Optimization of the AOD stainless steel processing cost by the UTCAS System

(1)

Optimization of the AOD stainless steel

processing cost by the UTCAS System

Chay Ta-uar

800521-T290

(2)

Acknowledgement

First and foremost, I would like to thank to my supervisor of this project, Mr. Carl Johan Rick and Mr. Mikael Engholm for the valuable guidance and advice. He inspired me greatly to work in this project. They willingness to motivate me contributed tremendously to my project. I also would like to thank them for showing me some examples that related to the topic of my project. Besides, I would like to thank the authority of Uvan Hagfors Teknologi AB at Kista and Uddelholm for providing me with a good

(3)

1. Introduction

1.1 General

Stainless steel is widely used in various applications this is due to that stainless steel has a CrO3

film coat on the surface for anti-corrosion when the steel is exposed to a corrosive environment. Therefore, all stainless steels are corrosion resistance as well as provide ranges of strength, formability, and high or low temperature service. In addition, other elements also have different advantages when they are added to the steel. Thus the stainless steels are primary classified as austenitic, ferritic, martensitic, duplex or precipitation hardening grades depending on the composition and treatment. Typical wrought alloy AISI series designations include 200 (high manganese austenitic), 300 (austenitic), and 400 (ferritic or martensitic). Martensitic and ferritic steels are magnetic. Martensitic steels are typically hardened by heat treatment and are not easily formable. Austenitic steel grades can be harder by cold work. Duplex grades (austenitic/ferritic) are more resistant to stress corrosion cracking than austenitic and are tougher than ferritic grades. Precipitation hardened grades (martensitic or austenitic) are strengthened during heat treatment by precipitation hardening [1] as shown in Table 1.

Table 1 stainless steels grades including properties and applications [2]

stainless steel grade properties application

304 Austenitic Cr-Ni stainless steel ,non-magnetic, low carbon content, weld ability

Cooking equipment, pressure vessel, kitchen sink, food processing equipment etc. 316 Austenitic stainless steel, excellent corrosion

resistance, elongation and ductility

Mirror, marine, medical, food industry etc.

409 Ferritic stainless steel, low Cr composition, weld ability without post weld annealing

Exhaust pipe, fuel filter, blade or vane in generator turbine, electrical transformer etc. 430 Magnetic in all condition, good physical and

mechanical characteristics, cost less than chromium-nickel stainless steel

Cabinet hardware, decorative appliance and automotive molding and trim, restaurant equipment etc.

201 Austenitic chromium-nickel-manganese stainless steel, non-hardenable by heat treatment but cold work be done, non-magnetic in the annealed condition

Cookware, hose clamps, piston rings, air bag container etc.

2205 High strength, excellent corrosion resistance, corrosion cracking resistance, pitting

resistance, low thermal expansion

Heat exchange, piping, paper production equipment etc.

(6)

Figure 1 Trend of stainless steel price from 2012 – 2013[3]

1.2 AOD/CLU

The stainless steel refining process is one step in a stainless steel production. This study will focus on the Argon Oxygen Decarburization/ Creusot Loire Uddeholm-process (AOD/CLU) process, which are used in many stainless steel plants such as Outokumpus Avesta Works and Acerinox’s Columbus Stainless Works. This technic was first introduced with an AOD installation in Joslyn,USA 1968[4] . However before the AOD installation in 1950, Union Carbide AB showed that a low carbon concentration could be reached without an excessive chromium oxidation by adding argon together with the oxygen during decarburization. The AOD process has basic features are decarburization, reduction, sulphur refining (desulphurization).

 Decarburization, the carbon removal is done by oxygen blowing. During oxygen is injected in to a steel bath, chromium and iron will oxidize. Decarburization occurs when dissolved carbon reduces the chromium and iron. If only chromium is considered, the overall reaction can be expressed as seen in equation ). The equilibrium equation for this reaction is given by

equation (

) ), where , and are the

activities of carbon, chromium and chromium oxide respectively, is the partial pressure of carbon monoxide (CO) in the gas phase and K is the equilibrium constant.

(

) )

Based on equation 2 and the assumption that equals unity, the equilibrium carbon concentration [ ] can be calculated from

equation[ ] [ ]

), where and are the activity

(7)

decreased. Under conditions where all injected oxygen has reacted with carbon,(CRE=100) that mean all oxygen react with C give Pco the partial pressure of CO can according to the gas law as

seen in equation

) where is the volumetric

amount of generated CO, is the volumetric amount of injected nitrogen or argon and is the total pressure. Then the ratio of oxygen to inert gas can be optimized for any combination of carbon with chromium concentration.

[ ] [ ]

The example of decarburization strategies are explained in Table 2 and Figure 2. Then the decarburization process based on a stepwise decrease of the oxygen with inert gas ratio strategies.[4]

Table 2 Ratio of oxygen with inert gas during decarburization

Oxygen/inert Condition

3/1 c ≥ 0.6

1/1 0.6 < c ≥ 0.25

1/3 0.25 < c ≥ 0.07

1/3 c < 0.07

Figure 2 shown Carbon Removal Efficiency (CRE) step during decarburization [5]

Not only carbon will be removed during the decarburization procedure, but also other elements with high affinity to oxygen such as silicon, manganese, chromium and nickel. Stainless steel is highly alloyed with chromium and nickel. Chromium and nickel, as well as iron, will oxidize during the decarburization. A major loss of valuable chromium to the slag in the form of oxides is not acceptable.[6]

 A reduction process will be activated after the amount of carbon reach the goal by adding a reducing agent such as FeSi to reduce valuable oxide materials such as Cr2O3 from slag back to

(8)

 Sulphur refining is done by using inert gas blowing such as argon or nitrogen in the steel and by controlling the sulphide capacity of the slag.

Nevertheless, the AOD process still requires a good inert gas supply, but some plant encounter with gas supply problem. Thus the CLU process was developed to substitute inert gas by using super heat steam as (5).

(5)

Δ H=+241.9 kJ/mol

The formed hydrogen acts as an inert gas to replace argon or nitrogen, while the oxygen oxidized the carbon in the steel. The use of 1 kg of steam substitutes for 1.25 Nm3 Ar (or N2) and 0.625 Nm

3

O2 in the

converter.[7] A boiler is used to produce steam. This steam must be dried using a super-heater before it becomes suitable as a steelmaking process gas. To enable blowing of a dry gas in to the converter, it is necessary to preheat all process gases. This is done in a heat exchanger. A logical process flow diagram is displayed in Figure 3

Figure 3 Typical configuration of a steam generator in relation to a refining converter [8]

1.3 Process model used for simulations

(9)

Figure 4 The UTCAS concept for process control and management [9]

The UTCAS system provides data to enable a level 1 control and it also receives feedback from a process shown on an operator screen, as seen in Figure 5.The main calculation steps to determine the required material amounts and distribution are as follow [9]

 An advance simplex kernel algorithm is used to calculate and compare the input data and optimize by relate to metallurgical process. This program is used to determine the total amounts of alloys, mass build up materials, reduction agents and slag formers in order to reach the defined final targets for steel chemistry, steel mass and slag composition

 The materials are initially distributed according to the different properties and conditions defined in the practice

 A repeated prediction of the process is used to adjust gas mixes, re-arrange additions and add extra cooling materials with the objective to balance the process within the boundary conditions of the practice

(10)

1.4 Aim of study

Raw materials may not only be a source of valuables but can also contain inclusions tramp elements and undesired reduction agents. This study attempt to

 Understand the benefits and loss in an AOD/CLU process.

 Optimize the stainless steel processing in an AOD/CLU process, which compares the time of the process and the raw material costs. Furthermore to present some aspects of process by the UTCAS program predictions.

 Study the least cost charge materials. The cost of materials amounts to as much as 85 percent of the total stainless melting and refining operation cost. [1]

2 Experimental set up

First of all, stainless steel process templates were set up in UTCAS program. These cover each stainless steel grades. This study focused on 304, 409,316, 430, 201, and 2205 stainless steel grade. This is due to that they are the most used in the stainless steel market share, as seen in Figure 6. A second goal was to optimize the process according to the specific parameter as composition, final weight output, and temperature. By just changing the targets, and/or the start conditions, the model will automatically generate an adapted process plan with respect to material distribution and heat control. Anyway, some raw materials have an uneven composition that will change over all material distribution. In some cases an operator should adjust input target to compromise with the process condition. Finally, all raw materials consumption is calculated as a processing cost (detail in 2.6).

Figure 6 Market share of stainless steel grade categories [11]

2.1 Stainless steel and raw materials composition

(11)

Table 3 Stainless steel grades analysis composition

Grade Start analysis Final analysis

%C %Si %Cr %Mn %Ni %Mo %N %C %Si %Cr %Mn %Ni %Mo %N

304 1.8 0.1 18 0.5 7 0.2 0.04 0.045 0.5 18 1.2 8.05 0.2 0.04 316 1.7 0.4 17 0.5 10 2 0.04 0.045 0.5 17 1.2 11 2.5 0.07 409 1.3 0.2 10.5 0.2 0.2 0.05 0.02 0.008 0.5 11 <0,015 430 1.7 0.2 16 0.2 0.2 0.05 0.04 0.025 0.5 17 0.04 201 1.7 0.2 20 0.2 4 0.05 0.04 0.08 0.4 17 6.5 5 0.2 2205 2.1 0.2 22 0.7 5 2.7 0.04 0.015 0.5 22 1.5 5.5 3.1 0.16

The raw materials composition is shown in Table 4. The data represents an average from various suppliers. All of the raw materials compositions including slag formers were assigned in the UTCAS database before the start of an optimization.

Table 4 Raw materials composition in UTCAS data base

Materials %C %Cr %Ni %Si %Fe %Mn %Mo %Al %P %S %Cu

(12)

(LCSiMn) Nickel Briquette

(NiBS) 0.014 98 1.949

Slag composition

(13)

2.2 Final weight controls

The final weight of the stainless steel should be controlled so that it will not exceed the furnace size. From

Figure 7, the weight will increase during processing due to alloy composition adjustment. In the converter refining, the main objective is to adjust the composition and mass delivered from EAF. It is often necessary to aim for a tap mass of between 1-8 % lower than the maximum allowable mass to avoid the risk of ending up with more steel melt than it is possible to bring to the caster in one ladle after refining.

[12]

To optimize the mass in a sequence it is necessary to consider the accuracy and predictability of the production.

Figure 7 Estimated offsets according to some former and present plant metallurgists [10]

This study focused on how to reach mass optimization as described in Table 5.

Table 5 Steel melt mass limitation with furnace size

Plant size (tons) mass limitation (tons)

10 8

30 24

100 72

(14)

2.3 Temperature controls

Each stainless steel grade has a different temperature, due to composition changes during a process. Besides, the refractory wear is influenced by the steel temperature. Thus a temperature control is necessary to avoid a destruction of the refractory. This optimization is possible by means of

 Adjusting gas mixture (oxygen, steam, inert gas ratio) over time  Calculated amount of alloys and slag formers

 Determining amounts and distribution of additional cooling additions 2.4 Parameters control

The following stainless steel grades were studied: 304, 409, 316,430, 201 and 2205. In addition, different conditions were studied. The 304 stainless steel grade was studied more extensively. Parameters such as input materials, alloys and processing were prediction. As seen in the appendix trials condition were assigned in the UTCAS program. The cases are presented in appendix trials condition, which all refining variations of the studied stainless steel grades. The starting conditions are different but the purpose is to optimize the stainless steel process. After optimization of the 304 stainless steel grade interesting cases were identified as seen in appendix trials condition. Stainless steel grade 409 was simulated in a similar except for alloy additions. Then set the trial conditions case for 316,430, 201 and 2205 the parameter will change as following.

2.4.1 Trial conditions for stainless steel grade 304

 Cases No. 1-3 simulated a case of 8 tons of steel melt, where the amounts of C, Cr input are varied. However the temperature was constant. These processes do not use an oxygen lance to increase the temperature during at the first stage, due to the low amount of steel melt.

 Cases No. 4-6 simulated a case of 24 tons of steel melt. These cases used the same condition as cases No. 1-3, but a different furnace size.

 Cases No. 7-15 simulated a case of 72 tons of steel melt. The temperature, amount of C were varied but keep constant of an amount of Cr input and without oxygen lance was used. Also in cases No. 16-24 the same simulations were done when using an oxygen lance.

 Cases No. 25-33 was simulated for a 72 tons steel melt by varying the temperature and amount of C, but maintaining a constant of Cr input, This process use H2O injection in the processing as an

AOD/CLU.

 Cases No.34-36 were simulated for a 72 tons steel melt by maintaining a constant temperature, but varying amounts of C and Cr.

 Cases No. 37-42 were simulated for a 72 tons steel melt by maintaining a constant temperature and amount of Cr input, but varying the amount of C. In cases No.37-39 the gas mixing was increased to 1.5 Quantity blowing (Qb). In cases No. 40-42 the gas mixing was decreased to 0.66 Quantity blowing (Qb).The aim was study how gas blowing effect with the processing.

 Cases No. 43-45 were simulated for a 72 tons steel melt by maintaining a constant temperature and amount of Cr input but varying the amount of C input. During these processes compressed air were added during the de-carburizing process, to study the cost difference between a normal processing and a compressed air injection.

(15)

 Cases No.49-51 were simulated for a 72 tons steel melt by maintain a constant temperature and amount of Cr input but varying the amount of C. To investigate the effect of an alloying element on the process these case study with Charge Chrome (ChCr) input without High Carbon

Chromium (HCCr) input adding in the process. Then compare the processing cost with cases No. 52-54.

 Cases No. 52-54 were simulated for a 72 tons steel melt by maintain a constant temperature and amount of Cr input, but varying the amount of C. To investigate the effect of an alloying element on the process Charge Chrome (ChCr) and High Carbon Chromium (HCCr) were added to the process. Also, change source of nickel from Ferro Nickel (FeNi.) to Nickel Briquettes (NiBs).  Cases No.55-57 were simulated for a 72 tons steel melt by maintain a constant temperature, but

varying the amount of carbon and Cr. To investigate alloying effect by adding Ferronickel (FeNi) input instead of Nickel Briquettes (NiBs). Also add iron of a Low Carbon Ferro-Manganese (LCFeMn) input instead of a Low Carbon Silicon Manganese (LCSiMn).

 Cases No. 58-60 were simulated for a 72 tons steel melt by maintain a constant temperature but varying the amount of carbon and Cr. In these processes Ferronickel (FeNi) were added instead of Nickel Briquette (NIBS). Also, Electrolytic Manganese (ElMn) was added instead of a Low Carbon Ferro-Manganese (LCFeMn).

 Cases No.61-63 were simulated for a 216 tons steel melt by maintaining a constant temperature and amount of Cr. These processes used an oxygen lance to investigate the effect of an oxygen lance on the process.

2.4.2 Trial conditions of stainless steel grade 409

The trial conditions of stainless steel grade 409 were set up similar to 304 conditions but without investigation of the alloying effect condition as described in detail.

 Cases No. 1-3 were simulated for a case of a 8 tons steel melt amounts of C, Cr input were varied, but temperature was kept constant. These processes did not use an oxygen lance to increase the temperature up to the first stage, due to that the amount of steel melt is low.

 Cases No. 4-6 were simulated case for a 24 tons steel melt. These cases were similar to cases No. 1-3, but using a different furnace size.

 Cases No. 7-15 simulated a case of a 72 tons of steel melt. The temperature and amount of C were varied but Cr input was kept constant. Without oxygen lance use. Also in cases No. 16-24 the same simulations were done, but the oxygen lance was used in cases No. 16-24.

 Cases 25-33 were simulated for a 72 tons steel melt by varying the temperature, amount of C and amount of Cr. Also H2O injection was used in the process as AOD/CLU.

 Cases No. 34-36 were simulated case for a 72 tons steel melt by maintain a constant temperature but low amount of C with low amount of Cr were set and set high amount of C with high amount of Cr were set to process are C/Cr = 0.9/9, 1.3/10.5, 1.7/12 to investigate parallel effect of C and Cr at the same process.

 Cases No. 37-42 were simulated for a 72 tons steel melt by maintaining a constant temperature and amount of Cr, but varying amount of C. In cases No.37-39 gas mixing are increased to 1.5 Quantity blowing (Qb). In cases No. 40-42 gas mixing are decreased to 0.66 Quantity blowing (Qb). To study effect of the gas mixing during the process.

(16)

injection during the de-carburizing. To study cost difference between the normal process and the compressed air injection effect.

 Cases No. 46-48 simulated for a 72 tons steel melt by maintain a constant temperature and amount of Cr input but varying the amount of C. During these processes, the de-slagging time was 3 minutes and the desulphurization time 8 minutes.

 Cases No.49-51 were simulated for a 72 tons steel melt by maintain a constant temperature and amount of Cr input but varying the amount of C. To investigate the effect of an alloying element on the process these case study with Charge Chrome (ChCr) input without High Carbon

Chromium (HCCr) input adding in the process. Then compare the processing cost with cases No.19-21.

 Cases No.52-54 were simulated for a 216 tons steel melt by varying amount of C. These

processes were injected an oxygen through the oxygen lance and the initial temperature was set at 1550°c. Then compare with the cases No.55-57, which were not injected oxygen through the oxygen lance, but set the temperature input at 1650°c to investigate effect of the initial temperature effect.

2.4.3 Trial condition of stainless steel grade 316

The stainless steel grade 316 was set the process only interesting cases, all of them with a 72 tons size steel melt detail as seen in appendix trial condition are described in detail.

 Cases No.1-3 were simulated by varying amount of Cr and set the refining process without oxygen injection through oxygen lance.

 Cases No. 4-6 were simulated similar to case No. 1-3, but set the oxygen injection through an oxygen lance.

 Case No. 7 was optimized process by using superheat steam during the decarburization process (CLU).

 Case No. 8 was simulated by attempt to increase gas blowing quantity 50% more.

 Case No. 9 was simulated by the compressed air injection during the decarburization process.  Case No. 10 was simulated by changing the Cr source from High Carbon Chromium (HCCr) to

Charge Chrome (ChCr), which is not only more amount of chromium but more expensive also.  Cases No. 11-13 were simulated by varying Mo initial composition to investigate initial Mo

effect during the refining process.

 Cases No. 14-16 were simulated by varying Ni initial composition to investigate initial Ni changing during the refining process.

The stainless steel grade 430 was set the process only interesting case, all of them with 72 tons size steel melt detail as seen in appendix trial condition are described in detail.

 Cases No. 1-3 were simulated by varying Cr in the refining process without oxygen lance input.  Cases No. 4-6 were simulated similar to case No. 1-3, but these processes were set with the

oxygen lance injection.

 Case No. 7 was optimized process by using superheat steam during the decarburization process (CLU).

 Case No. 8 was simulated by attempt to increase quantity of gas blowing 50% more.

(17)

 Case No. 10 was simulated by change Cr source form High Carbon Chromium (HCCr) to Charge Chrome (ChCr), which is not only more amount of chromium but more expensive also.

The stainless steel grade 201 was set the process only interesting case, all of them with a 72 tons size steel melt detail as seen in appendix trial condition are described in detail.

 Cases No.1-3 were simulated by varying Ni initial in the refining process without the oxygen lance input.

 Cases No. 4-6 were simulated similar to case No. 1-3, but these processes with the oxygen lance injection.

 Cases No. 7-9 were simulated by changing Ni source form Ferro Nickel (FeNi) to Nickel

Briquettes (NiBs), which is not only more amount of nickel composition but more expensive also.  Cases No.10-12 were simulated by varying Mn input. That the process refining was added Low

Carbon Silicon Manganese (LCSiMn) with the oxygen lance injection.

 Cases No.13-15 were simulated by changing Mn source from Low Carbon Silicon Manganese (LCSiMn) to Electrolytic Manganese (ElMn).

 Cases No.16-18 were simulated by changing Mn source from Low Carbon Silicon Manganese (LCSiMn) to Low Carbon Ferro Manganese (LCFeMn).

 Case No. 19 was optimized the process by the superheat steam injection during the decarburization process (CLU).

 Case No. 20 was simulated by increasing the quantity of gas blowing 50% more.

 Case No.21 was simulated by adding the compressed air injection during the decarburization process.

The stainless steel grade 201 was set the process only interesting case, all of them with a 72 tons size steel melt detail as seen in appendix trial condition are described in detail.

 Cases No.1-3 were simulated by varying Cr in the refining process without the oxygen lance input.

 Cases No. 4-6 were simulated similar to case No. 1-3, but with the oxygen lance injection.  Cases No.7-9 were simulated by varying Ni initial. The refining processes also use the oxygen

lance injection.

 Cases No.10-12 were simulated by varying Mo. The refining processes also use the oxygen lance injection.

 Case No. 13 was optimized process by the superheat steam injection during the decarburization process (CLU).

 Case No. 14 was simulated by increasing the quantity of gas blowing 50% more.

 Case No.15 was simulated by adding the compressed air injection during the decarburization process.

(18)

(19)

2.5 UTCAS interface set up

The UTCAS interface is easy to use. It can show various parameters on one screen page, as seen in Figure 8. This contains data such as process profiles, processing times, input-output values, gas supply layouts, raw material priorities. Also, trend lines for nitrogen-carbon-Cr2O3-temperature.

Figure 8 The UTCAS interface layout

Before setting up a process template it is important to study the alloying during the process. Before all of the process alloying is done, the main alloying is done before and/or during the decarburization period as seen in Figure 9. This is described by the raw materials adding rule in the process. [13]

 A high carbon containing material should be added early, to promote decarburization. Here an increased amount of carbon will increase the oxygen consumption.

 Si will form SiO2 when added. If Si is added prior to the reduction it will lead to a large consumption of oxygen, inert gas and slag formers. The added Si will break up the chromium oxide and through a chemical reaction cause Cr recovery to the melt. Si is cheaper than aluminum and it would be the preferred element to use unless there are circumstances that will require the use of aluminum. The most negative effect with a Si addition is probably an increase of the total slag amount, which in turn seems to have a negative influence on decarburization and lining wear. In addition, the adding of Si will also affect with de-sulphurization.

(20)

study worked with High Carbon Chromium (HCCr) and Charged Chrome (ChCr). Both of them have high carbon contents and they should be added early in the process.

 Ni will promote decarburization. Thus Ni should be added early. As for chromium alloying, there are a lot of different allotting materials for nickel. There are various grade materials with different nickel contents such as Ferro Nickel (FeNi) with 30-40 % Ni or Nickel Briquettes (NiBs) with over 90% Ni. NiBs was used when there was no need to build up a mass. However, NiBs is much more expensive than the other nickel source. Thus, the best Ni source should be considered for each case.

 Mn is has a very strong oxygen affinity, so it has a negative impact on decarburization. In

addition, Mn vaporizes when added to the process. Thus, Mn should be added at a late stage of the process. Also, a Mn source with a low carbon content should be used to avoid affect the carbon composition. Such as a very useful form of Mn is the Low Carbon Silicon Manganese (LCSiMn) what is contains 30%Si , 60%Mn and small amounts of iron and the other elements. This material has a high ability to alloy the steel with manganese without causing negative effect.

 High Mo concentrations have a negative impact on decarburization. Thus, high amounts of Mo should be avoided during decarburization.

 Oxide materials such as MoOx and NiOx are reduced during the process.

 Slag forming, Calcium oxide (CaO) and Magnesium oxide (MgO) are needed to protect the lining of the converter. Lime is used as a CaO source and dolomite as a MgO source. Lime and dolomite are normally added quite early in the process. Calcium fluoride (CAF2) is used to lower the viscosity of the slag to enchance the reduction of Cr2O3. Fluorspars are used for adding CaF2 and are normally added during the actual reduction.

 Additional of oxygen from a top lance during the first stage of the decarburization is a very efficient way to boost the oxidation of carbon and to quickly raise the temperature. There is particularly important in case of large alloy additions where the temperature will drop. The additional of oxygen from top lance techniques can be more or less eliminated and the total time will consequently be shorter. A higher temperature of the process will increase the refractory consumption.

 Cooling materials. This technique must be careful controlled with oxygen flows to be able to maintain a reasonable temperature during decarburization. If the ambition is to reduce the process time by the means of more aggressively a blowing with oxygen, a cooling material should be used for cooling purposes. Beneficial cooling materials are materials that are nearly neutral in chemical composition compared to stainless steel grade produced should be used. Additions should be made so that it is necessary to compensate with extra chromium, nickel, etc.

(21)

(22)

2.6 Cost model set up

After all the process were optimized by the UTCAS system. The cost models were set up to find out the processing cost by calculating the optimized results in primary cost and secondary cost. Primary cost, which do not calculate the profit or loss from EAF processing. The processing cost was calculated based on the average cost of raw materials as seen in the following

equation (6) and Table 6.

(6) This study will calculate the refractory consumption during the process. From the paper “Ferro alloy design, ferro alloy selection and utilization optimization with particular focus on stainless steel materials” C-J Rick et. al. has calculated refractory consumption by assume the refractory cost is a function of inserted oxygen where the average oxygen consumption 2056 nm3 gives a refractory cost of 1500 euro for the heat as in equation

7) calculated

from refractory life of 100 heats for a 150 ton lining.[14]

7) Table 6 Raw materials price

Materials Supplier Price (1 $=6.62 sek)

ChCr Xtrata 2.53 $/kg = 16.75 sek/kg

HCCr Vargön 2.134 $/kg= 14.13 sek/kg

FeNi Anglo-American 4 $/kg = 26.48 sek/kg

NiBS Norilsk 17.3$/kg = 114.5 sek/kg

LCCr EWW (Ruukki-group) 4.642$/kg= 30.7 sek/kg

FeMo Climax 29.7$/kg = 196.6 sek/kg

SS304 Outokumpu 4301 2.56 $/kg= 17 sek/kg

SS316 Outokumpu 4401 2.8 $/kg= 18.5 sek/kg

SS430 Outokumpu 4742 0.8 $/kg = 5.3 sek/kg

MSSC Klass 117 0.5 $/kg= 3.31 sek/kg

FeSi Fesil 1.52 $/kg=10.1 sek/kg

LCFeMn Eramet 1.21 $/kg=8.01 sek/kg

LCSiMn Fesil 1.4 $/kg =9.268 sek/kg

ElMn 2.5$/kg =16.55 sek/kg

Al-wire Eramet 2.041 $/kg=13.5 sek/kg

Lime 1 sek/kg

Dolo Nordkalk 1 sek/kg

CaF2 Nordkalk 1.5 sek/kg

O2 lance 1 sek/Nm3

(23)

H2O 0.45 sek/Nm3

N2 0.7 sek/Nm3

Argon 6 sek/Nm3

Compressed Air 0.3 sek/Nm3

In addition, secondary cost models were calculated by concern the profit and loss from earlier processing for this study is EAF process. Alloying cost will have the following components:

 When an alloy (Cr, Ni, Mo and Fe including stainless steel scrap) is added to the converter, the costs in the EAF by decreases 0.7 sek/kg.

 When Fe, Ni and Mo is added in the converter it improves the Fe, Ni and Mo-yield by 1% compared to the EAF. Thus the cost for these alloys can be decreased by 1%.

 When Cr is added in the converter it improves the Cr-yield by 3% compared to the EAF so the cost for Cr-alloy can be decreased by 3%.

(24)

3. Result

This study will discuss the following for each stainless steel grade in;  Comparing of furnace size in term of processing time

 Input parameters effected such as C, Cr, temperature, etc.  Alloying effect and limitation.

 Processing effect with CLU, compressed air, extra gas blowing and the time fixing. 3.1 Stainless steel grade 304 result

The Table 7 as follow is summary result of stainless steel grade 304 by average.

Table 7 Summary result from UTCAS optimization of stainless steel grade 304

Case No. 1-3 4-6 7-15 16-24 25-33 37-39 40-42 43-45 46-48 49-51 52-54 55-57 58-60 61-63 processing time 50.3 59.8 65.6 51.86 63.57 51.8 95.2 49.8 45.6 52.26 51.9 52.4 52.06 50.9 primary cost 2925.29 2715.7 2797.6 2791.8 3183.6 3508.3 3637.6 3589.5 8462.3 2964.5 5589.67 2958 2987.04 2954.02 secondary cost 2769.93 2505.8 2613.9 2588 2989.39 3295.6 3340.4 3370.6 8134.7 2794.3 5327.3 2783 2752.6 2782.2

 If cases No.1-3, No. 4-6 and No.7-15 are compared, the smaller furnace size has the less processing time but not difference too much. Then compare the size of production the process of the 72 tons steel melt use 15 minutes more but yield the production mass 9 times of 8 tons steel melt. With the 216 tons furnace size provide the production steel melt 3 times of the 72 tons furnace size but the more investment should be considered.[15] Then the 72 tons of steel melt is the best size for refining in this study.

 If the cases No. 7-15 are compared with No.16-24, the process with oxygen lance injection has processing time less than the process without oxygen lance injection. Then the oxygen lance is useful to save the processing time.

 If the cases No. 7-15 are compared with No.25-33(CLU). The CLU process can save the cost of inert gas, as seen in the appendix for the 304 result; page x but with the overall process cost is higher than for case No. 7-15 due to that the CLU process consume more FeSi during reduction process. Then in the plant where encounter with inert gas supply (argon or nitrogen) should be considered the CLU process with higher reduction agent cost.

 Comparing cases No.34-36 as seen in appendix 304 result; page x-xi these case show the result when the carbon input and the chromium input are low will make the processing time is less but processing cost is more expensive, due to the process consume more alloy materials to reach the aim composition. However, when the carbon input and the

chromium input are higher, then the process will need a longer processing time during decarburization but the process cost is lower.

(25)

 Cases No.43-45 were simulated by injecting compressed air. This process has the same processing time as for cases No.37-39 (extra gas blowing), but the process cost is a little bit higher. Then, the compressed air input can save the processing time.

 For cases No.46-48 the time was fixed during desulphurization to 11 minutes. However the cost is too high. Due to that UTCAS attempt to use NiBs which is expensive.  Case No.49-51 uses difference carbon values and uses Charged Chrome (ChCr) as an

alloying element, as seen in appendix 304 result; page xiii show the amount carbon input has affected with the processing cost. The lower carbon input is improved the processing time but the processing cost is higher, due to the UTCAS attempt to compensate the carbon by take more ChCr as the carbon source during the refining process.

 Cases No. 52-54 were simulated by using ChCr, NiBs and HCCr as alloying elements. The processing cost is double as high than cases No.49-51. That means that a good selection of alloying cost will save the processing cost more or less.

 Comparing case No. 55-57 (use LCFeMn as Mn source) with case No.58-60 (use ElMn as Mn source). The Mn source did not change significantly for the 304 stainless steel cases. This is due to the change of an Mn input with Mn aim composition is low.  Comparing case No.61-63 were simulated of 216 tons steel melt with No.16-24 the

processing time is the same but processing cost a little higher than No.16-24 which is 72 tons steel melt. These cases should be considered in term of furnace investment to produce the 216 tons plant.

3.2 Stainless steel grade 409 result

The Table 8 as follow is summary result of stainless steel grade 409 by average.

Case No. 1-3 4-6 7-15 16-24 25-33 37-39 40-42 43-45 46-48 49-51 52-54 55-57 processing time 55.2 64.4 68.1 60.8 64.6 54.8 97 53.5 45.6 52.1 61.8 74.1 primary cost 678.16 873.2 721.8 698.15 744.42 1290.1 1429.2 1231.2 1245.9 1382 675.1 678.7 secondary cost 657.49 822 699.7 677.3 718.65 1196.2 1328.3 1136.1 1152.1 1287.9 651.46 655.8  If the cases No.1-3, No. 4-6 and No.7-15 are compared the smallest furnace size has the

lowest processing time, but the difference is not so large. Then compare the size of production the process using a 72 tons steel melt uses 15 minutes more but yields a mass 9 times of 8 tons steel melt. Then the 72 tons steel melt is seem to be the best size for refining in this study.

 If the cases No. 7-15 are compared with 16-24, the process with an oxygen lance injection has a processing time where is less than the process without an oxygen lance injection. Thus, an oxygen lance is useful to save the processing time.

 Cases No. 7-15 are compared with No.25-33(CLU). The CLU process can save the cost of inert gas as seen in appendix 409 result; page xix. However the overall process cost is higher than for cases No. 7-15. This is the CLU process consume more FeSi as a

reduction agent during the reduction process.

(26)

processing time smaller but the processing cost higher. This is due to that alloy materials are consumed. In opposite, when the carbon and chromium inputs are high, the process need a longer processing time. This is because a higher carbon amount need more time during decarburization, but the processing cost is lower.

 Comparing cases No.37-39 (extra gas blowing) with cases No.40-42 (lower gas blowing). The cases No.37-39 have shorter processing time than the cases No.40-42. This is due to the differences in the gas blowing quantity. Here, a higher gas blowing quantity can save processing time. However, the processing cost is almost the same. Therefore, the process can save time by increasing gas blowing quantity. However, this should be avoided not to exceed the temperature during refining process.

 Cases No.43-45 were injected by compressed air. This process has the same processing time and the processing cost is a little bit higher when compared with cases No.37-39 (extra gas blowing). Then the compressed air input can save the processing time. However, the air compression system investment should also be taken into account.  Comparing case No.46-48 used a fixed time during de-sulphurization. The processing

cost was different for the cases No. 46-48 of stainless steel grade 304, due to a change of source from NiBs to FeNi instead. Then the time fixing is useful to optimize the

processing time.

 Case No. 49-51where ChCr was added as chromium source are compared with cases No. 19-21where HCCr was added as chromium source. The processes which added ChCr can improve the processing time by around 10%, but the processing cost is two times

compared to a HCCr adding.

 Cases No.52-54 were simulated using a 216 tons steel melt with oxygen injection through an oxygen lance and set the temperature input at 1550°c with cases No.55-57 were simulated of 216 tons steel melt without oxygen injection through oxygen lance and set the temperature input at 1650°c. Obviously, the processing time of the cases No.52-54 are shorter. Thus oxygen injection through oxygen lance can improve the processing time.

The Table 9 as follow is summary result of stainless steel grade 316.

Case No. 1-3 4-6 7 8 9 10 11 12 13 14 15 16 processing time 73 74.6 69.8 61.4 75.6 82 81.6 81.6 80.6 80.6 81.2 82.8 primary cost 3879.2 4007.6 4144.5 4263.3 4182.6 3897.5 8077.9 4068.9 2760.4 4434.9 4206.2 3149.9 secondary cost 3786.6 3917.1 4037.2 4155.9 4070.9 3808.7 7943.6 3978.9 2672.9 4330.7 4110.9 3089.8  Cases No.1-3 were simulated without using an oxygen injection through oxygen lance and cases

(27)

result in the appendix; page xxiv. Therefore, the exceeded temperature is affected with a process optimization.

 Case No.7 was simulated by adding superheated steam as in an AOD/CLU process. The

processing time was almost the same as for cases No.1-3, but this technique reduces the inert gas consumption as seen in appendix 316 result; page xxv. However, the AOD/CLU process consume more reduction agent compared to cases No. 1-3.

 In case No.8 50% more gas was using. This technique can save the processing time, but the processing cost is higher than for cases No. 1-3. This is due to that this process has consumed. Thus, this process can improve the processing time but with the drawback an incresed processing cost.

 In case No.9 compressed air was injected during the decarburization process. This technique can save an amount of inert gas as seen in appendix 316 result; page xxv. But this technique is used over a longer time, due to that the decarburization process consumes more time to reach the aimed carbon composition.

 Case No.10 is simulated by setting the chromium source to ChCr instead of HCCr as for the cases No.4-6. This simulation shows that ChCr had to be added during decarburization after the

temperature up process has ended. Thus, the decarburization process uses a longer time to reach the aimed carbon content. However, this technique consumes less alloying elements and therefore this process can lower the processing cost.

 Cases No.11-13 which are simulated by varying an amount of Mo initial input. These simulations have shown as seen in Table 9 the alloying effect with the processing cost when the Mo initial input is low. The process will consume more Mo source (FeMo), and then made the processing cost is high.

 Cases No. 14-16 are simulated by varying the amount of Ni input. These simulations show the effect of alloying on the processing cost, Table 9 when the Ni input is low the process will consume more Ni source (FeNi) then made the processing cost is higher than a high amount Ni input.

Table 10 summaries the results of stainless steel grade 430.

Case No. 1 2 3 4 5 6 7 8 9 10 processing time 77.2 74.8 73.2 78.8 77 74.2 71.8 64.2 74.6 82.6 primary cost 1092.2 830.9 595.8 1263.7 990.1 728.03 1014.1 882.9 1004.5 1152.7 secondary cost 1050.1 791.7 556.5 1222.1 951.1 689.5 967.9 842.9 964.1 1112.7  Cases No.1-3 were simulated by varying the amounts of Cr input and the process was set without

oxygen injection through the oxygen lance. The processing cost and the processing time are proportional to the amount of added Cr.

(28)

 Case No.7 was simulated by adding superheated steam as in the AOD/CLU process. The time of processing was almost the same as for cases No.2, but this technique reduces the inert gas consumption as seen in appendix 430 result; page xxviii. However, the AOD/CLU process consumes more reduction agents when compared to cases No.2.

 Case No.8 used a 50% higher gas addition. This technique can save the processing time, but the processing cost is higher than for cases No. 1-3, due to that this process has consumed a higher amount of gas. Thus, this process can improve the processing time but the drawback is a higher processing cost compared to case No.2.

 Case No.9 simulated injection of compressed air during the decarburization process. This technique can save an amount of inert gas, as seen in appendix 430 result; page xxviii. But this technique requires a longer time.

 Case No.10 simulated an addition of ChCr as a chromium source instead of HCCr as in case No.5. This simulation shows the effect an additional with the processing cost. Due to the ChCr is more expensive than the HCCr.

TTable 11 summaries the results for stainless steel grade 201.

Case No. 1-3 4-6 7-9 10-12 13-15 16-18 19 20 21 processing time 59.9 59.5 58.2 52.2 53.3 56.9 57 46.6 57 primary cost 2687.4 2780.8 2896.9 2525.8 2778.1 2391.6 2387.9 2302.1 2379.9 secondary cost 2611.7 2706 2851.4 2481.5 2719.8 2332.7 2323.9 2243.6 2321.1

 Cases No.1, No.2 and No.3 were simulated by varying the amounts of Ni input and the process was used without an oxygen injection through the oxygen lance, as seen in appendix 201 result; page xxix. The process which used a low amount of Ni input provided the higher processing cost. Due to the lower amount of Ni input consume more alloy materials (FeNi).

 Cases No.1-3 were simulated by varying the Ni amount and the process was used without oxygen injection through the oxygen lance as was used in cases No.4-6. Both of these processes have the same processing times. Due to the steel melt temperature input was set too high during the temperature up process, the UTCAS program stopped the oxygen injection through the oxygen lance during a short time when the process temperature reached the temperature process limit.  Cases No.4-6 used FeNi as the Ni source compared cases No.7-9 used NiBs as the Ni source.

Obviously, the cases No.7-9 had higher processing costs, due to that NiBs price is higher than FeNi.

 Cases No.10, No.11 and No.12 were set by varying the Mn input and the process used oxygen injection through the oxygen lance, as seen in appendix 201 result; page xxx. The process that used a lower amount of Mn resulted in a higher of processing cost, due to that the lower amount of Mn consume more alloy materials (LCSiMn).

(29)

No.13-15 used ElMn as a Mn source and cases No.16-18 used LCFeMn as a Mn source. The processing cost for cases No.13-15 are the highest, due to that ElMn has the highest prices of Mn source. The processing cost of case No. 16-18 is the cheapest, due to that LCFeMn has the lowest price. Nevertheless, the higher price provided a higher alloy composition. This is good for avoiding the mass build up after refining, due to the furnace size limitation.

 Cases No.19 was simulated by adding superheated steam, as used in the AOD/CLU process. The processing time is almost the same as for the other cases, but this technique reduces the inert gas consumption as seen in appendix 201; page xxxi. However, the AOD/CLU process consume more reduction agents compared to the other cases.

 Cases No.20 was simulated with a 50% higher gas addition. This technique can lower the

processing time. However, this process consumes more gas, but the processing cost is cheap. Due to that the overall process is done in a shorter time.

 Cases No.21 simulated injection of compressed air during decarburization process and oxygen was injected through an oxygen lance. This technique can save an amount of inert gas, as seen in appendix 201; page xxxi. Therefore the processing cost can be lowered.

Table 12 is a summary of the results for stainless steel grade 2205.

Case No. 1-3 4-6 7 8 9 10 11 12 13 14 15 16 processing time 75 75.3 75.4 74 73.6 74.4 74 74 78.8 59.4 72.2 77 primary cost 1180.4 1324.2 1030.8 701.58 702.9 1641.7 701.6 701.21 686.5 599.3 675.2 674.6 secondary cost 1156.1 1297.2 1007.5 690.4 691.7 1618.5 690.4 690.1 672.4 589.3 664.2 665.6  Cases No.1-6 were simulated by varying the amount of Cr input. Either the process was set

without or with oxygen injection through the oxygen lance. The processing cost are proportional to the amount of added Cr, as seen in appendix 2205; page xxxiii.

 Cases No.1-3 were simulated without an oxygen injection through the oxygen lance and cases No.4-6 simulated using an oxygen injection through the oxygen lance. Both the processing time and cost for cases No.4-6 is higher than cases No.1-3, due to that the oxygen injection made the oxidation rapidly at the temperature up state then the UTCAS attempt to balance the amount of Cr by put more HCCr at decarburization state. Therefore, the processing time has taken the longer time. The processing cost is higher, due to more the oxygen cost during injection and alloy materials.

 Cases No.7-9 simulated by varying the amounts of added Ni. As seen in Table 12, the alloying affects the processing cost. When the Ni input is low the process will consume more Ni source (FeNi). This will make the processing cost higher than a higher amount of added Ni.

 Cases No.10-12 simulated by varying the amounts of added Mo. These simulations are shown in Table 12. The alloying affects the processing cost when the Mo input is low. The process will consume more Mo (FeMo), and then made the processing cost higher than for a high amount of added Mo.

(30)

appendix 2205; page xxxv. However, the AOD/CLU process consumes more reduction agent compared to the other cases.

 Cases No.14 simulated a 50% higher gas addition. This technique can lower the processing time. Even through this process consumes more gas the processing cost is cheap, due to that the overall process is done during a shorter time.

 Cases No.15 simulated injection of compressed air during decarburization process and oxygen injection through an oxygen lance. This technique can lower the amount of inert gas, as seen in appendix 2205; page xxxv. Therefore the processing cost is cheap.

 Cases No. 16 simulated a changing in Ni source from FeNi to NiBs. The case No.16 is cheaper than case No.8. This is due to that the NiBs addition is less than the FeNi addition to reach the aimed composition. Thus the processing cost is cheaper.

4. Discussion

Each stainless steel grade requires a different optimization method. The optimization considers parameters such as initial steel melt composition, final steel melt composition, stainless steel grade, limitation of steel plant and etc. The result and discussion for each technique will be described below. 4.1The oxygen injection through the oxygen lance

This technique can save the processing time in some simulation as for stainless steel grade 304 and 409. But some stainless steel grade had a too high initial temperature and the gap between the initial

temperature and the final temperature is low. Thus this technique is less effecting with respect to the overall process as for stainless steel grades 316,201 and 2205.

4.2 The initial steel melt composition

The effect of the initial steel melt composition both the processing time and the processing cost were studied. The most effect i.e. alloy found in this study is Ni. For example the Ni initial composition was set too low when compared to the final composition such as from 1.5 to 5%Ni. For a plant size of 72 tons steel melt this will lead to a huge consumption of alloy material (FeNi). Approximate 40% of initial steel melt and encounter with mass built up from amount of Fe in FeNi. In addition, the initial steel mass must be set with a low initial mass, due to plant size limitation. Then, the blowing gas should be lowed, which cause a too long processing time and a high cost. Besides, another alloying materials selection can solve this problem such as use NiBs instead of FeNi. However the processing cost will increase too much. 4.3 Superheat steam injection (AOD/CLU)

This technique is suitable for the producer who encounters problems with inert gas supply. Furthermore, this technique requires an extra system investment (superheat steam supply) including a reduction agent cost. However, this technique can decrease the processing time more or less for every stainless steel grade.

4.4 Compressed air injection

(31)

4.5 Alloying materials selection

This technique relate to the other techniques such as the blowing plan, the initial composition, the initial temperature, and the processing cost. According to the specific stainless steel grade some alloy materials should not be used during the process. For example, stainless steel grade 201 which focus on the final Mn composition as in cases No.10-12. Here, the LCSiMn was added and the final Si composition was

increased over the aimed composition. Instead the LCFeMn or ElMn should be used for these cases. 4.6 Extra blowing

This technique improves the processing time for every cases, due to that the extra blowing has positive effects. Due to that the activity during the process such as entropy of steel melt, velocity of circulation are increased. However, this technique requires a higher gas consumption during the process, so the

processing cost will higher than for a lower the gas blowing. However, for stainless steel grade 2205 this technique has a positive effect on the overall processing cost, due to that the processing time is shortened compared to standard process.

4.7 The time adjustment

This technique was simulated with stainless steel grade 304 and 409. Stainless steel grade 304 uses NiBs as the alloying material while grade 409 uses FeNi as the alloying material. This technique has a positive effect on the processing time for both stainless steel grades. However for stainless steel grade 304 it has a negative affect with the processing cost, due to that the Ni source is expensive. On the other hand, the technique has a positive effect on the stainless steel grade 409 processing cost. Thus, this technique relates to the alloying selection technique also.

5. Conclusion

The AOD optimization is done by many factors as in the result and discussion. The optimized process will vary upon each situation such as

 The input parameters and amount C, Cr as well as on temperature, steel melt mass, alloys and other.

 Alloying materials selection  Processing strategy selection

(32)

6. Future work

This study did not focus on the off gas and the heat loss during the process. Because those values are more complicate than the basic setting and require much more detail calculations. Anyway, those values can be calculated by the UTCAS program. In a future work, the next study will include these aspects.

7. Reference

[1] Richard J. Choulet: Stainless Steel Refining,1997, page 1-5 [2] http://www.matweb.com, 2013

[3] http://agmetalminer.com, 2013

[4] M. Engholm, C-J Rick: Optimized AOD gas administration- a lean process for a green world, Baosteel BAC 2010, page 1

[5] CRE pic

[6] www.keytometals.com, 2008

[7] K Beskow, J-A Van Der Linde and C-J Rick: Steam as process gas brings economic benefits to

Columbus Stainless, page 1

[8] C-J Rick: Strategies for Use of Superheated Steam During stainless Steel Refining in Converters,2010 , Vol. 1, page 1107

[9] C. Rick, M. Engholm: Control and optimization of material additions throughout the AOD refining

cycle, Steelsim 09,2009

[10] C. Rick, M. Engholm: Stainless steel refining with the AOD-process,2011,page 48 [11] http://www.steelorbis.com,2013

[12] C. Rick et al: Increased stainless steel melt shop yield by improved converter tap weight

management,2011,page 2

[13] UHT distributed papers: AOD-process exercises with UTCAS simulation,2011

[14] C-J Rick et. al.: Ferro alloy design, ferro alloy selection and utilization optimization with particular

focus on stainless steel materials,2012

(33)

Appendix trials condition

304 condition

Size 8 tons 24tons 72 tons 216 tons

Case No. 1-3 4-6 7-15 16-24 25-33 34-36 37-42 43-45 46-48 49-51 52-54 55-57 58-60 61-63 Temp. maintain maintain vary vary vary maintain maintain maintain maintain maintain maintain maintain maintain maintain

%Cr maintain maintain maintain maintain maintain vary maintain maintain maintain maintain vary vary vary maintain

%C vary vary vary vary vary vary vary vary vary vary vary vary vary vary

O2lance off off off on off on on on on on on off off on

H2O on on off off on off off off off off off off off off

Air off off off off off off off on off off off off off off

O2-Bottom on on on on on on vary on on on on on on on

N on on on on on on vary on on on on on on on

Ar on on on on on on vary on on on on on on on

CHCr off off off off off off off off on on on off off off

HCCr on on on on on on on on on off on on on on

NiBs off off off off off off off off off off on off off off

LCSiMn on on on on on on on on on on on off off on

ElMn off off off off off off off off off off off off on off

(34)

409 condition

Size 8 tons 24tons 72 tons 216 tons

Case No. 1-3 4-6 7-15 16-24 25-33 34-36 37-42 43-45 46-48 49-51 52-57 Temp. maintain maintain vary vary vary maintain maintain maintain maintain maintain vary

%Cr maintain maintain vary vary vary vary maintain maintain maintain maintain maintain

%C vary vary vary vary vary vary vary vary vary vary vary

O2lance off off off on off on on on on on on

H2O on on off off on off off off off off off

Air off off off off off off off on off off off

O2-Bottom on on on on on on vary on on on on

N on on on on on on vary on on on on

Ar on on on on on on vary on on on on

CHCr off off off off off off off off off on off

HCCr on on on on on on on on on off on

Ni off off on on on on on on off off off

LCSiMn on on on on on on on on on on on

ElMn off off off off off off off off off off off

(35)

316 condition

Case No. 1-3 4-6 7 8 9 10 11-13 14-16

Temp. maintain maintain maintain maintain maintain maintain maintain maintain

start %Cr vary vary maintain maintain maintain maintain maintain maintain

start %C maintain maintain maintain maintain maintain maintain maintain vary

start % Mo maintain maintain maintain maintain maintain maintain vary maintain

start %Ni maintain maintain maintain maintain maintain maintain maintain vary

O2lance off on off off on on on on

H2O off off on off off off off off

Air off off off off on off off off

O2-Bottom on on on 1.5Qb on on on on

N on on on 1.5Qb on on on on

Ar on on on 1.5Qb on on on on

CHCr off off off off off on off off

HCCr on on on on on off on on

FeNi on on on on on on on on

LCSiMn on on on on on on on on

ElMn off off off off off off off off

LCFeMn off off off off off off off off

(36)

430 condition

Case No. 1-3 4-6 7 8 9 10

Temp. maintain maintain maintain maintain maintain maintain

start %Cr vary vary maintain maintain maintain maintain

start %C maintain maintain maintain maintain maintain maintain

start % Mo maintain maintain maintain maintain maintain maintain

start %Ni maintain maintain maintain maintain maintain maintain

O2lance off on off off on on

H2O off off on off off off

Air off off off off on off

O2-Bottom on on on 1.5Qb on on

N on on on 1.5Qb on on

Ar on on on 1.5Qb on on

CHCr off off off off off on

HCCr on on on on on off

FeNi on on on on on on

LCSiMn on on on on on on

ElMn off off off off off off

LCFeMn off off off off off off

(37)

201 condition

Case No. 1-3 4-6 7-9 10-12 13-15 16-18 19 20 21

Temp. maintain maintain maintain maintain maintain maintain maintain maintain maintain start %Cr maintain maintain maintain maintain maintain maintain maintain maintain maintain start %C maintain maintain maintain maintain maintain maintain maintain maintain maintain start % Mo maintain maintain maintain maintain maintain maintain maintain maintain maintain

start %Ni vary vary vary maintain maintain maintain maintain maintain maintain

start %Mn maintain maintain maintain vary vary vary maintain maintain maintain

O2lance off on on on on on off off on

H2O off off off off off off on off off

Air off off off off off off off off on

O2-Bottom on on on on on on on 1.5Qb on

N on on on on on on on 1.5Qb on

Ar on on on on on on on 1.5Qb on

CHCr off off off off off off off off off

HCCr on on on on on on on on on

FeNi on on off on on on on on on

NiBs off off on off off off off off off

LCSiMn on on on on off off on on on

ElMn off off off off on off off off off

LCFeMn off off off off off on off off off

(38)

2205 condition

Case No. 1-3 4-6 7-9 10-12 13 14 15 16

Temp. maintain maintain maintain maintain maintain maintain maintain maintain

start %Cr vary vary maintain maintain maintain maintain maintain maintain

start %C maintain maintain maintain maintain maintain maintain maintain maintain

start % Mo maintain maintain maintain vary maintain maintain maintain maintain

start %Ni maintain maintain vary maintain maintain maintain maintain vary

start %Mn maintain maintain maintain maintain maintain maintain maintain maintain

O2lance off on on on off off on on

H2O off off off off on off off off

Air off off off off off off on off

O2-Bottom on on on on on 1.5Qb on on

N on on on on on 1.5Qb on on

Ar on on on on on 1.5Qb on on

CHCr off off off off off off off off

HCCr on on on on on on on on

FeNi on on on on on on on off

NiBs off off off off off off off on

LCSiMn on on on on on on on on

ElMn off off off off off off off off

LCFeMn off off off off off off off off

(39)

Appendix 304 result 304 case No. 1 2 3 4 5 6 TempUp Time 16.2 16.6 22.6 21.8 26.4 31 Decarb Time 11 11 11.4 12 12.4 13.4 Reduction time 5.2 5.2 5.2 5.2 5.2 5.2 Deslagging time 6.2 6.2 6.2 6.2 6.2 6.2 Desulph time 9.4 9.4 9.4 9.4 9.4 9.4 ChCr 0 0 0 0 0 0 HCCr 21.534 22.376 18.913 16.205 16.266 16.398 FeNi 45.823 47.239 44.131 42.537 42.697 43.045 Ni 0 0 0 0 0 0 LCCr 0 0 0 0 0 0 FeMo 0 0 0 0 0 0 SS304 50.08 49.725 50.435 48.614 48.797 49.195 SS316 0 0 0 0 0 0 SS430 0 0 0 0 0 0 MSSC 5.258 0 0 0 0 0 FeSi 15.274 22.376 21.435 16.205 16.266 12.299 LCSiMn 10.392 10.442 11.096 10.047 10.41 10.413 Alum 0 0 0 0 0 0 LCFeMn 0 0 0 0 0 0 ElMn 0 0 0 0 0 0 Lime 52.333 38.91 36.061 24.55 23.056 21.154 Dolo 27.293 20.636 19.292 13.126 12.402 11.438 CaF2 9.64 7.334 6.935 4.74 4.473 4.182 O2-Lance 0 0 0 0 0 0 O2-bottom 20.745 18.684 24.65 21.584 25.602 29.763 H2O 0 0 0 0 0 0 N2 10.504 8.764 9.646 8.426 9.141 11.118 Ar 4.783 4.749 4.817 7.543 7.832 7.551 Compressed air 0 0 0 0 0 0 Processing time 48 48.4 54.8 54.6 59.6 65.2

(40)

304 case No. 7 8 9 10 11 12 13 14 15 TempUp Time 29.2 32.8 36.4 27.2 31.2 36 21.4 26 28.6 Decarb Time 11.4 12.2 13 12 12.8 13.6 14 14.8 23 Reduction time 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 Deslagging time 6.2 6.2 6.2 6.2 6.2 6.2 6.2 6.2 6.2 Desulph time 9.4 9.4 9.4 10 10.6 9.4 11.4 11 11.6 ChCr 0 0 0 0 0 0 0 0 0 HCCr 19.249 16.325 13.354 18.6 15.727 12.794 17.936 15.104 12.026 FeNi 45.371 42.951 40.492 44.762 42.379 39.971 44.131 41.797 39.21 Ni 0 0 0 0 0 0 0 0 0 LCCr 0 0 0 0 0 0 0 0 0 FeMo 0 0 0 0 0 0 0 0 0 SS304 49.889 49.976 49.973 50.045 50.103 49.739 49.975 50.023 49.879 SS316 0 0 0 0 0 0 0 0 0 SS430 0 0 0 0 0 0 0 0 0 MSSC 0 0 0 0 0 0 0 0 0 FeSi 26.081 23.6 21.266 20.24 18.26 16.248 14.007 12.478 9.394 LCSiMn 11.835 11.717 11.605 11.733 11.607 11.523 11.647 11.547 11.444 Alum 0 0 0 0 0 0 0 0 0 LCFeMn 0 0 0 0 0 0 0 0 0 ElMn 0 0 0 0 0 0 0 0 0 Lime 68.209 61.567 55.359 53.033 47.682 42.278 36.857 32.71 24.565 Dolo 30.363 27.445 24.709 23.674 21.308 18.928 16.534 14.715 11.126 CaF2 10.241 9.273 8.357 8.007 7.223 6.425 5.636 5.03 3.824 O2-Lance 0 0 0 0 0 0 0 0 0 O2-bottom 28.143 31.507 34.72 24.6 28.024 31.623 20.681 24.406 17.618 H2O 0 0 0 0 0 0 0 0 0 N2 12.367 13.471 14.503 14.635 16.066 17.121 14.701 16.102 16.432 Ar 5.46 5.47 5.619 5.894 6.319 6.041 6.858 6.898 7.449 Compressed air 0 0 0 0 0 0 0 0 0 Processing time 61.4 65.8 70.2 60.6 66 70.4 58.2 63.2 74.6

primary cost (sek/ton) 3049.5 6 2933.1 7 2815.5 2920.8 3 2815.8 9 2699.0 1 2780.9 6 2684.3 5 2560.3 secondary cost (sek/ton) 2930.6

(41)

304 case No. 16 17 18 19 20 21 22 23 24 TempUp Time 14.2 15.8 17.4 11.4 13 14.8 8.6 10.6 11.6 Decarb Time 13.4 14.4 15.4 15.2 16 17.4 16.8 18.4 23 Reduction time 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 Deslagging time 6.2 6.2 6.2 6.2 6.2 6.2 6.2 6.2 6.2 Desulph time 9.4 9.8 10.6 10.6 11.4 12.2 10.4 10.8 11.6 ChCr 0 0 0 0 0 0 0 0 0 HCCr 19.543 16.419 13.267 18.55 15.498 12.374 17.694 14.824 12.031 FeNi 45.642 43.027 40.411 44.695 42.137 39.549 43.888 41.51 39.214 Ni 0 0 0 0 0 0 0 0 0 LCCr 0 0 0 0 0 0 0 0 0 FeMo 0 0 0 0 0 0 0 0 0 SS304 50.003 50.048 49.959 49.985 49.949 49.939 49.999 50.062 49.954 SS316 0 0 0 0 0 0 0 0 0 SS430 0 0 0 0 0 0 0 0 0 MSSC 0 0 0 0 0 0 0 0 0 FeSi 26.115 22.789 19.232 19.518 16.198 12.859 11.986 10.04 9.394 LCSiMn 11.89 11.747 11.602 11.719 11.571 11.458 11.625 11.542 11.448 Alum 0 0 0 0 0 0 0 0 0 LCFeMn 0 0 0 0 0 0 0 0 0 ElMn 0 0 0 0 0 0 0 0 0 Lime 78.782 66.981 55.885 54.4 43.872 33.514 31.638 26.422 24.561 Dolo 35.03 29.834 24.938 24.271 19.633 15.065 14.236 11.945 11.115 CaF2 11.806 10.065 8.438 8.22 6.66 5.146 4.861 4.102 3.83 O2-Lance 14.001 15.615 17.186 11.197 12.787 14.582 8.4 10.413 11.39 O2-bottom 15.77 17.242 18.771 13.926 15.426 17.101 12.125 14.07 17.662 H2O 0 0 0 0 0 0 0 0 0 N2 11.831 12.987 14.002 12.666 13.544 15.059 12.652 14.343 16.457 Ar 5.473 5.756 6.3 6.304 6.854 7.408 6.789 7.075 7.443 Compressed air 0 0 0 0 0 0 0 0 0 Processing time 48.4 51.4 54.8 48.6 51.8 55.8 47.2 51.2 57.6

(42)

(43)

304 case No. 35 36 37 38 39 40 41 42 TempUp Time 15 15.4 20.6 22.2 23.6 57.4 60.2 60.2 Decarb Time 14.8 15.8 8.6 8.6 9.4 12.8 14.8 13.8 Reduction time 5.2 5.2 5.2 5.2 5.2 5.2 5.2 5.2 Deslagging time 6.2 6.2 6.2 6.2 6.2 6.2 6.2 6.2 Desulph time 10.4 11.2 9.4 9.4 9.4 10.4 10 12 ChCr 0 0 0 0 0 0 0 0 HCCr 31.942 0 56.727 31.297 6.121 56.267 31.314 6.125 FeNi 53.996 50.381 54.641 54.665 54.675 54.6 54.696 54.707 Ni 0 0 0 0 0 0 0 0 LCCr 0 0 0 0 0 0 0 0 FeMo 0 0 0 0 0 0 0 0 SS304 49.996 49.691 50.053 50.074 50.084 50.015 50.103 50.113 SS316 0 0 0 0 0 0 0 0 SS430 0 0 0 0 0 0 0 0 MSSC 28.317 47.924 1.988 30.072 58.153 0 27.571 55.278 FeSi 19.749 16.84 21.022 19.793 18.503 30.565 28.392 26.992 LCSiMn 12.305 12.147 12.388 12.338 12.298 12.129 12.359 12.292 Alum 0 0 0 0 0 0 0 0 LCFeMn 0 0 0 0 0 0 0 0 ElMn 0 0 0 0 0 0 0 0 Lime 52.163 43.549 56.31 52.342 48.206 71.925 71.021 70.089 Dolo 23.276 19.49 25.11 23.354 21.536 31.996 31.593 31.182 CaF2 7.888 6.612 8.495 7.901 7.304 10.781 10.661 10.524 O2-Lance 14.799 15.106 0 0 0 0 0 0 O2-bottom 16.713 17.119 27.044 28.847 30.638 32.327 34.6 35.988 H2O 0 0 0 0 0 0 0 0 N2 13.085 13.889 16.803 17.422 18.946 13.393 14.21 16.173 Ar 6.166 6.681 8.05 8.054 8.055 4.896 4.899 5.652 Compressed air 0 0 0 0 0 0 0 0 Processing time 51.6 53.8 50 51.6 53.8 92 96.4 97.4

Optimization of the AOD stainless steel processing cost by the UTCAS System