Carbohydrate Polymers 155 (2017) 173–181

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Hydroxypropyl methylcellulose based aerogels: Synthesis,

characterization and application as adsorbents for wastewater
pollutants
Bianca F. Martins, Paulo V.O. de Toledo, Denise F.S. Petri ∗
Department of Fundamental Chemistry, Institute of Chemistry, University of São Paulo, Av. Prof. Lineu Prestes, 748, 05508-000 São Paulo, SP, Brazil

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

Article history: Hydroxypropyl methylcellulose (HPMC) E4M and J5MS samples with degrees of substitution of methyl
Received 17 May 2016 groups (DS) 1.9 and 1.5, respectively, and molar substitutions of hydroxypropyl groups (MS) 0.25 and 0.75,
Received in revised form 25 August 2016 respectively, were used to create aerogels in the presence of citric (cit), oxalic (oxa) or terephthalic (ter)
Accepted 25 August 2016
acids, as crosslinkers. E4M-cit, J5MS-cit and J5MS-oxa aerogels presented outstanding stability in water,
Available online 26 August 2016
acid media, alcohols, acetone and hydrocarbons, density values of ∼(0.020 ± 0.002) g/cm3 , compressive
modulus ranging from (111 ± 9) kPa to (133 ± 11) kPa. On the other hand, E4M-oxa, E4M-ter and J5MS-
Keywords:
ter aerogels were not adequate adsorbents because they collapsed in water. The adsorption behavior of
Aerogels
Hydroxypropyl methylcellulose
17␣-ethinyl estradiol (EE) and methylene blue (MB) onto E4M-cit, J5MS-cit and J5MS-oxa aerogels were
Methylene blue fitted to Freundlich model, indicating higher affinity of EE for J5MS-cit and J5MS-oxa and higher affinity
17 ␣-ethinyl estradiol of MB for E4M-cit aerogels. HPMC aeorogels could be successfully reused.
Adsorption © 2016 Elsevier Ltd. All rights reserved.
Organic pollutants

1. Introduction because they present a combination of low density, large surface

Dyes, drugs (hormones, anti-inflammatories, antibiotics) and
phenolic compounds are examples of emerging organic contam-
inants found in wastewater, rivers and ground water, as result
of human and industrial activities. Advanced oxidation process,
nanofiltration, reverse osmosis, and adsorption are some strategies
for the removal of organic pollutants (Rashed, 2013). Adsorp-
tion is one of the most popular of them due to the efficiency,
selectivity, low cost, simplicity and possibility of adsorbent recy-
cling. The adsorbent choice should meet some criteria such as
large surface area, mechanical stability, possibility of recycling
and biodegradability. Grassi and co-workers (Grassi, Kaykioglu,
Belgiorno, & Lofrano, 2012; and references therein) showcased an
overview about the advantages and disadvantages of natural mate-
rials, agricultural waste and minerals (activated carbon, clays) as
adsorbents. Among the natural materials, cellulose and cellulose
derivatives are attractive due to their large adsorption capacity,
renewable origins and biodegradability. Particularly, aerogels made
of renewable materials constitute an interesting class of adsorbents
∗ Corresponding author.
E-mail addresses: bih.martins42@gmail.com (B.F. Martins),
pv.toledo@hotmail.com (P.V.O. de Toledo), dfsp@iq.usp.br, dfsp@usp.br
(D.F.S. Petri).
area and resilience (García-González, Alnaief, & Smirnova, 2011).
The physical properties of aerogels depend on the method used for
their preparation, solvent used, and presence of reinforcing parti-
cles (Siqueira, Bras, & Dufresne, 2010). The aerogel is produced by
extracting the solvent of a gel, while keeping the structure of the gel
network. Freeze-drying and supercritical drying are the most com-
mon methods; in the former the polymer solution is frozen and
the solvent is subsequently sublimated under reduced pressure,
while in the latter, the solvent used for the polymer dissolution
is first exchanged by CO2 under supercritical conditions and then
CO2 evaporates upon bringing the system back to environmental
conditions (Martins et al., 2015).
The application of cellulose based aerogels for solving environ-
mental issues has received much attention due to their low density,
low cost, nontoxic and biodegradable properties. For instance,
aerogels made of cellulose nanofibrils extracted from rice straw cel-
lulose behaved as super-absorbents for water and organic solvents;
after hydrophobization they served as selective adsorbents for oils
(Jiang & Hsieh, 2014). Hydrophobically modified cellulose aerogels
prepared in the absence (Feng, Nguyen, Fan, & Duong, 2015; Jin,
Han, Li, & Sun, 2015; Wang, Xu, Du, & Wang, 2016) or in the pres-
ence of poly(ethylene glycol) of high molecular weight (Wan & Li,
2015a) presented high affinity for oil. Hybrid aerogels composed
of cellulose and maghemite particles presented excellent ability

http://dx.doi.org/10.1016/j.carbpol.2016.08.082
0144-8617/© 2016 Elsevier Ltd. All rights reserved.
174 B.F. Martins et al. / Carbohydrate Polymers 155 (2017) 173–181
for Cr(VI) removal from contaminated water (Wan & Li, 2015b).
Hybrid aerogels composed of wheat straw cellulose, polypyrrole
and silver nanoparticles presented antibacterial properties upon
contact with microorganisms (Wan & Li, 2016). Cellulose aero-
gels are materials with outstanding mechanical properties and
large sorption capacities, but scarce literature about aerogels made
of cellulose ethers (He, Sui, He, & Li, 2015), such as hydrox-
ypropyl methylcellulose (HPMC), was found. HPMCs are interesting
materials because their hydrophobic/hydrophilic character can be
controlled by the degree of substitution (DS) of methyl groups
and molar substitution (MS) of hydroxypropyl groups in the cel-
lulose chains. Moreover, HPMCs nontoxicity and physicochemical
properties allow their application in food and drug formu-
lations (http://www.fda.gov/Food/IngredientsPackagingLabeling/
GRAS/NoticeInventory/ucm153856.htm). In the present work,
aerogels were prepared with two HPMC samples with different
DS and MS, in the presence of three different crosslinkers, by
slow freeze-drying. Oxalic, terephthalic and citric acids were used
as crosslinkers. Oxalic and terephthalic acids are hydrophilic and
hydrophobic diprotic acids, respectively, and citric acid is a triprotic
hydrophilic acid. Thus, one aim of the present study was to inves-
tigate how the chemical nature of crosslinkers would affect the
properties of the resulting aerogels made of HPMC chains with dif-
ferent hydrophobic/hydrophilic characters. The physicochemical
and mechanical properties of HPMC aerogels and their efficien-
cies to remove 17␣-ethinyl estradiol, a hormone, and methylene
blue, a dye, from contaminated water were evaluated, as well as
the possibility of aerogels recycling. The choice of contaminants
was based on their relevance for environmental issues. One major
environmental problem related to 17␣-ethinyl estradiol, which is
the principal component in contraceptive pills, is that its release to
ground water and rivers in concentration as low as 2 ng/L causes
alterations in reproduction or development of fishes because it acts
in their endocrinal systems (Snyder, Westerhoff, Yoon, & Sedlak,
2003; Ankley et al., 2009). Although methylene blue is not an
US FDA-approved drug due to the limited data about serotoner-
gic effects, it has been used for medical purposes, as for instance,
for treatment of methemoglobinemia, encephalopathy or cyanide
poisoning, with doses smaller than 10 mg/kg (http://www.fda.gov/
Drugs/DrugSafety/ucm276119.htm). On the other hand, the use of
methylene blue as a dye for textiles might cause ground water and
rivers contamination; depending on the level of contamination,
water consumers and species living in the contaminated medium
might suffer toxicological effects (Vutskits et al., 2008).
The present study considers the hypothesis that structural
aspects of HPMC chains, distinct DS and MS values, in combination
with different crosslinkers as di- or triprotic acids yield aerogels
with different physical chemical properties and affinities for 17␣-
ethinyl estradiol and methylene blue.
2. Experimental
2.1. Materials
Commercial HPMC samples coded as E4M and J5MS were kindly
supplied by The Dow Chemical Company (Brazil); their charac-
teristics are presented in Table 1. Citric acid (Labsynth, Brazil,
192.12 g/mol), oxalic acid (Labsynth, Brazil, 90.03 g/mol), tereph-
thalic acid (Labsynth, Brazil, 166.13 g/mol), sodium hypophosphite
(Labsynth, Brazil, 87.98 g/mol), methylene blue (MB, M9140 Sigma
Aldrich, 319.85 g/mol) and 17␣-ethinyl estradiol (EE, E4876 Sigma
Aldrich, 296.40 g/mol) were used without further purification.
MilliQ water was used in all experiments. The chemical structures
of HPMC, citric acid, oxalic acid, terephthalic acid, MB and EE are
schematically represented in Fig. 1.
2.2. HPMC aerogels preparation
First HPMC aqueous solutions at 20 g L−1 (2 wt%) in the pres-
ence of crosslinker (citric acid or oxalic acid or terephthalic acid)
at 1.0 g L−1 and sodium hypophosphite at 0.5 g L−1 were prepared.
One should note that using cold water (∼10 ◦ C) as solvent and
vigorous stirring favor the HPMC dissolution. At lower temper-
ature the dispersive forces among the HPMC chains are weaker,
favoring the contact among water molecules and polymer chains
and, thus the mixture enthalpy. At 25 ◦ C the solubilities of tereph-
thalic acid, oxalic acid and citric acid in water are 0.015 g/L,
98.1 g/L and 383 g/L, respectively (http://pubchem.ncbi.nlm.nih.
gov/compound/311#section=Computed-Properties). In the case of
terephthalic acid, it was first dissolved in a small volume (typically
0.5 mL/L) of NH4 OH 10% v/v, then it was added dropwise to the
HPMC solution and then the medium pH was adjusted to pH 7 with
HCl 10% v/v, without any visible aggregation or phase separation.
The concentrations of crosslinker and sodium hypophosphite
were chosen based on previous reports (Bueno, Bentini, Catalani, &
Petri, 2013; Bueno & Petri, 2014; Marani, Bloisi, & Petri, 2015; Reddy
and Yang, 2010;). The solutions were poured into molds made
of polypropylene with dimensions of 2.5 cm × 1.7 cm × 1.3 cm
(height) and frozen at − 18 ◦ C, during 4 h (slow freezing). After
that, the frozen samples were freeze-dried in a LT600 Terroni (Ter-
roni, São Carlos, Brazil) equipment for 24 h. After freeze-drying,
crosslinking of HPMC chains with the acid was achieved by heating
the aerogels at 165 ◦ C for 7 min (Bueno et al., 2013). The crosslink-
ing results from esterification reactions between crosslinker acid
groups and HPMC hydroxyl groups, catalyzed by sodium hypophos-
phite, with release of water molecules (Xinhao et al., 2014; Marani
et al., 2015). The temperature used is sufficient to promote the
esterification reaction and it is far below the polymer decompo-
sition temperature (Table 1). The unreacted sodium phosphite,
crosslinker and HPMC molecules were removed by dialysis against
MilliQ water. The conductivity of the external medium was mea-
sured periodically. The external MilliQ water was exchanged by
fresh MilliQ water until the conductivity achieved ∼5 ␮S/cm. After
that, the aerogels were freeze-dried again and characterized. Con-
sidering that two types of HPMC and three types of crosslinkers
were used, six different types of aerogels were prepared at all. They
were coded as E4M-cit, E4M-oxa, E4M-ter, J5MS-cit, J5MS-oxa and
J5MS-ter, where cit, oxa and ter stand for the crosslinkers citric acid,
oxalic acid and terephthalic acid, respectively.
2.3. HPMC aerogels characterization
The apparent density of aerogels was determined at (24 ± 1) ◦ C
by dividing the mass of freeze-dried aerogels by the correspond-
ing volume, which was estimated by their dimensions. The mean
apparent density values correspond to the average of at least 25
aerogels samples of same type. The chemical stability of HPMC
aerogels was tested by immersing them in methanol, ethanol, iso-
propanol, cyclohexane, chloroform, HNO3 2.0 M, NaOH 0.01 M and
water. All organic solvents were supplied by LabSynth (Brazil) with
99.9% purity. SEM analyses were performed in a Jeol microscope
FEG7401F equipped with a Field-Emission Gun. Samples were pre-
pared by cryo-fracturing freeze-dried aerogels. The cryo-fractured
surfaces were analyzed after gold coating (sputtering) with the
equipment operating at voltage of 5 kV. The compressive tests were
performed for 50 aerogels samples of same type using an Impac,
Digital Dynamometer IP-90DI, with a 10 N load cell, at strain rate
of 0.01 s−1 and at room temperature. The capillary constant (CW )
was determined with a Krüss K100 precision tensiometer (Krüss,
Hamburg Germany), in the sorption mode, at 20 ◦ C, using n-hexane
as test liquid, and the Washburn equation (Eq. (1)) (Galet Patry, &
 B.F. Martins et al. / Carbohydrate Polymers 155 (2017) 173–181 175

Table 1
HPMC characteristics: Degree of substitution (DS), molar substitution (MS), number average molecular weight (Mn ), weight average molecular weight (Mw ), polydispersity
(PD = Mw /Mn ), glass transition temperature (Tg ), decomposition temperature (Tdec ), ashes content (%). USP stands for United States Pharmacopeia. Data (a) provided by
producer, (b) determined by gel permeation chromatography, (c) determined by differential scanning calorimetry and (d) determined by thermogravimetric analysis.

HPMC code USP number DS(a) MS(a) Mn, g/mol (b) Mw , g/mol(b) PD Tg , ◦ C(c) Tdec , ◦ C(d) Ashes (%)(d)

E4M 2910 1.9 0.25 1.1 × 105 2.5 × 105 2.22 172 342 5.7
J5MS 1828 1.5 0.75 1.5 × 105 3.0 × 105 1.96 170 344 9.4

(b), (c) and (d) from Marani et al. (2015).

Fig. 1. Schematic representation of (a) HPMC chemical structure, with R = −H, –CH3 or −CH2 –CHOH–CH3 , (b) citric acid, (c) oxalic acid, (d) terephthalic acid, (e) methylene
blue (MB) and (f) 17␣-ethinyl estradiol (EE).

Dodds, 2010; Ogeda, Silva, Fidale, El Seoud, & Petri, 2012; Stauner,
Silva, El Seoud, Frollini, & Petri, 2013):
m2 2
= CW cos
t 
where m stands for the mass of liquid drawn up by capillarity, t and
 refer to time and contact angle, respectively; ,  and  refer to
the surface tension, viscosity and density, of n-hexane at 20 ◦ C; they
amount to 0.0184 N/m, 0.00031 Pa s and 0.6594 g/cm3 , respectively
(Weast, 1983). Considering that all aerogels samples are completely
wetted by n-hexane (cos  = 1.0), CW can be determined from the
initial slope of m2 as a function of t, as shown in the Supplementary
material Fig. SM1. The mean CW values correspond to the aver-
age of at least 5 aerogels samples of same type and were used for
the determination of the contact angle values  for water, replac-
ing the density, surface tension and viscosity values for water in
the Washburn equation. The contact angle values () for water can
be determined from the initial slope of m2 as a function of t for
water sorption, CW value, which was previously calculated from
sorption curves determined for n-hexane, and ,  and  water
values at 20 ◦ C; they amount to 0.0728 N/m, 0.0010016 Pa s and
0.9982 g/cm3 , respectively (Weast, 1983).
2.4. Adsorption isotherms
Adsorption isotherms were determined for 17␣-ethinyl estra-
diol (EE) and methylene blue (MB) onto HPMC aerogels at
(23 ± 1) ◦ C. The adsorption time was set as 24 h, in order to assure
equilibrium conditions (Kaewprasit, Hequet, Abidi, & Gourlot,
1998), occasionally the samples were manually shaken. Both E4M
and J5MS HPMC carry hydrophobic methyl groups, which inter-
act well with air. Upon inserting the aerogel samples in aqueous
media many air bubbles remain entrapped in the samples. These
(1)
air bubbles need at least 8 h to leave the samples.
The adsorbed amount () was calculated as the concentration of
adsorbed molecule (EE or MB) per mass of aerogel. First, a calibra-
tion curve of absorbance intensity as a function of adsorbate (EE or
MB) concentration was determined by means of spectrophotome-
try in a Beckmann Coulter DU640 spectrophotometer. Then the 
values were determined at (23 ± 1) ◦ C as the difference of concen-
tration of EE or MB in solution, calculated from absorbance data,
before (Ci) and after 24 h in contact with aerogels (Ce ), divided by
the mass of dried aerogel. The equilibrium concentration Ce was the
concentration of EE or MB free in solution after 24 h contact with the
aerogels. The mean  values correspond to the average of at least
duplicates. The removal efficiency of EE or MB by the HPMC aerogels
was calculated dividing the concentration of adsorbed molecule (EE
or MB) by its initial concentration.
In the case of EE the calibration curve was determined at 280 nm,
as a function of EE dissolved in a mixture of ethanol and water (1:4
in volume) (Supplementary material Figure SM2). This solvent mix-
ture allowed complete dissolution of EE. The adsorption isotherms
for EE onto HPMC aerogels were obtained in the initial con-
centration (Ci ) range of 3.374 ␮mol L−1 (1.0 mg/L)–33.74 ␮mol L−1
(10 mg/L). Desorption experiments were carried out by immersing
the EE coated aerogels in pure ethanol for 2 h at 55–60 ◦ C. After this
period of time, the aerogels were removed to avoid re-adsorption,
and the solvent was cooled to room temperature. Aliquots of sol-
176 B.F. Martins et al. / Carbohydrate Polymers 155 (2017) 173–181
vent were removed and the concentration of released EE was
determined by spectrophotometry. The desorption experiments
were performed under heating because it favors the desorption.
According to the Gibbs free energy, the adsorption process is spon-
taneous if the energy released compensates the loss in entropy of
the system, which makes the spontaneous adsorption an exother-
mic process.
The calibration curve for MB was determined at 662 nm, for solu-
tions of MB dissolved in 50 mM Tris-HCl, pH 7.0 (Supplementary
material Figure SM3). The spectra indicated the predominance of
positively charged MB monomers (Tardivo et al., 2005). The slope
of the calibration curve yielded the molar extinction coefficient
of 4.6 × 104 L mol−1 cm−1 , which is in agreement with reported
data (Kaewprasit, Hequet, Abidi, & Gourlot, 1998). For the adsorp-
tion isotherms of MB onto HPMC aerogels the initial concentration
(Ci ) of MB ranged from 1.56 ␮mol/L (0.5 mg/L) to 12.5 ␮mol/L
(4.0 mg/L), at pH 7 (50 mM Tris-HCl buffer), as indicated in the Sup-
plementary material Figure SM4. The buffer condition is important
to avoid the adsorption of MB molecules on the glass vial as well
as their aggregation. Desorption experiments were carried out by
immersing the MB coated aerogels in pure 50 mM Tris-HCl buffer
for 24 h. After this period of time, aliquots of buffer were removed
and the concentration of released MB was determined by spec-
trophotometry at 662 nm.
One-way analysis of variance (ANOVA) with post hoc test was
used to evaluate the differences of variables among groups. A value
of p < 0.05 was considered a significant difference. Analyses were
carried out in Excel 2013® for Windows® (Microsoft Office Home
and Student® , 2013).
3. Results and discussion
3.1. Characterization of HPMC aerogels
The HPMC E4M and J5MS aerogels were prepared by slow
freeze casting technique followed by the sublimation of water
under reduced pressure (freeze-drying). Figure SM5 (Supplemen-
tary material) shows the frontal and side views of J5MS-cit aerogels
to exemplify samples with homogeneous structure (A and B) and
samples with strong lamellar structure (C and D). The latter are due
to the large water volume expansion during crystallization process
(Dash, Li, & Ragauskas, 2012) or to impurities present in the system,
for this reason they were discarded. The homogeneous structure
(Figures SM5A and B) was achieved by rigorous cleaning of vials
and molds and by limiting the mold volume to avoid large water
volume expansion during freezing process, as shown in Supple-
mentary material Figure SM6. Only homogeneous aerogels samples
were considered for the characterization and adsorption studies.
One should note that the mean aerogel mass over 100 samples
was 0.12 ± 0.01 g. The production of homogeneous aerogels can
also be achieved by the use of the fast freezing technique, with liq-
uid nitrogen, for example. However, the use of extreme conditions
of temperature and pressure are commonly avoided in industrial
scale, since it increases the operational costs.
The samples E4M-cit, J5MS-cit and J5MS-oxa were very sta-
ble, whereas the samples crosslinked with terephthalic acid and
E4M-oxa collapsed in water. For this reason, only E4M-cit, J5MS-cit
and J5MS-oxa aerogels were characterized and applied as adsor-
bents. The collapsing of the aerogels in water was a combination
of the hydrophobicity of the walls of the aerogels and the poor
cross-linking efficiency. In contact with water, the thin walls of the
aerogel cannot resist the repulsion forces against the entrance of
water molecules in the structure. The walls collapse creating a gel
that avoids the entrance of water and can keep the inner part of the
aerogel dry even after being immersed in water for one week (Sup-
Table 2
Physical characteristics determined for E4M-cit, J5MS-cit and J5MS-oxa aerogels:
␳ap , CW ,  and ␧ stand for apparent density, capillary constant, contact angle for
water and Young modulus, respectively.
HPMC ␳ap (g/cm3 ) CW (10−14 m5 )  (◦ ) ␧ (kPa)
E4M-cit 0.018 ± 0.002 0.7 ± 0.2 84 ± 2 111 ± 9
J5MS-cit 0.023 ± 0.003 1.0 ± 0.3 75 ± 7 133 ± 13
J5MS-oxa 0.020 ± 0.003 1.0 ± 0.2 64 ± 5 120 ± 13
plementary material Figure SM7). This property could be explored
in applications were the protection of a substrate from water is
needed during a certain period, like in drug release control. The
poor crosslinking efficiency of E4M-ter and J5MS-ter was brought
about by the low solubility of terephthalic acid, even though it
was first dissolved in alkaline medium (see Experimental part for
details). Crystals of terephthalic acid, formed due to the low sol-
ubility in water, could be observed on the surface of E4M-ter and
J5MS-ter samples (Supplementary material Figure SM8). In the case
of E4M-oxa the crosslinking was hindered by the low MS value of
E4M HPMC. J5MS-oxa, J5MS-cit and E4M-cit aerogels resisted over
one year immersed in water, HNO3 2.0 mol/L, methanol, ethanol,
isopropanol, acetone, n-hexane, cyclohexane and chloroform, in
closed vials at (23 ± 2) ◦ C. They did not dissolve or lose the shape.
Noteworthy, the aerogels could be kept in a desiccator for one year
long without losing the shape and after this period no biofilm (fungi
or bacteria) formation could be observed on the surface. On the
other hand, chemical instability was observed in alkaline media,
which promoted the hydrolysis of ester bonds, making the aerogels
soluble.
J5MS-oxa, J5MS-cit and E4M-cit aerogels presented irregular
pore size, ranging from 5 ␮m to 500 ␮m, as evidenced by typical
SEM images in Fig. 2a, b and c, respectively. The apparent density
(␳ap ) values estimated for J5MS-oxa, J5MS-cit and E4M-cit aero-
gels amounted to (0.020 ± 0.002) g/cm3 , (0.023 ± 0.003) g/cm3 and
(0.018 ± 0.002) g/cm3 , respectively, as shown in Table 2. Similar
density values were reported for cellulose aerogels made of waste
newspaper dissolved in 1-allyl-3-methyimidazolium chloride and
subsequent solvent exchange with tert-butanol (Jin, Han, Li, & Sun,
2015) or for cellulose nanofibril aerogels prepared by freeze-drying
(Kim, Youn, & Lee, 2015). Comparing the mean values and the cor-
responding standard deviations there is no significant difference
among them. The aerogels presented an average porosity of 98%,
which is the fraction of the total volume filled by air.
The capillary constant (CW ) values calculated from hexane
sorption curves (see Experimental data in Supplementary material-
SM1) for J5MS-oxa, J5MS-cit and E4M-cit aerogels (Table 2) were
similar. Although capillary rise method used here is a convenient
technique to qualitatively compare the porosity of both aerogels,
quantitative analyses are not warranted because the exact pore
geometry is unknown. Noteworthy, the determination of pore size
and pore size distribution by BET analyses did not work for HPMC
aerogels because the pores collapsed during the measurements.
The contact angle values () presented in Table 2 were determined
from sorption curves of water in J5MS-oxa, J5MS-cit and E4M-cit
aerogels (Supplementary material-SM1). They indicated hydropho-
bicity E4M-cit > J5MS-cit ≈ J5MS-oxa, i.e., the largest DS favored the
hydrophobicity and the crosslinker type did not affect the aerogel
wettability. The average sorption capacity of the aerogels was 17 g
of water per gram of aerogel.
The compressive stress-strain curves determined for E4M-cit,
J5MS-cit and J5MS-oxa aerogels are presented in the Supplemen-
tary material Figure SM9, the corresponding Young compressive
modulus values (␧) were determined from the slopes of linear
regions as (111 ± 9) kPa, (133 ± 11) kPa and (120 ± 13) kPa respec-
tively. They were statistically similar, in agreement with the ␳ap and
 B.F. Martins et al. / Carbohydrate Polymers 155 (2017) 173–181 177

Fig. 2. SEM images of cryo-fracture surfaces obtained for (a) J5MS-oxa, (b) J5MS-cit and (c) E4M-cit aerogels. The bar corresponds to 100 ␮m.

CW values. Aerogels made of alumina and hydroxyl ethyl cellulose, Table 3

Fitting parameters determined from Fig. 3d (EE adsorption onto aerogels) and 4d
HEC, (1 wt%) presented comparable ␧ values (He, Sui, He, & Li, 2015).
(MB adsorption onto aerogels). KF is the Freundlich constant, n is related to the
Aerogels made of cellulose from native wheat straw prepared surface heterogeneity and R2 is related to the fitting quality.
by freeze-drying a 2 wt% solution of cellulose in NaOH/PEG-4000
 mol    1n
(9:1 wt/wt) presented ␧ values one order of magnitude larger than Aerogel adsorbate KF Lg
L
mol
n R2
those found for E4M and J5MS HPMC aerogels (Wan, & Li, 2015b).
J5MS-oxa EE 0.872 0.396 0.7513
These findings indicate that the substitution of cellulose hydroxyl J5MS-cit EE 0.920 0.515 0.9346
groups by methyl and hydropropyl groups weakens the hydrogen E4M-cit EE 0.511 0.439 0.8149
bonding among the chains, making them less stiff. Upon compres- J5MS-cit MB 1.2039 0.578 0.892
sion in the air the E4M and J5MS HPMC aerogels collapsed, and E4M-cit MB 13.24 1.893 0.763
shape recovery was slow, but upon immersing in water they recov-
ered the shape quickly (movie at https://youtu.be/SOhz TGWTaU).
with pure solvent at (24 ± 1) ◦ C was under 1%. Desorption of EE up
If the aerogels were compressed in water, they didn’t lose the origi-
to 37% was only achieved by immersing the EE-coated aerogels into
nal shape and resilience could be clearly observed (movie at https://
pure ethanol for 2 h at 55–60 ◦ C. This result indicates the possibil-
youtu.be/5QTfNYs5KQc). Similar behavior was observed for cellu-
ity of recycling the adsorbent, once higher desorption values can
lose nanofibrils aerogels (Jiang & Hsieh, 2014; Kim et al., 2015).
be achieved in industrial scale, by using strategies like continuous
So far it is possible to conclude that the successful formation of
solvent flow through the aerogel to increase extraction.
HPMC aerogels depends mainly on the high solubility of crosslinker
In order to compare the adsorption behavior on E4M-cit, J5MS-
in water and the availability of hydroxyl groups in the polymer
cit and J5MS-oxa aerogels adsorption isotherms were fitted to
chain. The physical properties of HPMC E4M and J5MS aerogels
adsorption models. Langmuir adsorption model considers homo-
were similar except for the wettability, which was slightly smaller
geneous surfaces (all binding sites have the same energy) and
for E4M aerogels than for J5MS aerogels, due to the higher DS.
dynamic monolayer formation, i.e. molecules adsorb and desorb
under equilibrium conditions. Facing the heterogeneous nature of
3.2. Adsorption of EE onto HPMC aerogels E4M-cit, J5MS-cit and J5MS-oxa aerogels and the extremely weak
desorption, Langmuir model is not adequate to fit the experimental
The adsorbed amount of EE (EE ) onto E4M-cit, J5MS-cit and data. On the other hand, Freundlich empirical model considers the
J5MS-oxa aerogels as a function of initial concentration of EE (Ci ) surface heterogeneity and does not require monolayer adsorption.
is presented in Fig. 3a. The (EE ) values increased linearly with It is described by Eq. (2):
the increase of Ci . However, the slope determined for J5MS-cit and
J5MS-oxa (0.24 ± 0.01 g−1 ) was ∼14% larger than that determined 1
ln  = ln KF + ln Ce (2)
for E4M-cit (0.21 ± 0.01 g−1 ). These results indicate that regarding n
the interaction between aerogels and EE molecules, the type of where KF is the Freundlich constant, the larger KF , the higher the
HPMC seems to be more important than the type of crosslinker. adsorption capacity, n is related to the surface heterogeneity, the
J5MS-cit and J5MS-oxa are more hydrophilic due to the large smaller n, the greater the heterogeneity, and Ce is the equilibrium
MS value, favoring the dipole-dipole interactions between matrix concentration measured by absorbance. The adsorbed amount ( )
hydroxyl groups and EE hydroxyl groups. was calculated as the difference between Ci and Ce , divided by the
Fig. 3b shows the removal (%) of EE by E4M-cit, J5MS-cit and mass of aerogel.
J5MS-oxa aerogels as a function of initial concentration of EE (Ci ). Fig. 3c and d shows the  values as a function of Ce and
In average one piece of E4M-cit, J5MS-cit and J5MS-oxa aero- the fitting to the Freundlich model, respectively, the corre-
gel (0.12 ± 0.01 g) inserted in 30 mL of EE solution presented the sponding fitting parameters are presented in Table 3. Although
removal capacity of (78 ± 5) %, (80 ± 2) % and (86 ± 8) %, respec- the fittings were not perfect, particularly in the case of J5MS-
tively. Comparison among these data performed with ANOVA and oxa (R2 = 0.7513), they indicated that surface heterogeneity
averages over six EE concentrations yielded p = 0.062 (p > 0.05), follows J5MS-oxa > E4M-cit > J5MS-cit and adsorption capacity
indicating that there is no statistical difference among them. Thus J5MS-cit > J5MS-oxa > E4M-cit, confirming the trends observed
all HPMC aerogels presented high removal capacity, which is com- in Fig. 3a. One practical application of HPMC, particularly
parable to the removal capacity of polyamide particles (Han, Qiu, J5MS-cit, aerogels could be the EE pre-concentrating. For envi-
Cao, Hu, & Gao, 2013), polyamide membranes (Han, Qiu, Meng, ronmental control, the typical concentration range of EE in
& Gao, 2012), duckweed (Shi, Wang, Rousseau, & Lens, 2010) or effluents may vary from 1 ng/L (∼3.4 × 10−12 mol L−1 ) to 50 ng/L
activated sludge (Feng et al., 2010). (∼1.7 × 10−10 mol L−1 ) (Shi et al., 2010). Although some commer-
Desorption experiments evidenced that regardless the aerogel cial ELISA kits (Hintemann, Schneider, Ler, & Schneider, 2006), or
type, the amount of desorbed EE molecules after 24 h in contact Chemical Activated LUciferase gene eXpression (CALUX) bioassays
178 B.F. Martins et al. / Carbohydrate Polymers 155 (2017) 173–181

Fig. 3. (a) The adsorbed amount () of EE onto E4M-cit, J5MS-cit and J5MS-oxa aerogels as a function of initial concentration of EE (Ci ). (b) EE removal capacity (%) of
J5MS-oxa (red column), J5MS-cit (gray column) and E4M-cit (blue column) aerogels as a function of Ci , mean mass of aerogel (0.12 ± 0.01) g. (c)  values as a function of
EE equilibrium concentration (Ce ) and (d) the corresponding fitting to the Freundlich adsorption model. Red, blue and black symbols correspond to J5MS-oxa, E4M-cit and
J5MS-cit, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

(Brand et al., 2013) allow the determination of such low EE concen-
tration, pre-concentrating EE onto HPMC aerogels would enable the
detection of EE by less sensitive analytical methods.
3.3. Adsorption of MB onto HPMC aerogels
The adsorption behavior of MB onto E4M-cit, J5MS-cit and J5MS-
oxa aerogels was investigated at pH 7 and (23 ± 1) ◦ C. Fig. 4a shows
the adsorbed amount of MB () onto E4M-cit, J5MS-cit and J5MS-
oxa as a function of initial concentration of MB (Ci ).  values
increased with the MB initial concentration (Ci ); which did not
exceed 12.5 ␮mol/L to avoid the formation of aggregates in the
bulk solution.  values determined for E4M-cit were higher than
those observed for J5MS-cit, while the  values determined for
J5MS-oxa were the lowest. The same tendency was observed for
the maximal removal capacities of E4M-cit, J5MS-cit and J5MS-oxa
(Fig. 4b), which were ∼60%, 40% and 18%, respectively. The capacity
of E4M-cit for MB removal was comparable to that determined for
cellulose nanocrystals alginate beads (Mohammed, Grishkewich,
Berry, & Tam, 2015) and magnetic cellulose/graphene oxide com-
posite (Shi, Li, Zhong, & Xu, 2014), at 298 K. The planar geometry of
MB molecules and the hydrophobic interaction might induce aggre-
gation in solution via ␲-stacking interaction among the aromatic
rings of MB molecules, reducing the MB activity. Such effect would
explain the reduction of removal capacity for MB concentration
larger than 10 ␮mol L−1 .
The charge density in J5MS-cit and E4M-cit aerogels is expected
to be higher than in J5MS-oxa, because citric acid is triprotic and
possibly only two carboxylic acid groups are involved in the esterifi-
cation reaction, remaining one unreacted acid group. The unreacted
carboxylic group is deprotonated at pH 7, increasing the charge
density in the aerogels. On the other hand, oxalic acid is diprotic
and both carboxylic acid groups are involved in the esterification
reaction, remaining no acid group free. The UV/Vis characteris-
tic absorbance bands of MB (Supplementary material Figure SM3)
at 662 nm and 291 nm correspond to the protonated monomeric
MB form (MB+ ) (Impert et al., 2003; Tardivo et al., 2005). There-
fore, the high affinity of MB molecules for J5MS-cit and E4M-cit
aerogels might be driven by electrostatic interaction between MB
positive charges and unreacted carboxylate (stemming from cit-
ric acid) groups present in both aerogels. Moreover, hydrophobic
interaction among the methyl groups belonging to MB and HPMC
molecules also favors the adsorption process (He et al., 2013). One
possibility to explain the more favorable interaction between MB
and E4M-cit is the more hydrophobic character of HPMC E4 M due
to higher DS value, which would favor the binding of MB methyl
groups and methyl groups on the polymer surface. The low affinity
of MB molecules for J5MS-oxa might be due to the lack of electro-
static interaction between MB and the aerogel.
Desorption experiments evidenced that for both aerogel types,
the amount of desorbed MB molecules after 24 h in contact with
pure buffer was under 1%. Thus considering that the adsorption
equilibrium was strongly dislocated to the adsorption direction and
that the surface chemical homogeneity is not granted due to the
crosslinking with citric acid, the experimental data were not fit-
ted to the Langmuir model. The adsorption isotherms of MB onto
 B.F. Martins et al. / Carbohydrate Polymers 155 (2017) 173–181 179

Fig. 4. (a) The adsorbed amount () of MB onto E4M-cit, J5MS-cit and J5MS-oxa aerogels as a function of initial concentration of MB (Ci ). (b) Removal capacity (%) of
J5MS-oxa (red column), J5MS-cit (gray column) and E4M-cit (blue column) aerogels as a function of Ci , mean mass of aerogel (0.12 ± 0.01) g. (c)  values as a function of
MB equilibrium concentration (Ce ) and (d) the corresponding fitting to the Freundlich adsorption model; blue, black and red symbols correspond to E4M-cit, J5MS-cit and
J5MS-oxa, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

E4M-cit and J5MS-cit aerogels presented in Fig. 4c were fitted with
Freundlich empirical model (Eq. (2)). The fitting parameters pre-
sented in Table 3 indicated that J5MS-cit is more heterogeneous
than E4M-cit and adsorption capacity of MB on E4M-cit is ∼10 times
larger than that on J5MS-cit, indicating that the high DS of E4M also
might favor hydrophobic interaction with MB methyl groups. The
n and KF values found for MB on E4M-cit were comparable to those
reported for MB on magnetic cellulose/graphene oxide composite
(Shi et al., 2014). The adsorption isotherm of MB onto J5MS-oxa
aerogels presented in Fig. 4c were converted to ln  and ln Ce , as
presented in Fig. 4d, but they were not fitted with Freundlich empir-
ical model (Eq. (2)) because the ln  values showed a very weak
dependence on the ln Ce values.
The possibility of adsorbent recycling is important for cost
reduction and environmental care. Particularly in the case of E4M-
cit MB, after MB removal from initial media, the samples can be
immersed for 30 min into pure ethanol, which is a good solvent
for MB, to remove MB molecules. After that the aerogels can be air
dried and reinserted in the MB aqueous solution, achieving ∼ the
original 60% removal capacity. This process can be repeated at least
5 times without losing efficiency. A similar behavior was reported
for cellulose nanocrystals and alginate beads and MB (Mohammed
et al., 2015). Another possibility to recycle the MB coated aerogels
was irradiating the samples with visible light to promote the pho-
tobleaching of MB molecules. Upon absorbing light MB molecules
achieve singlet and triplet excited states; the excited states can
be deactivated by spontaneous decay or by reaction with molec-
ular oxygen, as well described in the literature (Tardivo et al.,
2005). For instance, the MB molecules present strong absorbance
at 632.8 nm, as presented in the Supplementary material Figure
SM3a. After exposing a MB coated E4M-cit aerogel during 15 min
to a 3 mW laser beam, the irradiated spot was successfully decol-
orized, as shown in Fig. 5a. For comparison, an E4M-cit aerogel was
also exposed to the same laser beam during 15 min (Fig. 5b). It
is interesting to observe that in the absence of MB (Fig. 5b) the
beam light penetrated into the aerogel and the light was scattered
inside the aerogel. Nevertheless, it was not possible to observe the
beam light penetration into the MB coated aerogel surface (Fig. 5a)
because it was absorbed by the adsorbed MB molecules. Exposing
MB coated E4M-cit or J5MS-cit aerogels (samples corresponding to
the highest removal% in Fig. 4b) to a light-emitting diode for 48 h
promoted efficient decolorization (Fig. 5c). After photobleaching,
the samples were immersed in MilliQ water in order to remove
the main products (CO2 , SO4 2− , NH4+ and NO3 − ) resulting from
MB photodecomposion (Houas et al., 2001) and freeze-dried for
reuse. The photobleached aerogels could be reused at least 5 times
(Fig. 5c), retaining 90% of the original removal capacity.
4. Conclusions
Novel aerogels based on HPMC were created, characterized and
applied as adsorbents for EE and MB, two organic pollutants. The DS
and MS values of E4M and J5MS HPMC had no significant effect on
the density and compressibility modulus of the resulting aerogels,
180 B.F. Martins et al. / Carbohydrate Polymers 155 (2017) 173–181

Fig. 5. (a) MB coated E4M-cit and (b) bare E4M-cit aerogels exposed to laser beam (␭ = 632.8 nm, 3 mW), during 15 min and the resulting effects. (c) MB coated E4M-cit
exposed 48 h to a light-emitting diode became decolorized; this particular sample was reused 5 times.

but the wettability of E4M aerogels was slightly smaller than that of
J5MS aerogels, due to the higher DS of E4M HPMC. The crosslinkers
efficiency was directly dependent on their water solubility because
the aerogels stemmed from aqueous solutions; the higher the water
solubility the better the crosslinking. The E4M and J5MS HPMC
aerogels presented outstanding resilience and chemical stability
against alcohols, acetone, hydrocarbons and acid media.
The HPMC structural aspects (DS and MS values) affected sig-
nificantly the adsorption behavior of EE and MB on the aerogels.
The adsorption of EE onto J5MS HPMC aerogels was more favored
than that on E4M HPMC aerogels, revealing that more hydrophilic
(larger MS value) surfaces are more suitable for EE removal. On
the other hand, the interaction between MB molecules and E4M
HPMC aerogels was more favored than with J5MS aerogels, proba-
bly due to the binding of MB methyl groups and methyl groups on
the E4M HPMC (larger DS value) surface. Aerogels could be recycled
either by dissolving the adsorbate (EE or MB) in a good solvent or
by photodegradation, as in the case of MB.
Acknowledgments
Authors gratefully acknowledge financial support from Brazil-
ian Funding Agencies Fundação de Amparo à Pesquisa do
Estado de São Paulo (FAPESP Grant 2015/25103-2) and Con-
selho Nacional de Desenvolvimento Científico e Tecnológico (CNPq
Grants 448497/2014-0 and 305178/2013-0).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.carbpol.2016.08.
082.
References
Ankley, G. T., Bencic, D. C., Breen, M. S., Collette, T. W., Conolly, R. B., Denslow, N. D.,
et al. (2009). Endocrine disrupting chemicals in fish: Developing exposure
 B.F. Martins et al. / Carbohydrate Polymers 155 (2017) 173–181
indicators and predictive models of effects based on mechanism of action.
Aquatic Toxicology, 92, 168–178.
Brand, W., de Jongh, C. M., van der Linden, S. C., Mennes, W., Puijker, L. M., van
Leeuwen, C. J., et al. (2013). Trigger values for investigation of hormonal
activity in drinking water and its sources using CALUX bioassays. Environment
International, 55, 109–118.
Bueno, V. B., & Petri, D. F. S. (2014). Xanthan hydrogel films: Molecular
conformation, charge density and protein carriers. Carbohydrate Polymers, 101,
897–904.
Bueno, V. B., Bentini, R., Catalani, L. H., & Petri, D. F. S. (2013). Synthesis and swelling
behavior of xanthan-based hydrogels. Carbohydrate Polymers, 92, 1091–1099.
Dash, R., Li, Y., & Ragauskas, A. J. (2012). Cellulose nanowhisker foams by freeze
casting. Carbohydrate Polymers, 88, 789–792.
Feng, Y., Zhang, Z., Gao, P., Su, H., Yu, Y., & Ren, N. (2010). Adsorption behavior of
EE2 (17␣-ethinylestradiol) onto the inactivated sewage sludge: Kinetics,
thermodynamics and influence factors. Journal of Hazardous Materials, 175,
970–976.
Feng, J., Nguyen, S. T., Fan, Z., & Duong, H. M. (2015). Advanced fabrication and oil
absorption properties of super-hydrophobic recycled cellulose aerogels.
Chemical Engineering Journal, 270, 168–175.
Galet, L., Patry, S., & Dodds, J. (2010). Determination of the wettability of powders
by the Washburn capillary rise method with bed preparation by a centrifugal
packing technique. Journal of Colloid and Interface Science, 346, 470–475.
García-González, C. A., Alnaief, M., & Smirnova, I. (2011). Polysaccharide-based
aerogels—Promising biodegradable carriers for drug delivery systems.
Carbohydrate Polymers, 86, 1425–1438.
Grassi, M., Kaykioglu, G., Belgiorno, V., & Lofrano, G. (2012). Removal of emerging
contaminants from water and wastewater by adsorption process. In G. Lofrano
(Ed.), Emerging compounds removal from wastewater: Natural and solar based
treatments (pp. 15–38). London: Springer.
Han, J., Qiu, W., Meng, S., & Gao, W. (2012). Removal of ethinylestradiol (EE2) from
water via adsorption on aliphatic polyamides. Water Research, 46, 5715–5724.
Han, J., Qiu, W., Cao, Z., Hu, J., & Gao, W. (2013). Adsorption of ethinylestradiol
(EE2) on polyamide 612: Molecular modeling and effects of water chemistry.
Water Research, 47, 2273–2284.
He, X., Male, K. B., Nesterenko, P. N., Brabazon, D., Paull, B., & Luong, J. H. T. (2013).
Adsorption and desorption of methylene blue on porous carbon monoliths and
nanocrystalline cellulose. ACS Applied Materials & Interfaces, 5, 8796–8804.
He, F., Sui, C., He, X., & Li, M. (2015). Facile synthesis of strong alumina–cellulose
aerogels by a freeze-drying method. Materials Letters, 152, 9–12.
Hintemann, T., Schneider, C., Ler, H. F. S., & Schneider, R. J. (2006). Field study using
two immunoassays for the determination of estradiol and ethinylestradiol in
the aquatic environment. Water Research, 40, 2287–2294.
Houas, A., Lachheb, H., Ksibi, M., Elaloui, E., Guillard, C., & Herrmann, J. M. (2001).
Photocatalytic degradation pathway of methylene blue in water. Applied
Catalysis B: Environmental, 31, 145–157.
Impert, O., Katafias, A., Kita, P., Mills, A., Pietkiewicz-Graczyk, A., & Wrzeszcz, G.
(2003). Kinetics and mechanism of a fast leuco-Methylene Blue oxidation by
copper(II)–halide species in acidic aqueous media. Dalton Transactions, 350,
348–353.
Jiang, F., & Hsieh, Y.-L. (2014). Amphiphilic superabsorbent cellulose nanofibril
aerogels. Journal of Materials Chemistry A, 2, 6337–6342.
Jin, C., Han, S., Li, J., & Sun, Q. (2015). Fabrication of cellulose-based aerogels from
waste newspaper without any pretreatment and their use for absorbents.
Carbohydrate Polymers, 123, 150–156.
Kaewprasit, C., Hequet, E., Abidi, N., & Gourlot, J. P. (1998). Application of
methylene blue adsorption to cotton fiber specific surface area measurement:
Part I. Methodology. The Journal of Cotton Science, 2, 164–173.
Kim, C. H., Youn, H. J., & Lee, H. L. (2015). Preparation of cross-linked cellulose
nanofibril aerogel with water absorbency and shape recovery. Cellulose, 22,
3715–3724.
181
Marani, P. L., Bloisi, G. D., & Petri, D. F. S. (2015). Hydroxypropylmethyl cellulose
films crosslinked with citric acid for control release of nicotine. Cellulose, 22,
3907–3918.
Martins, M., Barros, A. A., Quraishi, S., Gurikov, P., Raman, S. P., Smirnova, I., et al.
(2015). Preparation of macroporous alginate-based aerogels for biomedical
applications. Journal of Supercritical Fluids, 106, 152–159.
Mohammed, N., Grishkewich, N., Berry, R. M., & Tam, K. C. (2015). Cellulose
nanocrystal–alginate hydrogel beads as novel adsorbents for organic dyes in
aqueous solutions. Cellulose, 22, 3725–3738.
Ogeda, T., Silva, I. B., Fidale, L. C., El Seoud, O. A., & Petri, D. F. S. (2012). Effect of
cellulose physical characteristics, especially the water sorption value, on the
efficiency of its hydrolysis catalyzed by free or immobilized cellulase. Journal of
Biotechnology, 157, 246–252.
Rashed, M. N. (2013). Adsorption technique for the removal of organic pollutants
from water and wastewater. In M. N. Rashed (Ed.), Organic pollutants –
monitoring, risk and treatment (pp. 167–194). Rijeka, Croatia: InTech.
Reddy, N., & Yang, Y. Q. (2010). Citric acid cross-linking of starch films. Food
Chemistry, 118, 702–711.
Shi, W., Wang, K., Rousseau, D. P. L., & Lens, P. N. L. (2010). Removal of estrone,
17␣-ethinylestradiol, and 17ß-estradiol in algae and duckweed-based
wastewater treatment systems. Environmental Science and Pollution Research,
17, 824–833.
Shi, H., Li, W., Zhong, L., & Xu, C. (2014). Methylene blue adsorption from aqueous
solution by magnetic Cellulose/Graphene oxide composite: Equilibrium,
kinetics, and thermodynamics. Industrial & Engineering Chemistry Research, 53,
1108–1118.
Siqueira, G., Bras, J., & Dufresne, A. (2010). Cellulosic bionanocomposites: A review
of preparation. Properties and Applications of Polymers, 2, 728–765. http://dx.
doi.org/10.3390/polym2040728
Snyder, S., Westerhoff, P., Yoon, Y., & Sedlak, D. (2003). Pharmaceuticals, personal
care products, and endocrine disruptors: Implications for the water industry.
Environmental Engineering Science, 20, 449–469.
Stauner, T., Silva, I. B., El Seoud, O. A., Frollini, E., & Petri, D. F. S. (2013). Cellulose
loading and water sorption value as important parameters for the enzymatic
hydrolysis of cellulose. Cellulose, 20, 1109–1119.
Tardivo, J. P., Del Giglio, A., de Oliveira, C. C., Gabrielli, D. S., Junqueira, H. C., Tada,
D. B. S., et al. (2005). Methylene blue in photodynamic therapy: From basic
mechanisms to clinical applications. Photodiagnosis and Photodynamic Therapy,
2, 175–191.
Vutskits, L., Briner, A., Klauser, P., Gascon, E., Dayer, A. G., Kiss, J. Z., et al. (2008).
Adverse effects of methylene blue on the central nervous system.
Anesthesiology, 108, 684–692.
Wan, C., & Li, J. (2015a). Cellulose aerogels based on a green NaOH/PEG solution:
Preparation, characterization and influence of molecular weight of PEG. Fibers
and Polymers, 16, 1230–1236.
Wan, C., & Li, J. (2015b). Facile synthesis of well-dispersed superparamagnetic
␥-Fe2 O3 nanoparticles encapsulated in three-dimensional architectures of
cellulose aerogels and their applications for Cr(VI) removal from contaminated
water. ACS Sustainable Chemistry & Engineering, 3, 2142–2152.
Wan, C., & Li, J. (2016). Cellulose aerogels functionalized with polypyrrole and
silver nanoparticles: In-situ synthesis, characterization and antibacterial
activity. Carbohydrate Polymers, 146, 362–367.
Wang, X., Xu, S., Du, T. J., & Wang, J. (2016). Synthesis and characterization of a
porous and hydrophobic cellulose-based composite for efficient and fast
oil–water separation. Carbohydrate Polymers, 140, 188–194.
Weast, R. C. (1983). Handbook of physical chemistry (64th ed.). CRC Press Inc.
Xinhao, F., Xiao, Z., Sui, S., Wan, Q., & Xie, Y. (2014). Esterification of wood with
citric acid: The catalytic effects of sodium hypophosphite (SHP). Holzforschung,
68, 427–433.