Fuel Processing Technology 120 (2014) 54–60

Contents lists available at ScienceDirect

Fuel Processing Technology

j ourna l homepage: www.e lsev ie r .com/ locate / fuproc

Trace metals emission in syngas from biomass gasification

Deepak Pudasainee ⁎, Hanns-Rudolf Paur, Sabine Fleck, Helmut SeifertKarlsruhe Institute of Technology, Institute of Technical Chemistry, Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany

⁎ Corresponding author. Tel.: +49 721 608 23832; fax:E-mail addresses: deepak.pudasainee@itc-tab.fzk.de, d

(D. Pudasainee).

0378-3820/$ – see front matter © 2013 Elsevier B.V. All rihttp://dx.doi.org/10.1016/j.fuproc.2013.12.010

a b s t r a c t

a r t i c l e i n f o

Article history:Received 12 September 2013Received in revised form 9 December 2013Accepted 13 December 2013Available online 29 December 2013

Keywords:Biomass gasificationMercuryMinamata conventionRenewable energySyngasTrace metals

In this paper, trace metals (As, Be, Cd, Cr, Hg Ni, Pb, Se and V) emission from the atmospheric entrained flowgasifier REGA at Karlsruhe Institute of Technology, Germany is presented. Elemental composition and metalsconcentration in glycol (model fuel) and straw char samples were determined. In order to achieve well definedand reproducible operating conditions, gasification experiments were carried out with glycol and slurry (strawchar mixed with glycol) as fuel. Trace metals emission and distribution into particle and gas phase in syngasfrom gasification were measured according to US EPA method 29. During glycol gasification, metals concentra-tion in syngas ranged from0.1 μg/Nm3 (V) to 4.6 μg/Nm3 (Cr). In slurry gasification, metals concentration in syn-gas ranged from 2.4 μg/Nm3 (Hg) to 53.2 μg/Nm3 (Ni). Trace metals tend to volatilize more in reducinggasification environment. Tendency of metals to speciate into particle phase increased in slurry gasificationthan glycol gasification due to increased in particle concentration and unburned carbon content in syngas. Theincrease of particle and carbon content in syngas possessed strong influence in distribution behavior of As, Beand Pb, however, such influence was less for other metals studied. Mercury in syngas was speciated predomi-nantly into elemental form followed by oxidized form whereas share of particle bond mercury was the least.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Biomass is a renewable, low carbon fuel that can be used as a substi-tute of fossil fuels promoting environmental and social benefits.Biomass is the fourth major sources of primary energy uses followingcoal, petroleum and natural gas [1]. The simplest utilization of biomassi.e. burning, largely practiced in the developing countries, is an unsus-tainable combustion practice which releases pollutants of serioushealth and environmental concern. Bio-chemical processes like bio-methanization and thermo-chemical methods such as combustion,pyrolysis and gasification are used in recovering energy from biomass[2–4]. Biomass gasification is one of such promising technologies toproduce syngas in an energy efficient manner. The gasification ofligno-cellulosic biomass has high conversion efficiency among variousthermo-chemical conversion technologies [5]. Due to several advan-tages, biomass gasification technology has drawn an attention aroundthe world [3,5,6]. As for example, over the past decades the NorthDakota Energy and Environmental Research Center has developed andapplied the gasification of solid fuels (coal, biomass, coal–biomassblends), cleaned syngas and converted syngas to useful energy [7].

Biomass based energy has been prioritized in current EuropeanUnion (EU) strategies to mitigate climate change and improve energy

+49 721 608 24303.eepu003@hotmail.com

ghts reserved.

security. The member states of the EU are committed to acquire 20%of their energy requirements from renewable sources, including bio-mass, by the year 2020 [8]. German government's Energy Concept(2010) aims to increase the share of regenerative energies up to 18%of the end-use energy by the year 2020. Influenced by these activities,biomass gasification research and development work has increasedintending on improving efficiency; reduce pollutants release duringthe process utilization. In spite increasing use, one of the environmentalaspects that have not been studied comprehensively is the fate of tracemetals during biomass gasification process.

Biomass contains trace quantity of metals [9]. Biomass burning, bothnatural and anthropogenic means has been reported as a source ofatmospheric metals includingmercury emissions [10–12]. Trace metalsintroduced into a gasification system, mainly from feedstocks, undergotransformation and removal within the system and the left over entersthe chemical conversion steps together with syngas or if syngas is usedas fuel in gas engines leaves the systemwith flue gas. Some tracemetalsin syngas cause fouling and corrosion of the gas turbine blades [13].Trace metals in syngas also affect the catalyst in the system and resultsin low biomass to fuel conversion. Specifically, as for example, alkalimetals (Na, K) andheavymetals V, and Zn species can damagematerialsin heat and power plants, cause abrasion, corrosion and erosion of inter-nal parts, influences energy andmass flow in the system. Further, depo-sition of alkali species such as NaCl, Na2SO4, andNa2CO3, can poison anddeactivate catalysts [14] and fuel cells. To achieve an economicallyacceptable service lifetime of catalyst, Hg and As concentration insyngas at the inlet should be reduced to b 5 ppb [15,16]. The catalystdeactivation by other trace heavy metals such as Pb and Cd in off-gas

55D. Pudasainee et al. / Fuel Processing Technology 120 (2014) 54–60

has been reported, however, the quantitative data is lacking. Never-theless, trace metals in biomass gasification are to be controlled, asminimum as possible, to increase the lifetime of the facility itself,increase biomass to fuel conversion and to create the cleanerenvironment.

To some extent, trace metals emission in combustion flue gas isknown [17–19]. However, their behavior in gasification environmentis unclear. Considering the difference in off gas chemistry, an apparentdifference in trace metal emission behavior between combustion andgasification atmosphere can be rationally anticipated. Behavior oftrace metals during gasification process utilization has been studied bysome researchers [20,21]. However, most of these studies are carriedwith fuels containing relatively higher amount of trace metals, mainlycoal. Very few studies [9] have investigated trace metals emission dur-ing gasification from fuels with relatively lower content of trace metalssuch as biomass and thus their behavior is still scarcely known. Beforebiomass gasification technology is used widely, understanding biomasscharacteristics and metals content in feedstocks and behavior in pro-duction process, input and output streams are vitally important, whichis lacking in the existing literature. In this paper for the first time wehave presented tracemetals emission and distribution from gasificationof straw char at 60 kW atmospheric entrained gasifier.

The Karlsruhe Institute of Technology (KIT), Germany has developeda two-stage process “bioliq®” for the production of synfuel frombiomass [22,23]. The process involves: 1) pyrolysis of straw and otherlignocellulosic agricultural byproducts at a regional plant, 2) bio-syncrude® gasification and syngas conversion in a central plant. Theenergy-rich suspension, obtained from pyrolysis plants is gasified inan entrained-flow gasifier and converted into syngas, a chemically reac-tive mixture of CO and H2. Pyrolysis is the starting point in the bio-mass gasification process, bioliq®. The biomass char, oil obtainedfrom pyrolysis are gasified to produce fuel gas that can be used toproduce heat, electricity, or converted into syngas. In this papercharacteristic of feedstocks namely glycol (model fuel) and strawchar is presented. Further, experiments were carried out to deter-mine emission concentration of trace metals in syngas and their dis-tribution into particle and gas phases at the atmospheric entrainedflow gasifier REGA.

Fig. 1. Process flow diagram of the

2. Experimental

2.1. Pilot-scale gasifier-REGA

Fig. 1 shows the process configuration of a 60 kW pilot-scale atmo-spheric entrained flow gasifier REGA (Research Entrained Flow GAsifi-er) at Karlsruhe Institute of Technology (KIT), Germany. At REGA basicresearch on the influence of atomization and fuel specification on thesyngas quality is carried out. The ceramic tube reactor of the gasifieris 3 m long with inner diameter of 28 cm. The gasifier walls can beheated up to 1200 °C. The adiabatic temperatures were 2000 °C(Test 1–1), 1700 °C (Tests 1–2, 1–3) and 1700 °C (Tests 2–1, 2–2, 2–3). Experiments were carried out with fuels: i) glycol ii) and slurry(glycol + straw char 10 wt.%), with feeding rate of 12.6 kg/h and12.4 kg/h in Test 1 and 10.05 kg/h in Test 2. The air numbers were0.57 and 0.46 in Test 1 and 0.52 in Test 2. Air enriched with oxygen(53.8 vol.% and 69.3 vol.% in Test 1 and 41.2 vol.%. in Test 2) wereused as gasification agent. Syngas after the reactor was cooled by atube-in-tube heat exchanger and a scrubber. During experiment, thegasification process was monitored by the reactor temperature andsyngas composition. Experimental parameters and syngas compositionin this study are summarized in Table 1. The syngas composition differsdue to different feedstocks: glycol and slurry and different experimentalconditions in Test 1 and Test 2.

2.2. Feedstocks characteristic

The feedstocks employed for biomass gasification were straw charand glycol. In order to achieve well defined operating conditions glycolwas considered as model feedstock. The energy-rich straw char (mixedwith glycol) obtained from pyrolysis were gasified in an entrained-flowgasifier and converted into synthesis gas, a chemically reactive mixtureof CO, H2 and so on. The main reason to use glycol as a model fuel wasthe oxygen content, which is comparable to pyrolysis oil from biomass[24]. The physico-chemical characteristics: elemental (C, H, N, S, O)composition, heating value, chlorine content, and volatiles in feedstocksused in the gasification process were analyzed.

pilot-scale gasifier REGA [22].

Table 1Experimental parameters and syngas composition in this study.

Parameters Glycol (Test 1) Slurry (Test 2)(Straw char 10 wt.%)

Test 1–1 Tests 1–2, 1–3 Test 2–1 Tests 2–2, 2–3

Fuel flow rate (Kg/h.) 12.60 12.42 10.05 10.05Stoichiometric ratio (λ) 0.57 0.46 0.523 0.523O2 content of gasificationagent (%)

53.70 69.3 41.82 41.82

CO (%) 22.86 29.41 21.79 21.36CO2 (%) 21.98 19.17 16.59 16.18H2 (%) 20.77 30.03 18.91 18.71CH4 (%) 0.019 0.49 0.44 0.52Corg (mg C/Nm3) 60 1990 1950 2200Syngas flow rate (Nm3/h.) 29 26 26 26

Table 2Elemental composition, heating value, chlorine content in straw char and glycol used ingasification.

Parameters Mass (%, dry basis)

Straw char Glycol

C 60.4 38H 3.4 8.9N 1.2 0.3S 0.2 b0.0001O 22.0 43.6Higher heating value (MJ/kg) 23.2 18.8Chlorine (mg/kg) 4183 36.5

Table 3Trace metals content in straw char and glycol used in gasification.

Metals Concentration (mg/kg)

Straw char Glycol

As b1 b0.1Be b0.5 b0.05Cd 0.2 b0.05Cr 1390 b0.5V b1 b0.1Ni 2190 b0.05Pb 41 b0.1Se b1 b0.1Hg 0.002 0.0002

56 D. Pudasainee et al. / Fuel Processing Technology 120 (2014) 54–60

2.3. Sampling and determination of selected metals in feedstocks andsyngas

Selected alkali, alkaline, trace metals content in straw char and eth-ylene glycol (glycol) used for gasification and in resulting syngas weredetermined. Concentration of selected alkali and alkaline metals (Na,K, Ca and Mg) in feedstocks were high enough to be analyzed byInductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES)(Optima 4300 DV, PerkinElmer Inc.). Trace metals concentrationswere determined using Inductively Coupled PlasmaMass Spectroscopy(ICP-MS) (X Series 2, Thermo Scientific).

Particulate and gas samples were collected to determine selectedtrace metals, including Hg speciation, in syngas according to US EPAmethod 29. US EPA method 29 is a multi metals analysis method, in-cludingmercury analysis with speciation at the same time. Syngas sam-plewaswithdrawn from the duct after gasifier (after cooler). Particulatematter was collected in a heated filter at the front end of the probe.Syngas withdrawn was passed through the filtration system followedby a series of impingers in an ice bath. All metals including oxidizedHg were collected in an aqueous acidic solution of hydrogen peroxide(5%HNO3/10%H2O2) solution. Elemental Hg was collected in an acidicsolution of potassium permanganate (4%KMNO4/10%H2SO4) solution.Any remaining moisture was removed in the last impinger, in a seriescontaining silica gel. Syngas temperature at the sampling point ranged169–196°C. Recovery of samples was carried out immediately aftersampling. The recovered samples were pre-treated, digested and ana-lyzed using ICP-MS. Syngas components such as CO, CO2, H2, Corg, CH4,were measured with analyzers installed at the gasifier.

2.4. Morphology of particulate matter in syngas

Fly ash particles were analyzed by scanning electron microscope(SEM) formorphological analysis. ZEISS Supra 55VP SEM,field emissiongun (FEG), secondary electron detector type instrument was used.Samples preparation and handling for particle analysis by SEM weredone according to standard operating procedures. Different magnifica-tions were applied to clearly view the morphology of particles.

3. Results and discussion

3.1. Feedstocks characteristic

Feedstocks characteristics are important parameter determiningmetals distribution, transformation and emission during biomass gasifi-cation. The physico-chemical characteristics: elemental (C, H, N, S, O)composition, higher heating value, chlorine content, in straw char andglycol used in the gasification process are shown in Table 2. From ele-mental analysis it can be seen that straw char from pyrolysis has highercarbon content, even higher than biomass because during pyrolysis vola-tiles are removed from the straw and carbon remains in the char. Higher

heating value of straw char (23.2 MJ/kg) and glycol (18.8 MJ/kg) washigher than of biomass (14–15 MJ/kg). Chlorine content in straw charand glycol were 4183 and 36.5 mg/kg, respectively. Average chlorinecontent of coal, biomass and straw pellets are reported 90–1100 mg/kg,100–6000 mg/kg [25] and 2700 mg/kg [26], respectively. Noticeably,chlorine content in straw char ismuch higher than coal.With the increas-ing use of biomass, mainly agricultural residues, the problem of chlorideinduced corrosion is increasing. Chlorine concentration in biomass hasan important role in determining alkali vaporization and corrosion. Chlo-rine aids the movement of many inorganic compounds, mainly K. Thus,high chlorine concentration in fuel possesses higher risk since alkali actas a Cl carrier to super heater, where elemental chlorine is released. Ashcontent of straw char at 815 °C was 19%. Volatile matters in straw charon an average were 25.6% to 25.8%. Moisture content and fix carbon instraw char sample were 4.3% and 51%, respectively. In general, volatilematter was higher in straw char compared to coal whereas heatingvalue and fixed carbon were lower.

Elemental content in straw char and glycol were, C 60.4% and 38%,S 0.25% and b0.0001%, H 3.4% and 8.9%, N 1.2% and b0.1% and O 22.0%and 43.6%, respectively. Sulfur content was lower. Sulfur content infuel has been reported effecting Hg emission and speciation in combus-tion flue gas. Sulfur content in fuel and the SO2 concentration in flue gashave an influential role on Hg0 oxidation in flue gas. The low sulfur fuelis more preferred because firstly it reduces SO2 emission and secondly itpromotes Hg removal in control devices.

3.2. Metals content in feedstocks

Trace metals concentration in biomass differs with place of origin,types, geology etc. Metals emission concentration in syngas from bio-mass gasification is without doubt related to their content in biomass.Alkali and alkaline metals are abundant in biomass. In general, K, Na,Ca and Mg content in biomass are higher than coal. Metals content instraw char and glycol used in gasification is presented in Table 3. Metalscontent in glycol and straw char were Na (1.5 and 395 mg/kg), K (b1and 42,800 mg/kg), Mg (b1 and 3560 mg/kg), and Ca (b1 and16,800 mg/kg). Reported value of Na and K in straw/grass ranged be-tween 70–15,000 mg/kg and 5000–30,000 mg/kg. Generally, biomass

57D. Pudasainee et al. / Fuel Processing Technology 120 (2014) 54–60

has high alkali content that may have catalytic influence in gasificationprocess [27,28]. Similarly, trace metals concentration in straw char andglycol were Cr (b0.5 and 1390 mg/kg), Ni (b0.05 and 2190 mg/kg), Pb(b0.1 and 41 mg/kg), As (b0.1 and b1), Be (b0.05 and b0.5), Cd (b0.05and 0.2), V (b0.1 and b1 mg/kg), and Se (b0.1 and b1). Hgconcentration in glycol and straw char ranged 0.0002 mg/kg, and0.002 mg/kg, respectively. Documented metal concentrations in bio-mass are As (0.1–6 mg/kg), Cr (0.15–25 mg/kg), Cd (0.1–0.3 mg/kg),Ni (0.2–10 mg/kg), Pb (1–5 mg/kg), V (0.1–3 ppmw), and Hg (0.01–0.05 mg/kg) [29,30].

3.3. Trace metals emission, distribution into particle and gas phase insyngas

Trace metals introduced into a gasifier, mainly from feedstocks(straw char and glycol) undergo transformation and removal withinthe system (boiler, cooler, syngas cleaning devices) and the remainingenters the chemical synthesis process (in case of bioliq®). Table 4shows trace metals concentration and distribution in syngas at twodifferent gasification feedstocks (i) glycol and (ii) slurry. The emissionconcentration of trace metals from glycol gasification were in theorder of Cr N Pb N Ni N Hg N As N Se N Be N V N Cd. In Test 1 (i.e. glycolgasification), metals concentration in syngas ranged from 0.1 μg/Nm3

(V) to 4.6 μg/Nm3 (Cr). In Test 2 (slurry gasification) the emissionconcentration of trace metals were in the order of Ni N Cr N Pb N As N

V N Be N Se N Cd N Hg. Metals concentration in syngas ranged from2.4 μg/Nm3 (Hg) to 53.2 μg/Nm3 (Ni). Metals concentration in syngasincreased moving from Test 1 to Test 2 because there were highercontent ofmetals in straw char compared to glycol. Themeasured emis-sion concentrations of metals in syngas are less compared to emissionfrom coal gasification. However, it should be noted that with theincreasing portion of char in glycol (future target of bioliq®) tracemetals emission concentrationwill increase. Unlike combustion system,gasification system is more sensitive with increased trace metals con-centration considering their effect in catalyst deactivation, so effortsneed to be directed to control trace metals as minimum as possible,e.g., Hg and As b 5 ppb [15,16]. In order tomeet the safe limit highly ef-ficient gas cleaning system is required. Based on the present study, re-moval efficiency more than 78% is required for As, which is one of thechallenges to meet.

All the heavy metals were more enriched in fly ash from slurrygasification compared to glycol gasification. In general, more portionsof trace metals were distributed into gas phase. In Test 1, V, As and Bewere mainly distributed into particulate phase whereas, Ni, Cd and Hgwere mainly distributed in gas phase (Fig. 2). Tendency of metals (Be,Pb, Cr, Hg) to speciate into particle phase increased in Test 2 than Test1, which is related to increase in fly ash particle and the unburnedcarbon content in it. In average, fly ash particles in syngas were59.5 mg/Nm3 (Tests 1–1, 1–2, 1–3) and 1511.7 mg/Nm3 (Tests 2–1, 2–2, 2–3). Similarly, carbon content in syngas were 60 mg C/Nm3 (Tests

Table 4Trace metals concentration (μg/Nm3) and distribution into particle and gas phase insyngas, average value of 3 measurements.

Metals Glycol gasification Slurry (straw char10 wt.%) gasification

Particulate Gas Particulate Gas

V 0.1 0.02 9.9 4.8As 0.8 0.1 18.5 3.8Be 0.3 0.1 9.9 0.04Se 0.3 0.3 3.0 4.8Pb 1.4 1.9 27.3 4.5Cr 1.8 2.8 28.7 5.8Ni 0.7 1.8 15.1 38.1Cd 0.03 0.1 0.9 4.3Hg 0.01 2.2 0.5 1.9

1–1) and 1900 mg C/Nm3 (Tests 1–2,1–3), whereas 1950 mg C/Nm3

(Test 2–1) and 2200 mg C/Nm3 (Tests 2–2, 2–3). Unburned carbon con-tains active carbon sites where metals can bind [31]. The increase in flyash particles in syngas increased the sorption sites for metals resultingincreased particle bond metals in Test 2 compared to Test 1. V and Sewere speciated more into gas phase even though the carbon content insyngas increased (Test 2). No significant differences in distribution behav-ior into particle and gas phase (in Test 1 and Test 2)weremeasured forNi,Cd and As. However, it is to be noted that metals in syngas, in general,were distributed more in gas phase compared to combustion flue gas.

Metals distribution into particle and gas phase in syngas from bio-mass gasification is affected by several possible factors such as operatingtemperature, reactor configuration, chlorine, sulfur, moisture content,oxygen availability, element constituents and so forth [32]. Consideringthe fact that trace metals behave differently in different gasificationenvironment (syngas components, operating conditions), the behaviorof metals species during gasification cannot be generalized for widerimplications. The effects of syngas constituents on metals distributioninto gas and particle phase are to be considered in the future study.However, one possible factor contributing higher portion of gas phasemetals in Test 1 is availability of very less amount of particle matterfor metals to be adsorbed. Whereas, might be due to formation ofmore metals chlorides species attributed by higher amount of chlorinecontent in straw char (Table 2), in Test 2.

3.4. Mercury emission and speciation in syngas

Mercury, unlike other trace metals behaved differently due to highvolatility. Hg speciation generally refers to its distribution into variouschemical forms. Emission of Hg compounds in gasification syngas existsin three main forms: elemental (Hg0), oxidized (Hg2+) and particlebond (Hgp). (i) Gaseous elemental mercury (Hg0): mostly associatedwithminerals, relatively non-reactive and insoluble inwater. Thus, usu-ally escapes emission control equipment and emitted into the atmo-sphere. (ii) Reactive gaseous or oxidized mercury (Hg2+): is soluble inwater and can be removed in wet scrubbers. (iii) Particulate mercury(Hgp): associated with particles (ash). Hgp can be removed in particu-late control devices. These all together comprises total Hg (HgT).

As expected, Hg in syngaswas speciated predominantly in elementalform (63.0% to 71.5%) followed by oxidized form (27.5% to 32.7%). Theshare of particle bondmercury was less (1 to 4.4%). Decreasing tenden-cy of Hgp in glycol than slurry gasification, which is due to the reductionof fly ash particles and carbon content in it, as discussed in Section 3.3.Elevated temperature and reducing atmosphere in gasifier resultedmost of the Hg in elemental form. Hg in fuel/feedstocks is believed tovaporize completely and converted into Hg0 in gasification zone, sinceHg0 is thermodynamically stable at high temperature in boiler. In com-bustion flue gas, as Hg0 passes the flue gas stream it may further reactwith acid gases, halogens, and thus mostly mercury in downstream ofboiler is found as Hg2+ compounds. The degree of conversion of Hg0

to Hg2+ and Hgp primarily depends on flue gas components, operatingconditions, process configuration, quenching rate, air pollution controldevices (APCDs) configuration etc.

In contrast to combustion environment, in gasification environment,due to the reducing atmosphere and reactionswith syngas componentssuch as CO and SO2; Hg2+ compounds may be reduced to Hg0 as it passdownstream of the boiler. Lack of oxygen at reducing atmosphere in thegasifier prevents the formation of chlorine (which may oxidize Hg)from hydrogen chloride. These might be the likely explanation forhigher portion of Hg0 in our measurements. The present measurementis also in agreement with the study by Li et al., [20] where it is reportedthat at the gasifier outlet mercury content in coal were released mainlyas Hg0. Variation in Hg emission concentration and speciation withinthe samples is related to the non-uniform distribution of Hg in feed-stock, fuel composition (especially Cl, S content), operating parameters,interaction with syngas components, fly ash properties etc.

Fig. 2. Trace metals distribution into particle and gas phase in syngas. T#1: Glycol gasification; T#2: Slurry (straw char 10 wt.% gasification.

Table 5REI of trace metals in fly ash from slurrygasification.

Metals REI

V 0.1As 1.4Be 1.5Se 0.3Pb 1.2Cr 0.3Ni 0.3Cd 0.1Hg 0.1

58 D. Pudasainee et al. / Fuel Processing Technology 120 (2014) 54–60

Theoretical chemical equilibrium calculation provides a valuableinsight to the stable species that are formed during combustion processas a function of temperature. Thermodynamic model estimation showsthat mercury was more volatile in reducing gasification condition thanoxidizing conditions [21]. This might be because volatile gaseous com-pounds like Cl−, SO4,− and OH− are more stable in reducing environ-ment. Study has shown that Hg0 oxidation by gas-phase reaction withindividual syngas components is very less. However, in simulatedsyngas conditions, Hg0 oxidation in gas phase reactions occurred onlyeither Cl2 or HCl + COS was introduced in syngas mixture. Hg0 oxida-tion was more notable after involvement of fly ash in syngas [33].

Hg distribution into particle phase was lesser. The adsorbing abilityof fly ash particles determines Hg speciation into particle and gasphase. The absorption capacity can be increased with treating fly ashwith chemicals such as bromine [34]. Fly ash (with unburned carbon)contains active carbon sites where Hg2+ can bind. Cl and S in fly ash(Hg\Cl and Hg\S bond respectively) and the temperature of flue gasare important in determining Hg sorption capacity [35]. On the otherhand, sulfur oxides such as SO2 and SO3 can bind to these active sitesinhibiting the binding of Hg [36]. Thus, sorption and desorption of Hgin fly ash depends on syngas condition and interactionwith several spe-cieswhich is bit complex, needs further studies in controlled conditions.

3.5. Relative enrichment of metals in fly ash

Relative Enrichment index (REI) reflects the change in concentrationof trace metals in combustion products. REI is also used to distinguishthe different classes of elements. REI was calculated using the followingformula [37].

REI ¼ Concentration of element in ash� Ash percentage in the feed fuelConcentration of element in feed fuel� 100

REI of trace metals in fly ash is shown in Table 5. During glycolgasification, there were very less particles in syngas and almost all thetrace metals were distributed into gaseous form, thus REI values weresignificantly lower and are not comparable to that in glycol slurry gasi-fication. Hence, the data is not presented here. All the tracemetals weremore enriched in fly ash from slurry gasification compared to glycol flyash. The REI of V, Se, Cr, Ni, Cd, and Hg in fly ash were lower than 1, im-plies that these metals tend to volatilize in reducing gasificationenvironment. Mercury is highly volatile and was almost distributed insyngas with least REI value (0.01) in fly ash. REI of As, Be, and Pb in flyash were 0.01 in glycol gasification, whereas more than 1.2 in slurry

gasification. This reveals that availability of particle and carbon contentin syngas has strong influence in determining the distribution behaviorof As, Be, Pb. However, the influence of particle and carbon content insyngas was not strong for V, Se, Cr, Ni, Cd and Hg. Depending on volatil-ization and REI, trace elements in coal ashes are classified into threeclasses [38,39]. Non-volatile elements are mainly collected with flyash or bottom ash.Whereas volatile elements like Hg and an intermedi-ate volatilemetals for example As and Se are emitted to the atmosphere.Ni, Pb, Cd, and Cr fall in class II, which are distributed into particles andmostly removed in the APCDs.

3.6. Morphology of fly ash particles

Fig. 3 (a) shows the morphology of straw char particles beforegasification showing irregular shape particles, those are rich in carboncontent as shown in elemental analysis. Fig. 3 (b) shows the morpholo-gy of particles in syngas emitted from slurry gasification. Particles withheterogeneous size and shapes were observed. No ash agglomerationwas observed. The irregular shaped particles are dominant, which areunburned char particles and mineral substances. The minerals presentin particle surface in different samples are almost the same. Basically,two types of particles existed in fly ash from gasification (i) irregular—mostly unburned carbon (char) particles and mineral substances, and(ii) spherical particles—formed by volatilization and condensation.Mostly, the spherical particles were lesser than 10 μm in geometricdiameter. This implies that melting might have occurred only in a verylimited region, e.g. on the surface of burning char [40] and feeding offine milled straw char in the gasifier.

Fig. 3 (c) shows the magnified view of the spherical particle. Spher-ical particles are formed bynucleation and condensation.Micro-spheres

Fig. 3. SEM photograph of fly ash particles in syngas emitted from slurry (straw char 10 wt.%) gasification. (a) Char particles before gasification (b) Overall view of particles at2000× magnification, (c) Spherical particles at 10,000× magnification (d) Char particles at 5000× magnification.

59D. Pudasainee et al. / Fuel Processing Technology 120 (2014) 54–60

were not seen to cluster each other. Fig. 3 (d) shows the closer view ofthe char particles. The char particles appear spongy and hole-like struc-tures appears in the surface due to release of volatiles from char parti-cles. Some fine spherical particles are seen trapped in char particles. Itis also evident from quantitative analyses that syngas contained signif-icantly higher levels of unburned carbon in slurry compared to glycolgasification. The unburned carbon has some oxidant/adsorption sitessimilar to activated carbon. The occurrence of more number of charparticles in syngas is favorable for reducing the emission of mercuryinto the atmosphere [38,41]. It can therefore be anticipated to havemore reduction of trace metals emission from gasification system thanthe similar combustion system.

4. Limitations and uncertainties

We are not aware of standard method for sampling and analysis ofmetals in syngas. In the absence of validated method for measuringmetals in syngas from gasification, the sampling methods developedfor metals measurement in flue gas from combustion system has beenused [9]. However, this has certain limitations such as lack of informa-tion about influence of fly ash cake particle, influence of syngas compo-nents etc. As presented in Section 3.3 syngas from slurry gasificationcontained higher fly ash particles and carbon content in it. Fly ash parti-cles collected in thefilter (filter cake layer) during samplingmight influ-ence the adsorption of metals in the cake layer affecting the distribution

ofmetals into particle and gas phase. In general, this effect is anticipatedto be more for Be, Pb, Cr, and Hg. The effects of filter cake on adsorptionbehavior of metals in gasification environment are to be quantified inthe future.

Due to variation in operating conditions, syngas chemistry, paramet-ric uncertainties (such as measurement, sampling and systematicerrors) etc. emission measurement is subjected to uncertainty. All theefforts were made to ensure that the data collected during this studyare of higher quality. Equipments and analyzers were calibrated andoperated according to the manufacturer's specification. Sampling, han-dling, storage, recovery, analysis were done as per standard operatingprocedure. Sampling, tests and analysis, at the identical condition,were carried out three times to ensure reliability.

5. Conclusions

The physico-chemical characteristics: elemental composition, heat-ing value, chlorine content, and selected trace metals in straw charand glycol used in the gasification process were analyzed. Trace metalsemission concentrations, distribution into particle and gas phase weremeasured, and particle morphology was observed. The following con-clusions can be drawn from the present study.

1. Concentrations of all the metals (As, Be, Cd, Cr, V, Ni, Pb, Se, Hg) stud-ied were very low in glycol. However, emission concentrations of

60 D. Pudasainee et al. / Fuel Processing Technology 120 (2014) 54–60

metals in syngas fromglycol gasificationwere detected. During glycolgasification, metals concentration in syngas ranged from 0.1 μg/Nm3

(V) to 4.6 μg/Nm3 (Cr). In slurry gasification, metals concentrationranged from 2.4 μg/Nm3 (Hg) to 53.2 μg/Nm3 (Ni). Consideringtheir contribution in catalyst deactivation, efforts are required tocontrol trace metals as minimum as possible.

2. In general, metals distribution into gas phase in syngas was higherthan combustion flue gas. Tendency of metals to speciate into parti-cle phase increased in slurry gasification than glycol gasificationdue to increase of fly ash particles and carbon content in it.

3. Mercury emission concentrations in syngas ranged from 2.2 to2.4 μg/Nm3. Mercury in syngas was speciated predominantly intoelemental form followed by oxidized form whereas share of particlebond mercury was the least.

4. The REI values showed the tendency of trace metals to volatilizemore in reducing gasification environment. Availability of particleand carbon content in syngas had strong influence in distributionbehavior of As, Be, and Pb than other metals studied.

Study on relation of tracemetals emission and speciationwith oper-ational variables (temperature and gas composition) are to be carriedout.

Acknowledgments

Dr. Deepak Pudasainee acknowledges his postdoctoral researchfellowship from Alexander von Humboldt Foundation, Germany.Authors thank Dr. Britta Bergfeldt, Dr. Günter Beuchle from Institute ofTechnical Chemistry, KIT for their help during experiment and sampleanalysis.

References

[1] R. Saxena, D. Adhikari, H. Goyal, Biomass-based energy fuel through biochemicalroutes: a review, Renewable and Sustainable Energy Reviews 13 (2009) 167–178.

[2] F. Trippe, M. Fröhling, F. Schultmann, R. Stahl, E. Henrich, Techno-economic assess-ment of gasification as a process step within biomass-to-liquid (BtL) fuel andchemicals production, Fuel Processing Technology 92 (2011) 2169–2184.

[3] T. Yoshioka, S. Hirata, Y. Matsumura, K.W. Sakanishi, Biomass resources and conver-sion in Japan: the current situation and projections to 2010 and 2050, Biomass andBioenergy 29 (2005) 336–346.

[4] A.C. Caputo, M. Palumbo, P.M. Pelagagge, F. Scacchia, Economics of biomass energyutilization in combustion and gasification plants: effects of logistic variables, Bio-mass and Bioenergy 28 (2005) 35–51.

[5] L. Devi, K.J. Ptasinski, F.J.J.G. Janssen, A review of the primary measures for tar elim-ination in biomass gasification processes, Biomass and Bioenergy 24 (2003)125–140.

[6] Z.A.B.Z. Alauddin, P. Lahijani, M. Mohammadi, A.R. Mohamed, Gasification ofligno-cellulosic biomass in fluidized beds for renewable energy development: areview, Renewable and Sustainable Energy Reviews 14 (2010) 2852–2862.

[7] B.C. Folkedahl, A.C. Snyder, J.R. Strege, S.J. Bjorgaard, Process development anddemonstration of coal and biomass indirect liquefaction to synthetic iso-paraffinickerosene, Fuel Processing Technology 92 (2011) 1939–1945.

[8] European Union (EU), Directive 2009/28/EC of the European parliament and of thecouncil, Official Journal of the European Union L 140 (2009) 16–62.

[9] H. Cui, S.Q. Turn, V. Keffer, D. Evans, T. Thai, M. Foley, Study on the fate of metal el-ements from biomass in a bench-scale fluidized bed gasifier, Fuel 108 (2013) 1–12.

[10] AMAP/UNEP, Technical background report for the global mercury assessment 2013,Arctic Monitoring and Assessment Programme, Oslo, Norway/UNEP ChemicalsBranch, Geneva, Switzerland2013.

[11] X. Huang, M. Li, H.R. Friedli, Y. Song, D. Chang, L. Zhu, Mercury emissions frombiomass burning in China, Environmental Science and Technology 45 (2011)9442–9448.

[12] H.D. Choi, T.M. Holsen, P.K. Hopke, Atmosphericmercury (Hg) in the Adirondacks: con-centrations and sources, Environmental Science and Technology 42 (2008) 5644–5653.

[13] K. Salo, W. Mojtahedi, Fate of alkali and trace metals in biomass gasification,Biomass and Bioenergy 15 (1998) 263–267.

[14] A.C. Larsson, J. Einvall, A. Andersson,M. Sanati, Targeting by comparison with labora-tory experiments the SCR catalyst deactivation process by potassium and zinc salts ina large-scale biomass combustion boiler, Energy and Fuels 20 (2006) 1398–1405.

[15] G.B. Hawkins, The impact on catalyst performance due to poisoning and foulingmechanisms, http://www.gbhenterprises.com/the%20impact%20on%20catalyst%20performance%20of%20poisons%20%26%20fouling%20mechanisms%20wsv.pdf2013((accessed April 17 2013)).

[16] J. Dunleavy, Mercury as a catalyst poison, Platinum Metals Review 50 (2006) 156.[17] F. Vejahati, Z. Xu, R. Gupta, Trace elements in coal: associations with coal and min-

erals and their behavior during coal utilization—a review, Fuel 89 (2010) 904–911.[18] D. Pudasainee, S.J. Lee, S.H. Lee, J.H. Kim, H.N. Jang, S.J. Cho, Y.C. Seo, Effect of selec-

tive catalytic reactor on oxidation and enhanced removal of mercury in coal-firedpower plants, Fuel 89 (2010) 804–809.

[19] D. Pudasainee, Y.C. Seo, J.H. Kim, H.N. Jang, Fate and behavior of selected heavymetals with mercury mass distribution in a fluidized bed sewage sludge incinerator,Journal of Material Cycles and Waste Management 15 (2013) 202–209.

[20] Y. Li, J. Zhang, Y.C. Zhao, C. Zheng, Volatility and speciation of mercury during pyrol-ysis and gasification of five Chinese coals, Energy and Fuels 25 (2011) 3988–3996.

[21] J.R. Bunt, F.B. Waanders, Trace element behaviour in the Sasol—Lurgi MKIV FBDBgasifier. Part 1—the volatile elements: Hg, As, Se, Cd and Pb, Fuel 87 (2008)2374–2387.

[22] T. Kolb, T. Jakobs, N. Zarzalis, Syngas from biomass-based slurry entrained-flow gas-ification, 10th Conference on energy for a clean environment, Lisbon, Portugal, July7–10 2009.

[23] H. Leibold, Dry high temperature sorption of HCl and H2S with natural carbonates,7th International Symposium on Gas Cleaning at High Temperatures (GCHT-7),Newcastle, Australia, June 23–25, 2008.

[24] H. Hartmann, M. Kaltschmitt, Energie aus Biomasse, Springer Verlag, BerlinHeidelberg NY, 2001.

[25] D.A. Tillman, D. Duong, B. Miller, Chlorine in solid fuels fired in pulverized fuelboilers-sources, forms, reactions, and consequences: a literature review, Energyand Fuels 23 (2009) 3379–3391.

[26] H. Kassman, J. Pettersson, B.M. Steenari, L.E. Åmand, Two strategies to reduce gas-eous KCl and chlorine in deposits during biomass combustion–injection of ammoni-um sulphate and co-combustion with peat, Fuel Processing Technology 105 (2013)170–180.

[27] D. Pudasainee, H.R. Paur, A. Bologa, K. Woletz, S. Fleck, T. Kolb, H. Seifert, Character-istics of feedstocks for biomass gasification, Chemie Ingenieur Technik 84 (8) (2012)1248.

[28] D.W. McKee, Mechanisms of the alkali metal catalysed gasification of carbon, Fuel62 (1983) 170–175.

[29] P. Monkhouse, On-line diagnostic methods for metal species in industrial processgas, Progress in Energy and Combustion Science 28 (2002) 331–381.

[30] P. Monkhouse, On-line spectroscopic and spectrometric methods for the determina-tion of metal species in industrial processes, Progress in Energy and Combustion Sci-ence 37 (2011) 125–171.

[31] S.D. Serre, G.D. Silcox, Adsorption of elemental mercury on the residual carbon incoal fly ash, Industrial and Engineering Chemistry Research 39 (2000) 1723–1730.

[32] D. Pudasainee, J.H. Kim, Y.S. Yoon, Y.C. Seo, Oxidation, reemission and mass distribu-tion of mercury in bituminous coal fired power plants with SCR, CS-ESP and wetFGD, Fuel 93 (2012) 312–318.

[33] D.Y. Lu, D.L. Granatstein, D.J. Rose, Study of mercury speciation from simulated coalgasification, Industrial and Engineering Chemistry Research 43 (2004) 5400–5404.

[34] S. Niksa, C.V. Naik, M.S. Berry, L. Monroe, Interpreting enhanced Hg oxidation withBr addition at plant miller, Fuel Processing Technology 90 (2009) 1372–1377.

[35] E.S. Olson, C.R. Crocker, S.A. Benson, J.H. Pavlish, M.J. Holmes, Surface compositionsof carbon sorbents exposed to simulated low-rank coal flue gases, Journal of theAir and Waste Management Association 55 (2005) 747–754.

[36] A. Presto, E. Granite, Impact of sulfur oxides onmercury capture by activated carbon,Environmental Science and Technology 41 (2007) 6579–6584.

[37] F. Huggins, F. Goodarzi, Environmental assessment of elements and polyaromatichydrocarbons emitted from a Canadian coal-fired power plant, International Journalof Coal Geology 77 (2009) 282–288.

[38] F. Goodarzi, Petrology of subbituminous feed coal as a guide to the capture of mer-cury by fly ash—influence of depositional environment, International Journal of CoalGeology 61 (2005) 1–12.

[39] R. Meij, Tracking trace elements at a coal fired plant equipped with a wet flue gasdesulphurization facility, Kema Science and Technology report, 7, 1989, pp. 269–355.

[40] J. Latva-Somppi, M. Moisio, E.I. Kauppinen, T. Valmari, P. Ahonen, U. Tapper, J.Keskinent, Ash formation during fluidized-bed incineration of paper mill wastesludge, Journal of Aerosol Science 29 (1998) 461–480.

[41] W.H. Gibbs, F. Clarke, A.K. Mehta, The fate of coal mercury during combustion, FuelProcessing Technology 65–66 (2000) 365–377.