AN APPLICATION OF EXERGY ACCOUNTING TO FOUR BASIC · 2008. 9. 30. · AN APPLICATION OF EXERGY...

AN APPLICATION OF EXERGYACCOUNTING TO FOUR BASIC

METAL INDUSTRIES

by

A. MASINI*and

R. U. AYRES**

96/65/EPS

This working paper was published in the context of INSEAD's Centre for the Management of EnvironmentalResources, an R&D partnership sponsored by Ciba-Geigy, Danfoss, Otto Group and Royal Dutch/Shell andSandoz AG.

* Research Associate at INSEAD, Boulevard de Constance, 77305 Fontainebleau Cede; France.

** Sandoz Professor of Management and the Environment at INSEAD, Boulevard de Constance, 77305Fontainebleau Cedex, France.

A working paper in the INSEAD Working Paper Series is intended as a means whereby a faculty researcher'sthoughts and findings may be communicated to interested readers. The paper should be consideredpreliminary in nature and may require revision.

Printed at INSEAD, Fontainebleau, France.

AN APPLICATION OF EXERGY ACCOUNTING TOFOUR BASIC METAL INDUSTRIES

Andrea Masini & Robert U. Ayres

CMER, INSEADFontainebleau, France

September 1996

Abstract

This paper aims to demonstrate that the thermodynamic quantity known as exergyprovides a systematic and uniform general-purpose indicator for both materials and energyresource inputs, useful product outputs, waste products and energy losses for any industrialprocess. In fact, an exergy balance groups together both mass and energy flows, thusproviding a concise means of characterizing any materials transformation process. It linksresource analysis, industrial engineering and environmental analysis. Moreover, it can be usedfor assessing process efficiency, identifying potential opportunities for future technologicaladvances, and, finally, it even provides a basis for first-order evaluation of the potential forharm associated with waste materials released into the environment.

To demonstrate these capabilities an exergy accounting has been carried out for fourmetal industries: steel, aluminum, copper and lead. Production processes have been analyzedand compared in order to identify major inefficiencies as well as areas of potential risk forthe environment. The results are presented in a common analytical framework, using data forthe US in 1988.

A. Masini & R. U. Ayres Exergy analysis of metal industries September 30, 1996 Page 2

Introduction: The Exergy Concept

The idea of available energy dates back to the last century, when it was first understoodby the French engineer Sadi Carrot for the specialized case of heat engines. In the nextdecades the concept of "available work" was further developed theoretically, especially byHerman Helmholtz and Willard Gibbs. It has been applied to many kinds of processes, fordifferent purposes, under several different names — availability, available work, essergy,physical information — but only recently a standard definition has been formulated and thename exergy definitely adopted [Rant 1956; Gyftopoulos et al 1974; Wall 1977; Szargut etal 1988]. However, for the purposes of this study it is sufficient to present only the essentialfeatures of the theory.

An adequate definition of exergy is the following: "Exergy is the amount of workobtainable when some matter is brought to a state of thermodynamic equilibrium with thecommon components of the natural surroundings by means of reversible processes, involvinginteraction only with the above mentioned components of nature" [Szargut et al 1988]. Inshort, exergy is an extensive non-conservative variable which synthesizes in a concise anduseful expression both the first and second law of thermodynamics. It is definable andcomputable (in principle) for any substance, or system, with respect to the real environmentin which the system is located and/or operates.

In principle, four different types of exergy B can be identified. These are denoted,respectively, as kinetic, potential, physical and chemical exergy, viz.

B = Bk + Bp + Bph + Bch

Kinetic and potential exergy have the same meaning as the corresponding energy terms.Kinetic exergy is relevant for analyzing a flywheel or turbine. Potential exergy is relevant forelectrical or hydraulic systems. But these two terms can safely be neglected for purposes ofanalyzing most common industrial processes. Physical exergy is "the work obtainable bytaking a substance through reversible physical processes from its initial state (temperature T,pressure p) to the state determined by the temperature T. and the pressure po of theenvironment"[Szargut et al 1988]. Physical exergy assumes an important role for purposesof optimization of thermal and mechanical processes, including heat engines and power plants.But it is of secondary — in fact negligible — importance when attention is focused on verylarge scale systems, such as chemical and metallurgical processes at the industry level. In thiscase chemical exergy plays a major role for purposes of resource accounting and environmen-tal analyses.

Chemical exergy is "the work that can be obtained by a substance having the parametersT. and po to a state of thermodynamic equilibrium with the datum level components of theenvironment"[Szargut et al 1988]. It has two components: a component associated withchemical reactions occurring in isolation and a component associated with the diffusion ofreaction products into the surroundings.

All the foregoing definitions stress the importance of defining a reference state, orsystem, when calculating both physical and chemical exergy. As a matter of fact, the exergyfunction is a measure of the difference between two states, namely the state of the "target"system and that of its surroundings (or, more precisely, the ultimate state of the combinedsystem + surroundings, after they have reached mutual equilibrium). In short, exergy cannotbe calculated without defining appropriate parameters for the environment where the targetsystem operates, in terms of temperature, pressure and chemical composition.

The importance of defining the parameters for the common environment emerges clearly

A. Masini & R. U. Ayres Exergy analysis of metal industries

September 30, 1996 Page 3

when we consider the analytical expressions for exergy. They also show that exergy is ameasure of the thermodynamic "distance" of the target system from equilibrium. Another wayof saying this is that exergy is a measure of the "distinguishability" of the target system fromits environment. These statements follow from the fact that exergy vanishes when the targetsystem under consideration has the same thermodynamic state as the environment.

In general, for a closed system with temperature T, pressure p, entropy S, and volume V,exergy can be written as:

B = S(T - Ta) - If(p - po) +1Ni ( - )

where Ni is the number of moles of the firth system and i is its chemical potential. Thesubscript "o" refers to the final state of equilibrium of the system plus the environment,combined together. Again, the exergy of a flow crossing the system boundaries of an opensystem can be written as the sum of three terms:

B = H - Ho - To(S - So) - - Nio)

where the letter H stands for enthalpy. The third term of these expressions takes account ofthe contribution due to the chemical transformation of the system.

In both of these expressions is straightforward to recognize how the choice of thereference state affects the value of the function B. For the purpose of calculating physicalexergy, this choice does not represent a major problem, as it is relatively easy to define anappropriate level for pressure and temperature of the environment, namely ambientatmospheric temperature and pressure. This is not the case for calculation of chemical exergy.The latter step requires knowledge of the detailed average chemical composition of thereaction products and the environmental sink with which the system interacts.

In this context, considerable efforts have been undertaken by a number of authors. Onepossible approach would be to assume, as the reference level, the average chemicalcomposition of the earth's crust after reaching an hypothetical (calculated) equilibrium withthe atmosphere and oceans [Ahrendts 1980]. However, the results vary dramatically accordingto the depth of the crustal layer that is assumed to be equilibrated. A more practical genericsolution has been proposed by Szargut et al [1988]. This approach recognizes that the threemain sinks — atmosphere, oceans and crust — are not in equilibrium with each other, butassumes that the reaction products in any given case must go to one of the three, dependingon whether they are volatile (to air), soluble in water (to oceans) or neither (to earth's crust).They calculate standard chemical exergy for a number of chemical compounds and pureelements. The latter procedure has been adopted and extended in several later works [Ayres& Martinas 1995; Ayres et al 1995, 1996; Ayres & Ayres 1996] and its results have been alsoused in the present study.

Exergy as a Tool for Resource and Waste Accounting

The intensive use of natural resources for anthropogenic activities is progressivelydepleting the reservoirs accumulated over the millennia. At the same time the large quantitiesof waste materials and effluents released to the atmosphere, the oceans, or to the land surfaceare altering the delicately balanced natural cycles that make the life possible on earth.Therefore minimization of resource use, as well as reduction of dangerous emissions

A. Masini & R. U. Ayres

Exergy analysis of metal industries September 30, 1996 Page 4

associated to industrial processes, constitute the primary objective of policies to be pursuedfor sustainable development. For this purpose it is of great importance to develop a generalmeasure capable of accounting both for materials use and waste residuals.

We suggest that exergy is the most suitable indicator for both resource accounting andwaste accounting. Nonetheless, up to now it has not been adopted for this role. For historicalreasons, resources have been always divided in two categories, namely fuels (measured inenergy units) and mineral, agricultural or forest resources (measured in a variety of massunits). This distinction leads to some incongruities and much confusion, as the choice of adifferent "currency" for each flux does not enable analysts to evaluate and compare all inputsand outputs on a common basis. In particular, non- fuel flows — such as minerals — areoften neglected.

The use of exergy as a general environmental indicator for resource accounting wouldimprove the situation in two important ways. First, an exergy balance automatically combinesboth mass and energy flows, thus providing a concise representation of the process. Thismakes processes easier to characterize. Second, the use of exergy enables the analyst to takeinto account automatically both the first law (energy conservation) and the second (entropy)law of thermodynamics. In addition, by virtue of reflecting second law constraints, exergyanalysis — rather than energy analysis — is a suitable tool to identify areas of potentialtechnological improvements. In fact, irreversibilities and process inefficiencies cause exergylosses (i.e. the difference between exergy of inputs and outputs of the process) which reflectincreasing entropy. Any exergy loss shows that the system under consideration could befurther improved — at least in principle — in order to increase its thermodynamic efficiencyand to reduce the use of natural resources. But more important, comparing the relativemagnitude of such losses, both within a complex process, and between alternative processes,is a useful guide to identifying the most promising technology choices and targets for R&D

A third advantage for exergy accounting, in contrast to conventional approaches, is thatit opens the possibility of comparative evaluation of different kinds of materials, not only inmass terms but also in terms of available energy "saved" in the sense of not being requiredfor separation and purification from the average composition of the environment. This isessentially equivalent to the "energy content" of an ore or mineral. (Here we use the termenergy in the familiar but inexact sense of normal language, rather than the language ofspecialists. In fact, it is exergy we are talking about). As already remarked, metal ores,minerals and even agricultural products, as well as chemical compounds, are all preciousresources for which heat of combustion (which is the usual measure of energy) is zero orirrelevant. These substances can easily be measured and compared in terms of exergy.

Finally, chemical exergy "content" can be used as a tool for a first-order evaluation ofthe environmental impact associated with the waste effluents of any industrial process. Infact, exergy is not only a natural measure of the resource inputs, it is also a measure of thematerial outputs. Indeed, as emphasized in a previous work by one of the authors, the exergycontent (or "potential entropy") of a waste residual can be considered as its potential for doingharm to the environment by driving uncontrolled reactions [Ayres & Martinas 1995; Ayreset al 1996].

The use of chemical exergy as a tool for evaluating the potential harm of wastes couldlead to over-simplification if directly applied to emissions. As a matter of fact, the chemicalexergy content of any substance cannot be directly related to its toxicity to humans or otherorganisms. On the other hand, it represents a measure of how far the substance is fromequilibrium with the common state of the environment. In this sense, an high exergy contentis a simple indication that the substance under consideration is likely to drive further chemicalreactions when it is discharged to the atmosphere, to watercourses, or deposited in landfills.



Or, under a different perspective, it suggests that the materials discharged could be furtherprocessed to extract potentially useful work.

The US Metallurgical Industries in 1988

We have conceptually divided the processes of mining, concentration (or winning),reduction or smelting and refining as shown schematically in Figure I. There are four stagesof separation or recombination. The first two, being physical in nature, are assigned to themining sector (Standard Industrial Category or SIC 11) or the quarrying sector (SIC 14). Thelast two, being chemical in nature, are assigned to the primary metals sector (SIC 33). At eachseparation stage, wastes are left behind and a purified product is sent along to the next stage.In principle, the wastes can be determined by subtracting useful outputs from inputs at eachstage, bearing in mind that inputs may include water and oxygen from the air.

Unfortunately, from the analytic point of view, published data is rarely available inappropriate forms. There are significant imports and exports of concentrates and crude metals(and even some crude ores) but trade data is often given in terms of metal content, rather thangross weight. Domestic data is also incomplete, due to the large number of data withheld forproprietary reasons. Thus, in a number of cases, we have been forced to work back fromsmelting or concentration process data to estimate the input quantities of concentrates.

Inputs to the U.S. primary metals sector consist of concentrates (produced in the miningsector, or imported), fuels, fluxes, and processing chemicals. As an example, Figure 2 showsthe inputs necessary to achieve the U.S. 1988 steel production by means of known unitprocesses and published scrap usage. This figure excludes ferroalloys, of which U.S.production in 1988 was about 1 million metric tons (MMT), because of the extremecomplexity of the subsector. We have decided to equally exclude fuel inputs, CO and CO2from our materials balance, as CO is a major pollutant of smelting processes, but it resultsfrom partial oxidation which is later completed in the atmosphere. (Thus, the materialsbalance approach is not applicable for estimating CO emissions.)

Major purchased material inputs, other than concentrates, are mostly fluxes. These aremineral substances that make molten metals and slag flow more easily. The most importantfluxes are limestone (calcium carbonate) and dolomite (magnesium carbonate), with fluorspar(calcium fluoride) being another. Manganese ore is also added to blast furnace feeds formetallurgical reasons that we need not consider in detail. In 1988 approximately 9.9 MMT(10.9 million short tons) of limestone and dolomite were used, mostly in blast furnaces. Itappears from data published by the U.S. Bureau of Mines that the limestone was mostlycalcined on site (to drive off carbon dioxide) and consumed as "slaked lime", or calciumoxide (4.8 MMT). Other mineral inputs to SIC 33 reported by the Bureau of Mines includesalt (0.33 MMT), manganese ore (0.123 MMT) and fluorspar (0.137 MMT).

The production of primary metals from concentrates is normally accomplished bycarbothermic reduction (smelting with coke), or electrolysis. By far the major product, intonnage terms, is pig iron. U.S. blast furnace output in 1988 was 49.8 MMT. Pig iron has aniron content of 94%. It is almost entirely used for carbon steel production, via the basicoxygen process. (There was a small amount of open hearth production in 1988, now phasedout. Electric steel "minimills" use scrap metal exclusively).

US blast furnace inputs in 1988 included about 3 MMT of scrap iron/steel, while sinteralso utilized about 6 MMT of upstream "reverts" (flue dust, mill scale). However, the ironcontent of U.S. ores in 1988 was reported as 57.515 MMT. Blast furnace inputs (pellets)averaged about 63% iron, 5% silica, 2% moisture and 0.35% other minerals (phosphorus,



manganese, alumina). The remainder was oxygen. In the reduction process the oxygencombines with carbon (actually carbon monoxide) from the coke. About 1 metric ton of coke(essentially pure carbon) was used per metric ton of pig iron, along with 0.142 metric tonsof miscellaneous materials, mostly fluxes (lime and limestone) for the sinter plants.

Slag consists of the silica and other non-ferrous minerals in the sinter and pellets, plusthe materials in the fluxes. Total iron blast furnace slag production in the U.S. was 14.2MMT, or 0.35 metric tons of slag per metric ton of pig iron. Slag is no longer considered awaste, since virtually all slag produced is marketed for a variety of uses. Subsequent refiningof pig iron and scrap iron to carbon steel is done in a further refining stage, normally thebasic oxygen furnace. Steel furnaces produced an additional 5 MMT of slag in 1988.

As noted, the oxygen in the iron-bearing concentrates reacts in the blast furnace withcarbon monoxide. The reduction process requires excess CO, so the emissions (known as"blast furnace gas"), consist mostly of unreacted CO. It is combustible, though of relativelylow heating value. Currently most of this gas is utilized elsewhere in an integrated steelcomplex as fuel e.g. for preheating blast air (Figure 2) or heating ingots for hot rolling.However, the capture of gaseous emissions from blast furnaces is not 100% efficient, so someCO escapes. Considering the iron/steel process as a whole, all of the carbon (from coke) iseventually oxidized to CO 2 except for a very small amount left in the raw steel. Carbon isadded back to the finished steel in carefully controlled amounts as "spiegeleisen" (ferromanga-nese) and other ferroalloys. In 1988, the US steel industry accounted for 182 MMT of CO2,from coke, which is included in the grand total from fossil fuel combustion, discussed later.(In addition, some other hydrocarbon fuels were used).

Coke ovens and steel rolling mills are significant sources of hazardous wastes, eventhough the coke oven gas is efficiently captured for use as fuel, and about 55 kMT ofammonium sulfate (N-content) is produced as a by-product. This material is used as fertilizer.Coke is cooled by rapid quenching with water, and some tars, cyanides and other contami-nants are unavoidably produced. Unfortunately, materials balances cannot be used to estimatethese wastes. However, they probably constitute a significant fraction of both water andairborne wastes from the primary iron and steel sector.

Also, in the rolling process steel is cleaned by an acid bath ("pickling"), resulting in aflow of dilute wastewater containing ferrous sulfate or ferrous chloride (depending on the acidused). The excess acid is usually neutralized by the addition of lime. In 1988 about 215 kMTof 100% sulfuric acid (74 kMT S content) were used for this purpose, producing 250-300kMT of ferrous sulfate mixed with calcium sulfate. Ferrous sulfate can, in principle, berecovered for sale. However the market (water treatment) is insufficient to absorb the quantitypotentially available, and most is discharged as waste. A simplified schematic of the steelprocess chain for the US in 1988 is shown in Figure 3. We use this simpler version forpurposes of the exergy analysis hereafter.

Light metals, mainly aluminum, phosphorus and magnesium are reduced electrolytically.Only aluminum is discussed here. The oxygen in the purified alumina (from the Bayerprocess) reacts with a carbon anode, made from petroleum coke. The reaction emits 0.65metric tons of CO2 per metric ton of primary aluminum produced. In addition, primaryaluminum plants emitted about 0.02 metric tons of fluorine, per metric ton of aluminum,partly as HF and partly as fluoride particulates. This is due to the breakdown of cryolite (analuminum-sodium fluoride electrolyte used in the process, in which the alumina is dissolved)at the anode. Total airborne emissions from primary aluminum production in the U.S. (3.944MMT) were, therefore, 2.564 MMT of CO2 (already counted), 0.08 MMT of fluorides andabout 0.17 MMT of particulates (Al203).

In the case of heavy metals from sulfide ores (copper, lead, zinc, nickel, molybdenum,



etc.), the smelting process is preceded by, but integrated with, a roasting process whereby thesulfur is oxidized to sulfur dioxide, S0 2. Roughly 1 metric ton of sulfur is associated witheach metric ton of copper smelted (see Figure 4), 0.43 metric tons of sulfur per metric tonof zinc, and 0.15 metric tons of sulfur per metric ton of lead. Most of this sulfur (90%) isnow captured and immediately converted to sulfuric acid. In 1988 1.125 MMT of by-productsulfuric acid (S- content) was produced at U.S. non-ferrous metal refineries. The sources wereas follows: copper smelters (0.946 MMT), zinc smelters (0.136 MMT), lead/molybdenumsmelters (0.043 MMT). In terms of sulfuric acid (100% H 2SO4) the quantity of by-product(100%) acid produced was 3.54 MMT.

In the case of copper, most smelters are now located near the mines and most of this by-product acid (1.2 MMT) was used by mines for "heap-leaching" copper ore to produceconcentrates. Leaching now accounts for a significant proportion (c. 30%) of copperconcentrates produced in the U.S. Leached copper sulfate is subsequently reducedelectrolytically, without an intermediate smelting stage. The combined process is called thesolvent extraction-electrowinning (SX-EW) process. In the case of copper smelting, typicalconcentrates fed to the roaster/smelter consist of about 35% Cu (23%-45%), 35% S and 30%other minerals. In addition, about 0.25 metric tons of limestone flux is added per ton of blistercopper. Thus, slag production amounts to roughly 0.55 metric tons per metric ton of primarycopper, or 0.77 MMT for the US in 1988.

In the case of zinc, a typical concentrate would be about 55% Zn (45%-64%), and 27%S, with other minerals accounting for 16%. For lead, the corresponding number appear to be(about) 60% Pb (50%-70%), 9% S, and 21% other. Thus, assuming flux per unit of slag tobe the same as for copper (1.2:1), slag output should have been roughly 0.3 metric tons/metricton for zinc and 0.38 metric tons/metric ton for lead. This implies total US slag output of 0.06MMT for zinc smelting and 0.14 MMT for lead smelting. Total 1988 US slag production forthe three main NF metals was therefore roughly 1 MMT in toto. Carbon monoxide and carbondioxide emissions are not known exactly, but they are quite small in comparison with othersources. The waste numbers for other metals are comparatively insignificant, on a nationalbasis. A simplified schematic of mass flows for the US non-ferrous metals sector, taken asa whole for the year 1988, is shown in Figure 5.

Altogether, based on mass-balance considerations, we estimate smelting and refiningwastes for primary metals, including CO 2, to have been 43.4 MMT in 1988, including theweight of limestone, manganese, calcium fluorite, and other materials used in the blastfurnaces and refineries. (This includes about 14.2 MMT of iron (blast furnace) slag, althoughmost of this material is marketed commercially, mainly for road ballast). In addition there wasabout 5.2 MMT of steel furnace slag, a denser material with a fairly high iron content.

As noted above, much of the sulfur from sulfide copper, lead, zinc and molybdenum oresis also recovered for use, and sold as sulfuric acid (1.125 MMT S content in 1988).Subtracting the B. F. slag, and the by-product sulfuric acid, we get 28.1 MMT as the residualwaste. Of this, only about 1 MMT was solid (N. F. slag) and the rest would represent theoxygen content of the original ores — mostly iron — which is released as CO 2)1 . We havenot included the wastes from coking, which we have not estimated. The major airborneemission other than CO 2 is probably CO and particulates. In both cases, blast furnaces are themajor sources. The coking quench waters and some spent acids used for pickling constitutethe major water-borne wastes.

Mass flows and wastes for the metallic mineral processing industries and the metallurgi-cal industries, taken together, are summarized in Figure 4 (ferrous) and Figure 5 (non-ferrous). These figures are normalized to U.S. production of the refined metals. Some of theflows are imputed from others. For example, pig iron (94% Fe) contains roughly 6% C, which



implies a carbon content of 3 MMT. The oxygen required to burn this carbon away istherefore approximately 8 MMT, implying a carbon dioxide output of 11 MMT for steelproduction. In the case of iron blast furnaces, we assumed that all of the input coke, less the3 MMT of C embodied in pig iron, is converted to carbon dioxide. This consumes 63.4 MMT02 and yields 87.1 MMT of CO 2. However, some oxygen is captured from the iron oxide inthe ore. So, balancing inputs and outputs, we calculate that the additional oxygen taken fromthe air must have been 31.5 MMT. Scrap flows are very approximate, partly because thestatistics of the scrap industry are poor and partly because we have lumped stock adjustmentsand scrap flows for convenience.

For comparison, EPA estimated the 1983 wastes from iron and steel production (probablywet) at 6.0 MMT and from non-ferrous metals at 6.5 MMT [SAI 1985]. Their estimates werenot specifically designated as "dry" so some water content can be presumed. EPA estimatesairborne emissions from the primary metals sector as a whole to be 2.8 MMT, includingparticulates and CO, but not including CO2 [USEPA 1991]. Both sets of estimates are roughlyconsistent with the above.

Exergy Analysis of the Metallurgical Industries

A number of efforts have been undertaken to apply exergy analyses to metal productionprocesses, [Gyftopoulos et al 1974], [Hall et al 1975],[Morris et al 1982; Wall 1986; Szargutet al 1988]. These studies presented exergy accounts for individual process stages, in orderto identify major losses and evaluate the potential for further technical improvements in themetallurgical processes. On the other hand, the results obtained for different stages of theoverall process are not typically combined to give an overall representation of themetallurgical industries from an exergy perspective. Without such an approach it is notpossible to evaluate the environmental burden associated with the use of metals from "cradleto grave".

In order to solve these problems and to provide an aggregate representation of the results,in the present work the exergy analysis has been used in an integrated framework and appliedto the overall production processes of steel, aluminum, copper and lead, "from ore to ingot".As a matter of fact, such a common framework enables the analyst to evaluate the exergyflows not only for each single step but for the whole chain of production, thus providing anhomogeneous representation which allows to compare the above mentioned processes on acommon basis.

Therefore, the same system boundaries have been chosen for the four metal industries.All the selected industries have been analyzed from ore mining and processing to finalcasting. Further processes such as shaping or product manufacturing, as well as by-productsprocessing have been excluded from the model. All data presented were first normalized to1 ton of final product. On a later stage we have calculated totals for the US in 1988.

It is worth stressing that, as most of the processes analyzed take place in different plantswith different operation times and yields, the presented flow charts have to be considered asan average representation of the major flows over a reasonable long period. Moreover, as theselected metal industries use different technologies, the figures are industry-level compositesthat not fully reflect the complexity of the processes to which they refer.

Each sub-process has been considered as a single unit-box, linked to the other sequencesby means of material or energy flows; for that reasons recovered flows are not shown in themodel and each stream entering or leaving the system boundaries represents a net value.

Many different sources of data have been utilized and compared, in order to build an


overall flow diagram "from ore to ingot" per each selected process. In addition to theaggregate flow data summarized in the previous section, data used in the four cases havedifferent features.

The case of steel production presented the major difficulties, because of the inherentcomplexity of this industry, characterized by a large number of multi-stage integratedoperations. The overall material balance is a composite of the balances of each single sub-process of the industry, for which some sources were available [Ayres' data base; IDEA1991]. In addition, in order to better represent the crucial sector of coking and its relatedemissions, some data were extrapolated and adapted from [Russell & Vaughan 1976].

In the case of aluminum a general analysis carried out by Tellus Institute [Tellus 1992]and revised by TME [TME 1995] has been used as a supplementary reference for the materialbalance. This study presented a detailed list of air and water emissions, but was incompletein terms of mass balance. Therefore data have been integrated with other sources [Altenpohl1982; Ayres & Ayres 1996, chapter 2] or crude estimates in order to represent with asatisfactory degree of accuracy the exergy flows related to materials and wastes of theprocess.

On the other hand, for copper and lead production, such a detailed list of emissions wasnot available, and the overall flow chart for the material balance is a composite of manydifferent sources. We used [Gaines 1980; Hall et al 1975; Ayres & Ayres 1996, Chapter 3]for the case of copper, and [Burkle 1980; OHMP 1977; Morris et al 1982] for leadproduction. The main consequence of the lack of detailed emissions data is that for the lattertwo industries the material balance is significant as concerns the major inputs/outputs flows,but only a first evaluation is possible to assess the environmental burden associated with airand water emissions. For that purpose, further efforts would be needed to evaluate wasteeffluents with higher accuracy.

Methodology for Exergy Calculations

In the following analysis only chemical exergy of the substances has been taken intoaccount, for reasons explained earlier. For purposes of resource/waste accounting at theindustry level the inherent error of this approximation is negligible. When attention is focusedon the overall processes, most of the materials involved in the production enter or leave thesystem boundaries at normal pressure and temperature so that physical exergy can beneglected and total exergy content of the main streams can be equated to the chemical exergycontent.

On the other hand it has to be stressed that when attention is focused on particularprocesses inside the system boundaries, the overall results cannot be used as second-lawefficiency indicators at the plant level, as the thermal inefficiency cannot be evaluated withoutconsidering the contribution of physical exergy at some stages of the process (particularly thehandling of molten materials).

The following methodology has been used in order to calculate chemical exergy of pureand complex substances. For pure chemical substances (i.e. substances composed by a simpleconstituent) tabulated values for standard chemical exergy have been directly used. Forcomplex substances (i.e. those substances or materials formed by several different purecompounds) chemical exergy can be computed by means of a simple formula [Szargut et al1988]:



Bch = Ifi tili

where:

Bch standard chemical exergy of the complex substance [kJ/kg]ti Szargut coefficient for the i-th pure substance [kJ/kg]zi mass fraction of i-th pure substance in the complex substance

Szargut coefficients are usually coincident with standard chemical exergy of the puresubstances, except when the latter have particular chemical bonds in the complex substanceunder consideration. When Szargut coefficients were not available, standard chemical exergieshave been directly utilized instead of them. Only for melted pig iron and steel Szargutsuggests the use of a slightly different expression, viz.

Bch = BFe + litizi

where:

B,,, standard chemical exergy of the complex substance [kJ/kg]BF, standard chemical exergy of iron [kJ/kg]ti Szargut coefficient for the i-th pure substance apart from iron [kJ/kg]zi mass fraction of i-th pure substance in the complex substance.

As already remarked by Wall, in the case of mixtures this is a crude approximation,inasmuch as the entropy of a mixture is not the sum of the entropy of its constituents, butdepends on a number of factors which is difficult to evaluate [Wall 1986]. However, therelative error is usually small and can be neglected.

Chemical exergies of the major inputs and outputs as well as byproducts and wastes forthe processes under consideration have computed using the above formulae. Particular carehas been taken to calculate chemical composition of the substances entering and leaving thesystem boundaries. Where not available in literature, the composition has been estimated.

Unfortunately values of chemical exergy are found in literature only for a limitednumbers of pure substances. However, when not available, the chemical exergy content of anypure substance can be computed by means of an approximate formula, viz.

Bch = AGE- + 1 i Ni bi

where

AG; standard Gibbs free energy of formation of the compoundbi chemical exergy of the -ith pure element of the compoundNi molar fraction of the -ith pure element of the compound

The Gibbs free energy of formation is available in standard reference sources for a largenumber of chemicals, so that the previous expression can be easily applied to calculate exergyonce the chemical composition of the substance is known.

A list of the major substances involved in the selected processes, together with theirchemical composition and chemical exergy content is presented in the Appendix.



A few words must be spent on the case of fuels and utility energy, for which a differentprocedure has been adopted in order to estimate the exergy content. Two types of energyinputs have been accounted in the model: electricity and fuels. Data for both of them wereusually given in energy units (gJ or BTU per ton of final product). In the case of fuels —namely coal, fuel oil and natural gas — the exergy content has been estimated by multiplyingthe net heating value by an appropriate coefficient. This coefficient can be used to estimatethe chemical exergy content of any energetic resource once its heating value is known[Szargut et al 1988]. It is a function of the ratio C/H and C/O of the substance and it isusually higher than 1, as it takes into account the contribution of diffusion. A short list offuels with their net heating value and chemical exergy content is presented in Table I.

An exergy content has also been attributed to electricity. For convenience, the exergycoefficient of electricity has been assumed to be equal to 1, so that 1 kJ of electrical energycorresponds to an exergy flow of 1 kJ. This assumption could lead to some incongruences,as electricity, which is "pure" exergy and is certainly the most useful form of energy, has aquality factor smaller than some of the fuels that are used for its production. For this reason,it would seem more appropriate to account for electricity in terms of the chemical exergycontent of the fuels used for its production, thus taking also into account the efficiency of theconversion process. Moreover this procedure would enable us to evaluate all the energy inputsin terms of primary resources.

Finally, the contribution of some energetic materials has been disaggregated, dependingupon whether the substance was used as a raw material (the case of coke in the blast furnace)or burned to produce heat. In the flow diagrams and in the tables values are expressed inmass units for raw materials, and in energy units for fuels. In the exergy balance they areboth accounted in terms of exergy units (MJ) but their contribution was still consideredseparately.

The Steel Industry

Process description

Many different processes are used in the steel industry, depending on availability offeedstocks (scrap and ores), the availability of different fossil fuels — especially good qualitycoking coal — and on the mechanical and chemical properties needed for the final products.The model here analyzed is comprehensive of the four major processes used in the US forthe production of steel, viz., Basic Oxygen Furnace, Electric Arc Furnace, Open HeartFurnace and Direct Casting of melted iron.

However, special attention has been paid to the production of carbon steel from pig ironin a basic oxygen furnace (BOF), as this sequence still plays a major role in the world steelindustry. It accounted for 53% of US production in 1988, the remainder being almost entirelysecondary (scrap) steel reprocessing via electric furnaces. (The old open hearth process wasalmost completely phased out by 1988 and has since been replaced completely, at least in thewestern countries). 2 Thus the BOF chain can be considered as representative of the wholeprimary steel industry [Tellus 1992].

No data were available to estimate the airborne and waterborne emissions from theprocess of steel production in the Open Hearth Furnace and Direct Casting. As a matter offact, these sequences account only for a small fraction of the overall production, therefore the


relative error can be considered negligible.The overall steel chain — from ore to ingot — involves five major steps, namely: ore

mining, pelletizing or sintering, pig iron production, and, finally, conversion of pig iron tosteel. Two additional processes have been included in our model, to take in to account cokingand lime production (calcination of limestone). A simplified flow diagram of the wholeindustry (US, 1988) was presented in Fig 2. However, for purposes of exergy analysis, a moredetailed scheme is helpful, as illustrated in Figure 6. Quantitative data are referred to theproduction of the 1988 "average" ton of steel, as a mix of BOF (53 %), EAF (33%), OHF(5%) and direct casting products (9%).

Brief process descriptions follow:

Iron ore mining and beneficiation: iron is mostly found in nature in the form of oxides(hematite: Fe203, magnetite Fe304). The iron content of the material mined may varyappreciably. A large part of the iron ore produced in the U.S. consists of taconite, a low gradeore with a 25% average iron content. Globally, 1336 kg of crude ore is extracted to produce1 ton of steel. As already remarked, this number includes the steel produced in the EAfurnace, which uses only scrap as feedstock. Therefore, when not considering the latterprocess, the amount of material required would be considerably higher. Of the materialmoved, 436 kg. is discarded as overburden and left at the mine site. The remaining materialfrom the mine is crushed ground and concentrated. This process, known as beneficiation, isneeded to separate the iron rich material from the useless gangue. It is usually accomplishedby means of magnetic separation or flotation. The residues from beneficiation are usuallydisposed of in ponds. The concentrates resulting from this process (866.4 kg) are conveyedeither to the pelletizing process (657 kg) or to the sintering machine (144 kg). A smallfraction of the concentrate (55 kg) is directly used in the blast furnace. In calculating theexergy flows, the contribution of overburden has been neglected as no accurate data wereavailable on its composition. However the relative error is certainly negligible.

Pelletizing and sintering: Pelletizing and sintering are agglomeration processes used to makethe iron ore fulfil the specific characteristic required by the blast furnace, especially in termsof particle size and hardness. In the pelletization process iron concentrate is mixed with abinder (usually bentonite) then hardened in a furnace. Sintering is also a form of recycling:it is used to process fine particles which are too small to pelletize or be directly charged inthe blast furnace. Iron pellets and sinter constitute the major constituents of the blast furnacecharge (645 and 165 kg per ton of steel, respectively).

Blast furnaces: The blast furnace is a large cylindrical tower lined with refractory bricks(based on chromite). It can be considered as the real "core" of steelmaking. The furnace ischarged with a batch of iron ore (sinter and pellets) coke and limestone. Approximately 283kg of coke and 80 kg of limestone are required for the production of 1 ton of steel in ourmodel process chain. During the reduction process the hot coke gases remove oxygen fromiron oxides in the top part of the furnace, while in the high-temperature bottom part of thefurnace the molten iron, mixed with a compound of iron and carbon is collected together witha slag. The latter contents silicon oxide and lime, together with small amounts of magnesiumoxide, and plays an important role to control the sulfur content of the liquid metal — called"pig iron". Pig iron and slag are periodically removed from the furnace. By-product off-gases(called blast furnace gas), consisting mainly of carbon monoxide and nitrogen, are usuallyrecovered and used as low heating value fuel for preheating the blast furnace feeds, or forother purposes. Finally, approximately 358 kg of CO2 per ton of steel are discharged into the



atmosphere at this stage of the process.

Steel production: The hot liquid pig iron (94% iron, 5% C) is converted to steel in the basicoxygen furnace (BOF) or in the open hearth furnace (OHF), while a small fraction is directlycasted without being further processed. The blast furnace is charged with molten pig iron (905kg/ton of BOF steel or 478 kg/ton of "average" steel) and steel scraps (302 or 159 kg/tonrespectively). Oxygen is added (hence the furnace name) to oxidize carbon and siliconimpurities (100 or 53 kg per ton of each type of steel respectively) without contaminationby atmospheric nitrogen.3 Lime and fluorspar are also added to form a fluid slag with theimpurities. It floats on the liquid steel, and is removed. The oxidation process is globallyexothermic and no external heat is needed. The liquid steel is then casted in to ingots orconveyed to a continuous casting machine. The amount of CO2 released by the BOF is about150 kg per ton of steel. The latter value has been estimated assuming that all the carboncontained in the pig iron is removed and converted into carbon dioxide. When calculating theexergy content of steel, no differences have been considered among the different types ofsteel. As a matter of fact, despite of the great importance that carbon and other alloy elementsrepresent for determining mechanical properties, a small variation of their content cannotstrongly affect the overall exergy content of the metal.

Lime and Coke production: Extraction and processing of limestone and coal are not actuallypart of the primary production process of steel. However, coke produced today is almostexclusively produced by steel companies and used in steelmaking: for each metric ton of steel,283 kg of coke is consumed by the blast furnace and 8 kg of coke breeze (dust) are consumedby the sintering machine. The steel industry also consumes a significant quantity of limestoneper metric ton of steel: 24 kg is consumed by the sinter process, 80 kg is consumed by theblast furnace and 82 kg is calcined to produce 39 kg of lime, used in the BOF. Hence therelated emissions should reasonably be attributed to the final steel product. This is consistentwith the "one step back rule" adopted in [Tellus 1992].

Exergy and exergy flows:

The main exergy flows of the substances entering or leaving the system boundary arepresented in Figure 7. The exergy analysis shows that the low-exergy content iron ore isprogressively upgraded in exergy terms during the process. At the same time, of course, alarge amount of exergy — mainly provided by coal — is lost as low grade heat or in processwastes and emissions. The largest loss occurs in those steps which involve a combustionprocess. About 10000 MJ of exergy per ton of steel are consumed in the steelmaking process.Most of this is discharged to the environment as low grade heat, most of the remainder beinglost in the form of carbon dioxide. Emissions of carbon dioxide have been also included: inspite of their relative importance in terms of mass, they account for a smaller exergy content.The emissions into water are considerably smaller, but cannot be completely neglected: themain contribution is due to ammonia (7.9 MJ) to phenol (about 5 Ml) and to the heavyhydrocarbons contained in the lubricants used in the plant (about 2 MJ).

It is worth stressing that, because of the extreme complexity of the steel industry, a largenumber of different possibilities can be used to represent the gas streams which are rejectedto the environment. We have decided to take account of such contributions at a disaggregatedlevel, i.e. by calculating the exergy content of the airborne emissions contained in the gasflows. In fact, on the one hand all available data were presented in such a format, and, on the



other hand, the complex recovery of coke oven and blast furnace off gas cannot be correctlyrepresented when analyzing the whole industry at a national level, because of the largetechnology differences occurring among different mills.

This approach only neglects the contribution to exergy due to atmospheric nitrogen, sothat the relative error can be certainly considered small.

The overall exergy flows of major material and energy inputs, outputs, by products andwastes are listed in Table II, and schematically presented in Figure 8.

The Aluminum industry

Process description

Aluminum never occur in nature in metallic form as it strongly tends to combine withother elements. Its most important mineral is bauxite, which contents 40% to 60% ofaluminum oxide (Al203). The production process of aluminum involves three major processes,namely: bauxite mining and processing, alumina production and aluminum production. A flowdiagram with material and energy balance is presented in Figure 9. Material inputs andoutputs for the process were adapted from [TME 1995] and from [Tellus 1992].

Bauxite mining: Bauxite is a mixture of aluminum oxide (A1,03) other oxides, such as Fe203,Si02, TiO2 and combined water. It is mostly mined by thy: open pit method; the largestdeposits are usually located in tropical regions. After mi n ng, the bauxite ore is crushed,washed and screened to separate clay and useless impuriti,,s, then it is dried in a rotary kiln,and conveyed to mills to be processed. Average composition — adapted from [Hall et al1975] — indicates an high content of aluminum oxide, mono- and trihydrate. The exergycontent of the material handled has been made equal to the exergy content of the bauxite ore,as the contribution of the overburden is certainly negligible.

Alumina production (Bayer process): In the standard Bayer process the hydrate aluminumoxide is transformed in to dehydrated alumina (Al 203). Major inputs of the process are bauxiteore (4788 kg/ton of aluminum), caustic soda (429 kg/ton of aluminum) and lime (87 kg/tonof aluminum). Row material are mixed together then conveyed to the digester, where hotsteam is added to the mixture. Due to the high pressure anq heat of the solution, the availablealumina dissolves in the caustic soda forming sodium aluminate (NaA102). The purifiedsolution of sodium aluminate, called green liquor, is further processed to remove suspendedsolids and other insoluble impurities, and, finally, to produce aluminum trihydrate by meansof precipitation. The latter is next filtered, washed and calcinated to drive off water andproduce alumina. The pure alumina is then conveyed to the aluminum production plant, whilethe remaining liquor solution is processed to recover caustic soda. The residual of thisprocess, call red mud, is a batch of insoluble metal oxides which are further processed to bedisposed or recovered. Approximately 1850 kg of red mud are produced per each ton of purealuminum.

Aluminum production (Hall-Herouli process): Pure aluminum is produced by means ofelectrolytic reduction of alumina in the standard process known as Hall-Heroult process. The



reduction plant is constituted of a series of electrolytic cells made of steel. Each cell isequipped with a carbon cathode and with a consumable cathode made of petroleum coke andpitch (the contribution to exergy due to the anode rests has not been considered, as no datawere available to estimate their chemical composition). Purified alumina from the Bayerprocess (approximately 1900 kg/ton of aluminum) is dissolved in a molten bath constitutedof cryolite (Na3A1F6), fluorspar (CaF2) and aluminum fluoride (A1F3) which serves as anelectrolyte. When a low-voltage direct current passes through the bath, alumina is reduced topure metal aluminum and oxygen. Aluminum is then removed from the bottom of the cell andtransferred to the cast house, while oxygen reacts with the anode carbon, forming CO 2, CO,and a certain amount of sulfur dioxide. The specific emissions of carbon dioxide were notreported in [TME 1995], therefore they have been estimated from [Ayres & Ayres 1996,Chapter 3].

Exergy and exergy flows

A simplified diagram with exergy content of the main material inputs and outputs foreach stage is presented in Figure 10. Among the analyzed industries, the aluminum sector isresponsible for the largest exergy losses. This is mainly due the large amount of electricityneeded in the final step of refining as well as to the considerable quantity of fuel oil used inthe Bayer process. The high exergy content of the energetic inputs is not totally transferredto the final product, and it is almost entirely destroyed. When focusing attention on wastesand emissions, it is easy to see that major emissions are discharged into air: carbon monoxideaccount for the largest exergy content, but also the contribution of carbon dioxide and sulfurdioxide seems to be relevant (364 and 295 MJ/ton of aluminum respectively). On the otherhand, emissions into water are of minor importance, amounting to for less than 20 MJ/ton ofaluminum. Figure 11 shows the overall exergy flows and the exergy content of water and airemissions. A list with the overall mass and exergy flows divided by category is also presentedin Table III.

The Copper Industry

Process description

Even though many systems have been carried out in the last decades for copperproduction — in order to improve both process efficiency and to reduce its environmentalimpact — the particular uses of this metal — such as electrical equipment production —require a high level of purity, which can be achieved only by means of a final electrolyticrefining. Thus, a standard process for the production of primary copper involves three majorstages, namely: ore mining and beneficiation, smelting and desulfurization and finalelectrolytic refining. The process presented in Figure 12 is derived from those ones analyzedin [Gaines 1980 and Ayres et al 1995].

Ore mining and beneficiation: Copper does not usually occur in nature in form of puremetal, but it is mostly found in sulfides, such as chalcopyrite (CuFeS2), chalcocite (Cu 2S) andbornite (Cu5FeS4). The copper content of ores hardly reaches 1%, so that large amounts of



inert material must be processed to obtain a small amount of copper (About 165000 kg ofmaterial are mined to produce 1 ton of refined metal). The main consequences of this fact arethe high specific energy consumption and the large amount of mine wastes to be stocked.Tailings have not been accounted for an exergy content, as their contribution is certainlynegligible. Beneficiation process involves two stages: sulfides ores are first crushed andground, then concentrated in froth flotation cells to 20-30% of Cu.

Smelting and desulfurization: In the smelting stage the 25% Cu copper ore is reduced tometallic copper containing about 98% Cu. Almost the total part of sulfur contained in theprocessed ore is converted into sulfur dioxide. The latter — emitted in the furnace off-gases— can be captured and conveyed to the acid plant where it is recovered as sulfuric acid(H2SO4). Although the chemical reaction is exothermic, the process is highly energy-intensive,as a large amount of fuel is required to heat the inputs materials to reaction temperature.

Due to the rapidly evolving technology, it is hard to identify a standard process forsmelting operations, as many types of furnaces are used, from fluid bed reactors, toreverbatory furnaces and continuous casting machines. The process usually starts with copperore smelting, whereby a molten copper and iron sulfide (known as copper matte) is produced.The latter is next processed and converted to produce metallic copper about 98% in Cu.

Electrolytic refining: Metallurgical copper (about 98% Cu)is not enough pure for electricalapplications, and need to be further processed. Molten copper from the smelting plant is castinto anodes which are placed in concrete cells, together with copper sheets as cathodes. Theelectrolyte is composed of copper sulfate (CuSO 4), sulfuric acid (H2SO4) and small quantitiesof additives. At the end of the process pure copper is collected on cathodes, together withsmall amounts of precious metals deposited as anode mud which can be easily recovered.


A simplified diagram with exergy content of the main material inputs and outputs foreach stage is presented in Figure 13. The exergy analysis shows that the copper industry haspartially the same features as the aluminum one, being responsible for the destruction of largequantities of exergy. This is due again to the process of electrolytic refining which is reallyenergy intensive. As already remarked further data are required in order to estimate emissionswith a sufficient degree of accuracy. Nonetheless, it is possible to argue that the majorpotential dangers for the environment are represented by the sulfur dioxide which is emittedto air during the smelting of copper concentrate. A list with the overall mass and exergy flowsof the process, divided by category is also presented in Table IV and illustrated in Fig. 14.

The Lead Industry

Process description

The process for the production of primary lead is constituted of five basic sub-processes,namely: ore mining and beneficiation, sintering, blast furnacing, drossing, final refining and



casting. Each sub process can be split again in many different operations. A general flowdiagram for lead production is shown in Figure 15. It is worth stressing that this diagram isonly an attempt to represent the whole chain of production "from ore to ingot" and cannotbe considered as the effective process of a plant really operating. Quantitative data wereobtained from many different sources [Burkle 1980; Morris et al 1982; Szargut et al 1988;OHMP 1977; US Bureau of Mines 1991] and integrated with gross estimations when notavailable in literature.

Ore Mining And Beneficiation: Lead is most found in nature as galena (PbS) which is itsprimary sulfide. Oxidized ores can also occur, such as anglesite (PbSO 4) and Cerussite(PbCO3) which are the weathered product of Galena. Lead ore often contains a certain amountof copper, iron and zinc sulfides or sulfates, as well as a little percentage of precious metalwhich can be recovered during the lead smelting. The average lead content of the materialmined is approximately 6% [US Bureau of Mines 1991].

The material mined has a low lead content, thus not making it suitable for a direct usein the lead smelter. Ore concentrating is then required before the smelter operations:concentrating is the process whereby the non-useful part of the material mined (gangue) isseparated from lead-containing ore. The lead-rich minerals (concentrate) are sent to thesmelter to be processed, while the fraction low in mineral content is discharged.

Lead ore is crushed, ground then concentrated by means of chemical and/or mechanicalprocess. In the chemical process ore dust is diluted in water, then treated with chemicaladditives that create a froth in which the mineral particles are floated from the gangue. In themechanical process separation is achieved thanks to the difference in specific gravity of thelead ore and the gangue particles. After concentration the lead content raises from 6% toabout 50% in weight.

Sintering: Sintering consists in the roasting of the ore concentrate mixtures. Metal sulfidesand sulfates are transformed in to oxides, more suitable to be processed in the blast furnacefor pure lead extraction. The sinter machine is charged with ore concentrate, limestone andsilica. Approximately 45% of the produced sinter is recovered and added to the charge inorder to control the sulfur content of the charge. The reaction is exothermic and it is usuallyignited by means of natural gas. About 85% of the sulfur is removed from the concentrateduring sintering under the form of sulfur dioxide. The latter is conveyed to the acid plantwhere it is used to produce sulfuric acid. No quantitative data were available to estimate theprocess yield. According to [Szargut et al 1988] about 25 kg of sulfuric acid are produced foreach tonne of sinter leaving the sintering plant. No data were available in order to estimatecomposition of the acid plant off-gases. Thus they were supposed to have the samecomposition as the sintering off-gases, except for sulfur dioxide, which has been consideredto be completely removed during the acid plant operations. In spite of the decrease in theexergy content of the material processed (due to the transformation of the metal sulfides intooxides) sintering is an important operation in order to obtain a material suitable forsubsequent processing in the blast furnace.

Blast furnacing: In the blast furnace lead oxides are reduced and transformed into metal leadbullion. the process is equivalent to that one already discussed for steel making. Accordingto [Morris et al 1982] about 4800 kg of sinter are needed to produce 1 ton of lead. The blastfurnace is also charged with coke (about 650 kg per ton of lead) and propane. The insolubleimpurities form a slag which floats over the liquid metal so that it can be easily removed.Slag is cooled and crushed, then partially recovered to the sinter machine. the remaining part



is discharged. Approximately 2900 kg of slag leave the blast furnace per each ton of lead andabout 50% of them are recovered. Other output streams are dust and off gases. Despite oftheir heat content, the latter are not usually reused in the process. Blast furnace utilities aresteam and electricity (approximately 1.5 Gj per ton of lead).

Drossing and final refining: Lead bullion as they leave the blast furnace need to be furtherprocessed and refined to fulfil the specific characteristics which are needed for industrialapplications. Therefore, smelted lead pass through a multi-step refining process, wherebymetal impurities are removed under form of slag. On that purpose silica and caustic soda (17and 78 kg per ton of lead, respectively) are added to the liquid metal, together with smallquantities of zinc, in order to accelerate slag formation. A matte rich in copper (coppermatte), a silver-gold alloy (known as dore) and small quantities of bismuth and zinc oxide arethe by-products of refining. Approximately 284 kg of copper matte are recovered per each tonof lead. No quantitative data were available to evaluate the amount of other by products aswell as the energy requirements for the last stage of refining.


A simplified diagram with exergy content of the main material inputs and outputs foreach stage is presented in Figure 16. The lead industry shows intermediate characteristicsbetween the steel and the aluminum ones. On one hand the exergy losses are considerablysmaller than those ones of the aluminum production, mainly because the refining processesare not based on electrolysis. On the other hand, a large amount of exergy is discharged tothe environment under the form of useless wastes, such as slug and dust. Even in the case oflead the lack of data does not allow to give an accurate evaluation of emissions. However themain contribution seems to come from sulfur dioxide emitted during the smelting of sulfideores.

The overall mass and exergy flows of inputs, outputs, by products and wastes is presentedin Table V and illustrated in Fig. 17.

Conclusions

The analysis clearly demonstrates how exergy can be successfully used for resource andwaste accounting purposes, as it enables to compare the systems under consideration on acommon basis, both to identify major inefficiencies and to provide a first evaluation of theirenvironmental burden.

A comparative list of the overall exergy flows for the selected metal industries ispresented in Table VI. Chemical exergy content of wastes, as well as absolute and relativeexergy losses per each sector are also computed and presented in this table: the relativeexergy losses represent the ratio between the absolute exergy losses and the exergy amountof the input streams. Finally, the cumulative exergetic efficiency of each process is expressedas the exergy content of useful outputs and by products, divided by the overall exergy contentof the input streams (raw materials and energetic inputs). Of course, a different definition ofefficiency - comprehensive of the waste streams and emissions - would lead to differentresults.

The data clearly indicates that the aluminum industry is characterized by a very large



exergy consumption, mainly because energy — intensive electrolysis is used as a finaloperation for refining. Exergy losses of aluminum industry, for instance, are about six timesthe losses of steel industry, thus emphasizing the need for a further technical improvementin the production processes currently under use. When considering the relative figures (i.e.values referred to the total amount of exergy in), such differences could appear less important,as the overall exergy requirements are remarkably high for the aluminum industry. In fact,about 100 gJ of exergy are required for the production of 1 ton of refined aluminum, whileonly 21 gJ are needed by the steel industry to produce the same quantity of metal.

On the other hand, when attention is focused on emissions, exergy analysis shows thatthe industry of steel is responsible for a larger environmental impact. This is especially dueto the difficulty of controlling air emissions, mainly under the form of carbon monoxide,which are released by the blast and oxygen furnaces.

The case of copper and lead industries represent an intermediate situation between steeland aluminum. The former industries shows the lowest exergy efficiency among the analyzedprocesses, but the absolute exergy consumptions are relatively smaller than those of thealuminum sector (less than 40 gJ per ton of metal in both cases). On the emission side, eventhough accurate data about the whole spectrum of pollutant were not available for theseindustries, one can state that sulfur dioxide emitted in the smelting process is more likely tobe the largest emission to air. For both copper and lead production, the exergy content of thissubstance has been accounted under the item "wastes and useless by-products".

In the case of lead production it was possible to estimate the overall flow of the blastfurnace off gas. However, instead of considering as an air emission the disaggregatedcontribution of each substances of the gas, we have preferred to include the overall exergycontent of the stream under the item "wastes". This choice was due to the fact that it was notclear whether the off gas was directly discharged to the atmosphere or further processed inany emission control system.

We have also calculated the overall exergy flows related to the selected metal industriesfor the year 1988. . The results, presented in Figure 17 and expressed in PJ (Petajoules, 1PJ= 1015 joules), have been computed by multiplying the specific exergy flows referred to theproduction of 1 ton of each metal by the actual production of primary metal occurred in thatyear.

The figures show that, in spite of a better relative efficiency, the steel industry is globallyresponsible for a bigger impact on the environment, because of the larger amount of materialproduced (about 99 MMT of steel vs. less than 6 MMT of non-ferrous materials in 1988).The main consequence of this fact is that, on a global scale, even small relative improvementsin the technology of steel production will have a significant effect for the reduction ofresource use and emissions of dangerous substances.

Endnotes

1. Assuming the iron in ore is mostly in the form Fe 203, the 57.5 MMT of iron content in ore (1988) would becombined with 25.55 MMT of oxygen.

2. The Russian and Chinese steel industries still use the open hearth process to a significant degree.

3. The BOF is very similar in concept to the old Bessemer process, except for the fact that the Bessemer processoxidized the excess carbon in the pig iron by blowing air through the molten pig iron. This procedure took placevery rapidly, and was difficult to control. If not done exactly right, the Bessemer process left trace quantitiesof nitrogen in solution in the steel, which made it brittle.


References

[Ahrendts 1980] Ahrendts, J., "Reference States", Energy 5(5), 1980 :667-677.

[Altenpohi 1982] Altenpohl, Dieter: Aluminum Viewed from Within, Aluminium-Verlag,Dusseldorf, 1982.

[Ayres & Ayres 1996] Ayres, Robert U. & Leslie W. Ayres, Industrial Ecology: Closing theMaterials Cycle, Edward Elgar, Aldershott, UK, 1996.

[Ayres & Ayres & Martinas 1996] Ayres, Robert U., Leslie W. Ayres & Katalin Martinas,Eco-Thermodynamics: Exergy & Life Cycle Analysis, Working Paper (96/./EPS),INSEAD, Fontainebleau, France, January 1996.

[Ayres & Martinas 1995] Ayres, Robert U. & Katalin Martinas, "Waste Potential Entropy:The Ultimate Ecotoxic?", Economie Appliquie XLVIII, 1995 :95-120.

[Ayres et al 1995] Ayres, Robert U., Leslie W. Ayres, Paolo Frankl, Howard Lee, NicholeWolfgang & Paul M. Weaver, Materials-Optimization in the Production of MajorFinished Materials, Final Report (Contract BRE2-CT93-0894), Center for the Manage-ment of Environmental Resources of INSEAD, Fontainebleau, France, January 25, 1995.

[Battelle 1975] Battelle - Columbus Laboratories, Evaluation of the Theoretical Potential forEnergy Conservation in 7 Basic Industries, (PB-244-772), Battelle - ColumbusLaboratories, Columbus OH, June 1975. (Prepared for U. S. Federal Energy Administra-tion)

[Battelle-a 1975] Battelle - Columbus Laboratories, Energy Use Patterns in Metallurgical &Nonmetallic Mineral Processing (Phase 5 - Energy Data & Flowsheets, IntermediatePriority Commodities), Interim Report (S0144093-5), Battelle - Columbus Laboratories,Columbus OH, September 16, 1975. (Prepared for U. S. Bureau of Mines)

[Battelle 1975] Battelle - Columbus Laboratories, Energy Use Patterns in Metallurgical &Nonmetallic Mineral Processing (Phase 4 - Energy Data & Flowsheets, High PriorityCommodities), Interim Report (S0144093-4), Battelle - Columbus Laboratories, ColumbusOH, June 27, 1975. (Prepared for U. S. Bureau of Mines)

[Burkle 1980] Burkle, J. 0. Industrial Process Profiles for Industrial Use: Chapter 27,Primary Lead Industry. Prepared for Industrial Environmental Research Lab., U.S. Dept.of Commerce, Cincinnati 1980.

[Gaines 1980] L. L. Gaines: Energy and Material Flows in the Copper Industry. TechnicalMemo, Argonne National Laboratory, Argonne IL, 1980. [Prepared for the U.S.Department of Energy]

[Gyftopoulos et al 1974] Gyftopoulos, E. P., L. J. Lazaridis & T. F. Widmer, Potential FuelEffectiveness in Industry (Series: Ford Foundation Energy Policy Project), Ballinger


Publishing Company, Cambridge MA, 1974.

[Hall et al 1975] Hall, E. H., W. H. Hanna, L. D. Reed, J. Varga Jr., D. N. Williams, K. E.Wilkes, B. E. Johnson, W. J. Mueller, E. J. Bradbury & W. J. Frederick, Evaluation ofthe Theoretical Potential for Energy Conservation in Seven Basic Industries, Final Report(PB-244,772), Battelle Columbus Laboratories, Columbus Ohio, July 11, 1975. [forUnited States Federal Energy Administration]

[IDEA 1991] Liibkert, Barbara et al, Life Cycle Analysis: International Database forEcoprofile Analysis (IDEA), Working Paper WP-91-30, International Institute for AppliedSystems Analysis, Laxenburg, Austria, 1991.

[Morris et al 1982] D. R. Morris, F. R. Steward, P. Evans: Energy Efficiency of a LeadSmelter. Energy — The International Journal, vol 8, No. 5 1983. Pergamon press.

[OHMP 1977] Operating Handbook of Mineral Processing, vol. 2 pp. 299-303. Herculaneum:tops in US refined lead output. R. Thomas (ed), McGraw-Hill, New York, 1977.

[Rant 1956] Rant, Z., "Exergy, a New Word for Technical Available Work", Forsch. Ing.Wis. 22(1), 1956 :36-37. [in German]

[Russel &Vaughan 1976] Russell, Clifford S., Vaughan, William J., Steel Production:Processes, Products, and Residuals. The Johns Hopkins University Press, 1976.

[SAI 1985] Science Applications International Corporation, Summary of Data on IndustrialNonhazardous Waste Disposal Practices, EPA Contract (68-01-7050), ScienceApplications International Corporation, Washington DC, 1985.

[Szargut et al 1988] Szargut, Jan, David R. Morris & Frank R. Steward, Exergy Analysis ofThermal, Chemical, & Metallurgical Processes (ISBN 0-89116-574-6), HemispherePublishing Corporation, New York, 1988.

[Tellus 1992] Tellus Institute, CSG/Tellus Packaging Study: Inventory of Material & EnergyUse & Air & Water Emissions from the Production of Packaging Materials, TechnicalReport (89-024/2), Tellus Institute, Boston MA, May 1992. [prepared for the Council ofState Governments & the United States Environmental Protection Agency]

[Tellus 1992a] Tellus Institute: Assessing the Impacts of Production and Disposal ofPackaging and Public Policy Measures to Alter its Mix. Boston, May 1992.

[TME 1995] Operational Indicators for Progress Towards Sustainability. Institute for AppliedEnvironmental Economics (TME), internal report, 1995.

[US Bureau of Mines 1991] Minerals Yearboolg volume 1, Metals and Minerals. U.S. Dept.of the Interior, Bureau of Mines, Washington D.C. 1991.

[USEPA 1991] United States Environmental Protection Agency Office of Solid Waste, 1987National Biennial Report of Hazardous Waste Treatment, Storage & Disposal FacilitiesRegulated Under RCRA, NTIS PB-87-114369 (EPA-530-SW-91-061), United States



Environmental Protection Agency Office of Solid Waste, Washington DC, 1991.

[USS 1971] The Making Shaping and Treating of Steel. H. E. McGannon Editor, UnitedStates Steel, Pittsburgh 1971.

[Wall 1977] Wall, Goran, Exergy: A Useful Concept within Resource Accounting, (77-42),Institute of Theoretical Physics, Chalmers University of Technology & University ofGoteborg, Goteborg, Sweden, 1977.

[Wall 1986] Wall, Goran, "Exergy Conversion in the Swedish Society", Energy 15, 1986:435-444.

.1:

Chemicals,Net energy

imports 'Overburden,

%■. /

Concentrate Coke, fluxes,carbon anodesNet

imports PelletsGangue, 1,

1 tailingsAlu-mina

Pel-lets

• •

CO2 .1SO2 I

Slag, \smelter Iwaste

Crude metal AlloyingelementsNet

imports Alumi-num

Pigiron

Ingotcopper

Refinedmetal or alloy

Netimports


Domesticuse

Figure 1.. Metals processing relationships

28.22

BASIC'XYGEAIJRNA`C

64.59

.91

BOILERS& GASI- COKE

FICATION OVEN &k B.F. GASPANTS

All material amounts adjusted from stock to produce final output amounts as if all production occurred in 1988.* Actual 1988 production = 50.53 MMT

COALMINING

BOILERFUELXXX

COKINGCOAL40.28

COKINGLECTRICURNACE

2.00STAIN

LESSSTEEL

2.00

8.89 SINTER16.42

16.42BLAST

FURNACE

LIME-STONE &

DOLOMITE9.89

15.88

PELLET-

IZINGSTEELASTINGS

0.91(stock) SCRAP

IRONIRON ORE

(86.3imputed) 3.27

IRONCASTINGS

8.07

FART7RNAC

CASTIN

COKEBREEZE

1.81i IRON \

•REVERTS(recycle) .;■

iviasirn ac u. Ivrea IL.zergy anatysts of metal tnaustrtes el:Itetnner .50, 199t rage Z4

Figure 2.. Process-product flows: inputs needed for U.S. 1988 ironlsteel production (MMT)

Anode mud Copper(25% Cu et al) salts

0.006 .0007

Tailings 0.138% Cu160.7

Wastewater

V

! gas, dust 0.35% Cu0.334

ORE3ENEFICIATION

82.5%RECOVERY

OF Cu

SMELTING

DESULFUR-IZATION

98%RECOVERY

OF Cu

1302 r SULFUR •— BPI DIOXIDE I

SO2

ELECTROLYTICREFINING

111 IMMO •



Electricity Fuel

Electricity Fuel

2703 kWh

15.26 MBTU

221 kWh 2.65 MBTU

• •=1•1• 01.■ •■ IMMO le MM/1M .I■1 1MM IM

Figure 3.. Composite unit process: copper

Overburden103

Concentration CO 2wastes

140

IRON ORE

MININGOre

CONCENTRATIONPELLETIZING PIG IRON

U.S. ores, pellets,& sinter 57.

SMELTING198 NODULIZING Imports .24.2•

•Scrap

:.•

N.B. Scrap consumption in iron & steel production is probably underestimated by up to4 million tonnes. Recirculated scrap may be underestimated by a similar amount.

CO2 Iron87.1 slag

14.2

Pi iron 50.9

(94%0 C) 52.8

STEELREFINING,PICKLING,

ETC.

Steel products:!;1.,:,:n.SED5::-• 90.65

culated(internal)scrap9.0 est

5.0

Obsoletescrap46.5 est

Ferrous SteelCO2 slagsulfate0.3 11.0 5.2

A. Masini & K. U. Ayres Exergy analysis of metal industries September 30, 1996 Page 26

Air, Lime Pureoxygenoxygen(Ca0)

Ferroalloys31.5 8. est1.0Coke,

limestone

Limestone,fluxes Coke

9.6 26.7

Figure 4.. Mass flows in the U.S. iron & steel sectors, 1988 (MMT)

Air-Oxygen

11.1Limestone,coke5.0

NON-FERROUS

METALMINING

Overburden862.2

\4_15_10.

Crude ores

467.9

FLOTATION

CONCENTRATION

ChemicalsH 2 SO 4 (for leaching)cyanide,flotation agents

Explosives

1.87

Concentrates 11.05

Imports 3.91

6.0

SO2N.F. slags & 0'3dry wastes

CO2 3.018.3

Concentrationwastes464.7


Aluminium 3.944

CopPer 1.406Lead 0.392Zinc 0.241

Sulfuric acid3.5

Figure 5.. Mass flows in the U.S. non-ferrous metals sector, 1988 (MMT)

FUELELECT. OIL

84 MJ

ELECT.

109 MJ

431ELECT. NAT. GAS

53.3 MJ 935 MJ

AIR

WASTES

ELECT. COAL

60 MJ

AIR

WASTES

NAT.ELECT. GAS

810 MJ 139 MJ

AIR WATER

WASTES WASTES

ELECT. COAL

14 MJ j 330.4 MJ

95

A. iviasini 6L K U. Ayres Exergy analysts of metal tnaustnes eptemper su, 1 WU rage zts

AIR

WASTES

Figure 6.. Steel production: mass balance [kg/ton of steel] and energy inputs [MJ/ton of steel]

FUELELECT. OIL

84ELECT.

109

ELECT. NAT GAS

53 973

NAT.GAS

810.7 108.2

88.34 0.611

AIR WATER

WASTES WASTES

ELECT. COAL

14.3 330.4

1.4



Figure 7.. Steel production: chemical exergy flows of major inputsloutputs [MJIton of steel]

STEEL PRODUCTIONAMMONIUM SULF. 20.2LIGHT OIL 128NAPHTALENE 1TAR 633

OVERALL

BLAST FURN. SLAG 227STEEL FURN. SLAG 68

AIR

EMISSIONS1122

WATER

EMISSIONS

17

AMMONIA 7.9

BOD n.a.

CHLORIDES 0.06

CYANIDES 0.58

HIDROFLUORIC ACID 0.65

PHENOL 5.32 10.3

OIL AND GREASE 2.62


FUEL OIL 557ELECT. 11243NATURAL GAS 1136

SCRAPS 3954.4

FLUORSPAR0.2

ORE 118.3

CALCIUM 0.56IDROXIDE

OXYGEN 10.3

MANGAN. 19.9

BENTON. 12.4

OIL 4172

COAL 10872.3

LIMEST. 2.8

AMMONIA 11.9BENZENE 15.6

CARBON DIOXIDE 248CARBON MONOXIDE 216

VOC 574MANGANESE 0.08

NOx 1.8PARTICULATE 63

SOx 15.7TOLUENE 4.7

TSP 27

Figure 8.. Steel production: overall exergy flows and chemical exergy content of emissions [MJIton of steel]

FUEL NAT.

ELECT. OIL GAS COAL

2334 MJ 1805 MJ 63 MJ 34 MJFUEL

ELECT. OIL

1594 MJ 16437 MJ

AIR

WASTES

FUELELECT. OIL

434 MJ +2432 MJ

DIESELELECT. FUEL

54000 MJ 3800 MJ

AIRWASTES

AIR WATER

WASTES WASTES

NAT.ELECT. GAS

;886 MJ 1261 MJ

DIESEL NAT. FUELELECT. FUEL GAS OIL COAL

i 1MJ 3 Mi 154 Mi 1 MJ AIR WATERWASTES WASTES

AIR

WASTES

AIR WATER

WASTES WASTES



Figure 9.. Aluminum production: mass balance ficglton of aluminum] and energy inputs [MJ/ton of aluminum]

36.6

AIR WATERWASTES WASTES

NAT.ELECT. GAS

;381 ;1311.4

FUELELECT. OIL

+159417636.9

ELECT.

_17

32.7

AIR

WASTES

FUELOIL

567.

DIESELELECT. FUEL

54000 4066

•

AIR

WASTES

2419.2 17.4

AIR WATER

WASTES WASTES

DIESEL NAT. FUELELECT. FUEL GAS OIL COAL

4 1 i3.2 11.0 t165.2 1.1

16.7

AIR

WASTES 70.9 0.03

AIR WATER

WASTES WASTES

NAT.GAS

32334 1936.8 65.5 337.0

FUELELECT. OIL COAL


Figure 10.. Aluminum production: chemical exergy of major inputsloutputs [MJIton of aluminum]

36776AMMONIA 0.26COD n.a.FLUORANTHENE 4.2CYANIDES 5HYDROFL. ACID 0.65

AIREMISSIONS

3136

WATEREMISSIONS

17

ALUM.FLUOR. 20.4

BAUXITE ORE 5430

COAL 1745

COKE 269PETR.COKES 11593

ROCK SALT 170

LIMESTONE 2.3

CARBON DIOXIDE 364CARBON MONOXIDE 2136

VOC 81NOx 2.0

PMIO 46.5

SOx 295.3CHLORINE 32.7

TSP 177.5

ALUMINUM PRODUCTION

OVERALL

ALUMINUM 32860

RED MUD 763



FUEL OIL 20306ELECT. 58317NATURAL GAS 1378DIESEL FUEL 4070

Figure 11.. Aluminum production: overall exergy flows and chemical exergy content of emissions [Milton of aluminum]

9730 MJ 16100 MJ 796 MJ 27% MJ

SMELTINGAND

DESULFURIZ:

A. Masini & H. U. Ayres Exergy analysis of metal industries September 30, 1996 Page 34

Figure 12.. Copper production: mass balance [kg/ton of copper] and energy inputs IM,/ton of copper]


Figure 13.. Copper production: chemical exergy of major inputs/outputs IMJIton of copper]

OLD SCRAPS 528COPPER ORE 3854

FLOTATION REAGENTS 23.2STEEL BALLS&RODS 1092

EXPLOSIVES 1233SILICA 26

LIMESTONE 3.2LIME 784

COPPER 2111

SLAG FROM SMELTING 3533SLAG FROM REFINING 2TAILINGS n.a.

EXPRGY:LOSSFS.:- .26200•


FUEL OIL 20306ELECT. 10526

COPPER PRODUCTION

OVERALL

COPPER 2.5SOx 98

OTHERS n.a.

AIR WATEREMISSIONS EMISSIONS

100.5 n.a.Figure 14. Copper production: overall exergy flows [liff Iton of copper]

Natural gas799 MJ

Natural gas Oil

t29 MJ 169 MJ

Electricity

Steam Electrkity153 1538 MJ

' •MINING••AND.

RENEFICIA.

• TION.•

Oil Electricity

IF



Figure 15.. Lead production: mass balance [kg/ton of lead] and energy inputs IMJIton of lead]

Electricity

Natural gas 011

1

4398.1 1469

Steam Electricity

153 1538

Oil Electricity

BENEFICIA--T1ON

Natural gas Electricity

1

830.96 808

A. masini u. Ayres Ixergy anatysts of metal maustrtes september su, 1990 rage szs

Figure 16.. Lead production: chemical exergy of major inputs/outputs [MJIton of lead]

COPPER MATTE 1484DROSS 31.7SULFURIC ACID 6459

DUST FROM SINTERING 697

BLAST FURNACE SLAG 2095TAILINGS 5175OFF GAS 7637


FUEL OIL 503ELECT. 2346NATURAL GAS 5229

LEAD ORE 8077

FLOTATION REAGENTS 568

CAUSTIC SODA 146PROPANE 1221

SILICA 36

LIMESTONE 22COKE 20340

PARTICULATE 25OTHERS n.a.

AIR WATEREMISSIONS EMISSIONS

25 n.a.

Figure 17. Lead production: overall exergy flows [MJ/ton of lead]

LEAD PRODUCTION

OVERALL

NON-FERROUS

METAL

INDUSTRY

EXERGY LOSSES

BY PRODUCTSWASTES

ALS



Figure 18. US metal industries: 1988 overall exergy flows [PJ] 1PJ = 1015J

1.088 21680 23587.841.06 28300 299981.073 39500 42383.51.04 44000 457601.07 39500 42265

FuelCoalCokeFuel oilNatural gasDiesel fuel

Ex. coeff. Net heat. value Chemical exergy!1I/k$1

[KJIke


Table I: Chemical exergy content of some fuels

A. Masini & R. U. Ayres Exergy analysis of metal industries September 30, 1996

Table Steel production: overall mass and exergy flows

Page 42

Substance Chem. exergy Mass in/out Exergy TOTALIkg or kJIken [kg/ton of steel] [MJ/ton of steel] [MJ/ton of steel]

IN

MATERIAL Bentonite 766.19 16.21 12.42 19407.79Calcium hydroxide 724.75 0.77 0.56Coal 23583.00 461.02 1087231Coke 29912.00 8.00 239.32Fluorspar 146.01 1.35 0.20Fluxes n.a. 20.15 0.00Iron ore. earth 88.58 1335.67 11831Limestone 12.69 218.12 2.77Make up oil 37450.00 0.12 4.49Manganese 8778.99 2.27 19.94Oil 39500.00 105.64 4172.78Oxygen 124.07 82.91 10.29Scraps 6958.89 568.25 3954.41

ENERGY Electricity 1.00 112337 1123.37 2315.06[MJ] Fuel oil 1.07 51.92 55.71

Natural gas 1.04 1092.29 1135.98OUTMAJOR Steel 6958.89 -919.80 -6400.79 -7063.96

Iron. castings 8272.04 -80.17 -663.18BY-PRODUCTS Ammonium sulfate 4999.29 -4.05 -20.25 -783.08

Light oil 35000.00 -3.66 -128.10Naphthalene 40999.42 -0.03 -1.23Tar 35000.00 -18.10 -633.50

WASTES Blast furnace slag 1481.98 -153.25 -227.12 -294.77Overburden 0.00 -435.56 0.00Steel furnace slag 1300.67 -52.02 -67.66Water waste 0.00 -1265.36 0.00

AIRBORNE Ammonia 19853.08 -0.60 -11.92 -1122.09EMISSIONS Benzene 4223939 -0.37 -15.61

Carbon dioxide 451.49 -439.76 -24833Carbon monoxide 981934 -22.03 -21633Flue dust n.a. -16.38 0.00NOx 1209.20 -1.47 -1.78PM10 634.82 -9.94 -6.31SOx 4893.17 -3.21 -15.68Toluene 42797.08 -0.11 -4.75TSP 634.82 -42.95 -27.27VOC 4223939 -1359 -574.01

WATERBORNE Ammonia 19853.08 -0.40 -7.90 -17.37EMISSIONS BOD n.a. -1.02 0.00

Chlorides 1252.04 -0.05 -0.06Cyanides 32477.70 -0.02 -0.58Hydrofluoric acid 3998.73 -0.16 -0.65Oil and grease 37450.00 -0.07 -2.62Phenol 33241 56 -0.16 -532Sulphates 1138.84 -0.20 -0.23TSS n.a. -24.38 0.00


Table III: Aluminum production: overall mass and exergy flowsSubstance Chem. exergy

[kJ/kg orkJIkJ]

Mass in/out[kg/ton of

aluminum]

Exergy in/out[Milton ofaluminum]

TOTAL[M./t

Date post:	15-Feb-2021
Category:	Documents
Upload:	others
View:	1 times
Download:	1 times

AN APPLICATION OF EXERGY ACCOUNTING TO FOUR BASIC · 2008. 9. 30. · AN APPLICATION OF EXERGY...

Documents