriculture, in civil engineering, in environmental protection, and in manifold chemical manufacturing processes. To address the problem of unavoidable process CO₂ emissions associated with the production of lime, efficient capture technologies need to be developed and implemented. The indirectly heated carbonate looping (IHCaL) process is an efficient candidate for this application because it utilizes lime as the sorbent for the CO₂ capture. In this work, a retrofit configuration of this process is presented and analyzed for net negative CO₂ emissions. This is done considering different fuels that provide the heat required for the regeneration of the sorbent. The different scenarios were simulated with an AspenPlus® model, key performance indicators were calculated, and the process was compared with other post-combustion capture methods. The results show that net negative CO₂ emissions as high as −1805 kg_CO2/t_CaO, calculated with a state-of-the-art coal power plant energy scenario (η_e = 44.2 %; e_ref,el = 770 kg_CO2/MWh_el), can be obtained. This represents an equivalent CO₂ avoidance of more than 230% with respect to the reference plant without capture (1368 kg_CO2/t_CaO). A specific primary energy consumption for CO₂ avoided (SPECCA) lower than 1.5 MJ_LHV/kg_CO2,av was achieved for the same energy scenario. Particularly promising results can be accomplished when applying fuels with high biogenic fraction and low specific CO₂ emissions, such as solid recovered fuels (SRFs) with a high calorific value.

1 Introduction

The latest report of the Intergovernmental Panel on Climate Change (IPCC) stated that carbon dioxide removal (CDR) needs to be deployed to achieve net zero greenhouse gas (GHG) emissions (IPCC, 2022). Furthermore, CDR is required in order to cap the global warming to 1.5 °C with no or limited overshoot (IPCC, 2018). CDR is a “key element” to limit global warming because it is the only means to counterbalance the so-called “residual emissions,” i.e., uneconomical to abate anthropogenic GHG emissions (Quader and Ahmed, 2017). Emission scenarios compatible with the 1.5 °C limit (67% probability) require huge amounts of CDR on top of deep emissions reductions; namely, 730 Gt_CO2 on average (IPCC, 2018; Merk et al., 2022). One of the most promising ways to achieve net negative CO₂ emissions, i.e., CDR, is the implementation of carbon capture and storage (CCS) into industrial processes which emit high amounts of biogenic CO₂ (Clarke et al., 2014; Fuss et al., 2018; Fuss et al., 2014; Quader and Ahmed, 2017). This approach is called bioenergy with carbon capture and storage (BECCS).

Waste-derived fuels have the potential to allow for more economical carbon capture systems due to their lower costs and at the same time enable CDR through BECCS (Haaf et al., 2020c). Refuse-derived fuels (RDFs) and solid recovered fuels (SRFs) are obtained by factions of municipal solid waste (MSW) that cannot be recycled. SRFs are fuels obtained from MSW, which comply with standards from the European Committee for Standardization (CEN) (Gerassimidou et al., 2020), e.g., DIN EN ISO (2021). The term RDF normally refers to waste-derived combustibles of high heating value¹, obtained through the selection of high-quality waste fractions (e.g., paper, wood, plastic, cardboard), which are not defined by CEN standards (Velis et al., 2010). Depending on the quality, consumers may have to pay for these fuels. For low-quality RDF/SRF, suppliers pay the consumers (Sarc and Lorber, 2013).

Lime plants are responsible for the production of raw materials that are widely used in agriculture and the industrial sector. Lime-related products are obtained from the calcination of limestone —mainly calcium carbonate (CaCO₃)— at high temperature (900–1200 °C). The calcination reaction is highly endothermic; thus, a heat input is required, e.g., from the combustion of fuels such as coal, coke, and secondary fuels. Carbon dioxide is emitted as a result of the combustion. Additional CO₂ is produced due to the chemical conversion of CaCO₃ into calcium oxide (CaO) during the calcination. This so-called “process CO₂,” which can only be avoided through CO₂ capture, represents approximately 65 % of the total CO₂ emissions (IEA, 2020a). Overall, the total CO₂ emissions per ton of burnt lime vary between 1 to 2 t_CO2/t_lime (Schorcht et al., 2013).

In order to capture the process and fuel CO₂ emissions, two groups of carbon capture technologies can be deployed, namely, post-combustion and oxyfuel combustion technologies (Plaza et al., 2020). Only few studies have analyzed carbon capture technologies specifically for the lime industry (Erans et al., 2016; Eriksson et al., 2014; Jafarian et al., 2022; Yang et al., 2020), whereas many works have been published recently that consider carbon capture for the cement industry (e.g., Busch et al., 2022; Nhuchhen et al., 2022; Carbone et al., 2022). There are similarities between both industries, like the calcination of CaCO₃, which is the main process in terms of energy consumption (Schorcht et al., 2013). Moreover, CO₂ capture —in particular post-combustion capture— from cement and lime plants have many common features. In both production processes, the majority of the CO₂ emissions come from the raw material, and the specific CO₂ emissions per unit of product are approximately the same. The other components of the flue gas (e.g., HCl, SO₂, moisture, NO_x and N₂, and residual dust) are also comparable if the same fuels are used ². A robust process is needed in both cases to capture the CO₂ from the flue gases. Even though this work focusses on the CO₂ capture from a lime plant, reference studies on carbon capture from cement kilns are used for comparison purposes.

Eriksson et al. (2014) proposed using oxyfuel combustion directly in a lime rotary kiln. They found that, with this system, the total CO₂ emissions may be reduced, but pointed out the technical challenges to control the temperature and, in this way, comply with the high-quality standards of rotary kiln lime products. The environmental and economic potential of oxyfuel combustion for cement production was analyzed by different authors (e.g., Rolfe et al., 2018; Barker et al., 2009). Carrasco et al. (2019) investigated oxyfuel carbon capture from the cement production in a 500 kW_th testing facility. This technology has good energy performance, but presents significant disadvantages when it comes to retrofitability (Voldsund et al., 2019b).

Post-combustion CO₂ capture technologies have a high CO₂ abatement potential and are more suitable for retrofitting compared to oxyfuel combustion (Voldsund et al., 2019b). Nonetheless, the majority of these technologies have very high energy requirements, which increase the costs of the final products and reduce the efficiency of the entire system considerably. Gardarsdottir et al. (2019) evaluated different post-combustion carbon capture processes for the cement production. They calculated that monoethanolamine-based absorption, the reference post-combustion carbon capture technology, has a specific primary energy consumption for CO₂ avoided (SPECCA) of 7.02 MJ/kg_CO2,av and a cost of CO₂ avoided of 80.2 €/t_CO2,av. Barker et al. (2009) estimated that the cost of CO₂ avoided would be higher than 100 €/t_CO2,av to retrofit a 1 Mt_cement/y cement plant located in North East Scotland with a solvent-based post-combustion capture unit.

One noteworthy post-combustion carbon capture technology is the carbonate looping (CaL) process (Shimizu et al., 1999), whereby the CO₂ capture is achieved by utilizing limestone as a sorbent, i.e., the raw material of the lime production facility. The sorbent binds CO₂ from the kiln flue gases in a carbonator and is regenerated through a temperature increase in a calciner, according to the reaction in Eq. (1) (Anantharaman et al., 2018).

$${CaCO}_{3(s)}\rightleftharpoons {CO}_{2(g)}+{CaO}_{(s)};\Delta {H}_{298K}=\pm 1,780\textrm{kJ}/\textrm{kg}$$

(1)

For the regeneration of the sorbent in the standard CaL process, fuel is burnt directly in the calciner. For this, technically pure oxygen is used, which requires an air separation unit (ASU) (Carrasco-Maldonado et al., 2016). CaL technology has the potential to efficiently capture CO₂ from lime plants by exploiting the synergies of the calcination.

The CaL process has been successfully operated up to the pilot scale in Stuttgart, Germany (200 MW_th) (Charitos et al., 2011; Dieter et al., 2014; Hornberger et al., 2021, 2020), in Darmstadt, Germany (1 MW_th) (Haaf et al., 2020b; Hilz et al., 2018, 2017; Kremer et al., 2013; Ströhle et al., 2020; Ströhle et al., 2014), and in La Pereda, Spain (1.7 MW_th) (Arias et al., 2017b; Arias et al., 2013; Diego et al., 2020; Diego et al., 2016b). For power plants, the CaL process has the potential to achieve high CO₂ capture rates with low energy penalties. Lasheras et al. (2011) estimated that a full-scale power plant could be retrofitted with CaL to capture 88% of the total CO₂ formed, with an energy penalty of less than 2.9%. Astolfi et al. (2019) calculated that a SPECCA of 2.16 MJ_LHV/kg_CO2,av could be achieved by the integration of the CaL process into power plants with thermochemical energy storage, and Haaf et al. (2020a) estimated a SPECCA of 5.72 MJ_LHV/kg_CO2,av for the integration into waste-to-energy plants.

Experimental investigations are being carried out to apply CaL technology into the cement industry. Arias et al. (2017a) achieved more than 90% CO₂ capture in a CaL 30 kW_th test facility at relevant conditions for cement plants. Within the CLEANKER project, a demonstrator CaL unit has been erected to capture CO₂ from an operating cement plant that produces 1.3 Mt_cement/y in Vernasca, Italy (Fantini et al., 2021). De Lena et al. (2022) investigated the application of different CaL configurations into the cement industry and reported SPECCA values between 2.8 and 3.0 MJ_LHV/kg_CO2,av for systems utilizing pure limestone as sorbent, and between 3.5 and 4.6 MJ_LHV/kg_CO2,av for systems that utilize cement raw meal.

The ASU in the CaL process increases the SPECCA by approximately 1 MJ_LHV/kg_CO2,av (De Lena et al., 2022). The requirement for technically pure O₂ can be avoided by indirectly heating the calciner, e.g., through solid looping (Diego et al., 2016a), and thus the energy penalty is reduced (Martínez et al., 2016). One excellent means to achieve this is through heat pipes (Hoeftberger and Karl, 2016), which transfer heat from an external combustor into the calciner via evaporation and condensation of a fluid. This indirectly heated carbonate looping (IHCaL) process (Epple, 2009) presents several advantages compared to the oxy-fired CaL process: reduced energy requirement, improved sorbent activity, lower sorbent attrition rates, and high purity of the captured CO₂. It has the potential to enable carbon capture with very low CO₂ avoidance costs³ (Junk et al., 2016).

The IHCaL process has been successfully operated for 400 h at the 300 kW_th facility of the Technical University of Darmstadt (Reitz et al., 2016) in operating conditions corresponding to CO₂ capture from coal-fired power plants. Additional test campaigns in Darmstadt were carried out during 2022 to prove the operability of the IHCaL process under lime plant conditions at the pilot scale with solid fuel feedstock (Hofmann et al., 2022a, 2022b; Ströhle et al., 2021). The facility was operated for more than 300 additional hours during the year 2022.

Furthermore, the utilization of secondary fuels has been successfully demonstrated up to the pilot scale (1 MW_th) for CaL operation (Haaf et al., 2020d; Haaf et al., 2020b). Regarding the IHCaL process, different solid fuels were fueled in the 300 kW_th heat pipe IHCaL testing facility of the Technical University of Darmstadt (Hofmann et al., 2022a, 2022b). The combustor was operated around 20 h with lignite and around 20 h with RDF pellets, with the compositions and heating values displayed in Table 5.

At the Technical University of Darmstadt, novel concepts for the integration of the IHCaL process into the lime production were developed and evaluated through process simulation (Greco-Coppi et al., 2021). The published results show that the direct CO₂ emissions can be reduced by up to 87% by utilizing dried lignite as fuel for both the lime kiln and the IHCaL combustor. Nevertheless, the application of waste-derived fuels into these concepts to enable CDR has not been discussed yet.

This work investigates the influence of four different fuels on the CO₂ emissions and energy requirements of a tail-end IHCaL process integrated into a lime plant. The objective of this paper is to unravel the potential of the IHCaL process to achieve net negative CO₂ emissions, thus enabling CDR. Furthermore, it aims to assess the energy performance of the IHCaL process, compared to other carbon capture technologies that are being considered for deployment in the cement and lime industries.

2 Methodology

2.1 Process integration

The IHCaL concept considered in this paper is referred to as the tail-end or retrofit configuration in previous works reported in the literature (Greco-Coppi et al., 2021; Junk et al., 2013). This process is suitable for capturing CO₂ from operating lime plants⁴. The configuration is shown schematically in Fig. 1. It consists of a host lime plant (left side) and an IHCaL facility (right side). In Europe, this configuration has the potential to decarbonize existing facilities with more than one kiln. The IHCaL facility replaces one kiln and, at the same time, captures the CO₂ of the remaining kilns.

Fig. 1

Tail-end concept for the integration of the IHCaL CO₂ capture process into an existing lime plant, introduced by Greco-Coppi et al. (2021)

The host facility for this work is the lime production line located in Germany described by Greco-Coppi et al. (2021). The rotary kiln is equipped with a limestone preheater (PRK) and is fueled with dried lignite (LHV = 21500 kJ/kg_wet). The burnt lime (mainly CaO) is cooled downstream of the kiln with the combustion air. The kiln flue gases are used to preheat the raw material. An air quench is used to reduce the temperature before the filter and the blower. The flue gases exit the host plant at 236 °C and high⁵ CO₂ concentrations (19.0 vol%_dry).

The IHCaL facility allows for the capture of CO₂ utilizing CaO as sorbent and increases the total production of the plant through the calcination of the make-up stream (CaCO₃). There are three main reactors: (i) a carbonator operating as a circulating fluidized bed (CFB) for the absorption of CO₂, (ii) a calciner operating in a bubbling bed regime (BFB) for the sorbent regeneration, and (iii) a BFB combustor providing the energy required to regenerate the sorbent.

The flue gases from the kiln and the combustor are cooled at HX-6 to reduce the propelling energy requirements. Afterwards, they enter the carbonator from the bottom by means of a blower. The same flue gas is the fluidizing agent that allows for the CFB operation. In the carbonator, the CO₂ from the flue gas is absorbed by the circulating sorbent (CaO) to form CaCO₃. The CO₂-depleted flue gas exits the IHCaL facility through the cyclone 1, and the CaCO₃ enters the calciner. In the calciner, the sorbent is regenerated, and the CO₂ is released in a high concentration (> 95 vol%_dry) stream. This CO₂ is then conditioned for transport and storage. The main assumptions for the downstream conditioning facility are presented in Section 2.3. The solids leaving the calciner enter the carbonator; thus, the calcium loop is established. Heat is supplied into the calciner from the combustor via heat pipes (Hoeftberger and Karl, 2016). The combustor can be fueled with lignite or waste-derived fuels, as explained in Section 2.5.

As a result of the deactivation of the sorbent, a constant make-up is required to maintain a high carbon capture rate (Grasa and Abanades, 2006). Make-up can be added into the process directly into the carbonator or the calciner or into the connecting elements (e.g., loop seals). The used sorbent (CaO) is removed from the system downstream of the calciner and may be sold as burnt lime⁶. The limestone composition from the host lime plant (see Table 1) is considered in this work. It is assumed that this limestone is used, not only for the rotary kiln, but also as make-up and sorbent for the IHCaL process.

Table 1 Composition of the limestone used in the reference plant and in the IHCaL carbon capture facility (Greco-Coppi et al., 2021)

2.2 Process model

The heat and mass balances were calculated with the software AspenPlus®, version V12. Custom routines in FORTRAN code were included. Steady-state conditions were assumed, and the cyclone separation was considered ideal. The ambient pressure and temperature were set to 1.013 bar and 15 °C, respectively, and a plant capacity factor of 91.3% was assumed in accordance with Voldsund et al. (2019a). The calculation of the material properties and the balances in the reactors was performed as explained by Greco-Coppi et al. (2021). For the combustor, an air-fuel equivalence ratio (λ) of 1.2 was specified.

The temperatures of the reactors and the main operating parameters for the calculations are displayed in Table 2. It was assumed that the reactors, heat exchangers, and ducts are adequately insulated, and thus the thermal losses are negligible. Accordingly, these components were modeled adiabatic (Chen et al., 2020).

Table 2 Main operating parameters of the IHCaL process

In this work, the make-up solid stream (F₀) and the circulating solid stream (F_R) are calculated from defined ratios (Λ, Φ) and the total CO₂ molar flow rate entering the carbonator (F_CO2), according to Eq. (2).

$${F}_0=\varLambda \cdot {F}_{CO2};\kern0.5em {F}_R=\varPhi \cdot {F}_{CO2}$$

(2)

The CO₂ capture efficiency (E, see Eq. 7), is given as an input: E = 90%. The required carbonator efficiency (E_carb) is calculated with Eq. (3), from the molar flow rates of CO₂ entering (F_CO2) and leaving (${F}_{CO2}^{carb}$) the carbonator.

$${E}_{carb}=1-\frac{F_{CO2}^{carb}}{F_{CO2}}$$

(3)

The calciner efficiency is an input for the model and is defined as:

$${E}_{calc}\equiv \frac{X_{carb}-{X}_{calc}}{X_{carb}}$$

(4)

Here, X_calc and X_carb are the fractions of CaCO₃ in the calcium (Ca) stream leaving the calciner and the carbonator, respectively. The composition of Table 1 was used to model the limestone streams. This includes the make-up stream and the raw material input into the rotary kiln.

The carbonator efficiency (E_carb) is calculated with the carbonator reactor model developed by Lasheras et al. (2011) that considers: (i) circulating fluidized bed hydrodynamics according to Kunii and Levenspiel (1991); (ii) the carbonation reaction model from Abanades et al. (2004); and (iii) sorbent deactivation as modeled by Abanades et al. (2005). The make-up rate was set to Λ = 0.1, and Φ was varied to achieve the necessary CO₂ capture efficiency of E = 90%. The carbonator reactor model assumptions and results are included in Appendix 1 and Appendix 2, respectively.

2.3 Power requirements

In this work, the battery limits for the analysis are the input of the flue gases from the reference facility before the stack and the exit from the CO₂ compression unit. The CO₂ emissions and primary energy consumption related with the transport and storage of CO₂, transport and pre-treatment of raw materials and fuels (e.g., fractioning), and the erection of the carbon capture facility are out of the scope of this paper⁷. The main assumptions and input parameters for the analysis are summarized in Table 3.

Table 3 Main assumptions and general input parameters for the calculation of power requirements

For the calculation of the electric power in the reference facility, the data from the best available technique reference document for cement, lime, and magnesium oxide (Schorcht et al., 2013) was considered: 17–45 kWh/t_lime for a lime rotary kiln. The mean value was used for the calculations: 31 kWh/t_lime. The power demand from the IHCaL facility and the downstream conditioning are used to obtain the net power generation (P_el) for the calculation of the indirect CO₂ emissions (Eq. 12) and indirect primary energy consumption (Eq. 10).

After a post-combustion carbon capture facility, downstream conditioning of the captured CO₂ is necessary. For oxy-fired CaL, purification is required due to the presence of combustion gases other than CO₂ (mainly O₂). Furthermore, the CO₂ stream is to be compressed up to a suitable temperature for transportation of around 110 bar. Such CO₂ compression and purification units (CPUs) have relatively high energy requirements that range from 80 to 120 kWh_e/t_CO2 (De Lena et al., 2018; Garðarsdóttir et al., 2018; Jackson and Brodal, 2019; Magli et al., 2022; Svensson et al., 2021). In the IHCaL process presented in this work, the combustion to generate the heat for the regeneration takes place in an external combustor, and thus the CO₂ stream after the calciner is almost pure. The only conditioning required is the cooling and the filtering, after which, the stream is ready for compression. The compression takes place in a 5-stage CO₂ compressor. The compression was simulated in Aspen Plus, based on the method reported by Posch and Haider (2012). The assumptions for the compression unit are reported in Table 3.

The power requirement of the blowers depends on the pressure drop (Δp) in the reactors and the auxiliary components, i.e., the nozzle grid, cyclone, cooler, filter, and ducts. The following values were assumed according to the experimental data of the research group: 100 mbar for the carbonator, 130 mbar for the calciner, and 150 mbar for the combustor (Reitz et al., 2016). For the blowers, the isentropic and mechanical efficiencies were set to 0.65 and 0.9, respectively (Grote and Feldhusen, 2007). It was assumed that the flue gases entering the carbonator and the combustion air act as fluidization agents for the corresponding reactors. For the calciner, the fluidization agent⁸ is a fraction of the pure CO₂ flow stream that is recirculated to allow for BFB operation. To calculate the amount of recirculation required, the following assumptions were made: (i) superficial velocity for the fluidization agent at inlet u_0,calc = 0.25; (ii) heat pipe properties as reported by Höftberger et al. (2016), namely, 3 m calciner width (i.e., 6 m heat pipes), 7.2 m calciner length/50 MW_th; (iii) and calciner heat input equal to 100 MW_th. Finally, the temperature of the fluidization agent (T_FA) before the blowers is defined. The air for the combustor is compressed from ambient temperature. The flue gases entering the carbonator are cooled down to 250 °C before the compression, and the recirculated gases for the fluidization in the calciner are cooled down to 450 °C.

2.4 Heat integration and power generation

The configuration displayed in Fig. 1 allows for efficient heat utilization. The combustion air for the combustor is preheated (T_preheat), and heat is exchanged between the solid streams to increase the temperature of the solids entering the calciner (T_sorb,calc,in). These design specifications minimize the total heat requirement, as shown by Greco-Coppi et al. (2021).

To achieve a high T_sorb,calc,in, a solid-solid heat exchanger (HX-SS) is required. Different configurations are possible for the design of this heat exchanger: (i) a concept that utilizes molten salt circulating inside of metal tubes; (ii) a concept with heat-pipes, similar to those presented by Hoeftberger and Karl (2016) to transfer heat into the calciner; (iii) a concept with high surface area metal walls separating the solid flows; and (iv) a concept consisting of two concentric L-valves (Greco-Coppi et al., 2021). For the considered inputs (see Table 2), a counter-current configuration of this heat exchanger yields a logarithmic mean temperature difference of around 90 °C.

The high operating temperatures (650–900 °C) make the IHCaL process particularly suitable for power generation through a heat recovery steam cycle (De Lena et al., 2018; Lasheras et al., 2011). Steam can be produced from the cooling of the carbonator and from the gas streams exiting the carbonator (650 °C), the calciner (900 °C), and the combustor (1000 °C). For the calculation of the power generation through a heat recovery steam cycle, the recovered heat is obtained from the AspenPlus® simulations. The temperatures assumed for this purpose are displayed in Table 4.

Table 4 Operating temperatures for the heat exchangers (HX), flue gas side

For the calculation of the power generation, the steam cycle was simulated with the software EBSILON Professional^TM. The main assumptions for the calculations were: (i) superheating of steam up to 565 °C and 130 bar; (ii) preheating of feed-water with steam extractions; (iii) isentropic turbine efficiency equal to 85%. The calculated heat-to-power efficiency (η_h2p) was 42.4%. This value corresponds to an equivalent net electrical efficiency of around 38% for a thermal power plant (e.g., pulverized coal), which is in agreement with values from the literature (IEA, 2020b). The total power generation from the IHCaL facility can be calculated with Eq. (5). Here, ${\dot{Q}}_{IHCaL, HRSG}$ is the recovered heat from the IHCaL unit.

$${P}_{el, out}={\eta}_{h2p}\cdot {\dot{Q}}_{IHCaL, HRSG}$$

(5)

2.5 Fuels and CO₂ emissions

The focus of this work lies on the investigation of the effect of implementing different fuels in the IHCaL process. The biogenic CO₂ capture and associated negative emissions are of special interest. Four fuels were selected for the analysis: (i) dried lignite from the reference process in the host plant in Germany (Greco-Coppi et al., 2021); (ii) RDF pellets, which are used in pilot test campaigns at the Technical University of Darmstadt (Ströhle et al., 2021); (iii) a class 3 SRF, according to EN ISO 21640:2021-11 (2021), that was successfully utilized in the 1 MW_th pilot plant at the Technical University of Darmstadt for CaL operation (Haaf et al., 2020d); and (iv) municipal solid waste (MSW), with the composition from the CaL techno-economic analysis from Haaf et al. (2020a). Dried lignite was maintained as the fuel of the reference plant for all cases, and only the fuel for the IHCaL combustor was varied.

The fuel CO₂ emissions index (Furimsky, 2007; Madejski et al., 2022) is also known as the fuel-specific CO₂ emissions, e_CO2,fuel (g_CO2/MJ_LHV). It indicates the mass of CO₂ produced by the combustion of fuel per unit of energy obtained. The Eq. (6) can be used to calculate it. Here, w_c,wet is the wet-basis mass fraction of carbon in the fuel, LHV_wet is the fuel lower heating value in wet basis, M_CO2 is the molar mass of CO₂, M_c is the molar mass of carbon. The input parameters of the fuels considered in this analysis, including the CO₂ emissions index, are presented in Table 5.

$${e}_{CO2, fuel}=\frac{w_{c, wet}}{LHV_{wet}}\cdot \frac{M_{CO2}}{M_C}$$

(6)

Table 5 Input parameters of the fuels used in this analysis

For the calculation of the negative CO₂ emissions, the biogenic and fossil emissions are distinguished. CO₂ emissions from pre-dried lignite are considered 100% fossil, as well as the emissions from limestone calcination. For the fuels burnt in the combustor, the biogenic carbon fractions (x_bio) are defined. According to Moora et al. (2017), x_bio varies considerably depending on the waste selection process and the region-dependent source segregation. The determination of the x_bio of the RDF pellets was performed according to the German standard DIN EN 15440:2011 (2011). For the other fuels, x_bio was assumed considering values from the literature (Astrup et al., 2009; Haaf et al., 2020a; Mohn et al., 2012; Mohn et al., 2008; Obermoser et al., 2009). Astrup et al. (2009) reported a range of 45–85% for MSW and explained that the biogenic carbon content of SRF is normally low, compared to MSW, because of the selective fractioning. The values of x_bio used in this work are displayed in Table 5.

2.6 Key performance indicators

Key performance indicators (KPIs) of the IHCaL process are the carbon capture efficiency (E), the heat ratios (HR_a, HR), and the product ratio (PR). The carbon capture efficiency of the IHCaL process (E) is defined as the ratio of CO₂ captured to total CO₂ generated. It can be calculated as follows:

$$E={\left(1+\frac{F_{CO2}^{carb}}{F_{CO2}^{calc}}\right)}^{-1}$$

(7)

Where ${F}_{CO_2}^{calc}$ and ${F}_{CO_2}^{carb}$ are the molar flow rates of the captured CO₂ leaving the calciner and the CO₂ leaving the carbonator, respectively. In this work, E was set as an input, and the required E_carb to achieve this efficiency was calculated.

The heat and product ratios are calculated with Eq. (8), where ${\dot{m}}_{CaO; prod}$ is the total lime production, ${\dot{Q}}_{in}$ is the total heat input from the fuel combustion, and the superscript ref indicates the lime production plant without carbon capture. PR is the product ratio, HR_a is the absolute heat ratio, and HR is the specific heat ratio that indicates the increase in heat input per unit of lime produced.

$$PR\equiv\frac{{\dot{m}}_{CaO, prod}}{{\dot{m}}_{CaO, prod}^{ref}}; \quad{HR}_a\equiv\frac{{\dot{Q}}_{in}}{{\dot{Q}}_{in}^{ref}};\quad HR\equiv\frac{HR_a}{PR}$$

(8)

The KPIs introduced above are specific of the IHCaL process. Other important KPIs, which allow to compare with other carbon capture technologies, are introduced hereunder. They were selected considering relevant work in post-combustion carbon capture from cement and lime plants (De Lena et al., 2017; Ströhle et al., 2021; Voldsund et al., 2019a), especially the work within the CEMCAP project, which established a framework for comparative analysis of CO₂ capture processes for cement plants (Anantharaman et al., 2018).

For the calculation of the specific primary energy consumption for CO₂ avoided (SPECCA), a procedure similar to the one considered by Haaf et al. (2020a) was adopted. The following power generation scenarios are taken into account: (i) the state-of-the-art for coal power plants (abbreviated “coal,” in this work) (De Lena et al., 2018; European Union, 2015), (ii) the European energy mix (abbreviated “energy mix,” in this work) calculated and used in CEMCAP (Anantharaman et al., 2018; De Lena et al., 2018), (iii) the renewable (Anantharaman et al., 2018), and (iii) the nuclear (Anantharaman et al., 2018). For each of them, a reference electrical efficiency (η_ref,el) and a reference CO₂ emissions factor for power production (e_ref,el) are defined (see Table 6). For the scenarios (i) and (ii), these parameters are within the range of the values used normally in the literature⁹. The scenarios (iii) and (iv) are zero-CO₂-emission with η_ref,el = 100% and η_ref,el = 33%, respectively. They were chosen to study the sensitivity of the results to η_ref,el.

Table 6 Main results and KPIs for the different fuels

The equivalent fuel consumption (q_eq) and the equivalent CO₂ emissions (e_CO2,eq) for the different cases can be calculated with Eq. (9).

$${\displaystyle \begin{array}{cc}{q}_{eq}=q+{q}_i;& {e}_{CO2, eq}\end{array}}={e}_{CO2}+{e}_{CO2,i}$$

(9)

The direct fuel consumption (q) is the primary energy entering the system through the combustion of the fuels in the rotary kiln and the combustor. The indirect fuel consumption (q_i) is the primary energy consumption related to the net electric generation (or consumption) in the entire facility (P_el). It depends on the reference electrical efficiency η_ref,el:

$${q}_i=\frac{P_{el}}{\eta_{ref, el}}$$

(10)

The direct CO₂ emission (e_CO2) is the sum of fossil CO₂ directly emitted at the stack of the facility per unit of produced lime. The CO₂ from the calcination, i.e., process emission, is considered fossil emission. For the retrofitted case with carbon capture (CC), e_CO2 can be calculated with Eq. (11), where ${\dot{m}}_{CO2, foss}$ (kg_CO2/h) is the total fossil CO₂ emissions generation, ${\dot{m}}_{CO2, capt}$ (kg_CO2/h) is the captured CO₂, and ${\dot{m}}_{CaO}$ (kg_CaO/h) is the total production from the retrofitted plant, including the product from the IHCaL unit.

$${e}_{CO2, CC}=\frac{{\dot{m}}_{CO2, foss}-{\dot{m}}_{CO2, capt}}{{\dot{m}}_{CaO}}$$

(11)

The indirect CO₂ emissions (e_CO2,i) are those associated to P_el. They can be calculated with Eq. (12), considering the reference CO₂ emissions factor for power production (e_ref,el) of the corresponding reference energy scenario (see Table 6).

$${e}_{CO2,i}={P}_{el}\cdot {e}_{ref, el}$$

(12)

The final equation for the calculation of the SPECCA is:

$$SPECCA=\frac{q_{eq}-{q}_{eq, ref}}{e_{CO2, ref}-{e}_{CO2}}$$

(13)

3 Results and discussion

In this section, the results for the analyzed cases are presented. Firstly, the specific CO₂ formation in each component is discussed. Afterwards, the results of the main KPIs are explained. Finally, the IHCaL process is compared with other post-combustion carbon capture processes.

Figure 2 shows the specific CO₂ formation in each component of the new integrated concepts (b–e), as well as the reference pilot plant without capture (a). The gray and green bars represent the fossil and biogenic specific CO₂, respectively. Direct negative CO₂ emissions are achieved for the scenarios that utilize waste-derived fuels in the combustor (Fig. 2c–e). In these cases, the total direct negative CO₂ emissions are depicted in the figure with a pink rhombus. For the reference facility and the carbon capture scenario with dried lignite (Fig. 2a–b), the direct CO₂ emissions balance is positive. The net direct CO₂ emissions are displayed with a white rhombus. No net negative direct emissions are achieved with dried lignite, since no biogenic emissions are captured.

Fig. 2

Specific CO₂ formation and capture by component for all five cases considered within this study: (a) reference facility; and IHCaL fueled with (b) dried lignite, (c) RDF pellets, (d) SRF, and (e) MSW. The biogenic CO₂ formation is indicated with green, whereas the gray represents the fossil ones. For the cases (a) and (b), the direct CO₂ balance is positive, and the total direct CO₂ emissions are displayed with a white rhombus (◇). For the remaining cases, direct negative CO₂ emissions are achieved. They are indicated with a pink rhombus (◇)

The total specific CO₂ formation increases with the addition of the IHCaL facility (Fig. 2b–e), compared to the reference case (Fig. 2a). This is because of the CO₂ generation associated with the additional energy requirement for the carbon capture. The additional formation correlates with the fuel CO₂ emissions index (e_CO2,fuel). For fuels with lower e_CO2,fuel, the total formation is also lower, as less amount of CO₂ is generated in the combustor to supply the heat to the calciner. For this reason, the scenario with SRF has the less total formation of all the carbon capture scenarios.

If dried lignite is burnt in the combustor (Fig. 2b), the specific direct fossil CO₂ generation is almost two times that from the reference case (Fig. 2a). This means that, for dried lignite, the direct fossil CO₂ generation associated with the carbon capture is approximately equal to the avoided CO₂. On the other hand, for the waste-derived fuels (Fig. 2c–e), the avoidance can be achieved without forming huge amounts of additional direct fossil CO₂ emissions (around 35% increase). The case with the lowest direct fossil CO₂ formation is the one of the RDF pellets (Fig. 2c), due to the combination of high x_bio with low e_CO2,fuel.

The highest variation of the formation with fuel type occurs in the combustor, where the fuel is burnt. The combustor is the most critical component regarding the direct formation of CO₂ in the IHCaL. Here, the formation is minimized by fuels with lower e_CO2,fuel. When dried lignite is used (Fig. 2b), the direct CO₂ formation in the IHCaL combustor is higher than the formation in the lime kiln. On the contrary, when RDF or SRF are utilized (Fig. 2c–d), the direct fossil generation in the combustor is much lower (61–65%). In the case of the MSW (Fig. 2e), the fossil emission of the combustor and the lime kiln are similar (84%).

The reduction of the specific CO₂ formation in the lime kiln with respect to the reference case is explained by the increase of the production, i.e., PR > 1. This reduction is stronger in the scenario with MSW (Fig. 2e), because of the higher PR. Nevertheless, due to the high e_CO2,fuel, more CO₂ is formed from the combustion in the IHCaL; thus, this case presents the highest total direct CO₂ formation.

Due to the high biogenic fractions of the waste-derived fuels, net negative direct CO₂ emissions can be achieved in all three cases (Fig. 2c–e). The total net direct CO₂ emissions can be read from the graph as the difference between the total capture and the total fossil formation. It is indicated with a pink rhombus. The values displayed in Fig. 2 correspond only to the direct emissions, whereas the equivalent emissions are illustrated in Fig. 4.

The main results and the KPIs of the simulated scenarios are reported in Table 6. The increase in specific heat requirement due to the carbon capture (see HR) ranges from 201 to 271%, with respect to the reference case. The difference results from the strong influence of the e_CO2,fuel in the heat consumption, illustrated in Fig. 3. This result is also highly dependent on Λ, as demonstrated by Greco-Coppi et al. (2021), who presented scenarios with less than 100% specific heat requirement increase for the CO₂ capture. The CO₂ formation in the combustor increases with e_CO2,fuel; thus, increasing the total captured CO₂. Due to the more demanding capture requirement, more heat is needed in the calciner and HR becomes higher. The direct fuel consumption, q, and the direct CO₂ emissions, e_CO2,d, increase with HR.

Fig. 3

Absolute heat ratio and CO₂ specific emissions for the fuels considered in this work. The circles represent the results of the simulations corresponding to each of the fuels. The gray area represents the theoretical increase of the heat requirement with the CO₂-specific emission, for a wide range of fuels

Due to the addition of the IHCaL facility, the total production increases (PR > 1). The increase is almost the same for lignite, RDF, and SRF and ranges from 36 to 38%. For the MSW, the production increases more (52%) because of the additional make-up requirements associated with higher CO₂ mass flows (see Fig. 2e). The product ratio can be increased by increasing the make-up rate (i.e., higher Λ). This has the effect of augmenting the production exponentially while reducing HR (Greco-Coppi et al., 2021).

Figure 3 illustrates the relation between HR_a and the specific CO₂  emissions for each fuel considered in this study. The HR_a rises with increasing CO₂ emissions index (e_CO2,fuel). This is due to the additional CO₂ that has to be captured from the carbonator. Other parameters such as LHV and the amount of hydrogen in the fuels also play a role in the variation of the heat requirements. Higher LHV and lower e_CO2,fuel minimize HR_a. The range of variation represented by the gray area was calculated with the energy balance of the calciner and the combustor (see Appendix 3).

The breakdown of specific CO₂ emissions per tonne of burnt lime (product) is displayed in Fig. 4. The emissions are separated in three categories: (i) direct fossil emissions, (ii) direct biogenic emissions, and (iii) indirect emissions. The sum of all three gives the equivalent CO₂ emissions (e_CO2,eq). The results are presented for all the calculated cases, and all the considered energy scenarios. The indirect and equivalent CO₂ emissions depend on the reference efficiency of the energy scenario (e_ref,el). The results are identical for the renewable and the nuclear energy scenarios because they both have e_ref,el = 0.

Fig. 4

Specific CO₂ emissions for the different fuels considered within this work. The breakdown of the emissions is depicted with bars. The power generation scenarios are indicted with the filling type (solid, hatched). The equivalent specific CO₂ emissions (e_CO2,eq) are displayed with a rhombus (◇)

The reference case, without carbon capture, presents the highest emissions level, e_CO2,eq, of 1344–1368 kg_CO2/tCaO. The major contribution comes from the direct fossil emissions corresponding to the calcination and combustion in the lime kiln. The indirect emissions are almost negligible. The results are similar for all the energy scenarios.

For the carbon capture scenarios, net negative equivalent CO₂ emissions can be achieved in every case, except when fueling lignite, for the renewables and nuclear energy scenarios. If waste-derived fuels are used, the highest contribution to the negative emissions corresponds to the captured biogenic CO₂, which is independent from the energy scenario. The indirect emissions are strongly dependent on e_ref,el because of the relatively high power generation in the retrofitted plants (42–63 MW_el). With waste-derived fuels, negative emissions as high as −1805 kgCO₂/t_CaO can be achieved. This corresponds to an equivalent CO₂ avoidance of over 230%.

The SPECCA is one of the most important thermodynamic KPIs. It represents the primary energy consumption associated with the CO₂ avoidance. In Fig. 5, the breakdown of SPECCA for all the carbon capture scenarios is displayed, considering (i) the CO₂ compression power requirement, (ii) the power requirement increase (without compression), (iii) the fuel requirement increase, and (iv) the power generation. The values are presented for the energy scenarios utilized throughout this work. Because of the high power generation (light blue bar in Fig. 5), the results depend strongly on the reference power generation efficiency (η_ref,el). The lower the η_ref,el, the better the results in terms of SPECCA. The values corresponding to the European energy mix scenario are higher than for the state-of-the-art coal power plant due to the lower associated CO₂ avoided (see Fig. 4). In the same way, the SPECCA values of the lignite scenario are relatively high because of the low CO₂ avoidance compared to the waste-derived fuels that allow for net negative CO₂ emissions. When compared with an oxy-fired CaL process, the SPECCA values presented in this work are in general much lower, mainly because of the avoided penalty of the ASU, which increases the total SPECCA by approximately 1 MJ_LHV/kg_CO2,av in the European mix scenario (De Lena et al., 2022). Nevertheless, for the renewable scenario, relatively high values were obtained because of the high η_ref,el. These values are discussed hereunder.

Fig. 5

SPECCA breakdown for all the carbon capture scenarios analyzed in this work. Each category of the breakdown is indicated in with solid bars (state-of-the-art coal power plant) and hatched bars (energy mix 2015). The SPECCA values are displayed with a rhombus (◇)

In Fig. 6, the SPECCA values of different post-combustion carbon capture processes are compared with the ones corresponding to the IHCaL scenarios, for the different energy scenarios considered in this work. The selected post-combustion CO₂ capture processes for this comparison are: (i) monoethanolamine absorption (MEA), a technologically ready process; (ii) chilled ammonia process (CAP); (iii) tail-end directly heated CaL; and (iv) entrained flow (EF) directly heated CaL. As a reference for the MEA, CAP, and CaL technologies, the SPECCA data from the CEMCAP project was used (Voldsund et al., 2019a). These values were calculated for cement production, which is similar to lime production, as discussed in the introductory chapter. It can be seen that the SPECCA of the IHCaL scenarios is considerably lower than the ones reported for the other carbon capture methods, except for the renewable scenarios. The IHCaL process allows for CO₂ capture with very low primary energy consumption, less than 2.1 MJ_LHV/kg_CO2,av, when using waste-derived fuel. Nevertheless, the SPECCA values increase drastically for the renewable scenario. Thereby, the electrical power is considered equivalent as the primary energy; thus, the additional power generation in the IHCaL process is not advantageous as in the other scenarios. Furthermore, the assumption of η_ref,el = 100% associated with this scenario is unfairly high considering the type of feedstock involved.

Fig. 6

SPECCA comparison between the different fuels considered within this work and other post-combustion carbon capture processes: monoethanolamine absorption (MEA), chilled ammonia process (CAP), tail-end directly heated carbonate looping (CaL), and entrained flow (EF) directly heated CaL. The reference data for MEA, CAP, CaL tail-end, and CaL EF was obtained from Voldsund et al. (2019a), who considered the application of these technologies in the cement production process. No data was available for the nuclear scenario

The main challenge of the IHCaL process is the significant increase in the absolute heat required for the capture, i.e., HR_a, which is around 30% higher than for an oxy-fired CaL process¹⁰. Nonetheless, dynamic investment models suggest that the IHCaL technology would be superior in terms of global economic performance, compared to other post-combustion CO₂ capture processes (Junk et al., 2016). Within the ANICA Project, the concepts presented in this work are being evaluated to assess their viability in terms of CO₂ avoidance costs and environmental impact (Ströhle et al., 2021).

4 Conclusion

An innovative CO₂ post-combustion carbon capture method, the IHCaL process, was analyzed in this work. The configuration presented is suitable for retrofitting lime and cement plants. To evaluate its performance with alternative fuel firing, mass and heat balances with different fuels were performed, and the most relevant KPIs were calculated.

From the direct emissions breakdown, it was shown that the combustor influences the direct formation of CO₂ the most. The direct CO₂ formation is minimized by fuels with a lower CO₂ emissions index, e_CO2,fuel. Additionally, the utilization of dried lignite yielded an increase of approximately 100% in the total direct fossil CO₂ formation. This means that the additional generation associated with the carbon capture was approximately equal to the avoided CO₂. With this consideration, it appears more reasonable to use waste-derived fuels for the tail-end IHCaL, whereby the increase in total direct fossil CO₂ formation linked to the avoidance is relatively low (around 30%).

The results show that very low SPECCA values can be achieved for three of the simulated scenarios¹¹: from 0.50 to 2.79 MJ_LHV/kg_CO2,av. In particular, SPECCA values between 0.50 and 1.98 MJ_LHV/kg_CO2,av were achieved for the scenarios utilizing waste-derived fuels in the combustor. By reason of its low primary energy requirements¹², the IHCaL process is a very promising retrofitting technology for carbon capture from lime and cement plants. It may be deployed in scenarios, in which the associated power generation is an advantage. This is not the case for the 100% renewables scenario, which assumes the same worth for generated power and primary energy (η_ref,el = 100%). For this scenario, the SPECCA values were higher than 4.6 MJ_LHV/kg_CO2,av.

Furthermore, it was demonstrated that the IHCaL process is suitable for achieving net negative CO₂ emissions; thus, carbon dioxide removal (CDR). For all the scenarios, the highest negative emissions were obtained with MSW fuel. Net negative emissions as high as −1805 kg_CO2/t_CaO were achieved¹³. This value represents an equivalent CO₂ avoidance of more than 230%, with respect to the reference plant without capture (1368 kg_CO2/t_CaO).

The IHCaL process is particularly suitable for fuels with a high biogenic fraction (x_bio) and low specific CO₂ emissions (e_CO2,fuel). This combination of properties can be found in high caloric SRF, such as the one considered in this work (class 3 SRF, according to DIN EN ISO, 2021). The utilization of these fuels in the IHCaL combustor allows for net negative CO₂ emissions (−1019 kg_CO2/t_CaO) with very low SPECCA (1.48 MJ_LHV/kg_CO2,av)¹⁴.

Appendices

Appendix 1. Additional input data

The fuel data used for the calculations in this work are provided in Table 7.

Table 7 Detailed fuel data used in this work

The carbonator was modeled based on the work of Lasheras et al. (2011). The main model assumptions are reported in Table 8. The governing equations for the carbonation model were:

$${\displaystyle \begin{array}{l}\frac{1}{K_r}=\frac{d_p}{6\cdot {K}_g}+\frac{1}{K_{ri}}\\ {}{K}_g=\frac{D_{CO2}}{d_p\cdot Sh}\\ {}{K}_{ri}={k}_s\cdot \frac{X_{b,N}\cdot {S}_0\cdot {\rho}_{CaO}}{M_{CaO}}\cdot {\left(1-X\right)}^{2/3}\end{array}}$$

(14)

K _r is the global reaction rate, K_g and K_ri are the diffusion-controlled rate and the chemical rate, respectively, d_p is the particle diameter, and Sh is the Sherwood dimensionless number. X is the conversion, i.e., fraction of active sorbent that has been carbonated. The rest of the constants are defined in Table 8. The active fraction of CaO (X_b,N) was calculated with the deactivation model of Abanades et al. (2005), according to Eq. (15).

$${X}_{b,N}=\frac{f_m\cdot \left(1-{f}_w\right)\cdot {F}_0}{F_0+{F}_R\cdot \left(1-{f}_m\right)}+{f}_w$$

(15)

Table 8 Inputs for the carbonator reactor model

Appendix 2. Detailed results

See Tables 9, 10, 11, 12, and 13.

Table 9 Results from the power calculations in MW_e

Table 10 CO₂ formation in kg_CO2/t_CaO, detailed results

Table 11 Result from CO₂ emission calculations in kg_CO2/t_CaO considering different energy scenarios

Table 12 Results of SPECCA breakdown in MJ_LHV/kg_CO2,av

Table 13 Results for the carbonator reactor model: variables from the process model and reactor model results

Appendix 3. Calculation of heat ratio for different fuels

For the calculation of the heat requirement in the combustor, for any fuel, the heat balance of the system calciner-combustor is performed.

$${\displaystyle \begin{array}{l}\left\{ comb ustion\kern0.5em heat\right\}=\left\{ heat\kern0.33em requirement\kern0.5em calciner\right\}+\left\{ sensible\kern0.5em heat\kern0.5em loss\kern0.5em in\kern0.5em combustor\right\}\\ {}{\dot{Q}}_{comb}={\dot{m}}_{fuel}\cdot {LHV}_{wet}={F}_{CO2}\cdot {\Psi}_1+{\dot{m}}_{fuel}\cdot {\Psi}_2\end{array}}$$

(16)

This equation can be solved using the specific heat capacities of the substances (c_p) and the operational parameters of the IHCaL facility:

$${\displaystyle \begin{array}{l}{\Psi}_1=\left(\varLambda +\varPhi \cdot {X}_{carb}\right)\cdot {E}_{calc}\cdot \Delta {H}_{calc}+\left({T}_{calc}-{T}_{sorb, calc, in}\right)\cdot \left[\left(\varLambda +\varPhi \cdot {X}_{carb}\right)\cdot {\overline{c}}_{p, CaCO3}+\left(1-{X}_{carb}\right)\cdot {\overline{c}}_{p, CaO}\right]\\ {}{\Psi}_2= AFR\cdot {c}_{p, air}\cdot \left({T}_{comb}-{T}_{preheat}\right)+{c}_{p, fuel}\cdot \left({T}_{comb}-{T}_0\right)\\ {}{\dot{Q}}_{comb}={F}_{CO2, plant}\cdot {\Psi}_1\cdot {\left(1-\frac{e_{CO2, fuel}}{M_{CO2}}\cdot {\Psi}_1+\frac{\Psi_2}{LHV}\right)}^{-1}\end{array}}$$

(17)

The values assumed for the calculation of the curves of Fig. 3 are reported in Table 14.

Table 14 Values for the calculation of typical heat ratios for different e_CO2,fuel