STATE OF THE DEBATE Cement Manufacture and the...

S T A T E O F T H E D E B A T E

httpmitpress mit edujie Journal of Industrial Ecology 93

q Copyright 2003 by theMassachusetts Institute of Technologyand Yale University

Volume 7 Number 1

Cement Manufacture andthe EnvironmentPart II Environmental Challengesand Opportunities

Hendrik G van Oss and Amy C Padovani

Keywords

alternative fuelscarbon dioxideclinkergreenhouse gases (GHG)industrial symbiosisportland cement

Address correspondence toHendrik G van OssUS Geological Survey983 National CenterReston VA 20192 USAhvanossusgsgovhttpmineralsusgsgovminerals

Summary

Construction materials account for a signicant proportion ofnonfuel materials ows throughout the industrialized worldHydraulic (chiey portland) cement the binding agent in con-crete and most mortars is an important construction materialPortland cement is made primarily from nely ground clinkera manufactured intermediate product that is composed pre-dominantly of hydraulically active calcium silicate mineralsformed through high-temperature burning of limestone andother materials in a kiln This process typically requires ap-proximately 3 to 6 million Btu (32 to 63 GJ) of energy and17 tons of raw materials (chiey limestone) per ton (t) ofclinker produced and is accompanied by signicant emissionsof in particular carbon dioxide (CO2) but also nitrogen ox-ides sulfur oxides and particulates The overall level of CO2

output about 1 tonton clinker is almost equally contributedby the calcination of limestone and the combustion of fuelsand makes the cement industry one of the top two manufac-turing industry sources of this greenhouse gas The enormousdemand for cement and the large energy and raw materialrequirements of its manufacture allow the cement industry toconsume a wide variety of waste raw materials and fuels andprovide the industry with signicant opportunities to symbi-otically utilize large quantities of by-products of other indus-tries

This article the second in a two-par t series summarizessome of the environmental challenges and opportunities facingthe cement manufacturing industry In the companion articlethe chemistry technology raw materials and energy require-ments of cement manufacture were summarized Because ofthe size and scope of the US cement industry the article reliesprimarily on data and practices from the United States


94 Journal of Industrial Ecology

Introduction

Construction materials constitute some 70of the nonfuel materials ows in the UnitedStates (Wernick et al 1997) Concrete andmortars are critically important constructionmaterials concrete is used as a bulk buildingmaterial in its own right and mortars are usedto bind together bricks stone or other blocksin masonry-type construction Concretes andmost mortars rely on hydraulic cement bindersfor their strength and durability but despitethis the dry cement component in these ma-terials is rather small (eg about 10 to 12by volume of the concrete mix) Most of theconcrete which is essentially an articial con-glomerate is a mix of sand and gravel or otherne and coarse aggregates (65 to 80) water(about 14 to 21) and air (05 to 8) Thecombination of cement and water in the con-crete mix is called cement paste Composition-ally mortars differ from concrete chiey in thefact that they contain only ne aggregates andthe hydraulic cement contains plasticizingagents Typically 1 ton (t)1 of cement sufcesfor about 3 to 4 cubic meters (m3) of concreteweighing about 7 to 9 t Current world outputof hydraulic cement exceeds 16 gigatons (Gt)

This article is the second of a pair As notedin part I (van Oss and Padovani 2002) hydrau-lic cements are those that can set and hardenunderwater through the hydration of the com-ponent cement minerals By far the most com-mon hydraulic cements in use today are eitherportland cements or similar-use cements (calledldquoblendedrdquo or ldquocompositerdquo cements) that aremade of a portland cement base plus cementi-tious or pozzolanic additives blended cementsare commonly included within the portland ce-ment designation in the economic research andtechnical literature A pozzolan is a siliceousmaterial that develops hydraulic cementitiousproperties when interacted with free lime(CaO) and water

Straight portland cement is made by grindingtogether portland cement clinker (the interme-diate product of cement manufacture) with asmall amount typically 5 by weight of calciumsulfate usually in the form of the mineral gyp-sum Summarizing from part I the chemical

composition of a typical portland cement clinkeris almost entirely just four oxides calcium oxideor lime (CaO) about 65 silica (SiO2) about22 alumina (Al2O3) about 6 and iron oxide(Fe2O3) about 3 In cement industry short-hand these four oxides are written as C S Aand F respectively and most clinkers do notshow deviations in these oxide proportions ofmore than 2 to 4 percentage points The remain-ing 4 or so of the clinker composition is dividedamong oxides of magnesium potassium sodiumsulfur and others Clinker is primarily made upof four clinker minerals denoted in shorthand asC3S C2S C3A and C4AF The C3S and C2S arethe main contributors to the performance ofportland cement and together make up about70 to 80 of the weight of the clinker Duringtheir hydration C3S and C2S combine with wa-ter by similar reaction paths to form calcium sil-icate hydrate (its variable composition is denotedldquoC-S-Hrdquo) plus lime the C-S-H is a colloidal gelthat is the actual binding agent in the concreteThe bulk of the C3S hydrates rapidly (hours todays) and provides most of the early strength ofthe concrete whereas the C2S hydrates slowly(days to weeks) and is responsible for most of theconcretersquos long-term strength The lime by-product of hydration activates any pozzolans thatmay be present in the concrete mix

As was reviewed in part I the manufacture ofclinker involves the thermochemical processingof large quantities of limestone and other rawmaterials typically about 17 tt clinker and re-quires enormous kilns and related equipmentsustained very high kiln temperatures (the ma-terials reach temperatures of about 14508C in or-der to form the key C3S mineral) and the con-sumption of large amounts of energy (fuels andelectricity) total energy consumption is about 3to 6 million British thermal units (Btu)t clinker(1 million Btu 4 1055 GJ) Clinker manufac-ture results in signicant emissions particularlyof carbon dioxide (CO2) Apart from the tech-nological aspects of cement manufacture part Idiscussed the main environmental considera-tions of the mining of cement raw materials Theremaining environmental challenges and oppor-tunities relating to clinker and cement manufac-ture are the subject of this article


van Oss and Padovani Cement Manufacture and the Environment Part II 95

Environmental Considerations

Where the public is aware of the cement in-dustry at all it is usually in an environmentallynegative context (ie pollution) less wellknown are the environmentally benecial as-pects of the industry The main environmentalissues associated with cement manufacture arediscussed rst in terms of those that are prob-lems2 and second in terms of those that arebenets

Cement manufacture involves both miningand manufacturing steps Although covered inpart I a few summary remarks concerning themining of nonfuel raw materials are warrantedhere About 17 t of nonfuel raw materials areconsumed to make 1 t of cement the bulk (about85) of the raw materials is limestone or similarrocks to which is added clay or shale and othermaterials to achieve the correct chemical pro-portions These are for the most part geochem-ically benign materials and their mining gener-ally does not lead to signicant problems ofacidic or otherwise chemically contaminateddrainage Although individual quarries and min-ing rates for cement raw materials are not par-ticularly large relative to mines for many otherminerals the existence of thousands of cementplants worldwide ensures that their quarriesrsquo cu-mulative yearly output of cement raw materialsis huge Current world cement output requiresalmost 3 Gtyr of nonfuel raw materials associ-ated fuel consumption is roughly 200 milliontons (Mt) per year in straight mass terms (ienot on a common fuel basis) or about 015 to02 t fuelt clinker The concrete and mortars(about 13 to 14 Gtyr) incorporating this cementrequire a total of about 15 Gtyr of raw materialsmostly aggregates Reserves of cement (and con-crete) nonfuel raw materials are geologicallyabundant although they may be quite limited forindividual plants for a variety of reasons

Although not discussed in part I and gener-ally not stressed in discussions of raw materialsfor concrete and mortars the current annualworldwide consumption of raw materials forthese includes about 1 Gt wateryr for cementhydration Water is also required in some cementplants especially to form the raw materials slurryfeed for wet-process kilns (this water is not

needed with dry-kiln technology) Slurry wateramounts to about 30 to 35 of the weight ofthe slurry or roughly 08 t watert (wet process)clinker and is ultimately evaporated (and thuslost) in the kiln line Lacking comprehensive in-ternational data distinguishing wet- from dry-process clinker production the total amount ofwater consumed worldwide for wet-process slurryis not known but would amount to about 16 Mtfor the United States in 2000 (table 5 in part Itable 7 in van Oss 2002) The main issue con-cerning water for cement or concrete is not pol-lution of it by the cement or concrete industriesbut its adequate supply and quality (the broaderissue of sediment loading or other contaminationof water bodies as a result of general constructionindustry activity is neglected here) For variousreasons water for concrete manufacture shouldbe of essentially potable quality (Kosmatka andPanarese 1988)

As was argued in part I and notwithstandingthe signicant tonnages involved the effects ofmining of cement raw materials are consideredto be local in impact at least compared to someother mining sectors Far more important are theenvironmental issues relating to cement manu-facturing itself specically the manufacture ofthe clinker intermediate product and the re-mainder of the article focuses on these issuesClinker manufacture has signicant emissions ofparticulates and gases of which one in particular(CO2) has garnered international attention andis routinely singled out in national and interna-tional emissions data compendia Althoughquantitatively small relative to CO2 emissionsby individual plants of the other substances canbe of considerable local concern especially forolder plants in countries (or in past times) wherestrong emissions regulations arewere lackingAnd even in modern state-of-the-art facilitiesso-called minor emissions can be of public con-cern where the emitted substance has gained no-toriety from instances perhaps elsewhere of ma-jor releases or poor handling where thesubstance is classied as toxic or where it has analarming appearance (eg visible nonsteamemissions plumes) Further emissions levels thatare ldquominorrdquo on an individual plant basis canreach substantial cumulative totals whensummed for the world Except for CO2 emissions



(in the sense of escaping the plant) can be con-trolled or reduced in modern cement plants al-though not all modern plants are necessarilyequipped to control all emissions

Particulate Emissions from theManufacturing Process

Particulate emissions including dust of vari-ous types derive intermittently and diffusely fromquarrying activities and more or less continu-ously on a point-source basis from the commi-nution circuits (ie crushing and grinding of rawmaterials and clinker) from the pyroprocessingor kiln line and from landlled cement kiln dust(see below) In general fugitive emissions ofcoarse particulates (particularly of particle di-ameters 10 l m) if not controlled are consid-ered to be more of a local nuisance than a healthhazard Fine particulates (those lt 10 l m and es-pecially lt 25 l m diameter known in US reg-ulatory parlance as ldquoPM10rdquo and ldquoPM25rdquo respec-tively) in contrast are of greater concernbecause of their respirable nature and becauseboth for cement raw materials and manufacturedproducts they may contain potentially harmfulconcentrations of toxic metals and compoundsEven where emissions of ne particulates by ce-ment plants do not exceed statutory limits theycan augment already high ambient particulatelevels (from other sources) in the air The USEnvironmental Protection Agency (US EPA)provides extensive summary tabulations most re-lated to plant process and control technologiesof emissions of particulates both in terms of totalmass and chemistry Most of the data in the tab-ulations are rated by the US EPA as havingbeen measured by techniques of low reliabilityand the agency cautions therefore that the dataare order-of-magnitude indicators only (USEPA 1994 1995)

The amount of dust from comminution ishighly variable from plant to plant and is depen-dent on the type and character (eg hard softwet dry) of the materials involved and on thedesign condition and operational practices ofequipment at individual plants With even ru-dimentary dust-control procedures generallysuch dust especially the PM10 fraction is notconsidered a problem or its effects do not extend

beyond or much beyond (a few hundred meters)the connes of the plant property Where cap-tured much of the comminution dust is suitablefor incorporation into the raw material feed (rawmix) for the kiln

Dust from the pyroprocessing line is looselycalled ldquocement kiln dustrdquo (CKD) and includesne particles of unburned and partially burnedraw materials clinker and material eroded fromthe refractory brick lining of the kilns As usedin this article CKD includes both the main stackparticulate emissions and emissions from the al-kali bypass system (see below) as well as emis-sions from the clinker cooler

Very few public data are available on na-tional or even plant-specic total generation ofCKD This is basically because there has beenlittle economic or regulatory incentive to collectsuch data in the past and in any case CKD gen-eration is not easily measured At many plantsas much CKD as possible is directly routed backwith return air to the kiln (effectively joining theraw mix stream) and the dust content of thisreturn ux would be very difcult to determineIn modern plants and most plants in countrieshaving particulate emissions restrictions plantsroute exhaust through electrostatic precipitators(ESPs) andor fabric ltration baghouses to re-move CKD The amount recovered this way isreadily measurable although where done tendsto be on an episodic basis (eg when the ltra-tion bags are purged or cleaned) Recovery byESP andor baghouses is generally quite efcient(commonly 99 or better with modern equip-ment based on measured emissions) (Duda1985) Modern scrubber systems are capable ofmeeting current US particulate emission stan-dards for kilns of 015 kgt (or 0015) of dry rawkiln feed (US EPA 1999a) which is roughlyequivalent to 0009 on a clinker weight basisemissions from clinker coolers are limited to 005kgt clinker Return of CKD to the kiln eithervia direct rerouting or after capture by ESPs orbaghouses makes sense chemically and eco-nomically because the CKD typically has a majoroxide composition very close to that of the rawmix feed or the clinker and such a return of CKDthus saves on raw materials and energy

Because of the difculty of completely mea-suring the material the relatively few data on



CKD output or production commonly are limitedto (1) that material rst captured by the ESP andor baghouse (2) perhaps only that fraction ofcaptured CKD that is returned to the kiln or(3) perhaps just that portion sent to landlls Inother words most CKD production data shouldat least be suspected of underrepresenting the truetotal or gross CKD generation Despite the scar-city of data it is generally agreed that the amountof CKD generation is highly variable amongplants and over time at individual plants Basedon limited informal data and conversations withvarious US plant personnel an estimate ofCKD generation as about 15 to 20 (byweight) of the clinker output is useful as a rstapproximation which has implications for rig-orous calculations of CO2 emissions as discussedbelow and in Appendix A

A 15 to 20 CKD to clinker ratio impliesa signicant disposal problem if only in terms ofquantity for plants that do not recycle the CKDto the kiln or that cannot nd outside customersfor it given the fact that most plant clinker ca-pacities fall in the range of 02 to 20 Mtyr Theinformal data from and conversations with pro-ducers noted above revealed that in the UnitedStates typically about two-thirds of the gener-ated CKD is returned to the kiln leaving one-third for landll disposal (the majority) or saleLandll disposal is becoming increasingly unsat-isfactory for environmental and cost reasons (ielandll space is increasingly at a premium and isunsightly some countries now require that newCKD pits be lined to prevent escape of leachate)Landll disposal also represents a loss of potentialrevenue from material that not only has beenmined and at least partially processed but is closeto the nished saleable product (ie cement) incomposition In this respect CKD waste differsfrom wastes of some other industries where thewastes are dissimilar to the saleable product

Some contaminants (trace elements or com-pounds) from the raw materials and fuels tend toconcentrate in the CKD and these contami-nants may constrain the degree to which a ce-ment plant can recycle the dust to the kiln if theclinker quality thus becomes compromised Thisis a particular problem with alkalis (eg sodiumand potassium) which can cause adverse effects(volume expansion and bond-weakening alkali-

silica reactions) between the cement paste andcertain amorphous silica-rich rock types used asaggregates for concrete in some areas (Kosmatkaand Panarese 1988 Lea 1970) Preheater andpreheater-precalciner dry plants having raw ma-terials with high alkali contents commonly in-corporate an alkali bypass system ahead of thekiln or precalciner to reduce condensation of al-kalis (coatings) in the kiln line and the alkalicontent of the clinker andor CKD

The presence of contaminants other than al-kalis may limit the ability of CKD to be used forother purposes notably the traditional use as aliming agent for soils (Palmer 1999) althoughthey would be less likely to affect the suitabilityof CKD for other common uses such as the sta-bilization of sludges wastes and soils as road llor as a cementitious additive in blended and ma-sonry cements (as yet a minor use) Further in-formation on alternative uses of CKD can be ob-tained from Bhatty (1995)

Health concerns regarding CKD relate to itsdispersal through the air (dust from the kiln linematerial disturbed during transportation or windaction on existing CKD piles) and to leachatefrom CKD piles and generally have to do withthe concentrations of heavy metals in the CKDitself or in leachate from CKD piles As notedearlier the US EPA (1994 1995) summarizeda number of studies into the mass and chemistryof particulate particularly airborne emissions Incement plants lacking dust controls particle sizeanalysis of emissions of particulates from wet-process kiln lines showed that 24 of the partic-ulates were of diameters of less than 10 l m and7 were smaller than 25 l m dry-process linesshowed 42 of emissions having particle diam-eters of less than 10 l m and 18 less than 25l m (US EPA 1994 table 116-5) For plantshaving dust-control technology very little coarsedust was escaping both wet and dry lines showedthat about 85 of the remaining escaping par-ticles were of diameters of less than 10 l m Wet-process plants using ESP scrubbers showed an av-erage of 64 of the particles at less than 25 l mdiameter and dry plants equipped with bagh-ouses showed 45 of escaping particles in theless than 25 l m size fraction

A summary of US EPA studies into healthand related environmental issues concerning



CKD (particularly that in landlls) as well asproposed CKD landll disposal and managementpractices is found in US EPArsquos proposed stan-dards for CKD (US EPA 1999b) The US EPAreport noted that whereas most metal concen-trations in CKD were at safe levels for use ofCKD as a soil liming agent this was equivocalfor cadmium (Cd) lead (Pb) and thallium (Tl)Accordingly maximum concentrations were setfor CKD for soil liming use at 22 ppm for Cd1500 ppm for Pb and 15 ppm for Tl Limits werealso placed on the concentration of dioxins andfurans (see below) Although no limits were pro-posed for hexavalent chromium in the US EPAreport general concerns about Cr` 6 toxicity andthe fact that it can be a component of CKD havecontributed to a decline in the use of ldquochromerdquo(magnesia chromite) refractory bricks in the kilnlines (Nievoll 1997) An overview of the chem-istry and utilization of CKD was given by Mc-Caffrey (1994) Apart from the studies cited bythe US EPA in various reports (US EPA 19941995 1999b) compendia of heavy metal andother trace elements and compounds in CKDcan be found in publications of Haynes and Kra-mer (1982) Delles and colleagues (1992) andPCA (1992) Gossman (1993) provides data oncertain toxic elements from particulate emissionsfor about 30 US cement plants all of whichburned hazardous waste fuels

Gaseous Emissions from the ClinkerManufacturing Process

Gaseous emissions from cement plants in-clude large quantities of CO2 (a major focus ofthis article) smaller amounts of carbon monox-ide (which is considered to ultimately oxidize toCO2 and is discussed along with CO2) sulfur andnitrogen oxides and trace amounts of dioxinsand furans These are discussed below In addi-tion cement plants can emit variable but gen-erally much smaller quantities of a variety ofother pollutants (eg volatile organic com-pounds other than dioxins and furans) but it isbeyond the scope of this review to cover theserelatively minor emissions publications by theUS EPA (1994 1995) provide some emissionsdata on these compounds All the pollutantsmentioned are all at least potentially subject to

emissions regulations and increasingly plantsare being designed or retrotted with variousmonitoring devices for these compounds Like-wise the operational practices of some plants arebeing modied to reduce some of these emissionsEmissions standards and testing procedures varyamong countries however it is beyond the scopeof this review to provide a comparison of thesedifferent standards and procedures

Sulfur Oxide Emissions fromClinker Manufacturing

Anthropogenic sulfur oxides (SOx) emissionsare of general interest primarily for their role inthe generation of acid rain and the bulk of theseemissions are generally attributed to fossil-fuel-red power plants and base-metals smelters Lo-cally (particularly in humid areas) major pointsources of SOx can generate acidic mists that canengender potential health concerns

In cement manufacturing SOx emissionsmainly derive from the combustion of sulfur-bearing compounds in the fuels (eg from pyrite[FeS2] in coal and various sulfur compounds inoil and petroleum coke) but can to a lesser ex-tent also come from pyrite sulfate minerals andkerogens in the nonfuel raw materials Fuel-derived SOx forms in the main burning zone ofthe kiln tube (gure 4 in part I) and in the in-dependently heated precalciner apparatus (if soequipped) whereas raw-material-derived SOxforms in the preheating apparatus or section ofthe kiln line Given the large quantities of coaland other sulfur-bearing fuels consumed in ce-ment manufacture (table 1) the cement industrywould be considered a fairly large SOx sourcewere it not for the signicant self-scrubbing na-ture of the clinker manufacturing process in-deed the ability to handle high-sulfur fuels isconsidered to be an asset of the industry Theamount and location of SOx formation and emis-sions in clinker kiln lines can vary with the kiln-line technology (eg wet versus dry lines) Abrief summary is provided below but a more de-tailed review of these variables and of SOxabatement strategies was given by BCA (1997)and by Terry (2000)

Although the proportions are quite variablefrom plant to plant many of the SOx and



Tab

le1

Fuel

cons

umpt

ion

and

carb

ondi

oxid

eem

issio

nsfo

rth

eU

Sc

emen

tin

dust

ry

Year

Fuel

cons

umpt

ion

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

1998

1999

2000

Coa

l(kM

t)7

206

791

87

591

828

87

227

650

910

601

100

879

098

824

19

066

920

610

095

Cok

e(f

rom

coal

)1(k

Mt)

ndnd

ndnd

ndnd

ndnd

nd45

543

234

344

2Pe

trol

eum

coke

1(k

Mt)

ndnd

ndnd

nd35

748

844

237

91

475

119

71

622

135

1Fu

eloi

l(M

L)83

61

352

641

710

159

41

166

653

120

299

4273

134

124

Nat

ural

gas

(Mm

3 )2

751

372

14

859

562

15

998

451

81

718

301

294

106

972

065

333

8T

ires

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

158

269

685

374

Oth

erso

lidw

aste

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

6874

816

101

6Li

quid

was

te(M

L)nd

ndnd

ndnd

ndnd

nd2

nd2

885

126

890

592

9

Car

bon

diox

ide

emis

sion

s(t

tcl

inke

r)C

ase

Af

uel

only

30

690

630

560

540

530

520

550

474

041

40

460

460

490

48C

ase

At

otal

35

120

114

107

105

104

103

106

098

40

924

097

097

100

099

Cas

eB

fue

lonl

y60

630

570

520

500

480

480

490

434

037

40

410

410

440

43C

ase

Bt

otal

56

114

108

103

101

099

099

100

094

40

884

092

092

095

094

Clin

ker

outp

ut(k

Mt)

386

7951

093

553

4963

991

671

2358

549

633

4160

941

643

5671

257

758

4277

337

796

56

Sour

ce

Tab

le4

ofva

nO

ssan

dPa

dova

ni(2

002)

Not

end

4no

data

lik

ely

smal

lor

nile

xcep

tfo

r19

90

1Fo

rye

ars

labe

led

ldquond

rdquoco

nsum

ptio

nif

any

may

bein

clud

edin

data

for

coal

2

Was

tefu

elda

taw

ere

not

colle

cted

unti

l199

3bu

tth

efu

els

wer

ebe

ing

cons

umed

begi

nnin

gin

the

mid

-190

8s

3C

alcu

late

dba

sed

onst

anda

rdgr

oss

(hig

h)he

atva

lues

for

fuel

sV

alue

sex

ceed

thos

eca

lcul

ated

usin

gne

t(l

ow)

heat

valu

esby

abou

t0

02un

it(1

950ndash

1975

)0

00to

001

unit

(198

0ndash19

90)

and

001

unit

(199

5ndash20

00)

4D

ata

are

unde

rval

ued

beca

use

ofth

ela

ckof

was

tefu

elda

tafo

r19

85an

d19

90s

eefo

otno

te2

5In

clud

esca

lcin

atio

nem

issi

ons

of0

51t

tcl

inke

r6

Cal

cula

ted

base

don

actu

alhe

atva

lues

(gro

sshe

atba

sis)

for

fuel

sre

port

edby

plan

tsto

the

USG

Sin

2000



volatile alkalis derived from the raw materialscombine within the preheating zone or apparatusin the kiln line to form stable alkali sulfates(eg Na2SO4) or calcium-alkali sulfates [egK2SO4(CaSO4)2] some of which wind up asbuildups or coatings in the cooler parts of thekiln line and some of which become incorpo-rated within the clinker andor the CKD Thekiln-line coatings help to protect the refractorybrick linings from damage but if allowed to buildup excessively can clog or otherwise impede themovement of material through the kiln

Some of the SOx formed during preheating isscrubbed by limestone or lime in the raw materialfeed and forms anhydrite (CaSO4) but althoughmuch of it can become part of the clinker at leastpart of the anhydrite tends to decompose andrerelease SOx as the feed enters the (much hot-ter) calcination zone or apparatus in the kilnline Anhydrite surviving in the clinker (pro-vided that the amount is neither too variable nortoo high) is generally viewed favorably as itspresence can reduce the need for gypsum addi-tion later in the nish mill Overall typicallymore than 70 of the original SOx winds upincorporated in one compound or another in thecoatings the clinker and the CKD The SOxfrom anhydrite decomposition in the calcinationzone and that derived from fuels in the sinteringzone of the kiln is carried back with the systemair into the preheating zone and can overwhelmthe lime and alkali scrubbing capacity of the rawmaterial feed Thus there can be a net evolutionof SOx in the exhaust gas in concentrations com-monly of 100 to 200 ppm but they are variableVery approximately 100 ppm SOx in the exhaustcorresponds to an emissions rate of about 05 kgSOxton clinker The US EPA noted typicalSOx emissions for wet and long dry kilns of 41to 49 kgt clinker whereas preheater andpreheater-precalciner kiln lines had much loweremissions of about 027 to 054 kgt (table 116-7 of US EPA 1994) US statutory emissionslimits are typically around 275 kg SOxt clinker(Schwab et al 1999) Where SOx emissions rou-tinely exceed local regulatory limits or wherethey frequently appear as visible detachedplumes cement plants can install scrubbers onthe exhaust gases (Olsen et al 1998) Similarbut of smaller scale to those for thermal power

plants these scrubbers react the SOx with lime-stone or lime to make gypsum such as by the netreactions (shown for SO3)

limestone scrubberCaCO3 ` SO3 ` 2H2O UCaSO4 ` 2H2O ` CO2(

lime scrubberCa(OH)2 ` SO3 ` H2O UCaSO4 ` 2H2O

Likewise this type of SOx scrubbing can occurif hot exhaust gases are used as a heat source fordrying the (calcareous) raw materials in the rawmilling circuit A cement plant can further re-duce SOx emissions by selecting low-sulfur rawmaterials and fuels but these may be of limitedavailability or high cost

Nitrogen Oxide Emissions fromClinker Manufacturing

High-temperature combustion of fuels in thekiln line releases nitrogen oxides (NOx) withthe nitrogen being mainly derived from the at-mosphere but also to some degree from the fuelsthemselves a minor contribution also comesfrom some types of raw materials The formationof NOx in cement kilns is complex and as yetincompletely understood useful reviews of thesubject are found in publications by Haspel(2002) Lanier and Hanson (2000) Smart andcolleagues (1998) Terry (2000) and Young andvon Seebach (1998) As noted in these studies90 or more of NOx emissions are NO with therest NO2 the cement industry generates almostno nitrous oxide (N2O) a powerful greenhousegas (GHG) (US EPA 2002) Four mechanismsof NOx formation are recognized thermal fuelfeed and prompt

Thermal NOx makes up about 70 or moreof total NOx from clinker kilns and is formed bydirect oxidation of atmospheric nitrogen throughthe dissociation of O2 and N2

O ` N2 U NO ` N andO2 ` N U NO ` O

Thermal NOx begins to form at temperatures aslow as 12008C but rapid formation requiresabout 16008C which is well below the burner-ame (not material) temperatures in clinkerkilns Thermal NOx formation increases rapidly


van Oss and Padovan i Cement Manufacture and the Environment Part II 101

with even small temperature increases whenwithin the range of 13708 to 18708C the high-end temperature approximates that of the gastemperatures in the kilnrsquos sintering zone Giventhe high kiln temperatures even small shifts inthe amount of combustion oxygen can have apronounced effect on the amount of thermalNOx formed

Fuel NOx is formed from the burning of ni-trogen compounds in the fuels most fuels con-tain at least some nitrogen Of the major fuelscoal the most common fuel contains the mostnitrogen and natural gas the least (essentiallynil) Fuel NOx forms throughout the entire rangeof combustion temperatures but mainly when inexcess of 8008C and the mechanisms of forma-tion are complex In the fuel-rich (reducing)zone of kiln ames fuel NOx is reduced to N2which remains stable typically until the tem-perature reaches about 16008C when it reoxi-dizes to NOx Based on the higher nitrogen con-tent in the fuel one would expect coal-red kilnsto have higher total NOx emissions than naturalgas-red kilns but the opposite is true because ofthe dominance of thermal NOx formation in thesintering zone and the fact that natural gas gen-erally generates higher ame temperatures thancoal As noted in part I precalciners have theirown burners and operate at lower temperaturesthan those in the sintering zone of the kilnsthemselves accordingly NOx formation in pre-calciners (alone) is dominated by fuel NOx

Feed NOx is derived from nitrogen com-pounds in the raw mix or feed to the kiln and isformed slowly during the preheating (3508 to7508C) phase of pyroprocessing Feed NOx pro-duction tends to be greater in wet and long drykilns because of the relatively slow rates of pre-heating with these older technologies

Prompt NOx refers to NO formed in the re-ducing (ie fuel-rich) ame in excess of thatwhich would be expected from thermal NOx-forming reactions Prompt NOx appears to beformed by the reaction of CH2 and similar fuel-derived radicals with atmospheric nitrogen toform cyanide (CN) radicals and N both of whichsubsequently oxidize to NO

As noted by Young and von Seebach (1998)overall output rates of NOx from individualplants are highly variable even over short to me-dium periods (minutes to days) their study de-

tailed the example of one long dry kiln that hadabsolute NOx output rates varying betweenabout 1 and 65 kg NOxt clinker (convertedfrom reported English units) or about 01 to07 of the weight of the clinker with most val-ues in the range of 015 to 045 and whatlooks like a 1 standard deviation range encom-passing NOx emissions of about 02 to 04 ofthe weight of the clinker These values illustratethe typical variability of NOx measurements tobe expected for kilns but absolute NOx emis-sions would likely show a somewhat larger rangefor a large population of plants or kiln technol-ogies The lower end of the range noted wouldbe fairly typical of precalciner-equipped kilns be-cause of the reduced amount of very high tem-perature fuel combustion in the kiln comparedwith that burned at lower temperatures in theprecalciner likewise the more modern kilns haveshorter material residence times (and hencelower unit emissions) An alternative generalmetric is that kilns produce about 05 to 2 kgNOx per million Btu (or per gigajoule [GJ])

A 02 to 04 (of the weight of the clinker)NOx emissions range would imply NOx emis-sions by the US cement industry within therange of 016 to 032 Mt in 2000 based on aclinker output of about 7966 Mt in that year(table 1) This may be compared with total non-agricultural US NOx emissions of about 22 Mtyr of which about 19 are so-called industrialand commercial sources (US EPA 1997 1998)Although an output of about 1 of total USnonagricultural NOx emissions is modest com-pared to that of motor vehicles and electricalutilities cement plants are nonetheless signi-cant point-source NOx contributors and are in-creasingly being required to install NOx-monitoring equipment and reduce emissionsThis is particularly true in regions that sufferfrom high levels of ambient ozone the mostwidespread urban air pollutant in the UnitedStates which is largely a secondary air pollutantresulting from the precursors NOx and hydro-carbons

Approaches to reducing NOx emissions in-clude technological upgrades to reduce fuel con-sumption and material residence times in thekilns installation of low NOx burners recyclingof CKD adoption of staged combustion to re-duce thermal NOx in precalciners midkiln ring



of fuels reduction of excess air (oxygen) switch-ing among major fuels (ie burning more coal)burning of waste fuels to induce reducing con-ditions and for precalciner kilns introductionof water injection to reduce ame temperaturesin the sintering zone (Haspel 2002) All reduc-tion strategies benet from improved kiln processcontrols (Lanier and Hanson 2000)

Dioxin Emissions fromClinker Manufacturing

Cement manufacturing releases small butvariable amounts of a variety of volatile organiccompounds the US EPA (1995) listed some ofthese and showed a general emission of thesecompounds in total in the range of only 0014to 0090 kgt clinker At their low individualemissions levels most of these compounds do notraise health concerns One class of these com-pounds dioxins and furans has attracted signi-cant scrutiny however

Dioxins and furans are general names appliedto a large complex group of polychlorinated or-ganic compounds many of which are highlytoxic even in trace amounts For simplicity thequantity and toxicity of individual dioxins andfurans as well as those of the similar polychlo-rinated biphenyls (PCBs) are commonly ex-pressed relative to that of the compound 2378-tetrachloro-dibenzo-p-dioxin (TCDD) the mosttoxic and well-studied member of the group(US EPA 2000) The toxic equivalency factor(TEF) of TCDD is assigned a value of 10 andmost of the other compounds have TEFs of nomore than 01 many are 2 to 4 orders of mag-nitude lower

Trace amounts of dioxins and similar com-pounds (hereafter collectively labeled ldquodioxinsrdquo)can be formed from the combustion of organiccompounds in fuels and raw materials in cementmanufacture especially as a result of the com-bustion of certain waste fuels The potential toincrease emissions of dioxins may inhibit aplantrsquos use of the offending fuel where emissionscannot be controlled by varying the combustionconditions in the kiln where this control pre-cludes efcient kiln operations or where obtain-ing permits to burn the fuel would be too timeconsuming or costly Dioxin emissions likely

would not be the sole criterion in a plantrsquos de-cision or ability to burn waste fuels however

Dioxin emissions by cement plants are intrace amounts only but there is not an abun-dance of plant-specic data available on the ac-tual outputs Emissions for a limited number(about 30) of US kilns were measured in 1995by the US EPA (2000) about half of the facili-ties burned a portion of hazardous waste fuelsBased on TEFs developed by the World HealthOrganization in 1998 the US EPA found thatkilns that did not burn hazardous wastes haddioxin emissions in toxicity mass equivalents(TEQ) relative to TCDD averaging 029 ngTEQkg clinker (1 ngkg 4 0001 ppb) Kilnsburning hazardous wastes (types unspecied)emitted an average of 2248 ng TEQkg clinker(with a range of 111 to 3070 ng TEQkgclinker) that is emissions from kilns burninghazardous waste were about 100 times higherthan those from kilns burning regular fuels(coal)

The US EPA also found that for kilns burn-ing hazardous wastes emissions differed signi-cantly between kilns having ldquohotrdquo exhaust gases(as measured at the CKD scrubber) 4508F(2328C) and those having ldquocoolrdquo exhaustslt 4508F The hot exhaust emissions averaged3069 ng TEQkg clinker whereas the cool emis-sions were just 111 ng TEQkg clinker Furtherpost-1995 measurements by the US EPAshowed that for hot exhaust systems scrubberoutlet emissions of dioxins could be signicantlyhigher than those at the scrubber inlet Evi-dently dioxins were being formed within the hotscrubber and this discovery has led since 1995to a number of plants installing water spray cool-ing to the exhaust gases ahead of the scrubbersto reduce scrubber emissions

Overall for 1995 the US EPA (2000) pro-jected total national emissions from US kilnsburning hazardous wastes of 1561 g TEQ (ofwhich 1547 g TEQ was from hot exhaust kilns)and just 178 g TEQ from kilns not burning haz-ardous wastes for a grand total of 1739 g TEQBy comparison total US airborne dioxin emis-sions in 1995 from all anthropogenic sourceswere estimated at 3125 g TEQ Importantly theUS EPA noted that because of the installationof exhaust cooling noted above the total US



cement industry emissions are now (post-1995)much lower closer to about 14 g TEQ annuallyAlthough considered representative and ade-quate for rst-order approximations the absolutelevels of these dioxin emission levels by cementplants must be considered to be of low statisticalcondence (US EPA 2002) Apart from coolingthe exhaust gases to prevent formation in thescrubbers overall dioxin generation from com-bustion is kept very low if fuel materials in thekiln are kept in excess of 12008C for several sec-onds or more limits that are in line with normalkiln operating conditions (Krogbeumker 1994)Dioxin limits in recovered CKD that is intendedto be sold as a soil liming agent in the UnitedStates have been set by the US EPA at 40 ngTEQkg clinker (reported as 004 ppb TEQ USEPA 1999b)

Carbon Dioxide Emissions fromClinker Manufacturing

In recent years there has been increased in-ternational concern about the long-term effectsof anthropogenic emissions of GHGs on globalclimate The most important of these gases is car-bon dioxide (CO2) not because it has the high-est unit heat retention of the GHGs but becausethe quantity of emissions is so large that its ef-fects overall are dominant For the UnitedStates the US EPA currently produces annualUS inventories of emissions data for GHGsother than water vapor In terms of warming po-tential CO2 accounted for about 835 of thetotal US GHG emissions in 2000 (US EPA2002) Unlike the cement industryrsquos emissions ofSOx and NOx (considered to be relatively mod-est) the emissions of CO2 by the cement industryare enormous and have led to the industry beingone of a very few singled out in the calculationof international GHG emissions levels Promi-nent in this attention to the cement industryis the Intergovernmental Panel on ClimateChange (IPCC) which has the responsibility ofhelping to promulgate the Kyoto Protocol and toderive methodologies for establishing nationalGHG emissions inventories (see below)

Industrial emission of CO2 arises universallyfrom the burning of fossil fuels but it is relativelyuncommon from other industrial pathways Both

the IPCC and the US EPA segregate emissionsresulting from fuel combustion (dominated bythose of power plants and motor vehicles) fromother pathway sources The logic behind thiscombustion segregation is that on a national ba-sis it is easier to determine the total quantity offuels burned (based on apparent consumptioncalculated from national data on fuel productionstockpile sales and trade) than to survey themyriad individual consumers of fuels This com-bustion segregation does not clearly demonstratethe full CO2 impact of the cement industry how-ever nor that of the two other industrial sources(the ironsteel and lime industries) identied ashaving major noncombustion CO2 emissions

As shown in gure 1 combustion of fuels ac-counted for about 97 of total US anthropo-genic CO2 emissions (about 58 Gt) in 2000 andalmost two-thirds of the total combustion emis-sions were from power plants and motor vehiclesAll of the remaining individual combustionsources were small by comparison but of thesesources the ironsteel and cement industrieswere the largest

Overall the US cement industry accountedfor only 06 of the countryrsquos total CO2 emis-sions from combustion or 13 of total emis-sions from all sources The cement contributionwould be 34 of total emissions from all sourcesexcluding motor vehicles and power plants Be-cause most countries do not have as proportion-ally large a thermal power generation infrastruc-ture as the United States or a comparableintensity of motor vehicle use the relative ce-ment contribution to total CO2 emissions islower in the United States than in many othercountries having substantial cement industriesA number of studies (eg Hendriks and col-leagues 1998 Worrell and colleagues 2001) havesuggested that worldwide the cement industrycontributes about 5 of total anthropogenicCO2 emissions an estimate of 24 was given byMarland and colleagues (1989) but this does notaccount for the combustion emissions by the in-dustry and so would need to be approximatelydoubled for the full emissions picture

Given the magnitude of the cement industryrsquosCO2 output considerable interest has been ex-pressed in quantifying these emissions TheIPCC (1996 2000) developed a detailed meth-



Figure 1 The contribution of the cement industry to total US anthropogenic CO2 emissions in 2000 Thecement industry is a large source of CO2 but its total contribution is only a small fraction of that from thecombustion of fuels by power plants and motor vehicles For countries having economies that are lesspower plant and motor vehicle intensive the proportion of total emissions from the cement industry isgenerally higher Source Based on data from US EPA (2002)

odology to estimate (to within 5 to 10) thecalcination CO2 emissions from cement manu-facture in individual countries The IPCCmethod is based on national statistics (but ap-plicable also to data for individual plants) onclinker production rather than cement itself Aproblem with a clinker approach is that clinkerproduction data are currently lacking or notreadily available for most countries although thedata probably could be easily collected in the fu-ture by national statistical agencies National hy-draulic cement production data are far moreavailable but commonly are rounded (see egtable 23 of van Oss 2002) Also as is discussedbelow the accuracy of the calculation using ce-ment is highly dependent on the clinker fractionof the cement which can be quite variable Onecould also estimate emissions derived from thecement industriesrsquo combustion of fuels providedthat data exist for the types and quantities of fu-els consumed Lacking such data for most coun-tries one can still make crude approximations forthem based upon existing combustion emissionsdata for the US industry owing to similar ce-ment plant technologies in use worldwide thisapproach is less valid for China because of itspreponderance of vertical shaft kiln (VSK)plants

CO2 Emissions from CalcinationAs noted in part I the calcination of calcium

carbonate (from limestone the major raw ma-terial in cement) releases CO2 by the simple re-action

CaCO ` heat (to about 9508C)U CaO CO (3 2

In theory the amount of calcination CO2 re-leased could be calculated from the componentformula weight ratios for this equation based onthe amount of limestone and similar rocksburned in the kiln (table 3 of part I) For thisone needs data on both the tonnage and com-position of the raw materials and these data aregenerally lacking on a national basis The onlypractical general country-level approach is towork backward from clinker production fromwhich may be derived a calcination emissionsfactor of 051 t CO2t clinker assuming a 65CaO content of the clinker Appendix A pro-vides a detailed discussion of this calculation

The 051 t CO2t clinker emissions factor forcalcination which has been adopted as a defaultby the IPCC (2000) is very similar to those usedin some other studies (eg US EPA 2002 Van-derborght and Brodmann 2001 Worrel et al2001) some of the differences disappear withrounding (component data quality does not war-rant precision to more than two decimal placesand most nal results should be rounded further)The methodologies in some of these other studies(eg Vanderborght and Brodmann 2001) how-ever work forward from the raw materialsmdashtheprocedures are proposed for individual plant re-portingmdashbut the equations are the same andthus subject to essentially the same error rangesAs noted by the IPCC (2000) and discussed fur-ther below unless one knows the net clinker frac-tion (about 95 for a straight portland cementtypically 55 to 80 for blended or compositecements and 45 to 60 for masonry cements)of a countryrsquos specic cement output mix there



is no CO2 emissions factor for cement productionthat can be used without the potential for sig-nicant error (up to about 35)

An assumption apparently common to allclinker-based methodologies for calculating CO2

emissions is that one is dealing with portland ce-ment clinker or something very similar to it Thisassumption which basically refers to the amountand source of CaO in the clinker would not holdvery well for clinker for aluminous cement (madeby burning a mix of bauxite and limestone) Alu-minous cement however is manufactured in tiny(lt1) quantities and in just a few countriesrelative to portland and related cements Pro-duction data on aluminous cement are almostinvariably proprietary and hence unavailableWithin any realistic assessment of the accuracyof current international data on clinker and ce-ment compositions and production levels satis-factory rst-order national or regional estimatesof calcination CO2 emissions for the cement in-dustry can be made using the assumption of a65 CaO content in clinker and a calcinationemissions factor of 051 t CO2 t clinker

CO2 Emissions from CombustionThe determination of CO2 emissions from

fuel combustion is complex and imprecise com-pared to that from calcination because the com-bustion emissions are dependent on the typesand quantities of the actual fuels burned to-gether with their interrelated carbon and heatfactors (contents) Specic data for all of thesefor the cement industry are generally lacking es-pecially on a country basis Further the calcula-tion of CO2 from combustion invariably makesthe assumption that all of the carbon monoxide(CO) generated through combustion ultimatelyis converted to CO2 and may thus be treated asCO2 for GHG calculation purposes The gener-ation of CO and the methodology used in thisarticle to calculate combustion CO2 are discussedin Appendix B Unlike the case with calcinationemissions the net combustion CO2 factors de-rived from data for one countryrsquos cement industrymay be only approximately applicable to othercountries because the quantity (and to some de-gree the type and composition) of fuels burned

depends signicantly on the kiln technologies inuse fuel availability and to a lesser degree on thenonfuel raw materials consumed

Table 1 shows the basic fuel consumptionbreakdown for the US cement industry for 1950through 2000 as well as the derived unit emis-sions of CO2 Two sets of CO2 emissions data areshown both utilize the gross or high heat con-tents (values) of the fuels to calculate the fuelsrsquocarbon contents Gross heat is the basis of fuelenergy reporting by the US cement industrywhereas net or low heat values are used almosteverywhere else Gross heats are preferred forCO2 calculations as they are more completeFortuitously there is very little difference be-tween gross and net heats for coal coke andmost other solid fuels and as these are the dom-inant fuels for the cement industry the CO2

emissions series in table 1 would not change sig-nicantly if it were recalculated using net heats

As seen in table 1 combustion emissions in2000 for the US industry were 043 t CO2 tclinker according to actual heat data supplied bythe plants or 048 t CO2 t clinker using standardgross heat assignations for each fuel the equiv-alent output using standard net heats is 047 tCO2 t clinker All of the CO2 data series show ageneral decline since 1950 although the data for1985 and 1990 are slightly and signicantly toolow respectively because they do not includewaste fuels Burning of signicant amounts ofwaste fuels in US cement kilns began aroundthe mid-1980s but data on this activity were notcollected before 1993

As noted in part I and as with unit energyrequirements the general decline in unit CO2

emissions from 1950 to 2000 reects the mod-ernization of the US industry over this periodspecically the increasing use of dry (vs wet)process technology in clinker manufacture aswell as technological upgrades at many existingdry-process kilns Data for the earlier years intable 1 could be viewed as comparable to valuesthat would be expected in countries presentlyrunning mostly old kilns Countries having allvery modern dry preheater-precalciner kilnshowever could be expected to have somewhatlower combustion emissions than the lowest val-ues shown in table 1



Summarizing CO2 Emissions fromCalcination and CombustionTable 1 also shows total emissions of CO2

which are simply the combustion emissions plusthe calcination emissions of 051 t CO2 t clinkernoted above Total unit emissions by US ce-ment plants amounted to about 094 t CO2 tclinker in 2000 Total emissions for the majorityof countries would likely be within the range of09 to 12 t CO2 t clinker perhaps slightly higherfor countries such as China and India that op-erate a large number of VSKs As noted earlierthe IPCC (2000) calculation methodology usedfor calculating CO2 is good to within 5 to 10for individual country calculations and this orgreater uncertainty would apply to most compet-ing methodologies Given likely uncertainties ininternational clinker production and other dataand in the CO2 calculation methodologies asimple ratio of 1 t CO2 t clinker production is areasonable rst-order approximation by which tocompare total emissions among countries

For a straight portland cement (typically 95clinker) the corresponding total emissionswould be about 095 t CO2 t cement and indeedthe casual literature commonly quotes ldquocementrdquoemissions of 1 t CO2 t cement A 11 ratio forcement however will not accurately allocate na-tional CO2 except for countries that produceonly straight portland cement because of thevariable clinker fraction that can be accommo-dated within the term ldquohydraulic cementrdquo Evenwhere data are available for a countryrsquos produc-tion mix of straight and blended cements theactual clinker fractions of the blended cementsis often unknown or available only as estimatedaverages or as a range of compositions For ex-ample the recipe for one of the common blendedcements in the United States (type IS general-use portland blast-furnace slag cement) allows fora ground granulated blast-furnace slag (GGBFS)content of 25 to 70 Assuming a 5 gypsumcontent in the portland cement fraction a typeIS cement could thus have a clinker factor of285 to 713

Where clinker fraction data are lacking theIPCC (2000) recommends using a default clinkerfraction of 75 Humphreys and Mahasenan(2002) gave world regional clinker factors esti-

mated for the mid-1990s by the InternationalEnergy Agency ranging from 81 (western Eu-rope average) to 89 (Africa and the MiddleEast) but there likely is variation within the re-gions as well These authors estimated totalworld output of CO2 from the cement industryoutput for 2000 at 14 Gt Worrell and colleagues(2001) used average clinker factors of 84 forindustrialized countries and 87 for the rest ofthe world and showed a breakout for 1994 ofabout 30 countries that revealed clinker fractionsof 74 to 99 and a world average of 85 Aregional tabulation of approximate CO2 emis-sions potentials based on data for clinker capac-ities is given in table 2 CO2 emissions3 for theworldrsquos cement industry are on the order of 17Gt about one-half of this potential resides inplants in Asia Figure 2 is a map showingcountry-level CO2 emissions potentials

CO2 Emissions from Electricity UseTable 1 does not include the CO2 derived

from the generation of electricity because it islikely that in the analysis of national GHG emis-sions such emissions would be allocated to thecommercial power plants in the case of cementmanufacturers purchasing all or most of theirelectricity (eg the US industry) Electricalgeneration by a cement plant itself generallywould make use of waste heat from the kiln andnot involve further fuel consumption If data areavailable on electricity purchases approximateoverall electricity CO2 can be calculated by as-suming that the cement industry accesses an ldquoav-eragerdquo national power grid and so has indirectCO2 emissions from electricity proportional tocementrsquos share of a countryrsquos total power gener-ation and derived emissions By this means unitelectrical consumption by the US cement in-dustry (see discussion and table 5 in part I) of151 kWht portland cement (144 kWht totalcement)4 in 2000 would correspond to about007 to 008 t CO2 t cement that is total CO2

emissions including purchased electricity areabout 7 to 8 more than the unit CO2 emis-sions levels shown in table 1 totaled just for cal-cination and combustion

Table 3 is a summary of emissions and otherdata for cement manufacture



Table 2 Regional tabulation of hydraulic cement production and carbon dioxide emissions potential

Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger

United States 895 55 802Canada 126 08 144Mexico 317 19 433Central America and Caribbean 128 08 130South America 753 46 915Western Europe 1899 116 1910Eastern Europe 421 26 567Former Soviet Union 469 29 1053Middle East 1036 63 1257Africa 718 44 787Asia and Pacic 9636 588 9405

Total world 16398 1000 17403

Source USBM (US Bureau of Mines) and USGS minerals yearbooks (cement) Cembureau (1996) and othersources (clinker capacities)

Data are for 2000

dagger Data are based on clinker capacities variously for the years 1996ndash2000

Figure 2 World distribution of annual cement industry CO2 emissions potential based on country clinkercapacities in the mid-1990s By far the greatest emissions potential is in Asia Source Based on data inCembureau (1996)

Environmental Benets and the IndustrialEcology of Cement ManufactureAlthough generally not widely discussed

there are environmental benets to cementmanufacturing and the use of cement or con-crete One of the general benets claimed by thecement and concrete industries is that concreteis ldquobetterrdquo than competing construction materi-

als In terms of overall construction tonnageequivalence it is difcult to quantitatively com-pare emissions levels (or other environmental ef-fects) of cement with those of its potential sub-stitutes and most of these materials do notsubstitute for cement or concrete on a 11 massbasis dening the appropriate basis of compari-son (ie the functional unit) is a challenge fa-



Table 3 Salient data for cement and clinker manufacture

Approximate quantity per metric ton

Item Clinker Cement1 Units

Raw materialsNonfuel2 173 184 TonsFuel 02 NA Ton

Energy emissions 475 496 Million British thermal units7

Cement kiln dust 028 NA TonSOx 05ndash1 NA KilogramsNOx 1ndash4 NA Kilograms

Carbon dioxideFrom calcination 0519 049 TonFrom combustion 04310 041 Ton

Dioxins11 300ndash3000 NA Nanograms TEQ12

Note NA 4 not applicable1 Portland cement2 Excluding water (used for wet kiln slurry)3 About 15 t of this is limestone or equivalent4 Assumes 5 gypsum5 Gross heat basis average for US industry in 2000 Excludes electricity Most plants are in the range of 3 to 6million Btuton clinker6 Includes electricity7 1 million Btu 4 1055 GJ8 Total generation not fugitive emissions9 Assumes CaO in clinker of 65 and 100 derivation from carbonate10 Gross heat basis US average in 2000 Most plants are in the range of 04 to 06 tonton clinker11 Dioxins furans and similar chemicals12 Nanograms toxic mass equivalents relative to the dioxin TCDD

miliar to life-cycle assessment (LCA) practition-ers A few points in this regard relative to steeland wood are made here

Structural steel is a common competitor withconcrete for projects such as ofce buildings andbridges In terms of a comparison of GHG emis-sions Price and colleagues (1999) cited CO2

emissions for a few countriesrsquo steel industries inthe range of 12 to 36 t CO2 t steel overall buttheir data include steel derived from both oresand scrap making comparisons complicated5

Here it is not important to provide exact CO2

emissions or emissions range for iron or steel butto note that steel derived from iron ores via ablast furnace has unit CO2 emissions that arehigher than those for cement

Another environmental comparison is the re-spective energy requirements of manufactureFreuhan and colleagues (2000) provided a num-

ber of theoretical energy requirements for iron-and steelmaking for several selected conditionsand compared them to actual energy require-ments (which are higher) Using hot metal(crude iron) from a blast furnace primary steelmade in a basic oxygen steel furnace has actualenergy requirements of about 24 to 26 GJt steelof which 13 to 14 GJt is the energy required toproduce the hot metal charge (ie the blast-furnace component) Steel from an electric arcfurnace (EAF) requires 21 to 24 GJt Price andcolleagues (1999) provided an energy compari-son between cement and steel manufacture for anumber of countries They showed energy re-quirements for primary steel (made from ironproduced in a blast furnace) of about 23 to 41GJt steel whereas steel from scrap feed (EAF)only requires about 55 GJt the latter is com-parable to the energy requirements to make ce-



ment (see part I) Their energy requirements (40to 58 GJt) for cement are comparable to therange we showed for the US industry in part I

Glover and colleagues (2002) compared theembodied energy in concrete steel and woodused in house construction and showed that thehighest embodied energy is for steel and the low-est is for wood their study also showed the greatsensitivity of the results to the many variables inthe analysis Their energy values for the cementin concrete are at the high end of the range thatwe showed in part I

In addition to embodied energy an environ-mental comparison of cement (concrete) andwood (mainly for houses) is of interest even injust qualitative generalizations because the ce-ment industry is actively trying to capture marketshare in housing construction (walls and roongnot just the almost ubiquitous concrete founda-tions) Concrete buildings are more durable thanwooden structures in terms of re resistance andthe strength needed for multioor construction(but concretersquos advantages in terms of the seis-mic resistance of houses and similarly sized struc-tures are less clear and depend on many construc-tion variables) Concrete is ldquoabundantrdquo (13 to 15Gtyr worldwide manufacture or consumption)as are its component mineral resources (but theyare nonrenewable) Wood particularly softwoodis a renewable resource but its availability is con-strained by severe deforestation andor slowgrowth rates in some parts of the world and it isnaturally scarce in other areas Wood is consid-ered to be more or less CO2 neutral because itsatmospherically derived carbon is modern (rela-tive to fossil fuels) and if burned to completionreturns all the carbon to the atmosphere (asCO2) Borjesson and Gustavsson (2000) pre-sented a GHG comparison of a wood- versusconcrete-framed house in a materials and energyow study which additionally well illustrates thepotential complexity of a quantitative compari-son of dissimilar construction materials6

Alternative Fuels forCement Manufacture

Apart from possible advantages in using ce-ment (concrete) instead of competing construc-tion materials cement manufacture itself has

environmental benets Although the practicevaries among individual plants cement manu-facture can consume signicant quantities of in-dustrial by-products as fuel (table 1) and nonfuelraw materials (table 3 in part I) This consump-tion reects the combination of long residencetimes and high temperatures in clinker kilns thatensures the complete breakdown of the raw ma-terials into their component oxides and the re-combination of the oxides into the clinker min-erals The ability to consume large quantities ofwaste products is what makes cement manufac-ture attractive from an industrial ecology view-point

Cement kilns after appropriate modica-tions are particularly adept at burning alterna-tive or waste fuels which can be almost any-thing solid or uid that can combust7 includingused motor vehicle tires municipal biowastesand hazardous (but not radioactive) wastes De-gre (1998) and Jenkins and Mather (1997) pro-vide lengthy lists of these materials8 Waste fuelburning in a cement kiln has several advantagesfor the cement plant and for the environmentand society in general Consumption of waste fu-els reduces a plantrsquos consumption of conven-tional fossil fuels In reality because of impuritiesin waste fuels the very large dimensions of kilnsa choice of introduction points for the waste fuels(main burner midkiln upper end) and certaintechnical complications that waste fuels may im-pose on kiln operations exact heat equivalencein fuel replacement may not always be possibleNonetheless the use of waste fuels in cementkilns allows the displaced fossil fuels to be utilizedby other industries that lack fuel exibilityWaste fuels cost less than conventional fuels ona heat equivalence basis (Boarder 1997 Gilling1999) and generally are used for the sake of econ-omy in some countries especially the UnitedStates the cement plants are in fact paid to takethe wastes According to plant personnel insome cases such revenues can completely offsetthe remaining conventional fuel costs (ie zeronet fuel costs) or even exceed them and thusfunction as a source of net income for the plantLower fuel costs from waste fuels are of especialbenet to wet and long dry process plants asboth are older technologies having relativelyhigh unit energy requirements (see part I) The



large dimensions of these older technology kilnsand long material residence times associated withthem contribute to their ability to efcientlyburn waste materials

A clinker kiln converts a worthless waste ma-terial or one that could otherwise incur substan-tial disposal or storage costs (especially for haz-ardous wastes) through its use as a fuel into aproduct typically worth $50 to $100 or more pert as nished cement by comparison a commer-cial waste incinerator burns fossil fuels to disposeof the wastes and may produce a modest amountof electricity Because the waste fuel (heat) com-ponent of total fuels is generally limited to 30or less (some plants are permitted to burn ahigher component even where they do not ac-tually do so) cement kilns can make use of fuelswith lower unit heat values than could be tol-erated by most other industrial furnaces

Burning waste fuels can convey additional ad-vantages in terms of reduced emissions (Mishu-lovich 2003) For example some hydrocarbon-containing waste fuels have a higher hydrogen tocarbon ratio than conventional fuels and thusyield less CO2 when burned Incorporating fuelsthat induce reducing conditions can reduce NOxemissions from kilns as noted earlier At leastone plant in the United States burns a propor-tion of semidry municipal sewage (ldquobiosolidsrdquo)solely for NOx control the material is energyneutral (Mayeld and Biggs 1997 Kahn 1998)Some forms of biomass fuel such as sawdust maygenerate CO2 that might be subtractable fromldquoregularrdquo process CO2 for a plantrsquos environmentalaccounting purposes where biomass is consideredto be CO2 neutral

In terms of community relations burning ofwaste fuels assuming that it is fully permittedsaves the community alternative disposal or stor-age costs and the cement plant could be consid-ered a safer destination for the wastes than cer-tain disposal alternatives A community mightonly see this as an advantage if the waste fuelsare generated locally however In cases wherekilns have been adapted to take whole (as op-posed to shredded) tires the plant may allow thegeneral public to individually deliver their worn-out tires directly to the facility a practice thatmay be more convenient and cheaper for thepublic than disposal at the local landll and for

the landll operator reduces the size and hencere risk of large tire accumulations

For a cement company a decision to burnwaste fuels involves many factors A potentiallylengthy and expensive environmental permittingprocess may be required which may involve pub-lic input The economic analysis of the decisionconsiders the suitability of the existing kiln tech-nology at the plant or which could be installed(eg modication to an existing kiln or con-struction of a new kiln) the straight cost savingsor other benets and the cost of complying withenvironmental restrictions (if any) on the fuel(such as storage and handling costs emissionsmonitoring costs penalties for emissions viola-tions and whether burning of waste fuels wouldforce emissions monitoring and limits for plantshitherto exempt from such requirements)

Fuel-handling considerations include physi-cally moving the material about any costs to pre-pare the fuels (crushing screening blendingquality control etc) safety issues (ammabilitytoxicity) and possibly aesthetic (odor) consid-erations A key consideration given the fact thatthe plant would prefer a consistent waste fuelmix is whether there is a sufcient and reliablesupply of the waste(s) available within a reason-able transportation (cost) distance Thus formany wastes (especially tires) a plant needs tobe fairly close to a large metropolitan area andfor most nonhazardous wastes the appropriatesource industry or other accumulation site shouldbe reasonably close to the cement plant Hazard-ous wastes generally involve high handling stor-age and transportation costs these costs can berelatively insensitive to distance

A related consideration is the cost and logis-tics of collecting the waste fuels even from aconvenient distance Commonly either the pro-ducers deliver to the cement plants or privatecontractors are engaged for this purpose Espe-cially with liquid and liquid hazardous wastesthere may exist (or the cement company mayhelp form) a company specically charged withcollecting the materials blending them to an ac-ceptable degree of consistent quality storing theindividual materials and blended fuel mix anddelivering it to the cement plant as requiredSuch fuel companies offer operational advan-tages to the cement plant are a practical way for



the cement plant to tap into an otherwise incon-venient diversity of fuel sources and may reducethe regulatory burdens of using the fuels on thecement company itself Likewise such a fuelcompany might provide an entire diverse indus-trial complex with a convenient means of dis-posing of part or all of its wastes (Vigon 2002)

A disadvantage to certain waste fuels is theirpotential to induce or increase kiln emissions(Gossman 1993) Certain hazardous wastes forexample may increase the chances of a plantemitting higher amounts of dioxins and furans(US EPA 2000) as noted earlier Highly het-erogeneous fuels may require extensive prepara-tion prior to burning to screen out or sufcientlydilute components such as toxic metals (egnickel chromium) and to blend to ensure con-sistent heat contents and ease of burning

Local or national building codes and projectspecications commonly cite cement types orperformance specications that indirectly maynot allow for burning waste fuels or may limitthe type(s) burned because of the potential forchanging the trace-element composition of thecement (or concrete) and potentially its perfor-mance For example a project to build aconcrete-lined municipal water reservoir couldindirectly discourage the use of waste fuels in ce-ment manufacture (for this project) that wouldincrease the cementrsquos content of potentiallyleachable heavy metals (see the particulates dis-cussion above) Wastes that might increase a ce-mentrsquos alkali content could preclude that ce-mentrsquos use for projects in areas where alkali-silicareactions might become a problem

Transportation storage and handling andburning of in particular hazardous wastes mayprove uneconomical or bothersome for the ce-ment plant and could invite lawsuits and gener-ally deteriorating public relations (it is unlikelythat a cement company can count on the auto-matic public approval of proposed waste fuelburning) Regarding the latter companies seek-ing to avoid public opposition to proposed wastefuel burning often provide local communitieswith factual data relating to the project and so-licit public commentary well in advance of theproposed startup date and commonly in ad-vance of the ofcial permitting process

Alternative Raw Materials

Clinker manufacture can make use of a widevariety of raw materials most of which are se-lected on a supplementary basis to make up forchemical deciencies of the primary limestonefeed As noted in part I these supplementary ma-terials may for example supply nearly all of thealumina or silica or they may be added as ldquosweet-enersrdquo to boost one oxide or another Examplesof sweeteners include high-purity limestone orcalcite itself to boost the lime content silicasand silica fume or diatomite to boost silica alu-mina or aluminum dross to supply alumina ormillscale for iron Apart from the oxide contentof a proposed alternative material plant opera-tors consider the materialrsquos thermochemical ac-cessibility (Mishulovich 2003) For example sil-ica sand is a commonly used supplementary silicasource yet it is hard to grind (thus increasingelectricity consumption) and the componentquartz (SiO2) requires high temperatures and along exposure in the kiln to activate the silicaIf available a more easily grindable andor morereactive silica source might be preferable ex-amples include diatomite ferrous slag or a ma-terial containing amorphous silica such as yash

Although many of the supplementary mate-rials are mined products any number of othermaterials including wastes are potentially suit-able especially if they are of low cost Some ma-terials contribute both oxides and energy for ex-ample deinking sludge from recycling andshredder nes from paper plants Some of thesematerials offer process advantages for examplecertain aluminum smelter by-products (pot lin-ers catalysts) not only contribute alumina butalso sufcient uorine or calcium uoride to actas a ux (Mishulovich 2003) Fly ash and bottomash from coal-red power plants as well as fer-rous slags are consumed in large quantities (seebelow and table 3 in part I) as supplementarysilica alumina and lime sources for clinkerNoncarbonate lime sources are of particular in-terest in an environmental context because theyreduce the calcination CO2 component of theprocess this is discussed in more detail later Thecriteria for selecting waste materials for the kilninclude appropriate chemistry (composition and



reactivity) resulting cement quality materialavailability material costs (base transportationstorage handling and preparation) regulatorycompliance and general environmental impacts(similar to those for waste fuels as discussedabove) and public and government acceptance

As discussed in part I pozzolans and similarcementitious extender materials may be added toproduce blended cements and masonry cementsWaste products used as extenders include sometypes of y ash GGBFS burned rice husk ashCKD burned clays (ldquometakaolinrdquo) and silicafume In regions or countries where cement spec-ications or building codes allow the practice(without renaming the product) extenders maybe incorporated as a minor bulking agent forportland cements a common example is the in-corporation of 1 to 3 GGBFS introduced (inunground form) in the nish mill as a grindingaid

Cementitious extenders allow the productionof a nished cement with lower clinker contentand hence proportionally lower unit equivalentenergy and raw materials inputs and emissionsoutputs Although blended cements can havelower early strength development their nalstrength (measured at 28 days) is generally com-parable to straight portland cements Blended ce-ments generally exhibit a reduced heat of hydra-tion which is usually but not always anadvantage Other advantages to using extendersinclude improved owability of concrete duringplacement and the fact that the hardened con-crete generally has lower porosity reduced limereactivity (the pozzolans consume the lime re-leased through cement hydration) and increasedresistance to other forms of chemical attack no-tably by sulfate-rich groundwater A large liter-ature on cement extenders exists for examplethe collected papers in Frohnsdorff (1986) andMalhotra (1989) Detwiler (1996) provided ashort overview

Cementitious extenders are generally addedon about a 11 clinker (or portland cement) sub-stitution basis These materials may or may notreduce the cost of making the nished cementdepending mainly on the procurement cost ofthe extender and whether and to what extent thematerial needs to be ground Granulated blast-furnace slag for example may be added in

ground form (GGBFS) to cement or concrete orin unground form (as a grinding aid) in the nishmill Because it is less hydraulically reactiveGGBFS needs to be ground considerably nerthan portland cement to make a satisfactory ce-ment extender Thus if a plant decided to grinda substantial quantity of slag it likely would in-cur higher unit grinding costs and electricityconsumption and the mill apparatus assigned tothe slag would have a lower output capacity thanif it had been kept on clinker-grinding dutySome cement plants operating large grinding fa-cilities for GGBFS produce a surplus and sellmuch or all of this material directly to the con-crete industry Fly ash generally does not requiregrinding and silica fume particles are alreadyvery much ner than cement

Industrial Symbiosis Involving Cement

In an industrial ecology context cementplants are of interest not only where they simplyconsume waste materials of other industries buteven more so to the degree that a cement plantmay be directly tied to a plant of a different in-dustry or vice versa (Nemerow 1995) Cementplant materials acquisition interactions can takeseveral forms The cement plant could simplyconsume wastes of other plants (as discussedabove) with every possibility of changing wastesuppliers or types as conditions warrant A vari-ation on this theme is where a parent companyseeks to link subsidiary cement plants with sub-sidiary plants manufacturing other products

Of greater industrial ecology interest is wherea cement plant is constructed to take advantageof the availability of wastes or by-products thelocation might be entirely or partially governedby the waste source depending on what thewaste contributes and the economics of its trans-portation An existing or proposed cement plantmight be modied to take advantage of wasteavailability Similarly a waste-generating indus-trial facility could be sited to take advantage ofproximity to a cement plant or a factory mightalter the composition or physical form of itswaste stream to make it attractive for use in ce-ment manufacture Both the industrial plant andthe cement plant might enter into partnership toform a waste-handling company More than one



facility can be involved in these relationshipsand some might be nonindustrial operations (afarm for example) In any industrial relationshipsound economic reasons may dictate subsequentswitches to other raw materials and sources ce-ment plants are remarkably exible in what theycan consume and would be well poised to takeadvantage of new industries in the market areaor to survive the cutoff or shortages of materialsbeing used

A comprehensive review of the industrialecology of cement was given by Vigon (2002)but a few examples here would sufce to illustratesome of the linkages that make cement an inter-esting industry from an industrial ecology stand-point A well-known case is that of Texas Indus-tries which operates a cement plant in Texasadjacent to the EAF steel plant of a subsidiarycompany The EAF steel slag had for some timebeen marketed as a road aggregate selling for afew dollars per t a typical price for such materialThe companyrsquos research into alternative uses re-sulted in the patenting of the CemStar processwhich introduces steel slag into the cement kilnraw mix as a source of CaO and SiO2 Use ofsteel slag was well known in the cement industrybut traditionally not favored because of the dif-culty in crushing it The key discovery behindCemStar was that the material did not in factneed to be crushed to ner than about 5 cm di-ameter to melt easily in the kiln Once intro-duced steel slag acts as a ux (indeed it is exo-thermic) and provides the important clinkermineral C2S Its use as a substitute for some ofthe limestone feed yields clinker in about a 11tonnage ratio to the slag input increases thethroughput of an existing kiln by several percentand has the added benet of supplying some orall of the Fe2O3 needed to form the C4AF clinkermineral Steel slag thus reduces the amounts oflimestone and fuel required to produce clinkerand allows for reduced emissions of both CO2

and NOx (Perkins 2000 Forward and Mangan1999)

One of the most popular cementitious ex-tenders is GGBFS This material differs from or-dinary (air-cooled) blast-furnace slag in that thegranulated slag has been cooled by quenchingthrough a water stream This quenching formssand-sized particles of glass Unground granu-

lated blast-furnace slag nds a modest marketamong cement producers as a grinding aid in ce-ment manufacture Finely ground into GGBFSit is used to make blended cements or is sold toconcrete manufacturers as a partial portland ce-ment substitute In the United States only about03 Mtyr of granulated slag is consumed by thecement industry itself and of this about two-thirds is as a grinding aid and the rest (asGGBFS) within blended cement (van Oss2002) But a substantially larger market forGGBFS exists within the concrete industry Al-though published data for the United States areincomplete the size of this growing marketwould probably be at least 3 Mtyr overall basedon the 23 Mt reported GGBFS sales in 2001(Prusinski 2002) for member companies of thenewly formed Slag Cement Association capaci-ties of granulated slag-grinding plants and USGeological Survey (USGS) data on the produc-tion and imports of slag The number of operat-ing blast furnaces in the United States is in majordecline however and most of them just produceair-cooled slag for aggregate accordingly USproduction of granulated slag is currently inade-quate to meet demand Some of the excess de-mand has been met by imports of unground ma-terial which is then ground at US facilities Toboost US production of GGBFS some cementcompanies have found it worthwhile to enterinto arrangements with steel companies whereingranulation cooling is installed at the blast fur-naces and the cement companies construct therequisite grinding plants two such granulatorswere installed in 2001 (Holcim 2002 CementAmericas 2001) In this way the previous air-cooled slag by-product worth a few dollars per tas an aggregate is converted to a GGBFS co-product worth about $30 to $40 per t when solddirectly or is incorporated within a blended ce-ment selling for substantially more This coolingconversion is not inexpensive the granulationequipment costs several million dollars and agrinding plant about $40 million or more

Coal-red electric power utilities have longsupplied y ash and bottom ash to the cementindustry for use as a raw feed and y ash to boththe cement industry and especially concrete in-dustry for use as a pozzolan Power plants are un-der pressure to reduce NOx emissions and one



way to do this is to lower combustion tempera-tures (which reduces thermal NOx) A disadvan-tage of this reduction is that it often results in aless complete combustion of fuels which causesan increase in the carbon content of the y ashwaste product This additional carbon makes they ash unsuitable for direct pozzolan use inblended cement and many concrete applicationsTo remedy this problem the y ash can bewashed but it adds to the cost of the ash Carbonin y ash or other ashes is not a problem how-ever when the ash is used as a raw material forcement kilns

Another ash that has potential use as raw feedfor cement kilns is that from municipal waste in-cinerators The pioneering application of thismaterial has been in Japan The concern in usingthis secondary waste is that it can contain highand variable levels of toxic metals (Suto and Ka-neko 2000)

Another linkage between electrical utilitiesand cement plants is that the SOx-scrubbing ap-paratus of the power plant commonly uses lime-stone or lime as a reagent As noted in the SOxdiscussion synthetic gypsum is thereby formed inthe scrubber (SOx scrubbers at cement plantscan also make this mineral) This synthetic gyp-sum if sufciently clean has market applicationsincluding substitution for natural gypsum in ce-ment manufacture Although consumption dataare incomplete for synthetic gypsum (it is com-monly reported in with natural gypsum on USGSannual cement plant surveys) in 2000 at least5 of the gypsum consumed by the US cementindustry was obtained from SOx scrubbers

One of the better examples of a cement plantbeing constructed to take advantage of an exist-ing by-product is that of the Cajati (formerly theJacupiranga) mine in Brazil The mine exploitsan apatite [Ca5(PO4)3F]-bearing carbonatite (anigneous carbonate complex) to supply phosphatefor an adjacent fertilizer plant The minersquos pro-cessing plant has a waste stream of carbonatiterock composed of calcite and dolomite Owingto the low grade of the phosphate reserves a ce-ment plant was constructed near the mine in theearly 1970s to use the calcite wastes and so pro-vide an additional revenue stream to the minecomplex the cement plant is now owned by an-other company Because portland cement man-ufacture cannot tolerate much MgO in the raw

materials (part I) to date both mining and oreprocessing have had to keep the calcite ade-quately segregated from the dolomite whichcontains signicant MgO (Alves 2002) Effortsare underway however to utilize some of the do-lomitic material as a cement feed by segregatinglow MgO dolomitic material (blendable into theraw feed in low proportions) from that of higherMgO content Cajati thus also illustrates a situ-ation whereby a waste-generating entity canmodify its operating procedures to favor thewaste consumer The Cajati cement plant is alsobeing converted over to the partial burning ofwaste fuels (Vigon 2002)

The Geopolitics and Mitigationof Emissions from CementManufacture

As noted earlier emissions from cement man-ufacture are relatively small compared with otherindustrial or economic sectors with the signi-cant exceptions of CKD (the effects of which arelocal) and CO2 But even for CO2 emissionsfrom cement manufacture are small comparedwith those from the combustion of fuels by powerplants and motor vehicles Thus it is fair to ques-tion why there is concern about CO2 emissionsfrom cement manufacture and why much of thatconcern is expressed by the cement industry it-self Part of the concern stems simply from thefact that the issue of global warming has beenwidely publicized as has the role of CO2 thereinIn addition the overwhelmingly dominant CO2

emissions from combustion have been segregatedby many major reporting agencies especially theIPCC from so-called industrial emissions Underthis segregation cement manufacture stands outfor scrutiny as one of the two largest industrialemitters of CO2 The cement industry fears thatit may be negatively impacted by CO2 reductionpolicies resulting from the Kyoto Protocol orother emissions-reducing treaties in the futureHere the worry is that efforts to force reductionsby the sectors emitting considerably larger quan-tities of CO2 may impact the cement industrydisproportionately to its more modest role (about5 of total anthropogenic emissions)

One of the reduction strategies particularlyfeared by the cement industry is the impositionof a large carbon tax on fossil fuels throughout



the economy Depending on the size and stipu-lations of the tax a signicant increase in pro-duction costs for clinker could be incurred Someof the proposals are on the order of $50t carbonor a proportionate sum expressed per t of fuel(Humphreys and Mahasenan 2002 Nisbet1996) Such a high tax on fuels would increasefossil fuel costs to a cement plant by approxi-mately 100 or typically about $12t clinker Al-though data on production costs are notoriouslyscarce unreliable and often proprietary costs inthe United States appear to be in the range of$27t to $44t (cement) and for some modernplants in developing countries as low as $15tWhatever the case a $12t production cost in-crease would be crippling for most plants becauseonly rarely can large production cost increases bepassed along to the customers given the highlycompetitive nature of the cement market inmany countries or regions Within individualUS market regions for example proprietaryUSGS data show that it is uncommon to ndcement prices among producers varying by morethan $1t to $2t Some carbon tax proposalshowever apply only for emissions levels above acertain amount and still others allow for emis-sions trading (Hoenig and Schneider 2002)

Depending on the actual carbon tax it islikely that some of the older least-efcient plantswould become uneconomic and would be closedwhereas for other plants the tax would certainlybe an added inducement (beyond those of thenormal market) to seek opportunities to reducefuel consumption levels or costs such as byswitching partially to waste fuels Not all CO2

reduction strategies by governments would nec-essarily involve carbon taxes Tax rebates and re-ductions for example could be offered to com-panies to reduce emissions

Because only a few countries would initiallybe bound to the Kyoto Protocol cement produc-ers in countries not so bound would be less likelyto incur added production costs related to emis-sions reductions and so if they are exportersthey would be at a competitive advantage to pro-ducers in signatory (developed) countries thatimport signicant quantities of cement The UScement market currently has an import depen-dence of more than 20 (van Oss 2002) andabsorbs approximately 25 of the total world ce-ment trade the domestic cement industry would

be vulnerable to international production costdifferentials resulting from the unequal applica-tion of emissions reduction strategies especiallycarbon taxes

In the case of unequal application of largeKyoto Protocolndashrelated carbon taxes Nisbet(1996) concluded that the effects on the UScement industry would be severe competitionfrom cheap imports would likely force the clo-sure at a minimum of smaller less fuel-efcientcement plants (especially wet-process plants)Further the resulting production shortfalls mighteven encourage the establishment of largeexport-oriented plants in nearby nonsignatorycountries Nisbetrsquos study however did not con-sider the mitigative effects of protective tariffsthat might be imposed on imports that are notcarbon taxed nor the fact that transportationcosts might shield some otherwise vulnerableproducers from import competition This latterprotection however might not be as widespreadas supposed because cement can in fact be eco-nomically transported over long distances bytrain and barge and large tonnages (of both do-mestic and imported cement) are so moved inthe United States (van Oss 2002)

In the face of import (or general) competi-tion cement companies can to protect theirmarket share instigate business strategies such asvertical integration the purchase of control ofimport terminals and growth through the pur-chase of other cement companies particularlythose overseas Although the potential replace-ment of older technology production capacity inthe United States by newer more cost-efcientcapacity elsewhere might be viewed as benecialfrom a global standpoint it would only be so ifthe replacement plants had lower net emissionsa mere transfer of emissions sources from one lo-cation to another would not directly benet theglobal climate

Emissions Reduction StrategiesMany cement companies are trying to reduce

fuel consumption and associated CO2 emissionsfor purely economic reasons but some of the ac-tions are also seen as an essential business strat-egy for the future Notable in this regard and interms of a shift in corporate philosophy is therecently released report Toward a Sustainable Ce-ment Industry (Battelle 2002) and a set of asso-



ciated substudies (eg Vigon 2002) sponsoredby ten of the worldrsquos largest cement companiesThese studies have identied many issues facingthe cement industry and have suggested ways inwhich the industry can improve its specic per-formances (operational practices emissions prof-itability) operate using concepts from industrialecology and through these improvements re-ceive more widespread and favorable public ex-posure The study also encourages governmentsto help their countryrsquos cement industries employstrategies based on the principles of industrialecology

A number of emission reduction strategies areavailable to the cement industry Given itschemistry the calcination of limestone emitsCO2 but the partial substitution of other CaOsources reduces the quantities of limestoneneeded Alternative CaO sources are beingsought the key requirement being that they notdemand even more energy for CaO access thanis needed for the calcination of limestone Onepromising material discussed above is steel slag(CemStar) which not only provides CaO with-out calcination but also plays a uxing role (iereduces the energy and hence fuel requirementsfor the sintering reaction in clinker manufac-ture) A number of uxes are available such asuorspar (Lea 1970) but these may have morelimited application

Similar to uxes would be materials that evenif they do not change the temperature require-ments of clinker formation are more thermody-namically accessible that is they require less res-idence time in the kiln An example of thiswould be the substitution of reactive slag for sil-ica sand (Mishulovich 2003) A decrease in res-idence time allows for a higher kiln throughputcapacity for the same fuel burned and thus lowersthe unit heat consumption and associated com-bustion CO2 and NOx Energy savings can alsobe achieved through choosing materials that re-quire less preparation (such as crushing) prior toburning

Reduction of CO2 can also be accomplishedby upgrading plants to more efcient technolo-gies and by switching to lower-carbon-contentfuels Many technological changes also reduceemissions of other pollutants Regarding tech-nological changes strategies can include the

physical conversion of wet kilns to dry technol-ogy the upgrade of dry kilns and precalciner sys-tems to more efcient versions the replacementof wet kilns and VSKs with entirely new mod-ern preheater-precalciner dry kilns the optimi-zation of kiln burner designs the upgrade ofclinker coolers and the installation of comput-erized control technology (expert systems)Goodreviews of these options include those publishedby Hendriks and colleagues (1998) and especiallyby Martin and colleagues (1999) These upgradesgenerally reduce unit energy consumption levelsas discussed in part I It should be noted how-ever that many of these options are very expen-sive commonly costing well in excess of $10 mil-lion and might not be economically feasible forsome small older plants or plants lacking suf-cient limestone reserves (approximately 50 yearsrsquoworth) The limestone reserve picture is impor-tant because the projects are generally seen ashaving very long payback periods and price in-creases for cement cannot always easily be passedon to customers Switching to lower carbon fuelssuch as from coal to natural gas is technicallyfeasible at many plants but it can involve prob-lems of fuel cost and availability may increasethermal NOx output and may require technicalalterations to the kiln line

Another CO2 reduction strategy would be toallow the use (in some applications) of lower-quality portland or similar cements One of theways to do this is by reducing the clinkering tem-perature either directly or effectively by increas-ing material throughput speed This would affectthe process chemistry resulting in less C3S andmore C2S in the clinker and thus resulting inreduced early strength This could result in a ce-ment that is satisfactory for some applicationswhere early strength is not essential Anotherbut indirect strategy and one that is in use inEurope would be to grind the cement less nelythereby reducing the unit electricity consump-tion A less nely ground cement would be lessreactive (lower overall surface area) and wouldthus tend to develop strength more slowly thana more nely ground cement Yet another strat-egy is using a bulked-out cement such asportland-limestone cement as discussed below

In the long run there is a limit to how muchCO2 can be reduced through technical or oper-



ational improvements at the plant As noted byRuth and colleagues (2000) the unit energy sav-ings through efciency improvements althoughsignicant (about 10) is much smaller thanthose achievable by incorporation of by-productextendersmdashespecially those not requiring addi-tional heat inputs to manufacturemdashin the n-ished cement (15 to 30 or more)

Extenders can be cementitious (ie blendedcements as discussed earlier) or inert or relativelyso The most common relatively inert extenderis ground limestone This can be added in smallproportions (about 3 or so) without signi-cantly altering the properties of the portland ce-ment this is as yet not allowed in the UnitedStates Ground limestone can also be added inlarger proportions (6 to 35) to makeportland-limestone cement this has capturedsignicant market share in some countries in Eu-rope and elsewhere (Moir 2003)

Extenders reduce the clinker component ofthe cement and allow a plant to increase pro-duction capacity without the expense of actuallyupgrading or adding new equipment Blended ce-ments are in common use in many parts of theworld but as noted in many studies there ismuch remaining growth potential worldwidenotably in the United States and China (egDetwiler 1996 Soule et al 2002) The usual sub-stitution ratio is 11 that is the extender dis-places an equal weight of portland cement A fewextenders (such as silica fume) may displace morethan their weight of portland cement Blendedcements are rarely sold with more than about30 extenders although concrete manufacturersmay themselves mix in up to 50 or more (someprojects up to 75 to 80) as a partial substitutefor portland cement Extenders may or may notbe locally available or abundant Of the commonextenders y ash is available in many countriesand could be utilized more but such use as notedearlier could be constrained if NOx restrictionson power plants become widespread resulting ina high-carbon product The use of GGBFS is sub-ject to fairly severe availability constraints asdiscussed earlier A potential growth area in ex-tenders is the use of CKD such a use would allowthe cement plant the added benet of being thesource of the extender The use of ground lime-stone extenders to make portland-lime cements

has a large potential for growth but they gener-ally exhibit lower long-term strength thanstraight portland cements or blended cements(Moir 2003)

Currently the United States is the worldrsquosthird largest cement market about 1 of cementsales are of blended cements and on averageextenders make up at least 10 of the ldquocementrdquoin US concrete (van Oss 2002) A 10 ex-tender substitution in concrete means for theUnited States a current pozzolan consumptionon the order of 11 to 15 Mt based on an annualplus blended portland cement consumption ofabout 111 Mt The true cement consumption inthe country therefore is about 10 higher thanUSGS reported sales data suggest The volumeof pozzolons consumed also implies that if theeconomics of cement and concrete shifted to fa-vor the concrete companies purchasing blendedcements for 100 of their extender needs thenthe cement companies could signicantly in-crease their effective cement capacities by pro-viding these cements If building codes were al-tered to allow the use of more blended cementsor similarly the ofcial cement specicationswere changed from the current composition basis(eg type V portland cement) to a performancebasis (eg sulfate-resisting cement) then themarket for blended cements or equivalent con-cretes in the United States could increase dra-matically with a proportional decrease in aver-age clinker factors and unit emissions of nishedcements In a growing cement market a shift toblended cements would allow growth withoutincreased clinker output assuming that the extrademand would not be met simply through im-ports

Beyond the estimated 10 reduction in CO2

emissions possible by technical upgrades atplants the 10 to 15 from noncarbonate CaOsubstitution in raw materials and the 30achievable through blending (not all of thesesavings would be additive) further reduction ofCO2 emissions from cement manufacturingwould require more extreme developments intechnology the widespread adoption of radicallydifferent more efcient technology the intro-duction of CO2 sequestration technologies (asyet the economics of this do not appear attrac-tive [Hendriks et al 1998]) or a shift to different



non-clinker-based cements Martin and col-leagues (1999) discussed some of the possiblenew production technologies It should be notedthat given the huge capital investments residentin individual cement plants (kiln lines them-selves typically cost $50 million or more) and thelong-lived nature of the equipment near-termwidespread shifts away from rotary kilns are un-likely

In the long term signicant reductions inemissions may also require the development ofentirely new cements for general constructionpurposes Both Martin and colleagues (1999) andUchikawa (2000) discussed several new cementsand there are a number of others some of whichallow blending An interesting new concept isthat of cements based on activated magnesia(MgO) derived from calcination of magnesite(MgCO3) or dolomite [CaMg(CO3)2] Magnesitecalcination occurs at a substantially lower tem-perature than does that of calcite which savesenergy No clinker is made but the magnesia isblended with a small amount of portland cement(for additional strength) and pozzolan (to tie upthe lime from portland cement hydration) andthe magnesia develops strength ultimatelythrough the carbonation of the material (orrather its hydroxide) back to magnesite (TecEco2002) Carbonation of activated magnesia is fas-ter and more pervasive than is the case for limebut remains a diffusive process and the reactionrate thus slows with time Thus the favored usesof an activated magnesia cement would be inhigh-surface-area applications such as buildingblocks The relatively high rate of carbonationof magnesia could allow for the reabsorbtion ofa signicant proportion of the calcination CO2

evolved in the cementrsquos manufactureA nal approach to reducing the overall en-

vironmental impacts of cement manufacture is todevelop stronger cements and more durable con-cretes either of which if not overly expensivewould allow for less material use for the sameapplication One approach to increased concretedurability is through the introduction into theconcrete mix of so-called engineered cementi-tious composites (ECCs) such as bers that im-part much of the tensile strength currently con-tributed by steel reinforcing bars (Li 2002)Among other attributes of ECC concretes the

elimination of rebar in concrete could dramati-cally reduce concrete failure owing to rebar cor-rosion a very common problem with bridges andsimilar structures exposed to deicing salts

Overall and per capita annual cement andconcrete production and consumption are likelyto increase steadily for the world overall per cap-ita production of concrete (and mortars) alreadyexceeds 2 t per year As indicated in part I (g-ures 1 and 2) and in gure 2 and table 2 of thiscurrent article the developing world has the bulkof current cement output and capacity and willlikely experience the greatest growth in bothAsia currently has more than 50 of world pro-duction China itself has approximately one-third of the total world cement output (van Oss2002) and capacity spread among several thou-sand cement plants

The size and growth of the industry makes thereduction of its environmental impacts very im-portant The cement industry is presently highlyconsolidated worldwide estimates vary but ap-proximately 40 of world production and capac-ity is controlled by just a dozen internationalcompanies As noted above ten of these com-panies have joined forces in investigating indus-trial ecology strategies and have committed totheir implementation Other companies arelikely to follow suit Industrial symbiosis or in-dustrial ecosystems tend to be thought of insomewhat geographically restrictive terms of in-dustrial complexes sometimes having a few out-lier facilities (Vigon 2002) however the cementindustry on the corporate level can expand theconcept to a global scale A single large multi-national cement company operates dozens ofplants worldwide Each of these plants is or canbe tied into a central information exchange andmarketing and credit network Each plant (andthe corporate headquarters) is thus able to drawon the expertise resident in the other plantsrsquo per-sonnel and to evaluate numerous technologiesoperational practices and raw materials perfor-mances at different plants and scales This ineffect is a linkage of at least the informationalaspects of multiple local industrial ecosystemswith potential linkages with respect to the move-ment of some materials Ultimately given appro-priate international agreements this globaliza-tion may allow for a worldwide system of



environmental charges and credits in the cementindustry so that emissions of a less well-performing individual plant might be chargedagainst emissions credits of a lower emissionsfacility

In that cement manufacture is a highlymaterials- and energy-intensive process and hassignicant associated emissions the industry canbe thought of as highly inefcient But these veryinefciencies coupled with the ability to utilizea wide variety and large quantities of waste rawmaterials and fuels can be thought of as vir-tuesmdasha veritable silk purse out of a sowrsquos earmdashin that they provide the opportunity and incen-tive for the cement industry to engage inindustrial ecology practices Indeed the cementindustry may be a better driver of such practicesthan any other manufacturing industry

Acknowledgments

Data compilations within Minerals Yearbooksand other publications of the US Bureau ofMines and the US Geological Survey were de-rived from annual canvasses of the US cementindustry extending back more than 120 years andrepresent an unparalleled record of cooperationin the provision of highly proprietary productionoperational and sales data by all of the US ce-ment companies as well as their industry orga-nization the Portland Cement Association Thiscooperation is gratefully acknowledged The ar-ticle beneted from the efforts of ve anonymousreviewers of the manuscript

Notes

1 Unless otherwise noted ldquotonrdquo refers to metric ton(1 metric ton 4 1 Mg [SI] and is about 11 shorttons)

2 This article does not examine environmental im-pacts of cement during the use or end-of-lifephases of the cement product (ie concrete) lifecycle

3 These emissions are described as potential emis-sions because they are based on clinker capacitiesrather than actual production

4 Electrical consumption for 2000 was assigned en-tirely to portland cement to be consistent withthe reporting practice for the 1950ndash1995 data intable 5 of part I Electricity consumption is better

expressed in terms of total (including masonry)cement and for this could be further broken outas 131 kWht for wet plants and 148 kWht fordry plants (table 2 of van Oss 2002)

5 Steel is basically an alloy of iron and a smallamount (rarely more than 1 and commonlymuch less) of carbon and other elements Theiron in steel can be derived from iron ores scrap(ie remelted) steel or a combination of the twoBlast furnaces in one form or another account forthe vast majority (93 in 2000) of world ironproduction from ores (Fenton 2002) the USfraction in 2000 was 97 Very simply a blastfurnace strips the oxygen from iron oxide ore (orits sinter) by reducing it at high temperatureswith carbon (generally from coke) to form carbonmonoxide (CO) which ultimately is oxidized toCO2 and removes silica and other impuritiesfrom the ore by reacting it with a limestone andor dolomite [(CaMg)CO3] ux to produce blastfurnace slag and CO2 (from ux calcination)Thus both ore reduction and slag formation ul-timately produce CO2 in large quantities (seebelow)

The iron ore ux and coke charges or burdento a blast furnace will vary somewhat dependingprincipally on the chemical composition of theiron ore and whether any ferrous scrap is beingadded (Bray 1942 USS 1964) Very approxi-mately for a furnace burden that does not includeferrous scrap the production of 1 t of crude ironwill involve the consumption of 05 to 10 t ofcoke as fuel and reductant and 02 to 05 t ofcarbonate ux Except for about 4 carbon re-maining in the crude iron all of the carbon incoke will ultimately become CO2 for the cokeproportion noted above and typical carbon con-tents therein the emissions yield would be about16 to 32 t CO2t crude iron The IPCC citesdefault emissions for coke of 31 t CO2t iron(IPCC 2000) For the carbonate ux the tonnagerange shown would yield calcination emissions(see methodology in Appendix A) of about 01to 02 t CO2t crude iron (at this level of round-ing limestone and dolomite yield the same emis-sions levels) Thus at a minimum the total unitCO2 emissions would be about 17 t CO2t crudeiron (again provided that ferrous scrap was notinvolved) which is almost double that for clinkeror cement as shown earlier Fruehan and col-leagues (2000) listed somewhat lower emissionsof 145 to 156 t CO2t crude iron but their cal-culations assumed a remaining carbon in thecrude iron of 5



In the conversion of crude iron to steel thecarbon content will be reduced by oxidation (toCO2) from about 4 to about 1 or less as notedabove which will yield further emissions of 0036t CO2t steel for every 1 (absolute) carbon re-duction Steel made from remelted scrap has con-siderably lower associated emissions

6 Part of the Borjesson and Gustavsson (2000)study appears awed by their estimates of theamount of CO2 capture by the ldquocarbonisationrdquo(carbonation) of hardened concrete The carbon-ation reaction is Ca(OH)2 ` CO2 U CaCO3 `

H2O and is a surcial process that very and in-creasingly slowly diffuses inward and along cracksurfaces in the concrete The authors appear toassume that all of the calcination CO2 releasedfrom the manufacture of cement would be reab-sorbed through carbonation whereas in fact theprocess principally affects free lime releasedthrough the hydration of clinker minerals signi-cant carbonation of the silicate minerals is ex-tremely slow (centuries) It can be shown that thenet free lime from clinker hydration is only aboutone-third of the total CaO contained in the origi-nal clinker minerals (having the mineralogical ra-tio shown in table 2) Accordingly only aboutone-third of the calcination CO2 could be reab-sorbed through carbonation if the process wentto completion but this is unlikely on a scale ofdecades and assumes that none of the free limehas been chemically bound by pozzolans In a re-connaissance study Gaijda (2001) provided datathat show that for a 50 year (1950ndash2000) massof concrete emplaced in the United States 50yearsrsquo worth of carbonation accounts for (very ap-proximately) only about 2 of the total calci-nation CO2 released in the manufacture of thecement in the concrete as a result of some am-biguous wording the study appears to claim thisamount as being annual absorption but thiswould be an incorrect interpretation of the datapresented

7 Most recently it has been proposed that cementkilns burn bonemeal from the carcasses of cowssuspected of carrying so-called mad cow disease(uncontaminated) bonemeal has already beenused as fuel for cement kilns in several Europeancountries for some time (ICR 2001)

8 Although a huge variety of waste materials canbe burned a given plant will generally use only afew materials or a preblended mix of materials(especially common with liquid wastes) to keepthe kiln burning conditions optimized and as uni-form as possible and so as not to compromise

clinker quality In this respect the kiln may beless exible than municipal incinerators

9 Vanderborght and Brodmann (2001) may havebeen misled by the fact that the IPCC report(IPCC 2000 311 footnote 1) misstated the totaloutput of CKD (itself) as 15 to 20 of theweight of the clinker this typographical error wasnot present in the original draft of the documentThe CO2 contribution from the CKD howeverremains correct in the footnote

10 USGS ldquoenergyrdquo data for cement have been citedin certain reports (eg Martin et al 1999) but infact were independent energy conversions of theactual USGS data on the types and quantities offuels and electricity consumed

References

Alves P R 2002 Varied applications of waste andtailings Case analysis of a low grade hard rockBrazilian phosphate mine Abstract In Procthirty-eighth forum on the geology of industrial min-erals April 28ndashMay 1 St Louis MO

Battelle 2002 Toward a sustainable cement industrySummary report Battelle study for the WorldBusiness Council for Sustainable DevelopmentColumbus OH Battelle

Bhatty J I 1995 Alternative uses for cement kiln dustReport RP 327 Skokie IL Portland Cement As-sociation

Boarder R W F 1997 The benets of alternativefuels Cement environmental yearbook 1997 Dor-king UK Tradeship Publications Ltd pp 99ndash104

Borjesson P and L Gustavsson 2000 Greenhouse gasbalances in building construction Wood versusconcrete from life-cycle and forest land-use per-spectives Energy Policy 28 575ndash 588

Bray J L 1942 Ferrous production metallurgy NewYork John Wiley and Sons

BCA (British Cement Association) 1997 Preventionand abatement of SO2emissions Cement environ-mental yearbook 1997 Dorking UK TradeshipPublications Ltd pp 112ndash 117

Cembureau 1996 World cement directory Brussels Eu-ropean Cement Association

Cement Americas 2001 Lafarge slag facility construc-tion moves ahead Cement Americas (SeptemberndashOctober) 3

Degre J P 1998 Strategy and principles for alterna-tive fuels and raw materials (AFR) coprocessingin the cement industry In Proceedings 34th inter-national cement seminar Salt Lake City UT En-glewood CO PRIMEIDA Intertec Exhibitionsand Conferences pp 125ndash 135



Delles J B H M Kanare S T Padiyara and D JBroton 1992 Trace metals in cement and kiln dustfrom North American cement plants Portland Ce-ment Association publication SP11001 SkokieIL Portland Cement Association

Detwiler R J 1996 Blended cement Now and forthe future Rock Products Cement Edition 99(7)27ndash32

Duda W H 1985 Cement data-book Vol 1 Thirdedition Berlin Bauverlag

Fenton M D 2002 Chapter in Vol 1 Iron and steelMinerals yearbook Metals and minerals RestonVA US Geological Survey

Forward G and A Mangan 1999 By-product syn-ergy The Bridge 29(1) 12ndash15

Frohnsdorff G ed 1986 Blended cements SpecialTechnical Pub 897 Philadelphia PA AmericanSociety for Testing and Materials

Fruehan R J O Fortini H W Paxton and R Bin-dle 2000 Theoretical minimum energies to producesteel for selected conditions Washington DC USDepartment of Energy Ofce of Industrial Tech-nologies

Gaijda J 2001 Absorption of atmospheric carbon dioxideby portland cement concrete Portland Cement As-sociation research and development serial 2255aSkokie IL Portland Cement Association

Gilling A 1999 Waste to riches International CementReview (July) 56ndash58

Glover J D O White and T A G Langrish 2002Wood versus concrete and steel in house con-struction A life cycle assessment Journal of For-estry 100(8) 34ndash41

Gossman D 1993 A comparison of metal emissionsfrom cement kilns utilizing hazardous waste fuelswith commercial hazardous waste incinerators InProc twenty-ninth international cement conferenceSan Francisco Chicago Rock ProductsIntertecPublishing pp 159ndash 168

Haspel D 2002 Lowering NOx for less InternationalCement Review (January) 63ndash66

Haynes B W and G W Kramer 1982 Characteriza-tion of US cement kiln dust US Bureau of MinesInformation Circular IC 8885 Washington DCUSBM

Hendriks C A E Worrell L Price N Martin LOzawa Meida D de Jager and P Riemer 1998Emission reduction of greenhouse gases from thecement industry In Proc fourth international con-ference on greenhouse gas control technologies Au-gust 30ndashSeptember 2 Interlaken

Hoenig V and M Schneider 2002 The effects of CO2

trading on energy-intensive sectors of industryZKG International 55(5) 64ndash73

Holcim Ltd 2002 Annual report for 2001 Jona Swit-zerland Holcim

Humphreys K and M Mahasenan 2002 Toward asustainable cement industry Substudy 8mdashClimatechange Battelle study for the World BusinessCouncil for Sustainable Development Colum-bus OH Battelle

ICR (International Cement Review) 2001 Morebonemeal to fuel European kilns International Ce-ment Review (January) 9

IPCC (International Panel on Climate Change) 1997Greenhouse gas inventory reference manual RevisedIPCC guidelines for national greenhouse gas inven-tories Vol 3 Bracknell UK IPCC

IPCC 2000 Good practice guidance and uncertaintymanagement in national greenhouse gas inventoriesHayama Japan Institute for Global Environmen-tal Strategies

Jenkins B G and S B Mather 1997 Fueling thedemand for alternatives Cement environmentalyearbook 1997 Dorking UK Tradeship Publica-tions Ltd pp 90ndash97

Kahn R D 1998 Considerations for evaluating a bio-solids injection program Proc thirty-fourth inter-national cement seminar Salt Lake City UT En-glewood CO PRIMEIDA Intertec Exhibitionsand Conferences

Kosmatka S H and W C Panarese 1988 Design andcontrol of concrete mixtures Thirteenth editionSkokie IL Portland Cement Association

Krogbeumker G 1994 Staying under the limit Inter-national Cement Review (May) 43ndash50

Lanier S and E Hansen 2000 Control of NOx emis-sions from cement manufacturing In Proceedings36th international cement seminar Charleston SCEnglewood CO PRIMEIDA Intertec Exhibi-tions and Conferences pp 175ndash185

Lea F M 1970 The chemistry of cement and concreteThird edition New York Chemical Publishing

Li V C 2002 Advances in ECC research ACI specialpublication on concrete material science to ap-plications SP 206ndash 23 Farmington Hills MIAmerican Concrete Institute pp 373ndash400

Malhotra V M ed 1989 Fly ash silica fume slagand natural pozzolans in concrete Proc third in-ternational conference of the American concreteinstitute Three volumes SP-114 Detroit MIAmerican Concrete Institute

Marland G T A Boden R C Grifn S F HuangP Kanciruk and T R Nelson 1989 Estimates ofCO2 emissions from fossil fuel burning and ce-ment manufacturing ORNLCDIAC-25 OakRidge TN Oak Ridge National Laboratory

Martin N E Worrell and L Price 1999 Energy ef-ciency and carbon dioxide emissions reduction op-portunities in the US cement industry LBNL-



44182 Berkeley CA Lawrence BerkeleyNational Laboratory

Mayeld L L and H O Biggs 1997 Non-adversarialenvironmentalism and the power of biosolids Ce-ment environmentl yearbook 1997 EnglewoodCO PRIMEIDA Intertec Exhibitions and Con-ferences pp 105ndash111

McCaffrey R 1994 CKD and the EPA reportInternationalCement Review (May) 68ndash74

Mishulovich A 2003 Alternative materials Interna-tional Cement Review (January) 59ndash62

Moir G K 2003 Gaining acceptance InternationalCement Review (March) 67ndash70

Nemerow N 1995 Zero pollution for industry Wasteminimization through industrial complexes NewYork John Wiley and Sons

Nievoll J 1997 Friendly alternatives Cement Envi-ronmental Yearbook 1997 Englewood CO PRI-MEIDA Intertec Exhibitions and Conferencespp 139ndash142

Nisbet M 1996 Will US climate change commit-ments shrink (sink) its cement industry In Pro-ceedings 32nd international cement seminar NewOrleans LA Chicago Rock ProductsIntertecPublishers pp 5ndash25

Nordqvist J and L J Nilsson 2001 Prospects for in-dustrial technology transfer in Chinese cementindustry In Proc ACEEE summer study on energyefciency in industry Tarrytown NY Vol 2 LundSweden Lund University

Olsen P B Y J Lee C C Leivo and K Visby-Kjaegaard 1998 Experience with FGD and heatrecovery system for a white clinker kiln over aperiod of seven years in Denmark In Proc thirty-fourth international cement seminar Salt Lake CityUT

Palmer G 1999 Beneting from CKD World Cement30(12) 78ndash85

Perkins D 2000 Increased production and loweremissions World Cement 31(12) 57ndash59

PCA (Portland Cement Association) 1992 An anal-ysis of selected trace metals in cement and kiln dustPublication SP109 Skokie IL Portland CementAssociation

PCA 2002 US and Canadian labor-energy input sur-vey 2000 Skokie IL Portland Cement Associ-ation

Price L E Worrell and D Phylipsen 1999 Energyuse and carbon dioxide emissions in energy-intensiveindustries in key developing countries LBNL-45292Berkeley CA Lawrence Berkeley National Lab-oratory

Prusinski J R 2002 The birth of the Slag Cement

Association Cement Americas (Septemberndash Oc-tober) 12ndash16

Rajbhandari C D 1996 In-line for a comeback In-ternational Cement Review (March) 47ndash50

Ruth M E Worrell and L Price 2000 Evaluatingclean development mechanism projects in the cementindustry using a process-step benchmarking approachLBNL-45346 Berkeley CA Lawrence BerkeleyNational Laboratory

Schwab J K Wilber and J Riley 1999 And SO2canyou International Cement Review (January) 54ndash55

Smart J P P J Mullinger and B G Jenkins 1998Combustion heat transfer and NOx World Ce-ment 29(12) 14ndash25

Soule M H J S Logan and T A Stewart 2002Toward a sustainable cement industry Trends chal-lenges and opportunities in Chinarsquos cement industryBattelle study for the World Business Council forSustainable Development Columbus OH Bat-telle

Suto K and Y Kaneko 2000 Using waste responsiblyInternational Cement Review (April) 75ndash82

Taylor H F W 1964 The chemistry of cements Twovolumes New York Academic Press

TecEco 2002 Information on Eco-Cement ^wwwtececocomau amp Accessed February 2003

Uchikawa H 2000 Approaches to ecologically be-nign system in cement and concrete industryJournal of Materials in Civil Engineering 12(4)320ndash329

US EPA (US Environmental Protection Agency)1994 Emission factor document for AP-42 section116 Portland cement manufacturing Final re-port EPA web document ^wwwepagovttnchiefap42ch11bgdocsb11s06pdfamp AccessedFebruary 2003

US EPA 1995 Portland cement manufacturing EPAweb document ^wwwepagovttnchiefap42ch11nalc11s06pdfamp Accessed February 2003

US EPA 1997 Summary report nitrogen dioxide EPAweb document ^wwwepagovairagtrnd97brochureno2htmlamp Accessed February 2003

US EPA 1998 NOx how nitrogen oxides affect theway we live and breathe EPA-456F-98-005Washington DC US EPA

US EPA 1999a 40 CFR Part 63 National emissionstandards for hazardous air pollutants for sourcecategories Portland cement manufacturing in-dustry nal rule Federal Register 64 (June 14)31898ndash 31962

US EPA 1999b 40 CFR Parts 259 261 266 and270 Standards for the management of cement



kiln dust proposed rule Federal Register 64 (Au-gust 20) 45632ndash 45697

US EPA 2000 Sources of dioxin-like compounds inthe United States In Draft exposure and humanhealth reassessment of 2378-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds EPA600P-00001 Bb Washington DC US EPA

US EPA 2002 Inventory of US greenhouse gas emis-sions and sinks 1990 ndash2000 Washington DCUS EPA

USGS Various years (1995ndash 1997) Cement chap-ter(s) in Minerals Yearbook Metals and mineralsVol 1 Reston VA USGS

US NEIC (US National Energy Information Cen-ter) 1977 Energy interrelationships mdashA handbookof tables and conversion factors for combining andcomparing international energy data WashingtonDC US National Energy Information Center

USS (United States Steel) 1964 The making shapingand treating of steel Eighth edition PittsburghPA United States Steel

Vanderborght B and U Brodmann 2001 The cementCO2protocol CO2emissions monitoring and re-porting protocol for the cement industry WorldBusiness Council for Sustainable Developmentworking group cement ^wwwwbcsdcementorgpdfco2protocolpdf amp Accessed 4 February 2002

van Oss H G 2002 Minerals yearbook Metals andminerals Vol 17 Cement Reston VA USGeological Survey

van Oss H G and A C Padovani 2002 Cementmanufacture and the environment Part 1Chemistry and technology Journal of IndustrialEcology 6(1) 89ndash105

Vigon B 2002 Toward a sustainable cement industrySubstudy 9mdashIndustrial ecology in the cement indus-try Battelle study for the World Business Councilfor Sustainable Development Columbus OHBattelle

Wernick I R Herman S Govind and J Ausubel1997 Materialization and dematerializationmeasures and trends In Technological trajectoriesand the human environment edited by J Ausubeland H D Langford Washington DC NationalAcademy Press pp 135ndash156

Worrell E L Price N Martin C Hendriks and LOzawa Meida 2001 Carbon dioxide emissionsfrom the global cement industry Annual ReviewEnergy Environment 26 303ndash329

Young G L and M von Seebach 1998 NOx vari-ability emissions and control from portland ce-ment kilns In Proceedings 34th international ce-ment seminar Salt Lake City UT EnglewoodCO PRIMEIDA Intertec Exhibitions and Con-ferences

Appendix A Methodology forCalculating CO2 Emissions fromthe Calcination Reaction inClinker Manufacture

Given the large amounts of CO2 evolved dur-ing cement manufacture and the desire to estab-lish national GHG inventories there has beenconsiderable interest in precisely calculatingCO2 emissions by the cement industry Essen-tially all emissions of CO2 during cement manu-facturing are from the manufacture of the clinkerintermediate product and is from two pathwaysthe calcination of calcium carbonate and thecombustion of fuels The methodology to calcu-late the emissions from calcination is presentedhere that for fuel combustion is discussed in Ap-pendix B Rival methodologies to calculate cal-cination emissions make use of identical equa-tions (both are based on the basic calcinationreaction CaCO3 U CaO ` CO2() but differ intheir approach (back-calculation from clinker orforward calculation from raw materials) and to aslight degree in the emissions factors utilized Fora country-level emissions calculation the ap-proach taken in this article data availability fa-vors the back-calculation from clinker approachalthough much of the discussion is applicable toa raw materials approach as well The method-ology described here is based on that adopted bythe IPCC (1997 2000)

As discussed in part I (see table 1 of part Ivan Oss and Padovani 2002) a typical portlandcement clinker has a CaO content of about 65(the range is 60 to 67 with most clinkers inthe range of 63 to 66 skewed toward theupper end) For the calcination CO2 calculationthe important assumption is made that all of theCaO comes from CaCO3 regardless of whetherthe CaCO3 comes from limestone (the main rawmaterial) or another type of rock The CaCO3

source assumption generally introduces only asmall (1 to 3) error on a country scale Onan individual plant basis however if it is knownthat a given plant burns a large quantity of rawmaterials containing noncarbonate CaO (such asCa silicates in igneous rocks or in slags andashes) then an adjustment for this source shouldbe made in the CO2 calculation Another as-sumption is that all of the CO2 comes from



CaCO3 and not from other carbonates such asthose of magnesium (see below) iron or man-ganese the error in this assumption is small un-less one knows that these other carbonates arepresent in the raw materials in signicant quan-tities

In the calcination equation given above theCaO fraction is 5603 of the original weight ofthe CaCO3 and the CO2 fraction is 4397 Ac-cordingly the weight (X) of CaCO3 required toyield 065 ton CaO in 1 ton of clinker (ie 65CaO) would be X 4 065 ton(05603) 4

11601 tons (unrounded) This weight of CaCO3

yields CO2 in the amount of 11601 ton CaCO3

(04397) 4 05101 ton CO2 (unrounded) or 051ton CO2 (rounded)

By comparison 1 ton of a 60 CaO clinkerwould back-calculate to 047 ton (rounded) ofcalcination CO2 and for a 67 CaO clinker theCO2 release would be 053 ton The ratio 051ton CO2ton clinker (of 65 CaO) was adoptedas the default calcination emissions factor by theIPCC (2000) Given imprecisions in the re-ported weights and chemical specications ofclinker produced in the assumptions regardingthe derivation of oxides as noted above and inthe inclusion of a CKD ldquocorrectionrdquo factor (seebelow) the use of the 051 ton calcination CO2ton clinker emissions factor yields a calcinationCO2 emissions estimate expected to be accurateto within 5 to 10 of actual emissions (IPCC2000)

Forward calculation from raw materials maybe preferable for individual cement plants cal-culating their own emissions as the plants havedetailed data on their raw materials These plant-specic raw materials data (both tonnage andchemical composition) however are essentiallyunavailable on a country basis As correctlynoted in an article by Vanderborght and Brod-mann (2001) describing the main rival meth-odology the IPCC (2000) method presentedabove does not consider the MgO content ofclinker Vanderborght and Brodmann prefer aslightly higher MgO modied calcination emis-sions default factor of 0525 ton CO2 (total)tonclinker based on the assumption that the MgOis from an MgCO3 (magnesite) phase in the lime-stone The average MgO content they used forclinker was 14 (versus the 1 shown in table

1 of part I) Although it is more likely that MgOwould be present as a dolomite [CaMg(CO3)2]phase rather than as magnesite (the latter beinga comparatively rare mineral) the stoichiometriceffect on CO2 is the same for both minerals ifone also corrects for the CaO contribution by thedolomite Using a comparable arithmetic ap-proach as for the pure CaCO3 example above itcan be shown that for an MgO bearing clinkerof 65 CaO the default emissions factor wouldbe (per ton of clinker) [05101 ` M(0011)] tonCO2 (result to be rounded to two decimal places)where M is the percent MgO in the clinker Thiswould be a maximum contribution and wouldneed to be reduced proportionally for any MgOfrom noncarbonate sources (highly likely) Be-cause the small MgO contribution to CO2 wouldbe subsumed in the overall 5 to 10 errorrange of the general calculation it may be arguedthat the additional CO2 from consideration ofMgO is insignicant at least for country-levelcalculations and for this reason MgO has beenignored in this article

Another criticism of the IPCC (2000) meth-odology made by Vanderborght and Brodmann(2001) was that the methodology may inade-quately correct for the contribution to CO2 emis-sions by the generation of any CKD not subse-quently recycled to the kiln 9 A high (butvariable) proportion of CKD represents materialincorporating calcined calcium carbonateWhere CKD is returned to the kiln it becomespart of the clinker and the CO2 emissions asso-ciated with this CKD are subsumed within thosecalculated for the clinker But the emissions cal-culated for the clinker would not include thosefrom ldquolostrdquo CKD which is CKD not returned tothe kiln because it was either landlled or usedfor other purposes a correction for this lost ma-terial is desirable in a rigorous calculation Al-though Vanderborght and Brodmann (2001)provided a complex formula for plant-level cal-culations of lost CO2 from CKD the calculationis impractical for country-level determinationsIn the absence of better CKD data the IPCCmethod provides for a default emissions additionfor lost CKD of 2 of the amount of CO2 cal-culated for the clinker itself In fact this 2 de-fault addition is not unreasonable assuming a to-tal generation of CKD equivalent to 15 to 20



of the weight of the clinker produced a calci-nation factor of the CKD of 33 and a 33nonrecycling (to the kiln) ratio Such ratioswould be applicable to modern rotary kiln linesbut could signicantly understate the lost CKDCO2 from VSKs or older rotary kilns lacking ef-cient CKD recovery systems Given the 5 to10 error of the IPCC method (this error rangewould also be applicable to the rival methodol-ogies) a 2 CKD correction can be safely omit-ted for rst-order estimations of country-levelemissions of calcination CO2

Appendix B Methodology forCalculating CO2 Emissions fromthe Combustion of Fuels inClinker Manufacture

Emissions of CO2 from cement manufacturingare both from calcination of limestone and fromfuel combustion Although calcination emissionscan be calculated fairly precisely (Appendix A)the calculation of CO2 from fuel combustion issignicantly more difcult and less precise themethodology is described here

Combustion (oxidation) efciencies are com-monly (if casually) expressed as being 100 incement kilns based on the high combustion tem-peratures and long residence times But unlikeclinker raw materials the residence times for thefuels and combustion gases are actually not allthat long (typically seconds to minutes) and notall parts of the kiln are at the high temperaturesof the sintering zone (part I) Further kerogen orother organic matter in the clinker raw materials(as opposed to the fuels) commonly burns in thepreheating and or precalcining sections of thekiln line where temperatures are more modestand combustion may therefore be incompleteIncomplete combustion can lead to the forma-tion of CO rather than CO2 The amount of COreleased is quite variable but it is extremely smallby comparison with CO2 According to the USEPA (1995) cement kilns typically emit CO inthe range of about 006 to 18 kgton clinker Aswith other studies of combustion contributionsto GHGs (eg IPCC 2000) it is assumed in thisarticle that any CO released by cement kiln linesultimately is converted to CO2 and may thus beignored as a distinct species (ie should be cal-culated as if it were CO2)

Theoretical energy (and hence fuel) require-ments to make clinker from a proportioned mixof pure oxides or from a theoretical mix of rawmaterials are invariably much less than the actualfuel requirements for the kiln The reasons forthis difference stem from variations in plant de-sign and operating parameters heat losses highlyvariable energy requirements to break down dif-ferent raw materials and recombine them intothe clinker minerals and variations in the chem-ical composition and energy yield of fuels Givenadequate data a rigorous approach to combus-tion CO2 based on actual fuel consumption ispractical on a plant-specic basis but these dataare generally lacking on a country basis For thevery few countries for which annual fuel con-sumption data (by type) are available for the ce-ment industry there are generally no speciccompositional data associated with the fuelsOne could however apply standard carbon con-tents data for each fuel to derive approximatecombustion CO2 emissions with the signicantcaveat that data on waste fuels (such as typetons dilution factors energy yields and carboncontents) are almost invariably poor or lackingentirely

Given type and tonnage data for fuels theCO2 calculation based on standard carbon con-tents of fuels is complicated by the fact that mostpublished conversion factors (ldquocarbon factorsrdquo)are not reported in terms of tons of carbon perton of fuel but are instead expressed as tons ofcarbon per unit energy yield of the fuel (eg coalat 2575 Mt carbon-equivalent per 1015 Btu)The carbon factors thus require fuel-specic en-ergy yield data in order to isolate the carbonwhich may then be converted to CO2 (tons car-bon 2 36642 4 tons CO2) Both the carbonfactor and energy data have signicant potentialfor rounding errors in actual use Further forsome fuels a range of standard energy values maybe provided and the carbon and energy datacommonly are based on laboratory testing con-ditions rather than on actual industrial combus-tion experience (which usually realizes lowerheat yields) For liquid and gaseous fuels in par-ticular the standard energy yield data vary sig-nicantly depending on whether the data are fornet (or low) heat contents values or gross (high)heat contents yet this distinction is rarely made



in the literature Net heats are more realistic forcomputing process-available energy yields of fu-els (they account for typical impurities and heatlosses from evaporation of residual moisture) andappear to be the basis of energy reporting whereavailable for the cement industries of most coun-tries other than the United States Gross heatvalues however which are used by the US ce-ment industry are probably better than net val-ues for computing the true total energy require-ments to make clinker (table 4 in part I) andespecially the CO2 yields of fuels as the energyvalues are more complete Gross heat valueswould yield higher CO2 emissions per ton ofliquid or gaseous fuel but where solid fuels aredominant (most US plants) the overall in-crease in combustion CO2 over that for net heatsis only 001 to 002 ton CO2 ton clinker

Because the US cement industry continuesto utilize a variety of rotary kiln technologies andhas plants varying widely in age that burn a broadvariety of fuels the range of energy consumption(table 4 in part I) and combustion CO2 emissions(table 1) per ton of product may be taken asbroadly comparable to the range for rotary plantsoutside the United States Two sets of combus-tion emissions data for the US industry areshown in table 1 Data for the rst set are basedon the carbon factor data published by the USEPA (2001) standard heat content data pub-lished by the US National Energy InformationCenter (US NEIC 1977) and Degre (1998) (forwaste fuels) data for the second set are based onplant-specic gross heat data reported to theUSGS in 2000 Estimates for heat yield and car-bon content were made for undifferentiatedwaste fuels not listed in these sources Data onthe type and quantity of fuels consumed werethose collected by the USGS as part of its annualcement industry surveys The quality of thesedata is considered to be good except possibly fornatural gas and for waste fuels in the early yearsof their use (mid-1980s through 1992) Naturalgas data for individual plants are subject to order-of-magnitude reporting errors that can be dif-

cult to identify because the fuel can be used bothfor warming up the kiln and for full kiln opera-tion Data on consumption of waste fuels was notcollected prior to 1993 despite their use actuallyhaving begun in the mid-1980s or possibly evenearlier at a few plants Further the waste fueldata other than tires are collected only in termsof broad categories (eg other solid wastes liq-uid wastes) that do not adequately distinguishamong the wide variety of waste fuels (hence car-bon and heat contents) burned

Data from the USGS surveys on the energyyields of fuels consumed by the cement industryhave not been routinely published10 Annual en-ergy consumption data for the US and Cana-dian cement industries are however availablefor recent decades from the Portland Cement As-sociation (eg PCA 2002) but the data onlycover its member companiesrsquo plants and sharesome of the same data collection and reportingproblems as the USGS surveys The average en-ergy consumption values for both US data setsare similar the USGS values are about 5higher (see part I)

Although it is evident from table 1 that dif-ferent combustion CO2 emissions values can bederived for the same fuel consumption data thedifferences among the values for any one yearamount to only 8 to 12 for combustion CO2

and only 5 to 6 for total (combustion pluscalcination) CO2 Given uncertainties in thebase fuel quantities the time series analysis of thedata in table 1 (and probably comparable datawhere available for other countries) should beconsidered accurate to within about 5 at bestthat is year-to-year variations of less than 5may not be real

About the Authors

Hendrik G van Oss an economic geologist is thecement commodity specialist with the US GeologicalSurvey in Reston Virginia USA Amy C Padovaniis a graduate student in the Department of Civil En-gineering at Stanford University California USA



Introduction

Construction materials constitute some 70of the nonfuel materials ows in the UnitedStates (Wernick et al 1997) Concrete andmortars are critically important constructionmaterials concrete is used as a bulk buildingmaterial in its own right and mortars are usedto bind together bricks stone or other blocksin masonry-type construction Concretes andmost mortars rely on hydraulic cement bindersfor their strength and durability but despitethis the dry cement component in these ma-terials is rather small (eg about 10 to 12by volume of the concrete mix) Most of theconcrete which is essentially an articial con-glomerate is a mix of sand and gravel or otherne and coarse aggregates (65 to 80) water(about 14 to 21) and air (05 to 8) Thecombination of cement and water in the con-crete mix is called cement paste Composition-ally mortars differ from concrete chiey in thefact that they contain only ne aggregates andthe hydraulic cement contains plasticizingagents Typically 1 ton (t)1 of cement sufcesfor about 3 to 4 cubic meters (m3) of concreteweighing about 7 to 9 t Current world outputof hydraulic cement exceeds 16 gigatons (Gt)

This article is the second of a pair As notedin part I (van Oss and Padovani 2002) hydrau-lic cements are those that can set and hardenunderwater through the hydration of the com-ponent cement minerals By far the most com-mon hydraulic cements in use today are eitherportland cements or similar-use cements (calledldquoblendedrdquo or ldquocompositerdquo cements) that aremade of a portland cement base plus cementi-tious or pozzolanic additives blended cementsare commonly included within the portland ce-ment designation in the economic research andtechnical literature A pozzolan is a siliceousmaterial that develops hydraulic cementitiousproperties when interacted with free lime(CaO) and water

Straight portland cement is made by grindingtogether portland cement clinker (the interme-diate product of cement manufacture) with asmall amount typically 5 by weight of calciumsulfate usually in the form of the mineral gyp-sum Summarizing from part I the chemical

composition of a typical portland cement clinkeris almost entirely just four oxides calcium oxideor lime (CaO) about 65 silica (SiO2) about22 alumina (Al2O3) about 6 and iron oxide(Fe2O3) about 3 In cement industry short-hand these four oxides are written as C S Aand F respectively and most clinkers do notshow deviations in these oxide proportions ofmore than 2 to 4 percentage points The remain-ing 4 or so of the clinker composition is dividedamong oxides of magnesium potassium sodiumsulfur and others Clinker is primarily made upof four clinker minerals denoted in shorthand asC3S C2S C3A and C4AF The C3S and C2S arethe main contributors to the performance ofportland cement and together make up about70 to 80 of the weight of the clinker Duringtheir hydration C3S and C2S combine with wa-ter by similar reaction paths to form calcium sil-icate hydrate (its variable composition is denotedldquoC-S-Hrdquo) plus lime the C-S-H is a colloidal gelthat is the actual binding agent in the concreteThe bulk of the C3S hydrates rapidly (hours todays) and provides most of the early strength ofthe concrete whereas the C2S hydrates slowly(days to weeks) and is responsible for most of theconcretersquos long-term strength The lime by-product of hydration activates any pozzolans thatmay be present in the concrete mix

As was reviewed in part I the manufacture ofclinker involves the thermochemical processingof large quantities of limestone and other rawmaterials typically about 17 tt clinker and re-quires enormous kilns and related equipmentsustained very high kiln temperatures (the ma-terials reach temperatures of about 14508C in or-der to form the key C3S mineral) and the con-sumption of large amounts of energy (fuels andelectricity) total energy consumption is about 3to 6 million British thermal units (Btu)t clinker(1 million Btu 4 1055 GJ) Clinker manufac-ture results in signicant emissions particularlyof carbon dioxide (CO2) Apart from the tech-nological aspects of cement manufacture part Idiscussed the main environmental considera-tions of the mining of cement raw materials Theremaining environmental challenges and oppor-tunities relating to clinker and cement manufac-ture are the subject of this article








































Tab

le1

Fuel

cons

umpt

ion

and

carb

ondi

oxid

eem

issio

nsfo

rth

eU

Sc

emen

tin

dust

ry

Year

Fuel

cons

umpt

ion

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

1998

1999

2000

Coa

l(kM

t)7

206

791

87

591

828

87

227

650

910

601

100

879

098

824

19

066

920

610

095

Cok

e(f

rom

coal

)1(k

Mt)

ndnd

ndnd

ndnd

ndnd

nd45

543

234

344

2Pe

trol

eum

coke

1(k

Mt)

ndnd

ndnd

nd35

748

844

237

91

475

119

71

622

135

1Fu

eloi

l(M

L)83

61

352

641

710

159

41

166

653

120

299

4273

134

124

Nat

ural

gas

(Mm

3 )2

751

372

14

859

562

15

998

451

81

718

301

294

106

972

065

333

8T

ires

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

158

269

685

374

Oth

erso

lidw

aste

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

6874

816

101

6Li

quid

was

te(M

L)nd

ndnd

ndnd

ndnd

nd2

nd2

885

126

890

592

9

Car

bon

diox

ide

emis

sion

s(t

tcl

inke

r)C

ase

Af

uel

only

30

690

630

560

540

530

520

550

474

041

40

460

460

490

48C

ase

At

otal

35

120

114

107

105

104

103

106

098

40

924

097

097

100

099

Cas

eB

fue

lonl

y60

630

570

520

500

480

480

490

434

037

40

410

410

440

43C

ase

Bt

otal

56

114

108

103

101

099

099

100

094

40

884

092

092

095

094

Clin

ker

outp

ut(k

Mt)

386

7951

093

553

4963

991

671

2358

549

633

4160

941

643

5671

257

758

4277

337

796

56

Sour

ce

Tab

le4

ofva

nO

ssan

dPa

dova

ni(2

002)

Not

end

4no

data

lik

ely

smal

lor

nile

xcep

tfo

r19

90

1Fo

rye

ars

labe

led

ldquond

rdquoco

nsum

ptio

nif

any

may

bein

clud

edin

data

for

coal

2

Was

tefu

elda

taw

ere

not

colle

cted

unti

l199

3bu

tth

efu

els

wer

ebe

ing

cons

umed

begi

nnin

gin

the

mid

-190

8s

3C

alcu

late

dba

sed

onst

anda

rdgr

oss

(hig

h)he

atva

lues

for

fuel

sV

alue

sex

ceed

thos

eca

lcul

ated

usin

gne

t(l

ow)

heat

valu

esby

abou

t0

02un

it(1

950ndash

1975

)0

00to

001

unit

(198

0ndash19

90)

and

001

unit

(199

5ndash20

00)

4D

ata

are

unde

rval

ued

beca

use

ofth

ela

ckof

was

tefu

elda

tafo

r19

85an

d19

90s

eefo

otno

te2

5In

clud

esca

lcin

atio

nem

issi

ons

of0

51t

tcl

inke

r6

Cal

cula

ted

base

don

actu

alhe

atva

lues

(gro

sshe

atba

sis)

for

fuel

sre

port

edby

plan

tsto

the

USG

Sin

2000

















































































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors









































Tab

le1

Fuel

cons

umpt

ion

and

carb

ondi

oxid

eem

issio

nsfo

rth

eU

Sc

emen

tin

dust

ry

Year

Fuel

cons

umpt

ion

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

1998

1999

2000

Coa

l(kM

t)7

206

791

87

591

828

87

227

650

910

601

100

879

098

824

19

066

920

610

095

Cok

e(f

rom

coal

)1(k

Mt)

ndnd

ndnd

ndnd

ndnd

nd45

543

234

344

2Pe

trol

eum

coke

1(k

Mt)

ndnd

ndnd

nd35

748

844

237

91

475

119

71

622

135

1Fu

eloi

l(M

L)83

61

352

641

710

159

41

166

653

120

299

4273

134

124

Nat

ural

gas

(Mm

3 )2

751

372

14

859

562

15

998

451

81

718

301

294

106

972

065

333

8T

ires

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

158

269

685

374

Oth

erso

lidw

aste

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

6874

816

101

6Li

quid

was

te(M

L)nd

ndnd

ndnd

ndnd

nd2

nd2

885

126

890

592

9

Car

bon

diox

ide

emis

sion

s(t

tcl

inke

r)C

ase

Af

uel

only

30

690

630

560

540

530

520

550

474

041

40

460

460

490

48C

ase

At

otal

35

120

114

107

105

104

103

106

098

40

924

097

097

100

099

Cas

eB

fue

lonl

y60

630

570

520

500

480

480

490

434

037

40

410

410

440

43C

ase

Bt

otal

56

114

108

103

101

099

099

100

094

40

884

092

092

095

094

Clin

ker

outp

ut(k

Mt)

386

7951

093

553

4963

991

671

2358

549

633

4160

941

643

5671

257

758

4277

337

796

56

Sour

ce

Tab

le4

ofva

nO

ssan

dPa

dova

ni(2

002)

Not

end

4no

data

lik

ely

smal

lor

nile

xcep

tfo

r19

90

1Fo

rye

ars

labe

led

ldquond

rdquoco

nsum

ptio

nif

any

may

bein

clud

edin

data

for

coal

2

Was

tefu

elda

taw

ere

not

colle

cted

unti

l199

3bu

tth

efu

els

wer

ebe

ing

cons

umed

begi

nnin

gin

the

mid

-190

8s

3C

alcu

late

dba

sed

onst

anda

rdgr

oss

(hig

h)he

atva

lues

for

fuel

sV

alue

sex

ceed

thos

eca

lcul

ated

usin

gne

t(l

ow)

heat

valu

esby

abou

t0

02un

it(1

950ndash

1975

)0

00to

001

unit

(198

0ndash19

90)

and

001

unit

(199

5ndash20

00)

4D

ata

are

unde

rval

ued

beca

use

ofth

ela

ckof

was

tefu

elda

tafo

r19

85an

d19

90s

eefo

otno

te2

5In

clud

esca

lcin

atio

nem

issi

ons

of0

51t

tcl

inke

r6

Cal

cula

ted

base

don

actu

alhe

atva

lues

(gro

sshe

atba

sis)

for

fuel

sre

port

edby

plan

tsto

the

USG

Sin

2000

















































































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors

































Tab

le1

Fuel

cons

umpt

ion

and

carb

ondi

oxid

eem

issio

nsfo

rth

eU

Sc

emen

tin

dust

ry

Year

Fuel

cons

umpt

ion

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

1998

1999

2000

Coa

l(kM

t)7

206

791

87

591

828

87

227

650

910

601

100

879

098

824

19

066

920

610

095

Cok

e(f

rom

coal

)1(k

Mt)

ndnd

ndnd

ndnd

ndnd

nd45

543

234

344

2Pe

trol

eum

coke

1(k

Mt)

ndnd

ndnd

nd35

748

844

237

91

475

119

71

622

135

1Fu

eloi

l(M

L)83

61

352

641

710

159

41

166

653

120

299

4273

134

124

Nat

ural

gas

(Mm

3 )2

751

372

14

859

562

15

998

451

81

718

301

294

106

972

065

333

8T

ires

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

158

269

685

374

Oth

erso

lidw

aste

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

6874

816

101

6Li

quid

was

te(M

L)nd

ndnd

ndnd

ndnd

nd2

nd2

885

126

890

592

9

Car

bon

diox

ide

emis

sion

s(t

tcl

inke

r)C

ase

Af

uel

only

30

690

630

560

540

530

520

550

474

041

40

460

460

490

48C

ase

At

otal

35

120

114

107

105

104

103

106

098

40

924

097

097

100

099

Cas

eB

fue

lonl

y60

630

570

520

500

480

480

490

434

037

40

410

410

440

43C

ase

Bt

otal

56

114

108

103

101

099

099

100

094

40

884

092

092

095

094

Clin

ker

outp

ut(k

Mt)

386

7951

093

553

4963

991

671

2358

549

633

4160

941

643

5671

257

758

4277

337

796

56

Sour

ce

Tab

le4

ofva

nO

ssan

dPa

dova

ni(2

002)

Not

end

4no

data

lik

ely

smal

lor

nile

xcep

tfo

r19

90

1Fo

rye

ars

labe

led

ldquond

rdquoco

nsum

ptio

nif

any

may

bein

clud

edin

data

for

coal

2

Was

tefu

elda

taw

ere

not

colle

cted

unti

l199

3bu

tth

efu

els

wer

ebe

ing

cons

umed

begi

nnin

gin

the

mid

-190

8s

3C

alcu

late

dba

sed

onst

anda

rdgr

oss

(hig

h)he

atva

lues

for

fuel

sV

alue

sex

ceed

thos

eca

lcul

ated

usin

gne

t(l

ow)

heat

valu

esby

abou

t0

02un

it(1

950ndash

1975

)0

00to

001

unit

(198

0ndash19

90)

and

001

unit

(199

5ndash20

00)

4D

ata

are

unde

rval

ued

beca

use

ofth

ela

ckof

was

tefu

elda

tafo

r19

85an

d19

90s

eefo

otno

te2

5In

clud

esca

lcin

atio

nem

issi

ons

of0

51t

tcl

inke

r6

Cal

cula

ted

base

don

actu

alhe

atva

lues

(gro

sshe

atba

sis)

for

fuel

sre

port

edby

plan

tsto

the

USG

Sin

2000

















































































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors























Tab

le1

Fuel

cons

umpt

ion

and

carb

ondi

oxid

eem

issio

nsfo

rth

eU

Sc

emen

tin

dust

ry

Year

Fuel

cons

umpt

ion

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

1998

1999

2000

Coa

l(kM

t)7

206

791

87

591

828

87

227

650

910

601

100

879

098

824

19

066

920

610

095

Cok

e(f

rom

coal

)1(k

Mt)

ndnd

ndnd

ndnd

ndnd

nd45

543

234

344

2Pe

trol

eum

coke

1(k

Mt)

ndnd

ndnd

nd35

748

844

237

91

475

119

71

622

135

1Fu

eloi

l(M

L)83

61

352

641

710

159

41

166

653

120

299

4273

134

124

Nat

ural

gas

(Mm

3 )2

751

372

14

859

562

15

998

451

81

718

301

294

106

972

065

333

8T

ires

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

158

269

685

374

Oth

erso

lidw

aste

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

6874

816

101

6Li

quid

was

te(M

L)nd

ndnd

ndnd

ndnd

nd2

nd2

885

126

890

592

9

Car

bon

diox

ide

emis

sion

s(t

tcl

inke

r)C

ase

Af

uel

only

30

690

630

560

540

530

520

550

474

041

40

460

460

490

48C

ase

At

otal

35

120

114

107

105

104

103

106

098

40

924

097

097

100

099

Cas

eB

fue

lonl

y60

630

570

520

500

480

480

490

434

037

40

410

410

440

43C

ase

Bt

otal

56

114

108

103

101

099

099

100

094

40

884

092

092

095

094

Clin

ker

outp

ut(k

Mt)

386

7951

093

553

4963

991

671

2358

549

633

4160

941

643

5671

257

758

4277

337

796

56

Sour

ce

Tab

le4

ofva

nO

ssan

dPa

dova

ni(2

002)

Not

end

4no

data

lik

ely

smal

lor

nile

xcep

tfo

r19

90

1Fo

rye

ars

labe

led

ldquond

rdquoco

nsum

ptio

nif

any

may

bein

clud

edin

data

for

coal

2

Was

tefu

elda

taw

ere

not

colle

cted

unti

l199

3bu

tth

efu

els

wer

ebe

ing

cons

umed

begi

nnin

gin

the

mid

-190

8s

3C

alcu

late

dba

sed

onst

anda

rdgr

oss

(hig

h)he

atva

lues

for

fuel

sV

alue

sex

ceed

thos

eca

lcul

ated

usin

gne

t(l

ow)

heat

valu

esby

abou

t0

02un

it(1

950ndash

1975

)0

00to

001

unit

(198

0ndash19

90)

and

001

unit

(199

5ndash20

00)

4D

ata

are

unde

rval

ued

beca

use

ofth

ela

ckof

was

tefu

elda

tafo

r19

85an

d19

90s

eefo

otno

te2

5In

clud

esca

lcin

atio

nem

issi

ons

of0

51t

tcl

inke

r6

Cal

cula

ted

base

don

actu

alhe

atva

lues

(gro

sshe

atba

sis)

for

fuel

sre

port

edby

plan

tsto

the

USG

Sin

2000

















































































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors














Tab

le1

Fuel

cons

umpt

ion

and

carb

ondi

oxid

eem

issio

nsfo

rth

eU

Sc

emen

tin

dust

ry

Year

Fuel

cons

umpt

ion

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

1998

1999

2000

Coa

l(kM

t)7

206

791

87

591

828

87

227

650

910

601

100

879

098

824

19

066

920

610

095

Cok

e(f

rom

coal

)1(k

Mt)

ndnd

ndnd

ndnd

ndnd

nd45

543

234

344

2Pe

trol

eum

coke

1(k

Mt)

ndnd

ndnd

nd35

748

844

237

91

475

119

71

622

135

1Fu

eloi

l(M

L)83

61

352

641

710

159

41

166

653

120

299

4273

134

124

Nat

ural

gas

(Mm

3 )2

751

372

14

859

562

15

998

451

81

718

301

294

106

972

065

333

8T

ires

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

158

269

685

374

Oth

erso

lidw

aste

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

6874

816

101

6Li

quid

was

te(M

L)nd

ndnd

ndnd

ndnd

nd2

nd2

885

126

890

592

9

Car

bon

diox

ide

emis

sion

s(t

tcl

inke

r)C

ase

Af

uel

only

30

690

630

560

540

530

520

550

474

041

40

460

460

490

48C

ase

At

otal

35

120

114

107

105

104

103

106

098

40

924

097

097

100

099

Cas

eB

fue

lonl

y60

630

570

520

500

480

480

490

434

037

40

410

410

440

43C

ase

Bt

otal

56

114

108

103

101

099

099

100

094

40

884

092

092

095

094

Clin

ker

outp

ut(k

Mt)

386

7951

093

553

4963

991

671

2358

549

633

4160

941

643

5671

257

758

4277

337

796

56

Sour

ce

Tab

le4

ofva

nO

ssan

dPa

dova

ni(2

002)

Not

end

4no

data

lik

ely

smal

lor

nile

xcep

tfo

r19

90

1Fo

rye

ars

labe

led

ldquond

rdquoco

nsum

ptio

nif

any

may

bein

clud

edin

data

for

coal

2

Was

tefu

elda

taw

ere

not

colle

cted

unti

l199

3bu

tth

efu

els

wer

ebe

ing

cons

umed

begi

nnin

gin

the

mid

-190

8s

3C

alcu

late

dba

sed

onst

anda

rdgr

oss

(hig

h)he

atva

lues

for

fuel

sV

alue

sex

ceed

thos

eca

lcul

ated

usin

gne

t(l

ow)

heat

valu

esby

abou

t0

02un

it(1

950ndash

1975

)0

00to

001

unit

(198

0ndash19

90)

and

001

unit

(199

5ndash20

00)

4D

ata

are

unde

rval

ued

beca

use

ofth

ela

ckof

was

tefu

elda

tafo

r19

85an

d19

90s

eefo

otno

te2

5In

clud

esca

lcin

atio

nem

issi

ons

of0

51t

tcl

inke

r6

Cal

cula

ted

base

don

actu

alhe

atva

lues

(gro

sshe

atba

sis)

for

fuel

sre

port

edby

plan

tsto

the

USG

Sin

2000

















































































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors




Tab

le1

Fuel

cons

umpt

ion

and

carb

ondi

oxid

eem

issio

nsfo

rth

eU

Sc

emen

tin

dust

ry

Year

Fuel

cons

umpt

ion

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

1998

1999

2000

Coa

l(kM

t)7

206

791

87

591

828

87

227

650

910

601

100

879

098

824

19

066

920

610

095

Cok

e(f

rom

coal

)1(k

Mt)

ndnd

ndnd

ndnd

ndnd

nd45

543

234

344

2Pe

trol

eum

coke

1(k

Mt)

ndnd

ndnd

nd35

748

844

237

91

475

119

71

622

135

1Fu

eloi

l(M

L)83

61

352

641

710

159

41

166

653

120

299

4273

134

124

Nat

ural

gas

(Mm

3 )2

751

372

14

859

562

15

998

451

81

718

301

294

106

972

065

333

8T

ires

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

158

269

685

374

Oth

erso

lidw

aste

(kM

t)nd

ndnd

ndnd

ndnd

nd2

nd2

6874

816

101

6Li

quid

was

te(M

L)nd

ndnd

ndnd

ndnd

nd2

nd2

885

126

890

592

9

Car

bon

diox

ide

emis

sion

s(t

tcl

inke

r)C

ase

Af

uel

only

30

690

630

560

540

530

520

550

474

041

40

460

460

490

48C

ase

At

otal

35

120

114

107

105

104

103

106

098

40

924

097

097

100

099

Cas

eB

fue

lonl

y60

630

570

520

500

480

480

490

434

037

40

410

410

440

43C

ase

Bt

otal

56

114

108

103

101

099

099

100

094

40

884

092

092

095

094

Clin

ker

outp

ut(k

Mt)

386

7951

093

553

4963

991

671

2358

549

633

4160

941

643

5671

257

758

4277

337

796

56

Sour

ce

Tab

le4

ofva

nO

ssan

dPa

dova

ni(2

002)

Not

end

4no

data

lik

ely

smal

lor

nile

xcep

tfo

r19

90

1Fo

rye

ars

labe

led

ldquond

rdquoco

nsum

ptio

nif

any

may

bein

clud

edin

data

for

coal

2

Was

tefu

elda

taw

ere

not

colle

cted

unti

l199

3bu

tth

efu

els

wer

ebe

ing

cons

umed

begi

nnin

gin

the

mid

-190

8s

3C

alcu

late

dba

sed

onst

anda

rdgr

oss

(hig

h)he

atva

lues

for

fuel

sV

alue

sex

ceed

thos

eca

lcul

ated

usin

gne

t(l

ow)

heat

valu

esby

abou

t0

02un

it(1

950ndash

1975

)0

00to

001

unit

(198

0ndash19

90)

and

001

unit

(199

5ndash20

00)

4D

ata

are

unde

rval

ued

beca

use

ofth

ela

ckof

was

tefu

elda

tafo

r19

85an

d19

90s

eefo

otno

te2

5In

clud

esca

lcin

atio

nem

issi

ons

of0

51t

tcl

inke

r6

Cal

cula

ted

base

don

actu

alhe

atva

lues

(gro

sshe

atba

sis)

for

fuel

sre

port

edby

plan

tsto

the

USG

Sin

2000

















































































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors


















































































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors





































































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors



























































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors
















































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors






































Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors





























Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors
















Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors





Region

Cementproduction

(Mt)Percent

of world production

Emissionspotential

(Mt CO2yr)dagger


Total world 16398 1000 17403


Data are for 2000



























































































































Acknowledgments


Notes


















References




























































































































About the Authors























































































































Acknowledgments


Notes


















References




























































































































About the Authors





































































































Acknowledgments


Notes


















References




























































































































About the Authors




























































































Acknowledgments


Notes


















References




























































































































About the Authors


















































































Acknowledgments


Notes


















References




























































































































About the Authors









































































Acknowledgments


Notes


















References




























































































































About the Authors































































Acknowledgments


Notes


















References




























































































































About the Authors























































Acknowledgments


Notes


















References




























































































































About the Authors












































Acknowledgments


Notes


















References




























































































































About the Authors

































Acknowledgments


Notes


















References




























































































































About the Authors
























Acknowledgments


Notes


















References




























































































































About the Authors















Acknowledgments


Notes


















References




























































































































About the Authors






Acknowledgments


Notes


















References




























































































































About the Authors












References




























































































































About the Authors



















































































































About the Authors


















































































About the Authors
















































About the Authors





























About the Authors


















About the Authors










About the Authors


Date post:	27-Mar-2018
Category:	Documents
Upload:	truongque
View:	215 times
Download:	1 times

STATE OF THE DEBATE Cement Manufacture and the...

Documents