lable at ScienceDirect

Applied Thermal Engineering 73 (2014) 236e242

Contents lists avai

Applied Thermal Engineering

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

Fluid-dynamic assessment of sugarcane bagasse to use as feedstock inbubbling fluidized bed gasifiers

Nestor Proenza P�erez*, Einara Blanco Machin, Daniel Travieso Pedroso,Julio Santana Antunes, Jose Luz SilveiraS~ao Paulo State University, Faculty of Engineering of Guaratinguet�a, Department of Energy, LOSE Laboratory, Brazil

h i g h l i g h t s

� The ideal range of diameters of sugarcane bagasse particles was proposed in the range between 0.8 and 1.21 mm.� The behavior of the main fluidization parameters for the range of diameters proposed was studied.� The composition of the producer gas was modeled, as well as its LHV reaching 4560 kJ/Nm3.

a r t i c l e i n f o

Article history:Received 28 March 2014Accepted 17 July 2014Available online 25 July 2014

Keywords:BiomassBubbling fluidized bed gasifierRenewable energySugarcane bagasse

* Corresponding author. Tel.: þ55 12 3123 2240.E-mail addresses: nestor@feg.unesp.br, nestorpro

einara@feg.unesp.br (E.B. Machin), danieltravieso@santana@feg.unesp.br (J.S. Antunes), joseluz@feg.unes

http://dx.doi.org/10.1016/j.applthermaleng.2014.07.041359-4311/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

The fluid bed gasification of a sugarcane bagasse is a promising option for the large-scale production offuel gas in order to increase the energetic efficiency of the sugar industry. A fluid-dynamic and ther-modynamic assessment of the use of sugarcane bagasse as feedstock for a bubbling fluidized bed gasifierwas performed. It was determined the required range of sugarcane bagasse particles sizes to permit theuse this biomass as feedstock in bubbling fluidized bed gasifiers and the main fluidization parameters forthe range of diameters proposed, with good agreements with the experimental values reported for thesame Geldart's type of particles. A range of particle sized between 0.8 and 1.21 mm was suggested. Theproducer gas composition was modeled using air as gasification agent, obtaining a theoretical LHV of4.56 MJ/Nm3.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Lignocellulosic biomass wastes are featured by their non-ediblecharacteristic; hence, they have no risk causing food shortage whenthe wastes are consumed for energy. For this reason, the utilizationof lignocellulosic biomass wastes as fuels has attracted remarkableinterest in the development of bioenergy.

Sugarcane is cultivated in more than 80 countries and the by-products obtained from the sugar production process represent agreat biomass potential. The harvest of sugarcane in the producingcountries is about 1.2 Gt and potentially its residue can be used foran electric power production of about 300 TWh y�1 [1]. Sugarcanehas a great capacity to produce biomass, yielding about 100 t/ha.

enza@yahoo.es (N.P. P�erez),feg.unesp.br (D.T. Pedroso),p.br (J.L. Silveira).

8

The sugarcane bagasse is undoubtedly the most representativebiomass in the Brazilian energy matrix, being responsible for thesupply of thermal, mechanical and electrical power to the industryunits that produce sugar and alcohol through cogeneration.

Bagasse is the crushed outer stalk material formed after thejuice extraction from sugarcane, in sugar mills. This by-productrepresents between 25% and 40% of the total processed materialin the juice extraction, depending of the sugarcane fiber contentsand the efficiency of the process.

Traditionally, sugar mills use bagasse and cane trash with a highmoisture content as fuel for low-pressure boilers to generate steam,using a conventional condensing-extraction steam turbine (CEST)technology, to provide the plant of heat, electricity and mechanicalpower. In recent years have seen moremodern systems for burningbagasse in suspension, that allow to raise the steam pressure andtemperature for the purpose of obtaining a higher electric powercycle cogeneration [2e4]. The thermal efficiency of the plant isusually in the range of 15e30% [5], consequently the size of con-ventional combined heat and power generation plants from

Nomenclature

LHV lower heating value [MJ Nm�3] [MJ kg�1]xnX Predicted value in mol of element X [mol]K1 methane formation constant [e]K2 constants of shift-gas reaction [e]DG0 standard Gibbs function of formation [kJ kmol�1]J,I integration constantsDH0 heat of formation [J mol K�1]A,B,C,D constants for determination specific heating [e]K equilibrium constant [e]εmf minimum fluidization porosity [e]εf bed porosity [e]Ar number of Archimedes [e]

dP particle diameter [e]rf density of the air at the temperature and pressure of

entry in the gasifier [kg m�3]rp density of bed material [kg/m�3]mf air viscosity at the temperature and pressure of entry

in the gasifier [Pa s]g gravity acceleration [m2 s�1]Vmf minimum fluidization velocity [m s�1]f sphericity of bed particles [e]Vt fluidization terminal velocity [m s�1]CD drag coefficient [e]Vf fluidization velocity [m s�1]Remf Reynolds's number at minimum fluidization estate [e]ra bulk density [kg m�3]

N.P. P�erez et al. / Applied Thermal Engineering 73 (2014) 236e242 237

bagasse, have been limited by these low efficiencies and theamount of fuel within an economical transportation radius.

In the wake of past developments of coal gasification, biomassgasification is one of the most scrutinized avenues to biomassthermochemical conversion where the produced gas can be usedeither for energy production (heat, electricity) or as a startingmaterial for the production of fuels and various chemicals [6].

The Biomass Integrated-Gasifier/Gas Turbine Combined Cycle(BIG/GTCC) technology has being identified by several authors [7,8]as an advanced technology with potential to be cost-competitivewith CEST technology using the biomass by-products ofsugarcane-processing as fuel, while dramatically increasing theelectricity generated per mass unit of sugarcane processed.

Numerous studies have identified fluid bed gasification of asugarcane bagasse as a promising option for the large-scale pro-duction of fuel gas from this biomass, in order to increase the en-ergetic efficiency of the sugar industry [5,9e11]. However, there arelimited information about the sugarcane bagasse requirements aswell as the thermodynamic and fluid-dynamic behavior of thisfeedstock in bubbling fluidized bed reactors.

Against this background, the goal of this paper is a fluid-dynamic assessment of the use sugarcane bagasse as feedstockfor a bubbling fluidized bed gasifier.

Fig. 1. a) Bubbling fluidized bed gasifier. b) Circulating fluidized bed gasifier.

2. Gasification in fluidized bed reactors

Gasification is a thermochemical process in which a carbona-ceous substrate (biomass) is transformed into a fuel gas (producergas), through a number of reactions that take place at a high tem-perature in the presence of a gasifying agent (air, oxygen or watervapor). The producer gas is the principal product of gasification,and its heating value (LHV) varies depending on the composition ofbiomass and the gasifying agent utilized. Using air as the gasifyingagent for biomass gasification, the LHV of the producer gas is in therange of 4e6 MJ/Nm3 and using water vapor and oxygen the LHV isbetween 8 and 20 MJ/Nm3 [12].

Fluidized bed reactors are those in which the gasification agentcirculates inside of the reactor bed at a rate in which the bed ma-terial remain in a fluidization state, increasing the energy transferintensity between the inert material and the fuel, as well as thehomogenization of the temperature in the reactor bed. There aretwo categories within these types of gasifiers: bubbling andcirculating (Fig. 1).

In the circulating fluidized bed (CFBG), the velocity of the gasi-fication agent is high, resulting in a fuel circulation, enabling a highenergetic efficiency of the gasification process in these reactors.

This solid is recirculated to the reactor by the use of a cyclone and areturn system to the gasifier.

In the bubbling fluidized beds (BFBG), the fluidization velocity ofthe gasification agents is low enough that do not causes a signifi-cant movement of the fuel in the bed. The size of the bubblesprincipally depend of the size of fuel particles and the height of thereactor bed.

The main advantages of fluidized beds include better controlof temperature and reaction rates, high specific capacity, poten-tial scaling to larger sizes than fixed bed reactors and possibilityto manage a wide biomass types. As drawback, this technologyshow moderate-high tars and particulates levels in the exhaustgas and the fuel conversion are not as high as in the fixed bedgasifiers.

The technology of CFBG is more adequate for big powergeneration systems (greater than 100 MW), due to its highoperation cost and complexity, compared to the BFBG. The BFBGare suitable for smaller power generation systems (less than100 MW), that is our case of study, where be used as feedstockthe surplus bagasse of the plant. The scaling of this technology toa larger capacity is limited by the operation difficulties observedduring the handling, transportation and feeds of large quantitiesof sugarcane bagasse to a fluidized bed gasifier, in previous ex-periences [13]. The sugarcane bagasse gasification in a BFBG isthe objective of this work.

N.P. P�erez et al. / Applied Thermal Engineering 73 (2014) 236e242238

3. Fuel properties of sugarcane bagasse

The bagasse characteristics vary in composition, consistency,and heating value depending on the climate, type of soil, variety ofcane, harvesting method, amount of cane washing, and the effi-ciency of the milling plant. Bagasse is not a homogeneous material.Around 30e35% of its structure is “pith” fibre, which is derivedfrom the core of the plant and “rind”, or “stem” fibre, which com-prises the balance. After crushing, the larger fraction, about 60%, isapparently cylindrical, while the smaller fraction, about 40%, ismore spherical in shape [1,14e16].

Bagasse is composed of fiber, and water soluble materials,mainly sugar and impurities. The fiber composition has an averagecomposition of 50% cellulose, 25% hemicellulose and 25% lignin[17e19].

Table 1 show the main physical and chemical properties ofsugarcane bagasse reported by several authors (Table 2).

The bagasse is classified as a fuel with high reactivity due to itshigh content of volatiles and low ash content, also the higherheating value of sugarcane bagasse is comparable with the value ofseveral woody biomasses [22], making of this fuel, a good feedstockfor gasification.

3.1. Ash composition

The low contents of alkaline salts in the chemical composition ofsugarcane bagasse increase the quality of this biomass as feedstockfor gasifiers. Those components are the reason for the rapid for-mation of unwanted deposits on the fireside surfaces of gasifiers.Such deposits retard the rate of heat transfer as a result of their lowthermal conductivity and high reflectivity.

The alkali index expresses the quantity of alkali oxide within thefuel per unit of fuel energy (kgAlkali/GJ). Although this index isincomplete as an overall descriptor of slagging and foulingbehavior, it gives useful hints for assessing the behavior of a certaintype of biomass as a fuel. Experiences have shown that above0.17 kgAlkali/GJ, fouling is probable and above 0.34 kgAlkali/GJ foulingis virtually certain to occur [30]. These threshold levels are deter-mined based on experience andmay thus vary considerably. Table 3shows the values determined here based on the data outlinedabove.

Thus, sugarcane bagasse show a low value of the alkali indexcompared with other types of biomass. It has to be considered thefact that bagasse is subjected to a leaching process during the juiceextraction procedure. Following these values, no major problemswith slagging and fouling are expected during fluidized bed gasi-fication of this fuel.

Table 1Main properties of sugarcane bagasse.

[20] [21] [22] [23] [1] [24] [25] [26] [27]

Proximate analysisVolatile matter

(wt.%, dry basis)87.06 83.0 85.61 83.0 88.7 82.1 85.43 79.35 n.a

Fixed carbon(wt.%, dry basis)

12.94 13.0 11.95 13.0 9.3 16.3 12.89 17.88 n.a

Ash (wt.%, dry basis) n.a. 4.0 2.44 1.9 2.0 1.6 1.68 3.66 6.8Higher heating value

(MJ kg�1, dry basis)18.6 18.9 18.99 n.a 18.7 19.19 19.14 19.41 18.85

Ultimate analysis (wt.%, dry basis)C 47.0 46.3 48.65 45.48 42.9 48.81 49.0 48.4 46.7H 5.9 6.4 5.87 5.70 5.9 6 5.87 6.01 6.2O (by difference) 45.81 43.0 42.82 45.21 49.0 43.1 43.27 41.61 39.8N 0.33 n.a. 0.16 0.40 0.20 0.46 0.1 0.17 0.2S 0.05 0.1 0.04 0.06 n.a. 0.1 0.06 0.02 0.02Cl n.a. n.a 0.03 n.a. n.a <0.01 0.02 n.a 0.06

The fluid-dynamic behavior of particles bed is determined bythe physical properties of bed material. In the case of sugarcanebagasse particles, the variety of grain sizes and shapes and its lowbulk density, make the fluidization of the sugarcane bagasseextremely difficult. According to [32,33], when the dimensions ofthe irregular and low density Switchgrass particles, are 31 timesbigger than the inert material of the bed; the cohesion and enlaceforces between the biomass particles are increased, formingchannels and requiring an additional external force to break downthe cohesion formed, meaning an increase of the fluidization ve-locity. There are not previous studies for the case for the sugarcanebagasse, but a similar behavior is expected.

3.2. Size distribution of the sugarcane bagasse particles

Physical properties of bagasse, which are related to its origin,make it a difficult material to process. Both the particle size dis-tribution and the behavior of assemblies of bagasse particles areimportant considerations for the selection and design of equipmentthat will be involved in the processing of this biomass (trans-portation, feeding, collection, thermal-conversion, separation ofdust, pelletizing and drying). A wide particle size distribution af-fects handling, influences the retention time and causes elutriationin the fluid bed and at the cyclones [34]. In the case of the sugarcanebagasse the particle size is expected to vary from each factory dueto different knifing, shredding and milling as well as samplingtechniques. Typically, the statistical parameters which describesparticle size distribution are mean, mode, standard deviation, andseveral cumulative percentile values (the size at which a deter-mined percentage of particles are larger or coarser) D10, D50, D90[35]. Polanco et al. [36]. report (Table 4) the particle size distribu-tion of sugarcane bagasse samples collected from different loca-tions and pretreatments during the 2010 sugarcane season inLouisiana, USA; using a Restch Sieving Machine (0.125e9.5 mm).

Other study of the size distribution of sugarcane bagasse par-ticles with similar results, have also been reported [37e39]. Thewide size distribution of the sugarcane bagasse particles (Table 4)influences the retention time, causes elutriation and channeling,(i.e. decreases the fluidization quality [40] of the reactor fluid bed);requiring a pretreatment process for the reduction of the gap in theparticles size distribution to use this biomass as feedstock. In orderto reduce the global energy losses in the generation process isrequired to know the higher size of particle that is possible tointroduce in a bubbling fluidized bed gasifier, without affect thefluidization quality of the bed.

4. Fluid-dynamic of the gasifier bed

The higher size of particle that is possible to introduce in abubbling fluidized bed gasifier, without affect the fluidizationquality of the bed, is function of the properties of the biomass usedas feedstock; in the case of sugarcane bagasse the maximum par-ticles size was determined based in the properties reported inprevious studies [36e39]. The determination of the maximumparticles size is needed in order to save energy during the biomasspretreatment process for the reduction of the gap in the particlessize distribution.

For this study were considered alumina as inert bed materialwith a bulk density of 1760 kg/m3 and sphericity of 0.4 with aparticle diameter of 379 mm; the fluidization agent considered wasair and reactor diameter of 500 mm.

The higher size of particle is determined from the study of thebehavior of a fluidized bed of sugarcane bagasse. The porosity of thebed is characterized by the bulk and the real density of the particles[41], defined as:

Table 2Ash composition of sugarcane bagasse.

Element in wt.% of ash

SiO2 CaO MgO Fe2O3 Al2O3 Na2O K2O SO3 P2O5 TiO2

Ref. [28] 45.88 4.31 3.22 15.45 20.55 0.96 1.67 0.40 0.89 3.77Ref. [25] 46.88 5.35 5.81 7.98 11.4 2.45 9.81 5.08 5.04 1.39Ref. [22] 46.61 4.47 3.33 14.14 17.69 0.79 0.15 2.08 2.72 2.63Ref. [27] 73.19 4.14 2.53 5.37 8.29 0.67 4.11 n.a. 0.91 n.a.Ref. [29] 84.0 1.2 1.7 1.8 3.6 0.9 2.3 0.9 n.a. 0.3

N.P. P�erez et al. / Applied Thermal Engineering 73 (2014) 236e242 239

εf ¼ 1� ra

rp

!(1)

where:

ra: bulk density.rp: real density.

The real density was considered, as the density of particlesincluding the voids within the individual particle and bulk density,as the overall density of material including interparticle distanceseparation.

For a free-standing bed there will exist a point, known as theminimum or incipient fluidization point, whereby the bed is sus-pended directly by the flow of the fluid stream. The correspondingfluid velocity is known as the “minimum fluidization velocity” [42]and for systems, where Reynolds number is lower than 20 (Re< 20),it can be calculated by the Ergun simplified equation as:

Vmf ¼d2p$�rp � rf

�$g

150$mf$ε3mf$f

2

1� εmf(2)

The fluidization terminal velocity (Vt) is the velocity that is largeenough to lift a single particle and carry it out of the fluidized bed.Vt can be determined from the relationship between the equipmentdimensions and the bed characteristics; also depend on the parti-cle's Reynolds number and the drag coefficient.

Vt ¼0@4$

�rp � rf

�$g$dp

3$CD$rf

1A

0:5

(3)

The fluidization velocity during the gasification can be definedas the superficial gas velocity to be used during operation of thegasifier. It has been established taking into account the height andthe characteristics of the fluidized bed [43].

Vf ¼" �

1:25� 1�,V0:937

mf $r0:126f

10:978$r0:376p $d1:006p

!#1:355þ Vmf (4)

Table 3Alkali index for the sugarcane bagasse and comparisonwith other typesof biomass.

Sugar cane bagasse [30] 0.06 kgAlkali/GJFir (wb) [30] 0.06 kgAlkali/GJWillow wood (wb) [30] 0.14 kgAlkali/GJPoplar (wb) [30] 0.14 kgAlkali/GJWheat straw [30] 1.07 kgAlkali/GJMarabú (wb) [31] 0.08 kgAlkali/GJRice straw [30] 1.64 kgAlkali/GJ

wb: with bark.

5. Results

5.1. Bed porosity of sugarcane bagasse

Based in experimental data reported [38,39] was developedmathematical models for the behavior of the bulk and real densityin function of the particle diameter using and exponential adjust asis shown in Fig. 1.

Based in the adjust equations was possible to determinate thebehavior of the porosity for the sugarcane bagasse bed for differentparticles diameters shown in Fig. 2.

For design parameters the static bed porosity is assumed as 0.4for solid materials [44] in the case of bagasse particles, a poly-disperse materials with very low density; this parameter must begreater, giving in values between 0.57 and 0.68 next to the exper-imental values reported [45] for different biomasses, with similarcharacteristics to the sugarcane bagasse. The values have a ten-dency to decreases with the increment of the particle diameter, dueto the fact that the real density increases more rapidly withdecreasing particle size, than the bulk density increase; until apoint where is observed the smaller variation of the porosity value(between 0.8 and 1.3 mm). Beyond this point, a change in themonotony is observed.

5.2. The fluidization velocities

The recommended fluidization velocity (Vf) for a good fluidiza-tion quality, high carbon conversion efficiency and safely functionof a bubbling fluidized reactor, based in experimental reports[9,46,41], is between 0.7 and 0.75 m/s. Fig. 3 shows the behavior ofthe Vmf, Vf and Vt as function of the particle radio for sugarcanebagasse (Fig. 4).

Commensurate with the results obtained for the Vf (Fig. 3) forthe sugarcane bagasse, the particles diameter must be lower than1.21 mm, considering a particle with a cylindrical form withsphericity of 0.70; in order to keep Vf equal or lower than 0.75 m/s.Seeing the bed porosity behavior for different particles sizes of thisfeedstock (Fig. 2), the ideal range of diameters is between 0.8 and1.21mm, where the porosity of the bed reach the lower value (0.58)and remain almost constant.

According to the results reported by Polanco et al. (Table 3), thevolume of particles with size larger than 1.21 mm in the sugarcane

Table 4Particle size distribution of different samples of sugarcane bagasse [36].

Sample D10 D50 D90 D90/D10 (D90eD10)/D90 Particles �0.125 mm

Description 10�3 m Ratio Span %Mass

Sample 1 0.16 0.90 5.36 34.1 5.8 8.6Sample 2 0.12 0.71 4.90 40.8 6.7 10.4Sample 3 0.12 0.78 6.36 52.6 8.1 10.4Sample 4 0.24 0.95 6.63 28.0 6.7 4.6Sample 5 0.30 1.28 3.92 13.1 2.8 4.0Sample 6 0.21 0.80 4.45 20.8 5.3 4.0

Fig. 2. a) Bulk density as function of particle radio, b) real density as function of particle radio.

N.P. P�erez et al. / Applied Thermal Engineering 73 (2014) 236e242240

bagasse produced in the sampled sugar industries; represent be-tween 40 and 50% of the total volume of sugarcane bagasse pro-duced. This indicate the necessity of a pretreatment process withthe aim of reduce the larger size of the sugarcane bagasse particlesto a size equal or below 1.21 mm.

The obtained parameters for the fluidization, in bubbling flu-idized bed regime, of the sugarcane bagasse with the mentionedcharacteristics are reported in Table 5.

Table 5 reports the fluid-dynamic parameters of the bagasseparticle, Vmf is very close to the experimental values reported[47,45] which reports values of Vmf for different biomass like sug-arcane bagasse. Otherwise Vf has a bit larger than the optimal valuefor the superficial gas velocity [46] for the safety operation of thereactor, around 0.75 m/s.

5.3. Prediction of product gas composition using thermodynamicproperties

The gases distribution in the producer gas is predicted using amodel developed by Zainal [48] from the fuel composition; throughseveral sets of equation as a function of gasifying temperature. The

Fig. 3. Bed porosity of the sugarcane bagasse in function of the particle diameter.

model was developed on the assumption that all reactions are inthermodynamic equilibrium. The global equation for gasificationbased on the biomass composition is defined as follows:

CH1:488O0:7 þwH2OþmO2 þ 3:76mN2

¼ x1H2 þ x2COþ x3CO2 þ x4H2Oþ x5CH4 þ 3:76mN2 (5)

According to the methodology developed by Zainal [48], thevalues of the model parameters DA, DB, DC, DD, I and J for thesugarcane bagasse were determined, as well as the values ofStandard Gibbs Function and the Heat of Formation. The obtainedvalues are reported in Table 6 (see Table 7).

After the determination of K1 and K2 for the sugarcane bagasse,the resolution of the set of equations generated with the model,produced a set of polynomial equations, where were fixed thevalues of m and w, considering a sub-estequiometric ration of 0.3and water content of bagasse of 10% w.b. The equations systemwassolved using a NewtoneRaphson method in order to determinatethe values of xn. The predicted values for the composition of theproducer gas from the sugarcane bagasse gasification are reportedin Table 5.

Fig. 4. Behavior of the Vmf, Vf and Vt as function of the particle radio.

Table 6Calculated values for the sugarcane bagasse.

Variable K1 K2

DA �6.567 �1.86DB 0.007466 0.000538DC �0.000002164 0DD 70,100 116,400DH� 298(J/mol) �74,520 41,166DG� 298(J/mol) �50,460 28,618J �58,886.8003 48,823.64I 32.541 18.014K 0.0690 0.781

Table 5Parameters and fluid-dynamic profiles of the gasifier.

Parameter Symbol Value Unit

Number of Archimedes Ar 7578 e

Fluidization velocity Vf 0.741 m/sFluidization minimum velocity Vmf 0.370 m/sTerminal fluidization velocity Vt 1.866 m/sReynolds number for minimum fluidization velocity Remf 16.87 e

N.P. P�erez et al. / Applied Thermal Engineering 73 (2014) 236e242 241

The lower heating value for the biomass producer gas is calcu-lated as follows [49]:

LHVgas ¼ 0:126$CCO þ 0:358$CCH4þ 0:018$CH2

(6)

where CCO, CCH4, and CH2

, are the percentage volumetric concen-trations of CO, CH4 and H2 in the producer gas; in the case of study,the LHV of the producer gas is:

LHVgas ¼ 4560kJ

Nm3

The predicted composition of the producer gas, as well as theLHV obtained, are typical for the values obtained with the airgasification of this type of biomass [50,51], showing that sugarcanebagasse have an adequate composition to be used as feedstock forbubbling fluidized bed gasification reactors.

6. Conclusion

The range of particle sizes allowing the use of sugarcane bagasseas feedstock in bubbling fluidized bed gasifiers with a good fluid-ization quality, high carbon conversion efficiency and safely func-tion of the gasifier was proposed, and was modeled the behavior ofthe main fluidization parameters for the range of diameters pro-posed. The ideal size of the particles should range between 0.8 and1.21mm, where the porosity of the bed reach the lower value (0.58)and remain almost constant with good agreements with theexperimental values ported for the same Geldart's type of particles.The composition of the producer gas was modeled, as well as the

Table 7Theoretical producer gas composition from gasification of sugarcane bagasse.

Element Variables Predictedvalues

Molefraction

% In wetbase

% In drybase

H2 x1 0.538 0.1643 16.43 18.21CO x2 0.669 0.2043 20.43 22.61CO2 x3 0.311 0.0949 9.49 10.5H2O x4 0.32 0.0977 9.77 0CH4 x5 0.02 0.0061 0.671 0.67N2 3.76 � m 1.419 0.4334 43.34 48.05Total 3.274 1.0000 100.00 100.00

LHV (4560 kJ/Nm3); the results are similar to the typical valuesobtained with the air gasification of this type of biomass, showingthat sugarcane bagasse have an adequate composition to be used asfeedstock for bubbling fluidized bed gasification reactors.

Acknowledgements

We are grateful to the Coordination for the Improvement ofHigher Education Personnel (CAPES) (process 5993105), from theBrazilian Ministry of Education (MEC) and to the National Councilfor Scientific and Technological Development (CNPq) (process162633/2013-0) from the Ministry of Science and Technology(MCT) for their generous financing support to this research.

References

[1] P. Filippis, C. Borgianni, M. Paolucci, Gasification process of Cuban bagasse intwo-stage reator, Biomass Bioenergy 27 (2004) 247e252.

[2] V. Rasul, M.G. Rudolph, Fluidized bed combustion of Australian bagasse, Fuel79 (2000) 123e130.

[3] R. Deshmukh, A. Jacobson, C. Chamberlin, D. Kammen, Thermal gasification ordirect combustion? comparison of advanced cogeneration systems in thesugarcane industry, Biomass Bioenergy 55 (Aug. 2013) 163e174.

[4] F.J. dos Santos, L. Goldstein, Experimental aspects of biomass fuels in abubbling fluidized bed combustor, Chem. Eng. Process. Process. Intensif. 47(9e10) (Sep. 2008) 1541e1549.

[5] E.D. Larson, R.H. Williams, M.R.L.V. Leal, A review of biomass integrated-gasifier/gas turbine combined cycle technology and its application in sugar-cane industries, with an analysis for Cuba, Energy Sustainable Dev. 5 (1) (Mar.2001) 54e76.

[6] A. Huber, G.W. Iborra, S. Corma, Synthesis of transportation fuels frombiomass: chemistry, catalysts, and engineering, Chem. Rev. 106 (2006) 4044.

[7] P. Babu, P. Suresh, Whaley, The Biomass Thermal Gasification Project, vol. 2,1992, pp. 299e306.

[8] W. Alonso Pippo, P. Garzone, G. Cornacchia, Agro-industry sugarcane residuesdisposal: the trends of their conversion into energy carriers in Cuba, WasteManage. 27 (7) (Jan. 2007) 869e885.

[9] E. G�omez, L.A.B. Cortez, Preliminary tests with a sugarcane bagasse fueledfluidized-bed air gasifier, Energy Convers. 40 (1999).

[10] L. Pellegrini, Exergy analysis of sugarcane bagasse gasification, Energy 32(2007) 314e327.

[11] M. Gabra, E. Pettersson, R. Backman, B. Kjellstr€om, Evaluation of cyclonegasifier performance for gasification of sugar cane residuedpart 1: gasifica-tion of bagasse, Biomass Bioenergy 21 (2001) 351e369.

[12] A. Reed, T.B. Das, Handbook of Biomass Downdraft Gasifier Engine Systems,The Biomass Energy Fundation Press, 1998.

[13] S. Babu, Biomass Gasification for Hydrogen Productioneprocess Descriptionand Research Needs, IEA, Gas Technol. Institute, IL, USA, Oct. 29, 2002, pp.1e10.

[14] Lijun Wang, Curtis L. Wellerb, David D. Jonesb, Milford A. Hanna, Contem-porary issues in thermal gasification of biomass and its application to elec-tricity and fuel production, Biomass Bioenergy 32 (7) (Jul. 2008) 573e581.

[15] E.O. G�omez, L.A.B. Cortez, G.R. Alarcon, G.J.D.M. Rocha, V.F.N. da Silva, E. deAlmeida, Some simplified geometrical properties of elephant grass and sug-arcane trash particles, Fuel Process. Technol. 104 (Dec. 2012) 234e244.

[16] C. Driemeier, M.M. Oliveira, F.M. Mendes, E.O. G�omez, Characterization ofsugarcane bagasse powders, Powder Technol. 214 (1) (Nov. 2011) 111e116.

[17] J. Bourgois, R. Guyonnet, Characterization and analysis of torrefied wood,Wood Sci. Technol. 22 (2) (1988) 143e155.

[18] J. Bourgois, M.C. Bartholin, R. Guyonnet, Thermal treatment of wood: analysisof the obtained product, Wood Sci. Technol. 23 (4) (1989) 303e310.

[19] J.M. Mesa-P�erez, J.D. Rocha, L.A. Barbosa-Cortez, M. Penedo-Medina,C.A. Luengo, E. Cascarosa, Fast oxidative pyrolysis of sugar cane straw in afluidized bed reactor, Appl. Therm. Eng. 56 (1e2) (Jul. 2013) 167e175.

[20] P. Rein, Cane Sugar Engineering, Verlag Dr Albert Bartens KG, Berlin, 2007, p.768.

[21] C.A. Camargo, H.A. Ushima, Conservaç~ao de Energia na Indústria do Açúcar edo �Alcool, Instituto de Pesquisas Tecnol�ogicas e IPT, Sao Paulo, Brazil, 1990, p.796.

[22] B. Jenkins, L. Baxter, T.R. Miles Jr., T. Miles, Combustion properties of biomass,Fuel Process. Technol. (1998).

[23] IPT, Apostilas do curso de Combust~ao Industrial. S~ao Paulo, 2007.[24] M. Garcıa-P�erez, A. Chaala, C. Roy, Co-pyrolysis of sugarcane bagasse with

petroleum residue. Part II. Product yields and properties, Fuel 81 (2002).[25] T.R. Miles, T.R. Miles Jr., L.L. Baxter, R.W. Bryers, B.M. Jenkins, L.L. Oden, Alkali

Deposits Found in Biomass Power Plants: a Preliminary Investigation of TheirExtent and Nature, vol. I, 1995. NREL/TP-433-8142 Sand 96-8225, Golden.

[26] F.A. Agblevor, H.L. Chum, D.K. Johnson, Compositional analysis of NISTbiomass standards from the IEA whole feedstock round robin, in: Energy fromBiomass and Wastes XVI: Proceedings of the Institute of Gas Technology

N.P. P�erez et al. / Applied Thermal Engineering 73 (2014) 236e242242

Conference, 2e6 March, Institute of Gas Techn.”, Orlando, Florida. Chicago, IL,1992.

[27] E. Natarajan, M. €Ohman, M. Gabra, A. Nordin, T. Liliedahl, A.N. Rao, Experi-mental determination of bed agglomeration tendencies of some commonagricultural residues in fluidized bed combustion and gasification, BiomassBioenergy 15 (2) (1998).

[28] S.Q. Turn, B.M. Jenkins, L. a. Jakeway, L.G. Blevins, R.B. Williams, G. Rubenstein,C.M. Kinoshita, Test results from sugar cane bagasse and high fiber cane co-fired with fossil fuels, Biomass Bioenergy 30 (6) (Jun. 2006) 565e574.

[29] J. Joyce, T. Dixon, J.C. Diniz da Costa, Characterization of sugar cane wastebiomass derived chars from pressurized gasification, Process Saf. Environ.Prot. 84 (6) (Nov. 2006) 429e439.

[30] T.R. Miles, T.R. Miles Jr., L.L. Baxter, R.W. Bryers, B.M. Jenkins, L.L. Oden, AlkaliDeposits Found in Biomass Power Plants: a Preliminary Investigation of TheirExtent and Nature, vol. II, 1995. NREL/TP-433-8142 Sand 96-8225, Golden.

[31] D.T. Pedroso, M. Kaltschmitt, Dichrostachys cinerea as a possible energycropdfacts and figures, Biomass Convers. Biorefin. 2 (1) (Nov. 2011) 41e51.

[32] R.R. Pattipati, C.Y. Wen, Minimum fluidization velocity at high temperatures,Ind. Eng. Chem. Process Des. Dev. 20 (4) (Oct. 1981) 705e707.

[33] V.S. Chok, A. Gorin, H.B. Chua, Minimum and complete fluidization velocity forsandepalm shell mixtures, part II: characteristic velocity profiles, criticalloading and binary correlations, Am. J. Appl. Sci. 7 (6) (2010) 773e779.

[34] R.B. Keey, Drying of Loose and Particulate Materials, Hemisphere, New York,1992.

[35] S.J. Blott, K. Pye, GRADISTAT, “a grain size distribution and statistics packagefor the analysis of un-consolidated sediments,”, Earth Surf. Processes Land-forms 26 (11) (2001) 1237e1248.

[36] Polanco Luz Stella, Kochergin Vadim, F. Alvarez Jose, Fluidized bed super-heated steam dryer for bagasse: effects of particle size distribution,J. Sustainable Bioenergy Syst. 3 (2013) 265e271.

[37] E. de Almeida, Study of the Pneumatic Separation of Sugarcane BagasseFractions and its Influence on Enzymatic Hydrolysis, Campinas, 2012.

[38] G.A.R. Alarcon, C.G. Sanchez, E.O. Gomez, L.A.B. Cortez, Caracterizaci�on delBagazo de ca~na de Azúcar. Parte I Características Físicas, in: Proceedings of

the 6. Encontro de Energia no Meio Rural, 2006, Campinas, Brazil, 2006, pp.1e10.

[39] P. Lenço, Caracterizaç~ao do Bagaço de cana-de-açúcar para Geraç~ao de Ener-gia, Universidade Estadual de Campinas, 2010.

[40] S.K. Wee, V.S. Chok, C. Srinivasakannan, H.B. Chua, H.M. Yan, Fluidizationquality study in a compartmented fluidized bed gasifier (CFBG)y, Energy Fuels22 (1) (Jan. 2008) 61e66.

[41] N.P. Cheremisinor, Hydrodynamics of Gasesolids Fluidization, Gulf PublishingCompany, Honston, 1984, pp. 2e10, 137e161.

[42] R.G. Holdich, Fundamentals of Particle Technology, Midland InformationTechnology & Publishing, 2002.

[43] N. Abatzoglou, N. Barker, P. Hasler, H. Knoef, The development of a draftprotocol for the sampling and analysis of particulate and organic contami-nants in the gas from small biomass gasifiers, Biomass Bioenergy 18 (2000).

[44] J.S. Santana, C.C. D'Avila, Design aspects of fluidized bed combustors, in:Special Topics in Particulate Systems, Internal Report, vol. 2, Department ofChemical Engineering, Sao Carlos, Brazil, 1986.

[45] M. Abdullah, Z. Husain, S.Y. Pong, Analysis of cold flow fluidization test resultsfor various biomass fuels, Biomass Bioenergy 24 (2003) 487e494.

[46] C. Sanchez, E. Lora, Biomass fluidized bed gasification research in the StateUniversity of Campinas, Energy Sustainable Dev. 1 (4) (1994) 31e34.

[47] E.O. Gomez, Design, Construction and Preliminary Evaluation of a FluidizedBed Reactor for Gasification of Sugar Cane Bagasse, State University of Cam-pinas, 1996.

[48] Z. Zainal, R. Ali, C. Lean, K. Seetharamu, Prediction of performance of adowndraft gasifier using equilibrium modeling for different biomass mate-rials, Energy Convers. 42 (2001).

[49] N. Sanford, K. Gregory, Thermodynamics, Cambridge University Press, 2011, p.1072.

[50] Rex T.L. Ng, Douglas H.S. Tay, Wan Azlina Wan Ab Karim Ghani, Denny K.S. Ng,Modelling and optimisation of biomass fluidised bed gasifier, Appl. Therm.Eng. 61 (1) (Oct. 2013) 98e105.

[51] R.I. Singh, A. Brink, M. Hupa, CFD modeling to study fluidized bed combustionand gasification, Appl. Therm. Eng. 52 (2) (Apr. 2013) 585e614.