Bioresource Technology 188 (2015) 265–272

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: [Link]/locate/biortech

Perennial grass (Arundo donax L.) as a feedstock for thermo-chemical

conversion to energy and materials
Ruprekha Saikia a, Rahul Singh Chutia a, Rupam Kataki a,⇑, Kamal K. Pant b
a
Department of Energy, Tezpur University, Tezpur 784028, Assam, India
b
Department of Chemical Engineering, Indian Institute of Technology Delhi, New Delhi 110016, India

h i g h l i g h t s g r a p h i c a l a b s t r a c t

 Reported a new feedstock (A. donax)

for alternate energy generation.
 Investigated pyrolysis of A. donax
Cut into
Ground
pieces

grass at 350–650 °C.

Fed to reactor
Dried Biomass

 Bio-oil showed promising fuel Arundo donax

properties.
 Perennial grass as a potent candidate
for futuristic energy utilities.
Aqueous & organic
fractions

Bio-oil

Biochar

a r t i c l e i n f o a b s t r a c t

Article history: In the present study, perennial grass species Arundo donax L. was pyrolysed in a fixed-bed reactor and
Received 30 November 2014 characterization was performed for the liquid and the solid products. The effect of process parameters
Received in revised form 20 January 2015 such as temperature (350–650 °C), heating rate (10 °C and 40 °C min1) and sweeping gas flow rate
Accepted 22 January 2015
(50–250 ml min1) was also investigated. Maximum bio-oil yield of 26% was observed at 500 °C for
Available online 31 January 2015
the heating rate of 40 °C min1. Chemical composition of the bio-oil was analysed through NMR, FTIR
and GC–MS. The biochar was characterized by Scanning Electron Microscopy (SEM), Fourier Transform
Keywords:
Infrared (FTIR) spectroscopy along with elemental analysis (CHN). The biochar produced as a co-product
Arundo donax L.
Perennial grass
of A. donax pyrolysis can be a potential soil amendment with multiple benefits including increased soil
Pyrolysis fertility and C-sequestration. Current investigation suggests suitability of A. donax as a potential feedstock
Bio-oil for exploitation of energy and biomaterials through pyrolytic route.
Biochar Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction of carbon from which value added chemicals, materials and fuels
can be produced using appropriate conversion pathways, is consid-
In recent years renewable energy has received increased impor- ered to be one of the most attractive and promising renewable
tance due to deminuting availability of the conventional fossil energy resource. In contrast to other renewable sources, which
based resources, regular hike in prices of fossil based fuels and give heat and power, biomass is the only source, which can be con-
increased emissions of anthropogenic greenhouse gases to the verted to liquid, solid and gaseous fuel by applying two main con-
atmosphere. In this regard, biomass, the only renewable resource version pathways viz., (i) biochemical and (ii) thermochemical
(Chutia et al., 2013). The thermochemical conversion process uses
the entire content of biomass unlike biochemical and biological
⇑ Corresponding author. Tel.: +91 9435380921; fax: +91 3712 267005/6. processes where individual conversion technologies favour specific
E-mail address: rupam@[Link] (R. Kataki). biomass types and pretreatment requirement before processing

[Link]
0960-8524/Ó 2015 Elsevier Ltd. All rights reserved.
266 R. Saikia et al. / Bioresource Technology 188 (2015) 265–272

(Mosier et al., 2005). Among the thermochemical conversion A. donax is a herbaceous, fast growing and low input-high yield-
technologies, pyrolysis is of great importance because it provides ing grass. It is well suited to subtropical and warm temperate
a liquid fuel that can substitute for fuel oil in any static heating regions and found in the Mediterranean basin for thousands of
or electricity generation application. The liquid fuel obtained from years. Different historical and archeological evidence indicate its
pyrolysis can be readily stored and it is beneficial when biomass origin as Asia, South Europe, North Africa, the Middle East, North
resources are remote from the location where the energy is and South America and also in Australia. It is a tall, erect, perennial
required. Pyrolysis process is the only renewable energy conver- C3 grass which may grow up to 9 m. It is often classified as an
sion technique which can convert various types of biomass into emergent aquatic plant due to its rapid growing in water
solid, liquid and gaseous fuels depending upon process parameters (Christou, 2013). It also grows very rapidly in marshy areas, peat,
namely heating rate, final temperature, vapour residence time, bog etc. It is not a native plant of India but during last few decades,
sweeping gas flow rate, pyrolysis environment etc. (Bridgwater it has spread rapidly and covered most of the wetlands in the
et al., 1999). North-Eastern region of India. The grass causes artificial flood in
However, availability of low cost and good quality non-edible the rainy season. During flood the stems and rhizomes of this grass
feedstock is a pre-requisite to large scale commercial use of any deposit in drainage system, small agricultural ditches and under
biofuel production process (Chutia et al., 2013). It is evident from bridges resulting in blocking of waterways, structural damage
literature that pyrolysis oils or bio-oils have been identified as and hazards to bridges and other structures (CABI datasheet).
inexpensive renewable liquid fuel that can be produced from ligno- The present work aims at utilization of this perennial grass through
cellulosic biomass. Fast pyrolysis of biomass has been shown to be the process of pyrolysis and study the product yields (biochar, bio-
two to three times cheaper than biomass conversion technologies oil & gas) at different temperature, varying heating rate and sweep-
based on gasification and fermentation (Vispute et al., 2010). ing gas flow rate to know the condition of optimum bio-oil yield.
Wright et al. (2010) and NREL (USA), 2013 estimated cost of bio-
mass fast pyrolysis oil at around $0.82 per litre. Another study by
Anex et al. (2010), estimated product value in the range of 2. Methods
$2.00–5.50 per gallon ($0.53–1.45 per litre) gasoline equivalent
for a standalone biomass-to-liquid fuel plants (pyrolysis, gasifica- 2.1. Raw material
tion and biochemical) to produce fuels, with pyrolysis the lowest
and biochemical the highest. In India, a number of perennial A. donax grass (Supplementary Fig. S1) collected from the forest,
grasses are available throughout the year, which often grows in on the bank of the Brahmaputra river was washed and sun dried
wild condition. These grasses can grow in a wide range of climatic and then was cut into small pieces (0.5–1.5 cm). The pieces were
conditions and are often characterised as low input-high yielding ground with a Willy mill (SECOR Scientific Eng. Co.) and then were
biomass. These grasses may be a potential source of bioenergy allowed to pass through 0.2 mm (70 meshes) (as per TAPPI T257
feedstock without adversely affecting the soil C stocks (Sarkhot Om-85 methods). Samples were then oven dried and kept in a des-
et al., 2012). Use of different kinds of grasses for pyrolysis process iccator for further analysis and pyrolysis experiments.
have already been reported in various literatures. Boateng et al.
(2006) investigated the conversion of switchgrass into gaseous
and liquid fuel at 600–1050 °C temperature range. An increase in 2.2. Characterization
condensable gas and decrease in non-condensable gas yield was
observed at temperature <750 °C. Boateng et al. (2007) performed 2.2.1. Determination of gross calorific value (GCV)
pyrolysis of bermudagrass for pyrolytic oil production and its With the help of a mechanical pelletizer, 1.0 g of the finely
potential use as biofuel. The experiments were carried out at tem- ground A. donax was pressed into a tablet. After that, gross calorific
peratures 500, 700 and 900 °C and the condensable product yield value (GCV) of the sample was determined by using an automatic
was observed in the range 72–83 wt.%. The char yield was observed adiabatic bomb calorimeter (Changsha Kaiyuan Instruments Co.,
in the range 5.5–16 wt.%. Strezov et al. (2008) investigated the 5E-1AC/ML) according to ASTM D 240. The sample was oxidized
thermal conversion of elephant grass to bio-gas, bio-oil and char- by combustion in an adiabatic bomb containing oxygen at
coal under two heating rates of 10 and 50 °C min1. It was 3.4 MPa pressure. Same procedure was also applied for determina-
observed that faster heating rate increased the amount of liquid tion of GCV of biochar. The assay was carried out in triplicates and
products by 10% but charcoal yields remained almost the same at the mean value was reported.
30%. Lee et al. (2010) studied the yield and chemical characteriza-
tion of pyrolytic products from the induction heating pyrolysis of
2.2.2. CHN analysis
napier grass. Promdee and Vitidsant (2012) reported the pyrolysis
The elements C, H and N of the sample as well as the biochars
of mixed grass (Cogongrass and Manilagrass) with the temperature
were analysed using a CHN analyser (Perkin Elmer, 2400 Series-
range of 400–500 °C with the feeding rate of 150, 350 and 550 rpm.
II). The percentage of oxygen was determined by means of
Greenhalf et al. (2013) characterised and compared the fast pyro-
difference.
lysis product yield for switchgrass and miscanthus using a fluid-
ized bed reactor. The maximum bio-oil yield for switchgrass and
miscanthus were found to be 60% and 52% respectively at 2.2.3. Determination of pH
500 °C. It was also observed that the bio-oil from switchgrass has pH of A. donax biochars were calculated following the method
the highest potential for upgrading to produce high value chemi- as described elsewhere (Kim et al., 2011) by using a pH metre
cals. Nanda et al. (2014) studied three lignocellulosic biomasses (EUTECH pH 510).
wheat straw, timothy grass and pinewood through slow
(2 °C min1) and high heating rate (450 °C min1) pyrolysis. The
yield of bio-oil were observed to be 40–48 wt.% through high heat- 2.2.4. Determination of electrical conductivity (EC)
ing rate pyrolysis and 18–24 wt.% through slow heating rate pyro- Electrical conductivity of the A. donax biochars were deter-
lysis. In the present study, a perennial grass Arundo donax L. is mined by using a conductivity metre (Digital TDS/Conductivity
investigated for its suitability as a prospective bioenergy feedstock Meter MK 509). The solutions were prepared following the same
for thermo-chemical conversion. method as for pH measurement.
 R. Saikia et al. / Bioresource Technology 188 (2015) 265–272 267

2.2.5. Proximate analysis and calorific values were determined by using ASTM methods fol-
The moisture, volatile matter and ash content of the A. donax lowing Bordoloi et al. (2015).
and its biochars were determined according to ASTM D 3173,
ASTM D 3175 and ASTM D 3174 protocols respectively on dry 2.3. Experimental procedure
basis. The fixed carbon content was calculated by difference.
The pyrolysis experiments were carried out by taking 10 g of the
2.2.6. Composition analysis dry biomass sample in a lab scale quartz fixed-bed reactor with a
Hemicellulose, lignin and cellulose content of A. donax were length of 30 cm and an internal diameter of 3 cm, equipped with
determined indirectly with a Fibertec I and M system (Foss Teca- an inert gas (nitrogen) connection. In order to remove the air inside
tor) as described elsewhere (Chutia et al., 2013). the reactor, nitrogen gas was passed for 30 min at a flow rate of
Hemicellulose ¼ NDF  ADF: 30 ml min1 before the start of the experiments and then flow rate
was set to the desired value. The reactor was heated externally by
Lignin ¼ ADL: an electric furnace, with the temperature being controlled by a
Ni-Cr-Ni thermocouple fitted inside the reactor. The thermocouple
was connected to a PID controller for controlling the temperature
Cellulose ¼ NDF  ðHemicellulose þ LigninÞ:
and heating rate. After the loading of the sample in the reactor, tem-
where NDF = Neutral Detergent Fibre, residue after treating the bio- perature was increased with heating rates of 10 and 40 °C min1
mass with neutral detergent solution (sodium lauryl sulphate and from room temperature to temperature range 350–650 °C and hold
EDTA) which contains (Cellulose + Hemicellulose + Lignin) accord- for 30 min at the desired pyrolysis temperature or until no signifi-
ing to EN ISO 16472:2006 and AOAC 2002:04. cant release of gas was observed. All volatile products from the
ADF = Acid Detergent Fibre, residue after treating the biomass reactor were allowed to condense by the water circulation at a tem-
with acid detergent solution (Cetyltrimethylammonium bromide perature of 8 °C. The condensed liquid product collected in a con-
in H2SO4 solution). tainer at the end of the condenser consisted of two phases; the
ADL = Acid Detergent Lignin, residue obtained by treating ADF aqueous phase and the organic phase (Supplementary Fig. S2).
with 72% H2SO4 according to EN ISO 13906:2008 and AOAC The organic portion was recovered with diethyl ether. The mixture
973.18, for Acid Detergent Fibre (ADF) and Acid Detergent Lignin was dried over anhydrous sodium sulphate, filtered and then evap-
(ADL), respectively. orated in a rotary evaporator at 30 °C. The residual was weighed as
bio-oil. The solid byproduct in the reactor was weighed as char. The
2.2.7. FTIR analysis gas yield was determined by taking difference.
The IR spectrum of the A. donax sample was recorded on a Nico-
let IR spectrometer at room temperature (26 ± 2 °C). A region in the 3. Results and discussion
spectral range of 4000–400 cm1 was used for analysis.
3.1. Raw material
2.2.8. NMR analysis
The bio-oil obtained at the pyrolysis conditions giving the max- A. donax grasses have properties as presented in Supplementary
imum bio-oil yield was selected for NMR characterization. NMR Table S3. It has a moisture content of 8.5% which is suitable for
spectra were recorded in a 400 MHz NMR spectrophotometer thermochemical conversion known from previous literatures. Ash
(JEOL, JNM ECS) by using deuterated chloroform (CDCl3) as the content, volatile matter and fixed carbon content were found to
internal standard and coupling constants were expressed in Hertz. be 5.3%, 74.3% and 11.7% respectively. The elemental content of
carbon, hydrogen, oxygen and nitrogen in A. donax grass were
2.2.9. Scanning Electron Microscopy (SEM) found to be 42.0%, 6.2%, 36.3% and 1.5% respectively on dry ash free
The surface morphology of the feedstock and the biochars were basis. The H/C and O/C ratios of the sample were 1.76 and 0.64
analysed by using Scanning Electronic Microscope and Energy-dis- respectively. Considering only the main elements C, H, N and O,
persive X-ray (EDX) JEOL (JSM-6390 LV) microscope with an accel- the molecular formulae of the sample can be written as
eration voltage of 20 kV. CH1.76N0.03O0.64. The higher heating value of the biomass sample

2.2.10. GC–MS analysis

The bio-oils of A. donax sample were analysed by GC–MS on a Per-
kin Elmer Claurus 600 equipped with TCD detector and DB-5 ms col-
umn (60.0 m  250 lm). The column temperature was held at 70 °C
for 2 min, then heated up to 290 °C where the rate of increase was
10 °C min1 and was held for 7 min at the final temperature. The car-
rier gas was helium maintaining a flow rate of 1 ml min1.

2.2.11. Thermal analysis

The thermal degradation of the A. donax sample was studied in a
TGA analyser (NETZSCHSTA 449F3 STA449F3A-0931-M) up to a
temperature of 1000 °C with heating rate of 10 °C min1. High pur-
ity N2 gas (99.99%) was used as a carrier gas with a flow rate of
100 ml min1. About 4–8 mg of the sample was used for the
experiment.

2.2.12. Analysis of physical properties

The physical properties of the A. donax bio-oil such as specific Fig. 1. Thermogravimetric (TG) and derivative thermogravimetric (DTG) curves at
gravity, density, flash point, fire point, pour point, cloud point 10 °C min1 heating rate.
268 R. Saikia et al. / Bioresource Technology 188 (2015) 265–272

was 17.14 MJ kg1. The cellulose, hemicelluloses and lignin con- (Varol et al., 2007). The FTIR band at 471 cm1 which is below
tent of the biomass sample was observed as 29.20%, 35.9% and 600 appears due to the metal oxygen stretching (Wen et al., 2007).
23.32% respectively. FTIR spectra were used to get useful information on the oxygen-
The thermal behaviour of the feedstock is shown in Fig. 1. It is ated compounds and their various forms. The quality of bio-oil is
evident from previous literature that the decomposition of hemi- dependent on the presence of oxygenated compounds. The above
cellulose occurs in the temperature range of 220–315 °C and that results reveal that the bio-oil contains high amount of oxygenated
of cellulose occurs at 315–400 °C after the initial weight loss due compounds and parts of oxygenated compounds present in the
to moisture evaporation at 30–150 °C (Yang et al., 2007). Lignin forms of alcohol and ether are confirmed by the peaks obtained
undergoes degradation in the wide temperature range of 150– in the range 950–1300 cm1. This was further correlated with
900 °C (Choudhuri et al., 2014). The temperature, in which the the biomass characteristics and bio-oil properties.
maximum rate of weight loss occurs, is given by the DTG curve.
For the heating rate of 10 °C, the major weight loss 39.3–42.2% 3.2. Influence of pyrolysis parameters on product yields
was occurred at 312.3 °C obtained from the DTG curve. The degra-
dation of the biomass occurs at three stages as observed from the 3.2.1. Effect of temperature and heating rate on pyrolysis product
TG/DTA curve. The first stage degradation occurs as the tempera- yields
ture increased from ambient temperature to a temperature of The effect of temperature (350–650 °C), heating rate (10 °C and
about 150 °C, in which a small weight loss of 15.10% due to the loss 40 °C min1) and sweeping gas flow rate (50–250 ml min1) were
of moisture and light volatile compounds present in the biomass investigated in this study. Table 1 shows the product distribution
sample which was also shown by the DTG curve with a peak at of pyrolysis of A. donax in different temperatures with different
67.6 °C. The major weight loss occurs at the second stage in the heating rates. The influence of temperature and heating rate on
temperature range of 150–400 °C which corresponds with a DTG the pyrolysis products yield are shown in Fig. 2(a) and (b). For
peak at 312.3 °C. The second stage of decomposition can be treated the heating rate of 10 °C min1, the char yield decreased from
as active pyrolysis zone due to high decomposition rate. The 45.77% to 40.51% when pyrolysis temperature increased from
weight loss in this stage is due to the breakdown of the hemicellu- 350 to 500 °C. The char yield was found to be decreased from
lose and cellulose molecule present in the biomass. After this a 42.38% to 33.50% for the temperature range of 350–500 °C at heat-
third stage from temperature range of 400–800 °C where a slow ing rate of 40 °C min1. Either the secondary decomposition reac-
continuous loss of weight occurs which is due to the slow decom- tions or the significant loss of volatile matter (indicating
position of lignin based compounds present in the biomass. significant loss of CH4, H2 and CO, dehydration of hydroxyl groups
Presence of wide range of organic chemicals in A. donax is evi- and thermal degradation of the lignocellulose structure) at higher
dent from the presence of functional groups identified from FTIR temperature may be responsible for the decreasing value of char
analysis (Supplementary Fig. S4). The absorption peak at yield with increasing temperature (Chutia et al., 2014). The forma-
3395 cm1 is due to O–H stretching indicating the presence of tion of lower char yields at higher heating rates and higher char
hydroxyl group which was attributed to moisture content of the yields at lower heating rates may be due to the fact that fast depo-
sample. The strong peak at 2919 cm1 is due to the axial deforma- lymerization of solid biomass to primary volatiles due to rapid
tion of C–H stretching vibration in CH2 groups. The peak at heating occurs at higher heating rate while dehydration to more
1638 cm1 is due to the amid-II band which is due to N–H bending stable anhydrocellulose is limited and slow at lower heating rate
vibration in conjunction with the aromatic ring (Fischer et al., (Chen et al., 1997). For the heating rate of 10 °C min1 the yield
2006). Presence of C@C stretching vibration at 1509 cm1 provides of bio-oil increased from 16.67% to 21.54% and for the heating rate
evidence of the presence of alkenes as well as aromatic com- of 40 °C min1 from 18.64% to 26.18% as the pyrolysis temperature
pounds. The peak at 1424 cm1 refers to the symmetric deforma- rise from 350 to 500 °C. But, with further increase in furnace tem-
tion of CH2 group of cellulose, 1244 cm1 is due to the C–O–C in perature from 500 to 650 °C, it was found that, bio-oil yield
cellulose chain (Mothe and Miranda, 2009). The peak at decreased to 16.65% and 21.02% for heating rate of 10 °C min1
1047 cm1 is due to the C–O–C symmetric stretching of dialkyl and 40 °C min1 respectively. In the present investigation, the
ether linkage and C–O stretching vibration in cellulose, hemicellu- bio-oil yield was found to be maximum at 500 °C for both the heat-
loses and minor lignin contribution (Bilba et al., 2007). The peaks ing rate of 10 and 40 °C min1 with increasing temperature from
between 800 and 600 cm1 are due to aromatic structures 350 to 650 °C. The higher pyrolysis temperatures have been

Table 1
Effect of temperature and heating rate on product yield.

Temperature (°C) Char (%) Bio-oil (%) Gas (%) Aqueous phase (%)
Heating rate 10 °C min1
350 45.77 ± 0.01 16.67 ± 0.01 28.74 ± 0.04 8.82 ± 0.04
400 43.87 ± 0.02 18.26 ± 0.11 30.61 ± 0.10 7.26 ± 0.07
450 42.91 ± 0.01 20.13 ± 0.02 32.97 ± 0.03 3.99 ± 0.01
500 40.51 ± 0.05 21.54 ± 0.39 34.58 ± 0.04 3.37 ± 0.03
550 35.24 ± 0.03 19.51 ± 0.09 41.03 ± 0.08 4.22 ± 0.02
600 33.11 ± 0.02 17.59 ± 0.20 44.59 ± 0.06 4.71 ± 0.03
650 32.98 ± 0.01 16.65 ± 0.04 45.16 ± 0.01 5.21 ± 0.01
Heating rate 40 °C min1
350 42.38 ± 0.05 18.64 ± 0.01 29.96 ± 0.04 9.02 ± 0.02
400 38.04 ± 0.04 22.49 ± 0.01 31.54 ± 0.06 7.93 ± 0.01
450 33.72 ± 0. 14 25.91 ± 0.07 33.75 ± 0.03 6.62 ± 0.02
500 33.50 ± 0.03 26.18 ± 0.10 36.29 ± 0.07 4.03 ± 0.05
550 32.15 ± 0.01 23.82 ± 0.01 38.11 ± 0.12 5.92 ± 0.01
600 30.18 ± 0.03 21.79 ± 0.04 40.12 ± 0.09 7.91 ± 0.03
650 29.8 ± 0.02 21.02 ± 0.12 41.23 ± 0.05 7.95 ± 0.01
 R. Saikia et al. / Bioresource Technology 188 (2015) 265–272 269

(a) (b) 45

45

char
40
bio-oil Char
40
gas Bio-oil
Gas
35
35

% Yield
% Yield

30 30

25
25

20

20

15
350 400 450 500 550 600 650 350 400 450 500 550 600 650
0 0
Temperature ( C) Temperature ( C)

Fig. 2. (a) Yield of pyrolysis products of A. donax at different temperatures at 10 °C min1 heating rate. (b) Yield of pyrolysis products of A. donax at different temperatures at
40 °C min1 heating rate.

associated with the secondary cracking reactions of the pyrolysis temperature of 500 °C and constant heating rate of 40 °C min1,
gases resulting in increased gas yield and reduced bio-oil yield the sample was pyrolysed with sweep gas flow rates of 50, 100,
(Choudhuri et al., 2014). It was observed that the gas product yield 150, 200 and 250 ml min1. Fig. 3 shows the N2 sweep gas flow rate
increased with the increase of pyrolysis temperature in both the effect on the product yields. The pyrolysis product distribution
heating rates. For the heating rate of 10 °C min1 the minimum with sweep gas flow rate is presented in Table 2. Sweeping gas
gaseous product obtained was 28.74% at 350 °C and maximum removes the volatile products formed during pyrolysis process
yield was 45.16% to at 650 °C. For 40 °C min1, the gas yield from the hot zone of the reactor where the thermal cracking, repo-
increased from 29.96% to 41.23%. The increasing yield of gaseous lymerization and recondensation reactions take place. The vapour
product probably due to both the secondary cracking of the pyro- residence time is very much significant for pyrolysis product yield.
lysis vapours and non-condensable gaseous products formed by Due to secondary cracking of vapour and recondensation reactions,
the secondary decomposition of char at higher temperatures and low N2 flow rate leads to higher vapour residence time giving low
heating rates (Encinar et al., 2000). The aqueous phase obtained yield of bio-oil in pyrolysis experiments. Experimental results
during pyrolysis process is due to the inherent moisture in the show that bio-oil yield first increases from 18.08% to 25.21% with
bio-oil and result of dehydration reaction occurring during pyroly- increase in N2 gas flow rate. But when flow rate of the sweep gas
sis. All the pyrolysis experiments were performed three times and was further increased up to 250 ml min1, the yield of bio-oil
the average value was reported. The maximum bio-oil yield was decreased to 19.41%. Thus, highest liquid product was obtained
found to be 26.18% at 500 °C with heating rate of 40 °C min1. for a sweep gas flow rate of 150 ml min1 during our study proba-
bly due to the minimum secondary reactions of the vapours
(Choudhuri et al., 2014).
3.2.2. Effect of sweep gas flow rate
Pyrolysis experiments were also carried out to investigate the
effect of sweep gas flow rate on the product yields. At a constant 3.3. Characterization of bio-oil

45 3.3.1. Physical properties

The A. donax bio-oil obtained at 500 °C is reddish brown in col-
our. The elemental composition, empirical formula, H/C and O/C
40
char ratio and higher heating value of A. donax bio-oil are presented
bio-oil in Supplementary Table S5. It was observed that the bio-oils con-
gas
35 tain less oxygen than the original feedstock. The carbon, hydrogen
and nitrogen content of the A. donax bio-oil pyrolysed at 500 °C
% Yield

30 were found to be 47.95%, 7.23% and 0.3% respectively. The H/C

molar ratio of the bio-oil was found to be 1.79 which is higher than
the original feedstock and quite comparable with the light petro-
25
leum products. The calorific value of the bio-oil increases to
24.7 MJ kg1 from the original biomass having calorific value of
20 17.14 MJ kg1.
The comparison of physical properties of the bio-oil with petro-
diesel is presented in Table 3. The bio-oil is denser in nature com-
15
50 100 150 200 250 pared to the transportation fuels. The flash point of the A. donax
Sweep gas flow rate(ml/min) bio-oil is in a comparable range with the transportation fuels.
The pour point has a value, which can be accepted for most of
Fig. 3. Effect of sweep gas flow rate on pyrolysis product yield. the climatic conditions.
270 R. Saikia et al. / Bioresource Technology 188 (2015) 265–272

Table 2
Effect of sweep gas flow rate on pyrolysis product distribution.

Sweep gas flow rate (ml min1) Char (%) Bio-oil (%) Aqueous phase (%) Gas (%)
50 36.78 ± 0.01 18.08 ± 0.13 9.32 ± 0.05 30.58 ± 0.01
100 32.46 ± 0.01 22.96 ± 0.03 8.51 ± 0.02 32.62 ± 0.03
150 31.61 ± 0.05 25.21 ± 0.04 6.59 ± 0.03 34.53 ± 0.01
200 31.36 ± 0.09 22.53 ± 0.01 6.49 ± 0.05 39.62 ± 0.07
250 30.81 ± 0.01 19.41 ± 0.03 6.27 ± 0.11 43.51 ± 0.01

Table 3
Fuel properties of A. donax pyrolytic oil.

Properties Standard test method Pyrolytic oil Gasoline Diesel

Appearance Dark brownish oil  
Density, 15 °C (kg/m3) ASTM D1298-99 0.9518  
Specific gravity ASTM D1298-99 0.9521 0.72–0.78 0.82–0.85
pH 3.8  
Pour point ASTM D5853-09 Plus 4 40 40 to 1
Cloud point (°C) ASTM D1310-01 (2007) 14  
Flash point (°C) ASTM D6450-05 (2010) 43 43 53–80
Fire point (°C) ASTM D1310-01 (2007) 57  

3.3.2. Chemical properties and 9.0 ppm is assigned due to protons of the aromatic ring
The FTIR spectra provide information about the presence of (18%). The integrated region between 4.0 and 6.0 ppm is due to
wide range of organic chemicals as evident from the functional the phenolic and olefinic protons (16%). The resonances of protons
groups identified in the spectra (Supplementary Fig. S6). The obtained in the integrated region 3.0–4.0 ppm is due to the aro-
absorption peak at 3367.8 cm1 indicates the presence of oxygen- matic ring-joining methylene groups (6%). CH3, CH2, CH proton
ated compounds along with moisture (Qiang et al., 2008). The present in alpha or further from an aromatic ring resonates in
peaks at 2853–2924 cm1 are due to the presence of C–H stretch- the range 2.0–3.0 ppm (24%). The protons between 0.5 and
ing vibration of alkane. The peak at 1743.8 cm1 indicates the pres- 3.0 ppm corresponds to the aliphatic protons (55%).
ence of carbonyl group of ketone or aldehyde. The peaks at 1514.6 The 13C NMR spectra of the A. donax bio-oil is shown in the Sup-
and 1612.1 cm1 provides the evidence of C@C bond of alkene and plementary Fig. S8. Table 5 represents the percentages of carbon
aromatics present in the bio-oil. The C–H deformation vibrations obtained based on 13C NMR analysis. Signals in the 055 ppm
have been observed at 1378–1464 cm1 due to alkane groups. range are due to the aliphatic carbon atoms that are separated
The C–H in plane benching vibration at 827–1025 cm1 confirms from oxygen atoms by at least two bonds (31%), although those
the presence of aromatic compounds. The peaks at 1116.1– adjacent to nitrogen atoms can be found in this region. The inte-
1270.2 cm1 indicate the presence of oxygenated compounds in grated region between 0–28 ppm is due to the short aliphatic
the form of alcohol and ether. (19%) present in the bio-oil. Signals in the region 28–55 ppm is
Bio-oil is a complex mixture of several hundreds of organic due to the long and branched aliphatic (12%) which could be due
compounds that exhibits a wide range of chemical functionality. to the presence of saturated and unsaturated free fatty acids and
Bio-oil contains water, guaiacols, catechols, syringols, vanillins, fatty acid derivatives. Signals in between 55 and 95 ppm are due
furancarboxaldehydes, isoeugenol, pyrones, acetic acid, formic acid
and other carboxylic acids (Ali et al., 2010). Other major groups of
Table 4
compounds, including hydroxyaldehydes, hydroxyketones, sugars, 1
H NMR spectral results of bio-oil obtained at condition of maximum yield.
carboxylic acids and phenolics are also present in bio-oil. Due to
the very complex nature, several analytical techniques such as Chemical Proton assignment % of total hydrogen
shift (ppm)
chromatographic (GC, HPLC, GPC) and spectroscopic (IR, MS) are
used for chemical characterization of bio-oil. But only 25–40% of 6.0–9.0 Aromatic 18
4.0–6.0 Phenolic (OH) or olefinic proton 16
bio-oil compounds are observable by GC analytical technique,
3.0–4.0 Ring-join methylene (Ar-CH2-Ar) 6
since the large fraction of the bio-oils consisting of lignin and car- 2.0–3.0 CH3, CH2 and CHa to an aromatic ring 24
bohydrate oligomers are not volatile enough to be observed by GC 0.5–3.0 Total aliphatics 55
(Mullen et al., 2009). FTIR and HPLC analysis can give only a quan-
titative measurement of the functional groups and water soluble
species present in the bio-oil respectively. Gel permeation chroma-
Table 5
tography (GPC) used for obtaining molecular weight distributions 13
C NMR spectral results of bio-oil obtained at condition of maximum yield.
on higher molecular weight phenolic species derived from lignin
Chemical Carbon assignment % of total
only characterizes only a portion of the bio-oil (Mullen et al.,
shift carbon
2009). Therefore, both 1H and 13C NMR spectroscopy was carried (ppm)
out to examine the different environmental hydrogen and carbon
0–28 Short aliphatics 19
present in the entire bio-oil rather than a selected fraction of it 28–55 Long and branched aliphatics 12
(Sensoz and Kaynar, 2006; Demiral et al., 2009). 55–95 Alcohols, ethers, phenolic-methoxys, carbohydrate 29
The 1H NMR spectra of A. donax bio-oil is given in the Supple- sugars
mentary Fig. S7. Table 4 shows the percentages of different types 95–165 Aromatics, olefins 38
165–180 Esters, carboxylic acids 3
of hydrogen present in the bio-oil obtained from chemical shift
180–215 Ketones, aldehydes 1
with the selected regions. In the spectra, regions between 6.0
 R. Saikia et al. / Bioresource Technology 188 (2015) 265–272 271

Table 6
Elemental analysis of biochar produced from A. donax.

A. donax biochar
350 °C 400 °C 450 °C 500 °C 550 °C 600 °C 650 °C
Water content 6.93 5.63 5.43 4.31 4.27 4.25 4.19
Volatile matter 40.89 34.02 28.96 23.02 22.63 19.95 17.91
Ash content 7.33 8.27 9.97 10.44 10.68 10.88 10.94
Fixed carbon 44.85 52.08 55.64 62.23 62.42 64.92 66.96
pH 9.1 9.12 9.33 9.93 10.01 10.42 10.45
Electrical Conductivity (mf) 0.98 1.20 1.37 1.51 1.69 1.75 1.92
Elemental analysis (wt %)
C 56.29 57.12 57.53 65.74 69.04 71.56 72.95
H 2.52 2.49 2.47 2.05 1.93 1.63 1.14
N 2.35 2.89 2.59 2.34 1.52 1.78 0.97
Oa 38.84 37.49 37.42 33.87 27.51 25.03 24.94
H/C 0.54 0.52 0.51 0.37 0.33 0.27 0.18
O/C 0.51 0.49 0.48 0.39 0.29 0.26 0.25
HHV (MJ kg1) 24.21
a
By difference.

to the carbon adjacent to heteroatoms (mostly oxygen) in alcohols, FTIR spectra of the biochars were taken at different final tem-
ethers and anhydrous carbohydrates (29%). All the methyl groups peratures of 350, 400, 450, 500, 550 and 600 °C with heating rate
present in this region are may be due to the methoxy groups on of 40 °C min1 and 150 ml min1 N2 flow rate (Supplementary
2 and 6 positions of phenolic rings (structures derived from lignin). Fig. S11). In the spectra, the peak at 3412 3432 cm1 was due
The peak for the solvent (CDCl3) also appear in this region. The to O–H stretching vibrations and hydrogen bond of hydroxyl
integrated region 95–165 ppm in the spectra represents aromatic groups (Yang et al., 2007). Peaks appeared at around 2926 cm1
carbon from both benzene rings and hetero aromatics (e.g. furans) are attributed to the C–H stretching vibration of alkanes from CH
(38%). The signals in the spectra 165–180 ppm includes carbonyl and CH2 in cellulose and hemicellulose components (Paiva et al.,
carbons representing carboxylic acids, esters and amides (3%). 2007). The presence of carbonyl group in the biochar is limited
The ketones and aldehydes present in the bio-oil appear in the as no intense peak is observed in the 1650–1750 cm1 region of
region 180–215 ppm (1%) (Chutia et al., 2014). the FTIR spectra. The peaks at 1578–1602 cm1 provides the evi-
The GC–MS analysis of the bio-oil shows the presence of acids, dence of C@C bond of alkene and aromatic present in the biochar.
alcohols, ketones, aldehydes, phenols, esters, sugars, furans and The presence of oxygen containing functionality in the biochar
other oxygenated hydrocarbon molecules (Supplementary docu- confirmed by the C–O stretching around 1072 cm1. The FTIR band
ment S9 and S10). In the GC–MS chromatograph more than 30 below 600 appears due to the metal oxygen stretching (Wen et al.,
peaks were obtained, but due to complex nature of the bio-oil 2007).
and limitation of the strength of MS library, perfect separation of The surface morphology of the biochar was studied by Scanning
all the peaks is not possible. Comparing with the mass spectra from Electron Microscopy (SEM) image obtained at 500 °C and 650 °C at
the NIST 98 spectrum library the peaks were identified. 40 °C min1 heating rate (Supplementary Fig. S12). The SEM
images confirmed the amorphous and heterogeneous structure of
the biochar obtained from A. donax. It was observed that the bio-
char obtained at 500 °C has a porous structure, indicating its prob-
3.4. Characterization of biochar able use as an adsorbent as well as development of solid acid
catalyst. From the EDX analysis (Supplementary Fig. S13), it was
The proximate and elemental analysis of A. donax biochar observed that the biochar from A. donax mainly consists of P, K
obtained at different temperatures at 40 °C min1 heating rate and Ca which are very important inorganic nutrient for soil fertility
are presented in Table 6. The biochar contains less oxygen and high and crop production (Chutia et al., 2014).
carbon content than the original feedstock. It was observed that
atomic H/C and O/C ratios of biochars decreased with an increase
in pyrolysis temperature, which may be due to dehydration, decar- 4. Conclusion
boxylation and decarbonylation. The oxygen containing function-
alities are decomposed by decarbonylation and decarboxylation The maximum bio-oil yield of 26.18% was obtained at a final
followed by transformation to the alkyl-aryl C–C bonds as a cross temperature of 500 °C, with a heating rate of 40 °C min1. It has
linking between small aromatic rings (Kim et al., 2011; Tang and a H/C atomic ratio of 1.79 with a heating value of 24.70 MJ kg1.
Bacon, 1964). There occurs a significant increase of the calorific The biochar obtained at the optimum condition has a calorific
value of the biochar i.e. 24.21 MJ kg1 from the original biomass value of 24.21 MJ kg1 with a porous structure. With a high pH,
having calorific value of 17.14 MJ kg1. Thus the biochar obtained it may be used as a liming material for correcting soil acidity. Thus,
from the A. donax biomass can be used as a solid fuel upon A. donax is a potential feedstock for pyrolytic conversion to energy
densifying. and materials.
The pH of the biochars obtained increased with increase in
pyrolysis temperature as at higher pyrolysis temperatures, concen- Acknowledgements
tration of alkali salts increases due to removal of organic materials
from the char. The pH is in the highly basic region and it can be The first author gratefully acknowledges the Senior Research
used for soil amendment of the soil of North-Eastern region of Fellowship from DBT, Govt of India through SAHYOG-EU-FP7-KBBE
India to neutralize soil acidity and increase soil pH (Bordoloi (BT/IN/EU/07/PMS/2011) along with Guwahati Biotech Park for the
et al., 2015). GC–MS and CSIR-NEIST for the NMR analysis.
272 R. Saikia et al. / Bioresource Technology 188 (2015) 265–272

Appendix A. Supplementary data Greenhalf, C.E., Nowakowski, D.J., Harms, A.B., Titiloye, J.O., Bridgwater, A.V., 2013. A
comparative study of straw, perennial grasses and hardwoods in terms of fast
pyrolysis products. Fuel 108, 216–230.
Supplementary data associated with this article can be found, in Kim, P., Johnson, A., Edmunds, C.W., Radosevich, M., Vogt, F., Rials, T.G., Labbe,
the online version, at [Link] N., 2011. Surface functionality and carbon structures in lignocellulosic-
derived biochars produced by fast pyrolysis. Energy Fuels 25 (10), 4693–
089.
4703.
Lee, M.K., Tsai, W.T., Tsai, Y.L., Lin, S.H., 2010. Pyrolysis of napier grass in an
References induction-heating reactor. J. Anal. Appl. Pyrol. 88, 110–116.
Mosier, N.S., Hendrickson, R., Brewer, M., Ho, N., Sedlak, M., Dreshel, R., Welch, G.,
Anex, R.P., Aden, A., Kazi, F.K., Fortman, J., Swanson, R.M., Wright, M.M., Satrio, J.A., Dien, B.S., Aden, A., Ladisch, M.R., 2005. Industrial scale-up of pH-controlled
Brown, R.C., Daugaard, D.E., Platon, A., Kothandaraman, G., Hsu, D.D., Dutta, A., liquid hot water pretreatment of corn fiber for fuel ethanol production. Appl.
2010. Techno-economic comparison of biomass-to-transportation fuels via Biochem. Biotechnol. 125, 77–97.
pyrolysis, gasification, and biochemical pathways. Fuel 89, S29–S35. Mothe, C.G., Miranda, I.C., 2009. Characterization of sugarcane and coconut fibers by
Ali, T., Hakki, A.M., Nasko, T., Sabrina, P., Elisabetta, F., 2010. Efficiency of bio-oil thermal analysis and FTIR. J. Therm. Anal. Calorim. 97, 661–665.
against wood destroying organisms. J. Biobased Mater. Bioenergy 4 (4), 317– Mullen, C.A., Strahan, G.D., Boateng, A.A., 2009. Characterization of various fast-
323(7). pyrolysis bio-oils by NMR spectroscopy. Energy Fuels 23, 2707–2718.
Bilba, K., Arsene, M.A., Ouensanga, A., 2007. Study of banana and coconut fibers Nanda, S., Mohanty, P., Kozinski, J.A., Dalai, A.K., 2014. Physico-chemical properties
botanical composition, thermal degradation and textural observations. of bio-oils from pyrolysis of lignocellulosic biomass with high and slow heating
Bioresour. Technol. 98, 58–68. rate. Energy Environ. Res. 3 (4), 21–32.
Boateng, A.A., Anderson, W.F., Phillips, J.G., 2007. Bermudagrass for biofuels: effect Paiva, M.C., Ammar, I., Campos, A.R., Cheikh, R.B., Cunha, A.M., 2007. Alfa fibres:
of two genotypes on pyrolysis product yield. Energy Fuels 21, 1183–1187. mechanical, morphological and interfacial characterization. Compos. Sci.
Boateng, A.A., Hicks, K.B., Vogel, K.P., 2006. Pyrolysis of switchgrass (Panicum Technol. 67, 1132–1138.
virgatum) harvested at several stages of maturity. J. Anal. Appl. Pyrol. 75 (2006), Promdee, K., Vitidsant, T., 2012. Received oil yields and chemical compounds of
55–64. mixed grass synthesized by pyrolysis process. J. Appl. Chem. 3 (2), 14–20.
Bordoloi, N., Narzari, R., Chutia, R.S., Bhaskar, T., Kataki, R., 2015. Pyrolysis of Mesua Qiang, L., Xu-lai, Y., Xi-feng, Z., 2008. Analysis on chemical and physical properties
ferra and Pongamia glabra seed cover: characterization of bio-oil and its sub- of bio-oil pyrolyzed from rice-husk. J. Anal. Appl. Pyrol. 82, 191–198.
fractions. Bioresour. Technol. 178, 83–89. Sarkhot, D.V., Grunwald, S., Ge, Y., Morgan, C.L.S., 2012. Total and available soil
Bridgwater, A.V., Meier, D., Radlein, D., 1999. An overview of fast pyrolysis of carbon fractions under the perennial grass Cynodon dactylon (L.) Pers and the
biomass. Org. Geochem. 30 (12), 1479–1493. bioenergy crop Arundo Donax L. Biomass Bioenergy 41, 122–130.
CABI datasheet. <[Link] (09.10.14). Sensoz, S., Kaynar, I., 2006. Bio-oil production from soybean (Glycine max L.) fuel
Chen, G., Yu, Q., Sjostrom, K., 1997. Reactivity of char from pyrolysis of birch wood. J. properties of bio-oil. Ind. Crops Prod. 23, 99–105.
Anal. Appl. Pyrol. 40–41, 491–499. Strezov, V., Evans, T.J., Hayman, C., 2008. Thermal conversion of elephant grass
Choudhuri, N.D., Chutia, R.S., Bhaskar, T., Kataki, R., 2014. Pyrolysis of jute dust: (Pennisetum Purpureum Schum) to bio-gas, bio-oil and charcoal. Bioresour.
effect of reaction parameters and analysis of products. J. Mater. Cycles Waste Technol. 99, 8394–8399.
Manage. 16 (3), 449–459. Tang, M.M., Bacon, R., 1964. Carbonization of cellulose fibers – I. Low temperature
Christou, M., 2013. Giant reed (Arundo donax L.) agronomy and yields in Europe. pyrolysis. Carbon 2 (3), 211–214.
<[Link] (09.10.14). Varol, E.A., Putun, E., Putun, A.E., 2007. Slow pyrolysis of pistachio shell. Fuel 86,
Chutia, R.S., Kataki, R., Bhaskar, T., 2014. Characterization of liquid and solid product 1892–1899.
from pyrolysis of Pongamia glabra deoiled cake. Bioresour. Technol. 165, 336–342. Vispute, T.P., Zhang, H., Sanna, A., Xiao, R., Huber, G.W., 2010. Renewable chemical
Chutia, R.S., Kataki, R., Bhaskar, T., 2013. Thermogravimetric and decomposition commodity feedstocks from integrated catalytic processing of pyrolysis oils.
kinetic studies of Mesua ferrea L. deoiled cake. Bioresour. Technol. 139, 66–72. Science 330, 1222–1227.
Demiral, I., Atilgan, N.G., Sensoz, S., 2009. Production of biofuel from soft shell of Wen, B., Zhang, J.J., Zhang, S.Z., Shan, X.Q., Khan, S.U., Xing, B., 2007. Phenanthrene
pistachio (Pistacia vera L.). Chem. Eng. Commun. 196, 104–115. sorption to soil humic acid and different humin fractions. Environ. Sci. Technol.
Encinar, J.M., Gonzalez, J.F., Gonzalez, J., 2000. Fixed-bed pyrolysis of Cynara 41, 3165–3171.
cardunculus L. product yields and compositions. Fuel Process. Technol. 68, 209–222. Wright, M.M., Daugaard, D.E., Satrio, J.A., Brown, R.C., 2010. Techno-economic
Fischer, G., Braun, S., Thissen, R., Dott, W., 2006. FT-IR spectroscopy as a tool for analysis of biomass fast pyrolysis to transportation fuels. Fuel 89, S2–S10.
rapid identification and intra-species characterization of airborne filamentous Yang, H., Yan, R., Chen, H., Lee, D.H., Zheng, C., 2007. Characteristics of
fungi. J. Microbiol. Methods 64 (1), 63–77. hemicellulose, cellulose and lignin pyrolysis. Fuel 86, 1781–1788.