JID: JTICE

ARTICLE IN PRESS [m5G;November 5, 2015;14:59]

Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–9

Contents lists available at ScienceDirect

Journal of the Taiwan Institute of Chemical Engineers

journal homepage: www.elsevier.com/locate/jtice

Characterization, swelling and slow-release properties of a new

controlled release fertilizer based on wheat straw cellulose hydrogel
Xiaodi Li a, Qian Li a,∗, Xing Xu a, Yuan Su a,b, Qinyan Yue a, Baoyu Gao a
a
Shandong Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering, Shandong University, 250100 Jinan,
China
b
School of Mathematic and Quantitative Economics, Shandong University of Finance and Economics, 250100 Jinan, China

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

Article history: A novel wheat straw based semi-interpenetrating polymer networks (semi-IPNs) hydrogel with slow-release
Received 17 May 2015 fertilizers of nitrogen and phosphorus was prepared by the method of solution polymerization. The influ-
Revised 4 October 2015
ences of particle sizes, salt solutions, ionic strength and pH changes on the swelling and fertilizer releasing
Accepted 13 October 2015
properties of the product were investigated. Effect of cations on both swelling and fertilizer release properties
Available online xxx
were observed in the order of Na+ > K+ > Ca2+ . The product could hold more water in the pH range of 6–9.
Keywords: The swelling behaviors of the product in various salt solutions, ionic strength and pHs conditions were eval-
Semi-IPNs uated by Schott’s second-order swelling kinetics model. The fertilizers diffusion parameters of the product
Hydrogel could be assessed by a cubic form perfectly according to the diffusion coefficients in the kinetic equation.
Wheat straw cellulose © 2015 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Fertilizer
Slow release
Swelling

1. Introduction
Fertilizer and water are two important limiting factors in increas-
ing demands for bioenergy crops. As a result, improving the utiliza-
tion efficiency of fertilizer nutrients and water resources has become
more and more important on driving an ever-increasing demand for
global agricultural production. Of all the nutrients required by crops,
nitrogen (N) is the one most often deficient in soil. However, due to
the high soluble characteristic of nitrogen fertilizer, it has shown the
greatest potential for losses. Leaching and flooding were normally
considered to be the major N loss pathways. Phosphorus was also an
important constituent element of protoplasm and nucleus which can
stimulate the growth and development of crops. Generally, only 10–
20% of the conventional phosphorus fertilizer applied to soil is ab-
sorbed by crops. Furthermore, the loss of nutrients (N and P) could
directly cause world-wide eutrophication problems in surface fresh-
water bodies and coastal ecosystems. The adoption of slow-release
fertilizers has been considered as a promising strategy to simultane-
ously improve the use efficiency of conventional fertilizer and reduce
the environmental hazards. Compared to the conventional way, the
slow-release fertilizers have many outstanding properties, such as
lessening the fertilizer loss rate, reducing the application frequency,
∗
Corresponding author. Tel.: +86 531 88361337; fax: +86 531 88364513.
E-mail address: qianli@sdu.edu.cn, qianli123sdu@aliyun.com (Q. Li).
providing the sustainable nutrients and lowering the potential nega-
tive effects on the overdosage [1,2].
Hydrogel is a kind of hydrophilic polymers with three-
dimensionally crosslinked structure. This structure is capable to ab-
sorb and hold a large number of water, and the preserved water can
be hardly removed even under certain pressure. Hydrogel could re-
duce the consumption of irrigation water, improve the amount of
fertilizer retained in soil and lower the death rate of plants [3]. As
a result, it has been extensively applied in agriculture and horticul-
ture due to their excellent properties [4]. Semi-interpenetrating poly-
mer networks (semi-IPNs) are a way to blend two polymers, and
the feature of this network is the penetration of some branched or
linear macromolecules onto a molecular dimension of networks [5].
The product generated by this blending method usually displays ex-
ceptional performances superior to either single polymer [6]. In this
study, poly(vinyl alcohol) (PVA) was involved in the semi-IPNs as lin-
ear macromolecule. As a water-soluble polymer, PVA has been widely
applied in biomedical and pharmaceutical field because of its perfor-
mance advantages such as non-toxic, non-carcinogenic, chemical re-
sistance and bioadhesive properties [7]. Besides, PVA is used in the
hydrogel preparation because that it can greatly enhance the me-
chanical toughness properties of hydrogel.
Hydrogels, composed of linear and crosslinked polymers, can be
utilized to enhance the characteristic of polymer composites own-
ing to the structure of semi-IPNs. However, the application of hydro-
gels also faced some problems such as poor biodegradability, high

http://dx.doi.org/10.1016/j.jtice.2015.10.027
1876-1070/© 2015 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Please cite this article as: X. Li et al., Characterization, swelling and slow-release properties of a new controlled release fertilizer based on
wheat straw cellulose hydrogel, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.10.027
JID: JTICE
ARTICLE IN PRESS
2 X. Li et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–9
production cost and weak salt resistance [8,9]. In this work, wheat
straw cellulose (WSC) was introduced to alleviate those problems.
The cellulose grafted hydrogel has gained increasing attention ei-
ther in academic or industrial field owning to its eco-friendly prop-
erty and expected biodegradability [10,11]. Wheat straw (WS), one
typical fiber crops containing 40–60% of natural cellulose, hemicel-
luloses and lignin, has a tremendous amount of readily available
hydroxyl groups. These groups can easily trigger a chain of chemi-
cal reactions (e.g. copolymerization, esterification and etherification),
and therefore the natural WS is suitable to be employed for hydrogels
preparation.
Currently, the research on methods of combining fertilizers with
hydrogels have been reported in other works to obtain composites
with slow-release and water-retention properties, including chem-
ically combined fertilizers [12,13], physically mixed fertilizers and
coated fertilizers [14–19]. However, some of these sustained release
methods only achieve the property of slow releasing fertilizers, but
not the function of water-retention, such as hydrogel coated fertil-
izers [20], polyethylene and paraffin. Therefore, it is important to
exploit an economical fertilizer containing both slow-release and
water-retention properties. Recently, Ni et al. developed a slow-
release nitrogen fertilizer chemically coated by modified cellulose
hydrogel, aiming at controlling the release rate of nutrient and im-
proving the water-holding ability of soil. However, its water swelling
capacity was obviously limited [21]. Wu et al. prepared a fertilizer
by double-coated method, which owned the inner coating (chitosan)
and out coating (P(AA-co-AM)) [19]. But there were two major prob-
lems for this pathway: (i) the nondegradable coating materials would
be left in the soil after the release of fertilizers, thereby resulting in
another environmental problem; (ii) this approach was achieved at
a high production cost. We should also point out that most of these
sustained fertilizers focused only on nitrogen nutrients, and a few re-
ports were concerned about other fertilizers, such as phosphorous
fertilizer. To overcome the above disadvantages, a new preparation
approach was required urgently.
In the present work, a novel semi-IPNs hydrogel with slow-
release fertilizers of nitrogen and phosphorus (WSC-g-PAA/PVA/NP),
on the base of wheat straw cellulose-g-poly (acrylic acid) (WSC-g-
PAA) network and linear poly(vinyl alcohol) (PVA), was prepared by
the method of solution polymerization. To be specific, this prepara-
tion method was developed by transferring nitrogen and phospho-
rus directly into the mixture during the procedure of synthesizing
semi-IPNs hydrogel, which was gained through graft copolymeriza-
tion between WSC and acyclic acid (AA) and the further interpenetra-
tion onto PVA. Fourior-transform infrared (FTIR) and scanning elec-
tron microscope (SEM) were carried out to investigate the surface
chemical functional groups and external morphologies of the semi-
IPNs hydrogel. In order to better understand the product’s swelling,
fertilizer release properties and evaluate its potential application in
soil, the swelling kinetics and fertilizer release behaviors of WSC-g-
PAA/PVA/NP in solutions with various pHs, ions and ionic strengths
were also determined.
2. Experimental
2.1. Materials
The reagents used in this work including urea, acyclic acid
(AA), potassium hydroxide (KOH), dipotassium hydrogen phosphate
(K2 HPO4 ), poly(vinyl alcohol) (PVA), N,N -methylene-bis-acrylamide
(MBA), sodium sulfite (Na2 SO3 ), potassium persulfate (K2 S2 O8 ), and
ammonium cerium nitrate ((NH4)2 Ce(NO3 )6 ) were of analytical
grade (provided by Dengke factory, Tianjin, China). Wheat straw
(WS) was gained from Liaocheng, Shandong, China. Distilled water
was applied in solution preparation, swelling and fertilizer release
experiments.
Please cite this article as: X. Li et al., Characterization, swelling and slow-release properties of a new controlled release fertilizer based on
wheat straw cellulose hydrogel, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.10.027
[m5G;November 5, 2015;14:59]
2.2. Preparation of WSC-g-PAA/PVA/NP
The wheat straw was crushed and sifted through a 100-mesh sieve
after being washed and dried. The powder sample was immersed in
10 wt% ammonia for about two days at a weight ratio of 1:12, and
then the mixture was filtered and washed several times with distilled
water in a vacuum filter. Thereafter, the filtered residue was soaked
in nitric acid (1 mol/L) at 100 °C for 45 min, still maintaining a 1:12
mass ratio. After that, the mixture was filtered and washed via the
same way, and then dried under a temperature of 75 °C to gain wheat
straw cellulose (WSC).
The semi-IPNs WSC-g-PAA/PVA/NP hydrogel based on WSC-g-PAA
network and linear PVA was prepared by a way of solution polymer-
ization. Approximately 1.0 g WSC was placed in a three-neck flask
equipped with a stirrer, and the flask was put in a water bath to make
sure that the whole process was carried out under a constant condi-
tion of 50 °C. About 10 ml K2 S2 O8 (20 g/L) and 4 ml (NH4)2 Ce(NO3 )6
(10 g/L) were transferred into the flask to trigger the reaction, stirring
for 15 min, and then 3.3 ml Na2 SO3 (20 g/L), 2 g urea and 10.0 g AA
with a 75% neutralization degree were added in proper order. After
15 min, during which the graft copolymerization reaction happened
between AA and WSC, 2 ml cross-linker MBA (20 g/L) was added. Fi-
nally, 1.6 g K2 HPO4 and 33.3 ml PVA (60 g/L) were put in after 45 min-
utes. Keeping stirring for additional 4 h, the polymerization reaction
was completed and the resulting product could be obtained after dry-
ing the mixture to a constant mass. The obtained products need to be
milled and sifted through three different screens of 20, 40 and 60
mesh, and then they were stored for further use.
2.3. Characterization
The Fourior-transform infrared (FTIR) of the prepared WSC-g-
PAA/PVA/NP was achieved from a NEXUS-470 series FTIR spectrom-
eter (Thermo nicolet, NEXUS). The pellets employed in FTIR analy-
sis were made by mixing powdered samples with KBr. The variation
of external morphology of samples was probed using a Hitachi S-
520 SEM instrument (Tokyo, Japan), and the particle samples were
oiled by gold under vacuum conditions before observing. The fertil-
izer (N and P) contents of WSC-g-PAA/PVA/NP were examined by an
Energy Dispersive Spectrometer instrument (Elementar Vario EL III,
Germany).
2.4. Study of water absorption and fertilizer release properties
2.4.1. Measurement of swelling and fertilizer release
The samples of 0.20 g was soaked in adequate distilled water for
about 5 h until the gels reached the swelling equilibrium under static
state (at room temperature). Then the swollen gels were separated
from water using a 100-mesh gauze and weighted. The water ab-
sorbency Qeq (g/g) of the samples could be calculated according to
the following equation:
M − M0
Qeq = (1)
M0
where M(g) and M0 (g) imply the mass of swollen and dried gels, re-
spectively.
To determine the release behavior of nitrogen and phosphorus fer-
tilizers from WSC-g-PAA/PVA/NP in water, the following procedures
were conducted: an accurate mass of 1.0 g dried samples were bathed
in 1000 ml distilled water bottled in Erlenmeyer flask equipped with
a stopper under unstirred condition. Approximately 2 ml of immer-
sion solution was drawn out from the flask at various moment to de-
tect the release quantity of fertilizer (nitrogen and phosphorus) using
a UV–visible spectrophotometer. Then, the percentage of fertilizer re-
lease can be calculated by:
0.5 × Ci
Fertilizer release (%) = × 100% (2)
0.2 × C
JID: JTICE
ARTICLE IN PRESS
X. Li et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–9
where Ci (g/L) and C (g/g) represent the contents of nitrogen and
phosphorus in immersion solution and samples, respectively. In this
work, C (nitrogen) = 0.061 g/g; C (phosphorus) = 0.021 g/g.
2.4.2. Swelling kinetics and fertilizer release behaviors
The swelling kinetics of the product in distilled water was car-
ried out as follows: an exact amount of dried samples (0.20 g) were
bathed in 500 ml distilled water in an Erlenmeyer flask. Then, the
swollen gels were leached at predetermined intervals and weighted.
The swelling absorbency of the product at a set timing could be ob-
tained by Eq. (1). Similar steps were carried out to study the fertilizer
release behaviors of WSC-g-PAA/PVA/NP, and the difference was that
2 ml immersion solution was needed to be taken out at set intervals
to determine the amount of fertilizer released.
2.4.3. Swelling and release at various pHs and salt solutions
Various desired pH solutions were prepared using NaOH
(0.1 mol/L) and HCl (0.1 mol/L) aqueous solutions to study the ef-
fects of pH on the swelling behaviors and fertilizer release. The mea-
surement conditions were same as depicted in Section 2.4.1. Four
saline solutions (concentration with 10 mmol/L), including NaCl, KCl,
Na2 SO4 and CaCl2 , were prepared to probe the effects of various
ions on swelling performance and fertilizer release behaviors of the
product.
2.4.4. Effects of particle sizes and ionic strength on swelling and release
As described in Section 2.2, the dried products were milled and
sieved through screens with 20, 40 and 60 mesh, through which par-
ticle samples could be obtained and applied to study the swelling
and fertilizer release behaviors of the product with different particle
sizes. The swelling behaviors of the product in four saline solutions of
NaCl, KCl, Na2 SO4 and CaCl2 with four concentrations of 10, 50, 100
and 500 mmol/L, were determined according to Sections 2.4.1 and
2.4.2. Effects of cations (Na+ , Ca2+ , K+ ) on fertilizer release behaviors
were also studied by soaking the product in solutions of NaCl, KCl and
CaCl2 with four concentrations of 10, 50, 100 and 500 mmol/L.
3. Results and discussion
3.1. FT-IR spectra analysis
The chemical structure differences between WSC and WSC-g-
PAA/PVA/NP [22] were characterized by FTIR and the results are dis-
played in Fig. 1. For the spectrum of WSC, several typical absorption
peaks of cellulose could be noticed, including peaks at 3405 cm−1
(stretching vibration of hydrogen bonded OH), 2918 cm−1 (caused by
methylene), 1650 cm−1 (stretching vibration of C=O of amide I band),
1430 cm−1 (stretching vibration of carbonyl) and 1058 cm−1 (caused
by b-1,4-glycosidic bond) [23,24]. In the curve of WSC-g-PAA/PVA/NP,
the peaks located at 3440 cm−1 and 1625 cm−1 , corresponding to
N-H and C=O stretching in -CONH2 , respectively; 1671 cm−1 , corre-
sponding to the carbonyl moiety in acrylamide unit [14]; 1167 cm−1 ,
ascribed to the bending vibration of N-H; these indicated that there
were acrylamides generated by the condensation reaction between
acrylic acid and urea. The peaks at 819 cm−1 and 1109 cm−1 were at-
tributed to the stretching vibration of P-O and P=O, respectively [25],
implying that the dipotassium hydrogen phosphate existed in the
product by its own form. Additionally, peaks situated at 1559 cm−1
and 1385 cm−1 , corresponding to the -COO− groups and the stretch-
ing of C-O-C in AA, respectively [23], suggesting that -COOH groups
of AA had partially transformed to -COO− groups [26]. By compari-
son, the peak situated at 897 cm−1 was not observed from the spec-
trum curve of WSC-g-PAA/PVA/NP, ascribed to the process of cross-
linking. Moreover, compared with the spectrum of WSC, the larger
wavenumber located at 2923 cm−1 (C-H stretching in acrylate unit)
[14], a weaker peak at 560 cm−1 , and a bending vibration observed at
Please cite this article as: X. Li et al., Characterization, swelling and slow-release properties of a new controlled release fertilizer based on
wheat straw cellulose hydrogel, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.10.027
[m5G;November 5, 2015;14:59]
3
531 cm−1 all demonstrated that during the synthesis process, a graft
copolymerization process occurred between hydroxyl groups of AA
and WSC. By the analysis of the curves displayed in Fig. 1, it could be
concluded that during the chemosynthesis, AA reacted with WSC by
graft copolymerization and the copolymer was combined with PVA
by semi-IPNs technology. Moreover, nitrogen and phosphorus fertil-
izer were integrated with the hydrogel by chemically reacting and
physically entrapping, respectively, endowing the resulting product
with function of controlled-release fertilizer. In conclusion, the ex-
pected product, a combination of hydrogel and fertilizer, was well
prepared.
3.2. SEM analysis
SEM, a widely applied technique, was utilized to detect and
determine the different surface morphology of WSC and WSC-g-
PAA/PVA/NP, and the results can be seen in Fig. 2. Compared with
the smooth surface and dense structure of WSC, WSC-g-PAA/PVA/NP
owns a coarse and porous surface, facilitating the penetration of
water into the network of the product and making the WSC-g-
PAA/PVA/NP own the properties of absorbing and holding water.
3.3. Swelling kinetics
The swelling behaviors of the product in various conditions were
displayed in Figs. 3(a), 4(a) and 5. As can be seen, the characteristics
of swelling curves could be concluded as following: a fast swelling
rate at first 60 min and then followed by a nearly equilibrium state.
In this section, Schott’s second-order swelling kinetics model was in-
troduced to explore the swelling behavior of the product. The model
can be indicated by the following equation:
t/Qt = 1/Kis + 1/Q∞ t (3)
In Eq. (3), Qt (g/g) and Q∞ (g/g) represented the swelling capacity
at any time t (s) and the theoretical equilibrium value, respectively;
Kis (g/g s) was a constant, indicating the initial swelling rate. The plot
of t/Qt versus t could give a straight line, and the linear correlation co-
efficients (Table 1) were pretty good, indicating that Schott’s second-
order swelling kinetics model could be adopted to assess the kinetic
swelling behaviors of the WSC-g-PAA/PVA/NP. Moreover, these fitted
straight lines implied the slope and intercept, through which Q∞ and
Kis can be calculated.
3.4. Release kinetics
Figs. 3(b), (c), 4(b), (c) and Fig. 6 display the release curves of phos-
phorus and nitrogen fertilizer from the product in various conditions.
As can be seen, the release curves of nitrogen and phosphorus exhib-
ited an apparent characteristic: a fast release rate at first and then
followed by a rapidly reducing release rate, which was a representa-
tive first-order release behavior. According to the theoretical effort of
Shavit et al. [27], a kinetic equation in a cubic form had been adopted
to illustrate the relationship between the diffusion and control re-
lease action of fertilizers toward a porous material (Mt /M0 < 60%):
Mt
ft = = 3Kt 1/2 − 3(Kt 1/2 )2 + (Kt 1/2 )3 (4)
M0
where ft is the fractional number of fertilizer released at time t, K rep-
resents the release rate constant. This equation can also be converted
into a linear form as following:
1 1
(1 − ft ) 3 = 1 − Kt 2 (5)
1 1
A straight line could be achieved by plotting (1 − ft ) 3 versus t 2 ,
and K was also obtained through the slope of the line.
JID: JTICE
ARTICLE IN PRESS [m5G;November 5, 2015;14:59]

4 X. Li et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–9

Fig. 1. FTIR spectra of WSC and WSC-g-PAA/PVA/NP.

Fig. 2. SEM of WSC (a) and WSC-g-PAA/PVA/NP (b).

Besides, K is defined as the following equation: of diffusion coefficient by the obtained release rate constant K:

1   12 (KC0 )2
D= (7)
K= × r0 D(2C0 − CS )CS (6) r0 (2C0 − CS )CS
2
C0
3.5. Effect of particle size on swelling and fertilizer release behaviors
where C0 represents the fertilizer weight per unit volume of sample;
Cs indicates the fertilizer solubility at equilibrium in dissolution liq- The swelling behaviors and water retention of WSC-g-
uid; r0 is the radius of granular sample; and D implies the fertilizer PAA/PVA/NP semi-IPNs hydrogel in different particle sizes are
diffusion coefficient in swollen sample. Based on the theoretical ef- displayed in Fig. 3. It could be seen from Fig. 3(a) that the water ab-
fort of Shavit et al., diffusion coefficient D contains the actions of two sorption of WSC-g-PAA/PVA/NP increased fast within initial 60 min
factors: porosity and tortuosity. It was possible to calculate the value and then almost reached equilibrium. Besides, the swelling rates of

Please cite this article as: X. Li et al., Characterization, swelling and slow-release properties of a new controlled release fertilizer based on
wheat straw cellulose hydrogel, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.10.027
JID: JTICE
ARTICLE IN PRESS [m5G;November 5, 2015;14:59]

X. Li et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–9 5

a b c
200 80 70

180
70
60
160

Phosphorus fertilizer released (%)

Nitrogen fertilizer released (%)

60
140
50
120 50
Qt (g/g)

100 10-20mesh 40
10-20mesh 40 10-20mesh
20-40mesh 20-40mesh
80 20-40mesh
40-60mesh 40-60mesh
40-60mesh
30 30
60

40 20
20
20
10
0 10
0 50 100 150 200 250 0 20 40 60 80 100 120 0 20 40 60 80 100 120
time (min) t (min) t (min)

Fig. 3. Effect of particle size on swelling behaviors (a), phosphorus fertilizer release behaviors (b), and nitrogen fertilizer release behaviors (c).

a b c
70

120 70 65

60
Phosphorus fertilizer released (%)

60

Nitrogen fertilizer released (%)

100
55

50
50
80 NaCl
NaCl 45 KCl
Qt (g/g)

KCl CaCl2
NaCl 40 40
60 CaCl2
KCl
Na2SO4 35
30
CaCl2 30
40
25
20
20 20

10 15
0 50 100 150 200 250 0 20 40 60 80 100 120 0 20 40 60 80 100 120
time (min) t (min) t (min)

Fig. 4. Effect of various salt solutions on swelling behaviors (a), phosphorus fertilizer release behaviors (b), and nitrogen fertilizer release behaviors (c).

Table 1
Swelling kinetic parameters and diffusion coefficients of fertilizers for WSC-g-PAA/PVA/NP in different conditions.

Conditions Q∞ Kis R2 Diffusion coefficient (×10−5 )

(g/g) (g/g s)
Nitrogen (cm2 /s) Phosphorus (cm2 /s)

Salt solutions NaCl 123.92 0.3985 0.9973 4.47272 4.37805

KCl 103.73 0.2124 0.9987 4.42555 4.31076
CaCl2 – – – 1.73057 3.13461
Na2 SO4 83.06 0.1518 0.9989 – –
Particle size (mesh) 10–20 183.15 0.3983 0.9914 1.21681 3.04834
20–40 228.83 0.5112 0.9849 1.33895 3.64364
40–60 243.31 0.5241 0.9996 3.2938 3.88832
NaCl concentration (mM) 10 123.92 0.3985 0.9973 2.01071 4.88426
50 66.98 0.4116 0.9990 2.68261 5.30878
100 58.96 0.2259 0.9965 2.85624 6.98786
500 32.54 0.1077 0.9934 4.47272 8.48744

small particles had an apparent raise than those of larger ones, and ascribed to the larger surface area owned by smaller particle, which
the water holding capacity showed a similar trend, namely that the could adsorb more water [28]. In addition, some space between
smaller ones could hold a litter more water than larger ones. Data the swollen particles existed when these particles were in contact
displayed in Table 1 also indicated that the values of swelling capac- with each other. The space named free or inter-stitial volume [29]
ity at equilibrium (Q∞ ) and initial swelling rate (Kis ) had an obvious could be attached with additional water through a simple capillary
improvement with the reducing of particle size. This result could be action [30]. Consequently, smaller particles of WSC-g-PAA/PVA/NP

Please cite this article as: X. Li et al., Characterization, swelling and slow-release properties of a new controlled release fertilizer based on
wheat straw cellulose hydrogel, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.10.027
JID: JTICE
ARTICLE IN PRESS [m5G;November 5, 2015;14:59]

6 X. Li et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–9

a b
105
120
90
100 10mM 10mM
50mM 50mM
75
100mM 100mM
80 500mM
500mM

Qt (g/g)
60
Qt(g/g)

60
45
40
30
20
15
0
0 50 100 150 200 250 0 50 100 150 200 250
time (min) time (min)
c d

80

70 10mM
50mM
60 100mM
500mM
50
Qt (g/g)

40

30

20

10
0 50 100 150 200 250
time (min)

Fig. 5. Effect of different ionic strengths on swelling: (a) NaCl solution, (b) KCl solution, (c) Na2 SO4 solution, (d) CaCl2 solution.

brought about bigger interstitial volume, faster initial swelling rate
and higher water absorbency.
The similar phenomenon was also presented in fertilizer release
of WSC-g-PAA/PVA/NP in distilled water. A static fertilizer release ex-
periment was carried out, and the results as well as the related pa-
rameters are shown in Fig. 3(b) and (c) and Table 1. As depicted in
Section 2.4.4, the resin samples with three different particle sizes
(10–20 mesh, 20–40 mesh and 40–60 mesh) were obtained by sieving
through sieves with different mesh sizes. Results in Table 1 indicated
that the fertilizer in hydrogel with smaller size released faster than
the larger ones; the total fertilizer release amount of smaller samples
(40–60 mesh) was also higher than those of larger ones. It is due to
that the samples with smaller size possessed both greater interstitial
volume and interfacial area between swollen gel and water.
3.6. Effect of various salt solutions on swelling and fertilizer release
behaviors
In order to investigate the effects of cations and anions (e.g.
Na+ , Ca2+ , K+ , Cl− and SO2−
4 ) on the swelling and fertilizer release
of WSC-g-PAA/PVA/NP, the swelling kinetics and fertilizer release
behaviors of the product were probed in several saline solutions
(Na2 SO4 , NaCl, KCl, and CaCl2 ) with a 10 mmol/L concentration
(Fig. 4). The swelling curves of the product shown in Fig. 4(a)
demonstrated that the existing of additional ions could obviously
lead to a slower swelling rate and a lower swelling capacity. The data
displayed in Table 1 also demonstrated that Q∞ and Kis in distilled
water were apparently higher than those in saline solutions, this was
a phenomenon commonly noticed in the static swelling experiment
of ionic hydrogels [31]. The reason could be ascribed to a charge
screening effect, which was caused by the additional cations. Mean-
while, the charge screening effect would also result in a non-perfect
electrostatic repulsion between anions. Besides, the osmotic pressure
gradient between the internal network and the external solution also
decreased due to the presence of ions. Conversely, data displayed
in Table 1 suggested that the existed cations can result in a greater
fertilizer diffusion coefficient D; this means that cations in water can
accelerate the release of fertilizers by competing adsorption sites,
competitive complexing, etc.
As shown in Fig .4(a), the water absorbency in NaCl, KCl, and
Na2 SO4 solutions increased with the prolongation of soaking time,
and then reached a swelling equilibrium within 45 min. But the wa-
ter absorbency in CaCl2 solution reached a peak and then decreased
to a certain value. The swelling capacities in four saline solution could
be ordered as NaCl > KCl > Na2 SO4 > CaCl2 and the cations acted on
the water absorbency in the order of Ca2+ < K+ < Na+ . The order of

Please cite this article as: X. Li et al., Characterization, swelling and slow-release properties of a new controlled release fertilizer based on
wheat straw cellulose hydrogel, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.10.027
JID: JTICE
ARTICLE IN PRESS [m5G;November 5, 2015;14:59]

X. Li et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–9 7

a1 a2 a3
75 75
70 70 3
70
65
65 65
60
Phosphorus fertilizer released (%)

Phosphorus fertilizer released (%)

Phosphorus fertilizer released (%)

55 500mM 60 60
50 100mM
55
45 50mM 55 500mM
40 10mM 50 500mM 100mM
100mM 50mM
35 45 50
50mM 10mM
30
40 10mM 45
25
20 35
40
15 30
10 35
25
5
0 20 30
0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120
1
t (min) 2 t (min) 3 t (min)
b b b
75 75 75
70 70 70 3
65 65 65
60 60
Nitrogen fertilizer released (%)
Nitrogen fertilizer released (%)

Nitrogen fertilizer released (%)

60
55 55
55
50 50 500mM
100mM 50
45 500mM 45 500mM
50mM
100MM 45 100mM
40 40 10mM
50mM 50mM
10mM 40 10mM
35 35
35
30 30
25 25 30

20 20 25

15 15 20
0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120
t (min) t (min) t (min)

Fig. 6. Effect of different ionic strengths on fertilizer release behaviors (a1 , b1 ) CaCl2 solutiona; (a2 ,b2 ) KCl solution; and (a3 ,b3 ) NaCl solution).

bivalent < monovalent cations could be ascribed to the formation of less than that of monovalent anion (Cl− ). This might be resulted from
complexes between divalent ion and carboxylate groups on WSC-g- the larger ionic strengths of SO2−4 , which brought a greater osmotic
PAA/PVA/NP hydrogel [32]. Furthermore, the carboxylate groups pos- pressure to the external solutions.
sessed some certain complexing abilities, inducing the production of As shown in Fig. 4(b) and (c) and Table 1, the total release amount
intramolecular and intermolecular complexes by reacting with bi- and the release rate of fertilizer in saline solution were ordered as
valent cations and therefore leading to less water penetrated into NaCl > KCl > CaCl2 and cations acted on the release of fertilizer in the
the internal networks of the product [33]. The particular tendency order of Ca2+ < K+ < Na+ . The phenomenon of bivalent < monova-
of swelling curve in CaCl2 solution mainly due to the more Cl− ions lent cations was similar to previous explanation on water absorbency.
existed in CaCl2 solution, which could cause a lower osmotic pressure Ca2+ ions entered the interior of gels with the water and then inter-
difference and the consequent smaller water absorbency. In addition, acted with the carboxyl groups, resulting in a denser network struc-
water absorbency observed in the swelling curve finally decreased to ture and the consequent slower release rate. The order of K+ < Na+
20 g/g; this low value phenomenon was practically understood as an was dependent on that K+ had a larger ion radius, which made it
overshooting effect. To be specific, the extending of polymer network harder for K+ ions to enter the interior network of gels. Consequently,
in CaCl2 solution mainly depended on the water penetration. How- the effects of competing adsorption sites on gels were stronger in
ever, the interaction of Ca2+ ions and carboxyl groups on gels led to a NaCl solutions, resulting in the faster release rate of phosphate and
greater cross-link degree, and thus caused a lower swelling capacity. nitrogen compounds.
Meanwhile, the special swelling curve also resulted in the inappli-
cability of pseudo second-order swelling kinetics model to evaluate 3.7. Effect of different ionic strength on swelling and fertilizer release
the swelling capacity and swelling rate of the product in CaCl2 so- behaviors
lution. In terms of monovalent cations, the swelling ratio abided by
the order of K+ < Na+ , which was mainly attributed to the difficulty The swelling curves of WSC-g-PAA/PVA/NP semi-IPNs hydro-
for ions with larger size to penetrate into the internal semi-IPNs of gel as well as the related parameters (Q∞ and Kis ) were displayed
the samples. Considering only on the anions (Table 1), it was obvious Fig. 5 and Table 1. The gel in solutions with low concentration swelled
that the influence of divalent anions (SO2−
4 ) on water absorbency was faster and held much more water compared to that with higher

Please cite this article as: X. Li et al., Characterization, swelling and slow-release properties of a new controlled release fertilizer based on
wheat straw cellulose hydrogel, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.10.027
JID: JTICE
ARTICLE IN PRESS
8 X. Li et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–9
a b
200
180
160
Phosphorus fertilizer release (%)
140
120
Qeq (g/g)
100
80
60
40
20
0 62
0 2 4 6 8 10 12
pH
Fig. 7. Effect of pH on swelling behaviors (a), phosphorus fertilizer release behaviors (b), and nitrogen fertilizer release behaviors (c).
solution concentration. Several factors might contribute to the water
absorbency, including the interaction of hydrophilic groups existed
in samples, the repulsion effect between these groups, and osmotic
pressure difference from the internal and external of samples. Of
all these factors, osmotic pressure difference might be the primary
one, which became weaker with the increased concentrations of
immersion solutions.
Donnan equilibrium theory could be adopted to assess the net
osmotic pressure (π ion ), and it could be indicated by following
equation:
 both the negative effect of H+ on electrostatic repulsive force and
πion = RT (Ci g − Ci s )
i
where “g” and “s” mean the gel and soaking solutions, respectively;
Ci means the concentration of movable ion of species i. In general, the
ion concentration (Ci g ) in the gel network is stable, so the increase of
Ci s (the concentration of saline solution) would undoubtedly lead to
the decrease of osmotic pressure (π ion ). The diffusion of water from
external soaking solution into polymeric network was the major rate
limiting factor at the initial stage of swelling. The osmotic pressure
difference just acted as a driving force at this stage. Therefore, higher
concentration of immersion solution would lead to a lower swelling
rate.
Results shown in Fig. 6 indicated that the release rate of nitro-
gen and phosphorus fertilizer was increased with the increase in ions
concentration. This phenomenon could be ascribed to the fact that
more cations penetrated into the polymeric network of gels at higher
ions concentration. These cations would be adsorbed onto the gels,
resulting in the so-called competing adsorption phenomena between
cations and phosphate, nitrogen compounds. As a result, phosphate
and nitrogen compounds were easier to be desorbed and released out
of gels in immersion solution with more cations. However, a particu-
lar tendency of phosphorus fertilizer release curves in CaCl2 solution
was observed when the concentration of Ca2+ reached 500 mM; the
release rate of phosphorus fertilizer did not keep increasing but de-
clined instead. This special tendency could be attributed to the reac-
tion between excessive Ca2+ and the released phosphate in solution.
3.8. Effect of pH on swelling and fertilizer release behaviors
The WSC-g-PAA/PVA/NP contains amounts of carboxyl groups, the
protonation extent of which mainly rely on the changes of pHs of
soaking medium. Therefore, the water absorbency is closely linked to
the pH values of the medium. The equilibrium water absorbency kept
Please cite this article as: X. Li et al., Characterization, swelling and slow-release properties of a new controlled release fertilizer based on
wheat straw cellulose hydrogel, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.10.027
[m5G;November 5, 2015;14:59]
c
74
80
72 78
Nitrogen fertilizer release (%)
76
70
74
68
72
66 70
68
64
66
64
2 4 6 8 10 12 2 4 6 8 10 12
pH pH
raising until pH reached 7, and then a decline appeared when pH was
higher than 7 (Fig. 7a); this demonstrated that WSC-g-PAA/PVA/NP
was stimuli-responsive to the pH. This pH stimuli-responsive prop-
erty was mainly attributed to the protonation of carboxylate anions
under an acidic circumstance, which could lead to the disappearance
of anion-anion repulsion. Furthermore, the hydrogen-bonding inter-
action among hydrophilic groups was also reinforced because of the
protonation. At the same time the physical crosslinking degree of
skeleton network was also heightened, resulting from the relatively
lower water holding capacities. With the sustained increasing pH,
(8)
the hydrogen-bonding interaction among carboxylate groups were
weakened by degrees and as a result more water could be absorbed
by WSC-g-PAA/PVA/NP. However, excessive Na+ in the immersion so-
lution could screen the charges of −COO− groups (so called “charge
shielding effect”). This process would weaken the electrostatic repul-
sion, consequently, the swelling capacities declined with increasing
pH from 7 to 12 [34].
From the curves displayed in Fig. 7(b) and (c), it can be seen that
the release curves of phosphorus and nitrogen fertilizers displayed
totally opposite tendencies. With the rising of pH from 2 to 12, the
release amount of phosphorus firstly increased and then decreased,
instead of which the release amount of nitrogen firstly decreased
and then increased. Under strongly acidic circumstances, the groups
of -COO− on WSC-g-PAA/PVA/NP transformed into -COOH, leading
to the decreased anion–anion repulsion between carboxyl and PO3− 4 .
Besides, the change from −COO− to −COOH would also bring about
the reinforcing of hydrogen-bonding interaction, which was not con-
ducive to the release of phosphate. On the contrary, the acrylamide
in WSC-g-PAA/PVA/NP was prone to react with H+ at low pH, and
then produce NH4 Cl. Thereby, the amount of nitrogen in solution was
increased. When the solution was in an alkaline conditions, the react
between sodium ions and carboxyl groups would be induced, result-
ing in the reduce of electrostatic repulsion between carboxyl and
PO3−
4 . This would lead to the lower release amount of phosphorus.
Instead, the total release amount of nitrogen at high pH was higher,
mainly due to the yielding of NH3 by the react between acrylamide
and OH— .
4. Conclusions
Through the method of solution polymerization, a novel slow-
release fertilizer based on wheat straw semi-IPNs hydrogel was
successfully prepared. FT-IR spectroscopy and SEM indicated that AA
JID: JTICE
ARTICLE IN PRESS
X. Li et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2015) 1–9
was successfully grafted onto the wheat straw, and PVA was involved
by semi-interpenetrating technologies. Additionally, the nitrogen
and phosphorus fertilizers were also combined with semi-IPNs
hydrogel as expected. The swelling and fertilizer release behaviors
of the product were affected by particle size, salt solutions, ionic
strength and various pHs. Schott’s second-order swelling kinetics
model could well evaluate the dynamic state for the swelling of
gels in various conditions. The results showed that the product own
a better swelling capacity in neutral condition, and the effects of
cations and anions on swelling abide by the order of Na+ > K+ > Ca2+
and Cl− > SO2− 4 , respectively. The diffusion coefficients of fertilizers
indicated that the actions of cations on fertilizer release was same
with that on swelling. Additionally, in neutral condition, nitrogen nu-
trient release slower and phosphorus release faster, compared with
conditions of strong acid or strong alkaline. The diffusion coefficients
of the product in different sizes suggested that the utilizing of WSC-
g-PAA/PVA/NP in larger size could achieve a better controlled-release
effect, just considering on lessening leaching loss. All these conclu-
sions could lay foundation for the application of WSC-g-PAA/PVA/NP
in agriculture. Researches on swelling and fertilizer release behaviors
of WSC-g-PAA/PVA/NP in soil will be done in the future.
Acknowledgments
The authors are grateful for financial support from National Nat-
ural Science Foundation of China (21477064), Natural Science Foun-
dation of Shandong Province (ZR2014EEM042), Universities Innova-
tion Projects of Jinan City (20130380) and Young Scholars Program
of Shandong University (2015WLJH34). This work was also supported
by grants from Tai Shan Scholar Foundation.
References
[1] Zhong K, Lin ZT, Zheng XL, Jiang GB, Fang YS, Mao XY, Liao ZW. Starch derivative-
based superabsorbent with integration of water-retaining and controlled-release
fertilizers. Carbohydr Polym 2013;92(2):1367–76.
[2] Guo MY, Liu MZ, Liang R, Niu AZ. Granular urea-formaldehyde slow-release
fertilizer with superabsorbent and moisture preservation. J Appl Polym Sci
2006;99:3230–5.
[3] Seetapan N, Wongsawaeng J, Kiatkamjornwong S. Gel strength and swelling
of acrylamide–protic acid superabsorbent copolymers. Polym Adv Technol
2011;22(12):1685–95.
[4] Thakur VK, Thakur MK, Gupta RK. Graft copolymers from cellulose: synthesis,
characterization and evaluation. Carbohydr Polym 2013;97(1):18–25.
[5] Bhattacharyya R, Ray SK. Enhanced adsorption of synthetic dyes from aqueous
solution by a semi-interpenetrating network hydrogel based on starch. J Ind Eng
Chem 2014;20:3714–25.
[6] Wang WB, Huang DJ, Kang YR, Wang AQ. One-step in situ fabrication of a granular
semi-IPN hydrogel based on chitosan and gelatin for fast and efficient adsorption
of Cu2+ ion. Colloids Surf B: Biointerfaces 2013;106:51–9.
[7] Roberts MJ, Bently MD, Harris JM. Chemistry for peptide and protein pegylation.
Adv Drug Delivery Rev 2002;54:459–76.
[8] Farag S, Al-Afaleq EI. Preparation and characterization of saponified delignified
cellulose polyacrylonitrile-graft copolymer. Carbohydr Polym 2002;48:1–5.
[9] Shogren RL, Willett JL, Biswas A. HRP-mediated synthesis of starch–
polyacrylamide graft copolymers. Carbohydr Polym 2009;75:189–91.
Please cite this article as: X. Li et al., Characterization, swelling and slow-release properties of a new controlled release fertilizer based on
wheat straw cellulose hydrogel, Journal of the Taiwan Institute of Chemical Engineers (2015), http://dx.doi.org/10.1016/j.jtice.2015.10.027
[m5G;November 5, 2015;14:59]
9
[10] Kono H, Fujita S. Biodegradable superabsorbent hydrogels derived from cel-
lulose by esterification crosslinking with 1,2,3,4-butanetetracarboxylic dianhy-
dride. Carbohydr Polym 2012;87(4):2582–8.
[11] Wang J, Zhou X, Xiao H. Structure and properties of cellulose/poly(N-
isopropylacrylamide) hydrogels prepared by SIPN strategy. Carbohydr Polym
2013;94(2):749–54.
[12] Jarosiewicz A, Tomaszewska M. Controlled-release NPK fertilizer encapsulated by
polymeric membranes. J Agric Food Chem 2003;51:413–17.
[13] Chandra PK, Ghosh K, Varadachari C. A new slow-releasing iron fertilizer. Chem
Eng J 2009;155:451–6.
[14] Ni BL, Liu MZ, Lü SY. Multifunctional slow-release urea fertilizer from ethyl cellu-
lose and super absorbent coated formulations. Chem Eng J 2009;155:892–8.
[15] Mathews AS, Narine S. Poly[N-isopropyl acrylamide]-co-polyurethane copoly-
mers for controlled release of urea. J Polym Sci Part A: Polym Chem
2010;48:3236–43.
[16] Teodorescu M, Lungu A, Stanescu PO, Neamtu C. Preparation and properties of
novel slow-release NPK agrochemical formulations based on poly(acrylic acid)
hydrogels and liquid fertilizers. Ind Eng Chem Res 2009;48:6527–34.
[17] Wu L, Liu M, Liang R. Preparation and properties of a double-coated slow re-
lease NPK compound fertilizer with superabsorbent and water-retention. Biore-
sour Technol 2008;99:547–54.
[18] Liang R, Yuan H, Xi G, Zhou Q. Synthesis of wheat straw-g-poly(acrylic acid) super-
absorbent composites and release of urea from it. Carbohydr Polym 2009;77:181–
7.
[19] Wu L, Liu M. Preparation and properties of chitosan-coated NPK compound fertil-
izer with controlled-release and water-retention. Carbohydr Polym 2008;72:240–
7.
[20] Al-Zahrani SM. Utilization of polyethylene and paraffin waxes as controlled deliv-
ery systems for different fertilizers. Ind Eng Chem Res 2000;39:367–71.
[21] Ni B, Liu M, Lu S, Xie L, Wang Y. Multifunctional slow-release organic–inorganic
compound fertilizer. J Agric Food Chem 2011;59:10169–75.
[22] Li XD, Li Q, Su Y, Yue QY, Gao BY, Su Y. A novel wheat straw cellulose-based semi-
IPNs superabsorbent with integration of water-retaining and controlled-release
fertilizers. J Taiwan Inst Chem Eng 2015;55:170–9.
[23] Liu J, Li Q, Su Y, Yue QY, Gao BY, Wang R. Synthesis of wheat straw cellulose-g-
poly (potassium acrylate)/PVA semi-IPNs superabsorbent resin. Carbohydr Polym
2013;94:539–46.
[24] Ma ZH, Li Q, Yue QY, Gao BY, Xu X, Zhong QQ. Synthesis and characterization of a
novel super-absorbent based on wheat straw. Bioresour Technol 2011;102:2853–
8.
[25] Pourjavadi A, Doulabi M, Soleyman R, Sharif S, Eghtesadi SA. Synthesis and char-
acterization of a novel (salep phosphate)-based hydrogel as a carrier matrix for
fertilizer release. React Funct Polym 2012;72:667–72.
[26] Zheng Y, Liu Y, Wang AQ. Fast removal of ammonium ion using a hydrogel opti-
mized with response surface methodology. Chem Eng J 2011;171:1201–8.
[27] Shavit U, Shaviv A, Shalit G, Zaslavsky D. Release characteristics of a new con-
trolled release fertilizer. J Contr Release 1997;43:131–8.
[28] Omidiana H, Hashemi SA, Sammes PG, Meldrum I. Modified acrylic based su-
perabsorbent polymers (dependence on particle size and salin-ity). Polymer
1999;40:1753–61.
[29] Bao Y, Ma JZ, Li N. Synthesis and swelling behaviors of sodium carboxymethyl
cellulose-g-poly(AA-co-AM-co-AMPS)/MMT superabsorbent hydrogel. Carbo-
hydr Polym 2011;84:76–82.
[30] Lipponen K, Liu Y, Stege PW, Öörni K, Kovanen PT, Riekkola M. Capillary elec-
tro chromatography and quartz crystal microbalance, valuable techniques in the
study of heparin–lipoprotein interactions. Anal Biochem 2012;424:71–8.
[31] Zhao Y, Su HJ, Fang L, Tan TW. Superabsorbent hydrogels from poly (as par-
ticacid) with salt-, temperature- and pH-responsiveness properties. Polymer
2005;46:5368–76.
[32] Zhang JP, Chen H, Wang AQ. Study on superabsorbent composite. IV.
Effects of organification degree of attapulgite on swelling behaviors of
polyacrylamide/organo-attapulgite composite. Eur Polym J 2006;42:101–8.
[33] Zhang J, Wang A. Study on superabsorbent composites. IX: synthesis, characteri-
zation and swelling behaviors of polyacrylamide/clay composites based on vari-
ous clays. React Funct Polym 2007;67:737–45.
[34] Lanthong P, Nuisin R, Kiatkamjornwong S. Graft copolymerization, characteri-
zation, and degradation of cassava starch-g-acrylamide/itaconic acid superab-
sorbents. Carbohydr Polym 2006;66(2):229–45.