Experimental test on a novel dual fluidised bed biomass gasifier for synthetic fuel production

The results show the dependency of bed material circulation rate on the superficial gas velocity in the combustor, the bed material inventory and the aeration of solids flow between the bo

Experimental test on a novel dual fluidised bed biomass gasifier

for synthetic fuel production

K Göransson⇑, U Söderlind, W Zhang

Department of Natural Sciences, Engineering and Mathematics, Mid Sweden University, SE-871 88 Härnösand, Sweden

a r t i c l e i n f o

Article history:

Received 15 September 2009

Received in revised form 22 December 2010

Accepted 29 December 2010

Available online 12 January 2011

Keywords:

Allothermal gasification

Biofuel

Biomass

Gasifier and syngas

a b s t r a c t

This article presents a preliminary test on the 150 kWthallothermal biomass gasifier at Mid Sweden Uni-versity (MIUN) in Härnösand, Sweden The MIUN gasifier is a combination of a fluidised bed gasifier and a CFB riser as a combustor with a design suitable for in-built tar/CH4catalytic reforming The test was car-ried out by two steps: (1) fluid-dynamic study; (2) measurements of gas composition and tar A novel solid circulation measurement system which works at high bed temperatures is developed in the pre-sented work The results show the dependency of bed material circulation rate on the superficial gas velocity in the combustor, the bed material inventory and the aeration of solids flow between the bot-toms of the gasifier and the combustor A strong influence of circulation rate on the temperature differ-ence between the combustor and the gasifier was identified The syngas analysis showed that, as steam/ biomass (S/B) ratio increases, CH4content decreases and H2/CO ratio increases Furthermore the total tar content decreases with increasing steam/biomass ratio and increasing temperature The biomass gasifi-cation technology at MIUN is simple, cheap, reliable, and can obtain a syngas of high CO + H2 concentra-tion with sufficient high ratio of H2to CO, which may be suitable for synthesis of methane, DME, FT-fuels

or alcohol fuels The measurement results of MIUN gasifier have been compared with other gasifiers The main differences can be observed in the H2and the CO content, as well as the tar content These can be explained by differences in the feed systems, operating temperature, S/B ratio or bed material catalytic effect, etc

Ó 2011 Elsevier Ltd All rights reserved

1 Introduction

Synthetic fuel production from biomass is an important issue

from the viewpoint of climatic conventions and energy shortage

crises Synthetic fuels such as methane, DME, FT-fuels, and alcohol

fuels, as the second generation bio-automotive fuels, can be

pro-duced via gasification and synthesis based on various forest and

agricultural biomass residues For biomass, a S&M (small or

med-ium) scale bio-automotive fuel plant is preferable as biomass

feed-stock is widely sparse, inhomogeneous in size and shape, difficult

to be pulverized, and has relatively low density, low heating value,

and high moisture content[1] Direct gasification with air produces

a fuel gas of heating value 4–7 MJ/Nm3– not suitable for synthesis

of bio-automotive fuels Pure oxygen gasification generates a fuel

heating value of 10–12 MJ/Nm3, but an oxygen plant is needed,

which can be economic only for large-scale bio-automotive fuel

production plants[2] However, indirect (or allothermal)

gasifica-tion, in a dual fluidised bed gasifier (DFBG) with steam as the

gasification agent, produces a syngas of 12–20 MJ/Nm3heating

va-lue Thus, DFBG turns out to be a promising biomass gasification technology for synthetic fuel production

There are a number of DFBGs or similar designs worldwide: the well known Güssing gasifier in Austria, e.g.[3], MILENA gasifier in The Netherlands[3–5], Trisaia gasifier in Italy by ENEA’s research center[3,6], Battelle Columbus Laboratories (BCL) gasifier in USA

[3] (now called The Rentech-SilvaGas Process [7]), and the CAPE FICFB Gasifier in New Zealand by the University of Canterbury[8]

A DFBG of 2–4 MWthwas recently built by Göteborg Energi at Chal-mers University of Technology in Sweden[9] In Japan there is a DFBG in Yokohama, by Xu et al.[10,11] and in China there are some examples in Beijing by the Chinese Academy of Sciences

[3]and in Hangzhou by Fang and co-workers[12] DFBGs can be designed with different combinations of the BFB and the CFB So far, the most attractive design is supposed to have biomass gasifi-cation in the BFB and char combustion in the CFB Examples of this design are the Trisia gasifier, the Güssing gasifier, the CAPE FICFB gasifier, the MIUN gasifier, etc The principle of the Chalmers gas-ifier is similar to these gasgas-ifiers as well, as a BFB gasgas-ifier is inte-grated into the loop of an existing 12 MWthresearch CFB-boiler

[13] The MILENA gasification process uses the riser for gasification and the BFB for combustion[4], and the Rentech-SilvaGas Process consists of two CFBs interconnected with each other[7]

0016-2361/$ - see front matter Ó 2011 Elsevier Ltd All rights reserved.

⇑Corresponding author Tel.: +46 70 289 26 30, +46 611 862 13; fax: +46 611 861

60.

E-mail address: kristina.goransson@miun.se (K Göransson).

Contents lists available atScienceDirect

Fuel

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

In general, the biomass gasification process occurs through

three steps: (1) pyrolysis, which produces volatile matter and char

residue; (2) secondary reactions, involving the volatile products;

(3) gasification reactions of the remaining carbonaceous residue

with steam and carbon dioxide[14] The pyrolysis of biomass

re-sults in volatiles and char that subsequently participate in a series

of complex and competing reactions The main conversion of the

biomass to syngas takes place within the bed, but some conversion

to syngas occurs in the freeboard section In DFBGs, biomass is

gas-ified in a bed fluidised with steam The composition of the syngas is

mainly dependent on the type of gasification agent used Steam

encourages the shift reactions with carbon and carbon monoxide

to increase the hydrogen content Shift reactions and steam

reforming of methane reaction are common reactions used to

pre-dict the composition of syngas

Synthetic fuel production requires a high quality syngas of high

CO + H2concentration with sufficient high ratio of H2 to CO and

low tar content Steam-to-biomass (S/B) ratios, gasification

temper-ature and bed material circulation rate, are important parameters in

determining the syngas composition and tar content The required

heat for biomass steam gasification is maintained by the char

combustion and the bed material circulation The bed material

circulation rate can control the temperature balance, and is an

important parameter to be considered in the gasifier design and

operation The operation of the gasifier needs to be optimised, and

hence, the operational behaviour of the gasifier needs to be studied

Attempt to measure the solid circulation rate have been

per-formed both in old proven ways like ‘‘bucket-and-stopwatch’’

and in new imaginative ways, e.g by using signals from an optical

mouse to PC[15]or a fibreglass spiral with a rotation electronics

[16] Unfortunately, most of the existing techniques for measuring

solid circulation rate are limited for various reasons A novel solid

circulation measurement system is developed in the presented

work, which is called here as Pressure-induced Measurement of

Circulation (PIMC), a technique that also works at high bed

tem-peratures under gasification conditions

The 150 kW allothermal biomass MIUN gasifier has been built

up in 2008 for the research on synthetic fuel production This paper

presents a preliminary test of the operational behaviour of the

MIUN gasifier The test was carried out by two steps: (1)

fluid-dy-namic study; (2) measurements of gas composition and tar The

experimental results of MIUN gasifier have been compared with

other DFBGs, which leads to a discussion on DFBG design and

oper-ation This paper shows a successful development of a pilot-scale

DFBG with internal reforming of tar and methane for synthesis

gas production

2 Experimental facilities

The BTL (biomass to liquids) system at MIUN is sketched in

on the gasifier is given below

2.1 The gasifier and test conditions The gasifier (seeFig 2) consists of an endothermic steam fluid-ised bed gasifier and an exothermal CFB riser combustor, and has a biomass treatment capacity of 150 kWth, i.e approx 30 kg wood pellet feed per hour

The heat carrier between the reactors is silica sand of about

150lm diameter The bed material circulation is controlled with the gas velocity in the combustor, the total solids inventory and aeration in the tube connecting the bottoms of the gasifier and the combustor The gas flow through the gasifier can also en-hance the aeration in the abovementioned tube, so that the bed material circulation increases with gas velocity in the gasifier

The PLC-operated (Process Logic Controller) feeding system is designed for constant biomass feeding The biomass is fed into a pneumatic oscillatory vane feeder to achieve high precision A coaxial left- and right-handed thread screw feeder thereafter makes a steady stream of biomass into the final screw feeder with-out plugging the pipe The final screw feeder is rapidly feeding bio-mass into the gasifier to avoid reaction in the screw feeder and particle congestion The feeding rate can be controlled by fre-quency-controlled motors of screw feeder Nearest to the gasifier, the final screw feeder is cooled by water flow through a cylindrical jacket

The gas distributors in both the gasifier and the combustor are own-produced sintered metal plates The fluidisation agent in the gasifier is steam (12 bar and 150 °C) and the synthesis gas is drawn off from the top of the gasifier The residual biomass char

is then transferred by bed material into the combustor through the lower pressure lock In the combustor, the fluidisation agent

is air, which results in an oxidation of the char and produces heat

at a temperature of 950–1050 °C The hot bed material separates from the flue gas in a cyclone to be recycled into the gasifier through the upper pressure lock (loop seal pot), which prevents gas leakage between the separate environments in the gasifier and the combustor The aeration medium in the upper pressure lock is steam

The steam gasifier is surrounded by electrical heaters (total ef-fect of 20 kW) and insulated There are no lining in the reactors The gasifier and the combustor have a height of 2.5 and 3.1 m, and inner diameters (i.d.) of 300 and 90 mm, respectively Temperatures and pressures, at a number of points from the distributor to the top of the gasifier and the combustor,

at the upper pressure lock and at the cyclone, as well as all gas flows, are registered through a computer data collection system

The test on the DFBG was carried out at the temperatures 750,

800 and 850 °C, in the fluidised bed gasifier The steam supply into the gasifier was held at 4 kg/h The biomass feedstock was wood pellets from SCA BioNorr AB, and the fuel analysis is given in

Nomenclature

CFB circulating fluidised bed

DFBG dual fluidised bed gasifier

Gs solid circulation rate (kg/m2s)

PIMC Pressure-induced Measurement of Circulation S/B steam/biomass ratio (kg/kg dry biomass)

Uc superficial gas velocity in the combustor (m/s)

Ug superficial gas velocity in the gasifier (m/s)

The steam/biomass ratios in weight were 0.3, 0.6 and 0.9 calcu-lated according to Eq.(1) The biomass input was 95, 39 and 27 kW respectively The surface of bed material in the gasifier was 0.7 m above the distributor, and about 0.2 m above the biomass feeding point

steam=biomass ratio ¼m_steamþ _mH2 O in biomass

_

mdry biomass

ð1Þ

where _msteam represents the mass flow of steam (kg/s), _mH 2 O represents the mass flow of water in the biomass (kg/s), _

Fig 2 The MIUN dual fluidised bed gasifier.

Table 1 Fuel analysis of wood pellets from SCA BioNorr AB [18]

Elementary analysis % db

Sulphur, S <0.01 Chlorine, Cl <0.01

Nitrogen, N <0.1 Oxygen, O (calc.) 42.7

Durability of pellets 98.1%

Bulk density 667 kg/m 3

Calorific value as rec 18.849 MJ/kg Calorific value db 20.159 MJ/kg Fig 1 The BTL system at MIUN with (A) biomass feeding system, (B) gasifier, (C) syngas cleaning system, and (D) catalysis reactor.

2.2 The solid circulation measurement system,

According to Bernoulli’s equation, the total pressure (pt) of a flow

is the sum of the static pressure (ps), the hydrostatic pressure (ph)

and the dynamic pressure (pd) in the flow The dynamic pressure is

associated with the gas velocity, and represents the kinetic energy

of the flow A common form of Bernoulli’s equation is as follows:

pt¼ psþ phþ pd

ph¼qgh pd¼qu

2 2 whereqrepresents the density of the fluid (kg/m3), g represents the

gravitational acceleration (m/s2), h represents the elevation above

the reference plane (m), u represents the velocity of the fluid (m/s)

As shown inFig 3, the solid circulation rate (Gs) is measured by

the solid circulation measurement system PIMC, with two pressure

transducers Purge gas is used to create a dynamic pressure (pd)

The positive side on each transducer is connected to a purge gas

nozzle inside the downcomer under the combustor cyclone (see

trans-ducer B is connected to purge gas nozzle B A glass tube, that is

holding a certain height of a water column, is connected to the

transducer’s negative side The water column imposes a constant

hydrostatic pressure so that the transducers indicate a negative

pressure in normal condition

By turning off the fluidisation of the upper pressure lock, the circulation in the gasifier is interrupted and the solids accumulate

in the upper pressure lock Hence, the level of bed material will start to rise in the downcomer When the bed material reaches and blocks the nozzle B and then the nozzle A in sequence, the purge gas entering the downcomer is stopped That creates a dy-namic pressure in nozzle B and then in nozzle A in sequence on the positive side of the transducer As a consequence, the pressure registered by the transducers will turn to positive in sequence Thus, the time delay of the pressure changes between the two noz-zles is registered and used to indicate the time used for bed mate-rial to fill the specific volume in the downcomer The distance between the two nozzle locations inside the downcomer (40 mm i.d.) is 200 mm

Hence, the solids circulation rate can be calculated as below

Gs¼ md

t  Ar where Gsrepresents the solid circulation rate (kg/m2s), md repre-sents the mass of the bed material in the downcomer between noz-zle B and noznoz-zle A (kg), Arrepresents the cross-section area of the riser (m2), t represents the time used for bed material to fill the downcomer between nozzle B, and nozzle A (s)

2.3 Measurements of gas composition The major syngas stream from the gasifier is led to a special gas burner for complete combustion of the syngas A minor stream of the syngas passes through the gas sampling and analysis system

To avoid tars condensation in the pipes, electrical heaters outside the pipes hold the temperature of the pipes at about 400 °C

An overview picture of the syngas sampling and analysis system can be seen inFig 4 The sampling procedure was repeated three times for each experimental run

For the measurements of the gas composition, the syngas is drawn by a vacuum pump through a gas conditioning step and sampled manually in a gas sampling bag (Cali-5-bond) and ana-lysed off-line in a parallel FID and TCD detection GC-system 2.4 Tars measurement

Tars can be divided into heavy tars and light tars Heavy tars comprise high molecular weight hydrocarbon compounds that have a boiling point higher than 350 °C Light tars comprise the range in molecular weight from 78 to 300, which are volatile and semi-volatile aromatics and phenolics The tar sampling takes place at the same time as the syngas sampling, as seen inFig 4 The total tar and light tar are measured in two different ways with gravimetrical analysis and SPA method respectively

The total tars are gravimetrically analysed First the product gas passes a high temperature filter usually kept at 380 °C during sam-pling, and cooled down to 40 °C while passing through two tar cap-ture glass fibre filters in series A known volume (4 l) of gas measured by a flow meter was drawn by a vacuum pump through the sampling during about 45 s After sampling the glass fibre fil-ters were frozen in aluminium foil envelopes, which later were placed in a clean round flask containing isopropanol After about

18 h the glass fibre filters and the parts of the foil were washed out with isopropanol The solution containing the tar sample to-gether with some particles and filter fibres is filtered through a glass fibre funnel to a new round flask that has been weighted The round flask is inserted into an evaporation condenser (rotation

60 rpm in water bath 45 °C) When the tar sticks on the inside of the round flask the evaporation is finished The weight difference, i.e the tar content, can finally be calculated

Fig 3 Pressure-induced Measurement of Circulation (PIMC), the solids circulation

Fig 4 An overview picture of sampling and analysis of syngas and tars, with (A) gasifier, (B) syngas burner, (C) syngas sampling, (D) total tars sampling (E) light tars sampling (SPA method), (F) ventilation (The syngas sampling with (1) impinger bottles, (2) cooler, (3) drying filters, (4) vacuum pump and (5) the outlet to GC-system; the total tar sampling with (6) hot gas filter, (7) glass fibre filters, (8) vacuum pump, (9) flowmeter and (10) total flowmeter; the SPA tar sampling with (11) Amino phase cartridge and (12) syringe 100 ml.).

The light tars were analysed according to the Solid-Phase

Adsorption (SPA) method The samples were taken by a septum

port of an electrically heated T-connection (400 °C) inserted in

the sampling line, in the same time as the total tars sampling as

shown inFig 4 A known amount of gas is extracted by using a

100 ml syringe filled at one minute The tar vapours are trapped

on amino propyl-bonded silica sorbent packed in a small cartridge

which was later sent to Royal Institute of Technology at Stockholm

for analysis Light tars can pass through a nonpolar GC-column

while heavy tars cannot The SPA method is used for compounds

which are enough volatile to be separated on a GC-column, and

is calibrated for 18 aromatics and 10 phenolics

3 Results and discussion

3.1 Bed material circulation

To maintain a sufficient heat transfer between the gasifier and

the combustor, the solid circulation rate, Gs, must be held at a

cer-tain high level Gsis mainly determined by the gas velocity in the

combustor, Uc, and the bed material inventory Gsis also affected

by the gas velocity in the gasifier, Ug, since it is serving as an

aer-ation of the lower pressure lock The lower pressure lock is

con-necting the gasifier and the combustor

con-stant at 3.5 m/s Gs increased gradually with Ug The results in

superficial gas velocity in the combustor and the bed material

inventory The bed inventory strongly influences Gs due to the

change of solids concentration in the upper part of the

combustor

A circulation rate of 20 kg bed material per 1 kg biomass is

re-quired in normal gasifier operation condition At the full load of

the 150 kW gasifier, i.e approx 30 kg wood pellets feed per hour,

a solid circulation rate of 600 kg bed material per hour is needed,

which corresponds to a bed material circulation rate of 26 kg/m2s

The temperature difference between the gasifier and the

com-bustor is strongly dependent on Gsas seen inFig 7 A high Gsgives

a small temperature difference and vice versa Gsvalues that

ex-ceeded 10 kg/m2s is required for a sufficiently low temperature

difference

3.2 Syngas composition

The pyrolysis of biomass is the first step of biomass gasification,

which results in volatiles and char The volatile and char

subse-quently participate in a series of complex and competing reactions,

as given below[14,19,20]

Oxidation:

C þ1

Boudouard:

Water–gas (steam oxidation):

Methanation:

Water–gas shift:

Steam reformation:

According to Franco et al.[14], the produced gas composition in steam gasification at temperatures in the range of 730–830 °C is dependent on the water–gas shift reaction(G), together with the reforming and cracking reactions

At temperatures higher than 830 °C, the water–gas shift reac-tion becomes less important as the Boudouard reacreac-tion (C) and the water–gas reactions(D) and (E)become dominant

The syngas measurement results from the MIUN gasifier are summarized inTable 2with the averaged values of syngas components

0

2

4

6

8

10

12

14

16

18

20

Gas Velocity in Gasifier (m/s)

2 s)

Bed inventory: 145 kg

Uc: 3,5 m/s

0,0 2,0 4,0 6,0 8,0 10,0 12,0 14,0 16,0 18,0

Gas Velocity in Combustor (m/s)

Bed inventory: 145 kg Bed inventory: 130 kg

Fig 6 The solid circulation rate as a function of bed inventory and the gas velocity

in the combustor.

60 80 100 120 140 160 180 200

4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0

Solid Circulation Rate (kg/m 2 s)

o C)

Fig 7 The temperature difference between the gasifier and the combustor as a function of the solids circulation rate.

for each experimental condition In the temperature range of 750–

850 °C, the H2 yield increases with S/B, while slightly decreases

with temperature The CO and CH4yields decrease with S/B, but

are fairly stable within the given temperature range

The decreasing trend of the H2yield with temperature may be

governed by the exothermic water–gas shift reaction as a higher

temperature pushes the reaction to the left hand at the expense

of H2 Those endothermic reactions including the Boudouard, the

water–gas and the steam reforming reactions are in favour of H2

yield, but have not been dominating reaction network under the

present low temperature experimental conditions A higher S/B

ra-tio means a higher steam partial pressure that pushes the water–

gas shift reaction to the right hand for more H2production at the

expense of CO Thus, the CO yield decreases with S/B All reactions,

including Boudoyard, water–gas, water–gas shift and steam

refor-mation reactions, suggest an increase of the CO yield with

temper-ature But the temperature effect have not been significant within

the given temperature range A higher gasification temperature

will be in favour of the H2and CO yields since the endothermic

gas-ification reactions(C), (D), (E), and (H)are enhanced

At low S/B ratio the carbon is only partially gasified A higher S/B

ratio results in greater carbon conversion as a result of the

en-hanced water–gas shift reaction and steam reformation reactions

Thus the methane and carbon monoxide contents decrease, and

the hydrogen and carbon dioxide contents increase When carbon

conversion is completed, an increase of the S/B ratio just results

in dilution of the syngas[21] Within a range of S/B ratios 0.3–

0.9, as seen inTable 2andFig 8, the methane and carbon

monox-ide contents decrease whereas the hydrogen and carbon dioxmonox-ide

contents increase with increasing S/B ratios This result is in good

agreement with the Refs.[22,23]

When the syngas is used for synthesis of transportation fuel such as FT-diesel, DME, and alcohols, a high CO + H2concentration (>80%) with a high H2/CO ratio in the syngas is required to ensure downstream synthesis smoothly The stoichiometric H2/CO ratio is

2 for the synthesis reaction of Fischer–Tropsch and Methanol, 2.5 for DME and 3 for Methanation (Bio-SNG) The stoichiometric ratio

is the optimal H2/CO ratio to obtain a maximum yield of the prod-uct In practise the ratio differs from the optimal ratio, but a certain high H2/CO is still necessary[24]

the higher concentration at higher S/B ratio, and the H2/CO ratios from 1.0 to 2.2, the higher ratio at higher S/B ratio as seen in

ratio of H2to CO at high S/B ratio That is insensitive to the temper-atures within the given temperature range

H2/CO adjustment can be achieved with water–gas shift reac-tion; either by an additional step in a separate reactor, or in terms

of DFBGs, by using the so-called Absorption Enhanced Reforming (AER) process The AER process uses calcium carbonate as bed material for in situ CO2capture during gasification The continuous

CO2removal during the gasification enhances the reforming reac-tions and the water–gas shift reaction towards H2production[24] The CH4 content is not sensitive to the temperature change within the given range (750–850 °C), but decreases significantly with increasing S/B, as seen inFig 8as a result of the enhanced steam reformation of methane The CH4range from 9.5% to 13.5%

in the syngas represent 1/3 of energy contained in the syngas, and would be lost in once through synthesis of transportation fuels

or chemicals Such CH4content must be reduced by thermal crack-ing and catalytic reformcrack-ing to CO and H2before syngas entering the synthesis reactors

3.3 Tar content Tars in the syngas not only results in a part of energy lost from syngas before synthesis, but also leads to many knotty problems to downstream processes, blocking of filters and pipeline passages, dirty working environment, heavy waste treatment, poisoning of catalyst in synthesis reactors and so on

A good gasifier should provide syngas with tar content under 1–

2 g/m3to facilitate the downstream syngas cleaning After syngas cleaning, tar content should be under 10 mg/m3for engine and tur-bine applications, and under 1 ppmv for most synthesis reactors However, as seen inFig 9, the content of tar from MIUN gasifier with sand bed material ranges from 10 to 65 g/Nm3, although the tar content decreases as S/B and temperature increase Increasing S/B means that the conversion of the H2O in the gasifier gradually decreases, and hence a greater proportion of steam bypass the gas-ifier unused[25]

Table 2

Syngas composition (vol.% dry.) as a function of the steam/biomass (S/B) ratio at 750,

800 and 850 °C.

0,0

2,0

4,0

6,0

8,0

10,0

12,0

14,0

16,0

Steam/Biomass ratio (S/B)

0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 4,5 5,0

H 2

850°C

800°C

750°C

750°C

800°C

850°C

Fig 8 The H 2 /CO ratio and the concentration of methane as a function of S/B ratio

0 10 20 30 40 50 60 70

Steam/Biomass ratio (S/B)

750 o C

800 o C

850 o C

As seen inFig 10, the light tars content decreases with

increas-ing S/B ratio, but unlike the total tars content, the light tars content

increases with increasing temperature – with the exception of the

case at S/B ratio 0.9, where the light tar content is lowest at the

temperature 800 °C

The light tars content increases with temperature because the

heavy tars are cracked to light tars as the temperature increases

This trend can be identified byFig 11, where the high molecular

weight compounds of the light tars like phenol are decreasing with

increasing temperature while the low molecular weight

com-pounds like naphthalene are increasing The exceptional case could

be attributed to the co-action of light tar reduction by thermal

cracking at higher temperatures and steam reforming

For MIUN gasifier, the tar content is too high to be used for

syn-thesis applications, and should be reduced further The tar

reduc-tion can be realized by (1) optimum of gasifiers as described

above, (2) catalytic in-bed material, and (3) downstream measures

Catalytic bed material (such as olivine, dolomite, nickel or iron) is

often used to crack down tar inside gasifiers as a primary method

Although primary measures are of importance, to keep the

thermo-dynamic efficiency losses to a minimum, downstream methods,

such as external reforming, thermal cracking, catalytic cracking

or mechanical separation (such as scrubber, filter, electrostatic

pre-cipitator or cyclone separator), are required when the syngas is

used for synthesis

A scrubber is easy to run and has beneficial effect on air

pollu-tion control The scrubber agent can be water or a scrubbing liquid

compatible with tar In terms of tar scrubbing with organic liquid

solvents, the tars are recycled to gasifier and destructed avoiding

wastewater production The mechanical separation technologies

are often applied in combination or together with catalytic tar

re-moval technologies

3.4 Comparison of the MIUN gasifier with other indirect gasifiers

A comparison of the MIUN gasifier to the Güssing gasifier, the MILENA gasifier, the CAPE gasifier, and the Chalmers gasifier, is gi-ven inTable 3, with regards to syngas composition and tar content The main differences in the syngas of the gasifiers can be observed

in the H2and the CO content The H2/CO ratios of the MIUN gasifier and the Güssing gasifier are high (1.7 and 1.8–1.9 respectively) The content of H2in the syngas of the MILENA gasifier, the CAPE gasifier and the Chalmers gasifier are quite low, which leads to the H2/CO ratio close to 0.5–0.8 The CH4content in the syngas of the MILENA gasifier is slightly higher than in the other gasifiers The Chalmers gasifier and the MIUN gasifier have similar tar con-tents (20 g/Nm3) The Güssing gasifier has the lowest tar content (4–8 g/Nm3) while the MILENA gasifier has the highest tar content (40 g/Nm3) The CAPE gasifier has a high content of N2in the syn-gas, probably caused by the nitrogen purge gas in the biomass feeding system

The water–gas shift reaction and the steam reforming of meth-ane reaction are common reactions used to explain changes in syngas composition The most important parameter governing the reactions are S/B as has been discussed previously The first reason behind the difference in H2/CO ratios is that the S/B ratio used in the MIUN gasifier and the Güssing gasifier is higher than that for the MILENA gasifier, the CAPE gasifier and the Chalmers gasifier

The second reason is attributed to the difference in the biomass feeding location Biomass were fed in the beds of the MIUN gasifier and the Güssing gasifier, but on the top of the beds for the Chal-mers gasifier and the CAPE gasifier The location of the feeding point to the gasifier influences the product distribution, since the heating rate of the biomass depends on the location where it is fed[25] High heating rates produce more light gases and less char and condensate Improved mixing of the biomass and the bed material, with higher temperature and longer residence times will drive the syngas composition in the direction of equilibrium[26] When the feed point is situated above the bed surface, the rela-tively low density of the biomass particles precludes good mixing with the bed material[26]

The poor contact between the volatiles and the bed material and extremely low temperature in the freeboard lead to a limited water–gas shift reaction so that a certain amount of CO has not been shifted to H2by steam, which gives a low H2/CO ratio The differences in the H2/CO ratio can also be attributed to the factors such as the type and the size of the biomass and of the bed material (for instance, olivine influences the water shift equilib-rium), and some variations of the sampling procedure

The CH4content in the syngas of the MILENA gasifier with a low S/B ratio is slightly higher than in the syngas of the other gasifiers

0

5

10

15

20

25

30

Steam/Biomass ratio (S/B)

850 800 750

Fig 10 The light tar content as a function of S/B ratio at 750, 800 and 850 °C.

0 1 2 3 4 5 6

Benz e

Tolu

ene m -Xyl

ene o-X yleneInda n

Inde

Naph tha

lene

2-Me th

napht halen e

1-M

hylna

phtha lene

Bip he l

Ac aph thy

lene

Ace

phth ene

Fluo

rene

Phe nth e

Anthr

ene

Fluo

rante ne

Pyr ene

Phen ol o-C res ol

m-C

resol p-C

resol

2, 3,5 -Xyle 2,4 -Xyl

enol 2,6 -Xyl

enol

3,

4-Xyl eno l

750°C S/B 0,3 800°C S/B 0,3 850°C S/B 0,3

That is reasonably when the carbon is only partially gasified at low

S/B ratio

The differences in the tar content can be explained based on the

influences of the bed material, the gasification temperature and

also the S/B ratio The Güssing gasifier uses a tar cracking active

bed material (olivine) at a high (850–900 °C) gasification

tempera-ture The MILENA gasifier uses sand and a low S/B ratio

In a fluidised bed gasifier, the temperature in the freeboard is

lower than in the bed At higher temperatures the endothermic

gasification reactions (such as the Boudouard reaction (C), the

water gas reactions(D) and (E)and the steam reforming reaction

(H)) are enhanced This give rise to a higher H2and CO contents

and a lower tar and methane contents

4 Conclusion

A preliminary test on the 150 kW allothermal biomass gasifier

at Mid Sweden University was carried out A novel method to

mea-sure solid circulation rate is developed, which works well at high

temperature condition The test provides basic information for

temperature control in the combustor and the gasifier by the bed

material circulation rate The circulation rate is, in turn,

deter-mined by the superficial gas velocity in the combustor, the bed

material inventory, and the aeration of the lower pressure lock

Measurements of gas composition as a function of temperature

and the steam/biomass ratio have been performed in the

temper-ature range 750–850 °C The results shows a high CO + H2

concen-tration (>80%) in the syngas The H2yield decreases while the CO2

yield increases slightly with increasing temperature The variation

of temperature in this range, has a negligible impact on the yields

of CO and CH4

Within a range of S/B ratios 0.3–0.9, the test shows that the

hydrogen content increases while the concentration of methane

decreases as the S/B ratios increase The total tar content decreases

with increasing S/B ratio and increasing temperature The light tars

content decreases with increasing S/B ratio, but unlike the total

tars content the light tars content increases with increasing

tem-perature Most likely some of the heavy tars are cracked to light

tars as the temperature increases

A comparison of the MIUN gasifier with other indirect gasifiers

has been carried out The main differences in the syngas of the

gas-ifiers can be observed in the H content and the CO content The

H2/CO ratios of the Güssing gasifier and the MIUN gasifier are high (1.8–1.9 and 1.7 respectively), while the MILENA gasifier, the Chal-mers gasifier and the CAPE gasifier have a relatively low content of

H2in the syngas That can partly be explained by differences in the feed systems and the operating temperature

Finally, differences in the tar content can be seen The Chalmers gasifier and the MIUN gasifier have similar tar contents (20 g/

Nm3) The Güssing gasifier has the lowest tar content (4–8 g/

Nm3) while the MILENA gasifier has the highest tar content (40 g/Nm3) The probable reason is differences in the bed mate-rial catalytic effect, the gasification temperature and the S/B ratio

Acknowledgements The authors would like to acknowledge the project support from the EU regional structure fond, Ångpanneföreningen Founda-tion for Research and Development, LKAB, Länsstyrelsen Väster-norrland, FOKUSERA, Härnösands Kommun, Toyota, SCA BioNorr

AB and SUNTIB AB

References [1] Zhang W, Forsman B, Söderlind U Biofuels in Sweden In: 7th World congress

on recovery, recycling and re-integration, Beijing, China; September 2005 [2] Zhang W Transportation fuels from biomass via gasification In: 2nd World conference on biomass for energy, industry and climate protection, Rome, Italy; 10–14 May 2004.

[3] Corella J, Toledo JM, Molina G A review on dual fluidized-bed biomass gasifiers Ind Eng Chem Res 2007;46:6831–9.

[4] van der Meijden CM, Veringa HJ, van der Drift A, Vreugdenhil BJ The 800 kWth allothermal biomass gasifier MILENA In: 16th Biomass conference & exhibition, Valencia, Spain; 2–6 June, 2008.

[5] van der Meijden CM, Veringa HJ, Bergman PCA, van der Drift A, Vreugdenhil BJ Scale-up of the MILENA biomass gasification technology In: 17th European biomass conference and exhibition, Hamburg, Germany; 2009.

[6] Molino A, Alvino E, Freda C, Mannarino V, Petrone MT, Blasi A, et al Simulation

of a molten carbonate fuel cell integrated into a biomass steam gasification pilot plant In: FP6 bigpower SSA project biogastech workshop, Gebze, Turkey; 9–11 April 2008.

[7] Sim Weeks Rialto renewable energy center – synthetic fuels and renewable electric power from biomass’’ presentation in biorefinery deployment collaborative, Durham, NC; 10/2/2009.

[8] Bull D, Gilmor I, Williamson C, Pang S The development and operation of a

100 kW dual fluidised bed biomass gasifier for production of high quality producer gas In: ICPS 09 (international conference on polygeneration strategies), Vienna, Austria; 2009.

Table 3

Steam gasification syngas composition (vol.% dry.).

8 MW th [23,24]

MILENA 800 kW th

[5,27,28]

CAPE FICFB

100 kW th [8,19]

Chalmers 2–4 MW th

[29,30]

No info.

Tar (g/Nm 3

Feedstock Fuel feeding Wood pellets into the side

of the bed in the gasifier

Biomass chips fed into the gasification-reactor bed

Wood pellets into the riser (i.e the gasifier)

Wood chips/pellets on the top of the bed of the gasifier

Wood chips/pellets fed from the top of the gasifier

0.02–0.35 c

a

Raw gas composition according to Fig 4 in [5]

b Basic design data MILENA pilot plant [5]

c Normal range for the MILENA lab-scale gasifier [28]

d Nitrogen purge gas feed into the biomass feeding system (10 l/min) and air fluidising of the siphon [19]

[9] Thunman H Co-gasification of biomass in a coal fired power plant Presented

at biomass energy workshop of IVA-CAE project on cooperation of renewable

energy and environment, Nanning, China; 2009.

[10] Xu G, Murakami T, Suda T, Matsuzaw Y, Tani H Two-stage dual fluidized bed

gasification: its conception and application to biomass Fuel Process Technol

2009;90:137–44.

[11] Xu G, Murakami T, Suda T, Matsuzawa Y Reactor siphon and its control of

particle flow rate when integrated into a circulating fluidized bed Ind Eng

Chem Res 2005;44:9347–54.

[12] Fang M, Yu C, Shi Z, Wang Q, Luo Z, Cen K Experimental research on solid

circulation in a twin fluidized bed system Chem Eng J 2003;94:171–8.

[13] Seemann MC, Thunman H The new Chalmers research-gasifier In: ICPS 09

(international conference on polygeneration strategies), Vienna, Austria; 2009.

[14] Franco C, Pinto F, Gulyurtlu I, Cabrita I The study of reactions influencing the

biomass seam gasification process Fuel 2003;82:835–42.

[15] Matsuda S Measurement of solid circulating rate in a circulating fluidized bed.

Powder Technol 2008;187:200–4.

[16] Ludlow JC, Monazam ER, Shadle LJ Improvement of continuous solid

circulation rate measurement in a cold flow circulating fluidized bed.

Powder Technol 2008;182:379–87.

[17] Göransson K, Söderlind U, Zhang W 16th European biomass conference &

exhibition, BTL Laboratory at Mid Sweden University, Valencia, Spain; 2–6

June, 2008.

[18] Stefan Rönnquist, SCA BioNorrAB, Härnösand, Sweden, via e-mail; June 2009.

[19] Brown J Biomass gasification: fast internal circulating fluidised bed gasifier

characterisation and comparison; 2006.

[20] Gao N, Li A, Quan C A novel reforming method for hydrogen production from

biomass steam gasification Bioresour Technol 2009;100:4271–7.

[21] Rutherford J, Williamson C Integrating advanced biomass gasifiers into the

New Zealand wood industry New Zeal J For 2006;51(3):35.

[22] Hofbauer H, Rauch R Stoichiometric water consumption of steam gasification

by the FICFB-gasification process Institute of Chemical Engineering, Fuel and Environmental Technology, Vienna, Austria.

[23] Pfeifer C Experiences and results derived from the dual fluidised bed gasifier

at Vienna University of Technology In: International conference on poly-generation strategies, Vienna, Austria; 2009.

[24] Hofbauer H Biomass for synthesis gas – overview In: 17th European biomass conference and exhibition, Hamburg, Germany; 2009.

[25] Corella J, Toledo JM, Molina G Biomass gasification with pure steam in fluidised bed: 12 variables that affect the effectiveness of the biomass gasifier Int J Oil Gas Coal Technol 2008;1(1/2).

[26] Rapagnà S, Jand N, Kiennemann A, Foscolo PU Steam-gasification of biomass in a fluidised-bed of olivine particles Biomass Bioenergy 2000; 19(3):187–97.

[27] MILENA gasification process ECN Energy research Centre of the Netherlands.

<http://www.ecn.nl/fileadmin/ecn/units/bio/Leaflets/b-08-021_MILENA.pdf> [28] van der Meijden CM, van der Drift A, Vreugdenhil BJ Experimental results from the allothermal biomass gasifier MILENA In: 15th European biomass conference, Berlin, Germany; 2007.

[29] Thunman H GoBiGas and other gasification activities at Chalmers, presentation <http://ciweb.chydenius.fi/project_files/HighBio-Projektiseminaari%20180510/Henrik%20Thunman%201805-2010.pdf> (18.05.10).

[30] Martin Seemann Chalmers University of Technology in Gothenburg, Sweden, via e-mail; August 2009.

[31] Doherty W, Reynolds A, Kennedy D Modelling and simulation of a biomass gasification-solid oxide fuel cell combined heat and power plant using Aspen Plus, Dublin Institute of Technology; 2009.