Study of lime flow rate on calcium looping carbon capture for steel plant

Abstract

Calcium looping carbon capture was simulated for a steel plant and three fuel cases were studied: coal, coke, and half coal/coke. The steelmaking production was about 1.2 t of liquid steel per hour at >99.85% purity by weight. This study evaluated the effect of carbon capture efficiency and raw material cost on the lime flow rate in the calcium looping section. Increasing the lime feeding flow rate achieved higher efficiency but at a higher cost. The overall captured efficiency was higher than the flue gas captured efficiency at a low lime feeding flow rate but lower at a high lime feeding flow rate. Then, the overall captured efficiency tended to decrease at high lime feeding flow rate because raw materials were used too much for carbon capture and caused overall carbon dioxide emission more than the content of captured carbon dioxide in the product stream. Coal as fuel plays an important role in carbon emission reduction as one of the raw materials, along with the utility cost and the raw material cost in calcium looping section.

Keywords

steelmaking process carbon capture calcium looping

Introduction

According to the World Steel Association, the total production of crude steel increased markedly by about 1878.5 Mt in 2022.¹ However, this industry has the highest carbon dioxide emission level among heavy industries. In 2019, the iron and steel industries emitted about 2.6 Gt of carbon dioxide annually, mainly from using coal to generate heat and produce coke.² To achieve climate goals, carbon dioxide emission should be reduced from 1.4 t_CO2/t_{crude steel} to 0.6 t_CO2/t_{crude steel} by 2050. Therefore, carbon capture technology plays a vital role in reducing carbon dioxide emissions.³

A blast furnace (BF) is central to the primary iron production process. Most simulations consider sub-reactors, representing the reaction zone at different temperatures, with the model containing four reactors from top to bottom. Gibbs's reactor models can represent low-temperature, middle-temperature, high-temperature, and raceway reactors.^4,5 A BF using the RGibbs reactor model, which considers the condition of Gibbs free energy minimization of the reacting system being at equilibrium to calculate the product mixture components, operates with a temperature range of 800–1600°C. Most of the ferric oxide (Fe₂O₃) and ferrous oxide (FeO) react with carbon monoxide and carbon to produce iron. Flue gas from the furnace is sent to an additional reactor to oxidize carbon dioxide that is subsequently sent to the carbon capture unit.^6,7 The combustion of fuels occurs in the base (the raceway section) of the BF, where a typical temperature is in the range 2000–2300°C.^8,9 The fuel type (coal and coke) and the amount are some of the factors that affect the air requirements in the iron and steel production process, the flue gas composition, and the amounts of carbon dioxide and flue gas. These can all affect the carbon capture unit; thus, accurate specification is required to achieve high efficiency.¹⁰ The basic oxygen furnace (BOF) is the main unit in the steel plant that produces products with low concentrations of carbon.¹¹ The liquid iron content is increased by adding scrap. Furthermore, contamination in the liquid iron is decreased using oxygen. However, oxide compounds can be produced from contamination, resulting in the production of flue gas, called basic oxygen flue gas (BOFG).¹²

The post-combustion carbon capture technique is suitable for steel plants. Calcium looping is one of the carbon capture techniques.¹³ Carbonation is the reaction of metal oxide and carbon dioxide, which separates carbon dioxide from the flue gas to form a metal carbonate. Subsequently, the metal carbonate is calcined to capture carbon dioxide, while metal oxide is regenerated in a looping process. Calcium is the main metal used in this technique.¹⁴ The calcium loop model concept has two main units, being the two reactors that represent the reaction inside the model. Flue gas from the furnace is sent to the carbonator reactor. By feeding lime (CaO) from the regenerator reactor, the reaction of the lime combined with CO₂ in the flue gas produces limestone (CaCO₃). Then, the combustion of limestone occurs in the calciner reactor. Next, the limestone is decomposed into CO₂ and lime using coal as the fuel. CO2 is sent to the carbon dioxide liquefaction unit. In addition, lime is recycled to the carbonator reactor as raw material to combine with CO₂ from the flue gas.¹⁵ Carbon dioxide liquefaction is an important unit for carbon dioxide transportation. The specific volume of carbon dioxide in the liquid phase is lower than in the gas phase; therefore, the cost is reduced by decreasing the size of the pipeline and using a pump instead of a compressor.¹⁶ The thermodynamics parameters were investigated which affect the performance of the carbon dioxide liquefaction process. The compression, cooling, and separation units are utilized for the purification of liquid carbon dioxide. The gas was compressed and cooled to the optimum condition for carbon dioxide condensation while other species were separated since they had different phases. Therefore, the acquired liquid carbon dioxide has high purity and is suited for transportation.¹⁷

The objective of this research was to evaluate the effect on carbon capture efficiency of the lime flowrate in the calcium looping section. Efficiency was studied in two classes. The first scoped only the carbon dioxide captured from the flue gas outlet stream from the steel plant, called the flue gas captured efficiency. The second scoped overall carbon dioxide emission using the raw material and utility in the calcium looping section, called the overall captured efficiency. The overall captured efficiency contains processed carbon dioxide from the flue gas and the carbon dioxide emission factor from the raw materials and utility manufacturing. In addition, the raw materials cost was evaluated to study how the cost depended on the amount of lime used. This research studied the flue gas from the steel plant based on different three cases: using coke, using coal to produce coke in the coke oven batteries unit, and using 50% coke and 50% coal as fuel.

Process simulation

The Aspen Plus V.12 software was used for the modeling and simulation of the following processes: ¹⁸ ironmaking in the BF; the desulfurization unit; steelmaking in the BOF using the coke oven batteries for coke production from coal; the carbon monoxide converter; and carbon dioxide capture based on membrane separation technology.

Modeling and simulation of steel plant with coke oven batteries

Figure 1 presents the process flow diagram of the steel plant with coke oven batteries. When using produced coke, coal is fed into the coke oven batteries to produce coke and coke oven gas (COG) for the ironmaking unit. The raw materials used in this process are iron ore, limestone, and quartzite from mining, coal and coke as fuels, calcium carbide for desulfurization, and air for combustion and conversion. The main product of the unit is desulfurized liquid iron, with by-products consisting of carbon, lime, magnesium oxide, calcium silicate, calcium sulfide, aluminum oxide, silicon, silica, and ash. The flue gas contains carbon monoxide, carbon dioxide, nitrogen, oxygen, hydrogen, and water vapor.

This research used a hematite type of iron ore from an Australian source. The composition and size distribution are referred to elsewhere.¹⁹ Coke can be produced in coke oven batteries using coal as the raw material based on the decomposition (pyrolysis) of coal that occurs inside the oven.²⁰ As a result, the products from the coke oven batteries include coke and COG. The COG is processed in the by-products recovery section to obtain outputs, such as tar and BTX (benzene, toluene, and xylene). Therefrom, the COG is sent to the ironmaking unit.²¹ This research used Australian bituminous coal with heating values of 24,000 kJ/kg as the raw material for the coke production unit with proximate and its ultimate composition analysis reported elsewhere.^22,23 The composition of the acquired coke, COG, coal tar, and BTX were referenced from the work of Qin and Chang.²⁴

Australian bituminous coal was used as the raw material for coke production. The coke oven batteries were represented using the RYield model. The temperature inside the oven was set at 1100°C. Coke was fed into the RACEWAY reactor, while the raw COG was sent to the treatment unit. The COG was sent to the RACEWAY reactor, while other gases were discharged from the system. Feeding 493.9 kg/hr of coal produced 353.26 kg/hr of coke and 80.88 kg/hr of COG from the coke oven batteries unit.

Australian iron ore, limestone, and quartzite at 1585.68, 216.65, and 129.52 kg/hr, respectively, were fed into the BF, represented by the reactors LOW, MIDDLE, HIGH, and RACEWAY, which had reactions occurring in different temperature zones and reaction types inside the BF. All these reactors were represented by RGibbs models.

In the LOW reactor, the temperature was set at 900°C and hematite (ferric oxide) and magnetite (ferrous ferric oxide) were converted to wüstite (ferrous oxide) by the presence of carbon monoxide, as shown in chemical equations (1) and (2):

3 {Fe}_{2} O_{3} + CO \to 2 F e_{3} O_{4} + C O_{2}

(1)

F e_{3} O_{4} + C O \to 3 F e O + C O_{2}

(2)

In the MIDDLE reactor, the temperature was set at 1200°C, resulting in some wüstite being converted to iron in the solid phase by carbon monoxide, as shown in chemical equation (3):

FeO + CO \to Fe + {CO}_{2}

(3)

In the HIGH reactor, the temperature was set at 1550°C, resulting in all the wüstite being converted to iron by the presence of carbon monoxide and coke from the RACEWAY reactor, as shown in chemical equations (3) and (4):

FeO + C \to Fe + CO

(4)

In the RACEWAY reactor, the temperature was set at 2200°C, where using air, complete and incomplete combustion of coke occurred based on the two reactions, as shown in chemical equations (5) and (6):

C + O_{2} \to {CO}_{2}

(5)

2 C + O_{2} \to 2 CO

(6)

In the DESULFU reactor, the temperature was set at 1550°C and 6 kg/hr of calcium carbide was added to the reactor. The carbon in the calcium carbide captured sulfur in the stream and converted it into calcium sulfide. Therefore, ferrous sulfide was converted and increased the iron content based on chemical equation (7):

FeS + {CaC}_{2} \to Fe + CaS + C_{2}

(7)

The liquid iron was separated from the slag in the solid phase. The liquid iron was sent to the basic oxygen steelmaking unit, where 200 kg/hr of steel scrap with the same composition of liquid steel and air were fed into this unit for contamination elimination and to increase the iron content. Inside the BOF, chemical equation (6) occurred. Then, the product stream from the BOF unit was separated. The gas product (BOF gas) was mixed with the BF gas stream.

In part of the gas phase, carbon monoxide was important in reducing iron oxide to iron inside the BF. This gas mainly occurred in the RACEWAY reactor due to the incomplete combustion of coke with oxygen in the AIRRACE stream. However, this process produced carbon dioxide by the complete combustion of coke inside the RACEWAY reactor and as the by-product of iron oxide reduction. Therefore, inside the BF, the amount of carbon monoxide decreased, while the quantity of carbon dioxide increased.

Then, the gas phase was sent to the plate heat exchanger (HX) as a hot stream for heating the AIRP stream that was compressed to five bars from the centrifugal compressor (COMP). Next, the air stream (3350 kg/hr for coal and 2830 kg/hr for coke) was separated into three streams. The first was sent to the RACEWAY reactor for coke combustion, the second was sent to the BOF unit to increase the iron purity, and the third was sent to the CONVERT reactor for oxidation of carbon monoxide.

When the GAS4 stream exited the BF, the stream was mixed with the BOFG stream from the BOF. Then, this mixed stream was fed into the CONVERT reactor type RGibbs model. In this reactor, the oxidation of carbon monoxide to carbon dioxide occurred in the presence of oxygen from the AIRCON stream at 1500°C, as shown in chemical equation (8):

2 CO + O_{2} \to 2 {CO}_{2}

(8)

The leftover air stream was fed into the CONVERT reactor. Lastly, the flue gas was sent to the carbon dioxide capture unit.

Modeling and simulation of carbon capture based on calcium looping methodology

Figure 2 presented the process flow diagram of carbon capture based on calcium loop methodology. Carbonation is the reaction between metal oxide and carbon dioxide. This technique separates carbon dioxide from the flue gas to form a metal carbonate. In the current study, the metal carbonate was calcined to capture carbon dioxide, while the metal oxide was regenerated in a looped system. Calcium was the primary metal for this technique.¹⁴ The calcium loop model concept has two central units, namely the two reactors that represent the reaction inside the model. Flue gas from the furnace was sent to the carbonator reactor. Lime (CaO) fed from the regenerator reactor combined with CO₂ in the flue gas to produce limestone (CaCO₃), which was combusted in the regenerator reactor, resulting in its decomposition into CO₂ and lime. The CO₂ was sent to the storage unit. Furthermore, the lime was recycled back to the carbonator reactor as raw material to combine with CO₂ from the flue gas.¹⁵

The flue gas was fed into the reactor (CARBO) type RCSTR model at 25 kg/hr lime from an external source and recycled inside the calcium looping unit. This reactor was set at 650°C (Figure 2). The carbonation reaction occurred as shown in chemical equation (9):

CaO + {CO}_{2} \to {CaCO}_{3}

(9)

Carbon dioxide from the flue gas mainly reacted with lime to form limestone, based on the particle size distribution of lime²⁵ and coarse limestone.²⁶ Therefore, the flue gas was leaned in carbon dioxide. The carbonation reaction was a first-order reaction with an activation energy of 28.3 kJ/mol and a pre-exponential factor of 2.22 × 10⁻⁴ mol/m²-s.²⁷

The product from the carbonator reactor was fed to the cyclone, wherein the solid and gas phase products were separated. The gas phase product (lean carbon dioxide) was cooled using a HX and discharged from the carbon capture unit. The solid stream fed to the reactor (CALCINE) type RCSTR model. This reactor was set at 900°C. Limestone decomposed to lime and carbon dioxide, as shown in the chemical equation (10):

{CaCO}_{3} \to CaO + {CO}_{2}

(10)

Additionally, coal and oxygen were fed for combustion. The calcination reaction was a first-order reaction with the activation energy of 203.8 kJ/mol and a pre-exponential factor of 1.1 × 10⁸ /s.²⁸

The calciner reactor products were fed into the cyclone to separate the solid and gas phases. Carbon dioxide was discharged from the carbon capture unit, while the solid phase product (lime) was purged from looping by 40% because of the presence of ashes and contaminants in the stream and the decreasing amount of accumulating lime in the calcium looping. The remaining lime was fed to the carbonator reactor to separate the carbon dioxide from the flue gas in the looped system.

The amount of lime affected the carbon captured efficiency and raw material cost. Adding more lime increased the carbon capture efficiency as well as resulting in higher raw material costs. Therefore, it was necessary to identify the proper amount of lime to feeding in that produced the highest carbon captured efficiency with as a low raw material cost as possible. Sensitivity analysis was studied by varying the lime feeding flow rate in the range 1–50 kmol/hr.

The overall captured efficiency is the percentage of captured carbon dioxide of the overall carbon dioxide emission factor from raw materials and utility manufacturing. The method of calculating overall captured efficiency uses the ratio of high-purity carbon dioxide flow rate to the overall carbon dioxide emission factor, as shown in equation (11). The flue gas captured efficiency is the percentage of carbon dioxide captured from the flue gas stream outlet of the steel plant. The flue gas captured efficiency is calculated as the ratio of the difference between the carbon dioxide in the flue gas stream and the carbon dioxide leaving the calcium looping unit in the lean gas stream to the carbon dioxide in the flue gas stream, as shown in equation (12).²⁹ The carbon dioxide emission factor of raw materials (lime, coal, and oxygen) and utility(electricity) are presented in Table 1.

Overall captured efficiency = \frac{{CO}_{2} flow rate in production stream}{Overall {CO}_{2} emission}

(11)

Flue gas captured efficiency = \frac{{CO}_{2} flow rate in flue gas stream - {CO}_{2} flow rate in lean gas stream}{{CO}_{2} flow rate in flue gas stream}

(12)

Table 1.

Carbon dioxide emission factor of raw materials and utility.

Raw material/Utility	CO₂ emission	Unit	Reference
Lime	0.75	ton_CO2/ton_lime	³⁰
Coal	94,600	kg_CO2/TJ	³¹
Oxygen	0.236	ton_CO2/ton_oxygen	³²
Electricity	0.394	kg_CO2/(kW-hr)	³³

Lime and coal were raw materials in the calcium looping section that were used for estimating raw material costs. Oxygen was specified as a product from the air separation unit in the plant which was not considered as a raw material cost. The lime and coal prices were THB 0.42/kg and THB 4.71/kg, respectively.^34,35 The plant operated 8000 h/yr.

Figure 3 presents the process flow diagram of carbon dioxide liquefaction unit. The carbon dioxide stream is cooled to 45°C. Then, the leftover solid phase and some condensed water in the stream are separated. Thus, only the gas phase stream is compressed and cooled down to −55°C and 10 bar, which is the optimum condition, acquired from sensitivity analysis of temperature and pressure, which has the highest possible percentage composition of liquid carbon dioxide stream. In this condition, carbon dioxide is liquefied which is ready for transportation.

Results and discussion

There were similar amounts of liquid steel produced from the steel plant in all three cases, being about 1200 kg/hr with iron purity of 99.85%wt. The flue gas stream results from the steel plant for the three cases (100% coal, 50% coal + 50% coke, and 100% coke) are presented in Table 2. Changing the fuel from coal to coke produced higher amounts of carbon dioxide but lower amounts of total flue gas. For model validation, a comparative study in coke oven batteries and a BF with coke oven batteries in the solid and gas phases were utilized from elsewhere.^4,10 The results of the model validation are presented in Table 3.

Table 2.

Flue gas stream results for three cases

	Coal	50:50	Coke
Temperature (°C)	750.5	738.82	723.88
Pressure (bar)	4.91	4.93	4.94
	Mass flow (kg/hr)
CO	1.81	1.76	1.6
CO₂	1274.08	1454.08	1639.94
H₂O	637.42	408.46	179.44
H₂	0.04	0.02	0.01
O₂	4.68	5.78	7.91
N₂	2489.26	2287.72	2101
Total mass flow	4407.29	4157.82	3929.9

Table 3.

Results of model validation of coke oven batteries (kg/t_coke) and blast furnace with coke oven batteries in solid phase (kg/t_{Liquid iron}) and gas phase (kg/t_BF)

Coke oven batteries	Coal	Coke	COG
¹⁰	1396.9	1000	195.4
Coal	1398.12	1000	228.95

Blast furnace(Solid)	Fe	C	Si	Mn	SiO₂	CaSiO₃	Al₂O₃	CaO	CaS	MgO	MnO
⁴	949.29	45.04	5.3	2.5	93.31	0	37.84	126.74	0	18.72	1.38
Coal	944.07	55.53	0.4	0	92.83	236.45	52.67	0	4.35	4.18	0

Blast furnace(Gas)	CO	CO₂	H₂O	H₂	N₂	CH₄
⁴	195.83	320.07	14.87	2.79	465.98	0.48
Coal	166	263.38	175.62	15.46	379.49	0

COG: coke oven gas.

The validation results indicated there were different concentrations in the coke oven batteries and the BF solid and gas streams because of the different compositions of iron ore, coal, and other raw materials. Additionally, the results for the RYield model in the coke oven batteries agreed with the work of Neto et al.¹⁷ and the RGibbs model in the BF depended on the Gibbs free energy potential. Furthermore, the different types and compositions of raw materials caused different results.

In this study, lime was used as a sorbent for calcium looping instead of limestone. By varying the mass flow rate of lime and limestone of the coal case as shown in Figure 4 indicated at the same amount, the percentage of flue gas captured by feeding the make-up lime case is higher than by feeding the make-up limestone case. Especially in the range of 300–2000 kg/hr, the flue gas captured efficiency is distinctly different. The 90% of flue gas captured efficiency was acquired by using about 1000 kg/hr of lime, however, using more than 2500 kg/hr of limestone still did not provide 90%. Furthermore, using lime as a make-up stream could decrease the required volume of production units because the amount of usage is lower. Therefore, this study decided to use lime as a make-up stream instead of limestone.

The percentage of the purge stream was set at 40%. By varying the %purge in the range of 5–50% solids at 95% flue gas carbon captured of the calcium looping in the coal case as shown in Figure 5 indicated, the purge flow rate and lime purity of the purge stream tended to increase when %purge was higher. In the 5–10% purge, the purity of the lime purge stream was 94.5–96.5% while 40% purge acquired the purge stream about 2300 kg/hr with 98% lime purity. At this point, the purge stream has a high %purity of lime which can be used in other processes such as using as a flux for ironmaking unit or else. Moreover, by specification the percentage of purge at 40%, this means the amount of recycled lime is only 60% that accumulated in the process. Therefore, the possible amount of deactivated sorbent tended to be fewer than in other cases. Moreover, feeding new make-up lime to keep the percentage captured helped to increase activated sorbent for carbon capturing. Therefore, the impact of thermal deactivation will be minimal because of 40% of purging and feeding make-up lime. Therefore, this study chose to use a 40% purge.

The results of varying the lime feeding flow rate in the three cases (100% coal, 50% coal + 50% coke, and 100% coke) for the flue gas captured efficiency, overall captured efficiency, carbon dioxide emissions, and raw material cost are shown in Figure 6. Carbon dioxide emission from raw materials and utility and overall captured efficiency of 100% coal, 50% coal + 50% coke, and 100% coke are presented in Figure 7. Flue gas captured efficiency, overall captured efficiency, carbon dioxide emissions, and raw material cost of three cases in Figure 8.

Figure 6(a) shows that increasing the lime feeding flow rate increased the flue gas captured efficiency, overall captured efficiency, and raw material cost. When the feeding lime flow rate was about 320 kg/hr, there were changes in the intersection points between the two efficiency lines. At a lime flow rate lower than 320 kg/hr, the overall captured efficiency was higher than the flue gas captured efficiency. However, at a lime flow rate higher than 320 kg/hr, the overall captured efficiency was lower than the flue gas captured efficiency. Above 320 kg/hr of lime used a high number of raw materials that caused excessive carbon dioxide emissions from the raw material manufacturing process, while carbon dioxide from the flue gas stream remained constant. Therefore, above 320 kg/hr of lime, increasing lime usage produced a greater amount of carbon dioxide captured from the flue gas stream which meant a greater increase in the flue gas captured efficiency. While above 320 kg/hr of lime, increasing lime usage caused excessive carbon dioxide emissions from the raw material manufacturing process, which meant a lower increase in the overall captured efficiency. The maximum point of the overall captured efficiency occurred at a lime flow rate of about 825 kg/hr. Even the flue gas captured efficiency still increases but the overall captured efficiency tends to decrease. This means above 825 kg/hr of lime, raw materials were used too much for carbon capture and caused overall carbon dioxide emission more than the content of captured carbon dioxide in the product stream. Therefore, the carbon capture unit may be inappropriate in terms of the overall captured efficiency at this point. At 710 kg/hr of lime, the raw material cost changed because of the changed amount of coal usage, as shown in Figure 6(a). The increase in coal usage tended to decrease when the lime flow rate was higher than 710 kg/hr because the increase in the required heat duty of the calciner tended to decrease.

Figure 6(b, c) shows similar patterns to Figure 6(a), albeit with the different intersection points being 365 and 415 kg/hr of lime flow rate, respectively. Moreover, the maximum point of the overall captured efficiency occurred at a lime flow rate of about 935 and 1045 kg/hr of lime flow rate, respectively.

In addition, Figure 7(b, c) had similar patterns to Figure 7(a), albeit with the different intersection points being 785 and 880 kg/hr of lime flow rate, respectively. The captured efficiency and raw material cost at the intersection point for all three cases (100% coal, 50% coal + 50% coke, and 100% coke) are presented in Table 4. The flue gas captured efficiency, overall captured efficiency, and raw material cost at the maximum point for all three cases (100% coal, 50% coal + 50% coke, and 100% coke) are presented in Table 5.

Table 4.

Captured efficiency and raw material cost at intersection point for three cases

Case	Lime flow rate (kg/hr)	% Captured efficiency	Raw material cost(MTHB/yr)
Coal	320	44.83	5.17
50:50	365	45.27	5.94
Coke	415	45.92	6.78

Table 5.

Captured efficiency and raw material cost at maximum point of the percentage overall captured efficiency for three cases

Case	Lime flow rate (kg/hr)	% Flue gas efficiency	% Overall efficiency	Raw material cost(MTHB/yr)
Coal	825	84.82	63.16	10.72
50:50	935	86.23	64.12	12.33
Coke	1045	87.6	65.06	14.02

Figure 8(a) shows the flue gas captured efficiency for 100% coal was the highest as was the overall captured efficiency, shown in Figure 8(b), while 100% coke had the lowest flue gas captured efficiency and overall captured efficiency. The flue gas captured efficiency for three cases tended to converge at 2000 kg/hr of lime, which was about 95% captured. The overall captured efficiency of 100% coal was the highest value in three cases at first however the overall captured efficiency of 100% coke became the highest value when the 100% coal case passed the maximum point of the overall captured efficiency. Figure 8(c) shows that the raw material cost for 100% coal was the lowest.

By specification of the percentage carbon capture at 95%wt. for evaluating, the amount of lime for coal, 50:50, and coke case were 2411.33, 2130.94, and 1962.71 kg/hr, respectively. The overall captured efficiency of coal, 50:50, and coke case were 48.02, 53.39, and 57.59%wt., respectively. The liquid carbon dioxide in coal, 50:50, and coke case are presented in Table 6. The liquid carbon dioxide purity of coal, 50:50, and coke case were 98.46, 98.71, and 98.73%wt., respectively.

Table 6.

Liquid carbon dioxide stream results of three cases

	Coal	50:50	Coke
Temperature (°C)	−55	−55	−55
Pressure (bar)	10	10	10
	Mass flow (kg/hr)
CO₂	1318.62	1516.17	1705.57
N₂	0.09	0.07	0.08
O₂	10.12	8.41	9.44
H₂O	10.29	11.28	12.4
H₂	0.08	0.05	0.06
Total mass flow	1339.2	1535.98	1727.55

Conclusions

This study investigated, simulated, and compared making iron and steel using coke from coal in the coke oven batteries process using purchased coke connected to a carbon dioxide capture unit based on membrane separation technology in terms of the production and CO₂ capture feasibility. These processes were analyzed using the Aspen Plus One V.12 industrial simulation software.

The production capacity of liquid steel in the steel plant was about 1200 kg/hr, with an iron purity of 99.85% wt. Flue gas levels from the steel plant for the three cases (100% coal, 50% coal + 50% coke, and 100% coke) were 4407.29, 4157.82, and 3929.9 kg/hr, respectively, with 28.91, 34.97, and 41.73% wt. of carbon dioxide purity, respectively.

The sensitivity analysis results in the calcium looping unit showed that an increase in the lime feeding flow rate resulted in higher flue gas captured efficiency, overall captured efficiency, and raw materials cost. The flue gas and overall captured efficiency lines had an intersecting point. The overall captured efficiency was higher than the flue gas captured efficiency at a low lime feeding flow rate, whereas, on the other hand, the overall captured efficiency was lower at a high lime feeding flow rate. Then, the overall captured efficiency tended to decrease. This means raw materials were used too much for carbon capture and caused overall carbon dioxide emission more than the content of captured carbon dioxide in the product stream.

The flue gas stream from the steel plant model using coal as the fuel had the highest flue gas captured efficiency and overall captured efficiency with the lowest raw material cost for the same amount of lime feeding. Therefore, installing coke oven batteries to produce coke from coal would be a better solution for reducing carbon dioxide in the flue gas stream of a steel plant. Additionally, using coal as fuel reduced the carbon dioxide emission from raw materials and reduced the utility and raw material costs in the calcium looping section.

Footnotes

Acknowledgments

Funding for this work was supported by a Master's Degree Study Plan A, Faculty of Engineering, Kasetsart University, Bangkok, Thailand (Research project code 65/16/CHEM/M.Eng).

Authors’ contributions

Chayut Sungsook: Resources, modeling, simulation, validation, and writing. Thongchai Rohitatisha Srinophakun: Supervision, conceptualization, investigation, and review & editing.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for this work was supported by a Master's Degree Study Plan A, Faculty of Engineering, Kasetsart University, Bangkok, Thailand (Research project code 65/16/CHEM/M.Eng).

ORCID iD

Thongchai Rohitatisha Srinophakun

References

WSA [internet]. Brussels (Belgium): 2022[cited 2023 May 14]. Available from: https://worldsteel.org/media-centre/press-releases/2023/ December-2022-crude-steel-production-and-2022-global-totals/

IEA [internet]. Paris (France): 2020[cited 2023 November 10]. Available from: https://www.iea.org/reports/ iron-and-steel-technology-roadmap

Lin

Yang

, et al. Low-carbon development for the iron and steel industry in China and the world: status quo, future vision, and key actions. Sustainability 2021; 13: 12548.

Suzuki

Hayashi

Kuribara

, et al. Quantitative evaluation of CO2 emission reduction of active carbon recycling energy system for ironmaking by modeling with aspen plus. ISIJ Int 2015; 55: 340–347.

Wang

Jiang

Wang

, et al. Process simulation of blast furnace operation with biomass syngas injection for clean production. Energy Proc 2019; 2: 996.

Zhang

Tang

, et al. Conceptual design and simulation analysis of thermal behaviors of TGR blast furnace and oxygen blast furnace. Sci China Ser E-Technol Sci 2010; 53: 85–92.

Luca

Petrescu

. Membrane technology applied to steel production: investigation based on process modelling and environmental tools. J Clean Prod 2021; 294: 126256.

Burgess

. Fuel combustion in the blast furnace raceway zone. Prog Energy and Combust Sci 1985; 11: 61–82.

Yilmaz

Wendelstorf

Turek

. Modeling and simulation of hydrogen injection into a blast furnace to reduce carbon dioxide emissions. J Clean Prod 2017; 154: 488–501. doi: 10.1016/j.jclepro.2017.03.162

10.

IEAGHG [internet]. Cheltenham (England): 2013[cited 2023 May 12]. Available from: https://ieaghg. org/publications/technical-reports/ reports-list/9-technical-reports/ 1001-2013-04-iron-and-steel-ccs-study-techno-economics-integrated-steel-mill

11.

Naidu

Sheridan

Dyk

LDV

. Basic oxygen furnace slag: review of current and potential uses. Miner Eng 2020; 149: 106234.

12.

Wang

. Energy saving technologies and optimisation of energy use for decarbonized iron and steel industry [dissertation]. Durham (England): Durham University, 2022.

13.

Koornneef

Harmelen

Horssen

, et al. Chemistry, Emission Control, Radioactive Pollution and Indoor Air Quality. InTech; 2011. Chapter 2, Carbon dioxide capture and air quality; p. 17–44. doi: 10.5772/1030

14.

Eide

Anheden

Lyngfelt

, et al. Novel capture processes. Oil & Gas Sci Technol 2005; 60: 497–508.

15.

Mateo

. Assessment of Calcium Looping as a solution for CO2 capture in the steel production process . master’s thesis, Universitat Politècnica de Catalunya, Barcelona (Spain), 2014.

16.

Zhang

Wang

Massarotto

, et al. Optimization of pipeline transport for CO2 sequestration. Energy Convers Manag 2006; 47: 702–715.

17.

Aliyon

Mehrpooya

Hajinezhad

. Comparison of different CO2 liquefaction processes and exergoeconomic evaluation of integrated CO2 liquefaction and absorption refrigeration system. Energy Convers Manag 2020; 211: 112752.

18.

AspenTech [internet]. Bedford (MA): 2020[cited 2023 March 14]. Available from: https://www. aspentech.com/en

19.

Moreno-Atanasio

O’Dea

, et al. Experimental study on the physical properties of iron ore granules made from Australian iron ores. ISIJ Int 2019; 59: 253–262.

20.

Sciazko

Mertas

Kosyrczyk

, et al. A predictive model for coal coking based on product yield and energy balance. Energies 2020; 13: 4953.

21.

Neto

GWF

Leite

MBM

Marcelino

TOAC

, et al. Optimizing the coke oven process by adjusting the temperature of the combustion chambers. Energ 2020; 217: 119419. doi: 10.1016/j.energy.2020.119419

22.

Gani

ZFA

Rathnam

Wall

, et al. Ignition and combustion characteristics of Australian coals in air (O₂/N₂) and oxy-fuel (O₂/CO₂) conditions. Proc. of the Aust. Combust. Symp. 2007:126–129.

23.

IEA [internet]. Paris (France): 2023[cited 2023 November 10]. Available from: https://iea.blob. core.windows.net/assets/ba921d99-b415-4d60-86fc-536ca71773a9/ Coal_Documentation_2023_July.pdf

24.

Qin

Chang

. Modeling, thermodynamic and techno-economic analysis of coke production process with waste heat recovery. Energy 2017; 141: 435–450. doi: 10.1016/j.energy.2017.09.105

25.

Hwidi

Izhar

TNT

Saad

FNM

, et al. Characterization of quicklime as raw material to hydrated lime: effect of temperature on its characteristics. Green Design Manuf 2018: 1. doi: 10.1063/1.5066668

26.

Majeed

Qudseih

Edens

, et al. Limestone particle size, calcium and phosphorus levels, and phytase effects on live performance and nutrients digestibility of broilers. Poult Sci 2020; 99: 1502–1514.

27.

Sedghkerdar

Mahinpey

Ellis

. The effect of sawdust on the calcination and the intrinsic rate of the carbonation reaction using a thermogravimetric analyzer (TGA). Fuel Process Technol 2013; 106: 533–538.

28.

Guler

Dollimore

Heal

. The investigation of the decomposition kinetics of calcium carbonate alone and in the presence of some clays using the rising temperature technique. Thermochim Acta 1982; 54: 187–199.

29.

Thimsen

Maxson

Smith

, et al. Results from MEA testing at the CO₂ technology centre mongstad part I: post-combustion CO₂ capture testing methodology. Energy Procedia 2014; 63: 5938–5958.

30.

Shan

Liu

Guan

. CO₂ Emissions from China’s lime industry. Appl Energy 2016; 166: 245–252.

31.

Gómez

Watterson

. 2006 IPCC guidelines for national greenhouse gas inventories. 2006. Chapter 2, Stationary combustion; p. 2.1-2.47.

32.

Praxair [internet]. Danbury (CT):2011 [cited 2023 August 13]. Available from: https://www.linde.com/-/ media/linde/merger/documents/sustainable-development/praxair-white-paper-impact-of-blast-oxygen-enrichment-w-disclaimer-r1.pdf?la = en

33.

EPPO [internet]. Bangkok (Thailand): 2023[cited 2023 September 4]. Available from: https://www.eppo. go.th/images/Energy-Statistics/ energyinformation/Energy_Statistics/Emission/09Emission.pptx

34.

Indiamart [internet]. Noida (India): 2023[cited 2023 August 13]. Available from: https://www. indiamart.com/proddetail/quick-lime-powder-11664652212.html? pos=10&kwd=quicklime&tags=BA||||7613.566|Price|proxy|||LSlc

35.

Asiaplus [internet]. Bangkok (Thailand): 2023[cited 2023 September 4]. Available from: https://www. asiaplus.co.th/asps/research_file.php?id=70064& file=1. Thai.