Renewable Energy 93 (2016) 691e699

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Characterization and productivity of cassava waste and its use as an

energy source
~o Paulo Soto Veiga a, 1, Teresa Losada Valle b, Jose
Joa  Carlos Feltran b,
Waldir Antonio Bizzo a, *
a
Faculty of Mechanical Engineering, University of Campinas e UNICAMP, 13083-970 Campinas, SP, Brazil
b
Agronomic Institute of Campinas e IAC, 13012-970 Campinas, SP, Brazil

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

Article history: This study sought to quantify and characterize cassava waste as fuel. The wastes from three cultivars
Received 15 May 2015 were collected to study and were divided into three distinct parts of the cassava plant: seed stem, thick
Received in revised form stalks, and thin stalks. Physical and chemical analyzes were carried out to determine the elemental
25 February 2016
composition of the waste: volatile matter; xed carbon; ash; moisture; lignin; cellulose; hemicellulose;
Accepted 27 February 2016
ash composition and higher heating value were determined. We conducted a thermogravimetric analysis
in oxidizing and inert atmospheres to study the behavior of the waste as fuel. The root productivity
obtained ranged from 7.7 to 13.0 t ha
1 yr
1 on a dry basis (db), and the ratio between waste and roots
Keywords:
Cassava
varied from 0.36 to 0.91. The physical and chemical properties of cassava waste are analogous to those of
Agricultural waste woody biomass regarding the elemental composition, the higher heating value, and thermogravimetric
Energy analysis. Ash content varied from 2.5% to 3.5%, reaching around 6.0% in samples unwashed. Approxi-
Biomass mately 60% of the ashes are alkali oxides, especially P2O5, K2O, and CaO, which have low melting points.
Biofuels The alkali index calculated suggests that there is a strong tendency that the combustion process leads to
ash fouling and the formation of ash deposits.
2016 Elsevier Ltd. All rights reserved.

1. Introduction
The world economy is characterized by a growing demand for
energy, which usually comes from non-renewable energy sources
(e.g., fossil fuels such as oil and coal), accounting for 81.1% of the
world's primary energy [28]. The use of such sources is assigned as
responsible for the increasing of CO2 concentration in the atmo-
sphere and the consequent negative impacts caused by climate
changes and global warming [29]. With the growth in research into
alternative energy sources, the use of biomass is an attractive op-
tion and has attracted considerable interest since the '90s as a way
of mitigating global climate change [11].
To avoid a conict of land use interests, agricultural wastes of
food crops has been the subject of research in order to determine
their potential for biomass to energy production, on a local scale
* Corresponding author.
E-mail addresses: jpsveiga@usp.br (J.P.S. Veiga), teresalv@iac.sp.gov.br
(T.L. Valle), feltran@iac.sp.gov.br (J.C. Feltran), bizzo@fem.unicamp.br (W.A. Bizzo).
1 ~o Paulo,
Current address: Luiz de Queiroz College of Agriculture, University of Sa
Piracicaba, SP, Brazil.
[20] and its characteristics as fuel for thermal processes [41,48].
These characteristics include the elemental composition and
heating value of the biomass and its behavior when subjected to
thermal treatment [13].
Several studies have determined the productivity and physical-
chemical characteristics of crop residues as corn stover and corn
cobs [25], cotton stalks [2,24] and wheat straw [17]. Literature data
on traditional residues is vast [16,32], but there is a lack of detailed
information about cassava (Manihot esculenta Crantz) waste.
Cassava is cultivated in the tropics to produce roots, used pri-
marily for human consumption but also for animal feed and the
extraction of starch. Cassava plants have a high biomass yield under
good environmental conditions and a great ability to adapt to en-
vironments with biotic and abiotic stresses, such as poor-fertility or
water-stressed soils where other cultures are not feasible. The
aerial part of the plant consists of thick and thin stalks, petioles and
leaves, and the underground part consists of roots and the seed
stem.
Seed stems are cassava stalks that were cut into approximately
200 mm-long sections and planted to reproduce the plant asexu-
ally, and after it grows it looks like woody biomass. The structure of

http://dx.doi.org/10.1016/j.renene.2016.02.078
0960-1481/ 2016 Elsevier Ltd. All rights reserved.
692 J.P.S. Veiga et al. / Renewable Energy 93 (2016) 691e699
the plant is shown in Fig. 1. Harvesting usually involves cutting the
stalks, uprooting the plant from the soil and removing the roots.
While the rst two operations can be done by mechanical equip-
ment, the seed stem and roots are separated manually as there is no
mechanical equipment available to carry out this operation.
After all the process of harvest, about half of the biomass pro-
duced is abandoned in the elds.
Worldwide, production of cassava in 2010 was approximately
230 million tons, with an average productivity of 12.4 t ha
1 on a
wet basis (wb), making it the ninth largest agricultural commodity
by weight produced. The world's largest producers are Nigeria (54
million tons per annum), Indonesia (24 million tons), Brazil (23.5
million tons) and Thailand (22.5 million tons) [19]. Cassava is
usually grown in tropical regions by low-income populations, to
meet their food needs, and any surplus is traded with small local or
regional businesses. Only a small percentage of world production is
traded internationally, and this trade is largely accounted by
Thailand, which serves the European and Asian markets.
Cassava, therefore, represents a major opportunity for
increasing productivity through technological innovation. Brazil
has a low average productivity of 13.8 t ha
1 [19], slightly higher
than the world's average (12.4 t ha
1).
This low yield is the result of a wide variety of production sys-
tems, from subsistence crops in highly stressful environments with
low productivity to technically highly advanced crops in favorable
environments with high productivities. Table 1 shows typical pro-
ductivity values in four regions of Sa ~o Paulo and other four regions
in Brazil, the largest cassava-producing region in Sa ~o Paulo has a
yield average twice higher than the national average. Improving
yield, biomass productivity can increase threefold among
producers.
In Brazil, the main use for the cassava root is for human con-
sumption. A negligible portion of the production of cassava is used
for ethanol production; however, we did not nd data available on
the amount of ethanol produced from cassava root in Brazil.
Authors that have studied cassava in the eld [39,42,50,57]
tested a range of production conditions including different vari-
eties, types of soil preparation, irrigation methods and resistance to
plant diseases. They carried out these works as a way to improve
Fig. 1. Cassava plant.
root yield, but little importance has been given to the productivity
of cassava crop residues.
To harness energy from agricultural waste, local and regional
parameters such as productivity and volume of waste produced
must be taken into consideration, especially in large countries and
environmentally diverse as Brazil. These factors are important
when planning and implementing projects like the Brazilian and
the United States projects of ethanol [23], the WBP/SIGAME uti-
lizing forest biomass [14], the UK government's target to generate
10% of the national electricity from renewable sources, being a
signicant part from biomass [37] and the EU Renewable Energy
Directive 2009/28 (RED) [55].
Since energy-production projects involve large amounts of in-
vestments, a project using cassava wastes, or any other biomass, as
an energy source, it cannot be justied by applying national pro-
ductivity because subsistence productions will always have an
economically prohibitive yield. Besides, to correctly assessing pro-
ductivity in the eld, the essential characteristics of cassava waste
as a feedstock for energy production must be determined in each
case.
Although it is not the main subject of this work, it is important
to point out that the root of the cassava has a high potential for
ethanol production, and can be used exclusively for fuel purposes,
and also there is the possibility to use the residues from cassava
starch and our industries. The production of ethanol from starch
involves cooking, liquefaction, saccharication, and then, the same
route as ethanol from saccharose: fermentation, and distillation [9].
Adelekan [1] recommends the production of ethanol from cas-
sava among the technologies to solve energy and environmental
problems in tropical regions of the world. In plantations with high
yield, 30 t ha
1 for example, it can be retrieved 6.2 m3 ha
1, this
value is comparable to the typical productivity of ethanol from
sugarcane, which is 7.0 m3 ha
1 [10].
The residues of the cassava industry can also be used to produce
bio-oils after pyrolysis or be treated before saccharication, opti-
mizing the process [59,60].
Few works in the literature discuss the use of cassava waste as a
source of energy. Antonio-Cisneros and Elizalde-Gonza lez [3]
separated rind, vascular system and pith from residues of cassava
stems and characterized it for activated carbon preparation. Pattiya
[45] characterized cassava waste to provide information to support
its use as fuel in Thailand, the author classied the stalks and seed
stem (called rhizome by Pattiya) as waste and characterized them
physically and chemically. Wei et al. [61] discuss the possibility to
extract starch from the branches to produce ethanol, evaluating
aspects as the region of production, varieties and time of harvest to
nd the better results.
Regions where there are large areas of cassava elds, its waste
can be used as a source of energy by root-processing mills, and, if
the investment is economically viable, a surplus of energy can be
sold, improving the prots. As the production around the world
increased 30% in the last decade [18], the potential to produce
energy with cassava wastes is also increasing.
The utilization of cassava wastes can also combine food and fuel
production, as described by Zhu et al. [63], reducing the competi-
tion between the land use for this two important issues in the
growing population of the world.
As cassava waste has the potential to make a signicant
contribution to Brazilian energy production because it is a wide-
spread crop in this country, the objective of the present work is to
characterize this type of waste as raw material for energy produc-
tion. Stalk and seed stem yields were assessed experimentally, and
both parts of the plant were characterized as raw material for
biofuel production.
 J.P.S. Veiga et al. / Renewable Energy 93 (2016) 691e699 693

Table 1
Cassava root productivity in various regions of Brazil.

Region Root productivity (t ha
1 yr
1) wb Dry matter (t ha
1 yr
1) Source
~o Paulo
Average in the state of Sa 15.0 4.8 IAE databasea
Latitude 1930' ~ 24300 S
Best producers, state of S~
ao Paulo 31.5 10.3 IAC databaseb
Latitude 1930' ~ 24300 S
Mogi-Mirim, SP 21.0 7.1 IAE databasea
Latitude 22260 S
Assis, SP 15.8 5.7 IAE databasea
Latitude 22300 S
Dourados, MS 33.3 8.3 [42]
Latitude 2213
Rio Branco, AC 40 13.2 [39]
Latitude 9580 S
ria da Conquista, BA
Vito 24.5 7.7 [50]
Latitude 14520
Araruna, PR 22.9 8.2 [57]
Latitude 23560
a
Institute of Agricultural Economics (SP) database, average for 2001e2010.
b
Variety IAC 14, 18e24 month harvest, Agronomic Institute of Campinas database (SP).

2. Material and methods
The experiment took place in the municipality of Echapora ~, SP,
at latitude 22150 , longitude 50 070 and a height of 700 m above sea
level. The region has medium-fertility acrisols predominance,
rainfall of 1375.8 mm per year, concentrated between October and
March, and is classied as Cwa in the Koppen system. It is a region
with high technological development in agriculture, and the pre-
dominantly cultivated crops are cassava, soybean, corn and
sugarcane.
Three cassava varieties were assessed, two of which are
commercially available (IAC 14 and IAC 90) and the third is a clone
still under development (clone 48/98). The varieties were selected
as they represent the most common ones used in industrial cassava
production and the clone 48/98 for its promising use soon as a
commercial variety. The main differences among the varieties
chosen are agronomic issues as the resistance of diseases and pests,
ease of sowing, cultivation and harvest.
Planting was performed on May 2009, using a four-row mech-
anized planter with a spacing of 0.90 m between rows and a within-
row spacing between plants of 0.90 m. We used a complete ran-
domized block design with three replications to arrange the vari-
eties on the experimental area, the total area planted was 2.5 ha.
Fertilizer was not used as the planting was done after soybean had
been harvested and the area had adequate levels of nutrients. The
cassava was harvested at three different phenological development
stages: 13 months after planting, when the plant has stopped
growing; 18 months after planting, when the plant is growing
again; and 26 months after planting, when it has stopped growing
again (after two growth cycles).
All the plants in two rows with 10 m in length in each block
were collected for the sampling, making a total of 19 specimens in
each sampling. During harvesting, the material was separated into
roots, seed stems and thick and thin stalks.
Each of these samples was weighed with a balance in the eld to
determine yield and washed to remove impurities (soil). Moisture
content was determined by measuring the weight difference after
drying a sample of each material in an oven at 60  C to constant
weight.
After the previous assessment, the three parts considered as
wastes (seed stem, thick stalks, and thin stalks) from the three
varieties were separated, making a total of 9 samples assessed. All
of them were comminuted in a knife mill with 20 mesh screen, and
the following analyzes were carried out.
Elemental analysis for carbon (C), hydrogen (H) and nitrogen (N)
was performed on a PerkineElmer 2400 CHNS Elemental Analyzer.
The sulfur (S) content was determined by plasma spectrometry and
optical analysis with a Sectro ICP-OES spectrometer, and chlorine
(Cl) content was established by titration according to [4]. The ox-
ygen (O) content was calculated based on the difference between
the results for the other elements. Proximate analysis was per-
formed according to ASTM E 1617-94 [7] and ASTM E 1755-01 [8] to
determine volatile-matter, xed-carbon, ash content, and moisture.
The ashes composition was analyzed by energy dispersive micro-
analysis (EDS) in a Zeiss EVO MA15 scanning electron microscope.
Higher heating value (HHV) was determined with a bomb calo-
rimeter according to ASTM D 2015 [6].
Extractive, lignin, cellulose and hemicellulose content were
determined according to TAPPI 264 cm-97 [51], TAPPI UM 250 [53]
and TAPPI 222 om-98 [52]. For these, analysis of carbohydrates was
performed by HPLC-PAD as described by Ref. [58].
Thermogravimetric analysis (TGA) was performed in a Netzsch
STA 409C to determine volatilization temperature, xed carbon,
and ignition temperature. All analysis were made with inert
(argonium) and oxidizing atmospheres [27] with a heating rate of
10 C min
1 and an atmosphere ow rate of 80 ml min
1. To study
the thermal behavior of cassava waste in industry, we prepared a
mix with all the plant parts to simulate the real waste that will be
used as fuel.
3. Results and discussion
3.1. Productivity
Cassava root and waste productivity are shown in Table 2 for the
different harvests and varieties studied. The average productivity of
the three varieties was 26.2 t ha
1 yr
1 (wb), slightly lower than the
value recorded by the best producers in the state of S~ ao Paulo but
higher than the values in the Mogi-Mirim and Assis regions (see
Table 1).
The waste-root ratio (db) ranged from 0.36 to 0.91 when the
three varieties and three harvests were taken into consideration.
The highest values were observed for IAC 14 (0.64e0.91), which
also showed the highest root productivity (7.7e13.0 t ha
1 yr
1 db).
The highest productivity (13.0 t ha
1 yr
1) was associated with a
waste-root ratio of 0.71. Root moisture at harvest ranged from
58.9% to 61.1% (wb), and the waste moisture ranged from 52% to
75%, the seed stems had the lowest water content.
694 J.P.S. Veiga et al. / Renewable Energy 93 (2016) 691e699

Table 2
Cassava root and waste productivity according to the time of harvest for the three varieties studied.

Variety Harvest Roots yield M Seed stem M Thick stalk M Thin stalk M Total waste Waste/Root Ratioa

Months t ha
1 yr
1 % t ha
1 yr
1 % t ha
1 yr
1 % t ha
1 yr
1 % t ha
1 yr
1

IAC 90 13 8.48 58.8 1.37 53.9 1.75 65.3 1.06 68.6 4.18 0.49
18 8.86 62.4 1.47 62.9 2.06 63.5 1.19 68.6 4.72 0.53
26 7.00 61.9 1.09 61.9 1.04 70.9 0.36 75.1 2.49 0.36
IAC 14 13 13.04 56.7 1.59 52.3 3.42 59.8 4.20 63.7 9.21 0.71
18 7.68 59.0 1.55 58.0 2.87 63.9 2.55 66.5 6.97 0.91
26 9.11 57.1 1.75 57.1 2.69 66.8 1.37 71.0 5.81 0.64
48/98 13 11.94 57.0 1.49 54.5 2.54 61.7 2.49 68.6 6.52 0.55
18 11.18 60.8 1.95 62.7 3.90 66.0 3.90 68.5 9.75 0.87
26 8.29 61.1 1.80 61.1 2.40 70.0 1.09 73.2 5.29 0.64

All productivities are given in dry basis.

M: moisture content at the time of harvesting (wet basis).
a
Waste considered: seed stem; thick stalk and thin stalk.

The highest values of root and waste productivity were observed Table 4
for IAC 14 at the 13-month harvest while for the other two varieties Elemental analysis of cassava waste for the different plant parts and the three va-
rieties studied.
the highest values were obtained at the 18-month harvest,
although these were only slightly higher than the values of the rst Months Variety Part of the plant Weight % (db)
harvest. Table 3 shows a comparison with typical productivity for C H Oa N S Cl
sugarcane, Brazil's most important energy crop. Sugarcane has a
13 48/98 Seed stem 43.08 6.58 49.32 1.01 <0.2 <0.3
dry matter yield about 50% higher than cassava, based on the IAC 14 18 IAC 90 42.91 6.50 49.97 0.61 <0.2 <0.3
yield. Although the processes involved in ethanol production using 26 IAC 14 44.60 6.52 48.08 0.80 <0.2 <0.3
these plants are slightly different because of the different feed- Average 43.53 6.53 49.13 0.80 <0.2 <0.3
stocks (sugars or starches), ethanol yield in both cases has the same
13 IAC 14 Thin stalk 44.62 6.36 48.47 0.55 <0.2 <0.3
order of magnitude. Since the Proa lcool program started in Brazil in
18 48/98 44.31 6.24 47.76 1.70 <0.2 <0.3
the 70s, intensive research has been carried out into genetically 26 IAC 90 45.77 6.60 46.81 0.82 <0.2 <0.3
improved sugarcane and new crop treatments, resulting in a
Average 44.89 6.40 47.69 1.02 <0.2 <0.3
growth in productivity from about 50 t ha
1 yr
1 in 1980 to the
current 85 t ha
1 yr
1, according to the 2012 AgriEnergy Statistical 13 IAC 90 Thick stalk 42.92 6.43 49.85 0.80 <0.2 <0.3
18 IAC 14 43.67 6.38 49.20 0.75 <0.2 <0.3
Yearbook [12]. 26 48/98 45.29 6.35 48.09 0.27 <0.2 <0.3

Average 43.95 6.39 49.05 0.61 <0.2 <0.3

3.2. Elemental analysis and chemical structures
Overall average 44.12 6.44 48.62 0.81 <0.2 <0.3
a
Table 4 shows the results of the elemental analysis for the By difference.
different harvests and varieties. Few variations were found among
the different parts of the plant except N, which was present in
greatest quantities in clone 48/98 (1.7% and 0.27% for thin and thick chlorine. At this point of view, cassava waste is an environmentally
stalks, respectively). In variety IAC 14, the concentration of N was friendly biofuel since it presents Cl and S contents lower than 0.3
between 0.55% and 0.80%. Knowledge of N content is necessary to and 0.2% respectively, avoiding emissions of dioxins, HCl, SOx and
estimate the formation of NO through the fuel-NO mechanism in particulate matter [40]. Chlorine can also be responsible for hot
waste combustion. Our results for C content (42%e44%) differ corrosion in superheaters and by the increase of fouling in the heat
slightly from those of [45]; who reported a C content about 51%. transfer surfaces, due to the formation of chlorides, along with
Dioxins can be formed in all combustion processes where sulfates, which vaporize and condense on cold surfaces later [5,38].
organic carbon, oxygen, and chlorine are present [35] and are very The small amount of S and Cl found are expected to be embedded in
hazardous to human health, and also, acid rain is usually a severe the ash, caused by the high contents of alkali and earth-alkali
problem due to fuel burning with high contents of sulfur and metals, thus xing the S and Cl, and avoiding problems with acid
rain [40].
Table 5 shows the chemical composition of the cassava waste
Table 3
Typical productivity gures for sugarcane and cassava in Brazil.
broken down into cellulose, hemicellulose, lignin, extractives and
minerals for the different parts of the plant and the different har-
Cassava rootsa Sugarcane vests. All the different types of waste have a high extractive content
t ha
1 yr
1 t ha
1 yr
1 (from 16% to 33%), and the cellulose content (33%e44%) is much
Productivity (wb) 30.1 85b higher than the hemicellulose content (around 10%e14%). These
Dry matter 13.1 36.0 values are useful for assessing the potential for ethanol production
Waste (db) 9.2 21.2c by hydrolysis followed by fermentation of the sugars formed.
Raw material for ethanol Production 9.8 13.17d
Ethanol production 4.9 5.36
a
Data from the present study for variety IAC 14 harvested at 13 months. 3.3. Proximate analysis and higher heating value
b
Mean productivity in Brazil in 2012 according to the 2012 Agrienergy Statistical
Yearbook [12].
c
Calculated based on productivity of 11% bers (bagasse) and 14% leaves and tops
Table 6 shows the results of the proximate analysis for each
in relation to stalk productivity (wet basis) [43]. plant part, the values shown are the averages of the values to the
d
Calculated based on productivity of 153 kg t
1 of sugarcane [34]. three different harvest periods. Volatile content (85%e87%) and
 J.P.S. Veiga et al. / Renewable Energy 93 (2016) 691e699 695

Table 5
Chemical composition of cassava waste for the different plant parts (average of the varieties IAC 14, IAC 90 and clone 48/98).

Weight % (db)

Harvest Months Part of the plant Cellulose Hemi cellulose Lignin Extractives Ash

13 Seed stem 38.09 10.47 16.08 33.23 2.13

18 37.15 10.46 15.82 33.27 3.30
26 44.55 14.26 21.73 16.79 2.69

Average 39.93 11.73 17.87 27.76 2.71

13 Thin stalk 40.67 13.50 23.81 19.62 2.41

18 33.72 10.28 24.38 29.03 2.60
26 38.62 11.54 19.64 26.61 3.59

Average 37.67 11.77 22.60 25.09 2.87

13 Thick stalk 38.60 10.96 18.94 28.53 2.97

18 41.97 14.32 24.41 17.06 2.23
26 41.64 11.14 16.86 26.87 3.49

Average 40.73 12.14 20.05 24.18 2.90

Overall average Overall average 39.45 11.88 20.17 25.68 2.83

Table 6
Proximate analysis for the three varieties of cassava studied (average values of the three waste parts).

Variety Ash(db) Volatile matter (db) Fixed carbon (db) Moisture (wb) High heat value

% MJ kg
1

IAC 90 3.40 85.64 10.96 9.55 16.59

IAC 14 2.50 86.75 10.74 9.58 16.30
48/98 2.83 85.22 11.96 9.73 18.75

xed carbon content (10%e12%) are similar to those in other woody
biomass. Ash content was approximately 2.5%e3.5%, contrasting
with the values of 4.0%e6.0% reported by Ref. [45]. In the present
study, seed stem samples that had not been cleaned off by washing
the dust of them had an ash content of 6.4%, which shows the
importance of cleaning to reduce ash content when using cassava
waste for combustion.
Higher heating value varied from 16.3 to 18.9 MJ kg
1 for the
three varieties studied, with the highest value observed for the seed
stem. These values differ from those reported by Ref. [45], who
reported HHVs of 17.6 MJ kg
1 and 23.4 MJ kg
1 for the seed stem
and stalk, respectively.
3.4. Ash analysis
Table 7 shows the results of the elemental analysis regarding
metal oxides of IAC 14 variety as it shows the major root and waste
yields, nonetheless, no signicant difference was found on the
other varieties.
Alkali oxides described in the literature as being responsible for
reducing the melting point of ash (K2O, CaO, MgO and Na2O)
[38,46] account for approximately 58%e67% of the total.
Some practical methods for evaluating ash fusibility character-
istics have been reported in the literature [36,56].
The alkali index is the ratio of potassium and sodium oxides to
the energy released in biomass combustion [30]:

Table 7
Elemental analysis of ash generated from cassava seed stem and thin and thick
 
stalks. The alkali index and the ratio B/A are also shown. AI 1=Q Yfa YKa2 O YNa
a
2 O

Oxide Seed stem (%) Thin stalks (%) Thick stalks (%) Average (%)

Na2O 0.10 0.49 0.25 0.31 where

MgO 15.55 11.64 14.12 13.46
Al2O3 3.41 2.91 0.88 2.37
SiO2 7.08 2.60 0.65 3.10 AI: amount of alkali oxide in the fuel per unit of fuel energy
P2O5 23.83 27.85 24.61 25.75 (kg GJ
1);
SO2

3 3.65 2.48 4.50 3.45 Q: HHV of the fuel (GJ kg
1);
ClO2 0.52 0.84 0.72 0.72 Yfa : mass fraction of ash in the fuel;
K2O 21.29 26.58 31.44 26.84
CaO 21.80 21.85 21.61 21.76
YKa2 O and YNa
a
2O
: mass fraction of potassium and sodium oxide in
TiO2 0.28 0.06 0.07 0.12 the ash, respectively.
MnO2 0.21 0.31 0.15 0.23
Fe2O3 1.66 1.65 0.67 1.33 Miles et al. [38] suggest that values from 0.17 kg GJ
1 to
Co3O4 0.09 0.04 0.02 0.05
0.34 kg GJ
1 increase the risk of fouling and above 0.34 kg GJ
1
NiO 0.02 0.01 0.02 0.02
CuO 0.28 0.39 0.17 0.29 fouling is almost certain to occur. Cassava waste ash has an AI be-
ZnO 0.08 0.07 0.07 0.07 tween 0.32 and 0.56 kg GJ
1, indicating that there is a high likeli-
MoO2 0.13 0.25 0.06 0.16 hood of fouling.
A.I. 0.32 0.48 0.56 Table 7 also includes an index representing the ratio of basic to
B/A 7.8 16.2 57.9
acid oxides in ash [47]:
696 J.P.S. Veiga et al. / Renewable Energy 93 (2016) 691e699

%Fe2 O3 CaO MgO Na2 O K2 O P2 O5

B=A
%SiO2 Al2 O3 TiO2
The index includes P2O5, since according to many authors, an
increase of phosphorus oxide also increases the tendency for
fouling to occur [46]. Kupka et al. [31] reported this tendency,
which is a consequence of the low melting point of phosphorus
oxide ash. It can be observed when the ratio B-A is between 0.75
and 2.0, and the tendency reduces when the ratio lies outside this
range. The values for B-A observed in our study are high above
those used in the correlation proposed by Ref. [31] based on
experimental data, which are a consequence of the large quantities
of alkali oxides present, and do not allow any conclusions to be
inferred.
Niu et al. [64] describe how the alkali metal aerosols, formed
during the biomass burning, condenses and form a sticky material
depositing on the heating surface and help the ying ashes forming
an alkali-induced slagging, reducing the heat transfer.
Kludze et al. [65] correlate agronomic practices to reduce slag- Fig. 3. DTG curves for the IAC 14 variety of cassava for oxidizing and inert atmospheres
(seed stem and thick and thin stalks in equal parts).
ging and fouling from herbaceous biomass used for combustion
which describes a reduction in the AI with small nitrogen fertilizer
applied as N is physiologically correlated with potassium, an
the samples (about 9% wb). Intense devolatilization starts at around
important alkali metal in both indexes (AI and B/A). Although no
238  C (inert atmosphere) and 226  C (oxidizing atmosphere), with
signicant differences in AI were found in this work among the
a peak at approximately 290  C to 300  C.
different varieties, time of harvest and plant parts, but other vari-
A second peak at 340  C was observed in the inert atmosphere,
eties of cassava and different fertilizers rates should be tested and
which [21] suggest may be the result of the onset of cellulose
lead to better results to use cassava wastes as fuel for combustion.
decomposition (345  C). Hemicellulose, in contrast, starts to
Thus, any possible tendency for fouling and corrosion caused by
decompose at lower temperatures (in this case around 300  C).
cassava waste ashes is an issue that requires further investigation.
Above 370  C, the rate of mass loss remains almost constant. This
The formation of deposits and fouling in combustion systems is
temperature can be attributed to the early formation of xed car-
particularly important on cooler surfaces such as heat transfer
bon. In the oxidizing atmosphere, xed-carbon combustion starts
surfaces in boilers and should be taken into consideration when
at 340  C, with constant mass loss until 618  C, and nishes at
using agricultural waste to generate energy by thermal processes.
648  C.
These values are similar to those found by Ref. [21] in an analysis
3.5. Thermogravimetric analysis
of the thermal decomposition of cane sugar bagasse, which, ac-
cording to [49]; can be explained by adding the behavior of the
Fig. 2 and Fig. 3 shows the results of TGA for a sample obtained
individual components and is more inuenced by the amount of
from the three parts of the cassava plant (seed stem and thick and
the components than by the form they take.
thin stalks, all in equal parts) in inert and oxidizing atmospheres.
The ignition point of the material can be determined by
The analysis shows that there is an initial peak of weight loss at
comparing the TGA curves in oxidizing and inert atmospheres and
80  C (in both atmospheres), which can be attributed to water loss.
observing when the rate of mass loss in an oxidizing atmosphere
At this temperature, the samples lost about 8e10% of their mass as
becomes higher than in an inert atmosphere [54]. However, this
water, a result that coincides with the moisture measurements of
point could not be clearly identied in Fig. 2, which shows the TGA
curves in both atmospheres. In contrast, the point at which xed-
carbon combustion begins (about 370  C), can be observed when
the mass loss curve in the inert atmosphere stabilizes while the rate
of mass loss in the oxidizing atmosphere remains unchanged in its
behavior.
Fig. 3, which shows the DTG curves of the same sample in the
inert and oxidizing atmospheres, shows a slightly higher increase
in the rate of mass loss in an oxidizing atmosphere compared with
the inert atmosphere, approximately at 235  C, which may be the
temperature at which ignition took place.
This temperature is close to that found for wheat (220  C) by
Ref. [26], who argued that the high volatile-matter content con-
tributes to a reduction in the ignition temperature of the material.
The low ignition temperature observed here may also be a result of
the plant high potassium content and the form in which the po-
tassium is present, as it acts as a catalyst not only for xed-carbon
combustion but also for the ignition of volatile matter [33].

3.6. Cassava waste in comparison with other biomasses

Fig. 2. TGA curves for the IAC 14 variety of cassava for oxidizing and inert atmospheres
(seed stem and thick and thin stalks in equal parts). Table 8 presents the characterization of cassava waste in
 J.P.S. Veiga et al. / Renewable Energy 93 (2016) 691e699 697

Table 8
Characterization of cassava waste in comparison of other typical biomasses.

Element Cassava wastea Cassava wasteb Sugarcane bagassec Rice strawd Yard wasted Switch-grassd Wheat strawd Eucalyptuse

Elemental composition % (db)

C 44.12 48.66 40.34 38.24 41.54 46.68 44.92 50.15
H 6.44 6.33 5.66 5.20 4.79 5.82 5.46 7.45
O 48.62 39.20 47.91 36.26 31.91 37.38 41.77 39.64
N 0.81 0.81 0.58 0.87 0.85 0.77 0.44 0.50
S <0.2 <0.1 0.17 0.18 0.24 0.19 0.16 0.02
Cl <0.3 <0.1 0.26 0.58 0.30 0.19 0.23 0.55
Proximate analysis % (db)
Moisture 9.62 11.93 e e e e e 6.50
Volatile matter 85.87 78.83 82.35 65.47 66.04 76.69 75.27 67.75
Fixed carbon 11.22 16.15 13.68 15.86 13.59 14.34 17.71 20.19
Ash 2.91 5.03 3.97 18.67 20.37 8.97 7.02 5.55
Higher Heating Value (MJ kg
1)
HHV 17.21 19.71 17.40 15.09 16.30 18.06 17.94 19.33
Ashes composition analysis % (db)
SiO2 3.10 e 41.91 74.67 59.65 65.18 55.32 0.09
Al2O3 2.37 e 8.32 1.04 3.06 4.51 1.88 3.77
TiO2 0.12 e 1.37 0.09 0.32 0.24 0.08 e
Fe2O3 1.33 e 4.85 0.85 1.97 2.03 0.73 10.34
CaO 21.76 31.55 16.95 3.01 23.75 5.60 6.14 46.79
MgO 13.46 13.36 5.60 1.75 2.15 3.00 1.06 5.00
Na2O 0.31 0.40 0.37 0.96 1.00 0.58 1.71 5.81
K2O 26.84 32.12 11.09 12.30 2.96 11.60 25.60 23.68
SO2

3 3.45 e 3.15 1.24 2.44 0.44 4.40 e
P2O5 25.75 18.27 3.80 1.41 1.97 4.50 1.26 4.53
ClO2 0.72 4.29 e e e e e e
MnO2 0.23 e 0.61 e e e e e
a
This work.
b
Ref. [45].
c
Ref. [66].
d
Ref. [30].
e
Ref. [15].

comparison with other biomass usually used as fuel, such as
bagasse, rice straw, and eucalyptus. In general, the characteristics of
the cassava waste resemble most of other biomasses, considering
elemental analysis, the higher heating value, and the volatile con-
tent. The main differences, even among the other biomasses pre-
sented, are in the ash content and the composition of the elements
from the ashes.
Cassava waste present 3e5% of ash, the same order of magni-
tude, for example, than eucalyptus (5.5%), but lower than rice straw
and yard waste. Cassava waste presented much lower silica content
than most biomass presented with the exception of eucalyptus.
However, cassava waste presented levels of potassium and phos-
phorus oxides larger than most of the biomass in comparison.
In general, its properties are very similar to the characteristics of
the eucalyptus, with the exception of the highest content of
phosphorus in the ashes. Of course, this comparison should not be
generalized, because changes in ash content and its composition
occurring among the same type of biomass depends on several
conditions such as variety, climate, plantation and fertilizing
method, and harvesting.
The results presented showed there is possible to have high root
and aerial biomass yields at the same time, despite the usual
genetical enhancement looks only for an improvement of root and
starch yields. Although the branches can be used to provide protein
and energy to feed livestock [62], cassava waste showed appro-
priate characteristics as a biofuel through thermal process like
combustion, gasication and pyrolysis, even with a relatively high
content of ashes. This kind of problem is already faced in traditional
energetic biomasses like wheat and rice straw or sugarcane
bagasse.
An important point to use cassava waste as fuel is its moisture,
which is inadequate at the time of harvest. However, it is similar to
most of the crop wastes just after the harvest. Dry this waste is a
key point and must be deeply analyzed. One way to dry this
biomass is ineld like already tested in crops like Arundo donax
[44]. The physical form of branches has the undesirable aspect of a
lower mass transfer area than rice or wheat straw, on the other side
it is possible to have a better aeration for drying and a lower ten-
dency to absorb rainwater after it is dry.
4. Conclusions
This study measured the productivity of three types of cassava
root and waste in a eld experiment. The best root productivity
obtained was 13 t ha
1 yr
1 (db), and the waste-root ratio varied
from 0.64 to 0.91 for IAC 14 harvested after 12 months. The mois-
ture content of the harvested product was about 60% for root and
52%e75% for eld waste (wb). Total production of dry matter from
cassava is still about 50% lower than the average values for sugar-
cane, the most important energy crop in Brazil.
The physical and chemical characteristics of cassava waste are
similar to those of woody biomass regarding HHV, proximate
analysis, elemental composition, and TGA. Ash content varied from
2.5% to 3.5% and was as high as 6.0% for unwashed samples.
Elemental analysis revealed that about 60% of the ash consisted of
alkali oxides, especially P2O5, K2O, and CaO, which have low
melting points. The alkali index calculated was indicative of a
strong tendency for ash fouling to occur during combustion.
In Brazil, cassava waste has potential for use as an energy source,
but some measures should be taken when used as fuel, to avoid the
high occurrence of incrustation due to the characteristics of the
ashes from this waste. Also, by the reason of the high moisture
content presented at the time of harvest, a drying process must be
provided before the waste is used as fuel. Other regions of cassava
production need an evaluation as it was developed in this study to
provide the right information about agronomic, chemical and
698 J.P.S. Veiga et al. / Renewable Energy 93 (2016) 691e699
physical characteristics of the cassava waste as a source of fuel.
Not counting ethanol, wood and sugarcane bagasse biomasses
already represent 17% of the Brazilian energy consumption [67],
thus, another kind of biomasses can be encouraged to be used as
modern biomass [22]. This measure would develop not only the
sugarcane sector but also many other industrial sectors with
different sources of agricultural wastes, typical of each region,
helping to improve the sustainability of the energy system and
avoiding uncertainties of the fossil-fuel market.
The use of cassava agricultural waste can increase the viability of
cassava root as a raw material for ethanol and food production in
integrated units with electricity cogeneration as it has been done in
Brazil to the culture of sugarcane. This study proved cassava is a
versatile crop and can provide food, ethanol, and thermal energy.
Acknowledgments
The authors would like to thank CAPES and CNPq, Brazilian
agencies for research funding.
References
[1] B.A. Adelekan, Investigation of ethanol productivity of cassava crop as a sus-
tainable source of biofuel in tropical countries, Afr. J. Biotechnol. 9 (35) (2010).
[2] R.C. Akdeniz, M. Acaroglu, A. Hepbasli, Cotton stalk as a potential energy
source, Energy Sources 26 (1) (2004) 65e75, http://dx.doi.org/10.1080/
00908310490251873.
[3] C.M. Antonio-Cisneros, M.P. Elizalde-Gonza lez, Characterization of manihot
residues and preparation of activated carbon, Biomass Bioenergy 34 (2010)
389e395, http://dx.doi.org/10.1016/j.biombioe.2009.12.001.
[4] APHA, American Public Health Association, Standard Methods for the Exam-
ination of Water and Wastewater, Method 4500-Cl B, twentieth ed., Water
Environment Federation, American Water Works Association, Washington DC,
USA, 1999.
[5] S. Arvelakis, H. Gehrmann, M. Beckman, E. Koukios, Studying the ash behav-
iour of agricultural residues using thermal analysis, J. Therm. Analysis Calorim.
72 (2003) 1019e1030.
[6] ASTM D2015-00 Standard Test Method for Gross Caloric Value of Coal and
Coke by the Adiabatic Bomb Calorimeter.
[7] E1617-09 Standard Practice for Reporting Particle Size Characterization Data.
Book of Standards Volume: 14.02.
[8] ASTM E1755-01-Standard Test Method for Ash in Biomass. Book of Standards
Volume: 11.06.
[9] P. Atthasampunna, P. Somchai, A. Eur-Aree, S. Artjariyasripong, Production of
fuel ethanol from cassava, MIRCEN J. 3 (2) (1987) 135e142.
[10] M. Balat, H. Balat, Recent trends in global production and utilization of bio-
ethanol fuel, Appl. Energy 86 (11) (2009) 2273e2282, http://dx.doi.org/
10.1016/j.apenergy.2009.03.015.
[11] G. Berndes, M. Hoogwijk, R. van den Broek, The contribution of biomass in the
future of global energy supply: a review of 17 studies, Biomass Bioenergy 25
(2003) 1e28, http://dx.doi.org/10.1016/S0961-9534(02)00185-X.
[12] Brazil, Ministe rio da Agricultura, Pecu aria e Abastecimento, in: Statistical
Yearbook of Agrienergy 2012, 2013 (Braslia).
[13] K.B. Cantrell, J.H. Martin II, K.S. Ro, Application of thermogravimetric analysis
for the proximate analysis of livestock wastes, J. ASTM Int. 7 (3) (2010) 1e13.
[14] E. Carpentieri, A. Silva, WBP/SIGAME the Brazilian BIG-GT demonstration
project actual status and perspectives, Biomass Bioenergy 15 (3) (1998)
229e232.
[15] L. Cuiping, W. Chuangzhi, H. Haitao, Chemical elemental characteristics of
biomass fuels in China, Biomass Bioenergy 27 (2) (2004) 119e130, http://
dx.doi.org/10.1016/j.biombioe.2004.01.002.
[16] C. Di Blasi, V. Tanzi, M. Lanzetta, A study on the production of agricultural
residues in Italy, Biomass Bioenergy 12 (5) (1997) 321e331, http://dx.doi.org/
10.1016/S0961-9534(96)00073-6.
[17] E. Donaldson, W.F. Schillinger, S.M. Dong, Straw production and grain yield
relationships in winter wheat, Crop Sci. 41 (2001) 100e106, http://dx.doi.org/
10.2135/cropsci2001.411100x.
[18] FAO e Food and Agriculture Organization of the United Nations, FAO Statis-
tical Database, 1990e2011, 2012 (accessed 28.11.12), http://www.fao.org/.
[19] FAO e Food and Agriculture Organization of the United Nations, FAO Statis-
tical Database, 2013 (accessed 15.09.15), http://faostat.fao.org/.
[20] V. Ferreira-Leita ~o, L.M.F. Gottschalk, M.A. Ferrara, A.L. Nepomuceno,
H.B.C. Molinari, E.P.S. Bon, Biomass residues in Brazil: availability and po-
tential uses, Waste Biomass Valoriz. 1 (1) (2010) 65e76, http://dx.doi.org/
10.1007/s12649-010-9008-8.
[21] M. Garcia-Perez, A. Chaala, J. Yang, C. Roy, Co-pyrolysis of sugarcane bagasse
with petroleum residue. Part I: thermogravimetric analysis, Fuel 80 (9) (2001)
1245e1258, http://dx.doi.org/10.1016/S0016-2361(00)00215-5.
[22] J. Goldemberg, S.T. Coelho, Renewable energydtraditional biomass vs. mod-
ern biomass, Energy Policy 32 (6) (2004) 711e714.
[23] J. Goldemberg, S.T. Coelho, P. Guardabassi, The sustainability of ethanol pro-
duction from sugarcane, Energy Policy 36 (6) (2008) 2086e2097.
[24] R.S. Gomes, P.N. Wilosn, W.E. Coates, R.W. Fox, Cotton (gossypium) plant
residue for industrial fuel. An economic assessment, Indus. Crops Prod. 7
(1997) 1e8, http://dx.doi.org/10.1016/S0926-6690(97)00006-X.
[25] R.L. Graham, R. Nelson, J. Sheehan, R.D. Perlack, L.L. Wright, Current and po-
tential U.S. corn stover supplies, Agron. J. 99 (2007) 1e11, http://dx.doi.org/
10.2134/agronj2005.0222.
[26] T. Grotkjr, K. Dam-Johansen, A.D. Jensen, P. Glarborg, An experimental study
of biomass ignition, Fuel 82 (7) (2003) 825e833, http://dx.doi.org/10.1016/
S0016-2361(02)00369-1.
[27] J.M. Heikkinen, J.C. Hordijk, W. De Jong, H. Spliethoff, Thermogravimetry as a
tool to classify waste components to be used for energy generation, J. Anal.
Appl. Pyrolysis 71 (2) (2004) 883e900.
[28] IEA e International Energy Agency, World Energy Outlook 2012, OECD/IEA,
Paris, 2012.
[29] IPCC, Intergovernmental Panel on Climate Change 2007: an Assessment of the
IPCC, 2007, pp. 12e17.
[30] B. Jenkins, Combustion properties of biomass, Fuel Process. Technol. 54 (1e3)
(1998) 17e46, http://dx.doi.org/10.1016/S0378-3820(97)00059-3.
[31] T. Kupka, M. Mancini, M. Irmer, R. Weber, Investigation of ash deposit for-
mation during co-ring of coal with sewage sludge, saw-dust and refuse
derived fuel, Fuel 87 (12) (2008) 2824e2837, http://dx.doi.org/10.1016/
j.fuel.2008.01.024.
[32] R. Lal, World crop residues production and implications of its use as a biofuel,
Environ. Int. 31 (2005) 575e584, http://dx.doi.org/10.1016/
j.envint.2004.09.005.
[33] T. Lang, R.H. Hurt, Char combustion reactivities for a suite of diverse solid fuels
and char-forming organic model compounds, Proc. Combust. Inst. 29 (2002)
423e431, http://dx.doi.org/10.1016/S1540-7489(02)80056-6.
[34] M.G.D.P. Lavanholi, Quality of sugarcane as feedstock for the production of
sugar and alcohol, chap. 23, in: L.L. Dinardo-Miranda, A.C.M. Vasconcelos,
M.G.A. Landell (Eds.), Sugarcane, Agronomic Institute, Campinas, SP, 2010 (in
Portuguese).
[35] E.D. Lavric, A.A. Konnov, J. De Ruyck, Dioxin levels in wood combustionda
review, Biomass Bioenergy 26 (2) (2004) 115e145, http://dx.doi.org/10.1016/
S0961-9534(03)00104-1.
[36] M.J.F. Llorente, J.E.C. Garcia, Comparing methods for predicting the sintering
of biomass ash in combustion, Fuel 84 (14e15) (2005) 1893e1900, http://
dx.doi.org/10.1016/j.fuel.2005.04.010.
[37] P. McKendry, Energy production from biomass (part 1): overview of biomass,
Bioresour. Technol. 83 (1) (2002) 37e46.
[38] T.R. Miles, J.R. Miles, L.L. Baxter, R.W. Bryers, B.M. Jenkins, L.L. Oden, Boiler
deposits from ring biomass fuels, Biomass Bioenergy 10 (2) (1996) 125e138,
http://dx.doi.org/10.1016/0961-9534(95)00067-4.
[39] G.M. Moura, N.L. Costa, Effect of frequency and height of pruning on root and
shoot productivity in cassava (in Portuguese with abstract in English), Pesqui.
Agropecua ria Bras. 36 (8) (2001) 1053e1059, http://dx.doi.org/10.1590/
S0100-204X2001000800007.
[40] I. Obernberger, T. Brunner, G. Ba rnthaler, Chemical properties of solid bio-
fuelsdsignicance and impact, Biomass Bioenergy 30 (11) (2006) 973e982,
http://dx.doi.org/10.1016/j.biombioe.2006.06.011.
[41] T.E. Odetoye, K.R. Onifade, M.S. Abubakar, J.O. Titiloye, Thermochemical
characterisation of parinari polyandra Benth fruit shell, Indus. Crops Prod. 44
(2013) 62e66, http://dx.doi.org/10.1016/j.indcrop.2012.10.013.
[42] A.A. Otsubo, F.M. Mercante, R.F. Silva, C.D. Borges, Soil tillage systems, cover
crops and productivity in cassava, Pesqui. Agropecua ria Bras. 43 (3) (2008)
327e332, http://dx.doi.org/10.1590/S0100-204X2008000300006 (in Portu-
guese with abstract in English).
[43] L.A.D. Paes, M.A. Oliveira, Potential trash biomass of the sugar cane plant, in:
S.J. Hassuani, M.R.L.V. Leal, I.C. Macedo (Eds.), Biomass Power Generation,
PNUD-CTC, Piracicaba, 2005, p. 216.
[44] L. Pari, A. Scarfone, E. Santangelo, S. Figorilli, S. Crognale, M. Petruccioli,
A. Suardi, F. Gallucci, M. Barontini, Alternative storage systems of Arundo
donax L. and characterization of the stored biomass, Indus. Crops Prod. 75
(2015) 59e65, http://dx.doi.org/10.1016/j.indcrop.2015.04.018.
[45] A. Pattyia, Thermochemical characterization of agricultural wastes from Thai
cassava plantations, Energy Sources, Part A Recover. Util. Environ. Eff. 33 (8)
(2011) 691e701, http://dx.doi.org/10.1080/15567030903228922.
[46] M. Pronobis, Evaluation of the inuence of biomass co-combustion on boiler
furnace slagging by means of fusibility correlations, Biomass Bioenergy 28 (4)
(2005) 375e383, http://dx.doi.org/10.1016/j.biombioe.2004.11.003.
[47] M. Pronobis, The inuence of biomass co-combustion on boiler fouling and
efciency, Fuel 85 (4) (2006) 474e480, http://dx.doi.org/10.1016/
j.fuel.2005.08.015.
[48] T.P. Protasio, L. Bufalino, G.H.D. Tonoli, J.R.M. Guimar~ aes, P.F. Trugilho,
L.M. Mendes, Brazilian Lignocellulosic wastes for bioenergy production:
characterization and comparison with fossil fuels, BioResources 8 (1) (2013)
1166e1185.
[49] K. Raveendran, A. Ganesh, K.C. Khilar, Pyrolysis characteristics of biomass and
biomass components, Fuel 75 (8) (1996) 987e998, http://dx.doi.org/10.1016/
0016-2361(96)00030-0.
[50] M.J.L. Souza, A.E.S. Viana, S.N. Matsumoto, R.C. Vasconcelos, T. Sediyama,
 J.P.S. Veiga et al. / Renewable Energy 93 (2016) 691e699
O.M. Morais, Cassava agronomical characteristics related to interaction among
irrigation, harvest time and mepiquat chloride (in Portuguese with abstract in
English), Acta Sci. Agron. 32 (1) (2010) 45e53, http://dx.doi.org/10.4025/
actasciagron.v32i1.720.
[51] TAPPI 264 cm-97, Preparation of Wood for Chemical Analysis, TAPPI PRESS,
Atlanta, 1988.
[52] TAPPI 222 om-02, Acid-insoluble Lignin in Wood and Pulp, TAPPI PRESS,
Atlanta, 1988.
[53] TAPPI UM 250, Acid-soluble Lignin in Wood Pulps, TAPPI PRESS, Atlanta, 1985.
[54] L. Tognotti, A. Malotti, L. Petarca, S. Zanelli, Measurement of ignition tem-
perature of coal particles technique using a thermogravimetric technique,
Combust. Sci. Technol. 44 (1985) 15e28, http://dx.doi.org/10.1080/
00102208508960290.
[55] Europa Union, Directive 2009/28/EC of the European Parliament and of the
Council of 23 April 2009 on the Promotion of the Use of Energy from
Renewable Sources and Amending and Subsequently Repealing Directives
2001/77/EC and 2003/30/EC, 2009.
[56] D. Vamvuka, D. Zografos, Predicting the behaviour of ash from agricultural
wastes during combustion, Fuel 83 (14e15) (2004) 2051e2057, http://
dx.doi.org/10.1016/j.fuel.2004.04.012.
[57] P.S. Vidigal Filho, M.G. Pequeno, C.A. Scapim, M.C.G. Vidigal, R.R. Maia,
E. Sagrilo, G.A. Simon, R.S. Lima, Evaluation of cassava cultivars in northwest
region of Parana  State, Brazil, Bragantia 59 (1) (2000) 69e75, http://
dx.doi.org/10.1590/S0006-87052000000100011.
[58] A.F.A. Wallis, R.H. Wearne, P.J. Wright, Chemical analysis of polysaccharides in
plantation eucalypt woods and pulps, Appita J. 49 (4) (1996) 258e262.
[59] P. Weerachanchai, S.S.J. Leong, M.W. Chang, C.B. Ching, J.M. Lee, Improvement
of biomass properties by pretreatment with ionic liquids for bioconversion
699
process, Bioresour. Technol. 111 (2012) 453e459, http://dx.doi.org/10.1016/
j.biortech.2012.02.02.
[60] P. Weerachanchai, C. Tangsathitkulchai, M. Tangsathitkulchai, Characteriza-
tion of products from slow pyrolysis of palm kernel cake and cassava pulp
residue, Korean J. Chem. Eng. 28 (12) (2011) 2262e2274, http://dx.doi.org/
10.1007/s11814-011-0116-3.
[61] M. Wei, W. Zhu, G. Xie, T.A. Lestander, S. Xiong, Cassava stem wastes as po-
tential feedstock for fuel ethanol production: A basic parameter study, Renew.
Energy 83 (2015) 970e978, http://dx.doi.org/10.1016/j.renene.2015.05.054.
[62] A. Westerbergh, J. Zhang, C. Sun, Cassava: a multi-purpose crop for the future,
in: C.M. Pace (Ed.), Cassava: Farming, Uses, and Economic Impact, Nova Sci-
ence Publishers, Inc, 2012. ISBN:978-1-61209-655-1.
[63]
W. Zhu, T.A. Lestander, H. Orberg, M. Wei, B. Hedman, J. Ren, S. Xiong, Cassava
stems: a new resource to increase food and fuel production, GCB Bioenergy 7
(1) (2015) 72e83, http://dx.doi.org/10.1111/gcbb.12112.
[64] Y. Niu, H. Tan, S. Hui, Ash-related issues during biomass combustion: alkali-
induced slagging, silicate melt-induced slagging (ash fusion), agglomeration,
corrosion, ash utilization, and related countermeasures, Prog. Energy
Combust. Sci 52 (2016) 1e61.
[65] H. Kludze, B. Deen, A. Dutta, Impact of agronomic treatments on fuel char-
acteristics of herbaceous biomass for combustion, Fuel Process. Technol 109
(2013) 96e102, http://dx.doi.org/10.1016/j.fuproc.2012.09.043.
[66] W.A. Bizzo, P.A. Leno, D.J. Carvalho, J.P.S. Veiga, The generation of residual
biomass during the production of bio-ethanol from sugarcane, its character-
ization and its use in energy production, Renew. Sustain. Energy Rev 29
(2014) 589e603, http://dx.doi.org/10.1016/j.rser.2013.08.056.
[67] EPE e Empresa de Pesquisa Energe tica (Brazil), Brazilian Energy Balance 2015
Year 2014, EPE, Rio de Janeiro, 2015, p. 292.