Industrial Crops & Products 122 (2018) 98–106

Contents lists available at ScienceDirect

Industrial Crops & Products

journal homepage: www.elsevier.com/locate/indcrop

Comparison of fermentation strategies for ethanol production from olive T

tree pruning biomass
⁎
Felipe Fernandes-Klajn, Juan Miguel Romero-García, Manuel J. Díaz, Eulogio Castro
Dpt. Chemical, Environmental and Materials Engineering, Center for Advanced Studies in Energy and Environment (CEAEMA), Universidad de Jaén, Campus las
Lagunillas, 23071, Jaén, Spain

A R T I C LE I N FO A B S T R A C T

Keywords: Olive tree pruning is a widely available and low cost lignocellulosic residue generated every year, being a
Olive tree potential source for bioproducts and renewable fuels production. In this context, this work has as main objective
Bioethanol to propose an efficient scheme for the production of second generation ethanol from olive tree pruning biomass,
Slurry focusing on the evaluation of different detoxification strategies (activated carbon, ammonium hydroxide and
Hemicellulosic sugars
overliming) of the liquid fraction obtained after pretreatment, as well as different configurations of enzymatic
Enzymes
hydrolysis and fermentation (including hydrolysis and sequential or simultaneous fermentation, pre-sacchar-
ification or co-fermentation) of the pretreated olive tree pruning solid. The best results of each fraction were
tested in slurry, at varying initial solid loadings and aeration levels. The use of NH4OH 5N as a detoxification
method and the pre-saccharification and co-fermentation configuration without aeration are proposed, resulting
in 13.86 g ethanol/100 g raw material after 46 h of process.

1. Introduction fermentative microorganisms. Some detoxification methods can be

applied to reduce the concentration of the inhibitors, based on che-
Biomass from pruning of olive trees is one of the most widely mical, physical or biological treatments (Mussatto and Roberto, 2004;
available lignocellulosic materials particularly in the Mediterranean Vallejos et al., 2016).
countries. Pruning, performed after fruit harvest, produced every year a Among chemical detoxification methods, examples can be found
huge amount of biomass that must be adequately handled from both using Ca(OH)2 or NaOH (Martínez et al., 2000), activated carbon
economic and environmental viewpoint. Olive tree pruning biomass (Vallejos et al., 2016), NH4OH (Geddes et al., 2013) or organic solvents
(OTPB) has been reported to be a suitable raw material for the pro- (Mateo et al., 2013).
duction of bioethanol as well as number of bioproducts (Romero-García For the production process of ethanol from lignocellulose to be ef-
et al., 2014). OTPB is composed of leaves, thin branches and thick ficient, a complete hydrolysis of the sugars, both from cellulose and
branches in different proportions, depending on several factors such as hemicelluloses, and an efficient fermentation of the sugars to ethanol is
the tree age, cultivation techniques applied or olive production required. In addition, the process must be done with a low energy de-
(Romero-García et al., 2014). OTPB contains structural sugars as cel- mand. To achieve these objectives, high concentrations of initial solids
lulose (26.1–36.6% by weight), hemicellulose (20.0–25.0%), lignin in the fermentation medium are necessary to obtain high final con-
(17.9–25.0%) and extractives (14.4–31.4%) (Romero-García et al., centrations of ethanol, together with an efficient integration of the
2014). stages of the process that allows achieving high yields and low invest-
The bioconversion of OTPB into biofuels and other compounds re- ment in equipment. This integration will have benefits in terms of
quires a pretreatment step to modify its complex and recalcitrant lig- higher yields, production rate and higher ethanol concentrations, which
nocellulosic structure thus facilitating access to the enzymes in charge will translate into lower capital and operating costs.
of releasing monomeric sugars (Sebayang et al., 2017a,b). As a parallel There are several options for fermenting the sugars contained in the
result of the pretreatment, a number of compounds such as furan de- lignocellulosic materials (Manzanares et al., 2011). For example, sugars
rivatives, phenolics and weak organic acids are also produced and en- contained in the liquid fractions can be fermented separately of the
countered in the liquid phase issued from pretreatment. These com- glucose released by enzymatic hydrolysis from pretreated solids or the
pounds may exert an inhibitory effect on both the enzymes and the whole slurry can be fermented without previous separation. Other

⁎
Corresponding author.
E-mail address: [email protected] (E. Castro).

https://doi.org/10.1016/j.indcrop.2018.05.063
Received 6 March 2018; Received in revised form 22 May 2018; Accepted 25 May 2018
0926-6690/ © 2018 Elsevier B.V. All rights reserved.
F. Fernandes-Klajn et al. Industrial Crops & Products 122 (2018) 98–106

Fig. 1. Work flow for ethanol production from OTPB.

options include simultaneous or separate saccharification and fermen- aeration level have been revealed as relevant factors whose influence
tation of the pretreated solid; and the same configurations preceded by on the ethanol yields from OTPB has rarely been addressed.
an additional and previous saccharification (Tomás-Pejó, 2011). This The objective of this work is to compare the main configurations for
work compares the production of ethanol from OTPB under six different the fermentation of sugars released from OTPB, including detoxification
fermentation configurations, which are briefly described next. of the liquid issued from pretreatment. The study of solid loading and
The so called Sequential Hydrolysis and Fermentation (SHF) and the aeration level when using the whole slurry as fermentation medium is
Sequential Hydrolysis and CoFermentation (SHCF) configurations in- also addressed.
clude two main steps, i.e., first the enzymatic hydrolysis of the pre-
treated solid and, second, the fermentation to ethanol. While both steps 2. Materials and methods
are conducted under the best options for each of them (i.e., higher
temperature for the enzymatic hydrolysis and then lower temperature 2.1. Raw material and pretreatment
for the fermentation), the main drawback is the glucose accumulation
produced by the hydrolysis that can result in enzyme inhibition and The OTPB was collected in an olive orchard after fruit harvesting, in
reduced ethanol yield (Vohra et al., 2014). The main difference be- March 2016. Once in the laboratory, this material was dried at room
tween both configurations is that the SHCF considers the fermentation temperature, reaching 7.8% moisture. Subsequently, the pruning was
of all sugars present, that is, glucose and hemicellulosic sugars (while grinded using a Retsch SM 100 blade mill until obtaining a particle size
SHF is directed to just glucose). Cofermentation requires a micro- between 1 and 8 millimeters, which was stored until its use. The high
organism able to ferment both types of sugars or the use of several extractives content in olive tree pruning are mainly derived from its
microorganisms. Several advantages have been reported for the single amount of leaves. Olive tree leaves are a significant source of bioactive
microorganism approach, e.g., lower cost and operation time and re- phenolic compounds such as iridoids, flavonoids, triterpenes and phe-
duced inhibitory effects and contamination risk (Rastogi and nolic compounds (Xynos et al., 2014). On the other hand, during the
Shrivastava, 2017). pretreatment, extractives can be condensed or precipitated and produce
In Simultaneous Saccharification and Fermentation (SSF) and condensation reactions between lignin and degraded sugars (Ballesteros
Simultaneous Saccharification and CoFermentation (SSCF) sugars are et al., 2011). For these reasons, in this work an alkaline extraction stage
fermented as they are released, in a single step. This eliminates the was carried out previous to pretreatment (Burkhardt et al., 2013). In
inhibition issues due to sugar accumulation. On the negative side, en- order to minimize the presence of extracts in the raw material, after
zymes and microorganisms do not present the same optimal operation milling, the OTPB was subjected to an alkaline extraction with NaOH.
temperatures (around 50 °C for enzymatic hydrolysis versus 28–35 °C The treatment with alkaline compounds, such as NaOH, favors the ac-
for the fermentation). cess of the enzymes in the subsequent hydrolysis stage (Umagiliyage
Variations of the SSF and SSCF are the so called Presaccharification et al., 2015). The conditions to carry out this extraction were selected in
and Simultaneous Saccharification and Fermentation (PSSF) and the a previous assay as 110 °C, 30 min, 1.4% NaOH and 20% S/L ratio. In
Pressacharification and Simultaneous Saccharification and these conditions the presence of non-structural compounds in olive
CoFermentation (PSSCF), in which a rapid liquefaction step is included, pruning was reduced by 75%. The characterization of the raw material
previous to the hydrolysis and fermentation steps. This allows reducing and extracted material was made based on the technical report of the
the viscosity of the slurry, resulting in an increase of ethanol yields. National Renewable Energy Laboratory of the United States (NREL,
This procedure is specially indicated for high solid loadings (Moreno 2012).
et al., 2017). The extracted olive pruning was subjected to a combined
As far as the use of the whole slurry is concerned, solid loading and thermal + acid pretreatment, at a temperature of 164 °C, for 10 min, in

99
F. Fernandes-Klajn et al. Industrial Crops & Products 122 (2018) 98–106

an S/L ratio of 15% in the presence of H2SO4 at 0.9% (Martínez-Patiño dissolving a lyophilized sample in 250 mL of basal medium (MLB)
et al., 2017a). A 2L Parr reactor (Parr Instrument Company, USA), was containing (in g/L) yeast extract (5), NH4Cl (2), KH2PO4 (1),
used. After the pretreatment, liquids and solids were separated by fil- MgSO4·7H2O (0.3) and glucose (20).
tration and analysed. The solid was washed with water until reaching a
pH close to neutral and placed in an oven between 30 and 40 °C to
eliminate excess moisture. Fig. 1 shows a schematic diagram of the 2.4. Fermentation of the liquid fraction
experimental procedure.
Two 300 mL capacity containers, called “minifermentors”, were
prepared for each liquor (a total of 8), adding 150 mL of liquor and the
2.2. Detoxification of the liquid issued from pretreatment
AM1 medium components, with the exception of the ammonium salts.
Sodium metabisulfite (1.5 mM) was added just before the micro-
Four fractions were prepared to apply different detoxification stra-
organism, proving that sodium metabisulfite is able to reduce the
tegies; a fraction of the liquid was not subjected to any detoxification
toxicity of the fermentative medium in addition to improving the me-
process to check if there was a real need for this stage in the process,
tabolism of xylose by E. coli (Nieves et al., 2011a). The pH was adjusted
passing the liquid simply by a pH adjustment to the fermentative con-
to 6.5 and E.coli SL100 was inoculated. The minifermentors were placed
ditions and centrifugation to remove solids. For each method, between
inside a thermostatic bath on agitation plates and immersed in water up
320 and 350 mL of liquid obtained from pretreatment (also referred to
to above the volume of the fermentative medium, maintaining a uni-
as prehydrolyzate) were used, which had an initial pH of 1.84, all tests
form temperature of 37 °C. An automatic pH adjustment system with
being conducted in 1L Erlenmeyer flasks and with orbital shaking. A
2 M potassium hydroxide was used (more details in Romero-García
second fraction of the liquid was submitted to detoxification through
et al., 2016).
overliming (Martínez et al., 2000); the pH was raised up to 10 with Ca
(OH)2 and kept for 30 min, at 50 °C and 200 rpm agitation and then
centrifuged at 5000 rpm for 5 min for removal of any precipitate that
2.5. Characteristics of the configurations adopted for the fermentation of
has formed. The amount of liquor was measured at the end to determine
the solid
the volume loss of the process. Activated carbon (100 mesh particle
size) was another method used, which consisted of adding 5% by mass
For each configuration, five 100 mL Erlenmeyer flasks were used,
of activated carbon with respect to the volume of liquid to be detox-
one triplicate of the sample being a cellulose control and another en-
ified, maintaining for 30 min at 50 °C and 200 rpm. The liquid was re-
zyme dissolution control. In the sampling, 1 mL was taken from each
covered through a vacuum filtration process. The last detoxification
flask, which was centrifuged at 10,500 rpm for 10 min and passed
method used consisted of a 5N solution of NH4OH, adding a certain
through a 0.45 μm filter before being taken to HPLC for determination
amount until reaching pH 9, observing the same conditions of tem-
of its composition. The S/L ratio used was 5% w/v (1.25 g substrate and
perature, time and speed of agitation of the previous methods (Geddes
25 mL of enzymatic solution) and the 0.05 M citric acid-citrate buffer at
et al., 2013). The recovery of the liquor was done through centrifuga-
pH 4.8. The cellulolytic complex used was the Cellic CTec-2, in a ratio
tion at 5000 rpm for 5 min.
of 15 FPU/g substrate, supplemented by β-glucanase in a ratio of 10%
of the Cellic Ctec-2 enzyme volume (Novozymes A/S, Denmark). The
2.3. Microorganisms and growth conditions initial concentration of microorganism was 0.25 g/L, S. cerevisiae or E.
coli SL100 as appropriate. Before inoculating E. coli SL100 the pH was
Due to the presence of hexoses (C6 compounds like glucose or ga- adjusted to 6.5 with 2 M KOH, the components of the medium AM1 and
lactose) and pentoses (C5 compounds such as xylose, arabinose or 1.5 M sodium metabisulfite (in proportion of 1 mL/L sample) were
mannose) in the raw material, two different microorganisms were as- added. The fermentation was carried out at 37 °C, 150 rpm and in the
sayed for the fermentation step. case of E. coli there was discontinuous adjustment of the pH with 2 M
Ethanologenic Escherichia coli SL100 was used to ferment all kinds KOH. A summary with the stages and conditions of temperature, agi-
of sugars. This strain was obtained from a modified E. coli MM170 tation and pH of each configuration is presented in Table 1.
which was reported as able to produce ethanol from both pentoses and
hexoses (Geddes et al., 2013). Inocula were prepared on AM1 medium
(Martínez et al., 2007) whose composition (in g/L) is xylose (16), glu- 2.5.1. SHF and SHCF – sequential hydrolysis and (Co) fermentation
cose (10), (NH4)2HPO4 (2.63), NH4H2PO4 (0.87), MgSO4·7H2O (0.246), For both configurations, the process began with the enzymatic hy-
KCl (0.149), betaine (0.117), Citric acid (0.1) and trace elements (dif- drolysis (EH). The agitation of the EH was 150 rpm, with a temperature
ferent concentrations). Growth conditions were 37 °C and 200 rpm for of 50 °C and a duration of 72 h. Following the EH, inoculum at 4% v/v
24 h. with culture of S. cerevisiae (SHF) was added, while SHCF was in-
Saccharomyces cerevisiae “Ethanol Red”, a kind gift from Fermentis, oculated with E. coli SL100 at 4% v/v.
France, was used for glucose fermentation. The inoculum was prepared

Table 1
Process variables (temperature, agitation and pH) for the different process configurations assayed on the pretreated solids.
Configuration Steps Variables

SHF Enzymatic hydrolysis 50 °C, 150 rpm, 4.8

Fermentation with S. cerevisiae 37 °C, 150 rpm, 4.8-5
SHCF Enzymatic hydrolysis 50 °C, 150 rpm, 4.8
Fermentation with E. coli 37 °C, 150 rpm, 6.5
SSF Enzymatic hydrolysis + Fermentation with S. cerevisiae 37 °C, 150 rpm, 4.8-5
SSCF Enzymatic hydrolysis + Co-Fermentation with E. coli 37 °C, 150 rpm, 6.5
PSSF Presaccharification 50 °C, 150 rpm, 4.8
SSF 37 °C, 150 rpm, 4.8-5
PSSCF Presaccharification 50 °C, 150 rpm, 4.8
SSCF 37 °C, 150 rpm, 6.5

100
F. Fernandes-Klajn et al. Industrial Crops & Products 122 (2018) 98–106

2.5.2. SSF and SSCF – simultaneous saccharification and (Co) 2.9. Analytical methods
fermentation
The SSF process, conducted at 37 °C, 150 rpm and pH 4.8 of the The raw material and the pretreated solid fractions were char-
citrate buffer, was put on the enzyme assembly and the culture of S. acterised by the NREL/TP-510-42618 procedure, for the determination
cerevisiae at 4% v/v was inoculated. The SSCF had as difference the of cellulose and hemicellulosic sugars and the lignin content (NREL,
adjustment of the pH to 6.5 (not citrate buffer), addition of the com- 2012). Concerning the liquid fractions, the composition in glucose and
ponents of the AM1 medium, 1.5 M sodium metabisulfite (1 mL/L) and hemicellulosic sugars (glucose, xylose, galactose, mannose and arabi-
inoculation of the E. coli to the 4% v/v. nose) and inhibitory compounds (acetic acid, formic acid, furfural and
HMF) was determined by HPLC. Total phenols were determined
2.5.3. PSSF and PSSCF – presaccharification and simultaneous through spectrophotometry. The methodology followed for the de-
saccharification and (Co) fermentation termination of total phenols was based on the procedure established by
Here, a presaccharification (EH) was performed for 6 h before the Singleton and Rossi (1965). An aliquot of 0.5 mL of each sample was
inoculation of the different microorganisms (a sample is taken at the mixed with 3.75 mL of water, 0.25 mL of Folin-Ciocalteau reagent
end of these 6 h and the temperature is adjusted to 37 °C). For in- (1:1 v/v) and 0.5 mL of 10% Na2CO3 w/v. After one hour in the dark,
oculation, unlike the previous configurations, the culture is centrifuged the absorbance of the mixtures was measured at 765 nm and a cali-
and resuspended with the presaccharified, PSSF S. cerevisiae and PSSCF bration curve for the gallic acid was constructed, the results being the
E. coli. total phenolic content in the prehydrolyzates expressed as g gallic acid
equivalent.
2.6. Preparation and detoxification of slurry The HPLC used for the measurement of inhibitors, sugars and
ethanol was an Agilent Technologies 1260 model (Santa Clara, CA,
The slurries were prepared at 5% w/v and at 8.8% w/v (con- USA) with Aminex HPX-87H column (Bio-Rad Hercules, CA, USA) op-
centration obtained after pretreatment) from the pre-treated olive erating at 65 °C with 5 mM H2SO4 solution as eluent and operating flow
pruning and the liquid obtained in the pretreatment, leaving the mix- of 0.6 mL/minute. This column is not able to separate the xylose, ga-
ture at rest for about 12 h to moisten the solid material better. The lactose and mannose, instead the sum of the three sugars is shown,
detoxification with NH4OH 5N has been applied to the slurries due to which will be called in the present work XGM (Xylose, Galactose and
their better results obtained when fermenting the liquid fraction in Mannose). A second HPLC (Waters, Milford, USA), equipped with
comparison to the other detoxification methods. The pH of the mixture CARBOSep CHO-782 PB column (Transgenomic Inc, Omaha, USA),
was adjusted to 9 with constant stirring 150 rpm and temperature of operating at 70 °C with ultrapure water as mobile phase and at a flow of
50 °C for 30 min, the pH being lowered to 5 with 96% sulfuric acid to 0.6 mL/minute was used to measure the exact composition of the XGM.
adjust to the conditions of enzymatic hydrolysis. In order to determine the concentration of oligomeric sugars in the
prehydrolyzates, a post-hydrolysis was carried out with 3% H2SO4 w/v
2.7. Enzymatic hydrolysis + fermentation of the slurry at 120 °C for 20 min, comparing the sugar content before and after the
post-hydrolysis.
After analyzing the results of the treatment stage of the solid frac-
tion and the composition of the prehydrolyzate, the SHCF and PSSCF
were selected as configurations, making a duplicate of each one. To 3. Results and discussion
each minifermentor was added the 5% w/v slurry previously prepared
and detoxified, containing 150 mL of liquor and 7.5 g dry solid. The 3.1. Composition of raw and pretreated materials
enzymatic hydrolysis was carried out under the same conditions de-
scribed for the pretreated solid. For the SHCF the EH (50 °C) lasted 72 h The composition of OTPB before and after the extraction stage are
while for the 6 h PSSCF. The fermentation occurred at 37 °C, 300 rpm summarized in Table 2. The raw olive pruning biomass contained about
and pH 6.5. The concentration of E. coli SL100 cells was 0.5 g cell per 45% of total sugars, with glucose being the main sugar with 25.82% by
liter of fermentative medium. weight of which 6.34% is found as non-structural within the extractive
fraction. Xylose is the main pentose of the hemicellulosic fraction, ac-
2.8. Enzymatic hydrolysis + fermentation in the best configuration in the counting for 86.6%. It should be noted the high proportion of ex-
highest capacity reactor and with slurry aeration tractives (about 30%), a value within the range normally found for
olive pruning biomass, which according to Ballesteros et al. (2011) can
In this stage the Dasgip Bioblock fermentation equipment be between 14 and 31% depending on the concentration of leaves and
(Eppendorf AG, Germany) was used, consisting of four 1-l fermentors
that allow temperature control and agitation through the DASGIP Table 2
TC4SC4 controller; pH and aeration control with the DASGIP PHPO Composition of OTPB before and after alkaline extraction.
controller, and pump control with the DASGIP MP8 controller. The Composition (%)
software DASware allows the modification of all the variables and
makes the capture and storage of the data. Compounds Raw Material Extracted solids
For this stage the PSSCF configuration was selected, in such a way
Glucose 25.82 ± 0.05 31.11 ± 0.43
that one of the fermenters received a slurry of 5% w/v (400 mL of liquor Hemicellulosic sugars 14.27 ± 0.76 16.36 ± 0.28
and 20 g dry solid) and the other a slurry of 8.8% w/v (400 mL of liquor Xylose 10.98 ± 0.43 12.35 ± 0.03
and 35.2 g dry solid). The slurries were initially detoxified with 5N Galactose 1.34 ± 0.13 1.33 ± 0.09
Arabinose 1.35 ± 0.16 2.68 ± 0.16
NH4OH, followed by the addition of Cellic Ctec-2 enzymes that per-
Mannose 0.6 ± 0.04 0 ± 0.00
formed a presaccharification for 6 h at 50 °C and a pH of 4.8. After this Lignin 17.43 ± 0.43 25.81 ± 0.99
time, the pH was adjusted to 6.5, the temperature to 37 °C and the AIL 15.03 ± 0.40 23.65 ± 0.74
agitation to 250 rpm, the E. coli SL100 being inoculated at a con- ASL 2.40 ± 0.03 2.16 ± 0.23
centration of 0.5 g cell for each liter of fermentative medium. The pH Ash 3.47 ± 0.05 0.44 ± 0.13
Acetil groups 2 ± 0.04 0 ± 0.00
control was performed automatically with 2 M KOH and the aeration
Extractives 29.46 ± 0.13 7.34 ± 0.17
maintained at a value of 0.1 vvm with manual rotameters (Geddes Glucose in extractives 6.340.04 0.23 ± 0.02
et al., 2013).

101
F. Fernandes-Klajn et al. Industrial Crops & Products 122 (2018) 98–106

the type of extraction method used. Cara et al. (2008) found a greater concentrations of acetic acid lower than 1 g/L for any of the 17 con-
amount of extractives (31.4%) and ash compared to the present work, ditions tested in their work.
but a concentration of acid insoluble lignin (AIL), acid soluble lignin
(ASL), cellulose and hemicellulosic sugars very similar to the compo- 3.2. Detoxification of liquids from pretreatment
sition detailed in Table 2. Martínez-Patiño et al. (2017a) found about
24.8% extractives and concentrations of glucose, AIL and acetyl groups Fig. 2 shows the effects on the liquors produced from pretreatment
also very similar to the present work (27.9%, 16.1% and 2%, respec- because of the different detoxification methods. In general, no method
tively). On the contrary, these authors observed a higher content in reduced significantly the concentration of initial sugars, but did have an
hemicellulosic sugars (21.3%), indicating some of the differences that effect on the formic acid and the HMF. The methods of detoxification
can be found when working with a raw material such as OTPB. that led to a rise of the pH above the fermentative conditions increased
The extraction with NaOH removed practically 75% of the extracts, the levels of acetic acid by 13.19% (overliming) (Díaz-Villanueva et al.,
and increased the concentrations of sugars and AIL. The glucose retired 2012) and 11.42% (NH4OH). In addition, these two methods of de-
from the extracted solid as part of the extractive fraction, is not actually toxification were not as efficient in the elimination of inhibitory com-
lost, i.e., not degraded, and could eventually be recovered. Ballesteros pounds as activated carbon, which decreased the concentration of
et al. (2011) and Martínez-Patiño et al. (2017a) reported a similar be- formic acid (60.5%), acetic acid (32.2%), HMF (100%), furfural
havior when using water for 1 h in an autoclave at 121 °C, where they (92.3%) (Díaz et al., 2009) and total phenols (79.8%), being the only
also verified a residual presence of extractives in the raw material of one to lower acetic acid levels. NH4OH was the second most efficient
8.5 ± 0.7% and 8.9 ± 0.5%, respectively. method, with a significant reduction of HMF (97.1%). Overliming,
As far as the pretreated solid obtained after filtration of the slurry is considered one of the most economic and high performance detox-
concerned, glucose was the main component (40.55%) followed by AIL ification techniques (Jönsson et al., 2013), was the least efficient
(about 38.55%). The hemicellulosic sugars such as arabinose, mannose technique in detoxifying the hydrolysates of OTPB (Mateo et al., 2013).
and galactose were practically solubilized during the pretreatment, not The detoxification with activated carbon, despite being the most
being identified in the solid. Still, a small amount of xylose remained at efficient method in the elimination of inhibitors, resulted in poorer
a concentration of 5.22%. This was also observed by Martínez-Patiño results as can be seen in Fig. 3, where the main parameters of the fer-
et al. (2017b) when they tested an acid pretreatment in OTPB using mentation process are displayed as a function of the detoxification
H2SO4 at different concentrations, reaching values from 1.9 to 17.3% of method. In the literature, no works were found that used activated
xylose in the pretreated solid. Negro et al. (2014) found similar results carbon for OTPB. For the overliming, Martínez-Patiño et al. (2017a)
to the present work when pre-treating the OTPB with steam explosion attained an ethanol yield between 70 and 88% of the theoretical one
impregnated with 1% w/w of H3PO4 at different temperatures, showing using strain E. coli MM160, while the present work reached a little more
a glucose concentration between 45.8% and 46.9%, xylose between 2.3 than 80% using strain SL100. According to the authors, low ethanol
and 4.7%, AIL between 44.5 and 42.9% and galactose, arabinose and yields can be explained by incomplete consumption of the XGM frac-
mannose close to zero, with the values corresponding to the tested tion, the same as reported by Geddes et al. (2011). In the tests carried
temperatures of 195 °C and 175 °C, respectively. out, this same phenomenon was observed, with an amount close to
The composition of the liquor obtained in the combined thermal- 10 g/L of XGM not consumed, the same amount reported by Martínez-
acid pretreatment, using H2SO4 0.9% (w/v) and 15% loading of solid is Patiño et al. (2017a). This behavior, according to Geddes et al. (2011) is
shown in Table 3. The values in parentheses indicate the concentration mainly due to the action of furfural, which prevents the complete
of sugars obtained in the post-hydrolysis, evidencing the presence of consumption of the XGM by the microorganism. However, the reason
oligomers in the liquor. The XGM is composed of 80.34% xylose, why there was not a high productivity in ethanol of the hydrolyzate by
18.89% galactose and 0.78% mannose. The galactose and arabinose are overliming for the present work does not fit in this explanation, since
in a concentration between 3 and 5 g/L due to an almost complete the furfural concentrations were low and practically of the same order
solubilization of the hemicellulose during the pretreatment, not being of the hydrolyzate detoxified with NH4OH, which obtained the best
visualized in the solid. A relatively high presence of glucose was ob- results. For some reason, there was a growth in the concentration of
served, probably due to a hydrolysis of starch fractions during the glycerol, which is a co-product typically produced by the yeast during
pretreatment of olive pruning and to a lesser degree of the glycosides the metabolization of sugar or as a result of some bacterial con-
present in the leaves, such as oleuropein (Ballesteros et al., 2011). tamination, ultimately affecting the final ethanol yield (Zabed et al.,
As can be seen in Table 3, in addition to sugars, other compounds 2017).
are present in the liquid obtained after pretreatment such as acetic acid The liquor detoxified with NH4OH produced the best results in the
(in a concentration of 0.23 g/L), coming from the hydrolysis of acetyl fermentation step, even though it is not the most efficient in the elim-
groups, in addition to formic acid (1.17 g/L), furfural (0.65 g/L) and ination of inhibitor compounds. It is important to note that the high
hydroxymethylfurfural (HMF) (0.05 g/L) resulting from the degrada- concentrations of acetic acid shown in Fig. 2 were not an impediment to
tion of sugars. Similar observations were made by Ballesteros et al. E. coli SL100, since it is a strain highly resistant to this inhibitor, and it
(2011), while Martínez-Patiño et al. (2017a) did not observed can even use acetate as a carbon source (Zaldivar and Ingram, 1999).
Palmqvist and Hahn-Hägerdal (2000) had previously reported that
ethanol production could be stimulated with high concentrations of
Table 3 acetic acid and low concentrations of furanic compounds, in a kind of
Composition of liquids obtained after pretreatment (liquors). Numbers in
positive synergistic effect of the inhibiting compounds. According to
brackets mean values obtained after post-hydrolysis.
Mussato and Roberto (2004), it is not possible to define a maximum
Composition (g/L) concentration of each inhibitor for a given microorganism precisely
because the inhibition effect depends very much on the type and con-
Glucose 8.07 (10.86) ± 0.01
XGM 15.33 (16.21) ± 0.01 centration of each compound, the type of medium and the possible
Arabinose 5.08 ± 0.15 synergy between compounds, being feasible high productions of
Formic acid 1.17 ± 0.05 ethanol in the presence of inhibitors.
Acetic acid 0.23 ± 0.01 Jennings and Schell (2011) also found a higher ethanol yield from
Hydroxymethylfurfural (HMF) 0.05 ± 0.00
liquors of pretreated corn stover detoxified with NH4OH, which
Furfural 0.65 ± 0.03
Phenolics 2.50 ± 0.05 reached 72% of the theoretical maximum and was about 7% more ef-
ficient than detoxification with overliming. The authors attributed this

102
F. Fernandes-Klajn et al. Industrial Crops & Products 122 (2018) 98–106

Fig. 2. Composition of liquors compared to initial composition after detoxification. Total sugars obtained as the sum of glucose, xylose, galactose, mannose and
arabinose.

Fig. 4. Ethanol yield (g etanol/100 g raw material) as a function of process

configuration. SHF, SSF and PSSF using S. cerevisiae “Ethanol Red”; SHCF, SSCF
Fig. 3. Ethanol yield, productivity and percentage of theoretical ethanol yield and PSSCF using E. coli SL100 as fermenting microorganism; * 72 h EH in op-
from the different liquors. timal conditions before fermentation; ** 6 h EH in optimal conditions before
fermentation.
greater efficiency to the removal of phenols or another toxic compound
between pH 6.8–8, contrary to a certain extent to the proposal of PSSCF, SSCF and SSF. By operating the hydrolysis and fermentation
Alriksson et al. (2006) that suggest a greater effectiveness of NH4OH stage under optimal conditions and after a long period of EH, higher
thanks to its ability to eliminate furanic and phenolic compounds, since yields were already expected from the SHF and SHCF, which in this case
Jennings and Schell (2011) did not observe an effective reduction of produced more ethanol in the first 24 h than all other configurations
furfural nor of HMF. An interesting point is that these authors observed after 72 h of fermentation process. The SHCF is the only one that pre-
an inhibition of the fermentative microorganism Zymomonas mobilis sented a decrease in the production of ethanol. For some reason, the
with detoxification at pH 9, attributed to a higher concentration of microorganism began to consume ethanol as a carbon source after the
inorganic salts resulting from a high amount of NH4OH to reach pH 9, initial 24 h. A possible hypothesis is that the fermentation reached its
which also occurred in the present work but had no effect on the E. coli maximum before the first sample was taken, since the consumption of
SL100 used. ethanol by E. coli indicates, in theory, that there are no more sugars
Geddes et al. (2013), working at different pH and using E. coli available in the medium to use as a source of energy, changing its
SL100, did not observe inhibition with detoxifications at pH 8 or 9, metabolism to use as a substrate the previously produced main product.
being even these pH those that resulted in a higher productivity in A slower growth of the ethanol yield can also be observed for the SSF
ethanol. This phenomenon can be attributed to the reduction of some and PSSF compared with their corresponding configurations, eviden-
unidentified toxic compounds, better use of NH4OH as a nitrogen cing the almost complete consumption of glucose by S. cerevisiae close
source by E. coli, or some synergistic effect. These same authors (Geddes to 24 h of the process, while E. coli SL 100 progressively increased
et al., 2013), regarding the consumption of sugars, observed a pre- performance due to its ability to metabolize some hemicellulosic sugars.
ference of E. coli SL100 to consume the monomers in the following The results shown in Fig. 4 differ in comparison to the only work
order: glucose > arabinose > galactose > xylose in the hydrolyzate found in the literature that reports on different fermentation config-
detoxified with NH4OH with the medium at pH 9. The chromatographic urations with OTPB. According to Manzanares et al. (2011), in the
monitoring of the present work allowed us to observe that glucose and configurations studied from the combined thermal-acid pretreatment
arabinose were consumed in their entirety in the first hours of fer- (H2SO4 at 1%, 180 °C and 10 min of process time) the SSF provided the
mentation, while xylose only began to decrease its concentration sig- best performance in ethanol compared to the SHF and PSSF at all solid
nificantly when the other two sugars were in values close to zero, in loadings (9–23%). Already in the case of pruning treated through hot
accordance to the mentioned reference. water under pressure, for the highest solid loading (17 and 23%) the
SHF showed better results than the SSF and the PSSF, with a better
3.3. Comparative summary of the fermentations of the pretreated solid mixing conditions between the hydrolyzed material and the micro-
organisms, while at the lowest load (9%) the SSF was far superior to the
The different configurations used for the fermentation of the solid other two configurations, reaching the highest performance in ethanol
fraction of OTPB are summarized in Fig. 4. The largest ethanol pro- (76%).
duction was verified for the SHF and SHCF, followed by the PSSF, Based on this, two configurations of the fermentation of the

103
F. Fernandes-Klajn et al. Industrial Crops & Products 122 (2018) 98–106

Table 4 process was carried out on a larger scale.

Production of glucose and xylose from the enzymatic hydrolysis of the SHCF The chromatographic analysis showed that initially all the available
configuration with the OTPB slurry. glucose was consumed in 16 h, being thereafter taken advantage of by
Time (h) g glucose/100 g raw material g XGM/100 g raw material the microorganism to the same extent that it is being released from the
hydrolysis. When the glucose level reached a balance of production and
0 7.85 ± 0.18 13.10 ± 0.31 consumption, xylose then began to decrease its concentration more
3 8.31 ± 1.58 12.24 ± 1.74
acutely, being consumed in 48 h, at which time it was being consumed
6 10.27 ± 0.34 14.06 ± 0.36
10 10.98 ± 0.23 14.16 ± 0.19 as it was released. After 92 h, the ethanol concentration began to de-
24 11.14 ± 0.27 14.28 ± 0.23 crease due to the end of the process of enzymatic hydrolysis and change
48 12.15 ± 0.61 15.03 ± 0.60 of the metabolism of E. coli, which began to consume the ethanol pre-
72 12.62 ± 0.29 15.40 ± 0.25
sent in the medium. It is interesting that there is a part of the arabinose
that was not consumed by the microorganism until the end of the
process. Geddes et al. (2011) also reported that a part of sugars (xylose)
was not consumed by the E. coli itself after 240 h of fermentation, which
was attributed to inhibition by furfural, more accentuated in the pre-
sence of xylose. Although the levels of HMF and furfural of the present
work remained at zero after 4 h of processing, indicating that in some
way E. coli was unable to access this small fraction of sugar.
Numerically, at the end of the process the SHCF had a higher yield,
which reached 16.99 g ethanol per 100 g raw material in 120 h, com-
pared to 14.77 g ethanol from the PSSCF after 92 h (Fig. 5). This higher
performance was obtained after 72 h of EH in optimal conditions, while
the PSCFS had just a period of 6 h under these conditions, the perfor-
mance of both configurations being practically equal up to 48 h, so that,
if one considers the result obtained at this time for the PSSCF and the
highest value obtained with the SHCF, the difference is only 19%; that
Fig. 5. Comparison of SHCF and PSSCF configurations for the OTPB slurry. was the reason why it was decided to proceed to the next stage with the
*72 h EH in optimal conditions; **6 h EH in optimal conditions. PSSCF.

pretreated solid with the highest potential for utilization of the hexoses 3.5. Enzymatic hydrolysis and fermentation of the slurry in the best
and pentoses were selected (SHCF and PSSCF), mixing the liquid and configuration at different solid loadings and aeration rates
solid fractions together to verify the production potential in ethanol
considering all the sugars available and using E. coli SL100. The fermentation with the use of aeration for the different con-
centrations of pretreated solids can be seen in Fig. 6. It is observed that
for 5% w/v the maximum production of ethanol was established after
3.4. Enzymatic hydrolysis and co-fermentation of the slurry at the highest 16 h, with 11.4 g ethanol per 100 g raw material (48.9% of theoretical
ethanol performance configurations yield). From there, the levels began to fall until reaching less than half
at 144 h of processing. The chromatographic analysis allowed to see
Table 4 shows the time evolution of the enzymatic hydrolysis, while clearly that the consumption rate of glucose and xylose were similar,
Fig. 5 compares ethanol yield for SHCF and PSSCF. In Table 4, as can be and that all the sugars were in concentrations close to zero just at 16 h
seen at 0 h, there were 7.9 g glucose and 13.1 g xylose for every 100 g of the process, just at the highest point of the ethanol curve, which then
raw material. After 72 h of enzymatic hydrolysis, glucose production began to fall. This reflects that the production of ethanol occurred from
increased only 4.8 g (referred to the same amount of raw material), a the sugars produced in the pre-saccharification stage and from those
production well below that verified for SHF and SHCF of the solid that already available in the hydrolyzed liquid, so that the enzyme was not
had already more than doubled this amount in 48 h of process. The able to produce more monomeric carbohydrates for E. coli SL100 con-
most probable hypothesis for this, in agreement with Xiao et al. (2004), sumption.
is that xylose (and other sugars, such as mannose and galactose) have Nieves et al. (2011b) obtained similar results regarding the con-
significant inhibitory effects on the action of cellulases during enzy- sumption of sugars, also observing a more efficient consumption of
matic hydrolysis and can lead to an incomplete conversion of carbo- xylose and a considerable decrease in fermentation time. These authors
hydrates, already being demonstrated by other authors (Kim et al.,
2011; Qing et al., 2010). In addition, enzymatic cocktails that in-
corporate enzymes capable of degrading xylose (as used in this study)
can have detrimental effects on glucose yields (Modenbach and Nokes,
2013).
The results of the fermentation show, for the SHCF, that E. coli
SL100 was able to efficiently convert to ethanol practically all the su-
gars of the medium, reaching 94.5% yield after 120 h of processing. The
most significant jumps in the production of ethanol were verified be-
tween 8 and 24 h and 24–50 h, when 92.6% of all ethanol had already
been produced (Fig. 5).
With respect to the PSSCF, the results for the slurry show better
yields compared to the fermentation performed on the solid, for both
the PSSF and the PSSCF. After 16 h of processing, the yield had already
practically doubled in relation to the same configurations made in the
solid, setting its maximum at 92 h. There was still a very insignificant Fig. 6. Ethanol yield at two different solid loadings in the slurry PSSCF con-
evolution after 40 h, this time being selected in the event that the figuration with aeration. ** 6 h EH in optimal conditions.

104
F. Fernandes-Klajn et al. Industrial Crops & Products 122 (2018) 98–106

verified an increase between 25% and 30% with the use of air for the
PSSCF configuration, contrary to that shown in the present work. This
increase caused by the injection of air, despite not being well known, is
attributed to the production of reducing compounds such as NADPH,
which produces a decrease in the effects of inhibitors such as furfural
and allows cell growth in conjunction with the increase in ethanol
production. From what is observed in the present work, E. coli has
developed well with aeration but a greater increase in ethanol yield was
not possible due to a paralysis of the enzymatic activity.
Already for an initial concentration of 8.8% w/v, in which there was
a higher concentration of initial sugars, the maximum yield in ethanol
was observed at 16 h, which was reduced until reaching almost half at
144 h of processing, behavior very similar to that performed for 5%
solids. The reason is the same as in the previous process, with a con-
sumption by the microorganisms of the available sugars of the hydro- Fig. 7. Comparison of ethanol production over time in the slurry PSSCF con-
lyzed liquid and of the six hours of pre-saccharification with the in- figuration at 5% solid loading with and without air injection. ** 6 h EH in
optimal conditions.
terruption of the functioning of the enzyme when the conditions were
modified. Glucose and xylose consumption rates were practically the
same one in comparison to the other, decreasing their concentration to
practically zero between 16 and 24 h of the process. This allowed a
higher solids loading to be used without the fermentation time being
longer.
Yields decreased compared to fermentation with 5% initial solids,
and are far from some data found in the literature. Nieves et al.
(2011b), with 6 h of pre-saccharification using E. coli MM170, 10%
solids loading and 0.01 vvm of air, obtained yields up to five times
higher than the PSSCF of slurry at 8.8% S/L, not reporting any type of
enzymatic inhibition until the end of the process (at 144 h). Liu et al.
(2016a) and Liu et al. (2016b) found a higher glucose consumption and
higher fermentative yields using aeration up to 0.02 vvm, observing a
decrease in the production of ethanol and residues of sugars not con-
sumed only above this volume of air. Varela et al. (1992) reported that
the rate of production of ethanol increased up to 3.3 times using small
amounts of oxygen in the fermentation of the serum using Kluyver-
omyces fragilis, and also the decrease in yield when large amounts of air
are used, which probably favored the biomass formation and co-pro-
duct generation.
In general, a greater amount of solids in the slurry can produce an
inhibition by excess of product in the enzymes, lowering the sugar
yields, in addition to produce viscosity problems that hinder the agi-
tation and the contact between the components of the medium. Also, an
excessive initial concentration of sugars can cause an interruption of the
fermentation by increasing the osmotic stress on the microorganism
(Zabed et al., 2017). However, these justifications are best applied
when working with high loads of solids, generally above 10–15% Fig. 8. Mass balance for the proposed scheme of ethanol production from
OTPB.
(Modenbach and Nokes, 2013) and do not explain exactly the differ-
ences observed in the present work, which were expected to be lower
and that ethanol yields were higher for both concentrations. sections, the use of NH4OH 5N did not degrade the sugars while pro-
Finally a comparison is drawn between the PSCFS at 5% initial so- viding fermentative yields more efficient than the other detoxification
lids for the slurry with and without the use of aeration, shown in Fig. 7. methods. The configuration of PSSCF without aeration has shown to
Compared to the process with air injection, the PSSCF without aeration provide good yields with shorter fermentation times, which are very
had a 30% higher yield in ethanol, with 14.77 g ethanol/100 g raw attractive characteristics for a possible up scale of the process. It is
material at 92 h compared to 11.41 g ethanol/100 g raw material pro- important to note that the use of aeration has a great potential because
duced in 16 h. In spite of the lower yield, the potential of the injection it increases the rate of consumption of sugars and decreases the fer-
of air in low quantities revealed to be able to decrease the fermentation mentation time. Further efforts must be directed to investigate the
time significantly, producing in 16 h what the process without aeration reasons why the interruption of enzymatic activity occurred.
took around twice. More research should be done focusing on the Finally, Fig. 8 summarizes the mass balance for the optimal scheme.
reason why the unexpected inhibition of the enzyme occurred with the It can be seen that a maximum of 20.45 g ethanol can be obtained from
injection of air, as the decrease in fermentation time is vital for a the amount of sugars present in 100 g pruning, producing after 46 h of
greater deployment of the facilities of second generation bioethanol. process 68% of the potential ethanol.

3.6. Mass balance for the production of second generation ethanol from 4. Conclusions
olive tree pruning biomass
The main conclusions obtained in this work are the following:
The proposed scheme for the production of second generation
ethanol from OTPB is shown in Fig. 8. As demonstrated in the previous - The extraction stage carried out with 1.4% NaOH at 110 °C for

105
F. Fernandes-Klajn et al. Industrial Crops & Products 122 (2018) 98–106

30 min eliminated about 75% of the extractives present in OTPB, Liu, C.-, Hao, X.-, Lin, Y.-, Bai, F.-, 2016a. Redox potential driven aeration during very-high-
contributing significantly to the sequence of the process. gravity ethanol fermentation by using flocculating yeast. Sci. Rep. 6.
Liu, Z., Ho, S.-, Hasunuma, T., Chang, J.-, Ren, N.-, Kondo, A., 2016b. Recent advances in yeast
- 5N NH4OH was the method of detoxification that promoted the best cell-surface display technologies for waste biorefineries. Bioresour. Technol. 215, 324–333.
fermentative yields for OTPB compared to activated carbon and Manzanares, P., Negro, M.J., Oliva, J.M., Saéz, F., Ballesteros, I., Ballesteros, M., Cara, C.,
Castro, E., Ruiz, E., 2011. Different process configurations for bioethanol production from
overliming. pretreated olive pruning biomass. J. Chem. Technol. Biotechnol. 86, 881–887.
- In the fermentation of the slurry, the PSSCF obtained similar yields Martínez, A., Rodriguez, M.E., York, S.W., Preston, J.F., Ingram, L.O., 2000. Effects of Ca(OH)2
(14.28 g ethanol/100 g raw material) in a shorter process time (6 h treatments ('overliming') on the composition and toxicity of bagasse hemicellulose hy-
drolysates. Biotechnol. Bioeng. 69, 526–536.
of EH and 48 h of fermentation) than SHCF (15.74 g ethanol/100 g Martínez, A., Grabar, T.B., Shanmugam, K.T., Yomano, L.P., York, S.W., Ingram, L.O., 2007.
raw material with 72 h of EH and 50 h of fermentation). Low salt medium for lactate and ethanol production by recombinant Escherichia coli B.
Biotechnol. Lett. 29, 397–404.
- The increase in initial solid loading in the PSSCF of the slurry with
Martínez-Patiño, J.C., Romero, I., Ruiz, E., Cara, C., Romero-García, J., Castro, E., 2017a.
the use of aeration (0.1 vvm) produced a decrease in the yield of Design and optimization of sulfuric acid pretreatment of extracted olive tree biomass using
ethanol production (5% S/L, 11.41 g ethanol/100 g raw material response surface methodology. BioResources 12.
Martínez-Patiño, J.C., Ruiz, E., Romero, I., Cara, C., López-Linares, J.C., Castro, E., 2017b.
versus 8.8% S/L, 7.81 g ethanol/100 g raw material). Combined acid/alkaline-peroxide pretreatment of olive tree biomass for bioethanol pro-
- The use of aeration in the PSSCF of the slurry increased the rate of duction. Bioresour. Technol. 239, 326–335.
consumption of sugars by the E. coli SL100 with a consequent re- Mateo, S., Roberto, I.C., Sánchez, S., Moya, A.J., 2013. Detoxification of hemicellulosic hy-
drolyzate from olive tree pruning residue. Ind. Crops Prod. 49, 196–203.
duction of the fermentative time (16 h), for both the 5% and the Modenbach, A.A., Nokes, S.E., 2013. Enzymatic hydrolysis of biomass at high-solids loadings –
8.8% solid loading. However, the yields obtained were lower a review. Biomass Bioenerg 56, 526–544.
Moreno, A.D., Alvira, P., Ibarra, D., Tomás-Pejó, E., 2017. Production of ethanol from lig-
(11.41 g ethanol/100 g raw material) compared to the slurry nocellulosic biomass. In: Fang, Z., Smith, J.R., Qi, X. (Eds.), Production of Platform
without air injection (14.77 g ethanol/100 g raw material). This Chemicals from Sustainable Resources. Springer, Singapore, pp. 375–410.
may be due to an unexpected interruption of the enzymatic activity Mussatto, S.I., Roberto, I.C., 2004. Alternatives for detoxification of diluted-acid lignocellulosic
hydrolyzates for use in fermentative processes: a review. Bioresour. Technol. 93, 1–10.
after 16 h of processing. NREL, 2012. National Renewable Energy Laboratory, Washington D.C., EE.UU. Biomass
- Considering the above, the following optimal scheme for the pro- Compositional Analysis Laboratory Procedures, NREL 2017.
Negro, M.J., Alvarez, C., Ballesteros, I., Romero, I., Ballesteros, M., Castro, E., Manzanares, P.,
duction of bioethanol from OTPB is proposed: extraction with 1.4%
Moya, M., Oliva, J.M., 2014. Ethanol production from glucose and xylose obtained from
NaOH and pretreatement with the combined thermal-acid method steam exploded water-extracted olive tree pruning using phosphoric acid as catalyst.
(H2SO4 0.9%, 110 °C for 30 min); then, a detoxification of the slurry Bioresour. Technol. 153, 101–107.
Nieves, I.U., Geddes, C.C., Miller, E.N., Mullinnix, M.T., Hoffman, R.W., Fu, Z., Tong, Z.,
with 5N NH4OH (pH 9, 50 °C, 30 min) and the use of the PSSCF Ingram, L.O., 2011a. Effect of reduced sulfur compounds on the fermentation of phosphoric
without aeration with 6 h of enzymatic hydrolysis at 50 °C, pH 4.8 acid pretreated sugarcane bagasse by ethanologenic Escherichia coli. Bioresour. Technol.
and fermentation for 40 h at pH 6.5 and 37 °C. Under these condi- 102, 5145–5152.
Nieves, I.U., Geddes, C.C., Mullinnix, M.T., Hoffman, R.W., Tong, Z., Castro, E., Shanmugam,
tions, the amount of ethanol that can be obtained from 100 g K.T., Ingram, L.O., 2011b. Injection of air into the headspace improves fermentation of
pruning is 13.86 g phosphoric acid pretreated sugarcane bagasse by Escherichia coli MM170. Bioresour.
Technol. 102, 6959–6965.
Palmqvist, E., Hahn-Hägerdal, B., 2000. Fermentation of lignocellulosic hydrolysates. I:
Acknowledgements Inhibition and detoxification. Bioresour. Technol. 74, 17–24.
Qing, Q., Yang, B., Wyman, C.E., 2010. Xylooligomers are strong inhibitors of cellulose hy-
drolysis by enzymes. Bioresour. Technol. 101, 9624–9630.
This work has been partially financed by the Junta de Andalucía Rastogi, M., Shrivastava, S., 2017. Recent advances in second generation bioethanol produc-
(Proyecto de Excelencia AGR-6103) and the Spanish Ministerio de tion: an insight to pretreatment, saccharification and fermentation processes. Renew.
Economía y Competitividad (Ref. ENE2014-60090-C2-2-R, including Sustain. Energy Rev. 80, 330–340.
Romero-García, J.M., Niño, L., Martínez-Patiño, C., Álvarez, C., Castro, E., Negro, M.J., 2014.
FEDER funds). Biorefinery based on olive biomass. State of the art and future trends. Bioresour. Technol.
159, 421–432.
Romero-García, J.M., Martínez-Patiño, C., Ruiz, E., Romero, I., Castro, E., 2016. Ethanol pro-
References duction from olive stone hydrolysates by xylose fermenting microorganisms. Bioethanol 2,
51–65.
Alriksson, B., Sjöde, A., Nilvebrant, N.-, Jönsson, L.J., 2006. Optimal conditions for alkaline Sebayang, A.H., Hassan, M.H., Chyuan, O.H., Dharma, S., Silitonga, A.S., Kusumo, F., Mahlia,
detoxification of dilute-acid lignocellulose hydrolysates. Appl. Biochem. Biotechnol. 130, T.M.I., Bahar, A.H., 2017a. Optimization of reducing sugar production from Manihot gla-
599–611. ziovii starch using response surface methodology. Energies 10, 35–48.
Ballesteros, I., Ballesteros, M., Cara, C., Sáez, F., Castro, E., Manzanares, P., Negro, M.J., Oliva, Sebayang, A.H., Hasan, M.H., Chyuan, O.H., Dharma, S., Bahar, A.H., Silitonga, A.S., Kusumo,
J.M., 2011. Effect of water extraction on sugars recovery from steam exploded olive tree F., 2017b. Enzymatic hydrolysis using ultrasound for bioethanol production from Durian
pruning. Bioresour. Technol. 102, 6611–6616. (Durio zibethinus) seeds as potential biofuel. Chem. Eng. Trans. 56, 553–558.
Burkhardt, S., Kumar, L., Chandra, R., Saddler, J., 2013. How effective are traditional methods Singleton, V.L., Rossi, J.A., 1965. Colorimetry of total phenolics with phosphomolybdic-phos-
of compositional analysis in providing an accurate material balance for a range of softwood photungstic acid reagents. Am. J. Enol. Vitic. 16, 144–158.
derived residues? Biotechnol. Biofuels 6, 90–99. Tomás-Pejó, E., 2011. Fermenting microorganisms for 2nd generation bioethanol production.
Cara, C., Ruiz, E., Oliva, J.M., Sáez, F., Castro, E., 2008. Conversion of olive tree biomass into In: Min-Tze, L. (Ed.), Bioprocess Sciences and Technology. Nova Science Publishers,
fermentable sugars by dilute acid pretreatment and enzymatic saccharification. Bioresour. Hauppage, pp. 187–208.
Technol. 99, 1869–1876. Umagiliyage, A.L., Choudhary, R., Liang, Y., Haddock, J., Watson, D.G., 2015. Laboratory scale
Díaz, M.J., Ruiz, E., Romero, I., Cara, C., Moya, M., Castro, E., 2009. Inhibition of Pichia stipitis optimization of alkali pretreatment for improving enzymatic hydrolysis of sweet sorghum
fermentation of hydrolysates from olive tree cuttings. World J. Microbiol. Biotechnol. 25, bagasse. Ind. Crops Prod. 74, 977–986.
891–899. Vallejos, M.E., Chade, M., Mereles, E.B., Bengoechea, D.I., Brizuela, J.G., Felissia, F.E., Area,
Díaz-Villanueva, M.J., Cara-Corpas, C., Ruiz-Ramos, E., Romero-Pulido, I., Castro-Galiano, E., M.C., 2016. Strategies of detoxification and fermentation for biotechnological production
2012. Olive tree pruning as an agricultural residue for ethanol production: fermentation of of xylitol from sugarcane bagasse. Ind. Crops Prod. 91, 161–169.
hydrolysates from dilute acid pretreatment. Span. J. Agric. Res. 10, 643–648. Varela, H., Ferrari, M.D., Loperena, L., Lareo, C., 1992. Effect of aeration rate on the alcoholic
Geddes, C.C., Mullinnix, M.T., Nieves, I.U., Peterson, J.J., Hoffman, R.W., York, S.W., Yomano, fermentation of whey by Kluyveromyces fragilis. Microbiologia 8, 14–20.
L.P., Miller, E.N., Shanmugam, K.T., Ingram, L.O., 2011. Simplified process for ethanol Vohra, M., Manwar, J., Manmode, R., Padgilwar, S., Patil, S., 2014. Bioethanol production:
production from sugarcane bagasse using hydrolysate-resistant Escherichia coli strain feedstock and current technologies. J. Environ. Chem. Eng. 2, 573–584.
MM160. Bioresour. Technol. 102, 2702–2711. Xiao, Z., Zhang, X., Gregg, D.J., Saddler, J.N., 2004. Effects of sugar inhibition on cellulases and
Geddes, C.C., Mullinnix, M.T., Nieves, I.U., Hoffman, R.W., Sagues, W.J., York, S.W., ß-glucosidase during enzymatic hydrolysis of softwood substrates. Appl. Biochem.
Shanmugam, K.T., Erickson, J.E., Vermerris, W.E., Ingram, L.O., 2013. Seed train devel- Biotechnol. – Part A Enzyme Eng. Biotechnol. 115, 1115–1126.
opment for the fermentation of bagasse from sweet sorghum and sugarcane using a sim- Xynos, N., Papaefstahiou, G., Gikas, E., Argyropoulou, A., Aligiannis, N., Skaltsounis, A.L.,
plified fermentation process. Bioresour. Technol. 128, 716–724. 2014. Design optimization study of the extraction of olive leaves performed with pres-
Jönsson, L.J., Alriksson, B., Nilvebrant, N.-, 2013. Bioconversion of lignocellulose: inhibitors surized liquid extraction using response surface methodology. Sep. Purif. Technol. 122,
and detoxification. Biotechnol. Biofuels 6. 323–330.
Jennings, E.W., Schell, D.J., 2011. Conditioning of dilute-acid pretreated corn stover hydro- Zabed, H., Sahu, J.N., Suely, A., Boyce, A.N., Faruq, G., 2017. Bioethanol production from
lysate liquors by treatment with lime or ammonium hydroxide to improve conversion of renewable sources: current perspectives and technological progress. Renew. Sustain.
sugars to ethanol. Bioresour. Technol. 102, 1240–1245. Energy Rev. 71, 475–501.
Kim, Y., Ximenes, E., Mosier, N.S., Ladisch, M.R., 2011. Soluble inhibitors/deactivators of Zaldivar, J., Ingram, L.O., 1999. Effect of organic acids on the growth and fermentation of
cellulase enzymes from lignocellulosic biomass. Enzyme Microb. Technol. 48, 408–415. ethanologenic Escherichia coli LY01. Biotechnol. Bioeng. 66, 203–210.

106