*3.3. Coke Oven Gas Injection*

In order to decrease the CO2 emissions of the blast furnace, coke oven gas was mixed with hot blast and injected from tuyeres. Table 5 shows the composition of the coke oven gas in Bayi Steel. The total amount of coke oven gas and blast gas was kept constant when injecting the coke oven gas. The proportion of coke oven gas in the gas mixture was from 3% to 15% at a step of 3%.


**Table 5.** Composition of coke oven gas.

The CH4 and O2 in coke oven gas were converted to CO and H2 for simplicity. The replacement ratio of coke oven gas to coke was 0.4 kg·Nm<sup>−</sup>3. Therefore, for each 3% increase of coke oven gas, the oxygen enrichment rate increased by 0.55%. The production increased by 246 ton HM and the coke ratio decreases by 16kg·ton HM<sup>−</sup>1. The temperature profiles at different coke oven gas injection ratios are shown in Figure 6. The carbon solution loss ratio under different coke oven gas injection ratios was also calculated after simulation, as shown in Figure 7.

**Figure 6.** Temperature profiles at different proportions of coke oven gas injected.

It can be seen from Figure 6 that the temperature decreases with the increase of the proportion of coke oven gas injected, and the temperature in the middle zone increase relatively more. When the proportion of coke oven gas injected is 15%, the facet average temperature at 14 m height decreases by 210 ◦C than the base model. This suggests that the cohesive zone widens with the increase of proportion of coke oven gas injected.

**Figure 7.** Carbon solution loss ratio at different amount of coke oven gas injected.

It can be seen from Figure 7 that the carbon solution loss ratio of the coke increases gradually as the coke oven gas injection increases. For each 3% increase in the coke oven gas injection ratio, the carbon solution loss ratio of coke increases by an average of 0.83%. The carbon solution loss ratio of the coke increases by 4.15% when the coke oven gas injection is 15%.

The reason may be that the replacement of hot blast with coke oven gas leads to an increase of H2 in the gas mixture due to the high H2 content of the coke oven gas. This results in a higher H2O content in the blast furnace due to the reduction reaction of the H2 and iron ore. The higher H2O content may be part in a water gas reaction, which increases the reaction of the carbon solution loss. The temperature of the strong reaction between the coke and the H2O is from 800 to 1300 ◦C. The average H2O content in the gas in the blast furnace increases from 2.45% to 8.67% when the coke oven gas ratio increases from 0% to 15%. Moreover, the high temperature zone shrinks after the injection of the coke oven gas, which leads to the expansion of the zone for the reaction of the carbon solution loss.

Therefore, on the one hand, coke oven gas can reduce the coke rate and CO2 emission of the blast furnace as a hydrogen-rich gas. On the other hand, the injection of coke oven gas causes the temperature in the lower part of the blast furnace to decrease and the carbon solution loss to increase, which impairs the operation of blast furnace. Therefore, coke oven gas should not be injected into the blast furnace in Bayi Steel in its present condition. Coke oven gas should only be injected only if the carbon solution loss is low enough, and it should be done a higher gas temperature and with better coke quality of a high M40 and a high CSR (coke strength after reaction).

#### *3.4. Steel Scrap Charging*

Since the steel market has been strong in China recently, steel scraps are charged so as to increase production. The metallic iron (MFe) of steel scrap is 52.29% and the FeO is 9.07%. The steel scrap is added when the total amount of iron-bearing materials is kept constant. The proportion of steel scrap is from 0% to 10% at a step of 2%, as shown in Table 6, where the absolute amount is also illustrated. For each 2% increase of steel scrap, the oxygen enrichment ratio increases 0.06%, the coke rate decreases 4 kg·ton HM<sup>−</sup>1, and production increases 70 ton HM while the coal rate and the amount of hot blast remain unchanged. The temperature profiles at different proportions of steel scrap added are shown in Figure 8. The carbon solution loss ratio is calculated after simulation, as shown in Figure 9.

**Table 6.** Amount of steel scrap added.

**Figure 8.** Temperature profiles at different proportions of steel scrap added.

It can be seen from Figure 8 that the temperature increases a little in the lower zone of the blast furnace with the increase of the proportion of steel scrap added, but the degree is very small. When the proportion of steel scrap added is 10%, the facet average temperature at 4 m height increases by 22 ◦C than the base model. Hardly any difference of temperature in the upper zone of the blast furnace can be found. This indicates that the cohesive zone only narrows a little with the increase of the proportion of steel scrap added.

**Figure 9.** Carbon solution loss ratio at different amounts of steel scrap added.

The carbon solution loss ratio decreases gradually as the proportion of steel scraps added increases; when 10% steel scrap is added, it decreases by 2.6%. The reason may be that the amount of iron ore that needs to be reduced decreases when steel scraps are added in the top of blast furnace since steel scrap provides iron content that does not need reduction. However, this leads the utilization ratio of the gas to decrease. The ratio decreases by 1.32% when the proportion of steel scraps increases to 10%. As a result, the CO content in the gas increases while the CO2 content decreases. This also leads to a decrease of the carbon solution loss ratio.

It should be noticed that the addition of steel scrap reduces the gas utilization ratio, and also the temperature in the upper part of the blast furnace. But the reduction in the gas utilization ratio and the temperature of the upper part of the blast furnace is small and within acceptable limits. Therefore, the addition of steel scrap is beneficial for the operation of blast furnace in the aspect of carbon solution loss. Steel scrap can be added to the blast furnace in appropriate amounts.

#### **4. Conclusions**

A three dimensional model was established to analyze the carbon solution loss of coke based on the practical operational parameters of blast furnace B in Bayi Steel. The model was then used to investigate the effects of the oxygen enrichment ratio, the injection of coke oven gas, and the addition of steel scraps on the carbon solution loss ratio of the coke. The conclusions are as follows.

(1) The carbon solution loss ratio of the blast furnace studied is 18.61% when the coke reaches 1300 ◦C.

(2) The carbon solution loss ratio decreases with the increase of the oxygen enrichment ratio or when a certain proportion of steel scraps is added, while it increases with an increase in the amount of coke oven gas injected.

(3) The oxygen enrichment ratio and the proportion of steel scraps added can be increased to 5% and 10%, respectively in order to reduce the carbon solution loss without affecting the operation of the blast furnace.

(4) The injection of coke oven gas is not recommended given the current condition of blast furnace B of Bayi Steel.

**Author Contributions:** Conceptualization, M.K. and S.W.; methodology, H.Z. and Z.H.; Software, M.K and H.Z.; validation, S.Y. and H.X.; formal analysis, L.P.W. and H.X.; investigation, M.K. and H.Z.; resources, H.X. and S.W.; data curation, Z.H. and S.Y.; writing, M.K and Z.H.; supervision, M.K and L.P.W.; funding acquisition, M.K., L.P.W. and H.X.

**Funding:** The authors would like to acknowledge the financial support by National Key R&D Program of China (grant number 2017YFB0603800, 2017YFB0603803), the National Natural Science Foundation of China (grant number 51804027) and USTB-NTUT Joint Research Program (grant number TW201909, NTUT-USTB-108-06).

**Acknowledgments:** The authors would like to appreciate much for the anonymous reviewers and editors for the improvement of this work, and Prof. Mark Buck for correcting language.

**Conflicts of Interest:** The authors declare no conflict of interest.
