Journal Information
Vol. 8. Issue 6.
Pages 5091-5102 (November - December 2019)
Download PDF
More article options
Vol. 8. Issue 6.
Pages 5091-5102 (November - December 2019)
Original Article
DOI: 10.1016/j.jmrt.2019.08.030
Open Access
Electrospun biocomposite: nanocellulose and chitosan entrapped within a poly(hydroxyalkanoate) matrix for Congo red removal
Chu Yong Soona, Norizah Abdul Rahmanb, Yee Bond Teec, Rosnita A. Talibd, Choon Hui Tana, Khalina Abdane, Eric Wei Chiang Chana,
Corresponding author

Corresponding author.
a Department of Food Science with Nutrition, Faculty of Applied Sciences, UCSI University, 56000 Cheras, Kuala Lumpur, Malaysia
b Department of Chemistry, Faculty of Science, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia
c Department of Science, Centre for Pre-University Studies, Stella Maris International School, Medan Damansara, 50490 Kuala Lumpur, Malaysia
d Department of Process and Food Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia
e Department of Biological and Agricultural Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia
Article information
Full Text
Download PDF
Figures (5)
Show moreShow less
Tables (4)
Table 1. Compositions of the Pickering emulsion mixtures in different formulations.
Table 2. The TGA profile of the electrospun biocomposites with the range of 25°C to 600°C at constant heating rate of 10°Cmin−1 under nitrogen atmosphere.
Table 3. Isotherm parameters for the adsorption of Congo red dye by PHA2NCC and PHA2Cts. (Adsorbent dosage=4g L−1).
Table 4. Kinetic parameters for the adsorption of Congo red dye by electrospun biocomposites. (Dye concentration=100mg L−1, adsorbent dosage=4g L−1).
Show moreShow less

Nano adsorbent possess notable adsorption capabilities but is difficult to recover in wastewater treatment processes. To overcome this limitation, the entrapment of nanocellulose (NCC) and chitosan (Cts) within poly(hydroxyalkanoate) (PHA) via electrospinning is proposed. The Pickering emulsion stabilized with Tween 80 formed a homogeneous NCC-Cts-PHA mixture prior to electrospinning. The resulting electrospun biocomposites were characterized with SEM, FT-IR, XrD and TGA. The electrospun biocomposites were with high porosity, rendering exposure of NCC and Cts to dye adsorption. The incorporation of nanocellulose and chitosan significantly increased the crystallinity of the electrospun biocomposites from 57.6% to 70.5%. The adsorption of Congo red dye by electrospun biocomposites fitted well with the Langmuir isotherm model and pseudo-second order kinetics, indicating a chemisorption nature. PHA2NCC (30.9%) has 3-fold higher dye removal percentage than that of PHA2Cts (10.5%). The results showed that Pickering emulsion is electrospinnable and recorded highest dye removal percentage in PHA3NCC1Cts (75.8%).

Pickering emulsion
Full Text

With the rapid industrialisation, untreated wastewater is becoming a global concern. Discharge of untreated wastewater causes irreversible disruption to ecosystem and would eventually affect human health [1]. The mobility and distribution of dyes has been studied due to their toxicity and ability to bio-accumulate, as well as their major impact on environmental aesthetics. They have been shown to bio-accumulate and react with metal ions to form micro-toxins [2]. Congo red, for instance, possesses a benzidine chromophore which is a known human carcinogen and mutagen, with the potential to bio-accumulate [3]. Hence, efficient dye removal remains an important aspect in wastewater treatment.

To date, several methods for dye removal have been explored using a range of physical, chemical and biological approaches. Methods such as reverse osmosis, ion exchange, precipitation, filtration, coagulation/flocculation, adsorption, catalytic and enzymatic oxidation have been well studied. Concurrently, these methods have their own drawbacks [4]. Adsorption is one of the most popular processes in advanced wastewater treatment as it is simple and highly efficient. The current research on adsorption focusses on bioadsorbents derived from natural materials, industrial solid waste, agricultural and aquacultural waste [5].

Cellulose and chitin are the most abundant polymer in the world. The cellulose microgel has been widely studied and proven as an excellent adsorbent for anionic dyes removal [6]. Furthermore, cellulose can be fabricated with graphene, alginate, synthetic polymers, grafted with amino-based functional groups, and so on to enhance the adsorption capability and efficiency [7–11]. Meanwhile, chitin is the precursor for the chitosan. Aside of being a low cost and easily available, chitosan has high contents of amino and hydroxyl functional groups that make it a high adsorption material in wastewater treatment [12,13]. Moreover, chitosan derivatives are preferably used in various industries due to its antimicrobial, antioxidant, biocompatible and eco-friendly properties [14,15]. Both cellulose and chitosan adsorbents could be regenerated via monoacid and monobasic washing, granting the potential of adsorbents as reusable materials [16–19].

In this context, the adsorption ability is dependent with the total adsorption active site at the surface area of adsorbent. However, the recovery of the nanoadsorbent for the regeneration purpose becomes prominently difficult. Entrapment of the nanoadsorbent in a matrix greatly simplifies the recovery process. Hence, nanocellulose and chitosan could be entrapped by poly(hydroxyalkanoate) (PHA) to form a porous biocomposite using electrospinning technique. Electrospun membranes are non-woven fabrics with high surface area to volume ratio, adjustable fiber topographies and morphologies [20]. PHA is selected to entrapment matrix because it is renewable, biocompatible and biodegradable [21].

However, a major hurdle for the entrapment of nanocellulose and chitosan within a PHA matrix is the occurrence of phase separation between PHA and the two composites during sample preparation. PHA is dissolved in chloroform, representing the hydrophobic oil phase whereas nanocellulose (NCC) and chitosan (Cts) are dispersed and dissolved in the water, representing the hydrophilic water phase.

This study uses an innovative approach to produce an electrospinnable homogeneous mixture of nanocellulose, chitosan and PHA by cold ultra-sonication of immiscible mixture into homogenized Pickering emulsion. This approach would allow for the efficient entrapment of substances within polymers matrix that exhibits low or no affinities to each other [20]. The rationales to this approach were the following: (1) nanocellulose and chitosan could be the colloidal particles for water-in-oil or oil-in-water Pickering emulsion that could render a stable and homogeneous emulsion, (2) PHA in high volatile chloroform crystallized faster than nanocellulose and chitosan, forming a porous matrix to entrap the adsorbents before agglomeration, (3) non-agglomerated particles are well-dispersed in PHA matrix with greater number of total active site and surface area for maximal adsorption.

The use of electrospun fabric with nano adsorbents is relatively uncommon. The PHA biocomposite with nanocellulose (NCC) and chitosan (Cts) was engineered using Pickering emulsion- electrospinning method. The morphological (SEM), physicochemical (FT-IR, XRD) and thermal (TGA) properties of the electrospun biocomposites were characterized. The Congo red dye adsorption processes of these electrospun biocomposites were evaluated. The output of this present research might widen both research and industrial selection of adsorbent filter production by using electrospinning of Pickering emulsion.


Poly(hydroxylalkanoate) (PHA) 3mm granule resins, condition elastomer (GF74310634) was purchased from GoodFellow. Nanocellulose (NCC) was prepared from corn cob obtained from Marine Gold Markerting Sdn. Bhd., Malaysia. The commercial chitosan (Cts) medium size (CAS: 9012-76-4) and Tween 80 (CAS: 9005-65-6) were purchased from Sigma Aldrich, USA. Congo red (CR) powder [1-naphthalenesulfonic acid, 3,3′-(4,4′-biphe-nylenebis (azo)) bis (4-amino-) disodium salt] was purchased from Merck. Analytical grade solvents and reagent used include sodium chlorite (NaClO2), acetic acid (CH3COOH), sodium hydroxide (NaOH), sulfuric acid (H2SO4) and chloroform (CHCl3).

2.2Preparation of emulsion solution for electrospinning

A 10% PHA stock was prepared via the dissolution of PHA resins in CHCl3. The corn cob waste was oven dried, grinded into powder using and sieved (40 mesh). The untreated corn cob powder was bleached with acidified 10% NaClO2 at the ratio of 1:20 (dry powder weight: volume of bleaching agent) under continuous stirring condition at 80°C for 2h. The delignified cellulose was washed with distilled water several times and then refluxed with 5% NaOH for 3h for the removal of hemicellulose. The excess NaOH was removed by several centrifugal washing with distilled water until the neutral pH of microcellulose pulp was obtained. The freeze-dried microcellulose powder was hydrolysed with 60% of H2SO4 at 45°C for 45min.

The particle size of the nanocellulose (1175.3±99.9nm) was determined via Zetasizer (Zetasizer nano ZS, Malvern). The 0.5% nanocellulose suspension was sonicated into NCC stock. The 0.5% chitosan was dispersed in pH 4 CH3COOH-acidified solution into Cts stock. Different formulations of Pickering emulsion were prepared according to Table 1. The Pickering emulsion was sonicated for 10min with 90% intensity, 10s pulse on and 5s pulse off sonication in the ice water bath to minimize the crystallization of PHA.

Table 1.

Compositions of the Pickering emulsion mixtures in different formulations.

Name  PHA, % wt  NCC, % wt  Cts, % wt  Tween 80, mL 
PHA  100 
PHA2NCC  98  0.1 
PHA2Cts  98  0.1 
PHA1NCC3Cts  96  0.1 
PHA2NCC2Cts  96  0.1 
PHA3NCC1Cts  96  0.1 

PHA, Poly(hydroxyalkanoate); NCC, Nanocrystalline cellulose; Cts, Chitosan.

2.3Fabrication of PHA electrospun biocomposite by electrospinning

The surgical syringe with 5mL of the Pickering emulsion was installed to the syringe pump with controlled flowrate 4mL/h. The syringe nozzle was attached with high-voltage electricity convertor by alligator clip. The voltage was slowly tuned to 20kV. The emulsion droplet at the nozzle tip was charged and collected at the static collector (aluminium foil) with 10cm distance from the nozzle tip. The temperature and relative humidity were controlled at 25.5±0.5°C and 55±5%, respectively throughout the electrospinning process. The electrospun biocomposite was peeled off from the aluminium foil and stored in dry area for physiochemical characterizations and adsorption test.

2.4Characterization tests

A scanning electron microscope (SEM) (S-3400N SEM Hitachi, Japan) was used to observe the microstructure and morphology of the electrospun biocomposites. The acceleration voltage was fixed at of 15V to ensure high resolution image. The structural study of the electrospun biocomposites was conducted using Fourier transform-infrared (FTIR) (Nicolet iS5 FT-IR Thermo Scientific, US) with the attenuated total reflectance (ATR) technique in the range of wavelength 4000 cm−1 to 600cm−1 with 4cm−1 spectral resolution. A total of 64 scans were conducted for each sample. The crystallinity of the electrospun biocomposite was identified using X-ray diffraction (XRD-6000 Shimadzu, Japan) employing Cu-Kα (λ=1.5406Å). The analysis was conducted at 30kV, 30mA and 2θ with the scan angle from 2.5° to 45° and scanning rate 1°/minute. Thermogravimetric analyser (TGA 7 Perkin Elmer, US) was employed to evaluate the thermal properties of the electrospun biocomposite. For TGA, the test was conducted in the ramp mode of temperature range from 25°C to 600°C at constant heating rate of 10°Cmin−1 under nitrogen atmosphere.

2.5Batch adsorption experiments

The Congo red dye was dissolved and diluted with distilled water into different concentrations ranging from 0 to 100mgL−1. The dilution standard curve of Congo red dye was prepared and found to be y=0.0158x+0.0129 and R2=0.9983, where y was the adsorbent at wavelength 470nm and x was the Congo red concentration (mg L−1). The batch adsorption studies were performed in 50mL Congo red dye solution of initial dye concentration 100mgL−1 with 200mg of electrospun biocomposites. The experiments were carried out at room conditions 297±1K with relative humidity of 50±3%. The adsorption removal percentage E% was calculated using Eqn. (1).

where C0 is the initial is the initial concentration of Congo red dyes, C1 is the final concentration of Congo red dyes.

2.5.1Effect of pH on adsorption of Congo red dye

The 100mgL−1 Congo red dye solution was adjusted to pH 4, 7 and 10, respectively. However, no adsorption test was done at pH 4, the Congo red dye turned into dark violet crystal and settled at the bottom of container. Congo red becomes protonated at lower pH and is much less hydrophilic and water-soluble than its non-protonated form [22]. The adsorption removal percentage E% at equilibrium was determined.

2.5.2Adsorption isotherms and kinetics

The adsorption isotherms of PHA2NCC and PHA2Cts were identified using the Congo red dye concentration in the range of 20mgL−1 to 100mgL−1. In this experiments, two models of adsorption isotherms were applied, namely Langmuir and Freundlich isotherms. For the adsorption kinetics, the electrospun biocomposites were mixed with 50mL Congo red dye solution of initial dye concentration 100mgL−1. The remaining Congo red dye concentrations were measured at different time intervals. The adsorption kinetics included pseudo-first kinetic model, pseudo-second kinetic model and intra-particle diffusion model. The amount of the adsorption at time t and at equilibrium, were calculated using Eqn. (2).

where qt (mg g−1) is the adsorbed amount per unit mass of adsorbent at time t and qe (mg g−1) is the adsorbed amount per unit mass of adsorbent at equilibrium; C0 is the initial dye concentration; Ct is the dye concentration at time t; Ce is the dye concentration at equilibrium; m is the mass of adsorbent (g) and v is volume of solutions (L).

3Results and discussion3.1Surface morphology of electrospun biocomposites

Fig. 1 showed the surface morphology of the electrospun biocomposites. From Fig. 1 (a), the PHA had less entangled straight fibres but a great number of micro-sized globular beads, resulting the membrane to rupture easily when peeling off from the aluminium foil static collector. The formation of beads in the electrospun PHA weakened the mechanical strength due to the reduction of intersection points among fibres and exposure surface area per volume. This suggested that the electrospinning conditions such as syringe injection speed, voltage used and the distance between the needle and the collector were not suitable for PHA. In contract, these electrospinning conditions worked well with the incorporation of 2% wt nanocellulose (PHA2NCC) and 2% wt chitosan (PHA2Cts).

Fig. 1.

SEM micrograph of (a) PHA, (b) PHA2NCC, (c) PHA2Cts, and (d) PHA3NCC1Cts of the electrospun biocomposite at x600 magnification. The micrographs showed relatively good porosity of electrospun biocomposite membranes.


In Fig. 1 (b), the electron micrograph showed a high number of intersected region between nanocellulose and PHA fibres PHA2NCC. Thick nanocellulose fibres were interwoven with numerous thinner PHA fibres resulted in better mechanical strength when compared to PHA alone. Electrospun PHA breaks easily when peeled of the aluminium collector while PHA2NCC remains relatively intact. Similar, enhancement of mechanical strength was observed with PHA2Cts but chitosan tends to form globules instead of thick fibres as shown in Fig. 1 (c). However, when the weight loading of the adsorptive fillers was increased as observed in PHA with 3% wt nanocellulose and 1% wt chitosan (PHA3NCC1Cts), Fig. 1 (d), the membrane had high amount of the irregular shape fillers which linked together by thick PHA fibres. Increase the load of nanocellulose and chitosan also resulted in a high degree of porosity. This high degree of porosity would later contribute to better adsorption of Congo red in subsequent adsorption studies.

3.2Structural study of the electrospun biocomposites

The composition and the crystallinity of the electrospun biocomposites were determined using FT-IR and XRD. From Fig. 2 (a), asymmetric and symmetric COC stretching vibration were shown as 1175, 1275, 1175, 1130, 1099, and 1054cm−1 peaks [23–25]. Meanwhile, another indicative peaks of PHA was 1719 and 1261cm−1, showing the stretching of ester carbonyl CO and CO bonding [24–26]. The moderate and peaks formed in the range of 1000 to 800cm−1 were the CC stretching vibrations, indicated the formation of the 21-helix structure of PHA.

Fig. 2.

(a) The FT-IR spectra of all electrospun biocomposites from wavelength 4000cm−1 to 600cm−1 and XRD diffractograms of (b) PHA, (c) PHA2NCC, (d) PHA2Cts, and (e) PHA3NCC1Cts with scanning angle from 2.5° to 45° at 30kV, 30mA. The addition of bioadsorbents into the membranes did not form new peaks in FT-IR spectrum.


The electrospun biocomposites were hydrophobic in nature by having 1452 and 1379cm−1 peaks that indicates the stretching of CH bond from CH2 and CH3, respectively [23]. This was further demonstrated by weak peaks in 2977 and 2934cm−1 which showed the asymmetric and symmetric stretching of CH3 and anti-symmetric CH2 vibration respectively [23,25]. Interestingly, the peak intensity of the 1379cm−1 increased with the addition of nanocellulose into the biocomposite. In short, the peaks formed in the spectra of PHA were close similar with the other electrospun biocomposite. This might be due to the small weight loading of adsorbent fillers into the electrospun biocomposite, hence, causing the absence of the peaks that showed the chemical functional groups of cellulose and chitosan.

From Fig. 2 (b–e), the electropun PHA and PHA electrospun biocomposites shared the similar peaks in XRD spectra. The three strong peaks were found to be around 13.86°, 16.73° and 24.35°, representing the diffraction planes of (020), (110), and (121) of the structure. The crystallinity (Xc) were computed by using the Segal method that involved the intensity difference of crystalline plane (110) and the amorphous plane (020) [27]. PHA is a semicrystalline thermoplastic [21]. The crystallinity of electrospun PHA membrane was the lowest (57.6%). With the addition of nanocellulose and chitosan into the biocomposite, the crystallinity of PHA2NCC and PHA2Cts was 70.9% and 68.1%, respectively.

From the literature, the nanocellulose from plant had relatively high crystallinity (70–78%) and comprised of mainly crystalline cellulose domains due to the treatment of strong alkaline and acid that removed the amorphous domains [28,29]. Meanwhile, the crystallinity of the commercial chitosan was recorded to be around 73%. Adding these highly crystalline fillers would increase the overall crystallinity of PHA from 57.6% to approximately 70%. Further increment of nanocellulose and chitosan did not result in an increase in crystallinity as observed in PHA3NCC1Cts with a crystallinity of 70.5%. The higher crystallinity of the membrane indicated closely-packed arrangement of the atoms in the polymer that resulted higher rigidity and enhanced tensile strength of the electrospun biocomposite [30].

3.3Thermal properties of the electrospun biocomposite

The thermal stability of the electrospun biocomposites was investigated via TGA as shown in Table 2. All the electrospun biocomposites had the first and most intense thermal decomposition process within the range of 200–300°C. Except for electrospun PHA membrane, the electrospun biocomposites had the second process in the range of 300–450°C, indicating the thermal decomposition of the adsorptive fillers. The temperature with the maximum decomposition rate (Tmax) of first process and 5% initial weight lost temperature (T5%) of electrospun PHA membrane (272.07°C and 241.74°C) and PH3NCC1Cts (273.55°C and 239.83°C) was closely similar and higher than PHA2NCC and PHA2Cts. This could be related to the structure morphology of the PHA and PHA3NCC1Cts that had the bead formation and thick interlinked fibres.

Table 2.

The TGA profile of the electrospun biocomposites with the range of 25°C to 600°C at constant heating rate of 10°Cmin−1 under nitrogen atmosphere.

MembraneFirst processSecond processT5%b (°C)Residue (%)
Tmaxa (°C)  Weight lost (%)  Tmaxa (°C)  Weight lost (%) 
PHA  272.07  98.34  –  –  241.74  1.66 
PHA2NCC  240.73  91.99  361.09  5.84  211.03  2.17 
PHA2Cts  255.45  92.25  353.03  6.45  227.50  0.80 
PHA3NCC1Cts  273.55  77.88  395.17  21.15  239.83  0.97 

Maximum decomposition rate temperature on that process (Tmax).


5% initial weight lost temperature (T5%).

Meanwhile, the Tmax of first process and T5% of PHA2NCC (240.73°C and 211.03°C) and PHA2Cts (255.45°C and 227.50°C) were lower due to the presence of massive and thin entangled fibres. The high surface area of the fibres in PHA2NCC and PHA2Cts rendered the higher exposure to heat and hence increased the heat transferred that caused the great reduction of the Tmax of first process and T5%.

The Tmax of second thermal decomposition process for PHA2NCC and PHA2Cts was marked at 361.09°C and 353.03°C, respectively. According to the TGA study by other researchers, the nanocellulose produced from acid hydrolysis of cotton linter showed two weight loss peaks at 287.50°C and 367.67°C as the result of cellulose sulfonation at the hydroxyl group [31]. Meanwhile, the commercial chitosan that was used in the present research showed the thermal decomposition peak of D-glucosamine group at 301.0°C [27]. The combination of the nanocellulose and commercial chitosan in PHA3NCC1Cts notably increased the Tmax of second thermal decomposition to 395.17°C.

All of the electrospun biocomposites were not hygroscopic. This was due to the absence of the thermal degradation process below 100°C that was usually referred to the loss of moisture content [32,33]. These properties could be explained by the presence of the intensive hydrophobic methyl group at the carbon polymer chains as indicated by the FTIR analysis.

3.4Dye adsorption isotherm, kinetics and removal ability3.4.1Effect of pH value

Congo red dye is a benzidine-based anionic diazo dye prepared by coupling tetrazotised benzidine with two molecules of napthoic acid [34]. It is a pH indicative dye where the transition pH is 3.9 to 5.0 and the colour changed from violet to red. Hence, in this study, the Congo red dye solution turned to dark violet crystal precipitate when tested in pH 4. The crystal precipitate settled to the bottom of solution which hindered the adsorption from occurring. The solution pH could affect the level of electrostatic or molecular interaction between the Congo red dye and the bioadsorbent [9]. The electrospun biocomposite was presented in Fig. 3. Before dye adsorption, the electrospun biocomposite was in white colour. After dye adsorption, PHA2NCC in Fig. 3 (c) had greater red intensity than PHA2Cts in Fig. 3 (b). This indicates PHA2NCC had better dye affinity to dye at the nanocellulose active site, resulting greater dye recovery from dye solution.

Fig. 3.

(a) The electrospun biocomposite before Congo red dye adsorption, and (b) PHA2Cts and (c) PHA2NCC after Congo red dye adsorption. PHA2NCC has higher red intensity, indicating greater Congo red adsorption.


From Fig. 4, the percentage of removal of Congo red dye by PHA2NCC in pH 7 (30.91±4.55%) was slightly higher than pH 10 (27.90±2.20%). The nanocellulose bioadsorbent in the PHA2NCC membrane had high negatively-charged hydroxyl group at the surface to be the active site for the adsorption reaction. At the alkaline region (pH 10), the electrostatic repulsion between the hydroxyl group and the anionic Congo red dye occurred and led to slight decrease of removal percentage. The results was consistent with the literature reported before, where the adsorption of Congo red dye had the highest removal percentage at the neutral pH [35,36].

Fig. 4.

Percentage of Congo red dye removal by PHA2NCC in pH 4, 7 and 10. Congo red was not soluble at pH 4 and precipitated as a violet crystal, adsorption could not be measured. (Dye concentration=100mg L−1, adsorbent dosage=4g L−1).

3.4.2Adsorption isotherm

Adsorption isotherm models are used to describe the interaction between the adsorbate and adsorbent. The equilibrium studies determine the capacity of the adsorbent and describe the adsorption isotherm by constants which the values express the surface properties and affinity of the adsorbents. The relationship between equilibrium data and either theoretical or practical equations is essential for the interpretation and prediction of the extent of adsorption [37].

Based on Table 3, the adsorption isotherm of PHA2NCC and PHA2Cts could be demonstrated by Langmuir parameters. The linear correlation coefficient (r2) in Langmuir isotherm has the r2 value closer to 1 for both PHA2NCC and PHA2Cts compared to Freundlich isotherm. This indicated the better fitting of the experimental result to the theoretical assumption according to Eqn. (4). The Langmuir isotherm assumed that the surface of the nanocellulose and chitosan in the PHA mesh membrane contained a fixed number of active sites. The adsorption reaction would be stopped in the condition of saturation of the Congo red dye adsorbate to the bioadsorbent. The monolayer Congo red dye that is attached to the active sites of the bioadsorbent hindered the further interaction between bioadsorbent and other dye molecules [34].

Table 3.

Isotherm parameters for the adsorption of Congo red dye by PHA2NCC and PHA2Cts. (Adsorbent dosage=4g L−1).

Electrospun biocompositesPHA2NCC  PHA2Cts 
Langmuir isothermaKL (1mg−10.0039  0.0002 
qm (mg g−1147.0588  434.7826 
R2  0.9884  0.9948 
RL  0.7169  0.9760 
Freundlich isothermKF (mg g−1)/(mg 1−1)1/n  0.0204  0.0152 
5.0505  11.6009 
R2  0.9530  0.8111 

The adsorption isotherm of the electrospun biocomposites fitted well to the Langmuir isotherm.

PHA2Cts (434.7826mgg−1) had notably higher qm than PHA2NCC (147.0588mgg−1), which may indicate that the chitosan had higher amount of maximum Congo red dye adsorption to the adsorbent surface. However, based on the microscopic view of PHA2Cts membrane, the globular chitosan was entrapped in the PHA mesh, reducing the surface that was exposed to the Congo red dye for the higher chances of adsorption interaction.

Furthermore, the Langmuir isotherm that was expressed by tested electrospun biocomposites were favourable due to the RL values of PHA2NCC (0.7169) and PHA2Cts (0.9760) that fell in the range of 0<RL<1. This range is desirable as it shows a good affinity of adsorption but not to high as to prohibit regeneration adsorbent. Several studies have shown that nanocellulose and chitosan adsorbents can be regenerated using monoacid HCl and monobasic NaOH solution [16–19]. However, viability of regeneration and reuse within an electrospun PHA matrix has yet to be demonstrated and this warrants further investigation.

The Langmuir isotherm model is based on a monolayer adsorption onto a surface with a finite number of adsorption sites of uniform adsorption energies [38]. It is expressed in the Eq. (3) and its linearized form is expressed in Eqn. (4).

where qm is the maximum amount of adsorption with complete monolayer coverage on the adsorbent surface (mg g−1), and KL is the Langmuir constant related to the energy of adsorption (l mg−1). The Langmuir constants KL and qm can be determined from the linear plot of 1/ Ce versus 1/ qe.

For the Langmuir model, the effect of isotherm shape is used to predict a favourability of an adsorption system under specific conditions. The favourable adsorption of Langmuir isotherm can be expressed in terms of a dimensionless constant separation factor or equilibrium parameter, RL. The RL values could be classified into four types of Langmuir adsorption favourites, which are (1) unfavourable when RL>1, (2) linear progression when RL=1, (3) favourable when 0<RL<1 and (4) irreversible isotherm when RL=0[37]. It can be represented by Eqn. (5).

The Freundlich isotherm model is commonly used to describe heterogeneous systems which involve multi-layer adsorption. It can be represented by Eqn. (6). The linearized form of Eqn. (6) is described by Eqn. (7) by showing the linear plot of logqe versus logCe. The values of KF and n could be obtained from the slope and intercept, respectively [11].

where KF is related to the Freundlich adsorption capacity, 1/n is heterogeneity factor which ranges from 0 to 1.

3.4.3Adsorption capacity

Based on Fig. 5 (a), The Congo red dye removal percentage of nanocellulose (30.9%) in PHA2NCC was 3-fold higher than that of chitosan (10.5%) PHA2Cts. This indicated the nanocellulose was better bioadsorbent than chitosan when was entrapped in PHA via electrospinning. When nanocellulose and chitosan were entrapped together, it resulted in higher porosity of the PHA matrix and corresponding Congo red dye adsorption. PHA1NCC3Cts, PHA2NCC2Cts and PHA3NCC1Cts were formulated by combining the nanocellulose and chitosan in the same PHA mesh membrane. As in Fig. 5 (b), the percentage of Congo red dye uptake in PHA3NCC1Cts (75.8%) that incorporated both nanocellulose and chitosan, showed a 2- and 7-fold increase compared to PHA2NCC and PHA2Cts, respectively. This shows that there is synergy between the two bioadsorbents.

Fig. 5.

Percentage of Congo red dye removal: (a) PHA2NCC and PHA2Cts; (b) PHA1NCC3Cts, PHA2NCC2Cts, PHA3NCC1Cts electrospun biocomposites. (Dye concentration=100mg L−1, adsorbent dosage=4g L−1).

3.4.4Adsorption kinetic models

The adsorption kinetics was studied to explain the rate of dye adsorption, adsorption mechanism and the rate limiting step. Three adsorption kinetic models were compared here, namely pseudo-first order, pseudo-second order, and intra-particle diffusion.

The findings of dye removal uptake in Fig. 5 correlated well with the kinetic parameters of all the electrospun biocomposites in Table 4. All the electrospun biocomposites showed the chemisorption nature by having the highest r2 in pseudo-second order. The adsorption process involves three consecutive mass transfers, which are bulk diffusion (movement of dye molecules in the bulk solution), external diffusion (transport of dye molecules to the active site) and intra-particle diffusion.

Table 4.

Kinetic parameters for the adsorption of Congo red dye by electrospun biocomposites. (Dye concentration=100mg L−1, adsorbent dosage=4g L−1).

Electrospun biocomposites  PHA2NCC  PHA2Cts  PHA1NCC3Cts  PHA2NCC2Cts  PHA3NCC1Cts 
Pseudo-first order
r2  0.9406  0.3455  0.7991  0.9260  0.9540 
k1 (min−10.0382  0.0183  0.0081  0.0116  0.0125 
qe,1 (mg g−17.1506  2.8284  3.0916  9.8296  10.8114 
Pseudo-second ordera
r2  0.9842  0.9995  0.9948  0.9985  0.9986 
k2 (mgg−1min−10.0080  0.2806  0.0098  0.0038  0.0036 
qe,2 (mg g−18.4459  2.6738  6.6357  18.6220  19.4175 
Intra-particle diffusion
r2  0.8858  0.3559  0.7350  0.7657  0.7676 
kipd (mgg−1min−1/20.5863  0.1003  0.2016  0.6426  0.6738 
qe. exp (mg g−17.7443  2.6863  6.5370  18.1239  18.9514 

The adsorption kinetic of the electrospun biocomposites fitted well to the Pseudo-second order parameters.

The pseudo-second order kinetic model deals with two sites occupancy adsorption, where one molecule of adsorbate can have interaction with dye at more than one adsorbent site. Therefore, this model is identified as chemisorption, which involves covalent bonding and ion exchange. The rate limitation in this kinetics model is the intra-particle diffusion [39]. However, the intra-particle diffusion was found to have the lowest correlation coefficient in the electrospun biocomposites.

In pseudo-second order kinetic model, qe,2 was compared with the qe. exp in the bivariate plot and found to have significant difference with p<0.0001. The addition of the bioadsorbent contents in the electrospun biocomposites from PHA2NCC to PHA2NCC2Cts and PHA3NCC1Cts increased the qe,2 by 120.5% and 129.9%, respectively. The qe,2 of PHA2NCC2Cts and PHA3NCC1Cts increased slightly from 18.6220mgg−1 to 19.4175mgg−1, indicating 80mgL−1 of nanocellulose was close to optimum dye adsorption capacity in equilibrium. However, the addition of more nanocellulose into biocomposite membranes generally decreased the rate of adsorption. Although the addition of chitosan had relatively small qe,2 compared to that of nanocellulose, the k2 of the PHA2Cts was the highest (0.2806mgg−1min−1) among all of the samples.

Pseudo-first order kinetic model was created with the same assumption used in Langmuir model, which is “one-site-occupancy” adsorption, where the adsorbate molecule can only “react” with one site of adsorbent. In pseudo-first order kinetic model, external diffusion is the rate limiting step and this mode of adsorption is usually physisorption. The pseudo-first order equation for adsorption reaction can be expressed in Eqn. (8).

where qt is the amount of adsorbate adsorbed at time t (mg g−1), qe,1 is the adsorption capacity in equilibrium (mg g−1), k1 is the rate constant of pseudo-first order model (min−1), and t is the time (min).

In order to identify the computational qe,1 and k1, the pseudo-first order of the kinetic equation can be linearized into Eqn. (9) by setting the plot of lnqe,1-qt against t[40].

For the pseudo-second order of adsorption, the equation model and its linearized form could be expressed in Eqn. (10) and (11), respectively.

where qt is the amount of adsorbate adsorbed at time t (mg g−1), qe,2 is the adsorption capacity in equilibrium (mg g−1), k2 is the rate constant of pseudo-second order model (mg g−1min−1), and t is the time (min). To identify the computational qe,2 and k2, the pseudo-second order of the kinetic equation can be linearized into Eqn. (11) by setting the plot of tqt against t[38].

For the intra-particle diffusion, the equation model could be expressed in Eqn. (12).

where qt is the amount of adsorbate adsorbed at time t (mg g−1), and kid is the rate constant of intra-particle diffusion model (mg g−1min−1/2).


Nanocellulose-chitosan-PHA electrospun biocomposites produced electrospinning demonstrated a good filter material for dye adsorption. The Pickering emulsion proved to be effective in producing a homogenous mixture of nanocellulose and chitosan with PHA which tends to be immiscible solutions. The simplicity and cost effectiveness of the Pickering emulsion makes this an easily scalable method for commercial scale production of the electrospun biocomposite. SEM analysis confirmed the highly porous structure of the electrospun biocomposite for higher adsorption capacity for Congo red dye. The addition of the nanocellulose and chitosan increased the crystallinity of the electrospun biocomposites but only limited to 70.5% in PHA3NCC1Cts. From the adsorption test, PHA2NCC had higher removal percentage in pH neutral than that of alkaline solution. The adsorption isotherm of PHA2NCC and PHA2Cts could be demonstrated by Langmuir model. Besides, the electrospun biocomposites showed the chemisorption nature with the pseudo-second kinetic order. The dye removal percentage of nanocellulose (30.9%) was 3-fold better than chitosan (10.5%) in PHA. The synergetic effect was reported when nanocellulose and chitosan was incorporated into the PHA mesh with enhancement of Congo red adsorption, ranging from 2- to 7-fold compared to individual bioadsorbents. This makes the electrospun biocomposite a suitable adsorbent for removal of Congo red dye from industrial wastewater.

Conflicts of interest

The authors declare no conflicts of interest.


The present research was financially supported by UCSI University Pioneer Scientist Incentive Fund (PROJ-in-FAS-030; PROJ-in-FAS-049; PROJ-in-FAS-052), Putra Graduate Initiative Grant (GP-IPS) under vote number 9590400, Universiti Putra Malaysia (UPM) and the Ministry of Higher Education (MOHE) Malaysia (FRGS/2/2014/SG01/UCSI/02/2; FRGS/1/2018/TK10/UCSI/02/1). The researchers also acknowledge the contribution of corn cob supply from Marine Gold Marketing Sdn. Bhd., Malaysia.

Y. Cao, Z. Pan, Q. Shi, J. Yu.
Modification of chitin with high adsorption capacity for methylene blue removal.
Int J Biol Macromol, 114 (2018), pp. 392-399
G. Vijayakumar, R. Tamilarasan, M. Dharmendirakumar.
Adsorption, kinetic, equilibrium and thermodynamic studies on the removal of basic dye rhodamine-B from aqueous solution by the use of natural adsorbent perlite.
J Mater Environ Sci, 3 (2012), pp. 157-170
S. Ghorai, A.K. Sarkar, A.B. Panda, S. Pal.
Effective removal of Congo red dye from aqueous solution using modified xanthan gum/silica hybrid nanocomposite as adsorbent.
Bioresour Technol, 144 (2013), pp. 485-491
R. Xu, J. Mao, N. Peng, X. Luo, C. Chang.
Chitin/clay microspheres with hierarchical architecture for highly efficient removal of organic dyes.
Carbohydr Polym, 188 (2018), pp. 143-150
M.T. Yagub, T.K. Sen, S. Afroze, H.M. Ang.
Dye and its removal from aqueous solution by adsorption: a review.
Adv Colloid Interface Sci, 209 (2014), pp. 172-184
L. Jin, Q. Sun, Q. Xu, Y. Xu.
Adsorptive removal of anionic dyes from aqueous solutions using microgel based on nanocellulose and polyvinylamine.
Bioresour Technol, 197 (2015), pp. 348-355
J. Xiao, W. Lv, Y. Song, Q. Zheng.
Graphene/nanofiber aerogels: performance regulation towards multiple applications in dye adsorption and oil/water separation.
Chem Eng J, 338 (2018), pp. 202-210
Z. Karim, S. Claudpierre, M. Grahn, K. Oksman, A.P. Mathew.
Nanocellulose based functional membranes for water cleaning: tailoring of mechanical properties, porosity and metal ion capture.
J Membr Sci, 514 (2016), pp. 418-428
L. Jin, W. Li, Q. Xu, Q. Sun.
Amino-functionalized nanocrystalline cellulose as an adsorbent for anionic dyes.
Cellulose, 22 (2015), pp. 2443-2456
N. Mohammed, N. Grishkewich, H.A. Waeijen, R.M. Berry, K.C. Tam.
Continuous flow adsorption of methylene blue by cellulose nanocrystal-alginate hydrogel beads in fixed bed columns.
Carbohydr Polym, 136 (2016), pp. 1194-1202
R. Batmaz, N. Mohammed, M. Zaman, G. Minhas, R.M. Berry, K.C. Tam.
Cellulose nanocrystals as promising adsorbents for the removal of cationic dyes.
Cellulose, 21 (2014), pp. 1655-1665
L. Cui, Z. Xiong, Y. Guo, Y. Liu, J. Zhao, C. Zhang, et al.
Fabrication of interpenetrating polymer network chitosan/gelatin porous materials and study on dye adsorption properties.
Carbohydr Polym, 132 (2015), pp. 330-337
B. Tanhaei, A. Ayati, M. Lahtinen, M. Sillanpää.
Preparation and characterization of a novel chitosan/Al2O3/magnetite nanoparticles composite adsorbent for kinetic, thermodynamic and isotherm studies of methyl Orange adsorption.
Chem Eng J, 259 (2015), pp. 1-10
R. Tang, Z. Yu, Y. Zhang, C. Qi.
Synthesis, characterization, and properties of antibacterial dye based on chitosan.
Cellulose, 23 (2016), pp. 1741-1749
X. Wang, R. Tang, Y. Zhang, Z. Yu, C. Qi.
Preparation of a novel chitosan based biopolymer dye and application in wood dyeing.
Polymers, 8 (2016), pp. 1-13
H. Yang, A. Sheikhi, T.G.M. Van De Ven.
Reusable green aerogels from cross-linked hairy nanocrystalline cellulose and modified chitosan for dye removal.
Langmuir, 32 (2016), pp. 11771-11779
K. Song, H. Xu, L. Xu, K. Xie, Y. Yang.
Cellulose nanocrystal-reinforced keratin bioadsorbent for effective removal of dyes from aqueous solution.
Bioresour Technol, 232 (2017), pp. 254-262
V.S. Munagapati, V. Yarramuthi, D.S. Kim.
Methyl orange removal from aqueous solution using goethite, chitosan beads and goethite impregnated with chitosan beads.
J Mol Liq, 240 (2017), pp. 329-339
M. Vakili, S. Deng, L. Shen, D. Shan, D. Liu, G. Yu.
Regeneration of chitosan-based adsorbents for eliminating dyes from aqueous solutions.
Sep Purif Rev, 48 (2019), pp. 1-13
G. Yazgan, A.M. Popa, R.M. Rossi, K. Maniura-Weber, J. Puigmartí-Luis, D. Crespy, et al.
Tunable release of hydrophilic compounds from hydrophobic nanostructured fibers prepared by emulsion electrospinning.
Polymer (United Kingdom), 66 (2015), pp. 268-276
B.S. Kushwah, A.V.S. Kushwah, V. Singh.
Towards understanding polyhydroxyalkanoates and their use.
A. Tor, Y. Cengeloglu.
Removal of congo red from aqueous solution by adsorption onto acid activated red mud.
J Hazard Mater, 138 (2006), pp. 409-415
Y.X. Weng, X.L. Wang, Y.Z. Wang.
Biodegradation behavior of PHAs with different chemical structures under controlled composting conditions.
M.R. Foroughi, S. Karbasi, R. Ebrahimi-Kahrizsangi.
Physical and mechanical properties of a poly-3-hydroxybutyratecoated nanocrystalline hydroxyapatite scaffold for bone tissue engineering.
J Porous Mater, 19 (2012), pp. 667-675
Y. Xu, L. Zou, H. Lu, Y. Wei, J. Hua, S. Chen.
Preparation and characterization of electrospun PHBV/PEO mats: The role of solvent and PEO component.
J Mater Sci, 51 (2016), pp. 5695-5711
F. Masood, P. Chen, T. Yasin, F. Hasan, B. Ahmad, A. Hameed.
Synthesis of poly-(3-hydroxybutyrate-co-12 mol % 3-hydroxyvalerate) by Bacillus cereus FB11: its characterization and application as a drug carrier.
J Mater Sci Mater Med, 24 (2013), pp. 1927-1937
C.Y. Soon, Y.B. Tee, C.H. Tan, A.T. Rosnita, A. Khalina.
Extraction and physicochemical characterization of chitin and chitosan from Zophobas morio larvae in varying sodium hydroxide concentration.
Int J Biol Macromol, 108 (2018), pp. 135-142
N.F. Souza, J.A. Pinheiro, A.I.S. Brígida, J.P.S. Morais, M. de Souza Filho M de sá, M. de Freitas Rosa.
Fibrous residues of palm oil as a source of green chemical building blocks.
Ind Crops Prod, 94 (2016), pp. 480-489
W. Li, X. Zhao, S. Liu.
Preparation of entangled nanocellulose fibers from APMP and its magnetic functional property as matrix.
Carbohydr Polym, 94 (2013), pp. 278-285
N.R. Savadekar, S.T. Mhaske.
Synthesis of nano cellulose fibers and effect on thermoplastics starch based films.
Carbohydr Polym, 89 (2012), pp. 146-151
J.P.S. Morais, M.D.F. Rosa, M. De Souza Filho M de sa, L.D. Nascimento, D.M. Do Nascimento, A.R. Cassales.
Extraction and characterization of nanocellulose structures from raw cotton linter.
Carbohydr Polym, 91 (2013), pp. 229-235
J.I. Morán, V.A. Alvarez, V.P. Cyras, A. Vázquez.
Extraction of cellulose and preparation of nanocellulose from sisal fibers.
Cellulose, 15 (2008), pp. 149-159
M. Mucha, A. Pawlak.
Thermal analysis of chitosan and its blends.
Thermochim Acta, 427 (2005), pp. 69-76
A. Mittal, J. Mittal, A. Malviya, V.K. Gupta.
Adsorptive removal of hazardous anionic dye “Congo red” from wastewater using waste materials and recovery by desorption.
J Colloid Interface Sci, 340 (2009), pp. 16-26
L.K. Akinola, A. Ibrahim, A.S. Chadi.
Kinetic and equilibrium studies of congo red adsorption on adsorbent from bambara groundnut hulls.
J Pure Appl Sci, 2 (2016), pp. 79-88
H. Patel, R.T. Vashi.
Removal of Congo Red dye from its aqueous solution using natural coagulants.
J Saudi Chem Soc, 16 (2012), pp. 131-136
V. Vimonses, S. Lei, B. Jin, C.W.K. Chow, C. Saint.
Kinetic study and equilibrium isotherm analysis of Congo Red adsorption by clay materials.
Chem Eng J, 148 (2009), pp. 354-364
V. Nair, A. Panigrahy, R. Vinu.
Development of novel chitosan-lignin composites for adsorption of dyes and metal ions from wastewater.
Chem Eng J, 254 (2014), pp. 491-502
J.N. Putro, A. Kurniawan, S. Ismadji, Y.H. Ju.
Nanocellulose based biosorbents for wastewater treatment: study of isotherm, kinetic, thermodynamic and reusability.
Environ Nanotechnol Monitor Manage, 8 (2017), pp. 134-149
I.D. Mall, V.C. Srivastava, N.K. Agarwal.
Removal of Orange-G and methyl violet dyes by adsorption onto bagasse fly ash - kinetic study and equilibrium isotherm analyses.
Dyes Pigments, 69 (2006), pp. 210-223
Copyright © 2019. The Authors
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.