EUROFER background paper on waste gases 27092008

CO2 connected to electricity produced from “waste gases” shall not be subject to auctioning.

For allocation purposes the carbon content in the “waste gases” must be allocated to the installation, where the respective “waste gas” is produced.

Allocation to installations in the steel sector shall be made by benchmarks and the amount of CO2 derived by combining benchmark and reference production volume shall be 100 % free of charge (which forces bad performers to buy allowances and allows good performers to sell allowances).

Benchmarks shall reflect the full carbon content of any associated waste gases. No subtractions for energy content in the respective “waste gases” shall be made.

All provisions must be anchored in the revised Emissions Trading Directive and not be entirely left to Comitology (which shall only be employed to define relevant technical details).

Justification in bullet points

Steel making from iron ore (“integrated steel making”) unavoidably gives rise to “waste gases”. It is not possible to produce steel by this route without producing “waste gases”.

The “waste gases” have a calorific value, which can be used, e.g. for heat or electricity production.

Amount and composition of waste gases are only governed by the steel production process. Neither is changed, if the energy content of the “waste gases” would not be used (e.g. for electricity production).

“Waste gas” production is limited by resource efficiency of steel making and in the future by process-related benchmarks.

If the energy content of “waste gases” is not used, because of their content of toxic and explosive gas components they must be flared before being released into the atmosphere. CO2-emission from the “waste gases” and the associated steel making processes is independent of the use or non-use of the energy contained in these “waste gases”.

Since the producer of “waste gases” cannot control and adjust their volume and composition according to the situation on the energy markets, in their application for electricity production the “waste gases” cannot be compared to commercial fuels.

Roughly 80 % of the overall CO2-emissions of integrated steel making is contained in the “waste gases”. Electricity production from “waste gases” releases roughly 50 % of the overall CO2-emissions of integrated steel making.

Independently from the point of its release (e.g. from stack of a power plant) the CO2 from “waste gas” use is connected to the steel making only. Thus exploitation of the energy content of “waste gases” is CO2-free.

The energy content of “waste gases” currently produced in EU27 can substitute natural gas equivalent to a CO2-emission of up to 50 Million tons per annum.

The use of “waste gases” for electricity production does not come for free, but is linked to huge investments and operating costs.

Justification and relevant background information in detail

To produce steel from virgin natural raw material carbon is needed to release the iron from its connection to oxygen as it is present in the iron ores.

This form of steel production is called the “integrated route” and is performed in a series of processes, which are called sintering, coke making, hot metal production and blowing. The installations used are called sintering strand, coke ovens, blast furnaces (“BF”) and basic oxygen furnaces (“BOF”).

These operations are usually concentrated at one site, which is called an “integrated site” or “integrated plant”. An integrated steel making site can be divided into two kinds of activities: the “liquid phase” and the “processing phase”. The “liquid phase” consists of the already mentioned processes: sintering, coke making, hot metal production and blowing. The processing phase of hot rolling, cold rolling, surface treatment operations and any other directly connected steel treatment operation or operation connected to the steel making (e.g. on-site power plants). The more of these operations are operated at one specific site, the higher the “integration” of this site.

During coke making (in the coke oven), hot metal making (in the BF) and blowing (in the BOF) not only coke, hot metal (also called “pig iron”) and crude steel are produced but inevitably also coke oven gas (COG), blast furnace gas (BFG) and basic oxygen furnace gas (BOFG).

For health protection (BFG and BOF gas contain large shares of Carbonmonoxide) and safety reasons (COG and BOFG contain Hydrogen) the “waste gases” must be fully oxidised before they can be discharged into the atmosphere. Therefore, if the energy content of the “waste gases” is not used in burners of power plants or industrial ovens, these gases must be flared. For this reason the CO2 emitted from the steel making operations are the same, independently of the use the “waste gases” are put to. The only difference is the possible resource saving effect when using the energy content of the “waste gases” and the associated mitigation of CO2.

Since these gases are inevitably linked to the respective production process they comply with one criterion (of several) of the definition of waste: There is no possibility to avoid their production, if coke, hot metal or crude steel are produced. Thus these gases are dubbed “waste gases” or “fatal gases”.

These gases have an energy content and can be burned to produce heat either for direct use (e.g. heating furnaces, air pre-heating, ...) or for electricity generation. Per ton of crude steel the “waste gases” from the production route contain roughly 7 000 MJ of energy. Roughly calculated shares are 60 % for BFG, 35% for COG and 5 % for BOFG.

The reason that BFG and BOFG have an usable energy content is the constraints that the laws of physics and chemistry (thermodynamics and kinetics) impose on the efficiency of the chemical reactions in hot metal making. Ideally, in the BF, each carbon atom will catch two oxygen atoms from their connections to iron atoms in the iron ore. Around three quarters of the carbon atoms entering the BF (as coke and coal) can do this and so the resulting gas has some carbon, which has still a possibility to catch an additional oxygen atom. This is equivalent to an energy potential, which can be released by adding these oxygen atoms (= burning the gas). Additionally, some part of the carbon used in the BF enters the hot metal and will be released during the blowing in the BOF as BOFG. This carbon also reacts only partly with oxygen, which again results in some energy potential, which can be extracted by burning this gas.

COG has a gestation history different to BFG and BOFG. COG represents the volatile component of coal, which has to be removed from the coal (the result of this process is coke) to give it enough strength, so that it will not be crushed by the weight of the iron ore, when both iron ore and coke are charged together into the BF.

The use of these gases as energy sources in diverse production processes has the effect that CO2-emissions, which from a chemical and process-related viewpoint stem from coke making, hot metal making or crude steel making are emitted not from the coke oven, the BF or BOF, but from other installations, e.g. a power plant or a hot rolling mill.

Keeping in mind their origin, it is evident that CO2 stemming from the burning of “waste gases” cannot be labelled as CO2 from the emitting point but only as CO2 from the source of the “waste gases”. Any other approach would also make the CO2 emissions from coke making, hot metal making and crude steel making incomparable (because the CO2-intensity would not be governed by the efficiency of the processes but the extent to which the “waste gases” are used inside or outside these processes).

The share of the CO2 emitted by an integrated steel plant, which is represented by the CO2 in the “waste gases” is rather high and not below 80%. The share of CO2 from the “waste gas”-connected electricity making is differing from site to site, but on average should hover around 50%.

This has a critical consequence with respect to any allocation mechanism. If the CO2 from the “waste gases” is not allocated for free (= without auctioning) to the producing process to the amount indicated by benchmarks, the steel making processes will receive allowances significantly below their needs. This will significantly increase risk of carbon leakage. Exactly to avoid this leakage risk, the use of benchmarks was introduced in the proposal for a revised ETS Directive. It is entirely unacceptable to invest resources in the benchmarking scheme, if any positive results thereof (reducing CO2 without generating leakage) is made impossible by destroying the benchmarks either directly (reduction of benchmarks in line with energy content of “waste gases”) or indirectly (not allocating 100% of the volume of allowances calculated by the benchmarks for free).

Using the “waste gases” as energy source (for electricity production or direct heat) is therefore not an activity giving rise to any CO2 attributable to the energy released by the burning of the “waste gases”. Composition and volume of the “waste gases” is a direct consequence of the steel making process and will not change if these gases would not be used! In this sense the production of heat or electricity is latently CO2-free.

This is of special importance to electricity production from “waste gases”. Since neither the energy potential nor the composition or the amount of the “waste gases” will change, if there would be no electricity production, both the CO2 and the energy contained are a result of steel making and in no way linked to the electricity, which may be produced from the “waste gases”. Giving free allowances to a power plant which uses the “waste gases” is therefore not a breach with the concept of the ETS Directive, which demands that electricity should bear a visible CO2-price: This electricity is made with no additional CO2-emissions!

The ETS Directive is focused on direct emissions. Thus the operator of an installation must submit allowances to the extent of CO2 physically leaving his plant. Therefore, an electricity generation plant using “waste gases” needs to have access to the allowances assigned for free to the steel making processes. In the light of the above, this does not constitute free allowances for electricity generation and is therefore not in conflict with the general principles of the ETS Directive. These free allowances are only an administrative effect, caused by the direct-emissions focussed nature of the ETS Directive.

The best method to guarantee that this allocation is in line with the actual use of the “waste gases” would be an ex-post allocation (only ex-post it is known which amount of which “waste gases” was produced and used in which installation). Since the ETS Directive is firmly built on an ex-ante principle this is not possible. The next best method is then the full allocation of the CO2 linked to the carbon in the “waste gases” to the primary source of these gases. If the integrated site wants an external operator to use these gases, it will have to submit respective allowances together with the gas. Due to the “waste characteristic” of the gases, it is the “waste gas”-user, who has the stronger market position, because the producer of the waste gases does not have the option of not producing these gases. There is therefore no possibility that an integrated site can pass-through any CO2-price together with the waste gas to the user.

In the framework of the auctioning rules as suggested by the proposal of the Commission for a revision of the ETS Directive an electricity consumer, who is supplied with electricity from “waste gases” has access to cheaper electricity than a consumer who does not have this access (to the extent that the integrated steel making indeed receives free allowances).This however is an inevitable consequence of any policy on resource conservation and CO2-mitigation.

Since the ultimate aim of climate change policies is a price signal which differentiates low-carbon and high-carbon electricity sources, this “unequal” situation is exactly in line with the overall political aim of environmental policy(because “waste gas”-related electricity bears no extra CO2 burden). This is not only true for the case of the “waste gases”, but also for any other situation, were energy potentials in industrial processes are exploited. Example is electricity production from BF top gas pressure turbines or any capture and use of waste heat.

It is important to note that the restricted access to electricity produced from “waste gases” stems from the restricted amount of “waste gases” available (which is off course necessary to reduce CO2-emissions from steel making). It is not linked to any physically restrictions to access (electricity contracts could be designed to contain electricity from “waste gas” use regardless of any physical proximity between electricity user and integrated site). A physical restriction applies to the use of the “waste gases”, which cannot be transported economically over long distances (due to relative low energy content per volume).

Currently, the annual energy content of the “waste gases” of the EU27 is equivalent to an amount of up to 50 Million tons of CO2 which would be emitted by the use of the amount of Natural Gas, which has the same energy content.

Given the significant amounts of CO2 which are associated to the energy potential of the “waste gases”, any kind of efficiency improvements on the use of “waste gases” should therefore be supported and not restricted. This is especially true for increasing efficiency of electricity generation equipment.

The integrated steel making route is characterised by a predefined sequence of production steps. First the liquid phase, then the downstream operation of hot forming, cold forming and surface treatment. The energy content of the “waste gases” from the “liquid phase” is largely equivalent to the energy needs downstream. In reality even fully integrated sites with highest energy efficiency rely on additional external energy supply, especially of electricity. Thus it can be said that all electricity and heat produced from “waste gases” are consumed by the downstream processes. The more integrated a steel making site is, the more of this energy can be consumed on-site.

An internal EUROFER survey covering 85% of integrated steelmaking in the EU 27 has shown that currently there is an additional net import of annually around 10 TWh of electricity (which is part of an estimated overall electricity demand of 45 TWh). For camparison: The electricity consumption of the EU27 in 2005 was roughly 2800 TWh the industry-part thereof roughly 1200 TWh. Even in the rare and exceptional cases where for historical reasons the “liquid phase” is on an isolated site, the intermediary products (“semis”) of this site will have to be processed on another site, on which the downstream processes are operated. Therefore “net exporter”-positions of electricity may exist, if judged from the viewpoint of a site, but not if judged from the viewpoint of the whole integrated steel making process up to the final products.

Linking the access to benchmark-based free allowances for the CO2 contained in “waste gases” used for electricity production to the geographical situation of the electricity generating plant only without taking into consideration the net-balance of the whole integrated production chain will create cases of unequal treatment within the steel industry. This may be relevant if an operator for historical reasons operates the “liquid phase” at a different location than the “processing phase”.

For at least two reasons, “waste gas” generation is always at the unavoidable minimum. First, respective changes in the processes reduce process efficiency of steel making, which is the prime source of profit for an integrated steel mill. Second, benchmarking will impede expansion of carbon use and thus “waste gas” production.

To be able to use these “waste gases” a complex infrastructure has to be built within the site of an integrated steel mill: capture (e.g. the investment for BOFG is around 5 Million Euro per MW), preparation, blending, transport.

The “waste gases” are also characterised by comparatively large volume of gas related to their energy content. This stems from the large shares of inert components and large amounts CO2 contained. The first element is due to the use of air in metallurgical processes the latter from the metallurgical processes themselves (which indeed strive to create as much CO2 as possible out of the carbon). The large volumes give rise to increased difficulties in electricity production, heat transfer and off-gas cleaning. This is another reason, why “waste gases” are not comparable to commercial energy carriers.

The huge volumes of “waste gases” produced do not allow for storage of gas production of more than a few minutes.

The “waste gases” are characterised by low energy content by volume and fluctuations in gas availability (due to the steel making process). The most robust way to produce electricity is a combination of boilers and steam turbine. This relatively simple arrangement can achieve high levels of availability and is suitable for gases with low calorific value. It also allows to bleed-off high temperature steam for the various steam consumers within the integrated steel making (mostly used as back-up if to level out fluctuations in waste-heat supply). Combustion turbines (= CHP) are thus not the first choice, since they are not ideal for low calorific gas feeds.

The “waste gases” are also characterised by large amounts of CO2 per energy unit. Compared to Natural Gas and its energy content, BOFG at least has the threefold carbon content and BFG the fourfold carbon content (COG is roughly equivalent to Natural Gas). If the direct link to the steel production processes were not taken into account, in a CO2 constricted environment the use of these gases would be impeded (which eventually would be counter-productive since the CO2-saving effect of “waste gas” use would not be taken into account).

The notion of reducing the benchmark to the extent of an energy content in the “waste gases” rests on the assumption that this energy has a market value, for which there is no need of support by free allocations. This kind of reasoning assumes that if a “waste gas” receives free allowances but would be flared, the operator, will have to buy electricity from the market (instead of producing it from the “waste gas”) and thus will lose the some of the free allowances received for the “waste gas” by the need to buy the CO2-costs contained in the electricity price. If this operator now receives less allowances for his “waste gas” to the extent of CO2-costs incurred by buying electricity, this will not disincentive him from electricity production from his “waste gas”, because in both cases (electricity making and flaring) the net financial effect is the same. This reasoning is not applicable, because the use of “waste gases” for electricity production does not come for free, but is linked to huge investments and operating costs. Thus a benchmark reduced by energy content constitutes a disincentive.

The production of these gases is directly linked to the production processes of steel and thus cannot be controlled seperately. A sophisticated on-site management system must be in place to make the best use possible of these gases as they are produced.

Waste gas composition and production in time are governed by the process characteristics of coke making, hot metal making and steel making. Both composition and volumes therefore show fluctuations, which give rise to the need for operation of the gas management system mentioned above.

This is also the most importantdifference between electricity production from “waste gases” and from commercial fuels. An operator who produces energy and heat from commercial energy carriers may optimise his operation according to the costs connected to the raw materials and market prices. Eventually, he has the choice of not producing heat or electricity and thus do not buy any energy carriers. This is not the case for the steel mill generating “waste gases”. These gases have to be used (or flared), independently of the question, if the market conditions are favourable for electricity production or not.

As the figures on the share of CO2 from waste gases on the overall CO2-emissions of integrated steel mills show, any reduction of associated allowances has a huge impact in these operations. This is an additional dimension compared to the first and second trading phase, where a mechanism to support “waste gas”-use in spite of the relatively higher carbon content was the main issue. Therefore, relevant provisions should be integrated in the text of the revised Emissions Trading Directive and not left to Comitology. The latter would delay any concrete stipulations and add an additional uncertainty.

Any benchmark for integrated sites must be able to allow equivalent treatment of such sites and point to the low-CO2-production, in a way which clearly reflects the applicability of mitigation options. To achieve this, one necessary characteristic of benchmarks for integrated steel-making is that all carbon (and thus CO2) present in “waste gases” must be assigned to the process generating these “waste gases”. Any deviation from this principle will introduce a political element in the benchmarks, which will make it impossible to set up a simple and robust benchmarking system. The Communication 870 from 2003 under point 92 allows Member States to do so optionally. The revised ETS Directive will move all decisions to the EU-level and thus the respective provisions of the Communication 870 from 2003 must be defined in the revised Directive itself.

Not allocating the carbon content of the “waste gases” or reducing a benchmark by significant parts of this carbon content -due to the allocation mechanism proposed in the Commission’s proposal for a revised ETS Directive- will set a level of allowed CO2-emissions, which cannot be met. Such a “reduced benchmark” will not be able to provide the service for which it was introduced: To avoid leakage.

Figure 1: Schematic explanation of “waste gas” generation, “waste gas” use and related electricity production (copyright with Salgzitter AG)

Figure 2: Schematic example of energy flows in an integrated site

Figure 3: Carbon flow through an integrated site (copyright with Salzgitter AG)

Figure 4: Aerial view of an integrated steel plant gas holders

Figure 4: Typical example for the source of CO2 and the point of discharge