Bioresource Technology 175 (2015) 51–58

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Co-gasification of biosolids with biomass: Thermogravimetric analysis and pilot scale study in a bubbling fluidized bed reactor Ming Ming Yu a, Mohammad S. Masnadi a,⇑, John R. Grace a, Xiaotao T. Bi a, C. Jim Lim a, Yonghua Li b a b

Clean Energy Research Centre, Department of Chemical & Biological Engineering, University of British Columbia, Vancouver, BC V6T 1Z4, Canada Highbury Energy Inc., Suite 1820 Cathedral Place, 925 West Georgia Street, Vancouver, BC V6C 3L2, Canada

h i g h l i g h t s  Switchgrass ash rich in potassium catalyzed and enhanced co-gasification reactions.  Biosolids minerals interacted with biomass minerals and inhibited gasification.  Increasing the feedstocks biosolids proportion adversely affected gasification.  No more than 25 wt% biosolids in the fuel feed is recommended.

a r t i c l e

i n f o

Article history: Received 7 August 2014 Received in revised form 6 October 2014 Accepted 9 October 2014 Available online 17 October 2014 Keywords: Biosolids Biomass Co-gasification Thermogravimetric analysis Bubbling fluidized bed

a b s t r a c t This work studied the feasibility of co-gasification of biosolids with biomass as a means of disposal with energy recovery. The kinetics study at 800 °C showed that biomass, such as switchgrass, could catalyze the reactions because switchgrass ash contained a high proportion of potassium, an excellent catalyst for gasification. However, biosolids could also inhibit gasification due to interaction between biomass alkali/alkaline earth metals and biosolids clay minerals. In the pilot scale experiments, increasing the proportion of biosolids in the feedstock affected gasification performance negatively. Syngas yield and char conversion decreased from 1.38 to 0.47 m3/kg and 82–36% respectively as the biosolids proportion in the fuel increased from 0% to 100%. Over the same range, the tar content increased from 10.3 to 200 g/m3, while the ammonia concentration increased from 1660 to 19,200 ppmv. No more than 25% biosolids in the fuel feed is recommended to maintain a reasonable gasification. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction In wastewater treatment plants (WWTPs), the solids in wastewater are separated, dewatered, and treated to meet the pollutant and pathogen (bacteria and viruses that cause diseases) requirements of local environmental protection agencies. The solids are called biosolids, composed mainly of water, organic matter and ash. According to the US Environmental Protection Agency (EPA, 1999), 60% of biosolids were being used in land application in 1999, while 40% were incinerated, landfilled, or disposed in other ways. Spreading of biosolids on land is controversial, as the public in some areas opposes application of biosolids because of the concern of perceived risks and odor concerns (Petersen and Werther, ⇑ Corresponding author at: Chemical and Biological Engineering Building, 2360 East Mall, Vancouver, BC V6T 1Z3, Canada. Tel.: +1 604 822 3121; fax: +1 604 822 6003. E-mail address: [email protected] (M.S. Masnadi). http://dx.doi.org/10.1016/j.biortech.2014.10.045 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.

2004). Incineration is often criticized because of secondary pollutants (Chun et al., 2011). Landfilling requires large area and sealing of the site boundary, so this method is also problematic (Seggiani et al., 2012). The above disposal methods are therefore far from perfect, and becoming less and less acceptable. Gasification of biosolids is an advantageous disposal method in many aspects compared to other disposal methods. During gasification, pathogens and pollutants are gasified or degraded at high temperatures. Thus, gasification can eliminate treatment processes such as the stabilization, digestion and composting, thereby reducing biosolids treatment costs. Also, for land applications, the public is worried about odors and risks, whereas gasification does not appear to worry the public. Compared to incineration, gasification is more efficient in terms of energy and causes less gas emission concern (Petersen and Werther, 2004; Saw et al., 2011). In addition, incineration extracts energy only in the form of heat, whereas the syngas produced from gasification has wider applications such as being burned in gas engines or converted to hydrogen and organic chemicals.

52

M.M. Yu et al. / Bioresource Technology 175 (2015) 51–58

Thermochemical conversion of fossil fuels and biomass mixtures has been investigated elsewhere (Habibi et al., 2012; Masnadi et al., 2015a,b; Masnadi, 2014; Masnadi et al., 2014). It is also likely to be necessary to mix biosolids with biomass like wood pellets before feeding to gasifiers. Several synergistic benefits might be achieved by biomass/biosolids co-feeding: (1) biosolids from WWTPs contain high moisture contents compared to other gasification feedstocks. For example, the moisture content of biosolids from Vancouver WWTPs is 70% according to a Greater Vancouver Sewerage and Drainage District Quality Control Annual Report (2011). Mixing biosolids with drier biomass can effectively reduce the average moisture content of the feedstock. (2) Biosolids usually have high ash content, typically 35% (i.e. Saw et al., 2011; Leckner et al., 2004; Nipattummakul et al., 2010). Co-gasifying biosolids with biomass of low ash content like wood pellets reduces the overall ash content in the feedstock. On the other hand, co-gasification of biosolids with fossil fuels does not seem to be feasible because of blend high ash content causing significant bed agglomeration (Dai et al., 2008). (3) The addition of biomass to an energy generation system lowers the CO2 footprint for that process. (4) Some components, such as alkali and alkaline earth metals (AAEM) in the biomass, may act as catalysts, promoting gasification of biosolids (Habibi et al., 2012). (5) Co-feeding may help to overcome some of the feeding difficulties commonly associated with biosolids feeding (Dai et al., 2012). In this work, co-gasification of biosolids and biomass was first studied in a thermogravimetric analyzer (TGA) at 800 °C in order to help understand the interactions between the fuels and their kinetic behavior. Next, pilot scale bubbling fluidized bed co-gasification of biosolids (0%, 10%, 25%, 50%, and 100% by weight) mixed with wood pellets was investigated. Results are presented showing measured syngas composition, syngas yield, char conversion, tar content and ammonia concentration. For 50% biosolids by mass in the fuel, bed temperature was varied from 720 to 830 °C in steps of 30 °C to investigate the influence of bed temperature on gasifier performance.

2. Methods 2.1. Materials Nexterra Systems Corp. of Vancouver, Canada, provided biosolids from a WWTP in Baltimore, USA. Two types of Canadian biomass samples were considered, wood pellets and switchgrass. The wood pellets were provided from a local supplier by Highbury Energy Inc. of Vancouver, BC. The switchgrass from Manitoba has been identified as having potential as an energy crop for Eastern Canada (Madakadze et al., 1996). Ash analysis were performed by Acme Labs in Vancouver, BC. Key properties of the biosolids and biomass samples are provided in Table 1. Biomass samples have much more oxygen than biosolids, whereas biosolids contain much more nitrogen (6.6% dry and ash free) than biomass samples. Although the wood pallets contain the highest calcium oxide content in its ash (21.4 wt%) which can catalyze gasification, its catalytic effect may not be significant because of the very low ash weight proportion (only 1.1%). The switchgrass has a higher ash content (6.3 wt%) and is rich in calcium and potassium (15.3 and 13.1 wt% oxides in its ash, respectively), and is expected to have the greatest catalytic effect on gasification. The biosolids ash also contains a high proportion of calcium (10.36 wt% oxide) which can enhance the fuel reactivity during gasification. The biosolids sample is also rich in phosphorous which is from the treated sewage sludge (Habibi, 2013). It

Table 1 Proximate, ultimate, and ash analysis of biosolids, wood pellets and switchgrass used.

a

Material

Biosolids

Wood pellets

Switchgrass

Water content (%)

9.2

5.9

6.0

Proximate (dry) Volatile (%) Ash content (%) Fixed carbon (%) Higher heating value(kJ/kg, dry)

82.3 10.9 6.8 22,100

83.6 1.1 15.4 19,300

76.9 6.3 16.8 19,600

Ultimate (dry and ash free) Carbon (%) Hydrogen (%) Oxygen (%) Nitrogen (%) Sulfur (%)

55.1 8.6 29.1 6.6 0.6

47.8 6.4 44.6 0.3 0.9

47.9 6.2 45.0 0.8 0.1

Ash analysis SiO2 Al2O3 TiO2 Fe2O3 CaO MgO K2O Na2O P2O5 LOIa

23.27 10.37 2.42 16.65 10.36 2.95 1.98 0.49 27.05 4.46

25.32 4.41 0.22 4.04 21.44 13.63 8.92 1.36 1.50 19.16

52.10 0.50 0.03 0.96 15.28 5.94 13.11 0.40 5.05 6.63

LOI, loss on ignition.

has been reported that phosphorous lowers the ash melting point during co-gasification (Coda, 2004). The deformation and flow temperatures of biosolids were measured and are 1136 and 1290 °C, much lower than for wood pellets, 1420 and 1450 °C respectively (Wilk et al., 2011). To prevent agglomeration and sintering, the temperature was kept below 1100 °C. 2.2. TGA experimental setup A Thermax500 high-pressure TGA was used for the kinetic study, as shown elsewhere (Masnadi, 2014). CO2 gasification of the different fuels was performed to compare their gasification rates at atmospheric pressure. In laboratory scale experiments (e.g. thermogravimetric analysis), CO2 is often used for kinetic studies. Catalysts which are active for the CO2 gasification have similar reactivity with steam (Pullen, 1984). The inlet gases were introduced from the bottom of the reactor. The outlet gases passed through a tar and moisture removal bucket. Fuel samples were loaded into a hemispherical quartz basket, 17 mm ID and 20 mm in height, connected to a load cell via a thin metal wire. A non-metal basket was chosen to minimize catalysis by the basket during the experiments. During the experiments, the weight of sample and temperature were monitored. For pyrolysis, the reactor was heated from room temperature to 800 °C at a heating rate of 25 °C/min and then maintained at 800 °C for half an hour. During this period, the carrier gas was nitrogen at a flow rate of 500 N mL/min. The purpose of pyrolysis was to yield 15 mg of char for gasification based on single fuel char yields. Masnadi et al. (2014) reported that mixing raw fuels before pyrolysis results in the same char yield as making char from each individual fuel. After the pyrolysis, the char samples were subjected to CO2 gasification. Hence, nitrogen was switched to CO2 with the temperature maintained at 800 °C throughout the gasification period. The experiments continued until the gasification was complete, i.e. until the weight of sample was no longer decreasing. The TGA experiments were replicated three times to verify the reproducibility of the data. At a 95% confidence level, the measurements for each case were found to be

53

M.M. Yu et al. / Bioresource Technology 175 (2015) 51–58

Fig. 1. Process flow diagram of the Highbury Energy Inc. bubbling fluidized bed gasification system.

statistically identical. The TGA load cell was calibrated and tared before any experiment using a 1 g standard reference weight. The gasification rate of each run is discussed by plotting the ash-free char conversion versus time during the gasification period. Ash-free char conversion is calculated from the mass of sample, i.e. where m0, mi, and mf are the initial, actual and final mass of the solid sample, respectively.

X¼

m0  mi mi  mf

ð1Þ

2.3. Pilot scale fluidized bed facility The pilot scale bubbling fluidized bed (BFB) facility was designed and built by Highbury Energy Inc. (HEI) (Watkinson et al., 2010). HEI developed calibrations, data acquisitions, methods/calculations, etc. The process flow diagram is shown in Fig. 1, and dimensions of the equipment are provided in Table 2. The hopper was filled with prepared feedstock before each run. During the gasification period, the screw feeder delivered feedstock to the bottom of the gasifier, assisted by conveying nitrogen. Building steam was heated by super-heaters to about 800 °C before entering the gasifier through a well-insulated steam pipe. The flow rate of steam was maintained at 3.5 kg/h. The gasifier was also

Table 2 BFB gasifier description. Hopper Steam super heater Gasifier Gasifier heaters

Bed material Bed height Cooler

0.22 m3 12  240 V  1800 W, 21.6 kW, 3.7 m long Cylindrical: ID = 100 mm, L = 1.2 m Two semi-cylindrical: 125 mm ID, 225 mm OD, 0.92 m long One full-cylindrical: 125 mm ID, 0.15 m long Silica sand 0.25 m Double pipe type: 12 m long, heat exchange area of 2.0 m2

well insulated, and its temperature was maintained at 850 °C by three electrical heaters. Producer gases leaving the top of the gasifier entered two cyclones in series, both insulated with glass fiber, where fly ash was separated and collected from the bottom. Then, produced gases passed through two coolers, the first cooled by ambient air and the second by water. Condensed water and tars were collected at the bottom of the coolers. After passing through the coolers, the producer gases entered a baghouse filter where the remaining ash particles were captured. The producer gases were combusted with natural gas in a roof-top burner. Temperatures and pressures along the path were monitored, displayed and recorded by a computer. Between the cyclones and coolers, gas samples were withdrawn for tar analysis and ammonia determination. Another sampling line between the coolers and baghouse extracted the producer gases for concentration analysis. Totally, five separate runs were performed, as shown in Table 3. All results are expressed on a dry weight basis, for example, syngas yield in m3/kg dry fuel. The fuel feed rate was maintained at 1.4 kg/ h for all five runs. Total feeding time was 2–3 h for each run. Tar sampling and ammonia determination were combined in one sampling line in series because they require similar conditions. The tar sampling equipment and protocol were set up by HEI. The Apex Instrument model XC-60 including a sample pump, control valves and dry gas meter extracted producer gases through the sampling line. Producer gases passed through impingement bottles containing solvents at around 0 °C in an ice bath. The first six impingement bottles containing acetone were used to absorb tar, while the next two contained 1 N sulfuric acid for ammonia Table 3 Operating conditions for five BFB runs at different biosolids proportions. Run number

1

2

3

4

5

Biosolids proportion in fuel (%) by mass Fuel feed rate (kg/h) Steam flow rate (kg/h) Steam/fuel mass ratio Bed temperature (°C)

10 1.27 4.04 3.18 857

0 1.27 3.48 2.74 855

100 1.5 3.43 2.29 854

50 1.29 3.52 2.73 825

25 1.47 3.28 2.23 828

54

M.M. Yu et al. / Bioresource Technology 175 (2015) 51–58

absorption. Ammonia was trapped and determined with a modified method CTM-027 of EPA (1997), with four impingers being used to capture ammonia, two with 0.1 N sulfuric acid and two empty. Due to limited space in one case, two impingers were used to absorb ammonia, one with 1 N sulfuric acid and one empty. After each gasification experiment, tar and ammonia were analyzed separately. Tar was separated by washing, evaporation and drying. Then, the tar content of the syngas was determined in units of g/m3 from the weight of the tar divided by the volume of syngas collected. The ammonia concentration was determined by back titration with 0.1 N NaOH. To measure gas concentrations of dry and tar-free gases, the gas analyzer was placed downstream of the condensers where water and tar in the producer gases had already been removed. Producer gases consisting mainly of CO, CO2, CH4, H2, and N2 were analyzed continuously by an on-line micro gas chromatograph (GC) with helium as the carrier gas. It took 4 min to analyze the gas concentrations for each gas sample, so a sample from the producer gases was injected and analyzed every 4 min. Syngas yield was determined from the syngas flow rate divided by the fuel feed rate in units of m3/kg. Char conversions were calculated from the char proportions in the feedstock and in the solids captured by the primary cyclone. Char is defined as the fixed carbon in proximate analysis in this study.

3. Results and discussion 3.1. TGA experiments The temperature gradient in the TGA sample bed due to endothermic gasification reactions was estimated and found to be insignificant. Also, mass transfer calculations showed that the internal and external resistances to mass transfer were negligible. See Masnadi (2014) for more details. Char conversions of biosolids, wood pellets, and switchgrass versus time are plotted in Fig. 2(a). The gasification rates of wood pellets and switchgrass were similar, but the gasification rate of biosolids was obviously slower, likely to be due to the lower micro-pore surface area of biosolids than biomass samples (Habibi, 2013). In addition, based on Table 1, the biosolids were rich in calcium and the switchgrass was rich in potassium. It is reported that the catalytic effect and mobility of potassium is higher than calcium during gasification (Walker et al., 1979; Wang et al., 2010; Radovic et al., 1984; Kannan and Richards, 1990; McKee, 1981; Masnadi, 2014). It took only 400 min for wood pellets and switchgrass to be completely converted, whereas it took 700 min for biosolids to complete the reaction. Consider the runs during which biosolids were co-gasified with wood pellets and with switchgrass, both at 50/50 weight proportions. The co-gasification rate is plotted and compared with gasification rate for the pure substances in Fig. 2(b) and (c). The cogasification rate of the mixture was much slower than expected. If no catalytic or inhibition occurred, the line for co-gasification would lie between the two lines for gasification of the pure materials as shown in Fig. 2(b) and (c). Clearly, inhibition must have occurred to slow down the co-gasification reaction. The catalytic effects of AAEM compounds such as K2CO3 in gasification have been documented for many years (Formella et al., 1986; Masnadi, 2014). These compounds can also undergo secondary reactions with mineral matter in biosolids. It is likely that the potassium and calcium of biomass samples were deactivated through interaction with biosolids minerals, like illite and kaolinite, to form a new mineral phase, e.g. Kalsilite (KAlSiO4) and Wollastonite (Ca3Si3O9) (Habibi et al., 2012; Formella et al., 1986; Masnadi, 2014; Kühn and Plogmann, 1983; Masnadi et al.,

2015a), as displayed in Fig. 3(a) and (b). Habibi (2013) conducted XRD analysis of biomass/biosolids partially gasified mixture samples and confirmed the formation of AAEM-Aluminosilicate crystals. Biosolids minerals can increase the reactivity of biosolids during gasification, and when they react with biomass AAEM, the AAEM is deactivated so that, inhibition was observed. Next, co-gasification tests were conducted on a 50:50 biosolids: switchgrass ash mixture. Masnadi (2014), Masnadi et al. (2015a,b) and Brown et al. (2000) reported that switchgrass ash contains high potassium content and should accelerate the reaction. Fig. 4(a) plots the co-gasification rate of biosolids with switchgrass ash. This shows clearly that there is a catalytic effect due to the switchgrass ash. Within 45 min, char conversion of mixture of biosolids with switchgrass ash reached 0.8. By comparison, 0.8 char conversion took switchgrass about 200 min and biosolids 330 min. The SG ash supplied enough potassium to satisfy the minerals in the biosolids ash. Then, the unreacted potassium acted as a catalyst for biosolids gasification. Habibi (2013) also reported formation of Whitlockite crystals, as presented in Fig. 3(c) (e.g. Ca9MgK(PO4)7), during biosolids/switchgrass ash co-gasification, which shows formation of multiple AAEM phosphate compounds as the potassium concentration increases. To investigate whether the inhibition effect might be due to the ash of biosolids, co-gasification of biosolids ash with biomass was performed, with the results being plotted in Fig. 4(b) and (c). From Fig. 4(b) and (c), inhibition occurred when co-gasifying biosolids ash with biomass, especially with switchgrass, possibly because of undesired interactions between the ash of biosolids and the ash of switchgrass. Clearly the inhibition effect by biosolids ash overcame the positive catalytic effect of the switchgrass ash, with the net result that co-gasification of biosolids with switchgrass was slower than gasification of each individual material. 3.2. Pilot scale BFB experiments Initially, it was planned to co-gasify switchgrass with biosolids in a pilot scale bubbling fluidized bed because of its catalytic effect. However, the TGA results indicated that neither wood pellets nor switchgrass could enhance co-gasification with low AAEM concentrations. It was therefore decided to use wood pellets instead of switchgrass as wood pellets are more likely to be used in gasification, at least in British Columbia. 3.2.1. Bed temperature effects In run 4 (see Table 3) with 50% biosolids and 50% wood pellets by mass, the impact of bed temperature was investigated over the range of 728–825 °C with steps of 30 °C, and approximately half an hour of near steady state operation after each temperature change. The syngas yield is plotted versus bed temperature in Fig. 5(a). As expected, higher temperature led to a dramatic increase in syngas yield because steam gasification is endothermic (Higman and Burgt, 2008). The syngas yield increased from 0.29 to 0.99 m3/kg as the bed temperature increased from 728 to 825 °C. Increasing the bed temperature could improve productivity, but the bed temperature was limited by melting and agglomeration of fuel ash in the gasifier (Petersen and Werther, 2004). Fig. 5(b) shows the impact of bed temperature on syngas composition. It is seen that H2 and CO did not exhibit obvious dependence on temperature. As temperature increased from 728 to 825 °C, the CO2 concentration decreased from 15.9% to 9.86%, whereas the CH4 concentration increased from 13.2% to 16.4%. Overall, the temperature had a relatively weak effect on the syngas composition for the limited temperature range covered in this study, similar to what Pinto et al. (2008) found from plotting syngas composition versus bed temperature.

M.M. Yu et al. / Bioresource Technology 175 (2015) 51–58

55

1.0

(a) Char conversion, X

0.8

0.6

0.4

0.2

biosolids wood switchgrass

0.0 0

100

200

300

400

500

600

700

Gasification Time (min)

1.0

(b) Char conversion, X

0.8

0.6

0.4 biosolids alone wood alone biosolids:wood 50:50 mixture expected gasification rate of biosolids:wood 50:50 mixutre

0.2

0.0 0

100

200

300

400

500

600

700

Gasification Time (min)

1.0

(c) Char conversion, X

0.8

0.6

0.4 biosolids alone switchgrass alone biosolids:switchgrass 50:50 mixture expected gasification rate of biosolids:switchgrass 50:50 mixutre

0.2

0.0 0

100

200

300

400

500

600

700

Gasification Time (min) Fig. 2. (a) Char conversion of biosolids, wood pellets and switchgrass vs. time. Cogasification of 50:50 biosolids by weight with (b) wood pellets; (c) switchgrass. Operating conditions: reactor temperature, 800 °C; sample size, 300–355 lm; initial mass of sample, 15 mg; CO2 flow rate, 500 mL/min.

Fig. 3. (a) Kalsilite (KAlSiO4); (b) Wollastonite (Ca3Si3O9); (c) Whitlockite crystalline structures. Elements: red: oxygen, pink: potassium, yellow: silicon, cyan: aluminum, silver: calcium, purple: magnesium, green: phosphorus (produced by VMD 1.9.1 software). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3.2.2. Biosolids proportion effects To study the effects of biosolids proportion in the fuel, operating conditions were maintained close enough for comparison to be made, except for the proportion of biosolids. The operating conditions are presented in Table 3.

Fig. 6(a) shows the impact of biosolids proportion in the fuel on syngas yield and char conversion. Char conversion decreased from 82% to 36% as the proportion of biosolids increased from 0% to 100%. In Fig. 6(a), syngas yield decreased from 1.38 to 0.47 m3/kg as the proportion of biosolids in the fuel increased over the same

56

M.M. Yu et al. / Bioresource Technology 175 (2015) 51–58 1.0

1.2

(a)

(a) 1.0

Syngas yield (m3/kg)

Char conversion, X

0.8

0.6

0.4

biosolids alone switchgrass alone biosolids:switchgrass_ash 50:50 mixture

0.2

0.8

0.6

0.4

0.0

0.2 0

100

200

300

400

500

600

700

740

760

Gasification Time (min)

820

800

820

60

(b) Gas molar concentration (%)

(b) 0.8

Char conversion, X

800

Temperature (°C)

1.0

0.6

0.4

biosolids alone wood alone biosolids_ash:wood 50:50

0.2

50

40

H2 CO CH4

30

CO2

20

10

0

0.0 0

100

200

300

400

500

600

700

760

780

Fig. 5. (a) Syngas yield and (b) syngas composition versus bed temperature. Operating conditions: feedstock, 50% biosolids with 50% wood pellets by mass; steam flow rate: 3.52 kg/h; fuel feed rate: 1.29 kg/h; steam/fuel mass ratio: 2.73. Error bars correspond to 95% confidence intervals. Temperature error bars are not plotted due to their very small values.

1.0

(c) 0.8

0.6

0.4

0.2

biosolids alone switchgrass alone biosolids_ash:switchgrass 50:50 mixture

0.0 0

740

Temperature (°C)

Gasification Time (min)

Char conversion, X

780

100

200

300

400

500

600

700

800

900

Gasification Time (min) Fig. 4. (a) Co-gasification of 50:50 biosolids with switchgrass ash. Co-gasification of 50:50 (mass:mass) biosolids ash with (b) wood pellets; (c) switchgrass. Operating conditions: reactor temperature, 800 °C; sample size, 300–355 lm; initial weight of sample, 15 mg; CO2 flow rate, 500 mL/min.

range. This decrease resulted from both higher ash content in the biosolids and lower char conversion in the gasification of biosolids. Several researchers (Saw et al., 2011; Peng et al., 2012) have found negative effects of increased biosolids proportion on char conver-

sion and syngas yield. From Fig. 6(a), the maximum syngas yield and char conversion occurred at 0% biosolids. However, since this study was motivated by the need for disposal of biosolids, blending of up to 25% biosolids to wood pellets did not cause significant decrease in biochar conversion and syngas yield. This is consistent with the literature where the optimal mixing ratio for co-gasification of biosolids with biomass has been reported to be 10–30% (Saw et al., 2011; Peng et al., 2012). Tar content and ammonia concentration are plotted versus biosolids proportion in Fig. 6(b). It is seen that the tar content in the syngas increased from 10.3 to 200 g/m3 as the biosolids proportion in the fuel increased from 0% to 100%, except for a proportion of 10%. The run with 10% biosolids proportion was the first run, and an inadequate number of impingers for tar absorption probably resulted in the low reported tar content, 2.3 g/m3. From Fig. 6(b), the ammonia concentration also increased with increasing biosolids proportion in the fuel, passing from 1660 to 19,200 ppmv as the biosolids proportion increased from 0% to 100%. These tar content and ammonia concentration trends are consistent with literature results (Saw et al., 2011; Peng et al., 2012; Pinto et al., 2008). Again, the presented results suggest that there will be only moderate increases in tar content and ammonia

57

M.M. Yu et al. / Bioresource Technology 175 (2015) 51–58 22000

250 80

Tar content in syngas NH3 concentration

60

1.0 40

0.8 0.6

Char conversion (%)

1.2

3

Syngas yield (m /kg)

3

Tar content in syngas (g/Nm )

1.4

20

Syngas yield Char conversion

0.4

(a)

0.2 0

20

40

60

80

0

20000 18000

200

16000 14000 150 12000 10000 100 8000 6000 50

4000

(b)

0

100

0

20

Biosolids proportion (%) in fuel

40

60

2000 0

100

80

Biosolids proportion (%) in fuel 60

60

H2

Gas molar concentrations (%)

H2

Gas molar concentrations (%)

NH3 concentration (ppmv)

1.6

CO CH4

50

CO2 40

30

20

10

(c)

0 0

20

40

60

80

CO CH4

50

CO2 40

30

20

10

(d)

0

100

0

20

40

60

80

100

Biosolids proportion (%) in fuel

Biosolids proportion (%) in fuel

Fig. 6. (a) Syngas yield and char conversion; (b) tar content and ammonia concentration (c) composition for runs 1–3; (d) composition for runs 4 and 5 vs. biosolids proportion in the feed. Error bars correspond to 95% confidence intervals.

concentration if the biosolids content in the feed does not exceed 25%. The syngas composition is plotted versus biosolids proportion in fuel in Fig. 6(c) and (d), with the results of runs 1–3 in (c) and results of runs 4 and 5 in (d). By splitting the plots, the trend of variation of gas concentrations on biosolids proportion can be observed more easily. In runs 1–3, the H2 concentration was always higher than the CO concentration. A jump in CO concentration and a drop in CH4 and H2 concentration were observed in runs 4 and 5 with 25% and 50% biosolids, and the CO:H2 ratio became higher than 1. Since the experimental system was slightly modified by HEI after run 3, the difference in the operating system between runs 1–3 and runs 4 and 5 appears to have caused the difference in the measured gas concentrations. Nevertheless, the slopes of decrease/increase for the same gas are almost the same for Figs. 6(c) and (d), with the H2 concentration decreasing with increasing biosolids proportion, while the CO and CH4 concentrations increased with increasing biosolids fraction in both cases. Consequently, the H2/CO ratio decreased with increasing biosolids proportion, while the CO2 concentration remained almost constant. CH4, the simplest hydrocarbon, can be an indicator of tars and other more complex hydrocarbons (Pfeifer et al., 2011). Therefore, the increase of CH4 concentration at higher biosolids fraction suggests an increase of tar content in the syngas, consistent with the above discussion of tar content and ammonia concentration.

4. Conclusions Kinetic study proved that switchgrass rich in potassium could catalytically accelerate gasification. However, this enhancement

can be overwhelmed by biosolids ash inhibition effect. Overall, biosolids tended to slow down gasification. From a study in the pilot scale BFB gasifier, increasing the proportion of biosolids in the fuel feedstock affected gasification performance negatively. For the purpose of biosolids disposal, the results suggest that the co-gasification mixing ratio should be chosen to be not more than 25% biosolids in the fuel feed in order to maintain a reasonable biochar conversion and syngas yield, as well as low tar and ammonia contents. Acknowledgements The authors thank Highbury Energy Inc. for their permission to use the HEI gasifier for this project. The authors are grateful to Nexterra Systems Corp. for their invitation to participate in gasification runs and for providing materials for this project. Financial support from the Natural Sciences and Engineering Research Council of Canada is also acknowledged with gratitude. References Brown, R.C., Liu, Q., Norton, G., 2000. Catalytic effects observed during the cogasification of coal and switchgrass. Biomass Bioenergy 18, 499–506. Chun, Y.N., Kim, S.C., Yoshikawa, K., 2011. Pyrolysis gasification of dried sewage sludge in a combined screw and rotary kiln gasifier. Appl. Energy 88, 1105– 1112. Coda, B., 2004. Impact of biomass co-firing on ash formation and ash behavior in IGCC plants. Netherlands Energy Research Foundation. Dai, J., Sokhansanj, S., Grace, J.R., Bi, X., Lim, C.J., Melin, S., 2008. Overview and some issues related to co-firing biomass and coal. Can. J. Chem. Eng. 86, 367–386. Dai, J., Cui, H.P., Grace, J.R., 2012. Biomass feeding for thermochemical reactors. Prog. Energy Combust. Sci. 38, 716–736. Environmental Protection Agency, 1997. Procedure for collection and analysis of ammonia in stationary sources, CTM-027.

58

M.M. Yu et al. / Bioresource Technology 175 (2015) 51–58

Formella, K., Leonhardt, P., Sulimma, A., Vanheek, K.H., Juntgen, H.D.A.O., 1986. Interaction of mineral matter in coal with potassium during gasification. Fuel 65, 1470–1472. Greater Vancouver Sewerage and Drainage District Quality Control Annual Report, Vancouver, Canada, 2011. Habibi, R., Kopyscinski, J., Masnadi, M.S., Lam, J., Grace, J.R., Mims, C.A., Hill, J.M., 2012. Co-gasification of biomass and non-biomass feedstocks: synergistic and inhibition effects of switchgrass mixed with sub-bituminous coal and fluid coke during CO2 gasification. Energy Fuels 27, 494–500. Habibi, R., 2013. Co-Gasification of Biomass and Non-Biomass Feedstocks (Ph.D. thesis). University of Calgary, Canada. Higman, C., Burgt, M., 2008. Gasification. Elsevier Science, Burlington, MA, the USA. Kannan, M.P., Richards, G.N., 1990. Gasification of biomass chars in carbon-dioxide dependence of gasification rate on the indigenous metal content. Fuel 69, 747– 753. Kühn, L., Plogmann, H., 1983. Reaction of catalysts with mineral matter during coal gasification. Fuel 62, 205–208. Leckner, B., Amanda, L., Luckeb, K., Wertherb, J., 2004. Gaseous emissions from cocombustion of sewage sludge and coal/wood in a fluidized bed. Fuel 83, 477– 486. Madakadze, I.C., Samson, D.L., Smith, D.L., Coulman, B.E., 1996. The phenological development and yield of warm season grasses in SW Quebec. Can. J. Plant Sci. 76, 354. Masnadi, M.S., 2014. Biomass/Fossile Fuel Co-Gasification With and Without Integrated CO2 Capture (Ph.D. thesis). University of British Columbia, Canada. Masnadi, M.S., Habibi, R., Kopyscinski, J., Hill, J.M., Bi, X., Lim, C.J., Ellis, N., Grace, J.R., 2014. Fuel characterization and co-pyrolysis kinetics of biomass and fossil fuels. Fuel 117 (B), 1204–1214. Masnadi, M.S., Grace, J.R., Bi, X., Lim, C.J., Ellis, N., 2015a. From fossil fuels towards renewables: inhibitory and catalytic effects on carbon thermochemical conversion during co-gasification of biomass with fossil fuels. Under Review. Masnadi, M.S., Grace, J.R., Bi, X., Lim, C.J., Ellis, N., 2015b. From coal towards renewables: catalytic/synergistic effects during steam co-gasification of switchgrass and coal in a pilot-scale bubbling fluidized bed reactor. Under Review. McKee, D.W., 1981. The catalyzed gasification reactions of carbon. In: Walker, P.L., Jr., Thrower, P.A. (Eds.), Chemistry and Physics of Carbon, vol. 16. Marcel Dekker, New York.

Nipattummakul, N., Ahmed, I., Kerdsuwan, S., Gupta, A., 2010. Hydrogen and syngas production from sewage sludge via steam gasification. Int. J. Hydrog. Energy 35, 11738–11745. Peng, L., Wang, Y., Lei, Z., Cheng, G., 2012. Co-gasification of wet sewage sludge and forestry waste in situ steam agent. Bioresour. Technol. 114, 698–702. Petersen, I., Werther, J., 2004. Experimental investigation and modeling of gasification of sewage sludge in the circulating fluidized bed. Chem. Eng. Process. 44, 717–736. Pfeifer, C., Koppatz, S., Hofbauer, H., 2011. Steam gasification of various feedstocks at a dual fluidized bed gasifier: impacts of operation conditions and bed material. Biomass Convers. Biorefin. 1, 39–53. Pinto, F., Andre, R.N., Lopes, H., Dias, M., Gulyurtlu, I., Cabrita, I., 2008. Effect of experimental conditions on gas quality and solids produced by sewage sludge cogasification. 2. Sewage sludge mixed with biomass. Energy Fuels 22, 2314– 2325. Pullen, J.R., 1984. Catalytic Coal Gasification. IEA Coal Research, London. Radovic, L.R., Walker Jr., P.L., Jenkins, R.G., 1984. Catalytic coal gasification: use of calcium versus potassium. Fuel 63, 1028–1030. Saw, W., McKinnon, H., Gilmour, I., Pang, S., 2011. Production of hydrogen-rich syngas from steam gasification of blend of biosolids and wood using a dual fluidised bed gasifier. Fuel 93, 473–478. Seggiani, M., Vitolo, S., Puccini, M., Bellini, A., 2012. Cogasification of sewage sludge in an updraft gasifier. Fuel 93, 486–491. U.S. Environmental Protection Agency, 1999. Biosolids generation, use, and disposal in the United States. Municipal and Industrial Solid Waste Division, Office of Solid Waste, EPA530-R-99-009. Walker Jr., P.L., Mahajan, O.P., Komatsu, M., 1979. Catalysis of lignite char gasification by various exchanged cations-dependence of activity on reactive atmosphere. Prepr. Div. Fuel Chem Am. Chem. Soc. 24, 10–16. Wang, J., Yao, Y., Cao, J., Jiang, M., 2010. Enhanced catalysis of K2CO3 for steam gasification of coal char by using Ca(OH)2 in char preparation. Fuel 89, 310–317. Watkinson, P., Li, Y., Haligva, J., 2010. IRAP Project 709261, High Grade Synfuels – Phase 1, Report to Accompany Final Claim. Wilk, V., Kitzler, H., Koppatz, S., Pfeifer, C., Hofbauer, H., 2011. Gasification of waste wood and bark in a dual fluidized bed steam gasifier. Biomass Convers. Biorefin. 1, 91–97.