Comparison of Energy Consumption and CO 2 Emission for Three Steel Production Routes — Integrated Steel Plant Equipped with Blast Furnace , Oxygen Blast Furnace or COREX

1 School of Energy and Environmental Engineering, University of Science and Technology Beijing, Beijing, 100083, China; songjiayuan2015@outlook.com (J.S.); baocheng@mail.tsinghua.edu.cn (C.B.) 2 School of Metallurgical and Ecological Engineering, University of Science and Technology Beijing, Beijing 100083, China; anjunxu@metall.ustb.edu.cn * Correspondence: zyjiang@ustb.edu.cn; Tel.: +86-10-6233-2741; Fax: +86-10-6233-2741


Introduction
Steel is the world's most popular construction material due to its durability, processability, and cost.However, producing steel creates high energy consumption and CO 2 emissions.According to the World Steel Association, the production of crude steel reached 1691 million tons in 2017 and the large amount of crude steel production resulted in about 3.1 × 10 11 GJ of energy consumption and 3043.8 million tons of CO 2 emissions [1].The blast furnace (BF)-basic oxygen furnace (BOF) route is the dominant steel production route in the world, and its ironmaking process contributes approximately 70% of the above energy consumption and CO 2 emissions [2].Many studies have shown that adopting commercially available energy-saving technologies for BF has significantly reduced the energy consumption of the ironmaking process [3] and almost reached the thermodynamic limits of BF [4].Thus, in order to minimize the energy consumption and CO 2 emissions of the ironmaking process, alternative liquid iron production technologies to BF such as top gas recycling-oxygen blast furnace (TGR-OBF) and COREX have been proposed and developed.
The oxygen blast furnace (OBF) is a type of metallurgical furnace used for producing liquid iron, which was proposed by Wenzel and Gudenau in 1970 [5].However, the first generation of OBF had the problem of overheating in the lower part of the furnace and thermal shortage in the upper part of the furnace.In order to solve these problems, some TGR-OBF processes such as Fink [6], Lu [7], Nippon Kokan Steel (NKK) [8], Tula [9], full oxygen blast furnace (FOBF) [10], and Ultra-Low CO 2 Steelmaking initiative (ULCOS) [11] have been proposed.In the TGR-OBF process, iron ore, coke, and flux are supplied through the top of OBF in succession, while normal temperature oxygen and pulverized coal are injected into the OBF by tuyeres at the bottom.The preheated top gas is recycled into the OBF at the lower stack, after the removal of CO 2 and H 2 O. Compared with the BF process, the two key characteristics of the TGR-OBF process are the injection of pure oxygen as an oxidant instead of a hot blast and the recycling of top gas as a reductant after the removal of CO 2 and H 2 O and preheating.These characteristics increase the amount of pulverized coal in the OBF to more than 300 kg/t-hot metal [12] and reduced the coke ratio to less than 200 kg/t-hot metal [13].As a result of the study of a pilot plant, an 8 m 3 experimental TGR-OBF built during the Ultra-low CO 2 steelmaking initiative phase I (ULCOS I) showed that the TGR-OBF process was feasible and could significantly reduce the consumption of coke and CO 2 emissions in the ironmaking process [14].Moreover, many researchers have conducted experiments and numerical simulations of the TGR-OBF process, which showed that the TGR-OBF process is a promising alternative liquid iron production technology given its low coke consumption and CO 2 emissions [15][16][17][18].
COREX is an industrially and commercially proven smelting reduction ironmaking process and was created by Siemens Voest-Alpine Industrieanlagenbau Gmbh & Co. (VAI) in the 1970s [19].There are eight COREXs in the world that have been successfully commercialized [20]; two of which are the latest generation of COREX with a capacity of 1.5 million tons of liquid iron per year and were built in China at the Baosteel Luojing steel plant [21].The characteristic of COREX is the separation of the iron reduction and smelting operations into two separate reactors, namely the upper reduction shaft and the lower melter-gasifier [22].In the upper reduction shaft, iron ore and flux are continuously charged from the top of the shaft.They descend by gravity and are reduced into approximately 95% direct reduced iron by the reduction gas generated from the melter-gasifier.In the melter-gasifier, the reduction shaft products-direct reduced iron-is fed from the top of the melter-gasifier for further reduction and melting.Non-coking coal and room temperature oxygen, as the main reductant and oxidant, are injected via the lock hopper system at the top of the melter-gasifier and tuyeres at the bottom of the melter-gasifier, respectively.After being purified by hot gas cyclones, a major part of the top gas from the melter-gasifier is cooled and subsequently added to the reduction shaft.The excess gas and the reduction shaft gas are mixed before the take-over point, called the COREX export gas.When compared with the BF or TGR-OBF process, the COREX process can directly produce liquid iron without using coking coal, sinter, or pellets.The COREX process combines the coking plant, sinter plant, and blast furnace or oxygen blast furnace into a single ironmaking process.Therefore, as an alternative liquid iron production technology to BF, further research into the TGR-OBF and COREX steel production processes are of great significance.
Most of the studies on TGR-OBF and COREX have focused on a reaction unit rather than the ISP.Wenlong et al. estimated the impact of different metallization rates and fuel structures on the energy consumption of the COREX process by using a modified Rist operating diagram [23].Hu et al. compared the CO 2 emissions between the COREX and blast furnace ironmaking system based on carbon element flow analysis [24].Lianzhi et al. used a two-dimensional numerical simulation model to analyze the effect of top gas recycling on the blast furnace status, productivity, and energy consumption.The results showed that TGR-OBF productivity increased by 5.3-35.3%and the energy saving was 27.3-35.9%when compared to blast furnaces [17].Wei et al. conducted exergy analyses of the TGR-OBF ironmaking process [18].Xuefeng et al. investigated the carbon saving potential of the TGR-OBF ironmaking process from the perspective of the relationship between the degree of direct reduction and carbon consumption [25].These studies demonstrated well the effect of operating parameters on energy consumption and CO 2 emissions in the COREX or TGR-OBF process.However, focusing only on a specific unit like the ironmaking process will make it difficult for policymakers to fully understand the alternative ironmaking process from a mill-wide perspective.Up to now, many studies conducted from a plant-wide perspective are based on only the one steel production route.Hooey et al. made a techno-economic comparison of an ISP equipped with the OBF and CO 2 capture by using process and economic models and found that the OBF with CO 2 capture offered a significant potential to reduce the overall CO 2 emissions from an ISP, achieving 47% CO 2 avoidance at a cost of ~$56/t CO 2 for the given assumptions [26].Huachun et al. analyzed the energy consumption and carbon emissions of a conventional ISP by using an integrated material flow analysis model [27].Jin et al. analyzed the energy consumption and carbon emissions of an ISP with the TGR-OBF route, according to an established heat transfer and reaction kinetics mathematical model for the TGR-OBF ironmaking process.The results indicate that the energy consumption of an ISP with the TGR-OBF route was reduced to 14.4 GJ per ton of crude steel; moreover, the direct CO 2 emissions were reduced by 26.2% per ton of crude steel when compared with a conventional ISP [28].Arasto et al. analyzed the technical assessment of the application of OBF with CCS to an ISP.The analysis showed that the CO 2 emissions from an ISP could be significantly reduced by the application of an oxygen blast furnace and CCS [29].Based on the technical analysis, the economic profitability was further evaluated in Reference [30], which found that the investment on OBF or CCS depended on CO 2 , the fuels, and electricity prices.However, the comprehensive energy consumption and CO 2 emissions analysis on three different routes based on the static process model such as an ISP with BF, an ISP with the TGR-OBF, and an ISP with the COREX were not analyzed in detail.
Therefore, in this study, TGR-OBF and COREX static process models were developed based on the mass and heat balance to obtain the material and energy flow parameters of the two ironmaking processes.Combining the operating data from the Jingtang steel plant with established static process models, this research presents a detailed analysis on the material flow, metallurgical gas generation and consumption, electricity consumption and generation, comprehensive energy consumption, and CO 2 emissions of the BF, TGR-OBF, and COREX in an ISP.

Static Process Models
The objective of this study is to compare different steel production routes under the same composition and temperature of the hot metal, as well as the same chemical composition of raw materials (For the TGR-OBF, the raw materials with the same composition as the BF are flux, coke, coal, sinter, and pellets.For the COREX, the raw materials with the same composition as the BF are flux, and coal).Thus, on the basis of the mass and heat balance principles, a TGR-OBF static process model based on our previous research-References [17,28,31] and a COREX static process model based on the Baosteel COREX3000 were established by using the gPROMS Modelbuilder platform.According to the calculation of the two static process models, the consumption of raw material and fluxes, the production and composition of slag and metallurgical gas per ton of hot metal can be obtained.Combining the two static process models with the ISP of the Jingtang, this research presents a detailed comparison and analysis of the three ISPs equipped with the BF, TGR-OBF, and COREX, respectively.

The TGR-OBF Process Model
The flow diagram of the TGR-OBF process is shown in Figure 1.In the TGR-OBF process, most of the top gas is recycled into the furnace as the recycled gas after CO 2 removal and preheating.Part of the recycled gas is injected at the hearth tuyeres, another part of the recycled gas is injected at the lower shaft.This kind of gas recycling in the TGR-OBF process could effectively supplement the heat and reducing atmosphere in the shaft area, and also reduced the combustion temperature in the tuyere area [17].In this study, the total amount of the recycled gas assumed was 600 m 3 /t-HM, the upper and lower recycled gas assumed were 300 m 3 /t-HM, respectively (Supplementary Materials Table S7).This condition could ensure the successful operation of the TGR-OBF and provide high calorific value gas for other processes, as well as ensure the balance of the metallurgical gas in the ISP [31].Moreover, based on our previous research [31], the temperature of the recycled gas at the hearth tuyeres had little effect on the burn-out rate of the pulverized coal (when the temperature of the recycled gas raised from 298 K to 1500 K, the burnout rate of the pulverized coal only changed by 1.5%).Therefore, considering the technical difficulty and the cost, the recycled gas was assumed to be preheated to 1173 K (Table S7).The main reactions in the furnace include the reduction of various iron oxides by CO/H 2 , the decomposition of carbonate, coke, and coal, and the combustion of carbon and reducing gas.In order to make the TGR-OBF process operate successfully under the certain conditions of the raw materials and fuel, it is very important to select reasonable slag-forming practices to make the slag have a better performance.The performance of the slag is closely related to its chemical composition.The basicity is an important parameter of measuring the viscosity of the slag, at the same time, the basicity has great influence on the melting and stability of the slag.Therefore, the binary basicity R 2 (Equation ( 3)) of the slag were included as the constrained equation.Based on the following balance Equations ( 1), (3), and (4) in the TGR-OBF process, the consumption of raw materials (sinter, pellets, limestone, and dolomite), the amount and chemical composition of slag can be obtained by giving the value of chemical composition of raw materials and hot metal (Tables S1, S2, S4, and S6).
• Mass balance: When the consumption of raw materials was calculated, the mass balance equations included the Fe balance equation, the S balance equation, the CaO balance equation, the SiO 2 balance equation, the Al 2 O 3 balance equation, the MgO balance equation, the MnO balance equation, and the P 2 O 5 balance equation.Taking the Fe balance equation as an example, • Binary basicity: • Quantity of slag: where m i is the mass of sinter, pellets, coke, coal, limestone, and dolomite for producing 1 ton of hot metal, kg; ω(Fe) i is the mass fraction of Fe in material i; ω(FeO) slag is the mass fraction of FeO in slag; m HM is the mass of hot metal, which in this study was 1 ton; m slag is the mass of slag, kg; and ω(x) i is the composition of x in material i, %.The utilization of hydrogen in the top gas and the oxidation degree of the top gas are important parameters that can affect the chemical composition of the top gas.Therefore, the utilization of hydrogen in the top gas and the oxidation degree of the top gas were included as the constrained equations.Based on the operating conditions and parameters (Tables S3 and S7) of the TGR-OBF process, the volume and chemical composition of top gas were obtained by calculating the following balance Equations (5)- (9), and (11).

• C balance:
Metals 2019, 9, 364 5 of 15 • The utilization of hydrogen in the top gas: • The oxidation degree of the top gas: here, V rg is the volume of recycled gas; V CO , V H2O , V CO2 , V H2 , V CH4 , and V N2 are the volume of CO, H 2 O, CO 2 , H 2 , CH 4 , and N 2 in top gas for producing 1 ton of hot metal, m 3 , respectively; V H2Or is the volume of water in the top gas, which was produced by the reduction, rg are the volume fraction of CO, CO 2 , H 2 , H 2 O, and N 2 in recycled gas, respectively; ω(C) i , ω(H) i , ω(N) I and ω(C) HM are the mass fraction of C, H, and N in material i and hot metal, respectively; and T gas is the temperature of the metallurgical gas, K.
• Heat balance: where h i , h j are the enthalpy of the chemical reaction, kJ/kg; q i , q j are the sensible heat, kJ/kg; and H l is the enthalpy of loss, kJ.Based on the above calculation results and the oxygen balance equation of the furnace, the volume of oxygen blast can be obtained.Based on the mass balance of the gas in top, bosh, and hearth of the furnace, and the heat balance (Equation ( 10)) of the whole furnace, the volume of recycled gas can be obtained.The volume of gas required for preheating the recycled gas is obtained by the heat balance (Equation ( 10)) of the heater unit.The parameters calculated by the TGR-OBF static process model are shown in Tables S1-S7 (Supplementary Materials).
In the VPSA unit, the recovery rate of CO 2 was set at 90% [32], and the energy consumption was set at 645.7 kJ/kg-CO 2 [33].In the heater unit, the thermal efficiency was set at 86% [11].
Metals 2019, 9, x FOR PEER REVIEW 5 of 16 • H balance: • N balance: • The utilization of hydrogen in the top gas:

•
The oxidation degree of the top gas: here, Vrg is the volume of recycled gas; VCO, VH2O, VCO2, VH2, VCH4, and VN2 are the volume of CO, H2O, CO2, H2, CH4, and N2 in top gas for producing 1 ton of hot metal, m 3 , respectively; VH2Or is the volume of water in the top gas, which was produced by the reduction, m 3 ; v(CO)rg, v(CO2)rg, v(H2)rg, v(H2O)rg, v(N2)rg are the volume fraction of CO, CO2, H2, H2O, and N2 in recycled gas, respectively; ω(C)i, ω(H)i, ω(N)I and ω(C)HM are the mass fraction of C, H, and N in material i and hot metal, respectively; and Tgas is the temperature of the metallurgical gas, K.
• Heat balance: where hi, hj are the enthalpy of the chemical reaction, kJ/kg; qi, qj are the sensible heat, kJ/kg; and Hl is the enthalpy of loss, kJ.Based on the above calculation results and the oxygen balance equation of the furnace, the volume of oxygen blast can be obtained.Based on the mass balance of the gas in top, bosh, and hearth of the furnace, and the heat balance (Equation ( 10)) of the whole furnace, the volume of recycled gas can be obtained.The volume of gas required for preheating the recycled gas is obtained by the heat balance (Equation ( 10)) of the heater unit.The parameters calculated by the TGR-OBF static process model are shown in Tables S1-S7 (Supplementary Materials).
In the VPSA unit, the recovery rate of CO2 was set at 90% [32], and the energy consumption was set at 645.7 kJ/kg-CO2 [33].In the heater unit, the thermal efficiency was set at 86% [11].

The COREX Process Model
The COREX process flow chart is shown in Figure 2. The COREX process is comprised of the shaft furnace and melter-gasifier.In the shaft furnace, the main reactions include the reduction of The COREX process flow chart is shown in Figure 2. The COREX process is comprised of the shaft furnace and melter-gasifier.In the shaft furnace, the main reactions include the reduction of various iron oxides and carbonate decomposition, while the main reactions inside the melter-gasifier include coal decomposition, carbon and reducing gas combustion, reduction of residual iron oxides, and decomposition of residual carbonate.In the COREX process, the raw gas produced by the melter-gasifier is first purified by hot gas cyclones.The dust after cyclone removal is recycled back to the melter-gasifier by dust burners.The major part of the purified raw gas is made available as reduction gas for the reduction shaft.Additionally, the rest of the purified top gas is scrubbed and mixed with the cleaned reduction shaft top gas to form COREX gas.The static process model and its operating parameters are based on Baosteel COREX 3000.
Based on the Tables S1, S2, S4, and S6, the consumption of raw materials (iron ore, limestone, and dolomite), the amount and chemical composition of slag can be calculated by using balance Equations ( 1), (3), and (4) in the COREX process.Based on the operating conditions and parameters (supplementary materials Tables S3 and S10) and the equation 11 of the COREX process, the volume and chemical composition of raw gas can be obtained by calculating balance Equations (9), and ( 11)- (14).
• C balance: • N balance: • The equilibrium constant of the water gas shift reaction: where, V CO , V H2O , V CO2 , V H2 , V CH4 , and

Integrated Steel Plant
As shown in Figure 3, the ISP consists of the steel manufacturing system and the energy system.The steel manufacturing system is comprised of the raw material handling process, ironmaking process, steelmaking process, and steel rolling process.In the steel manufacturing system, the iron ore is made into steel production with the help of energy inputs.At the same time, the generated Based on the volume, temperature, and composition of the raw gas, the volume of the cooling gas can be calculated by the heat balance (Equation ( 10)) of the hot gas cyclone.The consumption of coal is obtained by the heat balance (Equation ( 10)) of the melter-gasifier.The oxygen consumption can be obtained by the oxygen balance of the melter-gasifier.The volume of top gas can be obtained by the mass balance of the shaft furnace and gas oxidation degree.Finally, based on the mass balance of the metallurgical gas, the volume of COREX gas can be obtained.The parameters of the COREX process calculated by the COREX static process model are shown in Tables S1-S4, S6, and S9 (Supplementary Materials).

Integrated Steel Plant
As shown in Figure 3, the ISP consists of the steel manufacturing system and the energy system.The steel manufacturing system is comprised of the raw material handling process, ironmaking process, steelmaking process, and steel rolling process.In the steel manufacturing system, the iron ore is made into steel production with the help of energy inputs.At the same time, the generated metallurgical gases (including coke oven gas, blast furnace gas, and Linz-Donawitz gas), generated electricity, generated oxygen, and steam are controlled by the energy system.As a buffer, the power plant and oxygen plant can convert the surplus metallurgical gas from the steelmaking system into electricity and oxygen.For the BF and TGR-OBF steelmaking routes, the raw material handling processes include coking, sintering, and pelletizing, while the COREX steel production route does not.The analysis scale of an ISP was based on 1 t crude steel.In the analysis of a conventional ISP, the conventional ISP of Jingtang (which is located in the north of China) was taken as a reference case.By calculating the static process models of the TGR-OBF process and COREX process, the input and output parameters of per ton of hot metal in the two ironmaking processes can be obtained.On the basis of the hot metal consumption per ton of crude steel in the jingtang steel plant, combining the two static process models with the operation data of Jingtang steel plant (without BF process), an ISP with the TGR-OBF and an ISP with the COREX can be analyzed.In the analysis of the three ISPs, the following assumptions were made in this study: 1.
In the calculation of the TGR-OBF and COREX process models, the temperature and composition of hot metal, the chemical composition of raw materials were considered to be the same as that of the BF process in the Jingtang steel plant, and it was feasible for the two ironmaking processes.
For the TGR-OBF process, the raw materials with the same composition as the BF were flux, coke, coal, sinter, and pellets.For the COREX process, the raw materials with the same composition as the BF were flux and coal.

2.
For a better comparison of the three ironmaking technology in the ISP, the static process model of the TGR-OBF established in this paper was based on the current feasible technology, and the CCS technology was not considered in this study.

3.
In the analysis of an ISP with the TGR-OBF and an ISP with the COREX, the operation data for the raw materials handling process, steelmaking process, rolling process, lime making process, oxygen plant, and power plant were all from the Jingtang steel plant.The input and output values of the raw materials handling process were proportional to the raw material consumption of the TGR-OBF process and COREX process, respectively.4.
In an ISP, the metallurgical gas was preferentially supplied to the steel manufacturing system for use.The surplus gas was transported to the power plant, and the generating efficiency was 33%. 5.
The blast furnace blower was driven by electricity and the power consumption was 0.103 kWh/m 3 .6.
Since the calculation of TRT power generation is very complex.In this study, the generating capacity was simplified to be only related to the flow of top gas, it was 0.028 kWh/m 3 .7.
The electricity consumption of cryogenic oxygen process is 0.86 kWh/m 3 -oxygen.Therefore, based on the same composition and temperature of the hot metal, without changing the raw materials composition, the comparison and analysis of the three steel production routes can be obtained.

33%.
5. The blast furnace blower was driven by electricity and the power consumption was 0.103 kWh/m 3 .6. Since the calculation of TRT power generation is very complex.In this study, the generating capacity was simplified to be only related to the flow of top gas, it was 0.028 kWh /m 3 .7. The electricity consumption of cryogenic oxygen process is 0.86 kWh/m 3 -oxygen.
Therefore, based on the same composition and temperature of the hot metal, without changing the raw materials composition, the comparison and analysis of the three steel production routes can be obtained.

Methods to Calculate Comprehensive Energy Consumption
The calculation of comprehensive energy consumption of per ton steel is based on the energy balance of an ISP by using the energy equivalent value of each energy substance [34].In a conventional ISP or an ISP with TGR-OBF, the energy substances input into the system mainly include injection coal, coking coal, and electricity, and the outputs mainly include electricity, chemical by-products, and steam.In an ISP with COREX, the energy substance input into the system mainly includes coal and electricity, and the output mainly includes electricity and steam.The equation of comprehensive energy consumption of per ton steel is as follows: where in i e , out i e are the input and output energy substances i (kg/t crude steel); pi is the energy conversion coefficient of substance i (GJ/kg); and m, n are the number of energy substances.

Methods to Calculate CO2 Emissions
As

Methods to Calculate Comprehensive Energy Consumption
The calculation of comprehensive energy consumption of per ton steel is based on the energy balance of an ISP by using the energy equivalent value of each energy substance [34].In a conventional ISP or an ISP with TGR-OBF, the energy substances input into the system mainly include injection coal, coking coal, and electricity, and the outputs mainly include electricity, chemical by-products, and steam.In an ISP with COREX, the energy substance input into the system mainly includes coal and electricity, and the output mainly includes electricity and steam.The equation of comprehensive energy consumption of per ton steel is as follows: where e in i , e out i are the input and output energy substances i (kg/t crude steel); p i is the energy conversion coefficient of substance i (GJ/kg); and m, n are the number of energy substances.

Methods to Calculate CO 2 Emissions
As proposed in the WRI & WBCSD (World Resources Institute & World Business Council for Sustainable Development) guidelines, the net CO 2 emissions are calculated by the direct CO 2 emissions, the indirect CO 2 emissions and the CO 2 credit [35].In an ISP, the combustion of fossil fuel, metallurgical gas, and the decomposition of carbonate contribute to the direct CO 2 emissions, the purchased electricity and steam contribute to the indirect CO 2 emissions, and the excess steam, electricity, and chemical by-products can be considered as CO 2 credit.The net CO 2 emissions can be calculated by the following equations, (16) Metals 2019, 9, 364 9 of 15 (17) where CE, CE d , CE i , CE c represent the net, direct, indirect, and credit of the CO 2 emissions, t/t crude steel, respectively; and C in , C p , C byp represent the amount of total carbon input to the system, the amount of carbon fixed in the product, and the amount of carbon fixed in the byproduct, t/t crude steel.

Material Flows Analysis
Based on the operating data of the Jingtang steel plant, the material flows of a conventional ISP were investigated and presented in Figure 4.For a better comparison with a conventional ISP, the material flows of an ISP with the TGR-OBF and an ISP with the COREX were studied by combining the operating data in the Jingtang steel plant with established static process models, as shown in Figures 5 and 6.The parameters of the TGR-OBF process and the COREX process analyzed in the ISP are shown in Figures S1 and S2 (Supplementary Materials).As shown in Figure 4, a large amount of coking coal was consumed by a conventional ISP with 487.4 kg/t-crude steel, and the consumptions of coal and oxygen were 134.3 kg/t-crude steel and 96.9 m 3 /t-crude steel, respectively.In a conventional ISP, the total production of blast furnace gas (BFG) and coke oven gas (COG) were 1336.3 m 3 /t-crude steel and 166.8 m 3 /t-crude steel.Most were used to maintain production for each main process, and the surplus BFG and COG were transferred to the power plant for power supply and network pressure maintenance.As Figures 5 and 6 demonstrate, the material flows of an ISP with the TGR-OBF and an ISP with the COREX changed significantly in comparison with a conventional ISP.Compared with a conventional ISP, the coking coal consumption in the ISP with the TGR-OBF reduced by 39.7%.Since coking coal is not used in the production process of the COREX, the amount of coking coal was zero.However, the coal consumption in the ISP with the TGR-OBF and the ISP with the COREX increased by 23.6% and 535.2%, respectively.This is because the top gas recycling in the TGR-OBF reduced the coke consumption and oxygen blast increased the coal consumption.In the ISP with the COREX, there was no coking process and the consumption of coal and oxygen greatly increased in the melter-gasifier.At the same time, oxygen consumption increased by 174.7% and 521.5% in the TGR-OBF and COREX, respectively.In addition, due to the reduction in coke consumption and top gas recycling, the metallurgical gas produced in the ISP with the TGR-OBF was significantly reduced in comparison with the conventional ISP, i.e., the BFG and COG were reduced by 84% and 39.7%, respectively.In this steel production route, almost no surplus metallurgical gas was transported to the power plant for power generation.However, in the ISP with the COREX, a large amount of coal was used to make iron, so the generated COREX gas was very large, which was 23.9% more than the BFG process.

Metallurgical Gas Analysis
Figure 7 illustrates the generation and consumption of metallurgical gas by each process in a conventional ISP, an ISP with the TGR-OBF, and an ISP with the COREX.The order of metallurgical gas generation in the three ISPs is: ISP with the COREX (12.3 GJ/t-steel) > conventional ISP (7.8 GJ/tsteel) > ISP with the TGR-OBF (3.6 GJ/t-steel).This is because the oxygen consumption and coal consumption increased to 602.2 m 3 /t-crude steel and 853.1 kg/t-crude steel in an ISP with the COREX.The ISP with the TGR-OBF generated the least amount of metallurgical gas because of the use of recycled top gas.In addition, the ironmaking process of the ISP with the COREX contributed most of

Metallurgical Gas Analysis
Figure 7 illustrates the generation and consumption of metallurgical gas by each process in a conventional ISP, an ISP with the TGR-OBF, and an ISP with the COREX.The order of metallurgical gas generation in the three ISPs is: ISP with the COREX (12.3 GJ/t-steel) > conventional ISP (7.8 GJ/tsteel) > ISP with the TGR-OBF (3.6 GJ/t-steel).This is because the oxygen consumption and coal consumption increased to 602.2 m 3 /t-crude steel and 853.1 kg/t-crude steel in an ISP with the COREX.The ISP with the TGR-OBF generated the least amount of metallurgical gas because of the use of recycled top gas.In addition, the ironmaking process of the ISP with the COREX contributed most of

Metallurgical Gas Analysis
Figure 7 illustrates the generation and consumption of metallurgical gas by each process in a conventional ISP, an ISP with the TGR-OBF, and an ISP with the COREX.The order of metallurgical gas generation in the three ISPs is: ISP with the COREX (12.3 GJ/t-steel) > conventional ISP (7.8 GJ/t-steel) > ISP with the TGR-OBF (3.6 GJ/t-steel).This is because the oxygen consumption and coal consumption increased to 602.2 m 3 /t-crude steel and 853.1 kg/t-crude steel in an ISP with the COREX.The ISP with the TGR-OBF generated the least amount of metallurgical gas because of the use of recycled top gas.In addition, the ironmaking process of the ISP with the COREX contributed most of the metallurgical gas with 11.6 GJ/t-steel.This is because the ISP with COREX does not have a coking process, and a large amount of oxygen and coal react in the melter-gasifier to produce a large amount of high heat value COREX gas.In the coking process, the amount of COG in the conventional ISP and the ISP with the OBF were different.Compared with a conventional ISP (2.6 GJ/t-steel), the ISP with the TGR-OBF generated less COG with 1.6 GJ/t-steel.This is because the coke usage dropped to 294.1 kg/t-steel in the ironmaking process.The processes of metallurgical gas consumption included the power plant, ironmaking, coking, rolling, lime making, and steelmaking.In the ISP with COREX, 9.0 GJ/t-steel metallurgical gas was applied to the power plant, while only 2.9 GJ/t-steel and 0.027 GJ/t-steel were applied to power plants in a conventional ISP and an ISP with TGR-OBF.
Since an ISP with COREX does not include the coking process, sintering process, and pelletizing process, the amount of gas used in the steel manufacturing system is low and the amount of gas generated is large.Most of the surplus gas has to be sent to power plants for power generation.
the ISP with the OBF were different.Compared with a conventional ISP (2.6 GJ/t-steel), the ISP with the TGR-OBF generated less COG with 1.6 GJ/t-steel.This is because the coke usage dropped to 294.1 kg/t-steel in the ironmaking process.The processes of metallurgical gas consumption included the power plant, ironmaking, coking, rolling, lime making, and steelmaking.In the ISP with COREX, 9.0 GJ/t-steel metallurgical gas was applied to the power plant, while only 2.9 GJ/t-steel and 0.027 GJ/tsteel were applied to power plants in a conventional ISP and an ISP with TGR-OBF.Since an ISP with COREX does not include the coking process, sintering process, and pelletizing process, the amount of gas used in the steel manufacturing system is low and the amount of gas generated is large.Most of the surplus gas has to be sent to power plants for power generation.

Electricity Analysis
Figure 8 presents the generation and consumption of electricity by each process in a conventional ISP, an ISP with the TGR-OBF, and an ISP with the COREX.The electricity generated by the power plant was from the surplus metallurgical gas.Among the three steel production routes, the generation of electricity in an ISP with the COREX was the highest, followed by the conventional ISP, and the ISP with the TGR-OBF, which were 823.9 kWh/t-crude steel, 305.2 kWh/t-crude steel, and 35.4 kWh/t-crude steel, respectively.This is mainly because there is a large amount of surplus high heating value of metallurgical gas in an ISP with the COREX.The top gas recycling in an ISP with the TGR-OBF reduced the metallurgical gas production of the ironmaking and coking process, so almost no surplus metallurgical gas was transported to the power plant.Furthermore, the consumption of electricity in the ISP with the COREX was also the highest, followed by an ISP with the TGR-OBF, and the conventional ISP was the lowest.The increased demand for oxygen from the COREX and TGR-OBF was responsible for the significant increase in electricity consumption in the oxygen plant.As a result, the electricity consumed by the oxygen plant increased by 521.6% and 174.7%, respectively in comparison with the conventional ISP.Nevertheless, the ironmaking process of the ISP with the TGR-OBF consumed the most electricity in comparison with that of the other two steel production routes because the CO2 removal unit in the TGR-OBF process raises the demand for electricity.When compared with a conventional ISP, the electricity consumption in the ironmaking process of the ISP with the TGR-OBF increased by 89%.Therefore, as presented in Figure 8, the ISP with TGR-OBF purchased the most electricity, followed by the conventional ISP, and the ISP with COREX purchased the least, which were 710.1 kWh/t-crude, 212.4 kWh/t-crude, and 77.3 kWh/tcrude, respectively.

Electricity Analysis
Figure 8 presents the generation and consumption of electricity by each process in a conventional ISP, an ISP with the TGR-OBF, and an ISP with the COREX.The electricity generated by the power plant was from the surplus metallurgical gas.Among the three steel production routes, the generation of electricity in an ISP with the COREX was the highest, followed by the conventional ISP, and the ISP with the TGR-OBF, which were 823.9 kWh/t-crude steel, 305.2 kWh/t-crude steel, and 35.4 kWh/t-crude steel, respectively.This is mainly because there is a large amount of surplus high heating value of metallurgical gas in an ISP with the COREX.The top gas recycling in an ISP with the TGR-OBF reduced the metallurgical gas production of the ironmaking and coking process, so almost no surplus metallurgical gas was transported to the power plant.Furthermore, the consumption of electricity in the ISP with the COREX was also the highest, followed by an ISP with the TGR-OBF, and the conventional ISP was the lowest.The increased demand for oxygen from the COREX and TGR-OBF was responsible for the significant increase in electricity consumption in the oxygen plant.As a result, the electricity consumed by the oxygen plant increased by 521.6% and 174.7%, respectively in comparison with the conventional ISP.Nevertheless, the ironmaking process of the ISP with the TGR-OBF consumed the most electricity in comparison with that of the other two steel production routes because the CO 2 removal unit in the TGR-OBF process raises the demand for electricity.When compared with a conventional ISP, the electricity consumption in the ironmaking process of the ISP with the TGR-OBF increased by 89%.Therefore, as presented in Figure 8, the ISP with TGR-OBF purchased the most electricity, followed by the conventional ISP, and the ISP with COREX purchased the least, which were 710.1 kWh/t-crude, 212.4 kWh/t-crude, and 77.3 kWh/t-crude, respectively.

Comprehensive Energy Consumption Analysis
The comprehensive energy consumptions of the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX are shown in Figure 9.Among these three steel production routes, the ISP with the TGR-BOF consumed the least energy to produce 1 t of steel.The energy consumption was 14.16 GJ/t-steel, 16% and 16.5% lower than the conventional ISP and the ISP with the COREX, respectively.This is because the top gas recycling in the TGR-OBF ironmaking process can significantly reduce carbon consumption; despite requiring the most electricity to be purchased, the energy conversion coefficient of electricity is smaller than that of coal.Therefore, the comprehensive energy consumption in the ISP with the TGR-OBF is the lowest.Moreover, the energy consumed by the conventional ISP and the ISP with the COREX were 16.85 GJ/t-steel and 16.96 GJ/t-steel, respectively.However, the coking coal consumption accounted for 80.2% of the total energy input in the conventional ISP, which was the most out of the three steel production routes.

Comprehensive Energy Consumption Analysis
The comprehensive energy consumptions of the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX are shown in Figure 9.Among these three steel production routes, the ISP with the TGR-BOF consumed the least energy to produce 1 t of steel.The energy consumption was 14.16 GJ/t-steel, 16% and 16.5% lower than the conventional ISP and the ISP with the COREX, respectively.This is because the top gas recycling in the TGR-OBF ironmaking process can significantly reduce carbon consumption; despite requiring the most electricity to be purchased, the energy conversion coefficient of electricity is smaller than that of coal.Therefore, the comprehensive energy consumption in the ISP with the TGR-OBF is the lowest.Moreover, the energy consumed by the conventional ISP and the ISP with the COREX were 16.85 GJ/t-steel and 16.96 GJ/t-steel, respectively.However, the coking coal consumption accounted for 80.2% of the total energy input in the conventional ISP, which was the most out of the three steel production routes.

CO2 Emissions Analysis
Figure 10 shows the net, direct, indirect, and the credit of CO2 emissions in the three steel production routes with different CO2 emissions factors.CCS technology was not considered in this research.When considering the impact of the CO2 emissions factor of electricity, the International Energy Agency factor (0.504) and the China Grid factor (1.0302) were used to analyze the CO2 emissions.In the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX, the direct CO2 emissions accounted for the largest proportion of the total CO2 emissions.With the International Energy Agency factor, the direct CO2 emissions accounted for 93.2%, 79.8%, and 97.9% of the total

Comprehensive Energy Consumption Analysis
The comprehensive energy consumptions of the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX are shown in Figure 9.Among these three steel production routes, the ISP with the TGR-BOF consumed the least energy to produce 1 t of steel.The energy consumption was 14.16 GJ/t-steel, 16% and 16.5% lower than the conventional ISP and the ISP with the COREX, respectively.This is because the top gas recycling in the TGR-OBF ironmaking process can significantly reduce carbon consumption; despite requiring the most electricity to be purchased, the energy conversion coefficient of electricity is smaller than that of coal.Therefore, the comprehensive energy consumption in the ISP with the TGR-OBF is the lowest.Moreover, the energy consumed by the conventional ISP and the ISP with the COREX were 16.85 GJ/t-steel and 16.96 GJ/t-steel, respectively.However, the coking coal consumption accounted for 80.2% of the total energy input in the conventional ISP, which was the most out of the three steel production routes.

CO2 Emissions Analysis
Figure 10 shows the net, direct, indirect, and the credit of CO2 emissions in the three steel production routes with different CO2 emissions factors.CCS technology was not considered in this research.When considering the impact of the CO2 emissions factor of electricity, the International Energy Agency factor (0.504) and the China Grid factor (1.0302) were used to analyze the CO2 emissions.In the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX, the direct CO2 emissions accounted for the largest proportion of the total CO2 emissions.With the International Energy Agency factor, the direct CO2 emissions accounted for 93.2%, 79.8%, and 97.9% of the total

CO 2 Emissions Analysis
Figure 10 shows the net, direct, indirect, and the credit of CO 2 emissions in the three steel production routes with different CO 2 emissions factors.CCS technology was not considered in this research.When considering the impact of the CO 2 emissions factor of electricity, the International Energy Agency factor (0.504) and the China Grid factor (1.0302) were used to analyze the CO 2 emissions.In the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX, the direct CO 2 emissions accounted for the largest proportion of the total CO 2 emissions.With the International Energy Agency factor, the direct CO 2 emissions accounted for 93.2%, 79.8%, and 97.9% of the total CO 2 emissions in the three routes, respectively.With the China Grid factor, the direct CO 2 emissions accounted for 88.5%, 66.0% and 96.5% of the total CO 2 emissions of the three routes.The net CO 2 emissions in the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX increased from 1.94 t/t-steel, 1.73 t/t-steel, and 2.66 t/t-steel (International Energy Agency factor) to 2.05 t/t-steel, 2.11 t/t-steel, and 2.70 t/t-steel (China Grid factor), respectively.The increases were 5.7%, 22%, and 3.8%, respectively.This is because the indirect CO 2 emissions can be influenced by the CO 2 emissions factor.Furthermore, the more the electricity purchased by an ISP, the greater the impact of the CO 2 emissions factor.Therefore, if the International Energy Agency factor was used, the net CO 2 emissions of the ISP with the TGR-OBF was the lowest, followed by the conventional ISP, and the ISP with the COREX.However, if the China Grid factor was used, the net CO 2 emissions of the ISP with the TGR-OBF was not the lowest.That is to say, the net CO 2 emissions of the ISP with the TGR-OBF rely heavily on the electricity CO 2 emissions factor.When compared with the conventional ISP and the Metals 2019, 9, 364 13 of 15 ISP with the COREX, the ISP with the TGR-OBF had the lowest direct CO 2 emissions and the highest indirect CO 2 emissions.This is due to its lowest coal consumption and highest electricity consumption when compared with the other routes.The ISP with the COREX had the highest net CO 2 emissions for consuming a large amount of coal, which was 37.1% and 53.8% higher (International Energy Agency factor) as well as 31.7% and 28.0% higher (China Grid factor) than that of the conventional ISP and the ISP with the TGR-OBF, respectively.
2.11 t/t-steel, and 2.70 t/t-steel (China Grid factor), respectively.The increases were 5.7%, 22%, and 3.8%, respectively.This is because the indirect CO2 emissions can be influenced by the CO2 emissions factor.Furthermore, the more the electricity purchased by an ISP, the greater the impact of the CO2 emissions factor.Therefore, if the International Energy Agency factor was used, the net CO2 emissions of the ISP with the TGR-OBF was the lowest, followed by the conventional ISP, and the ISP with the COREX.However, if the China Grid factor was used, the net CO2 emissions of the ISP with the TGR-OBF was not the lowest.That is to say, the net CO2 emissions of the ISP with the TGR-OBF rely heavily on the electricity CO2 emissions factor.When compared with the conventional ISP and the ISP with the COREX, the ISP with the TGR-OBF had the lowest direct CO2 emissions and the highest indirect CO2 emissions.This is due to its lowest coal consumption and highest electricity consumption when compared with the other routes.The ISP with the COREX had the highest net CO2 emissions for consuming a large amount of coal, which was 37.1% and 53.8% higher (International Energy Agency factor) as well as 31.7% and 28.0% higher (China Grid factor) than that of the conventional ISP and the ISP with the TGR-OBF, respectively.

Conclusions
Based on two static process models, the material flows of the TGR-OBF and COREX ironmaking processes were investigated.Combining the operating data of the Jingtang steel plant, three steel production routes of the ISP with the BF, TGR-OBF, and COREX were analyzed comparatively on the material flows, metallurgical gas generation and consumption, electricity consumption and generation, comprehensive energy consumption, and CO2 emissions.Compared with the conventional ISP, coking coal consumption in the ISP with the TGR-OBF and the ISP with the COREX were reduced by 39.7% and 100%.The coal consumption in the ISP with the TGR-OBF and the ISP with the COREX increased by 23.6% and 535.2%.Among the three routes, the electricity purchased for the ISP with the TGR-OBF was the highest with 710.1 kWh/t-steel and the ISP with the COREX was the lowest with 77.3 kWh/t-steel.The energy consumption of the ISP with the TGR-OBF was 16% and 16.5% lower than that of the conventional ISP and the ISP with the COREX, respectively.The energy consumption of the ISP with the COREX was the highest with 16.96 GJ/t-steel.The ISP with the COREX had the highest net CO2 emissions, which was 37.1% and 53.8% higher (International Energy Agency factor) as well as 31.7% and 28.0% higher (China Grid factor) than that of the

Conclusions
Based on two static process models, the material flows of the TGR-OBF and COREX ironmaking processes were investigated.Combining the operating data of the Jingtang steel plant, three steel production routes of the ISP with the BF, TGR-OBF, and COREX were analyzed comparatively on the material flows, metallurgical gas generation and consumption, electricity consumption and generation, comprehensive energy consumption, and CO 2 emissions.Compared with the conventional ISP, coking coal consumption in the ISP with the TGR-OBF and the ISP with the COREX were reduced by 39.7% and 100%.The coal consumption in the ISP with the TGR-OBF and the ISP with the COREX increased by 23.6% and 535.2%.Among the three routes, the electricity purchased for the ISP with the TGR-OBF was the highest with 710.1 kWh/t-steel and the ISP with the COREX was the lowest with 77.3 kWh/t-steel.The energy consumption of the ISP with the TGR-OBF was 16% and 16.5% lower than that of the conventional ISP and the ISP with the COREX, respectively.The energy consumption of the ISP with the COREX was the highest with 16.96 GJ/t-steel.The ISP with the COREX had the highest net CO 2 emissions, which was 37.1% and 53.8% higher (International Energy Agency factor) as well as 31.7% and 28.0% higher (China Grid factor) than that of the conventional ISP and the ISP with TGR-OBF.Although the ISP with COREX had no advantages in energy consumption and carbon emissions, it had the least purchased electricity and coke consumption.In addition, the ISP with TGR-OBF had the lowest energy consumption.Therefore, the ISP with the COREX is suitable for the areas with high electricity prices and where the coke is not available.Moreover, the ISP with the TGR-OBF should be carefully considered when the CCS technology is not applied.

Supplementary Materials:
The following are available online at http://www.mdpi.com/2075-4701/9/3/364/s1,Table S1: The chemical composition of ore in the TGR-OBF process and the COREX process.Table S2: The chemical composition of flux in the TGR-OBF process and the COREX process.Table S3: The chemical composition of coal and coke in the TGR-OBF process and the COREX process.Table S4: The chemical composition of hot metal in the TGR-OBF process and the COREX process.Table S5: The chemical composition of oxygen blast furnace gas in the TGR-OBF process.Table S6: The distribution ratio of elements in the slag and hot metal (the TGR-OBF process & the COREX process) Table S7: The operating parameters of the TGR-OBF process.Table S8: The chemical composition of COREX gas in the COREX process.Table S9: The operating parameters of the COREX process.Figure S1: The parameters of the TGR-OBF process.Figure S2: The parameters of the COREX process.

Figure 2 .
Figure 2. The flow diagram of the COREX ® ironmaking process.

Figure 2 .
Figure 2. The flow diagram of the COREX ® ironmaking process.

Figure 3 .
Figure 3.The integrated steel plant (ISP) in this study.

Figure 3 .
Figure 3.The integrated steel plant (ISP) in this study.

16 Figure 4 .
Figure 4. Chart of the material flows of a conventional ISP.Figure 4. Chart of the material flows of a conventional ISP.

Figure 4 .
Figure 4. Chart of the material flows of a conventional ISP.Figure 4. Chart of the material flows of a conventional ISP.

Figure 4 .
Figure 4. Chart of the material flows of a conventional ISP.

Figure 5 .
Figure 5. Chart of the material flows of an ISP with the TGR-OBF.

Figure 6 .
Figure 6.Chart of the material flows of an ISP with the COREX.

Figure 5 .
Figure 5. Chart of the material flows of an ISP with the TGR-OBF.

Figure 4 .
Figure 4. Chart of the material flows of a conventional ISP.

Figure 5 .
Figure 5. Chart of the material flows of an ISP with the TGR-OBF.

Figure 6 .
Figure 6.Chart of the material flows of an ISP with the COREX.

Figure 6 .
Figure 6.Chart of the material flows of an ISP with the COREX.

Figure 7 .
Figure 7.The metallurgical balance in the conventional ISP, ISP with the TGR-OBF, and ISP with the COREX.

Figure 7 .
Figure 7.The metallurgical gas balance in the conventional ISP, ISP with the TGR-OBF, and ISP with the COREX.

Figure 8 .
Figure 8.The electricity balance in the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX.

Figure 9 .
Figure 9.The energy balance in the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX.

Figure 8 .
Figure 8.The electricity balance in the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX.

Figure 8 .
Figure 8.The electricity balance in the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX.

Figure 9 .
Figure 9.The energy balance in the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX.

Figure 9 .
Figure 9.The energy balance in the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX.

Figure 10 .
Figure 10.The CO2 emissions intensity in the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX with the CO2 emissions factor referring to the world average (a) or North China Grid (b).

Figure 10 .
Figure 10.The CO 2 emissions intensity in the conventional ISP, the ISP with the TGR-OBF, and the ISP with the COREX with the CO 2 emissions factor referring to the world average (a) or North China Grid (b).