Accepted Manuscript

Title: Chitosan-starch beads prepared by ionotropic gelation as

potential matrices for controlled release of fertilizers

Author: Jonas J. Perez Nora J. Francois

PII: S0144-8617(16)30415-5
DOI: http://dx.doi.org/doi:10.1016/j.carbpol.2016.04.054
Reference: CARP 10994

To appear in:

Received date: 3-12-2015

Revised date: 9-4-2016
Accepted date: 11-4-2016

Please cite this article as: Perez, Jonas J., & Francois, Nora J., Chitosan-starch
beads prepared by ionotropic gelation as potential matrices for controlled release of
fertilizers.Carbohydrate Polymers http://dx.doi.org/10.1016/j.carbpol.2016.04.054

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Chitosan-starch beads prepared by ionotropic gelation as potential matrices for
controlled release of fertilizers

Jonas J. Pereza,b,c and Nora J. Francoisa,b*

a
Grupo de Aplicaciones de Materiales Biocompatibles, Departamento de Química,
Facultad de Ingeniería, Universidad de Buenos Aires (UBA), Argentina
b
Instituto de Tecnología en Polímeros y Nanotecnología (ITPN), UBA-CONICET,
Argentina
c
Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
*
Corresponding author: nfranco@fi.uba.ar
Highlihgts

• A biodegradable fertilizer made of blends of starch and chitosan is proposed

• A dripping technique involving ionotropic crosslinking is used

• Starch content and crosslinking time highly affect the macrobeads characteristics
Abstract

The present study examines the agrochemical application of macrospheres prepared with
chitosan and chitosan-starch blends by an easy dripping technique, using a sodium
tripolyphosphate aqueous solution as the crosslinking agent. These biopolymers form
hydrogels that could be a viable alternative method to obtain controlled-release fertilizers
(CRFs). Three different concentrations (ranging from 20 to 100 wt/wt % of chitosan) and
two crosslinking times (2 or 4 hours) were used. The resulting polymeric matrices were
examined by scanning electron microscopy coupled with energy dispersive X-ray, X-ray
diffraction, Fourier transform infrared spectroscopy, solid-state nuclear magnetic
resonance, thermogravimetric analysis and differential scanning calorimetry. Ionotropic
gelation and neutralization induced the formation of the macrospheres. The crosslinking
time and the composition of the polymeric hydrogel controlled the crosslinking degree, the
swelling behavior and the fertilizer loading capability. Potassium nitrate-loaded beads were
shown to be useful as a controlled-release fertilizer. After 14 days of continuous release
into distilled water, the cumulative concentration in the release medium reached between
70 and 93 % of the initially loaded salt, depending on the matrix used. The prepared beads
showed properties that make them suitable for use in the agrochemical industry as CRFs.

Keywords: Chitosan; starch; macrospheres; fertilizer

1. Introduction

Agriculture is strongly dependent on the addition of fertilizers. Extensive usage of

fertilizers has been associated with pollution caused by high levels of nutrients in the
ground and surface waters (Valiela et al, 1992; Carpenter et al, 1998; Shaviv & Mikkelsen,
1992), causing health and environmental problems and increasing water purification costs
(Carpenter et al, 1998).
On the other hand, the degradation of soil quality after many cycles of intensive agricultural
use leads to an increased demand for fertilization, irrigation and energy to maintain the
productivity of those soils (Valiela et al, 1992; Carpenter et al, 1998; Shaviv & Mikkelsen,
1992).
One approach to reduce the problems associated with the excessive use of fertilizers is to
devise a method that will reduce the quantity and frequency of its application. One way to
achieve this goal is to use controlled-release fertilizers (CRFs). These agrochemicals use a
physical barrier to reduce the dissolution rate of the fertilizers. Their physical
characteristics (reservoir or matrix system), along with the mechanism that governs the
fertilizer release (diffusion, swelling or degradation of the polymeric matrix), can modulate
the release pattern of nutrients (Rashidzadeh et al, 2014; Zhong et al, 2013; Shaviv, 2001).
Cost is an important parameter when analyzing the viability of CRFs for industrial
manufacturing. For instance, the production of CRFs using hydrogels as carrier matrices
still has a considerably higher cost than the use of conventional mineral fertilizers
(Rashidzadeh et al, 2014; Zhong et al, 2013). Some polymers can be characterized as “low
cost,” as they require little processing, are abundant in nature and/or are byproducts or
waste materials from other industrial processes. Attempts to use hydrophilic biopolymers to
prepare CRFs have been well-documented and reviewed; however, their high cost/benefit
ratio hinders their wide use in agriculture. (Jamnongkan & Kaewpirom, 2010; Melaj &
Daraio, 2013; Azeem, 2014).
On the other hand, low-cost synthetic polymers currently used for CRFs often have
extremely slow or no decomposition in the soil (Ni, Lü, & Liu, 2012; Davidson & Gu,
2012; Trenkel, 1997), resulting in an accumulation of non-degradable residues. Taking this
enormous disadvantage into account, the use of biodegradable polymers would offer an
excellent solution to the problem.
Among hydrophilic polymers, we have selected two biodegradable polysaccharides,
chitosan and potato starch, that fulfill the desired expectations of low cost, non-polluting
characteristics and the ability to modulate fertilizer release.
Chitosan is a linear polysaccharide consisting of β(1→4) linked D-glucosamine residues
with a variable number of randomly located N-acetyl-glucosamine groups (Krajewska,
2004). It is the most important derivative of chitin, which is the main constituent of the
crustacean exoskeleton. It is obtained from partial deacetylation of chitin under alkaline
conditions or enzymatic hydrolysis in the presence of a chitin deacetylase (Krajewska,
2004; George & Abraham, 2006). Chitosan is a cationic, biocompatible, biodegradable and
non-toxic polymer (Kean & Thanou, 2009) that forms ionic complexes with a wide variety
of water-soluble anionic polymers or with anionic crosslinking agents, allowing the
formation of an insoluble gel (George & Abraham, 2006). It can be chemically crosslinked
with reagents, such as glutaraldehyde, genipin or epichlorohydrin (Jose et al, 2014). In this
study, physical crosslinking by electrostatic interactions with sodium tripolyphosphate was
selected instead of chemical crosslinking in order to avoid the possibility of soil
contamination with unreacted chemical crosslinkers. In addition to all of the properties
already mentioned, chitosan has excellent properties for agriculture applications because it
is a plant growth promoter and also protects against the actions of certain microorganisms
(Jose et al, 2014; Nge et al, 2006; Bautista-Baños et al, 2006). On the other hand, its cost
significantly diminishes its potential for application in the agrochemical industry.
To produce a low-cost matrix useful for the development of a controlled-release fertilizer
based on an ionically crosslinked chitosan, a blend of chitosan and starch was studied.
Starch possesses desirable properties, such as compatibility, hydrophilicity,
biodegradability, non-toxicity and low-cost (Frost et al, 2009). Using these blends, much
lower quantities of chitosan are needed to prepare the agrochemical.
Starch is a complex natural non-ionic polysaccharide consisting of two polymer fractions
that vary widely according to the starch source. The first fraction is a linear polymer called
amylose, which is made of β(1→4) linked D-glucose, and the second is a branched polymer
named amylopectin, which contains the same monomers that are also joined by C1–C6
linkages (Tripathi & Dubey, 2004; Pavlovic & Brandao, 2003; Rindlav-Westling, Stading,
& Gatenholm, 2002).
Therefore, the aim of this study was to investigate the feasibility of a simple method of
preparing non-polluting hydrogels based on blends of chitosan and starch to be used in the
agricultural industry as matrices for the controlled-release of fertilizers. Macrospheres were
prepared by ionotropic crosslinking of chitosan using sodium tripolyphosphate. The
preparation conditions (crosslinking time and composition of the blends) were analyzed to
produce low-cost macrospheres with the lowest possible content of chitosan. The prepared
hydrogels were characterized to elucidate their structural, thermal and morphological
properties. The final structure of the beads determined their behavior during swelling,
which is the main determinant of their fertilizer loading capability and release kinetics.

2. Materials and Methods

2.1. Materials

Chitosan of medium molecular weight (81% degree of deacetylation), native potato starch
and sodium tripolyphosphate (85 % purity) were purchased from Sigma-Aldrich (USA).
Lactic acid (85% wt/wt) and potassium nitrate were purchased from Cicarelli (Argentina).
All reagents used were of analytical grade and used as received.

2.2. Preparation of macrospheres

A 3 % wt/wt chitosan solution was prepared by dissolving chitosan (CS) powder in an

aqueous solution of lactic acid (1 % v/v) with mechanical stirring.
The potato starch gel was prepared by heating an 8 % wt/v starch (ST) solution in deionized
water with constant magnetic stirring. Gelatinization was achieved at 76 ºC in a boiling
distilled water bath.
The crosslinking solution was prepared by dissolving sodium tripolyphosphate (TPP) in
distilled water to produce a final concentration of 1 % wt/v (pH 8.6).
The blends were prepared by mixing the solution of chitosan and the starch gel with
mechanical stirring. Blends were prepared with CS/ST mass ratios of 100/0 (CS100), 30/70
(30CS/70ST) and 20/80 (20CS/80ST).
After they were obtained, the blends were kept at room temperature for 30 min before
preparing the beads using a dripping technique. The CS solution and the CS/ST blends
were dripped into the TPP solution using plastic tips. After 2 or 4 hours of continuous
stirring at room temperature, the macrospheres were removed from the TPP solution and
extensively washed with distilled water. To ensure complete cleaning, after a drying
process at low temperature, the macrospheres were rehydrated for 3 hours under magnetic
stirring and finally dried at 40 °C for 48 hours.

2.3. Soluble fraction

A known mass of dried beads was put into a 200 mesh stainless steel spherical net. After
being immersed in distilled water for 3 hours with magnetic stirring, the samples were dried
at 40 °C for 48 hours and weighed. The soluble fraction was defined as the ratio between
the mass lost as a consequence of the water immersion and the initial mass of the dried
material before immersion.
(1)

2.4. Characterization

2.4.1 Scanning electron microscopy (SEM) with energy dispersive X-ray analysis (EDS):
Crosslinked chitosan and chitosan-starch macrospheres were mounted on aluminum stubs
with double-sided adhesive tape and sputter-coated with gold. The surface morphology was
examined with a Karl Zeiss Supra 40 SEM (Germany) with a field emission gun operated at
3 kV. The micrographs were taken at magnifications between 70 and 25000 X. Elemental
analysis was performed using an Oxford Instruments EDS (United Kingdom) coupled to
SEM.

2.4.2. X-ray diffraction (XRD)

XRD patterns were obtained using a Rigaku diffractometer with Bragg Brentano geometry
and CuKα radiation (λ= 0.1542 nm, 40 kV, 20 mA) in the range of 2θ = 5 – 50 º at a
scanning rate of 1 º/min and a scan step of 0.05 º. The chart speed was set to 5 °/min.
Measurements were performed at ambient conditions.

2.4.3. Fourier transform infrared spectroscopy (FTIR)

FTIR spectra of pure chitosan, pure starch and chitosan and chitosan-starch macrospheres
were recorded on a Nicolet 380 FTIR spectrometer (Thermo Scientific, Japan) operating in
the range of 4000 – 400 cm-1 at a resolution of 4 cm-1. The samples were ground and mixed
thoroughly with potassium bromide at a 1:20 (sample:KBr) mass ratio. KBr discs were
prepared by compressing the powder mixture with a hydraulic press.
2.4.4. Nuclear magnetic resonance (NMR)

High-resolution 13C solid-state spectra were recorded using the ramp 1H−13C and the
combined techniques of proton dipolar decoupling (DD), magic angle spinning (MAS) and
cross polarization (CP). Experiments were performed at room temperature in a Bruker
Avance II-300 spectrometer equipped with a 4-mm MAS probe. The operating frequency
for carbons was 300.13 MHz. Glycine was used as an external reference for the 13C spectra,
and the Hartmann−Hahn matching condition was set in the cross-polarization experiments
13
C spectra. The recycling time was 4 s. Different contact times during CP were employed
in the range of 200 − 1500 μs for 13C spectra.

2.4.5. Differential scanning calorimetry (DSC)

Thermograms of chitosan, starch, cross-linked chitosan and chitosan-starch macrospheres

were obtained with a Shimadzu DSC-60-Plus instrument (Japan). The samples were
hermetically sealed in aluminum pans and heated at a constant rate of 10 °C/min. The DSC
tracings were performed from 25 to 350 °C. An inert atmosphere was maintained by
injecting nitrogen at a flow rate of 30 mL/min.

2.4.6. Thermogravimetric analysis (TGA)

Thermal degradation processes were investigated using a Shimadzu TGA-50 (Japan).

Measurements were carried out by heating the sample from 25 to 500 °C under an inert
atmosphere maintained by injecting N2 at a flow rate of 30 mL/min, with a heating rate of
10 ºC min-1 and using a sample weight of approximately 10 mg.

2.4.7. Equilibrium swelling

A known weight of macrospheres was dried to a constant value and then immersed in
distilled water at room temperature. The experimental equilibrium swelling degree ( )
was calculated as:

(2)

where, is the weight of the swelled macrospheres at equilibrium and is the weight
of the dry polymeric matrix before the swelling process.

2.4.8. Loading of potassium nitrate

Potassium nitrate in powder form was dissolved in deionized water at a final concentration
of 20 % wt/wt. Dry polymeric matrices were immersed in the fertilizer-saturated aqueous
solution for 4 hours at room temperature. After the swelling process, the macrospheres
were dried at 40 ºC for 48 hours. The loading percentage was calculated using the
following equation:
 (3)

Where and are the weights of the loaded and unloaded dry macrospheres,
respectively.

2.4.9. Release tests

The release of potassium nitrate-loaded in the matrices was performed using deionized
water as the release medium. A constant fixed weight of loaded beads was placed inside a
steel basket that was immersed in 100 mL of deionized water at a constant temperature of
25 ºC under un-stirred conditions (static experiment). At fixed time intervals, the
cumulative concentration of KNO3 released was evaluated by measuring the conductivity
of the release medium with a Hanna Instruments HI 9033 multi-range conductivity meter.
The cumulative concentration of KNO3 was determined from the calibration plot.

2.4.10. Statistical analyses

The values reported are the means and standard deviations of experiments carried out at
least three times. Data were analyzed using a one-way analysis of variance (t-test), and
p<0.05 was considered significant.

3. Results and discussion

3.1. Soluble fraction

As a consequence of the TPP dissolution and hydrolysis, the aqueous crosslinking solution
possesses OH− and tripolyphosphoric ions. Both ions can diffuse into hydrogels when the
polymeric material is in contact with the aqueous TPP solution. The OH− ions compete
with P3O105- ions for reaction with the protonated amino group of chitosan as soon as the
polymeric droplets come into contact with the sodium tripolyphosphate solution.
Taking into account that ionic crosslinking can be controlled by adjusting the pH of the
TPP aqueous solution, a pH of 8.6 was chosen in order to achieve a low crosslinking
density in the polymeric matrix (Mi et al, 1999; Shu & Zhu, 2001).
The macrospheres prepared from chitosan and from blends maintained their integrity (did
not dissolve or break apart) during the experiment. The calculated soluble fraction is
directly related to the selected polymeric composition and crosslinking time. The soluble
fractions are shown in Table 1.
A slight decrease in the soluble fraction was observed when the crosslinking time or the
chitosan content were increased. Bourtoom & Chinnan obtained similar experimental
results by preparing films made of rice starch-chitosan blends (Bourtoom & Chinnan,
2008). They reported a decrease in the soluble fraction with the augmentation of the CS wt
% that could be attributable to the interactions between both biopolymers. When the CS
percent or the crosslinking time (tc) are lowered, the ionic crosslinking between CS chains
decreases, and the interactions between biopolymers are affected, producing a larger
soluble fraction.

Table 1

3.2 SEM and EDS analysis

EDS was used to analyze the phosphorus content on the external surface of the
macrospheres after 2 or 4 hours of cross-linking in order to obtain information about the
effectiveness of the crosslinking. An increase in the amount of phosphorus would confirm
an increase in the crosslinking density (Mi et al, 1999). The phosphorus wt % measured on
the external surface of different matrices is shown in Table 1. The semi-quantitative results
revealed that the content of phosphorus is highly dependent on the tc. The experimental
results are in agreement with the lower crosslinking degree achieved at the minimum tc
tested.
It is also observed that this phosphorus wt % decreased after incorporation of ST. This fact
could indicate that, in the case of matrices prepared from blends, the added ST is not
involved in the ionic crosslinking with TPP. Using a room temperature setting in the
preparation method of macrospheres eliminates the possibility of a crosslinking reaction
between ST and TPP, which only would occur at a high temperature (>100 ºC) (Deetae et al
2008; Lim. & Seib, 1993).
Figure 1 shows the morphology of dried samples prepared with chitosan or polymeric
blends with different crosslinking times. The bead diameters, regardless the crosslinking
time, are 3.70 ± 0.03 mm for chitosan beads and 3.00 ± 0.05 mm for matrices prepared with
blends. These sizes are optimal for a final agrochemical application.
Other general properties were observed, including the findings that none of the analyzed
macrospheres were completely spherical in shape and that the crosslinking time does not
significantly affect the final size. According to the experimental procedure, the main factor
that would affect the size of the beads is the diameter of the plastic tip used in the dripping
technique.
The micrographs of the surfaces of the cross-sectioned beads show many cavities of
different sizes. The presence of these cavities could be caused by the preparation method.
They could be the consequence of occluded air bubbles incorporated during the preparation
of the polymeric material used to manufacture the beads. The observed inner structure is
useful in the preparation of a controlled-release fertilizer because it would enhance the
loading of the fertilizer inside the polymeric matrix.

Figure 1

In accordance with previous reports (Mi et al, 1999; Shu & Zhu, 2001), the pH chosen for
the TPP solution used in this work is adequate because, due to the low crosslinking density
achieved, it produces porous structures.

3.3. FTIR results

Fourier transform infrared spectroscopy studies were conducted in order to gain insight into
the structural changes of the macrospheres obtained with different compositions and using
either 2 or 4 hours of crosslinking time. Figure 2 shows all of the spectra obtained by this
method.
The FTIR spectrum of potato starch showed common features of polysaccharides. An
intense band at 1653 cm−1 was assigned to the O–H bending vibration (Capron et al, 2007;
Fang et al, 2002; Iizuka & Aishima, 1999). The bands at 1151 and 2930 cm−1 correspond to
the C–O and C–H stretching regions (Capron et al, 2007; Fang et al, 2002). The 3380 to
3420 cm-1 region (Capron et al, 2007) shows the weakest stretching frequency, that of the
O–H band (Capron et al, 2007; Fang et al, 2002).
The pure chitosan spectrum presented a characteristic band at 3449 cm-1 that is attributed to
an overlap of the stretching vibrations of the –NH2 and –OH groups. There is a band
located at 1657 cm-1 for amide I (Cui et al, 2008; Zawadzki & Kaczmarek, 2010; Pierog,
Gierszewska-Drużyńska, & Ostrowska-Czubenko, 2009; Pawlak & Mucha, 2003). There is
also a band at 1657 cm-1 corresponding to the CONH2 group and a signal at 1598 cm-1 due
to the NH2 group (Bhumkar & Pokharkar, 2006; Mi et al, 1999; Pierog, Gierszewska-
Drużyńska, & Ostrowska-Czubenko, 2009).
Macrospheres prepared with only chitosan showed a shoulder at 1240 cm-1, which can be
assigned to the –P=O stretching vibration, indicating the presence of the phosphate group
as a consequence of ionic crosslinking (Bhumkar & Pokharkar, 2006; Mi et al, 1999;
Pierog, Gierszewska-Drużyńska, & Ostrowska-Czubenko, 2009). It was found that the
intensity of this shoulder increased with increasing crosslinking time.
A comparison of the chitosan spectrum with that of the chitosan macrospheres shows that
the peak at 1657 cm-1 disappeared after gelation, and two new peaks, at 1650 cm-1 and 1540
cm-1, appeared in the spectrum after 2 or 4 hours of crosslinking. This can be attributed to
the linkage between the polyphosphoric and protonated amino ions (Bhumkar & Pokharkar,
2006; Qi et al, 2004; Qi & Xu, 2004). Additionally, the intensity of the peak at 1540 cm−1
increased with increasing crosslinking time, suggesting a higher crosslinking density (Mi et
al, 1999).

Figure 2

The starch gelatinization causes exposure of the OH groups. It has been established that
these groups can form hydrogen bonds with the protonated amino groups of chitosan
(Mathew, Brahmakumar, & Abraham, 2006; Pelissari et al, 2009; Wang et al, 2007). These
interactions causes changes in the spectra corresponding to the chitosan/starch hydrogels, as
the amino group peak of chitosan shifted from 1657 to 1629 cm−1. These changes are
evidence of the good molecular compatibility between the biopolymers selected to produce
the macrospheres.
The FTIR spectra of matrices obtained from blends also revealed a shoulder at 1240 cm-1,
demonstrating the existence of the –P=O group and indicating that crosslinking took place
through ionic interactions between the negatively charged –P=O group and the NH3+ group
in chitosan.

3.4 NMR assays

NMR was conducted using on the chitosan beads in order to analyze the influence of the
crosslinking time on the ionic crosslinking produced between chitosan and TPP.
The CP-MAS 13C NMR spectrum for chitosan, shown in Figure 3, was very similar to that
reported in the literature (Barbi et al; 2015). The following characteristic signals can be
identified: d = 60 ppm (two convoluted signals are observed and attributed to carbon C6
and C2); d = 76.4 ppm (carbons C5 and C3); d = 84 ppm (carbon C4) and d = 106.2 ppm
(carbon C1). It was found that for peaks corresponding to the C1 and C4 positions
decreased in intensity and became broader than the original signals, indicating that the
interaction between CS and TPP effectively occurred during the creation of the polymeric
matrices. A stronger interaction is observed when the crosslinking time is increased.

Figure 3

3.5 XRD studies

XRD diffractograms of pure chitosan, pure starch and macrospheres prepared with chitosan
or blends are shown in Figure 4.
Starch is a semicrystalline polymer. Depending on the starch source, its crystallinity can
vary from 15 to 51 % (Foresti et al, 2014).
Potato starch showed typical crystalline peaks centered at 5.75, 16.85, 23.71, 24.16, 30.61
and 35.14°. It exhibited the characteristic ‘B’ type X-ray pattern of tuber starches
(Cheetham & Tao; 1998). Chitosan exhibited two crystalline and defined peaks at 10.55
and 19.7° (Qi et al, 2004). The peaks found in the CS100 diffractogram decreased slightly
when the ionic crosslinking time was increased. The crosslinking inhibits a close packing of
the polymer chains by reducing the degrees of freedom in the 3-D conformation, limiting or
even preventing the formation of crystalline regions (Qi & Xu, 2004).
The method used to obtain chitosan beads causes simultaneous deprotonation ionic
crosslinking of the chitosan chains. This process could produce deprotonated chains but not
crosslinked or mobile deprotonated chain segments between crosslinking points, which are
capable of a certain rearrangement that could explain the XRD diffractogram of chitosan
beads (Bhumkar & Pokharkar, 2006).
In the case of polymeric macrospheres prepared with blends, the XRD pattern showed a
decreased intensity of the characteristic peaks corresponding to chitosan and starch, which
is indicative of a degree of interaction between the molecular chains of both
polysaccharides (Xu et al, 2005).
As the starch/chitosan ratio increases, there are more OH groups from the starch polymeric
chains that are able to interact with chitosan’s protonated amino groups. This increased
interaction between biopolymers caused a shift of the chitosan diffraction peak from 10.55°
to a lower value and then its disappearance with increasing starch content (Figure 4 (e), (f)
and (g)) (Xu et al, 2005).
Figure 4

3.6 Thermal analysis

The thermogravimetric analysis (TGA) results from different samples are shown in Figure
5.
 Figure 5

From the TGA results, is possible to appreciate three different stages. A first stage of
weight loss below and approximately 100 °C can be attributed to water evaporation.
Polysaccharides usually have a strong affinity for water and, therefore, may be easily
hydrated, resulting in macromolecules with rather disordered structures. The hydration
properties of these polysaccharides depend on their primary and supra-molecular structure
(Kittur et al, 2002). In our case, the matrices prepared exclusively with chitosan presented
the highest weight loss. The addition of starch caused a decrease in the ionic crosslinking
degree because of the lower chitosan concentration and also caused a decrease in the
amount of water lost compared to the results obtained from chitosan macrospheres. As a
result of the high hydrophilicity of starch, the water retention showed anomalous behavior
because it was found that the weight lost at 100 ºC was lower in macrospheres prepared
from blends. Water can act as a plasticizer for both biopolymers, but in the case of starch, it
can increase retrogradation with a consequent increase in crystallinity (Zhiqiang, Xiao Su,
& Yi, 1990). This phenomenon can generate a barrier effect that could prevent the loss of
water from the macrospheres obtained from blends.
There is a second stage of weight loss, observed between 240 and 340 ºC, that is due to the
depolymerization of chitosan and to the decomposition of starch for matrices prepared with
blends. This phenomenon can be ascribed to a complex process that includes the
dehydration of the saccharide rings and the depolymerization and decomposition of the
glucose units belonging to starch (Fang et al, 2002).
Finally, there is a third stage between 340 and 500 ºC caused by the pyrolysis of chitosan
and the vaporization and elimination of the volatile products of starch (Zawadzki &
Kaczmarek, 2010).
The results obtained from the derivative thermogravimetric analysis (dTGA) are shown in
Table 2.

Table 2

The dTGA curves showed that there are differences in the temperature associated with the
maximum mass loss rate (Tmax) for beads of different composition or prepared with
different crosslinking times. For macrospheres prepared with blends, the factor that mostly
affects the Tmax is the mass composition. The experimental Tmax is lower than the value
obtained by applying the mixture rule (He, Zhu, & Inoue, 2004). This observation points to
the existence of molecular interactions between the biopolymers used, which have already
been reported (Mi et al, 1999). Independent of the sample being analyzed, Tmax always
decreased with increasing crosslinking time. This fact could be related to the decrease in
crystallinity caused by the increase in ionic crosslinking (Mi et al, 1999). For blend
matrices, Tmax increased with the percentage of starch added, showing higher thermal
stability than the samples prepared exclusively with chitosan. This result could be attributed
to an increase in crystallinity, as already discussed for the XRD and TGA results.

The DSC curves are shown in Figure 6. All samples showed an endothermic transition
associated with water evaporation in the 50 - 150 °C range. Regarding the temperature peak
of the mentioned transition, increasing tc caused a shift to higher temperatures because
more energy was required to remove water from hydrogels with a greater degree of
crosslinking (Sarmento et al, 2006).

Figure 6

The chitosan content was a determining factor in the registered temperature shift of the
endothermic peak. As the chitosan content increased, the shift was more pronounced
because of the increase in the ionic interactions that affected the crystallinity of the
matrices, as already discussed with the TGA results. A peak corresponding to exothermic
decomposition (depolymerization) is observed for chitosan at 292.78 °C (Choi et al, 2007),
while for the crosslinked matrices, the same transition was observed at lower temperatures
but with lower intensity. Ionic crosslinking with TPP led to a decrease in crystallinity,
which led to lower thermal stability for chitosan macrospheres than for pure chitosan (Mi et
al, 1999, Lee et al, 2001). In the case of macrospheres prepared from blends, it was
observed that the temperature shift corresponding to the peak of the exothermic transition
was more pronounced than the one registered for crosslinked matrices prepared exclusively
with chitosan.
This result could be related to all the interactions occurring in the polymeric blend. By
increasing the starch content, the level of ionic crosslinking or the physical interactions via
hydrogen bonds between both biopolymers tends to decrease.
TGA and DSC results were indicative of the existence of ionic crosslinking and good
compatibility between chitosan and starch macromolecules.

3.6 Swelling and KNO3 loading assays

Water absorption produces swelling of the tridimensional polymeric network. In the case of
beads loaded with potassium nitrate, this phenomenon has a key role in governing fertilizer
release (Huanbutta et al, 2011).
Due to the direct relationship between the swelling and the structure of the material,
modification of the crosslinking time and/or the polymeric composition of the beads could
be an interesting approach for regulating the release of the loaded fertilizer.
The swelling experiments indicated that none of the samples dissolved during the time
interval assayed and that all reached an equilibrium swelling capacity after 3 hours.
An increase in the crosslinking time induced a decrease of approximately 12 % in the
swelling degree of all matrices tested, regardless of its polymeric composition (Yalinca,
Yilmaz, & Bullici, 2012). This result could be explained by the increase in the ionic
interchain linkages (Mi et al, 1999).
In the case of macrospheres prepared using blends at equal tc, the swelling equilibrium
degree is lower than that of the chitosan beads, reaching a minimum when a 30 % wt/wt of
chitosan is used. The introduction of a certain amount of starch induces changes in the final
matrix structure associated with the ionic crosslinking density and with the intermolecular
interactions between both biopolymers. XRD, FTIR and TGA results indicate that
increasing the starch content affects the interactions within the tridimensional polymeric
network, producing a larger swelling degree accordingly to the results presented in Table 2.
Because of the loading method, a direct relationship between the swelling degree and the
loading capacity is expected (Samanta & Ray, 2014). The experimental maximum fertilizer
loading shown in Table 2 indicates that chitosan beads possessed a maximum loading
capacity in accordance with the higher equilibrium swelling degrees measured for these
matrices. Surprisingly, in the case of the beads prepared with both blends, the fertilizer
loading capacity was approximately the same. Further studies are being conducted in order
to understand this experimental result.

3.7 Release kinetics

To use a conductimetric method to determine the concentration of the loaded fertilizer in

the release medium, potassium nitrate was selected because it has pH-independent
solubility and does not affect the pH of the release medium. The release kinetics of the
loaded beads is directly influenced by the swelling behavior (Samanta & Ray, 2014). For
CS100 matrices, the tc had a significant influence on the release pattern, changing the
percentage of cumulative KNO3 released after 14 days from 95 % to 73 % when the
crosslinking time was changed from two to four hours (experimental results not shown).
Fig. 7 shows the release pattern obtained from matrices prepared with a tc of two hours. The
cumulative percent release was calculated, for each matrix, considering the experimental
maximum fertilizer loading (Table 2) as a 100 %. After fourteen days, the cumulative
concentration of KNO3 released was 95 % for CS100, 73 % for CS30/ST70 and 80 % for
CS20/ST80.
These results correlate with the equilibrium swelling degrees shown in Table 2 because the
mechanism of fertilizer release involves swelling of the polymeric matrix when the loaded
bead makes contact with water (Rashidzadeh et al, 2014).
There are numerous reports of matrices prepared with hydrogels that are intended to be
used as controlled-release fertilizers, but the release was faster than that obtained in this
work. For example, Jamnongkan and Kaewpirom prepared a poly(vinyl alcohol)/chitosan
hydrogel that released up to 70 % of the same fertilizer used in this work over only four
days. Rui and coworkers prepared a coated fertilizer with a loaded core of alginate that
released approximately 90 % of the loaded potassium during six days.
Taking into account that the release experiments were conducted immersing the loaded
beads in water and observing in Fig. 7 the release profiles of matrices crosslinked over 2
hours, it is clear that the developed matrices would be potentially useful as controlled-
release fertilizers.

Figure 7

Conclusions

Chitosan and chitosan/starch macrospheres can be easily prepared from an ionic

crosslinking reaction between chitosan macromolecules using an aqueous TPP solution.
The ionic crosslinking and the starch incorporation produced a decrease in the crystallinity
of the material prepared.
FTIR, XRD, TGA and DSC experimental results suggested good molecular compatibility
between starch and chitosan.
Swelling experiments, fertilizer loading and release kinetics indicated that the
macrospheres prepared with blends are excellent candidates to be used for controlled-
release of fertilizer.
This approach to control the properties of hydrogels with various combinations of
composition and crosslinking densities may find a wide range of industrial applications,
such as in the agroindustry.

Acknowledgments

This work and the data herein contained was supported by Universidad de Buenos Aires,
Argentina (Grant 20020130200086BA UBACyT 2014-2017).

References

Azeem, B., KuShaari, K., Man, Z. B., Basit, A., & Thanh, T. H. (2014). Review on
materials & methods to produce controlled release coated urea fertilizer. Journal of
Controlled Release, 181, 11-21.
Barbi, M. D. S., Carvalho, F. C., Kiill, C. P., Da Silva, B., Santagneli, S. H., Ribeiro, S. J.
L., & Gremião, M. P. D. (2015). Preparation and Characterization of Chitosan
Nanoparticles for Zidovudine Nasal Delivery. Journal of Nanoscience and
Nanotechnology, 15 (1), 865-874.
Bautista-Baños, S., Hernández-Lauzardo, A. N., Velázquez-del Valle, M. G., Hernández-
López, M., Barka, E. A., Bosquez-Molina, E., & Wilson, C. L. (2006). Chitosan as a
potential natural compound to control pre and postharvest diseases of horticultural
commodities. Crop Protection, 25 (2), 108-118.
Bhumkar, D. R., & Pokharkar, V. B. (2006). Studies on effect of pH on cross-linking of
chitosan with sodium tripolyphosphate: a technical note. AAPS Pharmscitech, 7 (2), E138-
E143.
Bourtoom, T. & Chinnan, M. (2008). Preparation and properties of rice starch–chitosan
blend biodegradable film. LWT-Food Science and Technology, 41 (9), 1633-1641.
Capron, I., Robert, P., Colonna, P., Brogly, M. & Planchot, V. (2007). Starch in rubbery
and glassy states by FTIR spectroscopy. Carbohydrate Polymers, 68 (2), 249-259.
Carpenter, S., Caraco, N., Correll, D., Howarth, R., Sharpley, A., & Smith, V. (1998).
Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological
applications, 8 (3), 559-568.
Cheetham, N. W., & Tao, L. (1998). Variation in crystalline type with amylose content in
maize starch granules: an X-ray powder diffraction study. Carbohydrate Polymers, 36 (4),
277-284.
Chien, S., Prochnow, L., & Cantarella, H. (2009). Recent developments of fertilizer
production and use to improve nutrient efficiency and minimize environmental impacts.
Advances in Agronomy, 102, 267-322.
Choi, C. Y., Kim, S. B., Pak, P. K., Yoo, D. I., & Chung, Y. S. (2007). Effect of N-
acylation on structure and properties of chitosan fibers. Carbohydrate Polymers, 68 (1),
122-127.
Cui, Z., Xiang, Y., Si, J., Yang, M., Zhang, Q., & Zhang, T. (2008). Ionic interactions
between sulfuric acid and chitosan membranes. Carbohydrate Polymers, 73 (1), 111-116.
Davidson, D., & Gu, F. (2012). Materials for sustained and controlled release of nutrients
and molecules to support plant growth. Journal of agricultural and food chemistry, 60 (4),
870-876.
Deetae, P., Shobsngob, S., Varanyanond, W., Chinachoti, P., Naivikul, O., & Varavinit, S.
(2008). Preparation, pasting properties and freeze–thaw stability of dual modified crosslink-
phosphorylated rice starch. Carbohydrate Polymers, 73 (2), 351-358.
Fang, J., Fowler, P., Tomkinson, J., & Hill, C. A. S. (2002). The preparation and
characterisation of a series of chemically modified potato starches. Carbohydrate Polymers,
47 (3), 245-252.
Foresti, M., Williams, M., Martínez-García, R., Vázquez, A. (2014). Analysis of a
preferential action of α-amylase from B. licheniformis towards amorphous regions of waxy
maize starch. Carbohydrate Polymers. 102, 80–87.
Frost, K., Kaminski, D., Kirwan, G., Lascaris, E., & Shanks, R. (2009). Crystallinity and
structure of starch using wide angle X-ray scattering. Carbohydrate Polymers, 78 (3), 543-
548.
Kean, T., & Thanou, M. (2010). Biodegradation, biodistribution and toxicity of chitosan.
Advanced drug delivery reviews, 62 (1), 3-11.
George, M., & Abraham, T. E. (2006). Polyionic hydrocolloids for the intestinal delivery of
protein drugs: alginate and chitosan—a review. Journal of controlled release, 114 (1), 1-14.
He, Y., Zhu, B., & Inoue, Y. (2004). Hydrogen bonds in polymer blends. Progress in
Polymer Science, 29(10), 1021-1051.
Huanbutta, K., Sriamornsak, P., Limmatvapirat, S., Luangtana-anan, M., Yoshihashi, Y.,
Yonemochi, E., & Nunthanid, J. (2011). Swelling kinetics of spray-dried chitosan acetate
assessed by magnetic resonance imaging and their relation to drug release kinetics of
chitosan matrix tablets. European Journal of Pharmaceutics and Biopharmaceutics, 77 (2),
320-326.
Iizuka, K., & Aishima, T. (1999). Starch Gelation Process Observed by FT IR/ATR
Spectrometry with Multivariate Data Analysis. Journal of Food Science, 64 (4), 653-658.
Jamnongkan, T., & Kaewpirom, S. (2010). Potassium release kinetics and water retention
of controlled-release fertilizers based on chitosan hydrogels. Journal of Polymers and the
Environment, 18 (3), 413-421.
Jose, S., Fangueiro, J., Smitha, J., Cinu, T., Chacko, A., Premaletha, K., & Souto, E.
(2012). Cross-linked chitosan microspheres for oral delivery of insulin: Taguchi design and
in vivo testing. Colloids and Surfaces B: Biointerfaces, 92, 175-179.
Kean, T., & Thanou, M. (2010). Biodegradation, biodistribution and toxicity of chitosan.
Advanced drug delivery reviews, 62 (1), 3-11.
Kittur, F. S., Prashanth, K. H., Sankar, K. U., & Tharanathan, R. N. (2002).
Characterization of chitin, chitosan and their carboxymethyl derivatives by differential
scanning calorimetry. Carbohydrate Polymers, 49 (2), 185-193.
Krajewska, B. (2004). Application of chitin-and chitosan-based materials for enzyme
immobilizations: a review. Enzyme and microbial technology, 35 (2), 126-139.
Lee, S. T., Mi, F. L., Shen, Y. J., & Shyu, S. S. (2001). Equilibrium and kinetic studies of
copper (II) ion uptake by chitosan-tripolyphosphate chelating resin. Polymer, 42(5), 1879-
1892.
Lim, S. & Seib, P. (1993). Preparation and pasting properties of wheat and corn starch
phosphates. Cereal Chemistry, 70 (2), 137-137.
Mathew, S., Brahmakumar, M., & Abraham, T. E. (2006). Microstructural imaging and
characterization of the mechanical, chemical, thermal, and swelling properties of starch–
chitosan blend films. Biopolymers, 82 (2), 176-187.
Melaj, M. A., & Daraio, M. E. (2013). Preparation and characterization of potassium nitrate
controlled release fertilizer based on chitosan and xanthan layered tablets. Journal of
Applied Polymer Science, 130 (4), 2422-2428.
Mi, F. L., Shyu, S. S., Lee, S. T., & Wong, T. B. (1999). Kinetic study of
chitosan tripolyphosphate complex reaction and acid resistive properties of the
chitosan tripolyphosphate gel beads prepared by in liquid curing method. Journal of
Polymer Science Part B: Polymer Physics, 37 (14), 1551-1564.
Nge, K. L., Nwe, N., Chandrkrachang, S., & Stevens, W. F. (2006). Chitosan as a growth
stimulator in orchid tissue culture. Plant Science, 170 (6), 1185-1190.
Ni, B., Lü, S., & Liu, M. (2012). Novel Multinutrient Fertilizer and Its Effect on Slow
Release, Water Holding, and Soil Amending. Industrial & Engineering Chemistry
Research, 51 (40), 12993-13000.
Pavlovic, S., & Brandao, P. R. G. (2003). Adsorption of starch, amylose, amylopectin and
glucose monomer and their effect on the flotation of hematite and quartz. Minerals
Engineering, 16 (11), 1117-1122.
Pawlak, A., & Mucha, M. (2003). Thermogravimetric and FTIR studies of chitosan blends.
Thermochimica Acta, 396 (1), 153-166.
Pelissari, F. M., Grossmann, M. V., Yamashita, F., & Pineda, E. A. G. (2009).
Antimicrobial, mechanical, and barrier properties of cassava starch− chitosan films
incorporated with oregano essential oil. Journal of Agricultural and Food Chemistry, 57
(16), 7499-7504.
Pierog, M., Gierszewska-Drużyńska, M., & Ostrowska-Czubenko, J. (2009). Effect of ionic
crosslinking agents on swelling behavior of chitosan hydrogel membranes. Progress on
Chemistry and Application of Chitin and its Derivatives. Polish Chitin Society, Łódź, 75,
82.
Qi, L., & Xu, Z. (2004). Lead sorption from aqueous solutions on chitosan nanoparticles.
Colloids and Surfaces A: Physicochemical and Engineering Aspects, 251 (1), 183-190.
Qi, L., Xu, Z., Jiang, X., Hu, C., & Zou, X. (2004). Preparation and antibacterial activity of
chitosan nanoparticles. Carbohydrate Research, 339 (16), 2693-2700.
Rashidzadeh, A., Olad, A., Salari, D., & Reyhanitabar, A. (2014). On the preparation and
swelling properties of hydrogel nanocomposite based on Sodium alginate-g-Poly (acrylic
acid-co-acrylamide)/Clinoptilolite and its application as slow release fertilizer. Journal of
Polymer Research, 21 (2), 1-15.
Rindlav-Westling, Å., Stading, M., and Gatenholm, P. (2002). Crystallinity and
morphology in films of starch, amylose and amylopectin blends. Biomacromolecules, 3 (1),
84-91.
Rui Liang, R., Liu, M., Wu, L., (2007). Controlled release NPK compound fertilizer with
the function of water retention. Reactive & Functional Polymers, 67, 769–779.
Samanta, H., & Ray, S. (2014). Synthesis, characterization, swelling and drug release
behavior of semi-interpenetrating network hydrogels of sodium alginate and
polyacrylamide. Carbohydrate Polymers, 99, 666-678.
Sarmento, B., Ferreira, D., Veiga, F., & Ribeiro, A. (2006). Characterization of insulin-
loaded alginate nanoparticles produced by ionotropic pre-gelation through DSC and FTIR
studies. Carbohydrate Polymers, 66(1), 1-7.
Shaviv, A. (2001). Advances in controlled-release fertilizers. Advances in agronomy, 71, 1-
49.
Shaviv, A., & Mikkelsen, R. (1993). Controlled-release fertilizers to increase efficiency of
nutrient use and minimize environmental degradation-A review.Fertilizer Research, 35 (1-
2), 1-12.
Shu, X. Z., & Zhu, K. J. (2001). Chitosan/gelatin microspheres prepared by modified
emulsification and ionotropic gelation. Journal of microencapsulation,18 (2), 237-245.
Trenkel, M. (1997). Controlled-release and stabilized fertilizers in agricultura (Vol. 11).
Paris: International Fertilizer Industry Association.
Tripathi, P., & Dubey, N. K. (2004). Exploitation of natural products as an alternative
strategy to control postharvest fungal rotting of fruit and vegetables. Postharvest Biology
and Technology, 32 (3), 235-245.
Valiela, I., Foreman, K., LaMontagne, M., Hersh, D., Costa, J., Peckol, P & Lajtha, K.
(1992). Couplings of watersheds and coastal waters: sources and consequences of nutrient
enrichment in Waquoit Bay, Massachusetts. Estuaries, 15 (4), 443-457.
Wang, Q., Zhang, N., Hu, X., Yang, J., & Du, Y. (2007). Chitosan/starch fibers and their
properties for drug controlled release. European Journal of Pharmaceutics and
Biopharmaceutics, 66 (3), 398-404.
Xu, Y., Kim, K., Hanna, M., & Nag, D. (2005). Chitosan–starch composite film:
preparation and characterization. Industrial crops and Products, 21 (2), 185-192.
Yalinca, Z., Yilmaz, E., & Bullici, F. (2012). Evaluation of chitosan tripolyphosphate gel
beads as bioadsorbents for iron in aqueous solution and in human blood in vitro. Journal of
Applied Polymer Science, 125 (2), 1493-1505.
Zawadzki, J., & Kaczmarek, H. (2010). Thermal treatment of chitosan in various
conditions. Carbohydrate Polymers, 80 (2), 394-400.
Zhiqiang, L., Xiao su, Y., & Yi, F. (1999). Effect of bound water on thermal behaviors of
native starch, amylose and amylopectin. Starch Stärke, 51(11 12), 406-410.
Zhong, K., Lin, Z., Zheng, X., Jiang, G., Fang, Y., Mao, X., & Liao, Z. (2013). Starch
derivative-based superabsorbent with integration of water-retaining and controlled-release
fertilizers. Carbohydrate Polymers, 92 (2), 1367-1376.

Figure Captions

Figure 1. Scanning electron micrographs of beads prepared with 2 or 4 hours of

crosslinking time: (a) the entire bead, (b) the external surface and (c) a cross-section are
shown.

Figure 2. FTIR spectra of (a) chitosan, (b) CS100 with tc = 2 h, (c) CS100, tc = 4 h (d)
CS30/ST70 with tc = 2 h, (e) CS30/ST70 with tc = 4 h, (f) CS20/ST80 with tc = 2 h, (g)
CS20/ST80 with tc = 4 h and (h) ST

Figure 3. 13C CP-MAS NMR spectra of chitosan (a), CS100 with tc = 2 h (b) and CS100
with tc = 4 h (c).

Figure 4. X-ray diffractograms of starch (a), chitosan (b), CS100 with tc = 4 h (c), CS100
with tc = 2 h (d), CS30/ST70 with tc = 4 h (e), CS20/ST80 with tc = 4 h (f) and CS20/ST80
with tc = 2 h (g).
Figure 5. TGA curves of chitosan, starch, and CS100 with tc = 2 h, CS100 with tc = 4 h,
CS30/ST70 with tc = 2 h, CS30/ST70 with tc = 4 h, CS20/ST80 with tc = 2 h and
CS20/ST80 with tc = 4 h.

Figure 6. DSC thermograms of chitosan (a), starch (b), CS100 with tc = 2 h (c), CS100 with
tc = 4 h (d), CS30/ST70 with tc = 2 h (e), CS30/ST70 with tc = 4 h (f), CS20/ST80 with tc =
2 h (g) and CS20/ST80 with tc = 4 h (h).

Figure 7. Cumulative percent release of KNO3 as a function of time from matrices prepared
with a tc = 2 h.

Table 1. Soluble Fraction and EDS analysis of CS100, CS30/ST70 and CS20/ST80
prepared hydrogels by different tc.

tc (h) Sample Soluble Fraction % Phosphorous wt %

2 7.2±0.2 3.4±0.3
CS100
4 6.5±0.3 5.3±0.4
2 10.9±0.4 2.0±0.2
CS30/ST70
4 7.5±0.5 2.7±0.3
2 11.2±0.5 1.4±0.2
CS20/ST80
4 9.5±0.4 2.4±0.3
 tc
a) 85 X b) 25000 X c) Core 85 X
(h)

2
CS100

2
CS20/ST80

2
CS30/ST70