Influence of different bed material mixtures on dual fluidized bed steam gasification

Influence of different bed material mixtures on dual fluidized bed steam gasification

Accepted Manuscript Influence of Different Bed Material Mixtures on Dual Fluidized Bed Steam Gasification A.M. Mauerhofer, F. Benedikt, J.C. Schmid, ...

2MB Sizes 0 Downloads 37 Views

Accepted Manuscript Influence of Different Bed Material Mixtures on Dual Fluidized Bed Steam Gasification

A.M. Mauerhofer, F. Benedikt, J.C. Schmid, J. Fuchs, S. Müller, H. Hofbauer PII:

S0360-5442(18)31002-8

DOI:

10.1016/j.energy.2018.05.158

Reference:

EGY 12993

To appear in:

Energy

Received Date:

13 December 2017

Accepted Date:

24 May 2018

Please cite this article as: A.M. Mauerhofer, F. Benedikt, J.C. Schmid, J. Fuchs, S. Müller, H. Hofbauer, Influence of Different Bed Material Mixtures on Dual Fluidized Bed Steam Gasification, Energy (2018), doi: 10.1016/j.energy.2018.05.158

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Influence of Different Bed Material Mixtures on Dual Fluidized Bed Steam Gasification A. M. Mauerhofer1,*, F. Benedikt1, J. C. Schmid1, J. Fuchs1, S. Müller1, H. Hofbauer1 TU Wien, Institute for Chemical, Environmental and Biological Engineering, Vienna, 1060 Getreidemarkt 9/166 * Corresponding author: [email protected] E-Mail address: [email protected] (F. Benedikt), [email protected] (J. C. Schmid). [email protected] (J. Fuchs), [email protected] (S. Müller), [email protected] (H. Hofbauer) 1

Abstract Within this paper, investigations to convert softwood with four different types of bed materials in the 100 kWth dual fluidized bed steam gasification pilot plant at TU Wien are presented and discussed. The results of ten different experiments were compared. Quartz, olivine and feldspar were mixed with limestone in mass ratios of 100/0, 90/10, 50/50 and 0/100. Limestone was used due to its catalytic activity at high temperatures as CaO and thus enhanced tar, char and water conversion of quartz, olivine and feldspar. The admixture of limestone to quartz, olivine and feldspar shifted the product gas compositions towards higher hydrogen and carbon dioxide and lower carbon monoxide contents. By using 100 wt.-% limestone as bed material a hydrogen content of 47.4 vol.-% could be generated. Additionally, the tar concentrations as well as the tar dew points decreased and especially the heavy tar compounds could be reduced. Already small amounts of limestone (< 10 wt.-%) to the bed material mixture influenced tar reduction in a positive way. The low abrasion resistance of limestone resulted in increasing dust contents by increasing its amount. However, this could be balanced by the specific design of the separation system of the advanced pilot plant. Keywords: tar reduction, catalytic activity, limestone/CaO, olivine, quartz, feldspar

1. Introduction The worldwide coverage of electricity, heat and fuels increases the research on alternative feedstocks and technologies to enable a sustainable production in the future. Thus, the thermo-chemical conversion of biogenic feedstock is an auspicious way to promote an eco-friendly, sustainable supply of these basic goods in daily life. The dual fluidized bed (DFB) steam gasification, presented in Fig. 1, is a main subject of research at TU Wien. The DFB steam gasification consists of a gasification reactor (GR) and a combustion reactor (CR) and generates a nitrogen-free product gas by converting solid fuels. The nitrogen-free product gas mainly consists of hydrogen (H2), carbon monoxide (CO), carbon dioxide (CO2), methane (CH4) and ethylene (C2H4). This product gas can be utilized e.g. for generation of heat, electricity, energy, fuels for transportation [1] or synthetic chemicals, like mixed alcohols [2]. The two reactors, the combustion and the gasification reactor are connected through a circulating bed material. Via this bed material circulation it is possible to transport heat from the

combustion reactor to the gasification reactor, so that the endothermic gasification reaction is able to take place. The first 100 kWth DFB steam gasification pilot plant was established in the 1990s at TU Wien [3]. Afterwards, industrial-sized plants followed. In 2002, a demonstration plant with a fuel capacity of 8 MWth [4] in Güssing, Austria was built. Six years later a 8.5 MWth plant in Oberwart, Austria was constructed [5]. A 15 MWth plant in Senden, Germany [6] followed shortly afterwards. In 2014, a plant with a fuel power of 32 MWth was realized in Gothenburg, Sweden [7].

Fig. 1: Basic principle of the DFB steam gasification

ACCEPTED MANUSCRIPT List of abbreviations BTX CR DFB ECN GC/MS GR PLC vol.-% WGS wt.-% List of subscripts C CR db daf fuel GR h min PG th steam stp SC SF List of symbols a, b ṁ Q̇loss V̇ XH2O φSC φSF ηCG ηCG,o PGY LHV

Benzenz, toluene, xylene Combustion reactor Dual fluidzed bed Energy research Centre of the Netherlands Gas chromatography coupled with mass spectrometry Gasification reactor Programmable logic controller Volumetric percent Water gas shift Weight percent Carbon Combustion reactor Dry basis Dry and ash-free Fuel to gasification reactor Gasification reactor Hours Minutes Product gas Thermal Standard temperature and pressure Steam to carbon Steam to fuel Stoichiometric factors (-) Mass flow (kg/s) Heat loss (kW) Volumetric flow (m³/s) Steam-related water conversion (kgH2O/kgH2O) Steam to carbon ratio (kgH2O/kgC) Steam to fuel ratio (kgH2O/kgfuel,daf) Cold gas efficiency (%) Overall cold gas efficiency (%) Product gas yield Lower heating value

ACCEPTED MANUSCRIPT Typically, in industrial-sized operated plants, olivine is used as bed material [8], [9]. The reason for this is that olivine is a catalytic active mineral, which forms calcium-rich layers on the surface of the particles after days of operation. These layers, which are formed via the interaction of the bed material particles with the woody biomass ash, improve the catalytic activity and thus are favorable for tar reduction [10], [11]. However, due to the high price of olivine and limited regional existence, the research on alternative bed materials, which are cheap and easily available worldwide, is focused on within this paper. Therefore, limestone, quartz and feldspar were selected to be alternative materials to conventionally used olivine. In previous works, investigations of the DFB steam gasification with different bed materials were already carried out [12], [13]. However, the test runs were conducted in the conventional design, which is described in more

detail in [14]. Therefore, investigations in the advanced design of the 100 kWth DFB steam gasification pilot plant with different bed material types and bed material mixtures were carried out and compared within this work.

2. Materials and methods 2.1 Advanced DFB steam gasification In Fig. 2 the reactor concept of the advanced design of the dual fluidized bed steam gasification is displayed. Due to the continuous ongoing research on the reactor system, a solution was found to improve the gas-solid contact between the product gas and the hot bed material particles in the gasification reactor. This was one of the impulses to develop an advanced reactor design, which went into operation in 2014 [15].

Fig. 2: Sketch indicating dimensions (left) and 3D drawing of the reactor system (right)

Regarding the improvement of the gas-solid contact, the main focus of the advanced design of the pilot plant referred to the gasification reactor, which was divided into two parts. The lower part of the gasification reactor was designed as bubbling bed, while the upper part was operated as counter-current column with turbulent fluidized zones. Therefore, the column of the gasification reactor was equipped with constrictions, which led to an increased hold-up

of bed material over the height of the column. This allowed an increased residence time as well as an increased gas-solid contact of catalytic active bed material particles with the product gas. Supplementary, the prevailing higher temperatures in the counter-current column of the gasification reactor had a positive effect on tar reduction. Thus, the conversion efficiency was increased [16]. Another benefit of the advanced design related to the

ACCEPTED MANUSCRIPT separation system on top of the reactors. Due to the exchange of cyclones by gravity separators, the use of softer bed materials like limestone was possible. Limestone and CaO had a very low abrasion resistance, which could be balanced by using gravity separators instead of cyclones for the bed material separation. Arising fines (5 - 80 µm) after the gravity separators were removed via cyclones. Fig. 3 and 4 show pictures of the upper part with the fuel hoppers and the lower reactor part with the fuel feeding screw and several fine ash removal containers of the 100 kWth pilot plant at TU Wien. Coarse ash was removed in the lower part of the reactor system, which is especially essential in case of using ash-rich fuel types. First experimental results of the successful operation of the 100 kWth dual fluidized bed steam gasification of different fuels with different bed materials can be found in literature [17] - [20].

inner height of 4.7 m for the combustion and 4.3 m for the gasification reactor. The combustion reactor had an inner diameter of 125 mm. A control room was established, from which it was possible to control the pilot plant with a programmable logic controller (PLC). The PLC continuously measured and recorded data of all relevant flow rates, temperatures and pressures as well as the gas compositions (Rosemount NGA2000). Other gas components, like ethylene (C2H4) were analyzed by a gas chromatograph (Perkin Elmer ARNEL – Clarus 500) every 12-15 min. Before the product gas was measured online, it had to be cleaned. For this purpose it was filtered by a glass wool filter and washed with rapeseed methyl ester (RME) to remove condensable components, like water and tar. The setup and measurement devices are reported in detail within the work of Kolbitsch [21].

2.3 Offline measurement Tar was measured discontinuously by isokinetically taking samples with impinger bottles to condense and dissolve the condensable hydrocarbons (see Fig. 5).

Fig. 3: Upper part of the gasification reactor at TU Wien

Fig. 5: Measuring device of tar, char, dust and water sampling

Fig. 4: Lower part of the gasification reactor at TU Wien

2.2 Online measurement The advanced design of the DFB steam gasification pilot plant was implemented at TU Wien with an

The solvent, which was used for measuring these tar compounds, was toluene, because of the higher solubility of tar in toluene compared to isopropanol. The heavy (high molecular weight) tar compounds represented the mass of tar which was left after vacuum evaporation of the solvent. These compounds were quantified as gravimetric tar. Tar compounds with a medium molecular weight, like

ACCEPTED MANUSCRIPT naphthalene, were analyzed by a gas chromatograph coupled with mass spectrometry (GC/MS). The tar sampling and analysis procedure as well as the classification of tar is explained in detail in literature [22], [23], [24], [25]. In addition to the tar content, it was also possible to measure the water content at the same time by using toluene as solvent. However, by using toluene as solvent, its detection was precluded and the recording of benzene and xylene was difficult with this setup as well. Due to that, all tar contents are presented without benzene, toluene and xylene (BTX). Solid particles like char and dust were determined using a small cyclone and a quartz wool stuffed filter cartridge. Char is defined as carbonic fines, whereas dust represents mineral fines without carbon contents.

2.4 Classification of tar compounds and definition of the tar dew point Generally, tar compounds can be divided into different classes according to different parameters. One possible and quite common way is to classify tar compounds regarding their temperature of formation into primary, secondary and tertiary tar compounds,

which was proposed by Milne et al. [26]. During their investigations, they found out that primary tar products occur in a temperature range of 400 – 700°C and are characterized by oxygenated compounds. The secondary tar products include phenols and other monoaromatic hydrocarbons and are formed between 700 and 850°C. Regarding biomass gasification, phenols belong to the problematic compounds in the producer gas. The tertiary class, appears at a temperature range of 850 – 1000°C and includes aromatics as main products. The aromatics contain polyaromatic hydrocarbons (PAH). However, with an increasing content of PAH the toxicity of tar increases substantially [27], [28]. Another method presents the classification of tar by physical properties, which was proposed by Rabou et al. [29]. Table 1 presents the two classifications of detected GC/MS tar compounds. Regarding the classification of tar according to Milne et al., an assumption has to be mentioned about the occurrence of some tar compounds in two classes. If that was the case, it was decided to divide the total amount of this tar compound and count its halves to the referring classes.

Table 1: Detected GC/MS tar compounds according to Rabou and Milne et al. classification GC/MS compounds Rabou class Milne GC/MS compounds Rabou class phenol 2-methylphenol 4-methylphenol 2,6-dimethylphenol 2,5&2,4-dimethylphenol 3,5-dimethylphenol 2,3-dimethylphenol 3,4-dimethylphenol 2-methoxy-4-methylphenol benzofuran 2-methylbenzofuran dibenzofuran chinoline isochinoline 2-methylpyiridin 3,4-methylpyiridin eugenol isoeugenol phenylacetylene styrene mesitylene 1H-indene 1-indanone naphthalene 1-methylnaphthalene

Class II (heterocyclic aromatic)

Class III (1-ring aromatic)

p/s p/s p/s p p p p p p s s s s s s s p p t s/t t s/t s s/t s

2-methylnaphthalene 1-vinylnaphthalene 2-vinylnaphthalene biphenyl acenaphtylene acenaphthene fluorene anthracene 9-methylanthracene phenanthrene 4,5-methylphenanthrene pyrene benzo[a]anthracene chrysene benzo[b]fluoranthene benzo[k]fluoranthene benzo[a]pyrene benzo[e]pyrene benzo[g,h,i]perylene dibenz[a,h]anthracen indeno[1,2,3]pyrene perylen coronen fluoranthene

Class IV (light PAH; 2,3-ring)

Class V (heavy PAH; 2,3-ring)

Milne s s s s t t t t t t s t t t t t t t t t t t t t

p…primary; s…secondary; t…tertiary;

The tar dew point is defined as an important value for fouling as well as for long-term operation of biomass gasification systems regarding the impact on downstream equipment. The tar dew points of the detected GC/MS tar compounds were calculated with the calculation tool from the Energy Research Centre of the Netherlands (ECN) [30].

2.5 Investigated bed materials

As mentioned before, this paper focuses on the use of different bed materials as well as on mixtures of bed materials. For the experiments, which are presented in the following, limestone, quartz, feldspar and olivine were used as bed materials. Olivine was used as a reference material as it is used in most industrial-operated plants. Due to the fact, that limestone, quartz and feldspar are cheap and easy available world-wide they were selected as alternative materials to conventionally used olivine. Limestone mainly consists of calcite and aragonite,

ACCEPTED MANUSCRIPT which are two crystallization forms of calcium carbonate (CaCO3) and to a very low extent of other minerals. Before the test runs, limestone was filled into the pilot plant, which was transformed to the catalytic active product calcium oxide (CaO) during the test runs due to the high temperatures in the reactor system [19]. Calcium oxide (CaO) provided the catalytic activity for the ongoing gasification

parameter Al2O3 CaCO3 Fe2O3 K2O MgCO3 Na2O SiO2 trace elements (< 0.4 per element) hardness particle density

reaction, but exhibited a low abrasion resistance and had low heat transfer properties. Quartz conversely showed a good abrasion resistance and had much higher heat transfer properties. Olivine also showed a high catalytic activity and a good abrasion resistance as well as good heat transfer properties. In the presented test runs, feldspar with a high share of potassium was used.

Table 2: Investigated bed materials bed material type unit limestone quartz wt.-% wt.-% 95 - 97 wt.-% wt.-% wt.-% 1.5 - 4.0 wt.-% wt.-% 0.4 - 0.6 99 - 100 wt.-% ≤ 3.1 ≤1 Mohs 3 7 kg/m³ 2650, 1500* 2650

feldspar 17.5 – 18.5 14.0 – 15.0 0.5 – 1.0 65 - 66 ≤3 6 2600

olivine 8.0 - 10.5 48 - 50 39 - 42 ≤5 6-7 2850

*particle density after full calcination

In general, by using mixtures of bed materials, the supplementing properties of two different bed materials fulfilled the desired requirements for the gasification process. Table 2 shows the used bed materials. As mentioned before, CaCO3 of limestone was transformed to CaO and carbon dioxide (CO2) in the gasification system at temperatures higher than 750°C. Thus, a change of the bed material density occured.

2.6 Feedstock for the test runs All the experiments were carried out with softwood (SW) in the advanced 100 kWth pilot plant. The proximate and ultimate analysis of the used fuel type is given in Table 3. Wood pellets with a diameter of 6 mm according to the Austrian standard ÖNORM M 7135 were used for the gasification test runs. Table 4 presents the ratios of investigated bed material mixtures for the experiments presented within this paper. The results were generated in 10 experiments. Three different main bed materials

bed material limestone quartz olivine feldspar

with acceptable attrition resistance were selected and mixed with limestone in different ratios, starting from 100 wt.-% of the main bed material, going further with a mixture of 90/10 wt.-%, 50/50 wt.-% and finally using 100 wt.-% limestone in the gasification test runs. This scheme was carried out for quartz, olivine and feldspar. For all test runs softwood was used as fuel to ensure a comparability between the different types of bed material. Table 3: Proximate and ultimate analysis of feedstock parameter unit softwood ash content wt.-%db 0.2 carbon (C) wt.-%db 50.7 hydrogen (H) wt.-%db 5.9 nitrogen (N) wt.-%db 0.2 sulphur (S) wt.-%db 0.005 chlorid (Cl) wt.-%db 0.005 oxygen (O) wt.-%db 43.0 volatiles wt.-%db 85.4 fixed C wt.-%db 14.6 water content wt.-% 7.2 LHV (dry) MJ/kgdb 18.9 LHV (moist) MJ/kg 17.4

Table 4: Presented experiments with different bed material compositions experiment 1 2 3 4 5 6 7 unit L Q100 Q90 Q50 O100 O90 O50 wt.-% 100 10 50 10 50 wt.-% 100 90 50 wt.-% 100 90 50 wt.-% -

8 F100 100

9 F90 10 90

10 F50 50 50

L…limestone; Q…quartz; O…olivine; F…feldspar;

2.7 Validation of process data By using the process simulation software IPSEpro, it was possible to calculate mass and energy balances of the data, which were recorded during gasification test runs. IPSEpro is a flowsheet-based process simulation software, which originates from the power plant sector and offers the user stationary process simulation. Furthermore, IPSEpro enables to validate measured data, which represents results in a very high-quality and representative way. For the

calculations with IPSEpro, a detailed model library was developed at TU Wien by Pröll and Hofbauer [31]. Values, which could not be measured directly during the test runs, could be calculated via mass and energy balances with IPSEpro. A detailed process simulation flow sheet fort the evaluation of experimental data of the 100 kWth gasification pilot plant at TU Wien can be found in [19].

ACCEPTED MANUSCRIPT 2.8 Key figures for the gasification experiments Important key figures, which describe the gasification process, were calculated with the process simulation software IPSEpro. The most relevant key figures for the presented test runs were selected and are described in the following. Due to the fact, that steam is used for the gasification process of carbonaceous feedstock and to enable the comparison of gasification with different fuels, the steam to carbon ratio φSC is used (Eq. 1). φSC =

msteam,GR + mH2O,GR,fuel

Eq. 1

mC,GR,fuel

In Eq. 2 the sum of water and steam in relation to the total mass of dry and ash-free fuel introduced into the GR is expressed via the steam-to-fuel ratio φSF. φSF =

msteam,GR + mH2O,GR,fuel

Eq. 2

mGR,fuel,daf

The relation between water which is consumed for gasification and hydrogen production and the sum of water which is introduced into the gasification reactor is described via the steam-related water conversion XH2O, which is shown in Eq. 3. XH2O =

msteam,GR + mH2O,GR,fuel ‒ mH2O,PG msteam,GR + mH2O,GR,fuel

Eq. 3

The product gas yield describes the ratio between the dry product gas and the dry and ash-free fuel introduced into the GR. The calculation of the PGY is presented in Eq. 4.

PGY =

VPG

Eq. 4

mGR,fuel,daf

A further indicating key parameter of the gasification process is the cold gas efficiency ηCG, which is defined as the chemical energy content of gas components in the tar- and char-free product gas in relation to the chemical energy in the fuel, which is introduced into the GR. All values are based on the lower heating value (LHV), see Eq. 5.

ηCG =

VPG ∙ LHVPG mGR,fuel ∙ LHVGR,fuel

Eq. 5

∙ 100

The overall cold gas efficiency ηCG,o describes the quantity of chemical energy in the PG referred to the chemical energy in the fuel which is introduced into the gasification reactor and the combustion reactor based on the LHV minus heat losses (Eq. 6). ηCG,o =

VPG ∙ LHVPG mGR,fuel ∙ LHVGR,fuel + mCR,fuel ∙ LHVCR,fuel ‒ Qloss

∙ 100

Eq. 6

To describe the ongoing chemical gasification reaction, the water gas shift (WGS) reaction (Eq. 7) is the most important homogeneous gas-gas reaction, which takes place during gasification. CO + H2O ⇌ CO2 + H2

Eq. 7

The chemical reaction of the steam reforming of hydrocarbons is depicted in Eq. 8, which enables discussing tar decomposition. b CaHb + a H2O ⇌ a CO + (a + ) H2 2

Eq. 8

3. Results and discussion In Fig. 6 a typical trend of temperatures inside the gasification and combustion reactor is shown. The locations of the measurement points are visible in Fig. 2. In this experiment, the gasification of softwood in a mixture of 50 wt.-% olivine and 50 wt.-% limestone was investigated. At 2 pm the fluidization was switched from air to steam, which resulted in a temperature loss of about 30°C. From 2 to 11 pm the reactor system could be operated in a stationary phase. Afterwards the system was cooled down. The constant temperatures indicated the steady-state operation in the 100 kWth pilot plant at TU Wien, which was kept for several h during the experiments that are presented within this work. Fig. 7 displays a typical temperature trend over the height of the gasification and combustion reactors from the same experiment (gasification of softwood in a mixture of 50 wt.-% olivine and 50 wt.-% limestone).

ACCEPTED MANUSCRIPT

Fig. 6: Temperature trend in the gasification and combustion reactors during gasification of softwood with a mixture of 50 wt.-% olivine and 50 wt.-% limestone

The coloured areas around the results of the gasification of softwood with different bed materials/bed material mixtures illustrate the area of measured values. With these segments, the relatively narrow range of measured values with different bed materials/ bed material mixtures could be displayed.

50 50

H2

40 40 30 30

CO

20 20

CO2 10 10

CH4 00

mass fraction of total GC/MS tar [g/Nm³]

The measured temperatures at the top of the combustion reactor, which were approximately the temperatures of the hot bed material coming from the combustion reactor and entering the upper loop seal were about 950°C, which were in a typical range. The temperatures in the upper and lower gasification reactor were in the range of 800°C and 950°C and dropped down below the bubbling bed to around 750 - 800°C. The graphs of Fig. 6 and 7 were created with the software Matlab.

main product productgas gascomposition composition[vol.-% [vol.-%db ] main db]

Fig. 7: Temperature profile over the height of the DFB reactor system of the gasification of softwood with a mixture of 50 wt.-% olivine and 50 wt.-% limestone

7

Naphthalene

quartz olivine feldspar

5 4

Acenaphthylene Phenol Styrene Anthracene

3 2 1 0 0 100

00

H2_ H2_S CO2 CO2 H2_ H2_O CO_ CO_ CO2 CO2 CH4 CH4

1H-Indene

6

20 40 60 80 limestone/CaO [wt.-%] 80 60 40 20 quartz [wt.-%]

20 40 60 100 20 40 80 100 limestone [wt.-%] limestone/CaO [wt.-%] 100 80 60 40 20 100 80 60 00 quartz, quartz, olivine, olivine, feldspar [wt.-%] Fig. 8: Main product gas compositions

100 0

6 18 50 5 16

C2 H4

1.5 20 1.0

0 0.0 0.0

C2H4_S C2H4_S quartz C2H4_O C2H4_O olivine C2H4_F C2H4_F feldspar C2H6_S

5 4

C2H6_S C2H6_O C2H6_O C2H6_F C2H6_F

3 2

Phenol Styrene Anthracene

20 40 60 80 limestone/CaO [wt.-%] 80 60 40 20 quartz [wt.-%]

C2 H6

20 40 60 80 100 20 40 60 80 100 20 40 60 80 100 limestone [wt.-%] limestone/CaO [wt.-%] limestone/CaO [wt.-%] 100 60 40 20 000 100 80 80 60 40 20 100 80 60 40 20 quartz, olivine, feldspar [wt.-%] quartz, olivine, feldspar [wt.-%] quartz, olivine, feldspar [wt.-%]

Fig. 9: C2H4 and C2H6 contents in the product gases

In Fig. 9 the contents of C2H4 and C2H6 are depicted. Due to the fact, that C2H4 showed a similar trend like the tar content measured with the GC/MS measurement method (see Fig. 10) it could be regarded as an indicator for the formation of GC/MS tar. C2H6 was formed in very low amounts with decreasing values when limestone was added.

1H-Indene Acenaphthylene

quartz olivine feldspar

5 4

Phenol Styrene Anthracene

3

1 0 0

20

100

80

40

60

80

grav. tar limestone/CaO [wt.-%]

4 10 0 2 0

60 40 quartz [wt.-%]

H2_S gr CO2 G H2_O gr CO_ G CO2 gr CH4 100 G

20

0

20 40 60 80 100 limestone/CaO [wt.-%] 0 100 80 60 40 20 0 20 40 60 80 100 0 quartz, olivine, feldspar [wt.-%] limestone [wt.-%] limestone/CaO [wt.-%] 60 40 20 00 80 20 quartz, olivine, feldspar [wt.-%] [wt.-%]

Fig. 10 presents the tar contents from the experiments mentioned above. The high tar values of the experiments in which no limestone was added to H2_S the bed materialCO_S mixture could be traced back to the CO2_S activity CH4_S low catalytic in terms of steam reforming of hydrocarbons (see Eq. 8). Due to the formation of H2_O CO_O CO via reforming reaction, the WGS CO_Fthe steam CO2_O reaction took place more intensively, which resulted CO2_F CH4_O in higher CH4_FH2 contents H2_F (see Eq. 7). In general, the course of tar formation showed a decreasing trend in 100 line with an increasing content of limestone in the 0 material. It is remarkable that already small bed amounts of limestone addition led to significant effects on gas composition and tar reduction.

Acenaphthylene

0

000

Naphthalene

6

Fig. 10: Tar contents in the product gases Naphthalene

1 0

7

GC/MS tar 2

100

1H-Indene

6

100

10 0.5

12 3 30 10 2 8 20 6 1

7 50 12 12 6 10 10 40 5 88 4 30 6 36 20 4 24

dust

102 12

0000 0 000

20

40

mass fraction of total GC/MS tar [g/Nm³]

2.0 30

40 14 4

maindust product gas composition [vol.-% char content [g/m³ dust content [g/Nm³ ] ]db] db] and char content [g/Nm³ stp db,db

2.5 40

mass fraction of total GC/MS tar [g/Nm³]

main product composition [vol.-% ] CC22H content in H C22H H66gas content in PG [vol.-% [g/Nm³ 44,, C db] db

50 3.0

main composition [vol.-% tarproduct contentgas inchar product gas [g/Nm³ ]] content [g/m³ ,db db stp db]

In Fig. 8 the main product gas composition of the gasification experiments with softwood from 100 wt.-% quartz, 100 wt.-% olivine and 100 wt.-% feldspar to 100 wt.-% limestone used as bed materials is displayed. Regarding the main product gas compositions of the gasification of softwood with different bed materials, a trend could be identified. By increasing the limestone content, the H2 content showed an increasing effect with a maximum at 100 wt.-% limestone. At the same time, the CO content decreased in all experiments. These two phenomena can be explained by an enhanced WGS reaction due to the catalytic activity of calcium oxide (CaO). A comparable phenomenon can be found in literature [32]. CH4 showed a slightly decreasing trend. The gas compound CO2 indicated a lower value when no limestone was added to the bed materials quartz, olivine and feldspar, but when limestone was added, it slightly rose and showed a relative stable course afterwards. 7

mass fraction of total GC/MS tar [g/Nm³]

ACCEPTED MANUSCRIPT 7

7 6 5 4

H2_ CO2 H2_ CO_ CO2 CH4

Naphthalene

dust_S dust_S quartz dust_O dust_O olivine dust_F dust_F feldspar

1H-Indene Acenaphthylene Phenol Styrene Anthracene

3

60

80

100

2 20 40 80 100 20 40 60 100 20 40 60 60 limestone/CaO [wt.-%]80 1[wt.-%] limestone/CaO limestone [wt.-%] 100 80limestone/CaO 60 40 [wt.-%] 20 0 100 80 60 40 20 0 00 100 80 60 40 20 100 quartz, 80 olivine, 60 feldspar 40 20 0 [wt.-%] 0 20 40 60 80 quartz, quartz, olivine, feldspar [wt.-%] quartz,olivine, olivine,feldspar feldspar[wt.-%] [wt.-%] limestone/CaO [wt.-%]

Fig. 11: Dust contents in the product 100 gases 80

60 40 quartz [wt.-%]

20

100 0

ACCEPTED MANUSCRIPT

7 50 7 6 6 40 5 5

Class V tar compounds (heavy PAH) showed a decreasing trend in all experiments when more catalytic active bed material in form of limestone/CaO was added. Similar results were obtained by Devi with sand as bed material [35]. Contrarily, class IV indicated the major share of the classes and showed an increasing amount by increasing the content of limestone. This could be explained by the fact that on the one hand naphthalene is a very stable compound and on the other hand the decomposition of higher tar (e.g. class V) led to the formation of lighter PAH, like naphthalene. The highest proportion of class IV was also reported by van der Meijden et al. [36]. The content of class III remained relatively stable. The same could be seen for Class II (heterocyclic aromatics), which remained stable also at relatively low quantities.

00

7

Naphthalene 1H-Indene

6 5 4

quartz olivine feldspar

Acenaphthylene Phenol Styrene Anthracene

3

20 40 60 80 2 20 40 60 80 100 100 limestone/CaO [wt.-%] 1 [wt.-%] limestone [wt.-%] limestone/CaO 100 80 60 40 20 00 80 60 40 20 100 0 quartz, 0 [wt.-%] 20 40 60 quartz, olivine, olivine,feldspar feldspar [wt.-%]

80 limestone/CaO [wt.-%] Fig. 12: Char contents in the product gases 100 80 60 40 20 quartz [wt.-%]

In Fig. 12 the contents of char in the product gas for the gasification experiments are depicted. Compared to investigations in literature, where around 30 g/m3stp,db char was generated [34], [33], the observed char values in the product gas, which are presented in this paper, were in a range of 1 - 6 g/m³stp,db for all experiments. This could be explained by high char conversions and the improved separation system of the advanced design of the pilot plant. Due to the improvement of the gasification reactor, char particles stayed in the counter-current column longer, which resulted in a longer residence time and thus in a better interaction of gas and particles. Additionally, the product gas cyclone may lead to reduced fly-char contents. Fig. 13 and 14 present the classification of GC/MS tar according to their physical properties, which was proposed by Rabou et al. Fig. 15 displays the

100 0

CO_S CH4_S H2_Ochar_Q CO_O class IV light PAH CO_F char_Q CO2_O char_O 40 CO2_F char_O CH4_O 95 char_F CH4_F char_F H2_F

50 CO2_S 100

H2_ 2_ 4_ CO2 2_ H2_ 4_ 2_ CO_ 4_ CO2 2_ 4_ CH4 2_

30 90 20 85 10 80 0 75

00

mass fraction of total GC/MS tar [g/Nm³]

char

3 3 20 2 2 10 1 1

main product composition share of class gas in total GC/MS tar[vol.-% [wt.-%] db]

4 30 4

00

classification of GC/MS tar according to Milne et al., who categorized GC/MS tar compounds by their temperature of formation. A detailed explanation of both classification systems of GC/MS tar is given in chapter 2.4.

H2_S

mass fraction of total GC/MS tar [g/Nm³]

db db stp

main product composition db] char content [g/Nm³ chargas content [g/m³ [vol.-% ,] ]

Fig. 11 demonstrates the dust contents from 100 wt.% quartz, 100 wt.-% olivine and 100 wt.-% feldspar to 100 wt.-% limestone. The dust contents were measured for all experiments except for the experiment with 100 wt.-% quartz, because of a blocked cyclone down comer. In general, the dust contents increased with an increasing limestone content, which could be explained by the low abrasion resistance of limestone. However, for these experiments, in general quite low values of dust were observed, which was possible due to the advantageous separation system of the advanced design of the pilot plant compared to the conventional design [18]. The results of dust of the presented test runs can be compared with investigations carried out in the conventional 100 kWth dual fluidized bed steam gasification pilot plant at TU Wien. By converting softwood as fuel with pure olivine as bed material, dust contents of around 10 g/m³stp,db were generated [33], [34].

7

Naphthalene 1H-Indene

6 5 4

quartz olivine feldspar

Acenaphthylene Phenol Styrene Anthracene

3

2 20 40 60 80 100 20 40 60 80 100 limestone/CaO limestone [wt.-%] 1 [wt.-%] 100 60 40 20 0 100 80 80 60 40 20 0 quartz, olivine, feldspar [wt.-%] quartz, olivine, feldspar [wt.-%] 0 20 40 60

80

limestone/CaO [wt.-%] Fig. 13: GC/MS tar classified according to physical 80 60 40 20 properties (1) 100 quartz [wt.-%]

100 0

50 20 18 40 16

class V heavy PAH

14 30 12

7

Naphthalene 1H-Indene

6 5 4

quartz olivine feldspar

Phenol Styrene Anthracene

3

class II light aromatics2

10 208 6

1

Class III heterocyclic 0 20 40 60 80 aromatics limestone/CaO [wt.-%]

0

100

80

104

60 40 quartz [wt.-%]

20

2

00

00

20 40 60 80 100 20 40 60 80 limestone/CaO [wt.-%] limestone [wt.-%] 100 80 60 40 20 0 100 80 60 40 20 quartz, quartz, olivine, olivine, feldspar feldspar [wt.-%] [wt.-%]

main product composition [vol.-%db] Milnegas group [%] tar share in total GC/MS shareof ofclass class intar total GC/MS tar [wt.-%]

50 80 80 80

tertiary

40 60 60 60 30

77

Naphthalene of total GC/MS tar [g/m³stp,db] Naphthalene of total GC/MS tar [g/m³stp,db]

Fig. 14: GC/MS tar classified according to physical properties (2)

The tar classification according to Milne et al. showed a similar trend for both the gasification with quartz, olivine and feldspar in pure form and mixed with limestone. Primary tar compounds were merely detectable, which resulted from their low temperature of formation. Secondary tar compounds increased and tertiary tar compounds decreased when the limestone content was raised. The decreasing of tertiary tar can be compared to the declining trend of class V. Tertiary tar as well as class V tar included heavy and high molecular hydrocarbons and the formation of them could be reduced by increasing the favorable catalytic effect of limestone.

In Fig. 16 – 20 the behavior of the main observed tar compounds of the experiments versus the added limestone content to quartz, olivine and feldspar are 2_S 3_S H2_S CO_S shown. Naphthalene presented the main tar CO2_S 4_s with 5_sCH4_S compound a relatively high share according to H2_O CO_O the total GC/MS tar and showed a decrease (total 2_o 3_O amount) when5_o more limestone was added. However, CO_F CO2_O 4_o the relative of Naphthalene in class IV tar CO2_F CH4_O 2_kf share 3_kf compounds (see Fig. 13). Tertiary or CH4_F H2_F 4_kf increased 5_kf 100 partial tertiary tar compounds like 1H-Indene (s/t), acenaphthylene (t), styrene (s/t) or anthracene (t) 0 decreased when more limestone was added to the main bed material. Phenols (p/s) were extremely low and remained over all experiments near a value of 0 g/m³stp,db. This could be explained by the positive effect of the counter-current column of the advanced gasifier design compared to the conventional design, where the upper part of the gasification reactor was built as bubbling bed with a freeboard region and thus the conversion of tar compounds was not as sufficient as in the advanced design [37]. To give an example, investigations of the gasification of wood in the conventional design showed the formation of phenol indeed [32]. 7

Acenaphthylene

66 55

33 22

mass fraction of total GC/MS tar [g/Nm³]

20

7

5

p_S p_O 4 p_F

20 20 10

primary 3 000

0000

1H-Indene

quartz s_S t_S olivine s_O t_O feldspar s_F

t_F

Acenaphthylene Styrene Anthracene

2

Fig. 15: GC/MS tar classified according to temperature of quartz [wt.-%] formation

100 0

Acenaphthylene Phenol Styrene Anthracene

3

Na Na

2

Na Na

1

Na Na

0 0 100

0

20

Datenreihen1 Datenreihen1 Datenreihen2 Datenreihen2 Datenreihen3 Datenreihen3

Phenol

20 40 60 20 40 60 80 100 20 40 60 80 20 40 60 80 100 100 1 limestone/CaO [wt.-%] limestone [wt.-%] limestone/CaO [wt.-%] limestone/CaO [wt.-%] 0 100 80 60 40 20 100 80 60 40 20 100 80 60 400 20 20 40 00060 80 20 quartz, olivine, feldspar [wt.-%] limestone/CaO [wt.-%] quartz,olivine, [wt.-%] quartz, [wt.-%] quartz, olivine, feldspar feldspar 100 80 60 40 20

4

20 40 60 80 limestone/CaO [wt.-%] 80 60 40 20 quartz [wt.-%]

40

60

80

100

0 20 CO_S 40 60 80 100 H2_S limestone [wt.-%] CO2_S CH4_S limestone [wt.-%] 100 80 60 40 20 00 H2_O 80 CO_O 100 60 40 20 quartz, olivine, feldspar [wt.-%] CO_F CO2_O feldspar [wt.-%] quartz, olivine, CO2_F CH4_O Fig. 16: Naphthalene of total GC/MS tar CH4_F H2_F

Naphthalene

6

quartz olivine feldspar

5

11 00

Naphthalene 1H-Indene

6

Naphthalene

44

secondary

40 40

mass fraction of total GC/MS tar [g/Nm³]

mass fraction of total GC/MS tar [g/Nm³]

main product composition [vol.-% db] share of class gas in total GC/MS tar [wt.-%]

ACCEPTED MANUSCRIPT

100 0

1.5

1H-Indene

1.0

6

quartz olivine feldspar

5 4

Acenaphthylene Phenol Styrene

2.0 1H-Indene_O 1H-Indene_Q 1H-Indene_F 1.5

Anthracene

3 2 1 0 0 100

0.5

20 40 60 80 limestone/CaO [wt.-%] 80 60 40 20 quartz [wt.-%]

100

20

40 60 80 100 limestone [wt.-%] 100 80 60 40 20 0 quartz, olivine, feldspar [wt.-%]

1.0

0.5 2.0 2.5 0.4 1.5 2.0 0.3

quartz olivine feldspar

5 4

Acenaphthylene Phenol Styrene Anthracene

3 2

Acenaphthylene

1.0 0.21.5

1 0

0 100

0.5 0.11.0 0.0 0.0 0.5 0 0

1H-Indene

Styrene of total GC/MS tar [g/m³stp,db]

mass fraction of total GC/MS tar [g/Nm³]

Acenaphthylene of total GC/MS tar [g/m³ Styrene of total GC/MS Acenaphthylene stp,db] tar [g/m³stp,db] [g/m³stp,db]

0.6 2.5

1H-Indene Acenaphthylene

quartz olivine feldspar

5 4

Phenol Styrene Anthracene

An

3 2

An

1 0 0 100

20 40 60 80 limestone/CaO [wt.-%] 80 60 40 20 quartz [wt.-%]

100

An

0

0.0 0

40 60 80 100 limestone [wt.-%] 100 80 60 40 20 0 quartz, olivine, feldspar [wt.-%]

20

20 40 60 80 limestone/CaO [wt.-%] 80 60 40 20 quartz [wt.-%]

0.6 0.5

0.4 Acenaphthylene_O Styrene_O

40 60 80 100 40 60 80 100 limestone [wt.-%] limestone [wt.-%] 100 80 60 40 20 0 0.0 100 80 60 40 20 0 quartz, olivine,40 feldspar 0 20 60 [wt.-%] 80 100 quartz, olivine, feldspar [wt.-%] limestone Fig. 18: Acenaphthylene of total [wt.-%] GC/MS tar 100 80 60 40 20 0 quartz, olivine, feldspar [wt.-%] In Table 5, a selection of the most important indicating key parameters as well as the main operation parameters for the gasification of softwood with different types of bed materials are shown. The temperature “Tmean in GRlower” corresponds to the temperature in the bubbling bed and “T in GRupper” to the temperature of hot bed material re-entering the gasification reactor from the upper loop seal (see Fig. 2). As can be seen from Table 5, the product gas yields of the experiments were increased by adding limestone to quartz, olivine and feldspar. This fact can be proved by longer residence times of char particles in the gasifier promoted by the advanced design of the pilot plant as well as by the increasing share of limestone, which supported the WGS

7

0

0.1 Acenaphthylene_F

Naphthalene 1H-Indene

6 5 4

quartz olivine feldspar

Acenaphthylene Phenol Styrene Anthracene

S

3 2

Styrene 1 Acenaphthylene_Q 0.3Styrene_Q 0 Acenaphthylene_O Acenaphthylene_F 0 0.2Styrene_F 100 100 Acenaphthylene_Q

S

S 20 40 60 80 limestone/CaO [wt.-%] 80 60 40 20 quartz [wt.-%]

0.0 0

20 20

Naphthalene

6

Fig. 19: Anthracene of total GC/MS tar Naphthalene

6

7

0.5

Fig. 17: 1H-Indene of total GC/MS tar 7

Anthracene

0

0.0 0

2.5

mass fraction of total GC/MS tar [g/Nm³]

2.0

1H-Indene

mass fraction of total GC/MS tar [g/Nm³]

2.5

Naphthalene

Anthracene of total GC/MS tar [g/m³stp,db]

mass fraction of total GC/MS tar [g/Nm³]

1H-Indene of total GC/MS tar [g/m³stp,db]

ACCEPTED MANUSCRIPT 7

20

40 60 80 100 limestone [wt.-%] 100 80 60 40 20 0 quartz, olivine, feldspar [wt.-%] Fig. 20: Styrene of total GC/MS tar

reaction. The rising steam-related water conversion might have been favorably affected by more intense and longer contact times between gas and particles as well as by the increase of catalytic activity (see Fig. 21). This resulted in a rising water consumption via the steam reforming reaction as well as the water gas shift reaction. The cold gas efficiencies are approximately in the same range for all experiments. The tar dew point (TDP), which is an important indicator for formation of deposits in the product gas line, decreased by adding limestone to the bed material mixtures. A minimum value of 87°C was found for 100 wt.-% limestone (see Fig. 22).

100 0

ACCEPTED MANUSCRIPT

0.35

SRWconv_Q SRWconv_O SRWconv_F

0.30

0.40 0.25

0.40 0.15 0.30 0.35 0.10 0.25

Steam-related water conversion 7 XH2O

0.25 0.00 0.15 0 0.20 0.10 100 0.15 0.05 0.10 0.00 0 0.05

mass fraction of total GC/MS tar [g/Nm³]

0.30 0.05 0.20 20

Naphthalene 1H-Indene

6

40 60 80 quartz100 5 limestone [wt.-%] olivine 4 80 60 40 20 feldspar0 quartz, olivine, feldspar [wt.-%] 3

20

Acenaphthylene

tar dew point [°C]

0.35 0.20

Phenol Styrene Anthracene

2

220 200 SRWconv_Q 180 SRWconv_O 160 SRWconv_F 140 SRWconv_Q 120SRWconv_O 100SRWconv_F 80 60 40 20 0 0 20

Tar dew point

mass fraction of total GC/MS tar [g/Nm³]

steam-realted steam-realted water conversion water conversion [kgconversion steam-realted water H2O/kgH2O] [kgH2O/kgH2O] [kgH2O/kgH2O]

0.40

T T T

7

Naphthalene 1H-Indene

6

quartz olivine feldspar

5 4

Styrene Anthracene

3

unit kW m³stp,db/kgfuel,daf kgH2O/kgC kgH2O/kgfuel kgH2O/kgH2O

1 L 89 1.43 1.42 0.72 0.36

2 Q100 87 1.34 1.72 0.87 0.25

3 Q90 95 1.41 1.69 0.75 0.31

4 Q50 93 1.49 1.66 0.84 0.37

experiment 5 6 O100 O90 85 87 1.35 1.46 1.97 1.91 1.00 0.97 0.22 0.29

7 O50 87 1.42 1.49 0.76 0.32

8 F100 87 1.17 1.80 0.89 0.14

9 F90 92 1.36 1.65 0.83 0.29

10 F50 90 1.44 1.53 0.78 0.33

% %

88 73

88 70

94 70

93 71

93 71

92 72

87 74

87 71

97 72

89 71

kW kW kW °C °C °C °C

101 52 32 769 991 1010 87

101 46 20 805 944 947 211

101 65 31 790 947 948 181

101 56 26 813 967 974 104

91 46 19 836 951 954 213

95 51 25 848 965 954 154

102 31 13 815 941 950 123

98 53 29 772 934 923 205

95 61 29 771 933 942 201

101 44 20 797 967 977 115

L…limestone; Q…quartz; O…olivine; F…feldspar; b temperature in the GR at fuel feeding position

4. Conclusions Investigations with the dual fluidized bed steam gasification of softwood with different types of bed materials in the advanced 100 kWth pilot plant at TU Wien were carried out and compared. Experimental campaigns were performed, varying the mass ratios of quartz, olivine and feldspar with limestone from 100/0, 90/10, 50/50 and 0/100. To give an overview of the performed experiments, the obtained results can be summarized as follows.



Phenol

40 60 80 100 40 60 80 100 2 1 limestone [wt.-%] limestone [wt.-%] 1 100 80 60 20 0 0.00 040 100 80 60 40 20 0 0 20 40 60 80 100 100 80 60 [wt.-%] 0quartz,20 0 olivine,40feldspar quartz, olivine, feldspar [wt.-%] limestone/CaO [wt.-%] 0 20 40 60 80 Fig. 21: Steam-related water conversions [wt.-%] limestone 100 80 60 40 20 0 Fig. 22: Tar dew points limestone/CaO [wt.-%] quartz [wt.-%] 0 20 40 60 80 100 100 80 60 40 20 [wt.-%] for gasification of softwood with different types of bed materials quartz [wt.-%] feldspar quartz, olivine, Table 5: Performance indicating key parameters

product gas power product gas yield steam to carbon ratio steam to fuel ratio steam-related water conversion cold gas efficiency overall cold gas efficiency fuel input GR fuel input CR overall heat loss Tmeanb in GRlower T in GRupper T at CRoutlet TDP



Acenaphthylene

It was possible to shift the product gas compositions to higher H2 contents by increasing the limestone content. In parallel, the CO contents decreased and the CO2 contents increased. A relative constant trend was seen for CH4. The tar contents in the product gas decreased by increasing the limestone







content in the bed material. It was observed that C2H4 could be an indicator for the GC/MS tar content in the product gas, which decreased in the same trend as the GC/MS tar content. The tar dew points through all experimental campaigns could be reduced when more limestone was added to the bed material mixtures. The catalytic activity of limestone influenced the steam reforming reactions and thus reduced the tar dew points. Tar dew points lower than 180°C are very favorable for “dry” cleaning of the product gas e.g. via fabric filters. The dust contents increased with higher limestone addition, which could be traced back to the lower abrasion resistance of limestone. Olivine and feldspar showed good abrasion resistance resulting in a very low value for the dust content. The contents of char could be kept low, which could be explained by the improved

100 0

ACCEPTED MANUSCRIPT separation system and a high conversion of fine carbons in the upper gasification reactor of the advanced design of the pilot plant. A correlation between tar formation of different classes and the admixture of catalytic active bed material could be observed as well. The more catalytic active bed material was added, the more heavy tar compounds could be reduced. When investigating 100 wt.-% quartz, olivine or feldspar as bed material, thermal effects played a major role regarding tar cracking in the upper part of the gasification reactor. However, when limestone was added, the tar reduction was not only promoted by thermal destruction, but also by the catalytic activity of CaO.

[5]

Due to the cheap price of quartz, feldspar and limestone compared to olivine as well as because of the good heat transfer characteristics and the broad forthcoming all over the world, they present alternative bed materials, which can be used in gasification processes. In combination with limestone, the missing catalytic properties of the alternative bed materials can be compensated and thus a suitable bed material for the gasification process can be obtained. Another positive fact is presented by the advanced design of the dual fluidized bed steam gasification plant, which enables the use of soft materials like limestone. Additionally, the advanced design enhances thermal and chemical catalytic driving forces in the upper counter-current gasification reactor [38]. Concluding from these investigations, the use of alternative bed materials as well as bed material mixtures offer a promising alternative to commercially used olivine to produce a highly valuable product gas.

[9]





References [1] Müller S, Groß P, Rauch R, Zweiler R, Aichernig C, Fuchs M. Production of diesel from biomass and wind power – Energy storage by the use of the Fischer-Tropsch process 2017. doi:10.1007/s13399-0170287-1. [2] Weber G, Rauch R, Hofbauer H. Production of mixed alcohols from biomass-derived synthesis gas using a sulfidized molybdenum catalyst. Proc Int Conf Polygeneration Strateg (ICPS13), Vienna 2013. [3] Hofbauer H. Gasification of organic material in a novel fluidization bed system. Proc. of the 1st SCEJ Symposium on Fluidization. Tokyo, 1995. [4] Hofbauer H, Rauch R, Bosch K, Koch R, Aichernig C. Biomass CHP Plant Güssing A Success Story. Expert Meet Pyrolysis Gasif Biomass Waste 2002.

[6]

[7]

[8]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

Wilk V, Hofbauer H. Analysis of optimization potential in commercial biomass gasification plants using process simulation. Fuel Process Technol 2016;141:138–47. doi:10.1016/j.fuproc.2015.07.035. Kirnbauer F, Maierhans F, Kuba M, Hofbauer H. State of the art biomass gasification for CHP production - the Ulm plant. Proc 2nd Int Conf Renew Energy Gas Technol 2015;1:2–5. Gunnarsson I. The GoBiGas project efficient transfer of biomass to biofuels, in Proceedings of the International Seminar on Gasification, Gothenburg, Sweden 2010. Rauch R., Pfeifer C., Bosch K., Hofbauer H., Swierczynski D., Courson C. KA. Comparison of different olivines for biomass gasification n.d. Kirnbauer F, Hofbauer H. Investigations on bed material changes in a dual fluidized bed steam gasification plant in Güssing, Austria. Energy and Fuels 2011;25:3793–8. doi:10.1021/ef200746c. Kuba M, Havlik F, Kirnbauer F, Hofbauer H. Influence of bed material coatings on the water-gas-shift reaction and steam reforming of toluene as tar model compound of biomass gasification. Biomass and Bioenergy 2015;89:40–9. doi:10.1016/j.biombioe.2015.11.029. Kirnbauer F, Wilk V, Kitzler H, Kern S, Hofbauer H. The positive effects of bed material coating on tar reduction in a dual fluidized bed gasifier. Fuel 2012;95:553–62. doi:10.1016/j.fuel.2011.10.066. Pfeifer C, Koppatz S, Hofbauer H. Catalysts for dual fluidised bed biomass gasificationan experimental study at the pilot plant scale. Biomass Convers Biorefinery 2011;1:63– 74. doi:10.1007/s13399-011-0005-3. Pfeifer C, Koppatz S, Hofbauer H. Steam gasification of various feedstocks at a dual fluidised bed gasifier: Impacts of operation conditions and bed materials. Biomass Convers Biorefinery 2011;1:39–53. doi:10.1007/s13399-011-0007-1. Kern S, Pfeifer C, Hofbauer H. Gasification of wood in a dual fluidized bed gasifier: Influence of fuel feeding on process performance. Chem Eng Sci 2013;90:284– 98. doi:10.1016/j.ces.2012.12.044. Schmid JC. Development of a novel dual fluidized bed gasification system for increased fuel flexibility. TU Wien, doctoral thesis, 2014. Benedikt F, Fuchs J, Schmid JC, Hofbauer H. The Novel Dual Fluidized Bed Gasifier at TU Wien Part I : Parameter Variation with Wood Pellets. 12 Minisymp Für Verfahrenstechnik 2016:1–5. Benedikt F, Fuchs J, Schmid JC, Müller S,

ACCEPTED MANUSCRIPT

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25] [26] [27]

[28]

[29]

[30]

Hofbauer H. Advanced dual fluidized bed steam gasification of wood and lignite with calcite as bed material. Korean J Chem Eng 2017;34:2548–58. doi:10.1007/s11814-0170141-y. Schmid JC, Müller S, Hofbauer H. First Scientific Results with the Novel Dual Fluidized Bed Gasification Test Facility at TU Wien. 24th Eur Biomass Conf Exhib 2016:2–6. Müller S, Fuchs J, Schmid JC, Benedikt F, Hofbauer H. Experimental Development of Sorption Enhanced Reforming by the Use of an Advanced Gasification Test Plant. Int J Hydrogen Energy 2017;42:29697–707. doi:10.1016/j.ijhydene.2017.10.119. Kolbitsch M, Schmid JC, Diem R, Müller S, Hofbauer H. Influence of fuel feeding position on sorption enhanced reforming in a dual fluid gasifier 2014:1–6. Kolbitsch M. First Fuel Tests at a Novel 100 kWth Dual Fliuidized Bed Steam Gasification Pilot Plant. TU Wien, doctoral thesis, 2016. Schmid JC, Wolfesberger U, Koppatz S, Pfeifer C, Hofbauer H. Variation of feedstock in a dual fluidized bed steam gasifier - Influence on product gas, tar content, and composition. Environ Prog Sustain Energy 2012;31:205–15. doi:10.1002/ep.11607. Good J, Ventress L, Knoef H, Group BT, Zielke U, Hansen PL, et al. Sampling and analysis of tar and particles in biomass producer gases 2005;20002:1–44. Wolfesberger U, Aigner I, Hofbauer H. Tar Content and Composition in Producer Gas of Fluidized Bed Gasification of Wood — Influence of Temperature and Pressure 2009;28. doi:10.1002/ep. CEN/TS. 15439:2006 2006. Milne TA, Evans RJ. Biomass Gasifier “Tars”: Their Nature , Formation , and Conversion 1998. Morf P, Hasler P, Nussbaumer T. Mechanisms and kinetics of homogeneous secondary reactions of tar from continuous pyrolysis of wood chips. Fuel 2002;81. Ahrenfeldt J, Egsgaard H, Stelte W, Thomsen T, Henriksen UB. The influence of partial oxidation mechanisms on tar destruction in TwoStage biomass gasification. Fuel 2013;112:662–80. doi:10.1016/j.fuel.2012.09.048. Rabou LPLM, Zwart RWR, Vreugdenhil BJ, Bos L. Tar in Biomass Producer Gas, the Energy research Centre of The Netherlands (ECN) Experience: An Enduring Challenge. Energy & Fuels 2009;23:6189–98. doi:10.1021/ef9007032. Energy Research Center of the Netherlands (ECN), http://www.

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

thersites.nl/completemodel.aspx; [accessed 12 February 2017] n.d. Pröll T, Hofbauer H. Development and Application of a Simulation Tool for Biomass Gasification Based Processe Development and Application of a Simulation Tool for Biomass Gasification Based Processes *. Int J Chem React Eng 2008;6:Article A89. doi:10.2202/15426580.1769. Koppatz S, Pfeifer C, Hofbauer H. Comparison of the performance behaviour of silica sand and olivine in a dual fluidised bed reactor system for steam gasification of biomass at pilot plant scale. Chem Eng J 2011;175:468–83. doi:10.1016/j.cej.2011.09.071. Kern S, Pfeifer C, Hofbauer H. Synergetic Utilization of Renewable and Fossil Fuels: Dual Fluidized Bed Steam Co-gasification of Coal and Wood. APCBEE Procedia 2012;1:136–40. doi:10.1016/j.apcbee.2012.03.022. Wilk V, Aichernig C, Hofbauer H. Waste Wood Gasification : Distribution of Nitrogen , Sulphur and Chlorine in a Dual Fluidised Bed Steam Gasifier. Int Conf Circ Fluid Bed Technol CFB 10 2011:1–8. Devi L. Catalytic removal of biomass tars : olivine as prospective in-bed catalyst for fluidized-bed biomass gasifiers Catalytic removal of biomass tars ; Olivine as prospective in-bed catalyst for fluidized-bed biomass gasifiers. 2005. doi:10.6100/IR583960. Meijden CM Van Der, Veringa HJ, Drift A Van Der, Vreugdenhil BJ. The 800 kWth allothermal biomass gasifier MILENA. 16th Eur Biomass Conf 2008:2–6. Schmid JC, Pfeifer C, Kitzler H, Pröll T, Hofbauer H. A new dual fluidized bed gasifier design for improved in situ conversion of hydrocarbons. Proc Int Conf Polygeneration Strateg 2011:1–10. Schmid JC, Pröll T, Pfeifer C, Hofbauer H. Improvement of Gas – Solid Interaction in Dual Circulating Fluidized Bed Systems. Proc 9th Eur Conf Ind Furn Boil (INFUB), Estoril, Port 2011 2011:1–13.

ACCEPTED MANUSCRIPT Highlights:     

100 kWth dual fluidized bed steam gasification pilot plant at TU Wien Mixtures of quartz, olivine and feldspar with limestone as bed material Increasing tar destruction and H2 content by increasing limestone content Rising share of limestone in bed material mixture reduced heavy tar compounds Already low amounts of limestone showed significant effects on product gas and tar