Professional Documents
Culture Documents
a r t i c l e
i n f o
Article history:
Received 7 January 2014
Received in revised form 26 April 2014
Accepted 28 April 2014
Available online 9 May 2014
Keywords:
Ultraltration
Nanoltration
Ethanolic solutions
Bark extract
Polyphenols
Carbohydrates
a b s t r a c t
The aim of this work is to promote the enrichment of an ethanolic extract of Eucalyptus globulus bark in
polyphenolic compounds relatively to other compounds such as carbohydrates. Several at sheet membranes were tested with water and ethanol solutions (52% v/v and 80% v/v) assessing to permeability.
Rejections to gallic and tannic acids and maltose were evaluated for nanoltration membranes and for
the ultraltration membrane of lower cut-off. The dependence of permeability and rejection relative to
ethanol percentage is discussed giving new insights about the membrane performance towards
ethanol/water solutions. Among the tested membranes, two ultraltration (JW 30,000 Da and PLEAIDE
5000 Da) and one nanoltration (SolSep 90801) membranes were selected to the concentration process
of an ethanolic extract of Eucalyptus globulus bark produced at previously optimized conditions. The performance of the three membranes was evaluated concerning polyphenolic compounds and carbohydrate
composition. The volume reduction factor was 1.76. JW membrane revealed the lowest total decrease on
permeability (53%) relative to the initial. All the three membranes showed selective retention of polyphenolic compounds, however JW promoted the highest enrichment of formaldehyde-condensable tannins
(fcT) and proanthocyanidins (Pac) (17% and 28%, respectively). The nal composition of the retentate
(in % weight/dry weight) was: TPC 39%, fcT 46%, Pac 38%, GalT 3.2% and TC 15%. The detailed sugar analysis revealed that some arabinose- and rhamnose-containing oligo/polysaccharides are preferentially
retained, while those with glucose and galacturonic acid moieties are transported through the membrane
to permeate stream. Finally, cleaning performance of membranes was evaluated and 80100% ux
recoveries were attained.
2014 Elsevier B.V. All rights reserved.
1. Introduction
The development of biorenery platforms is currently undergoing rapid expansion. Pulp and paper industries have a privileged
position due to the availability of side-streams lignocellulosic materials usually classied as by-products, such as bark which is a disposal in mill site where the logs are debarked. This is the case of
pulp plants in Portugal which produces about 124,000 tons of Eucalyptus globulus bark per medium size industrial unit. Bark is further
integrated in the mill operation as energy source. The basic chemical
composition of bark and wood is similar concerning the major
macromolecular components: lignin, cellulose and hemicelluloses
[1,2]. However, the extractive and inorganic content is usually
Corresponding author. Address: LSRE Laboratory of Separation and Reaction
Engineering, Faculty of Engineering, Chemical Engineering Department, University
of Porto, Rua Dr. Roberto Frias s/n, 4200-465 Porto, Portugal. Tel.: +351 22 041
3606; fax: +351 22 508 1449.
E-mail address: ppinto@fe.up.pt (P.C.R. Pinto).
http://dx.doi.org/10.1016/j.seppur.2014.04.042
1383-5866/ 2014 Elsevier B.V. All rights reserved.
higher in bark than in wood. This is one of the reasons why bark
has not been used for pulp production. Among the undesired
extractive fractions is the polyphenolic fraction. This is composed
by simple phenolics such as gallic and ellagic acids, avonoids,
complex glycosides of phenolic compounds [3,4], hydrolysable tannins, and proanthocyanidins [5,6], often called condensed tannins.
The awareness on these compounds is growing up due to their
properties and biological activities with emerging applications on
cosmetics, nutricosmetic and fortied foods or supplements industries turning it on high added-value additives or active principles
[7,8].
In this perspective, E. globulus bark is a potential raw material to
produce polyphenolic enriched extracts. In our previous work, the
optimum conditions (time, temperature and ethanol %) for the
extraction of polyphenolic compounds from E. globulus bark were
reported. The extract produced at optimum conditions (OC extract)
was obtained in ethanol/water solution (52/48, v/v) and it
demonstrated important biological activity. The yield was 50 g of
material per kg of bark with 1/3 of the extracted material being
P.C.R. Pinto et al. / Separation and Purication Technology 132 (2014) 234243
235
Nomenclature
List of symbols
A
effective membrane area (m2)
Ara
arabinose
Cp
concentration in the permeate (g L1)
Cr
concentration in the retentate (g L1)
Gal
galactose
GalA
galacturonic acid
GalT
gallotannins (% w/w)
Glc
glucose
fcT
formaldehyde-condensable tannins (% w/w)
Jp
volumetric ux through membrane (L m2 h1)
Lp
membrane permeability coefcient (L m2 h1 bar1)
NF
nanoltration
pHPZC
point of zero charge
OC
optimum conditions
Pac
proanthocyanidins (% w/w)
Qp
permeate ow rate (L h1)
Rha
rhamnose
Man
Rm
Rj
SN
TMP
TPC
TS
TC
UF
Vf
Vr
VRF
Xyl
mannose
membrane resistance coefcient (m1)
apparent solute rejection coefcient
Stiasny number (% w/w)
transmembrane pressure (bar)
total phenolic compounds (% w/w)
total non-volatile solids (g L1)
total carbohydrates (% w/w)
ultraltration
feed volume (L)
retentate volume (L)
volume reduction factor
xylose
Greek letters
l
dynamic viscosity of water/solvent (kg m1 s1)
2. Experimental
2.1. Equipment, membranes and conditioning
Benchtop studies were conducted using a membrane cell system Sepa CF II Med/High Foulant System (GE Osmonics, USA) with
an effective area of 0.014 m2 plus a ow meter, a diaphragm pump
Hydra-Cell, model M-3/G-13, (Wanner Engineering, Inc.) with a
frequency inverter (MC07, MovitracB, SEW Eurodrive), and a
manual hydraulic pump (P19, SPX Corporation, USA). The NF/UF
unit withstands a maximum operating pressure of 69 bar, and a
maximum operating temperature of 177 C. The temperature of
the feed was assured by a Lauda thermostatic bath (Ecoline
Staredition Re 206) and a coil immersed on the feed tank. The feed
temperature was checked by an electronic contact thermometer
(VT-5 S40, VWR).
The UF and NF at sheet membranes studied are listed in
Table 1. Aqueous solutions of ethanol (Panreac) were prepared
on a volume/volume basis using deionized water. All membranes
were preconditioned according to the protocol recommended in
the literature [22]. Prior to use, the membranes were rst rinsed
with water and soaked overnight. Afterwards, the membranes
were soaked with ethanol solutions starting with 10% (v/v) ethanol
and then with increments of 1020% ethanol until 52% or 80% (v/v)
ethanol, depending of the programed assays. For the experiences
with water, membranes were simply soaked with water for three
times and left overnight. The SolSep membranes were directly
washed and conditioned in the working solvent as recommended
by the fabricant. Before operation, each membrane was prepared
by compressing it into the module by means of system hydraulic
pressure (about 1015 bar more than the operating pressure in
the experiments), using water or ethanol/water solutions at a
transmembrane pressure (TMP) of 1 bar for about 30 min to
remove material from the pores. Then, using fresh solution, the
membranes were submitted to compaction with a TMP 12 bar
higher than the operating pressure in the experiments. The permeate ux was measured and usually the time to ensure the steady
state was 1 h.
Ultra-pure water and analytical grade reagents were used for
membrane characterization.
236
P.C.R. Pinto et al. / Separation and Purication Technology 132 (2014) 234243
Table 1
Characteristics of the NF/UF membranes.
a
b
Membrane
Producer
Composition
Operational pH
Tmax (C)
Pmax (bar)
80105
90801
NF270
SolSep
SolSep
DowFilmtec
Polyamide derivative
Polyamide derivative
Polyamide thin lm composite
120
80
45
20
20
41
GE
PLEAIDE
JW
EW
1000
5000
30,000
60,000
GE Osmonics
Orelis Environnement
GE Osmonics
GE Osmonics
Polyamide composite
Polyethersulfone (PES)
Polyvinylidene uoride (PVDF)
Polysulfone (PS)
112
114
310a
112b
211
314
111
0.513
50
75
Continuous operations.
Short-term cleaning (30 min).
237
P.C.R. Pinto et al. / Separation and Purication Technology 132 (2014) 234243
Jp
Qp
A
Jp
TMP
lRm
J p Lp TMP
VRF
Vf
Vr
Rj 1
Cp
Cr
250
10
200
5
Water
150
100
Ethanol 52%
0
80105
90801
NF270
GE
Ethanol 80%
50
0
80105
90801
NF270
GE
PLEAIDE
JW
EW
membrane
Fig. 1. Permeability of polymeric membranes for water, ethanol 52% and ethanol
80% at 35 C and feed owrate 4.5 L h1.
238
P.C.R. Pinto et al. / Separation and Purication Technology 132 (2014) 234243
tension, molar volume and dielectric constant [33]. Surface tension and dielectric constant decrease in the order water, ethanol
52% and ethanol 80% [31,34,35], while molar volume increases
in this order. Molar volume could, at least, partially explain the
Lp decrease for NF membrane in the order water to ethanol 80%,
while surface tension would have the opposite effect on Lp. However, dielectric constant of the mixture and the related factor,
polarity, could have a practical effect on membrane due to the
interaction (or its absence) with the polymeric material. This
effect depends also on surface energy of the polymer as stated
in the literature [30].
Hence, for membranes with predominant hydrophilic character,
the ux would be considerably lower for ethanol/water mixtures
than for water due to the limited hydrogen bonding capability of
the ethanol delaying and impairing the transport through the
membrane. In accordance, other authors have concluded that the
hydrophilicity and porosity are the most important characteristics
of membranes affecting the uid transport, while viscosity and
polarity are the solvent properties with highest inuence on permeability [36].
From this preliminary study, it was possible to evaluate the
behavior of the different membranes concerning Lp as the net
result of different factors. Among the UF membranes, those of
medium cut-off and simultaneously presenting Lp values higher
than 35 L m2 h1 bar1 were selected for the concentration
assays: membranes JW and PLEAIDE. Among the NF membranes,
the 090801 is the membrane with highest Lp for 52% ethanol
(5.8 L m2 h1 bar1) and, thus, the most promising for the concentration process. In spite of this, a study of rejection coefcients for
three compounds was performed for NF membranes: two phenolic
compounds (gallic acid and tannic acid) and one disaccharide,
maltose, as general representatives of the families of compounds
present in the real extract. In this study, the membrane GE was
included due to its small cut-off. The overall results are presented
in Fig. 3.
Among all membranes, the NF270 presents the highest rejection
for the three compounds. NF 270 is a NF membrane stated to have
high rejection indexes for simple sugars and salts in water (about
90% for glucose in water at 37 C and at 8 bar [37] and more than
97% to MgSO4 (manufacturer data Table 1). However, Restolho
and co-workers [38] have reported rejections of 52% and 34% (in
water, 25 C, 18 bar and 2.0 L min1) for glucose and xylose,
respectively. The values found in this work for rejections are 40%
for gallic acid, 67% for maltose and 95% for tannic acid in 52%
ethanol. The membrane 80105 presented the lowest rejections in
52% ethanol (between 4% and 22%).
1000
EW
JW
Pleaide
-1
Jp (L m h )
750
-2
GE
NF270
500
20
250
GE
15
0
0.0
10
0.5
1.0
1.5
1/ (m s kg-1 )
Ethanol Ethanol
52%
80%
Water
2.0
0
0.0
0.5
1.0
1.5
Fig. 2. Correlation between Jp and l1 for membranes EW, JW, PLEAIDE, GE and NF270 for water, ethanol 52% and ethanol 80% at 35 C and feed owrate 4.5 L h1.
239
P.C.R. Pinto et al. / Separation and Purication Technology 132 (2014) 234243
100
Gallic acid
60
Maltose
Tannic acid
40
20
0
80105
90801
NF270
GE
Membrane
100
80
Rf (%)
Ethanol 52%
Gallic acid
60
Maltose
Tannic acid
40
20
0
80105
90801
NF270
Membrane
GE
Ethanol 80%
Fig. 3. Apparent rejection coefcients to gallic acid, tannic acid and maltose in
ethanol 52% (A) and in ethanol 80% (B) for the NF membranes and for the UF
membrane GE at 35 C and feed owrate 4.5 L h1.
In the conditions of this work, the relative rejection in the membranes 90801, NF270 and GE follows the relative molecular weight
order (gallic acid, maltose, tannic acid) in both solvents. However,
for the membrane 80105 a change was observed: the rejection to
gallic acid is higher than to the disaccharide. These observations
can only be explained by different membrane characteristics since
solventsolute interactions are the same as for the other membranes. The polarity differences between the molecules can lead
to a modication of the expected relative rejection based on molecular weight due to polarity itself (interaction with membrane) and
hydrodynamic volume (interaction with the solvent). The transport of the gallic acid relative to the disaccharide (the phenolic
compound is less polar than the disaccharide) was improved by
an additional afnity with this membrane. Among all membranes,
the 80105 seems to be the most hydrophobic. This is in accordance
with the increase of the rejection of the three compounds (but
maintaining the relative rejections) with the increase of ethanol
percentage (52% to 80%) (and consequently, with decrease on solvent polarity) observed for membrane 80105 Fig. 3A and B.
The most relevant difference on rejection between ethanol 52%
and ethanol 80% concerns to tannic acid: for GE and for 90801
membranes, the rejection decreases with the change of the percentage of ethanol from 52% to 80%; this means that the presence
of ethanol promotes the transport of this compound through the
membrane pores. This could be explained based on interaction of
solventmembrane and the solvation of the pores: for membranes
with some hydrophilic character, the hydration of the pore wall
occurs, leading to a decrease on their effective pore size [36]; considering that these membranes are not so effectively solvated by
ethanol as by water, the increase of ethanol percentage led to an
higher effective pore diameter and thus, to a lower hindrance to
the transport of the tannic acid reected in a higher rejection in
52% ethanol than in 80% ethanol. This effect is noticeable for tannic
acid probably due to its higher molecular weight. For membrane
NF270, rejection in 52% and 80% ethanol is similar for all compounds. Although for this membrane no effect was perceptible
from 52% to 80%, the lower rejections observed for this membrane
1.0
JW
0.8
Jt /J0
Rf (%)
80
PLEAIDE
90801
0.6
0.4
0.2
0.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
60
50
Jp (L h-1 m-2 )
JW
40
PLEAIDE
90801
30
20
10
0
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
VCR
Fig. 4. (A) Normalized permeate ux (ratio of instantaneous permeate ux at time
t, Jt, and at initial time, J0) along the operating time. (B) Instantaneous permeate ux
along the VRF for the extract OC. Conditions: concentration mode, 35 C, feed
owrate 4.5 L h1, TMP 4 bar for UF JW and PLEAIDE and 14 bar for NF 90801;
operating time for VRF 1.76: 4.3 h JW; 7.4 h PLEAIDE; 5.6 h 90801.
P.C.R. Pinto et al. / Separation and Purication Technology 132 (2014) 234243
due to solute adsorption onto the membrane surface or membrane pore plugging becomes noticeable and contributes to the
ux drop [9]. In the OC extract processing, the decrease pattern
of permeate uxes is different for the three membranes
(Fig. 4A). 90801 is the membrane with the highest absolute ux
but also with the higher initial decline, notorious up to the VRF
1.37 Fig. 4B which corresponds to about 2.5 h of operation.
From this point onward, the membrane 90801 presents lower
Jp value than the UF membrane JW (Fig. 4B). These differences
suggest that the 90801 membrane is the most susceptible one
to concentration polarization and fouling in the conditions of this
work, thus creating an additional resistance to permeate ux. At
the pH of the OC extract (4.3) the compounds are mainly neutral
and protonated in solution. In the conditions of the processing,
the membrane 090801 (pHPZC 3.3) is negatively charged and
electrostatic attractions should be expected between the surface
and some of the components of the extract. For JW (pHPZC 6.1)
and PLEAIDE (pHPZC 6.4) membranes, electrostatic repulsions are
expected due to the positively charged surface. Thus, considering
this data, fouling should be more severe in the case of PLEAIDE.
However, besides this effect, hydrophobicity of the membranes
would play also an important role in this process. The balance
between electrostatic repulsion and van der Waals interactions
determines the outcomes of membrane fouling, as well as the efciency of the membrane cleaning after processing.
The initial decline of normalized ux Jt/J0 (Fig. 4A) for UF membranes is the same, suggesting that the concentration polarization
effect is similar for both. However, after about 1 h of operation, the
membrane JW holds up higher normalized and absolute ux. At
the end of the concentration stage, for the same concentration
factor in volume, the ux reductions relative to the initial were
53%, 68% and 85% for JW, PLEAIDE and 90801 membranes, respectively. Among these, JW presents simultaneously the lowest
decrease of ux by fouling, reaching the steady state earlier than
the others, and the highest nal ux.
Table 2
Rj for TS, TPC and TC during the concentration of OC extract with the membranes JW,
PLEAIDE and 90801. Conditions: 35 C, feed owrate 4.5 L h1, TMP 4 bar for UF (JW
and PLEAIDE) and 14 bar for NF (90801); operating time for VRF 1.76: 4.3 h JW;
7.4 h PLEAIDE; 5.6 h 90801.
Membrane
TS
TPC
TC
JW
PLEAIDE
90801
0.58
0.77
0.78
0.75
0.85
0.92
0.12
0.17
0.15
50
Feed (OC extract)
40
% w/w
240
JW
PLEAIDE
30
90801
20
10
0
TPC
fcT
Pac
TC
Parameter
Fig. 5. Composition of feed (OC extract) and composition of retentates produced
with membranes JW, PLEAIDE and 90801. The values are represented as % weight/
dry weight of the OC extract (the feed) or dry weight of the retentate obtained for
each membrane. Conditions: concentration mode, 35 C, feed owrate 4.5 L h1,
TMP 4 bar for UF JW and PLEAIDE membranes and 14 bar for NF 90801 membrane;
VRF 1.76.
would be possible the elimination of sugar moieties in the permeate during UF and NF.
Besides TS, TPC and TC, nal retentates were also analyzed for
Pac, fcT and TC. The global results are depicted in Fig. 5.
For the same VRF, the membranes showed different performances as denoted by the composition of the retentate. The
common feature is an accentuated enrichment effect on fcT
(monoavonoids, biavonoids and some oligomers). However,
membranes with lower cut-off (PLEAIDE and 900801) retain more
material carrying phenolic hydroxyl groups, as indicated by TPC in
the respective retentates, probably those compounds of lower
molecular weight. This is not reected on Pac content, assessed
by butanol-acid assay, since this family is mainly composed by avonoid oligomers. In accordance, JW membrane promoted higher
enrichment of this parameter than did for TPC and fcT, probably
due to a noteworthy contribution of compounds with high
molecular weight that responds to the butanol-acid method. On
other side, the Pac content of PLEAIDE and 90801 retentate is similar to that of OC extract (which was not observed for TPC and fcT).
The reason for that could be a preferential entrapment or adsorption of Pac at the membranes contributing for fouling, as assessed
by the mass balance using the content on permeate for PLEAIDE
and 90801 (data not presented). This also occurs for JW, but in a
lower proportion. In accordance, JW showed the lowest ux
decline during the concentration process (Fig. 4).
Hydrolysable tannins (GalT) were quantied by the conventional
methodology: gallic acid analysis before and after methanolysis.
For this, gallic acid present in the extract is previously quantied
and them this value is deducted to the value quantied after acid
methanolysis. Acid methanolysis is the process of cleaving ester
linkages between monosaccharide and gallic acid units composing
GalT, one of the typical structures of hydrolysable tannins. The
results for OC extract, retentates and permeates from each
membrane are displayed in Fig. 6.
The contribution of GalT (estimated as gallic acid) as a parcel of
polyphenolic compounds in the OC extract, permeates and retentates is low (about 3% of the dried extract weigh). This could be
due to a natural low content on hydrolysable tannins in the extract
or due to ester bonds hydrolysis already in the extraction process.
Nevertheless, GalT quantication is a practical indication of the
membrane performance for this type of compounds. The membrane JW presents higher permeability to GalT than 90801 and
PLEAIDE as revealed by the lower content on JW retentate. This
result is in accordance with the differences already referred for
TPC, suggesting that GalT have signicant impact on the TPC
quantication.
P.C.R. Pinto et al. / Separation and Purication Technology 132 (2014) 234243
GalT (% w/w)
241
Fig. 6. Estimative of gallic acid liberated from GalT in the OC extract, retentates and
permeates produced with membranes JW, PLEAIDE and 90801. Conditions:
concentration mode, 35 C, feed owrate 4.5 L h1, TMP 4 bar for UF JW and
PLEAIDE membranes and 14 bar for NF 90801 membrane; VRF 1.76.
Ara
JW
PLEAIDE
25
90801
molar fraction
of TC %
molar fraction
of TC %
20
15
10
5
0
5
0
6
4
2
0
molar fraction
of TC %
30
Glc
20
10
10
25
20
6
4
2
15
10
GalA
10
5
0
molar fraction
of TC %
10
40
molar fraction
of TC %
molar fraction
of TC %
molar fraction
of TC %
Rha
PLEAIDE 90801
15
Gal
10
Xyl
JW
20
JW
PLEAIDE 90801
8
6
4
2
0
Man
PLEAIDE 90801
Fig. 7. Monosaccharide composition (molar fraction) of the OC extract, retentates (ret) and permeates (per) produced with membranes JW, PLEAIDE and 90801. Conditions:
concentration mode, 35 C, feed owrate 4.5 L h1, TMP 4 bar for UF JW and PLEAIDE membranes and 14 bar for NF 90801 membrane; VRF 1.76.
P.C.R. Pinto et al. / Separation and Purication Technology 132 (2014) 234243
100
242
80
60
JW
40
PLEAIDE
90801
20
0
0
Cleaning cycles
Fig. 8. The inuence of cleaning cycles on permeability recovery of membranes JW,
PLEAIDE and 90801.
4. Conclusions
The aim of this work was to test different membranes for the
concentration process of polyphenolic compounds from an ethanolic extract of E. globulus bark. Characterization of the selected
membranes concerning permeability to water, ethanol 52% and
80% and rejection to standards was the rst step. When changing
water to ethanol 52%, the permeability decreased between 47%
(PLEAIDE) and 80% (EW); further increase of ethanol to 80% led
to an additional decrease of permeability for all membranes,
except for EW (UF) and 80105 (NF) membranes. These effects
are related to ethanol/water properties and solvent interaction
with the membranes. 90801 and GE membranes showed a favorable
rejection ratio tannic acid/disaccharide for ethanol 52%, indicative
that some selective enrichment in polyphenolic compounds
would be possible. Among the tested membranes, JW (30 kDa),
PLEAIDE (5 kDa) and 90801 were selected for concentration
process of the bark extract. The best ux performance during
the concentration was found for JW. The three membranes
promoted an enrichment of fcT (2030%) for a VRF of 1.76. JW
promoted the highest concentration of Pac and lower rejection
for TC, what is advantageous considering the purpose of the membrane processing. GalT, the fraction of hydrolysable tannins
detected in the extract, were preferentially rejected by 90801
and PLEAIDE but not by JW membrane. The detailed carbohydrate
analysis showed some selective permeation to glucose and galacturonic acid-containing oligo-/polysaccharides. Higher rejection of
rhamnose and arabinose could be related to the association
of these moieties with polyphenolic compounds. Flux recoveries
of 80100% were attained for all the membranes. However, JW
was the easiest membrane to clean.
Membrane process was successfully applied for concentration
of an ethanolic bark extract achieving, with a VRF of 1.76, an
enrichment of polyphenolic compounds of avonoid nature.
This process could be the primary step in a separation process
envisaging the purication of the Pac fraction of this extract for
high-added value applications.
P.C.R. Pinto et al. / Separation and Purication Technology 132 (2014) 234243
Acknowledgements
This work was carried out under the Project BIIPP No.
11551 Integrated Biorenery in Pulp and Paper Industry funded
by the European Regional Development Fund (ERDF) through the
Operational Programme for Competitiveness Factors (POFC) of
the National Strategic Reference Framework (NSRF). This work
was co-nanced by FCT and FEDER under Programme COMPETE
(Project PEst-C/EQB/LA0020/2013).
Eng. Maria Eduarda Baptista and Dr. Sergio Morales Torres (LA
LSRE/LCM) are acknowledged for support in some of the
permeability and pHPZC assays, respectively.
References
[1] I. Mota, P.C. Rodrigues Pinto, C. Novo, G. Sousa, O. Guerreiro, A.R. Guerra, M.F.
Duarte, A.E. Rodrigues, Extraction of polyphenolic compounds from Eucalyptus
globulus bark: process optimization and screening for biological activity, Ind.
Eng. Chem. Res. 51 (2012) 69917000.
[2] I. Miranda, J. Gominho, I. Mirra, H. Pereira, Fractioning and chemical
characterization of barks of Betula pendula and Eucalyptus globulus, Ind. Crop.
Prod. 41 (2013) 299305.
[3] S.A.O. Santos, C.S.R. Freire, M.R.M. Domingues, A.J.D. Silvestre, C. Pascoal Neto,
Characterization of phenolic components in polar extracts of Eucalyptus
globulus Labill. bark by high-performance liquid chromatographymass
spectrometry, J. Agric. Food Chem. 59 (2011) 93869393.
[4] E. Conde, E. Cadaha, M.C. Garca-Vallejo, F. Toms-Barbern, Low molecular
weight polyphenols in wood and bark of Eucalyptus globulus, Wood Fiber Sci.
27 (1995) 379383.
[5] P.C.R. Pinto, G. Sousa, F. Crispim, A.J.D. Silvestre, C. Pascoal Neto, Eucalyptus
globulus bark as source of tannin extracts for application in leather industry,
ACS Sust. Chem. Eng. 1 (2013) 950955.
[6] E. Cadaha, E. Conde, B. Fernndez de Simn, M.C. Garca-Vallejo, Tannin
composition of Eucalyptus camaldulensis, E. globulus and E. rudis. Part II. Bark,
Holzforschung 51 (1997) 125129.
[7] T. Stevanovic, P.N. Diouf, M.E. Garcia-Perez, Bioactive polyphenols from
healthy diets and forest biomass, Curr. Nutr. Food Sci. 5 (2009) 264295.
[8] M. Royer, M. Prado, M.E. Garca-Prez, P.N. Diouf, T. Stevanovic, Study of
nutraceutical, nutricosmetics and cosmeceutical potentials of polyphenolic
bark extracts from Canadian forest species, PharmaNutrition 1 (2013) 158
167.
[9] M. Cheryan, Ultraltration and Microltration Handbook, Technomic
Publishing Co. Inc., Lancaster, USA, 1998.
[10] E. Conde, B. Daz Reinoso, M.J. Gonzlez-Muoz, A. Moure, H. Domnguez, J.C.
Paraj, Recovery and concentration of antioxidants from industrial efuents
and from processing streams of underutilized vegetal biomass, Food Public
Health 3 (2013) 6991.
[11] A. Cassano, C. Conidi, L. Giorno, E. Drioli, Fractionation of olive mill
wastewaters by membrane separation techniques, J. Hazard. Mater. 248249
(2013) 185193.
[12] C. Russo, A new membrane process for the selective fractionation and total
recovery of polyphenols, water and organic substances from vegetation waters
(VW), J. Membr. Sci. 288 (2007) 239246.
[13] E. Garcia-Castello, A. Cassano, A. Criscuoli, C. Conidi, E. Drioli, Recovery and
concentration of polyphenols from olive mill wastewaters by integrated
membrane system, Water Res. 44 (2010) 38833892.
[14] A. Cassano, C. Conidi, E. Drioli, Comparison of the performance of UF
membranes in olive mill wastewaters treatment, Water Res. 45 (2011)
31973204.
[15] B. Santamara, G. Salazar, S. Beltrn, J.L. Cabezas, Membrane sequences for
fractionation of polyphenolic extracts from defatted milled grape seeds,
Desalination 148 (2002) 103109.
[16] B. Daz-Reinoso, A. Moure, H. Domnguez, J.C. Paraj, Ultra- and nanoltration
of aqueous extracts from distilled fermented grape pomace, J. Food Eng. 91
(2009) 587593.
[17] B. Daz-Reinoso, N. Gonzlez-Lpez, A. Moure, H. Domnguez, J.C. Paraj,
Recovery of antioxidants from industrial waste liquors using membranes and
polymeric resins, J. Food Eng. 96 (2010) 127133.
[18] Y. Yazaki, Improved ultraltration of extracts from Pinus radiata bark,
Holzforschung 39 (1985).
[19] E.M. Tsui, M. Cheryan, Membrane processing of xanthophylls in ethanol
extracts of corn, J. Food Eng. 83 (2007) 590595.
243
[20] B.C.B.S. Mello, J.C.C. Petrus, M.D. Hubinger, Concentration of avonoids and
phenolic compounds in aqueous and ethanolic propolis extracts through
nanoltration, J. Food Eng. 96 (2010) 533539.
[21] J. Li, H. Chase, Applications of membrane techniques for purication of natural
products, Biotechnol. Lett. 32 (2010) 601608.
[22] E.M. Tsui, M. Cheryan, Characteristics of nanoltration membranes in aqueous
ethanol, J. Membr. Sci. 237 (2004) 6169.
[23] L.M. Pastrana-Martnez, S. Morales-Torres, S.K. Papageorgiou, F.K. Katsaros,
G.E. Romanos, J.L. Figueiredo, J.L. Faria, P. Falaras, A.M.T. Silva, Photocatalytic
behaviour of nanocarbonTiO2 composites and immobilization into hollow
bres, Appl. Catal., B 142143 (2013) 101111.
[24] S. Feng, S. Cheng, Z. Yuan, M. Leitch, C. Xu, Valorization of bark for chemicals
and materials: a review, Renew. Sust. Energy Rev. 26 (2013) 560578.
[25] P.M. Tahir, O.C. Musgrave, Z. Ashaari, Determination of polyphenolic content of
bark extracts for wood adhesives, Holzforschung 56 (2002) 267272.
[26] L.J. Porter, L.N. Hrstich, B.G. Chan, The conversion of procyanidins and
prodelphinidins to cyanidin and delphinidin, Phytochemistry 25 (1986) 223
230.
[27] A.E. Hagerman, L.G. Butler, Choosing appropriate methods and standards for
assaying tannin, J. Chem. Ecol. 15 (1988) 17951810.
[28] Z. Lei, J. Jervis, R.F. Helm, Use of methanolysis for the determination of total
ellagic and gallic acid contents of wood and food products, J. Agric. Food Chem.
49 (2001) 11651168.
[29] V. Freger, J. Gilron, S. Belfer, TFC polyamide membranes modied by grafting of
hydrophilic polymers: an FT-IR/AFM/TEM study, J. Membr. Sci. 209 (2002)
283292.
[30] D. Bhanushali, S. Kloos, C. Kurth, D. Bhattacharyya, Performance of solventresistant membranes for non-aqueous systems: solvent permeation results
and modeling, J. Membr. Sci. 189 (2001) 121.
[31] I. Khattab, F. Bandarkar, M. Fakhree, A. Jouyban, Density, viscosity, and surface
tension of water + ethanol mixtures from 293 to 323 K, Korean J. Chem. Eng.
29 (2012) 812817.
[32] R. Shukla, M. Cheryan, Performance of ultraltration membranes in ethanol
water solutions: effect of membrane conditioning, J. Membr. Sci. 198 (2002)
7585.
[33] D.R. Machado, D. Hasson, R. Semiat, Effect of solvent properties on permeate
ow through nanoltration membranes. Part I: Investigation of parameters
affecting solvent ux, J. Membr. Sci. 163 (1999) 93102.
[34] G. Akerlof, Dielectric constants of some organic solventwater mixtures at
various temperatures, J. Am. Chem. Soc. 54 (1932) 41254139.
[35] C. Wohlfarth, Dielectric constant of the mixture (1) water; (2) ethanol, in: M.D.
Lechner (Ed.), Supplement to IV/6, Springer, Berlin Heidelberg, 2008, pp. 520
523.
[36] J. Geens, K. Peeters, B. Van der Bruggen, C. Vandecasteele, Polymeric
nanoltration of binary wateralcohol mixtures: inuence of feed
composition and membrane properties on permeability and rejection, J.
Membr. Sci. 255 (2005) 255264.
[37] M. Mnttri, A. Pihlajamki, E. Kaipainen, M. Nystrm, Effect of temperature
and membrane pre-treatment by pressure on the ltration properties of
nanoltration membranes, Desalination 145 (2002) 8186.
[38] J.A. Restolho, A. Prates, M.N. de Pinho, M.D. Afonso, Sugars and
lignosulphonates recovery from eucalyptus spent sulphite liquor by
membrane processes, Biomass Bioenergy 33 (2009) 15581566.
[39] S.A. Lisboa, D.V. Evtuguin, C. Pascoal Neto, Characterization of non-cellulosic
glucans in Eucalyptus globulus Labill. wood and kraft pulp, Holzforschung 61
(2007) 478482.
[40] S. Quideau, K.S. Feldman, Ellagitannin chemistry, Chem. Rev. (Washington, DC,
U.S.) 96 (1996) 475503.
[41] A.A. Watrelot, C. Le Bourvellec, A. Imberty, C.M. Renard, Neutral sugar side
chains of pectins limit interactions with procyanidins, Carbohydr. Polym. 99
(2014) 527536.
[42] A.A. Shatalov, D.V. Evtuguin, C. Pascoal Neto, (2-O-a-D-Galactopyranosyl-4-Omethyl-a-D-glucurono)-D-xylan from Eucalyptus globulus Labill, Carbohydr.
Res. 320 (1999) 9399.
[43] M. Le Normand, U. Edlund, B. Holmbom, M. Ek, Hot-water extraction and
characterization of spruce bark non-cellulosic polysaccharides, Nord. Pulp Pap.
Res. J. 27 (1) (2012) 1823.
[44] D. Mohnen, Pectin structure and biosynthesis, Curr. Opin. Plant Biol. 11 (2008)
266277.
[45] E. Sjstrm, U. Westermark, Analytical methods in wood chemistry, pulping,
and papermaking, in: Chemical Composition of Wood and Pulps: Basic
Constituents and Their Distribution, Springer Series in Wood Science, 1999,
pp 119.
[46] P.C. Pinto, D.V. Evtuguin, C. Pascoal Neto, Structure of hardwood
glucuronoxylans: modications and impact on pulp retention during wood
kraft pulping, Carbohydr. Polym. 60 (2005) 489497.
[47] S. Lee, C.H. Lee, Microltration and ultraltration as a pretreatment for
nanoltration of surface water, Sep. Sci. Technol. 41 (2006) 123.