Waste and Biomass Valorization (2021) 12:3727–3740

https://doi.org/10.1007/s12649-020-01268-y

ORIGINAL PAPER

Improvements in the Extractive and Carbohydrate Analysis

of Sugarcane Bagasse
Paula S. Barbosa1 · Marcio H. P. Barbosa2 · Bruno de F. H. de Faria3 · Reinaldo F. Teófilo1

Received: 19 May 2020 / Accepted: 30 September 2020 / Published online: 7 October 2020
© Springer Nature B.V. 2020

Abstract
Generic procedures for the determination of extractives and carbohydrates (glucose, xylose, galactose, arabinose, mannose)
were improved specifically for sugarcane bagasse. The extraction time was optimized using a specific design of experiments.
The optimal condition for sugarcane biomass with up to 20% extractives was 11.5 h of extraction in water, followed by 9 h
in ethanol. For the carbohydrate analysis, a separation method using liquid chromatography was improved and optimized to
reduce the retention time of furanic compounds. The reduction obtained was approximately 30%. The optimal condition was
the isocratic elution of the mobile phase with an aqueous solution containing 15% acetonitrile. The percentages of glycans,
xylans, and arabinans in the extractive-free biomass was 33%, 22%, and 6.5%, respectively. The galactans and mannans were
not detected. Acetyl groups and uronic acids were not determined.

Electronic supplementary material The online version of this

article (https​://doi.org/10.1007/s1264​9-020-01268​-y) contains
supplementary material, which is available to authorized users.

* Reinaldo F. Teófilo
[email protected]
1
Chemistry Department, Universidade Federal de Viçosa,
Viçosa, MG 36570‑900, Brazil
2
Department of Plant Science, Universidade Federal de
Viçosa, Viçosa, MG 36570‑900, Brazil
3
Department of Forestry Engineering, Universidade Federal
de Viçosa, Viçosa, MG 36570‑900, Brazil

13
Vol.:(0123456789)
3728 Waste and Biomass Valorization (2021) 12:3727–3740

Graphic Abstract

Keywords Sugarcane · Holocellulose · Extractives · Lignocellulosic biomass · Carbohydrates

Abbreviations PDA Photodiode array

GVL Gamma-valerolactone ELSD Evaporative light scattering detector
HMF 5-Hydroxymethylfurfural ACN Acetonitrile
NREL National Renewable Energy Laboratory SRSs Sugar recovery standards
TAPPI Technical Association of the Pulp and Paper BMT Biomass of the RB867515 variety, used as
Industry a test biomass for the development of the
ASTM American Society of Testing and Materials procedure
tw Extraction time in water BMT25 Biomass of the RB867515 variety, used as a
te Extraction time in ethanol test biomass for the development of the proce-
CCD Central composite design dure, dried in open air
HPLC High-performance liquid chromatograph

13
Waste and Biomass Valorization (2021) 12:3727–3740 3729

BMT45 Biomass of the RB867515 variety, used as a Introduction

test biomass for the development of the proce-
dure, oven-dried at 45 °C for 24 h In the last decade, interest in producing bioproducts,
BMA32 Biomass of the RB127032 variety, used as a biochemicals, and biofuels from renewable sources has
application biomass for the validation of the increased worldwide. The environmental damage caused
procedure by the increased concentration of greenhouse gases,
BMA35 Biomass of the RB127035 variety, used as a together with the finite nature of oil reserves, has encour-
application biomass for the validation of the aged the use of renewable and sustainable inputs [1, 2].
procedure Thus, biomass produced with economic viability has
BMA39 Biomass of the RB127039 variety, used as a attracted the interest of governments and businesses [3, 4].
application biomass for the validation of the From a biorefinery standpoint, one of the most eco-
procedure nomically viable biomass sources is sugarcane bagasse,
BMA45 Biomass of the RB127045 variety, used as a due to its high production, logistical ease, and chemical
application biomass for the validation of the composition [5, 6].
procedure The chemical composition of sugarcane is approxi-
Extr Extractives mately 73 to 76% water, 10 to 16% soluble solids (primar-
TL Total lignin ily sucrose), and 11 to 16% (dry) fibers [7]. According to
TCarb Total carbohydrates: cellulose and Conab [8], approximately 110 to 160 kg of dry bagasse per
hemicellulose ton sugarcane are produced annually in Brazil.
IL Insoluble lignin The dry fiber fraction of sugarcane bagasse is highly ener-
SC Silica in the lignin getic and has the following composition: 35.0 to 46.4% cel-
SL Soluble lignin lulose, 20.6 to 35.8% hemicellulose, 15.7 to 28.6% lignin,
TA Total ash 1.6 to 57.9% extractives, and 0.8 to 5.96 ash [9–12].
RSD Relative standard deviation These compounds have a range of applications. For exam-
ANOVA Analysis of variance ple, cellulose, the principal component of the cell wall and
R2 Determination coefficient a source of glucose, has numerous applications, including
CELLO Cellobiose the production of high value-added chemical products such
GLU Glucose as gamma-valerolactone (GVL); hydroxymethylfurfural
XYL Xylose (HMF); succinic, levulinic, lactic, and propionic acids
ARA​ Arabinose [13–15]. Additionally, concerning the carbohydrates present
GAL Galactose in the biomass, when extracted in monomeric/oligomeric
MAN Mannose form, hemicelluloses can be used to manufacture high-value
FUR Furfural chemical products such as furfural and xylitol, as well as in
HMF Hydroxymethylfurfural the production of biofuels [15, 16].
GLN Glycan It is also relevant to note that the proportions of these
XLN Xylan components in the chemical composition of sugarcane bio-
ABN Arabinan mass can be modified through genetic breeding programs,
CELLU Cellulose which makes this grass strategically crucial to government
HEM Hemicellulose and industrial programs. Thus, the chemical characteriza-
tion of sugarcane bagasse is necessary for decision making
in breeding programs, and for indicating the best use of
this by-product.
Statement of Novelty Some studies have focused on the development of energy
cane, which can produce twice the fiber content relative
The novelty presented in this work is the improvement to traditional varieties, generating up to four times more
in generic procedures for quantifying extractives and car- bagasse per unit of the planted area [17–19]. A sugarcane
bohydrates to make them specific for sugarcane bagasse, variety with a higher fiber content is desired to increase
making the results more reliable and comparable. There energy production via the cogeneration of electricity, biofu-
is no specific procedure to characterize the main constitu- els, and chemical compounds of industrial interest, through
ents present in sugarcane bagasse generating higher energy biochemical or thermochemical processes [20–22].
consumption, time, and difficulty in comparing the results. Currently, there are no specific procedures in the litera-
ture for the chemical analysis of sugarcane bagasse. The

13
3730 Waste and Biomass Valorization (2021) 12:3727–3740

existing generics procedures are indicated for the biomass The best drying condition for the crude sample was stud-
of woody species such as eucalyptus and pine [23] as well ied, and the two experimental conditions were as follows:
as non-woody species such as grass, sugarcane, corn, and (i) open-air drying at ~ 25 °C and (ii) drying in an oven at
sorghum [24]. 45 °C for 24 h. For each condition, approximately 2 kg of a
Analyses of the chemical characterization of forest and crude sample was used. The samples were named according
agricultural biomass are based primarily on the procedures to the drying temperature (e.g., BMT25, biomass variety
of the National Renewable Energy Laboratory (NREL) RB867515 for test, dried in open air and BMT45, biomass
within the Technical Association of the Pulp and Paper variety RB867515 for test, oven-dried at 45 °C).
Industry (TAPPI) as well as the American Society of Testing The moisture content, as it is already well known, was
and Materials (ASTM). The TAPPI procedures are standard- determined in triplicate by the gravimetric method as rec-
ized for woods and pulps, whereas those of the ASTM and ommended in the NREL TP-510-42621 procedure until a
NREL are applied to biomass in general. moisture content below 10% was achieved [28].
Despite this knowledge, several studies have used the The crude samples were then ground in a Wiley mill and
TAPPI procedures for analyzing sugarcane bagasse, given sieved using 20 and 80-mesh sieves. The biomass was placed
that there is no specific procedure [25–27]. in the 20-mesh sieve, for 15 min. The material retained in the
The present study is justified by the need to define the 80-mesh sieve was used to analyze the extractives, lignin,
procedures for the determination of extractives and carbo- and carbohydrates. The fraction of fines (material retained
hydrates, specifically for sugarcane bagasse. in the capture pan) was used for an ash analysis. The sieved
This work aimed to propose improvements in the generic materials were stored in Ziplock bags and kept at − 20 °C
determination of extractives and carbohydrates from lig- [29].
nocellulosic biomass to make them specific for sugarcane The ash determination in the biomass was performed
bagasse. using the procedures presented by [30].
Specific characterization procedures for a type of bio-
mass, such as bagasse from sugarcane, are more reliable, Removal and Determination of Extractives
provide a better evaluation of the material, and allows com-
pare results. The BMT45 sample was used to improve the procedure
for analyzing the extractives following the NREL/TP-510-
42619 procedure as a reference [31]. In this step, the analy-
Materials and Methods sis was performed for 10 repetitions. A 3 g portion of this
sample (on a dry weight basis) was weighed and transferred
Samples of sugarcane stalks from the RB867515 variety (the to qualitative filter paper cartridges, which were then placed
commercial reference used in approximately 25% of the total in a continuous Soxhlet extractor (MA044/850, Marconi) for
sugarcane area cultivated in Brazil) were used to improve extraction. The volumes and temperatures of the extraction
extractive and carbohydrate procedures for the chemical solvents were as follows: 150 mL of water at 100 °C fol-
characterization of this biomass. The procedure was applied lowed by 150 mL of ethanol at 80 °C.
to other clones, namely RB127032, RB127035, RB127039, The levels of the variables, namely the extraction time
and RB127045, after obtaining the best conditions. All of in water (­ tw) and the extraction time in ethanol (­ te), were
the genotypes were obtained from the germplasm bank at studied using a complete ­22 factorial design with three
the Sugarcane Genetic Breeding Program at UFV, located in experiments at the central point. After the screening, a cen-
Oratórios, Minas Gerais, Brazil. The sugarcane was planted tral composite design (CCD) was applied to the sample to
in June of 2014 at 20° 44′ S latitude and 42° 50′ W lon- find the optimal levels of the variables that were capable of
gitude, at 400 m altitudes. The samples were harvested at removing the extractives within the shortest analysis time.
12 months of age in the form of ratoon cane (regrowth). The response variable in these designs was the extractives
content [31]. The procedures found in the literature stipulate
Preparation of the Sugarcane Bagasse Sample that the extraction time in the water should be between 6 and
24 h, and the extraction time in ethanol should be between
Fifteen stalks of RB867515 were randomly selected, har- 16 and 24 h [31]. Therefore, these times were the lowest and
vested, and manually stripped of their leaves and tops. The highest levels used for water and ethanol, respectively, in the
stalks were then chopped in an ensilage cutter, homogenized, first complete factorial design.
and pressed at 250 kgf cm−2 min−1 to remove the juice. After the extractions werw all completed, the cartridges
Immediately after the extraction of the juice from the containing the sample were removed from the Soxhlet appa-
stalks, the bagasse was dried. The crushed bagasse from the ratus. The material was dried in an oven at 50 °C for 24 h.
fifteen RB867515 stalks was defined as the crude sample. The extractives content was calculated using Eq. 1.

13
Waste and Biomass Valorization (2021) 12:3727–3740 3731

(
wi(s) − (wf (s+ct) − wct )
) Thus, for the results of the organic chemical composition
% Extractives = × 100 (1) determination of the biomass to be reliable, the overestima-
wi(s)
tion caused by the remaining silica in the insoluble lignin
where wf(s+ct) is the weight of the sample plus the cartridge was discounted, given that it co-precipitates with lignin dur-
after extraction, wct is the weight of the cartridge, and wi(s) ing the acid hydrolysis.
is the weight of the initial sample (dry basis). The silica content was determined according to the fol-
After the procedure was improved, the method was lowing procedure. With the aid of a spatula, the insoluble
applied to the oven and open-air drying condition. lignin was removed from the glass crucible and transferred
to porcelain crucibles. The pre-weighed crucibles were then
placed in a muffle furnace at 575 ± 25 °C for 24 h. After
Determination of the Lignin Content this time, the crucibles were removed from the furnace and
transferred to a desiccator until reaching room temperature.
The samples used for this analysis were first subjected to They were then weighed, and the weight was recorded. The
the extractive removal step (“Removal and Determination difference in the initial and final weight corresponds to the
of Extractives”) using the optimized procedure. inorganic compounds present in the sample. The determined
The amount of soluble and insoluble lignin contents was weight was used for lignin correction. The insoluble lignin
determined in triplicate using two steps: the treatment of the (IL) content was determined by gravimetry.
biomass with strong acid followed by hydrolysis with dilute To determine the soluble lignin content, three aliquots of
acid at high temperature, according to the NREL procedure the filtrate were diluted to 1:3 to keep the absorbance lower
TP-510-42618 [32]. than 1. Subsequently, the diluted solution of each aliquot
During the first step with strong acid, approximately 0.3 g was transferred to a 1 cm optical path quartz cuvette and
(dry weight basis) of the sample of extractive-free sugarcane placed in the sample chamber of a UV–visible spectropho-
bagasse was weighed in 100 mL tubes. To each tube, 3 mL tometer (model USB4000, Ocean Optics). The reading was
of 72% sulfuric acid was added. The tubes were then placed performed in absorbance mode at a wavelength of 240 nm.
in a temperature-controlled bath at 30 ± 3 °C for 120 min. The concentration of soluble lignin in the biomass was deter-
During this time, the mixture was stirred every 5 min with mined according to Beer’s law by considering an absorptiv-
a glass rod. ity at 240 nm equal to 25 L g−1 cm−1 [32].
A modification in the time of hydrolysis was required.
When the 60-min time indicated in NREL procedure Carbohydrate Analysis
TP-510-42618 [32] was used for the first hydrolysis, a high
concentration of cellobiose was observed in carbohydrate To develop the carbohydrate quantification method a high-
analysis. For the 120 min time, no cellobiose formation was performance liquid chromatograph (HPLC) system was
observed. Thus, the time of hydrolysis was increased from used (Prominence model, Shimadzu). A photodiode array
60 to 120 min. (PDA) was used to detect furanic compounds, and an evapo-
After the first step, the tubes were removed from the bath rative light scattering detector (ELSD) was used to detect
and the acid was diluted to a concentration of 4% and were carbohydrates.
autoclaved at 121 °C for 60 min. After cooling, the autoclave Experiments were conducted to select the best chromato-
was opened, and the tubes containing the mixtures were graphic column for separation and quantification. Two col-
removed and immediately subjected to vacuum filtration. umns were studied, specifically a Bio-Rad Aminex HPX-87P
To determine the acid-insoluble lignin content, the hydro- and HPX-87H, both with dimensions of 300 × 7.8 mm and
lyzed lignin obtained was vacuum-filtered in a sintered glass a particle size of 9 μm.
crucible. The chromatographic conditions for the HPX-87P col-
After the filtration, the crucibles were dried in an oven umn were an oven temperature of 80 °C and a flow rate of
at 105 ± 2 °C for 12 h to a constant weight. The weight of 0.85 mL min−1, whereas, for the HPX-87H column, an oven
the residue retained in the crucible was then recorded and temperature of 60 °C and a flow rate of 0.80 mL min−1 was
used to calculate the percentage of insoluble lignin in the used. For both columns, the injection volume of the sample
sample. The filtrate was used to analyze the soluble lignin was 10 μL.
and determine the carbohydrates. The mobile phase was developed by using a mixture of
To determine the insoluble lignin content, it was neces- water and acetonitrile (ACN). ACN was added to water with
sary to correct for the ash content. A large part of the inor- two objectives: (1) decrease the retention time of furanic
ganic constituents present in the bagasse is formed by silica compounds and (2) ensure continuous cleaning of the col-
­(SiO2), which is insoluble in common solvents such as water umn compounds that may hinder their use. The columns
and ethanol [9, 27]. for carbohydrate analysis are expensive, and the use of the

13
3732 Waste and Biomass Valorization (2021) 12:3727–3740

ACN is justified to ensure the stability and long-term use of the standards were collected for HPLC analysis. Shimadzu’s
the column. LabSolution software was used for the configuration of the
The mobile phase was vacuum-filtered through 0.45 μm method, injection, analysis, building calibration models, and
filters and subsequently degassed by ultrasound for 5 min. predictions. A set of sugar recovery standards (SRSs) con-
The ­H2O/ACN ratio and the time (t) to begin ACN addi- taining glucose, xylose, arabinose, galactose, and mannose to
tion at a defined ratio was determined using a complete ­22 correct carbohydrate loss during hydrolysis was also prepared.
factorial design with three experiments at the central point.
The assays were performed using a solution prepared using Preparation of the hydrolysate for HPLC analysis
standards of glucose, xylose, arabinose, galactose, and man-
nose, as well as the furanic compounds HMF and furfural. For the HPX-87P column, hydrolysates, samples, and SRSs
This experiment was conducted using the HPX-87P column were pH corrected to values between 5 and 6 through the slow
because of its strong ability to separate these carbohydrates. addition of calcium carbonate (­ CaCO3) under stirring. The pH
The variables and the studied levels are shown below in was measured with a Mettler Toledo InLab Expert Pro-ISM
Table 1. glass electrode connected to a S220 SevenCompact™ pH/Ion
The response was defined by Eq. 2 pH meter of the same brand.
n For the HPX-87H column, the pH corrections were not nec-
essary. The samples and SRSs were filtered through a 0.45-μm
∑
(tpi − tp1)
i=2 (2) syringe filter and transferred to a vial for subsequent HPLC
Rtr = × 100
tfurfural analysis.

where n is the number of peaks in the chromatogram, tpi is Determination of carbohydrate concentration
the retention time of each peak, tp1 is the retention time of in the biomass
the first peak, and tfurfural is the time spent until the exit of
the furfural, i.e., the last compound to leave the column. The The carbohydrate concentrations in the biomass were deter-
objective was to reduce the analysis time with good separa- mined using the concentrations found by HPLC analysis and
tion between the peaks, yielding the highest Rtr value. thus to obtain the carbohydrate values in an anhydrous form.
The optimized mobile phase levels for the HPX-87P col- The concentrations obtained by HPLC are measured in the
umn were applied to the HPX-87H column, with a slight form of monomers (i.e., glucose, xylose, galactose, arab-
change in the composition of the mobile phase. The opti- inose, and mannose), but the sugars constituting the sugar-
mized mobile phase was acidified with 0.005 mol ­L−1 acetic cane bagasse are bound together to form polysaccharides.
acid so that the HPX-87H column can be used. No other Therefore, the values were reported in their polymeric form
conditions were changed. to express them correctly. The bonding between the two mono-
mers to obtain a polymer occurs with the loss of a proton and a
Calibration models for carbohydrates analysis hydroxyl, for a total of 18 g mol−1 of mass lost, i.e., one mol-
ecule of ­H2O. The anhydrous correction factor for the hexoses
Calibration models were built after defining the composi- is 0.90 (162/180), whereas, for the pentoses, the value is 0.88
tion and optimal time for adding the ACN. In the HPX-87H (132/150) [12].
column, models for glucose, xylose, and arabinose were The carbohydrate concentration in the samples was cal-
built, covering 10 levels between 0.08 and 5 mg mL−1. The culated in triplicate using Eqs. 3–6, according to the NREL
calibration models were validated with validation standards procedure [32].
containing the carbohydrates at defined concentrations, i.e.,
0.4, 2.9, and 4.2 mg mL−1. Subsequently, the vials containing Conc.standard detected HPLC
% SRSx =
Conc.standard known before hydroly
× 100 (3)

Table 1  Variables and levels Variables Levels

Conc.sample HPLC × f
used in the complete 2­ 2 factorial Cx = %SRSx (4)
design for the studied mobile −1 0 +1 100
phase
t/min 0 6 12
% ACN 15 20 25 Cx.corrected = Cx × Anhydrous correction (5)
t is the time at which the addi-
tion of ACN begins, and %ACN
is the percentage of added ace-
tonitrile.

13
Waste and Biomass Valorization (2021) 12:3727–3740 3733

Cx .corrected × Vf ×
1g From these results, the central composite design (CCD)
× 100 (6) with two variables was set up to investigate the other levels
1000 mg
% Carbohydrates =
w for the variables ­tw and ­te. For ­tw, the levels (-) 4,5 and ( +)
Here w is the weight of the sample in grams without extrac- 11,5 h were investigated, and the levels studied for ­te were
tives (dry weight basis), Vf is the volume of the filtrate in (-) 4 and ( +) 9 h. The design levels and results are shown
mL, f is dilution factor and subscript x represents the differ- in Table 2.
ent sugars. Statistical analysis was performed based on the results
presented in Table 2. The evaluation of the coefficients
Validation of the Procedures for Other Varieties estimated for the CCD showed (at a significance level of
0.05) that the primary variables (­ tw and t­ e) were significant
After improving the procedures of extractives and carbohy- and positive. This result suggests that the higher levels of
drates, four different genotypes were selected for the appli- these variables should be used to obtain better extractions
cation of these procedures, namely RB127032 (BMA32), within the studied levels. The quadratic coefficient of the
RB127035 (BMA35), RB127039 (BMA39), and RB127045 t.w was also significant and indicates that there is a signifi-
(BMA45). cant quadratic variation in the responses within the studied
The total mass balance was compared with the results levels for this variable. The interaction between variables
found in the literature for sugarcane bagasse. t.w and t.e was also significant and positive, indicating that
their combination results in synergistic behavior.
The regression model built using analysis of variance
Calculation of the Total Mass Balance
(ANOVA) presented a proper fit at a significance level of
0.05 (Eq. 8). The determination coefficient (R2) indicates
To calculate the total mass balance considering all the con-
that the model can explain 97.5% of the total variation
stituents present in the sugarcane bagasse (i.e., extractives
around the mean; therefore, the model was satisfactory
[Extr], total lignin [TL], total carbohydrates: cellulose and
within the studied levels for explaining the extraction,
hemicellulose [TCarb], and ash [Ash]), based on the dry
using the variables t.w and t.e.
biomass with extractives, as shown in Eq. 7.
%Extr = 18.3 + 0.44 tw + 0.33 te + 0.49 t2w + +0.14 t.2e + 0.35 t�w te
%MB = Extr + TL + TCarb + Ash (7)
(8)
The response surface obtained from the model is shown
Calculations and Statistical Analyses (Fig. 1). The results indicated that the optimum time for
maximum extraction with water followed by ethanol is
All the calculations and graphs were completed using the 11.5 and 9 h, respectively.
following software: Statistica 7.0, Microsoft Excel 365,
Shimadzu’s LabSolution version 5.54 SP3, and Microcal
Origin 9.0.
Table 2  Variables, levels, and the response of the central composite
design (CCD) for optimizing the extractives determination
Experiment Variables Response Extractives/%
Results and discussion
t.w (h) t.e (h)

The genotypes obtained from the germplasm bank were 1 4.50 (−) 4.00 (−) 18.3
characterized concerning their fiber and sucrose contents 2 11.5 (+) 4.00 (−) 18.6
based on Brix and POL analysis. The results are presented 3 4.50 (−) 9.00 (+) 18.3
in Table S1 of supplementary material. 4 11.5 (+) 9.00 (+) 20.0
5 3.00 (− 1.41) 6.50 (0) 18.8
Determination of extractives 6 13.0 (+ 1.41) 6.50 (0) 19.9
7 8.00 (0) 3.00 (-1.41) 18.2
The use of a ­22 full factorial design, with three experi- 8 8.00 (0) 10.0 (+ 1.41) 19.1
ments at the central point, helped to obtain that the maxi- 9 8.00 (0) 6.50 (0) 18.3
mum extractives content expected when using the biomass 10 8.00 (0) 6.50 (0) 18.2
BMT45. The maximum extractives content obtained was, 11 8.00 (0) 6.50 (0) 18.3
on average, 19.5%. This amount is expected because, in the
tw is the extraction time in the water, and ­te is the extraction time in
screening, the studied levels were extreme, namely 6 to 24 h ethanol. The symbols (+) and (−) refer to the maximum and mini-
for water and 16 to 24 h for ethanol. mum levels of the independent variables.

13
3734 Waste and Biomass Valorization (2021) 12:3727–3740

attributed to variations in the weighing and weight loss

during the processing of the samples.
Different extractive content values were found in the lit-
erature, and they ranged from 1.6 to 57.9% [11, 33]. The
variation in the values found among the studies is due to the
characteristics of the analyzed sugarcane species, maturation
time, harvest type, sample preparation, climatic conditions,
region, and soil, among other factors.
It is important to emphasize that each study determined
the extractives content using different extraction times
and solvent types. The efficiency is affected by the solvent
polarity and the sequence in use. These make it difficult to
compare the data from this study with the literature. Thus,
a standard procedure for analyzing extractives specific to
sugarcane is desirable and necessary.
Fig. 1  Response surface that relates the extractives content according
to the extraction time in water ­(tw) and ethanol (­ te) Optimization of the removal of sugarcane bagasse extrac-
tives can be achieved using only water and ethanol, avoiding
the use of toxic solvents, such as hexane, toluene, benzene,
The procedures defined for the removal of extrac- which are quite common in the works found in the litera-
tives were applied in triplicate for the BMA32, BMA35, ture, these are not referring to a specific methodology for
BMA39, and BMA45 genotypes (Table 3). According to sugarcane bagasse.
RSD values obtained for the different samples (Table 3), The optimal times of 11.5 and 9 h found for maximum
it was concluded that the procedure developed could be extraction in water followed by ethanol, respectively, can be
reliably applied to other genotypes. The RSD values were applied to sugarcane bagasse that contains up to 20% extrac-
less than 10% for all the samples. This random error was tives. Therefore, if the extractives content found using the
indicated procedure is higher than 20%, it is recommended

Table 3  Application of the Extr/ % IL/% SC/% TS/% TL/% TA/% GLN/% XLN/% ARB/% 1
CEL/% 2
HEM/%
procedures for determining
the content of extractives, BMT45
carbohydrates and lignin in
MED 19.5a 21.8 2.29 2.45 22.0b 3.63 33.2 23.4 6.82 33.2ab 30.3d
sugarcane bagasse hydrolysate
of dry extractive-free samples RSD 1.90 0.540 4.70 2.30 0.660 2.20 3.30 1.90 0.550 3.30 1.50
of the bagasse from different BMA32
sugarcane genotypes MED 14.8b 24.2 1.03 2.01 25.2a 2.13 33.9 21.4 6.48 33.9ab 27.9a
RSD 8.00 0.980 2.20 1.20 1.00 1.30 3.00 0.660 0.120 3.00 0.490
BMA35
MED 12.0c 24.7 1.38 1.82 25.1a 2.62 32.7 21.3 6.47 32.7a 27.7a
RSD 6.40 1.30 2.10 1.80 1.40 1.80 2.50 0.550 0.810 2.50 0.420
BMA39
MED 8.67d 24.7 1.35 1.94 25.3a 2.29 35.8 22.1 6.52 35.8b 28.6b
RSD 9.00 0.890 1.30 4.00 0.65 3.80 3.80 0.270 0.390 3.80 0.160
BMA45
MED 19.5a 19.8 1.49 2.49 20.8c 2.60 33.7 22.8 6.65 33.7ab 29.5c
RSD 4.90 0.980 3.00 3.40 1.40 2.30 3.50 0.850 0.480 3.50 0.620

BMT45 (RB867515) dried at 45 °C; BMA32 (RB127032), BMA35 (RB127035), BMA39 (RB127039),
and BMA45 (RB127045). Extractives (Extr), Insoluble lignin (IL), Silica in the lignin (SC), Soluble lignin,
(SL), Total lignin (TL), Total ash (TA), Glycans (GLN), Xylans (XLN), Arabinans (ABN), Cellulose (CEL),
and Total hemicellulose (HEM). Relative standard deviation (RSD). Means followed by the same letter do
not differ according to Tukey’s test at 5% significance
1
The cellulose content was expressed according to the glycan content
2
The hemicellulose calculation was based on the sum of the xylans and arabinans present in the biomass.
Galactose and mannose were not detected, and the glucose content in the hemicellulose was not consid-
ered.

13
Waste and Biomass Valorization (2021) 12:3727–3740 3735

that the times in water and ethanol be increased by follow- condition is the one with the highest response because it is
ing the model presented in Eq. 14 as a reference. However, desirable to obtain a lower total analysis time and a good
according to the built and fitted model, the conditions found separation of peaks. The highest response was obtained for
here can be used for contents higher than 20%. However, no 15% ACN addition at time zero. Thus, the optimum condi-
experiment was conducted with sugarcane bagasse possess- tion was found to be isocratic elution.
ing extractives content higher than 20%. The effects of the studied variables on the factorial design
at the significance level of 0.05 indicated that the "acetoni-
Soluble and Insoluble Lignin Content trile addition start time" variable was statistically significant
and negative. This result indicates that the faster the ACN
As shown (Table 3), the samples have different mean values is introduced to the mobile phase, the better the response,
(ranging from 20.76 to 25.26%) for the total lignin content. at least within the studied levels, which is why we chose
The results obtained for the lignin fractions (IL, SL) are time zero. The %ACN was not significant within the studied
presented in Table 3. The results found in this study were levels.
reliable relative to those in the literature [34–37]. The chromatograms obtained by HPLC using the HPX-
87P column after optimization are shown (Fig. 2).
Determination of Carbohydrates A comparison between the results obtained without
mobile phase optimization and the results obtained (Fig. 2)
Method Optimization indicates that the total analysis time decreased from 30 to
19 min and that the retention time of the furanic compounds
The total analysis time in the HPLC with the HPX-87P col- decreased from 27 to 16.5 min. This finding confirms that
umn using only water as the mobile phase (i.e., the mobile the change in the mobile phase’s polarity through the addi-
phase was not optimized under this condition) was 30 min, tion of acetonitrile significantly improves the total time of
whereas the retention time of the furanic compounds was this analysis and improves the separation. The ion exchange
27 min. column used in this study, achieves separation based on the
The carbohydrates and furanic compounds were abbre- reversible electrostatic attraction between the ions of the
viated as follows: cellobiose (CELLO), glucose (GLU), mobile phase and the sample with the immobilized centers
xylose (XYL), arabinose (ARA), galactose (GAL), mannose of positive charge (­ Pb2+) present in the stationary phase of
(MAN), furfural (FUR), and hydroxymethylfurfural (HMF). the column.
The results for studying the effect of the acetonitrile addi- ACN is a polar aprotic solvent, and it has medium-polar-
tion start time, and its percentage in the mobile phase on ity, dissolving a wide range of ionic and nonpolar com-
the Rtr response indicated a range of 30.9 to 53.7. The best pounds. In the separation, ACN acts as an organic modifier,

Fig. 2  Chromatograms obtained

using the HPX-87P column
under the optimized condition
(Assay 2). a ELSD, b PDA
detector, (1) GLU, (2) XYL, (3)
GAL, (4) ARA, (5) MAN, (6)
HMF, and (7) FUR

13
3736 Waste and Biomass Valorization (2021) 12:3727–3740

hindering the interaction between the furanics and the sta- not detected in any of the analyzed samples. GLU and XYL
tionary phase since furanics dissolve better in acetonitrile. were the most significant carbohydrates.
Consequently, the total analysis time decreased significantly The chromatograms of the sugarcane bagasse hydro-
[38, 39]. lysates obtained using the HPX-87P column indicated a sig-
Since furanic compound retention times are longer nificant matrix effect; that is, the constituents present in the
than carbohydrates, the monitoring of these outputs may biomass (matrix) affect the response. Matrix effects affected
be neglected by the analyst. The non-removal of furanics the baseline of the chromatogram, and it could be observed
from the column will certainly cause a rapid decrease in after approximately 8 min. Statistical analyses showed that
separation efficiency and early loss of the chromatographic the matrix effect overestimated the concentration of the
column. In the quantification of insoluble lignin, carbohy- quantified carbohydrates, especially for ARA, which was
drates, and soluble lignin, hydrolysis with sulfuric acid is the most affected carbohydrate. Due to this effect, there was
performed. Furanics are generated from the acid hydrolysis a decision to replace the Bio-Rad HPX-87P column with the
step. 5-(hydroxymethyl)furfural (HMF) is originated from HPX-87H column.
6-carbon sugars (hexoses). Similar chemistry is observed The advantages of the HPX-87H column over the HPX-
with 5-carbon sugars (pentoses), which react with aqueous 87P column are (1) the use of an acid mobile phase (i.e., pH
acid to form furfural [40]. The use of ACN ensures high- between 1 and 3) and therefore it is not necessary to per-
resolution analysis and longer life of the chromatographic form pH correction of the samples with C ­ aCO3 and (2) not
column. exhibiting the matrix effect observed in HPX-87P column
The optimized conditions were applied to the HPX-87H (Fig. 3). However, mixtures with the carbohydrates GLU,
column. XYL, ARA, GAL, and MAN are not recommended for this
column because GAL and MAN co-elute with the first three
carbohydrates.
Separation of the Carbohydrates in Sugarcane Bagasse Given that only GLU, XYL, and ARA were detected in
Hydrolysates the sugarcane bagasse hydrolysate samples, the HPX-87H
column is the most recommended.
The typical chromatograms of the two sugarcane bagasse The results (Fig. 4) in which no matrix effect is observed
hydrolysates obtained using the HPX-87P column are shown illustrate the typical separation of the carbohydrates pre-
in Fig. 3. In all of the samples analyzed, only the peaks sent in the analyzed sugarcane bagasse hydrolysate samples.
related to glucose, xylose, and arabinose were detected. Thus, the predictions for the carbohydrate concentrations are
According to Fig. 3, the low-intensity cellobiose peak more reliable with this column.
was observed, which indicates that the hydrolysis was prac- The glycan (GLN), xylan (XLN), and arabinan (ABN)
tically complete. The GAL and MAN carbohydrates were contents, as well as the calculation of the cellulose (CELLU)

Fig. 3  Typical chromatograms

of two sugarcane bagasse
hydrolysates obtained by
HPLC-ELSD using the HPX-
87P column. a BMT45 sample
(RB867515 dried at 45 °C) and
b BMA45 (RB127045). Peaks:
(1) CELLO, (2) GLU, (3) XYL,
and (4) ARA​

13
Waste and Biomass Valorization (2021) 12:3727–3740 3737

Fig. 4  Typical chromatograms

obtained in the HPX-87H
column for the hydrolysates
of samples a BMA15 and b
BMA45. Peaks: (1) GLU, (2)
XYL, and (3) ARA​

and hemicellulose (HEM) values, are shown below 92.8%, and the RSD was less than 1.5%. These results are
(Table 3). The values are expressed basis extractive-free. similar to those reported by Masarin et al. [33] and Hoang
These results were obtained after the separation of the sugar- et al. [10], which found mean balance values of 92.4 and
cane bagasse hydrolysate samples in the HPX-87H column. 93.6%, respectively.
Among the carbohydrates analyzed, the GLN content The BMA39 genotype had higher total lignin and
was higher on average, with values ranging from 32.74 to total carbohydrate (CEL + HEM) contents, which can
35.78%, and the RSD was less than 4.0%. Regarding the be explained by the fact that this sample is classified as
HEM, the most representative was XLN, with values ranging energy cane and has a high fiber content (33.41%) in its
between 21.11 and 23.44% and an RSD lower than 2.0%. composition. By contrast, BMA45 had lower total lignin,
The cellulose contents obtained for the five genotypes cellulose, and hemicellulose content than the other geno-
were slightly below the range (35.0 to 46.4%) found in the types because it was the clone with the lowest fiber content
literature [10, 41–43]. However, to compensate for the lower (11.11%).
cellulose content, an increase of about 1.2 to 2.5% in ABN The concentrations of the acetyl group and uronic acids
content was observed [9, 12]. For the XLN and the total were not quantified. These groups are bound to hemicel-
HEM content, the values were similar to those reported in lulose, more specifically to the xylan chain, and they are
the literature [10, 27, 44, 45]. released during hydrolysis [46, 47]. The non-quantification
of these contents and the loss of volatile extractives during
Total Mass Balance sample drying are reasons why the mass balance did not
reach values close to 100%.
During the chemical characterization, achieving a mass bal- According to the literature, the acetyl group content
ance equal to 100% is not an easy task, which is due to the is significant, with values ranging from 1.9 to 3.7% [9,
difficulty of isolating the macromolecules constituting the 48]. By contrast, the uronic acid content is lower, ranging
cell wall (cellulose, hemicellulose, and lignin) without com- from 1.0 to 2.4%; therefore, they are often not analyzed
promising them. Furthermore, the high number of steps in [9, 26]. Consequently, the quantification of these compo-
the analyses increases the random error, making the results nents causes the total mass balance to be closer to the
susceptible to systematic errors. exact value.
For this study, the total mass balance of the analyzed The release of uronic acids during hydrolysis is undesir-
samples was obtained by adding the individual contents of able because it produces toxic compounds such as meth-
the extractives, total lignin, total ash, cellulose, and hemi- ane. Additionally, in the context of obtaining cellulose and
cellulose for each sample, based on the dry biomass with paper, these acids increase the consumption of reagents
extractives. The results are shown in Table 4. during cellulosic pulp bleaching and can cause whiteness
The results shown in Table 4 indicate that the total mass instability phenomena.
balance for the samples reached levels between 89.6 and

13
3738 Waste and Biomass Valorization (2021) 12:3727–3740

Table 4  Chemical constituents BMT45 BMA32 BMA35 BMA39 BMA45

of sugarcane bagasse biomass
(dry basis) with extractives Mean RSD Mean RSD Mean RSD Mean RSD Mean RSD
a b b b b
Extr/% 16.5 2.20 14.8 8.00 12.0 6.40 8.67 9.00 19.5 4.90
IL/%b 18.2 0.620 20.6 1.50 21.7 2.20 22.5 1.20 15.9 0.760
SC/%b 1.91 4.30 0.870 1.90 1.22 1.40 1.23 1.30 1.20 2.40
SL/% b 2.04 1.90 1.71 2.60 1.60 1.80 1.77 3.40 2.00 2.70
TL/%b 18.3 0.840 21.5 1.60 22.1 2.20 23.1 0.820 16.7 1.10
TA/%b 3.63 2.20 2.12 1.30 2.62 1.80 2.29 3.20 2.60 2.30
GLN/%b 27.7 3.10 28.9 4.00 28.8 3.20 32.7 3.30 27.1 3.10
XLN/%b 19.6 1.80 18.2 2.00 18.7 1.00 20.2 1.10 18.4 1.80
ABN/%b 5.69 0.900 5.52 1.30 5.69 0.260 5.95 0.640 5.35 0.710
1
CEL/%b 27.7 3.10 28.9 4.00 28.8 3.20 32.7 3.30 27.1 3.10
2
HEM/%b 25.3 1.50 23.8 1.80 24.4 0.750 26.1 1.00 23.7 1.50
Total/% 91.5 1.20 91.0 1.20 90.0 0.970 92.8 1.30 89.6 1.50

BMT45 (RB867515) dried at 45 °C. BMA32 (RB127032), BMA35 (RB127035), BMA39 (RB127039),
and BMA45 (RB127045). Extractives (Extr), Insoluble lignin (IL), Silica in the lignin (SC), Soluble lignin,
(SL), Total lignin (TL), Total ash (TA), Glycans (GLN), Xylans (XLN), Arabinans (ABN), Cellulose (CEL),
and Total hemicellulose (HEM).
a,b
Correspond to the mean obtained using ten and three replicates, respectively. RSD = relative standard
deviation
1
Cellulose is expressed according to the glycan content
2
The hemicelluloses calculation was based on the sum of the xylans and arabinans present in the biomass.
Galactose and mannose were not detected, and the glucose content present in the hemicellulose was not
considered

Conclusions bagasse in order to obtain more accurate and possible results

for comparison.
Procedures for the determination of extractive and carbo-
hydrates were improved, with significant gains in the anal- Acknowledgements This study was financed in part by the Coorde-
nação de Aperfeiçoamento de Pessoal de Nível Superior—Bra-
ysis time and cost. Additionally, more advanced knowl- sil (CAPES)—Finance Code 001 and by the Fundação de Amparo
edge of each procedure was achieved. The optimal time à Pesquisa no Estado de Minas Gerais (FAPEMIG)—Process
found to remove the extractives was significantly reduced. CEX—APQ-01424-13.
In addition, the use of environmentally clean and non-toxic
extracting solvents is advantageous. During the hydrolysis
step, a longer time in concentrated sulfuric acid is required
so that the cellobioses are broken down to free the glu- References
cose. The method developed to separate and quantify the
1. Baral, N.R., Sundstrom, E.R., Das, L., Gladden, J., Eudes, A.,
carbohydrates in the HPLC was efficient and precise. The Mortimer, J.C., Singer, S.W., Mukhopadhyay, A., Scown, C.D.:
chromatographic separation time was reduced by approxi- Approaches for more efficient biological conversion of ligno-
mately 30% with the addition of acetonitrile during the cellulosic feedstocks to biofuels and bioproducts. ACS Sustain.
mobile phase. The HPX-87H column presented excellent Chem. Eng. 7(10), 9062–9079 (2019). https​://doi.org/10.1021/
acssu​schem​eng.9b012​29
reliability for the quantification of the carbohydrates in the 2. Bezerra, T.L., Ragauskas, A.J.: A review of sugarcane bagasse for
hydrolysate of the analyzed samples. Furthermore, the use second-generation bioethanol and biopower production. Biofuel.
of the HPX-87H column does not require pH correction Bioprod. Biorefin. 10(5), 634–647 (2016). https:​ //doi.org/10.1002/
before injection into the HPLC instrument. The total mass bbb.1662
3. Ullah, K., Sharma, V.K., Dhingra, S., Braccio, G., Ahmad, M.,
balance for the sugarcane bagasse was determined to be Sofia, S.: Assessing the lignocellulosic biomass resources poten-
between 90 and 92.8%, which was considered satisfac- tial in developing countries: a critical review. Renew. Sustain.
tory, given that acetyl groups and uronic acids were not Energy Rev. 51, 682–698 (2015). https​: //doi.org/10.1016/j.
determined. rser.2015.05.044
4. Chen, H.Z., Fu, X.G.: Industrial technologies for bioetha-
Finally, with the development of this work, it is possible nol production from lignocellulosic biomass. Renew. Sustain.
to perceive the need and the importance of standardizing Energy Rev. 57, 468–478 (2016). https​: //doi.org/10.1016/j.
the chemical characterization methodology for sugarcane rser.2015.12.069

13
Waste and Biomass Valorization (2021) 12:3727–3740 3739

5. Chandel, A.K., Silveira, M.H.L.: Advances in Sugarcane 22. Chakraborty, S., Aggarwal, V., Mukherjee, D., Andras, K.: Bio-
Biorefinery: Technologies, Commercialization, Policy Issues mass to biofuel: a review on production technology. Asia Pac. J.
and Paradigm Shift for Bioethanol and By-products. Elsevier, Chem. Eng. 7, S254–S262 (2012)
Amsterdam (2017) 23. Tappi: Sampling and preparing wood for analysis. In: TAPPI
6. Chandel, A.K., Garlapati, V.K., Singh, A.K., Antunes, F.A.F., Standard Methods T257cm-12, Technical Association of the Pulp
da Silva, S.S.: The path forward for lignocellulose biorefiner- and Paper Industry, Atlanta, GA, USA (2012)
ies: Bottlenecks, solutions, and perspective on commerciali- 24. Sluiter, J.B., Ruiz, R.O., Scarlata, C.J., Sluiter, A.D., Templeton,
zation. Bioresour. Technol. 264, 370–381 (2018). https​://doi. D.W.: Compositional analysis of lignocellulosic feedstocks. 1
org/10.1016/j.biort​ech.2018.06.004 Review and description of methods. J. Agric. Food Chem. 58(16),
7. Eggleston, G.: Positive aspects of cane sugar and sugar cane 9043–9053 (2010). https​://doi.org/10.1021/jf100​8023
derived products in food and nutrition. J. Agric. Food Chem. 25. Alves, E.F., Bose, S.K., Francis, R.C., Colodette, J.L., Iakovlev,
66(16), 4007–4012 (2018). https ​ : //doi.org/10.1021/acs. M., Van Heiningen, A.: Carbohydrate composition of eucalyptus,
jafc.7b057​34 bagasse and bamboo by a combination of methods. Carbohydr.
8. CONAB: Acompanhamento da safra brasileira de cana-de- Polym. 82(4), 1097–1101 (2010). https​://doi.org/10.1016/j.carbp​
açúcar, Safra 2017/18, Terceiro levantamento, vol. 4. pp. 1–77. ol.2010.06.038
Brasília (2017) 26. Batalha, L., Han, Q., Jameel, H., Chang, H.M., Colodette, J.L.,
9. Andrade, M.F., Colodette, J.L.: Dissolving pulp production from Gomes, F.J.B.: Production of fermentable sugars from sugarcane
sugar cane bagasse. Ind. Crops Prod. 52, 58–64 (2014). https​:// bagasse by enzymatic hydrolysis after autohydrolysis and mechan-
doi.org/10.1016/j.indcr​op.2013.09.041 ical refining. Bioresour. Technol. 180, 97–105 (2015). https​://doi.
10. Hoang, N.V., Furtado, A., Donnan, L., Keeffe, E.C., Botha, org/10.1016/j.biort​ech.2014.12.060
F.C., Henry, R.J.: High-throughput profiling of the fiber and 27. de Carvalho, D.M., Sevastyanova, O., Penna, L.S., da Silva,
sugar composition of sugarcane biomass. BioEnergy Res. 10(2), B.P., Lindstrom, M.E., Colodette, J.L.: Assessment of chemical
400–416 (2017). https​://doi.org/10.1007/s1215​5-016-9801-8 transformations in eucalyptus, sugarcane bagasse and straw dur-
11. Kuchelmeister, C., Bauer, S.: Rapid small-scale determination ing hydrothermal, dilute acid, and alkaline pretreatments. Ind.
of extractives in biomass. BioEnergy Res. 8(1), 68–76 (2015). Crops Prod. 73, 118–126 (2015). https​://doi.org/10.1016/j.indcr​
https​://doi.org/10.1007/s1215​5-014-9493-x op.2015.04.021
12. Rocha, G.J.M., Nascimento, V.M., Goncalves, A.R., Silva, 28. Sluiter, A., Hames, B., Hyman, D., Payne, C., Ruiz, R., Scarlata,
V.F.N., Martin, C.: Influence of mixed sugarcane bagasse sam- C., Sluiter, J., Templeton, D., Wolfe, J.: Determination of Total
ples evaluated by elemental and physical-chemical composition. Solids in Biomass and Total Dissolved Solids in Liquid Process
Ind. Crops Prod. 64, 52–58 (2015). https​://doi.org/10.1016/j. Samples. NREL/TP-510–42621, pp. 1–6. National Renewable
indcr​op.2014.11.003 Energy Laboratory, Golden (2008)
13. Artz, J., Palkovits, R.: Cellulose-based platform chemical: The 29. Hames, B., Ruiz, R., Scarlata, C., Sluiter, A., Sluiter, J., Tem-
path to application. Curr. Opin. Green Sustain. Chem. 14, 14–18 pleton, D.: Preparation of Samples for Compositional Analysis.
(2018). https​://doi.org/10.1016/j.cogsc​.2018.05.005 NREL/TP-510-42620, pp. 1–9. National Renewable Energy Labo-
14. Dong, Y.X., Yu, H.Z., Iop: Research progress on biorefinery of ratory, Golden, CO (2008)
lignocellulosic biomass. In: 2019 5th International Conference 30. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Temple-
on Energy Materials and Environment Engineering, vol. 295. ton, D.: Determination of Ash in Biomass. NREL/TP-510–42622,
IOP Conference Series-Earth and Environmental Science. Iop pp. 1–5. National Renewable Energy Laboratory, Golden, CO
Publishing Ltd, Bristol (2019) (2008)
15. Liu, Y.R., Nie, Y., Lu, X.M., Zhang, X.P., He, H.Y., Pan, F.J., 31. Sluiter, A., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D.: Deter-
Zhou, L., Liu, X., Ji, X.Y., Zhang, S.J.: Cascade utilization of mination of Extractives in Biomass. NREL/TP-510-42619, pp.
lignocellulosic biomass to high-value products. Green Chem. 1–9. National Renewable Energy Laboratory, Golden, CO (2008)
21(13), 3499–3535 (2019). https​://doi.org/10.1039/c9gc0​0473d​ 32. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Temple-
16. Naidu, D.S., Hlangothi, S.P., John, M.J.: Bio-based products ton, D., Crocker, D.: Determination of Structural Carbohydrates
from xylan: a review. Carbohydr. Polym. 179, 28–41 (2018). and Lignin in Biomass. NREL/TP-510-42618, pp. 1–15. National
https​://doi.org/10.1016/j.carbp​ol.2017.09.064 Renewable Energy Laboratory, Golden (2012)
17. Matsuoka, S., Rubio, L.C.S.: Energy cane: a sound alternative 33. Masarin, F., Gurpilhares, D.B., Baffa, D.C.F., Barbosa, M.H.P.,
of a bioenergy crop for tropics and subtropics. In: Khan, M.T., Carvalho, W., Ferraz, A., Milagres, A.M.F.: Chemical composi-
Khan, I.A. (eds.) Sugarcane Biofuels, pp. 39–66. Springer, New tion and enzymatic digestibility of sugarcane clones selected for
York (2019) varied lignin content. Biotechnol. Biofuels. (2011). https​://doi.
18. Zhao, D.L., Momotaz, A., LaBorde, C., Irey, M.: Biomass org/10.1186/1754-6834-4-55
yield and carbohydrate composition in sugarcane and energy 34. Pitarelo, A.P., Silva, T.A.D., Peralta-Zamora, P.G., Ramos, L.P.:
cane grown on mineral soils. Sugar Tech. (2020). https​://doi. Efeito do teor de umidade sobre o pré-tratamento a vapor e a
org/10.1007/s1235​5-020-00807​-0 hidrólise enzimática do bagaço de cana-de-açúcar. Quím. Nova
19. Carvalho-Netto, O.V., Bressiani, J.A., Soriano, H.L., Fiori, C.S., 35(8), 1502–1509 (2012). https​://doi.org/10.1590/S0100​-40422​
Santos, J.M., Barbosa, G.V.S., Xavier, M.A., Landell, M.G.A., 01200​08000​03
Pereira, G.A.G.: The potential of the energy cane as the main 35. Chong, B.F., Purcell, D.E., O’Shea, M.G.: Diffuse Reflectance,
biomass crop for the cellulosic industry. Chem. Biol. Technol. Near-Infrared Spectroscopic Estimation of Sugarcane Lignocel-
Agric. 1(1), 20 (2014) lulose Components-Effect of Sample Preparation and Calibration
20. Matsuoka, S., Kennedy, A.J., Santos, E.G.D., Tomazela, A.L., Approach. BioEnergy Res. 6(1), 153–165 (2013). https​://doi.
Rubio, L.C.S.: Energy cane: its concept, development, charac- org/10.1007/s1215​5-012-9243-x
teristics, and prospects. Adv. Bot. 2014, 13 (2014) 36. Saha, K., Dasgupta, J., Chakraborty, S., Felipe, A.A.F., Sikder,
21. Barbosa, G.V.S., João, M.S., Diniz, C.A., Cursi, D.E., Hoff- J., Curcio, S., dos Santos, J.C., Arafat, H.A., da Silva, S.S.: Opti-
mann, H.P.: Energy cane breeding. Sugarcane Biorefinery Tech- mization of lignin recovery from sugarcane bagasse using ionic
nology and Perspectives, pp. 103–116. Elsevier, Amsterdam liquid aided pretreatment. Cellulose 24(8), 3191–3207 (2017).
(2020) https​://doi.org/10.1007/s1057​0-017-1330-x

13
3740 Waste and Biomass Valorization (2021) 12:3727–3740

37. Saha, K., Dwibedi, P., Ghosh, A., Sikder, J., Chakraborty, S., 44. Neureiter, M., Danner, H., Thomasser, C., Saidi, B., Braun, R.:
Curcio, S.: Extraction of lignin, structural characterization and Dilute-acid hydrolysis of sugarcane bagasse at varying condi-
bioconversion of sugarcane bagasse after ionic liquid assisted tions. Appl. Biochem. Biotechnol. 98, 49–58 (2002). https​://doi.
pretreatment. 3 Biotech 8(8), 12 (2018). https​://doi.org/10.1007/ org/10.1385/abab:98-100:1-9:49
s1320​5-018-1399-4 45. Saha, K., Maharana, A., Sikder, J., Chakraborty, S., Curcio, S.,
38. Yuan, J.P., Chen, F.: Simultaneous separation and determination Drioli, E.: Continuous production of bioethanol from sugarcane
of sugars, ascorbic acid and furanic compounds by HPLC—dual bagasse and downstream purification using membrane inte-
detection. Food chem. 64(3), 423–427 (1999) grated bioreactor. Catal. Today 331, 68–77 (2019). https​://doi.
39. Yuan, J.P., Wu, H., Zen, Y.S.: HPLC determination of 5-hydroxy- org/10.1016/j.catto​d.2017.11.031
methylfurfural in glucose injections. Chin. J. Pharm. 26, 505–506 46. McDougall, G.J., Morrison, I.M., Stewart, D., Weyers, J.D.B.,
(1995) Hillman, J.R.: Plants fibres - Botany, chemistry and processing
40. Catrinck, M.N., Barbosa, P.S., Filho, H.R.O., Monteiro, R.S., for industrial use. J. Sci. Food Agric. 62(1), 1–20 (1993). https​://
Barbosa, M.H.P., Ribas, R.M., Teofilo, R.F.: One-step process doi.org/10.1002/jsfa.27406​20102​
to produce furfural from sugarcane bagasse over niobium-based 47. Carvalheiro, F., Esteves, M.P., Parajo, J.C., Pereira, H., Girio,
solid acid catalysts in a water medium. Fuel Process. Technol. F.M.: Production of oligosaccharides by autohydrolysis of brew-
207, 11 (2020). https​://doi.org/10.1016/j.fupro​c.2020.10648​2 ery’s spent grain. Bioresour. Technol. 91(1), 93–100 (2004). https​
41. Sasaki, M., Adschiri, T., Arai, K.: Fractionation of sugarcane ://doi.org/10.1016/s0960​-8524(03)00148​-2
bagasse by hydrothermal treatment. Bioresour. Technol. 86(3), 48. Rocha, G.J.M., Martin, C., da Silva, V.F.N., Gomez, E.O., Gon-
301–304 (2003). https​://doi.org/10.1016/s0960​-8524(02)00173​-6 calves, A.R.: Mass balance of pilot-scale pretreatment of sug-
42. Saha, K., Verma, P., Sikder, J., Chakraborty, S., Curcio, S.: arcane bagasse by steam explosion followed by alkaline delig-
Synthesis of chitosan-cellulase nanohybrid and immobilization nification. Bioresour. Technol. 111, 447–452 (2012). https​://doi.
on alginate beads for hydrolysis of ionic liquid pretreated sug- org/10.1016/j.biort​ech.2012.02.005
arcane bagasse. Renew. Energy 133, 66–76 (2019). https​://doi.
org/10.1016/j.renen​e.2018.10.014 Publisher’s Note Springer Nature remains neutral with regard to
43. Saha, K., Maheswari, R.U., Sikder, J., Chakraborty, S., da Silva, jurisdictional claims in published maps and institutional affiliations.
S.S., dos Santos, J.C.: Membranes as a tool to support biorefiner-
ies: applications in enzymatic hydrolysis, fermentation and dehy-
dration for bioethanol production. Renew. Sustain. Energy Rev.
74, 873–890 (2017). https​://doi.org/10.1016/j.rser.2017.03.015

13