Production of biochar from rice husk: Particulate emissions from the combustion of raw pyrolysis volatiles

Production of biochar from rice husk: Particulate emissions from the combustion of raw pyrolysis volatiles

Accepted Manuscript Production of biochar from rice husk: Particulate emissions from the combustion of raw pyrolysis volatiles Lewis Dunnigan, Peter J...

1MB Sizes 0 Downloads 3 Views

Recommend Documents

No documents
Accepted Manuscript Production of biochar from rice husk: Particulate emissions from the combustion of raw pyrolysis volatiles Lewis Dunnigan, Peter J. Ashman, Xiangping Zhang, Chi Wai Kwong PII:

S0959-6526(16)31959-X

DOI:

10.1016/j.jclepro.2016.11.107

Reference:

JCLP 8496

To appear in:

Journal of Cleaner Production

Received Date: 30 March 2016 Revised Date:

11 August 2016

Accepted Date: 17 November 2016

Please cite this article as: Dunnigan L, Ashman PJ, Zhang X, Kwong CW, Production of biochar from rice husk: Particulate emissions from the combustion of raw pyrolysis volatiles, Journal of Cleaner Production (2016), doi: 10.1016/j.jclepro.2016.11.107. 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 Wordcount: 5491

2

Production of Biochar from Rice Husk: Particulate Emissions from the

3

Combustion of Raw Pyrolysis Volatiles

4

AUTHORS: Lewis Dunnigana, Peter J. Ashmana, Xiangping Zhangb, Chi Wai

5

Kwonga*

8 9

School of Chemical Engineering, The University of Adelaide, Adelaide, SA

5005, Australia b

SC

7

a

Institute of Process Engineering, Chinese Academy of Sciences, Beijing,

China

10

*Author

for

correspondence

11

[email protected])

12

16

17

18

8313

0724;

Email:

EP

15

+61

AC C

14

(Telephone:

TE D

13

M AN U

6

RI PT

1

19

20

1

ACCEPTED MANUSCRIPT 1

Abstract Increasing energy demands and waste management concerns have

3

motivated agricultural producers to consider the decentralized conversion of

4

agricultural by-products for energy and value-added product (biochar)

5

generation. Due to the variability of fuel properties, direct combustion of

6

agricultural by-products with high ash contents, such as rice husk, may suffer

7

from increased fouling and slagging issues with high particulate matter (PM)

8

emissions. Combustion of the raw pyrolysis volatiles (bio-oil and synthesis gas

9

(syngas) mixtures) produced from pyrolysis with the inherent separation of ash

10

in the biochar may potentially mitigate these issues. In this study, PM emissions

11

from the combustion of the raw pyrolysis volatiles derived from the pyrolysis of

12

rice husk were evaluated at laboratory scale by using a combined pyrolysis and

13

combustion facility. Pyrolysis temperatures ranging from 400 °C to 800 °C were

14

used to generate raw pyrolysis volatiles with differing bio-oil to syngas ratios

15

which were then combusted at 850 °C. It was found t hat bio-oil dominated the

16

higher heating value of the raw pyrolysis volatiles produced at low pyrolysis

17

temperatures. The combustion of such raw pyrolysis volatiles with high bio-oil

18

content substantially increased the yields of PM10 and PM2.1. Linear

19

dependence was observed between PM emissions and bio-oil fraction in the

20

raw pyrolysis volatiles. Nevertheless, the pyrolysis-combustion process, with

21

>96% of the ash retained in biochar prior to combustion, is more favorable than

22

direct combustion for high ash biomass as far as PM emissions are concerned.

AC C

EP

TE D

M AN U

SC

RI PT

2

2

ACCEPTED MANUSCRIPT Keywords: Rice Husk, Biochar, Ash, Particulate Emissions, Pyrolysis, Raw

2

Pyrolysis Volatiles.

3

1. Introduction

RI PT

1

Re-utilization of agricultural by-products as biomass resources has great

5

potential to contribute to the development of the bio-economy for the co-

6

generation of value added products and bioenergy (Allen et al., 2015).

7

However, limited availability of high quality agricultural by-products for energy

8

applications results in the need to utilize more problematic raw materials with

9

broader variations in fuel properties (Gilbe et al., 2008). Rice husk is an

10

agricultural by-product with low bulk density and high ash content. They are

11

abundant, with approximately 822 million tonnes produced annually worldwide,

12

about 97% of which is produced in developing countries (Naqvi et al., 2014).

13

Currently most rice husk is underutilized with limited options for recycling

14

(Zhang, et al., 2015). However, it can contribute to the bio-economy significantly

15

if options such as decentralized conversion systems are available for

16

agricultural producers/farmers to utilize rice husk as a renewable resource.

M AN U

TE D

EP

Direct combustion is often utilized for biomass to produce heat and

AC C

17

SC

4

18

power for energy services. However, combustion of biomass with high ash

19

content suffers from several disadvantages when compared to low ash fuels.

20

Firstly, the residual ash content can deposit on internal heat transfer surfaces,

21

leading to the formation of slags and clinkers (Gilbe et al., 2008; Jenkins et al.,

22

1998) which negatively affect heat transfer rates and decrease boiler efficiency.

23

Secondly, the abundant inherent inorganic species in biomass may potentially

3

ACCEPTED MANUSCRIPT lead to significant particulate matter (PM) emissions (Gao and Wu, 2011;

2

Johansson et al., 2003; Johansson et al., 2004). Direct combustion of

3

agricultural by-products with high silica ash content, such as rice husk, can

4

result in the release of fibrous particulate matter, including crystalline silica,

5

which can cause major health concerns (Gilbe et al., 2008).

RI PT

1

Considering the negative impact of ash on the performance of

7

combustion systems; ash components can instead be separated into the solid

8

biochar prior to combustion through the pyrolysis of biomass at mild

9

temperatures.

the

solid

product

of

biomass

M AN U

Biochar,

SC

6

pyrolysis,

has

demonstrated a high potential to offset carbon emissions by long-term carbon

11

sequestration with additional agricultural benefits (Williams and Nugranad,

12

2000). Gas-cleaning systems, such as condensation, are commonly used in

13

centralized pyrolysis operations to separate the bio-oil products from the raw

14

pyrolysis volatiles in order to generate a clean (bio-oil-free) syngas for power

15

generation (Ning et al., 2013). However, separation of the bio-oil substantially

16

reduces the effective energy output, as the production of syngas is limited at the

17

mild pyrolysis temperatures that favour biochar production. Furthermore, the

18

upgrading of bio-oil is currently economically prohibitive, due to issues with bio-

19

oil quality, catalyst instability, and deterioration during storage (Mortensen et al.,

20

2011; Lehto et al., 2014).

AC C

EP

TE D

10

21

This motivated the investigation of the combustion of raw pyrolysis

22

volatiles (bio-oil and syngas mixtures, hereafter called pyrolysis volatiles) for the

23

co-generation of value-added product (biochar) and low-emission bio-energy in

4

ACCEPTED MANUSCRIPT a combined pyrolysis and combustion (pyrolysis-combustion) process. This

2

potential approach eliminates the handling of the corrosive bio-oil product

3

(Mortensen et al., 2011) and utilizes the significant energy content of the liquid

4

product for power generation rather than being separated out for further

5

upgrading/disposal (Lehto et al., 2014). This approach is attractive for

6

decentralized biochar and renewable energy generation in agricultural sectors

7

(Mohammadi et al., 2016) as it minimizes the carbon emissions due to

8

transportation of the low-bulk density agricultural by-product and upgrading

9

processes (Roberts et al., 2010; Bazmi et al., 2015). However, combustion of

10

the pyrolysis volatiles is not entirely “clean”, as bio-oil is a mixture of complex

11

organic compounds, the combustion of which may result in higher emissions

12

when compared to syngas only. To the best of our knowledge, previous work

13

investigating the atmospheric emissions, particularly PM, resulting from

14

combustion of the pyrolysis volatiles during the co-production of biochar and

15

renewable energy is not available in the literature.

TE D

M AN U

SC

RI PT

1

The aims of this study are to investigate the energy-based yield of PM

17

from the combustion of pyrolysis volatiles with different fractions of bio-oil and

18

syngas generated from the pyrolysis of rice husk; and compare the PM

19

emissions from the pyrolysis-combustion process to the direct combustion of

20

other biomass.

21

2. Materials and Methods

22

2.1 Material Characterizations

AC C

EP

16

5

ACCEPTED MANUSCRIPT The rice husk used in this study was provided by Beerbelly Brewing

2

Equipment (Pooraka, South Australia). Once received the rice husk was ground

3

in a rotary mill and then sieved to 420 – 500 µm. It was then dried in an oven at

4

105 °C for a minimum of 15 hours. Biochar samples w ere obtained from the

5

pyrolysis of rice husk according to the procedure described in Section 2.2 and

6

then subjected to proximate and ultimate analysis. Proximate analyses were

7

carried out using a thermogravimetric analyser (TGA) (SETARAM, Labsys™).

8

Approximately 65 mg of sample was used to determine the weight fractions of

9

volatiles, ash, and fixed carbon according to ASTM D7582. The moisture

10

content of the rice husk was determined by the oven-drying method following

11

ASTM D4442. Ultimate analyses of the rice husk and biochar were carried out

12

using a CHNS determinator (PerkinElmer, 2400 Series II CHNS/O). The

13

ultimate analysis for C, H, N, S, and O was carried out following ASTM D5373.

14

The oxygen content was calculated by difference. The higher heating value

15

(HHV) of the rice husk and biochar were calculated using the Boie equation

16

(Boie, 1958).

17

2.2. Pyrolysis, Combustion, and Sampling

SC

M AN U

TE D

EP

AC C

18

RI PT

1

Fig. 1 shows a schematic diagram of the experimental system for the

19

pyrolysis-combustion study. Rice husk, after the pre-processing stage, was

20

loaded into a hopper and continuously fed into the main screw reactor at a

21

feeding rate of 1.3 g/min. The biomass was then transported along the entire

22

length of the screw reactor where it was heated by two sets of electrical heaters

23

and maintained at 400, 500, 600, 700, and 800 °C fo r the pyrolysis reactions. N2

6

ACCEPTED MANUSCRIPT at 0.25 l/min was provided through the hopper to remove oxygen from the

2

system. The solid biochar was collected under gravity in a collection vessel at

3

the end of the screw reactor. The yield of biochar is the recovered yield

4

expressed as a percent by weight of dry feed and was estimated using the

5

mass ratio of feedstock and biochar obtained in the collection vessel. An

6

additional 0.1 l/min of N2 was provided from the collection vessel to avoid

7

stagnation of the combustible volatiles. The raw pyrolysis volatiles were

8

premixed with the filtered air and transported to a burner situated at the bottom

9

of a quartz tube (150 cm height x 4.5 cm i.d.) within a vertical 3-zone tube

10

furnace (Carbolite®, GVC 12/1050). The furnace temperature was maintained at

11

850 °C for the combustion. The piping between the o utlet of the pyrolysis

12

reactor and the combustion region was insulated in order to maintain the

13

temperature above 350 °C and prevent condensation o f the bio-oil. The HHV of

14

the pyrolysis volatiles was estimated using a mass and energy balance based

15

on the yields of biochar and the HHVs of the rice husk and biochar.

16

2.2.1. Bio-Oil Sampling

EP

TE D

M AN U

SC

RI PT

1

The yield of bio-oil was obtained using a condensation train consisting of

18

eight 50 ml test tubes connected in series and immersed in an ice/water mixture

19

(≈ 0 °C). Piping between the pyrolysis heater and the first test tube was heated

20

to around 350 °C to prevent condensation of bio-oil . The difference in weight of

21

the condensation train before and after the experimental run was taken as the

22

mass of bio-oil.

23

2.2.2. Synthesis Gas (Syngas) Sampling

AC C

17

7

ACCEPTED MANUSCRIPT A Teflon gas-sampling bag was used to collect the syngas sample for

2

analysis using Gas Chromatography–Thermal Conductivity Detectors (Agilent,

3

490 Micro GC). The first channel of the gas analyser detects H₂, O₂, N₂, CH₄,

4

and CO with a Molecular Seive 5A column, the second channel detects CO2

5

with a Poraplot U column, and the third channel detects hydrocarbons (butane -

6

n-heptane) with a Silicon 5CB column. The HHV of the syngas was calculated

7

based on the heating values and concentrations of the individual species. The

8

mass of syngas produced was calculated as the difference between the mass of

9

the feedstock used and the combined mass of bio-oil and biochar.

SC

2.2.3. Flue Gas Sampling

M AN U

10

RI PT

1

Two samplings ports were positioned on the stainless steel adaptor

12

between the air dilution tunnel (ADT) and the top of the quartz pipe situated

13

above the vertical tube furnace. The CO2 analysis of the flue gas was carried

14

out using a portable non-dispersive infrared sensor (CO2meter.com, CM-0017)

15

with a resolution of 0.5 vol%. A portable CO/O2 analyser (Bacharach, Fyrite®

16

INSIGHT® Plus) was used to analyse the CO level in the flue gas with a

17

resolution of 20 ppmv and O2 level with a resolution of 0.3%.

18

2.2.4. Dilution and PM Sampling

EP

AC C

19

TE D

11

A cascade impactor (CI) (Copley Scientific, 8-stage Anderson cascade

20

impactor) was used to collect the PM and generate mass-size distributions. The

21

size range of the CI was 0.1 - 10 µm. PM10 and PM2.1 refer to particulates in the

22

size fractions 0.1 – 10 µm and 0.1 – 2.1 µm that presented in the flue gas,

23

respectively. The CI was situated immediately downstream of an ADT. The ADT

8

ACCEPTED MANUSCRIPT was used to supply a mixture of air and flue gas at a dilution ratio of 8 with a

2

fixed volumetric flowrate of 28.3 l/min to the CI. As all of the flue gas was

3

sampled, isokinetic sampling was not required (Zhang and Morawska, 2002).

4

Quartz fiber filter papers were used for all stages except the back-up stage,

5

which used a PTFE filter with a 0.1 µm pore size. The filters were prepared for

6

the collection of PM following an adapted methodology outlined in the USEPA

7

Method 5 (USEPA, 2000) and the State of California Air Resources Board:

8

Method 501 (CARB, 1990). The filter papers were dried and weighed before

9

and after the experiments. The average results from a set of three separate 18

10

min runs were obtained. The mass-based yields of PM10 and PM2.1 were divided

11

by the HHV of the pyrolysis volatiles and presented as the energy-based yields

12

in this study.

13

3. Results and Discussions

14

3.1. Feedstock and Biochar Characteristics

TE D

M AN U

SC

RI PT

1

The proximate analysis of the rice husk in Table 1 shows that they have

16

a high amount of ash (21.5%). The ash content of rice husk is significantly

17

greater than other types of biomass. Typical ash contents for other agricultural

18

by-products vary between 0.3 – 8.4% (Alper et al., 2015). This high ash content

19

clearly suggests that fouling and slagging could potentially occur in the direct

20

combustion of rice husk. In addition, it is well known that silica, the main

21

constituent of rice husk ash, is very abrasive and can react with alkalis to form

22

alkali silicates that melt or soften at low temperatures, promoting the formation

23

of slag and clinker (Gilbe et al., 2008; Jenkins et al., 1998). The ultimate

AC C

EP

15

9

ACCEPTED MANUSCRIPT analysis indicated that rice husk had a relatively low carbon content (38.1%)

2

when compared to other biomass types (Claoston et al., 2014). Biochars

3

produced from other forms of biomass typically have a carbon content between

4

60 – 95% (Antal Jr and Gronli, 2003; Claoston et al., 2014). The reduced

5

carbon content of the rice husk derived biochar (42 – 43%) compared to that

6

derived from other typical agricultural by-products is a direct result of its high

7

ash content. The proximate analysis results show that around 96 – 99% of

8

feedstock ash was retained in the biochar (Table 1). In most cases, the carbon

9

content of biochar increases with pyrolysis temperature (Alper et al., 2015).

10

However, this is often not the case for biochar produced from rice husk due to

11

the significant ash content which is not liberated during pyrolysis (Claoston et

12

al., 2014; Jindo et al., 2014). This agrees with (Enders et al., 2012) who found

13

that greater pyrolysis temperatures for low ash biochars increased fixed carbon,

14

but decreased for biochars with more than 20% ash. In addition, the hydrogen

15

and nitrogen contents were lower than other agricultural by-products, while the

16

sulphur content was relatively high (Antal Jr and Gronli, 2003). S content in the

17

char was reduced due to its volatilisation at higher temperatures.

18

3.2. Pyrolysis Products

SC

M AN U

TE D

EP

AC C

19

RI PT

1

Fig. 2a shows the effect of pyrolysis temperature on the yields of syngas,

20

bio-oil, and biochar. It was found that the yield of syngas increased with

21

pyrolysis temperature. The opposite trend was observed for the yields of bio-oil

22

and biochar, which decreased with temperature. This is consistent with other

23

studies of biomass pyrolysis (Williams and Nugranad, 2000; Antal Jr and Gronli,

10

ACCEPTED MANUSCRIPT 2003; Tsai et al., 2007). The maximum yield of bio-oil is typically obtained at a

2

low temperature, above which higher temperatures promote gas production.

3

The reason for this is that higher temperatures promote further cracking of the

4

condensable hydrocarbons which includes aromatic compounds, along with

5

other oxygen-containing hydrocarbons and complex polycyclic aromatic

6

hydrocarbons (PAHs) (Naqvi et al., 2014), leading to a greater production rate

7

and yield of gas (Tsai et al., 2007). The increasing gas yield reduced the mass

8

of the remaining char and therefore decreasing yields of biochar with pyrolysis

9

temperature was observed. Fig. 2b shows the effect of pyrolysis temperature on

10

the HHV of the pyrolysis volatiles and its contributions from syngas and bio-oil.

11

It was found that the energy content of the pyrolysis volatiles increased slightly

12

by about 7% from 400 °C (14.2 MJ/kg) to 700 °C (15. 2 MJ/kg). However, the

13

contribution of bio-oil to the overall HHV of the pyrolysis volatiles reduced

14

substantially from 90% to 44% at the pyrolysis temperatures of 400 and 800 °C,

15

respectively.

TE D

M AN U

SC

RI PT

1

Fig. 3 shows that the major constituents of the syngas at different

17

pyrolysis temperatures are CO, H2, CO2, and CH4. The concentrations of each

18

species at each pyrolysis temperature were normalized to 50% N2 for

19

comparison. It was found that the concentrations of CO and H2 in the syngas

20

increased significantly with pyrolysis temperature, while CO2 showed the

21

opposite trend. A slight increasing trend was observed for CH4, with its

22

concentration maintained at around 3% before 700 °C and then increasing to

23

around 6% at 800 °C . It has been suggested that the secondary reactions of

24

volatiles at high temperatures generate mostly CO, H2, and CH4 rather than

AC C

EP

16

11

ACCEPTED MANUSCRIPT CO2 (Xu et al., 2009; Luo et al., 2004). Aside from the gas species presented in

2

Fig. 3, trace amounts of ethane and propane were also detected at each

3

pyrolysis temperature. The HHV of the syngas and carrier gas mixture

4

increased from around 2.3 MJ/kg at 400 °C to 5.4 MJ /kg at 800 °C due to the

5

increase in H2 and CH4 concentrations.

6

3.3. PM Emissions during the Combustion of Pyrolysis Volatiles

7

3.3.1 Particle Size Distribution

SC

RI PT

1

Fig. 4a shows the normalized particle mass-size distribution resulting

9

from the combustion of the pyrolysis volatiles at 850 °C. In this study, a

10

relatively low combustion temperature of 850 °C was used as this is a typical

11

temperature found in small-scale combustion systems (Obernberger, 1998;

12

Miles et al., 1996). For each of the combustion experiments, the CO2 and CO

13

levels of the flue gas at 5% O2 were maintained at around 13% and 400 ppmv,

14

respectively. It was found that the coarse mode (≈ 9 - 10 µm) was the

15

predominant size range of PM generated from the combustion of pyrolysis

16

volatiles. In addition, no distinctive sub-micron peak was observed, suggesting

17

minimal ash vaporization. This result was expected as the combustion

18

temperature of 850 °C was not sufficiently high to vaporize the majority of the

19

ash. Furthermore, the majority of the ash (>96%) was retained in the biochar

20

prior to combustion. This result closely resembles the PM mass-size

21

distributions that can be found in fuel-oil combustion (Linak et al., 2000; Umbria

22

et al., 2004).

23

3.3.2 Energy-Based Yields of PM10 and PM2.1

AC C

EP

TE D

M AN U

8

12

ACCEPTED MANUSCRIPT 1

Fig. 4b shows the energy-based yield of PM10 and PM2.1 during the

2

combustion of the pyrolysis volatiles at 850 °C. Yi elds of PM10 and PM2.1 from

3

the combustion of the pyrolysis volatiles were estimated to be between 2 – 5

4

and 0.3 – 2 mg/gpyrolysis

5

yields of 154 – 370 mg/MJ for PM10 and 21 – 118 mg/MJ for PM2.1. The yields of

6

both PM10 and PM2.1 reduced substantially with the increasing pyrolysis

7

temperature. Between 400 °C and 800 °C there is a 5 8% decrease in the yield

8

of PM10 and 82% decrease in PM2.1. This demonstrated that the greatest PM

9

emissions occur when the pyrolysis-combustion system is optimized for biochar

10

production. In contrast, when the system is optimized for energy generation (i.e.

11

higher pyrolysis temperature), lower PM yields were observed.

respectively. This corresponds to energy-based

M AN U

SC

RI PT

volatiles

Despite similar combustion conditions being used, there is a strong linear

13

correlation between the energy-based yields of PM and the bio-oil fraction of the

14

pyrolysis volatiles (Fig. 5a). It is suggested that the increasing number of carbon

15

atoms in the fuel molecules would increase the unburned carbon contents

16

(soot) in premixed flames under similar combustion conditions (Calcote and

17

Manos, 1983; Olson et al., 1985). Therefore, increasing the bio-oil fraction has a

18

similar effect because bio-oil produced from biomass pyrolysis consists of

19

complex

20

anhydrosugars, and hydrocarbons (Naqvi et al., 2014; Alper et al., 2015). These

21

compounds contain much longer hydrocarbon chain lengths than the syngas

22

constituents (CO, CH4, H2). In addition, increasing aromatic contents in the fuel

23

strongly increases the tendency to soot due to greater ability to resist oxidation

24

and survive into the burned gas zone (Calcote and Manos, 1983). Bio-oil

AC C

EP

TE D

12

compounds

including

organic

acids,

carbonyls,

phenols,

13

ACCEPTED MANUSCRIPT possesses significant aromatic characteristics (Williams and Nugranad, 2000),

2

so greater bio-oil fractions leads to greater aromatic contents in the pyrolysis

3

volatiles. As a result, increased yields of PM from pyrolysis volatiles combustion

4

were observed with the increased bio-oil fraction.

5

3.3.3 PM10/PM2.1

RI PT

1

Even though the total yield of PM10 and PM2.1 increased with the bio-oil

7

fraction in the pyrolysis volatiles, a decrease in PM10/PM2.1 was observed (Fig.

8

5b). This suggested that higher bio-oil fractions in the pyrolysis volatiles

9

favoured PM2.1 formation at a greater rate than PM10 during combustion. This is

10

counterintuitive; as it would be expected that higher syngas fractions in the

11

pyrolysis volatiles would favour sub-micron particulate formation due to more

12

complete carbon burnout. However, both the sub- and super-micron size range

13

of particulates were included in PM2.1 (0.1 – 1 µm and 1 – 2.1 µm) and their

14

formation mechanisms were different (Linak et al., 2000). Therefore, it is

15

appropriate to examine not only the total PM2.1 mass, but also the contribution

16

from the sub and super-micron fractions in PM2.1. Table 2 summarises the

17

weight contribution of the smallest and largest particulate size fractions to the

18

overall yield of PM2.1 at different bio-oil fractions in the pyrolysis volatiles. It can

19

be seen that increased bio-oil fractions in the pyrolysis volatiles increase the

20

contribution of the largest particulates, while increased syngas fractions

21

increase the contribution of the smallest particulates. This suggests that

22

although PM10/PM2.1 decreased with the increasing bio-oil fraction, the

23

proportion of PM2.1 mass made up of super-micron particulates (1– 2.1 µm)

AC C

EP

TE D

M AN U

SC

6

14

ACCEPTED MANUSCRIPT increased. This agrees with the conclusion of (Linak et al., 2000), who

2

suggested that total super-micron PM fractions increase when incomplete

3

carbon burnout is the dominating PM formation mechanism, while sub-micron

4

PM fractions increase when ash vaporization and complete carbon burnout

5

dominates.

6

3.4. Comparison of PM Emissions of Pyrolysis-Combustion with Direct

7

Combustion

SC

RI PT

1

A meta-analysis was carried out in order to compare the energy-based

9

yield of PM10 when utilizing rice husk in the pyrolysis-combustion process with

10

other biomass types in small-scale combustion devices. Fig. 6 indicates that

11

lower PM emissions for a given energy output could be achieved when

12

compared to other small-scale direct combustion devices utilizing high ash

13

content biomass. Fig. 6 also indicated that, although the energy-based yield can

14

vary significantly between different combustion systems, the results from each

15

study showed a trend of increasing energy-based yield of PM with the ash

16

content of the biomass feedstock. This agrees with (Johansson et al., 2003;

17

Yani et al., 2014; Khalil et al., 2013), who all arrived at a similar conclusion

18

while investigating the relationship between fuel ash content and PM emissions

19

from raw biomass combustion. The substantial PM emission reduction during

20

pyrolysis-combustion is an important finding. It demonstrates a potential low

21

emission technique that enables the effective utilization of abundant high ash

22

biomass resources, such as rice husk, for simultaneous clean energy and

23

biochar generation.

AC C

EP

TE D

M AN U

8

15

ACCEPTED MANUSCRIPT 1

4. Conclusions An experimental comparison of the PM emissions from the combustion of

3

raw pyrolysis volatiles at 850 °C derived from the pyrolysis of rice husk at

4

different temperatures (400 - 800 °C) was carried o ut. It was found that

5

combustion of the raw pyrolysis volatiles generated at 400 °C with the highest

6

bio-oil fraction had substantially increased energy-based yields of PM10 and

7

PM2.1 by 58% and 82% respectively when compared to combustion of the raw

8

pyrolysis volatiles with the lowest bio-oil fraction (800 °C). This implied that the

9

PM emissions were higher if the pyrolysis-combustion process was optimized

10

for biochar production at lower pyrolysis temperatures. Despite a high ash

11

content feedstock (21.5%) being used, the energy-based yields of PM10 of 154

12

– 370 mg/MJ were comparable to or lower than those from the small-scale

13

direct combustion of low ash biomass.

14

Acknowledgements

TE D

M AN U

SC

RI PT

2

This funding of this research is supported by the Government of South

16

Australia through the Premier’s Research and Industry Fund - Catalyst

17

Research Grants (CGR72).

18

References

19

[1] – Allen, B., Nanni, S., Schweitzer, J-P., Baldock, D., Watkins, E., Withana,

20

S., Bowyer, C., 2015. International Review of Bio-economy Strategies with a

21

Focus on Waste Resources. Report prepared for the UK Government

22

Department for Business, Innovation and Skills. Institute for European

23

Environmental Policy, London.

AC C

EP

15

16

ACCEPTED MANUSCRIPT [2] - Alper, K., Tekin, K., Karagöz, S., 2015. Pyrolysis of Agricultural Residues

2

for Bio-Oil Production. Clean Technol. Envir. 17 (1), 211-223.

3

[3] - Antal Jr, M.J., Gronli, M., 2003. The Art, Science, and Technology of

4

Charcoal Production. Ind. Eng. Chem. Res. 42 (8), 1619-1640.

5

[4] - Bazmi, A.A., Zahedi, G., Hashim, H., 2015. Design of Decentralized

6

Biopower Generation and Distribution System for Developing Countries. J.

7

Clean Prod. 86, 209-220.

8

[5] - Boie, W., 1953. Fuel Technology Calculations. Energietechnik. 3, 309-316.

9

[6] - CARB, Method 50. 1990. Determination of Size Distribution of Particulate

M AN U

SC

RI PT

1

Matter from Stationary Sources. State of California Air Resources Board.

11

[7] - Calcote, H.F., Manos, D.M., 1983. Effect of Molecular Structure on Incipient

12

Soot Formation, Combust. Flame. 49, 289-304.

13

[8] - Claoston, N., Samsuri, A.W., Ahmad Husni, M.H., Mohd Amran M.S., 2014.

14

Effects of Pyrolysis Temperature on the Physicochemical Properties of Empty

15

Fruit Bunch and Rice Husk Biochars. Waste Manag. Res. 32 (4), 331-339.

16

[9] - Enders, A., Hanley, K., Whitman, T., Joseph, S., Lehmann, J., 2012.

17

Characterization of Biochars to Evaluate Recalcitrance and Agronomic

18

Performance. Bioresour. Technol. 114, 644-653.

19

[10] - Gao, X., Wu, H., 2011. Biochar as a Fuel: 4. Emission Behavior and

20

Characteristics of PM1 and PM10 from the Combustion of Pulverized Biochar in

21

a Drop-Tube Furnace. Energ. Fuel. 25 (6), 2702-2710.

22

[11] - Gilbe, C., Öhman, M., Lindström, E., Boström, D., Backman, R.,

23

Samuelsson, R., Burvall, J., 2008. Slagging Characteristics during Residential

24

Combustion of Biomass Pellets. Energ. Fuel. 22 (5), 3536-3543.

AC C

EP

TE D

10

17

ACCEPTED MANUSCRIPT [12] - Jenkins, B.M., Baxter, L.L., Miles Jr, T.R., Miles, T.R., 1998. Combustion

2

Properties of Biomass. Fuel Process. Technol. 54, 17-46.

3

[13] - Jindo, K., Mizumoto, H., Sawada, Y., Sanchez-Monedero, M.A., Sonoki,

4

T., 2014. Physical and Chemical Characterization of Biochars Derived from

5

Different Agricultural Residues. Biogeosciences. 11, 6613-6621.

6

[14] - Johansson, L.S., Tullin, C., Leckner, B., Sjovall, P., 2003. Particle

7

Emissions from Biomass Combustion in Small Combustors. Biomass Bioenerg.

8

25, 435-446.

9

[15] - Johansson, L.S., Leckner, B., Gustavsson, L., Cooper, D., Tullin, C.,

10

Potter, A., 2004. Emission Characteristics of Modern and Old-Type Residential

11

Boilers Fired with Wood Logs and Wood Pellets. Atmos. Environ. 38, 4183-

12

4195.

13

[16] - Khalil, R.A., Bach, Q.V., Skreiberg, Ø., Tran, K.Q., 2013. Performance of

14

a Residential Pellet Combustor Operating on Raw and Torrefied Spruce and

15

Spruce-Derived Residues. Energ. Fuel. 27, 4760−4769.

16

[17] - Lehto, J., Oasmaa, A., Solantausta, Y., Kytö, M., Chiaramonti, D., 2014.

17

Review of Fuel Oil Quality and Combustion of Fast Pyrolysis Bio-Oils from

18

Lignocellulosic Biomass. Appl. Energ. 116, 178-190.

19

[18] - Linak, W.P., Miller, C.A., Wendt, J.O.L., 2000, Fine Particle Emissions

20

from Residual Fuel Oil Combustion: Characterization and Mechanisms of

21

Formation. Proc. Combust. Inst. 28, 2651–2658.

22

[19] - Luo, Z., Wang, S., Liao, Y., Zhou, J., Gu, Y., Cen, K., 2004. Research on

23

Biomass Pyrolysis for Liquid Fuel. Biomass Bioenerg. 26 (5), 455-462.

AC C

EP

TE D

M AN U

SC

RI PT

1

18

ACCEPTED MANUSCRIPT [20] - Miles, T.R., Miles Jr, T.R., Baxter, L.L., Bryers, R.W., Jenkins, B.M.,

2

Oden, L.L., 1996. Boiler Deposits from Firing Biomass Fuels. Biomass

3

Bioenerg. 10 (2–3), 125-138.

4

[21] - Mohammadi, A., Cowie, A., Mai, T.L.A., de la Rosa, R.A., Kristiansen, P.,

5

Brandao, M., Joseph, S., 2016. Biochar Use for Climate-Change Mitigation in

6

Rice Cropping Systems. J. Clean Prod. 116, 61-70.

7

[22] - Mortensen, P.M., Grunwaldt, J.D., Jensen, P.A., Knudsen, K.G., Jensen,

8

A.D., 2011. A Review of Catalytic Upgrading of Bio-oil to Engine Fuels. Appl.

9

Catal. A-Gen. 407, 1-19.

M AN U

SC

RI PT

1

[23] - Naqvi, S.R., Uemura, Y., Yusup, S.B., 2014. Catalytic Pyrolysis of Paddy

11

Husk in a Drop Type Pyrolyzer for Bio-Oil Production: The Role of Temperature

12

and Catalyst. J. Anal. Appl. Pyrol. 106, 57-62.

13

[24] - Ning, S., Hung, M., Chang, Y., Wan, H., Lee, H., Shih, R., 2013. Benefit

14

Assessment of Cost, Energy, and Environment for Biomass Pyrolysis Oil. J.

15

Clean Prod. 59, 141-149.

16

[25] - Nussbaumer, T., Czasch, C., Klippel, N., Johansson, L., Tullin, C., 2008.

17

Particulate Emissions from Biomass Combustion in IEA Countries: Survey on

18

Measurements and Emission Factors. International Energy Agency (IEA)

19

Bioenergy Task 32 and Swiss Federal Office of Energy (SFOE): Zurich.

20

[26] - Obernberger, I., 1998. Decentralized Biomass Combustion: State of the

21

Art and Future Development. Biomass Bioenerg. 14 (1), 33-56.

22

[27] - Olson, D.B., Pickens, J.C., Gill, R.J., 1985. The Effects of Molecular

23

Structure on Soot Formation II. Diffusion Flames. Combust. Flame. 62, 43-60.

AC C

EP

TE D

10

19

ACCEPTED MANUSCRIPT [28] - Roberts, K.G., Gloy, B.A., Joseph, S., Scott, N.R., Lehmann, J., 2010. Life

2

Cycle Assessment of Biochar Systems: Estimating the Energetic, Economic,

3

and Climate Change Potential. Environ. Sci. Technol. 44 (2), 827-833.

4

[29] - Schmidl, C., Luisser, M., Padouvas, E., Lasselsberger, L., Rzaca, M.,

5

Cruz, C.R.S., Handler, M., Peng, G., Bauer, H., 2011. Particulate and Gaseous

6

Emissions from Manually and Automatically Fired Small Scale Combustion

7

Systems. Atmos. Environ. 45, 7443-7454.

8

[30] - Shen, G., Tao, S., Wei, S., Zhang, Y., Wang, R., Wang, B., Li, W., Shen,

9

H., Huang, Y., Chen, Y., Chen, H., Yang, Y., Wang, W., Wei, W., Wang, X., Liu,

10

W., Wang, X., Masse Simonich, S.L., 2012. Reductions in Emissions of

11

Carbonaceous Particulate Matter and Polycyclic Aromatic Hydrocarbons from

12

Combustion of Biomass Pellets in Comparison with Raw Fuel Burning. Environ.

13

Sci. Technol. 46 (11), 6409-6416.

14

[31] - Smith, K.R., Uma, R., Kishore, V.V.N., Lata, K., Joshi, V., Zhang, J.,

15

Rasmussen, R.A., Khalil, M.A.K., 2000. Greenhouse Gases from Small-Scale

16

Combustion Devices in Developing Countries Phase IIa: Household Stoves in

17

India; EPA-600/R-00-052. U.S. Environmental Protection Agency, Research

18

Triangle Park, NC, 27711.

19

[32] - Tsai, W.T., Lee, M.K., Chang, Y.M., 2007. Fast Pyrolysis of Rice Husk:

20

Product Yields and Compositions. Bioresour. Technol. 98 (1), 22-28.

21

[33] - Umbria, A., Galan, M., Munoz, M.J., Martin, R., 2004. Characterisation of

22

Atmospheric Particles: Analysis of Particles in the Campo de Gibraltar.

23

Atmosfera. 17 (4), 191-206.

AC C

EP

TE D

M AN U

SC

RI PT

1

20

ACCEPTED MANUSCRIPT [34] - USEPA, Method 5. 2000. Determination of Particulate Matter Emissions

2

from Stationary Sources. US Environmental Protection Agency, Washington

3

DC.

4

[35] - Williams, P.T., Nugranad, N., 2000. Comparison of Products from the

5

Pyrolysis and Catalytic Pyrolysis of Rice Husks. Energy. 25 (6), 493-513.

6

[36] - Xu, R., Ferrante, L., Briens, C., Berruti, F., 2009. Flash Pyrolysis of Grape

7

Residues into Biofuel in a Bubbling Fluid Bed. J. Anal. Appl. Pyrol. 86 (1), 58-

8

65.

9

[37] - Yani, S., Gao, X., Wu, H., 2015. Emission of Inorganic PM10 from the

10

Combustion of Torrefied Biomass under Pulverized-Fuel Conditions. Energ.

11

Fuel. 29, 800−807.

12

[38] - Zhang, H., Ding, X., Chen, X., Ma, Y., Wang, Z., Zhao, X., 2015. A New

13

Method of Utilizing Rice Husk: Consecutively Preparing D-xylose, Organosolv

14

Lignin, Ethanol and Amorphous Superfine Silica. J. Hazard Mater. 291, 65-73.

15

[39] - Zhang, J., Smith, K.R., Ma, Y., Ye, S., Jiang, F., Qi, W., Liu, P., Khalil,

16

M.A.K., Rasmussen, R.A., Thornelo, S.A., 2000. Greenhouse Gases and Other

17

Airborne Pollutants from Household Stoves in China: A Database for Emission

18

Factors. Atmos. Environ. 34, 4537-4549.

19

[40] - Zhang, J., Morawska, L., 2002. Combustion Sources of Particles: 2.

20

Emission Factors and Measurement Methods. Chemosphere. 49 (9), 1059-

21

1074.

AC C

EP

TE D

M AN U

SC

RI PT

1

22

21

ACCEPTED MANUSCRIPT Figure Captions:

2

Fig. 1. Schematic diagram of the pyrolysis-combustion and sampling systems

3

(components : 1. hopper; 2. motor; 3. feeding screw; 4. nitrogen cylinder; 5.

4

main screw reactor; 6. electric heaters; 7. char collection vessel; 8. in-line HEPA

5

filter for combustion air; 9. burner; 10; bio-oil collection vessel; 11. ice/water

6

mixture;

7

chromatography/thermal conductivity detector (GC/TCD) ; 15. combustion

8

furnace; 16. bio-oil/particulate filter; 17. CO2 analyser; 18. CO/O2 analyser; 19.

9

HEPA filter; 20. air dilution tunnel; 21. cascade impactor; 22. vacuum pump; T.

gas

collection

bag;

13.

coalescing

filter;

14.

gas

M AN U

SC

12.

RI PT

1

temperature control; F. flowmeter.

11

Fig. 2. Relationship between the pyrolysis temperature and a) the yield of

12

pyrolysis products, b) total raw pyrolysis volatiles HHV.

13

Fig. 3. Composition of the non-condensable syngas fraction of the raw pyrolysis

14

volatiles between 400 and 800 °C pyrolysis temperat ures (corrected to 50 vol %

15

nitrogen).

16

Fig. 4. a) Normalized mass-based particle-size distribution of PM from

17

combustion of the raw pyrolysis volatiles and b) energy-based yields of PM10

18

and PM2.1 from combustion of the raw pyrolysis volatiles. Raw pyrolysis volatiles

19

were produced at pyrolysis temperatures ranging from 400 to 800 °C and

20

combusted at 850 °C.

21

Fig. 5. a) Relationship between the contribution of bio-oil the overall pyrolysis

22

volatiles HHV and the PM10 and PM2.1 emissions b) relationship between

AC C

EP

TE D

10

22

ACCEPTED MANUSCRIPT PM10/PM2.1 ratio of collected PM in the cascade impactor and the contribution of

2

bio-oil the overall pyrolysis volatiles HHV.

3

Fig. 6. Relationship between ash content of feedstock and energy-based yield

4

of PM10 (mg/MJ). The energy-based yields of PM10 for this process optimized

5

for either biochar yield or energy generation are shown.

6

* Refers to total suspended particles (TSP).

SC

7

M AN U

8 9 10 11

16

17

18

EP

15

AC C

14

TE D

12 13

RI PT

1

19

20

21

23

ACCEPTED MANUSCRIPT 1

Table 1. Proximate and ultimate analysis of rice husk and biochar. Rice Husk

Yield (%) Ash (%)

8 9 10 11

800

9.2 51.3 18.4

-

-

-

-

-

38.1 5.00

43.0 2.21

43.5 1.92

43.4 1.58

41.6 1.19

41.7 1.19

0.26 0.52 56.1 15.4 21.5

0.47 0.41 53.9 17.0 44.5 47.3

0.30 0.09 54.6 17.0 40.0 53.8

0.35 0.01 56.9 15.7 39.1 54.0

0.37 0.00 56.7 15.9 37.6 55.7

RI PT

700

0.30 0.13 54.2 17.1 41.1 50.3

TE D

7

600

EP

6

(d.a.f = dry ash free) # Calculated by difference.

500

AC C

2 3 4 5

400

SC

N S O# HHV (MJ/kg)

-

M AN U

Pyrolysis Temperature (°C) Proximate Analysis (wt %) Moisture Volatiles Fixed Carbon Ultimate Analysis (wt % d.a.f) C H

Biochar

12 13 14

24

ACCEPTED MANUSCRIPT 1

Table 2. Contribution of smallest and largest PM size fractions to total PM2.1

2

yield with varying contributions of bio-oil to pyrolysis volatiles HHV. Bio-oil

Wt. % of PM2.1 Yield

Temperature

Contribution to

Contributed by Different Particle Sizes

(°C)

Pyrolysis Volatiles HHV (%)

6 7 8 9 10

28.4

90.3

40.8

500

74.0

48.6

600

63.9

700

53.0

800

43.6

M AN U

400

24.5

50.3

25.8

52.7

21.9

52.0

21.8

TE D

5

(Super-micron)

EP

4

(Sub-micron)

AC C

3

1.1 – 2.1 µm

SC

0.1 – 0.43 µm

RI PT

Pyrolysis

11 12 13 14

25

ACCEPTED MANUSCRIPT

5 6 7 8

Fig. 1.

AC C

4

EP

2 3

TE D

M AN U

SC

RI PT

1

9 10 11 12

26

ACCEPTED MANUSCRIPT 100 a) 90 80

RI PT

Syngas

60 50

Bio-Oil

40

Biochar

SC

30 20

M AN U

10 0 400

500 600 700 Pyrolysis Temperature (°C)

1 100

70

12 10

EP

60

40

b) 14

TE D

80

50

16

8

AC C

Contibution to Pyrolysis Volatiles HHV (%)

90

800

6

30

4

Syngas

2

Bio-Oil

0

Pyrolysis Volatiles HHV

20 10

0 400 2 3

Pyrolysis Volatiles HHV (MJ/kg)

Yield (%)

70

500 600 700 Pyrolysis Temperature (°C)

800

Fig. 2.

27

6

25

5

20

4

15

3

0 400

7

8

0 800

TE D

Fig. 3.

700

EP

6

600 Pyrolysis Temperature (˚C)

1

AC C

5

500

M AN U

5

4

2

SC

10

1 2 3

Syngas HHV (MJ/kg)

30

RI PT

Vol %

ACCEPTED MANUSCRIPT

9

10

28

ACCEPTED MANUSCRIPT a) 1200 400°C Pyrolysis Volatiles 500°C Pyrolysis Volatiles 1000 600°C Pyrolysis Volatiles

RI PT

800°C Pyrolysis Volatiles 600

SC

dM/dlogDp (mg/MJ)

700°C Pyrolysis Volatiles 800

400

M AN U

200

0 0.1 1

1 Aerodynamic Diameter (µm)

300 250 200

EP

350

b)

400°C Pyrolysis Volatiles 500°C Pyrolysis Volatiles 600°C Pyrolysis Volatiles 700°C Pyrolysis Volatiles 800°C Pyrolysis Volatiles

AC C

Energy-Based Yield of PM (mg/MJ)

400

TE D

450

10

150 100 50 0

2 3

PM10

PM2.1

Fig. 4.

29

ACCEPTED MANUSCRIPT 450 a)

350

RI PT

300 PM10

250

PM2.1

200 150

SC

Energy-Based Yield of PM (mg/MJ)

400

100

M AN U

50 0 40

50

60 70 80 Contribution of Bio-Oil to HHV (%)

1 8

6

b)

5

EP

PM10/PM2.1

100

TE D

7

90

4

AC C

3 2 1 0 40 2 3

50

60 70 80 Contribution of Bio-Oil to HHV (%)

90

100

Fig. 5.

4

30

ACCEPTED MANUSCRIPT

3500 3000

RI PT

2500 2000 1500 1000

SC

Energy-Based Yield of PM10 (mg/MJ)

4000

500 0

5

6

Fig. 6.

TE D

4

Biochar - Drop-Tube Furnace (Gao and Wu, 2011) Dung Cake - CookStove (Smith et al., 2000)* Rice Straw - Household Stove (Smith et al., 2000)* Root Fuel - Household Stove (Smith et al., 2000)* Corn Straw - Pellet Burner (Shen et al., 2012) Triticale Pellets - Biomass Boiler (Schmidl et al., 2011) Oak - Logwood Stove (Schmidl et al., 2011) Wood - Open Fireplace (Nussbaumer et al., 2008) Maize Residue - Brick Stove (Zhang, Y. et al., 2000)* Fuel Wood - Brick Stove (Zhang, Y. et al., 2000)* Rice husks - Pyrolysis-Combustion Optimized for Energy (This Study)

40

50

Wood - Cookstove (Smith et al., 2000)* Kerosene - Cookstove (Smith et al., 2000)* Acacia - Household Stove (Smith et al., 2000)* Char Briquette - Household Stove (Smith et al., 2000)* Pine Wood - Pellet Burner (Shen et al., 2012) Beech - Logwood Stove (Schmidl et al., 2011) Wood - Wood Stove (Nussbaumer et al., 2008) Brush Wood - Brick Stove (Zhang, Y. et al., 2000)* Wheat Residue - Brick Stove (Zhang, Y. et al., 2000)* Rice husks - Pyrolysis-Combustion Optimized for Biochar (This Study)

EP

3

20 30 Ash Content of Feedstock (%)

AC C

1 2

10

M AN U

0

31