Energy for Sustainable Development 42 (2018) 160–169

Contents lists available at ScienceDirect

Energy for Sustainable Development

Energy densification of sugarcane leaves through torrefaction under

minimized oxidative atmosphere
Angelique T. Conag, Jaye Earl R. Villahermosa, Luis K. Cabatingan ⁎, Alchris W. Go ⁎
Department of Chemical Engineering, University of San Carlos, Talamban, Cebu City 6000, Philippines

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

Article history: Worldwide, the annual energy content of sugarcane leaves (SCL) amounts to 4.21 EJ. However, its direct use as a
Received 21 July 2017 solid fuel is hindered by its inherent characteristics, such as its low bulk density, low energy density, and hygro-
Revised 28 November 2017 scopic nature. These characteristics of SCL could be improved through torrefaction. However, in addition to these
Accepted 28 November 2017
characteristics, combustion and flow characteristics, expressed in their respective indices, should also meet their
Available online xxxx
desired values as they have direct influence in the later performance of boilers and furnaces. Torrefaction of SCL at
Keywords:
various temperatures (250–350 °C) and times (0–75 min) under minimized oxidative atmosphere was investi-
Energy densification gated in this study to improve its characteristics as solid fuel. The torrefaction performance was not only assessed
Combustion indices through the mass and energy yields, but also changes in the combustion and flow indices. Torrefaction at 300 °C &
Higher heating value 45 min led to the increase in fixed carbon to as much as 21 g for every 100 g of raw SCL being torrefied, which
Sugarcane leaves originally contains 18 g of fixed carbon. Devolatilization during torrefaction not only involves the volatilization
Torrefaction of the volatiles but also the formation of additional fixed carbon. The higher heating value (HHV) of SCL increased
to as much as ~22 MJ/kg after the torrefaction. The resulting SCL torrefied at 300 °C for at least 45 min was found
to be suitable for industrial and domestic applications having fuel ratio (FR), combustibility index (CI), volatile
ignitability (VI), Hausner ratio (HR), and Carr compressibility index (CCI) falling within prescribed values of
0.5–2.0, 12–23 MJ/kg, N14 MJ/kg, b1.34, and b25, respectively.
© 2017 International Energy Initiative. Published by Elsevier Inc. All rights reserved.

Introduction problems and violate various environmental laws. During burning of

Sugarcane is the second most produced crop in the world after rice
paddy with a total production of 1.88 × 109 tons in 2014 (Food and
Agricultural Organization of the United Nations Statistics Division and
FAOSTAT, 2014). During harvesting of sugarcane, its leaves and tops,
which are undesirable during milling, are removed from the stalks.
Collectively, sugarcane leaves and tops are referred to as sugarcane
trash (SCT). For every ton of sugarcane harvested, about 140 to 180 kg
of SCT is produced (Paes & M., 2005). In order to prepare the fields for
the next cropping season, sugarcane leaves and tops left on the field
are usually burned. Pre-harvest burning is commonly practiced by
sugarcane farmers to remove the leaves prior to harvesting. However,
this is discouraged to avoid deterioration of the available sugars
(Mendoza, 2015). Both pre-harvest and post-harvest burning of SCT
are highly discouraged since these pose several environmental
Abbreviations: SCL, sugarcane leaves; TSCL, torrefied sugarcane leaves; HHV, higher
heating value; FR, fuel ratio; VI, volatile ignitability; CI, combustibility index; HR,
Hausner ratio; CCI, Carr compressibility index; M, moisture; VM, volatile matter; FC,
fixed carbon; db, dry basis.
⁎ Corresponding authors.
E-mail addresses: lkcabatingan@usc.edu.ph (L.K. Cabatingan), awgo@usc.edu.ph
(A.W. Go).
SCT, carbon dioxide (CO2) and other greenhouse gases (GHGs) are re-
leased to the atmosphere.
Based on the sugarcane production in 2014, ~ 264 million tons of
SCT, with an HHV of 15–17 MJ/kg, the total amount of energy which
could have been recovered would sum up to as much as 4.21 EJ of ener-
gy when burned. Sugarcane trash is mainly comprised of approximately
97% sugarcane leaves (SCL) or straw (Hassuani, Leal, & de Carvalho
Macedo, 2005). This amount of energy can be recovered when SCL is uti-
lized as a solid biofuel. However, the direct utilization of SCL as solid fuel
is hindered by its regular availability and physico-chemical characteris-
tics. Although sugarcane is produced in large quantities, it is, however,
a seasonal crop, with its harvesting only done from the months of
November to April. Its low bulk density and high moisture content
pose a problem in its transport and storage. Moreover, when tapped
as an energy source, its utilization would best be distributed so to
have a constant and stable supply of energy. The storage of SCL may po-
tentially be hindered by its high moisture content (13.5 to 70%)
(Hassuani et al., 2005) and hygroscopic nature. These characteristics
of sugarcane residues make it susceptible to biological degradation
through microbial activities during storage and results in poor burning
characteristics when used as solid fuel. In addition, the high moisture
content of SCL leads to a low heating value and would require an
energy-intensive drying step, resulting in a low energy conversion

https://doi.org/10.1016/j.esd.2017.11.004
0973-0826/© 2017 International Energy Initiative. Published by Elsevier Inc. All rights reserved.
 A.T. Conag et al. / Energy for Sustainable Development 42 (2018) 160–169
efficiency. Moreover, the direct use of wet SCL increases the risk of
corrosion in combustion systems owing to the wet flue gas produced
during combustion (Bach & Skreiberg, 2016). Although moisture in
SCL could be addressed through various strategies including sun
drying and/or thermal drying with the aid of waste heat, liberated gas
during pyrolysis and/or supplemented with excess low-grade biomass
(Kiel, Janssen, & Joshi, 2015), these strategies would not prevent the
later reabsorption of moisture during storage (Pimchuai, Dutta, &
Basu, 2010; Supramono, Devina, & Tristantini, 2015). The capability of
biomass, like that of SCL, to reabsorb moisture is attributed to the
presence of hydroxyl groups in its chemical structure which forms
hydrogen bonds with water. The capacity of the biomass to reabsorb
moisture is minimized through the elimination of the hydroxyl
groups and the formation of unsaturated structures which can
be achieved through a thermochemical treatment referred to as
torrefaction (Bergman, Boersma, RWR, & JHA, 2005). These problems
are not unique to SCL but are common among agricultural biomass
tapped for energy production. Torrefaction, a thermal pre-treatment
process, has been found to improve not only the hydrophobicity but
as well as the combustion characteristics of the torrefied biomass in
view of its use as fuel.
Coal is the most widely used solid fuel in power generation systems,
thus related combustion systems have been designed to operate based
on the ignition and combustion characteristics of coal. Considering
that the use of agricultural residues, like that of SCL, as a replacement
for coal, it would be best that their fuel characteristics are similar so
not to jeopardize the performance of the combustion system. Industri-
ally used coal typically has a fuel ratio (FR) ranging from 0.5 to 2.0
(Makino & Tanno, 2015; Ohm, Chae, Kim, & Oh, 2015). Presence of
high volatile matter (VM) content results in low FR and a fuel which
is prone to either emit high amounts of smoke or a fuel which rapidly
burns, posing a challenge in controlling the ignition flame. High FR as
a result of low VM is also not desired as it would entail difficulties in ig-
nition and flame stability (Engineering Foundation (U.S.). Conference
(1993: Solihull et al., 1994). In view of volatile ignitability (VI), this
measure estimates the gross calorific value of the total volatiles (volatile
matter and moisture) in the solid, considering that fixed carbon (FC) is
solely made up of carbon atoms. This measure is determined to be
N14 MJ/kg for most coals used (Asthana, 2015; Magasiner, van
Alphen, Inkson, & Misplon, 2001). Combustibility index (CI) on the
other hand, is a measure considered when mixing solid fuel alternatives
for co-firing with coal, which is found to range from 12 to 23 MJ/kg for
applications in coal fired power plant (Ohm et al., 2015). Apart from
combustion indices, its flow characteristics as described by Hausner
ratio (HR) or Carr compressibility index (CCI) should be b 1.34 and 25,
respectively, for use in fluidized bed combustion systems (Tannous,
Lam, Sokhansanj, & Grace, 2013).
In view of SCL, torrefaction studies have been limited. Although
its use has been recognized by researchers (Soponpongpipat,
Sittikul, & Sae-Ueng, 2015), detailed investigation on its torrefaction
behavior has not been looked into. A study by Soponpongipipat et al.
(Soponpongpipat et al., 2015) reported the improvement in HHV of
sugarcane leaves from 17.9 to 21.7 MJ/kg after torrefaction at 280 °C
for 60 min. However, the effects of time and temperature on the fuel
characteristics were not investigated in detail. Moreover, the mass
yield and energy recovery after torrefaction of SCL were not looked
into. These are important measures needed in the later assessment of
the process feasibility and practical application.
In this study, the effects of time and temperature of the
torrefaction of SCL under minimized oxidative conditions were
investigated in relation to mass and energy yields of the process.
Moreover, in this study, a possible mechanism and kinetics to de-
scribe the torrefaction process of sugarcane leaves is also proposed.
Furthermore, the hydrophobicity of torrefied SCL was also assessed
by looking into the equilibrium moisture reabsorption. Related fuel
quality indices (FR, CI, and VI) and the bulk density of the torrefied
161
SCL were also determined to assess its potential use for domestic
and industrial applications.
Material and methods
Sugarcane leaves from sugarcane (Saccharum officinarum x Saccharum
spontaneum) were obtained from the fields of Bogo, Cebu during the
month of September 2016 before the harvesting season. Collected SCL
samples were determined of its initial moisture content as received
and dried in an oven at 60 °C for 3 to 4 days or until the moisture con-
tent is 7.05 ± 1.07%w/w (dry basis). Dried SCL were milled using a
Wiley mill (Thomas-Wiley Model 4, New Jersey, USA) with a 2 mm
mesh aperture screen. This was done to have samples of uniform parti-
cle distribution (d = 0.55 ± 0.01 μm) for later torrefaction experiments.
Milled dry-leaves were then stored in polypropylene containers at
room temperature for subsequent characterization and torrefaction ex-
periments. Locally produced coconut shell charcoal briquette (Charcoal
A) and wood charcoal (Charcoal B) were also acquired for comparison.
Bulk density and compressibility index
Adopting the procedures described by Tannous et al. (Tannous et al.,
2013), around 1–2 g of biomass samples were introduced using a funnel
to a pre-weighed dry graduated cylinder (10 mL or 10 × 10− 3 dm3)
without compacting. The loose volume was read off directly from the
cylinder. The graduated cylinder containing the sample was then
weighed to the nearest 0.1 mg. The loose (untapped) bulk density
(ρLb) was calculated using Eq. (1):
  m −m
Loose Bulk Density; ρLb g=mL or 10−3 kg=m3 ¼
gs g
ð1Þ
VL
where mg is the mass of the empty dry graduated cylinder, mgs is the
mass of the graduated cylinder containing the sample and VL is the
loose (untapped) volume. For the tapped bulk density, the graduated
cylinder containing the sample was tapped at least 300 times or until
a constant volume reading was achieved. The tapped bulk density
(ρTb) was calculated using Eq. (2):
  m −m
Tapped Bulk Density; ρTb g=mL or10−3 kg=m3 ¼
gs g
ð2Þ
VT
where VT is the tapped volume. The Hausner Ratio (HR) was calculated
using Eq. (3) while the Carr Compressibility Index (CCI) was calculated
using Eq. (4).
ρTb
Hausner Ratio; HR ¼ ð3Þ
ρLb
 
ρ
Carr Compressibility Index; CCI ð%Þ ¼ 1− Lb  100 ð4Þ
ρTb
Proximate analysis
Proximate analysis was carried out according to the procedures de-
scribed by the American Society for Testing and Materials (ASTM),
method D1762–84 (ASTM International, 2013) for chemical analysis
of wood charcoal. About 2 g of raw or 1 g of torrefied sample (weighed
to the nearest 0.1 mg) was weighed into a pre-fired and pre-weighed
porcelain crucible and was placed in an oven at 105 °C for 2 h. The cru-
cible containing the dried sample was then transferred into a desiccator
and allowed to cool for an hour before weighing. The above drying pro-
cedure was repeated until the decrease in weight of consecutive
weighing was 0.0005 g or less. Moisture content (M) of the sample
was calculated using Eq. (5).


ms − mcs;105 °C − mc
% Moisture; M ð%w=wÞ ¼  100 ð5Þ
ms
162 A.T. Conag et al. / Energy for Sustainable Development 42 (2018) 160–169
where, ms is the mass of the sample weighed before drying, mc is the
mass of the pre-dried crucible and mcs, 105 °C is the mass of the crucible
containing the sample after drying at 105 °C.
After moisture determination, the dried samples in the crucibles
were further analyzed for their volatile matter content (VM). Covered
crucibles containing the samples were placed on the outer ledge of a
muffle furnace (pre-heated to 950 °C) for 2 min and for 3 min on the
edge of the furnace. Finally, the covered crucibles were placed at the
rear of the furnace for 6 min with the furnace door closed (ASTM
International, 2013). After the designated time, the covered crucibles
were taken out and allowed to cool in a desiccator for an hour before
weighing. Volatile matter content of the sample was calculated using
Eq. (6) and Eq. (7) with mcs, 950 °C being the mass of the crucible
containing the sample after firing at 950 °C.
mcs;105 °C − mcs;950 °C
% Volatile Matter; VM ð%w=wÞ ¼  100 ð6Þ
ms
1 ðFRÞ
% Volatile Matter ðdry basisÞ; VMdb ð%w=wÞ ¼ VM   100 ð7Þ
100 − M
For the determination of ash content, uncovered crucibles contain-
ing the samples after determination of volatile matter content were
fired in a muffle furnace heated to 750 °C at a rate of 10 °C/min. The tem-
perature was maintained at 750 °C for 6 h and allowed to cool until the
furnace temperature was below 150 °C but above 100 °C. The crucibles
were transferred, covered and further cooled in a desiccator for an hour
before weighing. The ash content of the sample was then calculated
using Eq. (8) and Eq. (9).
mcs;750 °C − mc
% Ash ð%w=wÞ ¼  100 ð8Þ
ms
1
% Ash ðdry basisÞ; Ashdb ð%w=wÞ ¼ Ash   100 ð9Þ
100 − M
where mcs, 750 °C is the weight of the crucible containing the sample after
combustion at 750 °C. Meanwhile, the fixed carbon was determined by
difference using Eq. (10) and Eq. (11) for wet and dry basis, respectively.
% Fixed Carbon; FC ð%w=wÞ ¼ 100 − M − VM − Ash ð10Þ
1
% Fixed Carbon ðdry basisÞ; FC db ðw=wÞ ¼ FC   100 ð11Þ
100 − M
Higher heating value (HHV)
Higher heating value, on the other hand, was determined using an
adiabatic bomb calorimeter (Parr 1108, Moline, Illinois, USA) following
the procedures prescribed by Parr Instruments (Parr, 2007). The calo-
rimeter was calibrated (Cbomb = 2235 J/K) using benzoic acid as the
standard reference material. Nickel chromium fuse wire (Parr 45C10)
was used to aid in the ignition of the sample, given the heat of combus-
tion of the fuse wire (−ΔH ^ fuse ) to be 1400 cal/g (5852 J/g). The correc-
^ HNO ) was based on standard sodium
tion for the formation of acids (ΔH
3
carbonate (0.0709 N, 4.18 J/mL) titrations with no correction for sulfur.
The heating value determined is expressed as “as determined” basis
(HHV) and was calculated through Eq. (12); for shifting basis to dry
basis, Eq. (13) was used.
Higher Heating Value; HHV ðMJ=kg
Þ 
^ HNO − mfuse −ΔH
^ fuse
CΔT− V titrant ΔH 3
¼ ð12Þ
ms
1
HHV db ðMJ=kgÞ ¼ HHV   100 ð13Þ
100 − M
Where Vtitrant and mfuse are the volume of the titrant and the mass of
the fuse wire, respectively.
Combustion indices
Changes in combustion indices such as fuel ratio (FR), combustibility
index (CI), and volatile ignitability (VI) were determined to assess the
torrefaction performance and fuel quality. These indices were calculated
using the following equations (Eq. (14) to Eq. (16)):
FC db
Fuel Ratio; FR ¼ ð14Þ
VMdb
HHV db 1
Combustibility Index; CI ðMJ=kgÞ ¼  ð115 − Ashdb Þ  ð15Þ
105
 
HHV db − 0:338FC db
Volatile Ignitability; VI ðMJ=kgÞ ¼  100 ð16Þ
VMdb þ M db
Fuels having FR values ranging from 0.5–2.0 are commonly used in
coal-fired power plants (Makino & Tanno, 2015; Ohm et al., 2015).
Ignition and flame stability problems are encountered with fuels having
FR values N 2.0 (Engineering Foundation (U.S.). Conference (1993:
Solihull et al., 1994). For the CI, it is recommended to have a value of
23 MJ/kg (Ohm et al., 2015). Meanwhile, for the VI, it is recommended
that at the minimum, the volatile matter should have a specific HHV
of at least 14 MJ/kg (Asthana, 2015; Magasiner et al., 2001).
Equilibrium moisture (moisture reabsorption)
Biomass samples (~ 1 g) were placed in 30 mL pre-dried crucibles
and placed in an oven at 105 °C with a drying time of at least 3 h to
remove the inherent moisture. The dried samples were then weighed
and taken as the initial weight of the dry biomass. The crucibles contain-
ing the dry biomass samples were exposed to atmospheric conditions
and weighed at predetermined time intervals, 1 h intervals for the
first 12 h and every 12 h in the succeeding days, for a duration of
15 days. A sling psychrometer (Bacharach 12-7013, Pennsylvania,
USA) was used to monitor the relative humidity (RH). Measurements
for RH were carried out at the same time interval when the samples
were weighed. The weight increase was then taken as the moisture
reabsorbed by the biomass sample.
Torrefaction of sugarcane residues
Medium walled porcelain crucibles (100 mL) were pre-fired in a
furnace at 350 °C for an hour and were allowed to cool in a desiccator
for an hour before weighing to determine their empty dry weight.
Pre-fired and pre-weighed porcelain crucibles were filled up to 90% of
their capacity with ~ 17 g of SCL weighed to the nearest 0.1 mg. The
covered crucibles containing the samples were loaded into the muffle
furnace (Vecstar PF 3, Derbyshire, UK) at a maximum of 6 crucibles
per batch. The furnace was heated from room temperature to the
desired furnace temperatures (250, 300 and 350 °C) at a heating rate
of 10 °C/min. A designated set of crucibles were immediately taken
out from the furnace at the onset temperature, taking this set as samples
at t = 0. The crucibles containing the samples were allowed to cool in a
desiccator for an hour before weighing. Subsequent crucibles were
taken out from the furnace at an interval of 15 min over a torrefaction
period of 75 min. This torrefaction method is adopted from a related
work on torrefaction of sugarcane bagasse (Conag, Villahermosa,
Cabatingan, & Go, 2017).
 A.T. Conag et al. / Energy for Sustainable Development 42 (2018) 160–169
Cooled samples after torrefaction were stored in 20 mL (20
× 10−3 dm3) screw-capped glass sample bottles for later characteriza-
tion and analysis. Triplicate trials were carried out for each torrefaction
condition. Torrefied sugarcane leaves (TSCL) were designated as SCL-
TTT-ttt, to indicate the torrefaction temperature (T) and time (t). Re-
sponses to the torrefaction treatment, mass yield (YM), energy density
ratio (RED), and energy yield (YE) were determined using Eq. (17) to
Eq. (19).
 
100 − M TSCL
ðmcTSL −mc Þ
100 mTSCL
Mass Yield; Y M ¼ ¼
100 − MSCL mSCL
mS 
100
HHV db;TSCL
Energy Density Ratio; RED ¼
HHV db;SCL
Energy Yield; Y E ¼ Y M  RED
where mcTSCL is the mass of the crucible containing the TSCL, mc is the
mass of the pre-fired empty dry crucible, mTSCL is the mass of the
torrefied sugarcane leaves, and mSCL is the mass of the dry raw sugar-
cane leaves.
Statistical analysis
Statistical analysis (Pearson's correlation and ANOVA) were carried
out using Microsoft Excel data analysis tool. Pearson's correlation was
employed for initial assessment of possible correlation between the re-
sponses investigated, while ANOVA was carried out to identify varia-
tions in the responses resulting from different treatments. In cases
when significant differences were observed between treatments,
Duncan's multiple range test was carried out to identify the specific
treatment and responses resulting in such difference.
Results and discussion
Sugarcane leaves as received were analyzed to have a moisture con-
tent of 63.43 ± 4.45%. After drying to a moisture content of b 10% for
storage, the samples were milled and further characterized for its prox-
imate components in weight percent (Mdb = 7.05 ± 1.07%, VMdb =
72.08 ± 0.56%, FCdb = 18.23 ± 0.49%, Ashdb = 9.69 ± 0.49%) and higher
heating value (HHVdb = 16.35 ± 1.00 MJ/kg), and bulk densities (ρLb =
0.15 g/mL and ρTb = 0.23 g/mL). A more objective assessment of the fuel
characteristics to qualify their use in industrial combustion systems is
through the determination of fuel combustions indices (FR = 0.26 ±
0.01, CI = 62.65 ± 3.17 MJ/kg and VI = 12.77 ± 1.18 MJ/kg) and flow
indices, Hausner ratio (HR = 1.51 ± 0.03) and Carr compressibility
index (CCI = 33.67 ± 1.16%). In view of the combustion and flow indi-
ces, dried SCL without any treatment do not meet current industrial
standards (0.5 b FR b 2.0 (Makino & Tanno, 2015; Ohm et al., 2015),
12 MJ/kg b CI b 23 MJ/kg (Ohm et al., 2015), VI N 14 MJ/kg (Asthana,
2015; Magasiner et al., 2001), HR b 1.34 (Tannous et al., 2013), and
CCI b 25 (Tannous et al., 2013)). These however, could be improved
through torrefaction. For instance, combustion indices (FR, CI, and VI)
are highly influenced by the decrease in VM, where a decrease in VM
could potentially result in the improvement of the combustion indices
of SCL.
Proximate composition and equilibrium moisture of torrefied SCL
Torrefaction temperature and time are found to significantly
(p b 0.05) influence the proximate composition of the torrefied SCL
(TSCL). Sample images of the torrefied SCL and a summary of the prox-
imate composition are shown in Figs. 1 and 2, respectively. As could be
observed, the volatile matter content of the TSCL decreases with the
163
increase in both time and temperature. Correspondingly, an increase
in both FC and ash is observed as volatiles are driven off during the
torrefaction process. Unlike FC, which is preferred to be in larger quan-
tities as it contributes more to the heating value, the increase in ash con-
tent should be kept at a minimum. For coal fired furnaces, it has been
suggested that the ash content should be kept below 20% on an as re-
ceived basis (Thomas, 2013), this is to avoid frequent maintenance of
the furnace to remove the ash and to avoid sensible heat losses in the
ash formed, whereby lowering the overall furnace efficiency. In view
of this, torrefaction of SCL at 300 and 350 °C should not be torrefied
beyond 45 min.
ð17Þ
In principle, moisture should also be volatilized and driven off at the
torrefaction conditions set. However, as the torrefied residues are
cooled and stored, moisture is reabsorbed, thus resulting in moisture
contents ranging from ~ 0.8 to 7%w/w. It should be noted that, with
ð18Þ
prolonged torrefaction time and as temperature is increased, the
amount of moisture reabsorbed generally decreases (Fig. 2). To have a
ð19Þ better understanding of the moisture reabsorption ability of the raw
and torrefied SCL, dried samples were exposed to atmospheric condi-
tions (T = 25.7 ± 1.8 °C and RH = 73.9 ± 6.2%). Shown in Fig. 3 are
the equilibrium moisture and moisture reabsorption time profile. In
general, dried samples rapidly reabsorb moisture once exposed to at-
mospheric conditions for the first 24-hour period. Considering that the
relative humidity of the room where the samples were exposed to
changes over the period of 15 days, moisture tend to be desorbed and
reabsorbed, however, these are more pronounced in the raw samples
and samples heated to the temperature of 300 °C. Thus, equilibrium
moisture was estimated to be the average moisture reabsorbed after
the first 24 h of exposure. Raw SCL exposed to the atmosphere were
found to have an equilibrium moisture of 9.5 ± 1.4% by weight of the
available dry biomass but could go as high as ~12%w/w (Fig. 3). These
results further support claims on the hygroscopic nature of raw biomass
residues, which may potentially lead to microbial activities during
storage and jeopardize the efficiency of the combustion system.
For solid fuels, it is recommended that the moisture content should
be kept below 10% (Obernberger & Thek, 2010), as to avoid producing
wet flue gas, which could again result in sensible and latent heat losses
as well as posing corrosion risks to the combustion system (Bach &
Skreiberg, 2016). Thus, in view of equilibrium moisture, torrefaction
would best carried out at least at 250 °C for 45 min or at 300 °C for
15 min, to have an equilibrium moisture of b6%w/w even after exposure
for over 15 days to the atmosphere. The improved hydrophobicity of the
TSCL may be attributed to the removal of components having functional
group (− COOH and –OH) having a strong affinity with water. These
components are often released as organic acids, ketones, and aldehydes
upon decomposition and volatilization (Chen et al., 2015). Thus, the hy-
drophobicity of TSCL may be associated with the decrease in VM content
after torrefaction, which is found to be positively correlated with the
decrease in the equilibrium moisture content of SCL (r = 0.68).
Mass yields, mechanism and kinetics of torrefaction
Torrefaction of SCL resulted in the decrease in mass yield as the tem-
perature was increased (Fig. 4A) from 250 to 350 °C and as time was in-
creased from 0 to 75 min approaching a final yield of ~ 42%w/w. In a
separate study by Joshi et al. (Joshi, Di Marcello, Krishnamurthy, & De
Jong, 2015) who investigated the torrefaction of sugarcane bagasse
under oxidative condition, reported a total burnout of the biomass
being treated at temperatures over 270 °C with oxygen concentration
over 3%v/v in the purge gas. Although purge gas was not employed in
this work, the torrefaction process may well be assumed to be under
minimized oxidative atmosphere, considering that chars were still re-
covered even at torrefaction temperatures of 300 °C and 350 °C. The de-
crease in mass yield is often attributed to the volatilization of the VM.
This is supported by the fact that the maximum decrease in mass yield
by 0.58 g/g is similar to that of the maximum loss in VM of about
164 A.T. Conag et al. / Energy for Sustainable Development 42 (2018) 160–169

Raw SCL TSCL-250-000 TSCL-250-015 TSCL-250-030 TSCL-250-045 TSCL-250-060 TSCL-250-075

Raw SCL TSCL-300-000 TSCL-300-015 TSCL-300-030 TSCL-300-045 TSCL-300-060 TSCL-300-075

Raw SCL TSCL-350-000 TSCL-350-015 TSCL-350-030 TSCL-350-045 TSCL-350-060 TSCL-350-075

Fig. 1. Sample images of raw and torrefied sugarcane leaves from various torrefaction temperatures (250 to 350 °C) and times (0 to 75 min).

Fig. 2. Proximate composition of raw and torrefied sugarcane leaves from various torrefaction temperatures (250 to 350 °C) and times (0 to 75 min).
 A.T. Conag et al. / Energy for Sustainable Development 42 (2018) 160–169 165

Fig. 3. Equilibrium moisture (RH = 73.9% and T = 25.7 °C or 298.7 K) of torrefied sugarcane leaves (A) and moisture reabsorption time profile of SCL torrefied at 300 °C (B).

0.56 g/g. Torrefaction at 250 °C resulted in the recovery of ~57%w/w of
the VM initially present and decreases to as low as ~19%w/w at 350 °C
(Fig. 4B). Further decrease in VM yield below 0.14 w/w at 350 °C were
not observed even at prolonged torrefaction time. This could be at-
tributed to the fact that the volatiles left require higher temperatures
for volatilization. However, the loss in VM is not solely because of its
volatilization but also due to the formation of solid combustibles or
char. This is supported by the fact that FC yield increases (Fig. 4C)
by as much as 16% of what was initially available, leading to a maxi-
mum FC yield of 0.21 w/w being achieved at 300 °C with a
torrefaction time of at least 45 min. A similar phenomenon was
also observed by Chen et al. (Chen et al., 2014) during the

FC = 0.184 ±0.003 g/g native SCL

A C

VM = 0.720 ± 0.005 g/g native SCL

B D
Fig. 4. Yields in torrefaction of sugarcane leaves under various times (0 to 75 min) and temperatures (250 to 350 °C): Mass (A), Volatile matter (B), Fixed carbon (C) and Ash (D).
166 A.T. Conag et al. / Energy for Sustainable Development 42 (2018) 160–169

torrefaction of cotton straw and corn stalks at temperatures of 200 to Table 1

290 °C for 30 min. Kinetic parameters of SCL torrefaction under minimized oxidative atmosphere.

Most literatures (Mesa-Pérez et al., 2013; Ohm et al., 2015; Pimchuai Biomass/component T k r2 k0 E r2
et al., 2010; Soponpongpipat et al., 2015; Supramono et al., 2015) focus (°C) (min−1) (min−1) (kJ/mol)
on the proximate composition of the torrefied biomass rather than the Total Mass (kT) 254 0.011 0.97 749.27 48.61 0.95
yield of the components, thus attributing the changes of the composi- rM = − kYM 305 0.037 0.96
tion to the volatilization of the VM. However, a careful evaluation of 344 0.051 0.95
VMs (K′) 254 0.010 0.96 1045.34 50.44 0.95
the yields of the components, supports the fact that FC is formed as
rVMs = − K′YVMs 305 0.036 0.94
the volatiles are reduced. Thus, the devolatilization of SCL is not solely rVMs = − k1YVMs − k2YVMs
due to the volatilization of VM but as well as the formation of FC s – solid phase, g – gaseous 344 0.049 0.94
from the VM initially present, which may have resulted because of phase
possible polymerization and Millard reactions at high temperatures. VMg, k1 = (K′ − k2) 254 0.009 – 1278.96 51.63 0.95
rVMg = k1YVMs 305 0.034 –
As for ash, the yields were found to be constant (Fig. 4D) as the mineral 344 0.048 –
matter available in the biomass would not change nor volatilize at the FC (k2) 254 0.0007 0.99 0.39 27.59 0.92
torrefaction temperatures employed in this work. rFC = k2YVMs 305 0.0014 1.00
Yields and proximate composition at the onset of the torrefaction 344 0.0016 1.00
temperatures (250, 300 and 350 °C) are not significantly different
(p N 0.05) between the heated biomass, as well as to that of the
raw or native biomass. During the torrefaction process, assuming Energy density, yield and recovery during torrefaction
that part of the VM is volatilized and part of which reacted to form
the additional FC, the proposed reaction mechanism may adequately One important aspect in the torrefaction of biomass is to upgrade the
be described by a pseudo first order isothermal kinetics. The proposed raw biomass to have fuels of higher energy density. Shown in Fig. 6A are
reaction scheme/mechanism is presented in Fig. 5. the effects of temperature and time on the HHV of TSCL. As could be
A summary of the kinetic parameters is summarized in Table 1. observed, HHV increases with increasing torrefaction temperature and
The apparent activation energy in the torrefaction SCL is found to be time. Generally, prolonged torrefaction time beyond 60 min did not
about 48.6 kJ/mol, suggesting that the torrefaction process is non- result in further increase of the HHV, which is similar to the changes
spontaneous in nature and the rate increases with the increase in observed in the VM and FC yield. Analysis of the experimental data
temperature. Likewise, activation energy required for the formation of revealed a strong correlation between the increase in HHV with the
FC (27.6 kJ/mol) from VCM are lower than the volatilization of VM decrease in VM yield (r = −0.9534) and with the increase in FC yield
(51.6 kJ/mol). Determination of activation energy is often times depen- (r = 0.9560). This may be viewed as the loss of energy-lean compo-
dent on the method and approach and to some extent, the biomass nents and the retention and/or formation of energy-rich components
source (Munir, Daood, Nimmo, Cunliffe, & Gibbs, 2009). Moreover, the (Anukam, Mamphweli, Reddy, Okoh, & Meyer, 2015).
pre-exponential factors for the volatilization of VM is found to be It could be observed (Fig. 6A) that the maximum HHV (22 MJ/kg) is
1278.96 min−1 while the formation of VM to FC is about 0.39 min−1, achieved under a torrefaction temperature of 300 °C and increasing the
which implies that the rate of volatilization of VM is much faster than temperature to 350 °C did not result in the further increase of HHV. This
its conversion to FC. is consistent with the results of Soponpongpipat et al. (Soponpongpipat

Volatiles
Biomass Torrefied Biomass
(Liquid, Gaseous and Moisture)

Volatiles (Liquid and Gaseous)

Volatile Matter (Solid/Semi-Solid) Fixed Carbon

Legends:

Biomass Matrix Moisture Fixed-Carbon Volatile Matter Ash

Fig. 5. Proposed mechanism for torrefaction of sugarcane bagasse and leaves.

 A.T. Conag et al. / Energy for Sustainable Development 42 (2018) 160–169 167

A B
Fig. 6. Higher heating value (A) and energy yield (B) of torrefied sugarcane leaves from various torrefaction times (0 to 75 min) and temperatures (250 to 350 °C).

et al., 2015), who reported an HHV of 21.65 MJ/kg after torrefaction for
60 min at 280 °C under nitrogen atmosphere. Sugarcane leaves, after
torrefaction under 300 and 350 °C over a period of 30 to 75 min, pro-
duced TSCL having a HHV comparable to locally available charcoal.
Apart from the added value provided by torrefaction on the HHV
of TSCL, it should be taken into consideration that torrefaction as a pro-
cess is considered energy intensive because of the high temperatures re-
quired by the process. Moreover, the removal of energy lean material
would also mean a loss in the total energy initially available. Energy
yield relative to the native dry biomass (Fig. 6B) or energy recovery
decreases to as much as 60% upon torrefaction at 300 °C and 60 min
or at 350 °C and 30 min. However, proper design of the process could
potentially result in a self-sustaining process taking into consideration
the possibility of recovering the energy lean volatiles to serve as heating
medium and fuel for the torrefaction process.
Combustion indices of torrefied residues
To provide a more objective evaluation or assessment of the
torrefaction performance and product quality, derived quantities,
referred to as combustion indices were determined. As has been
discussed, the energy density increases with increased severity
(temperature and time) of torrefaction. The degree of increase in the
energy density could be assessed through the determination of energy
density ratio (RED). The RED of the torrefied SCL were calculated to
have values ranging from 1.0 to 1.38, respectively (Table 2). These
values suggest an increase in HHV of the raw SCL by as much as
38%. Although an increase in RED is observed as a result of torrefaction,
high RED may not necessarily be an absolute measure for the torrefaction
performance. Thus, changes in combustion indices of the torrefied
residues such as FR, CI, and VI (Table 2) were also looked into to further
assess the torrefaction performance and product quality.
Torrefied SCL were determined to have an FR ranging from 0.26 to
1.76, while VI from 11.6 MJ/kg to as much as 19.9 MJ/kg and CI from
11.2 to 65 MJ/kg. Generally, torrefaction at 250 to 350 °C resulted in
the increase in FR & VI, and a decrease in CI. These changes in the com-
bustion indices after torrefaction resulted in solid fuels acceptable for
use in industrial applications. Fuel Ratio, CI, and VI are strongly influ-
enced by the torrefaction conditions, temperature and time. This is
due to the fact that these indices are dependent on VM content of the
solid fuel, which decreases as torrefaction temperature and time in-
duces the volatilization of VM and its conversion to FC. However, post
statistical test/evaluation of the determined VI revealed that this mea-
sure is not significantly influenced by the torrefaction temperature

Table 2
Combustion indices of raw and torrefied sugarcane leaves#.

RED FR (kg/kg) CI (MJ/kg) VI (MJ/kg)

Sample
AVE SD AVE SD AVE SD AVE SD
Raw SCL 1.00 - 0.26 0.00 62.65 3.17 12.77 1.18
SCL-250-000 0.94s 0.03 0.26u 0.00 60.01hh 1.77 11.55 0.49
SCL-250-015 1.07s 0.09 0.27u 0.00 65.35hh 6.25 14.80 1.99
SCL-250-030 1.14s 0.09 0.33u 0.02 56.16gg,hh 7.65 15.99 2.31
SCL-250-045 1.15s 0.04 0.40u 0.01 46.49ff,gg 3.17 15.84 1.05
SCL-250-060 1.23t 0.06 0.46v 0.00 41.94ee 2.55 17.83 1.79
Torrefied Sugarcane Leaves

SCL-250-075 1.27t 0.08 0.51w 0.01 39.30ee 2.29 18.77 2.40

SCL-300-000 0.97s 0.05 0.25u 0.00 63.81hh 4.77 12.49 1.19
SCL-300-015 1.09s 0.02 0.33u 0.03 53.73gg 6.21 14.97 0.82
SCL-300-030 1.30t 0.04 0.63x 0.06 32.46ee 2.65 18.76 1.06
SCL-300-045 1.31t 0.08 1.02y 0.09 19.65dd 3.18 17.67 3.20
SCL-300-060 1.32t 0.07 1.69cc 0.02 11.28dd 0.69 16.06 4.04
SCL-300-075 1.38t 0.07 1.76cc 0.02 11.21dd 0.51 19.91 3.87
SCL-350-000 1.00s 0.04 0.27u 0.00 61.05hh 2.15 13.00 0.85
SCL-350-015 1.08s 0.04 0.38u 0.04 45.65ff 2.76 14.46 0.63
SCL-350-030 1.29t 0.03 1.19z 0.08 16.10dd 1.46 16.18 1.82
SCL-350-045 1.32t 0.04 1.30aa 0.08 15.00dd 0.56 17.34 1.59
SCL-350-060 1.35t 0.01 1.39aa 0.04 14.14dd 0.62 18.50 0.15
SCL-350-075 1.35t 0.03 1.50bb 0.03 13.27dd 0.25 17.54 1.47
168 A.T. Conag et al. / Energy for Sustainable Development 42 (2018) 160–169

and time. This may be attributed to the observed variances in the deter-
mined VI values, owing to the ability of the biomass to reabsorb
moisture. The change in VI may be a useful indicator for the quality of
the solid fuel. However, its change may not be an appropriate measure
for evaluating a given torrefaction performance as it is influenced by
the moisture content of the torrefied biomass during storage after
torrefaction. For SCL, torrefaction at a temperature of 300 °C and
torrefied for at least 45 min is sufficient to meet all three combustion
indices.
Bulk density and flowability of torrefied residues
The bulk density of SCL (both untapped and tapped) generally
decreased as the torrefaction time and temperature was increased
(Table 3). During torrefaction, the biomass becomes more porous as
volatiles are removed while leaving the biomass matrix intact (Fig. 5),
thus resulting in the decrease of its density. However, at torrefaction
time beyond 45 min, an increase in the bulk density was also observed,
which results from the collapse of the biomass matrix owing to the
weakened structure of the torrefied biomass (Phanpanich & Mani,
2011). However, the increase in bulk density at prolonged torrefaction
time is not high enough to facilitate ease of storage. This would entail
further processing like that of briquetting so to improve the density,
while providing a rigid structure for later storage. Nevertheless, the
physical characteristics may be taken advantage of, depending on its
later application. Density data of solid fuels may be used in the evalua-
tion of flow characteristics through flow indices like HR and CCI as
a measure of flowability and ease of compaction of the biomass
(Sarraguça et al., 2010). From the results of the torrefaction experi-
ments, there seem to be no apparent observable trend, however, a de-
crease in both measures is observed when compared to the raw SCL.
Values of HR and CCI N1.34 and 25, respectively, are considered to
have poor flowability but are easily compacted and would be favorable
for production of briquettes for domestic and household applications.
The opposite is true for solids having good flowability and would be
favorable when used in industrial fluidized bed combustion and pyroly-
sis chambers (Tannous et al., 2013). A good compromise between both
applications as solid fuel, are chars obtained after torrefaction at 300 °C
for 45 min.
Comparison of chars form torrefaction and pyrolysis of straws
Fast pyrolysis, is an alternative to thermal pretreatment to upgrade
biomass residues with bio-oil as its main product, however, in the pro-
cess, chars are also produced. Chars obtained from pyrolysis of SCL
(Mesa-Pérez et al., 2013) have higher HHV of 24.4 MJ/kg (Table 4),
but would require much higher processing temperature of 470 °C
and resulting in lower mass yields and higher ash content. Although
torrefaction may also be carried out at a slightly lower temperature of
280 °C for 60 min to achieve the same HHV and of lower ash content
(Soponpongpipat et al., 2015), its use as fuel mix or substitute for coal
would not be possible as it does not meet all criteria for combustion
indices of solid fuel, where its CI is ~ 39 MJ/kg. This further shows the
importance of using combustion indices as a means of assessing the
torrefaction performance and product quality and applicability.
Apart from sugarcane, rice and corn are among the top agricultural
produce in the world, which also leaves behind large quantities of resi-
dues upon harvest. Compared to straws from rice and corn having been
torrefied under similar conditions, TSCL results in higher mass yields
when compared to residues of either corn or rice. In view of HHV,
TSCL has a higher HHV when compared to that of rice (18.7 MJ/kg)
(Deng, Wang, Kuang, Zhang, & Luo, 2009), while lower when compared
to that of corn (23.6 MJ/kg) (Chen et al., 2014). Moreover, TSCL torrefied
at 300 °C for 45 min has relatively lower ash content and is below the
limits for ash content of solid fuels used in industrial applications.
Conclusions
Torrefaction of SCL at 300 °C, for a minimum for 45 min, is required
to meet the prescribed values for combustion indices and flowability
indices. Torrefied SCL reached an HHV value as high as ~ 22 MJ/kg.
Increasing the torrefaction temperature from 300 °C to 350 °C did not
result in a significant increase in the fuel quality of TSCL. After
torrefaction, 45% of the initial dry weight of SCL was recovered which
corresponds to an energy recovery of at least 60%. Torrefied SCL were
found to be less hygroscopic having an equilibrium moisture below
6 g/100 g available dry biomass. The bulk densities of the TSCL were
found to be suitable for briquetting and fluidized combustion applica-
tions with HR and CCI values of b1.34 and b25, respectively. The
activation energy for the devolatization process of SCL is found to

Table 3
Bulk density and flowability indices of raw and torrefied sugarcane leaves.#

ρLb (g/mL or 10-3 kg/m3) ρTb (g/mL or 10-3 kg/m3) HR CCI

Sample
AVE SD AVE SD AVE SD AVE SD
Raw SCL 0.154 0.003 0.233 0.001 1.508 0.027 33.667 1.155
SCL-250-000 0.165rr 0.003 0.222zz 0.004 1.345bb,* 0.010 25.667gg 0.577
SCL-250-015 0.144ll 0.002 0.200vv,ww 0.003 1.389cc 0.000 28.000hh 0.000
SCL-250-030 0.138kk 0.004 0.196uu 0.005 1.422ee 0.012 29.667jj,kk 0.577
SCL-250-045 0.138kk 0.003 0.192tt 0.003 1.395cc,dd 0.011 28.333hh,ii 0.577
SCL-250-060 0.140kk 0.003 0.195uu 0.004 1.389cc 0.000 28.000hh 0.000
Torrefied Sugarcane Leaves

SCL-250-075 0.155oo 0.003 0.209xx 0.003 1.351bb,cc 0.000 26.000gg,hh 0.000

SCL-300-000 0.152mm,nn 0.002 0.215yy,zz 0.006 1.416dd 0.031 29.333ii 1.528
SCL-300-015 0.145ll,mm 0.002 0.193tt,uu 0.001 1.327bb,* 0.010 24.667gg 0.577
SCL-300-030 0.161qq,rr 0.002 0.208xx 0.001 1.293aa 0.019 22.667ff 1.155
SCL-300-045 0.159pp 0.002 0.213yy 0.001 1.340bb 0.028 25.333gg 1.528
SCL-300-060 0.151mm 0.003 0.206ww,xx 0.004 1.364cc 0.022 26.667hh 1.155
SCL-300-075 0.154nn 0.000 0.210xx,yy 0.003 1.364cc 0.022 26.667hh 1.155
SCL-350-000 0.136jj 0.004 0.185ss 0.001 1.364cc 0.028 26.667hh 1.528
SCL-350-015 0.134jj 0.004 0.191tt 0.003 1.429ee 0.020 30.000kk 1.000
SCL-350-030 0.126jj 0.001 0.169ss 0.001 1.333bb 0.018 25.000gg 1.000
SCL-350-045 0.138kk 0.004 0.188tt 0.004 1.370cc 0.019 27.000hh 1.000
SCL-350-060 0.144kk 0.001 0.186ss 0.003 1.293aa 0.010 22.667ff 0.577
SCL-350-075 0.150mm 0.002 0.188tt 0.002 1.255aa 0.018 20.333ff 1.155
 A.T. Conag et al. / Energy for Sustainable Development 42 (2018) 160–169 169

Table 4
Comparison of chars from the torrefaction or pyrolysis of various agricultural straws/leaves.

Agricultural Thermal Temp (°C)/time Mass yield Moisture Fixed carbon Volatile matter Ash HHVa Ref.
straw/leaves treatment (min) (w/w) (%w/w) (%w/w) (%w/w) (%w/w) (MJ/kg)

Sugarcane Torrefaction 300/45 0.52 1.8 40.4 39.9 17.9 21.6 This study
Sugarcane Pyrolysis 470/− 0.46 1.9 30.1 36.4 33.8 24.4 (Mesa-Pérez et al., 2013)
Sugarcane Torrefaction 280/60 – – (31.7)a (58.4)a (9.9)a 21.6 (Soponpongpipat et al., 2015)
Rice Torrefaction 300/30 0.37 – – – – 18.7 (Deng et al., 2009)
Rice Torrefaction 300/60 0.50 – (28.1)a (34.4)a (37.4)a 16.6 (Park, Lee, Ryu, & Park, 2014)
Corn Torrefaction 290/30 0.31 3.3 41.2 35.3 20.5 23.6 (Chen et al., 2014)
a
Dry basis.

be ~50 kJ/mol with a pre-exponential factor of 749.27 min−1, based on
pseudo-first order kinetics. The decrease in the volatile matter after
torrefaction is not solely due to the volatilization of the volatiles but
also due to the formation of the additional char, as supported by the in-
crease in the fixed carbon yield. Combustion indices are found to be of
importance when assessing torrefaction performance as well as the
quality of the solid fuel. Compared to other torrefied agricultural straws,
TSCL provides certain advantages such as a higher mass yield, higher
HHV and lower ash content.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.esd.2017.11.004.
References
Anukam, A., Mamphweli, S., Reddy, P., Okoh, O., & Meyer, E. (2015). An investigation into
the impact of reaction temperature on various parameters during torrefaction of
sugarcane bagasse relevant to gasification. Journal of Chemistry, 2015. https://doi.
org/10.1155/2015/235163.
Asthana, A. K. (2015). Coal blending and its effects on boiler performance.
ASTM International (2013). ASTM standard D1762–84(2013) “standard test method for
chemical analysis of wood charcoal.”.
Bach, Q. V., & Skreiberg, O. (2016). Upgrading biomass fuels via wet torrefaction: A review
and comparison with dry torrefaction. Renewable and Sustainable Energy Reviews, 54,
665–677. https://doi.org/10.1016/j.rser.2015.10.014.
Bergman, P. C. A., Boersma, A. R., RWR, Zwart, & JHA, Kiel (2005). Torrefaction for biomass
co-firing in existing coal-fired power plants.
Chen, Y., Yang, H., Yang, Q., Hao, H., Zhu, B., & Chen, H. (2014). Torrefaction of agriculture
straws and its application on biomass pyrolysis poly-generation. Bioresource
Technology, 156, 70–77. https://doi.org/10.1016/j.biortech.2013.12.088.
Chen, D., Zheng, Z., Fu, K., Zeng, Z., Wang, J., & Lu, M. (2015). Torrefaction of biomass
stalk and its effect on the yield and quality of pyrolysis products. Fuel, 159, 27–32.
https://doi.org/10.1016/j.fuel.2015.06.078.
Conag, A. T., Villahermosa, J. E. R., Cabatingan, L. K., & Go, A. W. (2017). Energy densifica-
tion of sugarcane bagasse through torrefaction under minimized oxidative atmo-
sphere. Journal of Environmental Chemical Engineering, 5. https://doi.org/10.1016/j.
jece.2017.10.032.
Deng, J., Wang, G. Jun, Kuang, J. Hong, Zhang, Y. Liang, & Luo, Y. Hao (2009). Pretreatment
of agricultural residues for co-gasification via torrefaction. Journal of Analytical and
Applied Pyrolysis, 86, 331–337. https://doi.org/10.1016/j.jaap.2009.08.006.
Engineering Foundation (U.S.). Conference (1993 Solihull, E., Williamson, J., & Wigley, F.
(1994). The impact of ash deposition on coal fired plants: Proceedings of the engineering
foundation conference held at the St. John’s Swallow Hotel, Solihull, England, June 20–25,
1993. Taylor & Francis.
Food and Agricultural Organization of the United Nations Statistics Division, & FAOSTAT
(2014). Production/crops processed.
Hassuani, S. J., Leal, M. R. L. V., & de Carvalho Macedo, I. (Eds.). (2005). Biomass power
generation: Sugar cane bagasse and trash (1st ed.). Brazil: Programa das Nacoes Unidas
para o Desenvolvimento; Centro de Tecnologia Canavieira.
Joshi, Y., Di Marcello, M., Krishnamurthy, E., & De Jong, W. (2015). Packed-bed
torrefaction of bagasse under inert and oxygenated atmospheres. Energy and Fuels,
29, 5078–5087. https://doi.org/10.1021/acs.energyfuels.5b00779.
Kiel, J. H. A., Janssen, A. H. H., & Joshi, Y. (2015). Thermochemical conversion:
Torrefaction. In W. de Jong, & J. R. van Ommen (Eds.), Biomass as a sustain. Energy
source Futur. Fundam. Convers. Process (pp. 388–401). New Jersey: John Wiley &
Sons Inc.
Magasiner, N., van Alphen, C., Inkson, M., & Misplon, B. (2001). Characterising fuels for
biomass - coal fired co-generation. Proc Annu Congr - south African sugar Technol
Assoc, 1. (pp. 282–291), 282–291.
Makino, H., & Tanno, K. (2015). Coal combustion for power production. In M. R. Riazi, & R.
Gupta (Eds.), Coal prod. Process. Technol (pp. 263–265). CRC Press, Taylor and Francis
Group.
Mendoza, T. (2015). Enhancing crop residues recycling in the Philippine landscape. In S. S.
Muthu (Ed.), Environ. Implic. Recycl. Recycl. Prod (pp. 79–100). Springer. https://doi.
org/10.1007/978-981-287-643-0.
Mesa-Pérez, J. M., Rocha, J. D., Barbosa-Cortez, L. A., Penedo-Medina, M., Luengo, C. A., &
Cascarosa, E. (2013). Fast oxidative pyrolysis of sugar cane straw in a fluidized bed
reactor. Applied Thermal Engineering, 56, 167–175. https://doi.org/10.1016/j.
applthermaleng.2013.03.017.
Munir, S., Daood, S. S., Nimmo, W., Cunliffe, A. M., & Gibbs, B. M. (2009). Thermal analysis
and devolatilization kinetics of cotton stalk, sugar cane bagasse and shea meal under
nitrogen and air atmospheres. Bioresource Technology, 100, 1413–1418. https://doi.
org/10.1016/j.biortech.2008.07.065.
Obernberger, I., & Thek, G. (2010). Physio-chemical characterisation of raw materials and
pellets. Pellet Handb. Prod. Therm. Util. Biomass pellets (pp. 58). Earthscan Ltd.
Ohm, T. I., Chae, J. S., Kim, J. K., & Oh, S. C. (2015). Study on the characteristics of biomass
for co-combustion in coal power plant. Journal of Material Cycles and Waste
Management, 17, 249–257. https://doi.org/10.1007/s10163-014-0334-y.
Paes, L., & de Oliveira, M. A. (2005). Potential trash biomass of the sugar cane plant. In S.
Hassuani, M. Leal, & I. Macedo (Eds.), Biomass power gener. sugar cane bagasse trash
(pp. 19–23) (1st ed.). Brazil: Programa das Nacoes Unidas para o Desenvolvimento;
Centro de Tecnologia Canavieira.
Park, J., Lee, Y., Ryu, C., & Park, Y. K. (2014). Slow pyrolysis of rice straw: Analysis of
products properties, carbon and energy yields. Bioresource Technology, 155, 63–70.
https://doi.org/10.1016/j.biortech.2013.12.084.
Parr (2007). Operating instructions for the 1108 oxygen combustion bomb, 1–12.
Phanpanich, M., & Mani, S. (2011). Impact of torrefaction on grindability and fuel charac-
teristics of forest biomass. Bioresource Technology, 102, 1246–1253. https://doi.org/10.
1016/j.biortech.2010.08.028.
Pimchuai, A., Dutta, A., & Basu, P. (2010). Torrefaction of agriculture residue to enhance
combustible properties. Energy and Fuels, 24, 4638–4645. https://doi.org/10.1021/
ef901168f.
Sarraguça, M. C., Cruz, A. V., Soares, S. O., Amaral, H. R., Costa, P. C., & Lopes, J. A. (2010).
Determination of flow properties of pharmaceutical powders by near infrared
spectroscopy. Journal of Pharmaceutical and Biomedical Analysis, 52, 484–492.
https://doi.org/10.1016/j.jpba.2010.01.038.
Soponpongpipat, N., Sittikul, D., & Sae-Ueng, U. (2015). Higher heating value prediction of
torrefaction char produced from non-woody biomass. Frontiers in Energy, 9, 461–471.
https://doi.org/10.1007/s11708-015-0377-3.
Supramono, D., Devina, Y. M., & Tristantini, D. (2015). Effect of heating rate of torrefaction
of sugarcane bagasse on its physical characteristics. International Journal of
Technology, 6, 1084–1093. https://doi.org/10.14716/ijtech.v6i7.1771.
Tannous, K., Lam, P. S., Sokhansanj, S., & Grace, J. R. (2013). Physical properties for flow
characterization of ground biomass from Douglas-fir wood. Particulate Science and
Technology, 31, 291–300. https://doi.org/10.1080/02726351.2012.732676.
Thomas, L. (2013). Coal as a substance. Coal Geol (pp. 114) (2nd ed.). John Wiley & Sons Inc.