Gasification of biomass and residues for electricity production

Gasification of biomass and residues for electricity production

Faaij, A., Ree, van, R., Waldheim, L., Olsson, E., Oudhuis, A., Wijk, van, A., Daey Ouwens, C., & Turkenburg, W. (1997). Gasification of biomass and residues for electricity production. Biomass and Bioenergy, 12(6), 387-407. https://doi.org/10.1016/S0961-9534(97)00010-X

DOI:

10.1016/S0961-9534(97)00010-X Document status and date: Published: 01/01/1997

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne

Take down policy

If you believe that this document breaches copyright please contact us at:

openaccess@tue.nl

providing details and we will investigate your claim.

~

Pergamon

P I h S0961-9534(97)00010-X 0961-9534/97 $17.00 + 0.00

G A S I F I C A T I O N O F B I O M A S S W A S T E S A N D R E S I D U E S F O R E L E C T R I C I T Y P R O D U C T I O N

ANDRI~ FAAIJ*, RENI~ VAN REE~', LARS WALDHEIM~, EVA OLSSON~, ANDRI~ OUDHUIS"f, AD VAN WIJK*, CEES DAEY-OUWENSll AND WIM TURKENBURG*

*Department of Science Technology and Society, Utrecht University, Padualaan 14, NL-3584 CH, Utrecht, The Netherlands

tNetherlands Energy Research Foundation, P.O. Box l, NL-1755 ZG, Petten, The Netherlands ++Termiska Processer AB, S-61182, Nyk6ping, Sweden

IlProvince of Noord-Holland, P.O. Box 3088, 2001 DB~ Haarlem, The Netherlands (Received 16 September 1996; revised 24 Januao~ 1997; accepted 31 Januao' 1997)

Abstract--The technical feasibility and the economic and environmental performance of atmospheric gasification of biomass wastes and residues integrated with a combined cycle for electricity production are investigated for Dutch conditions. The system selected for study is an atmospheric circulating fluidized bed gasifier-combined cycle (ACFBCC) plant based on the General Electric LM 2500 gas turbine and atmospheric gasification technology, including flue gas drying and low-temperature gas cleaning (similar to the Termiska Processer AB process). The performance of the system is assessed for clean wood, verge grass, organic domestic waste, demolition wood and a wood-sludge mixture as fuel input.

System calculations are performed with an ASPEN p~"' model. The composition of the fuel gas was derived by laboratory-scale fuel reactivity tests and subsequent model calculations. The net calculated efficiencies for electricity production are 35.440.3% (LHV) for the fuels studied, with potential for further improvement. Estimated investment costs, based on vendor quotes, for a fully commercial plant are 1500-2300 ECU per kWc installed.

Electricity production costs, including logistics and in some cases negative fuel price, vary between minus 6.7 and 8.5 ECUct/kWh. Negative fuel costs are obtained if current costs for waste treatment can serve as income to the facility. Environmental performance is expected to meet strict standards for waste incineration in the Netherlands. The system seems flexible enough to process a wide variety of fuels. The kWh costs are very sensitive to the system efficiency but only slightly sensitive to transport distance; this is an argument in favour of large power-scale plants. As a waste treatment option the concept seems very promising. There seem to be no fundamental technical and economic barriers that can hamper implementation of this technology. (~, 1997 Published by Elsevier Science Ltd

Keywords---atmospheric gasification; ASPEN'~°~; electricity production; biomass wastes and residues

1.

INTRODUCTION

At present, in the Netherlands various biomass wastes and residues are landfilled, incinerated, composted or digested. However, landfilling capacity is scarce and a ban on the landfilling of organic materials will be implemented in the short term. Composting gives rise to problems because supply exceeds demand)4 Furthermore, waste incineration combined with electricity production has low conversion efficiencies. This implies that the energy potential of biomass wastes and residues is poorly utilised.

However, biomass-fired integrated gasifier combined cycle (BIGCC) technology is a promising alternative for handling organic wastes. The potentially high efficiency corn-

pared with mass burning and the potentially low investment costs have been demonstrated in a number of ~tudies? ~0 This technology could therefore contribute significantly to the mitigation of CO2 emissions.

For BIGCC, Faaij et al. 6 and van Ree e t al. ~ have made an inventory of potential technologies. A preliminary feasibility study for the Province of Noord-Holland has also been made. 6 This province, supported by utilities and the Netherlands Ministry of Economic Affairs, has taken the initiative to set up a BIGCC plant. This technology will also be implemented in other countries. In this connection the Global Environment Facility World Bank project in Brazil should be mentioned especially) 2

As a waste treatment system, BIGCC technology should be capable of meeting the 387

very strict emission standards for waste treatment in t h e Netherlands. It should also be flexible enough to deal with a variety of different biomass fuels. In addition the system should be robust, be competitive and involve a minimum of technical risks.

BIGCC units however have not yet been constructed on a commercial basis. Cost estimates vary, 8' ~3, ~4 but the general conclusion is that the first plants will be expensive. A partial solution that can be proposed is to compensate for the initial high investment costs by using biomass wastes or residues that are available at very low or even negative costs. A disadvantage is that this complicates the conversion facility because residues and wastes have different properties and a higher degree of contamination compared with clean wood, e.g. from energy farming. The properties of various biomass wastes and residues in the Netherlands are discussed elsewhere.-" ~5 A detailed system analy- sis and cost assessment are necessary to provide more insight into the prospects and perform- ance of a BIGCC system, especially when it is utilised for a variety of biomass fuels. Such an analysis has been carried out for the Province of Noord-Holland and the results are presented in this paper.

2. SELECTION AND CHARACTERISTICS OF BIOMASS WASTES AND RESIDUES The characteristics of various biomass wastes and residues have been reported elsewhere?' ~5 It was shown that the costs of fuels that are available for energy production differ widely, ranging from a negative value of - 10 up to a positive value of + 5 ECU/GJ. Possible bio- fuels were found to differ substantially with regard to (chemical) composition, moisture content and ash, concentrations o f heavy metals and contents of nitrogen, sulfur and chlorine. It was concluded that in order to meet gas turbine constraints, the ash of the incoming fuel should not be > 10-20wt% of the dry matter content. A moisture content of --~ 70wt% (wet basis) was considered to be a maximum permissible value (for biomass of very low ash). Streams that exceed these limits have either to be treated by other conversion techniques or to be mixed with cleaner materials to meet the maximum permissible values.

The following fuels, representative of the wide variations in fuel characteristics (and prices) and

available in sufficient quantities, have been selected for this system analysis:

• Clean wood (forest thinnings): This stream represents a relatively large potential ( ~ 9 PJL,v/year), 2 but also has relatively high price per GJ. For this study the physical and composition data refer to poplar, which us considered to be representative of biomass residues from forest thinnings.

• Demolition wood: Demolition wood is currently available ( ~ 3 PJLnv/year) at low or negative costs. It is a drier fuel than thinnings. • Verge grass and organic domestic waste ( O D W ) : These streams have a negative value and will therefore reduce the electricity pro- duction costs. Both streams are available in large quantities, 2.1 and 5.3 PJLHv/year respect- ively. O D W could compensate for the absence of verge grass during winter months.

• Sludge: Sludge represents an energy value of about 4 PJL,v/year. Sludge is included in the analysis to illustrate the influence on the performance of a BIG/CL system when a contaminated fuel is used. High nitrogen, sulfur and heavy metal contents make sludge a very difficult fuel. Furthermore, the ash is too high when a G E LM2500 gas turbine is used, as shown in Faaij et al. 2 Consequently, sludge needs to be diluted by a cleaner material to reduce the average ash. F o r this purpose we select demolition wood.

Table 1 summarises the relevant parameters of the selected fuels. 2~5 Representative base values serve as input for the system calculations as well as for the gasifcation tests and gas composition calculations.

The composition of the fuel gas produced by the gasifier varies according to the fuel used. The gas compositions are derived from lab-scale fuel reactivity experiments and from subsequent separate gasifier model calculations. ~6 The results of this exercise for each selected fuel are given in Table 2. These results serve as input for further system modelling.

3. SYSTEM DESIGN AND PERFORMANCE 3.1. System selection and modelling

The selected gasification process is similar to Termiska Processer AB (TPS) technology, which makes use o f an atmospheric circulating ftuidized bed (ACFB) gasifier followed by a separate CFB tar cracker. 4"17-19 The main

Gasification of biomass wastes and residues for electricity production 389 Table 1. Characteristics, availability and costs of five selected biomass fuels (derived from Faaij et a l ? and van Doorn ~5)

Clean Verge Organic domestic Demolition

Fuel type woo& grass waste (ODW) wood Sludge Unit

Moisture' 50 60 54 20 20 b wt% of wet fuel

Ash ~ 1.3 8.4 18.9 0.9 37.5 wt% of dry fuel

LHV 7.7 5.4 6.4 13.9 8.8 MJ/kg a.r. (as-received) HHV 9.6 7.4 8.3 15.4 9.9 MJ/kg a.r. (as-received) Composition wt% dry, ash-free (daf) C 49. I 48.7 51.9 48.4 52.5 H 6.0 6.4 6.7 5.2 7.2 O 44.3 42.5 38.7 45.2 30.3 N 0.48 1.9 2.2 0.15 7.0 S 0.01 0.14 0.50 0.03 2.7 CI 0.10 0.39 0.3 0.08 0.19

Availability in the Netherlands

Gross 13 4 6 3 4 Net 9 4 3 2 4 Cost range Minimum 43 - 9 9 - 107 - 137 95 Maximum 50 11 - 46 - 11 - 38 PJu~v, year ECU/t dry

"Thinnings from commercial forestry are selected. Composition data for poplar wood are presented.

bThe moisture content of sludge from wastewater treatment plants is originally as high as 80-90wt%. After mechanical dewatering and drying, the moisture content is decreased. 20wt% is taken here as a representative value?'

~The quoted moisture and ash figures are considered representative for the biomass fuels as recieved at the conversion facility.

reasons for selecting this process with sub- sequent low-temperature gas cleaning are that it is expected to be able to deal with various biomass fuels with varying fuel properties and degrees o f contamination. Moreover, all parts of the system have been proven commercially. There are however still some technical uncer- tainties, particularly with regard to the inte- gration of various parts, such as the coupling of the gasifier to a gas turbine and a system-integrated dryer.L3 ~4

The gas turbine selected for this study is the General Electric LM 2500. This results in a system with a capacity of ~ 30 MW~. 2° Major arguments for selecting this turbine are that it is under development for low-CV gas appli- cations as part of the G E F World Bank project in Brazil, '~ it is relatively small in size and it therefore requires a relatively modest quantity of fuel. Furthermore, a S T I G version of this turbine (steam injected gas turbine) is available which allows larger differences in mass flows; this is necessary for operation on the low-CV gas produced by a direct gasifier. 2~ 24 Being an aeroderivative, this turbine combines a relatively high efficiency with a high turbine outlet temperature, which results in good conversion efficiencies of the com- bined-cycle plant. 22

The basic B I G C C design is shown in Fig. 1. After gasification o f the biomass, the resulting

fuel gas is cracked in a tar cracker using dolomite as a catalyst. The gas is cooled and particulates and alkalis are removed by a baghouse filter. Remaining contaminants, mainly a m m o n i a , are removed in a wet scrubber. Before combustion in the (modified) combustion chamber, the fuel gas is com- pressed. After steam production, the flue gas is led to a fuel gas dryer to dry wet fuels to required gasifier specifications. Table 3 summarises the main parameters of the se- lected system components. D a t a on these components have been derived partly from the literature, but more especially by con- sulting various suppliers. A more detailed description of the system configuration is given by van Ree e t a l . 2~

A S P E N r~°~ is used as a modelling tool for system calculations. With an A S P E N p~"~ model, mass flows, related emissions and the system performance have been calculated for various fuels. The gasification process itself is not modelled in A S P E N p~u~. The gasifier and tar cracker are modelled as a black box for which the input (parameters of incoming fuel) and output (calculated gas compositions on the basis of experiments) are known (see Table 2). The results of the calculations for each fuel are given in Table 4. Detailed descriptions of the process conditions are given in a background report. 25

Table 2. Fuel gas composition data for various biomass fuels, from fuel reactivity experiments and gasifier model calculations for each fuel (performed by TPS~6). These gas compositions serve as input data for ASPEN m~ modelling

20wt%

Clean Verge Organic Demolition Sludge + 80wt% dem.

wood grass domestic waste wood wood b Unit

Air

Flow rate 1.40 1.48 1.6 1.26 1.41 kg/kg wet

fueP

Temp. 400 400 400 400 400 °C

Dolomite

Flow rate 0.0268 0.0279 0.0279 0.0257 0.0261 kg/kg wet

fuel LCV gas

Flow rate 2.37 2.40 2.42 2.27 2.30 kg/kg wet

fuel

Temp. 900 900 900 900 900 °C

Composition vol% wet gas

C2H6 0.02 0.02 0.02 0.02 0.02 C2H4 0.94 0.87 0.77 0.98 0.88 CH4 2.82 2.61 2.81 2.93 2.63 CO 17.22 14.94 13.98 18.31 15.18 CO2 12.22 12.09 11.80 11.67 12.22 H2 13.25 12.42 11.27 15.07 12.37 H20 13.55 14.49 13.71 13.85 14.34 N2 39.20 41.64 44.59 36.64 41.04 02 0.00 0.00 0.00 0.00 0.00 Ar 0.47 0.50 0.54 0.44 0.49 NH3 0.27 0.33 1.00 0.07 0.78 H2S 0.00 0.03 0.03 0.01 0.04 HCN ppm level HCI 0.03 0.07 0.00 0.02 0.01 Molar 24.86 24.99 25.28 24.28 25.75 kg/kmol mass

Tar 12 11 I0 12 13 g/kg wet fuel

residues

Fly ash 0.036 0.083 0.152 0.032 0.045 kg/kg wet

fuel

Ash 65 87 95 61 84 wt% of fly

ash

LHV 5.22 4.74 4.39 5.59 4.82 MJ/m 3

(wet gas) c (s.t.p., wet)

LHV 5.77 5.31 4.86 6.21 5.6 MJ/m 3 (at

(30°C) d 30°C)

Gasifier ash

Flow rate 0.0158 0.0158 0.0357 0.00t7 0.0785 kg/kg wet

fuel

Ash 90 90 90 90 95 wt%

gasifier ash ~Moisture contents of all input fuels to the gasifier are set at 15wt% to permit comparison of the required heat demand for drying. Consequences for resulting low heating values of the fuel gas (in case o f verge grass and organic domestic waste) are discussed later.

bBoth dry, ash-free; mass fraction of the mixture determined by the minimal required heating value for the gas turbine. CHeating value of gas after tar cracker.

dHeating value of gas after wet scrubber (water condensed at 30°C).

3.2.

System efficiency

As shown in Table 4, the net overall energy conversion efficiency of the system (LHV basis) ranges from 35.4% for the wood-sludge mixture to 40.3% for clean wood. As expected, higher ash results in lower conversion effi- ciency. The same is found for fuels with a higher moisture content. In addition however, several remarks are needed on these results, as follows.

The calculated efficiencies are obtained for specific fuels and for system operation at a design point. In practice it might be that the dryer, feed system, gasifier, fuel gas compressor etc. would all have to be designed to specific boundary conditions, which could possibly result in a lower conversion efficiency.

For all the fuels, the heating values of the fuel gas, which serve as input for the calculations, exclude non-condensable tars, due to uncertainties in the measurements and

G a s i f i c a t i o n o f b i o m a s s w a s t e s a n d r e s i d u e s f o r electricity p r o d u c t i o n 391 Biomass ~ Sizing and screening A i ~ Solids Gas . . . Water/steam cleaning . . . Gas . .. . . .. . . . .. . Ash Gas coo~ing . . . - - - - - t - ; - _ ; . . . /eal s l ; a m ' l ~ Steam " " - . . . . " " Generator ~ t u r b i n e ... ...

©

. . . Gas Fuel gas compressor

Fig. 1. Scheme o f the considered integrated direct atmospheric gasification combined cycle system based on TPS gasification technology.

Table 3. Technical data on system components, derived from the literature and specific information from manufacturers.

More detailed information is given by Faaij et al. 34 and van Ree et al. "-5

Dryer: b Direct rotary drum dryer, 13.8 t/h water evaporation. Mass flows and temperatures for fuel of ~ 50wt% moisture: " d r y " flue gas, 78 kg/s, 2 0 0 C , 1.1 bar; "'wet" flue gas 81,5 kg/

s, 8 0 C , 1.1 bar. '-5

Gasifier: ~ ACFB type TPS technology, 1.3 bar, 9 0 0 C (depends on fuel), heat loss 2% of thermal input. Bed material: sand. Gasifier air: 1.3 bar, 4 0 0 C . ~6

Tar cracker: CFB reactor using dolomite, 1.3 bar, 9 0 0 C ? 6

Fuel gas coolers: 900-140 C (Q ~ 14-15 MW~h depending on the fuel), pressure drop 0.1 bar. -'5

Dust filter: Baghouse filter, pressure drop 0.05 bar?-'

Fuel gas scrubber: Spray tower using recirculating water; mass flow 73 kg/s, pressure 1.3 bar, temperature 2 5 C , pressure drop 0,05 bar. 4~

Fuel gas compressor: multistage compressor with intercooling. Cooling duty 2.3 MW,h, isentropic eft. 0.78, mechanical

eft. 0.998, pressure ratio Pm/P,,u~ + 33/1.1). -~544

Gas turbine: b General Electric LM 2500 (modified for LCV gas). Pressure drop over valves to inlet combustion chamber 10 bar, heat loss 2 MW~.

Compressor mass flow: 65 kg/s, To~, 459'C, mass flow turbine blade cooling 7 kg/s, isentropic eft. 0.91 Combustion chamber: pressure 23 bar, mass flows and To,, depending on fuel type.

Expander: Mass flow flue gas and T,, depending on the fuel type, inlet pressure 23 bar, isentropic eft. 0.89, outlet pressure flue gas 1.1 bar. Generator efficiency 0.99. 2°.2~,'-7

Ambient air: 1 5 C , 1 bar, composition (vol%) 1.01 H_,O, 77.29 N2, 20.7 02, 0.03 CO2, 0.92 Ar.

Heat recovery steam generator: c Superheater 1, 40 bar, 4 5 0 C ; superheater 2, 40 bar, 440-'C; air preheater for gasifier and tar cracker air, 4 0 0 C ; evaporator, 40 bar, 256C; economizer,

240C: minimum pinch air preheater (g/g), 15C; mimimum pinch (g/l), 2 0 C ; total pressure drop from feedwater to superheated steam, 4 bar.

Mass flow o f flue gas and steam produced depend on type of fuel. Steam conditions 450°C, 40 bar.

Steam turbine: Two-stage partly condensing steam turbine; 40 bar, 450-C to 8.1 bar to 0.07 bar. lsentropic eft. 0.735, mechanical eft. 0.99, generator eft. 0.99

Steam-water cycle: Condenser 0.07 bar, using surface water; water pump eft. 0.82. Deaerator: 3.6 bar, minor steam consumption of 8.1 bar.

Water pumps: pressures from 0.07 to 3.8 to 45 bar; eft. 0.99

~Mass flows o f gasifier air, dolomite consumption and ash production for selected fuels are given in Table 2. bTemperatures of incoming and outgoing gas for dryer, combustion temperatures and gas turbine expander outlet temperature depend on the type of fuel -'3 and are given with the results of the model calculations.

392

Table 4. Results of ASPEN p~"~ system calculations with various fuels

Clean Verge Organic Demolition Sludge~lemolition

wood grass domestic waste wood wood mixture"

Fuel input (kg/s) 9.30 12.71 12 5.27 total: 6.65

Moisture (wt%) 50 60 54 20 20

Ash (wt% db) 1.32 9.8 18.9 0.9 av. 11.1

LHV (MJ/kg a.r.) b 7.7 5.4 5.9 13.9 8.4- 13.9

HHV (MJ/kg a.r.) b 9.6 7.4 7.8 15.4 10.0- 15.4

Dryer

Moisture after drying (wt%) 15 15 15 15 15

Flue gas dryer T,, - Tou~ (~C) 195 - 71 276 - 67 292 - I 17 179 - 165 179 - 165

Fuel gas

LHV (M J/m3; 30'=C) 5.77 5.31 4.86 6.21 5.60

Flow (m3/s, s.t.p.) 10.55 11.46 12.50 9.79 10.87

E-input (MW) 60.85 60.85 60.75 60.80 60.87

Gas turbine expander inlet 1150 1136 1122 1160 1145

temperature (°C)

Steam production (kg/s) 11.8 9.85 9.50 12 11.60

Energy balance

Input: LHV (MW,h) 72.0 68.8 70.6 73.1 81.9

Input: HHV (MW,h) 89.6 94.2 93.2 81.2 92.3

Output: Gas turbine (MW0) 26.3 26.7 27.1 25.9 27.1

Steam turbine (MWe) 10.3 8.5 8.2 10.4 10.1

Gross (MWe) 36.6 35.2 35.3 36.3 37.2

Electricity consumption of system

Dryer (MWJ 0.33 0.44 0.39 0.19 0.19

Fuel gas compressor (MW¢) 6.53 7.27 8.10 5.94 7.29

Gasifier air compressor (MW~) 0.22 0.24 0.28 0.21 0.24

Pumps (MWo) 0.43 0.43 0.43 0.43 0.43

Total (MWe) 7.51 8.38 9.20 7.01 8.15

Net output (MWo) 29.0 26.8 25.6 29.3 29.0

Net system efficiency (LHV 40.3 39.0 36.3 40.0 35.4

a.r.) b,c

Net system efficiency (HHV 32.4 28,5 27.5 36.1 31.5

a.r.) b,c

"Ratio of sludge and demolition wood in mixture chosen as 20:80 w/w daf to give a fuel gas with a heating value of 5.6 MJ/m 3 (s.t.p.).

ba.r. implies fuel with moisture content as received at the gate of the facility.

CGenerally the system efficiency decreases with increasing ash content of the fuel. This is mainly due to increased work by the fuel gas compressor because the heating value of the fuel gas falls with increasing ash content; also the combustion temperature decreases with decreasing heating value of the fuel gas.

difficulties in extrapolating laboratory results to full-scale plant. It is therefore uncertain to what extent these tars (which are not removed during gas cleaning) actually appear in the gas. The tars could increase the heating value of the gas by 3-6%. ~6 Since this effect has not been taken into account in the calculations, the efficiencies presented are somewhat pessi- mistic. It should be kept in mind that a 6% increase in heating value of the gas could increase the net conversion efficiency by ~ 2 percentage points.

Another point is that the heat rate de- gradation of the gas turbine during its life- time will have a negative influence on the efficiency. The turbine is maintained at regular intervals, whereupon the efficiency is restored to its original level. However, even with a normal maintenance schedule a 3-4% drop in efficiency of the gas turbine during its lifetime is observed. 26 This is partly compensated by

a higher expander outlet temperature, which permits increased steam production. Overall, the loss in efficiency will be ,,, 2-3%.

The drop in efficiency as calculated for verge grass and organic domestic waste (see Table 4) is due to the steam system selected. The higher heat demand for drying these wet fuels means that the maximum amount of steam produced and superheated is limited by the minimum pinch point of 15°C for preheating air in the heat recovery steam generator (HRSG). If the gasifier air temperature were lowered somewhat (e.g. 380°C instead of the 400°C chosen), the steam system would operate at the selected design conditions. Lowering the gasifier air temperature would also cause a slight decrease in the heating value of the ga's, but the influence of this decrease on the conversion efficiency is very limited. ~6 These parameters are not optimised in this project.

Gasification of biomass wastes and residues for electricity production 393 quality of the incoming biomass are ,-~ 10-

20wt% ash (for dry biomass) and a moisture content of ~ 70% (for biomass of low ash). M o r e ash results in a leaner gas, which requires more compression work and lowers the combustion temperature of the gas turbine. Fuels that are too wet require so much waste heat for drying that steam production drops. Verge grass and especially organic domestic waste produce fuel gas with a heating value below the 5.6 M J / m 3 (s.t.p.) required for the gas turbine. This problem could be solved by more extensive drying. Verge grass meets the required heating value already at a moisture content of 12wt% instead of the 15wt% taken as the starting point in Table 4. This will have very little influence on the overall efficiency, as steam production is only slightly decreased.

Concerning organic domestic waste, a moist- ure content of < 3wt% is required to produce a fuel gas with a heating value of 5.6 M J / m 3 (s.t.p.). The required drying to achieve this will reduce the steam production drastically and might cause unacceptable emissions because the temperature in the dryer will rise and volatile fractions in the biomass might evaporate. However, there are several issues that must be kept in mind: non-condensable tars have been excluded, which could represent 3 - 6 % ad- ditional heating value. Also, the required heating value of 5.6 M J / m 3 might prove to be a conservative constraint. Lower heating values might be allowable with the LM 2500 and certainly with the use of specially developed combustion chambers. To make the processing of organic domestic waste feasible, one can also add w o o d of low ash (demolition wood). Another possible i m p r o v e m e n t option is heat recovery from the ash stream back to the gasifier, thus limiting heat losses and reducing the p r o b l e m of m a x i m u m permissible ash. However, the costs o f this option are not evaluated in this paper.

3.3. Environmental performance

The emissions after combustion have been investigated and c o m p a r e d with Dutch emission standards. First, Table 5 gives the standards for the required fuel gas quality for the LM 2500 gas turbine. The gas cleaning system will in any case have to meet these standards, to prevent excessive wear and corrosion of the gas turbine.

Table 6 shows the standards for gaseous emissions applicable in the Netherlands for

Table 5. Maximum permissible concentrations of contami- nants in flue gas stream to GE LM 2500 turbine ~-'~

Component

Calculated

Maximum maximum

allowable allowable concentration concentrations

in flue gas to in a typical

expander biogas (ppbw)" (ppbw) Solids < 10 lain 600 3000 1 0 < d < 1 3 lam 6 30 > 13 ~tm 0.6 3 Lead 20 100 Vanadium 10 50 Alkalis 4 20 (Na + K + Li) Calcium 40 200 Alkali metal 12 60 sulfates Chlorides 500 2500 Condensable tars - - 0.008 b

"Parts per billion by weight. These are the concentrations after combustion, so the permissible concentrations in the fuel gas are five times as high (values for operation on natural gas). Values in the second column are calculated from the first. When low-calorific-value gas is used. the dilution factor is ~ 6 7, depending somewhat on the composition of the biomass used.

hmg/m' at s.t.p.

waste incineration (the so-called BLA stan- dards) and power generation. The first column shows the strictest set of emission standards and is considered to be applicable to a unit that uses biomass wastes and residues.

Dust. Dust emissions are determined by the limits set by the gas turbine and are thus very low. The dust concentration after combustion is lower than the emission constraint in the BLA; the baghouse filter and scrubber together are capable of meeting this constraint.

Hydrogen chloride. U p to 90% of HCI is removed in the tar cracker and the remaining part is bound to (lime) particulates at 140°C in the baghouse filter. ,Lassing et al. ~6 concluded that almost 100% HC1 will be removed. Any HCI that remains will dissolve in the scrubber. HCI emissions from the system will therefore be negligible.

Hydrocarbons, CO, PCDD and PCDF.

Emissions of hydrocarbons, CO and polychlori- nated dibenzodioxins and dibenzofurans after combustion are determined completely by the specifications of the gas turbine. Because of the high temperature in the gasifier and reducing conditions before combustion in the turbine, these c o m p o u n d s are not formed. These specifications are such that all constraints stated in Table 5 are met when the gas turbine operates on natural gas. This is not expected to be

394 A. FAAIJ et al.

Table 6. Relevant emission standards for combustion of solid and gaseous fuels (mg/m 3 at s.t.p.)

Component BLA a BEES b

EU standards for stationary coal-fired plants Dust 5.0 HCI 10.0 HF 1.0 CO 50.0

Organic compounds (as C) 10.0

SO2 40.0

NO, 70.0

Total heavy metals 1.0 (Sb,Pb,Cu,Mn,V,Sn,As,Co,Ni,Te) Cd and compounds 0.05 Hg and compounds 0.05 Total PCDD and PCDF 0.1 ~ 5 20 35 d 200' 100

aBesluit Luchtemissies Afvalverbranding (Decision on Air Emissions from Waste Incineration); represents emission standards for waste incineration in the Netherlands, at present the strictest in the world.

bBesluit Emissie-eisen Stookinstallaties Milieubeheer (Decision on Emission Regulations for Heat Installations); applicable to boilers of electricity production facilities.

CFor 90% S removal with flue gas desulfurisation. d(65 g/GJ x gas turbine eft.)/30

~ng(l-TEQ)/m 3 at s.t.p.

different w h e n the t u r b i n e is fired with L C V gas. C O e m i s s i o n s h o w e v e r will be h i g h e r t h a n for n a t u r a l gas b e c a u s e o f the l o w e r c o m b u s t i o n t e m p e r a t u r e , b u t t h e y will n o t exceed the a b o v e - m e n t i o n e d s t a n d a r d ? 7

Nitrogen oxides. T h e r e a r e t w o sources o f

NOx; t h e r m a l NOx a n d c o m b u s t i o n o f a m m o n i a p r e s e n t in the fuel gas. W i t h r e g a r d to t h e r m a l NOx, s t a t e - o f - t h e - a r t G E gas t u r b i n e s h a v e emission f a c t o r s as low as 15 p p m v (with 15v01% o x y g e n in the flue gas). T h e l o w e r c o m b u s t i o n t e m p e r a t u r e o b t a i n e d b y using L C V gas ( , ~ 1150°C i n s t e a d o f 1230°C) will r e d u c e the t h e r m a l NOx f o r m a t i o n even b e l o w t h a t level. 26' 27

A m m o n i a is p r o d u c e d d u r i n g gasification. T h e NH3 c o n c e n t r a t i o n s in the fuel gas given in T a b l e 2 d o n o t i n c l u d e the r e m o v a l o f NH3 b y d o l o m i t e . F o r e a c h b i o m a s s s t r e a m the NH3

flOWS in the s y s t e m are given in T a b l e 7. T e s t results h a v e s h o w n that, d e p e n d i n g on the n i t r o g e n c o n t e n t o f the fuel, N is o n l y p a r t l y c o n v e r t e d to NH3. L a s s i n g et al. ~6 i n d i c a t e t h a t b e t w e e n 3 5 % (Miscanthus, waste w o o d ) a n d 8 0 % (sludge) o f the n i t r o g e n in the fuel is n o t f o u n d as NH3 in the gas. M e c h a n i s m s are n o t fully u n d e r s t o o d b u t a large f r a c t i o n is p r o b a b l y c o n v e r t e d to m o l e c u l a r n i t r o g e n .

T h e p e r m i s s i b l e level o f NO,. e m i s s i o n s is 70 m g / m 3 (s.t.p.) (see T a b l e 6). This s t a n d a r d will be exceeded b y all fuels, as s h o w n in T a b l e 7, even w h e n p a r t i a l c o n v e r s i o n to N2 is t a k e n i n t o a c c o u n t . T h e e s t i m a t e s o f the f r a c t i o n o f fuel N t h a t is c o n v e r t e d to N2 are also given. A m m o n i a dissolves well in w a t e r a n d c a n be r e m o v e d f r o m the fuel gas u s i n g a wet s c r u b b e r . T h e r e m o v a l efficiency o f the s c r u b b e r needs to be ~ 8 0 % (for o r g a n i c d o m e s t i c waste) to m e e t

Table 7. NH3 flows, estimated molecular nitrogen formation and and NO~ formation without a scrubber. Volume flows are derived from the ASPEN p~°s calculations

Organic

Clean Verge domestic Demolition Sludge-wood

Fuel wood grass waste wood mixture

Wet gas flow (ma/s at s.t.p.) 10.55 11.46 12.5 9.79 11.49

NH3 (vol% fuel gas) 0.27 0.33 1 0.07 0.78

Estimated N2 formation from fueI-N (%) 50 50 80 35 75

NH3 flow through scrubber (kg/h) 40 55 72 12 62

Gasification of biomass wastes and residues for electricity production Table 8. H2S flow in fuel gas and corresponding NaOH consumption for 100% S removal

395

Organic

Clean Verge domestic Demolition Sludge-wood

Fuel wood grass waste wood mixture

M a s s flow of H2S in fuel gas (g/s) 0.55 3.57 3.94 1.64 3.57

Corresponding N a O H c o n s u m p t i o n (kg/h) 2.31 15.08 16.64 6.92 15.08

the NO, standard. The efficiency can be increased by increasing the water flow or even by adding an acid (such as H2SO4) to the scrubber water. Discharge of the scrubber water to the (aerobic) wastewater treatment system for conversion to nitrate can be considered, but the costs involved are very high.* In this study it is assumed that ammonia is stripped from the scrubber water and removed. Ammonia can possibly be used as a fertiliser.

To some extent ammonia will interact with thermal N O , in the combustion chamber and reduce NOv emissions by the formation of molecular nitrogen. The degree of this interaction is not known.

SulJur dioxide. SO2 emission levels will

depend on the concentration of sulfur in the fuel and on the efficiency of removal in various gas cleaning stages. The sulfur contents of the fuel and the fuel gas (H2S) differ widely, as shown in Table 2. Part of this sulfur will react with lime in the cracker to form CaS. When the sulfur content exceeds 0.1wt% of dry matter in the biomass, chemical equilibrium is reached in the gasifier, which leads to an H2S con- centration of ~ 200-300 ppmv in the fuel gas) 6 This equilibrium state is reached for sludge, verge grass and organic domestic waste, leading to an SO2 concentration in the flue gas of 100 mg/m ~ (s.t.p.), which exceeds the limit of 40 mg/m ~, so measures have to be taken.

H,S dissolves very poorly in water. Adding a base ( N a O H ) to the water stream will convert H_~S to Na_,S, which dissolves well in water. Depending on the standards for surface water near the plant, the wastewater stream may or may not be discharged directly to the surface water. In the latter case, costs of operation will increase as a result of wastewater treatment at central facilities. It is also possible to

*Costs are determined by the oxygen d e m a n d in aerobic wastewater treatment plants; they a m o u n t to 0.47 E C U / k g O_,. which is typical for D u t c h conditions. A m m o n i a is converted to nitrate in wastewater treatment plants, giving an oxygen c o n s u m p t i o n of 4.57 g O,/g N. 2~ This will lead to wastewater treatment costs of ~ 106 E C U / y e a r for organic domestic waste and for a plant operating for 7400 h per year at full load.

precipitate Na2S, which produces a removable solid salt.

In this study it is assumed that all H2S is removed by sodium hydroxide in a wet scrubber. Table 8 shows the H2S concen- trations in the fuel gas for the selected fuels. For verge grass, organic domestic waste and the wood-sludge mixture a concentration of 200 ppmv is assumed? 6 For demolition wood and thinnings it is assumed that 70% of the sulfur in the fuel is bound to lime. The related N a O H consumption (for 100% reaction of H2S to Na + and S 2-) is also given. It is as- sumed that Na2S is removed from the scrubber as a solid salt.

Heavy metals. These will evaporate partly

in the gasifier, most probably to a far greater extent than would happen under combustion conditions. The reducing atmosphere will prevent oxidation of the metals, allowing more evaporation in metallic form. Cooling will condense the metals. All condensation tem- peratures exceed 140~=C, which is the tempera- ture to which the gas is cooled before it is passed through the baghouse filter. At the time of writing, no experimental data are available on the behaviour of heavy metals under gasifica- tion conditions, but it seems likely that all metals will condense during gas cooling. '-9 Possibly some remaining metals will be washed out in the scrubber.

The gasifier ash and the fly ash will contain the heavy metals that, were present in the fuels. The distribution of the various metals will depend on the gasification temperature and the type of fuel. Volatile metals (lead, cadmium, mercury) will concentrate in the fly ash since they evaporate to a greater extent and condense during gas cooling. 3°

Fluoride. No analysis was made of the

emission of fluorides, but the figures available for the fluoride content of demolition wood ( < 0'00003 wt% dry matter 3') are extremely low.

The most important unknown factor in the emissions is the flue gas dryer. The flue gas entering the dryer at ~ 200c'C will cause organic compounds to evaporate and will

396 A. FAAIJ et al. lead to the formation o f dust in the dryer. To reduce dust emissions the flue gas will be passed through cyclones (standard equip- ment for such dryers). The level of hydrocarbon emissions is unknown; possibly additional filters or water scrubbers will be required to meet emission standards.

Another option is to use a steam dryer, which has no effect on the system efficiency 25 but dries the fuel indirectly, thus preventing emission of dust and hydrocarbons. The main disadvantage of steam drying is that investment costs will increase. In addition, more waste- water will be produced (although this can be led to a central wastewater treatment facility). In this study we consider the conventional rotary dryer, taking investment costs for filters into account.

Emissions will also arise from storage (odour) and from the wet scrubber (wastewater). As already discussed, the scrubber water will contain ammonia, which is the main contami- nant. The presence of other compounds and possibly metals will depend on the fuel, although the foregoing gas cleaning steps in principle remove tars, dust and metals. Exper- imental data on this issue are lacking at the moment.

The ash stream from the gasifier and the baghouse filter is another emission from the system. Ash from clean wood such as thin- nings could be used as fertiliser, although this will depend on specific standards applicable. Contaminated fuels (e.g. waste wood and sludge) will produce ash that has to be land filled.

A BIGCC system capable of converting a wide variety of fuels needs to be equipped with a two-stage scrubber with two absorption units (one with water or acid for ammonia removal and the other with an alkali for sulfur re- moval). This will increase investment costs and, depending on the fuel, lead to the consumption of N a O H (and H2SO4).

The extent to which sulfur is bound to dolomite and the degree to which fuel nitrogen is converted to molecular nitrogen have to be investigated in more detail in relation to gasification conditions and dolomite quality. However, the behaviour of the tar cracker is very promising in these, respects? 2

With the proposed gas cleaning concept, the BIGCC system seems to be capable o f meet- ing the severe emission standards for waste incineration.

4. COST ANALYSIS*

In this section the electricity production costs are calculated and discussed for the different biofuels. Investment costs, operating and maintenance costs and logistics costs for collection and transport of the fuel are presented as minimum and maximum elec- tricity production or waste treatment costs. The discounting method used is based on annuity.

4.1. I n v e s t m e n t costs

The investment costs were mainly determined by consulting manufacturers o f various system components. Where possible, cost figures are presented in ranges so that uncertainties can be visualized. Total investment costs are determined by summing the lowest cost per system component and lowest engineering costs for the minimum-cost case and summing the highest component costs and highest engin- eering costs for the maximum-cost case. A first plant will involve high engineering costs. After a number of plants have been built, engineering costs are expected to drop. j2 The high-cost case should therefore reasonably represent the costs of a first c o m m e r c i a l plant, the low-cost case the costs of a plant after a number of similar (identical) plants have been built. However, the costs of a first unit may well lie above the maximum cost level given here, owing to uncertainties in the performance, required testing programmes and potential higher costs because of specified guarantees which are a crucial aspect for a new system.

Vendor quotes are used for all system components, except the gasifier and tar cracker, because only a small number of these com- ponents have been realized hitherto. For the gasifier, expert opinions are used to estimate the costs o f a gasifier based on the TPS concept. With additional information a b o u t the size, materials used and process conditions of an existing similar gasifier (in Greve, Italy) a cost estimate is made using known factors for steel and cement processing for comparable process equipment such as hot-blast furnaces. The investment costs o f the tar cracker are assumed to be the same as those of the gasifier since the design and size are also similar. The uncer- tainties o f such exercises are large but exclude the engineering and development costs.

Other relevant cost factors such as civil works, control systems and interest during

Table 9. Costs of system components Component costs (10 ~ ECU) Min Max Percentage of investment costs Min Max Explanatory notes Pretreatment Conveyors 0.26 0.26 0.6 Grinding 0.3 0.3 0.7 Storage 0.74 0.7 41.7 Dryer 3.5 5.6 7.8 Gasification system Gasifier 1.4 2.3 3.1 Tar cracker 1.4 2.3 3.1 Cyclones 0.9 1.9 2.1 Fuel feeding 0.3 0.3 0.6 Gas cleaning Gas cooling 2.1 2.1 4.8 Baghouse filter 1,2 1.2 2.6 Condensing scrubber 0.9 1.9 2. I Compressor 1.4 1.9 3.1 Combined cycle Gas turbine 9.3 11,6 20.8 Modifications LCV 0,5 0.9 1.0 gas HRSG 2.4 2.4 5.4 Steam turbine + 3,2 3.2 7.2 condenser Water + steam 0.3 0.3 0,6 system Cooling 0.3 0.3 0,6 Overall Control systems 4.7 2.3 10.4 Civil works 3.5 4.2 7.9 Electrical system 2.8 3.4 6.3 Buildings 0.3 0.3 0.7 Engineering (3-4% I. 1 1.7 2.4 of total investment costs) Project contingency -- 5+ 1 -- Interest during 0.4 0.8 1.0 construction I st year Idem 2nd year 1.3 2.3 2.9 Total investment costs 44.6 59.7 0.4 0.5 1.3 9.3 3.9 3.9 3.1 0.5 3.6 2.0 3.1 3.t _19.5 1.6 4.1 5.4 0.4 0.5 3.9 7.0 5.6 0.5 2.8 8.6 1.3 3.9 Assuming total 100 m of conveyors is required on the terrain. Cost figures from +5 Cost figures from "" Assuming storage capacity is suffÉcient for 5 days full load operation. Cost figures from '~ Cost range taken from 4~ (a wide cost range is found for dryers) Cost estimate based on estimated volumes of lining and steel, evaluated with comparable equipment such as hot-blast furnaces and data from the TPS gasification plant in Greve, Italy 47 Investment costs assumed same as those of gasifier, because of similar size, design and process conditions Cost estimate for four cyclones with same lining as gasifier and tar cracker Two double-screw feeders with rotary valves required. Cost figures from 46 Cost figures from ~ Range for a single or two-stage scrubber. Investment costs for wet gas cleaning are increased by 10 + ECU when additional measures are required to remove large quantities of sulfur and nitrogen. 4~ Cost figures from 44 Cost figures for the combined cycle and components from -'~-'+' STIG version LM .2500 including generator Additional costs for adaptations to LCV gas HRSG for modest steam conditions of 40 bar Presence of surface water is assumed, no cooling tower required Degree of automation determines costs of control system. For a combined cycle alone, 0.5 9 x l0 + ECU is possible= '~ The range given here is an assumption. More extensive control implies fewer operating personnel, which is why lower costs are given for the maximum-cost case. Percentage of total investments ~° Percentage of total investments" Assuming that an office, laboratory and porters' building are required. Based on 45 Engineering costs at level for electricity generation plants, assuming known technology is used 4" Project contingency is included only in the high cost cstimate; 10% of investments Expected construction time 2 years; investment in first year 25% of total Investment in second year 75% of total ran O O O ~z 3 ~ ,.-,g B O

398 A. FAAIJ et al. Project contingency ( 8 . 6 % ) Pre-treatment Building interes ~ (11.9%) (5.2 % ~ ~ / ~ ~ ~ s f f i c a t i o n ( l 1.3%)system ~ 1 ~ ~ G a s cleaning Control (8.6%) systems / ~ (3.9%) ~ ~ Compressor 7 (3.1%) Combined cycle (31.3%)

Fig. 2. Breakdown of investment costs (high-cost case) for the selected ACFB-CC system based on the GE LM 2500 as obtained in this study. Total investment costs amount to 60 million ECUI~9 4. "Overall"

covers civil works, engineering, buildings and piping.

construction are obtained from cost data o f comparable installations.

The investment costs o f the system com- ponents are given in Table 9. Where possible, ranges are given. Figure 2 presents the breakdown of investment costs for the high-cost case (clean wood as fuel). In this case a substantial project contingency is included, which is expected to be unnecessary in the low-cost case, where it is assumed that a num- ber have been built already. The total investment costs range from 45 to 60 million ECU.

The gasifier and cracker do not dominate the overall investment costs; they represent 9 - 1 2 % of the total. The combined-cycle unit, which represents one-third o f the investment costs, is the major component. The entire pretreatment system is also a significant cost factor (10-12% of total investment), although uncertainties here are large, However, when only one type of biofuel is used, the pretreatment could remain relatively straightforward. However, when a variety of very different fuels is to be used, different feeding lines might be required which will increase investment costs. In particular, densification or even pelletising equipment that might be required for a fuel such as verge grass would raise the pretreatment costs.* Costs

*Pelletising is an expensive pretreatment option. Feenstra et al. 33 report pelletising costs of ~ 8 ECU/t when done at the conversion facility itself. This excludes drying. Just pelletising (excluding drying) of wood and other biomass residues in a separate facility (20-40 kt/year capacity) costs

15 E C U / t . 34 However, densification may well be sufficient

for feeding fluffy biomass material to a gasifier operating at near atmospheric pressure. Such feeding would also be favoured from the energy point of view, since pelletising requires substantial electricity and heat inputs. More practical experience with fluff feeding of fuels such as verge grass and organic wastes is desirable.

might also increase because of the need for additional equipment attached to the dryer to prevent the emission o f dust and odour, although the investment costs o f the dryer already include various filters.

The costs of land and possibly of additional infrastructure are not taken into account. These factors depend strongly on the exact location o f a conversion unit.

4.2. O p e r a t i n g c o s t s

The costs of operation include personnel, maintenance and insurance. Variable costs relating to the operation of the plant are those o f the catalyst (dolomite) and of ash disposal, which can both be derived from the gas composition data in Table 2. Water use and costs o f wastewater treatment and additives are included when necessary. Relevant cost figures for the operation of the plant are given in Table 10. Figure 3 presents the annual operating costs for each fuel, assuming baseload operation (75% load factor in the maximum-cost case and 85% in the minimum-cost case).

It is assumed that NH3 and sulfur can be removed by several wet scrubbing steps. Although additional investment costs for extensive scrubbing are included in the econ- omic evaluation, a more detailed study o f this c o m p o n e n t is desirable.

4.3. L o g i s t i c s

The results of a logistic study of the supply of biomass waste streams for a B I G C C unit in the Province of N o o r d - H o l l a n d are u s e d ) 3 To calculate the costs of the fuel, including transport, a number o f assumptions were made regarding average transportation distances, location o f the conversion facility, source

Gasification of biomass wastes and residues for electricity production Table 10. Operating costs (input parameters for Dutch conditions)

399

Cost

category Costs Description, assumptions and sources

Maintenance 2% o f investment Personnel 32 500 E C U per p e r s o n - y e a r W a t e r 0.37-1.4 E C U / m ' Dolomite Ash disposal N a O H Insurance 27.9 E C U / t 46.5 E C U / t 1302 E C U / t 1% o f annual depreciation

2% of investment. A s s u m p t i o n based on normal operation of power plants? ~ 5 crews of 2-4 persons for shift work; 4 persons other activities. With a more advanced control system, fewer personnel are required.

W a t e r c o n s u m p t i o n expected to be minimal. The steam system is a condensing system and the waste water stream from the scrubber is expected to equal the condensed water from the fuel gas. 45

Dolomite c o n s u m p t i o n per stream is given in ~6

Ash disposal costs will vary with location and degree of contamination. Tariffs for landfilling will be increased to the level of waste incineration (116 E C U / ) ? 9 Cost figure for bulk quantities of solid N a O H Y '

Data from composting and digestion p l a n t s ) ~

location of the fuel, supply patterns of fuels and type of transport. Other relevant aspects taken into account are drying during storage, costs and capacity of storage and pretreatment (chipping or pelletising) of fuel before it reaches the conversion facility. These data have been calculated for a number of potential fuels (thinnings, prunings, demolition wood, waste paper and sludge). To determine the average transportation distances, several locations for the BIGCC system and various source locations were selected. In general these distances are substantial (75 km one way for thinnings, which covers a large part of the Netherlands). In general it is concluded that transport by road, central storage and pretreatment (at the conversion facility) is the cheapest route. Here, the minimum-cost scenarios for transport, storage and pretreatment are used for further calculations. For the collection and transport costs of organic domestic waste and verge grass, other sources are u s e d . 3'35"36

Table 11 summarizes the minimum-cost scenarios for logistics for the selected fuels. Pretreatment of waste before it reaches the central facility is logically possiblC 7, but in all cases central pretreatment is cheaper, and costs of chipping and drying are therefore included in the conversion costs.

4.4. Cost of electricity and waste treatment The calculated minimum and maximum costs of electricity and waste treatment, based on the real interest rates, lifetime, load factor and construction time in Table 12 are presented in Table 13. The minimum-cost scenarios are the cases in which all parameters (investments, fuel

costs, costs of logistics, load factor etc.) are the lowest. In the maximum-cost cases, all par- ameters result in the highest costs. For verge grass, organic domestic waste and the sludge- wood mixture, additional investments are included to cover a more extensive scrubbing unit.

Figure 4 shows the breakdown of (annual) electricity production costs into capital cost, operating and maintenance cost, fuel cost and logistics. For thinnings the fuel costs represent half the electricity production costs. All other fuels in the minimum-cost case show that strongly negative costs of biomass wastes compensate all other costs. Figure 5 shows the electricity production costs in ECU/kWh assuming that the negative value of the fuel (that represents waste treatment costs) serves as income to the plant. This leads to wide ranges in, and potentially negative, electricity production costs.

The costs of electricity (COE) cover a wide range, namely from minus 6.7 up to plus 8.6 ECUct/kWh. When the fuel costs are set at zero, electricity costs are 2.9-4.8 ECUct/ kWh, compared with 4 ECUct/kWh for average Dutch electricity production in 1994. 38

Figs 6 and 7 show the sensitivity of the electricity production costs to variation in various parameters. The best way of reducing the COE is to increase the system efficiency. An increase in efficiency to 50% will bring the COE down by 25%. Such an improvement is possible with improved system integration and gas turbines (possibly with intercooling). A high load factor (and high reliability) is crucial for obtaining low electricity production costs.

c- O e ~ O 4000000 3000000 2000000 1 0 0 0 0 0 0 l

0 ~1_ 1l_

12

max min max

Personel I l , y'r

N

i =

i2

min max rain max

[ ] Dolomite consumption NaOH consumption • Insurance

Fig. 3. Breakdown of calculated minimum and maximum operating and maintenance costs o f electricity production with the selected A C F B - C C system as a function o f the fuel. Differences are caused particularly by ash disposal costs, which for thinnings can be zero when the ash is used as fertilizer,

although this is not shown in the graph.

Another important outcome is the low sensitivity of kWh costs to the transportation distance. Selected scenarios for various fuels already include substantial transport distances, but even when biomass is transported from all over the Netherlands (100 km diameter) the kWh costs are only modestly affected. The COE are obviously dominated by the fuel costs.

Waste treatment costs are calculated by considering the value of electricity produced by the plant. Reimbursement levels for decentrally produced power are 2.42 ECUct per kWh produced and 98 ECU per kW installed per year

in the Netherlands. 39'4° The results for waste treatment are given in Table 13, although for thinnings this stream should not be seen as waste. In several cases the reimbursements paid for decentralised power production in the Netherlands, which thus serve as income to the facility, outweigh the costs of the plant operation. This results in negative waste treatment costs for all minimum-cost cases.

The waste treatment costs (and efficiency) are compared with other state-of-the-art waste treatment options for organic waste in Table 14. From the point of view of both efficiency and

Table I 1. Minimum cost scenarios for logistics (all transport by road and all pretreatment centralized at the gasification plant)

Demolition

Thinnings" Verge grass O D W ¢ wood b Sludge

Assumed moisture content (wt%) d 50 60 54 15 20

Density (t/m 3 db) 0.15 0.16 0.5 0.213 0.56

Average transport distance (km) (two-way) 150 30-50 89 58

Transport costs (ECU/t wet) ¢ 5.44 4.65-9.30 6.97-11.62 3.22 2.11

Transfer & storage costs (ECU/t wet) f 0.29 0.32 0.22 0.63 0.23

Total costs o f logistics 5.73 4.97-9.62 7.19-11.84 3.85 2.34

~Thinnings are expected to be delivered as chips. Partial storage at lower landing (in the forest) is assumed. bWaste wood is expected to be delivered in shredded form. Costs o f shredding are already included in the fuel costs (presented in Table 1). The material is supplied by specialized companies? -'.53

q~ransport costs for O D W are relatively high since they include the collection o f waste in residential areas and the high moisture content 36. Also for verge grass the costs are relatively high because o f high moisture content and inclusion o f hauling costs (mowing). 3.3~

dMoisture content assumed for transport costs. Especially for sludge and verge grass the moisture content can vary considerably.

'Road transport is in all cases more economic. Specific data for road transport in the Netherlands: for a capacity o f 25 t or 80 m 3 the costs are 0.91 ECU/km with an average speed o f 50 km/h. 33

fTransfer costs are 0A6 ECU/m 3 (capacity 170 m3/h) for a shovel and 0.11 ECU/m 3 (capacity 275 m3/h) for a crane. 33 One transfer is assumed for all fuels.

Gasification of biomass wastes and Table 12. General economic parameters and a s s u m p t i o n s

used in this study

M i n i m u m - c o s t M a x i m u m - c o s t

case case

Real interest rate (%) 4 6

Expected lifetime of plant

(years) 25 20

Load factor 0.85" 0.75 h

Construction time (years) 2 2

"7400 h h6750 h

costs, gasification appears favourable compared with the main alternatives currently available.

5. D I S C U S S I O N

The main arguments for selecting the GE LM 2500 and an ACFB gasification process were that the BIGCC system could be constructed in the near future and the system should be flexible enough to treat various biomass residues and wastes. This fixes the scale of the system and excludes other (pressurized and indirect) gasifi- cation processes. In the longer term, other systems should be considered as well, especially systems that are used for clean fuels only or for the production of methanol and hydrogen.

The modelling has been performed relatively statically. For example, the behaviour of the gas turbine (combustion temperature, mass flows, behaviour in part-load conditions and operation on LCV gas below 5.6 MJ/m 3) is dealt with relatively simply. However, more elaborate dynamic modelling is not useful at this stage, because further experimental data first need to be collected. Dynamic aspects of the system, such as behaviour with fluctuating fuel gas composition and heating value, have to be investigated by testing, e.g. on the pilot scale. A related issue is the extent to which the dryer can produce biomass with a constant moisture content and can be regulated to respond to fluctuations in the compostion (moisture and ash) of the biomass delivered. There is also the (slight) risk of dust explosion under certain conditions. Drying with flue gas is selected here since it seems to be the cheapest and simplest way to reduce the moisture content of bio- mass fuels. However, steam drying can pro-

vide a good (though somewhat more

expensive) alternative when flue gas drying meets problems.

Further improvements in the system investi- gated are possible. One-shaft arrangement, a modified turbine combustion chamber and

residues for electricity production 401

expander inlet, allowing higher combustion temperatures, higher steam temperatures and pressures, and especially scale-up are the main options to obtain higher efficiency and lower costs per kWh. Further system integration can lead to a better use of the available waste heat. In the longer term, intercooling of the gas turbine compressor can be an interesting improvement option. The constraints on ash and moisture might be relaxed to some extent by further system improvements. These im- provements include heat recovery from the gasifier ash, allowing higher-ash fuels, use of various waste heat sources for fuel drying that reduce waste-heat requirements from the flue gas and especially modified combustor design that could allow fuel gas with lower heating values.

From an environmental point of view the flue gas dryer is the most uncertain factor. Dust emissions can be controlled by using cyclones. Emission of hydrocarbons and possibly ammo- nia and other compounds might be too high unless precautions are taken. Experimental data are needed so that the emission levels for drying can be confirmed. Additional filters (for reducing dust and odour) might be necessary. Steam drying can also be considered, for it will hardly influence system efficiency, although it will increase the investment costs to a limited extent. 2•

Investment costs are mainly based on vendor quotes. Uncertainties are included by present- ing cost ranges and differences in engineering costs. The high-cost estimate (2300 ECU/kW, for the most contaminated fuel) seems repre- sentative for a first fully commercial plant. The low-cost figure (1500 ECU/kW) is an estimate of the obtainable cost level after a number of identical plants with this capacity (30 MWe) have been constructed. For comparison: Elliott and Booth t2 suggest a cost level of 3000 U$/kWo installed (2230 ECU/kW) for a first BIGCC (25 MWe), potentially falling to 1300 U$/kWo installed (970 ECU/kW0) for the tenth identical plant.

Not only lower investment costs but es- pecially a further increase in efficiency will have a significant influence on the costs per installed kWe. The low-cost estimate presented should therefore not be seen as the cost level for the longer term. To obtain insight into such figures, further study on long-term developments of system components and further system inte- gration as discussed above is required.