ARTICLE IN PRESS

0360-5442/$ - se

doi:10.1016/j.en

�CorrespondE-mail addr

Energy 32 (2007) 549–567

www.elsevier.com/locate/energy

CO2 abatement by co-firing of natural gas andbiomass-derived gas in a gas turbine

Daniele Fiaschi�, Riccardo Carta

Dipartimento di Energetica ‘‘Sergio Stecco’’, University of Florence, Via C. Lombroso, 6/17, 50134 Firenze, Italy

Received 1 November 2005

Abstract

In this work, a possible way for partial CO2 emissions reduction from gas turbine exhausts by co-firing with biomass is investigated.

The basic principle is the recirculation of a fraction of the exhausts (still rich in oxygen) to a gasifier, in order to produce syngas to mix

with natural gas fuel. As biomass is a CO2 neutral fuel, the fraction of replaced natural gas is a measure of CO2 removal potential of the

powerplant.

The investigated solution considers the conversion of solid fuel to a gaseous fuel into an atmospheric gasifier, which is blown with a

recirculated fraction of hot gas turbine exhausts, typically still rich in air. In this way, the heat content of the exhausts may be exploited to

partially sustain the gasification section.

The produced syngas, after the tar removal into the high temperature cracker, is thus sent to the cooling section, consisting of three

main components: (I) gas turbine recuperator, (II) heat recovery steam generator and (III) condensing heat exchanger to cool down the

syngas close to the environmental temperature before the subsequent recompression and mixing with natural gas fuel into the

combustion chamber. The water stream produced within the condensing heat exchanger upstream the syngas compression is vaporised

and sent back to the gasifier.

If very limited modification to the existing gas turbine has to be applied in order to keep the additional costs limited, only a relatively

reduced fraction of the low calorific value syngas may be mixed with natural gas. The analysis at different levels of co-firing has shown

that no appreciable redesign has to be applied to the target GE5 machine up to 25–30% (heat rate based) renewable fraction. With an

accurate heat recovery from the cooling/cleaning system of the syngas, the same levels of efficiency of the original machine have been

achieved, in spite of the relatively large power consumption of the syngas recompression. Very interesting results have been obtained

within the 10–30% range of biomass co-firing, with CO2 removal levels between 30% and 50% with reference to the values of the base

GE5 gas turbine powerplant.

The economic analysis has shown that, in spite of the high investment required for the syngas fuel production chain (gasifier, coolers,

cleaners and fuel compressor), approximately at the same level of gas turbine itself, there is an interesting attractiveness due to the

possibility of selling high-value green certificates and CO2 allowances, which reduce the payback time to 2–4 years.

The uncertainty on the calculated economic parameters are greatly influenced by the uncertainty on actual biomass availability and

yearly working time of powerplant, whereas off design operation, which affects mainly the uncertainty of compressor and turbine

efficiency, is mainly reflected on the uncertainty of electric power output and efficiency.

r 2006 Elsevier Ltd. All rights reserved.

Keywords: Gasification; Co-firing; Gas turbine; Biomass; Green certificates; CO2 emissions reduction

1. Introduction

The Kyoto Protocol subscription led many countries tomake efforts toward the research and proposal of systems

e front matter r 2006 Elsevier Ltd. All rights reserved.

ergy.2006.07.026

ing author. Tel.: +39055 4796776; fax: +39 055 4224137.

ess: danif@de.unifi.it (D. Fiaschi).

and techniques for CO2 capture and sequestration frompowerplants. Most of these studies investigate the field ofmassive CO2 emissions capture (80% plus), by applying preor post combustion technologies [1–11]. The proposedsolutions often imply relevant changes in existing turbo-machinery equipment, which are highly expensive andgenerally discourage the electricity providers in taking

ARTICLE IN PRESS

Nomenclature

AFst actual air/fuel ratioAFact stoichiometric air/fuel ratioaGT sound speed at inlet turbine (m/s)a moles of H per mole of biomassb moles of O per mole of biomassDFGE5 variation of dimensional flow coefficient at

turbine inlet relative to the nominal GE5 designvalue

ZGT GT efficiencyZEL net electric efficiencyZfilm film cooling effectivenessZGEN electric generator efficiencyCquota price of CO2 allowance per ton of CO2 avoidedER equivalence ratio ¼ AFact/AFst

eH blade cooling effectivenessf fraction of exhausts recirculated to the gasifierFGE5 nominal dimensional flow coefficient of the

target GE5 gas turbine (m2)FGT dimensional mass flow coefficient at turbine

inlet (m2)Fren fraction of gas turbine heat rate provided with

syngas fuelGC price of green certificates (h/kWh)Ggreen yearly incomes from sale of green certificates

(h)GCO2

yearly incomes (or missed outcomes) due toCO2 quota (h)

GCH4yearly savings, over the basic GT, due toreduced natural gas consumption (h)

HP total energy of products (kW)HR total energy of reactants (kW)

hY yearly number of working hours (years)LHV generic lower heating value (kJ/kg)LHVCC lower heating value of fuel mixture at combus-

tion chamber inlet (kJ/kg)LHVsyn lower heating value of syngas (kJ/kg)LpGAS relative pressure loss (Dp/p) at the gasifier+tar

crackerLpHRSG relative pressure loss (Dp/p) at the heat recovery

steam generator (HRSG)+baghouse filterLpREC relative pressure loss (Dp/p) at the GT recup-

eratorLpSEP relative pressure loss (Dp/p) at the condensing

heat exchanger (SEP)+scrubbermbio biomass flowrate (kg/s)mIG mass flowrate of exhausts recirculated to the

gasifier (kg/s)mGT inlet turbine mass flowrate (kg/s)msteam mass flow of reinjected steam into the gasifier

(kg/s)msyn syngas flowrate (kg/s)PBT payback time (years)PR compressor pressure ratioQ heat added or released to gasifier (kW)R powerplant size scale coefficientrGT gas density at turbine inlet (kg/m3)T0 reference temperature (K)Tg gasification temperature (K)Tin gasifier inlet temperature (K)TOSEP syngas temperature at the SEP outlet (K)Wcmain main compressor power consumption (kW)Wcsyn power demand for syngas fuel recompression

(kW)

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567550

some measures against CO2 emissions [10,11]. Even thecheapest solutions, aimed to reduce, as far as possible, theredesign of existing equipment (i.e. semi-closed gas turbinecycles SCGT), have shown additional costs of electricityaround 60–70% compared to traditional layouts with noCO2 removal [12,13], even if most recent studies promise toreach the low 30–40% levels [14]. Further unknownsconnected to costs of transportation and storage of liquidcompressed CO2 and the related environmental safetymade application of CO2 capture systems unattended [15].

It is well known that renewable fuel sources have, globally,zero CO2 emissions to the environment, thus they might havean interesting potential to approach the greenhouse issue. Onthe other hand, their extensive application to existingpowerplants often involves deep and expensive modificationsto current technologies. The most mature are those involvingIGCC, the largest fraction of which are currently fed withcoal but might be converted (at least partially) to biomassfuels with no appreciable changes in equipment. They are,anyway, applicable in the field of large power generation, ofthe order of few hundred MW electric power [16].

The upcoming 2006 CO2 emissions trading into theEuropean Community should encourage all Member Statesto provide substantial investments for tackling CO2

emissions. The market of CO2 allowances is planned tostart by April 2006. Companies covered by the EmissionsTrading Scheme need to record and report their CO2

emissions as of January 2005. They also need to deliver forthe first time in April 2006 a sufficient number ofallowances to cover emissions during 2005. If a companydelivers no allowances—or not enough allowances—asanction of h40 per non-delivered allowance will beimposed by the Member State. In this way, the adoptionof systems with even partial CO2 emissions reductionpotential (15–50%) might lead to a consistent reduction inelectricity production costs and encourage companies tomake investments for CO2 abatement. The partial integra-tion of biomass fuels with natural gas implies a propor-tional reduction of CO2 emissions, playing, globally, animportant role if its application was extended to severalpowerplants. The co-firing of gas turbines with natural gasand biomass-derived fuel has been a largely investigated

ARTICLE IN PRESSD. Fiaschi, R. Carta / Energy 32 (2007) 549–567 551

issue in the last decades [16–18]. Generally speaking, thebiomass integrated gasification/gas turbine (BIG–GT) hasthe potential of improving the efficiency of electricityproduction from external combustion of steam power-plants (ranging from 15% to 30%), approaching the 40%level with atmospheric gasifiers at reduced scale [19–21].One of the main problems connected with couplingbiomass gasification and gas turbines is the low heatingvalue of the produced gas (typically 10–25% of the naturalgas heating value, which is the target level in design of mostcurrent gas turbines [16]), which leads to an high increase ininlet turbine flowrate at fixed combustion chamber outlettemperature. When current production machines areconsidered, this increase in mass flowrate raises thecompressor pressure ratio, pushing it toward the surgeline. Pressure ratios higher than 4–5% of the design valueare not acceptable, thus some measures must be adopted,such as partial closing of the compressor’s IGV, bleedingair from the compressor or derating the power cycle byreducing the combustor outlet temperature. The last one isthe cheapest option, but leads to a noticeable reduction ofGT performance [22,23]. The other option is a deep turbineredesign, at the price of a great economic effort. If theintegration of fuel with renewables is not so high, reducedoff design problems on existing turbomachinery equipmentmay be achieved and limited additional costs would beexpectable. In fact, an adaptation of the compressor—turbine system by the means of reduction of turbine inlettemperature by 20–30 1C might be a reasonable option at arelatively limited efficiency price [22,23]. Co-firing of gasturbines with biomass-derived syngas is generally consid-ered, at present, the most promising option for a consistentenhancement of biomass use in electricity generation [23].The use of biomass-based fuels as a mean for mitigation ofCO2 release to the environment has been investigatedrecently for a large-scale IGCC, showing the potential forefficient power generation and negative CO2 emissionsbalance at a relatively competitive investment cost [24]. Theapplication of LCA even showed the environmentaladvantages of biomass utilisation for reduction ofboth greenhouse gas emissions and natural resourcedepletion [25].

The present proposal aims to apply not very complexsolutions for reduction of CO2 emissions from powerplantsby co-firing with biomass-derived fuel. The basic idea is thepossibility of an ‘‘internal’’ capture of CO2 emissions, bythe integration of biomass fuel with natural gas to a currentproduction gas turbine: a small fraction of the exhausts,typically rich in fresh air, is recycled and used as theoxidiser of a gasification process, which should produce theamount of gas to replace the saved fraction of natural gas.The recirculation of a fraction of the exhausts to thegasifier is advantageous because it is a hot stream, whichgives an important contribution to sustain the endothermicgasification reactions, thus helping the conversion ratio ofthe solid fuel. Moreover, recirculation represents a sort of‘‘internal recycling’’ of a fraction of produced CO2. In fact,

an important feature of the proposed powerplant over thepower cycles with CO2 capture upstream or downstreamcombustion is that no additional recompression and/ordisposal of the CO2 is needed. The objective of this work isto assess the chances of applying this technique to remove afraction of the CO2 emissions from the traditional naturalgas-fuelled GT cycle without recurring to redesign of themain components and without the need of devices for CO2

concentration and managing. In this way, even if arelatively small amount of CO2 may be avoided (typicallyfrom 10% to 30% with minor modifications to the gasturbine), such solutions might become attractive, in theshort to medium term, due to their rather limitedadditional cost and encouraged by the application of CO2

emissions trading and green certificates. These kinds ofsolutions, with only limited potential CO2 emissionsreduction, deserve special attention even in the view ofKyoto Protocol agreement signed by several Europeancountries (for example, the objective for Italy is thereduction of CO2 emissions by 7% related to 1990s value).

2. Description of the power cycle

The most of BIG–GT powerplant proposals and therelated studies are based on integrated gasificationcombined cycles (IGCC), that represent the best solutionfor efficiency (44–48% efficiency levels are achieved[16–18,24,25]). Unfortunately, they become economicallyaffordable only in large sizes (50MWe plus) and theapplication of co-firing to an existing fossil-fuelled power-plant is, often, economically discouraging. What weinvestigate here is the possibility of applying an economicsolution, which is able to convert an existing gas turbinepowerplant, designed to be natural gas fuelled, into a co-fired powerplant. This should be achieved by the addition,to the existing installation, of a gasifier for the conversionof solid to gaseous fuel followed by the related syngascleaning equipment. Here, this is not done in theconventional way for gas turbine, which is applying apressurised gasifier fed by the main compressor. Anatmospheric gasifier is considered, where the endothermicreduction reactions are sustained by partial combustion ofbiomass fuel, which accomplished through recirculation ofa fraction of the gas turbine exhausts, having typically astill large air content. In this way, two advantages areachieved: (I) the heat content of the recirculated exhausts isuseful for sustaining (partially) the gasification reactionsand increase the higher heating value of syngas [26] and (II)the CO2 and vapour content of the exhausts (normally notpresent in air-blown reactors) should favour tri-reformingreactions, thus enhancing the conversion of C to COand H2 [27].The basic scheme of the proposed power cycles is shown

on Fig. 1. The biomass-derived fuel (syngas) comes fromgasification process. Rather than blowing environmentalair as gasification agent, a fraction of the gas turbineexhausts, having a still large air content, is recirculated into

ARTICLE IN PRESS

Fig. 1. Schematic of co-fired power cycle and main working data.

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567552

the gasifier. In this way, as above discussed, the heatcontent is exploited to partially sustain the endothermicgasification reactions. Moreover, the CO2 of the exhaustshas a beneficial effect on the gasification of carbon. Thesyngas coming out from the gasifier is thus cooled downinto three different sections:

1.

In the first one (recuperator at high temperature) theheat is transferred to the compressed air of the GT,upstream the combustion chamber. To do this, agas–gas heat exchanger must be adopted, which is,typically, a large and relatively expensive component. Inthis case, the low mass flowrate of the syngas shouldallow a limited size anyway (the warming of compressedair is relatively modest, see values of temperatureincrease on air side in Fig. 1).

2.

In the second one, a further cooling is done into a heatrecovery steam generator (HRSG) and superheatedatmospheric steam is produced. The necessary watercomes from the third syngas cooling step.

3.

Finally, the syngas is further cooled down into thecondensing heat exchanger (SEP in Fig. 1), near toenvironmental temperature, which is below the dewpoint of the gas mixture. This operation leads to thecondensation of some water, which is vaporised againinto the upstream HRSG and sent back to the gasifier.The water condensation allows about 10% reduction of

the syngas mass flowrate sent to the syngas compressor,thus the related power demand is reduced and thehydrogen content of the biomass-derived fuel is in-creased. The heat released at the SEP might be used forcogeneration purposes, but its contribution to theimprovement of plant performance has not beenaccounted here.

For the proposed power cycle, heat recovery on thesyngas line is fundamental to avoid dramatic drops inefficiency, thus the tar removal from hot gas is a crucialissue. The possibility of achieving satisfactory levels of tarremoval by high-temperature cracking using dolomite as acatalyst is proven and has reached commercial level attemperature ranging between 800 and 1000 1C [16,17,33],like those found at the gasifier outlet here proposed (seeFig. 1). It may be even integrated to the gasification processitself. Recent studies [34] have shown the interestingpotential of applying nickel-based catalytic filters tointegrated high-temperature removal of tars and particlesfrom biomass-derived gaseous fuels. After the tar cracking,the final gas cleaning is done at low-temperature withbaghouse filter and water scrubbing, downstream theHRSG. Actually, water scrubbing would not be requiredif the final cleaning of gas was done into the SEP, whichworks under syngas saturated conditions. Anyway, giventhe uncertainty related to the application of this rather

ARTICLE IN PRESSD. Fiaschi, R. Carta / Energy 32 (2007) 549–567 553

unconventional component for final gas cleaning, a classicscrubber has been added too.

These technologies should be able to deal with variousbiomass fuels with different properties and degrees ofcontamination. Moreover, most parts of the system havebeen proven commercially [33], thus they are attractive inthe present proposal where control of additional costs is amajor issue. If tar removal at cracker is not satisfactory,possible problems related to their condensation might ariseat the sections where temperature is below 400–500 1C (i.e.at HRSG, see Fig. 1). On the other hand, the eventualresidual tars may be completely removed downstream theSEP, where the saturation conditions favour the coales-cence of particulate and residual tars into the followingscrubber [16]. Finally, if the whole amount of SEPcondensed water is sent back to the gasifier, severalproblems related to the treatment of contaminated waste-water may be reduced.

Fig. 2 shows the main performance data of the referencetarget gas turbine (GE5) working with simple Joule cycle.The reference catalogue data of the target gas turbine (GE5), taken from GE Power site [35], are reported on Table 1.They have been evaluated by tuning unknown parameters(such as turbine and compressor polytropic efficiencies,blade and film cooling efficiency and maximum cycletemperature) in order to match the catalogue data (i.e.exhausts mass flowrate mexh and temperature Texh, heatrate and design pressure ratio).

As it is evident from data shown on Fig. 1, thepowerplant efficiency might be further improved ifthe large heat content of hot exhausts was recovered into

19.59 [kg/s]

2.35

Wcmain = 7787 [kW

0.3587 [kg/s

ηpc = -0.8

288 [K]

718.7 [K]

N2c = 0.7504

O2c = 0.23

CO2c = 0.01279

H2Oc = 0.0063

1.013 [bar]

ηEL = 0.3066

WEL = 5500 [kW]

288 [K

HEAT rate = 11740 [kJ/kWh]

CO2Spec.emiss. = 816.9 [g/kWh]

coolrat

Mass composition

PR= 14.8 14.99 [bar]

Fig. 2. Schematic of target GE5 gas

the heat recovery steam generator of a bottoming steamcycle or, either, to increase the heat recovery on thecompressed air upstream the combustion chamber. Thispossibilities have not been considered here, as the maingoal of this work is to assess the effects of co-firing withpartial recirculation of the exhausts applied to gas turbinealone. Additional expensive components such as bottomingsteam cycle and HRSG, which investment costs oftenrequire large power outputs to be paid off within reason-able time, have not been considered in this analysis.Without entailed CO2 removal systems, the saved CO2

emission with respect to a standard gas turbine cyclecorresponds to that which would be emitted by thereplaced natural gas amount. In fact, the energy produc-tion from renewables like biomass, leads to globally zeroCO2 emissions to the environment. The addition of adownstream CO2 removal system might further increasethe CO2 abatement, but it has not been investigated here.The analysis has been carried out referring to a biogas/

natural gas integration ranging from 10% to 100%. Theupper limit is very hard to exceed due to the extremely lowsyngas heating value, so that the gas turbine wouldundergo serious off design problems.

3. Modelling gasifier and powerplant

Two different modules, developed with EES software(a tool for solving systems of equations with built inlibraries of several substances thermodynamic properties,[28]), model the gasifier and the whole powerplant.

1 [kg/s]

]

WGT = 5612 [kW]

]

847 [K]ηGT = 0.3129

ηpnc = -0.9371

LpCC = 0.035

ηfilm = 0.23

εH = 0.4

1.013 [bar]

Tb = 1135 [K]

ηgen = 0.98

19.95 [kg/s]

]

ηpt = -0.8

Φ Φ GT = 0.01259 [m2]

io = 12 [%]

14.47 [bar]1519 [K]

turbine and main working data.

ARTICLE IN PRESS

Table 1

GE-5 gas turbine design working data and main data of the co-fired power

cycle

Net electric power output (kW) 5500

Compressor pressure ratio 14.8

Exhausts flowrate (kg/s) 20

Exhausts temperature (1C) 574

Heat rate (kJ/kWh) 11,740

Inlet compressor temperature (K) 288

Inlet compressor pressure (bar) 1.013

Assumed combustion chamber pressure loss (Dp/p) (%) 3.5

Design inlet turbine dimensional flow coefficient (m2) 0.01285

Assumed blade cooling effectiveness eH 0.4

Assumed film cooling effectiveness Zfilm 0.23

Assumed compressor polytrophic efficiency Zpc 0.8

Assumed cooled turbine polytrophic efficiency Zpt 0.8

Calculated uncooled turbine polytrophic efficiency Zpt 0.938

Assumed blade metal allowable temperature Tb (K) 1135

Calculated maximum cycle temperature (K) 1519

Exhausts composition

Mass fraction O2 0.1473

Mass fraction N2 0.735

Mass fraction CO2 0.06868

Mass fraction H2O 0.04902

Further parameters of co-fired power cycle

Biomass composition (% dry basis)

C 49.60

H 6.20

O 44.20

Pressure loss into the GT recuperator (Dp/p) (%) 3

Pressure loss into the gasifier+tar cracker (Dp/p) (%) 3

Pressure loss into the SEP+scrubber (Dp/p) (%) 5

Pressure loss into the HRSG+baghouse filter (Dp/p) (%) 5

Syngas cooler approach temperature (outlet syngas–inlet air) (K) 60

HRSG approach temperature (inlet syngas–outlet steam) (K) 20

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567554

3.1. Gasification/pyrolysis module

Which follows a chemical equilibrium approach for theevaluation of syngas composition, once the inlet biomass,oxidant and reducer composition, equivalence ratio (ER)and pressure are fixed. The ER is here defined as the actualair/fuel ratio (AFact) divided by the theoretical stoichio-metric air/fuel ratio (AFst), which depends, obviously, onthe fuel chemical composition: ER ¼ AFact/AFst.

The general chemical equation, referred to one mole ofbiomass, is the following:

CHaOb þ a O2 þ79

21N2

� �þ bH2Oþ cCO2

) n1O2 þ n2N2 þ n3CO2 þ n4COþ n5H2O

þ n6H2 þ n7OHþ n8NOþ n9NO2 þ n10CH4, ð1Þ

where CHaOb is the chemical composition of the biomass,resulting from the ultimate analysis of the feedstock (seeTable 1 for reference values here adopted).

Two calculation approaches may be followed:

�

Adiabatic, by fixing ER and calculating the outletsyngas composition and the related equilibrium tem-perature Tg, following the energy balance:

HR ¼ HP þQ,

where HR and HP are the total energies of reactants andproducts respectively at the related temperatures,defined as

HR ¼ LHVCHaOb þ a

Z T in

T0

cpO2dT þ

79

21a

Z T in

T0

cpN2dT ,

(2)

HP ¼X

i

ðni

Z Tg

T0

cpi dTÞ þ n4LHV co þ n6LHVH2

þ n10LHVC þ n11LHVCH4, ð3Þ

where LHV are the lower heating values of thecombustible species; Q is the heat needed or releasedby the gasification reactions, which is set to zero in theadiabatic process.

� With fixed gasification temperature Tg and ER, by the

same heat balance, determining the heat Q to bereleased/provided from/to the gasifier.

The number of moles of products n1,y, n10 of Eq. (1) isfound by minimisation of the Gibbs’ free energy. It is doneinto a procedure that is the core of the gasifier module.The pyrolysis is implemented into same module, with the

addition of a semi-empirical model, which is able to predictthe gas composition and char production as functions ofthe inlet biomass composition and the temperatureavailable at the pyrolyzer. Three main assumptions havebeen done:

(1)

pyrolysis is isothermal; (2) the temperature range within the reactor must be within

the 400–800 1C range and the required heat may beprovided externally (for example, by cooling the GTexhausts [31]);

(3)

residence time of the reactants into the reactor isenough to achieve complete reactions.

In particular, the model is provided with a polynomialsecond-order correlation based on fitting of experimentaldata [32], allowing the determination of the char massproduced in pyrolysis (char yield as percentage of inletbiomass flowrate), which is a function of reactor tempera-ture:

mchar ¼ 0:000075T2outpyr � 0:125Toutpyr þ 67.

The temperature at the pyrolyzer outlet is determined atequilibrium, by minimisation of Gibbs free energy, with thesame routine as the gasifier module. The chemical speciesconsidered in pyrolysis are CO, CO2, CH4 and H2. Tarsmass flow, here considered as made of H, C and Oelements, is calculated by mass balance of each elementalspecie.

ARTICLE IN PRESSD. Fiaschi, R. Carta / Energy 32 (2007) 549–567 555

3.2. Gas turbine (GT)

Which is modelled by the means of well-tested software,developed at Energy Engineering Department of Univer-sity of Florence [29,30] with EES software, including theblade cooling [28]. The two modules are joined together intwo different sections:

1.

Fuel section: the syngas coming from the gasifier ismixed up with the natural gas and the related fuelmixture is sent to the GT combustion chamber.

2.

Exhausts section: where the GT exhausts are partiallyrecirculated and blown to the gasifier as oxidisers for thegasification process.

4. Effects of blowing gas turbine exhausts on the gasification

reactions

The basic idea of recirculating a fraction of GT exhauststo the gasifier is related to the exploitation of theirconsiderable heat content to the gasifier, where endother-mic reduction reactions take place. This fraction ofexhausts is mainly determined by the working conditionsof the gasifier (i.e. ER). As above remarked, the CO2

content of the exhausts should favour the Bouduardreaction [27]:

Cþ CO2 ¼ 2CO.

The reference composition and temperature of exhaustgas is shown on Table 1.

The volumetric syngas composition vs. ER, related tothe gasification with air and GT exhausts, is shown onFig. 3. The gasification with exhausts leads to a morenitrogen-diluted syngas for a fixed ER, due to the loweroxygen concentration of the exhausts (around 8–9%). Itleads to a lower heating value of syngas (LHVsyn) when the

0.2 0.225 0.25 0.275 0.3 0.325 0.35 0.375 0.40

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

0.55

0.6

ER

Vol

umet

ric S

ynga

s co

mpo

sitio

n. w

et b

asis

CH4

N2

CO2

H2O

COH2

ExhaustsAir

Fig. 3. Syngas composition vs. ER for air and exhausts oxidisers.

exhausts are used, in spite of the higher adiabatictemperature Tg, as shown on Fig. 4. On the other hand,the adoption of the hot gas stream as oxidiser suggests thepossibility of working at lower ERs, as a consistent fractionof the heat needed for the endothermic gasificationreactions is provided by the hot stream itself. Actually,the conversion ratios (moles of gas species per mole of inletbiomass) of the main syngas combustible species (H2 andCO) increase in gasification with hot exhausts, as shown onFig. 5. It is mainly due to the higher inlet temperature andto the content of steam and CO2 of the exhausts.

5. Performance analysis of the co-fired GT power cycle with

integrated gasification

The analysis has been carried out referring to anintegration of syngas into natural gas to cover from 10%to 100% of the overall GT heat rate., However, the 50%bound is very hard to exceed, due to the extremely lowheating value of the produced syngas, which would lead thecompressor–turbine system to a consistent increase inpressure ratio, pushing the compressor toward the surgeline. As previously remarked, it would imply a deepredesign of the expander, or, either, a consistent reductionof maximum cycle temperature, with remarkable perfor-mance loss. As the main objective of this proposal is amoderate reduction of CO2 emissions with reducedinterventions on the existing plant and minimisation ofthe additional costs, the solutions involving deep modifica-tions to the existing equipment fall out of scope of thiswork, so they have not been considered here.In the co-fired powerplant (Fig. 1), the gasification

temperature is calculated with fixed ER (the reference valueis ER ¼ 0.3 in this analysis) and adiabatic reactor model,thus the outlet gasifier temperature is determined, atchemical equilibrium, by the heat balance betweenreactants and products. Even the fraction of syngas fuelreplacing natural gas (Fren) is fixed (the reference value is0.3 here). Fren is defined on heat rate basis, as the fractionof the total gas turbine heat rate provided with biomass-derived syngas:

F ren ¼msyn LHV syn

msyn LHV syn þmCH4LHVCH4

.

As syngas is derived from renewables, Fren is also ameasure of the saved fraction of CO2 emissions withrespect to the standard GT. ER and Fren allow thedetermination of the fraction f of exhausts to berecirculated to the gasifier.From Figs. 1 and 2, showing the reference working data

of the biomass integrated powerplant and the basicthermodynamic data of the target GE5 gas turbine,respectively, it is evident that the introduction of co-firingwith a moderate heat recuperation from the syngas leads toa modest increase in electric efficiency and to a reduction ofelectric power output of about 10%, mainly due to thesyngas recompression. Its power demand has been kept,

ARTICLE IN PRESS

0.2 0.225 0.25 0.275 0.3 0.325 0.35 0.375 0.40

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

ER

Co

nve

rsio

n r

atio

[mo

l/mo

l of

bio

mas

s]

CO

CO2

H2O

H2

CH4Exhausts

Air

Fig. 5. Conversion ratio vs. ER for air and exhausts oxidisers.

0.2 0.225 0.25 0.275 0.3 0.325 0.35 0.375 0.4550

600

650

700

750

800

850

900

950

1000

3000

3500

4000

4500

5000

5500

6000

6500

7000

7500

8000

ER

Tg

[°C

]

LH

Vsy

n [

kJ/N

m3 ]

Tg

LHVsyn

AirExhausts

Fig. 4. Syngas LHV and adiabatic gasification temperature vs. ER for air and exhausts oxidisers.

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567556

anyway, limited by cooling the syngas down to 300K (12Kmore than the environmental temperature, here fixed atISO conditions). This value is below the dew point, thus thecondensation of a fraction of vapour takes place into theSEP, with a further reduction of the power consumption ofsyngas compressor, due to the reduction of the mass flowto be recompressed of about 23%. It has also a beneficialeffect on the syngas heating value, which is raised by about10%. With 30% integration of syngas fuel, the CO2 specificemissions are reduced by about 48% compared withstandard gas turbine, which is a very interesting result. Inorder to assess the possibility of applying the cycle schemeof Fig. 1 directly to GE5 gas turbine with no majormodifications of the existing machine and with minimumperformance loss, the dimensional flow coefficient at theturbine inlet has been calculated:

FGT ¼mGT

rGTaGT½m2�,

where mGT, rGT and aGT are the inlet turbine massflowrate, density and sound speed, respectively. Thedimensional form of the flow coefficient is here enough,as the comparison of working conditions is referred to thesame machine. As it is seen on Figs. 1 and 2, at thereference F ren ¼ 0:3 the increase in flow coefficient of co-fired powerplant relative to the design value of GE5 islimited to about 2.5%, which allows low off design levels.Thus, neither modifications in existing equipment, norreduction in Tmax are required [22]. Moreover, thevolumetric hydrogen content of the natural gas/syngasmixture at the combustion chamber inlet is about 16%,which is well within the suggested range to achieve flamestability and avoid back stream flame propagation [23].Finally, the heating value of the fuel mixture to the inletcombustion chamber is about 25% of natural gas value,which is above the minimum required heating value for gasturbines fuels [16,23].The parametric analysis of co-fired gas turbine vs.

fraction of renewable fuel Fren (ranging between 10% and100%) is shown on Figs. 6 and 7. Fig. 6 reports thebehaviour of electric efficiency ZEL, electric power outputWEL and specific CO2 emissions. The electric efficiency andpower output show a linear decay of about 10% movingfrom 10% to 100% biomass fuel, which is mainlyattributable to the increasing syngas flowrate and thesubsequent increase power demand of the related com-pressor. Fig. 7 shows the behaviours of the heating value ofnatural gas/syngas fuel mixture at combustion chamberinlet (LHVCC) and dimensional flow coefficient at turbineinlet relative to the nominal GE5 design value DFGE5,defined as

DFGE5 ¼FGT � FGE5

FGE5,

where FGE5 is the nominal dimensional flow coefficient ofthe target GE5 gas turbine (Fig. 2). This parameter has a

ARTICLE IN PRESS

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

0.55

0.6

4500

4550

4600

4650

4700

4750

4800

4850

4900

4950

5000

Fren

ηEL

WEL

ηEL

.CO

2sp

ec.e

mis

s. [

kgC

O2/

kWh

]

WE

L [k

W]CO2 spec. emiss.

Fig. 6. Electric efficiency, electric power output and specific CO2 emissions vs. fraction of renewable fuel Fren.

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

4000

6000

8000

10000

12000

14000

16000

18000

20000

22000

24000

-2.5-1.5-0.50.51.52.53.54.55.56.57.58.59.510.511.512.513.514.515.516.517.5

ΔΦΔΦGE5 [%]

LH

VC

C [k

J/kg

]

Fren

ΔΦΔΦG

E5

[%]

LHVCC

Fig. 7. Behaviour of heating value of natural gas/syngas fuel mixture at combustion chamber inlet (LHVCC) and dimensional flow coefficient at turbine

inlet relative to the nominal GE5 design value (DFGE5) vs. fraction of renewable Fren.

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567 557

linear increase over the whole field, ranging from �1.5% toabout 15.5% at full renewable fuelling. The LHVCC has aheavy drop at low levels of syngas integration and isreduced to values lower than 20% of natural gas for Fren

higher than 0.4. This concise analysis suggests that, if theco-fired powerplant has to be built around the original gasturbine without important changes and without appreci-able reduction of performance, the value of Fren must bekept below 0.30–0.35.

6. Sensitivity analysis: effect of ER and gasifier injected

steam

The possibility of water recuperation from syngas at theexit of condensing heat exchanger (SEP) is an interestingfeature of the proposed powerplant. This water, vaporised

within the HRSG, may be conveniently reinjected into thegasifier, in order to enhance the biomass conversion tohydrogen. Otherwise, the produced steam might be usedfor cogeneration purposes. The possibility of pressurisingthe condensed water and inject steam into the combustionchamber is not so convenient in this case, as the relatedincrease in turbine inlet mass flowrate would add acontribution to off design in the same direction of lowheating value syngas fuel. The combined effects of ER andsteam reinjection into the gasifier on the electric efficiencyof the power cycle are shown on Fig. 8a). Three levels areconsidered: full reinjection of the produced steam, 50%reinjection and no reinjection. Within the field of interest(i.e. ER ¼ 0.2–0.35) the efficiency is only a little sensitive toER, showing a minimisation at ER ¼ 0.26. At low ER,steam injection into the gasifier leads to a very modest

ARTICLE IN PRESS

0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.340.14

0.1450.15

0.1550.16

0.1650.17

0.1750.18

0.1850.19

0.1950.2

0.2050.21

0.2150.22

ER

0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.340.1

0.110.120.130.140.150.160.170.180.190.2

0.210.220.230.24

ER

0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.340

0.020.040.060.080.1

0.120.140.160.180.2

0.220.240.260.280.3

ER

Syn

gas

H2c

once

ntra

tion

[%vo

l]

Syn

gas

CO

con

cent

ratio

n [%

vol]

Syn

gas

H2

Oco

ncen

trat

ion

[%vo

l]

Full steam injection50% steam injectionNO steam injection Full steam injection

50% steam injectionNO steam injection

Full steam injection 50% steam injection NO steam injection

(a) (b) (c)

Fig. 9. (a) Volumetric fraction of H2 in the syngas at the gasifier outlet vs. ER at different levels of steam injection, (b) volumetric fraction of CO in the

syngas at the gasifier outlet vs. ER at different levels of steam injection, (c) volumetric fraction of H2O in the syngas at the gasifier outlet vs. ER at different

levels of steam injection.

0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.340.304

0.306

0.308

0.31

0.312

0.314

0.316

ER

0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34450

500

550

600

650

700

750

720

730

740

750

760

770

780

790

ER(a) (b)

η EL

Wcs

yn [

kW]

TC

C [

K]

Full steam injection

50% steam injection

NO steam injection

Full steam injection

50% steam injection

NO steam injection

Fig. 8. (a) Electric efficiency of the power cycle vs. ER at different levels of steam reinjection into the gasifier, (b) syngas compressor power consumption

and inlet combustion chamber temperature of compressed air vs. ER at different levels of steam reinjection into the gasifier.

0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34600

650

700

750

800

850

900

950

1000

1050

1100

1150

1200

ER

Tg

[°C

]

Full steam injection

50% steam injection

NO steam injectino

Fig. 10. Gasifier outlet temperature vs. ER at different levels of steam

reinjection.

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567558

detrimental effect, whereas above 0.26 full steam injectionshows rather more consistent advantages and efficiency isimproved of 0.5 points at the reference value ER ¼ 0.3.The minimisation of the cycle efficiency vs. ER is mainlythe result of two opposite effects: (I) the increased powerdemand of fuel compressor at higher ER, due to the highersyngas mass flowrate, and (II) the temperature of thecompressed air at the syngas cooler outlet (which isincreased due to the higher recuperation). BelowER ¼ 0.26 the first one dominates, whereas at higher ERthe second one prevails. These effects are exalted with fullreinjection of steam into the gasifier (Fig. 8b).

The volumetric fractions of the chemical species of thesyngas that are more sensitive to steam injection at thegasifier outlet are reported on Fig. 9, showing a slightoptimisation of hydrogen concentration at ER ¼ 0.26. Thisoptimisation is more evident at higher steam injectionlevels (Fig. 9a). The volumetric fraction of CO decreaseswith increasing ER (Fig. 9b). The vapour content of syngas(Fig. 9c) shows an opposite behaviour with ER. Thesetrends are justified by observing that an increase in ERleads to higher recirculated mass flow of exhausts andhigher gasifier equilibrium temperature (due to the larger

extent of partial combustion reactions, see Fig. 10). As thegas turbine pressure ratio and the approach temperaturedifference between the outlet syngas (i.e. inlet HRSG) and

ARTICLE IN PRESS

0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.340

0.050.1

0.150.2

0.250.3

0.350.4

0.450.5

0.550.6

0.650.7

0.750.8

0.850.9

ER

mst

eam

. mIG

[kg

/s]

msteam

mIG

Fig. 11. Mass flowrate of exhausts and steam reinjected to the gasifier vs.

ER at different levels of steam reinjection.

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567 559

the inlet compressed air are fixed, the steam production isincreased with ER due to the higher mass flowrate, thus alarger amount of steam may be produced and sent back tothe gasifier (Fig. 11). Indeed, a larger amount of water maybe condensed at SEP and thus recirculated to the gasifier.When the whole amount of this water is recirculated to thegasifier, it gives a valuable contribution to the conversionof biomass to hydrogen at ER higher than 0.24. It isevident on Fig. 12, reporting the volumetric fractions ofH2, CO and CO2 vs. ER, downstream the SEP, at differentsteam injection levels. The optimisation of H2 concentra-tion is confirmed at ER ¼ 0.26, whereas at high steaminjection levels the H2 and CO2 content of the syngas(Figs. 12a and b, respectively) is enhanced, at the expensesof CO (Fig. 12c). As the volumetric fraction of watervapour into the syngas is constant with varying ER andsteam injection (due to the same imposed SEP outlettemperature which leads to the same gas saturationconditions), it can be stated that the higher content of H2

and CO2 and the lower content of CO into the syngas athigher reinjection levels is attributable to the larger extentof water gas shift reaction:

COþH2O ¼ CO2 þH2.

Obviously, at higher ER, the CO2 content increases dueto the dominant effect of the combustion reactions.

On the whole, the influence of steam injection into thegasifier on the syngas heating value is negligible at all(Fig. 13) and is further dimmed when natural gas fuel isadded.

7. Evaluation of additional costs due to the gasification

system and analysis of the payback time

This analysis has been carried out on the basis ofliterature data, which are referred to a nominal 30MWpowerplant size [36,37] and later rescaled down with the

following relationship, which was introduced by Faaij:

costsize 2=costsize 1 ¼ ðsize2=size1ÞR,

where R is a scale coefficient, variable within the 0.6–0.85range and depending on the specific components consid-ered.The costs of the main additional components required

for the integration of biomass derived fuel and the relatedcooling/cleaning equipment to the standard natural gas-fuelled GT are reported on Table 2 and are taken directlyfrom published data [33,36]. When the costs are scaleddown from the reference 30MW powerplant, which iscompletely biomass fuelled, the level of biomass integrationof the present case is taken into account, considering thatthe scale factor is given by the ratio between the fraction ofnominal power output produced by renewable source andthe reference 30MW case. Thus, referring to the nominal5MWe, the power capacity of gasification system and therelated devices are reduced by a factor 6 multiplied by Fren

and not 6 times only, as would be if the case under studywas completely biomass fuelled.The current cost of the 5MW GE5 here proposed

powerplant is considered around 800 h/kW, which seems tobe a plausible value for that scale (500–600 h/kW aretypical of large combined cycles [17]). The modificationsrequired by the GT combustion chamber in order to allowthe burning of LCV syngas is considered variable with Fren

as a percentage of the gas turbine overall cost [36]:

�

2% for burner modification at Freno1%; � 5% for replacement of burner with a dual fuel one when

1%oFreno25%;

� 20% for complete replacement of combustion chamber

when Fren425%.

The cost of the gas cooling equipment of the reference30MW plant, which is assumed on [36] to be 2.6Mh as awhole, here is differentiated for each component of thesyngas cooling chain and the related values are taken from[33]. The sum of costs is well above the assumption done in[36] and leads, probably, to an overestimation of the syngascooling costs, which we consider justified for two mainreasons:

1.

the gas turbine recuperator is a gas/gas heat exchanger(with typically low heat transfer efficiency) which workson a still not completely particulate cleaned gas, thus itshould be equipped with automated mechanical systemfor periodical soot removal;

2.

the condensing heat exchanger (SEP), which brings thesyngas below the dew point, is a well known componentin chemical industry, but the content of aggressivesubstances, the managing of the condensed watercontaining several contaminants and its back piping tothe gasifier requires the use of special alloys, which arerather unconventional in the heat transfer practise.

ARTICLE IN PRESS

0.2 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.28 0.29 0.3 0.31 0.32 0.33 0.340.08

0.085

0.09

0.095

0.1

0.105

0.11

0.115

0.12

0.125

0.13

0.135

ER(c)

CO

2vo

l. co

nce

ntr

atio

n

Full steam injection

50% steam injection

NO steam injection

0.2 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.28 0.29 0.3 0.31 0.32 0.33 0.340.15

0.155

0.16

0.165

0.17

0.175

0.18

0.185

0.191

0.195

0.2

0.205

0.21

0.215

0.22

ER(a)0.2 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.28 0.29 0.3 0.31 0.32 0.33 0.34

0.14

0.15

0.16

0.17

0.18

0.19

0.2

0.21

0.22

0.23

0.24

ER(b)

H2v

ol.

con

cen

trat

ion

CO

vol.

con

cen

trat

ion

Full steam injection

50% steam injection

NO steam injection NO steam injection

50% steam injection

Full steam injection

Fig. 12. (a) Volumetric fraction of H2 vs. ER downstream the SEP at different steam injection levels, (b) volumetric fraction of CO vs. ER downstream the

SEP at different steam injection levels, (c) volumetric fraction of CO2 vs. ER downstream the SEP at different steam injection levels.

0.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34

4000

5000

6000

7000

8000

9000

10000

11000

12000

13000

14000

15000

16000

ER

LH

Vsy

n [

kJ/k

g]

Full steam injection

50% steam injection

NO steam injection

Fig. 13. Heating value of syngas at the SEP outlet and heating value of

fuel mixture at the combustion chamber inlet.

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567560

Additional costs related to biomass logistics have beenestimated, with reference to [33], in 6 h per ton of drybiomass. The biomass scrubber might be, theoretically,

omitted, as the SEP brings the syngas under saturationconditions and the residual particulates, alkaline metalsand ammonia should be removed here. Anyway, as such acondensing heat exchanger is not proven to be applicableto replace the scrubber, the latter has been accounted aswell (it might be even regarded as a way for increasing thecost of the SEP, which is one of the most critical andlargely uncertainty affected components of the proposedpowerplant).With the assumed data of Table 2, the cost of

gasification equipment to achieve 50% natural gasreplacement represents about 107% the total cost of gasturbine. With integration of 30% renewable, the additionalcosts of gasification island are about 95% of the whole gasturbine. Finally, if a modest 15% of renewable is used inco-firing, the additional costs are reduced to about 50% ofstandard powerplant cost (no combustion chamber repla-cement is required). When the gas turbine plus gasifierequipment costs are summed, at each of the three levels ofFren, the resulting overall powerplant costs are on line withthose extrapolated from Bridgewater regression [17] when

ARTICLE IN PRESS

Table 2

Estimation of gasification island costs of the co-fired powerplant (Mh)

R F ren ¼ 0:5 F ren ¼ 0:3 F ren ¼ 0:15

Fuel

Biomass storage

and distribution

system

0.6 0.080 0.059 0.039

Biomass dryer 0.8 0.9 0.6 0.35

Biomass

conveyors

0.8 0.043 0.029 0.017

Biomass feeding

system

0.7 0.055 0.039 0.024

Reactor

Gasifier 0.7 0.49 0.34 0.21

Syngas cooling/cleaning system

Cyclones 0.7 0.38 0.27 0.16

Metal removal 0.7 0.048 0.034 0.021

Tar cracker 0.7 0.48 0.34 0.21

GT recuperator 0.7 0.48 0.34 0.21

HRSG 0.7 0.45 0.31 0.19

SEP 0.7 0.34 0.24 0.15

Scrubber 0.7 0.24 0.17 0.10

Particulate filters 0.65 0.43 0.31 0.20

Syngas recompression

Compressor 0.85 0.26 0.17 0.095

Combustion chamber modifications

Replacement Replacement Burner

modification

0.88 0.88 0.22

Overall 5.56 4.13 2.20

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567 561

the scaling down on the basis of biomass input is done.Actually, the analysis of Bridgewater is referred topressurised gasification in large powerplants, which shouldlead to higher costs than the atmospheric one hereproposed, but the economy of scale referred to this smallsize should, approximately, compensate the higher specificcosts due to pressurised gasification.

The evaluation of investment profitability has beencarried out referring to the payback time (PBT) of theinitial investment, which should be an effective parameterin this analysis, considering that one of the mainattractiveness of the present proposal is the possibility ofapplication to existing powerplants within short term.

Even if the proposed powerplant comes out as a possibleproposal for partial fighting of greenhouse effect within theshort term, since 2002 the European Community entitlesthe producers of electricity from renewables to get thegreen certificates, which are additional incomes to en-courage the production of electricity from renewable fuels(Directory 2001/77/CE of European Community [38]). Onthe basis of this directory, the goal of Italy is to reach 22%energy production from renewables referred to the overallinternal consumption. To this aim, the Italian Managementof Electricity Board (GRTN, [38]) has fixed the value of

green certificates, for the year 2005, in 108 h/MWhproduced from renewable source. The converted power-plants, or the novel ones, are entitled to get greencertificates for 8 years time from their startup. Originally,the proposal of 2001/77/CE fixed the price of greencertificates at about 80 h/MWh. They represent a funda-mental income and may greatly improve the attractivenesstoward the conversion of powerplants to partial biomassfuelling. The costs analysis here presented has been carriedout vs. variable price of green certificates from 0 to0.1 h/kWh produced from renewable fuel.In the recently introduced CO2 emissions trading, the

proposed power cycle may also be entitled to be a CO2

allowances seller. Given the relative uncertainties related toemissions trading, which has not started yet, the values ofCO2 quota (here referred as Cquota symbol) are determinedon the basis of the old carbon tax, considered variable fromthe last applied values (5–10 h/t) to possible values of25–50 h/t [39,40]. The main difference between carbon taxand emission allowance is that the first one considers CO2

emissions as a source of outcomes for the powerplantmanagement, whereas the second one sets the avoidedemissions, referred to the current levels, as a source ofoutcomes to be paid. Thus, the calculations must be doneon the basis of avoided CO2 emissions, referred to a givenstate of the art. The mechanism of emissions trading shouldassign a specific amount of allowances to each electricityproducer. If it is able to reach lower levels, it may sellallowances. On the contrary, if its emissions exceed theassigned allowance limit, it is charged for the difference.Thus, CO2 emissions quota can be regarded as a possibleincome or outcome source. In the present paper, theavoided CO2 emissions with co-fired plant are referred tothe nominal emissions of GE5 gas turbine (see Fig. 2).The average price of natural gas and biomass fuel are

assumed to be 20 and 5 ch/kg, respectively. The yearlysavings or incomes of the co-fired cycle with respect to thestandard gas turbine are mainly due to three terms:

1.

lower natural gas consumption; 2. sale of green certificates; 3. CO2 quota (sale of allowances and/or avoided penal-

ties).

The sensitivity analysis of PBT vs. price of greencertificates, integrated fraction of renewable Fren and CO2

quota price has been carried out. A discount rate of 10%and a plant working time of 7000 h/year have beenassumed. The latter, generally set at about 8000 h/year instandard largely diffused fossil-fuelled powerplants, hasbeen reduced because of the uncertainty related to thereliability and actual utilisation coefficient of somecomponents, such as tar cracker (that has reached a fullcommercial level only recently) and syngas cooling system,especially the sections at lower temperature.Fig. 14 shows the behaviour of PBT vs. fraction of

renewable Fren at different values of green certificates

ARTICLE IN PRESSD. Fiaschi, R. Carta / Energy 32 (2007) 549–567562

(ranging from 0 to 0.1 h/kWh) with fixed CO2 quota at 10 hper avoided ton referred the simple gas turbine cycle. Withincreasing Fren, PBT increases at low values of Fren, thendecreases above F ren ¼ 0:25. It is due to the competition oftwo effects: the investment cost, increasing with theintegrated fraction of renewable, and the incomes fromCO2 quota and green certificates, increasing with Fren. Theinfluence of green certificates price is fundamental inmaking PBT attractive or unattractive and, obviously, indriving a decision about the conversion of an existingfossil-fuelled powerplant to co-firing. Generally, within therange of 50–100 h/MWh electric power from renewablefuel, the return of initial investment is accomplished withina reasonable time, which varies between 1.5 and 5.5 years.Referring to the Italian directive established by GRTN[38], the current price of green certificates, as abovementioned, is about 108 h/MWh from renewables. Theplant is entitled to get green certificates for eight years fromstartup. This case is represented on the lowest line ofFig. 14, and shows the appealing possibility of achievingPBT within 1.5–3.5 years. The PBT is maximised, for anyprice of green certificates, at 25% Fren. It is due, as aboveremarked, to the assumption that the complete combustionchamber replacement is needed when the co-firing levelexceeds 25%. When no income from green certificates isaccounted, the PBT increases dramatically and reachesvalues above 10 years at Fren40.2.

The influence of CO2 quota on PBT is shown on Fig. 15, atthree different values of Fren. Its effect on PBT is important,but less than that due green certificates. A reduction of up to3 years PBT is observed if CO2 quota is increased from 0 to45h/t. The sensitivity is higher at lower co-firing levels, due tothe less relative influence of green certificates.

It is interesting to notice, on Fig. 16, the relativeweight of the three terms contributing to the overall yearly

0 0.05 0.1 0.15 0.2 01

1.52

2.53

3.54

4.55

5.56

6.57

7.58

8.59

9.510

F

PB

T [

Yea

rs]

GC=0

Fig. 14. Behaviour of PBT vs. fraction of renewab

gain of the co-fired plant over the standard natural gasfired plant:

1.

.25

ren

/kW

le F

natural gas savings;

2. incomes from green certificates; 3. incomes (or missed outcomes) from sale of CO2

allowances.

At low Fren, the savings due to cheaper fuel and CO2

quota play a dominant role, which is exalted when lowprice of green certificates is considered. Above 20% co-firing level, their contribution is dominant. Anyway, CO2

quota still plays an important role in reducing PBT.

8. Main uncertainties connected with the performance and

economic results of the proposed co-fired power cycle

When dealing with biomass co-fired powerplants, severaluncertainties affect the calculated performance and costs.Often, they are due to the difficulties in matching closelythe actual and assumed input data on which theperformance and costs analysis have been carried out.The most important output data, on which basis thethermodynamic and economic assessment of the power-plant is carried out, are the power output and efficiency. Inthe case of the proposed co-fired power cycle, otherfundamental outputs (some of which are directly connectedwith the main performance data) are the CO2 specificemissions, the fraction of input power coming fromrenewable source and the payback time. These parametersare, more or less directly, affected by the uncertaintiesrelated to the following input data:

�

compressor and turbine efficiencies, as the existing GE5is run under co-firing mode, thus, even if the above

0.3 0.35 0.4 0.45 0.5

h

GC= 0.05 /kWh

GC=0.08 /kWh

GC=0.1 /kWh

ren at different values of green certificates.

ARTICLE IN PRESS

0 10 15 20 25 30 35 40 45

1.41.61.8

22.22.42.62.8

33.23.43.63.8

44.24.44.6

Cquota [ /t]

PB

T [

year

s]

Fren=0.15

5

Fren =0.3

Fren=0.5

Fig. 15. Influence of CO2 quota on PBT at three different values of Fren.

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

Fren

Rel

ativ

e co

ntr

ibu

tio

n t

o y

earl

y in

com

es [

%] GC=0.05 /kWh

GC=0.08 /kWh

GC=0.1 /kWh

Savings due to cheaper fuel [%]

Savings due toCO2quota [%]

Savings due to green certificates [%]

Fig. 16. Relative weight of the green certificates, CO2 quota and fuel cost to the overall yearly gain of the co-fired plant over the standard natural gas-fired

plant at different levels of green certificates price.

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567 563

analysis has shown no high off design levels at moderateintegration of biomass fuel, some performance loss incompressor and turbine have to be expected. To keepthem into account, an uncertainty of 5% on thereference assumed values (refer to Table 1) has beenadopted.

� Syngas temperature at the SEP outlet (TOSEP), as actual

reliability of this component at commercial level has tobe proven yet. It affects the powerplant performancethrough the power consumption of syngas compressor(directly related to TOSEP, which represents the inlettemperature). Referring to the basic value of Fig. 1, apossible uncertainty of 720K has been assumed.

�

Pressure losses on the syngas cooling line, as theprogressive accumulation of soot can lead to someplugging effects. It should be minimised by periodicalcleaning (manual or automated), but a 20% relativeuncertainty referred to the Dp/p values of Table 2 hasbeen assumed. � Biomass availability (here accounted as an uncertainty

on input biomass flowrate mbio), as the proposed co-fired cycle, when working at the reference F ren ¼ 0:3,needs the remarkable amount of 7200 tons of biomassper year. It implies a very accurate evaluation of theavailable resource at acceptable distance from thepowerplant site. Thus, a fluctuation of 20% availability

ARTICLE IN PRESS

ToSEP

300720

5.28

0.31

10.28

1.91

2.61

33.52

7.91

0.54

8.05

9.72

D. Fiaschi, R. Carta / Energy 32 (2007) 549–567564

is accounted in the uncertainty analysis. It reflectsdirectly on the incomes from green certificates and CO2

quota, as lower amounts can be sale.

uts

0

�

Table

3

Effects

oftheuncertainties

ofthebasicdata

ontheuncertainties

ofthermodynamic

andeconomic

outputs

ofco-firedcycleandrelativepercentageinfluence

ofeach

inputontheoutp

Outputvariable7uncertainty

Inputvariable7uncertainty

Z pc

Z pnc

Z pt

Lpgas

LpHRSG

LpREC

LpSEP

mbio

hY

0.870.04

0.93770.047

0.870.04

0.0370.06

0.0570.01

0.0370.06

0.0570.01

0.288770.0577

70007100

Relativecontributionsto

uncertainties

onoutputparameters(%

)

Z EL

0.308370.0403

52.16

19.23

22.34

00

0.82

00.16

0

WEL(kW)

48977781.6

69.43

13.70

15.92

00

0.60

00.03

0

CO

2spec.em

iss.(gXkWh)

425.7765.98

23.42

13.26

15.40

00

0.56

037.07

0

Fren

0.300170.0642

3.61

0.01

0.01

00

00

94.46

0

LHVCC(kJX

kg)

1228871927

3.47

00

00

00

93.91

0

GCH4(khXY)

380.27107.8

32.88

0.07

0.08

00

0.01

08.04

25.40

GCO2(khXY)

168.6730.43

7.76

0.02

0.02

00

00

21.68

62.62

Ggreen(khXY)

822.87230.8

10.84

4.16

4.83

00

0.18

053.52

25.93

GTOT(khXY)

13727309.9

0.03

2.00

2.33

00

0.08

047.51

39.99

PBT(Y

)3.10470.583

12.62

0.15

0.17

00

0.01

01.04

76.30

Yearly number of working hours (hY), due to the highercomplexity and unpredictable events related to themanagement of the gasifier and the syngas cleaningsystem, which depend also on variables that are difficultto predict with high accuracy, such as the biomasscomposition, ashes content, kind of contaminants andso on. All these factors may lead to a consistentvariability of the number of plant stops during the year.The possibility of having 1000 h/year absolute uncer-tainty, referred to the basic 7000 h/year working timehere assumed, has been taken into account.

The effects of the above-listed uncertainties on the mainthermodynamic performance, yearly gains over the refer-ence natural gas-fired GT cycle and payback time of the co-fired powerplant are shown on Table 3. On the first up row,the reference values of the basic assumed workingparameters and the related uncertainties are reported.On the first left column, the power cycle performance,yearly incomes terms and PBT at the reference conditions,together with the related propagated uncertainties,are shown. On the other cells, the percentage relativecontributions of uncertainties of the basic data oneach thermodynamic and economic output parameterare reported. It is clear that the electric efficiencyand power output suffer of compressor performancereduction (i.e. due to off design). Among all the consideredvariables affected by the given uncertainties, the compres-sor polytropic efficiency contributes by 52% to theuncertainty of power cycle efficiency, 69% to the un-certainty of power output and 23% on CO2 specificemissions. As these variables are directly connected withfuel consumption, the savings due to reduced natural gasconsumption, CO2 emissions and incomes from greencertificates are affected as well, respectively by 32%, 7.7%and 10.8%.

The contribution of the uncertainty on polytropicturbine efficiency (cooled and non-cooled sections Zpncand Zpt) to the uncertainty on cycle efficiency and poweroutput are rather smaller than that due to uncertainty oncompressor efficiency and are on the order of 20% for thefirst one and 15% for the second one. The uncertainties onZpnc and Zpt weigh 15% on the overall uncertainty of CO2

specific emissions. The mostly affected economic para-meters is the income from green certificates sale (Ggreen), asa lower amount of power is produced.

Another important thermodynamic variable, which maybe a source of uncertainties due to the rather unconven-tional component it is connected to and the nature ofbiomass and its kind and content of contaminants, tars,metals, ashes and so on, is the temperature at SEP outlet(T0SEP). Besides to the heat exchanger role, it has to allowthe final cleaning of syngas, with partially (or even totally)replacing the function of scrubber. The uncertainty on

ARTICLE IN PRESSD. Fiaschi, R. Carta / Energy 32 (2007) 549–567 565

T0SEP reflects directly on the power consumption of thesyngas compressor, thus on the efficiency and CO2

emissions of the power cycle. An uncertainty of 720Kgives an overall contribution of 5% on the uncertainty ofnet electric efficiency and around 10% on the uncertaintyof CO2 specific emissions. The highest relative contributionof uncertainty on T0SEP is on the natural gas savings, whichare affected by 33%, that is the same level as compressorefficiency. The influence on the uncertainty of incomes (ormissed outcomes) due to CO2 quota is almost 8%, as wellas on the overall yearly incomes (GTOT).

The uncertainty on biomass availability over the work-ing time of the powerplant has the biggest influence on theuncertainty of co-firing level (94% on Fren) and, obviously,on CO2 specific emissions (37%), which reduction is solelylinked to the use of renewable fuel. The uncertainty onbiomass availability has a minor effect on the otherperformance parameters like electric power output andefficiency. On the contrary, they might improve due to thereduced off design level (LCVCC is positively influenced).Obviously, the uncertainty of biomass availability has thebiggest effect on the uncertainty of the main estimatedeconomic parameters, giving a contribution of 8% on thenatural gas savings, 21% on incomes (or missed outcomes)from sale of CO2 quota and 53% on the incomes from saleof green certificates. The overall effect of the uncertainty onbiomass availability on the uncertainty of yearly incomesof co-fired powerplant over the basic natural gas-fired GTamounts to about 47.5%. It means that an accurateestimation of the biomass availability over the time is afundamental issue to assure the economic return of theinvestment, as the incomes are closely connected to theactual use of renewable fuel.

The uncertainty on number of yearly working hours ofthe powerplant only affects the economic parameters: it isresponsible for 25% uncertainty on natural gas savings,62% on incomes from CO2 quota, 26% on greencertificates and, finally, 40% on the overall yearly incomes.Moreover, it contributes to more than 76% of theuncertainty on PBT. It means that the reliability ofpowerplant components to avoid stops is a major issuefor the profitability of the investment. Thus, an accurateselection of input biomass even regarding ash meltingpoint, type and content of contaminants and potential tarproduction, as well as an accurate design of the mostcritical components are of primary importance to avoidlarge uncertainties on the actual exploitation of powerplantpotential electricity production.

The relative uncertainty on PBT due to the uncertaintieson the considered input parameters is rather remarkable.Anyway, its absolute value amounts to half year for thereference case, which is a definitely low uncertainty in theanalysis of investments. It means that the uncertainties onthe input parameters have, in practise, a modest effect onPBT but a large effect on the powerplant performance andprofitability once the initial investments for the conversionto co-firing is paid off.

The influences of uncertainties on pressure losses of theheat exchangers are negligible on practically all of theparameters. Only that of recuperator has a very modestinfluence, which, however, is below 1%.On the whole, we can say that the uncertainty on

compressor efficiency is the main source of uncertainty onnet power output and efficiency, while the uncertainty ofturbine efficiency has about half contribution. The relativeinfluence on the uncertainty of economic parameters ismodest compared to the effects of uncertainties on biomassavailability and yearly number of working hours. Theuncertainty on outlet SEP temperature is important onthermodynamic and economic parameters, as it influencesdirectly the syngas compressor power.

9. Conclusions

A relatively not complex solution for reducing a fractionranging from 10% to 50% of CO2 emissions from gasturbine-based powerplants has been proposed. It adopts anintegrated gasification system, which recirculates a smallfraction of the oxygen-rich exhausts to the gasifier, in orderto exploit their heat content and partially sustain theendothermic reduction reactions. In spite of the higheramount of gas flow required with respect to air gasificationdue to the lower oxygen concentration, the related heatcontent and, partially, the larger reducing capacity offluegas due to presence of CO2, lead to an increase ofbiomass to syngas fuel conversion ratio.When compared to the standard gas turbine, the co-fired

cycle with 30% integrated biomass fuel exhibits the samelevel of efficiency with a drop in power output of about8–10%, mainly due to the syngas fuel recompression. Itspower consumption is considerably reduced with theadoption of syngas condensing cooler (SEP), which bringsthe temperature close to the environmental value. The SEPalso helps in the final process of syngas cleaning, beingable, in principle but probably not yet in practise, toreplace the scrubbing with external cooling water.The flow coefficient at the gas turbine inlet, here adopted

as a first approach parameter to predict the off design levelof the co-fired machine, shows an increase of about 2.5%above the GE5 design value, which seems to be acceptableto avoid redesign of the compressor and turbine andimportant performance losses. The heating value and theoverall hydrogen concentration of the syngas plus naturalgas fuel mixture fall within the acceptable limits for gasturbines.The recycle of the steam produced from condensed

contaminated water into the gasifier allows a drasticreduction of the problems related to the managing,treatment and disposal of wastewater. Moreover, thereinjection of steam into the gasifier enhances the hydrogenconcentration of the syngas.Even if the primary idea of the proposed co-fired cycle

was to try a way for reducing CO2 emissions fromgas turbine-based powerplants without redesign or

ARTICLE IN PRESSD. Fiaschi, R. Carta / Energy 32 (2007) 549–567566

replacement of the existing equipment, an economicallyfundamental feature is the possibility of getting greencertificates, which gives a great contribution to return ofthe initial investment. The economic analysis shows thatthe proposed solution might be applied to the existingpowerplant within the very short time, as it allows therecuperation of the initial investment within 2–4 years atthe current levels of green certificate price and CO2 quota.The first ones play a dominant role when the integration offuel from renewables is above 25%.

The main uncertainties connected with the proposedanalysis are those related to the compressor and turbineefficiencies, due to the possible off design conditions, andto the biomass availability and powerplant yearly workingtime: the first ones have direct consequences on theuncertainty of estimated power output, efficiency andCO2 emissions, whereas the second ones affect theuncertainties on the evaluation of co-fired powerplantprofitability. Finally, the uncertainty on SEP outlettemperature influences the uncertainties of both economicand thermodynamic parameters.

References

[1] Adelman ST, Hoffman MA, Baughn JW. A methane-steam reformer

for a basic chemically recuperated gas turbine. ASME J Eng Gas

Turbines Power 1995;117:16–23.

[2] Andersen T, Kvamsdal M, Bolland O. Gas turbine combined cycle

with CO2-capture using auto-thermal reforming of natural gas.

Proceedings of ASME Turbo Expo, Munich, Germany, 8–11 May

2000.

[3] Chiesa P, Consonni S. Shift reactors and physical absorption for low-

CO2 emission IGCCs. ASME J Eng Gas Turbines Power 1999;121:

295–305.

[4] Lozza G, Chiesa P. Natural gas decarbonization to reduce CO2

emission from combined cycles. Part A: partial oxidation. ASME

IGTI conference and exhibition, Munich, paper 2000-GT-0163, 2000.

[5] Lozza G, Chiesa P. Natural gas decarbonization to reduce CO2

emission from combined cycles. Part B: steam-methane reforming,

ASME IGTI conference and exhibition, Munich, paper 2000-GT-

0164, 2000.

[6] Fiaschi D, Lombardi L, Tapinassi L. The recuperative auto thermal

reforming and recuperative reforming gas turbine power cycles with

CO2 removal—part I: the recuperative-auto thermal reforming cycle.

J Eng Gas Turbines Power 2003;125(4):933–9.

[7] Fiaschi D, Lombardi L, Tapinassi L. The recuperative auto thermal

reforming and recuperative reforming gas turbine power cycles with

CO2 removal—part II: the recuperative reforming cycle. J Eng Gas

Turbines Power 2004;126(1):62–8.

[8] Corti A, Failli L, Fiaschi D, Manfrida G. Exergy analysis of two

second-generation SCGT plant proposals. Proceedings of ASME

IGTI 43rd gas turbine and aeroengine congress and exhibition,

Stockholm, Sweden, 1998.

[9] Facchini B, Fiaschi D, Manfrida G. SCGT/CC: an innovative cycle

with advanced environmental and peakload shaving features. Energy

Convers Manage 1997;38(15–17):1647–53 (ScienceDirect).

[10] DOE. Carbon sequestration technology roadmap. Office of Fossil

Energy, NETL, US Department of Energy; 2002.

[11] Corti A, Lombardi L, Manfrida G. Absorption of CO2 with amines

in a semi-closed GT cycle: plant performance and operating costs.

Proceedings of ASME IGTI 43rd gas turbine and aeroengine

congress and exhibition, Stockholm, Sweden, June 1998.

[12] Corti A, Manfrida G. Economic analysis of a semi-closed gas turbine/

combined cycle (SCGT/CC) with CO2 removal by amines

absorption. GHGT-4, fourth international conference on greenhouse

gas control technologies, InterLaken, Switzerland, 30 August–2

September 1998.

[13] Corti A, Fiaschi D, Manfrida G. Thermo-economic evaluation of the

SCGT cycle. Energy Convers Manage 1999;40:1917–29.

[14] IPCC special report on carbon dioxide capture and storage. In: Metz

B, Davidson O, de Coninck H, Loos M, Meyer L, editors. Full

report. Cambridge, New York, Melbourne, Madrid, Cape Town,

Singapore, Sao Paulo: Cambridge University Press; 2005.

[15] IEA. Carbon dioxide capture and storage. Report from Department

of Trade and Industry, September 2000, on www.dti.gov.uk/cct.

[16] Bridgwater AV. The technical and economic feasibility of biomass

gasification for power generation. Fuel 1995;74(5):631–53.

[17] Bridgwater AV, Toft AJ, Brammer JG. A techno-economic

comparison of power production by biomass fast pyrolysis with

gasification and combustion. Renew Sustain Energy Rev 2002;6(3):

181–246.

[18] Marbe A. New opportunities and system consequences for biomass

integrated gasification technology in CHP. PhD.thesis, Heat and

Power Technology Department of Energy and Environment,

Chalmers University of Technology, Goteborg, Sweeden, 2005.

[19] Andre Faaij, Rene van Ree, Lars Waldheim, Eva Olsson, Andre

Oudhuis, Ad van Wijk, et al. Gasification of biomass wastes and

residues for electricity production. Biomass Bioenergy 1997;12(6):

387–407.

[20] Consonni S, Larson ED. Biomass-gasifier/aero-derivative gas turbine

combined cycles. Part A—technologies and performance modelling.

J Eng Gas Turbines Power 1996;118(3):507–15.

[21] Babu SP. Thermal gasification of biomass technology developments:

end of task report for 1992–1994. Biomass Bioenergy 1995;

9(1–5):271–85.

[22] Simon Harvey and Bruno Facchini. Predicting black liquor gasifica-

tion combined cycle powerhouse performance accounting for

off-design gas turbine operation. Appl Thermal Eng 2004;24(1):

111–26.

[23] Rodrigues M, Walter A, Faaij A. Co-firing of natural gas and

Biomass gas in biomass integrated gasification/combined cycle

systems. Energy 2003;28(11):1115–31.

[24] Rhodes JS, Keith DW. Engineering economic analysis of biomass

IGCC with carbon capture and storage. Biomass Bioenergy 2005;

29(6):440–50.

[25] Matteo Carpentieri, Andrea Corti, Lidia Lombardi. Life cycle

assessment (LCA) of an integrated biomass gasification combined

cycle (IBGCC) with CO2 removal. Energy Convers Manage

2005;46(11–12):1790–808.

[26] Sugiyama S, Suzuki N, Kato Y, Yoshikawa K, Omino A, Ishii T,

et al. Gasification performance of coals using high temperature air.

Energy 2005;30(2–4):399–413.

[27] Song C, Pan W. Tri-reforming of methane: a novel concept for

catalyticproduction of industrially useful synthesis gas with desired

H2/CO ratios. Catal Today 2004;98:463–84.

[28] Klein SA. Engineering equation solver. Academic Professional

V7.431 3D, F-Chart Software, 1992–2005, http://www.fchart.com/.

[29] Fiaschi D, Lombardi L. Cellular thermodynamic model for gas

turbine blade cooling. Proceedings of ECOS’01 conference, Istanbul,

Turkey, 4–6 July 2001.

[30] Fiaschi D, Tapinassi L. Comparison of gas turbine power cycles

performance using air/steam blade cooling. ECOS 2003 conference,

Copenhagen, 2003.

[31] Klass DL. Biomass for renewable energy, fuels and chemicals. New

York: Academic Press; 1998.

[32] Pedersen ST, Andersen SM. Pyrolyse af Traeflis, Eksamensproject

MEK-ET-EP-2001-04, DTU, Lyngby, Denmark, May 2001.

[33] Faaij A, van Ree R, Waldheim L, Olsson E, Oudhuis A, van Wijk A,

et al. Gasification of biomass wastes and residues for electricity

production. Biomass Bioenergy 1997;12(6):387–407.

ARTICLE IN PRESSD. Fiaschi, R. Carta / Energy 32 (2007) 549–567 567

[34] Ma L, Verelst H, Baron GV. Integrated high temperature gas

cleaning: tar removal in biomass gasification with a catalytic filter.

Catal Today 2005;105(3–4):729–34.

[35] GE Power Official Web Site. http://www.gepower.com/prod_serv/

products/gas_turbines_cc/en/sm_heavy_duty/ge5.htm, 2006.

[36] Rodrigues M, Faaij APC, Walter A. Techno-economic analysis of co-

fired biomass integrated gasification/combined cycle systems with

inclusion of economies of scale. Energy 2003;28(12):1229–58.

[37] Michiel J, Tijmensen A, Faaij APC, Hamelinck CN, van Hardeveld

MRM. Exploration of the possibilities for production of Fischer

Tropsch liquids and power via biomass gasification. Biomass

Bioenergy 2002;23(2):129–52.

[38] GRTN, 2006, http://www.grtn.it/ita/fontirinnovabili/certificativerdi.asp

[39] Grubb M. Implications of EU ETS for competitiveness of energy

intensive industry. Presentation to IEA/IETA/EPRI workshop on

greenhouse gas emissions trading, Paris, 4–5 October 2004.

[40] Reinaud J. Emissions trading and its possible impacts: the case of

energy-intensive industries in the EU-ETS. Presentation to IEA/

IETA/EPRI workshop on greenhouse gas emissions trading, Paris,

4–5 October 2004.