Bioresource Technology 169 (2014) 284290

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Effect of liquid hot water pre-treatment on sugarcane press mud

methane yield
Lisbet Mailin Lpez Gonzlez a,, Ileana Pereda Reyes b, Jo Dewulf c, Jrn Budde d, Monika Heiermann d,
Han Vervaeren e
a

Universidad de Sancti Spritus Jos Mart Prez (UNISS), Centro de Energa y Procesos Industriales (CEEPI), Avenida de los Mrtires 360, CP 60100 Sancti Spritus, Cuba
Instituto Superior Politcnico Jos Antonio Echeverra (Cujae), Centro de Estudio de Ingeniera de Procesos (CIPRO), Calle 114 No. 11901 e/Rotonda y Ciclova,
Marianao CP 19390, La Habana, Cuba
c
Ghent University, Department of Sustainable Organic Chemistry and Technology, Coupure Links, 653, 9000 Ghent, Belgium
d
Leibniz-Institute for Agricultural Engineering Potsdam-Bornim, Max-Eyth-Allee 100, 14469 Potsdam, Germany
e
Laboratory of Industrial Water and Eco-technology (LIWET), Faculty of Bioscience Engineering, Ghent University Campus Kortrijk, Graaf Karel de Goedelaan 5, B-8500
Kortrijk, Belgium
b

h i g h l i g h t s
 Improved biomethanation by LHW pre-treatment compared with untreated press mud.
 A 33% COD solubilisation resulted in the best methane yield (>55% increase).
 Increase of methane yield by a maximum of 63% at 150 C for 20 min.
1

 Highest furfural concentration of 1214.17 mg L

was found at 200 C for 5 min.

 HAc release was inuenced more by temperature than by pre-treatment severity.

a r t i c l e

i n f o

Article history:
Received 28 April 2014
Received in revised form 27 June 2014
Accepted 28 June 2014
Available online 5 July 2014
Keywords:
Anaerobic digestion
LHW pre-treatment
Press mud
Filter cake
Inhibitors

a b s t r a c t
Sugarcane press mud was pretreated by liquid hot water (LHW) at different temperatures (140210 C)
and pre-treatment times (520 min) in order to assess the effects on the chemical oxygen demand (COD)
solubilisation, inhibitors formation and methane yield. The experimental results showed that a high
degree of biomass solubilisation was possible using LHW. Higher methane yields were obtained at lower
severities (log(Ro) = 2.172.77) with (i) mild temperatures (140150 C) and long contact times
(12.5 min, 20 min) or (ii) mild temperatures (175 C) with short contact time (2 min). The highest
increase in methane yield (up to 63%) compared to the untreated press mud was found at 150 C for
20 min. At temperatures of 200 C and 210 C, low methane efciency was attributed to the possible
formation of refractory compounds through the Maillard reaction.
2014 Elsevier Ltd. All rights reserved.

1. Introduction
Sugarcane cultivation has increased dramatically with a world
production level of about 1.7 billion tons in 2011. The global
expansion of sugarcane has been in response to a rising sugar
demand for food consumption and as a feedstock for ethanol
production (FAO, 2013). That increase leads to larger quantities
of by-products and waste generated by sugar production. One of
them is press mud, a solid residue obtained by the vacuum ltration of the settled cake in the clarication process of the cane sugar

Corresponding author. Tel.: +53 (41)336118.

E-mail address: lisbet@uniss.edu.cu (L.M. Lpez Gonzlez).
http://dx.doi.org/10.1016/j.biortech.2014.06.107
0960-8524/ 2014 Elsevier Ltd. All rights reserved.

juice. Each ton of milled cane generates 2845 kg of press mud

(Velarde et al., 2004).
Press mud is a potential source for methane production. However, the presence of a resilient biomass matrix (9.03%, 21.67% and
17.23% on dry basis for lignin, hemicelluloses and cellulose, respectively) and a low COD solubilisation (around 21%) explains its low
efciency (40%) for bio-conversion processes (Lopez et al., 2013).
Previous pre-treatment studies have focused on improving the
digestibility of lignocellulosic material by many different means:
mechanical, thermal, chemical, and through combined measures
(Hendriks and Zeeman, 2009; Taherzadeh and Karimi, 2008;
Zheng et al., 2014). Among these methods, liquid hot water
(LHW) pre-treatment utilises water at elevated temperatures as
the only solvent (usually between 120 C and 230 C) and various

L.M. Lpez Gonzlez et al. / Bioresource Technology 169 (2014) 284290

pressures, conditions at which water exists in subcritical conditions (Nitsos et al., 2013). Water under high pressure and temperature can penetrate the biomass, hydrate cellulose, and remove
hemicellulose and a share of lignin. LHW pre-treatment enhances
the accessible and susceptible surface area of the cellulose and
makes it more accessible to hydrolytic enzymes. The major advantages offered by this technique are: no additional chemicals
needed and no corrosion resistant materials required for the
hydrolysis reactors. Furthermore, since much less amounts of
chemical are needed for hydrolysed neutralisation, lesser amounts
of neutralisation residues are generated when compared to other
processes (Taherzadeh and Karimi, 2008).
A signicant drawback is the formation of phenolic compounds,
as well as furfural and hydroxymethylfurfural (HMF). These byproducts are commonly toxic and inhibit the growth of bacteria
and archea (Hendriks and Zeeman, 2009). The optimisation of
operating parameters is therefore required in order to reduce the
formation of these inhibitory compounds, and obtain a pre-treated
substrate that can be easily decomposed to biogas.
To our knowledge, LHW pretreatment on press mud have not
been published yet. LHW pre-treatment has been studied as part
of the biochemical conversion of lignocelluloses into biogas from
a variety of feedstocks with positive results, including the solid
and liquid cattle manure (Budde et al., 2014) and agricultural
waste such as wheat straw (Chandra et al., 2012b; Menardo
et al., 2012), rice straw (Chandra et al., 2012a; Menardo et al.,
2012), oil palm empty fruit bunches (O-Thong et al., 2012), sunower oil cake (Fernandez-Cegri et al., 2012), sunower stalks
and palm oil mesocarp bre (Costa et al., 2013), sorghum bagasse
and ensiled sorghum forage (Sambusiti et al., 2013), etc. Milder
temperatures (120170 C) with times between 5 and 30 min,
were most effective for substrates as wheat straw, solid cattle
manure, liquid cattle manure and sunower oil cake. On the contrary, LHW pre-treatments at higher temperatures (200230 C,
1015 min) were needed in order to observe a higher methane
yield when using rice straw and oil palm empty fruit bunches. As
can be seen, depending on the chemical compositions and structural properties of the different biomass materials the effectiveness
of LHW pre-treatment varies considerably.
The purpose of the present study is to evaluate the inuence of
time and temperature on the effectiveness of LHW pre-treatment
of press mud for methane production, using a central composite
design (CCD) statistical experimental design.
2. Methods
2.1. Substrate characterisation
Fresh press mud (2013 harvest) was provided from the Sugar
Mill Melanio Hernndez (Sancti Spiritus, Cuba). Press mud was
air-dried and stored in plastic bags at 4 C until use. Dry press
mud contained 90.48% and 72.22% of total solids (TS) and volatile
solids (VS), respectively. Furthermore, the compositional analysis
resulted in 11.34% cellulose, 27.13% hemicellulose and 9.30% lignin
on dry weight basis. Concentrations of sugar, protein and
extractable material containing wax and fat of the press mud were
determined as 8.14%, 11.05%, and 9.32% on dry weight basis,
respectively. The fresh press mud also contains various micronutrients such as nitrogen, phosphorous, potassium, calcium, magnesium, manganese and zinc (Lopez et al., 2013).
2.2. Analytical methods
Total chemical oxygen demand (tCOD), TS, VS, ashes and pH
were determined according to standard methods (APHA et al.,

285

1995). Liquid samples were centrifuged for 30 min at 6000 rpm

and further used for soluble COD (sCOD) and volatile fatty acids
(VFA) analyses. COD analysis was carried out by standard closed
reux, colorimetric method 5220 D (APHA et al., 1995). VFA (acetic
acid, propionic acid, butyric acid, valeric acid and iso-valeric acid)
were determined by gas chromatography using a previously
published methodology (Lopez et al., 2013).
Liquid aliquots were analysed for carbohydrates mono-and
disaccharides (xylose, arabinose, fructose, galactose, mannose, glucose, sucrose, lactose, cellobiose and maltose). The sugars were
determined by a gas chromatograph (GC Varian 3380) coupled to
a FID: CP Sil-5CB column (25 m L  0.32 lm internal diameter  0.25 lm particle size). Oven initial T = 120 C, T
ramp = 20 C min1 to a nal temperature of 290 C held for
3 min, split/splitless inlet at 280 C. Hydrogen was the carrier
gas, detector T = 325 C.
Lignin and structural carbohydrates were determined according
to the NREL procedure (Sluiter et al., 2008), using GCFID as
pointed out above.
Other organic compounds found in the liquid product of hydrothermally pre-treated biomass, such as gallic acid, hydroxymethylfurfural (HMF), furfural, vanillic acid, syringic acid, p-coumaric acid
and ferulic acid were analysed by High Performance Liquid Chromatography (HPLC)-Agilent 1100 Series-equipped with a diode
array detector-Agilent 1200 Series-. The analysis was done using
an Inertsil-ODS column, 5 ll injection volume, 1 ml/min ow rate
in a gradient run using 1% v/v acetic acid in distilled water and
1% v/v acetic acid in methanol as mobile phase.
Press mud Soxhlet extraction was performed with water followed by ethanol. The aqueous and ethanolic extracts were dried
under vacuum in a rotary evaporator at 80 and 60 C respectively.
The obtained solids were further dried to a constant weight at
60 C, and maintained under vacuum in a desiccator with tetraphosphorus decaoxide. Wax and fats were quantied as ethanol
extracts. Proteins were calculated from the Total Kjeldahl nitrogen
(TKN) content, using a conversion factor of 6.25.

2.3. Pre-treatment conditions

The LHW press mud pre-treatment assays were conducted in a
600 ml Mini Reactor System, Model number 4568 (Parr Instruments, Moline, USA) (Budde et al., 2014). A press mud sample of
100 g was mixed with 500 g of deionized water with a liquid to
solid ratio of 5.5 and heated at varying temperatures (140
210 C) under constant stirring (350 rpm) for different reaction
times (223 min). The pressure values were equivalent to the temperature-specic saturated water vapour pressure. The heating
rate was 3 C/min. Table 1 shows detailed experimental conditions
according to a CCD setup. Experiments were replicated two times
for factorial and axial points, and four times for center points. In
total, 29 trials were performed.
After completion of the pre-treatment time the heater was
removed and the reactor was cooled down to less than 50 C by
immersing it into room temperature water. The pre-treated press
mud slurry was taken out of the reactor and stored in containers
at 4 C for subsequent biochemical methane potential test. A subsample was used to separate liquid and solid fraction by centrifugation. The solid fraction was washed with hot deionized water at
60 C several times until the ltrate had a neutral pH, dried in air
and stored at 4 C until used for further analysis. The liquid product
was passed through a 0.2 lm lter and stored at 20 C for further
analysis. The COD solubilisation (SCOD) was calculated as sCOD/
tCOD and expressed in %.
The severity factor Ro was determined for each experiment. This
parameter quanties the hydrothermal pre-treatment combined

286

L.M. Lpez Gonzlez et al. / Bioresource Technology 169 (2014) 284290

Table 1
Range of variables for the central composite design.
Variable

Axial point (1.41)

Low level (1)

Central level (0)

High level (+1)

Axial point (+1.41)

Temperature, T (C)
Time, t (min)

140
2

150
5

175
12.5

200
20

210
23

effect of time and temperature on the matrix, and it is dened by

Eq. (1).

Ro t  e

Tr  Tb
14:75

t is the pre-treatment time in min, Tb is the base temperature

(100 C), Tr is the pre-treatment temperature in C and 14.75 is
the conventional energy of activation assuming the overall reaction
is hydrolytic and the overall conversion is rst order (Overend and
Chornet, 1987).

SCOD, sVFA, ymax, and k, as well as to estimate the size of the experimental errors. The means between pre-treatment pairs were statistically compared using a test of multiple ranks (Duncans test).
Data normality and homogeneity of variances were determined
using the KolmogorovSmirnov and the Fischer tests respectively.
The data sets were analysed using Statgraphics Centurion XVI.2
with a condence interval of 95%.
3. Results and discussion
3.1. Effect of LHW pre-treatment on COD solubilisation

2.4. Anaerobic digestion

2.4.1. Biochemical methane potential (BMP) test
Anaerobic digestion experiments with untreated and pre-treated press mud for measuring methane potentials were carried
out in accordance with (VDI 4630, 2006). The temperature was
set to 37.5 C. Schott bottles (500 mL) with four port lids were used
as reactors. One port was connected to a graduated cylinder lled
with acidied water at pH 2, in order to measure the cumulative
biogas production by liquid displacement. From a second lid port
with a septum, headspace gas samples from each bottle were collected (5 mL pressure-tight syringe) and analysed by gas chromatography. The other two ports were kept hermetically closed.
Each trial was conducted for 30 days. Once daily, reactors were
shaken manually. Each reactor received a mixture of substrate and
inoculum in a 2:1 ratio, calculated on the basis of volatile solid content (Lopez et al., 2013; VDI 4630, 2006). The digestate collected
from an anaerobic digestion plant (Inagrovzw, Belgium) fed with
swine manure and maize silage was used as inoculum after one
week of degassing. The inoculum used in the tests contained
3.56% TS with a 65.90% VS content on dry basis. The VFA/Alkalinity
ratio was 0.27. Eleven batch assays were performed in triplicate
and controls, including inoculum (blank reactor) and untreated
press mud (control reactor).
The gas pressure was calculated according to the liquid column
height and subtracted from the atmospheric pressure before standardization (273 K and 101.29 kPa). Methane yields are given in
STP mL per g of initial volatile solids content in the untreated press
mud (mL CH4 g1 VS1
in ).
2.4.2. Kinetics models
The rst-order kinetic model has been widely applied to assess
biogas or methane production performance for various biomass
types under different conditions (Lopez et al., 2013). In this study,
the biogas production of raw and pre-treated biomass was
assumed to follow a rst-order rate of decay according to Eq. (2):

yt ymax :1  ekt

2
1

VS1
in ),

where y(t) is the cumulative methane yield (mL CH4 g

ymax
is the ultimate methane yield taken as residence time tends to innity, k is the kinetic constant (d1), t is time (days). Experimental
data were tted by a non-linear regression analysis (Statgraphics
Centurion XVI.2).
2.5. Statistical analyses
One-way ANOVA (Analysis of Variance) was used to test significant differences in mean values for the response variables

Different temperatures and contact times were applied to

sugarcane press mud through LHW pre-treatment using central
composite design. COD solubilisation increased linearly with the
pretreatment severity, with the exclusion of the point with highest
log(Ro) = 4.3. The most severe pre-treatment of log(Ro) = 4.35
(210 C, 12.5 min) did not yield the highest SCOD. The lowest solubilisation found at that severity could be due to the recondensation
of previous solubilised components from lignin, as well as to the
degradation of organic matter possibly lost in the gas phase when
the reactor was open.
The values for SCOD ranged between 28.79% and 42.75%, in comparison with 23.98% for the untreated press mud (Table 2). In all
cases values for SCOD were signicantly higher than the SCOD for
untreated press mud. The highest SCOD were obtained when
200 C during 5 and 20 min were applied with 41.19 and 42.75%,
respectively. As a result increases up to 78% with respect to
untreated press mud were found for those conditions.
At lower temperatures (140 C, 150 C) dissolved compounds
can be due to easily extractable compounds present in biomass like
sugars, proteins and fats. The extractive form a protective layer,
which has to be washed out before the hydrolysis reaction starts
(Ingram et al., 2009).
Higher levels of solubilisation have been reported at elevated
temperatures (175220 C) in comparison with milder temperatures (140170 C) (Fernandez-Cegri et al., 2012; Nitsos et al.,
2013). During the hydrothermal pre-treatment of beech wood
the COD solubilisation was increased up to 40% when 220 C for
15 min were applied (Nitsos et al., 2013). Soluble compounds were
also increased by the hydrothermal pre-treatment of sunower oil
cake at 200 C (Fernandez-Cegri et al., 2012). During pre-treatment, water can penetrate into the biomass cell structure, hydrating cellulose, solubilising hemicellulose, and slightly removing
lignin. LHW pre-treatment is highly effective for enlarging the
accessible and susceptible surface area of cellulose and improving
cellulose degradability to microbes and enzymes (Zheng et al.,
2014).
3.2. Effect of LHW pre-treatment on the chemical composition of the
liquid fraction
The liquid fractions were analysed for pH, acetic acid, monomeric sugars, furfural, HMF, organic acids and aromatic acids from
solubilised lignin, as described in the experimental section.
Xylose, galactose, mannose and arabinose were the main
hemicellulose components recovered and degraded to monomers
after oligomers formation. Xylose, as a monomer, was present at
low concentrations (0.030.14 g 100 g1 press mud, Fig. 1). A

287

L.M. Lpez Gonzlez et al. / Bioresource Technology 169 (2014) 284290

Table 2
Composition of soluble fraction (sCOD), total COD (tCOD), COD solubilisation (SCOD), acetic acid (HAc) and pH in the hydrolysates.
Run

T (C)

Untreated
1
2
3
4
5
6
7
8
9

t (min)

150
150
200
200
140
210
175
175
175

sCOD (g L1)

log(Ro)
0.00
2.17
2.77
3.64
4.25
2.26
4.35
2.49
3.57
3.31

5
20
5
20
12.5
12.5
2
23
12.5

tCOD (g L1)

38 1
46 1
53 1
69 2
48 1
44 1
38 2
45 2
47 1
46 2

HAc (mg L1)

SCOD (%)
a

159 48
139 1
139 1
167 35
113 11
136 15
132 8
144 2
132 7
135 4

pH

23.9
33.2cd
37.9f
41.2g
42.7g
32.6cd
28.8b
31.5c
35.7e
34.4d

265 13
392a 8
577a 11
4408d 186
6132e 83
394a 9
5787e 263
1607b 340
2530c 64
2127bc 12

6.2 0.1
6.1 0.1
5.5 0.1
4.2 0.1
4.0 0.1
5.9 0.0
3.9 0.0
4.9 0.0
4.5 0.1
4.7 0.1

Data expressed as mean standard deviation.

Groups with different superscripts differ (p < 0.05).

af

degradation of pentoses, hexoses, and the lignin present. These

products can include furfural, acetic acid, HMF, and formic acid
among others. Furans, organic acids and phenols can be inhibitors
to the anaerobic digestion process according to its concentration
and availability in the medium (Yu et al., 2010b).
In the experiments presented, several by-products were found
in the hydrolysates from the degradation of sugars (furfural and
HMF), aliphatic acids (mainly acetic acid) and phenolic acids produced by the solubilisation of the lignin. Their relationship with
the pre-treatment severity differs from each other (Figs. 1 and 2).
The acetic acid concentration increases almost linearly with
increasing severity factor, reaching values of 6132 and
5787 mg L1 for pre-treatment severities of 4.25 (200 C for
20 min) and 4.35 (210 C for 12.5 min), respectively (Table 2).
The acetic acid formation was better correlated with temperature
(R2 = 0.96) than with the severity pre-treatment (R2 = 0.88), showing its high dependence from temperature (Eqs. (3) and (4)). Hot
water cleaves hemiacetal linkages of hemicellulose and liberates
O-acetyl and other acid moieties from hemicellulose so that acetic
acid is formed (Perez et al., 2007). Related to increased formation
of VFA as the temperature rises, pH of the liquid hydrolysate
decreased from 6.21 to 3.96 (Table 2).

HAc 2599:2  logRo  5657:7

pH 0:030 T 10:19

180
Aromatic acids (mg L-1)

maximum value of xylose monomer concentration was detected at

log(Ro) = 3.64 (200 C, 5 min) with 0.37 g 100 g1 press mud.
The objective of the LHW pre-treatment is to solubilise mainly
the hemicellulose to make the cellulose more accessible to
hydrolysis and to avoid the formation of inhibitors by keeping
the pH between 4 and 7. Maintaining the pH between these limits
minimises the formation of monosaccharides, and therefore the
formation of degradation products that can further catalyse hydrolysis of the cellulosic material during pre-treatment (Hendriks and
Zeeman, 2009).
However, as glucose, fructose and sucrose were present in raw
press mud extracts, sugar degradation started at lower severities
(log(Ro) = 2.17) with a pH higher than 4 (Fig. 1). A signicant
decrease of sucrose was found when more severity was applied
(log(Ro) = 2.49). Since sucrose is hydrolysed to glucose and
fructose, this reaction was completed at higher severities (log(Ro)
P 3.31), according to low amount of sucrose found in the hydrolysed (Fig. 1).
Fructose and glucose were degraded to form a wide range of
products, including: anhydrosaccharides, 5-hydroxymethyl-2-furaldehyde (HMF), furfural and organic acids (Fig. 3). In addition,
the Maillard reaction could take place accompanied by formation
of a class of compounds known as Maillard products. These are
coloured recalcitrant compounds produced by polymerisation at
elevated temperatures of low molecular weight intermediates,
such as carbohydrates and amino compounds (Bougrier et al.,
2008). As it was observed a dark brown colouration in the pre-treated samples with log(Ro) P 3.31, it can be assumed this was the
case.
During the pre-treatment water penetrates the biomass cells,
and both hemicellulose and lignin are solubilised by this liquid
hot water acting as an acid (Kim et al., 2009; Yu et al., 2010b). This
process results in a solubilised product, consisting primarily of
oligosaccharides derived from hemicellulose, lignin and a minor
amount of cellulose, but also by-products that are seen to be
formed within the liquid fraction of this process due to the

160
140
120
100
80

3
4

A
Ferulic acid
p-coumaric
Syringic acid
Vanillic Acid
Gallic acid

60
40
20
0
0.00

2.17

2.26

2.49

2.77

3.31

3.57

3.64

4.25

4.35

log(Ro)

Mono and Di-saccharides

(g100g-1press mud)

lactose

sucrose

glucose

mannose

galactose

fructose

arabinose

xylose

12
10
8
6
4
2

Furfural and HMF (mgL-1)

1400

14

HMF

Furfural

1200
1000
800
600
400
200
0
0.00

0
0.00

2.17

2.26

2.49

2.77

3.31
log (Ro)

3.57

3.64

4.25

4.35

Fig. 1. Composition of Mono and Di-saccharides in the hydrolyzates.

2.17

2.26

2.49

2.77
3.31
log(Ro)

3.57

3.64

4.25

4.35

Fig. 2. Concentration of aromatic acids (A) and furans (B) according to the pretreatment severity (log(Ro)).

288

L.M. Lpez Gonzlez et al. / Bioresource Technology 169 (2014) 284290

Among the aromatic acids analysed in the hydrolysate,

p-coumaric acid predominated in all samples, whereas ferulic acid
was detected at lower levels (Fig. 2). These data agree with previous results of sugarcane bagasse characterisation (Masarin et al.,
2011). The small quantity of ferulic acid released during this stage
could be due to its localisation in the cell wall. Both acids
decreased signicantly for higher severities (log(Ro) of 4.25 and
4.35). The hydroxycinnamic acids could be ester-linked to methylglucurono-arabino-xylans and solubilised at mild severities and
the remaining hydroxycinnamic acids are linked to lignin through
ether linkages involving the phenolic oxygen of the hydroxycinnamic acid (Xu et al., 2005). Syringic and vanillic acid concentrations did not differ at milder severities (log(Ro) = 2.173.57), but
when the temperature was 200 C or higher severities (log(Ro)
P 3.64), they increased due to lignin solubilisation. Gallic acid
behaviour was not clearly related with the pre-treatment severity.
Other phenolic compounds like vanillin and syringaldehyde were
not determined; they have been reported in other studies using
similar conditions (Masarin et al., 2011).
Furfural and HMF were found in all the samples analysed.
Higher concentrations of furfural were released at log(Ro) = 3.31
(175 C,
12.5 min),
log(Ro) = 3.57
(175 C,
23 min)
and
log(Ro) = 3.64 (200 C, 5 min), with a marked increase at 200 C
for 5 min (Fig. 2). In the case of HMF concentrations varied from
34.08 to 362 mg L1, with the maximum value at log(Ro) = 2.49
(175 C, 2 min) (Fig. 2). The capability of the hydrothermal
treatment of producing considerable amounts of furfural in the
liquid product under certain experimental conditions has been
reported before (Abril et al., 2012; Hendriks and Zeeman, 2009;
Nitsos et al., 2013; Suryawati et al., 2009). Nevertheless, HMF concentration remains lower due to limited cellulose solubilisation
and consequent glucose formation (Hendriks and Zeeman, 2009).
However, some amount of HMF was found to milder severities
formed from a readily soluble glucose and sucrose proportion that
accounts for up to 8.94% of the untreated press mud on dry basis.
Consequently, the heat-up time (lasting from 60 to 80 min,
depending on nal process temperature) may also be responsible
for the formation of degradation products, especially from
extracted soluble glucose.
The maximum furfural concentration determined was 1.21 g L1
(0.73 g/100 g press mud) at 200 C for 5 min. This result is in accordance with the 0.72 g/100 g switchgrass calculated from the values
reported by Suryawati et al. (2009), when switchgrass was pretreated at 200 C for 10 min. The furfural values reported in literature at temperatures from 200 to 220 C for 515 min are between
0.23.1 g/100 g biomass (Nitsos et al., 2013; Perez et al., 2007;
Suryawati et al., 2009; Yu et al., 2010a). The furfural concentration
obtained here represents about 9.02% of the theoretical stoichiometric yield for the conversion of all available C5 sugars into furfural,
which is lower than the 30% reported by Nitsos et al. (2013).

Furfural concentration decreased with increasing pre-treatment

time at 200 C for 20 min, which could be due to further degradation of furans to aldehydes and organic acids (Yu et al., 2010b). A
similar tendency was observed by Suryawati et al. (2009) at
210 C prolonging the time from 10 to 15 min.
3.3. BMP tests results
The hydrothermal pre-treatment of press mud increased the
methane yield for pretreatments with severities up to
log(Ro) = 3.64, with signicant differences when compared with
the untreated press mud (Table 3). Higher methane yields were
obtained at lower severities (log(Ro) = 2.172.77), which correspond to temperatures of 140 C for 12.5 min, 150 C for 5 and
20 min, and 175 C for 2 min. The methane yields ranged from
321.95 to 340.80 mL CH4 g1 VS1
in , when those pre-treatments
were applied (Table 3). The maximum yield was attained at
150 C and 20 min, with a 63% increase over the untreated press
mud. Specically, low pre-treatment temperatures do not break
signicantly lignocellulosic bonds, but promote organic matter
hydrolysis and breakdown during anaerobic digestion (Wang
et al., 1997). This effect benets organic matter degradation by
anaerobic bacteria, and consequently improves methane
production.
Time spans higher than 2 min at 175 C led to a reduction of
methane yield up to 11%, which suggests the formation of complex
recalcitrant substances and/or toxic compounds at more severe
pre-treatment conditions. That behaviour was harsher at the
highest pre-treatment severities (3.64, 4.25 and 4.35), which corresponds to 200 C for 5 min, 200 C for 20 min and 210 C for
12.5 min, respectively (Fig. 3). In comparison with the untreated
press mud, the methane yield increased only by 20% and 2% when
200 C for 5 min and 210 C for 12.5 min were applied, respectively. At 200 C for 20 min the methane yield decreased by 9%
compared to untreated press mud.
Although furfural was found up to concentrations of 1.21 g L1
it can be assumed that no inhibition is possible according to
reported inhibitory values (Barakat et al., 2012; Monlau et al.,
2012, 2013b). In the same way, HMF and phenolic acids were
found at lower concentrations than that reported as inhibitory.
This is 3 g L1 and 10 g L1 respectively, according to Monlau
et al. (2013b).
It is assumed that the formation of complex recalcitrant substances in the substrate matrix called melanoidins might explain
these ndings. Based on current data we cannot conclude whether
Maillard occurred, however, such reactions are highly probable in
the case of press mud which is rich in proteins and carbohydrates.
Moreover, that performance has already been ascribed in the case
of thermal pre-treatment at 170 C of sunower oil cakes (Monlau
et al., 2013a).

Table 3
Maximum methane yield, rst order rate constant (k) and initial and nal pH during the tests of both untreated press mud and under different pretreatment conditions.
Run

log(Ro)

ymax (mL CH4 g1 VS1

in )

k (d1)

R2

pHi

pHf

Untreated
1
2
3
4
5
6
7
8
9

0.00
2.17
2.77
3.64
4.25
2.26
4.35
2.49
3.57
3.31

209.17b 10.46
323.44e 3.06
340.80f 5.31
252.89c 4.12
189.76a 5.88
321.95e 8.01
214.68b 7.53
320.94e 7.95
288.03d 4.12
299.53d 4.67

0.17d 0.017
0.18d 0.005
0.17c,d 0.003
0.17b 0.023
0.12a 0.000
0.15b,c 0.009
0.11a 0.006
0.18e 0.001
0.17d 0.005
0.16c,d 0.002

0.96
0.99
0.99
0.97
0.97
0.99
0.97
0.99
0.99
0.99

7.69
7.67
7.66
7.27
7.20
7.80
7.14
7.68
7.50
7.72

7.87
7.77
7.75
7.84
7.87
7.82
7.81
7.69
7.82
7.81

Data expressed as mean standard deviation.

Groups with different superscripts differ (p < 0.05).

af

289

400

400

Methane yield (mL CH4 g1VSin-1)

Methane yield (mL CH4 g1 VSin-1)

L.M. Lpez Gonzlez et al. / Bioresource Technology 169 (2014) 284290

350
300
250
200
150
100
50

350
300
250
200
150
100

10

15
20
Time (days)

25

25

30

Fig. 3. Cumulative methane production (STP) for untreated and pre-treated press
mud. Untreated press mud (s), 150 C_5 min (), 150 C_20 min (e), 200 C_5 min
(+), 200 C_20 min (), 140 C_12.5 (d), 210 C_12.5 min (N), 175 C_2 min (j),
175 C_23 min (h), 175 C_12.5 min (D).

Fig. 4. Response surface plot for the effect of Temperature (T) and pre-treatment
time (t) on methane yield.

The effect of pre-treatment time span and temperature on

methane yield was evaluated by response surface methodology
(Fig. 4). The mathematical regression for yCH4 as a function of temperature (T) and pre-treatment time span (t) is obtained as follows
(Eq. (5)), leaving out the non-signicant quadratic term t2):

yCH4 451:98 9:36  T 16:88  t  0:028  T2  0:10T  t

The square of the determination coefcient (R2) and the

adjusted determination coefcient (R2adj) for the model were 0.95
and 0.94 respectively. These data indicate that only 8.37% of the total
variation remained unexplained by the quadratic model. Moreover,
DurbinWatson statistics showed a value lower than 2 (1.91) indicating a good model t with no correlation of errors. The normal
probability plot of standardised residuals (Figure not shown) shows
no outliers for this experiment demonstrating that there is no abnormality in the study. The linear and interactive model terms are signicant (p < 0.05). For the quadratic model terms only the
temperature is signicant. The negative effect of temperature is 4
times higher in comparison with other effects. The optimal point
obtained from equation 7 is found for the following conditions:
T = 150 C and t = 20 min, with an estimated methane yield of
337.04 mLCH4 g1 VS1
in .
The maximum increase found here was similar to that reported
by Budde et al. (2014) at a mild temperature (160 C), but lower
contact time (5 min) and faster heating time (6 C/min). In a study
on sunower oil cake, an increase of methane yield of 20% was
observed at 170 C for 5 min (Monlau et al., 2013a). More severe
pre-treatment conditions (230 C for 15 min) were effective in

27

29

31
33
35
37
39
COD solubilization (%)

41

43

45

Fig. 5. Dependence of methane yield with the COD solubilisation (B). Experimental
() and modelling (__) curve.

increasing the biochemical methane potential of oil palm empty

fruit bunches by 29% (O-Thong et al., 2012). Nevertheless, interstudy comparisons are almost impossible due to the variability in
substrates characteristics, equipment used for the pre-treatment,
as well as operational conditions (e.g., solid/liquid ratio, mass/volume ratio, heating time, pressure, temperature, pre-treatment
time, stirring). More research on the topic of heat-up phase, time
at set-point temperature and cooling-down phase is suggested.
The methane yield plotted versus SCOD is shown in the Fig. 5. The
methane yield increased with the SCOD to a certain extent, further
on a decrease in methane yield is observed when increasing the
SCOD. The relationship between them was found with an R2 = 0.81.
Regarding reaction kinetics, the k values ranged from 0.107 to
0.181 d1 (Table 3). It is expected that reaction kinetics on pretreated biomasses decreases k-values as it will take longer to
obtain a complete digestion as more and complex substances are
made available for digestion. For pre-treatments with higher severities k was found slightly lower. A long lag phase and a decrease in
the methane production rate have been reported as the main
effects caused by inhibitory compounds such as furans and phenolic compounds that were produced by LHW pre-treatment Budde
et al. (2014) and by accumulation of VFAs.
The pH for all batch experiments was in the range from 7.14 to
7.80 at the beginning of the test (Table 3). These values are well
within the operational range of 6.58.5 reported previously for an
anaerobic digestion process (Weiland, 2010). The nal pH values
obtained after the digestions were in the range from 7.69 to 7.87
(Table 3), according to the buffer capacity developed by CO2 release
as well as that supplied by ammonia released during proteins degradation. The concentration of VFA at the end of the test was lower
than 100 mg L1, which demonstrates that the methanogenic stage
was not disturbed in the nal stage of the experiment.
4. Conclusions
The better pre-treatment conditions were found at lower severities (log(Ro) = 2.172.77) when coupling (i) low temperature
(140 C, 150 C) with a long contact time (12.5 min, 20 min) and
(ii) milder temperature (175 C) with a short contact time (2 min).
Future research could aim at investigating the effect of prolonged times (t P 20 min) at 150 C for press mud solubilisation
and anaerobic digestion. Furthermore, low pre-treatment times
(t < 5 min) at a temperature of 175 C could be looked at in more
detail when screening for a fast LHW pre-treatment.
Acknowledgements
This research was supported by the VLIR-UOS project entitled
Biogas production from waste from local food, wood and

290

L.M. Lpez Gonzlez et al. / Bioresource Technology 169 (2014) 284290

sugarcane industries for increasing self-sufciency of energy in

Sancti Spiritus, Cuba. We thank Soe Coelus from Research Group
Food Chemistry and Human Nutrition at Ghent University and
Anka Thoma from Leibniz-Institute for Agricultural Engineering
Potsdam-Bornim for technical assistance for sugar analysis and
the LHW press mud pre-treatment assays, respectively.

References
Abril, D., Medina, M., Abril, A., 2012. Sugar cane bagasse prehydrolysis using hot
water. Braz. J. Chem. Eng. 29 (01), 3138.
APHA, Awwa, WEF, 1995. Standard Methods for the Examination of Water and
Wastewater, nineteenth ed. American Public Health Association/America
Water Works Association/Water Environment Federation, Washington, DC,
USA.
Barakat, A., Monlau, F., Steyer, J.-P., Carrere, H., 2012. Effect of lignin-derived and
furan compounds found in lignocellulosic hydrolysates on biomethane
production. Bioresour. Technol. 104, 9099.
Bougrier, C., Delgens, J.P., Carrre, H., 2008. Effects of thermal treatments on ve
different waste activated sludge samples solubilisation, physical properties and
anaerobic digestion. Chem. Eng. J. 139, 236244.
Budde, J., Heiermann, M., Surez-Quiones, T., Plchl, M., 2014. Effects of
thermobarical pretreatment of cattle waste as feedstock for anaerobic
digestion. Waste Manag. 34 (2), 522529.
Chandra, R., Takeuchi, H., Hasegawa, T., 2012a. Hydrothermal pretreatment of rice
straw biomass: a potential and promising method for enhanced methane
production. Appl. Energy 94, 129140.
Chandra, R., Takeuchi, H., Hasegawa, T., Kumar, R., 2012b. Improving
biodegradability and biogas production of wheat straw substrates using
sodium hydroxide and hydrothermal pretreatments. Energy 43 (1), 273282.
Costa, A.G., Pinheiro, G.C., Pinheiro, F.G.C., Dos Santos, A.B., Santaella, S.T., Leito,
R.C., 2013. Pretreatment strategies to improve anaerobic biodegradability and
methane production potential of the palm oil mesocarp bre. Chem. Eng. J. 230,
158165.
FAO, 2013. FAO Statistical Yearbook 2013. World Food and Agriculture, pp. 289.
Fernandez-Cegri, V., De la Rubia, M.A., Raposo, F., Borja, R., 2012. Effect of
hydrothermal pretreatment of sunower oil cake on biomethane potential
focusing on bre composition. Bioresour. Technol. 123, 424429.
Hendriks, A.T.W.M., Zeeman, G., 2009. Pretreatments to enhance the digestibility of
lignocellulosic biomass. Bioresour. Technol. 100 (1), 1018.
Ingram, T., Rogalinski, T., Bockemuhl, V., Antranikian, G., Brunner, G., 2009. Semicontinuous liquid hot water pretreatment of rye straw. J. Supercrit. Fluid 48 (3),
238246.
Kim, Y., Mosier, N.S., Ladisch, M.R., 2009. Enzymatic digestion of liquid hot water
pretreated hybrid poplar. Biotechnol. Prog. 25 (2), 340348.
Lopez, G.L.M., Vervaeren, H., Pereda Reyes, I., Dumoulin, A., Romero Romero, O.,
Dewulf, J., 2013. Thermo-chemical pre-treatment to solubilise and improve
anaerobic biodegradability of press mud. Bioresour. Technol. 131, 250257.
Masarin, F., Gurpilhares, D.B., Baffa, D.C., Barbosa, M.H., Carvalho, W., Ferraz, A.,
Milagres, A.M., 2011. Chemical composition and enzymatic digestibility of
sugarcane clones selected for varied lignin content. Biotechnol. Biofuels 4 (1),
55.
Menardo, S., Airoldi, G., Balsari, P., 2012. The effect of particle size and thermal pretreatment on the methane yield of four agricultural by-products. Bioresour.
Technol. 104, 708714.

Monlau, F., Barakat, A., Steyer, J.P., Carrere, H., 2012. Comparison of seven types of
thermo-chemical pretreatments on the structural features and anaerobic
digestion of sunower stalks. Bioresour. Technol. 120, 241247.
Monlau, F., Latrille, E., Da Costa, A.C., Steyer, J.-P., Carrre, H., 2013a. Enhancement
of methane production from sunower oil cakes by dilute acid pretreatment.
Appl. Energy 102, 11051113.
Monlau, F., Trably, E., Barakat, A., Quemeneur, M., Steyer, J.P., Carrre, H., 2013b. Do
by-products of thermochemical treatment of lignocellulosic materials inhibit
anaerobic mixed cultures? Overview of recent ndings. In: 13th World
Congress on Anaerobic Digestion. Recovering (bio) Resources for the World,
2528 June 2013, Santiago de Compostela, Espana.
Nitsos, C.K., Konstantinos, A.M., Kostas, S.T., 2013. Optimisation of hydrothermal
pretreatment of lignocellulosic biomass in the bioethanol production process.
ChemSusChem 6, 110122.
O-Thong, S., Boe, K., Angelidaki, I., 2012. Thermophilic anaerobic co-digestion of oil
palm empty fruit bunches with palm oil mill efuent for efcient biogas
production. Appl. Energy 93, 648654.
Overend, R.P., Chornet, E., 1987. Fractionation of lignocellulosics by steam-aqueous
pretreatments. Philos. Trans. R. Soc. Lond. A 321, 526533.
Perez, J.A., Gonzalez, A., Oliva, J.M., Ballesteros, I., Manzanares, P., 2007. Effect of
process variables on liquid hot water pretreatment of wheat straw for
bioconversion to fuel-ethanol in a batch reactor. J. Chem. Technol. Biotechnol.
82 (10), 929938.
Sambusiti, C., Monlau, F., Ficara, E., Carrere, H., Malpei, F., 2013. A comparison of
different pre-treatments to increase methane production from two agricultural
substrates. Appl. Energy 104, 6270.
Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D., 2008.
Determination of structural carbohydrates and lignin in biomass. Technical
report NREL/TP-510-42618.
Suryawati, L., Wilkins, M.R., Bellmer, D.D., Huhnke, R.L., Maness, N.O., Banat, I.M.,
2009. Effect of hydrothermolysis process conditions on pretreated switchgrass
composition and ethanol yield by SSF with Kluyveromyces marxianus IMB4.
Process Biochem. 44 (5), 540545.
Taherzadeh, M.J., Karimi, K., 2008. Pretreatment of lignocellulosic wastes to
improve ethanol and biogas production: a review. Int. J. Mol. Sci. 9 (9), 1621
1651.
VDI 4630, 2006. Fermentation of Organic Materials. Characterisation of the
Substrates, Sampling, Collection of Material Data, Fermentation Tests.
Handbuch Energietechnik.
Velarde, S.E., Len, O.M., Cullar, A.I., Villegas, D.R., 2004. Produccin y aplicacin de
compost, rst ed. La Habana, Cuba.
Wang, Q., Noguchi, C., Hara, Y., Sharon, C., Kakimoto, K., Kato, Y., 1997. Studies on
anaerobic digestion mechanism: inuence of pretreatment temperature on
biodegradation of waste activated sludge. Environ. Technol. 18 (10), 9991008.
Weiland, P., 2010. Biogas production: current state and perspectives. Appl.
Microbiol. Biotechnol. 85, 849860.
Xu, F., Sun, R.-C., Sun, J.-X., Liu, C.-F., He, B.-H., Fan, J.-S., 2005. Determination of cell
wall ferulic and p-coumaric acids in sugarcane bagasse. Anal. Chim. Acta 552
(12), 207217.
Yu, G., Yano, S., Inoue, H., Inoue, S., Endo, T., Sawayama, S., 2010a. Pretreatment of
rice straw by a hot-compressed water process for enzymatic hydrolysis. Appl.
Biochem. Biotechol. 160 (2), 539551.
Yu, Q., Zhuang, X.S., Yuan, Z.H., Wang, Q., Qi, W., Wang, W., Zhang, Y., Xu, J.L., Xu,
H.J., 2010b. Two-step liquid hot water pretreatment of Eucalyptus grandis to
enhance sugar recovery and enzymatic digestibility of cellulose. Bioresour.
Technol. 101 (13), 48954899.
Zheng, Y., Zhao, J., Xu, F., Li, Y., 2014. Pretreatment of lignocellulosic biomass for
enhanced biogas production. Prog. Energy Combust. Sci., http://dx.doi.org/
10.1016/j.pecs.2014.01.001.