Apostila Furanic Resin

GGoooodd PPrraaccttiiccee GGuuiiddee oonn

MMIINNIIMMIIZZIINNGG DDIIOOXXIINN AANNDD FFUURRAANN EEMMIISSSSIIOONNSS

IINN FFOOUUNNDDRRIIEESS

GGoooodd PPrraaccttiiccee GGuuiiddee oonn MMIINNIIMMIIZZIINNGG DDIIOOXXIINN AANNDD FFUURRAANN

EEMMIISSSSIIOONNSS IINN FFOOUUNNDDRRIIEESS Table of contents

Preamble ______________________________________________________ 1

1. Why this guide ______________________________________________________________ 1

2. Objective of this Good Practice Guide ____________________________________________ 1

Note to users ___________________________________________________ 1

Part 1: Dioxin and furans essentials _________________________________ 2

1. What are dioxins? ____________________________________________________________ 2

2. Potential sources_____________________________________________________________ 8

3. Dioxins in foundry melting areas _______________________________________________ 11

4. Chemical inhibition of dioxin formation __________________________________________ 36

5. End of pipe techniques to capture dioxins ________________________________________ 39

6. Bibliography _______________________________________________________________ 61

Part 2: Good practice manual _____________________________________ 63

1. Introduction _______________________________________________________________ 63

2. Block diagram: a previous guide to know if actions are required _______________ 63

3. Task guidance sheets: finding recommendations for your furnace and raw

materials ___________________________________________________________________ 65

4. Mind map: focusing the most important issues_______________________________ 12

5. Check list: concepts to check before the production __________________________ 15

Good Practice Guide on MINIMIZING DIOXINS AND FURANS EMISSION IN FOUNDRIES

1

Preamble 1. Why this guide

This guide is the result of a collection of existing knowledge and acquired information in a Project developed within SMEs, RTDs and IAGs which belong to the foundry sector and have knowledge of the problems of dioxins. The publication of this guide is a contribution of the industry (employers and employees) towards the protection of the population from dioxin exposure and its potential health effects. This guide has focused on the most important aspects of the dioxin formation and emission in different furnaces used in ferrous foundries. 2. Objective of this Good Practice Guide Dioxins and furans can be produced in ferrous foundries during combustion processes. They could have a clear impact on human health. A person is exposed to them in the food primarily. Dioxins are soluble in fatty tissues, so once they are consumed by a person, the body processes accumulate them very slowly creating a build-up of the chemical in those fatty tissues. That is why it is necessary to eliminate the sources: the industrial emissions like those from ferrous foundries. The objective of this guide is to give foundries guidance on the practical application of the knowledge to minimise and/or eliminate dioxin in/from their emissions.

Note to users This guide represents a summary of information collected from a number of sources, including existing documents providing information on the dioxin formation and emission, legal documents and expertise of people working in the industry. It also has to be mentioned that this good practice guide is one of the deliverables of the European Project Diofur. The partners agree that dioxin formation is a complex process that requires conditions such as:

Presence of chloride ions (scrap having plastics, paintings etc.) Available organic carbon Temperatures between 250 and 450ºC and enough residence time in this range Presence of a catalyst such as copper Presence of oxygen

In this short document it is not possible to cover all of the topics comprehensively, nor is it possible to cover in detail all areas of concern regarding dioxin emissions. Users, customers, workers, and readers are advised to consult further documents for more detailed information.


2

Part 1: Dioxin and furans essentials 1. What are dioxins? 1.1 Definition Dioxin is a general term that describes a group of hundreds of chemicals that are highly persistent in the environment. Dioxins (or PCDD) and furans (or PCDF) are compounds with the general chemical structure shown in the following figure. There is a large number of individual dioxin and furan compounds (210) called congeners. There can be as few as one or as many as eight chlorine atoms substituted on the dioxin or furan ring compound.

Figure 1: General molecular structure of dioxin and furan compounds. Source: EPA Sources of dioxins and furans include waste incinerators, cement production, fossil-fuel-fired combustors, and forest fires. The dioxin and furan compounds having from four to eight chlorine atoms are considered especially toxic. All of the dioxin and furan compounds are considered potentially toxic. For sampling and control dioxin and furan compound emissions are given in two different ways:

As the total dioxin and furan compound concentrations As the Toxic Equivalency Quotient (TEQ) concentration

The TEQ value for dioxin and furan emission is calculated according to a toxicity weighting scale. The compound 2,3,7,8 tetrachlorinated dibenzo-p-dioxin is considered the most toxic compound and is assigned a weighting factor of 1.0. Sixteen other dioxin-furan compounds are assigned weighting factors ranging from 0.5 down to 0.001. The observed concentrations of these seventeen dioxin-furan congeners are multiplied by these weighting factors to determine the total concentration of dioxin-furan compounds that have a toxic equivalent to 2,3,7,8 tetrachlorinated dibenzo-p-dioxin. This concentration is usually expressed as dioxin-furan compounds in TEQ nanograms per cubic meter (ng TEQ/Nm³).


3

Figure 2: Structure of the 2,3,7,8-tetrachlorodibenzo-p-dioxin. Source: Tohoku University The TEQ value is often used in regulatory limits because it is most closely related to the adverse health effects believed to be associated with dioxin and furan compounds. 1.2 Formation mechanism Dioxin is formed by burning chlorine-based chemical compounds with hydrocarbons. The major source of dioxin in the environment comes from waste-burning incinerators of various sorts and also from backyard burn-barrels. However, the formation mechanisms for dioxin-furan compounds have not been fully identified. It is believed that there are at least three different types of formation mechanisms that are possible. All of these depend on the availability of chlorinated precursor compounds in the fuel and/or waste being burned and the appropriate gas temperature conditions. One of the proposed formation mechanisms for dioxin-furan compounds involves reactions on the surfaces of particles entrained in the gas stream. Dioxin-furan concentrations appear to increase in the temperature range from 250 to 450ºC. However, at temperatures well above 450ºC, dioxin-furan compounds are readily oxidised. In other words, their formation occurs between 250-450 ºC and they are destroyed at temperatures above 450 ºC. Some dioxins and furans are formed and also destroyed (i.e. oxidised) in the burner flames of combustion chambers. Most of the chlorinated precursor compounds volatilize and move with the gas stream through the combustion process until they reach the temperature range favourable for dioxin and furan formation (250 to 450ºC). A small percentage of dioxins and furans can form in boilers where the economizers and heat exchange equipment are located. Since most dioxins and furans tend to form in control devices where temperatures well above 450ºC are reached, the dioxins and furans are destroyed. Therefore, gas streams leaving combustion processes should be cooled to temperatures below 450ºC to assure a reformation of dioxins and furans. This reformation process of the compounds due to the cooling of the gases to temperatures between 250-450 ºC is called de novo synthesis. In fact, oxidation of these compounds is completed at lower temperatures than for some other forms of partially oxidised compounds as indicated in the diagram of figure 3. These temperatures usually exist in the combustion zones of incinerators and fossil-fuel-fired boilers.


4

Figure 3: Destruction temperature for dioxins and furans. Source: EPA modules The formation mechanisms decrease to negligible rates when the gas stream temperature decreases below 204.45 ºC (400°F). Accordingly, ensuring that the gas stream is sufficiently cooled prior to going through the air pollution control system can eliminate this formation mechanism. Cooling is accomplished in the heat recovery equipment or in the incinerator waste heat boilers. The best way to control dioxin and furan emissions is preventing their formation by reducing or eliminating the chlorine in the fuel and waste material being burned. 1.2.1 Iron casting In the particular case of melting in ferrous foundries, dioxins may be produced if the conditions that give rise to such pollutants are present at the same location and time in the process. These conditions are:

The presence of chloride ions – these can arise from contaminated scrap, from the use of coal, coke, fuel oil or from certain fluxes

The presence of organic carbon – this may arise from contaminated scrap and from coal, coke or oil used as a fuel

Temperature conditions between 250 ºC and 450 ºC A sufficient gas residence time in this temperature interval The presence of a catalyst such as copper The presence of oxygen.

In evaluating the risk of dioxin formation, a distinction can be made between non-ferrous and ferrous foundries: Non-ferrous foundries: In as far as when only ingots and internal scrap are melted, the risk

of dioxin formation in the melting stage is very low. The melting of pure non-ferrous metals avoids the presence of both the chlorine and carbon required for dioxin (re)formation. However, the re-melting of external non-ferrous scrap materials (with paintings or plastic) for metal production may involve a risk of dioxin formation.

Ferrous foundries: Depending on the furnace type and metal load, the conditions for dioxin

formation could occur. Considering the high temperatures in the melting furnace, dioxin emission (if occurring at all) will mainly generate from de-novo synthesis. The above mentioned conditions can be used to evaluate the risk of dioxin formation.


5

1.3 Effects Much of the knowledge about the health effects of TCDD, other dioxins, and dioxin-like compounds (henceforth DLCs) in humans comes from studies of relatively highly exposed workplace populations. Widespread use of certain herbicides containing dioxins and DLCs, as well as some types of industrial emissions, resulted in local and global contamination of air, soil, and water with trace levels of these compounds. These trace levels built up in the food chain because these components do not readily degrade. Instead, they persist in the environment and accumulate in the tissues of animals. The general public is exposed to them primarily by eating such foods as beef, dairy products, pork, fish, and shellfish.

Figure 4: Dioxin cycle. Source: www.vadscorner.com Studies suggest veterans and workers exposed occupationally to dioxins and DLCs experience an increased risk of developing a potentially disfiguring skin lesion (called chloracne), liver disease, and possibly cancer. Animal and human studies also demonstrate that dioxins and DLCs might contribute to thyroid dysfunction, lipid disorders, neurotoxicity, cardiovascular disease, and metabolic disorders. The potential adverse effects of these components from long-term, low-level exposures to the general public are not directly observable and remain controversial. One major controversy is the issue of estimating risks at doses below the range of existing reliable data. 1.4 Measuring methods The complete measurement procedure of PCDD/F (dioxin and furans) consists of: sampling, extraction and clean-up, identification and quantification, specified in the parts 1, 2 and 3, respectively, of the European Standard EN 1948:2006.


6

Referring to the sampling step, gas is sampled isokinetically in the duct. The dioxins and furans both adsorbed on particles and in the gas phase, are collected in the sampling train. The collecting parts can be a filter, a condensate flask and a solid or liquid adsorbent appropriate to the sampling system chosen. The Standard establishes that the user has the possibility to choose between three different methods: filter/condenser, dilution and cooled probe methods. The Standard EN 1948:2006 includes representations of the sampling trains for each method. Extraction is necessary to isolate the dioxins and furans from the sample and to collect them in an appropriate solvent volume. Extraction procedures are normally based on soxhlet extraction of filters and adsorbents, and liquid extraction of condensates. Sample clean-up is usually carried out by multi-column chromatographic techniques using a range of adsorbents. The main purpose of cleaning the raw sample extracts is to remove sample matrix components, which may overload the separation method, disturb quantification or otherwise severely impact the performance of the identification and quantification method. In principle any clean-up method can be used which recovers the analytes in sufficient quantities. Furthermore, the final sample extract should not affect adversely the performance of the analytical system or the quantification step. For the identification and quantification the Standard is based on the use of high resolution gas chromatography/high resolution mass spectrometry for separation

Figure 5: Diagram of the complete measuring method for dioxins and furans. Source: Inasmet-TECNALIA

11.. SSAAMMPPLLIINNGG

22.. EEXXTTRRAACCTTIIOONN AANNDD CCLLEEAANN--UUPP

33.. IIDDEENNTTIIFFIICCAATTIIOONN AANNDD QQUUAANNTTIIFFIICCAATTIIOONN


7

and detection, combined with isotope dilution of samples at different stages for quantification of dioxins and furans in emission samples. At present, this technique is the only analytical technique that can provide sufficient sensitivity, selectivity and specificity for the determination of minute amounts of dioxins and furans in emission samples. The gas chromatographic parameters offer information which allow the identification of isomers (position of Cl substituents) whereas the mass spectrometric parameters allow the differentiation between congeners with different numbers of chlorine substituents and between dioxins and furans. The main difference between an isomer and a congener is that isomers have the same molecular formula but a different molecular structure (for this case, the same number of Cl atoms but substituted in different positions); on the other hand congeners are each of the configurations for a same chemical structure (in this case it is referred to the compounds with a different number of Cl atoms, also taking into account the different positions for each number of Cl atoms substituted).


8

2. Potential sources 2.1 Industrial activities in general The major identified sources of environmental release have been grouped into the following classes:

Combustion Sources: PCDD/PCDFs are formed in most combustion systems. These can include waste incineration (such as municipal solid waste, sewage sludge, medical waste, and hazardous wastes), burning of various fuels (such as coal, wood, and petroleum products), other high temperature sources (such as cement kilns), and poorly controlled combustion sources (such as building fires).

Metals melting, refining and processing Sources: PCDD/PCDFs can be formed

during different types of metals operations, both primary and secondary operations. In this report have been studied the emission ranges specifically in the ferrous foundry sector, secondary process, which is not considered to be a very potential source of these compounds.

Chemical Manufacturing: PCDD/PCDFs can be formed as by-products from the

manufacture of chlorine bleached wood pulp, chlorinated phenols (e. g., pentachlorophenol - PCP), PCBs, phenoxy herbicides (e. g., 2,4,5-T), and chlorinated aliphatic compounds (e. g., ethylene dichloride).

Biological and Photochemical Processes: Recent studies have suggested that

PCDD/PCDFs can be formed under certain environmental conditions (e. g., composting) from the action of microorganisms on chlorinated phenolic compounds. Similarly, PCDD/PCDFs have been reported to be formed during photolysis of highly chlorinated phenols.

Reservoir Sources: Reservoirs are materials or places which contain previously formed

PCDD/PCDFs or dioxin-like PCBs and have the potential for redistribution and circulation of these compounds into the environment. Potential reservoirs include soils, sediments, vegetation, and PCP-treated wood. Recently, PCDD/PCDFs have been discovered in ball clay deposits. Although the origin of the PCDD/PCDFs in these clays has not been confirmed, natural occurrence is a possibility.

The possibility remains that truly unknown sources exist. Many of the sources which are well accepted today were only discovered in the past 20 years. For example, PCDD/PCDFs were found unexpectedly in the wastewater effluent from bleached pulp and paper mills in the mid 1980s. Ore sintering is now listed as one of the leading sources of PCDD/PCDF emissions in Germany, but was first reported in the early 1990s. Another potentially important source is reservoirs. In this context, reservoirs are places such as soils, sediments, vegetation or other media which contain dioxin-like compounds originally formed some time in the past and the potential for emissions in the present. The dioxin-like compounds in these "reservoirs" can be re-released to the environment by processes such as volatilization and particle resuspension. Such releases may (or may not) add significantly to the mass of dioxin-like compounds circulating in the environment and potentially contributing to human exposure. Two of the largest potential reservoirs are soils and pentachlorophenol (PCP) treated wood. PCP contains low levels of PCDD/PCDFs and wood which has been treated with this pesticide represents a large reservoir of PCDD/PCDFs. PCDD/PCDFs may be released from


9

the PCP-treated wood to the air by volatilization or to surrounding soils by leaching. Although hypothesized to occur, no reliable measurements have been made. Similarly, no empirical evidence exists on the possible magnitude of reservoir emissions from soil to air.

Table 1: Order of magnitude estimates of PCDD/PCDF air emissions from sources not quantified in the National Inventory (Reference Year 1995). Source: US EPA

2.2 Foundries in particular (BREF document) Ferrous foundries produce high strength iron and steel castings used in industrial machinery, pipes, and heavy transportation equipment. Iron and steel castings are solid solutions of iron, carbon, and various alloying materials. Castings are produced by injecting or pouring molten metal into cavities of a mould made of sand, metal, or ceramic material. Metallic raw materials are pig iron, iron and steel scrap, foundry returns, and metal turnings. Iron and steel foundries, particularly those using EAFs, are highly dependent on iron and steel scrap. Thus, foundries face the same potential for PCDD/PCDF emissions as EAFs because of use of scrap containing chlorinated solvents, plastics, and cutting oils. The potential for formation and release of PCDD/PCDFs during the casting process (i.e., pouring of molten metal into moulds and cores comprised of sand and various organic binders and polymers) is not known. The first results of emissions testing were reported for only one U.S. ferrous foundry by the US EPA (CARB, 1993a - as reported in U.S. EPA, 1997b). The tested facility consisted of a batch-operated, coke-fired cupola furnace charged with pig iron, scrap iron, scrap steel, coke, and limestone. Emission control devices operating during the testing were an oil-fired afterburner and a baghouse. The calculated TEQ emission factor for this set of tests is 0.37 ng/kg of metal charged to the furnace. The Umweltbundesamt (1996) reported stack testing results for a variety of ferrous foundries in Germany. Sufficient data were provided in the Umweltbundesamt (1996) to allow calculation of


10

TEQ emission factors for eight of the tested facilities. Three facilities had emission factors exceeding 10 ng/kg of metal charge, and four facilities had emission factors less than 0.1 ng TEQ/kg of metal charge; the emission factors span more than four orders of magnitude. The mean emission factor was 1.26-ng TEQ/kg of metal feed. Based on the wide range of emissions for the tested German foundries reported in the Umweltbundesamt (1996), the confidence in the degree to which the one tested U.S. facility represents the mean emission factor for the approximate 1000 U.S. foundries is considered very low. Therefore, the limited data available were thus judged inadequate for developing national emission estimates that could be included in the national inventory. However, a preliminary order of magnitude estimate of potential TEQ annual emissions from U.S. ferrous foundries can be made by combining the mean emission factor derived from the data reported in the Umweltbundesamt (1996) for eight foundries (1.26-ng TEQ/kg of metal feed) with an activity level for U.S. foundries. In 1995, U.S. shipments from the approximate 1000 U.S. ferrous foundries were 13.9-million metric tons of which about 90 percent were iron castings and 10 percent were steel castings (Fenton, 1996). This calculation yields an annual emission estimate of 17.5 g of TEQ in 1995, which, when rounded to the nearest order of magnitude to emphasize the uncertainty in this estimate, results in a value of 10-g TEQ/yr. This estimate should be regarded as a preliminary indication of possible emissions from this source category; further testing is needed to confirm the true magnitude of these emissions. 2.3 Others (induction furnaces why not measured) The electric induction furnace, as well as using clean charge, does not produce a ducted high temperature flue-gas stream that cools down slowly. Therefore, there are no potential sources of chlorine and de-novo synthesis is not likely to occur. The BREF document on Smitheries and Foundries mentions that induction furnaces show a low risk of dioxin formation and therefore emission rates from this type of furnace have not been measured in the Diofur project (main data source of this Good Practice Book). Although this type of furnace was not included in the Project (because of the BREF statement), we obtained the measurement data for a modern ‘dual-track’ furnace giving lower emissions differing in two orders of magnitude. Its charge consisted of pig iron, packed steel and returns. Consulted data in other similar foundries show similar results but the decision whether fewer measurements are required for foundries that have low emission rates depends on the authorities. . In other cases, the authorities could treat these furnaces like any other furnace or change their requirements based on the results of the emission rate measurements (in case they have been obtained repetitively). The following table includes data from the BREF document:

Table 2: Emission data for induction furnaces. Source: Inasmet-TECNALIA

Foundry 1 Emission limit value Source Induction furnace -

Depuration system YES -

Date 28/06/07 Day1 0.003 Emitted dioxins

ng I-TEQ/Nm³ Day2 0.010 0.1

ng I-TEQ/Nm³


11

3. Dioxins in foundry melting areas Although it is known in general terms that ferrous foundries are not an important dioxin source, it was thought that a situation study in depth for the most common furnaces was necessary, because of the fact that they have processes where these compounds are likely to form and the lack of reliable and detailed databases. Basically, it was the mentioned ‘leit motiv’ for the DIOFUR Project and for its accepting by the European Commission. It would be a great success for the DIOFUR project if from now on the required values for the foundries were logical, realistic and based in the obtained results. This is one of the main objectives of this document and of the project as a whole. In this Good Practice Book there would be given solutions for those cases where the emissions are above the target value although the foundries are working in the most efficient conditions. This is another main purpose of the project and therefore comes from it. As mentioned, foundries use different melting capacities what makes necessary to classify and study them separately. The dioxin formation will be different in each furnace depending on the different types of charge, fuel etc. they use, despite the final product is the same or at least similar. According to the existing background in dioxin formation in industrial processes, all the works show that there is a high variability in the dioxin formation rate. This is mainly due to the complex formation mechanism of this type of compounds and to the industrial process itself. During the scale running occur many problems and differences which can lead to very different emission rates. However, based on the existing background (BREF), a priori, some furnaces where thought to be more relevant than others. For example cupolas have the higher risk due to the intrinsic or essential properties of the process. They have the capacity of giving a good final product from charge having a high percentage of scrap. Scraps charged in foundries have paintings, oil, plastics etc. which is the chlorine source for the dioxin formation. However the most relevant property is that these types of furnaces have a lot of carbon available because they need coke in the process. On the other hand rotary furnaces as well as electric arc furnaces do not use coke in their processes and the first ones charge less scrap. Therefore they were not thought to be the most relevant furnaces in this matter. If we compare both, the rotary furnace is the one that uses less scrap as charge because of the essential properties of the process but EAFs are not a very high source of dioxins because they do not have such high carbon sources as cupolas. These results make us conclude that rotary furnaces and electric arc furnaces should be considered in regulations as the electric induction furnace is: no potential source of dioxins. All these facts were established before checking the situation and measuring the emission rates of each furnace and therefore were a mere hypothesis. In this section are explained the properties of each type of furnace.


12

3.1 Cupola melting 3.1.1 Furnace operation (cold/hot blast) Description The cupola is a refractory lined shaft furnace where the metal charge is heated by the combustion of coke, which takes place in the lower part of the shaft. Combustion air, supplied by fans, is injected in the hearth through a number of exhaust nozzles (“tuyeres”). A ring and control valve allows a controlled and evenly distributed flow of combustion air through the tuyeres. The metal (pig iron, scrap steel, scrap iron, foundry returns), coke, alloying elements (e.g. FeSi, SiC), slag forming (SiO2) and fluxing agents (e.g. CaCO3) are added to the shaft through a charging door at the upper part of the shaft. The combustion gases move upward from the hearth and exchange heat with the charge, before leaving the furnace through the cupola stack. In its basic configuration, the cupola is called a cold blast cupola (CBC). This is a cupola which uses the blast at atmospheric pressure and at normal environmental temperature. In order to optimize the efficiency of the cupola furnace, the combustion air may be preheated. This principle is used in the hot blast cupola (HBC) which uses a preheated blast. Two main groups of iron foundries often use cupola at least as primary melting: - small foundries, which use cupolas as a means of obtaining economic operation for a modest capital investment cost. Usually, less than 1000 mm diameter cupolas operate in these foundries, prevalent cold blast system is used and not significant changes in cupola operation system from improvements in peripheral equipment are recorded (especially for small cupolas). A divided blast system is a typical modernization of these cupolas. - large-tonnage foundries use cupola as primary melting with highest known melting rate (20... 110 t/h, metallurgical type cupolas), usually as water-cooled furnaces, hot blast operation or/and oxygen enrichment and more and more improved in cupola operations, especially concerning computerized processes, automated activities, pollution control, recuperative heat and waste systems. For medium-sized foundries producing up to 2000 t/month of good castings, the hot blast cupola is difficult to consider, in particular because of the large investment it requires. In these instances, the cold blast cupola prevails for some types of production. The hot blast cupola remains the most widely applied melting device for mass production foundries, e.g. for parts for the automobile industry, centrifugal casting, road accessories. Hot blast cupolas are normally set up for long campaign operation, in order to minimize process switch overs and maintenance time and effort. Foundry coke The most common fuel used in cupola furnaces is foundry coke. The coking operation lasts for 24 to 48 hours, and consists in heating coal in the absence of air at temperatures of the order of 1.100°C. The characteristics of the resulting coke are dependent on the raw material used, the coal, and on the heating conditions. Few coals have coking properties. Careful selection of the raw materials allows the coking plant to prepare the mixture which will provide a finished


13

product exhibiting the required properties. Foundry coke is prepared from a mixture of semi-bituminous coal, with the addition of about 20% coke fines or anthracite, to ensure the cohesion of the coke. For cupola furnaces, the coke plays three main roles:

a thermal: it must be able to heat, melt and superheat the metal; a chemical: it must provide some of the carbon for the cast iron; a mechanical: it must support and aerate the charge in the cupola furnace.

The following size classes are generally recognized: < 60 mm; 60-90 mm;> 90 mm; 90-150 mm; >150 mm; 90-250 mm. The size of the coke exerts a major influence on its combustion. For a given quantity of coke, the smaller the size of the lumps of coke the greater the contact surface with the gas. Consequently, the greater its reactivity to the Boudouard reaction: CO2 (g) + C = 2 CO (g) is promoted, and also the combustion index “n” diminishes: n = % CO2 / (% CO + % CO2) The combustion of coke thus generates less heat and the temperature of the melt is lower. Increasing the coke rate reduces the flow of metal running over it and raises the temperature level. This increases the carbon pick up. Fusion begins higher up the cupola furnace and the run-off path lengthens. The CO content increases and the atmosphere is less oxidizing, which reduces the melting losses. In cold blast operation, the coke consumption between the charges is generally 90 - 120 kg/tmetal charge, but can be less than 70 kg/tmetal charge in some types of pieces (e.g. counterweights). Calculating the quantity of coke in the bed gives a total coke consumption of 110— 140 kg/tmetal charge. As the calorific value of European cokes is 8.5 kWh/kg, this corresponds to a calorific input of 950 – 1200 kWh/tmetal charge. The total coke ratio in a hot blast cupola is generally 110 - 145 kg/tmetal charge. However, as the average steel percentage is 50 %, and the recarburization consumes about 1.5 %, the real burned coke ratio is 95 - 130 kg/tmetal charge, which is 810 to 1100 kWh/tmetal charge. This corresponds to a thermal efficiency of 35 to 45 %. Melting When the preheated charge reaches the combustion zone, the metallic parts melt due to the high temperatures, and the charged coke starts to burn in the presence of oxygen. The molten metal droplets run through the coke bed and gather in the zone called the well, which is below the combustion zone. All the impurities are trapped in the slag, which is mostly formed by SiO2, CaO, Al2O3 and FeO. Due to its lower density, the slag floats on the molten metal in the well. Once the liquid metal in the well has reached a certain level, a tap-hole is opened. The metal flows discontinuously through the tap hole, via a refractory lined channel or launder into a separate collection vessel or ladle. Alternatively, the molten metal can be continuously directed to a holding furnace. The slag, due to its lower density floats on the molten metal and is tapped separately by means of a dam and a slag spout placed at a higher level. It is collected discontinuously in pots, or continuously granulated in a water stream, or in a special installation for dry granulation.


14

Post combustion Post combustion of the waste gases is used to optimize heat recovery (chemically bonded as CO) and to provide cleaner exhaust gases. In burning CO, any residual carbonaceous material is simultaneously oxidised to CO2 and H2O. The generated heat can be recovered using a heat exchanger and then transported to an internal user (e.g. blast air preheating). Typical arrangements are:

a post combustion chamber placed before (bag filter) or after (wet-filter) the dedusting unit (for below charge-hole off take),

(natural gas) burners or controlled air injection in the cupola shaft (for above charge-hole off take).

The design of the system needs to ensure that the waste gases remain at a temperature above 800 °C and with a suitable residence time to guarantee the complete oxidation of the waste gases. The post combustion of CO allows (additional) heat recovery from the cupola off-gas. Additionally, it allows the melting of scrap contaminated with oil and grease without additional environmental effects and thus stimulates the recycling of metals. Exhaust capture and cleaning is a necessary measure to reduce the emission products from coke combustion such as PCDD/F and dust. However, despite meaning energy recovery and elimination of CO, this represents a possible risk of dioxin reformation in the posterior cooling step. It is necessary to assure minimum residence time in the de novo window (250-450 ºC) performing a rapid enough cooling. It is not recommended post combustion of CO without heat recovery due to its negative impacts on the environment, since powerful burners with a power of tens of kW need to be installed. The burners generate emissions of combustion gases and consume additional oxygen. Cold blast cupola Cold blast cupola (CBC) uses the blast at atmospheric pressure and at normal environmental temperature and, currently, post combustion system is not usual amongst them. The BREF-note considers post-combustion in a CBC as follows: Post combustion limits the emissions of CO and eliminates the majority of organic compounds. If not combusted, they would be captured in the dust or emitted through the chimney. Furthermore, post combustion is recommended in order to reduce the risk of fire in the filter. If the post combustion is in the cupola shaft, the gases are combusted by an injection of air into the upper part of the charge or at a position above the charge top level. The airflow is adjusted so that the off-gases ignite spontaneously, due to their CO content and temperature. The injection nozzles can be placed on one or two levels. The partitioning of the airflow over the various levels, the choice of diameter and the position of the nozzles is based on experience. The goal of the optimization is to burn the CO without ignition of the coke. The draught will also suck in air from the charge door. This air excess allows a more complete burn-out of the CO.


15

A supporting burner may be provided to maintain the flame. When using very low coke charges (i.e. < 6 - 8 %) the precautionary measure is reasonable. In case of post combustion of the off-gas, this must be combined with a gas cooling, if a bag filter is used. In cold blast operation, a rapid cooling may be applied using water injection in the furnace shaft. Alternatively, an (off gas - air) heat-exchanger may be used. Post combustion is also known to avoid explosion risks under certain circumstances. The positive environmental effect is limited to those cases when the off-gas is burning autothermally most of the time. Otherwise, the energy consumption will counterbalance the CO reduction. If the cupola off-gas does not ignite spontaneously, the installation of ignition or support burners is necessary. These incur a significant power use and increase the overall flue-gas volume. The BREF-note defines post combustion in the cupola shaft of a CBC as BAT only if the off-gases can burn autothermally and then recover the heat for internal use. Hot blast cupola Hot blast cupola (HBC) uses a preheated blast. In hot blast cupola, two methods of heating are: - Recuperative heating: this involves the transfer of the residual (“latent”) heat of the flue gases to the combustion air. The flue-gases are collected at the top of the furnace, mixed with sufficient air and then burned in a post combustion unit. This provokes the exothermic oxidation of CO. The burnt gases are led through a heat-exchanger (recuperator) where the heat is transferred to the combustion air. Typically the blast air is heated at temperatures of 500 to 600°C. Above these temperatures, problems arise with the sintering of furnace dust on the surface of the recuperator. - External heating: here the combustion air is heated by some external means, e.g. by a gas or fuel burner, by electrical resistance or by a plasma torch. The combination of these two heating methods permits the superheating of the blast air up to 1000°C. These high temperatures, however, require the use of more expensive refractory materials and may cause too high a melt temperature. Recuperative systems offer increased energy and thermal efficiencies. It should be noted that the coke quality may affect the overall blast efficiency. The advantages of the hot-blast operation may be summarized as follows:

reduced coke consumption increased metal temperature increased melting rate reduced sulphur pick-up (as a consequence of the coke reduction) reduced melting losses increased carbon pick-up and hence the ability to substitute scrap steel for pig

iron in the furnace charge. The advantages of the hot blast cupola are a high flame temperature, allowing good thermal efficiency and the charging of a higher level of steel than in the cold blast cupola. Experience has shown that increasing the blast temperature by 200°C, from 550 to 750°C, which takes 60


16

kWh per ton of iron, saves 10 kg of coke per ton melted. The main benefit, more important than a saving of coke, is flexibility: hourly output can be increased by 30% without modifying the melting bed. A post combustion chamber with a burner is installed after the cupola. If there is a separate post combustion unit it should be preheated with a natural gas burner. Once the cupola is up and running either a smaller burner sustains the combustion of the waste gases or the gases self-ignite. The type and position of the chamber can vary according to the process composition. Both horizontal and vertical combustion chamber types exist: - Hot blast cupola with a recuperator and wet scrubber: in this arrangement, the gases are dedusted prior to combustion. This reduces dust build-up in the recuperator, which improves the rate of heat transfer. One disadvantage is the higher energy consumption in the post combustion unit, caused because the gases are cooled down in the wet scrubber. Early cooling of the offtake gases is continuously carried out to reduce the size of the dedusting unit - Hot blast cupola with a recuperator and bag filter: the hot, dust laden, top gases are fed directly into the post combustion unit. Close process control is necessary to prevent sintering of the dust particles to the walls of the recuperator, which need to be cleaned regularly. The gases need further cooling before entering the bag filter since they leave the recuperator at temperatures of 500 to 600°C. Burning the fumes in the post combustion chamber does not consume much energy, providing there is sufficient carbon monoxide in the fumes, which is generally the case. But the whole system for treating the fumes (combustion chamber + heat-exchanger + filter or wet scrubber + fans) also needs electrical energy and regular maintenance. For economic reasons, the post combustion chamber is only used on hot blast cupolas. Nature of atmospheric emissions Cupolas can be charged with a wide range of materials, many of which may contain loose particles such as rust, sand and non-ferrous materials. The metallurgical-coke can break and produce small pieces, as can the added fluxing materials. Breakage and mechanical abrasion during charge preparation, as well as during charging itself, generate particles, some of which are immediately emitted. During melting, abrasion of the charge against the refractory lining will also generate dust. A third source of particulate matter is coke ash, generated in the melting zone, which is not trapped by the slag phase. Particulate matter of various sources, if light enough, can be entrained in the combustion gases of the cupola. Under certain conditions metallurgical-fume may be generated from the melting zone, leading to a visible plume from the cupola stack. The smoke particles consist of submicron agglomerates of spherical soot particles and metallic oxides, such as ZnO, PbO, etc., if the metals are present in the charged steel or iron scrap, such as in galvanized or painted scrap. The smoke emission will increase with the proportion of coke and contaminants in the charge, the blast temperature, and the oxygen injection rate.


17

Carbonaceous smoke is airborne matter formed by the incomplete combustion of organic matter in the cupola. Scrap contaminants such as oil and grease, wood, textiles and, rubber will form oily vapors in the stack gases. Vapors and partially burnt organic matter may carry unpleasant smells. Again, scrap cleanliness and its nature significantly affect the nature of the emissions. The burning of coke creates odorous gas emissions containing CO2, CO and SO2. Decreasing the proportion of the coke charge (by increasing the thermal efficiency) or (partial or complete) substitution of the coke by natural gas can help reduce the levels of these substances. Waste gases Coke fired cupola gas is composed primarily of N2, CO2, H2O, and CO, with smaller amounts of SO2. For conventional cupolas, where the off-gas is collected above the charging door, a distinction has to be made between the condition of the flue-gases below and above the charging door, since ambient air is entrained through the open charge door. This input significantly changes the total airflow. If the cupola gases are hot enough, and if there is enough CO present, the gases may burn spontaneously together with drawn-in air (2CO + O2 → 2 CO2) and temperatures may rise to 900°C. Little or no CO will then be left in the exhaust. If no combustion takes place, the air intake will result in a cooling effect, of between 100 and 300°C, and the CO/CO2 balance will remain unchanged. The temperature of the gases just below the charging door is primarily dependent on the charge height; the input of ambient air is determined by the fan capacity or the natural draught available. The flow of the undiluted cupola top gases is proportional to the coke consumption. Increasing the coke proportion in the charge will decrease the production rate (ton molten metal/h) if the same blast airflow is maintained. It may then be necessary to increase the blast to maintain the production. The metal temperature will also increase. With reference to the combustion rate (CO + O2 → 2 CO2), more coke and blast air will result in an increased flow of exhaust gases. For a given furnace, the coke and blast air consumption depend on the targeted melting rate and the metal temperature, which can vary on an hourly basis. Typical flow rates reported in literature vary from 600 to 800 Nm³/tmetal charge for CBC and 500 to 700 Nm³/tmetal charge for HBC. Above the charging door, the flow rates of the exhaust gases may be two to five times higher, depending on the target temperature (which depends on the kind of dust arrestment system applied) of the diluted gases and, in the case of a recuperative hot blast cupola, the presence of a post combustion chamber. Typical flow rate values are 3000 to 4000 Nm³/tmetal charge melted for CBC and 900 to 1400 Nm³/tmetal charge when post-combustion is performed. The composition of the gases is determined by the rate of dilution (natural draught or fan power), the degree of spontaneous combustion of the CO, or the post combustion itself, which can be executed on the complete or partial flow. Cooling Following collection, the gases may need cooling depending on the dust abatement system used. In the hot blast operation, the heat recovered from cooling may be used for preheating the blast air.


18

Several options are possible for cooling the collected gases, including: - using tube coolers: running the collected gases through long ducts, decreases the temperature by natural convection and radiation. This system is simple but takes up a lot of space and does not offer controlled cooling (therefore there is a risk of condensation) - using a forced air/gas heat-exchanger cold ambient air is forced through an arrangement of tubes or plates to cool down the gases. Dust collection and the subsequent need for cleaning the heat-exchanging surfaces may lead to a complex and expensive design of the system. One advantage of this system is the possible use of the heated air for external heating purposes. Recuperative hot blast cupolas are equipped with a post combustion unit and a heat-exchanger (recuperator) to heat the blast air. - using an oil/gas heat-exchanger. This is similar to the above system but more expensive because of the need for a secondary cooling system. The heat-exchanger is generally cooled with a circulation of mineral oil. Cooling with a water/gas heat-exchanger is not (or only very rarely) practiced. - saturation with water : here the gases are cooled by the evaporation of the water sprayed into the gas stream. Wet scrubbers perform better if the gases are cooled in a saturation chamber prior to cleaning. When using fabric filters only, partial saturation is possible to prevent clogging of the fabric due to the condensation of water. A good control system is necessary to guarantee correct functioning of the system. Quenching the gases has the advantage that rapid cooling reduces the risk of dioxin formation. Dedusting Dust capture equipment of various types can be used to remove particulate matter from the waste gases. With regard to dry systems, the following remarks can be made: - Multi cyclones are often used in conjunction with a fabric filter, acting as coarse dust arrestors. They help to prevent incandescent coke particles from reaching the filter cloth. Provided refractory lining and high grade steel are used in the design of the cyclone, they can operate at high temperatures. The collection efficiency from cyclones alone is not sufficient to meet today’s regulations; hence they are usually used in combination with other dedusting systems. - Bag filters are ideal when the gases are burned prior to the dedusting. This avoids problems of the deposition of carbonaceous material or fire hazards. Bag filters can be designed to provide good efficiency for collecting metallurgical fume particles such as ZnO. - Electrostatic precipitators are less common in the European foundry industry. This system is best suited to more or less constant working conditions, such as in long campaign cupolas, because of its sensitivity to variations in gas temperature, flow and humidity. There is an explosion hazard when dedusting unburned gases mixed with air, due to the relatively large volume of the precipitator. The precipitator therefore needs to be flushed before applying electrical power. Generally wet scrubbers have low capital costs and maintenance, but require a high energy input to achieve acceptable collection efficiencies. The removal of the sludge is difficult and the


19

scrubber water has to be treated prior to discharge. Dry collection systems have more expensive capital costs and need better control of the inlet gas conditions (temperature, condensation of water or organic vapors, CO/O2 ratio, sparks) but usually use less energy than that needed for wet scrubbing. Both venturi scrubbers and disintegrators are used with cupola systems. A separator to remove small particles entrained in water droplets, is located after the wet scrubber. 3.1.2 Results Remembering that the results of this document belong to the Diofur Project, the situation of the emissions in cupolas operating in normal conditions is placed in the following range:

Table 3: Dioxin emission range for cupolas Source: Diofur Project

FURNACE TYPE ng I-TEQ/Nm³ (EU Project goal:<0.1)

Hot Blast Cupola 0.0017÷0.4166 Cold Blast Cupola 0.0010÷0.3290

It can be observed that the emissions show a high variability and both emission levels above or below the target (0.1 ng I-TEQ/Nm³, ELV1 of Incineration Directive) could be achieved. Therefore during the mentioned Project some trials of parameter modifications were performed giving the following results:

Table 4: Emission data of emissions in normal and modified conditions Source. Diofur Project

Furnace CBC HBC

Type of measurement CURRENT SITUATION PARAMETER

MODIFICATION CURRENT

SITUATION PARAMETER MODIFICATION

Date 19/12/06 20/12/06 21/12/06 30/10/07 13/03/07 04/04/07 29/11/07

ng/Nm³ 0.329 0.0013 0.0025 0.207 0.0017 0.4166 0.2711

Interruptions Max Middle Low Low None None None Injected O2 Normal Normal 1.5% 0% 530 Nm³/h 530 Nm³/h 836 Nm³/h

Change of coke No No No Chinese No 20% Anthracite No

Process disruptions No No Cyclone

breaks No

quenching spray

No No Changes in blast flow

Most important causes of dioxin formation were the stoppage and quenchers temperature control. It was recommended a good and

accurate temperature control of the exhaust in order to maintain working the sprayers and especially to avoid as more as possible the

interruptions.

In this case there were no

dioxins, but it is noted that in

previous samplings

dioxins were found. See

recommendations about

continuous temperature

control, particulate matter and

clean charges.

Regarding this result,

anthracite seems to have

influence in the process

and it is noted that oxygen injection was not enough. It is an option to

reduce the dust level and to increase the

oxygen injection.

The effect of coke as chloride provider makes

almost impossible to avoid dioxins formation in the exchangers after

the post combustor, even when the post combustion

destroys more than 95% of

arriving dioxins. More oxygen injection was

used and it has been interpreted

that it is not enough.

1 Emission limit value


20

Therefore, after observing the results in the table above, it could be said that for the CBC, process stoppages and not enough high temperatures of the exhaust could influence the dioxin emissions2. However the control of these parameters does not assure that emissions will be below the emission rate in any case. Other issues also have to be controlled because they are thought to be influencing the dioxin emission:

Raw materials. It is recommended to check wether the use of clean raw materials has an effect on dioxin emissions. Especially the scrap has oil, painting, grease contents which are a chlorine source for the dioxin formation.

Material storing places. It is very important to store each material always in the same place not to pollute clean raw materials. Moreover, the storing places itself, should be cleaned regularly.

Gas line. Long and horizontal ducts can lead to problems with reformation and the memory effect. Therefore they have to be avoided as long as it is possible. Moreover, ducts have to be kept clean.

Dust depuration system. A high percentage of dioxins are trapped in the dust particles. Therefore an exhaust treatment eliminates them avoiding the emission of dioxins with these characteristics.

All available information proves that also CBC are able to meet the limit value of 0.1 ng I-TEQ/Nm³ if the necessary measures are taken. In other cases, secondary measures are necessary.

3.1.3 Additional beneficial effects (metallurgy, reduction of other contaminants) Avoiding stoppages users can obtain additional benefits such as chemical analysis stability which leads to a better metallurgical quality and to fulfil standards. Finally, fewer stoppages mean not only less memory effect and dioxin content in other wastes, but also better energy efficiency.

2 This had a direct influence in the gas line of the equipment, where the scrubbers did not work avoiding the first step of gas and particle depuration. These facts had a direct influence from our point of view on the sampling of that day. It is NOT possible to generalize this situation but it has to be taken into account for installations with a similar gas line.


21

3.2 Rotary furnace melting 3.2.1 Furnace operation Description The rotary furnace consists of a horizontal cylindrical vessel, generally with two trunk cones in its sides, in which the metallic charge is heated by a burner located at one side of the furnace. The flue-gases leave the oven through the opposite side. To generate the required heat, fuel or natural gas is used combined with air or pure oxygen. The newest equipment available, (pure oxygen), comes with a built-in high efficiency oxygen burner where thanks to the combustion with pure oxygen it is possible to reach high flame temperatures of up to 2600ºC (in the flame core). The pure oxygen burner is water-cooled with an exterior jacket (with water or part of the flow of cold oxygen or gas). The outlet for the combustible and comburent mixture can vary depending on the morphology and length of the flame, or even on the design of each supplier. These elements can be found in tubular or multiflame form, the tubulars being the preferred ones given their high performance and their simplicity of design. A tilting mechanism allows the furnace to be lifted to a certain angle or into a vertical position. This position is used for lining repair and renewal lining while the position lifted to an angle is for the charging of the furnace with a drop bottom bucket or a vibrating chute. During heating and melting the furnace is rotated slowly, continuously or intermittently respectively, to allow the optimum heat transfer and distribution. The furnace atmosphere is controlled by the air (oxygen)/fuel ratio. Once the metal is melted, and after a composition and temperature check and adjustment, a tap-hole in front of the furnace is opened and the melt in the furnace is discharged into ladles. Because of its lower density, the slag floats on the metal bath in the furnace and remains in the furnace if the tap hole is well kept and finally, at the end of the casting operation, is collected through the tap-hole into slag pots. A melting cycle spans from 1 to several hours depending on the furnace capacity. For continuous molten metal production, foundries install 2 or more rotary furnaces, which are operated consecutively. Considering the process itself, the thermal efficiency of the rotary furnace is very high, i.e. at 50 to 65 %, depending on the capacity. This high yield is achieved by using pure oxygen instead of air as the combustion medium. Melting practice Control of process parameters; Influential parameters:

Refractory. It is composed of high purity pisé of quartzite and agglomerated with special refractory clay. After a good compaction, a curing and stabilization with a controlled warm-up is necessary in order not to modify the thermal phases of the quartzite as the electric furnaces do.

Pouring temperature. As occurs in any foundry, pouring temperature depends on the

product. Therefore, big pieces made in 20 t furnaces that have a thick section need slightly


22

SPECIFIC POWER‘OXIGAS’ ROTARY FURNACES

CHARGES (t)

SPECIFIC POWER‘OXIGAS’ ROTARY FURNACES

CHARGES (t)

lower tapping temperature (from 1350 to 1400ºC) than small pieces with a slightly higher temperature (from 1480 to 1520ºC).

Charging and casting. It is important to always abide to the charging protocol, so that

the raw materials are added in the correct order to the furnace. This will ensure a good use of the energy, a high metallurgical efficiency and therefore also a high environmental efficiency.

The casting operation. There have been developed several specific lines of work for the

last 15 years, which allowed to create a new work methodology and to transfer it to the industrial stage.

Current casting ratios. The next chart shows the specific consumption values of the

existing optimized foundries:

Figure 6: Diagram representing the power generated for the charge introduced in an Oxigas RF. Source: Inasmet-TECNALIA

The lifespan of the refractory is largely dependent on its running use as well as on the overheating temperature or the charge composition. In the charging operation, mechanical shocks and cold start-ups need to be prevented. The furnace atmosphere (oxidative or reductive), the holding time, rotational speed in each phase and the burner position also affect the refractory life. In normal conditions the refractory life is 100 to 300 melting cycles without renewals. With renewals of the lining it has been possible to reach more than 2000 melting cycles. Metallurgy This type of furnace has been used in non-ferrous melting for many years. In this application traditional oil-air burners can provide the relatively low flame temperatures. The development of oxygen-air burners has enabled the introduction of cast iron production (of a higher melting point), using a higher relative amount of scrap steel and applying graphite for recarburization. A significant disadvantage of the rotary furnace is its high oxidative capacity and that it also burns Fe, C, Si, Mn or S. These losses have to be compensated for by the addition of alloying elements before or after melting. Depending on the alloying element the efficiency of uptake of these elements is usually rather low. Concentration gradients may occur between the front and


23

the back of the metal bath due to the absence of axial motion and due to inhomogeneities in radiation and the atmosphere above the wide bath surface. Application Due to its batch character, the rotary furnace provides the same flexibility than the coreless induction furnace in the cast iron foundry. The investment costs however are generally lower. A 5 t furnace costs EUR 500000 – 600000, of which 30 % are for the exhaust system and dedusting. The rotary furnace is also a good alternative for the small-scale cold blast cupola, due to its higher flexibility and lower environmental costs. Rotary furnaces are currently used for melting volumes of 2 to 20 t (real range 0.6 to 30t), with production rates of 1 to 6 t per hour. Advantages

Quick change of alloy possible. Melting without contamination, e.g. without sulphur pick-up. Less investment costs. Small dedusting system because of low flue gas rate. Easy to use and maintain.

Disadvantages

If there is not an appropriate control high burn-out of different elements as Fe, C, Si, Mn, etc.

Gas and oxygen use can be high if not operated continuously. The energy consumption increases if more steel is added to the charge.

Increasing the furnace efficiency Description All measures that increase the thermal efficiency of the furnace will in turn lead to a lower CO2 output. A major improvement has been achieved by introducing oxygen instead of air as the combustion medium. Further improvements of furnace yield can be obtained through tight control and optimization of:

Burner regimen. Burner position. Charging. Metal composition. Temperature.

Achieved environmental benefits The optimization results in lower amounts of dust and residues and in higher energetic efficiency. Cross-media effects


24

No cross-media effect applied. Operational data An optimization program on a 3 tons/h cast iron melting furnace suggested the following as good operational practice:

Use of clean scrap and loading in the following order: (1) ingots and materials with low Si-content; (2) internal return material and foundry scrap; (3) alloying elements and melt protection; (4) scrap steel.

Melt protection: use of anthracite for melt protection (2 % of metal charge) and silica

sand (2 %).

Rotations: there is a specific work methodology that consists in increasing gradually the turning speed (from a quarter of a turn every 3 minutes to half a turn every 2 minutes), which is discontinued when the charge is still solid and continued and increasing in the overheating phase (recommended a speed from 1 to 3 rpm).

Power and angle of burner: use a parallel (with 10º of inclination downwards) burner-

head position for the lower injectors. Start at maximum power for 20 minutes, reducing 10 % every 20 minutes until change of phase (60 minutes after start).

Using these measures, a metal efficiency (molten metal/charged metal) at >95 % could be maintained. Applicability The principles of the optimization are generally valid for any furnace. Driving force for implementation Optimization of the furnace operation in order to increase the melting efficiency. Example plants Process optimization measures are commonly applied in European foundries using rotary furnaces. Reference literature: Use of an oxyburner Description Flame temperatures are increased by the application of pure oxygen instead of air in the burners used for melting or preheating the pouring ladles. This allows a more efficient heat transfer to the melt and reduces the energy use. If the air supply is blocked by a tight closure of the recipient, no NOx can be formed through the oxidation of atmospheric nitrogen. Additionally, the total flow of flue-gases from an oxyburner is smaller due to the absence of nitrogen ballast. This allows the application of a smaller dedusting installation. Achieved environmental benefits


25

The techniques reduce energy consumption and lower the emissions of NOx and CO2, through higher combustion temperatures. Cross-media effects The production, storage and use of oxygen increase the safety risks. Oxygen production is done through cryogenic distillation or Vacuum (Pressure) Swing Adsorption, which both involve electricity consumption. The consumption of the latter technique is 0.35 – 0.38 kWh/Nm³ O2. Oxygen production is often done by an external supplier, who delivers the oxygen to a storage tank or directly through a pipeline. Fuel or heavy oil gives rise to SO2 or NOx emissions, depending on their S or N content. The use of cleaner carburants such as natural gas and propane will not cause any additional pollution, except for CO2, as is the case for all combustion processes. Operational data Table 5 gives, for the "oxygas" melting of cast iron and various furnace capacities, the theoretical consumption of several fuels and of oxygen per ton melted:

Table 5: Table of energy consumptions (minimum melt). Source: BREF Document on Smitheries and Foundries

Oxygen enrichment used in conjunction with a recuperator generally achieves a 30 % energy saving. Additionally, the higher combustion temperature assists in reducing the overall emissions. The exhaust gas volume is also reduced. Full oxy/fuel firing may offer energy savings of up to 50 %, and can reduce the exhaust gas volume by up to 72 %. Applicability This technique can be applied on any rotary furnace and in the preheating of pouring ladles. Oxyburners do not find implementation in non-ferrous foundries, although they are used e.g. in secondary copper smelting. Economics Investment costs EUR 3400 – 4500. Operational costs: dependent on process operation. Driving force to implement an optimisation of the furnace operation and to increase melting efficiency. Example plants This technique finds wide application in ferrous foundries using rotary furnaces. Selection of the type of furnace in foundries Description


26

Given the fact that various melting techniques show overlapping fields of application and taking into account reference to this issue in the BREF document, a basic technique is the selection of the melting technology. In this selection, there are many influential parameters including the environmental. Before having any emission measurement result it is not possible to conclude if a certain furnace type is better than another exclusively from an environmental point of view.

BEST AVAILABLE TECHNIQUES FOR FOUNDRIES Ferrous metal melting: Rotary furnace melting of cast iron For the operation of rotary furnaces, BAT is all of the following:

To implement measures to optimize furnace yield as discussed in Section 4.2.4.1 of BREF in the Smitheries and Foundries Industry:

To use an oxyburner (Section 4.2.4.2 of BREF in the Smitheries and Foundries Industry).

To collect the off-gas close to the furnace exit, apply post combustion, cool it using a heat-exchanger and then to apply dry dedusting (Section 4.5.5.1 of BREF in the Smitheries and Foundries Industry), taking into account the BAT associated emission levels as given in Table 5.1 and Table 5.4.

To prevent and minimize dioxins and furan emissions to a level below 0.1ngI-TEQ/Nm³, using a combination of measures as given in Section 4.5.1.4. of BREF in the Smitheries and Foundries Industry. In some cases this may result in a preference for wet scrubbing. Industry has expressed doubts on the implementation of secondary measures that have only been proven in other sectors and in particular questions the applicability for smaller foundries.

3.2.2 Results Remembering that the results of this document belong to the Diofur Project, the situation of the emissions in RFs operating in normal conditions is placed in the following range:

Table 6: Dioxin emission range for rotary furnaces Source: Diofur Project


Rotary (Small Size) 0.0160÷0.0190 Rotary (Big Size) 0.0019÷0.0022

It can be observed that the emissions are below the emission target (0.1 ng I-TEQ/Nm³, ELV of Incineration Directive). Therefore, after analysing the situation of some RFs, it has been seen that they are not potentially dioxin generating installations. However, these results do not assure that in every RF will be achieved emissions below the limit value. For that reason, in this document are given good practices tips and recommendations to operate correctly in foundries using this type of furnaces. (See part 2: block diagram, task guidance sheets and check list)


27

3.2.3 Additional effects (metallurgy, reduction of other contaminants) An efficient conduction avoids refractory line oxidation and wear (main sources of dust emissions) and due to this fact has lower energy consumption and dust emissions (Carnicer 2001). Moreover, the melting time also decreases. Lower energy consumption means a lower source of organic carbon; and in the same way, lower melting time means a higher metallurgical quality and less oxidations. On the other hand, the oxidations in this type of furnace lead to high CO2 production which is considered to be a carbon source. Therefore, the available protocols for an industrial optimization (from both metallurgical and energetic perspectives) have to lead to lower dioxin emission rates, which are by nature quite low.


28

3.3 Electric arc furnace melting 3.3.1 Furnace operation (acid/basic lining) Description Most of the commonly used electric arc furnaces (EAF) in foundry practice are of 1.5 – 30 t capacity. In metallurgy industry EAFs of much bigger capacity are used. The Electric Arc Furnace is a 3 phase furnace in which an electric arc is obtained between vertically hanged electrodes and metallic charge. In each of the 3 circuits a current floats as follows: electrode I – arc – slag – metal – slag – arc – electrode II. The scheme of the EAF is presented in figure 7:

Figure 7: Scheme of an EAF. Source: Encyclopedia Britanica

The furnace consists of a spherical hearth (bottom), cylindrical shell and a swinging water-cooled dome-shaped roof. The roof has three holes for consumable graphite electrodes held by a clamping mechanism. The mechanism provides independent lifting and lowering of each electrode. The water-cooled electrode holders serve also as contacts for transmitting electric current supplied by water-cooled cables (tubes). The electrode and the scrap form the star connection of three-phase current, in which the scrap is common junction. The furnace is mounted on a tilting mechanism for tapping the molten steel through a tap hole with a pour spout located on the back side of the shell. The charge door, through which the slag components and alloying additives are charged, is located on the front side of the furnace shell. The charge door is also used for removing the slag (de-slagging). The scrap is charged commonly from the top part of the furnace. The roof with the electrodes is swung aside before the scrap charging. The scrap arranged in the charge basket is transferred to the furnace by a crane and then dropped into the shell.


29

Refractory lining of an Electric Arc Furnace Refractory linings of Electric Arc Furnaces are made generally of resin-bonded magnesia-carbon bricks (basic lining). Fused magnesite grains and flake graphite are used as raw materials. When the bricks are heated the bonding material is coked and turns into a carbon network binding the refractory grains, preventing wetting by the slag and protecting the lining from erosion and chemical attack of the molten metal and slag. Basic refractory linings of EAFs are made of magnesite without carbon addition bricks, of magnesite-chromite bricks or of roast dolomite, too. Refractory in the EAF’s roof are generally made of quartz, schamotte or mullit bricks. In some countries EAF’s hearth and shell lining are made from acid refractory (quartz) too, but furnaces with this type of lining are used only occasionally. Operation of an Electric Arc Furnace • Scrap charging; • Melting; • Sampling and chemical analysis of the melt; • Oxidizing slag formation; • Oxidation of C, P, Mn, Si, Al; • Sampling and temperature measurement; • De-slagging; • Basic slag formation; • Deoxidizing (”killing”); • Alloying; • Tapping the steel; • Refractory lining maintenance. Furnace Charging The first step in the production of any heat is to select the grade of steel to be made. Usually a schedule is developed prior to each production shift. Thus the foundryman will know in advance the schedule for his shift. The scrap yard operator will prepare buckets of scrap according to the needs of the foundryman. Preparation of the charge bucket is an important operation, not only to ensure proper melt-in chemistry but also to ensure good melting conditions. The scrap must be layered in the bucket according to size and density to promote the rapid formation of a liquid pool of steel in the hearth while providing protection for the sidewalls and roof from electric arc radiation. Other considerations include minimization of scrap cave-ins which can break electrodes and ensuring that large heavy pieces of scrap do not lie directly in front of burner ports which would result in blow-back of the flame onto the water cooled panels. The charge can include lime and carbon or these can be injected into the furnace during the heat. Many operations add some lime and carbon in the scrap bucket and supplement this with injection. The first step in any tap-to-tap cycle is "charging" of the scrap. The roof and electrodes are raised and are swung to the side of the furnace to allow the scrap charging crane to move a full bucket of scrap into the furnace. The bucket bottom is usually a clam shell design - i.e. the bucket opens up by retracting two segments on the bottom of the bucket. The scrap falls into the furnace and the scrap crane removes the scrap bucket. The roof and electrodes swing back


30

into place over the furnace. The roof is lowered and then the electrodes are lowered to strike an arc on the scrap. This is done to begin the melting step of the cycle. The number of charge buckets of scrap required to produce a heat of steel is dependent primarily on the volume of the furnace and the scrap density. Most operations in foundries with EAFs have capacities between 1.5 – 6 t achieved with a single bucket charge. Most of the modern furnaces are designed to operate with a minimum of back-charges. This is advantageous because charging is a dead-time where the furnace does not have power on and therefore is not melting. Minimizing these dead-times helps to maximize the productivity of the furnace. Melting The melting step is the heart of EAF operations. The EAF has evolved to a highly efficient melting apparatus and modern designs are focused on maximizing the melting capacity of the EAF. Melting is accomplished by supplying energy to the furnace interior. This energy can be electrical or chemical. Electrical energy is supplied via the graphite electrodes and is usually the largest contributor in melting operations. Initially, an intermediate voltage tap is selected until the electrodes bore into the scrap. Usually, light scrap is placed on top of the charge to accelerate bore-in. Approximately 15% of the scrap is melted during the initial bore-in period. After a few minutes, the electrodes will have penetrated the scrap sufficiently so that a long arc (high voltage) tap can be used without risk of radiation damage to the roof. The long arc maximizes the transfer of power to the scrap and a liquid pool of metal will form in the furnace hearth. At the start of melting the arc is erratic and unstable. Wide swings in current are observed accompanied by rapid movement of the electrodes. As the furnace atmosphere heats up, the arc stabilizes and once the molten pool is formed, the arc becomes quite stable and the average power input increases. Temperature of the arc reaches 3500ºC. During this operation the gas emission depends on the charged scrap type and consequently the dioxin formation. Chemical energy is supplied via oxygen lances. In some operations, oxygen is injected via a consumable pipe lance to "cut" the scrap. The oxygen reacts with the hot scrap and burns iron to produce intense heat for cutting the scrap. Once a molten pool of steel is generated in the furnace, oxygen can be lanced directly into the bath. This oxygen will react with several components in the bath including aluminum, silicon, manganese, phosphorus, carbon and iron. All of these reactions are exothermic (i.e. they generate heat) and supply additional energy to aid in the melting of the scrap. The metallic oxides that are formed will end up in the slag. The reaction of oxygen with carbon in the bath produces carbon monoxide, which either burns in the furnace if there is sufficient oxygen, and/or is exhausted through the direct evacuation system where it is burned and conveyed to the pollution control system. This step has a high emission rate of residual gases and metallic oxide particles. Once enough scrap has been melted to accommodate the second charge, the charging process is repeated. Once the final scrap charge is melted, the furnace sidewalls are exposed to intense radiation from the arc. As a result, the voltage must be reduced. Alternatively, creation of a foamy slag will allow the arc to be buried and will protect the furnace shell. In addition, a greater amount of energy will be retained in the slag and is transferred to the bath resulting in greater energy efficiency. Once the final scrap charge is fully melted, flat bath conditions are reached. At this point, a bath temperature and sample will be taken. The analysis of the bath chemistry will allow the melter to determine the amount of oxygen to be blown during refining. At this point, the melter


31

can also start to arrange for the bulk tap alloy additions to be made. These quantities are finalized after the refining period. Refining Refining operations in the electric arc furnace have traditionally involved the removal of phosphorus, sulfur, aluminum, silicon, manganese and carbon from the steel. In recent times, liquid gases, especially hydrogen and nitrogen, have been recognized as a concern. Traditionally, refining operations were carried out following meltdown i.e. once a flat bath was achieved. These refining reactions are all dependent on the availability of oxygen. Oxygen was lanced at the end of meltdown to lower the bath carbon content to the desired level for tapping. Most of the compounds which are to be removed during refining have a higher affinity for oxygen than the carbon. Thus the oxygen will preferentially react with these elements to form oxides which float out of the steel and into the slag. Phosphorus and sulfur occur normally in the furnace charge in higher concentrations than are generally permitted in steel and must be removed. Unfortunately the conditions favorable for removing phosphorus are the opposite of those promoting the removal of sulfur. Therefore once these materials are pushed into the slag phase they may revert back into the steel. Phosphorus retention in the slag is a function of the bath temperature, the slag basicity and FeO levels in the slag. At higher temperature or low FeO levels, the phosphorus will revert from the slag back into the bath. Phosphorus removal is usually carried out as early as possible in the heat. Hot heel practice is very beneficial for phosphorus removal because oxygen can be lanced into the bath while its temperature is quite low. Early in the heat the slag will contain high FeO levels carried over from the previous heat thus aiding in phosphorus removal. High slag basicity (i.e. high lime content) is also beneficial for phosphorus removal but care must be taken not to saturate the slag with lime. This will lead to an increase in slag viscosity, which will make the slag less effective. Sometimes fluorspar is added to help fluidize the slag. Stirring the bath with inert gas is also beneficial because it renews the slag/metal interface thus improving the reaction kinetics. In general, if low phosphorus levels are a requirement for a particular steel grade, the scrap is selected to give a low level at melt-in. The partition of phosphorus in the slag to phosphorus in the bath ranges from 5 to 15. Usually the phosphorus is reduced by 20 to 50 % in the EAF. Sulfur is removed mainly as a sulfide dissolved in the slag. The sulfur partition between the slag and metal is dependent on slag chemistry and is favored at low steel oxidation levels. Removal of sulfur in the EAF is difficult especially given modern practices where the oxidation level of the bath is quite high. Generally the partition ratio is between 3 and 5 for EAF operations. Most operations find it more effective to carry out desulfurization during the reducing phase of steelmaking. This means that desulfurization is performed during tapping (where a calcium aluminate slag is built) and during ladle furnace operations. For reducing conditions where the bath has a much lower oxygen activity, distribution ratios for sulfur of between 20 and 100 can be achieved. Control of the metallic constituents in the bath is important as it determines the properties of the final product. Usually, the melter will aim at lower levels in the bath than are specified for the final product. Oxygen reacts with aluminum, silicon and manganese to form metallic oxides, which are slag components. These metallics tend to react with oxygen before the carbon. They will also react with FeO resulting in a recovery of iron units to the bath. For example:


32

Mn + FeO = MnO + Fe Manganese will typically be lowered to about 0.06 % in the bath. The reaction of carbon with oxygen in the bath to produce CO is important as it supplies a less expensive form of energy to the bath, and performs several important refining reactions. In modern EAF operations, the combination of oxygen with carbon can supply between 30 and 40% of the net heat input to the furnace. Evolution of carbon monoxide is very important for slag foaming. Coupled with a basic slag, CO bubbles are tapped in the slag causing it to "foam" and helping to bury the arc. This gives greatly improved thermal efficiency and allows the furnace to operate at high arc voltages even after a flat bath has been achieved. Burying the arc also helps to prevent nitrogen from being exposed to the arc where it can dissociate and enter into the steel. If the CO is evolved within the steel bath, it helps to strip nitrogen and hydrogen from the steel. Nitrogen levels in steel as low as 50 ppm can be achieved in the furnace prior to tap. Bottom tapping is beneficial for maintaining low nitrogen levels because tapping is fast and a tight tap stream is maintained. A high oxygen potential in the steel is beneficial for low nitrogen levels and the heat should be tapped open as opposed to blocking the heat. At 1600 C, the maximum solubility of nitrogen in pure iron is 450 ppm. Typically, the nitrogen levels in the steel following tapping are 80 - 100 ppm. Decarburization is also beneficial for the removal of hydrogen. It has been demonstrated that decarburizing at a rate of 1 % per hour can lower hydrogen levels in the steel from 8 ppm down to 2 ppm in 10 minutes. At the end of refining, a bath temperature measurement and a bath sample are taken. If the temperature is too low, power may be applied to the bath. This is not a big concern in modern melting shops where temperature adjustment is carried out in the ladle furnace. The refining operation is the step with higher emission rate of gases and particles and therefore it is a decisive parameter for the gas line design. De-Slagging De-slagging operations are carried out to remove impurities from the furnace. During melting and refining operations, some of the undesirable materials within the bath are oxidised and enter the slag phase. It is advantageous to remove as much phosphorus into the slag as early in the heat as possible (i.e. while the bath temperature is still low). The furnace is tilted backwards and slag is poured out of the furnace through the slag door. Removal of the slag eliminates the possibility of phosphorus reversion. During slag foaming operations, carbon may be injected into the slag where it will reduce FeO to metallic iron and in the process produce carbon monoxide which helps foam the slag. If the high phosphorus slag has not been removed prior to this operation, phosphorus reversion will occur. During slag foaming, slag may overflow the sill level in the EAF and flow out of the slag door. The following table shows the typical constituents of an EAF slag:

Table 7: Range of compositions for EAF slag. Source: PFRI

Component Source Composition Range


33

CaO Charged 40 - 60 %

SiO2 Oxidation product 5 - 15 %

FeO Oxidation product 10 - 30 %

MgO Charged as dolomite 3 - 8 %

CaF2 Charged - slag fluidizer

MnO Oxidation product 2 - 5%

S Absorbed from steel

P Oxidation product

Tapping Once the desired steel composition and temperature are achieved in the furnace, the tap-hole is opened, the furnace is tilted, and the steel pours into a ladle for transfer to the next batch operation (usually a ladle furnace or ladle station). During the tapping process bulk alloy additions are made based on the bath analysis and the desired steel grade. De-oxidisers may be added to the steel to lower the oxygen content prior to further processing. This is commonly referred to as "blocking the heat" or "killing". Common de-oxidisers are aluminum or silicon in the form of ferrosilicon or silico-manganese. Most carbon steel operations aim for minimal slag carry-over. A new slag cover is "built" during tapping. For ladle furnace operations, a calcium aluminate slag is a good choice for sulfur control. Slag forming compounds are added in the ladle at tap so that a slag cover is formed prior to transfer to the ladle furnace. Additional slag materials may be added at the ladle furnace if the slag cover is insufficient. Cooling of alloy in the ladle could be realized by blowing argon into the molten metal. Furnace Turn-around Furnace turn-around is the period following completion of tapping until the furnace is recharged for the next heat. During this period, the electrodes and roof are raised and the furnace lining is inspected for refractory damage. If necessary, repairs are made to the hearth, slag-line, tap-hole and spout. In the case of a bottom-tapping furnace, the taphole is filled with sand. Repairs to the furnace are made using gunned refractories or mud slingers. In most modern furnaces, the increased use of water-cooled panels has reduced the amount of patching or "fettling" required between heats. Many operations now switch out the furnace bottom on a regular basis (2 to 6 weeks) and perform the hearth maintenance off-line. This reduces the power-off time for the EAF and maximizes furnace productivity. Furnace turn-around time is generally the largest dead-time (i.e. power off) period in the tap-to-tap cycle. With advances in furnace practices this has been reduced from 20 minutes to less than 5 minutes in some newer operations. Off-gas Direct Evacuation System Early off-gas evacuation systems were installed so that the furnace operators could better see what was happening in and around the furnace. Since the early days of EAF steelmaking, the off-gas system has evolved considerably and most modern EAF shops now use a "fourth hole" direct furnace shell evacuation system (DES). The term fourth hole refers to an additional hole


34

other than those for the electrodes, which is provided for off-gas extraction. It is important to maintain sufficient draft on the furnace for the following reasons: 1. To provide adequate pollution control. 2. Excessive shop emissions make it difficult for the crane operator to charge the furnace. 3. Excessive emissions around the electrode ports can result in damage of hoses, cables,

the electrode holder, the furnace delta, roof refractory, accelerated electrode wear, damage to the electrode spray cooler etc.

4. Emissions at the roof ring can result in warping of the roof ring structure. 5. Excessive emissions of carbon monoxide to the secondary canopy system may result in

explosions in the ductwork downstream. 6. Excessive dust build-up may cause arcing between electrode phases.

Most DES systems consist of water-cooled duct, spray cooling, dry duct and may or may not have a dedicated DES booster fan.

Main factors influencing the formation and emission of PCDDs/PCDFs The primary off-gas system receives the off-gas directly from the furnace and is considered the primary source of PCDD/PCDF emission from the process. In order to obtain flow conditions suitable for sampling it would be necessary to build a special installation for sampling on top of the baghouse. Emission factors of EAF plants in Europe show values from 0.07 to 9 µg I-TEQ/t of produced steel (including metallurgical plants). The formation of PCDD/PCDF in an electric arc furnace is not totally understood yet, but there are two main formation mechanisms under discussion: • Certain organic precursors react (e. g. chloro-phenols) on the surface of dust particles at a temperature exceeding 300°C. These precursors enter in the process with scrap impurities like lubrication and cooling oil, paint and plastics. • The de-novo synthesis without organic precursors can occur, i.e. formation out of carbon, oxygen and metal chlorides (e. g. copper chloride) at a temperature of about 250–500 °C in the off-gas. Copper chloride acts as a catalyst in this case. PCDDs and PCDFs are highly viscous liquids at temperatures below 300 °C and can therefore be easily adsorbed at the surface of dust particles. PCDD/PCDF emissions from an EAF plant Most of the existing EAF plants extract the emissions of an electric arc furnace by the 4th hole of the furnace roof (2nd hole in case of DC-furnace). The fumes are mixed with air for post combustion of CO and unburned organic compounds. After cooling, these primary fumes are mixed with the, so-called, secondary fumes coming from the melt-shop building. The combined off-gas flux is then cleaned with fabric filters. 3.3.2 Results Remembering that the results of this document belong to the Diofur Project, the situation of the emissions in EAFs operating in normal conditions is placed in the following range:

Table 8: Dioxin emission range for electric arc furnaces Source: Diofur Project


Electric Arc 1 0.0019÷0.0023


35

Electric Arc 2 0.0020÷0.0370

It can be observed that the emissions show a considerable variability even though they are below the emission target (0.1 ng I-TEQ/Nm³, ELV of Incineration Directive). Therefore, after analysing the situation of some EAFs, it has been seen that they are not potentially dioxin generating installations. However, these results do not assure that in every EAF will be achieved emissions below the limit value. For that reason, in this document are given good practices tips and recommendations to operate correctly in foundries using this type of furnaces. (See part 2: block diagram, task guidance sheets and check list)


36

4. Chemical inhibition of dioxin formation Inhibiting compounds can be added to the flow of stack gases to achieve compliance with the threshold value of 0.1 ng I-TEQ/Nm³ dry without incurring the significant investment costs of other dioxin abatement techniques. The foundry BREF states "sulphur has an inhibiting effect on the formation of dioxins through the depletion of molecular chlorine. It has been shown that the use in large combustion installations of coal having high sulphur content can lower the concentrations of polychlorodibenzodioxins and polychlorodibenzofurans. The inhibiting effect is related to the S/Cl ratio, with a critical ratio of 0.64. Increasing the sulphur content does not reduce the dioxin and furan concentrations. This effect has not been demonstrated in foundries, but can be investigated". Some authors have studied the inhibition of dioxin formation by the addition of chemical compounds in the gases. Globally, 4 classes of inhibitors can be distinguished:

Compounds containing urea. Compounds containing sulphur. Compounds containing nitrogen and sulphur. Metallic oxides.

Their relative inhibition effectiveness is thought to be as follows, in decreasing order: compounds containing nitrogen and sulphur > sulphur compounds > nitrogen compounds > metallic oxides. No tests of the injection of inhibitors to reduce dioxin emissions from foundry cupolas are mentioned. 4.1 Action of urea In steel-making, a dioxin treatment system based on the injection of urea has been set in place. In this type of business, the agglomeration stage is identified as a major emitter of dioxins. The mechanisms of formation in the case of agglomeration lines are still poorly known, but it has been shown that the principal dioxin formation site is situated in the layer to be agglomerated, under the flame front. One of the first applications was at CORUS (UK), where it has been reported that dioxin discharges could be halved by adding a small quantity of urea to the mixture of raw materials. Note that the process involved is totally unlike the foundry cupola process. The injection of urea is relatively simple to implement and inexpensive compared to the other possible dioxin treatment options. Urea is incorporated in the mixture to be agglomerated; its chemical decomposition releases ammonia, which inhibits the formation of chlorinated organic compounds. The urea is added to the mixture in the form of pellets, by means of a flat-bottomed hopper with a proportioning screw. The injection of urea (approximately 0.04% by weight in the mixture) must nevertheless be finely controlled in order to avoid an increase of dust emissions, because ammonium salts are likely to form. Initially, these arrangements were expected to reduce dioxin emissions by 50% and to guarantee maximum concentrations of 1 ng I-TEQ/Nm³ dry.


37

Various studies have been carried out at laboratory scale to evaluate the action of urea during the incineration of household wastes. Injected, with the wastes, directly into the hearth, this compound decomposes rapidly. At the scale of a pilot unit, authors have inhibited the formation of PCDD/PCDF by injecting urea into household waste combustion gases at precise points after the furnace. Nevertheless, in spite of the large number of studies of the subject, the mechanisms of action of nitrogen compounds are still unknown. 4.2 Action of sulphur compounds Researchers at the Institute for Research in Ecological Chemistry of the GSF Centre near Munich have developed a process that can substantially reduce dioxins in the stack gases of incinerators. They have managed to reduce dioxins by as much as 99% by adding non-toxic sulphur compounds. According to the literature, the inhibiting effect of sulphur studied in incinerators could occur in two ways:

- through the formation of hydrochloric acid and of SO3 by the following reaction:

Cl2 + SO2 + H2O 2 HCl + SO3

- through a reduction of the catalytic activity of copper on the ashes by the formation of copper sulphate (CuSO4)

In studies on household waste incinerators, it has been observed that the addition of sulphur compounds (or of coal having a high sulphur content) can reduce PCDD/F emissions. It would seem that adding sulphur compounds to the fuel is more effective than adding gaseous sulphur compounds. 4.3 Action of basic adsorbents The inhibiting effect of such compounds as Ca(OH)2 and NaOH has been studied in steelmaking. Reductions of dioxin discharges by as much as 63% have been observed with the addition of 2% Ca(OH)2 in the gases and by as much as 93% with 5% Ca(OH)2; a similar order of magnitude is possible with NaOH. 4.4 Transposition of the laboratory results to industrial sites The great majority of the experiments have been carried out in laboratories, under conditions that are not representative of industrial sites. For example, in the laboratory, the fly ash itself can be impregnated with the inhibiting products. In an industrial waste incinerator, a larger quantity of the products is needed to overcome problems of the transport of the active molecule to the fly ash. The excess product added may be toxic and/or corrosive, or lead to the formation of other harmful compounds [9]. The choice of the point of injection and temperature of injection of the products makes it all the more difficult to adapt the laboratory experiments to industrial installations.


38

Transfers of pollution or secondary reactions producing other regulated or toxic pollutants (dust, nitrogen oxides, etc.) are possible. There has been very little study of this point so far.


39

5. End of pipe techniques to capture dioxins If concentrations of dioxins above the target value are currently measured, the gas cleaning system has to be checked for:

The temperature profile of the gas. If there is a long residence time in the temperature window of 450 - 250°C, the cooling steps have to be adjusted in a way that one of the cooling steps includes the window of 450°C to 250°C.

Dust deposits in the ducting. Historic dioxin formation can be adsorbed to that dust. Cleaning of the ducting removes this source of dioxins.

Check if a better dust removal can achieve the limit value of 0.1 ng I-TEQ/Nm³ When these primary measurements are not sufficient, there are 3 possible schemes to abate the PCDD/DF:

Install better filter material to achieve a higher efficiency of dust removal. Since dioxins adsorb readily on the dust (values above 99% have been determined during the Project with minimum values of 80%)it may be possible to achieve the limit value of 0.1 ng I-TEQ/Nm³.

Injection of adsorbent ( activated carbon, lignite, brown coal, zeolite, etc.) before the baghouse.

Placement of a fixed bed adsorption after the baghouse. Important point of attention is that when carbon containing adsorbents are used special attention has to go to fire safety. Sparks have to be absent but more important build-up of carbon adsorbent at the bottom of the bag filter has to be avoided. In this build-up of dust, ignition can take place because of exothermal reactions. 5.1 Techniques already used in other industrial sectors In this chapter different techniques that can be used for reduction of PCDD/PCDF are evaluated. Primary measures – preventing the formation of dioxins – are only summarized here. Primary techniques to avoid dioxin formation are a better combustion of the organic compounds (optimization of combustion parameters) and the removal of dust at high temperatures from the flue gas. Also some chemical “inhibitors” can be injected in the flue gas to inhibit the PCDD/PCDF formation. Some examples of inhibitors are urea, NH3, CaO, NaOH, KOH, S, amines etc. The most experience with abatement technologies for dioxins is available for municipal solid waste incineration. This is mostly because, to comply with legal requirements additional measures have to be taken to meet the limit value. Following techniques are capable of reducing dioxin concentrations to the value of 0.1 ng I-TEQ/m³:

Fixed bed adsorption of PCDD/PCDF on e.g. active carbon, brown coal or coke. Injection of e.g. lime + active carbon, followed by fabric filter. Selective catalytic reduction (SCR) using a TiO2-DENOX catalyst.


40

5.1.1 Techniques based on adsorption Activated carbon is widely used in treatment facilities of air and waste waters. The process by which activated carbon removes substances from air is called adsorption. It is a removal process where certain compounds are bound to an adsorbent particle surface by either chemical or physical attraction. The reason that activated carbon is such an effective adsorbent material is due to its large number of pores. These provide a large surface area relative to the size of the actual carbon particle and its visible exterior surface. An approximate ratio is 1 gram = 500 m2 of surface area. Pilot scale experiments indicate that each gram activated carbon adsorbs 105 – 115 ng PCDD/DF. The surface area of the activated carbon is a critical factor affecting PCDD/DF adsorption capacity. Filters with activated carbon are usually used in compressed air and waste gas purification to remove oil vapor, odor, and other hydrocarbons from compressed air and waste gas. The most common designs use a 1 stage or 2 stage filtration principle where activated carbon is embedded inside the filter media. Disadvantages of these techniques are the increased disposal cost (PCDD/DF are moved from air to solid waste) and also the large area required (treatment + storage). Besides activated carbon, other adsorbents can be used in similar technologies. These other adsorbents are brown coal, activated coke, peat and ceramic adsorbents (e.g. AlSiO). Classic design for the adsorbent technique is a packed bed system as final step is the flue gas cleaning system. The temperature of the flue gases at this point is typically 130 to 170°C. Different designs for the filter are proven technology and on the market. Besides packed bed systems, the adsorbent can also be injected directly into the flue gases. This means injection of powder activated carbon into the flue gases in upstream of the bag filter. The amount of carbon injected varies between 50 and 200 mg/Nm³. Advantages of the system are that it is compact, efficient and can be easily integrated.

Figure 8: injection of activated carbon

Removal of dioxins from flue gases of a municipal waste incinerator in Barcelona was investigated. This study evaluates different abatement strategies: ESP, ESP + adsorbers + fabric filters and adsorbers + fabric filters + activated carbon. Using only ESP, the PCDD concentration in the flue gas was between 44 – 111 ng I-TEQ/Nm³. With a semi-dry scrubber, this level was reduced to 15 ng I-TEQ/Nm³. After installing the fabric filter dioxin levels further


41

dropped to 0.3 – 0.4 ng I-TEQ/Nm³. The limit of < 0.1 ng I-TEQ/Nm³ was reached when activated carbon (100 mg/Nm³) was injected in upstream of the filters. A special system is the ADIOX-process. This is based on a material (ADIOX) where carbon particles are dispersed in a polymer matrix. PCDD/PCDF are absorbed in the polymer and then diffused to the surface of the carbon particles. Here they are adsorbed irreversibly, where absorption on pure plastics can result in absorption-desorption. Also the polymer acts as a selective barrier, protecting the carbon filter from other contaminants (e. g. Hg).

Figure 9: ADIOX-process. Source: www.gmab.se

The ADIOX-packings can be used in wet gas cleaning scrubbers as in a dry absorber. The removal efficiency in wet scrubbers is lower due to the water film which limits the mass transfer of the PCDD/PCDF. The dry absorber is located downstream of an ESP, wet scrubber and reheater (see figure). Typical raw concentrations are 6 – 10 ng I-TEQ/Nm³ before entering the ADIOX. At the stack, the clean gas concentrations was <0.1 ng I-TEQ/m3 (n, dg), despite elevated raw gas concentration typical for start-up conditions. Following table shows some differences between the wet and the dry ADIOX.

Table 9: wet versus dry ADIOX-system

Wet Dry Installation in existing scrubbers simple absorber construction Operation scrubber circulation keeps

material clean intermittent rinsing

Performance multifunctional (PCDD/PCDF, HCl, SO2, etc.)

more effective, less material required for same efficiency

Operating temperatures are 20 – 40°C above dew point (dry absorber). The filter can be used as a polishing step, but also in front of e.g. the SCR (selective catalytic reduction, explained in the following section) dioxin removal system. It can reduce the concentration of dioxins into the SCR. Also memory effects can be prevented by using this filter.


42

5.1.2 Techniques based on catalytic reduction An alternative for adsorption is the decomposition of PCDD/DF. This can be done by catalytic reaction, resulting in the formation of H2O, CO2 and HCl. Due to the catalyst the decomposition can be performed at low temperature (150 – 300°C). The catalyst is based on the catalyst used for De-NOx (e.g. TiO2 / V2O5). This means that by adding ammonia (see figure) also de-NOx takes place. When SOx is present in the flue gas, temperature needs to be higher (T > 300°C) to prevent formation of (NH4)2SO4. Dust needs to be removed from the flue gas to avoid deactivation of the catalyst.

Figure 10: catalytic reactor

These systems are compact and can easily be integrated in existing units. Also the dioxins are decomposed, meaning that no secondary treatment/disposal is required. Typical performances are > 90% removal for dioxins. Different reactor constructions are available on the market (e. g. the TSK-system). The TSK Dioxin Destruction System consists of a high active catalyst and a special type of reactor called a Lateral Flow Reactor, or LFR. The so-called Lateral Flow Reactor (LFR) is a system of gas channels and catalysis slabs. The gas then passes through a thin catalyst slab. The fact that the gas only passes through a thin catalysis layer is reason for the very low-pressure drop possible with this system. The Shell De-Dioxin System (SDDS) [18] uses a catalyst S-090 which is a high metals loaded catalyst with high surface area and porosity, resulting in its low temperature activity. Once the flue gas passes through the catalyst layers, dioxins are oxidised and destroyed by the active metals. High inlet concentrations (100 ng I-TEQ/Nm³) can be reduced to the standard of less than 0.1 ng I-TEQ/Nm³. Engineering the catalyst layers, the pressure drop can be very low. If necessary multiple modules can be installed in series to meet the emission limit.


43

A disadvantage of catalysts is their deactivation due to catalyst poisons present in the flue gases. Compounds that could poison the catalyst are: heavy metals, sulphur compound, halogens... Catalytic filters (Remedia) are a combination of a filtration and a catalytic reaction. PCDD/DF are removed and destructed at the same time. Due to the catalyst the reaction can take place at low temperatures (150 – 250 °C). Less waste (> 90% dioxin is destroyed) results in a lower disposal cost.

Figure 11: enlargement of a catalytic filter

The filters can be used in an existing filter house. The technology has been worldwide used and proven in e.g. municipal -, industrial - and medical waste incinerators. 5.2 Techniques proposed for foundry furnaces 5.2.1 Foundries with Rotary Furnace The air leaves the rotary furnace at a temperature of 1300 – 1500°C. By means of dilution with cooling air for first cooling followed by a long duct, a heat exchanger or quench the gases are cooled to below 200°C for the bag filter. The dust removal is achieved by cyclones or a baghouse. Dust removal is not always part of the air treatment. Between the furnace and the cooling a thermal oxidiser can be placed to reduce organic particles, TOC and reduce the risk of primary dioxins. Because of the high temperature in the rotary furnace most dioxin – when present - were formed by de novo during cooling. Rotary furnaces with no dust abatement use cooling air directly after the furnace to cool the gases to below 250 C. Due to the fast cooling these furnaces have no problems with de novo synthesis of dioxins. The measurements show that concentrations of dioxins are below 0.1 ng I-TEQ/Nm³. Rotary furnaces with a bag filter show that dioxins can be present (BREF Foundries). This is due to slow cooling of the gases. As the cost of a baghouse is determined by the volume of air, big volumes of cooling air will not be used. Therefore other means of cooling must be used. If this cooling is not fast enough, de novo synthesis of dioxins will occur.


44

Air treatment for dioxin removal

cooling

stack

Furnace fan

Furnace cyclonequench adsorbentinjection

bagfilter fan

stack

cyclonequench fixedbedbag filter

stack

Furnace fan

Figure 12: top: common installation without dust depuration system; intermediate and bottom:alternatives in case

of needing end of pipe techniques One of the sampled foundries in the DIOFUR Project had a low capacity furnace, a small daily production and no dust depuration system. Only due to the effect of the natural dilution at the outlet of the furnace together with the stack (vertical position and 8 m of length) it was able to reduce the temperature of the exhaust gases (from 1400-1600ºC to 200-250ºC). The rapid cooling and the used clean scrap allowed to have dioxin emission rates below 0.1 ng I-TEQ/Nm³. Therefore in installations without depuration system, if a rapid enough cooling is assured it could mean that there is no risk of dioxin reformation. When dust removal is included the system consists of:


45

Combined air cooling (with or without quench or heat exchanger) to cool below 250°C Cyclone to remove coarse dust and sparks Bag filter

The cyclone can also be placed between the injection of cooling air and the quench to achieve a high temperature dust removal. If dioxins are formed there are 3 possible schemes: better filter material, injection of adsorbent or a fixed bed. A point of attention is fire safety (for the cases that use carbon adsorbents). Therefore, for installation without dust removal dioxins will most likely not exceed the limit value of 0.1 ng I-TEQ/Nm³ because of rapid cooling with cooling air. For installations with dust removal the dioxin concentration depends on the cooling trajectory and the efficiency of dust removal since most of the dioxins are adsorbed on the dust. In some installations due to the aspiration effect from the filter there is a natural dilution comparable to the situation of a natural draught. This dilution effect could mean an enough rapid cooling preventing the dioxin formation by de novo synthesis. When this effect occurs, a cyclone could be enough before the filter, in order to eliminate the coarse particles and sparks. To totally prevent dioxin formation there should be a high dilution flow meaning that the gas volume in the filter would be also higher. Therefore, the filter dimensions would be also bigger. This is the main cost factor to consider in the design of the gas line. During the DIOFUR Project there was studied the situation for two different rotary furnaces. On the one hand, there was a foundry with only a small furnace (3 t), low production rate and no dust depuration system. On the other hand, there was a foundry with three large furnaces (2 x 24 t + 1 x 12 t) working simultaneously, high production rate, a dust depuration system consisting of a cyclone and a filter but no more gas depuration devices (as scrubbers e. g.). For none of the sampling days in both furnaces were determined emission rates above the established limit. In fact, the emission rates were still below the target value having made a simulation for the second foundry where there were taking into account the dioxins in the eliminated dust in both the cyclone and the filter (simulating the situation with no cyclone and no filter). Nevertheless, if concentrations of dioxins above the target value were currently measured, the gas cleaning system has to be checked for: temperature profile, dust deposits and for knowing if better dust removal can be achieved. If these measures are not sufficient and if a baghouse is already present, injection of adsorbent seems to be the best way to reduce the dioxin emission. For smaller installation a fixed bed system could be more cost effective. Also for this system an especially efficient dust removal by means of a bag filter is necessary. 5.2.2 Foundries with Electric Arc Furnace An EAF has a primary off gas from the furnace and secondary off gas from hoods from the ladle and the feeding system. The primary off gas contains most of the dioxins. The primary gases are cooled with the secondary gases followed cooling by a long duct, a heat exchanger or


46

quench to cool the gases to below 200°C for the bag filter. The dust removal is achieved by a baghouse. In an EAF the dust is very fine which means a highly efficient dust filtration is needed. To protect the bag filter against damage from sparks, a cyclone can be installed before the bag filter. Between the furnace and the cooling a thermal oxidiser can be placed to reduce organic particles, TOC and reduce the risk of primary dioxins. The dimensions of this oxidiser are very critical to guarantee a complete combustion (at 900°C) during al stages of the melting. As occurs in other furnaces, it is not in the furnace but in its gas line where most dioxins will be formed by de novo during cooling. Measurements show that the concentrations of dioxins in EAF could well be below 0.1 ng I-TEQ/Nm³. If this is the case no actions have to be taken. Nevertheless, the existing risk a priori is higher because the scrap charged in every process is not the same. The electric arc furnaces use as raw materials machining residues which usually have oil layers. This type of scrap can not be used in rotary furnaces or cupolas because it would be difficult to obtain good efficiency rates. Also in induction furnaces this scrap type is not used due to the explosion risk. As machining residues can only be charged in electric arc furnaces, their use in this type of furnace is not surprising and due to this fact exists and additional risk of dioxin emission in them. However, results obtained in the Diofur project have shown that it is possible to obtain emission rates below the target value.



bagfilter fan

stack

cyclonequench fixedbedbag filter

stack

Furnace fan

Figure 13: top: air treatment with injection of adsorbent; bottom: treatment with fixed bed

Low emissions of dioxins can be achieved with an air treatment consisting of:


47

Cooling air or quench to below 250°C Cyclone to remove coarse dust and sparks Bag filter

If dioxins are formed there are 3 possible schemes: better filter material, injection of adsorbent or a fixed bed. Point of attention is fire safety (for the cases that use carbon adsorbents). EAF-installations will most likely not exceed the limit value of 0.1 ng I-TEQ/Nm³. However, if high concentrations of dioxins are currently measured, the gas cleaning system has to be checked for: temperature profile, dust deposits and checking if better dust removal can be achieved. Also here the injection of adsorbent seems to be the best way to reduce the dioxin emission for bigger installations that already have a baghouse. For smaller installation a fixed bed system could be more cost effective. Also for this system an efficient dust removal by means of a bag filter is necessary. 5.2.3 Foundries with Cold Blast Cupola Some CBC use post combustion to optimize heat recovery and cleaner off gases. Post combustion of the off gases can occur in the cupola shaft or in a separate chamber. In the first case air is injected into the upper part of the cupola shaft so the gases ignite spontaneously (post combustion). Sometimes ignition or support burners are necessary. CO and residual carbonaceous material is oxidised to CO2 and water. Due to the combustion the temperature of the gases increases to ~ 900°C, meaning that no dioxins are present at this time. Dioxins will be formed during cooling (de novo) the off gases. When no post combustion occurs, dioxin formation in the primary off gases is posible. The heat in the off gases (CBC with post combustion) can be recovered in a heat exchanger. Alternatively the temperature is reduced by quenching the gases. The advantage of quenching is that the temperature drops quickly, chlorides are washed-out and the dust content is reduced, lowering the risk of dioxin formation. Finally the dust is commonly removed by a cyclone followed by a bag filter.



bagfilter fan

stack

postcombustion


48

Furnace cycloneheatexchanger

fixedbedbag filter

stack

fanpostcombustion

Figure 14: top: air treatment with injection of adsorbent; bottom: treatment with fixed bed If dioxins are present an appropriate air treatment will be necessary. Most CBC will have dust removal which consists of:

Air cooling or quench to an acceptable temperature for a cyclone. Cyclone to remove coarse dust and sparks. Bag filter.

When dioxins are formed there are 3 possible schemes: better filter material, injection of adsorbent or a fixed bed. Point of attention is fire safety (for the cases that use carbon adsorbents). In cupola furnaces all ingredients for de novo synthesis are present. Mainly, dioxins are formed during cooling of the off-gases. When a quench is used for cooling between 450°C and 250°C, the critical temperature window is passed rapidly. Also chlorides and dust are washed out. This reduces the risk of de novo synthesis. What to do when dioxins are present? Mostly a baghouse is already present, which makes the injection of adsorbent seems to be the best way to reduce the dioxin emission. For smaller installation a fixed bed system could be more cost effective. Also for this system an efficient dust removal by means of a bag filter is necessary. If high concentrations of dioxins are currently measured, the gas cleaning system has to be checked for: temperature profile, dust deposits and for finding out if better dust removal can be achieved. Economical impact Primary measures do normally not involve additional investment costs. Operational costs are restricted to the use and the costs concerning

• coke; • oxygen; • metallic raw materials.

In the project DIOFUR different types of coke have been used:

• European coke; • Chinese coke;


49

• Coke with 20 % Anthracite In the last years the coke prices vary much. In 2008 the price reaches highest level, beyond 500.00 € per ton. Some years ago Chinese coke was roughly 20 % cheaper than European coke. But today we have to recognize that there is no relevant difference. The prizes of each type of coke are all on a high level. In the project stage with parameter modifications more oxygen than in current situation was injected in the hot blast cupola. The price of oxygen (including delivery, storage etc.) is roughly 100.00 € per ton or 120.00 € per 1000 Nm³. The additional 306 Nm³ O2 per hour increases costs 36.00 € per hour. That means roughly 2.00 € per ton liquid iron. A small increase in the cost. And it has to be taken in account that there was an increase of the production rate of 12 %. Oxygen injection is a recommendable primary measure. Similar to coke prices the prices of metallic raw materials has varied in 2008 very much. In the summer 2008 prices were on a very high level, e. g. scrap beyond 500.00 € per ton. The price level varies and also the ratio of prices of the different kind of metallic raw materials. In general, it can be noted that there are differences in prices of up to 20%: scrap iron, scrap steel and especially chips are relatively cheaper, pig iron and packed sheets are relatively more expensive. Given that raw-materials account for roughly 25 % of the cost of production, a change e. g. from a cheaper to a more expensive kind of raw material can increase costs of a foundry extremely. Simply by using defined raw materials (scrap) the targeted cost increase could be exceeded and thus the economic impact would not be suitable. This would not be acceptable for a ferrous foundry, especially a SME in competition with low cost countries, e. g. in East Asia. It should be noted that the average profit margin in the European foundry industry is about 2 %. When deciding on the metallic raw material, metallurgical aspects and prices must also be taken into account. 5.2.4 Foundries with Hot Blast Cupola For the hot blast cupola, a post combustor is placed after the cupola. In some cases, the off-gases are fed directly to the combustion chamber. Others de-dust the off-gases prior to combustion. They use wet (scrubber) or dry (cyclone) de-dusting. Using the scrubber, more energy is needed in the post combustion. After combustion the heat in the off-gases (T ~ 900°C) is recuperated in first (recuperative heating) and secondary recuperators or quenching. When a wet de-dusting was used before combustion, the risk of de novo synthesis during cooling down the off-gases is strongly reduced. This means that the off-gases probably need no further treatment after cooling. In the other case, the off-gases must pass a baghouse and in some cases first a cyclone to remove sparks.


50


cyclone /scrubber

fixedbed

bagfilter

stack

Furnace fanpost

combustionheat

exchangers

quenchcyclone /scrubber

bagfilter

stack

Furnace fanpost

combustionadsorbentinjection

Figure 15: top: treatment with fixed bed; bottom: air treatment with injection of adsorbent

Typically, the air treatment for HBC consists of:

optional: de-dusting (dry or wet) the off-gases prior to post combustion Post combustion Heat exchangers and an optional quench An optional cyclone to further remove dust and reduce the risk of carryover of sparks

to the baghouse. Bag filter

When it is expected that de novo synthesis could occur, the following schemes are possible: better filter material (mostly not sufficient to meet the standard), injection of adsorbent or a fixed bed. The important points of attention are: general remarks and fire safety (for the cases that use carbon adsorbents). Primary formed dioxins are destroyed during the post combustion. The flue gases leave the combustion chamber at 900°C which means that de novo can take place during cooling. The potential of dioxin formation is strongly reduced when wet de-dusting is done before post combustion because the catalysts are removed. In the other cases, dioxins will probably be formed.


51

The best way of reducing dioxin emission in a HBC seems to be injection of adsorbent, when a baghouse is already present. Also the high volume of the flue gases is in favour for injection compared to fixed bed adsorption. If high concentrations of dioxins are currently measured, the gas cleaning system has to be checked for: temperature profile, dust deposits and for knowing if better dust removal can be achieved, as it has been included in detail in § 2.1.2. For the HBC an end-of-pipe technique shall always be necessary to meet the standard. 5.3 Test: fixed bed adsorption 5.3.1 Methodology In the DIOFUR project the participants have investigated the efficiency of a fixed bed of adsorbent (activated carbon or brown coal). The principle is to put a mobile fixed bed unit – AIRCON 2000 (see photo below) – after the baghouse.

bag-house

stack

Fan

max: 3000 m³/h

Measuringchannel

Figure 16: schema of the AIRCON-unit in the air treatment. Photo courtesy of DESOTEC.

The pilot installation has a limited flow of max 3.000 m³/h. Therefore we need to draw off a smaller flow from the main emissions. We foresee this diversion after the baghouse. At this stage dioxins are found in the gas phase and adsorbed on the remaining dust. Because the size of the dust after the filter is small and the dust concentration is limited, isokinetic diversion is not crucial. Nonetheless we will try to achieve an isokinetic diversion at nominal flow rate. 5.3.2 Adaptation conditions to furnaces


52

Considering the fixed bed system, the adsorbent is placed into a filter house. The filter is directly connected to the flue gases to be treated by means of flexible hoses and quick fasteners. Due to limitations regarding dust concentration (must be below 5 mg/Nm³), the filter will always be connected after a baghouse. It is recommended to install a continuous dust measurement in front of the fixed bed to prevent the filter from blocking. When the dust concentrations are to high, the fixed bed shall be by-passed. Actions must be taken to optimize the baghouse.

Figure 17: example of an AIRCON-unit. Photo courtesy of DESOTEC.

5.3.3 Adsorbents Different granular adsorbents can be used in the fixed bed system. Due to limitations for dust concentration towards the fixed bed, these units shall always be located after a baghouse. This implicates that the temperatures of the gases won’t be a restriction for the use of organic adsorbents like activated carbon, brow coal, peat, … . In this project the fixed bed was filled with PARADIOX 30-KBK, a brow coal based adsorbent developed and used to remove dioxins from flue gases in incineration plants. 5.3.4 Results The results of this project indicated the proper and stable working of this unit. The used adsorbent determines the effectiveness of the system for removal of PCDD/F. Based on the results obtained in the Diofur Project, a good removing efficiency is expected with PARADIOX 30-KBK. The following results were obtained in a sampling campaign on an HBC, after observing that PCDD/DF are most probable with these type of furnaces.

Table 10: results for tests using fixed bed technology

Furnace HBC

Type of measurement PILOT PLANT: FIXED BED

Date 08/11/07 09/11/07

Stack (non treatment) ng I-TEQ/Nm³ 0.1224 0.0866


53

After pilot plant ng I-TEQ/Nm³ 0.0005 0.0001

% removal > 99.5 > 99.5

Interruptions None Yes

Injected O2 576 Nm³/h 575 Nm³/h

Change of coke No No

Process disruptions No No operation in 3 of 6 hours measured

In this case anthracite produces more ultrafine particles which go through the bag filter with trapped dioxins within them (based on a previous measurement in these conditions).

For the tests a mobile fixed bed unit – AIRCON 2000 – was placed on a diverted air flow after the baghouse. The pilot system consists of connection ducts, a ventilator, measuring ducts and the AIRCON 2000 (fixed bed of active carbon). Measurements of dioxins were made at the stack (non diverted air flow) and after the pilot plant in order to compare the dioxin emission rate with and without end of pipe technique. During the trial 20% of the coke was replaced by anthracite. To reduce the potential memory effect, the cupola was operated in those conditions for the 10 days preceding the campaign. Good scientific results were obtained for this end of pipe system. However it has a limitation; a low dust concentration is necessary to prevent the fixed bed from blocking. In a full scale unit precautions must be taken. Via a continuous dust measurement, the fixed bed will automatically be by-passed when high dust concentrations are indicated. Therefore, based on these results and taking into account that this technology is successfully applied in other industries the fixed bed technology can technologically be introduced in foundry furnaces with dioxin emission rates above the limit value. Main point of attention is the baghouse. The maximum dust concentration towards the fixed bed system is limited at 5 mg/Dry Nm³ to prevent blocking of the filter. 5.3.5 Additional beneficial effects (reduction of other contaminants) The adsorbent (activated carbon or brown coal) is not selective, meaning that besides PCDD/DF also other volatile organic compounds (PAH’s), odours, heavy metals etc. can be removed from the flue gases. This will depend on the characteristics of the compound, its concentration and the temperature of the flue gases. Since the temperature of the flue gases at the filter will be below 150°C, only compounds with high boiling point (e.g. PAHs) will be adsorbed. 5.3.6 Economic impact For a fixed bed solution for a foundry with a nominal air flow of 35000 Nm³/h, it has been proposed the installation of 2 filters parallel, after the baghouse. The construction is an isolated mobile filter with activated carbon AIRCON®-H XL-I. In this approximate economical calculation a flow of 35000 Nm³/h was obtained (situation like the HBC of the project). It is proposed to use the activated carbon containing lignite coke PARADIOX 30-KBK. The mobile fixed bed filters are used already to adsorb the PCDD/PCDF, mercury, lead and other heavy metals. In this calculation piping, the valves of safety, the by-pass… are not included.


54

Figure 18: example of 6 AIRCON-units parallel. Photo courtesy of DESOTEC.

AIRCON® - Horizontal XL – Insulated (prices of 2009)

• Cost for rent: 42.00 €/filter per calendar day • Quantity of activated carbon: approximately 8000 kg for a 50 cm thickness • Delivery: 1450€/filter

(Based on costs for transport between the foundry and deposit of waste treatment) • Pressure drop: is calculated for filter AIRCON, 50 cm of an activated carbon with a

granulometry comparable to PARADIOX 30-KBK and an operational flow of 26.150 m³/h to approximately 1500 Pa.

PARADIOX 30-KBK

• cost: 935.00 €/ton (recovery/elimination excluded)

For an assessment the following factors have to be taken into account:

• Durability of the fixed bed material: This depends on different parameters (e.g. metal

compounds in the flue gases) and is difficult to estimate at this time. Tests must illustrate

the point of “breakthrough” of the fixed bed. Roughly estimated it could take 1 year till

breakthrough, however depending strongly on the composition of the flue gases.

• Costs for additional equipment: piping, valves of safety etc.; This depends on the amount of

piping and requested automation. Roughly the cost is estimated on 40.000 - 50.000 EUR.

• Costs for treatment and/or deposition of used adsorbents; This saturated adsorbent will not

be regenerated. Costs for treatment/elimination depend mainly on the concentration of Hg

on the saturated adsorbent. An indication of the elimination cost is 1,4 EUR/kg, when

< 1500 ppm Hg. For higher concentrations up to 5000 ppm Hg, costs can increase up to 2,0

- 2,5 EUR/kg.


55

• Costs for energy and maintenance; To overcome the pressure drop of ~ 2000 Pa an extra

fan has to be installed. The power is estimated at 30 - 40 kW.

Possibly: additional costs for a new dust filter system, that permanently ensures <5 mg dust / Nm3. 5.4 Test: injection of adsorbent 5.4.1 Methodology Another way of bringing the dioxins in contact with an adsorbent is the injection system. The project partner there were Typhoon performed actions necessary to evaluate the efficiency of injection of an adsorbent into the flue gases. The principle is to inject the adsorbent after a cyclone (to avoid sparks) and just in front of the baghouse. From practical experience it is known that the most dioxins are adsorbed on the adsorbent in the filter cake and not during contact in the channels before the bag filter. Also just in front of the baghouse the temperature of the flue gases is reduced to below 180°C. Circulation and re-use of the mixture filter dust and adsorbent could be considered. Whether it’s useful will depend on the amount of dust collected in the bag. In most circumstances re-use is not possible because of high dust loads to the baghouse. The amount of adsorbent will depend on the type used. For carbon based adsorbents, we foresee 50 – 200 mg/Nm³. The injection system consists of a hopper with a dosing screw which regulated the amount of adsorbent dosed to the gases. To facilitate the installation, the dosing system will be placed at ground level. The ducts however are mostly located on the roof of the foundry. By means of a small blower and an abrasion resistant flexible pipe the adsorbent is transported into the flue emissions at the roof. This avoids complex constructions at the roof and makes it easier to exchange the hopper when the adsorbent is used up.


56

silo for adsorbent + transport screw

transport screw + injection system

Figure 19: Injection system. Photo courtesy of TYPHOON. Point of attention is possible clogging of the adsorbent due to humidity. This can be avoided by placing the bag inside or installing a closed hopper. Also bridge formation in the hopper can be a problem. This can be avoided by placing a vibrating motor on the dosing unit. When an organic adsorbent is used, some safety aspects must be considered depending on the type of organic adsorbent. Some organic adsorbent are more prone to ignite than others. To protect the sleeves from pinholes an effective spark removal is needed. With the use of activated carbon the spark removal is still more important than normal. Mostly accumulation of dust in the bottom of the baghouse must be prevented because of fire risk due exothermal reactions in the dust. When using the injection technique, the filtration afterwards is very important. Filtration must be very effective. 5.4.2 Adaptation conditions to furnaces As described above, the injection system exists of:

1. Station for unloading the Big Bag 2. Screw of transport towards volumetric batcher 3. Volumetric batcher 4. System of injection containing pneumatic transport 5. Electrical equipment box


57

Figure 20: View of the installation. Photo courtesy of TYPHOON.

The only adaptation which has to be made is the connection with the pipe for the flue gases towards the baghouse. Point of attention is the place of injection so that optimal mixing in the flue gases is realized. If the mixing is not optimal not everywhere in the baghouse a layer of adsorbent will be on the sleeves. To eliminate the risk of fire, a good spark remover must be installed in front of the baghouse. For new treatment systems this shall not be a problem. Older units which are adjusted with an injection system, must pay attention to this point. 5.4.3 Adsorbents Also for injection different types of adsorbent may be used. Literature reveals application of organic and inorganic adsorbents for the removal of dioxins in several industries. Among the inorganic adsorbents, zeolites are natural, microporous, aluminosilicate minerals. In this project SIALSORB NM050 was used as adsorbent. Also the product DIOXORB 866 - an inorganic product based on lime and clay – was examined. Most commonly used adsorbent are activated carbon or brown coal. Both have a very big internal surface (~500 m2/g for PARADIOX 70) which makes them ideal for adsorption. Literature also reports mixtures of zeolite, carbon and inert materials for the removal of dioxins. 5.4.4 Results

1

23

5

4


58

In the DIOFUR-project four adsorbents were examined at the HBC. Two inorganic adsorbents - SIALSORB NM050 and DIOXORB 866 – showed bad results for dioxin removal. This was not expected since both products are developed and used especially for the removal of dioxins from flue gases, even in the foundry sector. Based on our findings the adsorption capacity of these adsorbents is less than for organic adsorbents. In some cases they can work but in tests in the DIOFUR project showed no proof for sufficient additional dioxin removal. The emissions were above 0.1 ng iTEQ/Nm³. Organic based adsorbents PARADIOX 30 and 70 were also tested at the HBC. Based on the results of the fixed bed, PARADIOX 30 is effective in the removal of dioxins. However due to safety concerns from the foundry, no further tests were performed with PARADIOX 30 for injection. PARADIOX 70 was developed during the course of this project. It’s a coke based adsorbent with higher ash content, making it more suitable for high(er) temperature applications, without concern for safety aspects. Burning behaviour at 20°C and 100°C was reported as “brief ignition and rapid extinction”, the ignition temperature was 800°C. Also the PARADIOX 70 was not glowing up to 450°C. Results showed that when using this adsorbent the results for PCDD/DF were below the standard of 0.1 ng I-TEQ/Nm³.

Table 11: results for tests using injection of adsorbent technology

Furnace HBC

Type of measurement PILOT PLANT: ADSORBENT INJECTION

Date 15/04/2009 16/04/2009

At the stack ng I-TEQ/Nm³ 0.0592 0.0163

Interruptions None None

Injected O2 576 Nm³/h 575 Nm³/h

Change of coke No No

Process disruptions Reloading of the furnace between 9h25 – 10h25

Stop of the furnace between 10h35-11h45

Based on the performance of the pilot installation, we conclude that the adsorbent injection system works reliable in the foundry sector. The selection of a suitable adsorbent is however not obvious. Based on our results the organic adsorbents favour for dioxin removal but safety aspects must be considered. For this reason new, safer organic products (with high ash content) may offer the solution. Main point of attention is effective dust removal in the baghouse. The effectiveness of the baghouse shall determine the final result of dioxin removal. When PCDD/DF are adsorbed on the dust/adsorbent, they have to be eliminated in the baghouse. 5.4.5 Additional effects (reduction of other contaminants) The same benefits as mentioned in § 5.3.5 can be expected when carbon based adsorbent is used.


59

5.4.4 Economic impact The installation cost for a unit (prizes of 2009) is estimated at 35000 – 45000 EUR. This is the cost for a unit as described in § 5.4.2. Without the Big Bag discharger (this could be excluded for smaller foundries) the estimated cost is 25000 – 35000 EUR. Not included and at cost of the client:

• Work of civil engineering like the concrete bases, mural drilling, … • Protection of the material (electric instruments, detections, etc…) • The access for the crane and the elevators (nacelles) on the building site • The cable of power for ours electrical equipment box

For an assessment the following factors have to be taken into account:

• The costs related to the four points above; typically: Additional cost could be expected max.

100%: 30% installation cost, 10% instrumentation/control, 30% piping and 30% civil work.

• Costs for treatment and/or deposition of dust with a higher content of dioxin, PAHs, heavy

metals etc.; This will be the same as current situation (~200 - 270 EUR/ton). Additional cost

will be for the amount of injected adsorbent.

• Costs for energy and maintenance; Estimated power for the unit is 3.5 – 5.0 kW.

• Possibly: additional costs for a new dust filter system, that permanently ensure <5 mg dust

/ Nm3.

Detailed Information is given in a comparison between PAC Injection or fixed bed filtration for

the foundry FIDAY GESTION (Table 24).

Table 9: Secondary measures for reducing dioxins emission at FIDAY GESTION OPERATIONAL PARAMETERS FOR HOT BLAST CUPOLA FIDAY GESTION T [°C] at stack 155 - 165 T [°C] after exchangers 225 - 250 Dust [mg/Nm³] at stack <10 Dust [mg/Nm³] after exchangers >10 PCDD/PCDF [ng I-TEQ/Nm³] after exchangers >5 PCDD/PCDF [ng I-TEQ/Nm³] in stack <5 Flow [Nm³/h] at stack 35,000 - 40,000 ~ 60,000 m³/h Production [t good castings/year] 45,000 INJECTION OF PAC PARADIOX 70 (yearly basis - 8000 hours) - € Injection system + piping 50,000.00 PARADIOX 70 (150 mg/m³ - 72 tons/year; 750 €/ton) 54,000.00 Additional landfill cost (270 €/ton) 19,440.00 123,440.00 INVESTMENT (years for investment amortization = 5 => 10,000.00 €/year) 50,000.00OPERATIONAL COST (without energy costs) 73,440.00 Additional costs: 1.85 €/t; 0.19 cent€/Kg


60

LOCATION FIXED BED FILTRATION PARADIOX 30-KBK (yearly basis - 8000 hours) - EURO Location 2xIsolated Aircon-H + transportation 31,560.00 Piping + ventilators + manual valves (budgetted) 40,000.00 PARADIOX 30-KBK (16 tons/year, 935 €/t) 14,960.00 Additional landfill cost (depending from Hg-concentration <1500ppm; 24,000.00 costs for landfill ~ 1.5 €/kg) 110,520.00 INVESTMENT (years for investment amortization = 5 => 8,000.00 €/year) 40,000.00OPERATIONAL COST (without energy costs) 70,520.00 Additional costs: 1.74 €/t; 0.17 cent€/Kg PURCHASE FIXED BED FILTRATION PARADIOX 30-KBK (yearly basis - 8000 hours) - EURO purchase 2xIsolated FILTRAFIX-H + transportation 80,650.00 Piping + ventilators + manual valves (budgetted) 40,000.00 PARADIOX 30-KBK (16 tons/year, 935€/t) 14,960.00 Additional landfill cost (depending from Hg-concentration) <1500ppm 24,000.00 costs for landfill ~ 1.5 €/kg) 159,610.00 INVESTMENT (years for investment amortization = 5 => 24,130.00 €/year) 120,650.00OPERATIONAL COST (without energy costs) 38,960.00 Additional costs: 1.40 €/t; 0.14 cent€/Kg The operational cost for PAC is higher - considered 1 year estimated lifetime (¡Error! No se encuentra el origen de la referencia.).

ATEX-zone: the unit will be out of these zones. The amount of adsorbent used for the removal of dioxins will be negligible compared to the total dust concentration. This means that the disposal cost of the filter dust will not increase significantly. Extra costs rise mainly from used adsorbent and daily follow up activities.


61

6. Bibliography [1] EUROPEAN COMMISSION (2004) – Reference Document on Best Available Techniques in

the Smitheries and Foundry Industry, 365 p. http://eippcb.jrc.es. [2] UNEP (2003). "Formation of PCDD and PCDF - an overview" Stockholm convention on

persistent organic pollutants, expert group on BAT and BEP, Research Triangle Park (USA).

[3] PANDELOVA M., LENOIR D., KETTRUP A., SCHRAMM K.W., 2005 – Primary measures for

reduction of PCDD/F in co-combustion of lignite coal and waste : effect of various inhibitors. Environ. Sci. Technol., 39, 3345-3350.

[4] LEBLANC O., PONS A., 2004 – Résultats d’études technico-économiques pour le traitement

des dioxines dans la sidérurgie. Journées Techniques Nationales, Dioxines & Polluants Organiques Persistants. 10 – 11 mars. ADEME Ed.

[5] SCHFIELD N., FISHER R., ANDERSON D.R., 2004 – Environmental challenges for the iron

and steelmaking process. Ironmaking and Steelmaking, 31, 6, 428-431. [6] RUOKOJÄRVI P., ASAKAINEN A., RUUSKANEN J. 2001 - Urea as a PCDD/F inhibitor in

municipal waste incineration. Air & Waste Management Association, 51, 422-431. [7] G.S.F., 2004 – Intelligent use of waste material reduces formation of dioxins. Vigie

Environnement, 97, 32 [8] VISEZ N., 2005 – Etude des processus hétérogène de formation et de destruction des

dioxines. Dissertation, Univ. Lille; U.F.R. Chimie, n°3724. [9] HELL K., STIEGLITZ L., DINJUS E., SEGERS P., BUEKENS A., 2000 – Inhibition of PCDD/F

“de novo” formation by addition of basic compounds to dust from metallurgical plants : experimental results and discussion of inhibition mechanisms. Organohalogen Compounds, 46, 252 – 256.

[10] VITO (2001). "Beste Beschikbare Technieken voor de Gieterijen", Vito, ISBN 90 382 0312

8. [11] MAASKANT, O.L. (2001) The Shell System for NOx Removal and Dioxin Destruction from

Incineration Flue Gas. 3rd International Symposium on Incineration and Flue Gas Treatment Technologies. 2-4 July, Brussels.

[12] Remedia catalytic filters web page, www.gore.com [13] WEBER, R., PLINKE, M. AND XU, Z. Dioxin destruction efficiency of catalytic filters

evaluation in laboratory and comparison to field operation. [14] Database on Japanese advanced environmental equipment.

www.gec.jp/JSIM_DATA/index.html#AIR [15] Typhoon, www.typhoon.be


62

[16] Desotec, www.desotec.be [17] JIN-SOO CHOI, JEONG-EUN OH AND YOON-SEOK CHANG. Reduction of PCDD/Fs

emissions in the flue gas of full-scale incinerators with activated carbon injection-dust removal systems in Korea.

[18] LEMMENS, B., ELSLANDER, H., CEULEMANS, J., PEYS, K., VAN ROMPAEY, H. AND

HUYBRECHTS, D. (2004) Gids luchtzuiveringstechnielken. Academia Press, Gent.


63

Part 2: Good practice manual 1. Introduction

CHARGE SAND EARTH PAINT OIL PLASTIC OTHER METALS

BC (%)

EAF (%)

RF (%)

IF (%)

Pig iron x 0-10 0-10 40-60 0-10 Foundry returns x 30-50 40-60 30-50 30-60

Scrap iron x x x x x 10-30 0-30 10-20 0-20

Packed sheets x 0-30 0-30 -* 40-60

Scrap Steel x x x 10-30 20-40 0-20 0-20

Chips x 0-10 0-10 -* -*

*Not usual

In the previous table there are indicated an estimation of the type of pollutants currently contained in each type of charge as well as one theoretical charge configuration of each type of furnace. Therefore, based on the knowledge that some pollutants are more relevant than others, it can be estimated which are the types of furnace with the higher dioxin emissions. 2. Block diagram: a previous guide to know if actions are required This block diagram helps to identify the actions to take in the foundry. The method is simple: consists on answering the questions and trading the individual path of the foundry. This path will head to a final recommendation. There are three possibilities: no actions required, a dioxin abatement system is required or only changes in the process (cyclone, bag filter…). Therefore the following diagram makes possible a first identification of the degree of the problem:


64

BLOCK DIAGRAM

YES

NO ACTIONS NEEDED (Until next regulated test)

YES

NO

YES

NO NO

YES Can you minimize them following the

technical sheets?

NO

YES

Is your filter, cyclone in good

condition?

YES

NO

Is the furnace a CUPOLA?

YES

NO

Is the furnace an EAF?

NO

Is the furnace a RF?

Do raw materials include oils, paintings or

plastics?

YES Is the furnace an IF?

Are PCDD/Fs above the Standard?

NO ACTIONS REQUIRED

NEW MEASUREMENTS

REQUIRED *Because of previous indications

AN ABATEMENT SYSTEM REQUIRED

TAKE ACTIONS BASED ON CORRESPONDING CHECK LIST and

TECHNICAL SHEETS

YES

NO Are PCDD/Fs above the Standard?

NO

YES

Is the cupola a CBC?

NEW MEASUREMENTS


NO ACTIONS REQUIRED

YES

NEW MEASUREMENTS


AN ABATEMENT SYSTEM REQUIRED

TAKE ACTIONS BASED ON CORRESPONDING CHECK LIST and

TECHNICAL SHEETS

NO

Dust removal before post combustion?

YES

DO IT!

NO

Are PCDD/Fs above the Standard?

TAKE CARE OF IT!

If there is an indication, measure PCDD/F emission rate (According to related Administration

Regulations) *See also BREF document

FILTER MAINTENANCE REQUIRED

*Consider change of fuel


65

3. Task guidance sheets: finding recommendations for your furnace and raw materials The second step designed for the minimization and/or elimination of dioxin emissions in foundries are the task guidance sheets. They are a series of technical sheets where are analyzed on the one hand each type of raw material used in foundries and on the other hand, the foundry furnaces. Each task guidance sheet consists of one part referring to the characteristics and another one of recommendations. 3.1 Task guidance sheets for raw material


1

This sheet is part of the Good Practice Guide on prevention of dioxin formation in iron foundries and it is for public use.

This guide offers advice, recommendations and individual data to perform correctly use, dosage and selection of the raw materials for iron casting, in order to minimise dioxin formation. All the data included in this guide have been obtained from the Diofur Project, carried out under EU’s Sixth Framework Programme.

RRAAWW MMAATTEERRIIAALL:: PPIIGG IIRROONN

TASK GUIDANCE SHEET 1

CLASSIFICATION OF THE MATERIAL ACCORDING TO DIOXIN FORMATION Scale (0-5): 5 0=MOST DANGEROUS 5= SAFEST

COMPOSITION

C Si Mn P S 3.5-4.5 0-3 0.1-1 <0.15 <0.1

High carbon percentage can be observed that could be one of the carbon sources for dioxins’ formation.

POLLUTANT CONTENT

Usually has no high pollutant content. Can have traces of pollutants in layers of soil and/or traces of oxides that got there due to storage in places where other raw materials have been before.

APPEAREANCE (usual cleanliness grade)

Usually ingots of 5-20 kg.

SIZE (Bulk density)

Has high density because it is a compact material. Due to its shape easy to pile up so no large storing place necessary.

PRICE

Is usually an expensive raw material because it is iron of the first melting; has a good quality.

EFFICIENCY

Has a very high efficiency due to its high density .

MANIPULATION

Special manipulation system not required because it is moved by an electric magnet.

ADVICES

Store in places only used for pig iron ingots to avoid addition of other pollutants. Also, due to its small surface area it is not easily polluted.

SUPPLIERS

It is recommended to work with reliable suppliers with enough traceability in their products and tidy installations to avoid pollutant transfer from usually polluted materials to usually no polluted ones. In other words, suppliers should not mix materials likely to form dioxins with the raw materials.

RREE C

C OO

MMMM

EE NN

DDAA

TT II OO

NNSS

CC

HHAA

RRAA

CCTT E

E RRII SS

TT II CC

SS


2



RRAAWW MMAATTEERRIIAALLSS:: FFOOUUNNDDRRYY RREETTUURRNNSS



COMPOSITION

Type C Si Mn P S Iron return 2.5-3.5 1.0-3.0 0.15-1.0 <0.15 <0.1

Steel return 0.1-0.3 0.2-0.3 0.4-0.6 <0.03 <0.02

Composition of the returns depends on process but average values have been included. High carbon content.

POLLUTANT CONTENT

There are two possibilities: Returns of the foundry: Material that has not left the foundry so at the most it can contain sand. If the

sand is not burned enough it can contain organic carbon being a source for dioxin formation. Consists basically of pouring systems and rejected castings.

External returns (currently extremely low quantities used): It has been rejected and returned by the customer and it can return painted, partially machined and/or treated by zinc. Paintings or cooling fluids (oils) can be the source for dioxin formation.


May vary, depends on the kind of pieces and parts cast in the foundry.

SIZE (Bulk density)

Has low density because they are pieces with different geometry. Breaking them strongly recommended in order to increase the density.

PRICE

Usually not considered as a cost in raw materials. They currently mean good quality.

EFFICIENCY

Oxide and impurity free, therefore has high efficiency.

MANIPULATION

Manipulation with an electric magnet can be complicated due to the variety of geometries among the pieces. This matter can be solved by machines which reduce the size and compact the material by breaking.

ADVICES

The only precaution is not to concentrate the external returns in order to keep as low as possible the chlorine level.

SUPPLIERS

It is recommended to work with reliable suppliers with enough traceability in their products and tidy installations to avoid pollutant transfer from usually polluted materials to usually no polluted ones. In other words, suppliers should not mix materials likely to form dioxins with the raw materials.

RREE C

C OO

MMMM

EE NN

DDAA

TT II OO

NNSS

CC

HHAA

RRAA

CCTT E

E RRII SS

TT II CC

SS


3


COMPOSITION

C Si Mn P S 2.5-3.5 1.0-3.0 0.15-1.0 <0.15 <0.1

POLLUTANT CONTENT

Pollutant content depends on the origin but can contain: paintings, oil and plastics which can be a potential source of chlorine, other metals or even soil. Recommended to buy the cleanest possible scrap.


Can have a wide range of appearances depending on origin.

SIZE (Bulk density)

Scraps with the highest possible density are recommended to obtain a high efficiency rate. Size depends directly on the used furnace.

PRICE

Raw material of medium price range. Normally the cleanliness has an influence on the scrap.

EFFICIENCY

Medium to high efficiency depending on scrap quality.



MANIPULATION


ADVICES

Purchasing clean scrap eliminates potential dioxin sources before process.

SUPPLIERS

It is recommended to work with reliable suppliers with enough traceability in their products and tidy installations to avoid the pollutant transfer from usually polluted materials to usually no polluted ones. In other words, suppliers should not mix materials likely to form dioxins with the raw materials. RREE C

C OO

MMMM

EE NN

DDAA

TT II OO

NNSS

CC

HHAA

RRAA

CCTT E

E RRII SS

TT II CC

SS

RRAAWW MMAATTEERRIIAALLSS:: SSCCRRAAPP IIRROONN



4



RRAAWW MMAATTEERRIIAALLSS:: NNEEWW PPRROODDUUCCTTIIOONN CCOOMMPPRREESSSSEEDD

SSTTEEEELL SSHHEEEETT BBAALLEESS** AAllssoo ccaalllleedd NNEEWW AARRIISSIINNGGSS oorr PPAACCKKEEDD SSHHEEEETTSS



COMPOSITION

C Si Mn P S 0.1-0.3 0.2-0.3 0.4-0.6 <0.03 <0.02

POLLUTANT CONTENT

Has usually no high pollutant content. Can have some oil (very low quantities because the sheets containing oil come from hobbing). Usually not painted. Due to the low pollutant in combination with the low carbon content not a raw material with a high risk of dioxin formation.


Must be dry to avoid accident risk. Appearance as means to ensure safety at work.

SIZE (Bulk density)

Of average density because it is a compacted material. Size tends to be standard.

PRICE

Has usually a medium-high price because of good quality.

EFFICIENCY

Good efficiency due to improved density.

MANIPULATION


ADVICES

Keep dry to avoid accidents during the process.

SUPPLIERS


C OO

MMMM

EE NN

DDAA

TT II OO

NNSS

CC

HHAA

RRAA

CCTT E

E RRII SS

TT II CC

SS


5



RRAAWW MMAATTEERRIIAALLSS:: SSCCRRAAPP SSTTEEEELL


CLASSIFICATION OF THE MATERIAL ACCORDING TO DIOXIN FORMATION Scale (0-5): 4 0=MOST DANGEROUS 5= SAFEST COMPOSITION

C Si Mn P S 0.1-0.3 0.2-0.3 0.4-0.6 <0.03 <0.02

POLLUTANT CONTENT

Usually contains paintings because there sometimes are pads with antioxidant primers. Therefore, there is a risk of dioxin formation but not a potential risk.


Have usually a very clean appearance because they have been previously cut using heat so that any possible type of impurity will have been vaporised.

SIZE (Bulk density)

Of average density because it is a compacted material. Size can vary depending on source.

PRICE

Has usually medium price so that a percentage of this raw material is used as charge in every kind of furnace.

EFFICIENCY

Average efficiency due to its middle density.

MANIPULATION


ADVICES

No special precautions required. Can be dried easily if delivered in wet condition by supplier.

SUPPLIERS


C OO

MMMM

EE NN

DDAA

TT II OO

NNSS

CC

HHAA

RRAA

CCTT E

E RRII SS

TT II CC

SS


6



RRAAWW MMAATTEERRIIAALLSS:: CCHHIIPPSS



COMPOSITION

Type C Si Mn P S Iron chips 2.5-3.5 1.0-3.0 0.15-1.0 <0.15 <0.1 Steel chips 0.1-0.3 0.2-0.3 0.4-0.6 <0.03 <0.02

Currently the same material as returns. Therefore can be one of the carbon sources for dioxins’ formation.

POLLUTANT CONTENT

Are made of swarfs cleared in the machining operation so have a high share (up to 3%) of oil, which can be a potential chlorine source for dioxin formation. Are a recycled product currently used in foundries with machining shop.


Oily appearance due to its high oil content..

SIZE (Bulk density)

Has good enough density, especially chips of iron scrap. Recommended to make chips from the original swarfs to increase density a little and facilitate manipulation. Size can be small (pellets) or medium (chips).

PRICE

Like returns, a very inexpensive raw material because it is waste of the machining operation.

EFFICIENCY

Low efficiency due to low mechanical strength . Efficiency of chips higher than that of swarfs.

MANIPULATION

Manipulation of chips complicated because of breaking during transport.

ADVICES

Manipulation by electric magnet difficult if oil content very high . Recommended:

Increase bulk density if chips are bought. Compact swarfs before making chips, for swarfs cleared in machining of own plant.

SUPPLIERS

It is recommended to work with reliable suppliers with enough traceability in their products and tidy installations to avoid the pollutant transfer from usually polluted materials to usually no polluted ones. In other words, suppliers should not mix materials likely to form dioxins with the raw materials.

RREE C

C OO

MMMM

EE NN

DDAA

TT II OO

NNSS

CC

HHAA

RRAA

CCTT E

E RRII SS

TT II CC

SS


7

3.2 Task guidance sheets for furnaces


8


TYPICAL CHARGE

Pig iron Returns Iron Scrap Packed sheet Steel scrap Chips 50% 40% 5% - 5% -

90% of scrap are pig iron and returns which have no high risk of dioxin formation. For more details see task guidance sheets of each raw material. DIOXIN EMISSION IN ROTARY FURNACES Because of the furnace characteristics normally clean raw materials are used and therefore there is no high emission rate of dioxins. However, as dioxins are semivolatile compounds, they can be emitted as gas or particles so sometimes a depuration system is required not to exceed the emission limit value as it is the case for Foundry 1. A bigger furnace size and thereby, a higher production rate does not involve a higher emission rate. Gas line and process design are the main influence on dioxin formation (see recommendations in the following lines).

FOUNDRIES Dioxin emission rate ng I-TEQ/Nm³

WITHOUT DEPURATION SYSTEM 0.02 - 0.10

WITH DEPURATION SYSTEM ≈0.002

BAT associated emission level 0.1 ng I-TEQ/Nm³

FFUURRNNAACCEESS:: RROOTTAARRYY FFUURRNNAACCEE


CCHH

AARR

AACC

TT EE RR

II SSTT I

I CCSS

DIOXIN CAPTURE SYSTEM According to current legislation, specific dioxin capture systems are not required because low emission rates have been measured. PROCESS DESIGN: GAS LINE

It is very important that gas lines should be designed to avoid dioxin formation by ‘de novo’ synthesis (occurs when gases reach temperatures between 250-450ºC with low cooling rate). However, this phenomenon is almost impossible to avoid in cases where interchangers or long lines are required. In these cases a special waste treatment and handling regime must be followed. If possible, exhausts quenching is recommended.

PROCESS PARAMETERS CONTROL

Make sure that charged raw materials are clean. RREE C

C OO

MMMM

EE NN

DDAA

TT II OO

NNSS


This guide offers advice, recommendations and individual data to perform correctly melting process in furnaces used in iron foundries, in order to minimise dioxin formation. All the data included in this guide have been obtained from the Diofur Project, carried out under EU’s Sixth Framework Programme.


9


TYPICAL CHARGE

Pig iron Returns Iron Scrap Packed sheet Steel scrap Chips 15% 40% 15% 0% 20% 10%

Iron scrap and chips containing oil and paintings are a chlorine source for the dioxins formation. For more details see the task guidance sheets of each raw material. DIOXIN EMISSION IN CUPOLA FURNACES This type of furnace has a risk of dioxin formation. The high rate of charged chips, iron scrap and steel scrap provide oil and paintings, which are the major source of chlorine for dioxin formation. For blast cupolas it is absolutely necessary to use a depuration system in order to capture dioxins condensed and adsorbed on particles. A bigger furnace size and thereby, a higher production rate does not involve a higher emission rate. The gas line and process design are the main influence on dioxin formation (see recommendations in the following lines).


WITHOUT DEPURATION SYSTEM + NO CONTROL OF PROCESS PARAMETERS 4.0 - 7.0

WITH DEPURATION SYSTEM + CONTROL OF PROCESS PARAMETERS 0.0010 - 0.5000

WITH DEPURATION SYSTEM + CONTROL OF PROCESS PARAMETERS + ABATEMENT SYSTEM 0.0001 - 0.0600*

*See explanatory note at the end of the document


FFUURRNNAACCEESS:: BBLLAASSTT CCUUPPOOLLAA


CCHH

AARR

AACC

TT EE RR

II SSTT I

I CCSS

DIOXIN CAPTURE SYSTEM According to current legislation, in order to assure emission rates below the established limit it can be necessary to install dioxin capture systems. It has to be studied if it is possible to control the situation by controlling the process and its parameters. PROCESS DESIGN: GAS LINE

It is very important that gas lines should be designed to avoid dioxin formation by ‘de novo’ synthesis (occurs when gases reach temperatures between 250-450 ºC with low cooling rate). However, this phenomenon is almost impossible to avoid in cases where exchangers or long lines are required. In these cases a special waste treatment and handling regime must be followed.. If possible, exhausts quenching is recommended.

PROCESS PARAMETERS CONTROL

The main parameters to check are: the use of raw materials, a steady and continuous process, a high enough temperature at the stack to allow the spraying of water. However the control of these parameters may not be enough to assure an emission rate below the target. For those cases an end of pipe solution would be necessary.

RREE C

C OO

MMMM

EE NN

DDAA

TT II OO

NNSS




10


TYPICAL CHARGE

Pig iron Returns Iron Scrap Packed sheet Steel scrap Chips 15% 50% - 30% 5% -

Due to the limited possibility of alloying, the induction furnace charge generally consists of clean material. For more details see the task guidance sheets of each raw material. DIOXIN EMISSION IN INDUCTION FURNACES As well as using clean charge, the furnace does not produce a ducted high temperature flue-gas stream that cools down slowly. Therefore, there are no potential sources of chlorine and de-novo synthesis is not likely to occur. The BREF document on Smitheries and Foundries mentions that induction furnaces show a low risk of dioxin formation and therefore emission rates from this type of furnace have not been measured in the Diofur project (main data source of this Good Practice Book). The following table includes data from the BREF document:

Dioxin emission rate ng I-TEQ/Nm³

Foundries WITH DEPURATION SYSTEM 0.003 - 0.010

Nm³ BAT associated emission level 0.1 ng I-TEQ/Nm³

FFUURRNNAACCEESS:: IINNDDUUCCTTIIOONN FFUURRNNAACCEE


CCHH

AARR

AACC

TT EE RR

II SSTT I

I CCSS

DIOXIN CAPTURE SYSTEM According to current legislation, specific dioxin capture systems are not required because low emission rates have been measured. PROCESS DESIGN: GAS LINE In this type of furnaces, a ducted high temperature flue-gas stream that cools down slowly is not produced so the process design does not affect dioxin formation. However, an uncommon design involving long ducts and other parameters where slow cooling processes could occur should be avoided. PROCESS PARAMETERS

Process parameter values related to a normal furnace working day ensure emissions below the

limit. RREE C

C OO

MMMM

EE NN

DDAA

TT II OO

NNSS




11


TYPICAL CHARGE

Pig iron Returns Iron Scrap Packed sheet Steel scrap Chips - 50% 10% 5% 20% 10%

The risk of dioxin formation is due to the use of charge polluted by oil and paintings mainly iron scrap, steel scrap and chips. However, this type of raw materials represents about 40% so the risk is lower than that for blast furnaces. DIOXIN EMISSION IN ELECTRIC ARC FURNACES This type of furnace represents a mid-low emission rate of dioxins. However, as dioxins are semivolatile compounds they can be emitted as gas or particles so sometimes a depuration system is required not to exceed the emission limit value. A bigger furnace size and thereby, a higher production rate does not involve a higher emission rate. The gas line and process design are the main influence on dioxin formation (see recommendations in the following lines).


WITHOUT DEPURATION SYSTEM 0.035 - 0.800

WITH DEPURATION SYSTEM 0.002 - 0.040


FFUURRNNAACCEESS:: EELLEECCTTRRIICC AARRCC FFUURRNNAACCEE


CCHH

AARR

AACC

TT EE RR

II SSTT I

I CCSS

DIOXIN CAPTURE SYSTEM According to current legislation, specific dioxin capture systems are not required because low emission rates have been measured. PROCESS DESIGN: GAS LINE It is very important that gas lines should be designed to avoid dioxin formation by ‘de novo’ synthesis (occurs when gases reach temperatures between 250-450 ºC with low cooling rate). However, this phenomenon is almost impossible to avoid in cases where interchangers or long lines are required. a special waste treatment and handling regime must be followed. . (Comment: Please check, if this is still the intended meaning). If possible, exhausts quenching is recommended. PROCESS PARAMETERS CONTROL

Make sure that charged raw materials are clean .

*If EAF is used to melt iron, the recommendations and advices were the same as the given in IF sheet.

RREE C

C OO

MMMM

EE NN

DDAA

TT II OO

NNSS




12

4. Mind map: focusing the most important issues This tool helps to the foundrymen to focus on the most important issues to achieve low dioxin emissions: 4.1 Cupolas


13

4.2 Rotary furnaces

Raw material

Use uniform charge

Charge in the correct order

Minimize or eliminate oil, paintings or plastics

Avoid contamination during storage

Eliminate mould sand on returns

Use gaseous fuel Furnace

Optimize combustible/comburent ratio

Control the refractory wear

Adjust well the burner support and the furnace

Heat exchanger Rapid cooling between 450-250ºC

Avoid long and horizontal ducts

Optimize duct diameters

Keep ducts clean

Gas line

Rotary Furnace

Dust removal

Dry

Wet

Baghouse

Cyclone

< 5 mg/Nm³

Repair faulty equipment

Implement procedures

Keep good working order Tightness

Sleeves

Supports

Exhausts


14

4.3 Electric Arc furnaces


15

5. Check list: concepts to check before the production A check list is a list of items to be checked, noted or remembered. In this case this tool has been included in order to help the verifying of the process to the foundrymen. Checking one per one the foundrymen can mark the parameters which do not fulfil the recommendation. After finishing the checking, it will be easy to change the parameters previously marked. Therefore this tool has the purpose of helping the foundrymen to identify the problems, parameters… to check before starting the process.


16

CHECK LIST FOR GOOD OPERATION AND DIOXIN MINIZATION IN FERROUS FOUNDRIES

PART 1: COMMON FOR EVERY FOUNDRY

CHECKING RAW MATERIALS FOR ALL TYPE OF FURNACES

MATERIAL STORING PLACE

□□ Do the raw materials arrive clean? It is important if we could choose a clean supplier.

□□ Do you store each material always in the same storing place? It is very important not to pollute clean raw materials with residues of oil, paintings… of previously stored materials.

□□ Are the storing places clean? Do you clean them frequently?

RETURNS

□□ Are they free of mould sand (binder residues)?

SCRAPS (Steel or iron)

□□ Are they free of oil, grease, plastic, wood and paintings?

□□ If they are oily, greasy… is it possible to carry on a cleanliness procedure?

PIG IRON

□□ Are the storing places clean? Do you clean them frequently?

□□ If not, is it possible to clean them?

MACHINING CHIPS

□□ Are they free of oil? (Not usual for induction furnaces)

COKE FOR CUPOLAS

□□ Is it dry? Avoid too wet cokes.

□□ Is the ash content low enough? It has to be as low as possible.

□□ Is the fine fraction of coke low? It has to be as low as possible.

□□ Is the breakage index low enough? It is necessary a good MICUM index

LIQUID OR GASEOUS FUELS FOR RFs or POST COMBUSTORS

□□ Is the fuel in liquid form? It is interesting using as gaseous for reducing some dioxin formations.


17

CHECKING THE DESIGN AND EQUIPMENT OF THE GAS LINE

GAS LINE

□□ Are duct diameters optimized / minimized?

□□ Are avoided long and horizontal ducts that can lead to problems with memory effect?

□□ Do you keep the ducts clean?


EXHAUST TREATMENT

□□ Do you have post-combustors?

□□ Do you remove dry dust (with dry or wet systems) before the post combustor or stack?

□□ Do you use water sprays or air/air exchangers for the cooling process?

POST COMBUSTION

□□ Do you eliminate CO? CO is a precursor of dioxins.

□□ Do you need to implement burners? Implement them if necessary. Minimize them due to energy and new dioxins source reasons.

□□ Do you cool the gases before the fabric filter? At least to avoid burnings.

COOLING SYSTEM 1: QUENCHING BY SPRAYED WATER

□□ Do you use a permanent spraying even having low temperature in exhausts?

□□ Is the cooling process until 100ºC quick enough? It is necessary a quick pass between 250 and 450ºC temperature window to avoid de-novo synthesis; dioxin reformation.

□□ Have you removed deposits or quencher pipes? It is necessary to remove them (sludge included) in order to avoid memory effect.

COOLING SYSTEM 2: EXCHANGERS AT THE STACK

□□ Do you cool gases before removing the dusts? Do you cool them quick enough?

□□ Do you keep clean the exchangers? Do you remove deposits or sludges?

□□ Do you keep clean the ducts?


DUST REMOVAL 1: DRY SYSTEM

□□ Is the cyclone effective? It is necessary an effective cyclone to limit the presence of coarse particles rich in unburned matter, as well as to avoid burnings.

□□ Do you have a bag filter? Do you have an effective cleaning?

□□ Are dust removers sized to filter stack gases in process peaks (of volume flow, temperature, dust load…)?

□□ Do you follow strictly supplier recommendations for operation set-points?


18

□□ Do you check regularly the equipment? In case of failures, do you repair it immediately?

□□ Do you fulfil the lowest goals of particle emission (<5 mg/Nm³) included in Foundry BREF?

DUST REMOVAL 2: WET SYSTEM

□□ Is it efficient enough for your process? They are in general less efficient and more problematic than dry systems.

EQUIPMENT MAINTENANCE

□□ Do you carry out preventive maintenance of the equipment?

□□ Do you keep it in good working conditions?

TRAINING

□□ Are the operators informed of the risks associated to dioxin formation and emission?

□□ Have the operators been trained in prevention of dioxin emissions? (Satisfactory furnace operation, checking of the stack gas treatment systems…)

SUPERVISION

□□ Do you have checking procedures?

□□ Do you have operation procedures in accordance with recommendation of technical sheets?

WASTE MANAGEMENT

□□ Do you store removed dust correctly? Big bags or closed containers avoid rain and diffuse emissions.

□□ Do you analyze periodically the dioxin content of the removed dust?

□□ In case of having dioxins, do you handle removed dust as hazardous waste?


19

PART 2: CHECKING EACH TYPE OF FURNACE

CHECKING THE CUPOLA FURNACE

CHARGE

□□ Is it uniform?

□□ Has it got a good distribution?

THERMAL EFFICIENCY

□□ Will O2 injection improve your energy efficiency? It could be interesting using an optimized oxygen injection.

□□ Is there a secondary row of nozzles?

PRODUCTION ORGANIZING

□□ Try to minimize interruptions in the melting (stoppage), unburned gases, memory effect...

□□ Is furnace operation regular?

□□ Are air losses avoided along the process?

CHECKING THE ROTARY FURNACE

CHARGE

□□ Is the charge uniform? Uniform charge allows an easier melting, saving fuel and time.

THERMAL EFFICIENCY

□□ Do you charge the materials in the correct order? A correct order of the charge means thermal efficiency.

□□ Is it optimized the combustible (fuel or natural gas)/comburent (air or O2) ratio?

□□ Have you got an oxyburner?

□□ Is well kept the adjustment between the burner support and the furnace during the melting operation? Little maladjustments mean increasing needs of fuel and decreasing efficiencies.

FURNACE MAINTENANCE

□□ Is well controlled the refractory wear, in particular, at the exhausts outlet zone? Uncontrolled wears mean exhaust escapes to the plant with a high dust generation containing dioxins.


20

CHECKING THE ELECTRIC ARC FURNACES

CHARGE

□□ Is the charge uniform? Uniform charge allows an easier melting, saving fuel and time. Strongly recommend mixing of the charge with various sizes of pieces.

THERMAL EFFICIENCY

□□ Do you charge the materials in the correct order? A correct order of the charge means thermal efficiency.

□□ Have you got an oxyburner? It's necessary, if you use it as heat booster in melting process.

FURNACE MAINTENANCE

▫ Is well controlled the refractory wear, in particular, at the exhausts outlet zone? Uncontrolled wears mean exhaust escapes to the plant with dust and other pollutants generation.

▫ Have you got indirect exhaust gases? If yes, during melting period check leakliness of hood and collecting pipe of the dedusting installation.

▫ Did you check the condition of the graphite electrodes before the melting? Bad condition of electrodes induces increase of energy consumption and possibility of dioxin formation.

Documents

Apostila Furanic Resin