See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/320581572

Preparation of Porous 2,3-dialdehyde Cellulose Beads Crosslinked with

Chitosan and their Application in Adsorption of Congo Red Dye

Article  in  Carbohydrate Polymers · October 2017

DOI: 10.1016/j.carbpol.2017.10.072

CITATIONS READS

16 234

3 authors:

Chang-Qing Ruan Maria Strømme

Southwest University in Chongqing Uppsala University
20 PUBLICATIONS   307 CITATIONS    361 PUBLICATIONS   7,138 CITATIONS   

SEE PROFILE SEE PROFILE

Jonas Lindh
Uppsala University
44 PUBLICATIONS   640 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Conducting Redox Polymer Batteries View project

All content following this page was uploaded by Chang-Qing Ruan on 27 October 2018.

The user has requested enhancement of the downloaded file.

 Contents lists available at ScienceDirect

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

Preparation of porous 2,3-dialdehyde cellulose beads crosslinked with

chitosan and their application in adsorption of Congo red dye
⁎
Chang-Qing Ruan, Maria Strømme, Jonas Lindh
Nanotechnology and Functional Materials, Department of Engineering Sciences, Uppsala University, Box 534, 75121 Uppsala, Sweden

A R T I C L E I N F O A B S T R A C T

Keywords: Micrometer sized 2,3-dialdehyde cellulose (DAC) beads were produced via a recently developed method relying
Cladophora nanocellulose on periodate oxidation of Cladophora nanocellulose. The produced dialdehyde groups and pristine hydroxyl
2,3-Dialdehyde cellulose beads groups provided the DAC beads with a vast potential for further functionalization. The sensitivity of the DAC
Chitosan beads to alkaline conditions, however, limits their possible functionalization and applications. Hence, alkaline-
Crosslink
stable and porous cellulose beads were prepared via a reductive amination crosslinking reaction between 2,3-
Congo red dye
dialdehyde cellulose beads and chitosan. The produced materials were thoroughly characterized with different
methods. The reaction conditions, including the amount of chitosan used, conditions for reductive amination,
reaction temperature and time, were investigated and the maintained morphology of the beads after exposure to
1 M NaOH (aq.) was verified with SEM. Different washing and drying procedures were used and the results were
studied by SEM and BET analysis. Furthermore, FTIR, TGA, EDX, XPS, DLS and elemental analysis were per-
formed to characterize the properties of the prepared beads. Finally, the alkaline-stable porous chitosan cross-
linked 2,3-dialdehyde cellulose beads were applied as adsorbent for the dye Congo red. The crosslinked beads
displayed fast and high adsorption capacity at pH 2 and good desorption properties at pH 12, providing a
promising sorption material.

1. Introduction which is known to be a human carcinogen (Ahmad & Kumar, 2010).

The pollution caused by dyes, including Congo red, from waste streams
The contamination of the environment by chemical waste is a pro- generated by the textile industry is a health and environmental concern.
blem of great concern and solutions to limit the release are urgently Hence, efficient treatment of dye containing waste is essential before
needed. An efficient way to sequester the waste is to adsorb it using the release into the environment (Ghorai, Sarkar, Panda, & Pal, 2013).
suitable materials with good sorption properties (Mahfoudhi & Boufi, As Congo red is fairly stable, it is not easily removed by biodegradation,
2017). To be in line with the concept of green processes the sorption which renders adsorption of Congo red an attractive way to limit its
materials should be prepared in an environmentally benign way and release into nature. Many materials have been investigated to be used
preferably rely on a renewable feedstock (Divakaran & Pillai, 2002; as adsorbents for Congo red, such as carbon based materials
Habiba et al., 2017). Such materials also need to be designed in an (Ahmad & Kumar, 2010; Purkait, Maiti, DasGupta, & De, 2007), silica
appropriate way to efficiently meet the practical demands of the ad- composites (Ghorai et al., 2013), cellulosic materials (Jin, Li, Xu, & Sun,
sorption process, e.g. to adsorb waste from the liquid phase it should be 2015) and chitosan related materials (Vakili et al., 2014). Among them,
convenient to shape the sorption material as a membrane or as sphe- low cost, green, and sustainable cellulose and chitosan would be sui-
rical particles (Ruan, Strømme, & Lindh, 2016). Furthermore, the ma- table candidates for producing the desirable adsorbent for Congo red.
terials should have suitable coordinating groups to efficiently capture The bead shape is beneficial compared to membranes or irregularly
and bind to the waste compounds. A common type of chemical waste in shaped particles in many applications as it lowers the backpressure
the liquid phase is dyes. Dyes are used in a number of industrial pro- under flow condition. Cellulose beads are commonly defined as sphe-
cesses and often generate large amounts of waste (Shi, Li, Wang, rical particles with a diameter of a few micro- or millimeters prepared
Feng, & Tang, 2007; Wan Ngah, Teong, & Hanafiah, 2011). from cellulose and have been used in many areas, such as chromato-
Congo red, disodium-3,3′-([1,1′-biphenyl]-4,4′-diyl)bis(4-amino- graphy (Wang & Sun, 2007), controlled drug release (Trygg, Yildir,
naphthalene-1-sulfonic acid) is a benzidine-based anionic biazo dye, Kolakovic, Sandler, & Fardim, 2015), sodium-ion batteries (Dingfeng

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

http://dx.doi.org/10.1016/j.carbpol.2017.10.072
Received 12 June 2017; Received in revised form 19 September 2017; Accepted 22 October 2017
C.-Q. Ruan et al.

et al., 2016) and noble metal adsorption (Ruan et al., 2016). Since the The aim of the present study was to prepare a material capable of
first report of preparation of cellulose pellets published in 1951 (O’neill, efficient adsorption of the dye Congo red made from the renewable
1951), various approaches have been developed to prepare cellulose polymers cellulose and chitosan.
beads (Gericke, Trygg, & Fardim, 2013). A typical procedure for the
preparation of cellulose beads follows three steps (Gericke et al., 2013): 2. Experimental
(i) dissolution of cellulose (or related derivatives) in special solvent e.g.
alkali/urea aqueous solution (Luo & Zhang, 2013), (ii) cellulose bead 2.1. Materials
shaping in solution, (iii) coagulation and solidification of formed cel-
lulose solution droplets. Recently, a new method for preparing 2,3- Nanocellulose from Cladophora algae was provided by FMC
dialdehyde cellulose (DAC) beads via spontaneous formation of the Biopolymer (USA). Sodium metaperiodate (NaIO4), chitosan (low mo-
beads at high degrees of oxidation during periodate oxidation of Cla- lecular weight, 75–85% deacetylated), sodium cyanoborohydride
dophora cellulose, was report (Lindh, Carlsson, Strømme, & Mihranyan, (NaBH3CN), Congo red (C32H22N6Na2O6S2) and other chemicals used
2014). The method relies on a one-pot procedure conducted under were of analytical or reagent grade and were used as received from
aqueous conditions, which avoids some tedious steps generally in- Sigma-Aldrich. Deionized water was used throughout the experimental
volved in cellulose beads production such as special solvents, chemical procedures.
reagents and sophisticated equipment. The produced DAC beads are,
however, not stable under alkaline conditions, which might limit the 2.2. Preparation of DAC (2,3-dialdehyde cellulose)
beads applications (Lindh, Ruan, Strømme, & Mihranyan, 2016). Dia-
ldehyde cellulose is sensitive to beta-elimination (Potthast, Schiehser, The preparation of porous DAC microbeads was performed ac-
Rosenau, & Kostic, 2009); a possible mechanism explaining the beta- cording to a previously published method with minor modifications
elimination is shown in Scheme 1a. Under alkaline condition, base (Lindh et al., 2014). Briefly, Cladophora nanocellulose, 20 g dispersed in
catalyzes the formation of enolate in the DAC unit while the beta- 1.8 L of water was added into 134 g of sodium metaperiodate (5 mol
elimination makes the DAC chain break. In order to enhance the sta- per mol of anhydroglucose units) dispersed in 0.2 L water. The peri-
bility of DAC beads and improve their usability, we have previously odate-containing reaction mixture was carefully wrapped in aluminum
demonstrated that crosslinking the beads with diamines via reductive foil to avoid light exposure. The reaction mixture was gently stirred at
amination can render the beads stable to alkaline conditions (Lindh room temperature in the dark for 10 days. The reaction mixture was
et al., 2016). In the herein presented work we further developed this quenched via the addition of ethylene glycol and washed four times by
concept by employing chitosan as a polyamine crosslinker as shown in centrifugation/washing/redispersion in water to provide pure DAC.
Scheme 1b to produce DAC-chitosan (DACeCS). DAC samples were kept in water at 4 °C for further use.
Chitosan is a pseudo natural polysaccharide, derived from chitin
shells of crustaceans and composed of glucosamine and N-acetyl glu-
2.3. Preparation of porous chitosan cross-linked DAC beads
cosamine (Rinaudo, 2006). As chitosan is derived from chitin, one of
the most abundant biopolymers in nature, it is almost inexhaustible.
Never-dried DAC was dispersed in deionized water to obtain a DAC
Chitosan and chitosan-based composite materials have been frequently
dispersion with good dispersibility at a concentration of 2 mg/mL.
employed as sorption materials (Wan Ngah et al., 2011). Glutar-
Chitosan was dissolved in HCl solution with a concentration of
aldehyde is frequently used as a crosslinker for chitosan
19.2 mg/mL and the pH of the solution was controlled at a range of
(Monteiro & Airoldi, 1999) and in analogy there are plenty of aldehyde
2.5–3.5, a process which required > 3 h to obtain a transparent solu-
groups in DAC, which could be crosslinked by chitosan (Teotia,
tion. After the dissolution, the chitosan solution was added into the
Ikram, & Gupta, 2012; Yunhui & Yongjin, 2011). Furthermore, chitosan
DAC dispersion dropwise (about 1 mL/min) with stirring of the DAC
can be dissolved in acidic solutions if the pH is lower than 6 and the
dispersion in a round-bottom flask. Sodium cyanoborohydride
average degree of acetylation (DA) is not higher than 0.5 (Rinaudo,
(1.25 mol per mol of chitosan) was added into the reaction mixture.
2010). The dissolved chitosan would be more prone to undergo cross-
After the desired reaction time, the product, named DACeCS, was
linking reaction with DAC dispersion. Moreover, the introduced amine
purified two times by centrifugation/washing/redispersion in diluted
groups from the chitosan will enhance the functionality and possible
HCl solution (about 8 mM) and two times by centrifugation/washing/
applications of DAC beads, such as an adsorption material for dyes.
redispersion in water and kept in water for further use. All procedures

Scheme 1. a) Possible mechanism of beta-elimination of DAC in aqueous alkaline solution, b) reaction scheme for the preparation of DACeCS.
C.-Q. Ruan et al.

were performed in a fume hood. 2.12. Adsorption and desorption experiments for Congo red

2.4. Alkaline stability A Congo red dye in water at a concentration of 100 mg/L was used
as adsorbate in the sorption experiments. Congo red solutions of pH 2,
Samples were treated with 1 M NaOH (aq.) to verify their stability 5, 7, 9 and 12 were prepared and adjusted with HCl and NaOH. For
under alkaline aqueous conditions. DAC sample, 50 mg, was immersed adsorption experiments at different pH, 20 mg DACeCS beads were
in 1 M NaOH (aq.) for 1 h and then washed by centrifugation/washing/ added into 40 mL Congo red solution at specific pH. Aliquots were
redispersion in water to obtain the purified sample, which was then withdrawn and centrifuged after 3 and 22 h of contact time. The con-
dried at 45 °C under vacuum overnight. The weight of the dried sample centration of supernatants was determined by a UV–vis spectro-
was used to calculate the loss percentage after NaOH (aq.) treatment. photometer (Shimadzu, UV-1800, Japan). As the absorbance peaks for
The treated samples were named DAC-CS-NaOH. Congo red at different pH are different from each other (Supporting
information, Fig. S1), the measurements were done at wavelengths of
2.5. Scanning electron microscopy (SEM) and energy-dispersive X-ray 570, 499, 489, 500 and 495 nm for solutions at pH 2, 5, 7, 9 and 12,
(EDX) analysis respectively. pH-specific standard calibration curves were prepared at
these wavelengths and used for determining concentration of Congo red
The morphology of the produced materials was observed with a LEO at different pH.
1550 field-emission SEM instrument (Zeiss, Germany) at an accel- The effect of contact time on the percentage of adsorption was
erating voltage of 1 kV with an in-lens secondary electron detector. studied in detail. DACeCS beads (20 mg) were added into 40 mL of
Samples were mounted on aluminum stubs by double-sided adhesive 100 mg/L Congo red solutions with pH 2 or 7 to undergo adsorption.
carbon tape and coated with Au/Pd to reduce charging effects. EDX After 10, 20, 30, 60, 120, 180, 1200 and 1620 min, aliquots were
analysis was performed on the same LEO 1550 field-emission SEM in- withdrawn and centrifuged and the concentrations of supernatants
strument (Zeiss, Germany) but at an accelerating voltage of 20 kV with were determined by UV-vis.
a map model to obtain the content of the relative elements of the Desorption experiments were performed with NaOH aqueous solu-
samples. tion (pH ∼12). DACeCS beads being exposed to Congo red at pH 2
(about 20 mg DACeCS beads loaded with 4 mg Congo red) were dis-
2.6. Nitrogen sorption isotherm analysis persed in NaOH (aq., pH ∼12) for 3 h. Aliquots were withdrawn and
centrifuged, and the concentration of supernatants was determined by
Nitrogen gas adsorption and desorption isotherms of samples were UV–vis. The DACeCS beads desorbed with NaOH were purified with
recorded with an ASAP 2020 instrument (Micromeritics, USA). The deionized water and exposed to another cycle of adsorption and deso-
specific surface area (SSA) was determined based on the BET (Brunauer, rption following the same procedure.
Emmett, & Teller, 1938) method during adsorption with the ASAP 2020 The adsorption and desorption capacity was determined by calcu-
software while the pore size distribution was calculated according to lating the differences between the values before and after adsorption
the density functional theory (DFT) method with the model for nitrogen and desorption (Ca = [Co − Cf1] * V/Wo, Cd = Cf2 * V/Wo, Ca and Cd
at 77 K. are the adsorption and desorption capacity, and Co is the concentration
of Congo red solution before adsorption, Cf1 is the final concentration of
2.7. Fourier transform infrared (FTIR) spectroscopy Congo red solution after adsorption, V is the volume of solution, W0 is
the weight of adsorbent, Cf2 is the final concentration of Congo red
FTIR spectra were recorded on a Tensor 27 FTIR spectrometer solution after desorption.)
(Bruker, Germany). The resolution was 4 cm−1 with 64 scans at a range A simple adsorption column was prepared by adding never-dried
of 4000–400 cm−1. Samples were ground with KBr at a percentage of DACeCS into a glass pipette with a diameter of about 5 mm (about
1 wt% and pressed into pellets for FTIR measurements. 1 mm at bottom) with a glass fiber plug in the bottom to prevent
leakage of DACeCS. The adsorption and desorption was performed
2.8. Thermal gravimetric analysis (TGA) with Congo red solution at pH 2 and in NaOH aqueous solution (pH
∼12) respectively, and recorded by camera.
Samples were analyzed with a TGA/SC 3+ instrument (Mettler
Toledo, Switzerland) at a temperature range of 25–800 °C with a tem- 3. Results and discussion
perature increase of 5 °C per minute under nitrogen atmosphere.
3.1. Synthesis of DACeCS beads and SEM characterization
2.9. CHN elemental analysis
The spherical morphology of materials used as matrix under flow
CHN elemental analysis of purified and dried samples was per- conditions offer low backpressure, where DAC beads are highly sui-
formed by Analytische Laboratorien (Lindlar, Germany). table. However, during the desorption of Congo red, aqueous alkaline
solution is conveniently used, so high stability of the matrix in aqueous
2.10. X-ray photoelectron spectroscopy (XPS) NaOH solution is an essential property. Upon exposure to alkaline so-
lution (1 M NaOH solution) the DAC beads were almost completely
High-resolution XPS spectra for C, N and O were recorded with an dissolved after 1 h of alkaline treatment and the color of the mixture
ESCA instrument (Physical Electronics, U.S.) equipped with an Al Kα changed to brown (Lindh et al., 2016). Hence, crosslinking of DAC
source (1486.6 eV, 45 W). The analysis was performed at 23.5 eV pass beads to enhance their stability in aqueous NaOH solution is crucial for
energy with electron and Ar+ Gun neutralization to compensate for making them reusable.
surface charging effects. The preparation of DAC beads was conducted as in previously
published work (Lindh et al., 2014) and the never-dried DAC beads
2.11. Particle size measurement were used for the synthesis of DACeCS beads, where several para-
meters, including the amount of chitosan, reaction temperature and
Samples were dispersed in water and sonicated and then analyzed time, and different synthesis approaches were investigated to obtain
with laser diffraction using a Mastersizer 3000 instrument (Malvern, useful DACeCS beads with suitable properties, such as suitable por-
England). osity, SSA and high stability under alkaline conditions.
C.-Q. Ruan et al.

During the screening of synthesis conditions, the following condi- consideration to study its effect on the stability of DACeCS. Reaction
tions were selected as starting points based on a rough first screen: the times of 2, 4, 6, 8, 24, 48 and 96 h were selected. Fig. 1 depicts beads of
amount of chitosan was 2 equivalents, the reaction time was 24 h, the DACeCS samples produced at different reaction times and the corre-
temperature was room temperature (22.5 °C), the chitosan solution was sponding samples treated with 1 M NaOH (aq.), where samples ob-
added drop-wise with in-situ reduction using NaBH3CN. When one tained from reactions lasting 6, 8 and 24 h were observed to have a
parameter was altered, the other parameters were fixed. more intact bead morphology compared with those resulting from re-
To investigate the effect of the amount of crosslinking agent used, actions lasting 2, 4, 48 and 96 h, which indicated that 2 and 4 h were
0.1, 0.5, 1, 2 and 5 equivalents of chitosan relative to DAC were em- too short reaction times whereas 48 and 96 h were too long. The
ployed, where the corresponding SEM images of the products after crosslinking reaction between DAC and chitosan was insufficient in
NaOH treatment are shown in Fig. S2. In these images, beads with short reaction times such as 2 h, while side reactions might take place to
different degrees of morphological disruption could be observed, and destroy the beads if too long reaction time, such as 96 h was employed.
the fraction of broken beads decreased as the amount of chitosan in- Interestingly, it was found that the yields of samples obtained from 6
creased, which means that the stability of beads in 1 M NaOH (aq.) and 24 h reaction after the first exposure to 1 M NaOH (aq.) were both
increased as the amount of chitosan increased. Satisfyingly, the results about 40%, whereas yields of the second time NaOH treatment were
verified that cross-linking DAC with chitosan could enhance the stabi- almost 100% for both samples. As the yields of samples from the 6 and
lity of the beads in alkaline solution. No obvious differences with regard 24 h reactions after 1 M NaOH treatment were similar, 6 h was selected
to intact beads could be observed when increasing the chitosan amount as the reaction time for subsequent experiments. These results de-
from 2 to 5 equivalents, hence 2 equivalents of chitosan was used in monstrated that the crosslinking reaction between DAC and chitosan
subsequent experiments. was not completely successful in keeping the DAC intact in alkaline
Three different reaction temperatures, 22.5 °C, 40 °C and 80 °C, solution, and some parts, which might not be highly cross-linked, dis-
were employed during the cross-linking reaction to investigate the ef- solved in 1 M NaOH, but after the alkaline treatment DACeCS became
fect of temperature on the coupling efficiency. The corresponding SEM stable towards further degradation in alkaline solution even after pro-
images (Fig. S3) of the samples before and after treatment with NaOH longed exposure to 1 M NaOH for up to 24 h (Fig. S12).
(aq.) demonstrated that the products produced at room temperature In addition, different synthesis procedures were screened to find
had the highest stability. In the work by Kim and co-workers (Kim, suitable conditions. Firstly, the effect of reduction with NaBH3CN was
Wada, & Kuga, 2004), they reported severe decomposition of dialde- studied. In Scheme 1b, it is shown that after the Schiff’s base reaction
hyde cellulose in water at high temperature (80 °C), which might be the between DAC and chitosan (product named DAC]CS), the formed
reason why the product produced at room temperature keep the most imine is reduced to an amine with NaBH3CN, so it was deemed inter-
intact bead morphology after 1 M NaOH treatment. Thus, subsequent esting to study whether the reduction reaction was efficient. DAC]CS
preparations were performed at room temperature. (imine) was collected and treated with NaOH, and the SEM images
The reaction time for the reductive amination was also taken into before and after NaOH treatment are show in Fig. 2a and c,

Fig. 1. SEM micrographs of DACeCS samples ob-

tained at different reaction times: (a) 2 h, (b) 4 h, (c)
6 h, (g) 8 h, (h) 48 h (i) 96 h and the corresponding
samples after treatment with 1 M NaOH (aq.): (d),
(e), (f), (j), (k), (l).
C.-Q. Ruan et al.

Fig. 2. SEM micrographs of sample (a) DAC]CS and

(b) DAC-CS-2 and the corresponding samples (c), (d)
treated with 1 M NaOH (aq.).

respectively. The never-dried imine (DAC]CS) was reduced to the Finally, we checked the result of drop-wise adding the DAC dis-
corresponding amine using NaBH3CN (named DAC-CS-2, 2 referring to persion into the chitosan solution. The SEM micrographs of samples
the two-pot synthesis) in deionized water, which was also treated with produced before and after treatment with 1 M NaOH (aq.) are shown in
1 M NaOH and then studied using SEM (images are shown in Fig. 2b Fig. S4, where some irregular products and beads could be found before
and d). As shown in Fig. 2a, DAC]CS beads could be observed before NaOH treatment and almost no intact beads could be observed after
NaOH treatment, but the beads were disrupted after NaOH treatment NaOH treatment. These results indicate that this reaction route resulted
(Fig. 2c) and the yield was low (< 5%). Some broken beads could be in low efficiency for the cross-linking reaction between DAC and chit-
found in Fig. 2b for DAC-CS-2 samples before NaOH treatment, which osan, which might be due to the high viscosity of chitosan solution
indicated that unstable DAC]CS beads were destroyed to some extent hindering its surface contact with the DAC beads. Thus, the route of
during the reduction process due to the alkaline reductive conditions in adding chitosan solution into the DAC dispersion drop-wise was sub-
the presence of NaBH3CN. Although the yield after NaOH treatment of sequently used.
DAC-CS-2 is much higher than that of DAC]CS, intact bead mor- Based on the previous experiments and results presented, the fol-
phology could not be maintained after NaOH treatment (Fig. 2d). Thus, lowing reaction conditions and procedures were selected for the re-
the in-situ reduction process was selected as reduction method. ductive amination between DAC and chitosan: 2 equiv. of chitosan,
Further, the effect of adding chitosan solution into the DAC dis- room temperature (22.5 °C) as reaction temperature, 6 h reaction time,
persion batch-wise instead of drop-wise was studied. The SEM images drop-wise addition of chitosan solution into the DAC dispersion and in-
of the product before and after treatment with 1 M NaOH are shown in situ reduction using NaBH3CN.
Fig. 3. In the SEM images, some destroyed beads could be observed both Fardim and co-workers (Gericke et al., 2013) stated that an appro-
in Fig. 3a and b, whose morphology was not as intact as those in Fig. 1c priate drying technique and procedure are required in preparation of
and f. Interestingly, some hollow beads and bowl-like objects could be cellulose beads samples for SEM and gas adsorption analysis. In order to
observed after NaOH treatment, which might have been formed as the study the effects of different purification and drying routes on the
highly crosslinked surface of the beads tolerated the alkaline treatment properties of the products, DAC, DAC-CS and DAC-CS-NaOH samples
whereas the insufficiently crosslinked interior of the beads was dis- were purified and dried via three different routes after having been
solved. The interesting morphology of these samples is currently under washed two times with deionized water; 1) washed by centrifugation/
investigation. washing/redispersion in deionized water 6 times and dried from water

Fig. 3. SEM micrographs of DACeCS samples produced by batch-wise

addition of chitosan before (a) and after (b) 1 M NaOH (aq.) treatment.
C.-Q. Ruan et al.

Table 1 observed although porous beads could still be found in Fig. 4d and g
Pore volume and specific surface area (SSA) for DAC, DACeCS and DAC-CS-NaOH pur- depicting DACeCS and DAC-CS-NaOH samples, which indicate that the
ified with and dried from different solvents.
chitosan modified DAC beads possessed properties that could resist
Samples Pore volume (cm3/g) SSA (m2/g) hornification. DAC is prone to undergo hornification due to its ability to
form hydrogen bonds and hemiacetals. By introducing substituents such
DAC-H2O 0.003 8.93 ± 0.23 as chitosan on the cellulose fibrils hornification can be sterically hin-
DAC-EtOH 0.013 7.81 ± 0.17
dered and its effects limited. The effect of the solvent from which the
DAC-Ether 0.068 18.89 ± 0.22
DAC-CS-H2O 0.060 12.29 ± 0.11 beads are dried also has a pronounced effect and the extensive hy-
DAC-CS-EtOH 0.094 25.49 ± 0.15 drogen bonding occurring during drying of samples in water is less
DAC-CS-Ether 0.131 44.25 ± 0.11 pronounced in less polar solvents. It might as well be related to the
DAC-CS-NaOH-H2O 0.007 9.12 ± 0.35 solvent’s Hildebrand solubility parameter, which was recently reported
DAC-CS-NaOH-EtOH 0.190 44.88 ± 0.32
to correlate with the swelling of cellulose nanocrystals (Reid,
DAC-CS-NaOH-Ether 0.298 79.67 ± 0.31
Villalobos, & Cranston, 2016). It is possible that the less swollen mate-
rial is less prone to form hydrogen bonds during the drying process and
under ambient atmosphere; 2) washed by centrifugation/washing/re- therefore provides a more porous texture in the dry state.
dispersion in ethanol 6 times and dried from EtOH under ambient at- The DACeCS beads also increased in size compared with the DAC
mosphere; 3) washed by centrifugation/washing/redispersion in beads according to laser diffraction particle size measurements (Fig.
ethanol 2 times, in acetone 2 times and in diethyl ether 2 times and S5). The unmodified DAC beads had a d50 value of 32 μm while the
dried from diethyl ether under ambient atmosphere. The samples were DACeCS had a d50 of 61 μm. The increased size of the beads might be
named by adding eH2O, eEtOH and –Ether following their original beneficial as it could reduce the backpressure if the beads were used as
name (Table 1). The SEM images of the samples purified and dried via column material under flow conditions.
the different approaches are shown in Fig. 4. In Fig. 4a–c, it could be
found that the DAC beads became more porous after washing and 3.2. Surface area and porosity analysis
drying from non-polar solvent, while the trend was not as profound for
the crosslinked samples DACeCS and DAC-CS-NaOH (Fig. 4d–i). As The observation that the yield after treatment of DACeCS with 1 M
commonly known the evaporation of water from cellulose might result NaOH was 40% although the morphology of beads was unaltered led to
in hornification (Gericke et al., 2013). The solvent exchange procedure, the assumption that other properties such as pore volume, SSA and pore
where the product is transferred to increasingly less polar solvents size distribution might be affected. Hence, nitrogen gas sorption iso-
followed by drying from non-polar solvents, seems capable of main- therms at a temperature of 77 K and pore size distribution for DAC,
taining the porous morphology after drying, as can be seen in Fig. 4. For DACeCS and DAC-CS-NaOH obtained by different drying routes are
the samples described herein, DAC-H2O might have undergone the given in Figs. S6 and S7, respectively. The pore volume and SSA of
mentioned hornification and a smooth surface of the beads was samples are summarized in Table 1. For almost all samples, DAC,

Fig. 4. SEM micrographs of samples DAC (a)(d)(g), DACeCS (b)(e)(h) and DAC-CS-NaOH (c)(f)(i) washed via different routes and dried from different solvents.
C.-Q. Ruan et al.

DACeCS and DAC-CS-NaOH, the SSA increased after purification with 3.6. CHN elemental analysis
and drying from less polar solvents, which could be observed from the
nitrogen sorption isotherms in Fig. S6 and the data in Table 1. Further, The CHN elemental analysis for DACeCS samples showed that the
both the amount of micropores and mesopores increased after pur- weight percentages of C, H, O and N elements were 43.2%, 6.5%, 48.0%
ification with less polar solvents, which could be observed from the and 2.4%, respectively, which indicates that the grafted percentage of
pore size distribution in Fig. S7. On the other hand, the DACeCS beads chitosan on 2,3-dialdehyde cellulose is about 40% (the ratio of mole
became more porous compared to the unmodified DAC beads. For amount of chitosan to cellulose). CHN elemental analysis was also
DACeCS, the SSA and the pore volume also increased. Further, more performed according to a previously described procedure (Lindh et al.,
porous DACeCS beads (DAC-CS-NaOH) were obtained after the treat- 2014) on the DAC material in order to estimate the degree of oxidation
ment with 1 M NaOH, which could explain the phenomenon of the and a degree of oxidation of 98% was calculated.
weight decreasing of DACeCS after NaOH treatment.
3.7. Energy dispersive X-ray (EDX) spectroscopy

3.3. FTIR spectroscopy The results of EDX measurement are show in Fig. S10. Nitrogen
could be detected in the mapping, which indicates that the coupling
The FTIR spectra of unmodified Cladophora cellulose, DAC and reaction between DAC and chitosan occurred successfully.
DACeCS are shown in Fig. S8. The absorbance peak at 1736 cm−1
assigned to the carbonyl groups (Kim, Kuga, Wada, Okano, & Kondo, 3.8. Adsorption and desorption of Congo red
2000) could be observed in the spectrum of DAC but disappeared in
that of DACeCS. The new peak at 1736 cm−1 in the spectrum of DAC The adsorption of Congo red is an interesting topic, which has been
compared to that of Cladophora cellulose indicates that carbonyl groups extensively studied (Chatterjee, Chatterjee, Chatterjee, & Guha, 2007;
were produced after periodate oxidation, while the disappearance of Kumari, Mankotia, & Chauhan, 2016; Wong, Szeto, Cheung, & McKay,
this peak in the spectrum of DACeCS illustrates the crosslinking reac- 2003). Congo red aqueous solutions are red at pH > 5 and blue at pH 3
tion between DAC and chitosan proceeded as shown in Scheme 1b. (Purkait et al., 2007). In UV–vis spectroscopy the absorbance shift de-
pends on the pH of the Congo red solution (Fig. S1), so different stan-
dard calibration curves were produced at different pH. The percentage
3.4. Thermal gravimetric analysis (TGA) of Congo red adsorption at different pH after 3 and 20 h is shown in
Fig. 6a and the adsorption decreased when the pH increased both for 3
Thermal stability of DAC and DACeCS was studied by TGA mea- and 20 h. Particularly, the adsorption is almost 100% at pH 2 but only
surement and compared with unmodified Cladophora nanocellulose, 3% at pH 12. Fig. 6b shows the relationship between adsorption per-
and the thermograms are depicted in Fig. 5. The initiating temperature centage and contact time at pH 2 and pH 7, which indicates that the
of degradation for DAC was lower than that of Cladophora nanocellu- adsorption at pH 2 was much faster and higher than that at pH 7. The
lose, which is to be expected as the periodate oxidation disrupts the adsorption capacity of DACeCS beads for Congo red is about 200 mg/g
ordered bonding of the highly crystalline Cladophora nanocellulose at pH 2 in 100 g/L dye solution. When DACeCS beads loaded with
(Lindh et al., 2014). In addition, DACeCS has a higher starting tem- Congo red were subjected to NaOH (aq.) of pH 12 the desorption was
perature of degradation compared to DAC, which demonstrates the about 93%. The desorbed DACeCS beads were then subjected to Congo
crosslinking reaction enhanced the thermal stability of DAC and also red solution of pH 2 a second time followed by NaOH (aq.) of pH 12 and
confirms that the crosslinking reaction was efficient (Lindh et al., the adsorption and desorption were about 100% and 89%, respectively.
2016). The high percentage of adsorption and desorption indicates that the
DACeCS has the potential to be reused for Congo red adsorption for
several times. The adsorption and desorption processes shown in Fig.
3.5. X-ray photoelectron spectroscopy (XPS) S11 and the videos in the supporting information indicate that both the
adsorption and desorption were efficient and demonstrate that the
The XPS spectra in wide scans of DAC and DACeCS are presented in DACeCS beads have the potential to be used as a useful material for
Fig. S9. The characteristic peaks of both C 1s (285 eV) and O 1s Congo red adsorption.
(531 eV) can be observed in the spectra of DAC and DACeCS, while the
additional peak of N 1s (398 eV) can only be found in the spectrum of 4. Conclusions
DACeCS which indicates the existence of nitrogen in the product of
DACeCS. In this study, micrometer sized DACeCS beads were produced,
which were stable in basic aqueous solution when prepared according
to the following reaction conditions and procedures for the reductive
amination between DAC and chitosan: 2 equiv. of chitosan, room
temperature (22.5 °C) as reaction temperature, 6 h reaction time, drop-
wise addition of chitosan solution into the DAC dispersion and con-
ducting the reduction with NaBH3CN in-situ. DACeCS beads dried from
nonpolar solvents appeared more porous and had larger specific surface
area. The disappearance of the absorbance peak at 1736 cm−1 of the
FTIR spectrum proved the DAC was modified by chitosan. XPS, EDX and
elemental analysis showed presence of nitrogen in DACeCS beads,
which confirmed that the DAC modification was successful.
Additionally, the enhanced thermal stability according to TGA and
morphological stability in 1 M NaOH (aq.) according to SEM further
support efficient crosslinking. The fast and high adsorption capacity at
pH 2 and good desorption properties at pH 12 of DACeCS beads for
Fig. 5. Thermograms obtained in nitrogen atmosphere for dried Cladophora cellulose, Congo red indicated possible applications as a suitable material in
DAC and DACeCS samples.
chemical waste adsorption.
C.-Q. Ruan et al.

Fig. 6. DACeCS beads adsorption percentage for

Congo red (a) at specific pH after 3 and 20 h, (b) for
different time at pH 2 and 7. Initial concentration of
Congo red is 100 mg/L.

Acknowledgements formation of 2,3-dialdehyde cellulose beads via periodate oxidation of cellulose in

water. Biomacromolecules, 15(5), 1928–1932.
Lindh, J., Ruan, C., Strømme, M., & Mihranyan, A. (2016). Preparation of porous cellulose
Ollie and Elof Ericssońs Foundation are gratefully acknowledged for beads via introduction of diamine spacers. Langmuir, 32(22), 5600–5607.
their financial support. Chang-Qing Ruan thanks the China Scholarship Luo, X., & Zhang, L. (2013). New solvents and functional materials prepared from cel-
lulose solutions in alkali/urea aqueous system. Food Research International, 52(1),
Council (CSC) for financial support. 387–400.
Mahfoudhi, N., & Boufi, S. (2017). Nanocellulose as a novel nanostructured adsorbent for
Appendix A. Supplementary data environmental remediation: A review. Cellulose, 24(3), 1171–1197.
Monteiro, O. A., & Airoldi, C. (1999). Some studies of crosslinking chitosan–glutar-
aldehyde interaction in a homogeneous system. International Journal of Biological
Supplementary data associated with this article can be found, in the Macromolecules, 2, 119–128.
online version, at http://dx.doi.org/10.1016/j.carbpol.2017.10.072. O'neill Jr John, J. (1951). Method of producing cellulose pellets. US 2543928 A.
Potthast, A., Schiehser, S., Rosenau, T., & Kostic, M. (2009). Oxidative modifications of
cellulose in the periodate system – Reduction and beta-elimination reactions.
References Holzforschung, 63(1), 12–17.
Purkait, M. K., Maiti, A., DasGupta, S., & De, S. (2007). Removal of congo red using
Ahmad, R., & Kumar, R. (2010). Adsorptive removal of congo red dye from aqueous activated carbon and its regeneration. Journal of Hazardous Materials, 145(1–2),
solution using bael shell carbon. Applied Surface Science, 257(5), 1628–1633. 287–295.
Brunauer, S., Emmett, P. H., & Teller, E. (1938). Adsorption of gases in multimolecular Reid, M. S., Villalobos, M., & Cranston, E. D. (2016). Cellulose nanocrystal interactions
layers. Journal of the American Chemical Society, 60(2), 309–319. probed by thin film swelling to predict dispersibility. Nanoscale, 8, 12247–12257.
Chatterjee, S., Chatterjee, S., Chatterjee, B. P., & Guha, A. K. (2007). Adsorptive removal Rinaudo, M. (2006). Chitin and chitosan: Properties and applications. Progress in Polymer
of congo red, a carcinogenic textile dye by chitosan hydrobeads: Binding mechanism, Science, 31(7), 603–632.
equilibrium and kinetics. Colloids and Surfaces A: Physicochemical and Engineering Rinaudo, M. (2010). New way to crosslink chitosan in aqueous solution. European Polymer
Aspects, 299(1), 146–152. Journal, 46(7), 1537–1544.
Dingfeng, X., Chaoji, C., Jia, X., Bao, Z., Ling, M., Jie, C., et al. (2016). A hierarchical N/S- Ruan, C. Q., Strømme, M., & Lindh, J. (2016). A green and simple method for preparation
codoped carbon anode fabricated facilely from cellulose/polyaniline microspheres of an efficient palladium adsorbent based on cysteine functionalized 2,3-dialdehyde
for high-performance sodium-ion batteries. Advanced Energy Materials, 6(6) [1501929 cellulose. Cellulose, 23(4), 2627–2638.
(1501927 pp.)-1501929 (1501927 pp.)]. Shi, B., Li, G., Wang, D., Feng, C., & Tang, H. (2007). Removal of direct dyes by coagu-
Divakaran, R., & Pillai, V. S. (2002). Flocculation of river silt using chitosan. Water lation: The performance of preformed polymeric aluminum species. Journal of
Research, 36(9), 2414–2418. Hazardous Materials, 143(1), 567–574.
Gericke, M., Trygg, J., & Fardim, P. (2013). Functional cellulose beads: Preparation, Teotia, A., Ikram, S., & Gupta, B. (2012). Structural characterization of chitosan and
characterization, and applications. Chemical Reviews, 113(7), 4812–4836. oxidized carboxymethyl cellulose based freeze-dried films. Polymer Bulletin, 69(2),
Ghorai, S., Sarkar, A. K., Panda, A. B., & Pal, S. (2013). Effective removal of Congo red dye 175–188.
from aqueous solution using modified xanthan gum/silica hybrid nanocomposite as Trygg, J., Yildir, E., Kolakovic, R., Sandler, N., & Fardim, P. (2015). Solid-state properties
adsorbent. Bioresource Technology, 144, 485–491. and controlled release of ranitidine hydrochloride from tailored oxidised cellulose
Habiba, U., Siddique, T. A., Joo, T. C., Salleh, A., Ang, B. C., & Afifi, A. M. (2017). beads. Macromolecular Materials and Engineering, 300(2), 210–217.
Synthesis of chitosan/polyvinyl alcohol/zeolite composite for removal of methyl Vakili, M., Rafatullah, M., Salamatinia, B., Abdullah, A. Z., Ibrahim, M. H., Tan, K. B.,
orange: Congo red and chromium(VI) by flocculation/adsorption. Carbohydrate et al. (2014). Application of chitosan and its derivatives as adsorbents for dye re-
Polymers, 157, 1568–1576. moval from water and wastewater: A review. Carbohydrate Polymers, 113, 115–130.
Jin, L., Li, W., Xu, Q., & Sun, Q. (2015). Amino-functionalized nanocrystalline cellulose as Wan Ngah, W. S., Teong, L. C., & Hanafiah, M. A. K. M. (2011). Adsorption of dyes and
an adsorbent for anionic dyes. Cellulose, 22(4), 2443–2456. heavy metal ions by chitosan composites: A review. Carbohydrate Polymers, 83(4),
Kim, U. J., Kuga, S., Wada, M., Okano, T., & Kondo, T. (2000). Periodate oxidation of 1446–1456.
crystalline cellulose. Biomacromolecules, 1(3), 488–492. Wang, D.-M., & Sun, Y. (2007). Fabrication of superporous cellulose beads with grafted
Kim, U. J., Wada, M., & Kuga, S. (2004). Solubilization of dialdehyde cellulose by hot anion-exchange polymer chains for protein chromatography. Biochemical Engineering
water. Carbohydrate Polymers, 56(1), 7–10. Journal, 37(3), 332–337.
Kumari, S., Mankotia, D., & Chauhan, G. S. (2016). Crosslinked cellulose dialdehyde for Wong, Y. C., Szeto, Y. S., Cheung, W. H., & McKay, G. (2003). Equilibrium studies for acid
Congo red removal from its aqueous solutions. Journal of Environmental Chemical dye adsorption onto chitosan. Langmuir, 19(19), 7888–7894.
Engineering, 4(1), 1126–1136. Yunhui, X., & Yongjin, D. (2011). Crosslinking properties on oxidized cotton cellulose and
Lindh, J., Carlsson, D. O., Strømme, M., & Mihranyan, A. (2014). Convenient one-pot chitosan with low molecular weight. Advanced Materials Research, 332–334, 100–103.

View publication stats