UDK 669.162.2:662.737 Professional article/Strokovni članek
ISSN 1580-2949 MTAEC9, 46(6)673(2012)
INJECTION OF BROWN-COAL TAR IN A STATISTICAL RELATIONSHIP WITH THE SIGNIFICANT PARAMETERS OF A BLAST FURNACE
STATISTIČNO RAZMERJE MED VPIHOVANJEM KATRANA IZ RJAVEGA PREMOGA IN POMEMBNIMI PARAMETRI PLAVŽA
Pavlina Pustejovska1, Simona Jursova2, Silvie Brožova1, Michal Pivko3
!VSB - Technical University of Ostrava, Faculty of Metallurgy and Material Engineering, Czech Republic 2VSB - Technical University of Ostrava, Centre ENET, Czech Republic 3VSB - Technical University of Ostrava, Faculty of Mining and Geology, Czech Republic pavlina.pustejovska@vsb.cz
Prejem rokopisa — received: 2011-10-26; sprejem za objavo - accepted for publication: 2012-08-28
The extremely large energy requirements of blast-furnace iron production and the high levels of pollution in terms of carbon dioxide have resulted in systematic efforts aimed at decreasing the specific consumption of fuels and deoxidizing agents during this process. This paper deals with the injection of alternative fuels into a blast furnace and their effect on the blast furnace's operation. In this case oil injection was replaced by coal-tar injection. A comparison analysis of the brown-coal generator tar injection and oil injection was applied to assess the effect of the injection on the reduction of CO2 emissions. The daily parameters of the inputs and outputs were available. The statistical relations for the injection of brown-coal tar and the significant parameters of the aggregate are studied. Correlation coefficients indicating the tightness in the tar injection and other significant technological parameters, such as the specific coke consumption, theoretical combustion temperature and the emissions of the blast furnace are calculated.
Keywords: blast furnace, tar, injection
Ekstremno visoke zahteve po energiji pri proizvodnji železa v plavžu in velika polucija z ogljikovim dioksidom zahtevajo sistematično prizadevanje za zmanjšanje specifične porabe goriv in dezoksidacijskih sredstev pri tem postopku. Ta članek obravnava vbrizgavanje alternativnih goriv v plavž in njihov učinek na delovanje plavža. Vbrizgavanje olja je bilo nadomeščeno z vbrizgavanjem katrana iz premoga. Izvršena je bila primerjalna študija vbrizgavanja generatorskega katrana iz rjavega premoga ter vbrizgavanja olja za oceno učinka vbrizgavanja na zmanjšanje emisije CO2. Na voljo so dnevni parametri vhodnih in izhodnih snovi. Studirano je bilo statistično razmerje med vbrizgavanjem katrana iz rjavega premoga in pomembnimi procesnimi parametri plavža. Izračunani korelacijski koeficienti kažejo tesno povezavo med vbrizgavanjem katrana in drugimi pomembnimi tehnološkimi parametri, kot so poraba koksa, teoretična temperatura zgorevanja in emisije iz plavža.
Ključne besede: plavž, katran, vbrizgavanje
1 INTRODUCTION
An essential prerequisite for specific fuel consumption is coke substitution with an alternative fuel, a reducing agent. Nowadays, coal or gas is widely used as a substitute fuel. In the Czech Republic, oil was also used in past. Today's price of oil has caused oil injection to be very expensive. Coal, gas and the by-products of coke making, such as tars of brown and black coal. are exploitable as alternative fuels. However, the injection of alternative fuels has to respect the heat demands of the blast furnace. It is not possible for it to cause disorders in the subside of the blast-furnace burden. The unburned remainder of the fuel has to be consumed through the blast-furnace aggregate. A decrease in chemical energy exploitation in the blast-furnace gas is not allowed1. The augmentation in furnace gases and hydrogen input into the deoxidizing gaseous agent is caused by the injection, which probably results in a reduction of the specific coke consumption and emissions.
2 EXPERIMENT AND RESULTS
Up-to-date physical simulation procedures make it possible to study the behavior of materials under conditions very close to real industrial processing or applications.2 For a comparison analysis some statistical calculations were carried out. The statistical evaluation of brown-coal tar injection contributes to a technological analysis of the blast-furnace order. The statistical relationships between the process parameters of the blast furnace describe, at the same time, ecological aspects of the utilization of alternative fuels for Czech blast furnaces. This might present possibilities for a decrease of the CO2 emissions in metallurgy.
The computations were made with the automatic software Statgraphic. The data for the computations were acquired by a Czech metallurgical company; they refer to the real iron-making process, which is summarized in the equations below3 4:
3 Fe2O3 + CO = 2 Fe3O4 + CO2	(1)
Fe3O4 + CO = 3 FeO + CO2	(2)
FeO + CO = Fe + CO2	(3)
Table 1 shows a chemical analysis of the generator tar5 produced by Czech brown-coal pyrolysis.
Table 1: Chemical composition of brown-coal generator tar5 Tabela 1: Kemijska sestava generatorskega katrana iz rjavega pre-
moga5
Element	C	HO	O	N	S	H2O
Amount in mass fractions, w/%	83	9.4	3.8	0.5	0.4	1.2
The results of the statistical analysis are presented in Table 2.
Table 2: Correlation coefficients indicating tightness in the tar injection and other significant technological parameters of the blast furnace Tabela 2: Korelacijski koeficienti, ki kažejo tesno povezavo med vbrizgavanjem katrana in drugimi pomembnimi tehnološkimi parametri plavža
Parameter	Tar	W(H2)/ %	O2 in wind	Coke	Emission	Humidity	TCT
Tar		0.64	0.59	-0.35	-0.36	0.1	-0.1
w(H2 top gas)/ %	0.64		0.55	-0.5	-0.31	0.53	0.12
O2 in wind	0.59	0.55		-0.27	-0.43	0.20	0.29
Coke	-0.35	-0.5	-0.27		0.56	-0.10	-0.27
Emission	-0.36	-0.31	-0.43	0.56		0.01	-0.18
Humidity	0.10	0.53	0.20	-0.10	0.01		-0.66
TCT	-0.10	0.12	0.29	-0.27	-0.18	-0.66	
TCT - Theoretical Combustion Temperature w - mass fraction (%)
3 DISCUSSION
The first statistical evaluation concerns the relationship between the injected amount and the specific coke consumption. The statistical relationship is quite insignificant because the injected amount is low, stable and there are other aspects of the blast-furnace burden and technological properties affecting the coke consumption. The value of the coefficient of the independent variable in the regression equation determines the coefficient of substitution. It is 0.77, which is rather lower than the figure from theoretical calculations and practical experience.
Tar injection results significantly and quickly in hydrogen content augmentation in blast-furnace gas and furthermore in the outgoing furnace gas. Its correlation coefficient is 0.64. At the same time, the linear-regression dependence shows that 10 kg t-1 per pig-iron augmentation in tar injection makes a 6 % increase in the hydrogen content. So the speed of hydrogen-content augmentation is unlikely to be caused by the tar injection. For an explanation it is necessary to look at other parameters.
For wind enriched by oxygen, which often accompanies tar injection, the augmentation is R = 0.59. The linear relationship in the changes of oxygen content in wind
and the hydrogen content in the gas show a medium statistical relationship of R = 0.55. As the effect of oxygen was included, the multiple regression H2-tar, O2 is used. The correlation coefficient reaches a slightly higher value (R2 = 0.45), the coefficient of the tar variable shows a more probable value of 0.016 (10 kg t-1 of tar augmentation results in 0.16 %, 100 kg t-1, 1.6 % augmentation of hydrogen). Similar results were computed by the multiple correlation H2-tar (R2 = 59 %) because the more the wind is enriched, the less hydrogen is brought by the wind into the blast furnace.
Humidification of the wind by the water steam affects the hydrogen content in the gas (R = 0.53). With the tar included in the regression relationship, the correlation level is better. (R2 = 63 %). However, including the oxygen effect in the regression equation does not cause a significant augmentation in the correlation level (R2 = 65 %).
There is an effect of injection, oxygen enrichment and other parameters in the coke's specific consumption. The correlation coefficient H2-coke, i.e., R = 0.50, refers to the relationship between injection and the hydrogen content in the top gas
The tar included in the regression relationship between the coke consumption and the wind humidification (so H2-humidification, tar) significantly increases the correlation level (R2 = 0.17).
The statistical relationship between the oxygen enrichment and the specific coke consumption is rather featureless (R = -0.27) because the oxygen intensification variously affects a wide range of technological aspects of blast furnace's operation. It possibly affects the coke consumption as a global parameter only partially. Within the statistical evaluation, the relation between coke-tar, oxygen analyzed by a method of multiple regression shows a higher coefficient, in comparison with the linear regression without the oxygen content.
The total amount of CO2 emission includes the CO emission in the top gas, although the CO burns out of the blast furnace process in wind heaters, a coke plant or in other metallurgical appliances. The total emission sources from CO + CO2 ratio in the top gas, specific gas amount counted by the balance computations. This parameter of the blast-furnace process is very variable and affected by all kinds of items; therefore, it is not statistically significant in low tar injection.
There is a significant correlation between the emissions and the specific coke consumption where the correlation coefficient is 0.56. A polynomial regression results in a slight augmentation of the relationship's tightness (R2 = 34.7 %). The augmentation includes the oxygen-enriched wind effect, where the regressive relationship shows the correlation R2 = 40 %.
The tar-injection effect on the emissions statistically confirms a significant relationship (R = 0.36); however, the relationship is weak and is not affected by the polynomial regression. A cause of the augmentation is the
Figure 1: Effect of tar injection as a hydrocarbon fuel on blast-furnace operation and its total CO2 emission
Slika 1: Učinek vbrizgavanja katrana kot goriva iz ogljikovodikov na delovanje plavža in skupno emisijo CO2
other parameters included in the regressive relationships. The effect of oxygen enrichment increases the correlation level to R2 = 20.2 %.
The significant indicator of the technological operation is the theoretical combustion temperature (TCT). It is counted from the input and output data values. The statistical analysis confirms a significant effect of wind humidification (R = -0.66) and a rather lowly tight relationship between TCT and oxygen enrichment of the wind (R = 0.29). The lowest tightness is for the tar (R = -0.1). The combination of oxygen effect, humidification and tar in the regression relationship to TCT represents the correlation level R2 = 0.75, so a 75 % change in the TCT indicator is possible to explain with changes in the parameters, such as wind, enrichment, humidification and tar injection.
4 CONCLUSION
Tar, as an alternative fuel, brings to the blast furnace more hydrogen into the reduction gas than the fuel coke, which causes a decrease in the specific coke consumption and blast-furnace emissions (CO + CO2). The compensation of oxygen wind enrichment is intensified by the tar injection. The effect of specific coke consumption on the emission of CO + CO2 is statistically closest, even in the current conditions of a low level of tar injection. The graphical conclusion of the relationship between tar injection and other technological blast-furnace processes is summed up in Figure 1.
From the economical point of view, the alternative fuels injection is beneficial because it decreases the costs for fuels.6 Modern operating and verification methods in ironmaking and steelmaking metallurgy contribute to a further decrease in emissions.7
Acknowledgement
This work was supported by the Czech Ministry of Education, Youth and Sports, research project MSM 6198910019 and project CZ.1.05/2.1.00/03.0069.
5 REFERENCES
1 V. Roubicek, J. Buchtele, Coal: sources, processes, utilization, Mon-tanex Ostrava 2002, 173 2S. T. Madziej, Physical simulation of metallurgical processes, Mater. Tehnol., 44 (2010) 3, 105-119 31. B. Risteski, New discovered paradoxes in theory of balancing
chemical reactions, Mater. Tehnol., 45 (2011) 6, 503-522 41. B. Risteski, A new complex vector method for balancing chemical
equatations, Mater. Tehnol., 44 (2010) 4, 193-203 5 P. Hudecek, Utilization of Vresova brown coal gasification products for metallurgy, Diplomova prace, Fakulta metalurgie a materialoveho inzenyrstvi, Katedra metalurgie, VSB - TU Ostrava, 2008, 77 6E. Kardas et al., Selected elements of economics of blast furnace process, Acta Metallurgica Slovaca, 13 (2007), 497-501 7 E. Mihok, D. Baricova, Modern operating and verification methods in ironmaking and steelmaking metallurgy, Elfa, Kosice 2008, 137