lable at ScienceDirect

Journal of Environmental Management 91 (2010) 1864e1871

Contents lists avai

Journal of Environmental Management

journal homepage: www.elsevier .com/locate/ jenvman

Cement replacement by sugar cane bagasse ash: CO2 emissions reductionand potential for carbon credits

Eduardo M.R. Fairbairn a,*, Branca B. Americano b, Guilherme C. Cordeiro c, Thiago P. Paula a,Romildo D. Toledo Filho a, Marcos M. Silvoso a

aDepartment of Civil Engineering, COPPE/Universidade Federal do Rio Janeiro, 21941-972 Rio de Janeiro, RJ, BrazilbMinistry of the Environment e Federal Government, Brasília, DF, Brazilc Laboratory of Civil Engineering, Universidade Estadual do Norte Fluminense Darcy Ribeiro, Campos dos Goytacazes, RJ, Brazil

a r t i c l e i n f o

Article history:Received 21 August 2009Received in revised form23 March 2010Accepted 19 April 2010Available online 20 May 2010

Keywords:Sugar cane bagasse ashCarbon creditsCO2 emissionsConcreteCement

* Corresponding author.E-mail address: eduardo@coc.ufrj.br (E.M.R. Fairba

0301-4797/$ e see front matter � 2010 Elsevier Ltd.doi:10.1016/j.jenvman.2010.04.008

a b s t r a c t

This paper presents a study of cement replacement by sugar cane bagasse ash (SCBA) in industrial scaleaiming to reduce the CO2 emissions into the atmosphere. SCBA is a by-product of the sugar/ethanol agro-industry abundantly available in some regions of the world and has cementitious properties indicatingthat it can be used together with cement. Recent comprehensive research developed at the FederalUniversity of Rio de Janeiro/Brazil has demonstrated that SCBA maintains, or even improves, themechanical and durability properties of cement-based materials such as mortars and concretes. Brazil isthe world’s largest sugar cane producer and being a developing country can claim carbon credits. Asimulation was carried out to estimate the potential of CO2 emission reductions and the viability to issuecertified emission reduction (CER) credits. The simulation was developed within the framework of themethodology established by the United Nations Framework Convention on Climate Change (UNFCCC) forthe Clean Development Mechanism (CDM). The State of São Paulo (Brazil) was chosen for this case studybecause it concentrates about 60% of the national sugar cane and ash production together with animportant concentration of cement factories. Since one of the key variables to estimate the CO2 emissionsis the average distance between sugar cane/ethanol factories and the cement plants, a genetic algorithmwas developed to solve this optimization problem. The results indicated that SCBA blended cementreduces CO2 emissions, which qualifies this product for CDM projects.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Concrete consists of cement, aggregates, water, and eventually,mineral and chemical admixtures. When all these materials aremixed, cement particles upon contact with water undergo a hard-ening reaction that bonds the aggregates together. Concrete is theworld’s most consumed construction material because it combinesgood mechanical and durability properties, pleaceability, work-ability and it is relatively inexpensive. However, cement productioninvolves significant CO2 emissions, which is known as the green-house gas mostly important for the global warming. Each tonne ofcement produces approximately one tonne of CO2 (Malhotra, 2002;Hewlett, 2005) and the cement industry is responsible for about 5%of global anthropogenic CO2 emissions (Worrell et al., 2001).

Cement is manufactured in more than 80 countries and themost commonly used cement is Ordinary Portland Cement

irn).

All rights reserved.

(referenced as OPC). For its production, limestone mixed with claysand small quantities of other materials needs to be heated up to1450 �C. As a result of this process, clinker (about 95% in mass) isobtained. The clinker is then ground and mixed to gypsum (about5% in mass).

In the cement production greenhouse gas (GHG) emissionscome from both industrial process and fuel combustion. During theindustrial process CO2 is emitted due to the heating of limestone(CaCO3 / CaO þ CO2) to obtain calcium oxide (CaO), which is themain oxide in the OPC. To reduce emissions, several recommen-dations can be followed in the cement industry. Concerning theindustrial process an alternative is the replacement of clinker bymineral additions that can also act as cementitious materials suchas blast furnace slag and pozzolans (Malhotra and Mehta, 1996).

The main pozzolans currently used in cement industry are flyash, a by-product of coal-fired power plants, and silica fume, a by-product of metallurgical processes. Other pozzolans have also beenused in a reduced scale, such as natural pozzolans, metakaolin, andagro-industrial ashes such as rice husk and sugar cane bagasseashes.

40

50

60

70

80

0 50 100 150 200

Curing time (days)

Com

pres

sive

stre

ngth

(MPa

)

Reference concrete10% SCBA concrete

15% SCBA concrete20% SCBA concrete

Fig. 1. Evolution of compressive strength of reference and SCBA concretes (Cordeiroet al., 2008a).

E.M.R. Fairbairn et al. / Journal of Environmental Management 91 (2010) 1864e1871 1865

This paper presents a case study that simulates the industrialuse of sugar cane bagasse ash (SCBA) as a cement mineral additionin the Southeastern region of Brazil, estimating the potential forgranting carbon credits. Brazil is the world’s largest sugar caneproducer (515million tonnes in 2007), which corresponds to 33% ofthe world’s production (FAO, 2008). Of this total, 45% is destined toproduce sugar, seeds and animal food, whereas the other 55% isintended to ethanol production, being 92% of the produced ethanolused as vehicle fuel. Today, 90% of new cars sold in the country areflex-fuel and represent around 25% of the fleet. It is important tohighlight that even the gasoline sold in Brazil is a blend of 76%gasoline and 24% of anhydrous ethanol. The increased awarenessabout climate changes and the recent strong oscillations of oilprices indicate that ethanol plays a new strategic role in terms ofenvironmental and energy safety, not only in Brazil but also all overthe world.

Bagasse is an important by-product of the sugar cane industryand most of it is burned to produce steam and electricity in a co-generation plant at the ethanol factory. The SCBA is the result fromthe bagasse combustion and consists mainly of silica (SiO2), whichindicates its potential as mineral admixture. Brazilian SCBApotential amounts to more than 2.5 million tonnes per year, cor-responding to about 6% of the cement production in the country.

Carbon credits, implemented through a mechanism called CleanDevelopment Mechanism (CDM) is a great incentive for the imple-mentation of projects that reduces CO2 emissions in developingcountries (Americano, 2008). The CDM, which has been establishedby the Kyoto Protocol, allows emission-reduction or emission-removal to earn Certified Emission Reduction (CER) credits, eachequivalent to one tonne of CO2. These CERs can be traded and sold,and used by industrialized countries to meet a part of their emis-sion reduction targets under the Kyoto Protocol. The mechanismstimulates sustainable development and emission reductions,while giving industrialized countries some flexibility in how theymeet their emission reduction limitation targets. Hence, Brazilianprojects involving clinker replacement by ultrafine SCBAmay claimthe benefits of the CDM, if it brings real, measurable and longtermadvantages of CO2 emissions reductions.

The simulation presented in this paper considers the technicalviability for using SCBA as partial replacement of clinker, as well asthe conditions to implement CDM projects within the methodol-ogies established by the Kyoto Protocol. It is shown that SCBA canbe a suitable choice for reducing cement related CO2 emissions andthat its industrial implementation is entitled for the emission ofCER credits.

2. Sugar cane bagasse ash: a new cement blending material

The use of SCBA as a pozzolan is possible due to the presence ofreactive silica in the ash. Since the late 90s studies have beencarried out showing that SCBA presents adequate behavior inblended-based cementitious materials (Martirena Hernández et al.,1998; Singh et al., 2000; Morales et al., 2009). Moreover, Ganesanet al. (2007) showed that the addition of SCBA as partial cementreplacement (5e30%, in mass) could increase the mechanical anddurability properties of concrete.

Recently, a comprehensive research has been carried outconsidering several thermo-chemo-mechanical aspects of the SCBAproduced in Brazil (Cordeiro, 2006; Cordeiro et al., 2008a,b, 2009a,b). In these works, the ashes were collected during the boilercleaning operation of a sugar cane/ethanol plant in the South-eastern region. The SCBA was submitted to grinding to improve itshomogeneity and reactivity (Cordeiro et al., 2009a). After grinding,different concretes were mix designed with a cement replacementlevel ranging from 0 to 20%, and their performance was studied

based on rheological, mechanical, durability and calorimetric tests(Cordeiro et al., 2008a). The results indicated that, in general, theSCBA-based concretes presented better rheological behavior (i.e.,workability) than the reference concrete.

The mechanical behavior synthesized in Fig. 1 does not showa significant difference after 180 days of curing.

However, the concrete containing SCBA presented best perfor-mance in rapid-chloride ion permeability tests (Ganesan et al.,2007; Cordeiro et al., 2008a) according to ASTM 1202 standard(1997). The investigation has also shown that the maximum adia-batic temperature rise of concrete was decreased by about 11%.Other possibilities of mixing SCBA to cement have been studied. Inthis way, the thermo-chemo-mechanical simulation of a damconstruction (Cordeiro et al., 2007) indicated that a concrete madewith rice husk ash/SCBA blended cement performs better thana reference concrete. Therefore, it can be concluded that sugar canebagasse ash is a pozzolan which, when mixed to cement, improvesthe performance of the material.

3. Methodology for CDM implementation

3.1. CDM basic concepts

The CDM (UNFCCC, 2009a) allows that an emission-reductionproject in developing countries (named within the Kyoto Protocolas non-Annex I) generates CER credits, which can be counted byindustrialized countries towards meeting Kyoto targets in gener-ating a carbon market. However, it is important to avoid the situ-ation where carbon credits are given to projects that would occuranyway. In this case, emissions in developing countries will notreduce beyond “the business as usual” and industrialized countrieswill continue to emit compromising the Protocol environmentalintegrity.

The CDM projects must be certified and must qualify througha rigorous and public registration and issuance process. “Theconcepts of applicability, additionality and baseline are fundamentalto guarantee the environmental integrity of the Protocol. Regardingprojects related to blending of mineral additions to cement, theseconcepts could be explained as follows” (UNFCCC, 2005, 2009b).

The applicability is limited to “projects that increase the share ofadditives (i.e. reduce the share of clinker) in the production ofcement types beyond current practices in the country”. It should beinsured, among other restrictions, that: (i) “there is no shortage of

E.M.R. Fairbairn et al. / Journal of Environmental Management 91 (2010) 1864e18711866

additives related to the lack of blending materials”; and that (ii)“the methodology is applicable to domestically sold output of theproject activity plant and excludes export of blended cement”.

Additionality is defined as follows: “A CDM project activity isadditional if anthropogenic emissions of greenhouse gases bysources are reduced below those that would have occurred in theabsence of the registered CDM project activity”. “The applicants toCDM projects must demonstrate that there exist real and demon-strable barriers to the increase in the additive blend. Such barriersmay include, among others: (i) a substantial research effort that isrequired to enable the increase in blending; (ii) lack of infrastruc-ture for implementation of the technology; (iii) lack of access tofinancing; (iv) perception by the market that high additive blendedcement is of inferior quality; (v) lack of awareness of customers onthe use high additive blended cement”.

The baseline for a CDM project activity is the scenario thatreasonably represents the anthropogenic emissions by sources ofgreenhouse gases that would occur in the absence of the proposedproject activity.

A CDM Project Activity (CPA) is by definition established ina project-by-project basis. This means that the applicability, addi-tionality and baseline are established for each and every project.Programmatic CDM permits a replication of similar units (i.e.,similar CPAs), but in general the foundation is the same.

CDM is a project-based mechanism, which means that baselinesand project scenarios are to be established for one and each case.However, to make an assessment of the potential for CDM projectsconcerning the implementation of SCBA in a given region, generalaveraged emission factors have been estimated within the frame-work of the present study.

Themethodologypresented in the followingparagraphs is theoneapproved and consolidated by UNFCCC (2009b). It follows UNFCCguidelines and is based on results published in research papers.Furthermore, the data suggested for baseline determination is fromauthentic sources and readily verifiable. Justificationanddetails of themethodology can be found elsewhere (see, for instance, new meth-odologies, UNFCCC, 2004). Within the framework of this method-ologythebasicnomenclatureof thevariables isestablishedas follows:thefirst letters,BandP, stand, respectively, forBaselineandProject; thesecond letter, E, stands for Emissions; BC is the abbreviation of BlendedCement and ER of Emissions Reduction; and Lmeans Leakage.

The equations corresponding to emissions reduction, baselineand project emissions, and leakage are detailed in the followingparagraphs.

3.2. Emission reductions

The emission reductions provided by the use of the blendedcement are calculated by equation (1).

ERy ¼ ��BEBC;y � PEBC;y

�$BCy þ Ly

�$�1� ay

�(1)

where:

ERy e CO2 emissions reductions in the year (kilotonnes);BEBC,y e Baseline CO2 emissions in the year y (tonnes CO2/tonnecement);PEBC,y e Project CO2 emissions in the year y (tonnes CO2/tonnecement);BCy e Total cement production (kilotonnes);Ly e Leakage emissions for transport of additives (kilotonnes ofCO2);ay e Additional leakage coming from the diversion of mineraladmixtures from existing uses. The project proponents shalldemonstrate that additional amounts of additives used are

surplus. If the project proponents do not substantiate that xtonnes of additives used in the project activity are surplus, theproject emissions reductions are reduced by the factor a definedas x tonnes of mineral admixtures in year y divided by the totaladditional additives used in year y.

3.3. Baseline emissions

The baseline emissions for the cement production are expressedby equation (2):

BEBC;y ¼�BEclinker$BBlend;y

�þ BEele ADD BC (2)

where:

BEBC,y emass of CO2 emitted per mass of blended cement in theyear y (tonnes of CO2/tonne of blended cement);BEclinker e mass of CO2 emitted per mass of clinker (tonnes ofCO2/tonne of clinker);BBlend,y e baseline benchmark of share of clinker per tonne ofblended cement (tonnes of clinker/tonne of blended cement);BEele_ADD_BC e baseline emissions due to the use of electricity ineventual mineral admixtures used in the cement (tonnes of CO2/tonne of blended cement).

The terms of equation (2) are expressed in equations (3) to (7).

BEclinker ¼ BEcalcin þ BEfossil fuel þ BEele grid CLNK þ BEele sg CLNK

(3)

where:

BEcalcin e emissions due to the process of calcinations of calciumcarbonate and magnesium carbonate (tonnes of CO2/tonne ofclinker);BEfossil_fuel e emissions due to the combustion of fossil fuelsduring the clinker production (tonnes of CO2/tonne of clinker);BEele_grid_CLNK e emissions due to electricity for grinding (tonnesof CO2/tonne of clinker).BEele_sg_CLNK e emissions from self-generated electricity forclinker production (tonnes of CO2/tonne of clinker)

BEcalcin ¼ ½C1ðOutCaO� InCaOÞ þ C2ðOutMgO� InMgOÞ�ðCLNKBSL$C3Þ

(4)where:

C1 ¼ 0.785 e stoichiometric emission factor for CaO (tonnes ofCO2/tonne of CaO)C2 ¼ 1.092 e stoichiometric emission factor for MgO (tonnes ofCO2/tonne of MgO)OutCaO e fraction of CaO in the produced clinker (tonnes);InCaO e fraction of CaO in the raw material (tonnes);OutMgO e fraction of MgO in the produced clinker (tonnes);InMgO e fraction of MgO in the raw material (tonnes);CLNKBSL e total production of clinker (kilotonnes of clinker).C3 ¼ 1000 e constant (tonnes/kilotonne)

BEfossil fuel ¼ðP FFi BSL$EFFiÞðCLNKBSL$C3Þ

(5)

where:

FFi_BSLemass of the different fossil fuels used for the productionof clinker (tonnes of fuel i);

E.M.R. Fairbairn et al. / Journal of Environmental Management 91 (2010) 1864e1871 1867

EFFie emissions factor for fossil fuel of the type i. (tonnes of CO2/tonne of fuel) (given by UNFCCC, 2009c)

�BELEgrid CLNK$EFgrid BSL

�

BEele grid CLNK ¼ ðCLNKBSL$C3Þ

(6)

where:

BELEgrid_CLINK e baseline grid electricity for clinker production(MWh);EFgrid_BSL e baseline emissions factor for the electricity used inthe grinding (tonnes of CO2/MWh). (given by UNFCCC, 2009d)

BEele sg CLNK ¼�BELEsg CLNK$EFsg BSL

�ðCLNKBSL$C3Þ

(7)

where:

BELEsg_CLNK e amount of self-generated electricity (MWh);EFsg_BSL e emissions factor for the self-generated electricity(tonnes of CO2/MWh) e (given by UNFCCC, 2009c).

3.4. Project activity emissions

The UNFCCC methodology for increasing the blend in cementproduction (UNFCCC, 2009b) states that the boundary of the CDMproject includes the cement production plant, any onsite powergeneration and the power generation in the grid. Besides, themethodology defines admixtures as materials blended with clinkerto produce blended cement types and include gypsum, fly ash, slag,etc. For such materials the only GHG emissions associated arerelated to grinding, preparation, and eventual use of fossil fuels forthe transportation of such admixtures, defined by themethodologyas leakage (see Section 3.5).

The project emissions are given by equation (8).

PEBC;y ¼�PEclinker;y$Pblend;y

�þ PEele ADD BC;y (8)

where:

PEBC,y e CO2 emissions per tonne of the SCBA in the projectactivity (tonnes of CO2/tonne of blended cement);PEclinker,y e CO2 emissions per tonne of clinker (tonnes of CO2/tonne of clinker);Pblend,y e share of clinker per tonne of blended cement (tonnesof clinker/tonne of blended cement);PEele_ADD_BC,y e electricity emissions for grinding and prepara-tion of the mineral admixture (tonnes of CO2/tonne of blendedcement).

3.5. Leakage

The leakage emissions are given by the following equation:

Ly ¼ Ladd trans

�Ablend;y � Pblend;y

�BCy (9)

where:

Ablend,y ¼ Baseline benchmark share of additives per tonne ofblended cement updated for year y (tonnes of additives/tonne ofblended cement);

Pblend,y ¼ Share of additives per tonne of blended cement in yeary (tonnes of additives/tonne of blended cement);BCy ¼ Production of blended cement in year y (kilotonnes ofblended cement).

The term Ladd_trans that corresponds to the transportation of themineral additions, treated in the methodology as leakages, iscalculated by the equation (10).

Ladd trans ¼ ðTFcons$Dadd source$TEF$C4ÞQadd

þ�ELEconveyor ADD$EFgrid

�ADDy

(10)

where

Ladd_trans e factor of emissions of the SCBA due to the transport(tonnes of CO2/tonne of additive);TFcons e consumption of fuel oil of the vehicle used in thetransport (kg of fuel oil/km);Dadd_source e distance of transportation (km);TEF e emission factor of the fuel oil (kg of CO2/kg of fuel oil);C4 ¼ 0.001 e constant (tonnes/kg)Qadd e quantity of additive carried per trip in one vehicle(tonnes of additive);ELEconveyor_ADD e Annual electricity consumption for conveyorsystem for additives (MWh);EFgrid e grid electricity emission factor (tonnes of CO2/MWh);ADDy e annual production of SCBA in year y (tonnes of additive).

4. Case study

4.1. General considerations

The State of São Paulo (Brazil) together with bordering regionsin the States of Minas Gerais, Paraná and Rio de Janeiro was chosenfor this case study (see Fig. 2a).

In the State of São Paulo, with an area of 248,000 km2, about 60%of the national sugar cane and ash production are concentrated in201 agro-industrial units, which produced in 2005 over 222milliontonnes of cane. Considering that one tonne of sugar cane supplies6.6 kg of ash (Cordeiro et al., 2008a), the production potential ofSCBA can be estimated by approximately 1.5 million tonnes for thestudied region.

The State mentioned above produced about 8.5 million tonnesof cement in 2005, considering 13 cement plants, including theones which are no further than 100 km from São Paulo’sboundaries.

The capacity of the cement industry to receive and incorporateSCBA as mineral admixture was estimated by establishing twoaverage scenarios for the cement production. Scenario #1 is thebaseline and corresponds to cement production of the delimitedregion in recent years, considering 2005 statistical data. Thescenario #2 corresponds to the project activities for which it isassumed that SCBA is used as mineral addition.

In scenario #1, for the region delimited by the present study, theaverage Brazilian baseline benchmark of clinker inclusion wasconsidered to be BBlend,y ¼ 0.80 tonnes of clinker/tonne of cement(OECD/IEA, 2000). This ratio determines the composition of anaverage hypothetical cement, cement “type A” with the followingcontents (by mass): 80% clinker, 5% gypsum and 15% mineraladditions (pozzolans, slags and fillers) denoted here as madd1.Considering that the total cement production of the region is 6.8million tonnes of clinker, 1.275 million tonnes of madd1 and 0.425

Fig. 2. Project area: (a) State of São Paulo. (b) Localization of the sugar cane/ethanolplants (red circles) and cement factories (blue circles). (c) Result of the optimizationprocedure used to determine the average distance. (For interpretation of the referencesto colour in this figure legend, the reader is referred to the web version of this article.).

Table 1Baseline.

Term Average value Unit

OutCaO 4.3 106 tonnesInCaO 252 103 tonnesOutMgO 263 103 tonnesInMgO 14.3 103 tonnesFFcoke_BSL 604 103 tonnesFFfueloil_BSL 5.3 103 tonnesFFcoal_BSL 2.7 103 tonnesEFFcoke 3.5066 tonnes CO2/tonne cokeEFFfuel oil 3.0753 tonnes CO2/tonne fuel oilEFFcoal 2.713 tonnes CO2/tonne coalBELEgrid_CLNK 1925 MWhEFgrid_BSL 0.2767 tonnes CO2/MWh

BEcalcin 0.512 tonnes CO2/tonne clinkerBEfossil_fuel 0.315 tonnes CO2/tonne clinkerBEele_grid_CLNK y0 tonnes CO2/tonne clinkerBEele_sg_CLNK y0 tonnes CO2/tonne clinkerBEclinker 0.827 tonnes CO2/tonne clinker

E.M.R. Fairbairn et al. / Journal of Environmental Management 91 (2010) 1864e18711868

million tonnes of gypsum are needed for the cement “type A”production.

The recommendations of the Brazilian standards (ABNT, 1991)are observed. These standards allow the production of cementswith mineral addition contents varying from: 6% to 44% (compositeslag cement type II-E); 6% to 24% (composite pozzolanic cementtype II-Z); 35% to 75% (slag cement type III); 15% to 55% (pozzolaniccement type IV). Within scenario #2, following the principles ofCDM implementation, it is considered that the cement industry willmaintain the same consumption (in mass) of madd1 as in scenario#1. Therefore, although up to 75% of additives are allowed in thecement production it was decided to use, within a conservativehypothesis, for the present simulation, a hypothetical cement “typeB” which consists of 30% mineral addition madd1, 65% clinker and5% gypsum. Hence, cement “type B” would be produced using allthe 1.275 million tonnes of madd1 that are used in scenario #1.Therefore, the total production of cement “type B” would be 4.25million tonnes, which corresponds to half of the total cementproduction. Since in scenario #2, the total cement production of 8.5million tonnes shall be maintained, the percentage of madd1 in thetotal cement production would be 15%.

With regards to the other half of the total production (4.25million tonnes), the production of a hypothetical cement “type C” isconsidered, containing SCBA in the mixture. Although percentagesof cement replacement vary up to 20% (in mass) of SCBA it has beenproven to be technically viable (see Section 2) to use an optimalcement replacement of 15%. In order to maintain the same quantityof gypsum the two scenarios for the cement “type C” (to simplify)the contents presented in Section 2 are rounded up to 15% SCBA,80% clinker and 5% gypsum. Therefore, within the hypothesis ofscenario #2, the cement industry will consume 0.6375 milliontonnes of SCBA, and the percentage of SCBA in the total cementproduction will be of 7.5%.

Blending of cement with SCBA fulfills the requirements ofapplicability and additionality of CDM projects. As established byscenarios #1 and #2 there is no shortage of additives since the samequantity of mineral additions madd1 is used in both scenarios.Furthermore, the cement industry in Brazil aims at its internalmarket. In what concerns additionality, since SCBA is a newmaterial related to new technologies and industrial processes, itwould be easy to demonstrate that there exists technological andmarket barriers for its implementation.

4.2. Baseline emissions

The above described scenario #1 establishes the main charac-teristics for computing the emissions corresponding to the base-line. The term BEele_ADD_BC of equation (2) corresponds to themineral additions madd1 and to gypsum, i.e.:

BEele ADD BC ¼ BEele ADD BCmadd1 þ BEele ADD BCgypsum (11)

Since the baseline and the project (i.e., scenarios #1 and #2respectively) use the same quantity of these two additions (seeSection 4.1), the terms BEele_ADD_BCmadd1 and BEele_ADD_BCgypsum willbe subtracted by equivalent terms for the project (see equations (1),(13) and (14)), and it is only relevant to calculate the clinkerbaseline emissions, BEclinker, which are expressed by equation (3).

The results for the clinker baseline are presented in Table 1. Theaverage values were calculated according to the data supplied bythe plants investigated in this work and according to the method-ological tools provided by UNFCCC for calculating the emissionfactors (UNFCCC, 2009c,d).

It is important to note that in Brazil the emissions calculated byequations (6) and (7) are extremely low due to the fact that elec-tricity is predominantly generated based from renewable andnoncarbon intensive sources.

Since the term BBlend,y of equation (2) is estimated as 0.80 tonnesclinker/tonne cement (OECD/IEA, 2000) for the average Braziliancement production, the baseline emissions (tonnes of CO2/tonne ofclinker) can be written as:

BEBC;y ¼ 0:662þ BEele ADD BCmadd1 þ BEele ADD BCgypsum (12)

4.3. Project activity emissions

The mineral addition used to reduce the percentage of clinker inthe total production is SCBA, which is a residue and presents zero

E.M.R. Fairbairn et al. / Journal of Environmental Management 91 (2010) 1864e1871 1869

CO2 emissions, once it is originated from vegetable organic matter,being the carbon of the ash inside the carbon cycle. Therefore, SCBAcan be equated to other additives (fly ash, slag, etc.) described in theUNFCCC methodology (UNFCCC 2009b) and equation (8) is appliedto the present project activity.

The term PEele_ADD_BC,y of equation (8) corresponds to themineral additions, madd1, gypsum and SCBA as indicated byequation (13).

PEele ADD BC ¼ PEele ADD BCmadd1 þ PEele ADD BCgypsum

þ PEele ADD BC SCBA (13)

As stated in Sections 4.1 and 4.2, the baseline and the project (i.e.,scenarios #1 and #2) use the same quantities of gypsum and of theadditions madd1. Therefore, it can be written that:�BEele ADD BCmadd1 þ BEele ADD BCgypsum

���PEele ADD BCmadd1

þ PEele ADD BCgypsum

�¼ 0 ð14Þ

Hence, it is only relevant to compute the last term of equation (13).Similarly to the calculation of the baseline emissions (equations (6)and (7)), it can be considered that, since emissions related toelectricity generation are extremely low in Brazil, the term PEel-e_ADD_BC SCBA y 0.

Since the amount of clinker used in scenarios #1 and #2 is thesame, themass of CO2 emitted permass of clinker is the same in thebaseline and in the project, i.e., PEclinker,y ¼ BEclinker,y.

Otherwise, the clinker to cement ratio of the total production forthe hypothetical cements “types B” and “C” within scenario #2 canbe calculated by equation (15):

PBlend;y ¼�0:65$4:25$106 tonþ 0:80$4:25$106 ton

�8:5$106 ton

¼ 0:725

(15)

Finally, the project activity emissions (tonnes of CO2 per tonne ofclinker) can be written as:

PEBC;y ¼ ð0:827$0:725Þ þ PEele ADD BCmadd1

þ PEele ADD BCgypsum

¼ 0:600þ BEele ADD BCmadd1 þ BEele ADD BCgypsum (16)

4.4. Leakage emissions

The first term of equation (10) stands for the emissions relatedto the transportation of SCBA from the sugar/ethanol plants to thecement plants, corresponding to the main emissions related to theuse of SCBA. The second term concerning the utilized energy in thetransport is negligible for the same reason explained in Section 4.2regarding the low emissions calculated by equations (6) and (7).

To estimate the values of TFcons, TEF and Qadd, the most viableway of transport of SCBA chosen for the study, and in accordancewith the characteristics of the project’s area, was the road transportby trucks. For the present simulation a standard diesel fuel oil truck(TEF ¼ 3.1 tonnes of CO2/tonne of fuel oil, UNFCCC, 2009c), mostcommonly used in Brazil was chosen. According to ANFAVEA(2008) and manufacturer data, this truck has a fuel consumption(TFcons) equal to 0.4 kg fuel oil/km and can transport (Qadd) 16.46tonnes of SCBA per trip.

The parameter Dadd_source was assumed to be the averagedistance between producers of SCBA and cement plants. Since this

parameter is important to estimate the emissions related to theimplementation of SCBA, the procedure used in its determination isshown in the next paragraphs.

To simplify, the 201 sugar/ethanol plants were grouped into 35municipalities denominated spmk (K¼ 1, nspm; nspm¼ 35), and thecement plants (13 units) were grouped into 8 municipalitiesdenominated cpml (l¼ 1, ncpm; ncpm¼ 8), as can be seen in Fig. 2b.In both sugar/ethanol and cement plants groups, the units weregrouped within a 30 km radius circumference. Considering thissimplification, there are 280 possible paths between cement plantsand sugar/ethanol plants.

Therefore, the calculation of the average distance is an optimi-zation problem that can be stated as follows:

� Find the minimum average distance between sugar/ethanolplants and cement plants in such a way that the cementindustry receives as much SCBA as possible.

A genetic algorithm (GA) was used to solve this problem as it hasdemonstrated great ability to provide effective solutions to discreteoptimization problems, allowing a robust search for the globaloptimal point (Fairbairn et al., 2004).

To simplify the formulation of the GA, the SCBA mass unit wasestablished to be 104 tonne. Hence, the SCBA production of a givenmunicipality spmk is given by a number MS(spmk) (104 tonne),where MS(spmk) is rounded up to zero decimal places.

Each municipality was subdivided into nd(spmk) ¼ MS(spmk)/104 virtual districts, denominated spdi, of which each one produces104 tonnes of SCBA, and are located in the same geographicallocation of the original municipality. The total number of districtproducers of SCBA is then:

nspd ¼Xnspmk¼1

ndðspmkÞ (17)

The mass of SCBA that a cement producer municipality is able toreceive, MCSlim(cpml), is also a number expressed in (104 tonne),rounded up to zero decimal places. It is based on the hypothesisestablished by the scenario #2 and is given by equation (18).

MCSlim ¼ ð0:15Þ$ð0:50Þ$MCðcpmlÞ (18)

where MC(cpml) is the cement production of the municipality.A “dummy” municipality cpmncpmþ1 having distance zero from

all the spdi was created to receive the excess of SCBA within theframework of the GA.

The SCBA producer districts and the cement producer munici-palities are organized, respectively, within the discrete sets:spd˛spd1; spd2;.spdnspd and cpm˛fcpm1; cpm2;.cpmncpmþ1g. Let~S be a tentative solution set composed of nspd pairs (spdi, cpmj(i)),where i ¼ 1, 2,., nspd and jðiÞ˛f1;2;.;ncpmþ 1g is randomilychosen. LetMCS~SðcpmlÞsMCSlimðcpmlÞ be the quantity of SCBA thatis received by cpmi within the set ~S.

The GA minimizes the following fitness function:

F ¼ 1nspd

Xnspdi¼1

D�spdi; cpmjðiÞ

�þXncpml¼1

PðcpmlÞ (19)

where D(spdi, cpmj(i)) is the road distance (km) between spdi andcpmj(i), and P is a function that penalizes F ifMCS~SðcpmlÞsMCSlimðcpmlÞ. In this way, for the optimal solution setS we have:

F ¼ Dav;min (20)

E.M.R. Fairbairn et al. / Journal of Environmental Management 91 (2010) 1864e18711870

where Dav,min is the minimum average distance between theproducers of SCBA and the cement plants, assuming that the plantsreceive a quantity of SCBA compatible with their cement produc-tion within the hypothesis of scenario #2. Therefore, besides theminimum of the average distance, the solution set S has also thefollowing properties:

MCSSðcpmlÞ ¼ MCSlimðcpmlÞ; l ¼ 1; ncpm (21)

and

MCSS�cpmncpmþ1

� ¼ Xnspm

k¼1

MSðspmkÞ �Xncpml¼1

MCSlimðcpmlÞ!

(22)

For the problem at hand, the GA converged for Dav,min ¼ 153 km,which is assimilated to the parameter Dadd_source of equation (10).

As all the outstanding parameters are available in the equation(10), it is possible to calculate the transport related emissions pertonne of additives:

Ladd trans ¼ 0:012 tonnes of CO2

tonne of SCBA(23)

Therefore, the leakage emissions Ly can be computed using equa-tion (9), knowing that (Ablend,y�Bblend,y)BCy is a negative numberequal to the mass of SCBA introduced in scenario #2, i.e., �637.5kilotonnes of SCBA:

Ly ¼ �0:012$637:5 ¼ �7:7 kilotonnes of CO2 (24)

4.5. Emissions reduction

The SCBA potential in the area of the project is estimated to be1.5 million tonnes, which is surplus compared to the maximumquantity that could be absorbed by the factories, estimated to be0.635 million tonnes, considering the average production ofscenario #2. Therefore, the term a can be considered negligible.

The emissions reduction can finally be calculated by applyingequation (1):

ERy ¼h�

0:662þ BEele ADD BCmadd1 þ BEele ADD BCgypsum

���0:600þ BEele ADD BCmadd1 þ BEele ADD BCgypsum

�i8500

� 7:7 ¼ 519:3 kilotonnes of CO2 ð25Þ

5. Discussion

For the case study analyzed in this paper, the total production ofSCBA is 1.5 million tonnes but only 0.6375million tonnes have beenused for increasing the blend in cement production. Hence,a surplus of 0.8625 million tonnes of SCBA would exist. Severalscenarios can be imagined for using this surplus, such as: (i) anincreasing of cement production in Brazil in the next few years; (ii)sending the ashes to more distant cement factories consideringeconomical and emission reduction restrictions; (iii) increasing thequantity of the additives madd1 beyond the 30% of the presentsimulation; (iv) increasing the quantity of SCBA in cement beyondthe 15% used in the present simulation, by developing new tech-nologies for the preparation and grinding.

The authors deliberately avoided the estimation of costs for theindustrial implementation of SCBA because it was out of the scope

of the present paper which considers only the balance of theaveraged CO2 emissions. An economical analysis should considerlocal costs and market variables that are beyond the purpose of thepresent study. However, it must be pointed out that the maincomponents of the cost for the implementation of SCBA shouldinclude the costs of the ash, transportation, preparation andgrinding, and the revenues obtained with the carbon credits. SinceSCBA is substituting clinker it can be estimated, in a very firstapproach, that the costs for preparation and grinding are similar tothat of the material to be substituted. The cost of the ash, in a veryfirst moment, is zero or even negative because it is a residue andshould be disposed by the sugar cane/ethanol plant. However, likeother residual pozzolans (e.g., fly ash and silica fume), the cost ofthe raw material tends to increase with the increasing demand ofthemarket. Inwhat concerns the CER credits, the value of the tonneof CO2 has varied in the last years from V15 to V35.

The exercise proposed in this paper was to assess the potentialfor CDM in Brazil related to cement blended with SCBA, taking intoaccount the main production region in the country. This was nota proposed project activity under CDM or a programmatic CDM. ACDMmethodology which was conceived to be applied in a project-by-project basis was used. However, the application of the meth-odology in a regional basis is a good proxy for the real emissionsreduction potential for this technology.

6. Conclusions

� Sugar cane bagasse ash (SCBA) is a pozzolan that can partiallyreplace clinker in cement production and reduces emissions ofCO2 into the atmosphere. SCBA is an agro-industrial residueavailable in several countries, and it was proved by previouscomprehensive studies that its use generally improves thebehavior of the cementitious construction materials.

� The case study presented in this paper simulates the use ofSCBA in industrial scale for the southern eastern region ofBrazil. The emissions reduction estimated using the method-ology of UNFCCC was of 519.3 kilotonnes of CO2 per year.

� The increasing of blend in cement production using SCBA, forthe simulation carried out in this study, fulfill the conditions tobe candidate for CDMprojects, respecting the constraints of theUNFCCC approved and consolidated methodologies. Hencethere is potential for the issuance of Certified EmissionReduction credits.

Acknowledgements

The authors acknowledge the Brazilian agencies CNPq, CAPESand FAPERJ for their financial support, and the Universidade Federaldo Rio de Janeiro for the scholarship of the fourth author.

References

ABNT, 1991. Brazilian Association for Technical Standards, NBR5732 e CommonPortland cement, NBR11578 e Composite Portland cement, NBR5735 e Slagcement, NBR5736 e Pozzolanic cement, Rio de Janeiro.

Americano, B.B., 2008. CDM in Brazil: towards structural change for sustainabledevelopment in some sectors. In: Olsen, K.H., Fenham, J. (Eds.), A ReformedCDM: Including New Mechanisms for Sustainable Development. UNEP RisoCentre, Roskilde, pp. 23e33.

ANFAVEA, 2008. Associação Nacional de Veículos Automotores. http://www.anfavea.com.br/Index.html (accessed on 18.08.08.).

ASTM, 1997. American Society for Testing and Materials, Standard Test Method forElectrical Indication of Concrete’s Ability to Resist Chloride Ion Penetration.ASTM, Philadelphia. C1202.

Cordeiro, G.C., 2006. Use of sugar cane bagasse ash and rice husk ashes as mineraladmixture in concrete. D.Sc. thesis, Universidade Federal do Rio de Janeiro,Brazil. Available at: http://www.coc.ufrj.br.

Cordeiro, G.C., Silvoso, M.M., Fairbairn, E.M.R., Toledo Filho, R., Ribeiro, F.L.B.R.,Ferreira, I.A., 2007. User friendly concretes in Brazil: simulation of dam

E.M.R. Fairbairn et al. / Journal of Environmental Management 91 (2010) 1864e1871 1871

construction using cement blended with sugar cane bagasse and rice-huskashes. In: 18th Engineering Mechanics Division Conference. ASCE, Blacksburg.

Cordeiro, G.C., Toledo Filho, R.D., Fairbairn, E.M.R., 2008a. Use of ultra-fine sugarcane bagasse ash as mineral admixture for concrete. ACI Mater. J. 105, 487e493.

Cordeiro, G.C., Toledo Filho, R.D., Tavares, L.M., Fairbairn, E.M.R., 2008b. Pozzolanicactivity and filler effect of sugar cane bagasse ash in Portland cement and limemortars. Cement Concr. Compos. 30, 410e418.

Cordeiro, G.C., Toledo Filho, R.D., Tavares, L.M., Fairbairn, E.M.R., 2009a. Ultrafinegrinding of sugar cane bagasse ash for application as pozzolanic admixture inconcrete. Cem. Concr. Res. 39, 110e115.

Cordeiro, G.C., Toledo Filho, R.D., Fairbairn, E.M.R., 2009b. Effect of calcinationtemperature on the pozzolanic activity of sugar cane bagasse ash. Construct.Build. Mater. 23, 3301e3303.

FAO, 2008. The Food and Agriculture Organization of the United Nations. http://www.faostat.org (accessed on 08.05.08.).

Fairbairn, E.M.R., Silvoso, M.M., Toledo Filho, R.D., Alves, J.L.D., Ebecken, N.F.F., 2004.Optimization of mass concrete construction using genetic algorithms. Comput.Struct. 84, 281e299.

Ganesan, K., Rajagopal, K., Thangavel, K., 2007. Evaluation of bagasse ash assupplementary cementitious material. Cement Concr. Compos. 29, 515e524.

Hewlett, P.C., 2005. Lea’s Chemistry of Cement and Concrete, fourth ed. Elsevier,Oxford.

Malhotra, V.M., 2002. Introduction: sustainable development and concrete tech-nology. Concr. Int. 24, 22.

Malhotra, V.M., Mehta, P.K., 1996. Pozzolanic and Cementitious Materials, first ed.Gordon and Breach Publishers, Amsterdan.

Martirena Hernández, J.F., Middendorf, B., Gehrke, M., Budelmann, H., 1998. Use ofwastes of the sugar industry as pozzolana in lime-pozzolana binders: study ofthe reaction. Cement Concr. Res. 28, 1525e1536.

Morales, E.V., Villar-Cociña, E., Frías, M., Santos, S.F., Savastano Jr., H., 2009. Effects ofcalcining conditions on the microstructure of sugar cane waste ashes (SCWA):Influence in the pozzolanic activation. Cement Concr. Compos. 31, 22e28.

OECD/IEA, 2000. Emission Baselines: Estimating the Unknown. Organization forEconomic Co-operation and Development/International Energy Agency, NewYork.

Singh, N.B., Singh, V.D., Rai, Sarita, 2000. Hydration of bagasse ash-blended portlandcement. Cement Concr. Res. 30, 1485e1488.

UNFCCC (United Nations Framework Convention on Climate Change), 2004.Proposed new methodology baseline (CDM-NMB) version 1. http://cdm.unfccc.int/UserManagement/FileStorage/FS_220777679.

UNFCCC, 2005. Decision 3/CMP.1, modalities and procedures for a clean develop-ment mechanism as defined in Article 12 of the Kyoto Protocol. http://unfccc.int/resource/docs/2005/cmp1/eng/08a01.pdf#page¼6.

UNFCCC, 2009a. About CDM. http://cdm.unfccc.int/about/index.html.UNFCCC, 2009b. Revision to the approved consolidated baseline methodology

ACM0005 e consolidated baseline methodology for increasing the blend incement production. http://cdm.unfccc.int/methodologies/PAmethodologies/approved.html.

UNFCCC, 2009c. About calculation of emission factor. Tool to calculate the emissionfactor for an electricity system. http://cdm.unfccc.int/methodologies/PAmethodologies/tools/am-tool-07-v1.1.pdf.

UNFCCC, 2009d. Revision to the approved consolidated baseline methodologyACM0005 e consolidated methodology for grid-connected electricity genera-tion from renewable sources e version 9. http://cdm.unfccc.int/UserManagement/FileStorage/CDMWF_AM_71ZC14NVE4V5DHA3TUT3896PFLPVGG.

Worrell, E., Price, L., Martin, N., Hendriks, C., Meida, L.O., 2001. Carbon dioxideemissions from the global cement industry. Annu. Rev. Energ. Env. 26, 303e329.