See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/245076796

Energy efficiency and the influence of gas burners to the energy related
carbon dioxide emissions of electric arc furnaces in steel industry

Article in Energy · September 2009

DOI: 10.1016/j.energy.2009.04.015

CITATIONS READS

113 6,709

3 authors, including:

Marcus Kirschen Herbert Pfeifer

RHI Magnesita RWTH Aachen University
86 PUBLICATIONS 781 CITATIONS 279 PUBLICATIONS 1,921 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Marcus Kirschen on 18 October 2017.

The user has requested enhancement of the downloaded file.

This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
 Author's personal copy

Energy 34 (2009) 1065–1072

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Energy efficiency and the influence of gas burners to the energy related carbon
dioxide emissions of electric arc furnaces in steel industry
Marcus Kirschen*, Victor Risonarta, Herbert Pfeifer
Institute for Industrial Furnaces and Heat Engineering, RWTH Aachen University, Kopernikusstraße 16, D-52074 Aachen, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Determining the complete energy balance of an electric arc furnace (EAF) provides an appropriate
Received 6 June 2008 method to examine energy efficiency and identify energy saving potentials. However, the EAF energy
Received in revised form balance is complex due to the combined input of electrical energy and chemical energy resulting from
14 January 2009
natural gas (NG) combustion and oxidation reactions in the steel melt. In addition, furnace off-gas
Accepted 20 April 2009
measurements and slag analysis are necessary to reliably determine energy sinks. In this paper 70 energy
Available online 20 May 2009
balances and energy efficiencies from multiple EAFs are presented, including data calculated from plant
measurements and compiled from the literature. Potential errors that can be incorporated in these
Keywords:
Energy efficiency calculations are also highlighted. The total energy requirement of these modern EAFs analysed ranged
CO2 emission from 510 to 880 kWh/t, with energy efficiency values (h ¼ DHSteel/ETotal) of between 40% and 75%.
Electric arc furnace Furthermore, the focus was placed on the total energy related CO2 emissions of EAF processes comprising
Steel industry NG combustion and electrical energy input. By assessing multiple EAF energy balances, a significant
Gas burner correlation between the total energy requirement and energy related specific CO2 emissions was not
evident. Whilst the specific consumption of NG in the EAF only had a minor impact on the EAF energy
efficiency, it decreased the specific electrical energy requirement and increased EAF productivity where
transformer power was restricted. The analysis also demonstrated that complementing and substituting
electrical energy with NG was beneficial in reducing the total energy related CO2 emissions when
a certain level of substitution efficiency was achieved. Therefore, the appropriate use of NG burners in
modern EAFs can result in an increased EAF energy intensity, whilst the total energy related CO2
emissions remain constant or are even decreased.
Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction the high proportion of recycled scrap and the use of electrical
energy (1389–4250 kWh/tsteel and 0:15—1:08 tCO2 =tsteel , respec-
32% of total world steel production is based on melting of scrap, tively [56]). Detailed calculations and estimations of energy
direct reduced iron, ferro-alloys and additives in the EAF (40.5% in intensities in the steel industry and life cycle analysis have been
EU-25 2006 [19], and 31% in Germany 2007 [62]). The trend for EAF published [1,13,18,26,37,39,43,45,50,51,57,63–65]. The total CO2
steelmaking is increasing due to the high flexibility in producing emissions comprise significant contributions from process related
low alloyed to high alloyed steel grades, due to the range of emissions (e.g., steel liquid decarburization due to oxygen injection,
insertable input materials from scrap, sponge iron or direct reduced electrode graphite consumption, carbon fines injection for slag
iron and hot metal, and due to the combination of electrical and foaming, and process slag reduction) and energy related emissions
chemical energy input. A global LCI study defined total primary (e.g., NG burners and indirect emissions resulting from the
energy consumption of steelmaking to 5958–8806 kWh/t with production and provision of electrical energy, Fig. 1). Dissolved
a total CO2 emission between 1:61 and 2:60 tCO2 =tsteel [56]. In EAF carbon from input materials (e.g., scrap, sponge iron, direct reduced
steelmaking, the total energy consumption and CO2 emissions iron, ferro-alloys, oil contamination, charge coal, and injected
levels are significantly lower than for oxygen steelmaking due to carbon) is oxidized in the steel melt in order to achieve the required
low carbon concentration, avoid chromium oxidation during
oxygen injection in stainless steelmaking, and to generate foaming
slag for high energy conversion efficiency. E.g., for the EU emission
* Corresponding author. Present address: RHI AG, Technology Center,
Magnesitstraße 2, A-8700 Leoben, Austria. Tel.: þ43 (0)502 13 5406; fax: þ43
trading period 2005–2007, 101.5 Mio t emission certificates were
(0)502 13 5182. allocated to 39 steel plants in Germany (6.8% of total 1485 Mio. t
E-mail address: marcus.kirschen@rhi-ag.com (M. Kirschen). CO2). 81% of the emissions were classified as process related

0360-5442/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.energy.2009.04.015
 Author's personal copy

1066 M. Kirschen et al. / Energy 34 (2009) 1065–1072

Post-combustion EAF vessel (Fig. 1) any additional energy input, for example from
Electrical with air scrap pre-heating systems, must be accounted for separately as
Gas energy
well as the additional energy contributions from the input of hot
oxygen metal and hot direct reduced iron. The EAF energy balance is given
burner
To coarse separator, by Eq. (1):
gas cooling and filter
system Tapping
Z
2
ETotal ¼ PElectric dt þ DHOxygen injection þ DHNG burners
1 Oxygen
injector Charging
Z Z
System boundary ¼ DHSteel þ DHSlag þ DH_ Off -gas dt þ DQ_ Cooling dt
for energy balance Z
Fig. 1. Process related (1) and energy related (2) CO2 emissions in an EAF.
þ DQ_ Radiation; other losses dt (1)

Time integration of the enthalpy and heat flow rates is performed

emissions, and 19% as combustion related emissions from NG, oil or from the beginning of charging the furnace to the end of tapping,
R
coal combustion. i.e. tap-to-tap time. The electric energy input EElectrical ¼ Pdt is
Precisely determining the total energy input into an EAF is exactly determined from operation data of the steel plant. The steel
complex because energy is supplied to the EAF from multiple and slag enthalpies are dependent on the tapped masses and
sources: Electrical energy as well as chemical energy that is specific enthalpies, namely the actual chemical compositions. Heat
released from the combustion of NG, liquefied petroleum gas or oil, losses from the system are due to water-cooling of the furnace
and due to the oxidation of elements in the melt during refining, for vessel, roof, and hot gas duct as well as radiation losses to the
example C, Si, Al, Fe, Cr, and Mn (Fig. 1). The latter energy contri- surroundings during charging and lining repair when the EAF roof
bution is not only dependent on the oxygen input but also on the is open. Ohmic losses from the high current system between the
chemical composition of the raw steel melt and slag. The electrical transformer and electric arc account for few %. The chemical energy
energy requirements of EAF steel plants in Germany are in the input is the sum of the reaction enthalpies (Eq. (2)) due to metal
range of 500 kWh/t. Decreasing the electrical energy requirement (Me) oxidation (e.g., C, Si, Mn, Al, P, Cr, Ni, and Fe):
and decreasing the mean specific NG consumption (i.e., 21.4 m3/t in X
1991 to 10.5 m3/t in 2007, Fig. 2) correlated with an increased DHOxygen injection ¼ mMe DhMe (2)
specific oxygen consumption (i.e., 24.4 m3/t to 35.6 m3/t, respec- Me
tively). Similar values have been reported for EAFs in the U.S. mMe is the mass of oxidised element Me. DrhMe is the specific
(500 kWhel/t, 23 kg/t coal, 34 m3/t oxygen and 8.5 m3/t NG [57], reaction enthalpy of Me oxidation. For carbon, mC in Eq. (2) denotes
and 481 kWhel/t [63]). the process related emission due to decarburization of the steel
In this paper EAF energy balances are examined that were melt, and oxidation of injected coal fines for slag foaming
calculated from actual measured plant data as well as energy (FeO þ C ¼ Fe þ CO). The specific reaction enthalpies range from
balances compiled from the literature. Using these figures the 1.32 kWh/kgFe for FeO, 2.55 kWh/kgC for CO, to 8.94 kWh/kgSi for
influences of the electrical energy requirement and NG consump- SiO2 (Table 1). Energy and CO2 emission levels resulting from the
tion on EAF energy efficiency and energy related CO2 emissions combustion of different NG grades with oxygen, DHNG burner ¼ huVG,
were determined. are detailed in Table 2.
The total energy intensities in the EAFs examined ranged from
2. Internally consistent energy balances of the EAF 510 kWh/t to 880 kWh/t (Fig. 3). An increased energy efficiency
requires decreased energy losses to the EAF off-gas and water-
EAF energy sources comprise electrical energy and energy cooling systems. The large variation in energy losses to these sinks
generated from oxidation reactions during refining (Fig. 1, Eq. (1)). indicates a high potential for energy savings. The CO2 emission
NG burners are used to increase energy intensity in the EAF, avoid intensities of EAFs ðmCO2 =ETotal Þ are in the order of 0:15 kgCO2 =kWh
cold spots in eccentric furnace designs, accelerate the scrap melting
time, and increase productivity in the case of restricted transformer
power. When the energy balance boundary encompasses only the
Table 1
Exothermic oxidation reactions and energy released during steel melt refining.
600 80
Electrical energy input [kWh/t]

Energy released
Specific consumption [m3/t]

Electrical energy input

561.9 Spec. oxygen consumption Reactions in melt
Si þ O2 / SiO2 8.94 kWh/kgSi 11:2 kWh=m3O2
550 Spec. gas consumption 60
Mn þ 0.5O2 / MnO 1.93 kWh/kgMn 9:48 kWh=m3O2
2Cr þ 1.5O2 / Cr2O3 3.05 kWh/kgCr 9:42 kWh=m3O2
533.1
2Fe þ 1.5O2 / Fe2O3 2.05 kWh/kgFe 6:80 kWh=m3O2
500 40 Fe þ 0.5O2 / FeO 1.32 kWh/kgFe 6:58 kWh=m3O2
35.6 C þ 0.5O2 / CO 2.55 kWh/kgC 2:73 kWh=m3O2
24.4
2Al þ 1.5O2 / Al2O3 5.29 kWh/kgAl 13:84 kWh=m3O2
450 20 Mo þ O2 / MoO2 1.70 kWh/kgMo 7:29 kWh=m3O2
20.7 10.5 S þ O2 / SO2 2.75 kWh/kgS 3:94 kWh=m3O2
2P þ 2.5O2 / P2O5 5.54 kWh/kgP 8:58 kWh=m3O2
400 0
Reactions in gas phase
1990 1995 2000 2005
C þ O2 / CO2 9.10 kWh/kgC 4:88 kWh=m3O2
CO þ 0.5O2 / CO2 7:01 kWh=m3O2
Fig. 2. Mean specific electrical energy input and specific consumption of oxygen and
H2 þ 0.5O2 / H2O 5:99 kWh=m3O2
natural gas in EAF steel plants in Germany [7,62].
 Author's personal copy

M. Kirschen et al. / Energy 34 (2009) 1065–1072 1067

Table 2
Typical caloric and CO2 emission levels resulting from the combustion of different NG grades.

NG Grade hu [kWh/m3] Gas density [kg/m3] CO2 emissionsa ½kgCO2 =MJ CO2 emissionsa ½kgCO2 =kWh CO2 emissionsa ½kgCO2 =m3  CO2 emissionsa [kgC/kgG]
Combined L 9.28 0.8161 0.0562 0.2023 1.87 0.627
Holland L 9.33 0.8288 0.0563 0.2027 1.89 0.622
Russia H 9.97 0.7304 0.0549 0.1976 1.97 0.735
Mixed H 10.48 0.8128 0.0564 0.2030 2.12 0.714
North Sea H 10.70 0.8147 0.0565 0.2034 2.17 0.729
a
Combustion with air, additional indirect emissions from production and provision of oxygen to oxy-fuel burners are approx. 0.2 kg/m3G (Table 4).

[63]. Recently measured CO2 intensities from EAFs in Germany 2.2. Complete EAF energy balances
ranged from 0:11 to 0:21 kgCO2 =kWh [27].
Complete EAF energy balances were determined for EAFs with
and without scrap pre-heating based on plant measurements [28–
2.1. Potential underestimations of the complete EAF
30] and from the literature (Table 3). Since energy sources as well as
energy requirements
energy sinks were measured, calculated or estimated, the calcu-
lated total energy requirement was supported twofold and there-
When estimating the complete energy balance of an EAF there
fore self-consistent. Any compensations necessary to equate the
are several factors that can result in underestimating the total
energy balances were typically only a few percent. From the
energy requirement. The energy that is released from the oxidation
assessment of complete energy balances it was evident that the use
of C is the enthalpy of formation of CO2 (i.e., 9.10 kWh/kgC, Table 1)
of NG burners in EAFs decreased the specific electrical energy
although C is oxidized to CO in the steel melt. The remaining CO
requirement by increasing the productivity (Fig. 4) although the
concentration in the off-gas accounts for the chemical energy loss
specific total energy requirement decreased only slightly. However,
to the off-gas extraction system. If the enthalpy of formation of CO
the energy efficiency of the EAF did not increase automatically with
(i.e., 2.55 kWh/kgC) is accounted for in the energy balance instead
the use of NG burners (Fig. 5) but required further measures to
of CO2 (i.e., 9.10 kWh/kgC), then both the off-gas enthalpy and the
use the NG burners efficiently in the EAF, for example the use of
total energy requirement are underestimated. The energy contri-
burners only during the first minutes of melting resulting in high
bution from efficient internal post-combustion is related to the
heat transfer to the cold scrap pile.
oxidation of CO to CO2 using O2 injectors located inside the EAF.
Since the specific enthalpy of steel at tapping was defined as the
Determining the input from chemical energy is complex due to
process benefit, which depends only on the tapping temperature at
the oxidation of different elements, for example C, Si, Al, Cr, and
a given steel composition, the total energy efficiency (h ¼ DHSteel/
Fe (Table 1). Accurately determining the oxidation products
ETotal) decreased with increasing total energy input (Fig. 6). For this
requires off-gas measurements (e.g., CO and CO2) and analysis of
case, the enthalpy of the slag at tapping was considered as heat loss.
slag samples. However, energy models that use mean values for
Additional energy input exceeding the enthalpy of steel at 1600  C
the chemical energy as a function of injected O2 tend to under-
tapping temperature (i.e., 361 kWh/t for low alloyed carbon steel
estimate the chemical energy contribution. Besides the
and 372 kWh/t for high alloyed stainless steel) decreased the
complexity of oxidation reactions, an additional reason for this
energy efficiency of the EAF (Fig. 5).
shortcoming may be neglecting the O2 present in the infiltrated
Fruehan et al. [13] determined the practical minimum energy
air involved in post-combustion of CO that is not accounted for by
of an EAF to be 444 kWh/t. Whereby 444 kWh/t was the
the mass of injected O2.
minimal energy requirement to melt the scrap and superheat the
melt and basic slag (i.e., 25% FeO, CaO/SiO2 ¼ 2.5) to 1600  C and
this corresponded to an energy efficiency of 82% if the process
benefit was defined as the specific enthalpy of the steel melt. If
Gas/Oil Burners 2% - 10% the sum of enthalpies of tapped steel and slag are both
Electrical considered as the process benefit, Fruehan’s practical minimum
Energy Oxygen Injection 20% - 50%
energy demand of 444 kWh/t corresponds to an energy effi-
40% - 65%
ciency of 100%. Fruehan’s minimum energy demand of 361 kWh/
t [13], is the specific enthalpy of unalloyed steel at 1600  C.
However, enthalpy sinks to the off-gas and slag cannot be
avoided completely due to CO emissions during steel melt
refining in the EAF, as well as the separation of SiO2, Al2O3, and
Total 650-850 kWh/t P2O5 for example from the melt by deslagging, and the necessary
direct dedusting in a modern EAF.

3. Influence of NG combustion in EAFs on the total energy

related CO2 emissions

Steel Off-gas 15% - 35%

NG burners are commonly used only during the first 5–15 min
45% - 60% Slag & Dust 5% - 10% for melting after charging scrap into the EAF, because the heat
Cooling 10% - 20% exchange to the scrap decreases drastically with increasing scrap
Radiation few % temperature and a decreasing specific surface of the melting scrap
Fig. 3. Energy sources in an EAF (e.g., electrical energy and chemical energy resulting
pile in an EAF. The lowest heat transfer rates from NG burners occur
from oxidation reactions) and energy sinks (e.g., off-gas extraction and furnace cooling under flat bath conditions when the steel is fully covered with
systems). foamy slag. The caloric values of NGs vary between
 Author's personal copy

1068 M. Kirschen et al. / Energy 34 (2009) 1065–1072

Table 3
Complete energy balances of electric arc furnaces.

Tapping Ref. Electrical Oxid. react. Gas burners Scrap preh. Steel Slag Off-gas Cooling, Total
weight [t] Energy [kWh/t] [kWh/t] [kWh/t] [kWh/t] [kWh/t] [kWh/t] [kWh/t] Radiation [kWh/t] [kWh/t]
100 [11] 541 234 23 423 80 168 127 798
55 [2] 571 155 415 59 100 152 726
60 [46] 408 204 20 51 367 68 109 139 683
80 [46] 560 118 382 50 81 165 678
100 [2] 577 155 415 59 33 225 732
150 [12] 450 199 385 35 102 127 649
[42] 390 170 60 12 345 50 170 67 632
100 [3] 557 233 21 427 83 158 142 810
115 [34] 487 217 397 46 127 134 704
125 [34] 462 94 34 70 401 70 48 140 660
100 [14] 459 295 354 62 138 199 754
100 [14] 413 378 354 62 140 235 791
100 [14] 401 397 354 62 141 241 798
100 [14] 445 341 355 47 141 243 786
100 [14] 423 325 61 355 47 141 266 809
100 [58] 91 249 182 106 385 50 68 125 628
80 [16] 380 195 15 388 52 85 65 590
100 [41] 365 222 33 5 431 110 84 625
100 [41] 287 223 33 63 431 97 78 606
75 [49] 459 230 7 362 49 90 195 696
75 [49] 422 237 362 40 105 152 659
70 [54] 477 187 392 62 96 114 664
150 [52] 312 278 88 320 47 229 83 678
150 [52] 300 377 97 355 53 232 135 775
55 [9] 347 399 86 407 26 317 82 832
[21] 368 348 38 380 32 155 187 754
[21] 482 170 25 360 31 160 126 677
60 [6] 465 215 36 408 72 150 86 715
60 [6] 429 215 21 50 408 72 150 86 715
[10] 450 130 90 395 35 122 118 670
135 [60] 400 190 40 385 50 135 50 630
135 [60] 285 190 50 55 385 50 65 70 580
92 [5] 410 180 40 380 50 140 60 630
92 [5] 340 180 40 380 50 58 72 560
50 [23] 338 198 385 46 47 58 536
[23] 420 190 30 365 60 130 78 633
[23] 580 125 365 80 50 210 705
[15] 400 190 40 385 50 135 60 630
[40] 450 195 60 382 76 135 112 705
[40] 320 236 80 382 76 58 120 636
[40] 260 192 60 382 69 21 40 512
125 [35] 490 123 160 386 78 158 40 773
145 [29] 477 281 408 41 128 181 758
100 [27] 394 335 55 388 57 170 181 784
140 [27] 431 342 429 89 127 128 773
120 [27] 497 343 399 46 125 270 840
120 [20] 343 285 118 375 59 116 196 746
30 [27] 570 204 382 118 153 121 774
85 [28] 494 248 12 383 30 79 262 754
145 [28] 510 271 365 36 121 259 781
100 [36] 485 114 342 48 66 144 600
100 [36] 390 112 119 356 50 69 150 624
120 [4] 487 217 397 46 127 135 704
60 [53] 427 256 15 31 381 48 131 169 729
120 [61] 323 305 119 380 52 238 75 745
[35] 578 85 42 402 56 120 127 705
[35] 557 251 430 81 154 146 811
[8] 410 213 40 379 53 160 73 665
[8] 340 201 47 378 47 83 83 590
[8] 310 202 50 381 45 50 84 560
40 [38] 371 296 63 393 105 114 118 730
128 [38] 413 297 15 407 63 137 118 725
30 [48] 350 417 292 46 123 306 767
36 [33] 372 446 65 381 96 167 239 883
[25] 368 346 38 380 32 155 185 752

hu,G ¼ 9.28 kWh/m3G and hu,G ¼ 10.70 kWh/m3G (Table 2). If a mean [47], the electrical energy savings realizable through the substitu-
efficiency of heat transfer from the burners to scrap, hG ¼ 50–60%, is tion of electrical energy by NG combustion can be determined [47]:
considered [22], with an electrical efficiency of energy transfer in
the high current system, hel ¼ 90–95% [47], and the efficiency of hG hu;G kWh
Deel;G ¼ ¼ 11:5 to  6:9 3 (3)
heat transfer from electric arc to melt and scrap, helharc ¼ 60–80% hel hG mG
 Author's personal copy

M. Kirschen et al. / Energy 34 (2009) 1065–1072 1069

900 100
800 90

Energy Efficiency [%]

80
Energy Input [kWh/t]

700
70
600
y = -0,185x + 701 60
500 Fruehan (2000) [13]
50
400 40 Enthalpy Steel at Tapping (1600°C):
300 30 361 to 372 kWh/t

200 20
Total Energy Input [kWh/t] y = -1,236x + 461
100 10
Electrical Energy Input [kWh/t]
0
0
350 450 550 650 750 850
0 50 100 150 200
Natural Gas Enthalpy [kWh/t] Total Energy Input [kWh/t]

Fig. 6. Relationship between total energy input and energy efficiency of an EAF if the
Fig. 4. Total energy input and electrical energy requirement as a function of NG burner
enthalpy of steel is considered as the process benefit.
enthalpy for EAFs determined from complete energy balances.

These values are supported by other studies (Deel,G ¼ 8.5 to Besides the 2 kg/m3G direct CO2 emissions of NG, indirect emis-
7.8 kWh/m3G [31] and 11.2 kWh/m3G [35], respectively). Köhle sions from the production and provision of oxygen, with
determined the correlation coefficient between the electrical a maximum estimation of 400 kWhel =tO2 [24], i.e. 0:2 kgCO2 =m3O2 or
energy requirement and NG input by linear regression of process 0:4 kgCO2 =m3G , was considered (Table 4). The substitution of elec-
data from 60 EAF plants to be 8 kWh/m3G [32]. trical energy with NG burners in an EAF was beneficial at a corre-
Additional negative correlation coefficients between NG lation coefficient lower than 6.6 kWhel/m3G for 0:365 kgCO2 =kWhel
consumption and the electrical energy requirement are depicted in (Fig. 8). The correlation coefficient decreased to 10.7 kWh/m3G
Fig. 7 for 2 cases: 9.6 kWh/m3G for both EAF B with a low specific with a decreasing electrical energy CO2 emission level, for example
gas consumption and EAF A also operating at a gas consumption 0:224 kgCO2 =kWhel in Canada (Table 4 [17]). Whilst, a correlation
level lower than 7 m3G/t. The efficient substitution of electrical between energy efficiency and total CO2 emissions could not be
energy by NG in EAF A was not maintained at gas consumption established for EAFs using this data due to the high proportion of
levels higher than 7 m3G/t, as additional production parameters process related carbon CO2 emissions, calculations performed with
influence the tap-to-tap time and the electrical energy require- the data indicated that the total energy related specific CO2 emis-
ment. Under these conditions the efficiency of the NG burners sions did not increase with the use of NG burners in the EAF but
decreased significantly below hG ¼ 50–60% [22] with a resulting decreased slightly (Fig. 9) if the NG burners were used efficiently.
low NG substitution potential and high heat losses to the off-gas
and cooling systems. 4. Correlations between total CO2 emission and specific
The efficiency of NG combustion in EAFs was also compared to electrical energy requirement
the decreasing electrical energy requirement for a given specific
electrical energy CO2 emission level (e.g., 0:365 kgCO2 =kWhel in Besides CO2 emission from the combustion of NG, the oxidation
Table 4). When the substitution of electrical energy by NG burners of carbon due to metallurgical reasons provides additional CO2
achieved a certain substitution efficiency (Eq. (4)) in the EAF, then emissions in the EAF (Fig. 1). Carbon is introduced to the EAF with
the total specific CO2 emissions of NG burners in the EAF and the ferrous input raw materials (e.g. direct reduced iron, dissolved
electrical energy, DmCO2 ; NG þ DmCO2 ; Electrical energy input ¼ 0, in ferro-chromium, -nickel or -silicon alloys, and oil and grease
remained the same, for example: contamination of scrap), and must oxidized below a certain level
before refining and casting. Coal is injected into the EAF to generate
2:4 kgCO2 =m3G kWhel a foaming slag during melting and added to high alloyed steel melts
KG ¼ ¼ 6:6 (4) in order to decrease chromium oxidation during refining of the
0:365 kgCO2 =kWhel m3G
melt. For EAF steelmaking, total CO2 emission values have been

100 Table 4
90 Direct and indirect CO2 emission levels resulting from gas combustion and electrical
energy input in EAFs.
Energy Efficiency [%]

80
Specific CO2 emissions kgCO2 =m3 kgCO2 =kWh Required
70
substitution
60 efficiency
50 Natural gas (NG) 2.0
Oxygena 0.2
40
Total oxy-fuel burners 2.4 0.24
30 Electrical energyb
Germany, gas combustion 0.365 6.6
20
Germany, total mix 0.596 4.0
10 Great Britain 0.473 5.1
0 China 0.771 3.1
0 50 100 150 200 Canada 0.224 10.7
U.S. 0.575 4.2
Natural Gas Enthalpy [kWh/t]
a
Assumed electrical energy requirement: 400 kWhel =tO2 max.
b
Fig. 5. Influence of NG input on EAF energy efficiency. Data from UBA Germany and [17].
 Author's personal copy

1070 M. Kirschen et al. / Energy 34 (2009) 1065–1072

Electrical Energy Requirement [kWh/t]

700 0,4

Energy Related CO2 emission

WR M
E0 9.558 3 G
kWh / t m /t
600
0,3
500

[tCO2/tSteel]
400
0,2
300
WR M
E0 1.881 3 G
kWh / t m /t 0,1
200
EAF A
100
EAF B
0,0
0 0 50 100 150 200
0 5 10 15 20 Natural Gas Enthalpy [kWh/t]
Natural Gas Consumption [m3/t]
Fig. 9. Specific energy related CO2 emissions and NG input in EAFs (data from Fig. 4,
Fig. 7. Correlation between specific NG consumption and electrical energy input for with e.g., 0:365 kgCO2 =kWhel ).
two EAFs.

Electrical energy demand [kWh/t]

800
reported from 364 to 416 kgCO2 =t [13], 441 kgCO2 =t [18], and
459 kgCO2 =t [37] for steel scrap charges and above 632 kgCO2 =t for
combined EAF feeding with steel scrap, direct reduced iron and pig 600 1
iron in mini steel mills [18]. Mean values of total CO2 emission were
in the order of 250 kgCO2 =t for EAF steel plants in Germany and near
400
350 kgCO2 =t in Europe [55]. Detailed carbon mass balances and EAF
off-gas measurements, e.g. [27–30], provided specific CO2 emis-
sions from 84 to 161 kgCO2 =t. Requested and allocated CO2 emis- 200
sion certificates to EAF steel plants in Germany and Great Britain, EAF 1, high alloyed
2005, were in the same range. Whilst the influence of NG EAF 2, high alloyed
combustion to the electrical energy requirement was obvious, e.g. 0
Fig. 4, the correlation between measured total CO2 emission and 0 5 10 15 20 25 30 35
electrical energy requirement was more complex due to the high Emitted carbon mass in off-gas [kg/t]
proportion of process related carbon input.
At high carbon content of the melt and with restricted oxygen
Electrical energy demand [kWh/t]

800
injector technology, the increasing time period for oxygen injection
and refining of the melt increases the tap-to-tap time and, there-
fore, time-dependent energy losses to off-gas and cooling systems 600 3
and by radiation to the surrounding environment.
The emitted carbon mass was measured in the off-gas systems
of 6 EAFs and given in Fig. 10 versus the electrical energy require- 400 4
ments for multiple heats. For the production of low alloyed and
high alloyed steels with and without gas burners in EAF no. 1, 3, 5,
and 6, the electrical energy requirement is increasing with total 200
EAF 3, high alloyed
oxidized carbon mass. Kühn [35] also reported a positive correla-
tion coefficient between injected coal fines and electric energy EAF 4, low alloyed
input, 1.804 kWhel/kgC, for a 125 t EAF. However, a negative corre- 0
0 5 10 15 20 25 30 35
lation between carbon emission and electric energy requirement
Emitted carbon mass in off-gas [kg/t]
Electrical energy demand [kWh/t]

800
Electrical Energy Input [kWh/t]

Inefficient substitution of electrical energy by NG

regarding for a specific CO2 emission level
5
600

6
Efficient substitution of electrical energy 400
by NG regarding for a specific CO2 emission level

-3.0 kWh/m3
200
-6.6 kWh/m3 EAF 5, low alloyed
-8.0 kWh/m3 [Köhle 2002]
EAF 6, high alloyed
-11.2 kWh/m3 [Kühn 2003]
0
0 5 10 15 20 25 30 0 5 10 15 20 25 30 35
Natural Gas Consumption [m 3/t] Emitted carbon mass in off-gas [kg/t]

Fig. 8. Correlations between electrical energy input and specific gas consumption in an Fig. 10. Correlations between electrical energy requirement and measured emitted
EAF for a defined CO2 emission level ð0:365 kgCO2 =kWhel Þ. total carbon mass in EAF off-gas.
 Author's personal copy

M. Kirschen et al. / Energy 34 (2009) 1065–1072 1071

was shown for EAF no. 4 indicating a substitution of electrical electric furnace conference proceedings. Orlando, Florida: Iron and Steel
Society; 2000. p. 39–48.
energy by chemical energy from carbon oxidation. A clear corre-
[10] Fedele Delloste E, Galluzzo S, Garzitto V. Heat recovery from an electric
lation between electric energy input and carbon emission was not arc furnace for scrap preheating. In: 1st European electric steelmaking
evident for EAF no. 2. Fig. 10 indicates that the input of carbon and conference proceedings. Aachen: Verein Deutscher Eisenhüttenleute VDEh;
the corresponding CO2 emission during refining yields an energetic 1983. p. H3.
[11] Fett F, Pfeifer H, Siegert H. Energetische Untersuchung eines Hochleis-
and economic disadvantage to the producers of high alloyed and tungslichtbogenofens. [Investigations on energy balance of a UHP arc furnace].
special steel grades, independently from additional costs to CO2 Stahl und Eisen 1982;102(9):461–5.
emission. Decarburization rates of the applied oxygen injector [12] Franzen H, Reichelt W. Umwandlung und Nutzung der Restenergien an hüt-
tentechnischen Anlagen. [Conversion and use of residual energy of metallur-
technology (oxygen lances, late oxygen injection, low specific gical plants]. In: 12. Metallurgisches GDMB Seminar Wärmetechnik im
oxygen consumption) seemed to be too low for an energy-efficient Hüttenwesen; 1981 [Bad Neuenahr].
process. A shift of carbon oxidation from the EAF to the subsequent [13] Fruehan R, Fortini O, Praxton H, Brindle R. Theoretical minimum energies to
produce steel for selected conditions. In: Report for U.S. Department of Energy,
metallurgical aggregates for decarburization of high alloyed steel Office of Industrial Technologies. Carnegie Mellon University Pittsburgh;
melts, e.g. argon oxygen decarburization converter or vacuum ladle March 2000.
furnace with higher oxygen efficiency and higher rate of decarbu- [14] Grant M. Principle and strategy of EAF post-combustion. In: 58th electric
furnace conference proceedings. Orlando, Florida: Iron and Steel Society;
rization, is beneficial for a lower electric energy requirement and 2000. p. 15–28.
higher energy efficiency in the EAF. On the other hand, off-gas [15] Haissig M, Fuchs G, Auer W. Electric arc furnace technology beyond the year
monitoring at the EAF is an appropriate tool to control CO post- 2000. MPT International 1999;(1):56–62.
[16] Hamy M, Koller F. Performance of UHP-EAF at Badische Stahlwerke. In: 6th
combustion and CO2 emission and to increase NG and oxygen
European electric steelmaking conference proceedings. Düsseldorf: Verein
conversion and EAF energy efficiency, e.g. [9,14,20,33,49,66]. Deutscher Eisenhüttenleute VDEh; 1999. p. 29–34.
[17] Höhne N, Phylipson, Moltman S. Factors underpinning future action 2007
update. Report for the DEFRA. UK: Ecofys; 2007.
5. Conclusion [18] Hornby-Anderson S, Metius G, Mcclelland J. Future green steelmaking tech-
nologies. In: 60th electric furnace conference proceedings. San Antonio, Texas:
Iron and Steel Society; 2002. p. 175–91.
The assessment of multiple EAF energy balances did not reveal [19] International Iron and Steel Institute. Steel statistical yearbook 2007; 2007.
a significant influence of NG consumption on the total energy input Brussels [Report].
[20] Jannasch O. Steuerung der Nachverbrennung im Elektrolichtbogenofen
in an EAF but did demonstrate an associated decrease in the elec- durch Abgasmessung mittels Infrarotspektroskopie (Control of post-combus-
trical energy requirement. Furthermore, a decrease in energy losses tion in an electric arc furnace by use of infrared spectroscopy). RWTH
to off-gas systems (e.g., by off-gas monitoring and extraction Aachen University, Faculty for Mining, Metallurgy and Earth Sciences, PhD
diss.; 2002.
control) and cooling systems (e.g., using high quality refractory [21] Jones J. Understanding energy use in the EAF. In: 60th electric furnace
products) will increase energy efficiency and thereby decrease CO2 conference proceedings. San Antonio, Texas: Iron and Steel Society; 2002.
emissions. p. 141–54.
[22] Jones J. New steel melting technologies: part II, oxy-fuel burner application in
It was also demonstrated that the substitution of electrical
EAF. Iron and Steelmaker 1996;23(5):63–5.
energy by NG in EAFs is beneficial in reducing total CO2 emissions if [23] Jones J. New steel melting technologies: part XIV, Comelt, Conarc and Con-
a certain level of substitution efficiency is achieved, namely DEel tiarc. Iron and Steelmaker 1997;24(10):109–12.
[24] Kato T, Kubota M, Kobayashi N, Suzuoki Y. Effective utilization of by-product
[kWhel/t] < 3.1 to 10.7 VGas [m3G/t] for a specific electrical energy
oxygen from electrolysis hydrogen production. Energy 2005;30:2580–95.
CO2 emission level. Comprehensive investigations have reported [25] Kendall M, Thys M, Horrex A, Verhoeven J. A window into the electric arc
correlation coefficients between the electrical energy requirement furnace, a continuous temperature sensor measuring the complete furnace
and NG input of 8 to 11.2 kWhel/m3G [31,32,35]. Therefore, the cycle. In: 9th European electric steelmaking conference proceedings. Krakow:
Archives of Metallurgy and Materials, vol. 53; 2008. p. 451–4.
highly efficient use of NG in EAFs is beneficial to maximize material [26] Kim Y, Worrell E. International comparison of CO2 emission trends in the iron
and energy conversion whilst minimizing CO2 emissions during and steel industry. Energy Policy 2002;30:827–38.
EAF steelmaking. [27] Kirschen M, Pfeifer H, Risonarta V, Velikorodov V, Voj L. Off-gas measurements
at electric arc furnaces. RWTH Aachen University – Institute for Industrial
Furnaces and Heat Engineering; 2000–2005.
[28] Kirschen M, Pfeifer H, Wahlers F. Mass and energy balances of stainless
References steel electric arc furnace. In: 7th European electric steelmaking confer-
ence proceedings. Venice, Italy: Associatione Italiana di Metallurgia; 2002.
[1] Aichinger M. 5. CO2-Monitoring-Fortschrittsbericht der Stahlindustrie in p. 2.3–2.12.
Deutsch-land für die Berichtsjahre 2000 bis 2003. [CO2 monitoring report of [29] Kirschen M, Pfeifer H, Wahlers F, Mees H. Off-gas measurements for mass and
the steel industry in Germany for 2000 to 2003]. Stahlinstitut VDEh Düssel- energy balances of a stainless steel EAF. In: 59th electric furnace conference
dorf; 2005. proceedings. Phoenix, Arizona: Iron and Steel Society; 2001. p. 737–45.
[2] Altfeld K, Schneider A. Energiebilanzen von Elektrolichtbogenöfen - Auswir- [30] Kirschen M, Velikorodov V, Pfeifer H, Kühn R, Lenz S, Loh J, et al. Off-gas
kungen wassergekühlter Ofenelemente. [Energy balances of electric arc furnaces- measurements at the EAF primary dedusting system. In: 8th European electric
implications of water cooled panels]. Stahl und Eisen 1982;102(20):979–84. steelmaking conference proceedings. Birmingham: Institute of Materials,
[3] Ameling D, Baum R, Köhle S, Kreutzer H. Entwicklungsrichtungen bei der Minerals and Mining; 2005. p. 563–76.
Stahlerzeugung in Lichtbogenöfen. [Trends of electric arc furnace steel- [31] Klein H, Schindler J. Metallurgie bei Schrotteinsatz. In: Heinen K, editor.
making)]. Stahl und Eisen 1981;101(4):197–207. Elektrostahlerzeugung. Verlag Stahleisen Düsseldorf; 1997. p. 473–502.
[4] Brod H, Kempkens F, Strohschein H. Energierückgewinnung aus einem UHP- [32] Köhle S. Recent improvements in modelling energy consumption of electric
Elektrolichtbogenofen. [Energy recovery from an UHP electric arc furnace]. arc furnaces. In: 7th European electric steelmaking conference proceedings.
Stahl und Eisen 1989;109(5):229–38. Venice, Italy: Associatione Italiana di Metallurgia; 2002. p. 1.305–1.314.
[5] Clayton J, Ehle J, Twiselton J, Knapp H. The Sheerness shaft furnace. In: SEACI [33] Krassnig H, Kleimt B, Voj L, Antrekowitsch H. EAF post-combustion control by
conference proceedings. Sendai, Japan; 1993. p. 1–7. on-line laser-based off-gas measurement. In: 9th European electric steel-
[6] Dening W, Klein K, Paul G, Schindler J. Schrottvorwärmung bei Badische making conference proceedings. Krakow: Archives of Metallurgy and Mate-
Stahlwerke. [Scrap pre-heating at Badische steel works]. In: 1st European rials, vol. 53; 2008. p. 455–62.
electric steelmaking conference proceedings. Aachen: Verein Deutscher [34] Krüger K, Timm K. Energieverbrauch und Produktivität von Lichtbogenöfen.
Eisenhüttenleute VDEh; 1983. p. H1. (Energy consumption and productivity of arc furnaces). Elektrowärme
[7] Statistical Agency Germany. Statistical year book, <http://www.destatis.de>; International 1996;54(4):B179–85.
2008 [version]. [35] Kühn R. Untersuchungen zum Energieumsatz in einem Gleichstromlicht-
[8] Ehle J. Neuere Entwicklungen des Drehstrom-Lichtbogenofens. [New devel- bogenofen zur Stahlerzeugung (Investigations of energy efficiency of a DC arc
opments of the AC electric furnace]. In: VDEh-Seminar 14/97 Elektrotechnik furnace), Technische Universität Clausthal-Zellerfeld, PhD diss.; 2003.
des Licht-bogenofens. Düsseldorf: Verein Deutscher Eisenhüttenleute VDEh; [36] Linninger A, Hofer M, Patuzzi A. DynEAF – a dynamic modeling tool for
1997. integrated electric steelmaking. Iron and Steel Engineer 1995;(3):43–53.
[9] Evenson E, Goodfellow H, Kempe M. EAF process optimization through off-gas [37] Macedo-Costa M, Schaeffer R, Worrell E. Exergy accounting of energy and
analysis and closed-loop process control at Deacero, Saltillo, Mexico. In: 58th materials flows in steel production systems. Energy 2001;26:363–84.
 Author's personal copy

1072 M. Kirschen et al. / Energy 34 (2009) 1065–1072

[38] Mair A. Economic investigations of water cooled panels vs. entire refractory [52] Pujadas A, Mccauley J, Iacuzzi M. EAF energy optimisation at Nucor Yamato
wall lining of an electric arc furnace. Montanuniversität Leoben, Institut für Steel. In: AISTech2003 conference proceedings. Indianapolis, Indiana: Iron and
Verfahrenstechnik des industriellen Umweltschutzes, Master thesis; 2000. Steel Society; 2003. p. 391–402.
[39] Margolis N, Brindle R. Energy and environmental profile of the U.S. iron and [53] Reichelt W, Hofmann W. Contiarc – ein energieoptimiertes und umweltscho-
steel industry. Columbia: U.S. Department of Energy, Office of Industrial nendes Schrottschmelzverfahren. [Contiarc – the energy optimized and envi-
Technologies; August 2000 [Report of Energetics Inc.]. ronment friendly scrap melting process]. Stahl und Eisen 1996;116(5):89–92.
[40] Meierling P, Meger P, Espendiller B, Schmale K. Contiarc – the energy efficient [54] Reinhardt M, Reiber D. Practice of ALARC-PC post-combustion at Von Roll Ltd.,
EAF. Steel Times 1999;227(5):14–5. Gerlafingen. In: 5th European electric steelmaking conference proceedings.
[41] Nakano H, Uchida S, Arita K, Kuroda H. New scrap preheating system for Paris: Association Technique de la Siderurgie Francaise; 1995. p. 117–25.
electric arc furnace (UL-BA). In: 6th European electric steelmaking confer- [55] Risonarta V. Carbon mass balance and CO2 emission reduction in the arc
ence proceedings. Düsseldorf: Verein Deutscher Eisenhüttenleute VDEh; furnace. RWTH Aachen University, Institute for Industrial Furnaces and Heat
1999. p. 84–90. Engineering, Master thesis; 2005.
[42] Nikko Industry Co Ltd.. Outline of Nikko-type scrap preheating equipment; [56] Sandberg H, Lagneborg R, Lindblad B, Axelsson H, Bentell L. CO2 emissions of the
1995. Swedish steel industry. Scandinavian Journal of Metallurgy 2001;30:420–5.
[43] Ozawa L, Sheinbaum C, Martin N, Worrell E, Price L. Energy use and CO2 [57] Stubbles J. Energy use in the U.S. steel industry: an historical perspective and
emissions in Mexico’s iron and steel industry. Energy 2002;27:225–39. future opportunities. Columbia: U.S. Department of Energy, Office of Industrial
[45] Pfeifer H. Energy balance of the electric arc furnace. In: VDEh seminar – Technologies; September 2000 [Report of Energetics Inc].
electrical engineering of arc furnaces. Luxembourg: Verein Deutscher Eisen- [58] Toulouevski Y, Zinurov I. Outlook for reduction in energy consumption of electric
hüttenleute VDEh; 2007. arc furnaces. In: 7th European electric steelmaking conference proceedings.
[46] Pfeifer H, Fett F, Heinen K. Möglichkeiten zur Verbesserung der Energiebilanz Venice, Italy: Associatione Italiana di Metallurgia; 2002. p. 1.57–1.64.
des Lichtbogenofens. [Options to increase the energy efficiency of the arc [60] Vai Fuchs Systemtechnik. Electric steelmaking. ArcTrends 1998;5(2):1–8.
furnace]. Elektrowärme International 1988;46:B72–7. [61] Vallomy J, Fuse T, Nakamura S. Consteel process successful in USA – a 120 t/h
[47] Pfeifer H, Kirschen M, Simoes J. Thermodynamic analysis of EAF electrical unit started up in Japan. In: 50th electric furnace conference proceedings. Iron
energy demand. In: 8th European electric steelmaking conference proceed- and Steel Society; 1992. p. 309–13.
ings. Birmingham: Institute of Materials, Minerals and Mining; 2005. p. [62] German Iron and Steel Institute VDEh. Statistical yearbook of the steel
211–32. industry. Stahlinstitut VDEh Düsseldorf; 2008.
[48] Prakash S, Mukherjee K, Singh S, Mehrotra S. Simulation of energy dynamics [63] Worrell E, Martin N, Price L. Energy efficiency and CO2 emissions reduction
of electric arc furnace steelmaking using DRI. Ironmaking and Steelmaking opportunities in the US iron and steel sector. Lawrence Berkeley National
2007;34(1):61–8. Laboratory – Environmental Energy Technologies Division; July 1999 [Report
[49] Prestidge J, Lee M, Wibberley L. EAF process monitoring at OneSteels Sydney no. LBNL 41724].
Steel Mill. Steel Research 2002;73(5):5–8. [64] Worrell E, Martin N, Price L. Energy efficiency opportunities in electric arc
[50] Price L, Sintona J, Worrell E, Phylipsen D, Huc X, LID J. Energy use and carbon steelmaking. Iron and Steelmaker 1999:25–32.
dioxide emissions from steel production in China. Energy 2002;27:429–46. [65] Worrell E, Price L, Martin N. Energy efficiency and carbon dioxide emissions
[51] Price L, Worrell E, Phylipsen D. Energy use and carbon dioxide emissions in reduction opportunities in the US iron and steel sector. Energy 2001;26:513–36.
energy-intensive industries in key developing countries. Lawrence Berkeley [66] Wu Q, Thomson M, Chanda A. Tunable diode laser measurements of CO, H2O,
National Laboratory – Environmental Energy Technologies Division; and temperature near 1.56 mm for steelmaking furnace pollution control and
September 1999 [Report no. LBNL 45292]. energy efficiency. Metallurgical and Materials Transactions 2005;36B:53–7.

View publication stats