International Journal of Biological Macromolecules 266 (2024) 131354

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

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

A floating biosorbent of polylactide and carboxylated cellulose from

biomass for effective removal of methylene blue from water
Sasimaporn Ampawan , Jiratchaya Dairoop , Mallika Keawbanjong , Watchanida Chinpa *
Division of Physical Science, Faculty of Science, Prince of Songkla University, Hat Yai, Songkhla 90110, Thailand

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

Keywords: A floating adsorbent bead was prepared from polylactide (PLA) and maleic anhydride (MAH)-modified cellulose
Polylactide in a one-pot process (OP bead). Cellulose was extracted from waste lemongrass leaf (LGL) and modified with
Cellulose MAH in the presence of dimethylacetamide (DMAc). PLA was then added directly into the system to form sorbent
Biosorbent
beads by a phase separation process that reused unreacted MAH and DMAc as a pore former and a solvent,
Dye adsorption
respectively. The chemical modification converted cellulose macrofibres (55.1 ± 31.5 μm) to microfibers (8.8 ±
1.5 μm) without the need for grinding. The OP beads exhibited more and larger surface pores and greater thermal
stability than beads prepared conventionally. The OP beads also removed methylene blue (MB) more effectively,
with a maximum adsorption capacity of 86.19 mg⋅g− 1. The adsorption of MB on the OP bead fitted the pseudo-
second order and the Langmuir isotherm models. The OP bead was reusable over five adsorption cycles, retaining
88 % of MB adsorption. In a mixed solution of MB and methyl orange (MO), the OP bead adsorbed 96 % of the
cationic dye MB while repelling the anionic dye MO. The proposed method not only reduced time, energy and
chemical consumption, but also enabled the fabrication of a green, effective and easy-to-use biosorbent.

1. Introduction Na-Alg/cellulose/clay [11] and Na-Alg/Halloysite/Hemp hurd [12]

were applied to remove methylene blue (MB). Microspheres of NaCMC
Adsorption is more commonly used to eliminate dyes from effluent loaded with modified aminated lignin removed MB with a maximum
than methods such as flocculation, coagulation, photodegradation and adsorption capacities (qm) of 118.30 mg g− 1 [13]. In another study, CS/
membrane separation. It is a simple and efficient process that can be kC/activated bentonite (ABTN) was prepared in membrane form and
used to treat concentrated dye solutions [1]. Moreover, the adsorbents removed 98 % of MB [14].
used in the process can be prepared from a variety of materials. Acti­ Although composite adsorbents simplify the separation and recovery
vated carbon (AC) is frequently used to adsorb dyes because its large process, polymer matrix swelling can make upscaling the process diffi­
surface area and micropore volume promotes rapid adsorption. How­ cult [15]. Therefore, hydrophobic polymers such as polyethersulfone
ever, since AC is expensive [1], cheaper adsorbents made from clays and (PES) and polypolyvinylidene fluoride (PVDF) are selected as matrix
clay minerals or lignocellulosic biomass such as cellulose and lignin, and materials for composite adsorbents. They have good mechanical prop­
their modified forms [2–8], are used to remove cationic pollutants. erties, can be incorporated in porous materials and, importantly, do not
Although abundant, efficient and environmentally friendly, these ad­ swell in water. These properties enable the fabrication of adsorbent
sorbents are made of tiny particles, which complicate the separation of beads that float on the surface of water to facilitate separation and re­
the dye-loaded adsorbent from the dye solution, typically requiring covery. A floating adsorbent bead was prepared from PES with atta­
centrifugation [9]. To overcome these problems, biocomposite sorbents pulgite (ATP) by using non-solvent induced phase separation (NIPS)
have been prepared by combining powdered adsorbents with bio­ [15]. The ATP/PES bead was easy to use for removal of Cu(II) and Cd(II).
polymers such as carboxymethyl cellulose (CMC), k-carrageenan (kC), A composite bead for the removal of malachite green (MG) was prepared
sodium alginate (Na-Alg), and chitosan (CS). These biocomposite sor­ from zeolitic imidazolate frameworks (ZIFs) and PES [16]. The devel­
bents have been produced as beads [9–13] and membranes [14]. Com­ oped adsorbent removed 99.2 % of MG while pure PES beads showed
posite beads from CMC/kC/activated montmorillonite (AMMT) [10], negligible adsorption of MG. A novel composite bead of UiO-66 MOF

* Corresponding author.
E-mail address: watchanida.c@psu.ac.th (W. Chinpa).

https://doi.org/10.1016/j.ijbiomac.2024.131354
Received 26 December 2023; Received in revised form 17 March 2024; Accepted 1 April 2024
Available online 2 April 2024
0141-8130/© 2024 Elsevier B.V. All rights reserved.
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

and PVDF removed 97 % of tartrazine [17]. However, these sorbents are point DMAc was added and the mixture was refluxed. After refluxing the
produced with petroleum-based polymers that are not biodegradable mixture for 8 h, the MLGL was separated and washed with acetone and
and therefore create secondary pollution. DI water to remove MAH. The obtained MLGL was dried and stored in a
Polylactide (PLA) is an inexpensive, easily processed biodegradable desiccator until used. The preparation of MLGL from raw LGL was pre­
polymer with good mechanical strength and low toxicity [18]. PLA is sented in Fig. 1a.
also hydrophobic and its solubility in organic solvents could facilitate
NIPS. Recently, a PLA-based composite membrane loaded with hy­ 2.3. Preparation of biocomposite adsorbent bead
droxyapatite (HAp) was produced by NIPS and removed 97 % of Pb and
82 % of As from water [19]. Cellulose is a biopolymer that is abundant, The biosorbent beads were prepared by two different methods.
inexpensive, and biodegradable. Moreover, it contains many hydroxyl
groups which readily bind with dye pollutants and could be chemically (i) In a conventional, or post modification (PM), method, dry MLGL
modified to increase adsorption efficiency [6,20,21]. In our previous obtained as described in 2.2 was mixed with PLA in DMAc at
work, we used NIPS to prepare a biocomposite membrane of PLA and 70 ◦ C until MLGL microfibers were well dispersed. The mixture
poly(butylene adipate-co-terephthalate) (PBAT) with modified cellulose solution of PLA, MLGL, and DMAc was added dropwise into DI
extracted from waste empty fruit bunch (MEFB) (PLA/PBAT-MEFB). The water and PM biocomposite beads were formed, which were
loading of MEFB enabled efficient removal of MB from water [22]. washed with DI water.
The abovementioned composite adsorbents are conventionally pre­ (ii) In the proposed one-pot (OP) method, the DMAc used in the
pared in two main steps. The first step is the preparation and/or modi­ preparation of MLGL (2.2) was reused to dissolve PLA. Therefore,
fication of an adsorbent powder that is processed by isolation, after the reflux reaction, the mixture of MLGL and unreacted
purification and drying. The second step is the preparation of the com­ MAH (uMAH) and DMAc was allowed to cool to 70 ◦ C, and PLA
posite adsorbent by combining the processed adsorbent powder and a was then directly added to the mixture. When the PLA was
polymeric matrix. Although some composite adsorbents prepared in this completely dissolved, the mixture solution was added dropwise
way are highly efficient, the synthesis involves many steps and/or into DI water to form OP adsorbent beads. The OP beads were
chemicals. washed with ethanol followed by DI water until pH was neutral.
In our previous work [22], cellulose extracted from EFB was first
modified using maleic anhydride (MAH) dissolved in dimethylaceta­ After washing, both bead types were dried for the following study.
mide (DMAc) and the modified cellulose was then loaded into a solution The loading of MLGL in the PLA matrix of the OP bead was 15 wt%
of PLA and DMAc. The aim of the present work is to prepare the floating (OP15). The PM bead was fabricated with loadings of 15 and 20 wt%
biosorbent developed in our previous work in a simpler way. Cellulose (PM15 and PM20). The preparation procedures of the adsorbent beads
was extracted from lemongrass leaf (LGL) and chemically modified as were presented in Fig. 1a.
we previously modified cellulose from LGL [23]. However, the separa­
tion, purification and drying of the modified cellulose from LGL were
2.4. Sample characterizations
eliminated, and the MAH and DMAc used to modify the cellulose were
reused in the production of porous beads by NIPS. Therefore, this
The morphology of samples was observed by scanning electron mi­
alternative method is easier and faster. Less time and energy and fewer
croscopy (SEM, Quanta 400, FEI).
chemicals are used since the adsorbent bead is fabricated in one step.
The functional groups of LGL-derived adsorbents were identified by
The morphology, water absorption and thermal stability of the proposed
using Fourier transform infrared spectroscopy (FTIR, Nicolet iS5,
adsorbent bead were characterized for comparison with the composite
Thermo Scientific) in the range of 4000 to 600 cm− 1.
bead prepared by a conventional method. The effects of contact time,
Water uptake was determined from the following equation:
adsorbent dose, dye solution pH, initial dye concentration, and tem­
perature on the adsorption of MB on the proposed bead were deter­ Ww − Wd
Water uptake (%) = × 100 (1)
mined. Adsorption kinetics and isotherms, and the reusability and Wd
selectivity of the adsorbent bead were also studied.
where Wd and Ww are respectively the weight of the dried adsorbent and
2. Material and methods the adsorbent immersed in distilled water for 24 h.
The thermal stability of PLA-based adsorbent beads was determined
2.1. Materials by thermal gravimetric analyzer (STA8000, Perkin Elmer) from 30 to
700 ◦ C at a heating rate of 10 ◦ C min− 1 under N2.
Lemongrass leaves (LGL) were collected from a local garden. Poly­ The Brunauer–Emmett–Teller (BET) surface area and the Bar­
lactide (PLA 4043D) was supplied by NatureWorks LLC (Cargill-Dow, ret–Joyer–Halenda (BJH) pore size distributions of the beads were
Minneapolis, MN). Dimethylacetamide (DMAc, 99.5 %), sodium hy­ analyzed using a high throughput surface area and porosity analyzer
droxide (NaOH, 99 %), acetone (98 %), and methyl orange (MO) were (Micromeritics, ASAP2460).
from RCI Labscan. Methylene blue (MB, 95 %) was from Loba Chemie. The point of zero charge (pHPZC) was determined following the
Maleic anhydride (MAH, 99 %) was purchased from Sigma-Aldrich. method reported by Jia et al. [24]. Solutions of 0.01 M NaCl were pre­
pared at pH 2–10 adjusted with HCl or NaOH solutions. Their pH values
2.2. Preparation of carboxylated microfiber from LGL (MLGL) were recorded as initial pH (pHinitial). A fixed dose of the biosorbent was
placed in bottles containing the NaCl solutions. After shaking for 24 h,
Modified lemongrass leaf (MLGL) was prepared as in our previous the pH values of supernatants were measured and recorded as final pH
report [23] with a slight adaptation. Briefly, LGL was cleaned in water, (pHfinal). The values of pHfinal were plotted against pHinitial and pHpzc
dried and cut into 1–2 cm lengths. The pieces of LGL were pretreated by was determined from the point at which pHfinal was equal to pHinitial.
immersion for 2 h in a NaOH aqueous solution of 5 wt% at 50 ◦ C. After
immersion, the extracted macrofibres (LGLF) were washed repeatedly 2.5. Adsorption study
with deionized water (DI water) until neutral pH was obtained. LGLF
was then dried until its weight was constant. To produce MLGL, LGLF The adsorption study was performed in a batch adsorption process. A
was mixed with MAH at an LGLF–MAH mass ratio of 1:10 and the measured amount of adsorbent was added to 10 mL of MB dye solution,
mixture was heated at 50 ◦ C until all the MAH was melted, at which and the mixture was agitated at a constant rate of 200 rpm at ambient

2
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

Fig. 1. (a) Preparation procedure of modified cellulose from lemongrass leaf (MLGL) (steps i–vi) for use in biocomposite adsorbent beads prepared by post
modification (PM) and one-pot (OP) methods. (b) The proposed reaction between cellulose in lemongrass macrofibers (LGLF) and maleic anhydride (MAH). (c) The
formation mechanism of the bead by non-solvent induced phase separation (NIPS) in the one-pot method (i-iii) and the proposed adsorption mechanism of MB on the
adsorbent bead (iv).

temperature. The effects on adsorption efficiency of contact time (0− 30 method of Ampawan et al. [22]. Adsorbent beads were placed in 10 mL
h), adsorbent dose (2–100 mg), and MB solution initial pH (3− 10) and of 10 mg L− 1 MB solution and agitated at ambient temperature for 24 h.
concentration (10–200 mg L− 1) were determined. The residual con­ To desorb the adsorbed dye, the MB-loaded beads were shaken in 0.5 M
centration of MB in the dye solution was estimated using UV–vis spec­ of acetic acid solution until the solution was colorless. The beads were
trophotometer (Lambda 25, PerkinElmer). The removal percentage (R, then washed with DI water, dried and used for the next adsorption cycle.
%) and the adsorption capacity (q, mg g− 1) of the adsorbent were
calculated from the following equations: 3. Results and discussion
C0 − Ct
R (%) = × 100 (2) 3.1. Characterization of fibers and adsorbent beads
C0

C0 − Ct The morphological transition from raw LGL to carboxylated-

q= ×V (3) cellulose was investigated by using SEM. LGL comprised tight bundles
m
of lignocellulose and non-lignocellulosic substances (Fig. 2a). The
where C0 and Ct (mg L− 1) are the concentration at the initial and selected smaller macrofibers (LGLF) obtained by pretreatment with NaOH were
times, respectively. V (mL) is the dye solution volume and m (g) is the around 55.13 ± 31.54 μm in diameter (Fig. 2b). This result indicated the
adsorbent mass. partial removal of lignin and hemicellulose by the cleavage of ester
The reusability of the adsorbent bead was investigated following the linkages between the two components during pretreatment [25]. The

3
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

Fig. 2. SEM images of (a) raw lemongrass leaf (LGL), (b) macrofiber extracted from LGL (LGLF), and (c) maleic anhydride-modified LGLF (MLGL). (d) FTIR spectra of
LGL-based fibers.

modification of LGLF with MAH to produce MLGL converted the mac­ obtained MLGL microfiber produced new peaks at 1724 cm− 1 and 1618
rofibres to microfibers with an average diameter of 8.79 ± 1.48 μm cm− 1 related to carbonyl and carboxyl stretching vibrations, respec­
(Fig. 2c). This result was consistent with our previous finding that MAH tively [28]. The appearance of these peaks indicated the presence of
not only converted hydroxyl groups on cellulose into carboxyl groups carboxyl groups and the successful synthesis of carboxylated cellulose.
but also eliminated residual hemicellulose and lignin during modifica­ The reaction scheme between cellulose in LGLF and MAH was proposed
tion [23]. in Fig. 1b [21,28].
The FTIR peaks of LGL at 3340, 1732 and 1032 cm− 1 were charac­ The biosorbent beads were prepared via NIPS from the solution of
teristic of lignocellulose [26] and respectively associated with the PLA loaded with MLGL by two different methods (Fig. 1a). The solution
stretching vibrations of hydroxyl (–OH), carbonyl (–C=O) and ether for the PM bead was prepared by adding dry MLGL into a solution of PLA
(–C–O) groups (Fig. 2d). LGL also produced peaks at 1516 cm− 1 and and DMAc. Therefore, MLGL had to be separated from the refluxed so­
1238 cm− 1 that were respectively related to the C– –C stretching of the lution, purified and dried before being added to the PLA and DMAc
aromatic ring and the C–O of the aryl group in lignin [26,27]. After LGL solution. The solution for the OP bead was prepared by adding PLA
was pretreated with NaOH to produce LGLF, the intensity of both these directly into the cooled refluxed solution containing MLGL, DMAc and
peaks and the peak at 1732 cm− 1 was reduced, indicating the partial uMAH. Therefore, DMAc and MAH were reused as a solvent and addi­
elimination of hemicellulose and lignin, and confirming the observa­ tive, respectively for the OP bead solution. Since the PLA-based solution
tions of the SEM study (Fig. 2b). After modifying LGLF with MAH, the for the PM and OP beads was prepared with different methods, the

4
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

maximum practical loading of MLGL was not the same for each solution. polymer-rich phase and a polymer-poor phase. These phases solidified,
The adsorption study showed that the PM bead disintegrated under forming the polymer matrix and the pores, respectively (Fig. 1c-iii) [32].
agitation if more than 20 wt% of MLGL was loaded in the PLA matrix, The pores within the polymer matrix of the beads and the non-swelling
and the OP bead disintegrated under agitation if more than 15 wt% of property of the PLA matrix (Fig. 4) enabled the beads to float on the dye
MLGL was loaded. The neat PLA, PM15, and OP15 adsorbent beads solution.
showed quite a similar appearance (Fig. S1a) with average diameters of In the NIPS process, the pores at the surface are generated with the
around 3.26 ± 0.21 mm. aid of a pore former (MAH in the present work) dissolved in water that
The adsorption capacities of LGL-based adsorbents and various leaches out during the phase inversion process (Fig. 1c-ii) [31,33].
adsorbent beads toward MB were investigated. The adsorption test was Therefore, the OP bead exhibited larger surface pores in comparison
performed using an adsorbent dose of 0.006 g and an initial MB con­ with other beads (Fig. 4a-c). The presence of hydrophilic MLGL together
centration of 10 mg L− 1. The high adsorption capacity of MLGL powder with MAH increased the rate of diffusional exchange between the sol­
was due to the carboxyl groups present on the cellulose (Fig. 3a). The vent and non-solvent [30,34], giving the OP bead a different cross-
LGL-based fibers powder exhibited higher adsorption capacities than the sectional morphology from the other beads (Fig. 4). The microfibers
adsorbent beads since the surface area of the powder was larger. How­ embedded in the matrix of the OP bead partially interacted with PLA in
ever, after adsorption, the dye loaded powders could not be easily the matrix (Fig. 4c). During the dissolution of PLA in MAH and DMAc,
separated from the solution (Fig. 3b) and could therefore present a anhydride groups of MAH might react with hydroxyl groups of PLA [35],
secondary pollutant problem. Centrifugation was necessary to separate and hydrogen bonding between the maleic acid groups and carboxyl and
the adsorbent and the process was impractical. Although the PM and OP hydroxyl groups in MLGL promoted the interaction between fiber and
beads removed less MB than the LGL and MLGL powders, both types of matrix. The presence of larger pores and MLGL on the surface of the
floating bead were easy to remove from the solution, facilitating their OP15 beads facilitated binding with the cationic MB dye.
regeneration and reuse (Fig. 3b). The PLA bead could slightly adsorb MB The nitrogen adsorption-desorption isotherm curves and pore size
through electrostatic interaction between positive charges in MB and distribution of all beads were presented in Fig. S2. The specific surface
partially negative charges on O atoms in carbonyl groups [19,29]. The area, pore size, and pore volume were listed in Table S1. The pore size
PM and OP beads adsorbed MB much better than the PLA bead due to the distribution of PLA-based beads was centered around 0.75 nm. The
presence of MLGL. ANOVA statistical comparisons indicated at a 95 % specific surface area of OP15 was around two fold the area of the PLA
confidence interval that the adsorption capacity of PM20 was higher and PM beads, indicating that uMAH in the solution structurally
than that of PM15. Although this result indicated that a higher MLGL changed the bead.
loading improved the adsorption capacity of the composite bead, the In the thermal analysis, all the beads demonstrated one decomposi­
adsorption capacity of the OP15 bead was higher than that of the PM20 tion step in the TGA and DTG curves (Fig. 5). The thermal decomposition
bead (p < 0.05). As well as showing a higher adsorption capacity than in the range of 300–420 ◦ C was attributed to the decomposition of cel­
the PM bead, the OP bead could be prepared using fewer steps and more lulose and PLA [36,37]. The temperatures at decomposition onset
quickly because the separation, washing and drying of MLGL were (Tonset) of PLA, PM15 and OP15 were 351.5, 336.0, and 357.0 ◦ C,
eliminated and some chemicals were reused. respectively, and the temperatures at maximum decomposition rate
The surfaces of PLA and PM15 beads were similar (Fig. S1b). At a (Tmax) were 369.8, 370.0 and 377.4 ◦ C, respectively. The OP bead
magnification of 3000×, many small pores of around 0.320 ± 0.007 μm showed the highest Tonset and Tmax. The increased thermal stability
could be seen (Fig. 4a and b). The OP15 bead (Fig. 4c) presented a larger demonstrated by these results was due to the partial compatibility be­
pore surface of around 1.69 ± 0.10 μm. At a magnification of 300×, tween PLA and MLGL observed in the SEM study (Fig. 4c).
pores could be seen covering the whole surface of the OP bead (Fig. S1b). The water uptake of the adsorbent beads was investigated by im­
The cross-sections of all beads (Fig. 4d-i) revealed a thin outer skin layer mersion in water for 24 h. The water absorption of the PLA bead was the
supported by finger-like structures that were generated by the NIPS lowest at 10.39 ± 4.08 %, while the water absorption of the PM15 and
process [30,31]. The inner porous structure comprised a discontinuous OP15 beads was 87.48 ± 7.04 % and 193.42 ± 12.59 %, respectively.
PLA matrix of flake-like structures (Fig. 4j-l). The formation mechanism The water absorption of the OP15 bead was twice that of the PM15 bead
of the OP bead during phase inversion represented how all the beads due to the modified microfiber and larger pores on the surface (Fig. 4c
were formed (Fig. 1c). After adding the PLA-based solution dropwise and Fig. S1b). Both properties facilitated the penetration of dye solution.
(Fig. 1c-i) into the non-solvent (DI water), the high mutual affinity be­ The results showed that the easier OP preparation method could
tween DMAc and water enabled the exchange between DMAc in the produce a floating bead with a high adsorption capacity. The one-pot
polymer solution drops and water in the coagulation bath (Fig. 1c-ii). method reduced preparation time, and energy and chemical consump­
Therefore, the demixing of previously solution was induced into a tion. Therefore, the OP15 bead was selected for further study.

3.2. Effect of contact time on adsorption

The adsorption of MB on the OP15 bead was studied from 0 to 30 h.

The adsorption capacity of OP15 bead increased rapidly at first (Fig. 6a),
owing to the large number of unoccupied binding sites, which were
favorable for dye adsorption. The adsorption rate slowed down and
reached equilibrium at 24 h since the binding sites were fully occupied
[38].
The adsorption kinetics of MB on the OP15 bead was studied by
fitting the data from the contact time study with the pseudo-first order
(PFO), pseudo-second order (PSO), and intraparticle diffusion (IPD)
kinetics models in Eq. (4) [39], (5) [40], and (6) [41], respectively:
ln (qe − qt ) = ln qe –k1 t (4)

Fig. 3. (a) Adsorption capacity of various biosorbents toward MB and (b) t 1 1

digital photograph of MB dye solution after adsorption by MLGL and OP15 = + t (5)
qt k2 q2e qe
adsorbent beads.

5
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

Fig. 4. SEM images of the surface morphology of various adsorbent beads (a-c) at a magnification of 3000×. SEM images of cross-sections of the beads at different
magnifications of (d-f) 30×, (g-i) 300×, and (j-l) 1000×. PLA = neat PLA; PM = PLA + 15 wt% MLGL; OP = PLA+ 15 wt% MLGL.

Fig. 5. (a) TGA and (b) DTG curves of PLA-based adsorbent beads.

bead (mg g− 1) at specific and equilibrium times, respectively; k1

qt = kid t0.5 + h (6)
(min− 1), k2 (g mg− 1 min− 1), and kid (mg g− 1 min-1/2) are respectively the
In these equations, qt and qe are the amount of dye adsorbed on the PFO, PSO, and IPD rate constants; and h (mg g− 1) is the intercept of

6
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

Fig. 6. (a) Effect of contact time on adsorption capacity of the OP15 bead toward MB. The kinetics adsorption on the OP15 bead by fitting data to (b) the pseudo-first
order (PFO), (c) the pseudo-second order (PSO), and (d) the intraparticle diffusion (IPD) models (C0 = 10 mg L− 1, dye solution pH = 7, and adsorbent dose =
0.006 g).

intraparticle diffusion at different stages. 3.3. Effect of adsorbent dose and dye solution pH
The experimental data fitted well with the PSO model (Fig. 6b and c).
The R2 value from the PSO model was higher than the R2 of the PFO A bigger adsorbent dose provides more active sites to bind with MB
model and close to 1, and the experimental qe was close to the qe molecules resulting in the increased removal of MB from water [44]. The
calculated from the PSO equation (Table 1). Therefore, the adsorption of removal of MB in the present study was insignificantly changed on an
MB on the OP15 bead was better explained by the PSO kinetics equation, adsorbent dose above 40 mg (Fig. 7a) since the available binding sites
implying that the adsorption process was governed by chemisorption exceeded the number of dye molecules in solution.
[42]. The influence of dye solution pH on the adsorption of MB was
The intraparticle diffusion model was used to analyze the diffusion investigated from pH 3–10. The adsorption of MB was lower in the acidic
mechanism during adsorption. As shown in Fig. 6d, the plot of qt vs. t0.5 condition than in the basic condition (Fig. 7b). To understand the effect
was non-linear and the fitting line did not pass through the origin. of dye solution pH on the adsorption of MB on the OP bead, pHpzc was
Intraparticle diffusion was, therefore, not the only rate-determining step evaluated. The pHpzc of the OP15 was at pH 5.1 (Fig. S3). When the
[43]. The two linear fitting lines in Fig. 6d indicated that MB was solution pH was lower than 5.1, the surface of the OP bead was posi­
adsorbed on the OP15 bead in two steps. The R2 and kid from the fitted tively charged. On the other hand, when the solution pH was higher than
lines were presented in Table 1. The initial stage with a steep slope was a 5.1, the surface of the bead became negatively charged. At low pH (pH
fast adsorption process involving the attraction of dye molecules by the < pHpzc), H+ ions were generated that competed with MB for active
active functional groups on the surface of the OP15 bead. The slow sites on the adsorbent, reducing dye removal [11]. At higher dye solu­
process in the second stage was related to intraparticle diffusion tion pH, the bead became negatively charged due to deprotonation. The
[12,42,43]. positively charged MB could then easily bind with the bead through
electrostatic forces and the removal of MB increased.

Table 1 3.4. Effect of initial dye concentration

Kinetic parameters of MB adsorption on the OP15 biosorbent.
Kinetics model Parameter Value The influence of initial dye concentration on the adsorption capacity
Experimental, qe (mg g¡1) – 12.07
of the OP15 bead was investigated from 10 to 200 mg L− 1. The
Pseudo-first order qe (mg g¡1) 6.572 adsorption capacity of a fixed dose of the adsorbent bead increased with
k1 (min− 1) 0.002 increments in dye concentration from 10 to 120 mg L− 1 (Fig. 8a). At
R2 0.978 these concentrations, the available binding sites were sufficient for the
Pseudo-second order qe (mg g¡1) 12.903
number of MB molecules in solution [43]. At higher initial dye con­
k2 × 103 (g mg− 1 min− 1) 0.567
R2 0.996 centrations, the adsorption capacity of the adsorbent did not increase
Intraparticle diffusion model kid,1 (mg g¡1 min-1/2) 0.235 since the number of available binding sites was insufficient to adsorb the
R2 0.947 increased number of MB molecules [13].
kid,2 (mg g¡1 min-1/2) 0.095 The adsorption behavior between MB and the OP15 bead was studied
R2 0.914
by fitting the experimental data to the Langmuir and Freundlich

7
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

Fig. 7. (a) The effect of adsorbent dose on the adsorption of MB at pH 7 and (b) the effect of MB solution pH on adsorption using an adsorbent dose of 0.006 g (C0 of
10 mg L− 1 and contact time of 24 h).

Fig. 8. (a) The effect of initial MB concentration on adsorption capacity at equilibrium on the OP15 bead. Adsorption isotherms of the OP15 bead by fitting data to
(b) the Langmuir and (c) the Freundlich isotherm models (dye solution pH = 8 and adsorbent dose = 0.006 g).

isotherm models, expressed by Eq. (7) and Eq. (8) [45,46], respectively:
Table 3
Ce Ce 1 The qm values of various composite adsorbents toward MB calculated from the
= + (7)
qe qm qm KL Langmuir isotherm model.
Adsorbent Adsorbent qm (mg T pH Ref
1 form g− 1) (◦ C)
lnqe = lnKF + lnCe (8)
n
Hexamethylenetetramine Bead 127.23 ~25 ~5 [9]
functionalized graphene
where Ce and qe are the dye concentration (mg L− 1) and adsorption
oxide-alginate
capacity (mg g− 1), respectively, in the equilibrium state; qm is the Carboxymethyl cellulose/k- Bead 10.75 30 6 [10]
maximum adsorption capacity of the adsorbent (mg g− 1), and KL is the carrageenan/ activated
Langmuir constant (L mg− 1). KF and n are, respectively, the Freundlich montmorillonite
adsorption constant ((mg g− 1)(L mg− 1)1/n) and the adsorption intensity. Sodium Microsphere 116.95 25 7 [13]
carboxymethylcellulose/
The fitted curves (Fig. 8b and c) and calculated parameters (Table 2) modified aminated lignin
indicated that the adsorption of MB on the OP15 bead followed the Chitosan/κ-carrageenan Membrane 18.8 50 4 [14]
Langmuir isotherm model. The qmax calculated from the Langmuir reinforced with activated
equation was 90.09 mg g− 1, which was close to the experimental qmax bentonite
PLA/PBAT/carboxylated Membrane 35.97 25 ~6 [22]
(86.19 mg g− 1). This result indicated that the adsorption of MB on the
cellulose microfiber from
homogeneous surface of the OP15 bead occurred in a monolayer manner empty fruit bunch (MEFB)
[13]. Alginate/compact discs waste- Bead 11.78 Troom 6 [47]
The maximum adsorption capacity of the proposed adsorbent bead derived activated carbon
was compared with the qm of other composite adsorbents (Table 3). The Poly(MAA-co-AAm)/cloisite N/A 32.83 25 8 [48]
30B nanocomposite
OP15 bead could be fabricated easily using few chemicals and its hydrogel
Cellulose acetate-acrylic acid- N/A 68.97 25 7 [49]
acrylamide
Table 2 PLA/carboxylated cellulose Bead 90.09 30 8 This
Isotherm parameters of MB adsorption on the OP15 biosorbent. from LGL (OP15) work
Isotherm model Parameter Value

Experimental, qm (mg g¡1) – 86.193 adsorption efficiency was comparable to other adsorbents.
Langmuir qm (mg g¡1) 90.090
KL (× 103) (L mg− 1) 0.702
R2 0.999 3.5. Effect of temperature
Freundlich KF (mg g− 1)(L mg − 1)1/n 18.946
n 2.823 The effect of temperature on the adsorption of MB by OP15 was
R2 0.855
investigated at 30, 40, 50, and 60 ◦ C. The adsorption capacity of the OP

8
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

bead increased with increments of dye solution temperature (Fig. 9a). same as the initial MO concentration. These results indicated that the
This result indicated that the adsorption of MB on the OP bead was OP15 bead did not adsorb the anionic MO dye and implied that the
endothermic [21]. Gibb’s free energy (ΔG◦ ) was then evaluated from Eq. charge of the bead was similar to the charge of MO and they repelled
(9), and enthalpy (ΔH◦ ) and entropy (ΔS◦ ) were determined from the each other.
slope and the intercept of the plots of ln Kc versus 1/T (Fig. 9b), The selectivity of the OP15 bead was studied by using a binary dye
respectively, using Eq. (10): solution of MB and MO at 10 mg L¡1 that was green in color (Fig. 11b).
After adsorption, the color and UV–Vis spectrum of the solution were
(9)
◦
ΔG = − RT ln Kc
similar to these of the original MO solution. The UV–Vis spectrum of MB
◦ ◦ was very low since approximately 96 % of MB had been adsorbed. This
ΔS ΔH
ln Kc = − (10) result is consistent with results reported for the adsorption of MB and
R RT repulsion of MO by a PVDF/GO nanofibrous membrane (NFM) that was
where Kc is the ratio of qe and Ce, R is an ideal gas constant (8.314 J also negatively charged [51]. A similar pattern was also observed in a
mol− 1 K− 1) and T is the absolute temperature (K). All calculated pa­ ternary dye solution of the cationic dyes MB and crystal violet (CV) and
rameters were listed in Table S2. the anionic dye MO (Fig. S5a). After adsorption, the intensity of the
The positive values of ΔH◦ and ΔS◦ signified that the adsorption absorption peaks corresponding to cationic MB and CV was significantly
process was endothermic and random at the solid-liquid interface. The lower while that of MO had hardly changed. The color of the ternary dye
negative values of ΔG◦ implied a spontaneous adsorption process. The solution after adsorption was slightly different from the color of MO
increases in absolute ΔG◦ values with increments of dye solution tem­ solution (Fig. S5b) because some residual MB and CV remained
perature indicated that the adsorption of MB on the bead was easier at (Fig. S5c). This result confirmed that the OP15 bead favorably adsorbed
higher temperatures. Similar results were reported for the adsorption of the positively charged dyes but repulsed the negatively charged MO.
MB on nonionic surfactant-modified MMT [2], CS/kC/activated BTN Regarding the pHpzc value, the charge of the OP15 bead in these studies
[14], and ZnO/starch biodegradable hydrogel [50] composite was negative since the studies were performed at dye solution pH >
adsorbents. pHpzc. Based on these findings, it was concluded that the negatively
charged OP15 bead selectively adsorbed the positively charged cationic
dye due to electrostatic interaction.
3.6. Reusability
3.8. Adsorption mechanism
The reuse of the floating adsorbent bead was studied by repeating the
adsorption of MB five times (Fig. 10a). An aqueous solution of 0.5 M
To further elucidate the adsorption mechanism, the OP15 was
acetic acid was the desorbing agent since MB adsorbed on beads could
analyzed by FTIR spectroscopy before and after adsorption of MB. The
be substituted by the H+ ion in acetic acid. The adsorption of MB
spectra (Fig. S6) showed that peak of the C– –O stretching vibration in
decreased slightly from the first to fifth adsorption cycles. The OP15
carboxyl groups at 1618 cm− 1 was shifted to 1623 cm− 1 after adsorption
bead could remove 88 % of MB over five cycles. The gradual decrease in
and the characteristic peak of the –OH group at 3340 cm− 1 was shifted
adsorption capacity over the five cycles was caused by the loss of
to 3330 cm− 1. The overlapping peaks associated with the –C=O group of
binding sites. The beads retained their shape without damage after the
PLA and MLGL at 1749 cm− 1 were shifted to 1756 cm− 1. Moreover, the
fifth cycle (Fig. 10b and c). As shown in Fig. S4, the surface morphology
intensity of some peaks was lower after adsorption and a new peak was
of the OP15 bead after the 5th cycle of MB adsorption-desorption was
observed at 1604 cm− 1, attributed to the CH– –N of MB [52]. The
similar to the original bead (Fig. 4c). The stability of the OP15 bead was
changes in the IR spectra demonstrated that –OH, –COOH, and –C=O
studied by using the same method used in the reusability test but using
groups on MLGL in the OP15 bead were involved in the interaction with
water, with a similar pH to the dye solution, instead of dye solution to
MB [53].
eliminate weight gained from adsorbed dye on the bead. After the fifth
An FTIR analysis and the studies of the effects of pH dye solution and
cycle, the weight loss of beads was only 1 %. These results suggested that
pHpzc value revealed that the OP bead became negatively charged in the
the OP15 bead could be reused.
basic condition due to the deprotonation of –COOH groups. Therefore,
the adsorption of MB occurred through electrostatic interactions be­
3.7. Dye selectivity tween the negative charges of –COO− groups in the bead and positive
charges (N+) of MB. Hydrogen bonding could also occur between H
The adsorption of anionic methyl orange (MO) on the OP15 bead was atoms of the –OH and –COOH groups in the bead and N or S atoms in MB
studied in a 10 mg L¡1 solution of MO (Fig. 11a). The color of the MO molecules, and ion-dipole forces could generate electrostatic in­
solution was similar before and after adsorption as was the UV–Vis teractions between positive charges in MB and partially negative
spectrum. Moreover, the MO concentration after adsorption was the charges on O atoms of the –C=O groups in the bead. The proposed

Fig. 9. (a) Effect of temperature on adsorption capacity and (b) Plot of the ln Kc versus 1/T (C0 = 10 mg L− 1, dye solution pH = 7, and adsorbent dose = 0.006 g).

9
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

Fig. 10. (a) Removal of MB over five cycles using 0.5 M acetic acid solution as a desorbing agent, (b) virgin OP15 beads and OP15 beads after the fifth adsorption
(C5), and (c) dye solution after the first (C1) and fifth (C5) adsorption on OP15 beads.

Fig. 11. UV–Vis spectra and colors of (a) methyl orange (MO) solution and (b) mixed MB and MO solutions before and after adsorption on the OP15 adsorbent bead
(dose of 0.04 g, C0 = 10 mg L¡1, pH 6.5).

adsorption mechanism was presented in Fig. 1c-iv. CRediT authorship contribution statement

4. Conclusion Sasimaporn Ampawan: Writing – original draft, Investigation.

Jiratchaya Dairoop: Investigation. Mallika Keawbanjong: Investiga­
A floating adsorbent bead was prepared by a one-pot method using tion. Watchanida Chinpa: Writing – review & editing, Writing – orig­
polylactide and maleic anhydride-modified cellulose. Polylactide was inal draft, Supervision, Project administration, Methodology,
directly combined with modified cellulose, avoiding the usual need to Conceptualization.
purify and dry the modified cellulose before incorporating the two parts
of the composite. The unreacted maleic anhydride and dimethylaceta­
Declaration of competing interest
mide remaining from the cellulose modification process were reused in
the non-solvent induced phase separation process that formed the bead.
The authors declare that they have no known competing financial
The modified cellulose microfiber exhibited higher adsorption capacity
interests or personal relationships that could have appeared to influence
than unmodified fiber. The loading of modified cellulose in the poly­
the work reported in this paper.
lactide matrix increased the adsorption of methylene blue compared
with pure polylactide beads. In the one-pot synthesis, maleic anhydride
Acknowledgements
in the polylactide solution increased surface pore size and produced a
bead whose structure facilitated the adsorption of dye. The OP bead
Ms. Sasimaporn Ampawan was supported by Graduate Fellowship
exhibited better water absorption, thermal stability and methylene blue
(Research Assistant), Faculty of Science, Prince of Songkla University,
adsorption than conventionally prepared adsorbent beads. The adsorp­
Contract no. 1-2565-02-006. We also acknowledge Mr. Thomas Coyne
tion behavior of methylene blue on the bead was best described by the
for his kind assistance in editing the English text.
pseudo second-order model and Langmuir isotherm model with a
maximum adsorption capacity of 90.09 mg⋅g− 1 at pH 8. The proposed
Appendix A. Supplementary data
bead could be used to adsorb methylene blue over five cycles without
disintegration. In a binary and ternary dye solution of cationic dyes
Supplementary data to this article can be found online at https://doi.
(methylene blue and crystal violet) and anionic methyl orange, the
org/10.1016/j.ijbiomac.2024.131354.
proposed bead selected to bind with cationic dye, rejecting anionic
methyl orange. The proposed fabrication process reduces operation time
and chemical use, producing a reusable, green, floating adsorbent bead References
that effectively removes cationic dye from water and is easy to recover
[1] A. Benhouria, H. Zaghouane-Boudiaf, R. Bourzami, F. Djerboua, B.H. Hameed,
from the dye solution. M. Boutahala, Cross-linked chitosan-epichlorohydrin/bentonite composite for
reactive orange 16 dye removal: experimental study and molecular dynamic
simulation, Int. J. Biol. Macromol. 242 (2023) 124786, https://doi.org/10.1016/j.
ijbiomac.2023.124786.

10
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

[2] G. Wang, S. Wang, Z. Sun, S. Zheng, Y. Xi, Structures of nonionic surfactant [24] Y.F. Jia, B. Xiao, K.M. Thomas, Adsorption of metal ions on nitrogen surface
modified montmorillonites and their enhanced adsorption capacities towards a functional groups in activated carbons, Langmuir 18 (2002) 470–478, https://doi.
cationic organic dye, Appl. Clay Sci. 148 (2017) 1–10, https://doi.org/10.1016/j. org/10.1021/la011161z.
clay.2017.08.001. [25] S. Sun, S. Sun, X. Cao, R. Sun, The role of pretreatment in improving the enzymatic
[3] K. Tohdee, L. Kaewsichan, Asadullah, Enhancement of adsorption efficiency of hydrolysis of lignocellulosic materials, Bioresour. Technol. 199 (2016) 49–58,
heavy metal Cu(II) and Zn(II) onto cationic surfactant modified bentonite, https://doi.org/10.1016/j.biortech.2015.08.061.
J. Environ. Chem. Eng. 6 (2018) 2821–2828, https://doi.org/10.1016/j. [26] H. Kargarzadeh, I. Ahmad, I. Abdullah, A. Dufresne, S.Y. Zainudin, R.M. Sheltami,
jece.2018.04.030. Effects of hydrolysis conditions on the morphology, crystallinity, and thermal
[4] G.J. Vadakkekara, S. Thomas, C.P.R. Nair, Maleic acid modified cellulose for stability of cellulose nanocrystals extracted from kenaf bast fibers, Cellulose 19
scavenging lead from water, Int. J. Biol. Macromol. 129 (2019) 293–304, https:// (2012) 855–866, https://doi.org/10.1007/s10570-012-9684-6.
doi.org/10.1016/j.ijbiomac.2019.02.037. [27] F. Fahma, S. Iwamoto, N. Hori, T. Iwata, A. Takemura, Isolation, preparation, and
[5] E. Khalilzadeh Shirazi, J.W. Metzger, K. Fischer, A.H. Hassani, Removal of textile characterization of nanofibers from oil palm empty-fruit-bunch (OPEFB), Cellulose
dyes from single and binary component systems by Persian bentonite and a mixed 17 (2010) 977–985, https://doi.org/10.1007/s10570-010-9436-4.
adsorbent of bentonite/charred dolomite, Colloids Surf. A Physicochem. Eng. Asp. [28] S. Cichosz, A. Masek, K. Wolski, M. Zaborski, Universal approach of cellulose fibres
598 (2020) 124807, https://doi.org/10.1016/j.colsurfa.2020.124807. chemical modification result analysis via commonly used techniques, Polym. Bull.
[6] Y. Tang, T. Lin, C. Jiang, Y. Zhao, S. Ai, Renewable adsorbents from carboxylate- 76 (2019) 2147–2162, https://doi.org/10.1007/s00289-018-2487-7.
modified agro-forestry residues for efficient removal of methylene blue dye, [29] A. Nouri, W.L. Ang, E. Mahmoudi, S.F. Chua, A.W. Mohammad, A. Benamor, M.
J. Phys. Chem. Solids 149 (2021) 109811, https://doi.org/10.1016/j. M. Ba-Abbad, C.P. Leo, Decoration of polylactic acid on graphene oxide for
jpcs.2020.109811. efficient adsorption of methylene blue and tetracycline, Chemosphere 322 (2023)
[7] C. Song, C. Gao, P. Fatehi, S. Wang, C. Jiang, F. Kong, Influence of structure and 138219, https://doi.org/10.1016/j.chemosphere.2023.138219.
functional group of modified kraft lignin on adsorption behavior of dye, Int. J. Biol. [30] P. Lin, C. Liu, J. Li, J. Qi, R. Luo, X. Sun, J. Shen, W. Han, L. Wang, Nanosized
Macromol. 240 (2023) 124368, https://doi.org/10.1016/j.ijbiomac.2023.124368. amine-rich spheres embedded polymeric beads for Cr (VI) removal, J. Colloid
[8] X. Dai, C. Jing, K. Li, X. Zhang, D. Song, L. Feng, X. Liu, H. Ding, H. Ran, K. Zhu, Interface Sci. 508 (2017) 369–377, https://doi.org/10.1016/j.jcis.2017.08.064.
N. Dai, S. Yi, J. Rao, Y. Zhang, Enhanced bifunctional adsorption of anionic and [31] L. Li, J. Yao, P. Xiao, J. Shang, Y. Feng, P.A. Webley, H. Wang, One-step fabrication
cationic pollutants by MgAl LDH nanosheets modified montmorillonite via acid- of ZIF-8/polymer composite spheres by a phase inversion method for gas
salt activation, Appl. Clay Sci. 233 (2023) 106815, https://doi.org/10.1016/j. adsorption, Colloid Polym. Sci. 291 (2013) 2711–2717, https://doi.org/10.1007/
clay.2023.106815. s00396-013-3024-8.
[9] D. Lotfy, G.S. El-Sayyad, N. Shehata, Hexamethylenetetramine functionalized [32] J.A. Ronner, S.G. Wassink, C.A. Smolders, Investigation of liquid-liquid demixing
graphene oxide-alginate beads nanocomposite as efficient sorbent for dye from and aggregate formation in a membrane-forming system by means of pulse-
aqueous solution, Int. J. Biol. Macromol. 228 (2023) 754–772, https://doi.org/ induced critical scattering (PICS), J. Membr. Sci. 42 (1989) 27–45, https://doi.org/
10.1016/j.ijbiomac.2022.12.208. 10.1016/S0376-7388(00)82363-7.
[10] C. Liu, A.M. Omer, X.-k. Ouyang, Adsorptive removal of cationic methylene blue [33] C.-C. Ho, J.F. Su, Boosting permeation and separation characteristics of
dye using carboxymethyl cellulose/k-carrageenan/activated montmorillonite polyethersulfone ultrafiltration membranes by structure modification via dual-PVP
composite beads: isotherm and kinetic studies, Int. J. Biol. Macromol. 106 (2018) pore formers, Polymer 241 (2022) 124560, https://doi.org/10.1016/j.
823–833, https://doi.org/10.1016/j.ijbiomac.2017.08.084. polymer.2022.124560.
[11] A. Kausar, S.U. Rehman, F. Khalid, A. Bonilla-Petriciolet, D.I. Mendoza-Castillo, H. [34] A.I. Nazri, A.L. Ahmad, M.H. Hussin, Microcrystalline cellulose-blended
N. Bhatti, S.M. Ibrahim, M. Iqbal, Cellulose, clay and sodium alginate composites polyethersulfone membranes for enhanced water permeability and humic acid
for the removal of methylene blue dye: experimental and DFT studies, Int. J. Biol. removal, Membranes 11 (2021) 660, https://doi.org/10.3390/
Macromol. 209 (2022) 576–585, https://doi.org/10.1016/j.ijbiomac.2022.04.044. membranes11090660.
[12] G. Viscusi, E. Lamberti, G. Gorrasi, Design of a hybrid bio-adsorbent based on [35] X. Li, Q. Yang, K. Zhang, L. Pan, Y. Feng, Y. Jia, N. Xu, Property improvement and
Sodium Alginate/Halloysite/Hemp hurd for methylene blue dye removal: kinetic compatibilization mechanism of biodegradable polylactic acid/maleic anhydride-
studies and mathematical modeling, Colloids Surf. A Physicochem. Eng. Asp. 633 based/polypropylene spunbonded nonwoven slices, J. Clean. Prod. 375 (2022)
(2022) 127925, https://doi.org/10.1016/j.colsurfa.2021.127925. 134097, https://doi.org/10.1016/j.jclepro.2022.134097.
[13] Z. Zhang, F. Li, J.W. Heo, J.W. Kim, M.S. Kim, Q. Xia, Y.S. Kim, Decoration of [36] F. Tamaddon, S. Hosseinzadeh, Anionic micro-cellulose (AMC): preparation,
sodium carboxymethylcellulose gel microspheres with modified lignin to enhanced characterization, and application as a novel heterogeneous base catalyst, Cellulose
methylene blue removal, Int. J. Biol. Macromol. 242 (2023) 125041, https://doi. 25 (2018) 5277–5287, https://doi.org/10.1007/s10570-018-1939-4.
org/10.1016/j.ijbiomac.2023.125041. [37] T. Suparanon, N. Phusunti, W. Phetwarotai, Properties and flame retardancy of
[14] A. Ulu, M. Alpaslan, A. Gultek, B. Ates, Eco-friendly chitosan/κ-carrageenan polylactide composites incorporating tricresyl phosphate and modified
membranes reinforced with activated bentonite for adsorption of methylene blue, microcrystalline cellulose from oil palmempty fruit bunch waste, Int. J. Biol.
Mater. Chem. Phys. 278 (2022) 125611, https://doi.org/10.1016/j. Macromol. 253 (2023) 127580, https://doi.org/10.1016/j.ijbiomac.2023.127580.
matchemphys.2021.125611. [38] R.S. Bangari, A. Yadav, J. Bharadwaj, N. Sinha, Boron nitride nanosheets
[15] Y. Feng, Y. Wang, Y. Wang, S. Liu, J. Jiang, C. Cao, J. Yao, Simple fabrication of incorporated polyvinylidene fluoride mixed matrix membranes for removal of
easy handling millimeter-sized porous attapulgite/polymer beads for heavy metal methylene blue from aqueous stream, J. Environ. Chem. Eng. 10 (2022) 107052,
removal, J. Colloid Interface Sci. 502 (2017) 52–58, https://doi.org/10.1016/j. https://doi.org/10.1016/j.jece.2021.107052.
jcis.2017.04.086. [39] S. Lagergren, About the theory of so called adsorption of soluble substances, in:
[16] J. Abdi, H. Abedini, MOF-based polymeric nanocomposite beads as an efficient K. Sven (Ed.), Vetenskapsakad vol. 24, Handlingar Band, 1898, pp. 1–39.
adsorbent for wastewater treatment in batch and continuous systems: modelling [40] Y.S. Ho, G. Mckay, Pseudo-second order model for sorption processes, Process
and experiment, Chem. Eng. J. 400 (2020) 125862, https://doi.org/10.1016/j. Biochem. 34 (1999) 451–465.
cej.2020.125862. [41] T. Hu, Q. Liu, T. Gao, K. Dong, G. Wei, J. Yao, Facile preparation of tannic acid-
[17] H. Singh, A. Goyal, S.K. Bhardwaj, M. Khatri, N. Bhardwaj, Highly robust UiO-66@ poly(vinyl alcohol)/sodium alginate hydrogel beads for methylene blue removal
PVDF metal–organic framework beads for tartrazine removal from aqueous from simulated solution, ACS Omega 3 (2018) 7523–7531, https://doi.org/
solutions, Mater. Sci. Eng. B 288 (2023) 116165, https://doi.org/10.1016/j. 10.1021/acsomega.8b00577.
mseb.2022.116165. [42] A. Saning, C. Thanachayanont, L. Suksai, W. Watcharin, S. Techasakul,
[18] V. Vatanpour, A. Dehqan, S. Paziresh, S. Zinadini, A.A. Zinatizadeh, I. Koyuncu, L. Chuenchom, D. Dechtrira, Green magnetic carbon/alginate biocomposite beads
Polylactic acid in the fabrication of separation membranes: a review, Sep. Purif. from iron scrap waste for efficient removal of textile dye and heavy metal, Int. J.
Technol. 296 (2022) 121433, https://doi.org/10.1016/j.seppur.2022.121433. Biol. Macromol. 261 (2024) 129765, https://doi.org/10.1016/j.
[19] E. Shokri, B. Khanghahi, E. Esmizadeh, H. Etemadi, Biopolymer-based adsorptive ijbiomac.2024.129765.
membrane for simultaneous removal of cationic and anionic heavy metals from [43] S. Mallakpour, E. Azadi, M. Dinari, Removal of cationic and anionic dyes using Ca-
water, Int. J. Environ. Sci. Technol. 19 (2022) 4167–4180, https://doi.org/ alginate and Zn-Al layered double hydroxide/metal-organic framework,
10.1007/s13762-021-03592-9. Carbohydr. Polym. 301 (2023) 120362, https://doi.org/10.1016/j.
[20] F.A. Pavan, E.S. Camacho, E.C. Lima, G.L. Dotto, V.T.A. Branco, S.L.P. Dias, carbpol.2022.120362.
Formosa papaya seed powder (FPSP): preparation, characterization and [44] F. Haq, A. Farid, N. Ullah, M. Kiran, R.U. Khan, T. Aziz, S. Mehmood, M. Haroon,
application as an alternative adsorbent for the removal of crystal violet from M. Mubashir, A. Bokhari, L.F. Chuah, P.L. Show, A study on the uptake of
aqueous phase, J. Environ. Chem. Eng. 2 (2014) 230–238, https://doi.org/ methylene blue by biodegradable and eco-friendly carboxylated starch grafted
10.1016/j.jece.2013.12.017. polyvinyl pyrrolidone, Environ. Res. 215 (2022) 114241, https://doi.org/10.1016/
[21] Y. Zhou, Y. Min, H. Qiao, Q. Huang, E. Wang, T. Ma, Improved removal of j.envres.2022.114241.
malachite green from aqueous solution using chemically modified cellulose by [45] I. Langmuir, The adsorption of gases on plane surfaces of glass, mica, and platinum,
anhydride, Int. J. Biol. Macromol. 74 (2015) 271–277, https://doi.org/10.1016/j. J. Am. Chem. Soc. 40 (1918) 1361–1403, https://doi.org/10.1021/ja02242a004.
ijbiomac.2014.12.020. [46] H.M.F. Freundlich, Over the adsorption in solution, J. Phys. Chem. 57 (1906)
[22] S. Ampawan, N. Phreecha, S. Chantarak, W. Chinpa, Selective separation of dyes by 385–470.
green composite membrane based on polylactide with carboxylated cellulose [47] S. Mallakpour, V. Behranvand, Methylene blue contaminated water sanitization
microfiber from empty fruit bunch, Int. J. Biol. Macromol. 225 (2023) 1607–1619, with alginate/compact discs waste-derived activated carbon composite beads:
https://doi.org/10.1016/j.ijbiomac.2022.11.218. adsorption studies, Int. J. Biol. Macromol. 180 (2021) 28–35, https://doi.org/
[23] S. Madsulaiman, S. Ampawan, P. Boonsuk, W. Chinpa, Preparation and 10.1016/j.ijbiomac.2021.03.044.
characterization of cationic dye scavenger from lemongrass leaf biomass, Biomass [48] H. Safarzadeh, S.J. Peighambardoust, S.H. Mousavi, R. Foroutan, R. Mohammadi,
Convers. Bioref. (2023), https://doi.org/10.1007/s13399-023-04369-5. S.H. Peighambardoust, Adsorption ability evaluation of the poly(methacrylic acid-

11
S. Ampawan et al. International Journal of Biological Macromolecules 266 (2024) 131354

co-acrylamide)/cloisite 30B nanocomposite hydrogel as a new adsorbent for [51] A. Ghaffar, L. Zhang, X. Zhu, B. Chen, Porous PVdF/GO nanofibrous membranes for
cationic dye removal, Environ. Res. 212 (2022) 113349, https://doi.org/10.1016/ selective separation and recycling of charged organic dyes from water, Environ.
j.envres.2022.113349. Sci. Technol. 52 (2018) 4265–4274, https://doi.org/10.1021/acs.est.7b06081.
[49] J. Rana, G. Goindi, N. Kaur, S. Krishna, A. Kakati, Synthesis and application of [52] A.A. Alshehri, M.A. Malik, Biogenic fabrication of ZnO nanoparticles using
cellulose acetate-acrylic acid-acrylamide composite for removal of toxic methylene Trigonella foenum-graecum (Fenugreek) for proficient photocatalytic degradation of
blue dye from aqueous solution, J. Water Process Eng. 49 (2022) 103102, https:// methylene blue under UV irradiation, J. Mater. Sci. Mater. Electron. 30 (2019)
doi.org/10.1016/j.jwpe.2022.103102. 16156–16173, https://doi.org/10.1007/s10854-019-01985-8.
[50] K. Jumnong, P. Kongseng, P. Maijan, S. Suwanboon, S. Chantarak, Double-function [53] P.D. Saha, S. Chakraborty, S. Chowdhury, Batch and continuous (fixed-bed
ZnO/starch biodegradable hydrogel composite for methylene blue adsorption and column) biosorption of crystal violet by Artocarpus heterophyllus (jackfruit) leaf
photocatalytic degradation, Int. J. Biol. Macromol. 253 (2023) 127533, https:// powder, Colloids Surf. B: Biointerfaces 92 (2012) 262–270, https://doi.org/
doi.org/10.1016/j.ijbiomac.2023.127533. 10.1016/j.colsurfb.2011.11.057.

12