Ulvan dialdehyde-gelatin hydrogels for removal of heavy metals and methylene blue from aqueous solution

Carbohydrate Polymers 249 (2020) 116841

Contents lists available at ScienceDirect

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

Ulvan dialdehyde-gelatin hydrogels for removal of heavy metals and
methylene blue from aqueous solution

T

Niklas Wahlströma, Sophie Steinhagenb, Gunilla Tothb, Henrik Paviab, Ulrica Edlunda,*
a
b

Fiber and Polymer Technology, KTH Royal Institute of Technology, Teknikringen 56, SE-100 44, Stockholm, Sweden
Department of Marine Sciences, Lovén Centre for Marine Sciences – Tjärnö, University of Gothenburg, SE-452 96, Strömstad, Sweden

A R T I C LE I N FO

A B S T R A C T

Keywords:
Ulvan
Ulva
Hydrogel
Heavy metal
Methylene blue
Adsorption

Hydrogels based on the polysaccharide ulvan from the green macroalgae Ulva fenestrata were synthesized and
evaluated as an adsorbent for heavy metals ions and methylene blue. Ulvan was extracted from Ulva fenestrata
using diluted hydrochloric acid and recovered by precipitation with EtOH. The extracted ulvan was converted
into ulvan dialdehyde via periodate-oxidation and subsequently combined with gelatin yielding hydrogels. The
hydrogels showed good water-uptake capacity with a maximum swelling degree of 2400 % in water and 900 %
in PBS buffer. Adsorption tests of methylene blue showed a maximum adsorption capacity of 465 mg/g. The
adsorption data of methylene blue followed the pseudo-second order kinetics and agreed with the Langmuir
adsorption isotherm. The maximum adsorption capacity of heavy metal ions was 14 mg/g for Cu2+, 7 mg/g for
Co2+and 6 mg/g for Ni2+and Zn2+ indicating that the hydrogels have a stronger affinity for Cu2+ than for
Co2+, Ni2+, and Zn2+.

1. Introduction
Water pollution is a global environmental problem worldwide,
especially in developing countries where it causes the death of a million
people every year (Ebenstein, 2012). Many heavy metals are highly
problematic water contaminants, often derived from manufacturing
facilities such as metalworking operations, mining, and textile industries. Emission of heavy metals is of great concern for both human
health and the environment, due to toxicity and bioaccumulation in
living organisms (Marcoveccio, Botte, & Freije, 2007; Phillips & Russo,
1978). Because of their toxicity to humans, several heavy metals and
metalloids are listed in the Agency for Toxic Substances and Disease
Registry (ATSDR)’s Substance Priority List (SPL) (Agency for Toxic
Substances & Disease Registry (ATSDR), 2017). Dyes are other problematic water contaminants, typically originating from the textile industries. (Mondal, 2008; Rastogi, Sahu, Meikap, & Biswas, 2008;
Yaseen & Scholz, 2019). The emitted dyes are often resistant to photodegradation, biodegradation, and oxidation, so they will persist in the
environment over time. Dyes also discolor the seawater, which impedes
the light penetration and retards the photosynthesis of water-living
organisms. Methylene blue is a common cationic dye used in the textile
industry. Methylene blue is harmful to humans at higher doses and can
cause increased heart rate, vomiting, and shock (Ahmad & Kumar,
2010). For the above-mentioned reasons, the development of efficient

⁎

strategies for the removal of water contaminants such as heavy metals
and dyes is an important field of research. Research regarding development of more efficient water purification techniques is in line with
the UN Sustainable Development Goals. which targets universal and
equitable access to safe and affordable drinking water for all. Different
strategies for removal of water contaminants are used industrially including physical adsorption, precipitation, ion-exchange resins, electrochemical methods and membrane filtration (Barakat, 2011). Among
these methods, physical adsorption is one of the most commonly used.
Activated carbon is one of the most commonly used absorbents due to
its high surface area and ability to adsorb a wide variety of contaminants from water including heavy metals and dyes (Barakat, 2011),
however activated carbon is less efficient for heavy metals. Organic
hydrogels may offer a potent alternative. Hydrogels are three-dimensional cross-linked polymer networks with a high water-uptake capacity. The water-adsorption capacity of a hydrogel depends on factors
such as porosity and the available amounts of binding sites or functional groups in the hydrogel. The presence of polar functional groups
such as −OH and −COOH in hydrogels also enables physical adsorption of heavy metal ions (Ngah & Hanafiah, 2008). Polysaccharides
have gained high interest as a key-component in biobased hydrogels
due to their hydrophilicity, high abundance, low cost, and environmental benefits. Hydrogels based on polysaccharides have been evaluated as potential adsorbents for heavy metal ions or dyes. Various

Corresponding author.
E-mail address: (U. Edlund).

/>Received 21 February 2020; Received in revised form 30 June 2020; Accepted 28 July 2020
Available online 06 August 2020
0144-8617/ © 2020 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
( />

Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

polysaccharides have been evaluated, including cellulose (Dai & Huang,

2016; O’Connell, Birkinshaw, & O’Dwyer, 2008; Zhao et al., 2019;
Zhou, Wu, Lei, & Negulescu, 2014), chitin (Bartczak et al., 2017;
Wysokowski et al., 2014), starch (Yu et al., 2018), chitosan (Huang,
Hsieh, & Chiu, 2015; Kandile & Nasr, 2009), pectin (Guilherme et al.,
2010), and hemicelluloses (Ferrari, Ranucci, Edlund, & Albertsson,
2015; Sun, Liu, Jing, & Wang, 2015). Polysaccharides from the extracellular diatom Didymosphenia geminate has also been reported as an
efficient adsorbent of heavy metals (Wysokowski et al., 2017).
However, one issue with many commercial hydrogels is that the
crosslinker is sometimes either toxic or derived from fossil-based
sources (Hu, Wang, Xiao, Zhang, & Wang, 2019). For this reason, the
development and implementation of biobased and environmentally
friendly crosslinkers is an important field of research. Gelatin is a
polypeptide derived from collagen isolated from animal tissue such as
porcine skin, and may serve as a biobased crosslinker for polysaccharide-based hydrogels. Gelatin is non-toxic, biodegradable, and
biocompatible, which makes it a promising crosslinker candidate. The
key strategy for crosslinking between gelatin and polysaccharides is a
two-step process. The polysaccharide is first oxidized with sodium
periodate leading to the formation of aldehyde groups along the polysaccharide backbone. The oxidized polysaccharide is then mixed with
gelatin in aqueous solution which leads to covalent crosslinking between the aldehyde groups in the polysaccharide and the free amine
groups in the lysine amino acid residues in gelatin. This type of reaction
is generally known as a Schiff-base reaction. This strategy has been used
in previous studies to synthesize hydrogels based on carboxymethylcellulose (Li, Ye, Li, Li, & Mu, 2016), pectin, (Gupta,
Tummalapalli, Deopura, & Alam, 2014), pullulan (Zhang et al., 2019)
and alginate (Sarker et al., 2014; Yuan et al., 2017).
Sustainable marine biomass such as macroalgae is a so far largely
unused source of biomass, which stands out as an alternative feedstock
for polysaccharides aside land-based biomass. Macroalgae offer several
benefits compared with land-based biomass. Macroalgae are a low-cost
source of biomass that grows quickly and does not require the use of
either irrigation or fertilizers, the latter being important in mitigating

eutrophication. Furthermore, the cultivation of macroalgae does not
compete with valuable land areas (Kraan, 2013). The green macroalgae
of the genus Ulva is one of the most popular edible seaweeds in the
world. It is widely distributed along the coastlines across the world.
Ulva spp. are particularly suitable for cultivation due to its ability to
thrive under many different growing conditions (Ye et al., 2011).
However, the cultivation of Ulva spp. is still quite limited. It is mainly
tank-cultivated (mainly in South Africa and Southeast Asia) rather than
cultivated in the wild. Aside from being used for food consumption,
Ulva spp. was recently identified as a potential feedstock for material
applications due to its high content of polysaccharides (Taboada,
Millán, & Míguez, 2010). Ulvan is such a polysaccharide. Ulvan has a
sulfated and quite complex structure consisting of a linear backbone
with rhamnose (Rha), xylose (Xyl), glucuronic acid (GlcA), and iduronic
acid (IdoA) as the main building blocks (Fig. 1). The main repeating
units in ulvan are β-D-GlcA (1 → 4)-α-L-Rha-3-sulfate, α-L-IdoA (1 →
4)-α-L-Rha-3-sulfate, β-D-Xyl (1 → 4)-α-L-Rha-3-sulfate, and β-D-Xyl-2sulfate (1 → 4)-α-L-Rha-3-sulfate. Smaller amounts of other sugars such
as galactose (Gal), mannose (Man), and glucose (Glc) can also occur.
The chemical composition of ulvan varies between different species and
is also dependent on the cultivation conditions and the season of harvest (Lahaye & Robic, 2007; Lahaye, 1998; Quemener, Lahaye, & Bobin
Dubigeon, 1997; Robic, Gaillard, Sassi, Lerat, & Lahaye, 2009; Robic,
Rondeau-Mouro, Sassi, Lerat, & Lahaye, 2009; Wahlström et al., 2020).
Ulvan has shown to exhibit both antioxidant (Morelli & Chiellini,
2010) and anticoagulant properties (Zhang et al., 2008). It has also
shown high potential as a building block in hydrogels (Morelli &
Chiellini, 2010; Morelli, Betti, Puppi, Bartoli et al., 2016; Morelli, Betti,
Puppi, & Chiellini, 2016; Dash et al., 2018). Since ulvan can be extracted from the macroalgae Ulva which is one of the most widely

Fig. 1. The characteristic disaccharide motifs in ulvan, a) β-D-GlcA (1 → 4)-αL-Rha-3-sulfate. b) α-L-IdoA (1 → 4)-α-L-Rha-3-sulfate. c) β-D-Xyl (1 → 4)-α-LRha-3-sulfate, and d) β-D-Xyl-2-sulfate (1 → 4)-α-L-Rha-3-sulfate.

distributed seaweeds in the world, ulvan has great potential as a valuable feedstock for the global and large-scale production of bio-based
materials. Like other seaweed-derived polysaccharides, ulvan also show
affinity to heavy metal ions (Lahaye & Robic, 2007), which we hypothesize could merit ulvan as a possible candidate as a component in
heavy metal adsorbing hydrogels. Still, there are no reported studies
regarding the use of ulvan in hydrogels for the adsorption of heavy
metals or dyes.
Our aim was to investigate the potential use of ulvan as a building
block in hydrogels for adsorption of heavy metals and dye from aqueous
solution, using methylene blue as a model dye. The hydrogels were
prepared from oxidized ulvan (ulvan dialdehyde) and gelatin. Previous
studies have shown that ulvan is accessible to periodate-oxidation (de
Carvalho et al., 2018). The devised strategy involves 1) to extract ulvan
from the green macroalgae Ulva fenestrata, Linneaus 2)to oxidize the
extracted ulvan to ulvan dialdehyde, and 3) crosslinking with gelatin in
aqueous solution. Our hypothesis is that ulvan dialdehyde will react
with gelatin in a Schiff-base reaction leading to hydrogel formation. We
believe that the high content of free carboxylic groups in the formed
hydrogels originating from the uronic acid moieties in ulvan dialdehyde
could work as binding sites for both heavy metals and cationic methylene blue. The formed hydrogels were thoroughly with respect to
water-uptake capacity, mechanical properties, and adsorption capacity
of heavy metals and methylene blue. Correlations of properties and the
gelatin:ulvan dialdehyde ratio were investigated.
2. Experimental
2.1. Chemicals and materials
Hydrochloric acid (37 %, CAS nr: 7647.01-0), gelatin from porcine
skin (Bloom-number 170–190, Mw = 40−50 kDa, pl = 7–9, CASnumber: 9000-70-8), methylene blue (CAS-number: 122965-43-9)
2


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

Table 1
Composition and appearance of ulvan dialdehyde-gelatin hydrogels.

Amount of gelatin (g)
Amount of ulvan dialdehyde (g)
Mass ratio gelatin: ulvan dialdehyde

0.200
0.050
80:20

0.200
0.085
70:30

0.200
0.133
60:40

0.200
0.200
50:50

0.133
0.200
40:60

500 mL round bottom flask, 2.4 g ulvan and 500 mL of deionized water
was added. The solution was stirred until complete dissolution of the
ulvan. To the solution was added 5.33 g of sodium periodate and the
solution was allowed to stir at 20 °C under dark conditions for 48 h. The
reaction was quenched by the addition of 10 mL of ethylene glycol. The
solution was dialyzed against deionized water for 48 h and freeze-dried.
The product was recovered as a fluffy white powder with a mass of 1.60
g. The sample was stored in a closed container at 20 °C until further use.

sodium (meta)periodate (99 %, CAS- number: 7790-28-5), ethylene
glycol (99 %, CAS-number: 107-21-1), cupper nitrate trihydrate (98 %,
CAS-number: 10031-43-3), cobalt nitrate hexahydrate (98 %, CASnumber: 10026-22-9), nickel nitrate hexahydrate (99 %, CAS-number:
13478-00-7) zinc nitrate hexahydrate (99 %, CAS-number: 10196-18-6)
deuterium oxide (99 atom% D, CAS-number: 7789-20-0), 2,4 dinitrophenylhydrazine (98 %, CAS-number: 119-26-6) and PBS buffer tablets (MDL-number: MFCD00131855) were obtained from Sigma
Aldrich. Ethanol (99 %, CAS nr: 64-17-5) was purchased from VWR
Chemicals. All chemicals where used as received.
Ulva fenestrata, previously known as Ulva lactuca (Hughey et al.,
2019), was collected from Inre Vattenholmen (58°52′37.4″N
11°6′52.1″E) and brought back to the lab within 2 h of collection.
Collected Ulva fenestrata individuals were rinsed several times in natural seawater to remove grazers and loose epiphytes. The seaweeds
were placed into cultivation tanks (90 L) under a neutral light cycle (16
h daylight, 8 h darkness) at a light intensity of 140 μE m−2 s-1. The light
source was an INDY66 LED 60 W 4000 K 6000 lm. The seaweeds
continuously received filtered seawater that was passed through 1 μm
filters. No additional medium or chemicals were added to the water.
The natural seawater used in the flow-through system was pumped in
from the bay outside the Tjärnö Marine Laboratory (58°52′36.4″N
11°6′42.84″E). Thus, the salinity and temperature fluctuated depending
on the prevailing weather and seasonal conditions. During the cultivation period (March to May 2019), the salinity ranged from 11 to 25
PSU and the temperature increased from 4 to 19 °C. The biomass used

during our experiments was molecularly identified by DNA barcoding.
Sequences of the tufA gene unequivocally identified the algal tissue as
Ulva fenestrata. Molecular identification followed the protocol described
by Steinhagen, Karez, and Weinberger (2019).

2.2.3. Preparation of ulvan dialdehyde-gelatin hydrogels
The hydrogels were prepared in 10 mL glass vials. Five different
hydrogels were prepared using different mass ratios of ulvan dialdehyde and gelatin to investigate how the relative ratio between ulvan
dialdehyde and gelatin affects the hydrogel properties. Three replicates
of each hydrogel were prepared. The compositions of the hydrogels are
given in Table 1. For example, the sample denoted GU-80−20 was
prepared by dissolving 0.050 g ulvan dialdehyde in 2 mL 0.01 M PBS
buffer (pH = 7.4) at 80 °C. In a separate container, 0.20 g gelatin was
dissolved in 2 mL 0.01 M PBS buffer (pH = 7.4) at 60 °C. Both solutions
were cooled to 20 °C before mixing. The ulvan dialdehyde solution was
added dropwise to the gelatin solution under vigorous stirring. The
glass vials were sealed and placed on a shaking board at 20 °C overnight
to ensure complete gelation of the samples. The formed hydrogels were
removed from the glass vials by breaking the vials. The gels were allowed to dry in air at 20 °C for 48 h in a fume-hood.
2.3. Characterization techniques
2.3.1. Attenuated total reflectance Fourier transform infrared spectroscopy
(ATR-FTIR)
FTIR spectra of ulvan, ulvan dialdehyde, and the dried hydrogels
were recorded between 4000-600 cm−1 at 20 °C with a spectral resolution of 4 cm−1 using a Perkin-Elmer Spectrum 100 FTIR with a
triglycine detector and equipped with an attenuated total reflectance
crystal accessory (ATR Golden Gate) from Graseby Specac LTD (Kent,
England). Corrections were made for atmospheric water and carbon
dioxide. The obtained spectra were calculated as an average of 64
scans. The data were analyzed using PerkinElmer Spectrum software.

2.2. Synthesis of ulvan dialdehyde-gelatin hydrogels
2.2.1. Extraction and purification of ulvan from Ulva fenestrata
Ulvan was extracted from Ulva fenestrata according to a protocol
from our previous study (Wahlström et al., 2020). Briefly, freeze-dried
Ulva fenestrata (50 g) was grinded to a fine powder using a Bosch
MKM6003 coffee miller (Clas Ohlson, Sweden) and suspended in 1 L of
0.01 M HCl (pH = 2). The mixture was heated to 90 °C for 4 h under
continuous stirring. The solution was cooled to 20 °C and centrifuged at
5000 rpm for 5 min. The supernatant was collected and freeze-dried.
The freeze-dried sample was re-dissolved in 200 mL of deionized water
and dialyzed against deionized water for 48 h. After dialysis, the ulvan
was precipitated by addition of 800 mL absolute EtOH. The mixture was
centrifuged at 5000 rpm for 5 min and the supernatant was discarded.
The precipitated ulvan was washed with absolute EtOH and air-dried
overnight. The product was recovered as a white fluffy powder. Yield:
9.0 g (18 % (w/w) of the starting biomass). The sample was stored in an
airtight container at 20 °C until further use.

2.3.2. Nuclear magnetic resonance spectroscopy (13C-NMR)
The ulvan and the ulvan dialdehyde fractions were analyzed by 13CNMR. Freeze-dried samples (80 mg) were dissolved in 1.0 mL of D2O,
and NMR spectra were recorded at 20 °C on a Bruker DMX-500 NMR
spectrometer at 500 MHz. Spectra were calculated as an average of
16384 scans and the data were analyzed using MestReNova software.
2.3.3. Thermal gravimetric analysis (TGA)
The thermal stability of the hydrogels was estimated using a Mettler
Toledo TGA/DSC. Approximately 4 mg of each sample was heated in
alumina cups from 40 °C to 800 °C at a heating rate of 10 °C/min under
an N2 atmosphere using a flow rate of 50 mL/min. The residual water
content in the dried samples were calculated as the mass loss in the
temperature interval 40−110 °C. The data were processed and

2.2.2. Synthesis of periodate-oxidized ulvan (ulvan dialdehyde)
The synthesis of ulvan dialdehyde was performed as described by de
Carvalho et al. (2018), with the below described modifications. To a
3


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

glucuronic acid and iduronic acid was used as the reference. The data
were processed and analyzed using Chromelon 7.1

analyzed using STARe software. From the TGA data, the residual water
content in the dried hydrogels was estimated by measuring the weightloss of the samples between 40−110 °C.

2.3.8. Rheological behavior of ulvan dialdehyde-gelatin hydrogels
Hydrogels were prepared as cylindrical discs with a thickness of 0.2
cm and a diameter of 0.8 cm. The storage modulus (G’) and the loss
modulus (G’’) of the hydrogels were measured by a TA Discovery
Hybrid 2 (DHR-2) rheometer equipped with a 0,8 cm stainless steel
Peltier plate at 20 °C. The applied strain was set to 0.3 % and the oscillating frequency range was from 0 to 100 rad/s. A solvent trap
containing deionized water was used during all the measurements to
prevents drying of the gels. The loss tangent tan(δ) of the hydrogels was
calculated with Eq. (1). tan(δ) expresses the ratio of the energy loss to
the energy stored which provides information about the overall viscoelasticity of the hydrogels.

2.3.4. Field-emission scanning electron microscopy (FE-SEM)
The morphology of the hydrogels was observed by ultra-high-resolution field emission scanning electron microscopy (FE-SEM) using a

Hitachi S-4800 operating at 5 kV. The hydrogels were swollen in
deionized water for 48 h after which they were frozen in liquid nitrogen
and freeze-dried overnight. The dried samples were then attached to the
sample supports using double-sided adhesive carbon tape and sputtercoated with a 7 nm Pt/Pd layer using a Cressington 208HR under an
inert atmosphere.
2.3.5. Estimation of aldehyde group content in ulvan dialdehyde
The content of aldehyde groups in ulvan dialdehyde was determined
colorimetrically by using 2, 4 dinitrophenylhydrazine (DNPH) as described in a previously reported protocol (Tummalapalli & Gupta,
2015) with some modifications. First, 2,4 dinitrophenylhydrazine (20
mg) was dissolved in a mixture of 0.3 mL 98 % H2SO4, 3.0 mL EtOH and
6.7 mL H2O. To this solution was added 50 mg ulvan dialdehyde and
the solution was allowed to react for 1 h at room temperature. The
solution was filtered through 0.45 μm PTFE filters. The absorbance of
the solution before and after reaction with ulvan dialdehyde was
measured at 370 nm with UV–vis spectroscopy. UV–vis spectra were
recorded on a UV–vis spectrophotometer (UV-2410). The data was
processed using UVProbe. The content of aldehyde group was calculated from the decrease in absorbance of the solution after reaction with
ulvan dialdehyde.

G′′
G′

tan(δ ) =

(1)

2.3.9. Swelling behavior of ulvan dialdehyde-gelatin hydrogels
The swelling behavior of the hydrogels was tested separately in
deionized water and 0.01 M PBS buffer. Dried hydrogels (0.1−0.3 g)
were immersed in 100 mL distilled water or 100 mL PBS buffer and

allowed to swell for 48 h at 20 °C. The swollen hydrogels were weighed
after wiping off excess water with a paper towel. The degree of swelling
(SD) was calculated using Eq. (2).

SD (%) =

me − m 0
× 100
m0

(2)

where me is the mass of the swollen hydrogel and m0 is the initial mass
of the dry hydrogel before swelling.

2.3.6. Size-exclusion chromatography (SEC)
The molecular weights of the ulvan before and after oxidation was
measured by SEC as described in our previous study (Wahlström et al.,
2020). Dried samples of ulvan and ulvan dialdehyde (20 mg) were
completely dissolved in 10 mL of 10 mM NaOH. Before injection, the
samples were filtered through 0.20 μm PTFE filters. The HPLC system
was equipped with a WPS-3000SL autosampler, an LPG-3400SD gradient pump, three PSS Suprema columns with pore sizes of 30 Å, 1000
Å and 1000 Å in series (300 × 8 mm, 10 μm particle size) together with
a guard column (50 × 8 mm, 10 μm particle size) and a Waters-410
Refractive Index Detector (Wafers, Milford, CA, USA). The mobile phase
was 10 mM NaOH and the temperature was kept at 40 °C during the
run. Pullulan samples with molecular weights averages (Mw) ranging
from 342 g/mol to 708,000 g/mol were used as reference samples.

2.3.10. Heavy metal adsorption of ulvan dialdehyde-gelatin hydrogels

The heavy metal adsorption test was made for Cu2+, Co2+, Ni2+and
2+
Zn . Specimens of dry hydrogel (10 mg each) were placed in 5 mL of
heavy metal solution containing 30 mg/L (30 ppm) of each heavy
metal. The pH of the solution was 6.5. Adsorption tests were also done
with solutions containing 150 mg/L (150 ppm) of each heavy metal,
and for a solution containing 30 mg/L (30 ppm) of each heavy metal
where the pH value was adjusted to 3. The samples were left on a
shaking board at 20 °C for 24 h after which the hydrogels where removed from the solution. The heavy metal mass concentration in each
solution, before and after adsorption, was measured by induced coupled
plasma optical emission spectroscopy (ICP-OES, iCAP 6500 Thermo
Scientific). All heavy metal solutions were diluted 10 times prior to the
measurements. The percent of adsorption was calculated by using Eq.
(3).

2.3.7. High-performance anion exchange chromatography with pulsed
amperiometric detection (HPAEC-PAD)
The composition of monosaccharides and uronic acids of the extracted ulvan were measured by HPAEC-PAD as described in our previous study (Wahlström et al., 2020). Ulvan was hydrolyzed using a
modified method described in a previous study (De Ruiter, Schols,
Voragen, & Rombouts, 1992). A dried ulvan sample (1.0 mg) was
placed in an oven-dried Pyrex tube and 1 mL of 2 M HCl solution in
methanol (dried with Na2SO4) was added to the tube. The tube was
sealed and heated to 100 °C for 5 h in a heating block. The solution was
neutralized by adding 200 μL of pyridine to the reaction tube, cooled to
20 °C and the solvent was evaporated under flowing N2 gas. Hydrolysis
was performed by adding 1 mL of 2 M trifluoroacetic acid (TFA) to the
dried sample, and the solution was heated to 120 °C for 1 h in a heating
block followed by cooling to room temperature.
The composition of monosaccharides and uronic acids was determined using a HPAEC-PAD (HPAEC-PAD, ICS-3000 Dionex). The
column was a CarboPac PA1 (4 × 250 mm) column. The eluent was

pumped at 1.5 mL/min using a program starting with 0.10 M NaOH and
increasing to 0.16 M NaOH and 0,19 M NaAc during the run. A mixture
of arabinose, rhamnose, galactose, glucose, xylose, mannose,

A (%) =

c 0 − ce
× 100
c0

(3)

where ce is the mass concentration (given in mg/L) of heavy metals
remaining in the solution after adsorption and c0 is the initial heavy
metal mass concentration (given in mg/L). The adsorption capacities
(qe) of the hydrogels (given as the amount in mg of adsorbed heavy
metal per gram of hydrogel) were calculated using Eq. (4).

qe =

(c0 − ce )⋅V
m

(4)

where ce is the mass concentration (in mg/L) of heavy metals remaining
in the solution after adsorption, c0 is the initial heavy metal mass
concentration (in mg/L), V is the volume of the solution, and m is the
mass of the hydrogel.
2.3.11. Methylene blue adsorption of ulvan dialdehyde-gelatin hydrogels

The adsorption of methylene blue was tested using an aqueous solution containing either 30, 300, or 1500 mg/L of methylene blue.
4


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

mechanical and thermal properties of the hydrogels were investigated.

Adsorption tests at higher methylene blue mass concentrations (2000
mg/L and 3000 mg/L) were performed with the sample named GU-4060 (Table 1). Dried hydrogels (10 mg each) were placed in 5 mL methylene blue solution and the solutions were placed on a shaking board
at 20 °C for 24 h, after which the hydrogels where removed from the
solution. The mass concentration of methylene blue in each solution,
before and after adsorption, was measured by UV–vis spectroscopy by
measuring the absorbance at 668 nm. UV–vis spectra were recorded on
a UV–vis spectrophotometer (UV-2410). The data was processed using
UVProbe software. The percent of adsorption and adsorption capacities
where calculated using Eqs. (3) and (4).

3.1. Characterization of extracted ulvan and oxidized ulvan (ulvan
dialdehyde)
Ulvan was extracted from the green macroalgae Ulva fenestrata
using diluted hydrochloric acid and the supernatant was dialyzed to
remove salts and other low-molecular weight compounds. The ulvan
was recovered from the supernatant by precipitation with EtOH. No
further purification steps of the ulvan were performed. This method was
used in our previous study, yielding a ulvan fraction containing 76.8 %
(w/w) carbohydrates. (Wahlström et al., 2020). The ulvan was recovered as a white solid and the overall yield was 18 % (w/w) from the
starting biomass. The chemical structure was confirmed with FTIR and

13
C-NMR. Monosaccharide and uronic acid analysis with HPAEC-PAD
showed the presence of rhamnose, glucuronic acid, iduronic acid and
xylose which are all typical building-blocks in ulvan. Glucose and trace
amounts of mannose was also detected. The presence of glucose may be
due to co-extracted starch (Lahaye & Robic, 2007). The monosaccharide
and uronic acid composition of the extracted ulvan is shown in Supporting Information, Table S1.
The extracted ulvan was converted to ulvan dialdehyde by oxidation
with sodium periodate. The reaction scheme is shown in Supporting
Information, Fig. S1. This reaction introduces aldehyde groups along
the backbone of ulvan. More specifically, the reaction takes place at the
uronic acid moieties in ulvan by conversion of the vicinal diols into
aldehyde groups. The reaction took place under dark conditions to
prevent the decomposition of the periodate solution upon exposure to
light. One side-effect of periodate oxidation is that the reaction also
leads to depolymerization of the polysaccharide due to cleavage of the
glycosidic bonds. To evaluate the change in molecular weight before
and after periodate oxidation, the molecular weight of ulvan was
measured before and after oxidation. The extracted ulvan before oxidation showed a high molecular weight (Mw = 761 500 g/mol).
However, the molecular weight of the oxidized ulvan (ulvan dialdehyde) was significantly lower (MW = 10 400 g/mol) indicating that the
ulvan was depolymerized to a large extent during the oxidation. This is
in line with a previous study which showed that ulvan undergoes a
significantly reduction in molecular weight upon periodate-oxidation
(de Carvalho et al., 2018). The total content of aldehyde groups in
ulvan dialdehyde was estimated colorimetrically by 2,4 dinitrophenylhydrazine according to a previously reported protocol
(Tummalapalli & Gupta, 2015). The aldehyde content was estimated to
1.58*10−3 mol of aldehyde groups per gram of sample.
Ulvan and ulvan dialdehyde were analyzed by FTIR and 13C-NMR
spectroscopy to confirm if the periodate-oxidation was successful. The
FTIR spectra show many characteristic peaks for polysaccharides. The

broad peak at 3400-3300 cm−1 corresponds to OHe stretching and the
peak at 2900 cm−1 corresponds to C–H stretching. The peaks at 1620
cm−1 and 1420 cm−1 correspond to symmetric and asymmetric
stretching of carbonyl groups in the uronic acid moieties. The strong
peaks around 1070−1030 cm−1 corresponds to C-O-C stretching. The
peaks at 1215 cm−1 and 845 cm−1 corresponds to S=O stretching and
C-O-S stretching of sulfate groups (Robic, Bertrand, Sassi, Lerat, &
Lahaye, 2009). By comparing spectra of ulvan and ulvan dialdehyde it
is clear that a new peak appears at 1725 cm−1 in the spectrum of ulvan
dialdehyde (curve b, Fig. 2, left) which corresponds to C]O stretching
of carbonyl groups in aldehyde, confirming that the oxidation reaction
was successful. However, the ulvan dialdehyde sample still has a band
at 3300 cm−1, which indicates that there are unreacted OH groups left.
This can be explained by the fact that the different building blocks in
ulvan show different reactivity towards sodium periodate. A previous
study showed that the periodate oxidation of ulvan mainly take place at
the uronic acid moieties, while the non-sulfated rhamnose and xylose
residues are more resistant towards periodate oxidation (de Carvalho
et al., 2018) which leads to unreacted OH groups.

2.3.12. Kinetic studies and adsorption isotherms for methylene blue
adsorption
Adsorption kinetic studies were performed with the sample named
GU-40-60 (Table 1). Each sample specimen (45 mg each) was placed in
90 mL of 30 mg/L methylene blue solution. The methylene blue concentration was measured after 30 min, 1 h, 2 h, 4 h, 6 h, and 24 h. The
pseudo-first order model (Eq. (5)) and the pseudo-second order model
(Eq. (6)) were used to describe the adsorption (Zhou et al., 2017).

ln(qe − qt ) = ln(qe ) − k1 t

(5)

t
t
1
=
+
qt
qe
k2 qe2

(6)

where t is the time, k1 is the pseudo-first order rate constant, and k2 is
the pseudo-second order rate constant. qt is the adsorption capacity (in
mg/g) at a specific time and qe is the adsorption capacity after 24 h. k1
was estimated by plotting ln(qe-qt) as a function of t and the slope of the
line was calculated. k2 was estimated by plotting t/qt as a function of t
and estimate the slope and intercept of the line.
The adsorption isotherm of the methylene blue adsorption was estimated with the sample GU-40-60 by placing 10 mg dry hydrogel
specimens in separate containers containing 10 mL methylene blue
solution with starting mass concentrations of 300, 1500, 2000, and
3000 mg/L, respectively. After 24 h, the hydrogels were removed from
the solution and the adsorption capacities (qe) were calculated with Eq.
(4). Two adsorption isotherms were evaluated, the Langmuir isotherm,
Eq. (7) and the Freundlich isotherm, Eq. (8).

ce
c
1

= e +
qe
qm
KL qm
ln(qe ) = ln(KF ) +

(7)

1
ln(ce )
n

(8)

where qe is the adsorption capacity after 24 h, ce is the mass concentration (in mg/L) of methylene blue remaining in the solution after
24 h, qm is the theoretical maximal adsorption capacity. KL and KF are
the Langmuir and Freundlich constants, respectively, and n is a constant which is related to the heterogeneity of the hydrogel. KL was estimated by plotting ce/qe as a function of ce and estimating the slope
and intercept of the line. KF was estimated by plotting ln(qe) as a
function of ln(ce) and estimate the intercept of the line (Zhou et al.,
2017).
3. Results and discussion
We developed a method for the fabrication of hydrogels based on
gelatin and the polysaccharide ulvan extracted from the green macroalgae Ulva fenestrata collected along the Swedish west coast. The hydrogel preparation was a three-step process where the ulvan was first
extracted from Ulva fenestrata using diluted hydrochloric acid(aq). The
isolated ulvan was further oxidized to ulvan dialdehyde with sodium
periodate. After purification and drying, the ulvan dialdehyde was
crosslinked with gelatin in PBS buffer, producing hydrogels. The adsorption capacity of heavy metals and methylene blue as well as the
5

Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

Fig. 2. FTIR spectra (left) and

13

C-NMR spectra (right) of ulvan, a) before periodate oxidation, and b) after periodate oxidation (ulvan dialdehyde).

The 13C-NMR spectrum of ulvan (curve a, Fig. 2, right) shows many
of the characteristic signals for ulvan. A table of assigned peaks are
given in the Supporting Information, Table S4. The strong signal at 17
ppm corresponds to the carbon in the methyl group in rhamnose moieties. The anomeric region (100−105 ppm) shows the characteristic
signals for the anomeric carbons in rhamnose, glucuronic acid and
iduronic acid. The weak signal around 175 ppm corresponds to carbons
in the carboxyl groups located at the uronic acid moieties (de Carvalho
et al., 2018). By comparing the spectrum of ulvan and ulvan dialdehyde, some significant differences can be observed. No signals for aldehyde groups in the range 170−200 ppm were observed in the
spectrum for ulvan dialdehyde. However, new signals in the range
88−91 ppm appeared in the 13C-NMR spectrum of ulvan dialdehyde
which indicates the presence of hemiacetals or gem-diol groups (de
Carvalho et al., 2018). Previous studies have shown that periodateoxidized polysaccharides such as ulvan dialdehyde tend to form
hemiacetals or gem-diols (hydrated aldehydes) in aqueous solution
(Amer et al., 2016; de Carvalho et al., 2018; Fan, Lewis, & Tapley, 2001;
Jejurikar et al., 2012) due to the high reactivity of the aldehyde groups
formed during periodate oxidation. The main reason why the aldehyde
peak was observed in the FTIR spectrum and not in the 13C-NMR
spectrum is probably because the aldehyde peak is only visible if the
sample is free from water (Amer et al., 2016; Fan et al., 2001; Jejurikar
et al., 2012). The FTIR spectra was recorded on freeze-dried samples

which were substantially free from water and therefore, the aldehyde
peak is visible. On the other hand, the 13C-NMR spectra were recorded
with an aqueous solution of ulvan dialdehyde which may have led to
formation of hemiacetals and gem-diols resulting in the appearance of
the peaks at 88−91 ppm instead of aldehyde peaks around 170−200
ppm. The overall conclusion from the FTIR and 13C-NMR analysis is
that the aldehydes groups were formed during the periodate oxidation
of ulvan dialdehyde, but the aldehyde groups are in equilibrium with
the corresponding hemiacetals and gem-diols in aqueous solution.

react with the primary amine groups (-NH2) in gelatin originating from
the lysine amino acid residues. The reaction leads to the formation of
covalent C]N bonds between ulvan dialdehyde and gelatin which is
known as imine bonds or Schiff-bases. Since the crosslinking reaction
took place in PBS buffer at pH = 7.4, and the pKa value of the amine
groups in lysine is approximately 10.5, the amine groups will be partially protonated (-NH3+) under these reaction conditions, so another
possible crosslinking mechanism is physical crosslinking between the
NH3+ groups in gelatin and the COO− groups in ulvan dialdehyde.
Physical crosslinking between the NH3+ groups in gelatin and the SO3−
groups in ulvan dialdehyde is also possible. Finally, gelatin itself can
form a gel in aqueous solution by physical entanglement of the of the
gelatin molecules. The hydrogel network formed in the reaction between ulvan dialdehyde and gelatin could therefore possibly contain
two types of crosslinks and also regions consisting of entangled gelatin
chains.
FTIR spectroscopy was used to investigate if a Schiff-base reaction
occurred during the reaction between ulvan dialdehyde and gelatin.
The FTIR spectra of the hydrogels show characteristic peaks for both
ulvan dialdehyde and gelatin (Fig. 3). The broad peak at 3400-3300
cm−1 corresponds to OHe stretching and the peak at 2900 cm−1 corresponds to C–H stretching. The peaks at 1620 cm−1 and 1420 cm−1
corresponds to symmetric and asymmetric stretching of carbonyl

groups in the uronic acid moieties of ulvan dialdehyde. The bands at
1215 cm−1 and 845 cm−1 correspond to S=O stretching and C-O-S

3.2. Characterization of the chemical structure of ulvan dialdehyde-gelatin
hydrogels
Hydrogels were synthesized by mixing ulvan dialdehyde and gelatin
in PBS buffer at pH = 7.4. Gelation of the samples occurred after ∼ 1 h,
but the samples were left for 16 h to ensure complete crosslinking. All
the hydrogels appeared as soft and pliable materials that could be easily
removed from the reaction containers and handled without damaging
the gels (Table 1). The hydrogels became weaker in the water-swollen
state but did still not break when handled even after 48 h of swelling.
The suggested crosslinking reaction is shown in Supporting
Information, Fig. S2. The aldehyde groups (−CHO) in ulvan dialdehyde

Fig. 3. FTIR spectra of hydrogels, a) pure gelatin, b) GU-80-20, c) GU-70-30, d)
GU-60-40, e) GU-50-50, f) GU-40-60, g) pure ulvan dialdehyde.
6


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

of gelatin shows only one degradation peak at 320 °C corresponding to
cleavage of peptide bonds and degradation of gelatin (Sarker et al.,
2014).
TGA was also used to measure the thermal stability of the hydrogels.
The TGA curves and corresponding DTG curves are shown in
Supporting Information, Figs. S5–S6. The DTG curves of the hydrogels

show four main degradation peaks. The peak around 100 °C corresponding to evaporation of bounded water, the peak at 200 °C corresponding to depolymerization and degradation of ulvan dialdehyde, the
peak at 320 °C corresponding to cleavage of peptide bonds and degradation of gelatin, and the carbonization peak at 700 °C. Mass loss
below 100 °C is due to evaporation of bound water and not due to
degradation of the hydrogel. Therefore, the hydrogels are thermally
stable up to around 200 °C where the ulvan dialdehyde chains starts to
degrade. From the TGA curves, the residual water-content of the dried
hydrogels were estimated to 5% (w/w). The values were calculated by
calculating the mass loss of the hydrogels in the temperature interval
40−110 °C.

stretching of sulfate groups in ulvan dialdehyde (Robic, Bertrand et al.,
2009). The peaks at 3100 cm−1 and 1540 cm−1 correspond to N-H
stretching and bending in amine groups in gelatin, respectively (Sarker
et al., 2014). No clear peak for C]N stretching in Schiff-bases around
1650 cm-1 was observed, probably due to overlap with the carbonyl
peak at 1620 cm−1. Furthermore, no peak for aldehyde groups around
1725 cm−1 was observed for the samples GU-80-20 and GU-70-30 indicating that the aldehyde groups in ulvan dialdehyde was consumed
during the crosslinking reaction which suggest that a Schiff-base reaction did occur between ulvan dialdehyde and gelatin. However, a
shoulder peak around 1725 cm-1 was observed for the samples GU-6040, GU-50-50 and GU-40-60 probably due to unreacted aldehyde
groups in ulvan dialdehyde. This is an indication that the consumption
of aldehyde groups in ulvan dialdehyde is incomplete at higher ulvan
dialdehyde concentrations.
3.3. Rheological behavior of ulvan dialdehyde-gelatin hydrogels
The mechanical stability of the hydrogels during stress was evaluated by performing a frequency-sweep experiment. All samples show
typical behavior for crosslinked networks indicated by the storage
modulus (G’) always being higher than the loss modulus (G’’) and the
value of tan(δ) is below 1 during the whole frequency-sweep (Fig. 4).
This is an indication that the elastic response from the hydrogels
dominates over the viscous response, which is characteristic for hydrogels. Fig. 4 also shows that the storage modulus increases with increasing angular frequency indicating that the elastic response from the
hydrogels becomes more and more dominant over the viscous response

with increasing frequency.
The storage modulus is strongly dependent on the hydrogel composition. By comparing the storage modulus for the samples, GU-80-20,
GU-70-30, GU-60-40, and GU-50-50 (Fig. 4) with their compositions
(Table 1), it can be observed that the storage modulus increases as the
mass percentage of ulvan dialdehyde increases. One possible explanation is that when the content of ulvan dialdehyde increases, the content
of aldehyde groups increases, which increases the number of chemical
crosslinks between ulvan dialdehyde chain leading to a higher storage
modulus. However, the sample with the highest mass percentage of
ulvan dialdehyde, GU-40-60 had a lower storage modulus than the
samples GU-60-40 and GU-50-50, probably due to that the fact that the
sample GU-40-60 contained a lower amount of gelatin (Table 1) which
give rise to fewer crosslinks and a lower storage modulus. Overall, the
data from the rheology experiments suggests that the crosslinking reaction between ulvan dialdehyde and gelatin was successful.

3.5. Swelling-behavior of ulvan dialdehyde-gelatin hydrogels
The water-uptake capacity of the hydrogels was investigated by
swelling in deionized water for 48 h (Fig. 6). The sample GU-80-20 had
the lowest swelling capacity (SD = 1000 %) and the sample GU-40-60
had the highest (SD = 2400 %). By comparing the hydrogel compositions in Table 1 with the corresponding SD values, it is clear that SD
increases as the mass percentage of ulvan dialdehyde in the hydrogel
increases. One possible explanation is that by increasing the amount of
ulvan dialdehyde in the hydrogel, we also increase the number of
available binding sites for water molecules, which leads to a higher
driving force for water uptake and therefore a higher swelling. The
swelling behavior was also investigated in PBS buffer by swelling the
hydrogels in 0.01 PBS buffer for 48 h. Clearly, the hydrogels had lower
SD in PBS buffer than in deionized water, which was expected given the
higher ionic strength of the PBS solution compared with deionized
water. GU-80-20 had the lowest swelling capacity (SD = 300 %) and
GU-40-60 had the highest (SD = 900 %). The same trend was observed

for swelling in deionized water. The observed SD in this study is within
the range that was reported in previous studies on swelling of ulvanbased hydrogels in PBS buffer (Morelli, Betti, Puppi, Bartoli et al., 2016;
Morelli, Betti, Puppi, Chiellini et al., 2016; Morelli & Chiellini, 2010;
Dash et al., 2018). Likewise, the reported values of SD are within the
range for what has been reported for oxidized polysaccharides crosslinked with gelatin (Sarker et al., 2014; Yuan et al., 2017).

3.4. Morphology and thermal behavior of ulvan dialdehyde-gelatin
hydrogels

3.6. Adsorption properties of ulvan dialdehyde-gelatin hydrogels
The ulvan dialdehyde-gelatin hydrogels were tested as a potential
absorbent for the dye methylene blue. Dried hydrogels were immersed
in 3 individual methylene blue solutions with starting mass concentrations of 30, 300, and 1500 mg/L, respectively. Adsorption tests
with higher methylene blue mass concentration (2000 mg/L and 3000
mg/L) were performed with the sample named GU-40-60. The hydrogels were removed from the solution after 24 h and the remaining
methylene blue mass concentration was quantified with UV–vis spectroscopy. The percent of adsorptions and the adsorption capacities were
calculated using Eqs. (3) and (4) and are reported in Fig. 7.
GU-80-20 had the lowest adsorption capacity and GU-40-60 had the
highest (Fig. 7). By comparing the gel compositions in Table 1 with the
corresponding adsorption capacities, it is clear that the adsorption capacity increases as the mass percentage of ulvan dialdehyde in the
hydrogel increases. One possible explanation is that by increasing the
amount of ulvan dialdehyde in the hydrogel, we also increase the
number of available COO− and SO3− groups, which could act as
binding sites for the cationic methylene blue. More available binding
sites in the hydrogel will result in higher adsorption. It is also evident

The morphology of the hydrogels was studied by SEM. The hydrogels were swollen in deionized water for 48 h after which they were
frozen in liquid nitrogen and immediately freeze-dried overnight to
preserve the morphology. All samples exhibit a similar morphology
consisting of many small pores separated by thin walls (Fig. 5). This

porous structure is believed to be convenient for penetration of water
and small molecule into the hydrogel network leading to efficient water
uptake and adsorption of dye and heavy metals.
The thermal stability of extracted ulvan, ulvan dialdehyde and gelatin was measured by TGA. The TGA curves and corresponding DTG
curves are shown in Supporting Information, Figs. S3–S4. The TGA
curves for ulvan and ulvan dialdehyde look similar, indicating that the
oxidation of ulvan did not have any impact on the thermal stability. The
DTG curves of ulvan and ulvan dialdehyde show three degradation
peaks, one small peak around 100 °C corresponding to evaporation of
bound water, one major peak at 200 °C corresponding to depolymerization and degradation of ulvan, and one small peak at 700 °C, probably due to carbonization of the degradation products. The DTG curve
7


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

Fig. 4. Rheological behavior of hydrogels showing the storage modulus (G’), loss modulus (G’’) and tan(δ) as a function of angular frequency, a) GU-80-20, b) GU-7030, c) GU-60-40, d) GU-50-50, e) GU-40-60.

which will strongly influence the overall adsorption capacity. Hence,
the adsorption capacities of hydrogels prepared from different polysaccharides cannot be directly compared.
The adsorption kinetics of methylene blue in ulvan dialdehyde hydrogels was investigated by measuring the adsorption capacity of the
sample GU-40-60 as a function of time. The hydrogel was placed in 30
mg/L methylene blue solution and the adsorption capacity was measured after 30 min, 1 h, 2 h, 4 h, 6 h, and 24 h. As shown in Fig. 8a, the
adsorption capacity increases quickly during the first 4 h. After 4 h, the
absorption capacity increases much more slowly. Between 6 h and 24 h,
the adsorption capacity of the hydrogel is almost constant. The pseudofirst order and the pseudo-second order kinetic model were used to
investigate the kinetics of the methylene blue adsorption (Fig. 8b-c). By
comparing Fig. 8b and c, it is clear that the adsorption data fits very
well with the pseudo-second order kinetic model (Fig. 8c). From

Fig. 8b-c, the kinetic parameters in Eqs. (5) and (6) were calculated.
The kinetic parameters are given in Table S2, Supplementary information.
The adsorption isotherm for methylene blue adsorption was also
evaluated. Two different adsorption isotherms were evaluated: the

from Fig. 7 that the adsorption capacity also depends on the starting
concentration of the methylene blue solution. For example, the adsorption capacity of the sample GU-40-60 increased from 15 mg/g to
465 mg/g as the concentration of the methylene blue solution increases
from 30 mg/L to 3000 mg/L. This is most probably because a more
concentrated solution of methylene blue induces a stronger driving
force for adsorption. On the other hand, the percent of adsorption decreased from 98 % to 30 % as the methylene blue concentration increases from 30 mg/L to 3000 mg/L, which indicated a limited number
of binding sites in the hydrogels. The adsorption capacity of the sample
GU-40-60 did not change when the methylene blue concentration was
increased from 2000 mg/L to 3000 mg/L indicating that 465 mg/g is
the maximum adsorption capacity. The maximum adsorption capacity
(465 mg/g) obtained in this study is in the range of what has been
reported for polysaccharide-based hydrogels in previous studies
(Table 2) (Dai & Huang, 2016; Liu, Zheng, & Wang, 2010; Sun et al.,
2015; Wang, Wang, & Wang, 2013; Zhou et al., 2014). However, the
adsorption capacity of a hydrogel depends on factors such as the
number of available binding sites. Different polysaccharides may differ
largely in chemical structure and number of available binding sites
8


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

Fig. 5. SEM-images of ulvan dialdehyde-gelatin hydrogels at x200 magnification, a) GU-80-20, b) GU-70-30, c) GU-60-40, d) GU-40-60.

The ulvan dialdehyde-gelatin hydrogels were also tested as a potential absorbent for heavy metals (Fig. 10). Dried hydrogels were immersed in a solution containing four different heavy metals, Cu2+,
Co2+, Ni2+, and Zn2+ for 24 h. The adsorption capacity was tested at
two different initial mass concentrations (30 mg/L and 150 mg/L) and
two different pH values (3 and 6.5). Adsorption tests at pH = 6.5 was
performed without adjusting the pH after dissolving the metal salts to
the solution to assess the adsorption capacity when no other ions are
present in the solution. However, heavy metal polluted industrial effluents are typically more acidic and adsorption tests were hence also
conducted at pH = 3 to assess the hydrogels performance under more
acidic conditions. The adsorption of Cu2+ was constantly higher than
the adsorption of Co2+, Ni2+, and Zn2+, so the hydrogels seems to have
a stronger affinity towards Cu2+ ions, in agreement with a previous
study on ulvan (Lahaye & Robic, 2007). The hydrogels have almost
equal affinity towards Co2+, Ni2+, and Zn2+. The adsorption of heavy
metals is overall lower at pH = 3 than at pH = 6.5. One possible explanation is that the carboxylic groups in the uronic acid moieties in
ulvan dialdehyde are protonated to a higher extent at pH = 3 than at
pH = 6.5. The carboxylic groups act as binding sites for heavy metal
ions. The interaction between the carboxylic groups and the heavy
metal ions in solution will be weaker when the carboxylic groups are
protonated. At pH = 3, the amine groups in gelatin will also be protonated to a higher extent. The protonated amine groups (NH3+) could
cause electrostatic repulsion of the heavy metal ions, which prevents
heavy metal adsorption at lower pH values. Consequently, the affinity
to heavy metal ions will be lower at pH 3 than at pH 6.5. By comparing
the adsorption capacities of the hydrogels, we can see that the sample
GU-80-20 had the lowest adsorption capacity and the sample GU-40-60
had the highest. The hydrogels reached a maximum adsorption capacity
of 14 mg/g for Cu2+, 7 mg/g for Co2+, 6 mg/g for Ni2+, and 6 mg/g for
Zn2+ (Fig. 10, bottom-right) which is in the lower range of what was
previously reported in previous studies for polysaccharide-based hydrogels (Ferrari et al., 2015; Guilherme et al., 2010; Kandile & Nasr,

Fig. 6. The degree of swelling (SD) of ulvan dialdehyde-gelatin hydrogels in
deionized water (white bars) and in 0.01 PBS buffer (grey bars). The values are
given as mean values ± standard deviation of three replicates.

Langmuir isotherm, Eq. (7), and the Freundlich isotherm, Eq. (8). The
Langmuir isotherm model assumes that the surface of the hydrogel is
homogenous and that every binding site can adsorb one molecule of
methylene blue. It also assumes that only a monolayer of methylene
blue is adsorbed on the surface (Liu et al., 2010). The Freundlich adsorption isotherm model assumes that the surface of the hydrogel is
heterogeneous and it also allows multilayer adsorption (Liu et al.,
2010). By comparing Fig. 9a and b, it is clear that the Langmuir isotherm model (Fig. 9a) gives a much better linear fitting to the data than
the Freundlich isotherm model (Fig. 9b). This is an indication that the
adsorption of methylene blue onto ulvan dialdehyde-gelatin hydrogels
follows the Langmuir isotherm model which indicates that only one
monolayer of methylene blue is adsorbed. The calculated adsorption
isotherm parameters are given in Table S3, Supplementary information.
9


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

Fig. 7. The percent of adsorption (left) and the adsorption capacities (right) of ulvan dialdehyde-gelatin hydrogels for the adsorption of methylene blue at different
starting mass concentrations. Data are given as the mean values from triplicate measurements.

4. Conclusion

Table 2
Comparison between the maximum adsorption capacity (qe) of methylene blue

obtained in this study and the adsorption capacity of other polysaccharidebased hydrogels.
Hydrogel composition

qe (mg/g)

Reference

Ulvan dialdehyde/gelatin
Cellulose/sepia ink
Chitosan/polyacrylic acid
Xylan/polyacrylic acid/Fe3O4
Alginate/organo-illite smectite
Cellulose nanocrystals/polyacrylamide

465
138
1800
438
1843
326

This work
Dai and Huang (2016)
Liu et al. (2010)
Sun et al. (2015)
Wang et al. (2013)
Zhou et al. (2014)

This study involved the preparation of hydrogels based on gelatin
and the polysaccharide ulvan extracted from the green macroalgae Ulva

fenestrata. Ulvan was extracted from Ulva fenestrate, converted to ulvan
dialdehyde by oxidation with sodium periodate, and then crosslinked
with gelatin in a Schiff-base reaction leading to the formation of hydrogels. Hydrogels with different mass ratios of ulvan dialdehyde and
gelatin were prepared to investigate how the composition affects the
hydrogel properties. The hydrogels were tested as a potential absorbent
for heavy metals and the dye methylene blue. Other important parameters such as morphology, swelling behavior, mechanical properties,
and thermal stability were also investigated. The mass ratio of ulvan
dialdehyde and gelatin in the hydrogel had a huge impact on the hydrogel properties. The degree of swelling of the hydrogels increases as
the mass percentage of ulvan dialdehyde in the hydrogel increases. The
highest recorded degree of swelling was 2400 % and 900 % in deionized water and 0.01 M PBS buffer, respectively. The adsorption capacity of methylene blue and heavy metals was improved by a higher
mass percentage of ulvan dialdehyde in the hydrogel. The hydrogels
adsorbed up to 465 mg/g of methylene blue. The adsorption of

2009; Huang et al., 2015; O’Connell et al., 2008; Sun et al., 2015; Yu
et al., 2018; Zhao et al., 2019). However, the hydrogels prepared in
these studies were based on either cellulose, chitin, chitosan or hemicellulose which chemical structures differs largely from ulvan dialdehyde in terms of functional groups and monosaccharide composition.
As discussed previously, the adsorption capacities of hydrogels prepared from different polysaccharides cannot be directly compared since
they may differ largely in terms of chemical structure and number of
binding sites etc.

Fig. 8. a) The adsorption capacity of methylene blue as a function of time, b) linear fitting curve for the pseudo-first order adsorption kinetic model, c) linear fitting
curve for the pseudo-second order adsorption kinetic model.
10


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

Fig. 9. Linear fitting curves of a) Langmuir adsorption isotherm model, and b) Freundlich isotherm model.

Fig. 10. The percent of adsorption (left) and the adsorption capacities (right) of ulvan dialdehyde-gelatin hydrogels for the adsorption heavy metal solution at
different pH values and starting mass concentrations. Data are given as the mean values from triplicate measurements.

thermally stable up to 200 °C. Future work will involve development of
methods for improving the adsorption properties.

methylene blue follows the pseudo-second order kinetics and the adsorption data was in good agreement with the Langmuir adsorption
isotherm indicating monolayer adsorption. The adsorption of heavy
metal ions reached a maximum of 14 mg/g for Cu2+, 7 mg/g for Co2+,
6 mg/g for Ni2+, and 6 mg/g for Zn2+. Rheology measurements
showed typical behavior for hydrogels indicated by G’ > G’’ for the
whole frequency sweep. TGA analysis showed that the hydrogels were

CRediT authorship contribution statement
Niklas
Wahlström:
Conceptualization,
Data
curation,
Methodology, Formal analysis, Validation, Investigation, Visualization,
11


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

Writing - original draft, Writing - review & editing. Sophie Steinhagen:
Methodology, Data curation, Formal analysis, Writing - original draft,

Writing - review & editing. Gunilla Toth: Conceptualization,
Methodology, Supervision, Funding acquisition, Writing - review &
editing. Henrik Pavia: Funding acquisition, Methodology, Supervision,
Writing - review & editing. Ulrica Edlund: Conceptualization,
Methodology, Data curation, Supervision, Funding acquisition, Writing
- review & editing.

Jejurikar, A., Seow, X. T., Lawrie, G., Martin, D., Jayakrishnan, A., & Grøndahl, L. (2012).
Degradable alginate hydrogels crosslinked by the macromolecular crosslinker alginate dialdehyde. Journal of Materials Chemistry, 22(19), 9751–9758.
Kandile, N. G., & Nasr, A. S. (2009). Environment friendly modified chitosan hydrogels as
a matrix for adsorption of metal ions, synthesis and characterization. Carbohydrate
Polymers, 78(4), 753–759.
Kraan, S. (2013). Mass-cultivation of carbohydrate rich macroalgae, a possible solution
for sustainable biofuel production. Mitigation and Adaptation Strategies for Global
Change, 18(1), 27–46.
Lahaye, M. (1998). NMR spectroscopic characterisation of oligosaccharides from two Ulva
rigida ulvan samples (Ulvales, Chlorophyta) degraded by a lyase. Carbohydrate
Research, 314(1-2), 1–12.
Lahaye, M., & Robic, A. (2007). Structure and functional properties of ulvan, a polysaccharide from green seaweeds. Biomacromolecules, 8(6), 1765–1774.
Li, D., Ye, Y., Li, D., Li, X., & Mu, C. (2016). Biological properties of dialdehyde carboxymethyl cellulose crosslinked gelatin–PEG composite hydrogel fibers for wound
dressings. Carbohydrate Polymers, 137, 508–514.
Liu, Y., Zheng, Y., & Wang, A. (2010). Enhanced adsorption of Methylene Blue from
aqueous solution by chitosan-g-poly (acrylic acid)/vermiculite hydrogel composites.
Journal of the Environmental Sciences, 22(4), 486–493.
Marcoveccio, J. E., Botte, S. E., & Freije, R. H. (2007). In L. M. Nollet (Ed.). Handbook of
water analysis (pp. 275). (2nd edition). London: CRC Press.
Mondal, S. (2008). Methods of dye removal from dye house effluent—An overview.
Environmental Engineering Science, 25(3), 383–396.
Morelli, A., & Chiellini, F. (2010). Ulvan as a new type of biomaterial from renewable
resources: Functionalization and hydrogel preparation. Macromolecular Chemistry and

Physics, 211(7), 821–832.
Morelli, A., Betti, M., Puppi, D., Bartoli, C., Gazzarri, M., & Chiellini, F. (2016).
Enzymatically crosslinked ulvan hydrogels as injectable systems for cell delivery.
Macromolecular Chemistry and Physics, 217(4), 581–590.
Morelli, A., Betti, M., Puppi, D., & Chiellini, F. (2016). Design, preparation and characterization of ulvan based thermosensitive hydrogels. Carbohydrate Polymers, 136,
1108–1117.
Ngah, W. W., & Hanafiah, M. A. K. M. (2008). Removal of heavy metal ions from wastewater by chemically modified plant wastes as adsorbents: A review. Bioresource
Technology, 99(10), 3935–3948.
O’Connell, D. W., Birkinshaw, C., & O’Dwyer, T. F. (2008). Heavy metal adsorbents
prepared from the modification of cellulose: A review. Bioresource Technology, 99(15),
6709–6724.
Phillips, G. R., & Russo, R. C. (1978). Metal bioaccumulation in fishes and aquatic invertebrates: A literature review. Environmental Protection Agency, Office of Research
and Development, Environmental Research Laboratory.
Quemener, B., Lahaye, M., & Bobin Dubigeon, C. (1997). Sugar determination in ulvans
by a chemical-enzymatic method coupled to high performance anion exchange
chromatography. Journal of Applied Phycology, 9(2), 179–188.
Rastogi, K., Sahu, J. N., Meikap, B. C., & Biswas, M. N. (2008). Removal of methylene blue
from wastewater using fly ash as an adsorbent by hydrocyclone. Journal of Hazardous
Materials, 158(2-3), 531–540.
Robic, A., Bertrand, D., Sassi, J. F., Lerat, Y., & Lahaye, M. (2009). Determination of the
chemical composition of ulvan, a cell wall polysaccharide from Ulva spp.(Ulvales,
Chlorophyta) by FT-IR and chemometrics. Journal of Applied Phycology, 21(4),
451–456.
Robic, A., Gaillard, C., Sassi, J. F., Lerat, Y., & Lahaye, M. (2009). Ultrastructure of ulvan:
A polysaccharide from green seaweeds. Biopolymers, 91(8), 652–664.
Robic, A., Rondeau-Mouro, C., Sassi, J. F., Lerat, Y., & Lahaye, M. (2009). Structure and
interactions of ulvan in the cell wall of the marine green algae Ulva rotundata
(Ulvales, Chlorophyceae). Carbohydrate Polymers, 77(2), 206–216.
Sarker, B., Papageorgiou, D. G., Silva, R., Zehnder, T., Gul-E-Noor, F., Bertmer, M., et al.
(2014). Fabrication of alginate–gelatin crosslinked hydrogel microcapsules and

evaluation of the microstructure and physico-chemical properties. Journal of
Materials Chemistry B, 2(11), 1470–1482.
Steinhagen, S., Karez, R., & Weinberger, F. (2019). Cryptic, alien and lost species:
Molecular diversity of Ulva sensu lato along the German coasts of the North and
Baltic Seas. European Journal of Phycology, 54(3), 466–483.
Sun, X. F., Liu, B., Jing, Z., & Wang, H. (2015). Preparation and adsorption property of
xylan/poly (acrylic acid) magnetic nanocomposite hydrogel adsorbent. Carbohydrate
Polymers, 118, 16–23.
Taboada, C., Millán, R., & Míguez, I. (2010). Composition, nutritional aspects and effect
on serum parameters of marine algae Ulva rigida. Journal of the Science of Food and
Agriculture, 90(3), 445–449.
Tummalapalli, M., & Gupta, B. (2015). A UV-vis spectrophotometric method for the estimation of aldehyde groups in periodate-oxidized polysaccharides using 2, 4-dinitrophenyl hydrazine. Journal of Carbohydrate Chemistry, 34(6), 338–348.
Wahlström, N., Nylander, F., Malmhäll-Bah, E., Sjövold, K., Edlund, U., Westman, G.,
et al. (2020). Composition and structure of cell wall ulvans recovered from Ulva spp.
along the Swedish west coast. Carbohydrate PolymersArticle 115852.
Wang, Y., Wang, W., & Wang, A. (2013). Efficient adsorption of methylene blue on an
alginate-based nanocomposite hydrogel enhanced by organo-illite/smectite clay.
Chemical Engineering Journal, 228, 132–139.
Wysokowski, M., Bartczak, P., Żółtowska‐Aksamitowska, S., Chudzińska, A., Piasecki, A.,
Langer, E., et al. (2017). Adhesive stalks of diatom Didymosphenia geminata as a
novel biological adsorbent for hazardous metals removal. CLEAN–Soil, Air, Water,
45(11), Article 1600678.
Wysokowski, M., Klapiszewski, Ł., Moszyński, D., Bartczak, P., Szatkowski, T., Majchrzak,
I., et al. (2014). Modification of chitin with kraft lignin and development of new
biosorbents for removal of cadmium (II) and nickel (II) ions. Marine Drugs, 12(4),

Acknowledgements
We thank Swedish Foundation for Strategic Research (SSF), project
number RBP14-0045 for financial support. We greatly thank our colleagues Göran Nylund, Friedrike Eimer and Gunnar Cervin at Göteborg
University for help with collection of the seaweed samples. We also

thank Adbusalam Uhedia and Fei Ye at KTH Royal Institute of
Technology for the help with the ICP-OES analysis.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi: />References
Agency for Toxic Substances and Disease Registry (ATSDR) (2017). The ATSDR 2017
substance priority list. Accessed 2020-01-30.
Ahmad, R., & Kumar, R. (2010). Adsorption studies of hazardous malachite green onto
treated ginger waste. Journal of Environmental Management, 91(4), 1032–1038.
Amer, H., Nypelö, T., Sulaeva, I., Bacher, M., Henniges, U., Potthast, A., et al. (2016).
Synthesis and characterization of periodate-oxidized polysaccharides: Dialdehyde
xylan (DAX). Biomacromolecules, 17(9), 2972–2980.
Barakat, M. A. (2011). New trends in removing heavy metals from industrial wastewater.
Arabian Journal of Chemistry, 4(4), 361–377.
Bartczak, P., Klapiszewski, Ł., Wysokowski, M., Majchrzak, I., Czernicka, W., Piasecki, A.,
et al. (2017). Treatment of model solutions and wastewater containing selected hazardous metal ions using a chitin/lignin hybrid material as an effective sorbent.
Journal of Environmental Management, 204, 300–310.
Dai, H., & Huang, H. (2016). Modified pineapple peel cellulose hydrogels embedded with
sepia ink for effective removal of methylene blue. Carbohydrate Polymers, 148, 1–10.
Dash, M., Samal, S. K., Morelli, A., Bartoli, C., Declercq, H. A., Douglas, T. E. L., et al.
(2018). Ulvan-chitosan polyelectrolyte complexes as matrices for enzyme induced
biomimetic mineralization. Carbohydrate Polymers, 182, 254–264.
de Carvalho, M. M., de Freitas, R. A., Ducatti, D. R., Ferreira, L. G., Gonỗalves, A. G.,
Colodi, F. G., et al. (2018). Modication of ulvans via periodate-chlorite oxidation:
Chemical characterization and anticoagulant activity. Carbohydrate Polymers, 197,
631–640.
De Ruiter, G. A., Schols, H. A., Voragen, A. G., & Rombouts, F. M. (1992). Carbohydrate
analysis of water-soluble uronic acid-containing polysaccharides with high-performance anion-exchange chromatography using methanolysis combined with TFA
hydrolysis is superior to four other methods. Analytical Biochemistry, 207(1),
176–185.

Ebenstein, A. (2012). The consequences of industrialization: Evidence from water pollution and digestive cancers in China. The Review of Economics and Statistics, 94(1),
186–201.
Fan, Q. G., Lewis, D. M., & Tapley, K. N. (2001). Characterization of cellulose aldehyde
using Fourier transform infrared spectroscopy. Journal of Applied Polymer Science,
82(5), 1195–1202.
Ferrari, E., Ranucci, E., Edlund, U., & Albertsson, A. C. (2015). Design of renewable poly
(amidoamine)/hemicellulose hydrogels for heavy metal adsorption. Journal of
Applied Polymer Science, 132(12).
Guilherme, M. R., Reis, A. V., Paulino, A. T., Moia, T. A., Mattoso, L. H., & Tambourgi, E.
B. (2010). Pectin‐based polymer hydrogel as a carrier for release of agricultural nutrients and removal of heavy metals from wastewater. Journal of Applied Polymer
Science, 117(6), 3146–3154.
Gupta, B., Tummalapalli, M., Deopura, B. L., & Alam, M. S. (2014). Preparation and
characterization of in-situ crosslinked pectin–gelatin hydrogels. Carbohydrate
Polymers, 106, 312–318. Accessed 2020-02-13 />sustainabledevelopment.
Hu, W., Wang, Z., Xiao, Y., Zhang, S., & Wang, J. (2019). Advances in crosslinking
strategies of biomedical hydrogels. Biomaterials Science, 7(3), 843–855.
Huang, C. H., Hsieh, T. H., & Chiu, W. Y. (2015). Evaluation of thermally crosslinkable
chitosan-based nanofibrous mats for the removal of metal ions. Carbohydrate
Polymers, 116, 249–254.
Hughey, J. R., Maggs, C. A., Mineur, F., Jarvis, C., Miller, K. A., Shabaka, S. H., et al.
(2019). Genetic analysis of the Linnaean Ulva fenestrata (Ulvales, Chlorophyta) holotype and related type specimens reveals name misapplications, unexpected origins,
and new synonymies. Journal of Phycology, 55(3), 503–508.

12


Carbohydrate Polymers 249 (2020) 116841

N. Wahlström, et al.

crosslinked gelatin hydrogels with high strength for biomedical applications.
Carbohydrate Polymers, 216, 45–53.
Zhang, H. J., Mao, W. J., Fang, F., Li, H. Y., Sun, H. H., Chen, Y., et al. (2008). Chemical
characteristics and anticoagulant activities of a sulfated polysaccharide and its
fragments from Monostroma latissimum. Carbohydrate Polymers, 71(3), 428–434.
Zhao, B., Jiang, H., Lin, Z., Xu, S., Xie, J., & Zhang, A. (2019). Preparation of acrylamide/
acrylic acid cellulose hydrogels for the adsorption of heavy metal ions. Carbohydrate
Polymers, 224, Article 115022.
Zhou, Z., Liu, Y. G., Liu, S. B., Liu, H. Y., Zeng, G. M., Tan, X. F., et al. (2017). Sorption
performance and mechanisms of arsenic (V) removal by magnetic gelatin-modified
biochar. Chemical Engineering Journal, 314, 223–231.
Zhou, C., Wu, Q., Lei, T., & Negulescu, I. I. (2014). Adsorption kinetic and equilibrium
studies for methylene blue dye by partially hydrolyzed polyacrylamide/cellulose
nanocrystal nanocomposite hydrogels. Chemical Engineering Journal, 251, 17–24.

2245–2268.
Yaseen, D. A., & Scholz, M. (2019). Textile dye wastewater characteristics and constituents of synthetic effluents: A critical review. International Journal of
Environmental Science and Technology, 16(2), 1193–1226.
Ye, N. H., Zhang, X. W., Mao, Y. Z., Liang, C. W., Xu, D., Zou, J., et al. (2011). “Green
tides” are overwhelming the coastline of our blue planet: Taking the world’s largest
example. Ecological Research, 26(3), 477–485.
Yu, C., Tang, X., Liu, S., Yang, Y., Shen, X., & Gao, C. (2018). Laponite crosslinked starch/
polyvinyl alcohol hydrogels by freezing/thawing process and studying their cadmium
ion adsorption. International Journal of Biological Macromolecules, 117, 1–6.
Yuan, L., Wu, Y., Gu, Q. S., El-Hamshary, H., El-Newehy, M., & Mo, X. (2017). Injectable
photo crosslinked enhanced double-network hydrogels from modified sodium alginate and gelatin. International Journal of Biological Macromolecules, 96, 569–577.
Zhang, L., Liu, J., Zheng, X., Zhang, A., Zhang, X., & Tang, K. (2019). Pullulan dialdehyde

13