Separation and Purification Technology xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Separation and Purification Technology

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

Entrapment of micro-sized zeolites in porous hydrogels: Strategy to

overcome drawbacks of zeolite particles and beads for adsorption of
ammonium ions
⁎
Rio Nugraha Putra, Young Haeng Lee
Center for Water Resource Cycle Research, Korea Institute of Science and Technology, Seoul 02792, South Korea
Division of Energy and Environment Technology, KIST School, Korea University of Science and Technology (UST), Seoul 02792, South Korea

A R T I C LE I N FO A B S T R A C T

Keywords: We have demonstrated that the entrapment of micro-sized zeolite particles (ZPs) in porous hydrogels is a pro-
Micro-sized zeolite particle mising strategy toward overcoming several drawbacks of ZPs and zeolite beads (ZBs) for adsorption of ammo-
Hydrogel bead nium ions. The polyvinyl alcohol (PVA)-alginate-ZPs (PAZ) composite hydrogel beads with the same size as that
Entrapment of ZBs were prepared by physically entrapping the ZPs. On the basis of structural characterization results, the
Adsorption capacity
entrapped ZPs are well distributed with minimal agglomeration in the highly permeable and porous hydrogels.
Adsorption mechanism
Ammonium ion
Unlike the ZPs, the PAZ hydrogels had good settleability and showed no significant further agglomeration of the
entrapped ZPs even under high zeolite dosage conditions. Due to still high accessible surface area of the en-
trapped ZPs in the hydrogel, the maximum ammonium adsorption capacities of PAZ hydrogels from Langmuir
isotherm were 3.4–4.3 times higher than those of ZBs. Thus, we propose that the entrapment of ZPs in highly
porous hydrogel has much less adverse effects on the adsorption capacity, compared to the binder commonly
used in the synthetic ZBs. Similar to the ZPs and unlike the ZBs, the ammonium adsorption process in the PAZ
hydrogels was governed by chemical ion exchange mechanism and tended to be rate-limited by boundary layer
diffusion and mainly intra-particle diffusion. Owing to the above-mentioned key features toward overcoming
drawbacks of both the ZPs and the ZBs, the PAZ hydrogels have great potential as promising alternative zeolite-
type adsorbents for a broad range of industrial applications in water purification and wastewater treatment.

1. Introduction channel structure and a maximum cavity size of 0.41 × 0.41 nm, make
this zeolite widely applied in adsorption and ion exchange processes
Zeolite is a commonly used adsorbent in batch or column adsorption [53].
process for the removal of ammonium ions which can cause environ- Although zeolite particles (e.g., micro-sized zeolites) have a high
mental problems such as eutrophication and toxicity to aquatic life, adsorption capacity due to their highly available surface area [24], the
because its ion exchange process is fast, simple, and less dependent on usage of micro-sized zeolite particles (ZP) has been limited due to long
environmental conditions [3,12,23,34,42,50,54]. Zeolites are micro- settling time [14], difficulty to be separated from the solution for re-
porous crystalline aluminosilicates, composed of AlO4 and SiO4 tetra- cycling [18], and high pressure drop in column adsorption process
hedral with O atoms connecting neighboring tetrahedral. When Al is [7,32]. Moreover, a high dosage of ZPs could impose particle interac-
incorporated into silica framework, the positive charge on the Al makes tions, which resulted in the agglomeration and overcrowding of ZPs.
the framework negatively charges, thus require the presence of extra Such agglomeration would lead to a significant decrease in the acces-
framework cations within the structure to keep the overall framework sible surface area, lowering the adsorption capacity [40]. It was ob-
neutral. This extra framework cations are ion exchangeable [3]. Among served that zeolite particles formed cluster when the size was reduced
many types of zeolites which can be classified according to their below 100 μm, which was caused by an increase in static charge on the
structural properties, single-phase zeolite 4A has been of great interest surface [32]. This cluster of micro-sized ZPs could eventually reduce
because of its high cation exchange capacity and large number of the available adsorptive sites, adversely affecting the adsorption effi-
acidity sites. These properties resulted from its 8-ring, 6-ring, and 4-ring ciency.

⁎
Corresponding author.
E-mail address: [email protected] (Y.H. Lee).

https://doi.org/10.1016/j.seppur.2019.116351
Received 12 June 2019; Received in revised form 23 November 2019; Accepted 24 November 2019
1383-5866/ © 2019 Elsevier B.V. All rights reserved.

Please cite this article as: Rio Nugraha Putra and Young Haeng Lee, Separation and Purification Technology,
https://doi.org/10.1016/j.seppur.2019.116351
R.N. Putra and Y.H. Lee Separation and Purification Technology xxx (xxxx) xxxx

Because of these drawbacks of ZPs, granulated zeolite beads (ZBs) 2.3. Batch adsorption experiments
which are commonly shaped using a binder, glue, and filler [35], have
been generally applied in wastewater treatment [6,26]. One of the most In the literature, various concentration of ammonium nitrogen was
noteworthy drawbacks of ZBs is a significant reduction of the adsorp- reported with typical concentration range of ammonium nitrogen from
tion capacity mainly due to the much low accessible surface area 20 mg-N L−1 in the domestic wastewater [28] to 400 mg-N L−1 in the
[4,33], which has prompted a growing research interest in the pro- agricultural wastewater [29]. Thus, in this study, the adsorption ex-
duction of bead-sized zeolite-type adsorbents with less adverse effect on periments were conducted under various initial ammonium nitrogen
the adsorption capacity especially for the application concerning was- concentration in the range of 20–400 mgNH4+-N L−1.
tewater treatment. In order to investigate the adsorption behavior, batch adsorption
In this study, to overcome the above-mentioned serious drawbacks experiments were carried out under various initial ammonium nitrogen
of micro-sized ZPs and ZBs for adsorption of ammonium ions, we de- (NH4+-N) concentration (20, 100, 200, and 400 mgNH4+-N L−1),
monstrate the entrapment of micro-sized ZPs in porous hydrogels henceforth referred to as ammonium concentration. Ammonium
without any addition of commonly used binders (e.g., clays, silica-based chloride and deionized water were used to prepare a stock solution
or aluminum-based compounds, etc.), as an alternative strategy for which subsequently diluted to obtain desired ammonium concentra-
both the ZPs and the ZBs. We prepare the polyvinyl alcohol (PVA)- tion. Operating temperature and initial pH were about 20 °C and 5.7,
alginate-ZPs (PAZ) composite hydrogel beads by physically entrapping respectively. Adsorption experiments were performed using 50 mL
the ZPs. The as-prepared PAZ hydrogels were structurally character- conical tubes, shaken on a digital orbital shaker (Wisd WiseShake SHO-
ized, and their adsorption capacity toward ammonium ions as a func- 2D) at 100 rpm for 180 min. The volume of ammonium solution and
tion of zeolite dosage at various initial ammonium concentrations were PAZ hydrogels applied throughout the batch adsorption experiments
investigated and discussed. In addition, the adsorption isotherm models was 40 and 5 mL respectively.
and the adsorption diffusion model were comprehensively investigated Because 0, 2, 4, 10, and 20 g of ZPs were added to each 100 mL of
to understand the behavior and mechanism of the ammonium adsorp- PA solution during the preparation of PAZ hydrogels, the resulting
tion process. Furthermore, to experimentally verify the applicability of amount of ZPs added in 5 mL of hydrogel corresponded to 0, 0.1, 0.2,
PAZ hydrogels as a promising adsorbent, a comparison was made with 0.5, and 1.0 g of ZPs, respectively, which hereafter will be denoted as
commercially available ZPs and ZBs. PA, PAZ-0.1, PAZ-0.2, PAZ-0.5, and PAZ-1.0, respectively. Therefore,
for the comparison study, equal dosage (0.1, 0.2, 0.5, and 1.0 g) of ZPs
and ZBs were used. Samples were extracted at certain time intervals
2. Materials and methods
from different conical tubes, filtrated through 0.45 μm microporous
membrane filter, and then acidified by a drop of 0.1 N HCl before the
2.1. Materials
measurement of ammonium concentration using Ion Chromatography
(Dionex ICS-1000, Thermo Scientific) equipped with Dionex IonPac™
Synthetic 4A micro-sized ZPs, synthetic 4A ZBs with a minimum
SCS-1 column (Thermo Scientific).
purity of 90%, and PVA with polymerization degree about 2000 were
Ammonium adsorption capacity (qt) at time t for ZPs and ZBs was
purchased from Wako Pure Chemical Industries Ltd., Japan. Sodium
calculated using the following equation:
alginate was purchased from Showa Chemical Co., Ltd., Japan.
Ammonium chloride (98.5%), boric acid (99.5%) and calcium chloride (C0 − Ct )·V
qt =
anhydrous (96%) were purchased from Samchun Pure Chemical Co., m (1)
Ltd., South Korea.
While qt for PAZ hydrogels was calculated after excluding the con-
tribution of PA by using the following equation:
2.2. Preparation of PAZ composite hydrogel beads [(C0 − Ct )PAZ − (C0 − Ct )PA]·V
qt =
m (2)
The PAZ composite hydrogel beads were prepared as follows: PVA
(8 g) and sodium alginate (1.3 g) were stirred in 80 °C deionized water Where qt is the amount of ammonium ions adsorbed on unit mass of
(total volume 100 mL) followed by autoclaving for 30 min at 120 °C for the zeolite (mg g−1) at time t, C0 and Ct are the initial ammonium
complete dissolution. Then, 0, 2, 4, 10, and 20 g of ZPs were added to concentration at time t (mg L−1), respectively. V is the solution volume
each 100 mL of PVA-alginate (PA) solution and mixed thoroughly. PVA (L) and m is the zeolite weight (g).
is a non-toxic synthetic polymer and has high physical-chemical stabi-
lity, and has been used to entrap cells, enzymes, chemicals and ad- 2.4. Analytical methods
sorbents [1,5,9,22]. In the absence of sodium alginate, the formed
beads had a strong tendency to agglomerate and were very difficult to The diameter of spherical PAZ hydrogels and ZBs adsorbents were
break [9]. Completely mixed solutions of PVA, alginate, and ZPs were measured using a digital caliper, while the size of micro-sized ZPs at
transferred in a dropwise manner into a gentle-stirred solution of boric 0.1 g L−1 concentration was analyzed by Zetasizer Nano ZS (Malvern
acid (5.5% w/v) and calcium chloride (3% w/v). Instrument Ltd.). Environmental scanning electron microscopy (E-SEM
As illustrated schematically in Fig. 1, boric acid and calcium Philips XL-30 FEG, FEI Corporate) was used to observe the surface and
chloride act as a crosslinking agent to form hydrogel for PVA and al- cross-section obtained by cutting the PAZ hydrogels using a sharp knife.
ginate, respectively. When sodium alginate contacted with calcium Energy dispersive X-ray spectroscopy (EDS, EDAX TSL, AMETEK, Inc.)
chloride, the formation of calcium alginate gel occurred almost in- was used to determine the chemical composition of samples.
stantaneously [48]. This instantaneous gelation is expected to suffi-
ciently hold the dispersed ZPs in place, and prevent the agglomeration 3. Results and discussion
of particles and beads during the PVA crosslinking process. In general,
the PVA-gelling reaction occurred immediately on the surface of the 3.1. Structural characterization
immobilized beads. Subsequent gelling reaction inside the beads was
accomplished with the further diffusion of the boric acid into the beads Before the adsorption experiments, all tested adsorbents were
[9]. The beads were stirred gently in the solution for 24 h at 4 °C to structurally characterized. As schematically shown in Fig. 1, the micro-
complete solidification and then washed with distilled water to remove sized ZPs were physically entrapped in PA hydrogel matrix. The re-
any excess boric acid and calcium chloride. sulting hydrogel beads were strong, highly elastic, and of spherical

2
R.N. Putra and Y.H. Lee Separation and Purification Technology xxx (xxxx) xxxx

Fig. 1. Schematic illustration of the preparation of PAZ composite hydrogel beads.

Table 1 porous matrix support. The EDS results are presented in Fig. S1 and the
Diameter, density, and water content of ZPs, PAZ hydrogels, and ZBs. percentage of each element calculated from EDS spectra is summarized
Adsorbent Diameter Wet Density Dry Density Water in Table 2. The Si to Al ratio was 1.31 for entrapped ZPs both on and in
(g cm−3) (g cm−3) Content (%) the PAZ hydrogels, which was very close to the Si/Al ratio of 1.37 for
the ZPs. As clearly seen in Fig. 2, the micro-sized ZPs are well entrapped
Zeolite Particles 4.68 ± 1.22 µm n.a.a 0.49 n.a.
and distributed in the PAZ hydrogel, and the observed size of entrapped
(ZPs) 4.07 ± 0.12 mm 1.18 0.18 84.7
PAZ-0.1
ZPs is in the range of 2–20 µm which is up to 4 times larger than the
PAZ-0.2 4.01 ± 0.11 mm 1.19 0.23 80.7 measured diameter of ZPs. Thus, the micro-sized ZPs are successfully
PAZ-0.5 4.01 ± 0.10 mm 1.35 0.32 74.1 entrapped in the matrix of PA hydrogel with minimal agglomeration
PAZ-1.0 4.07 ± 0.14 mm 1.46 0.39 73.3 during the PA-gelling reaction and subsequent solidification. In ac-
Zeolite Beads 3.91 ± 0.23 mm n.a. 1.45 n.a.
cordance with the high water content, the highly porous structure of
(ZBs)
PAZ hydrogels may allow rapid passage and diffusion of ammonium
a
Not available. ions into the internal pores for contact with the adsorptive sites of the
entrapped ZPs through the water-filled porous channels.
shape. The measured diameter of ZPs, PAZ hydrogels, and ZBs is
4.68 ± 1.22 μm, 4.04 ± 0.13 mm, and 3.91 ± 0.23 mm, respec- 3.2. Effect of zeolite dosage and ammonium concentration
tively, which shows that the size of PAZ hydrogels is almost same as
that of ZBs (Table 1). With an increase in the amount of entrapped ZPs, The adsorption capacity (qt at 180 min) of ZPs, PAZ hydrogels, and
we observe a corresponding increase in the dry density of PAZ hydro- ZBs was studied as a function of zeolite dosage at various initial am-
gels. It is noteworthy that similar to the ZBs, the PAZ hydrogels in the monium concentrations. An amount of ammonium ions adsorbed on
solution were completely settled down immediately because the wet PAZ hydrogels was calculated after excluding the contribution of PA in
density of PAZ hydrogels was higher than that of water (Table 1), while hydrogel. The adsorbed ammonium amount per unit volume of PA
the ZPs in the solution needed hours to be settled down completely. The hydrogel prepared without the addition of ZPs was in the range of 0.04
above results confirm the successful entrapment of micro-sized ZPs in to 0.42 mg mL−1 throughout this study, which showed no significant
the hydrogel beads which can be easily separated from aqueous solu- ammonium adsorption capacity of PA hydrogels. As shown in Fig. 3, the
tion, unlike the ZPs which requires additional solid-liquid separation adsorption capacity of all the absorbents increases with an increase of
techniques such as extensive filtration and sedimentation [19]. It is the initial ammonium concentration due to high driving force for mass
generally accepted that the PA hydrogel acts as a porous and permeable transfer at high ammonium concentration, and decreases with an in-
support matrix and allows water to penetrate the matrix without crease of the zeolite dosage due to an increase of absorbent mass for a
washing out the trapped particles [2,38]. High water content (73–85%) fixed amount of ammonium ions.
in PAZ hydrogels is in good agreement with previously reported values Regardless of the initial ammonium concentrations, the PAZ hy-
of PVA hydrogel [2,36,38], which indicates that the PAZ hydrogels drogels showed lower ammonium adsorption capacity than the ZPs
prepared in this study is highly water permeable. most likely due to less accessible surface area of the entrapped ZPs
A more detailed structural characterization of the representative which were blocked and hindered by PA polymer matrix during the
PAZ-1.0 was investigated by SEM and is presented in Fig. 2. The SEM entrapment of ZPs in the hydrogel. It is noted that the micro-sized ZPs
images reveal the highly porous structure of hydrogel in cross-section showed a significant decrease of the adsorption capacity (43.7 mg g−1)
(Fig. 2a and 2c) and indicate that PA hydrogel is favorably acting as a with an increase of the zeolite dosage from 0.1 to 1 g at the initial

3
R.N. Putra and Y.H. Lee Separation and Purification Technology xxx (xxxx) xxxx

Fig. 2. SEM images of (a) cross-section of PA hydrogel, (b) ZPs, (c) cross-section of PAZ-1.0 hydrogel, and (d) surface of PAZ-1.0 hydrogel.

Table 2 mechanical strength to the beads, are known to significantly reduce the
Elemental composition of PA hydrogel, ZPs, and entrapped ZPs from EDS effective surface area and working adsorption capacity [30,37]. Thus,
analysis. the higher adsorption capacity of PAZ hydrogels is most likely attrib-
Element Cross-section of ZPs Entrapped ZPs on Entrapped ZPs in the uted to the highly porous structure of PA hydrogel which allows much
PA hydrogel (wt%) the surface of PAZ cross-section of PAZ higher accessible surface area of the entrapped ZPs in the hydrogels
(wt%) hydrogel (wt%) hydrogel (wt%) than those in the ZBs. These results reveal that the PAZ hydrogels have
great potential as promising alternative zeolite-type adsorbents to the
C 36.67 27.60 15.70 15.05
O 34.09 11.47 26.79 23.10 ZPs and the ZBs for practical applications involving the adsorption of
Na 0 0 1.90 1.20 cation ions.
Al 0 25.73 15.56 15.23
Si 0 35.20 20.16 20.03
3.3. Adsorption isotherms
Cl 14.42 0 1.07 0.75
Ca 14.82 0 18.81 24.64
In the present work, Langmuir and Dubinin-Radushkevich isotherm
models were used to analyze the experimental results. Correlation
ammonium concentration of 400 mg L−1, which was mainly due to an coefficient (R2) was used to indicate the validity of the model to the
unavoidable significant agglomeration of the ZPs, especially under high experimental data. The Langmuir isotherm model assumes a monolayer
adsorbent dosage conditions. Similar results were previously observed surface adsorption on a finite number of adsorption sites with uniform
with micro-sized ZPs which formed clusters and caused clogging [32], adsorption energy [49]. The linear form of Langmuir isotherm equation
and clinoptilolite nanoparticles which showed agglomeration and sigi- is expressed as:
nicant decrease in available active sites in the case of high adsorbent Ce C 1
loads [51]. Especially, for high zeolite content adsorbents (0.5 and = e +
qe qm KL·qm (3)
1.0 g), the adsorption capacity of PAZ hydrogels is only 14.3 to 36.6%
lower than that of the micro-sized ZPs under all operational conditions. Where Ce is the equilibrium concentration of ammonium ions in
Therefore, unlike the ZPs, the PAZ hydrogels showed no significant solution (mg L−1), qe is the ammonium sorption capacity at equilibrium
agglomeration of the entrapped ZPs even under high zeolite dosage per unit mass of zeolite (mg g−1), qm is the maximum monolayer ad-
conditions. During all adsorption experiments with the PAZ hydrogels, sorption capacity (mg g−1), KL is the Langmuir adsorption constant (L
any noticeable zeolite powders in the solution were not observed. In the mg−1).
literature, high physical–chemical stability of polyvinyl alcohol (PVA)- The values of qm and KL were obtained by plotting Ce/qe against Ce
alginate (PA) hydrogel as supporting matrix has been reported [16,31]. under various experimental conditions (Fig S2). The obtained iso-
Therefore, we suggest that this high stability of PA supporting matrix thermal constants and the correlation coefficients are presented in
resulted in no observable sign of leaching of entrapped ZPs in PAZ Table 3. The adsorption of ammonium ions on all the adsorbents cor-
hydrogel beads during our adsorption experiments for 180 min. related well (R2 > 0.95) with Langmuir isotherm equation, indicating
Although the PAZ hydrogel has a similar size to the ZBs (Table 1), its that the ammonium adsorption under the studied concentration range
ammonium adsorption capacity is up to 3 times higher than that of ZBs is a monolayer adsorption. These high R2 values indicate that the ob-
(Fig. 3). According to the manufacturer, the binder content in the ZBs tained ammonium adsorption results using all the adsorbents under the
employed in this study was less than 10 wt%. In general, the binders studied concentration range are reliably validated by the Langmuir
commonly used for making the ZBs and giving the necessary isotherm model. The Langmuir isotherm model assumes that the forces
of interaction between the adsorbed adsorbate molecules are negligible

4
R.N. Putra and Y.H. Lee Separation and Purification Technology xxx (xxxx) xxxx

and once the adsorbate molecule occupies the adsorption site, no fur-
ther adsorption will take place at that site [49].
The maximum ammonium adsorption capacities obtained by the
model are found to decrease with increasing the zeolite dosage, due to
the increase of adsorption sites on the zeolite surface for the fixed
amount of ammonium ions. The highest value of the maximum ad-
sorption capacities was 60.6, 28.2, and 6.6 mg g−1 for ZPs, PAZ hy-
drogels, and ZBs, respectively (Table 3). Regardless of the zeolite do-
sage, the maximum adsorption capacities showed the following order:
ZPs > PAZ hydrogels > ZBs. The maximum adsorption capacities of
PAZ hydrogels are 49.8–53.5% lower than those of the micro-sized ZPs,
which suggests that about 50% of the accessible adsorption sites of the
entrapped ZPs in the hydrogel are blocked and hindered by PA polymer
matrix during the entrapment process. However, the maximum ad-
sorption capacities of PAZ hydrogels are 3.4–4.3 times higher than
those of ZBs, mainly due to much higher accessible adsorption sites of
the entrapped ZPs in the hydrogels than those in the ZBs. In agreement
with the results mentioned in the above section and literature findings
[37], we propose that the highly porous structure of PA hydrogel has
much less adverse effects on the adsorption capacity compared to the
binders used in the preparation of commercial synthetic ZBs.
The maximum ammonium adsorption capacities of all the ad-
sorbents tested in this study are compared with those of natural and
synthetic zeolite-type adsorbents with various particle sizes
(1–3,200 μm) reported in the literatures
[12,17,19,20,21,25,27,40,41,43,45,47,52,54]. As shown in Fig. 4, it is
evident that the maximum adsorption capacity is inversely proportional
to the reported highest value of adsorbent particle size (Table S1). The
maximum adsorption capacities of ZPs and ZBs are close to those of
other zeolite-type adsorbents with the similar particle sizes, which
further justifies our results for commercial ZPs and ZBs. The maximum
adsorption capacities of PAZ hydrogels are even higher than those of
other natural and synthetic zeolite-type adsorbents with particle sizes
ranging up to 75 μm ([28,34,44,54]. It should be noteworthy that the
PAZ hydrogel beads exhibit remarkably higher adsorption capacity
than other bead-sized zeolite-type adsorbents [12,23,50], owing to
their unique properties such as highly porous structure and much
Fig. 3. Effect of zeolite dosage and initial ammonium concentration on the higher active adsorption sites still in the entrapped ZPs, and comparable
adsorption capacity of each adsorbent (a) ZPs, (b) PAZ hydrogels, and (c) ZBs. higher adsorption capacity with other micro-sized zeolite-type parti-
cles. These results suggest that the PAZ hydrogels are highly promising
adsorbents and can be used to replace other bead-sized and even micro-
Table 3
sized zeolite-type adsorbents.
Isotherm parameters for the ammonium adsorption on ZPs, PAZ hydrogels, and
Dubinin-Radushkevitch (D-R) isotherm model is generally used to
ZBs.
describe the adsorption process of microporous adsorbents, such as
Adsorbent Langmuir isotherm Dubinin-Radushkevich isotherm activated carbon and zeolite, expressed as the following linear equation
[8,10]:
qm KL R2 K E R2
(mg g−1) (L mg−1) (mol2 kJ–2) (kJ mol−1) lnq e = lnqm − Kε 2 (4)
Zeolite
1 ⎞
particles ε = RT·ln ⎛1 +
⎜ ⎟

(ZPs) ⎝ Ce ⎠ (5)
ZP-0.1 60.61 0.04 0.98 0.0050 10.01 0.99
−1
ZP-0.2 45.05 0.03 0.96 0.0057 9.39 0.96 where qm is the theoretical maximum adsorption capacity (mg g ), K
ZP-0.5 42.74 0.01 1.00 0.0067 8.62 0.93 is a D-R constant associated to the adsorption energy (mol2 kJ−2), qe
ZP-1.0 35.97 0.01 0.96 0.0068 8.61 0.90 and Ce are the same as those in the Langmuir isotherm model, ε is the
PAZ hydrogels Polanyi potential (kJ mol−1), R is the gas law constant (kJ mol−1 K−1),
PAZ-0.1 28.17 0.01 0.99 0.0051 9.90 0.92 and T is the absolute temperature (K). The values of qm and K were
PAZ-0.2 22.62 0.01 0.99 0.0076 8.11 0.99
obtained by plotting ln qe against ε2 under various experimental con-
PAZ-0.5 20.49 0.01 0.99 0.0093 7.33 0.99
PAZ-1.0 16.84 0.01 0.99 0.0105 6.90 0.99 ditions (Fig. S3). The D-R constant K can give an information regarding
the mean free energy E (kJ mol−1) of adsorption per molecule of ad-
Zeolite beads
(ZBs) sorbate, and E can be obtained using the following relationship [11,39]:
ZB-0.1 6.60 0.02 0.97 0.0052 9.81 0.95 1
ZB-0.2 5.86 0.06 0.95 0.0053 9.71 0.87 E=
¯
√ 2K (6)
ZB-0.5 5.08 0.03 0.95 0.0085 7.67 0.95
ZB-1.0 5.02 0.05 0.99 0.0085 7.67 0.91
Based on the principle of this model, it is known that the value of E
could determine the type of adsorption mechanism. If the E value is in
the range of 8–16 kJ mol−1, the adsorption process follows by chemical

5
R.N. Putra and Y.H. Lee Separation and Purification Technology xxx (xxxx) xxxx

Fig. 4. Comparison of the maximum ammonium adsorption capacities of all the adsorbents tested in this study to the reported values of zeolite-type adsorbents with
various particle sizes in the literatures (see Table S1 for detailed information).

ion exchange, whereas if the E value lies between 1 and 8 kJ mol−1, the
physical forces may affect the adsorption mechanism [11]. As listed in
Table 3, an average value of the obtained mean free energies (E) for the
adsorption of ammonium ions onto ZPs, PAZ hydrogels, and ZBs under
all operational conditions was 9.16 ± 0.68, 8.06 ± 1.32, and
8.72 ± 1.21 kJ mol−1, respectively, which are within the energy
ranges of chemical ion exchange reaction. This result indicates that the
ammonium adsorption process in all the adsorbents tested in this study
is governed by chemical ion exchange mechanism, which is consistent
with the previously reported results with synthetic zeolite adsorbents
[11,18]. The similar E values for all the adsorbents suggest that the
entrapment of ZPs in the hydrogel beads does not significantly affect
the adsorption mechanism of ammonium ions onto the zeolite.

3.4. Adsorption diffusion mechanism

In most cases, the rate-limiting step for the batch adsorption process
could be one of these steps: boundary layer diffusion (external mass
transfer), intra-particle diffusion, and sorption of solute in the adsorp-
tion site, or could be mix of these steps. In order to identify the diffusion
mechanism of ammonium adsorption on the zeolite, the Weber and
Morris intra-particle diffusion model is used when the rate-limiting step
is a combination of boundary layer diffusion and intra-particle diffusion
mechanisms [15,46]:

qt = kint ·t0.5 + C (7)

where kint is the rate constant of intra-particle diffusion

(mg g−1 min−0.5) and C is a constant (mg g−1) related to the thickness
of the boundary layer. If the plot of qt versus t0.5 is single-linear and
passes through the origin, the adsorption process is only controlled by
intra-particle diffusion. However, if the data exhibit multi-linear plots,
two or more steps influence the adsorption process [23].
Fig. 5 shows that the intra-particle diffusion model fits well with the
experimental data obtained under various zeolite dosage conditions at
initial ammonium concentration of 400 mg L−1, which indicates that
the overall adsorption process is divided into two linear segments for Fig. 5. Intra-particle diffusion plots for the ammonium adsorption on (a) ZPs,
ZPs and PAZ hydrogels, and one linear segment for ZBs. Thus, the (b) PAZ hydrogels, and (c) ZBs (C).
overall adsorption process of ammonium ions from aqueous solution

6
R.N. Putra and Y.H. Lee Separation and Purification Technology xxx (xxxx) xxxx

Table 4 agglomeration of particles under high zeolite dosage conditions, the

Intra-particle diffusion constants for the ammonium adsorption on ZPs, PAZ PAZ hydrogels showed no significant agglomeration of the entrapped
hydrogels, and ZBs. ZPs in the hydrogels. Due to still high accessible surface area of the
Adsorbent Intra-particle diffusion model entrapped ZPs in the hydrogels, the maximum ammonium adsorption
capacities of PAZ hydrogels from Langmuir isotherm were up to 4.3
kint,1 R2 kint,2 R2 times higher than those of ZBs. Thus, we propose that the entrapment of
(mg g−1 min−0.5) (mg g−1 min−0.5)
ZPs in highly porous hydrogel has much less adverse effects on the
Zeolite particles (ZPs) adsorption capacity, compared to the binder used in the synthetic ZBs.
ZP-0.1 16.94 1.00 0.17 0.95 It should be noted that the PAZ hydrogels exhibit even and remarkably
ZP-0.2 12.23 1.00 0.03 0.90 higher ammonium adsorption capacity than other micro-sized and
ZP-0.5 6.44 1.00 0.05 0.95
bead-sized zeolite-type adsorbents in the literature, respectively.
ZP-1.0 3.59 1.00 0.03 0.96
Similar to the ZPs, the ammonium adsorption process in the PAZ hy-
PAZ hydrogels drogels was governed by chemical ion exchange mechanism and tended
PAZ-0.1 2.72 0.96 0.29 0.97
PAZ-0.2 2.24 0.99 0.33 0.94
to be rate-limited by boundary layer diffusion and mainly intra-particle
PAZ-0.5 1.75 0.99 0.38 0.92 diffusion. These results reveal that the PAZ hydrogels have great po-
PAZ-1.0 1.19 0.97 0.27 0.96 tential as promising alternative zeolite-type adsorbents to both the ZPs
Zeolite beads (ZBs) and the ZBs. We anticipate that this strategy can be expanded to the
ZB-0.1 0.46 0.98 entrapment of other micro-sized adsorbents in highly porous hydrogels
ZB-0.2 0.41 0.96 for a broad range of industrial applications in water purification and
ZB-0.5 0.44 0.99
wastewater treatment.
ZB-1.0 0.38 0.97

Declaration of Competing Interest

onto the ZBs is solely controlled by the intra-particle diffusion as the
rate-limiting process. The diffusion rate constants of ZBs were sig- The authors declare that they have no known competing financial
nificantly lower than those of ZPs and PAZ hydrogels, because of the interests or personal relationships that could have appeared to influ-
adverse effects of binders which were used in the preparation of com- ence the work reported in this paper.
mercial synthetic ZBs and commonly known to hinder the diffusion of
solute into the zeolite [13]. In contrast, the PA hydrogel, which can Acknowledgements
adsorb a large amount of water due to its hydrophilicity, is known to
diminish mass transfer resistance and allow the solute to diffuse easily This work was financially supported by the Korea Institute of
through the porous hydrogel structure [55]. Science and Technology Institutional Program (Project No. 2E29660).
The multi-linearity coming from two linear segments suggests that The authors would like to express gratitude to Ms. Ji-Su Lee for her
the rate-limiting step of the ammonium adsorption process using ZPs assistance in laboratory work during this study.
and PAZ hydrogels is attributed to the boundary layer diffusion (the
first sharper linear stage) and the intra-particle diffusion (the second Appendix A. Supplementary material
linear stage), which was consistent with the reported results with nat-
ural and synthetic zeolite-type adsorbents [20,52]. Thus, the boundary Supplementary data to this article can be found online at https://
layer diffusion and the subsequent intra-particle diffusion take place doi.org/10.1016/j.seppur.2019.116351.
during the gradual adsorption process. As summarized in Table 4, the
kint,1 values for both ZPs and PAZ hydrogels were considerably higher References
than the kint,2 values, indicating that the intra-particle diffusion was the
dominant rate-limiting step for both adsorbents and their adsorption [1] S. Al-Zuhair, M. El-Naas, Immobilization of Pseudomonas putida in PVA gel parti-
cles for the biodegradation of phenol at high concentrations, Biochem. Eng. J. 56
was mainly occurred at the surface of ZPs. Similar results were reported (1–2) (2011) 46–50.
for zeolite-type adsorbents at various ammonium concentrations [2] B. Amsden, Solute diffusion within hydrogels, Mech. Models. Macromol. 31 (23)
[15,23]. This result implies that the highly porous structure of PAZ (1998) 8382–8395.
[3] S.M. Auerbach, K.A. Carrado, P.K. Dutta, Handbook of Zeolite Science and
hydrogels allows a rapid transport of ammonium ions to occupy the Technology, Science, Marcel Dekker, Inc., New York, 2003.
adsorption sites on the surface of the entrapped ZPs, and then diffusion [4] A. Charkhi, M. Kazemeini, S.J. Ahmadi, S. Ammari Allahyari, Effect of bentonite
of ammonium ions into the internal pores at slower pace. Although the binder on adsorption and cation exchange properties of granulated nano NaY
zeolite, Adv. Mater. Res. Trans. Tech. Publ. (2011) 423–428.
diffusion rate constants of PAZ hydrogels were lower than those of ZPs, [5] J.H. Chen, G.P. Li, Q.L. Liu, J.C. Ni, W.B. Wu, J.M. Lin, Cr(III) ionic imprinted
the entrapment of micro-sized ZPs in the hydrogel did not affect the polyvinyl alcohol/sodium alginate (PVA/SA) porous composite membranes for se-
diffusion mechanism of ammonium adsorption process in the ZPs. lective adsorption of Cr(III) ions, Chem. Eng. J. 165 (2) (2010) 465–473.
[6] C. Covarrubias, R. Arriagada, J. Yanez, M.T. Garland, R. Garcia, Natural mordenite
Therefore, utilization of the PAZ hydrogels in the adsorption process
derived zeolites: Synthesis, formation, and their evaluation in Cr(III) removal from
makes the rate-limiting diffusion processes less effective on the overall tannery wastewater, Environ. Eng. Sci. 24 (10) (2007) 1443–1455.
rate of adsorption process, compared to the granulated ZBs. [7] A. Cybulski, M.M. Sharma, R.A. Sheldon, J.A. Moulijn, Fine Chemicals
Manufacture: Technology and Engineering, Gulf Professional Publishing, 2001.
[8] J. Dai, H. Yan, H. Yang, R. Cheng, Simple method for preparation of chitosan/poly
(acrylic acid) blending hydrogel beads and adsorption of copper(II) from aqueous
4. Conclusion solutions, Chem. Eng. J. 165 (1) (2010) 240–249.
[9] R. Dave, D. Madamwar, Esterification in organic solvents by lipase immobilized in
We have demonstrated that the entrapment of micro-sized ZPs in polymer of PVA–alginate–boric acid, Process Biochem. 41 (4) (2006) 951–955.
[10] R. Donat, A. Akdogan, E. Erdem, H. Cetisli, Thermodynamics of Pb2+ and Ni2+
porous hydrogels is a promising strategy toward overcoming drawbacks adsorption onto natural bentonite from aqueous solutions, J. Coll. Interf. Sci. 286
of ZPs and ZBs for adsorption of ammonium ions. In this study, we (1) (2005) 43–52.
prepared the PAZ hydrogel beads by physically entrapping the ZPs in [11] A.M. El-Kamash, A.A. Zaki, M.A. El Geleel, Modeling batch kinetics and thermo-
dynamics of zinc and cadmium ions removal from waste solutions using synthetic
highly water permeable and porous PA hydrogel matrix. Unlike the ZPs zeolite A, J. Hazard. Mater. 127 (1–3) (2005) 211–220.
which needed hours for complete settling, the PAZ hydrogels in the [12] B.C. Erdoğan, S. Ülkü, Ammonium sorption by Gördes clinoptilolite rich mineral
solution were settled down immediately. While the ZPs showed a sig- specimen, Appl. Clay Sci. 54 (3–4) (2011) 217–225.
[13] T. Fakin, A. Ristić, A. Horvat, V. Kaučič, Water Adsorption Study on the Zeolite Lta
nificant decrease of the adsorption capacity due to an unavoidable

7
R.N. Putra and Y.H. Lee Separation and Purification Technology xxx (xxxx) xxxx

Granules, Proceedings of the 5th Serbian-Croatian-Slovenian Symposium on (1) (2007) 258–263.

Zeolites, May 30th-June 2nd, (2013). [35] K. Shams, S.J. Mirmohammadi, Preparation of 5A zeolite monolith granular ex-
[14] G.W. Gee, D. Or, 2.4 Particle-size analysis, Meth. Soil Anal. Part 4 598 (2002) trudates using kaolin: Investigation of the effect of binder on sieving/adsorption
255–293. properties using a mixture of linear and branched paraffin hydrocarbons, Micropor.
[15] D. Guaya, C. Valderrama, A. Farran, C. Armijos, J.L. Cortina, Simultaneous phos- Mesopor. Mater. 106 (1–3) (2007) 268–277.
phate and ammonium removal from aqueous solution by a hydrated aluminum [36] S. Sharma, S.S. Kanwar, Organic solvent tolerant lipases and applications, Sci.
oxide modified natural zeolite, Chem. Eng. J. 271 (2015) 204–213. World J. 2014 (2014) 1–15.
[16] U. Habiba, T.A. Siddique, T.C. Joo, A. Salleh, B.C. Ang, A.M. Afifi, Synthesis of [37] J.A.C. Silva, K. Schumann, A.E. Rodrigues, Sorption and kinetics of CO2 and CH4 in
Chitosan/polyvinyl alcohol/zeolite composite for removal of methyl orange, Congo binderless beads of 13X zeolite, Micropor. Mesopor. Mater. 158 (2012) 219–228.
Red and Chromium(VI) by flocculation/adsorption, Carbohydrate Polym. 157 [38] J. Stasko, M. Kalnińš, A. Dzene, V. Tupureina, Poly(vinyl alcohol) hydrogels, Proc.
(2016) 1568–1576. Est. Acad. Sci. 58 (1) (2009) 63–66.
[17] E. Ivanova, M. Karsheva, B. Koumanova, Adsorption of ammonium ions onto nat- [39] I.A.W. Tan, A.L. Ahmad, B.H. Hameed, Adsorption isotherms, kinetics, thermo-
ural zeolite, J. Univ. Chem. Technol. Metall. 45 (3) (2010) 295–302. dynamics and desorption studies of 2,4,6-trichlorophenol on oil palm empty fruit
[18] F. Ji, C. Li, B. Tang, J. Xu, G. Lu, P. Liu, Preparation of cellulose acetate/zeolite bunch-based activated carbon, J. Hazard. Mater. 164 (2–3) (2009) 473–482.
composite fiber and its adsorption behavior for heavy metal ions in aqueous solu- [40] P. Vassileva, D. Voikova, Investigation on natural and pretreated Bulgarian clin-
tion, Chem. Eng. J. 209 (2012) 325–333. optilolite for ammonium ions removal from aqueous solutions, J. Hazard. Mater.
[19] C.D. Johnson, F. Worrall, Novel low density granular adsorbents - Properties of a 170 (2–3) (2009) 948–953.
composite matrix from zeolitisation of vermiculite, Chemosphere 68 (6) (2007) [41] Y.-F. Wang, F. Lin, W.-Q. Pang, Ammonium exchange in aqueous solution using
1153–1162. Chinese natural clinoptilolite and modified zeolite, J. Hazard. Mater. 142 (1–2)
[20] D. Karadag, Y. Koc, M. Turan, B. Armagan, Removal of ammonium ion from aqu- (2007) 160–164.
eous solution using natural Turkish clinoptilolite, J. Hazard. Mater. 136 (3) (2006) [42] Y. Wang, S. Liu, Z. Xu, T. Han, S. Chuan, T. Zhu, Ammonia removal from leachate
604–609. solution using natural Chinese clinoptilolite, J. Hazard. Mater. 136 (3) (2006)
[21] L. Lei, X. Li, X. Zhang, Ammonium removal from aqueous solutions using micro- 735–740.
wave-treated natural Chinese zeolite, Sep. Purif. Technol. 58 (3) (2008) 359–366. [43] Y. Watanabe, H. Yamada, J. Tanaka, Y. Komatsu, Y. Moriyoshi, Ammonium ion
[22] X. Li, Y. Li, Z. Ye, Preparation of macroporous bead adsorbents based on poly(vinyl exchange of synthetic zeolites: the effect of their open-window sizes, pore struc-
alcohol)/chitosan and their adsorption properties for heavy metals from aqueous tures, and cation exchange capacities, Sep. Sci. Technol. 39 (9) (2005) 2091–2104.
solution, Chem. Eng. J. 178 (2011) 60–68. [44] Water Environment Federation, 2010. Design of Municipal Wastewater Treatment
[23] L. Lin, Z. Lei, L. Wang, X. Liu, Y. Zhang, C. Wan, D.-J. Lee, J.H. Tay, Adsorption Plants: WEF Manual of Practice No. 8 ASCE Manuals and Reports on Engineering
mechanisms of high-levels of ammonium onto natural and NaCl-modified zeolites, Practice No. 76. McGraw-Hill Education.
Sep. Purif. Technol. 103 (2013) 15–20. [45] L.R. Weatherley, N.D. Miladinovic, Comparison of the ion exchange uptake of
[24] C.H. Liu, K.V. Lo, Ammonia removal from composting leachate using zeolite. I. ammonium ion onto New Zealand clinoptilolite and mordenite, Water Res. 38 (20)
Characterization of the zeolite, J. Environ. Sci. Health. A. Tox. Hazard. Subst (2004) 4305–4312.
Environ. Eng. 36 (9) (2001) 1671–1688. [46] W.J. Weber, J.C. Morris, Kinetics of adsorption on carbon from solution, J. Sanit.
[25] H. Liu, Y. Dong, H. Wang, Y. Liu, Ammonium adsorption from aqueous solutions by Eng. Div. 89 (2) (1963) 31–60.
strawberry leaf powder: Equilibrium, kinetics and effects of coexisting ions, [47] D. Wen, Y.S. Ho, X. Tang, Comparative sorption kinetic studies of ammonium onto
Desalination 263 (1–3) (2010) 70–75. zeolite, J. Hazard. Mater. 133 (1–3) (2006) 252–256.
[26] I.D.R. Mackinnon, G.J. Millar, Practical applications for clay-derived zeolite N: an [48] K.-Y.A. Wu, K.D. Wisecarver, Cell immobilization using PVA crosslinked with boric
ammonium-selective ion exchanger for industry use, 50th Meeting of the Clay acid, Biotechnol. Bioeng. 39 (4) (1992) 447–449.
Minerals Society, 5-9 October 2013, University of Illinois, Urbana-Champaign, IL, [49] Y. Yu, J.G. Shapter, R. Popelka-Filcoff, J.W. Bennett, A.V. Ellis, Copper removal
2013. using bio-inspired polydopamine coated natural zeolites, J. Hazard. Mater. 273
[27] E. Marañón, M. Ulmanu, Y. Fernández, I. Anger, L. Castrillón, Removal of ammo- (2014) 174–182.
nium from aqueous solutions with volcanic tuff, J. Hazard. Mater. 137 (3) (2006) [50] A.M. Yusof, L.K. Keat, Z. Ibrahim, Z.A. Majid, N.A. Nizam, Kinetic and equilibrium
1402–1409. studies of the removal of ammonium ions from aqueous solution by rice husk ash-
[28] Metcalf and Eddy, Inc. 2003 Wastewater Engineering: Treatment, Disposal, and synthesized zeolite Y and powdered and granulated forms of mordenite, J. Hazard.
Reuse. McGraw-Hill, Inc.: New York. Mater. 174 (1–3) (2010) 380–385.
[29] M.L. Nguyen, C.C. Tanner, ammonium removal from wastewaters using natural [51] H. Zabihi-Mobarakeh, A. Nezamzadeh-Ejhieh, Application of supported TiO2 onto
New Zealand zeolites, New Zealand J. Agri. Res. 41 (3) (1998) 427–446. Iranian clinoptilolite nanoparticles in the photodegradation of mixture of aniline
[30] O. Öhrman, J. Hedlund, J. Sterte, Synthesis and evaluation of ZSM-5 films on and 2, 4-dinitroaniline aqueous solution, J. Ind. Eng. Chem. 26 (2015) 315–321.
cordierite monoliths, Appl. Catal. A Gen. 270 (1–2) (2004) 193–199. [52] M. Zhang, H. Zhang, D. Xu, L. Han, D. Niu, B. Tian, J. Zhang, L. Zhang, W. Wu,
[31] L.R. Rad, A. Momeni, B.F. Ghazani, M. Irani, M. Mahmoudi, B. Noghreh, Removal of Removal of ammonium from aqueous solutions using zeolite synthesized from fly
Ni2+ and Cd2+ Ions from aqueous solutions using electrospun PVA/zeolite na- ash by a fusion method, Desalination 271 (1–3) (2011) 111–121.
nofibrous adsorbent, Chem. Eng. J. 256 (2014) 119–127. [53] Y. Zhang, L. Zhao, Y. Chen, Synthesis and characterization of starch-g-Poly(acrylic
[32] V.R. Rao, S. Farooq, Experimental study of a pulsed-pressure-swing-adsorption acid)/Organo-Zeolite 4A superabsorbent composites with respect to their water-
process with very small 5A zeolite particles for oxygen enrichment, Ind. Eng. Chem. holding capacities and nutrient-release behavior, Polym. Compos. 38 (9) (2015)
Res. 53 (33) (2014) 13157–13170. 1838–1848.
[33] G. Rioland, H. Nouali, T.J. Daou, D. Faye, J. Patarin, Adsorption of volatile organic [54] H. Zheng, L. Han, H. Ma, Y. Zheng, H. Zhang, D. Liu, S. Liang, Adsorption char-
compounds in composite zeolites pellets for space decontamination, Adsorption 23 acteristics of ammonium ion by zeolite 13X, J. Hazard. Mater. 158 (2–3) (2008)
(2–3) (2017) 395–403. 577–584.
[34] K. Saltali, A. Sari, M. Aydin, Removal of ammonium ion from aqueous solution by [55] Y. Zheng, A. Wang, Preparation and ammonium adsorption properties of biotite-
natural Turkish (Yildizeli) zeolite for environmental quality, J. Hazard. Mater. 141 based hydrogel composites, Ind. Eng. Chem. Res. 49 (13) (2010) 6034–6041.