creșterea productiei de bioetanol din trestie de zahar prin cogenerare.pdf

8/11/2019 Creterea productiei de bioetanol din trestie de zahar prin cogenerare.pdf

1/13

Improving bioethanol production from sugarcane: evaluation of distillation,thermal integration and cogeneration systems

Marina O.S. Dias a,b,*,Marcelo Modesto c, Adriano V. Ensinas c, Silvia A. Nebra d, Rubens Maciel Filho a,Carlos E.V. Rossell a,b

a School of Chemical Engineering, University of Campinas, P.O. Box 6066, 13083-970 Campinas, SP, Brazilb Laboratrio Nacional de Cincia e Tecnologia do Bioetanol (CTBE), P.O. Box 6170, 13083-970 Campinas, SP, Brazilc Centre of Engineering, Modeling and Applied Social Sciences, CECS, Federal University of ABC, Rua Santa Adlia, 166, 09210-170 Santo Andr, SP, Brazild Interdisciplinary Centre of Energy Planning - NIPE, University of Campinas, P.O. Box 6192, 13400-970 Campinas, SP, Brazil

a r t i c l e i n f o

Article history:

Received 1 March 2010Received in revised form9 September 2010Accepted 10 September 2010Available online xxx

Keywords:

BioethanolProcess integrationSugarcaneExergetic analysis costPower generation

a b s t r a c t

Demand for bioethanol has grown considerably over thelast years. Even thoughBrazil hasbeen producingethanolfrom sugarcaneon a largescalefor decades, this industry is characterized bylow energyefciency,using a large fraction of the bagasse produced as fuel in the cogeneration system to supply the processenergy requirements.The possibility of selling surplus electricity to the gridor usingsurplusbagasseas rawmaterial of other processeshas motivatedinvestmentson moreefcient cogeneration systems and processthermal integration. In this work simulations of an autonomous distillery were carried out, along withutilities demand optimization using Pinch Analysis concepts. Different cogeneration systems wereanalyzed: a traditional Rankine Cycle, with steam of high temperature and pressure (80 bar, 510 C) andback pressure and condensing steam turbines conguration, and a BIGCC (Biomass Integrated GasicationCombinedCycle),comprised bya gas turbine setoperatingwith biomass gas producedin a gasierthatusessugarcane bagasse as raw material. Thermoeconomic analyses determining exergy-based costs of elec-tricity and ethanol for both cases were carried out. The main objective is to show the impact that these

process improvements can produce in industrial systems, compared to the current situation.2010 Elsevier Ltd. All rights reserved.

1. Introduction

Increased interest on alternative fuels has been observed in thepast few years, as a result of increasing energy demand and fore-casted depletion of fossil resources[1,2]. Global warming and theconsequent need to diminish greenhouse gases emissions haveencouraged the use of fuels produced from biomass [3], which isthe only renewable carbon source that can be efciently convertedinto solid, liquid or gaseous fuels[4].

Bioethanol is presently the most abundant biofuel for automobiletransportation [5]. It is produced from fermentation of sugarsobtained from biomass, either in the form of sucrose, starch orlignocellulose. Sugarcaneis so far the most efcient raw material forbioethanol production: the consumption of fossil energy duringsugarcane processing is much smaller than that of corn[6]. One ofthe main by-products generated during sugarcane processing issugarcane bagasse, whichis usually burnt in boilers forproduction ofsteam and electrical energy, providing theenergy necessaryto fulll

the process requirements. The use of efcient technologies forcogeneration coupled with optimization of bioethanol productionprocess allows the production of surplus bagasse[7], which may beused as a fuel source for electricity generation or as raw material forproducing bioethanol and other biobased products [8,9]. Differentcogeneration systems, such as the Rankine Cycle with condensingsteam turbines and those based on gasication technologies, mayimprovetheamount of electricity produced fromsugarcanebagasse,thus allowing its sell to the grid.

Optimization of bioethanol production process can reduce energyconsumption, consequently increasing sugarcane bagasse avail-ability. Thermal integration using Pinch Technology can reduce hotand coldutilitiesrequirements, thusoptimizingenergyconsumptionof the bioethanol production process[5].

Separation is the step where major costs are generated in processindustry [10]. In the case of bioethanol production, distillationprocess may be optimized to reduce steam consumption on columnreboilers, therefore reducing production costs. The distillationprocess commonly employed in Brazilian distilleries is based on thesame conguration used for decades, on which atmospheric pres-sures are adopted; in a new conguration presented in this work,

* Corresponding author. Laboratrio Nacional de Cincia e Tecnologia do Bio-etanol (CTBE), P.O. Box 6170, 13083-970 Campinas, SP, Brazil. Fax: 55 19 3518 3164.

E-mail address:[email protected](M.O.S. Dias).

Contents lists available atScienceDirect

Energy

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m/ l o c a t e / e n e r g y

0360-5442/$e see front matter 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.energy.2010.09.024

Energy xxx (2010) 1e13

Please cite this article in press as: Dias MOS, et al., Improving bioethanol production from sugarcane: evaluation of distillation, thermal inte-gration and cogeneration systems, Energy (2010), doi:10.1016/j.energy.2010.09.024
mailto:[email protected]://www.sciencedirect.com/science/journal/03605442http://www.elsevier.com/locate/energyhttp://dx.doi.org/10.1016/j.energy.2010.09.024http://dx.doi.org/10.1016/j.energy.2010.09.024http://dx.doi.org/10.1016/j.energy.2010.09.024http://dx.doi.org/10.1016/j.energy.2010.09.024http://dx.doi.org/10.1016/j.energy.2010.09.024http://dx.doi.org/10.1016/j.energy.2010.09.024http://www.elsevier.com/locate/energyhttp://www.sciencedirect.com/science/journal/03605442mailto:[email protected]


2/13

wine is fed to distillation columns operating under vacuum pres-sures, where ethanol-rich streams containing from 40 to 50 wt%ethanol areproduced; thesestreams arefed to atmospheric-pressurerectication columns, where hydrous ethanol (approximately 93 wt%) is produced. Reduction of steam consumption on distillationcolumn reboilers is then achieved by the use of a double-effectdistillation system, sincedifferent temperature levels are obtained incolumn condensers and reboilers, allowing the thermal integrationof these equipments[11].

In order to evaluate the consequences of introducing a double-effect distillation system in ethanol production from sugarcane,simulations of bioethanol production processes were carried outusing software UniSim Design from Honeywell. Steam demand wasevaluated for each case (bioethanol production using conventionalor double-effect distillation columns), after thermal integrationusing Pinch Analysis was carried out. The results obtained wereused to assess two different cogeneration systems: Rankine Cycleoperating with steam at high temperature and pressure (480 Cand80 bar) and a BIGCC, on which sugarcane bagasse is gasied. Eventhough the gasication technology is not yet commercially feasible,the possibility of improving electricity production in this caseshould be evaluated. Simulations of the cogeneration systems were

carried out using Gate Cycle and EES (Engineering Equation Solver).Thermoeconomic analyses and determination of exergy-basedcosts for ethanol and electricity were assessed for both cases.

2. Bioethanol production process

In this work an autonomous distillery, on which all sugarcaneprocessed is used to produce ethanol, for the production of onemillion liters of anhydrous ethanol per day was considered. This isa typical industrial scale unit, where around 12,000 ton of sugar-cane (TC) per day are crushed, producing around 85 L of anhydrousethanol per ton of sugarcane.

Simulations were carried out using the commercial softwareUniSim Design from Honeywell.

2.1. Chemical and physical properties

In order to represent sugarcane composition (displayed inTable 1) more accurately, some hypothetic components that are notpart of UniSim database were created as described in a previouswork [12]: bagasse components (cellulose, hemicellulose andlignin); sand, with properties considered equal to those of SiO2;impurities, represented by potassium salts and aconitic acid; inputmaterials such as phosphoric acid and lime; calciumephosphate,themain salt formedduring limingoperation, of greatimportance inremoval of impurities during juice settlement; minerals, repre-sented by K2O [13]; and yeast. Properties for these hypotheticcomponents were obtained in Wooley and Putsche[14]and Perry

and Green[15]. All reducing sugars are considered dextrose; allothercomponents(water, sucrose, ethanol, carbondioxide, glycerol,succinic acid, acetic acid, isoamyl alcohol, sulfuric acid, MEG(monoethyleneglycol)) are part of UniSim database.

NRTL (Nom-Random Two Liquid) was the model chosen forcalculating activitycoefcientoftheliquidphase,andtheequationofstate SRK (SoaveeRedlicheKwong) for the vapor model. It wasveried that the NRTL model was theone that calculated elevation ofthe boiling point of sucrose solutions with larger accuracy, whencompared to UNIQUAC (Universal Quasi Chemical) or Pen-geRobinson equation of state, as represented in Fig. 1. For theextractive distillation process, employed in ethanol dehydration, themodel UNIQUAC and equation of state SRK were used.

2.2. Main aspects of the production process

Production of anhydrous bioethanol from sugarcane iscomprised by the following main steps: reception and cleaning ofsugarcane, extraction of sugars, juice treatment, concentration andsterilization, fermentation, distillation and dehydration. All thesesteps were represented on the simulation [12]. Some processimprovements were considered in this work, such as: for sugarcanecleaning,useofadry-cleaningsystem,insteadofonebasedonwateruse, what decreases sugar losses as well as water consumption;efcient juice treatment, considering addition of phosphoric acid

and lime, what increases juice purity and consequently improvesthe fermentation stage; concentration of juice on multiple effectevaporators, decreasing process steam consumption; decreasedfermentation temperature (28 C), which allows the production ofwine with higher ethanol content, consequently diminishing sugarand ethanol losses, as well as vinasse generation; a double-effectdistillation process, which allows thermal integration betweencolumns condensers and reboilers; optimization of the extractivedistillation process with MEG for anhydrous bioethanol production,considering feed of hydrous ethanol on the vapor phase (in theindustry,hydrousbioethanolisobtainedinacondensedphaseintheconventional distillation process, followed by vaporization prior tothe extractive distillation process, what leads to increased andunnecessary steam consumption). A block-ow diagram of thebioethanol production process is depicted in Fig. 2. Description ofthe different steps of the bioethanol production process fromsugarcane is presented in the following subsections.

2.3. Sugarcane reception and cleaning, extraction of sugars and

juice treatment

A dry-cleaning system is used to clean sugarcane received in thefactory, removing about 70% of dirt before it is fed to the mills.Extraction of sugars is done using mills, where sugarcane juice and

Table 1

Composition of sugarcane received in the factory.

Component Content (wt%)

Sucrose 13.30Fibers 11.92Reducing sugars 0.62Minerals 0.20Impurities 1.79Water 71.57Sand 0.60

0 10 20 30 40 50 60 70 80 90

100

102

104

106

108

110

ExperimentalNRTLPeng RobinsonUNIQUAC)

C(tnioP

gnilioB

Mass fraction of sucrose (%)

Fig.1. Boiling point of sucrose solutions: experimental data [16]and values calculated

using the process simulator with NRTL, Penge

Robinson and UNIQUAC.

M.O.S. Dias et al. / Energy xxx (2010) 1e132



3/13

bagasse are obtained. Water is used to improve sugars recovery, ina process termed imbibition[13]. A physical treatment is used to

remove sand and ber from the juice, in which screens and hydro-cyclones are used[13]. Juice then receives a chemical treatmentconsisting of addition of phosphoric acid, lime, heating and settle-ment to remove other impurities. In the settler, mud and claried

juice are obtained. A lter is used to recover some of the sugarscontainedin the mud, and the ltrateis recycled to theprocess priorto the second heating operation.

2.4. Juice concentration and sterilization

Claried juice contains around 15 wt% solids, so it must beconcentratedbefore fermentation in order for the product to containan adequate ethanol content that allows reduction of energyconsumption during purication steps. Concentration is done ona 5-stage multiple effect evaporators (MEE) up to 65 wt% sucrose.Vapor purges may be done if needed, in order tosupply heatto otheroperations. Only part of the juice is concentrated, and nal juicecontains around 22 wt% sucrose. In order to avoid contamination infermentation, juice is sterilized and cooled till fermentationtemperature (28 C).

2.5. Fermentation

Simulations were based on the Melle-Boinot (feed-batch withcell recycle) process. Sterilized juice is added to the fermentorsalong with yeast stream. In the fermentor sucrose is inverted toglucose and fructose, which are consumed by the yeast producing

ethanol, CO2and other products, such as higher alcohols, organic

acids, glycerol and yeast. Fermentation gases are collected andwashed in an absorber for ethanol recovery, while wine containingyeast cells are centrifuged to recover yeasts. The yeast milk streamreceives an acid treatment prior to be fed back to the fermentor, inorder to avoid contamination. Wine is mixed with alcoholicsolutions from the absorber before being fed to the distillationcolumns.

Fermentation was carried at 28 C, in order to obtain wine withhigher ethanol content (around 10 wt%), decreasing ethanol andsugar losses and energy consumption during the purication step[17]. An alternative cooling method (absorption with lithium-ebromide) was considered to supply cool water to maintain thislow fermentation temperature.

2.6. Distillation and dehydration

The conventional distillation process consists of a distillationcolumn comprised by columns A, A1 and D, and a recticationcolumn comprised by columns B and B1. Wine obtained in fermen-tation is pre-heated and fed to column A1, which is located abovecolumn A and below column D, as shown inFig. 3. Pressure on thedistillation columns range from 133.8 to 152.5 kPa, and on the

rectication columns from 116 to 135.7 kPa.In the distillation columns ethanol-rich streams containing

around 40 wt% ethanol (phlegms) are obtained, as well as vinasseand 2nd grade ethanol. In the rectication column hydrous ethanol

Fig. 2. Block-ow diagram of the bioethanol production process from sugarcane in an

autonomous distillery.

A1

VapourPhlegm

Vinasse

R

R

TopD

Liquidphlegm

R

R

R

D

Top Drecycle

R

Gases

2nd gradeethanol

A

B,B1

HydrousEthanol

Phlegmasse

Fuseloil

Warmwine

Fig. 3. Conventional distillation process con

guration.

M.O.S. Dias et al. / Energy xxx (2010) 1e13 3



4/13

and residues like phlegmasse and fusel oil are obtained. Hydrousethanol in vapor phase is produced on top of column B.

Since water and ethanol form an azeotrope with concentrationof 95.6 wt% ethanol at 1 atm, an extractive separation process withMEG was used to produce anhydrous bioethanol. In this processtwo columns are used: an extractive column, where hydrousethanol and a suitable solvent is used, anhydrous bioethanol isproduced on the top and a solvent solution on the bottom, and

a recovery column, in which extractive solvent is recovered.Extractive column operates at atmospheric pressure (101.325 kPa),

while the recovery column operates at 20 kPa, in order to avoidhigh temperature and solvent decomposition. The solvent is cooledbefore fed to the extractive distillation. Since the bottom temper-atures of both extractive and recovery columns are relatively high(136 and 150 C, respectively), both reboilers need to operate withhigh pressure (6 bar) steam.

A1

Vapourphlegm

BottomA

R

R

LiquidphlegmR

R

R

R

Warmwine

Hydrousethanol.

Fuseloil

PhlegmasseB,B1

Extractive

Recovery

MEG

AnhydrousEthanol

Water

Solvent

R

D

Gas

A

Boilup-A VinasseR

2ndgradeethanol

Fig. 4. Double-effect distillation e integration of column A reboiler to column B and extractive column condensers.

Table 2

Initial and nal temperature, heat ow of streams considered for thermal integration.

Streams Conventional distillation Double-effect distillationHot streams Tinitial

(C)Tnal(C)

Heat ow(kW)

Tinitial(C)

Tnal(C)

Heat ow(kW)

Sterilized juice 130 28 27,606 130 28 27,602Fermented wine 28 24 5440 28 24 5063

Vinasse 112 35 27,426 65 35 10,547Anhydrous ethanol cooling 78 35 9070 78 35 8983Vapor condensates 107 50 10,499 109 50 12,418Condenser column B 82 82 24,297 82 82 23,656Condenser extractive column 78 78 9224 78 78 13,734Condenser column D 81 30 662 41 26 4482Condenser recovery column 82 64 290 78 63 295Cold streams Tinitial(C) Tnal(C) Heat ow (kW) Tinitial(C) Tnal(C) Heat ow

(kW)Raw juice 30 70 21,062 30 70 21,096Limed juice 76 105 20,535 74 105 20,667

Juice for sterilization 96 130 9287 96 130 9309Centrifuged wine 28 82 23,161 28 48 8636Reboiler column A 112 112 38,553 65 65 37,389Reboiler column B 108 108 6725 108 108 5248Reboiler extractive column 111 137 8432 111 137 12,881Reboiler recovery column 150 150 1704 150 150 1794




5/13

In the double-effect distillation system, the distillation columnsoperate under vacuum pressures (19e25 kPa), while the recticationcolumnoperates at atmospheric pressures (101.325e135.7kPa),whatmakes temperature on thebottom of column A reachabout 65 Candtemperature on the top of column B reach 78 C. Consequently,integration of column B condenserand column A reboileris possible,and the rectication column condenser may work as the distillationcolumn reboiler,whatprovides a reduction in steam consumption ondistillation. Nevertheless, since phlegms are obtained in the vaporphase in column A, they must be compressed before being fed tocolumn B, what is done by using steam compressors. Since column A

reboiler duty is greater than that of column B condenser, this inte-gration cannot supply all the heat necessary to the adequate opera-tion of the distillation columns. Given that temperature on top of theextractive column from the extractive distillation process reaches

78 C, it is possible to useanhydrousbioethanolvapors to providetheremaining heat necessary to an adequate operation of column A. Thisintegration in the double-effect distillation is shown inFig. 4.

3. Process integration analysis

The Pinch Point Method described by Linnhoffet al. [18] was usedto analyze the streams of the process which are available for thermalintegration. The method developed by works of Hohmann [19],Umeda et al. [20] and Linnhoff and Flower [21,22] uses ent-halpyetemperature diagrams torepresent theprocessstreams andto

nd the thermal integration target for them, considering a minimumapproach difference of temperature (Dtmin) for the heat exchange.

The thermal integration target indicates the minimumrequirements of external hot and cold utilities for the process. The

Fig. 5. Composite Curves for the conventional distillation case.

Fig. 6. Composite Curves for the double-effect distillation case.




6/13

analysis is useful to represent all the streams available for heatexchange in one diagram, being possible to evaluate the effect ofmodications in the process parameters before the heat exchangenetwork design.

Thermal integration analysis of the bioethanol productionprocess considered the hot and cold streams parameters deter-

mined in the simulation of the autonomous distillery for bothconventional and double-effect distillation. Thermal integrationbetween rectication and distillation columns has an importantimpact on the steam balance, promoting changes in the integrationpossibilities. Data of the streams available for thermal integrationare presented inTable 2.

Some streams with less than 1000 kW of heat ow were notconsidered in the analysis because of their low thermal integrationpotential. The evaporation system was considered as a separatedsystemwhichis integrated with the backgroundprocess in a secondstep. This procedure previouslyapplied by Ensinas[23] in sugarcanemills simplies theanalysis, maximizing thevaporbleeding use andreducing the total exhaust steam consumption in the process.

As it was previously mentioned, in this analysis a single-effectlithium bromide absorption system, with COP 0.65, was consideredto provide the cold utility requirements of the fermentation step,using for that vapor from the third effect of the evaporation systemas heating source to produce the necessary cold water.

Thus, following the procedure presented by Linnhoff et al. [18]the CC (Composite Curve) (Figs. 5 and 6) could be drawn and thetargets determined.Fig. 7shows the GCC (Grand Composite Curve)of double-effect distillation case considering the integration of thebackground with the evaporation system after the thermal inte-gration. The Pinch Point temperature in this case was found at113 C, considering a global Dtminof 10 C for all the streams of thebackground process, and a Dtminof 4 C for the evaporation vaporstreams. The software presented by Elsevier [24] was used forcalculation of the targets and drawing of CC and GCC curves.

4. Cogeneration systems

Traditionally, cogeneration systems employed in Brazilian sugar-cane mills are based on the Rankine Cycle[25]. Sugarcane bagasse isused as a fuel to supplythermal, mechanical andelectricaldemand ofthe sugar and ethanol production process. These plants used tooperatewith steamat lowlevels of pressure andtemperature (20barand 350 C) and back pressure steam turbines, resulting in lowenergetic efciency, high bagasse consumption, low bagasse surplusand a small electricity surplus, or even none. After the Brazilianelectrical crisis in 2001, independent energy producers were allowedto sell electricity to the grid, thus providing the sugarcane industrythe possibility of improving electricity generation from biomass. Asa consequence,new projectsof sugarcane mills considered the useofRankine Cycles with steam at higher levels of temperature andpressure, condensing steam turbines and the use of all the bagassegenerated in the mills as a fuel to produce electricity, selling thesurplus electricity to the grid. In the following sections a descriptionof the cogeneration systems studied in this work is made.

Fig. 7. Grand Composite Curves for the double-effect distillation case (background/

foreground process integration).

Fig. 8. Con

guration of the cogeneration system based on Rankine Cycle: back pressure steam turbines.




7/13

4.1. Traditional cogeneration system in sugarcane mills (Rankine

Cycle)

Figs. 8 and 9 show two basic congurations of the Rankine Cycleused in cogeneration systems traditionally employed in Brazil. Therst conguration (Fig. 8) displays a boiler, a back pressure steamturbine, deaerator and pumps; the second conguration (Fig. 9)shows the same set of equipment; however, a condensing steamturbine is employed, therefore a condenser is added to system. Thecogeneration system provides steam (2.5 and 6 bar of pressure) andelectricity to supply the anhydrous ethanol production process. Theethanol production process receives sugarcane (stream 11 on Fig. 8)producing anhydrous ethanol (stream 12), vinasse (stream 13) andsugarcane bagasse (stream 14), and requires 6 bar (stream 5) and2.5 bar steam (stream 3) besides electrical energy (We). The

bagasse produced in the mills has a moisture content of 50% and isused to supply energy to the cogeneration system (stream 10). 7% ofthe bagasse is separated to provide energy for start-ups (stream 15)and the remaining fraction is available for other uses (stream 16).Sugarcane bagasse ow was calculated using Eq.(1).

mbagasse x

wmcane (1)

where:xis the percentage ofber in sugarcane;wis the moisturecontent of the bagasse. The values used for ber content of sugar-cane and moisture of sugarcane bagasse are 12% and 50%,respectively.

Nowadays, bagasse surplus is used to increase the production ofelectricity, mainly in condensing systems as shown in Fig. 9.However, whenproduction of ethanol from lignocellulosic materials(second generation ethanol) becomes feasible, bagasse surplus maybe used to increase ethanol production from sugarcane, using thesamecultivatedarea throughpretreatment and hydrolysisprocesses.

The use of a Rankine Cycle was assessed through a thermoeco-nomic analysis using software EES; the main parameters used inthe simulation are shown inTable 3.

4.2. Cogeneration systems based on BIGCC (Biomass Integrated

Gasication Combined Cycle) in sugarcane mills

The use of the BIGCC in sugarcane mills has been investigatedfor the past 15 years [26e33]. Other works assess the combined useof biomass gas and natural gas in cogeneration plants, in order to

overcome the problems found in the BIGCC applied in sugarcane

mills[34e39]. The use of gas turbines with biomass is an option as

well[40e46].A conguration of the BIGCC system employed in sugarcanemills is illustrated in Fig. 10. The system consists of a gasier,a dryer, a gas turbine set, an HRSG (Heat Recovery Steam Generator)and back pressure steam turbines that provide steam to supply thedemands of the process at both 2.5 and 6 bar. Similarly to theprevious case, sugarcane bagasse is produced during anhydrousethanol production (stream 10 on Fig.10) and fed to an atmosphericgasier, on which biomass gas production takes place. The biomassgas that leaves the gasier (stream 15) is compressed in a biomassgas compressor and fed to a combustion chamber, on which itreacts with compressed air (stream 2) producing stack gases(stream 3) and supplying power for compressor and generator and,consequently, producing electric energy. Stack gases from the gas

turbine (stream 4) are fed in the HRSG producing steam at 480

Cand 80 bar, which supplies the thermal demand of the ethanolproduction process through steam turbines that produce lowpressure steam (stream 17, which consists of 6 bar steam andstream 19, 2.5 bar steam). The stack gases that leave the HRSG(stream 5) are used to dry the bagasse before it is fed in the gasier.

One of the most important steps of the simulation of BIGCC isthe modeling of the gasier. In order to simulate the behavior of theBIGCC two important parameters must be determined: lowerheating value of biomass gas and adequate moisture content of thebagasse after the drier. In order to obtain these parameters,a computational tool kindly provided by Pellegrini and Oliveira Jr.[47]is employed. The model is based on a study carried out by Fockand Thomsem [48]. The input variables are sugarcane bagassecomposition (47.7% carbon, 5.8% hydrogen and 46.5% oxygen)[49]

Fig. 9. Conguration of the cogeneration system based on Rankine Cycle: condensing steam turbines.

Table 3

Main parameters adopted in the simulation of the Rankine Cycle.

Parameters Value

Sugarcane processed (ton/h) 493Anhydrous ethanol production (L/TC) 85Vinasse production (L/L of anhydrous ethanol) 11Lower heating value of bagasse (kJ/kg) 7500Isentropic efciency of pumps 0.85Isentropic efciency of steam turbines 0.80First law efciency of boiler 0.85Electric power demand in the process (kWh/TC) 28Fraction of bagasse for start-ups (%) 7Efciency of electrical generators 0.98




8/13

and moisture. Gasication was considered to occur in the presenceof atmospheric air. In fact, the best gasication conditions arepresent when a mixture of oxygen and steam is employed;however, costs for separating oxygen from atmospheric air for use

in power generation are prohibitive.Thecompositionofbiomassgasasafunctionofsugarcanebagassemoistureis represented in Fig.11. Theproductsofbagassegasicationare CO, CO2, H2O,CH4, N2and H2. Molar fraction of CH4is consideredconstant and that of the other components vary with bagasse mois-ture. Decreaseof bagassemoisture from50 to40% leadsto anincreaseof the H2fraction; lowermoisture values lead to a decrease of the H2fraction. A different behavioris observed forthe molar fractionof bothCO and N2, which increase when moisture content decreases. Molarfractions of H2OandCO2 decreasealong withmoisture; thisbehavior

affects the lower heating value of biomass gas as well as the gasi-cation temperature, as shown inFig. 12.

Proles of the gasication temperature and lower heating valueof the biomass gas as a function of bagasse moisture are depicted in

Fig. 12. Increase of the CO molar fraction and decrease of watermolar fraction gives rise to an increase on the lower heating valueof biomass gas. In addition, if the temperature is too low, there willnot be enough energy to start up the gasicationprocess. Accordingto Reed and Gaur[50], the gasication process requires a minimaltemperature of approximately 800 C. Thus, the minimal value ofsugarcane bagasse moisture is 25%, as illustrated in Fig. 12.

Sugarcane bagasse produced during sugarcane processing hasa moisture content of 50%. Therefore, in order to be employed in

Fig. 10. Conguration of the cogeneration system based on the BIGCC cycle.

Fig. 11. Biomass gas composition as a function of bagasse moisture. Fig. 12. Lower heating value and biomass gas temperature versus bagasse moisture.




9/13

a gasication cycle, the moisture content of bagasse must bereduced to 25%. The bagasse drying process takes place at the drier(equipment IV depicted inFig.10). The drier makes use of the stackgases from the HRSG (stream 5), and its main parameters aretemperature and composition of the stack gases and the outletminimal temperature of the gases leaving the device (which isassumed to be equal to the stack gases adiabatic saturatedtemperature). The energy balance in the drier is shown in Eq.(2)

m5h5m10h10m6h6m14h14 0 (2)

where:m10mbdmw10,mbdrepresents the dry bagasse ow andmw10 theamountof waterpresent in thebagasse; m14 mbd mw14,mw14 corresponds to the amount of water in the bagasse that leavesthe drier andm6m5mw6, on whichmw6is the amount of waterremoved from the bagasse.

It is convenient to dene the moisture of bagasse throughstreams 10 (bagasse fed to the drier) and 14 (dried bagasse), asshown in Eq.(3).

w10 mw10

m10and w14

mw14m14

(3)

The values adopted in this simulation consider the moisturecontent of the stream 14 equal to 15% [47]. However, since thetemperature (75 C, considered constant) of stream 6 (stack gasesleaving the drier) depends on the composition of the stack gases,and it is necessary to reach 15% moisture on stream 15 (bagasseleaving the gasier), energy must be provided to reach the requiredmoisture level through a convenient value of temperature of thestack gas from the HRSG (stream 5). Usually, the value used in theHRSG is close to 150 C [51]. However, this value is too low to assurea convenient bagasse drying, as shown inFig. 13. Increasing stackgases temperature leads to suitable levels of moisture for gasica-tion; however, decrease of this temperature implies reduction onthe capacity of steam production for the ethanol productionprocess. In order to assure bagasse moisture of 15%, the minimum

temperature of thestack gases must be 215

C for this simulation, asdepicted inFig. 13.Inapreviousstudy,Modestoetal. [52] showed three proposals to

use the BIGCC cycle in ethanol plants. The rst two proposals useda commercial gas turbine (GE LM 2500) adapted to use biomass gasas fuel; however, the thermodynamic conditions adopted could notsupply the steam demand (340 kg/TC) without using a supplemen-tary fuel (natural gas) in the HRSG. The third proposal, whichconsiders a gas turbine designed for biomass gas, allowed theelimination of the supplementary fuel; however, the operatingconditions of the gas turbine penalize the performance of electric

generation. According to Modesto et al. [52], a suitable value ofsteam demand to make a BIGCC cycle feasible is 230 kg/TC. Thus,thermal integration and the use of a double-effect distillationsystem, as described in this work, will improve the feasibility of theBIGCC cycle by not penalizing electric generation.

In order to execute the thermoeconomic analysis of the BIGCCcycle the software Gate Cycle (mass and energy balances) andEES (exergy and exergy cost balances) were used. The mainparameters used in the simulation of BIGCC cycle are shown inTable 4. In this simulation, a gas turbine using only biomass gas issimulated using the Gate Cycle Software, developed by GePower.Palmer et al.[42]provided some suggestions for the use of biomassgas in the operation of this gas turbine. The main parametersanalyzed were pressure ratio and temperature at the outlet of thecombustion chamber. The biomass gas from gasier is compressedin a compressor and enters a combustion chamber along with an airow at the same pressure. Both react in a combustion chamber andthe gases produced expand in a power turbine producing electricalenergy. It is important to mention that the exclusive use of biomassgas in the gas turbine needs a catalytic combustion chamberdesigned to operate with fuels of low heating values. Proposals ofcatalytic combustion chambers can be found at Witton et al. [53]

and Forzatti [54]. Another aspect that needs to be mentioned isthat this study does not consider the energetic consumption of animportant step of the gasication process, the cleaning of gases thatleave the gasier. The cleaning of stack gases is a very importantstep to allow the use of these gases in a gas turbine, in order toassure suitable levels of impurities to avoid corrosion and scorchingin the blades of the turbine.

5. Exergetic cost analysis

In order to assess the two cogeneration systems proposals, anexergetic cost analysis was performed. This analysis employs mass,energy, exergy and exergy costs balances in all components of the

plant. Thus, it is possible to determine energetic consumption,irreversibility generation and the exergetic cost of the products,through a convenient cost allocation method.

Eqs. (4)e(6) show mass, energy and exergy balances fora generic control volume, respectively, according to Kotas[55].

X _min

X _mout 0 (4)

_Q _WX

_minhinX

_mouthout 0 (5)

_Q

1

T

To

_W

X _minein

X _mouteout _I (6)

In order to determine steam, stack gases, air, biomass gas and

water exergy, Eq.(7)was used

051

071

091

012

032

09

011

031

5045403530252015

Stackgasestemperaturefrom

HRSG

(C

)

)%(erutsiomessagaB

Fig. 13. Bagasse moisture as a function of stack gases temperature from HRSG.

Table 4

Parameters used in simulations of BIGCC cycle.

Parameter Value

Production of biomass gas (kg/kg of bagasse) 2.23Lower heating value (kJ/kg) 5100Pressure ratio 18Isentropic efciency of compressors 0.85Isentropic efciency of power turbine 0.92Isentropic efciency of steam turbine 0.9Isentropic efciency of pumps 0.85Temperature of gases leaving the HRSG (C) 215Temperature of gases leaving the drier (C) 75




10/13

e hihoTosiso e0ch (7)

where: hi enthalpy at point i; ho enthalpy of reference;si entropy at point i,so entropy of reference; ech chemicalexergy of water. The values used for the temperature and pressureof reference are 25 C and 1 bar, respectively. The chemical exergyvalues are determined by Szargut et al. [56].

The exergy values of sugarcane bagasse were calculated by themethodology described in Sosa-Arnao and Nebra [57]; exergy ofthe ethanolewater mixture employed the methodology describedin Modesto et al.[58].

The methodology used to perform the exergetic cost analysisis the Theory of Exergetic Cost proposed by Lozano and Valero[59]. This methodology can be used to determine the exergeticand monetary cost of each ow that compose the system. Theevaluation of the exergetic and monetary costs of a cogenerationsystem in a sugar plant evaluating the inuence of the price ofthe main fuel (sugarcane bagasse) in steam production andelectricity costs was presented by Sanchez and Nebra [60]. The

use of the thermoeconomic methodology to assess the exergeticcost of sugar in the production process was described by Fer-nandez Parra[49].

The exergetic cost calculation is made through cost balanceequations in each component, as shown by Eq. (8).

XkinEin

XkoutEout 0 (8)

where: kdenes the exergy-based cost and Ethe total exergyow. The subscript inand outindicate the ows that enter andleave the control volume, respectively.

The application of Eq.(8)in all control volumes forms a linearequations set, where the number of variables is greater than the

number of equations. In order to obtain a set with a unique solutionit is necessary to add some additional equations. Cerqueira andNebra[61]reported in a simple way the postulates of the meth-odology to dene these additional equations.

In the case of anhydrous ethanol production process, the exer-getic cost balance equation is written by Eq. (9) (for the RankineCycle) and Eq.(10)(BIGCC):

_m11e11k11 _m3e3k3 _m5e5k5Weke _m12e12k12

_m13e13k13 _m14e14k14 0 9

_m11e11k11 _m17e17k17 _m19e19k19Weke _m7e7k7

_m12e12k12 _m13e13k13 _m14e14k14 0 10

The set of additional equations were included followingthe considerations proposed by Lozano and Valero [59]. For theexergy-based costs of the inputs (sugarcane) a unitary value isassigned, and therefore

k11

1 (10a)

In BIGCC, stack gases that leave the drier have the exergy-basedcost null

k6 0 (11)

All the irreversibility generation in the turbines must be carriedout by the exergy-based cost of electric power, and consequentlythe exergy-based costs of the steam entering and leaving theseturbines are considered equal. Therefore, Eq.(12)(Rankine Cycle)and Eq.(13)(BIGCC) are obtained

k1 k5 k2 k17 k18 (12)

k3 k4 and k9 k17 and k18 k19 (13)

In the splitters, where no irreversibility generation takes place,ows entering and leaving the valves have the same exergy-basedcost. Thus, Eqs. (14) and (15) areobtained, for the Rankine Cycle andBIGCC, respectively.

k2 k3 k4; k4 k10 k15 k16 (14)

k8 k9 k18 (15)

In the BIGCC, the exergy-based cost of stack gases that enter andleave the HRSG (streams 4 and 5, respectively), is the same. Thus, allthe irreversibility generated in the HRSG is carried by exergy-basedcost of steam produced by HRSG (stream 8)

k4 k5 (16)In the anhydrous ethanol production process, the following

considerations were made:

Table 5

Hot utilities demand before and after thermal integration for the studied congurations (A: conventional distillationwithout thermal integration; B: double-effect distillationwithout thermal integration; C: conventional distillation with thermal integration; D: double-effect distillation with thermal integration).

Operation Saturated steam consumption (kg/h)

Conguration A Conguration B Conguration C Conguration D

2.5 bar 6 bar 2.5 bar 6 bar 2.5 bar 6 bar 2.5 bar 6 bar

1st juice heating 34,632 e 34,687 e e e e e

2nd juice heating 33,765 e 33,981 e e e e eMulti-effect evaporation 56,270 e 60,706 e 59,254 e 68,519 eJuice sterilization e 15,843 e 15,916 4056 4711 5427 4729Reboiler column A 63,390 e e e 63,390 e e eReboiler column B 11,058 e 8629 e 11,058 8629 eReboiler extractive column e 14,426 e 22,039 e 14,426 e 22,039Reboiler recovery column e 2915 e 3069 e 2915 e 3069Total 199,115 33,184 138,003 41,024 137,758 22,052 82,575 29,837

Table 6

Thermal and electric energy demand of anhydrous ethanol production process.

Conguration 2.5 bar steamconsumption(kg/s)

6.0 bar steamconsumption(kg/s)

Electric energyconsumption(kW)

SteamDemand(kg/TC)

A 55.31 9.22 0 468B 38.33 4.43 4328 310C 38.26 6.12 0 322D 22.94 8.29 4328 226




11/13

i) The exergy-based cost of the steam that enters the process isthe same of the condensate obtained after heat exchange:For the Rankine Cycle:

k3 k5 k6 (17)

For the BIGCC cycle:

k17 k18 k7 (18)

ii) The exergy-based cost of bagasseand vinasse (streams 14 and13, respectively) is the same as that of the sugarcane (stream11) that enters the process. Thus, all the irreversibilitygeneration is carried by the exergy-based cost of the ethanol.As a result:

k12 k14 k11 (19)

Thus, with the set equations above, the number of equations isequal to the number of variables. The system was solved using theEES software.

6. Results and discussions

6.1. Process utilities requirements

After integration of the background process and the distributionof the vapor bleeding, the hot utilities requirements could bedetermined for each case. Four different congurations for theanhydrous ethanol production process steam demand wereanalyzed, considering conventional and double-effect distillationwith and without thermal integration. Two levels of steam pressurewere considered for hot utilities, derived from extractions at 6.0 barand 2.5 bar of pressure in the steam turbine. Steam consumption ofthe analyzed cases is shown in Table 5, which summarizes the nal

results of hot utilities demand for each case.As can be seen in Table 5, the thermal integration promoted

important steam reductions in the anhydrous bioethanol produc-tion. For the mills with conventional distillation system scheme thereduction of the steam consumption at 2.5 bar can reach 31% withthe integration and 34% for the 6.0 bar steam.

When double-effect distillation columns are used in the millsthe reduction of steam consumptionwith the thermal integration is40% for the 2.5 bar steam and 27% for the 6.0 bar steam.

6.2. Evaluation of the cogeneration systems

Information regarding steam consumption, gathered fromTable

5, and electric energy demand from the anhydrous ethanol process

are shown in Table 6. Exergetic costs and summarized results of theevaluation are showed inTable 7.

The exergetic cost analysis in the Rankine Cycle with backpressure steam turbine shows that the decrease of steam demand,through the use of thermal integration and double-effect distilla-tion (conguration D), leads to an increase of bagasse surplus;however, lower electricity surplus is produced, mainly due to the

increase on electricity consumption of compressors in the double-effect distillation system as well as to decreased steam consump-tion. In fact, this conguration is the one that makes the best use ofbagasse for other applications, such as bagasse hydrolysis forsecond generation bioethanol production: the exergy-based cost ofprocess steam and electricity remains constant, while exergy-basedcost of ethanol decreases due to the lower steam demand of theprocess.

The case on which the Rankine Cycle is employed along withcondensing steam turbines shows higher surplus electricity thanthe previous system. The possibility to use all the bagasse availableto generate electricity allows the production of an electricitysurplus of 79.61 kWh/TC, which corresponds to an increase of about27%. However, the exergy-based costs are different than those ofthe conventional cogeneration system used in Brazil: for electricity,they are 5% higher, while for ethanol they are 0.5% lower.

The exergetic cost analysis of the BIGCC cycle was performedonly on conguration D, since this conguration has the loweststeam demand (226 kg/TC), a value lower than the minimal(230 kg/TC) foreseen by Modesto et al. [52]. The main difference ofthe BIGCC cycle is the expressive electricity surplus produced in thisconguration. The value of 144.3 kWh/TC is 130% higher thantraditional Rankine Cycle with condensing steam turbines. Inaddition, the exergy-based cost of electricity and ethanol is lower(56 and 19%, respectively) than those of the traditional congura-tion. Thus, the use of the BIGCC cycle leads to a higher electricitysurplus with low production exergy-based costs of electricity andethanol. Composition of costs of ethanol is found in Modesto et al.[62]. All results are shown inTable 7.

7. Conclusions

A study of the bioethanol process modeling and thermal inte-gration was described in detail in this paper. The use of thecommercial software UniSim Design as a modeling tool proved tobe useful for analysis of the energy integration opportunities,particularly considering different distillation systems.

The thermal integration of the rectication and distillationcolumns promoted important energy savings in the bioethanolprocess as a whole. The analysis of the integrated plant witha double-effect distillation system showed that this alternativepresents a lower process steam demand, when compared with

traditional distillation schemes. On the other hand, it requires

Table 7

Results of different congurations of cogeneration systems (STBP: Rankine Cycle with back pressure steam turbine; STCOND: Rankine Cycle with condensing steam turbine;BIGCC: Biomass Integrated Gasication Combined Cycle).

Cogeneration System Process conguration Bagasse surplus (%) Electricity surplus (kWh/TC) Exergy-based cost

Steam Electricity Ethanol

STBP A 5.00 62.66 3.300 3.823 2.018STBP B 37.05 23.86 3.300 3.824 1.861STBP C 34.65 34.53 3.300 3.823 1.875

STBP D 54.03 6.23 3.300 3.817 1.790STCOND A 0 62.62 3.402 3.841 2.238STCOND B 0 71.74 3.536 3.966 2.093STCOND C 0 78.8 3.525 3.956 2.017STCOND D 0 79.61 3.611 4.036 2.007BIGCC D 0 144.3 1.833 1.676 1.806




12/13


13/13

optimization, simulation and environmental impact of energy systems, vol. 1.Padova: Sevizi Graci Editoriali; 2007. p. 829e38.

[39] Walter A, Llagostera J, Oliveira Jr S, Pellegrini LF. Gas turbine integration tosugarcane trash boilers: comparative energy analysis. In: The 20th interna-tional conference on efciency, costs, optimization, simulation and environ-mental impact of energy systems, vol. 1. Padova: Sevizi Graci Editoriali;2007. p. 801e8.

[40] Walter AC, Llagostera J. Exergetic and thermoeconomic analysis ofbiomass integrated gasier/gas turbine power cycles. In: Proceeding ofthe ASME COGEN-turbo power conference. Vienna, Austria; Aug 23 e25

1995.[41] Walter A, Llagostera J. Off-design operation of cored combined cycles. In:

Proceedings ECOS 2003 e the 16th international conference on efciency,costs, optimization, simulation and environmental impact of energy systems.Copenhagen, Denmark; June 30e2 July 2003. p. 1443e50.

[42] Palmer CA, Erbes MR, Pechtl PA. Gate cycle performance analysis of theLM2500 gas turbine utilizing low heating values. In: Proceedings of IGTI ASMECogen-Turbo; 1993.

[43] Kapat JS, Agrawal AK, Yang T. Air extraction in a gas turbine for integratedgasication combined cycle (IGCC): experiments and analysis. Journal ofEngineering for Gas Turbines and Power 1997;119(1):20e6.

[44] Korobitsyn MA, Jellema P, Hirs GG. Possibilities for gas turbine and wasteincinerator integration. Energy 1999;24:783e93.

[45] Ferreira SB, Nascimento MAR, Pilidis P. The use of biomass fuels in gas turbinecombined cycles: gasication vs externally red cycle. In: III world climate &energy event, vol. 1. Rio de Janeiro, Brazil; 2003. p. 1e8.

[46] Nascimento MAR, Ferreira SB, Pilidis P. A comparison of different gas turbineconcepts using biomass fuel. New Orleans: The American Society ofMechanical Engineers; 2001.

[47] Pellegrini LF, Oliveira Jr. S. Exergy analysis of sugarcane bagasse gasication.In: Proceedings of ECOS 2005-the 18th international conference on efciency,costs, optimization, simulation and environmental impact of energy systems.Trondheim, Norway; June 20e22 2005. p. 393e400.

[48] Fock F, Thomsen KBP. Modelling a biomass gasication system by means ofEES. Kgs. Lyngby, Denmark: The Scandinavian Simulation Society, TechnicalUniversity of Denmark; 2000.

[49] Fernandez Parra MI. Exergoeconomic methodology in sugar process [Meto-dologia de analise exergoeconomica do processo de fabricao de acar]. PhD

Thesis: School of Mechanical Engineering, State University of Campinas; 2003[in Portuguese].

[50] Reed TB, Gaur S. A survey of biomass gasication 2000. The NationalRenewable Laboratory and The Biomass Energy Foundation; 2001.

[51] Kehlhofer R, Bachmann R, Nielsen H, Warner J. Combined cycle gas &steam turbine power plant. Tulsa, Oklahoma: PennWell PublhishingCompany; 1999.

[52] ModestoM, Nascimento MAR,Nebra SA.Proposals forBIGCC cycle utilization inan ethanol production plant from sugarcane. In: ECOS 2007 ethe 20th inter-national conference on efciency, costs, optimization, simulation and envi-

ronmental impact of energy systems.Padova,Italy;25e28June2007.p.761e8.[53] Witton JJ, Noordally E, Przybylski JM. Clean catalytic combustion of low heatvalue fuels from gasication process. Chemical Engineering Journal 2003;91(2e3):115e21.

[54] Forzatti P. Status and perspectives of catalytic combustion for gas turbine.Catalysis Today 2003;83(1e4):3e18.

[55] Kotas TJ. The exergy method of thermal plant analysis. Malabar Florida:Krieger Publishing Cobarny; 1995.

[56] Szargut J, Morris DR, Steward FR. Exergy analysis of thermal, chemical andmetallurgical process. New York, USA: Hemisphere Publishing Co.; 1988.

[57] Sosa-Arnao JH, Nebra SA. Exergy of sugarcane bagasse. In: Proceedings of 14thEuropean biomass conference & exhibition. biomass for energy, industry andclimate protection, 17e21 October 2005. Paris, France; 1995.

[58] Modesto M, Nebra SA, Zemp RJ. A proposal to calculate the exergy of non idealmixtures ethanolewater using properties of excess. In: Proceedings of the14th European conference and exhibition: biomass for energy, industry andclimate protection. Paris, France; 17e21 October 2005.

[59] Lozano MA, Valero A. Theory of the exergetic cost. Energy 1993;18(9):939e60.

[60] Sanchez MG, Nebra SA. Thermoeconomic analysis of a cogeneration system ofa sugar mill plant. In: Proceeding of ECOS 2002 e 15th international confer-ence on efciency, costs, optimization, simulation and environmental impactof energy systems, vol. I. Berlin, Germany; 2002, p. 258 e7.

[61] Cerqueira SAAG, Nebra SA. Cost attribution methodologies in cogenerationsystems. Energy Conversion and Management 1999;40(15e16):1587e97.

[62] Modesto M, Zemp RJ, Nebra SA. Ethanol production from sugarcane: assess ofpossibilities of decrease of thermal energy consumption through exergeticcost analysis. Heat Transfer Engineering 2009;30(4):272e81.


Pl it thi ti l i Di MOS t l I i bi th l d ti f l ti f di till ti th l i t

creșterea productiei de bioetanol din trestie de zahar prin cogenerare.pdf

Documents