Separation and Purification Technology 279 (2021) 119679

Contents lists available at ScienceDirect

Separation and Purification Technology

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

Biocompatible magnetic flocculant for efficient harvesting of microalgal

cells: Isotherms, mechanisms and water recycling
Zhihong Yin a, Lingbo Zhang a, Dan Hu a, Shuangxi Li a, Ruoyu Chu a, Chenchen Liu a,
Yuanfei Lv a, Jianfeng Bao a, Mingdeng Xiang b, *, Liandong Zhu a, *
a
School of Resource and Environmental Sciences, Hubei Key Laboratory of Biomass-Resources Chemistry and Environmental Biotechnology, and Hubei International
Scientific and Technological Cooperation Base of Sustainable Resource and Energy, Wuhan University, Wuhan 430079, China
b
State Environmental Protection Key Laboratory of Environmental Pollution Health Risk Assessment, South China Institute of Environmental Sciences, Ministry of
Ecology and Environment, Guangzhou 510655, China

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

Keywords: As an environmentally friendly bioflocculant, chitosan has a great application potential for microalgae har­
Fe3O4-chitosan nanocomposites vesting due to its biocompatibility. The rapid development of magnetic separation technology has attracted
Magnetic separation extensive attention because of its reusability, energy and time saving. In this work, Fe3O4-chitosan composites
Microalgae harvesting
flocculant with high magnetic harvesting ability and fast separation was successfully prepared with sodium
Flocculation
Modification
tripolyphosphate as crosslinking agent, and applied to the harvesting and separation of microalgae cells. The
harvesting efficiency of 97.63% for Chlorella vulgaris was achieved within 3 min using chitosan/Fe3O4 with the
weight ratio of 3:1 and the dosage of 0.5 g/L under pH 3.0. Freundlich isotherm was more suitable for Fe3O4-
chitosan to harvest microalgae, indicating the microalgae harvesting by Fe3O4-chitosan composites was a het­
erogeneous multilayer process. The highest adsorption ability of the Fe3O4-chitosan on Chlorella vulgaris could be
reached 75.43 g/g. Double-layer compression, charge neutralization and adsorption bridging are primarily
harvesting mechanisms between Fe3O4-chitosan composites and microalgae cells. The second-order polynomial
model well simulates the experimental data, establishing the relationship between Fe3O4-chitosan and algae
slurry pH, concentration of microalgae cells and settling time. The supernatant after Fe3O4-chitosan harvesting
could be reused, saving water resources and cost, showing its biocompatibility. The Fe3O4-chitosan composites
had excellent reusability, and the harvesting efficiency could reach 85.10% after four cycles, which improved the
economic feasibility of magnetic harvesting. Fe3O4-chitosan has the advantages of high efficiency, fast separa­
tion, low economic cost, and low energy consumption. This work provides basic guidance and data analysis in
the study of Fe3O4-chitosan for microalgae harvesting on a large scale.

1. Introduction of biofuels such as biodiesel, biohydrogen and bioethanol [9]. It has the
advantages of fast growth, high photosynthetic efficiency, high oil
Biofuels are the promising alternative to exhaustible and environ­ content and carbohydrates accumulation [10]. However, the surface is
mentally unsafe fossil fuels [1]. The resources of biofuels can be divided negatively charge, which caused the microalgae cells to be stably sus­
into three generations: biomass from food crops rich in sugar or starch pended in the culture medium [11]. The harvesting process of suspen­
(such as stems of sugar cane and sugar beet) and oil (such as soybeans, sion microalgae accounts for 20%-30% of the whole production process.
sunflower seeds and rapeseeds) are the feedstock to produce the first In addition, it has been reported that 3726 kg of water is needed for
generation of biofuels. Second generation biofuels are produced from producing 1.0 kg of microalgae biodiesel [12]. The substantial amount
lignocellulosic biomass [1–5]. As a source of the third-generation bio­ of freshwater usage limited the biofuel production from microalgae cells
fuels, microalgae are considered as an effective resource to solve the on a large scale.
global problems, such as global warming, energy crisis and water Several common harvesting methods are mainly included filtration,
pollution [6–8]. Microalgae are considered to be a promising producer centrifugation, flotation, and flocculation, etc. [13,14]. Flocculation,

* Corresponding authors.
E-mail addresses: [email protected] (M. Xiang), [email protected] (L. Zhu).

https://doi.org/10.1016/j.seppur.2021.119679
Received 17 June 2021; Received in revised form 2 September 2021; Accepted 5 September 2021
Available online 8 September 2021
1383-5866/© 2021 Elsevier B.V. All rights reserved.
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

which small cells form larger aggregates are separated from water by the effects of some parameters (ratio of chitosan/Fe3O4, dosage, settle
gravity sedimentation, is a popular harvesting method due to its high time, pH and concentration of microalgal cells) on harvesting process
separation efficiency and simple operation [15,16]. Commonly used were investigated; (ii) the flocculation mechanisms of magnetic chitosan
flocculants include metal salts, organic polymers and natural bio­ were evaluated by the Langmuir and Freundlich isotherm model; and
polymers [17,18]. However, chemicals flocculants which are used for (iii) the relationship between of Fe3O4-chitosan and external operating
harvesting process pose a serious challenge in downstream processing conditions (microalgae concentration, pH and the dosage of flocculants)
for production of high-value products such as biodiesel, cosmetics, and was established based on the second-order polynomial model. (iv) the
nutraceuticals. Ang et al. [18] investigated the performance of natural re-cultivation performance after water recycling and the reusability of
flocculants in microalgae harvesting and demonstrated that natural magnetic flocculants after harvesting process was evaluated.
flocculants outperformed inorganic salt. The use of metal salts has a
toxic effect on the utilization of downstream biomass, and the super­ 2. Experimental section
natant after harvesting cannot be reused. In addition, the other draw­
back of flocculation is the difficulty of separation the additives from the 2.1. Culture conditions for microalgae growth
microalgae biomass [19]. It is urgent to develop an effective harvesting
method to recover the microalgae. The Chlorella vulgaris used in this study was obtained from School of
Chitosan, which produced from chitin as raw materials, have become Resource and Environmental Sciences, Hubei Key Laboratory of
a promising choice because it is non-toxic, biodegradable, and Biomass-Resources Chemistry and Environmental Biotechnology, and
biocompatible. It has a high cationic density, which could be harvested was cultivated on BG11 medium after inoculation [4]. The composition
negatively microalgae cells to its surface through polymer adsorption of BG11 medium consisted mainly of NaNO3 (1.5 g/L), K2HPO4 (0.04 g/
bridging and charge neutralization. Commercially, chitosan is extracted L), MgSO4⋅7H2O (0.075 g/L), CaCl2⋅2H2O (0.036 g/L), citric acid (0.006
from chitin, which is the waste product of seafood processing (shrimp, g/L), ferrous ammonium citrate (0.006 g/L), Na2⋅EDTA (0.001 g/L),
crab and squid, etc.) [20]. A study by Vu et al. [21] investigated the Na2CO3 (0.02 g/L) and 1.0 mL of trace metal solution per liter. Trace
harvesting efficiency of chitosan and inorganic salts could reach >80% metals included the following species: H3BO3 (2.86 g/L), MnCl4⋅4H2O
for microalgae cells. Ahmad et al. [22] found that chitosan could harvest (1.86 g/L), ZnSO4⋅7H2O (0.22 g/L), CuSO4⋅5H2O (0.08 g/L),
99% of microalgae cells at mixing time of 20 min. However, chitosan is a Na2MoO4⋅4H2O (0.39 g/L) and Co(NO3)⋅6H2O (0.05 g/L). The pH was
kind of expensive bioflocculant, which requires a high dosage to harvest adjusted to 7.20, and the culture medium was autoclaved at 121 ◦ C for
microalgae. Moreover, the sedimentation process of aggregates is rela­ 20 min, and the microalgae cells were cultivated in a light incubator.
tively time-consuming, which increases the harvest time. The temperature is 28 ± 1 ◦ C, and the light conditions are about 200
Magnetophoretic particles harvesting is regarded as a time-saving μmol/m2⋅s (the ratio of light/dark was 12 h/12 h). The microalgae
and energy-efficient, fast technology for harvesting suspension micro­ culture was artificially shaken four times a day to avoid the growth of
algae due to its high specific area, biocompatibility and super­ microalgae cell along the wall [27]. The Chlorella vulgaris was then
paramagnetic [23]. The principle of Fe3O4 nanoparticles harvesting of transferred to large tubes to grow on a large scale, which shown in Fig
microalgae is to use magnetic particles to make microalgae cells, and use S1. The initial dry weight concentration of Chlorella vulgaris was 0.07 g/
an external magnetic field to quickly separate them from the water [24]. L, after 15 days of cultivation, the final dry weight concentration of
However, the collecting efficiency of bare magnetic nanoparticles is biomass was 1.08 g/L. The cultivated Chlorella vulgaris was directly used
relatively low, which limits the development of microalgae cells har­ in the harvest test. All chemicals in the experiment were purchased at
vesting. Modification of magnetic nanoparticles by positive compounds analytical purity. All subsequent harvesting experiments used deionized
can improve the interaction between microalgae cells and magnetic water (18.2 MΩ cm, Milli-Q).
nanoparticles, thereby increasing the harvest performance. Recently,
several magnetic particles with functionalized modification, such as 2.2. Preparation of Fe3O4-chitosan composites
cationic polyacrylamides (CPAMs), diallyldimethylammonium chloride
(PDDA), amino-riched dendrimers (PAMAM) and polyethyleneimine According to previous literature studies, some modifications are
nanocomposites (PEI) were synthesized and utilized for microalgae made to the magnetic chitosan composites [28]. The preparation
harvesting [23–25]. Gerulová et al. synthesized that the Fe3O4-PEI methods are as follows: chitosan (0.9 g) is dissolved in citric acid solu­
nanoparticles for harvesting negatively charged microalgae cells tion (36 mL of 1 mol/L). Then, different doses of Fe3O4 nanoparticles are
(Chlorella zofingiensis and Chlorella vulgaris, etc.) [25]. Under the con­ uniformly dispersed in chitosan, and a colloidal solution is formed under
dition of an additive dose of 200 mg/L, the harvest efficiency reached ultrasonic stirring. After that, the mixture solution was added dropwise
68–97% within 1 min. Wang et al. synthesized magnetic cationic poly­ into sodium tripolyphosphate (STPP) solution (50 mg/mL) and mixed
acrylamide (CPAM, 0.1 mg/mL) to efficiently recover microalgae cells for 120 min in 4 ◦ C conditions. STPP is a common non-toxic crosslinking
suspensions. The results showed that the harvesting efficiency of agent. Firstly, Fe3O4 was evenly dispersed in chitosan solution. In acidic
Chlorella ellipsoidea reached more than 95% within 1 min [23]. How­ medium, phosphate ion (anion) in STPP could be crosslinked with
ever, the modified functional materials are slightly toxic effect on protonated amino group (cations) of magnetic chitosan through ionic
downstream products, affecting the production of high products, and the bond, which improved the chemical ability of magnetic chitosan.
capture efficiency and economic feasibility of magnetic particles need to
be improved. By improving the harvesting efficiency and reducing the 2.3. Characterization of Fe3O4-chitosan and microalgae flocs
cost of collecting process, the economic feasibility and scope of appli­
cation of the magnetic flocculant in the production of microalgae could The surface morphologies and elemental compositions of the Fe3O4-
be improved. According to reports, because the harvested supernatant chitosan and magnetic flocs were tested by scanning electron micro­
has energy-saving and nutritional benefits value, it is a recyclable scopy (SEM) and energy-dispersive X-ray spectroscopy (EDS) (Zeiss
resource [26]. However, thus far, the application of magnetic chitosan in SIGMA, Britain). The crystallinity of the flocculants was tested by X-ray
microalgae harvesting is already reported in previous literature, but has diffraction (XRD) (XPert Pro, Netherlands). The specific surface area
not been deeply investigated. The analysis between of magnetic floc­ (SSA) of the magnetic flocculants was measured by automatic surface
culants and water recycling in microalgae harvesting remain to be analyzer (Belsorp-minII, MicrotracBEL, Japan). Zeta potential values of
explored. microalgae cells suspensions and modified flocculants were obtained
This work aims to design and explore a novel magnetic polymeric from a zeta potential meter (Zetasizer Nano-ZS90, Malvern, Britain).
bioflocculant by Fe3O4 and chitosan. The special goal of this work: (i) Fourier transform infrared spectrum (FTIR) was carried out to determine

2
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

the surface functional groups of flocculants and microalgae flocs (IR To analyze the fitting results of isotherm model for the harvesting
Tracer-100, Shimadzu, Japan). The chemical compositions were exam­ experiment, the correlation fitting coefficient (the degree of difference
ined by the X-ray photoelectron spectroscopy (XPS) (ESCALAB250Xi, χ2 and the root mean square error (RMSE))were calculated using the Eq.
America). The magnetic properties of Fe3O4-chitosan were character­ (6) [23].
ized by vibrating sample magnetometer (VSM) analysis (MPMS (SQUID) ∑[( ) ]
XL, America). χ2 = Qexp cal 2
e − Qe /Qcal
e (6)

√̅̅̅̅̅̅̅̅̅̅̅̅ P
2.4. Magnetic harvesting procedures 1 ∑ exp
(7)
2
RMSE = (Q − Qcal
e )
P − 2 i=1 e
In the process of harvesting, the flocculant was mixed with micro­
algae cells. The mixed solutions were first stirred at 200 rpm for 2 min, where p presents the number of parameters, Qexp e is the measured
then stirred at 50 rpm for 10 min, and finally stood for 3 min in a adsorbate concentration at harvesting equilibrium (mg/g), and the Qcal e
magnetic slow mixer. The absorbance of the solution at 2–3 cm was is calculated adsorbate concentration at harvesting equilibrium (mg/g).
tested under the condition of OD680 through UV. The effect of different In order to estimate the influence of external operating conditions
ratio of chitosan/Fe3O4 (α = 0.5, 1, 2 and 3) and dosage (0.1, 0.2, 0.3, further quantitatively (i.e., concentration of microalgae, pH, and the
0.4, 0.5, 0.6 and 0.7 g/L) of magnetic flocculants on microalgae har­ dosage of magnetic chitosan) on the harvesting microalgae, according to
vesting were measured. At different separation time (0.5, 1, 2, 3, 5 and the phenomenological theory, a second-order polynomial model was
10 min) was measured to investigate the optimal separation time. The introduced, as shown in Eq. (8), to reveal the operating factors-activity
effect of different pH on harvesting process was studied within the range relationship between magnetic flocculant and suspension microalgae
of 2–12 by using NaOH and HCl (0.1 M) solutions. The influence of cells [29].
different concentration of microalgae cells on microalgae harvesting by
Fe3O4-chitosan was studied. The linear relationship between OD680 and ∑ ∑n ∑n− 1 ∑n
n
Y = b0 + bi xi + bii x2i + bij xi xj (8)
cell dry weight was established by gravity method. And then the con­ i=1
i=1 i=1 j=i+1

centration of microalgae was calculated by testing the absorbance of

microalgae. The curve of microalgae cell dry weight is shown in Eq. (1)
and Fig. S2, and the detailed gravity test method was described in
Y = b0 +b1 x1 + b2 x2 +b3 x3 +b11 x21 +b22 x22 +b33 x23 +b12 x1 x2 +b13 x1 x3 + b23 x2 x3
Supplementary information. The harvesting efficiency (R%) and
adsorption ability (Qe, g-DCW/g-magnetic flocculants) of Fe3O4-chito­ (9)
san for Chlorella vulgaris could be calculated according to Eqs. (2) and The above three external operating factors in Eq. (8), (n = 3, x1 = the
(3), respectively. concentration of microalgae cells, x2 = solution pH, x3 = the dosage of
y = 0.3797x − 0.0539 (1) flocculant) are taken as independent variables, and Y = harvesting ef­
ficiency is taken as a dependent variable. The specific equation is shown
C0 − Ct in Eq. (9). Where Y refers to the harvesting efficiency, b0 is a constant
R(%) = × 100% (2) coefficient, b1, b2 and b3 are the linear coefficients corresponding to
C0
microalgae concentration, pH, and the dosage of flocculant, respec­
(C0 − Ct ) × V tively. b11, b22, b33, b12, b13 and b23 are the corresponding quadratic
Qe = (3)
m coefficients, respectively.

where C0 (g/L) is the initial concentration of microalgae cells in the 2.6. Reusability of spent medium and Fe3O4-chitosan
culture broth before harvesting, Ct (g/L) is the biomass concentration of
supernatant at harvesting equilibrium t (min). V (L) represents the Water resource is a crucial factor that affecting the cost of harvesting
volume of suspension microalgae cell, the m (g) is the amount of Fe3O4- process. The culture medium after harvesting of Chlorella vulgaris by
chitosan used for microalgae cells harvesting. Fe3O4-chitosan were reused for the re-cultivation of next batch of
microalgae cells. In the supernatant re-cultivation experiment, 100 mL
2.5. The modeling analysis supernatant after harvesting and 50 mL supernatant + 50 mL BG11 were
used as culture medium for microalgae re-cultivation, and the 100 mL
In order to further analyze the flocculation mechanism of Fe3O4- BG11 fresh culture medium was used as blank control group. The growth
chitosan on microalgae cells suspensions, Langmuir and Freundlich process of the re-cultivation microalgae in the recycled culture medium
adsorption isotherm models were used to fit the experimental data. The was measured daily by measuring the OD680.
linear equations of the two isotherm models are shown in Eqs. (4) and To investigate the reusability of Fe3O4-chitosan and reduce the
(5), respectively. microalgae harvesting cost, the regeneration experiment was carried out
Ce 1 Ce to recycle for microalgae harvesting as follows according to the modified
= + (4) methods by [24] method: first, the magnetic flocs were suspended in 10
Qe Qm KF Qm
mL NaOH solution of pH (10) for 20 min, and next, the mixed solution
1 were ultrasound treatment for 10 min under 40 kHz. Then, the methanol
logQe = logKF + logCe (5) (2 mL) and chloroform (2 mL) solution were added to the mixed solution
n
under ultrasound treatment for 20 min. The regenerated flocculant
Ce (g/L) and Qe (g/g) represent the equilibrium concentration of
(Fe3O4-chitosan) was collected by an external permanent field for 20
microalgae cells and the adsorption capacity of magnetic flocculant after
min and washed repeatedly by deionized water. Subsequently, it was
harvesting, respectively. Qm (g/g) represent the maximum adsorption
reused after drying. The microalgae harvesting experiments by regen­
efficiency of Fe3O4-chitosan. The KL (L/g) is the constant of Langmuir
erated Fe3O4-chitosan were constructed for 4 times. The harvesting ef­
model, and KF and 1/n are the constants of the Freundlich. These con­
ficiency of each cycled was measured by OD680.
stants are related to adsorption capacity, strength, and heterogeneous
sorption sites.

3
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

3. Results and discussion 79-0418), these peaks can be assigned to the pure Fe3O4-chitosan with a
spinel structure. It can be observed that the crystal structure of Fe3O4-
3.1. Properties of microalgae and nanoparticles chitosan was not significantly changed after the modification process,
suggesting that the cross-linking process of chitosan did not lead to the
3.1.1. SEM and EDS analysis phase transformation of Fe3O4 particles. The six characteristics peaks of
The images of SEM exhibited the surface morphology of Fe3O4, 30.1◦ , 35.5◦ , 43.2◦ , 53.5◦ , 57.0◦ and 62.8◦ were correspond to (2 2 0),
Fe3O4-chitosan nanocomposites before and after harvesting process, (3 1 1), (4 0 0), (4 2 2), (5 1 1) and (4 4 0), respectively, which further
respectively (Fig. 1a–c). SEM images of Fe3O4 showed a relatively uni­ indicated the ferromagnetic properties of Fe3O4-chitosan. It can be
form distribution and less aggregation. The Fe3O4-chitosan image pre­ concluded that the magnetic flocs could be easily separated from the
sented an obviously loose interlaced porous structure. And there are tiny culture medium quickly through extent magnetic field conditions. The
particles that are loaded onto the surface. It is worth noting here that the results are similar to those of [32] et al. study. In order to further
abundant pores and uneven surface can provide a platform for the evaluate its magnetic properties, the relative magnetic performance
microalgae collecting. It is expected that this large porous structure may curves were measured by VSM in the following part.
be beneficial to the flocculation of microalgae cells. In addition, the EDS The magnetic properties of Fe3O4-chitosan and its after harvesting
analysis results are also confirmed that the presence of C, O, N and Fe were studied by VSM analysis at room temperature. As shown in Fig. 2,
elements, indicating that Fe3O4-chitosan nanoparticles have been suc­ the saturation magnetization of Fe3O4-chitosan and its after harvesting
cessfully prepared. The SEM images of magnetic microalgae flocs after were 11.574 and 7.275 emu/g, respectively. It can be clearly found that
harvesting by Fe3O4-chitosan flocculant were shown in Fig. 1c. It can be the Fe3O4-chitosan was superparamagnetic properties, which indicated
found that algal cells could be attached on the surface of Fe3O4-chitosan that the composite materials-flocs were easy to be separated from cul­
to form flocs, which indicated that the Fe3O4-chitosan had a good floc­ ture medium. Notably, the regenerated Fe3O4-chitosan also showed
culation effect with Chlorella vulgaris cells. excellent magnetic properties, which proved that the Fe3O4-chitosan
could not only improved the harvest speed and efficiency, but also could
3.1.2. Surface area, crystal phases and VSM analysis be resulted the reduction of cost of harvesting process by external
The specific surface area (SSA) and pore volume of Fe3O4 and Fe3O4- magnetic field. In addition, the separation performance of magnetic
chitosan nanoparticles were calculated to be about 6.96 and 20.14 m2. flocs from the culture medium could be verified by the insets shown in
g− 1, 0.01 and 0.03 cm3.g− 1, respectively, demonstrating the increase in Fig. 2c. It was confirmed that the harvested magnetic flocs could be
SSA and pore volume by the cross-linking process of chitosan (Fig. 2a). separated from the culture medium by permanent magnet. Therefore, it
The adsorption/desorption image of Fe3O4-chitosan was attributed to could be concluded that the magnetic process makes flocs easier sepa­
IV-type isothermal curve with a large hysteresis loop, which indicated ration from the following culture medium.
that the Fe3O4-chitosan has a mesoporous structure. The results showed
that the average pore diameter was calculated to be 5.03 nm. Large SSA, 3.1.3. Functionalization of Fe3O4-chitosan
pore volume and appropriate pore diameter are favorable for the In order to study the existence of magnetic functional groups coating,
adsorption of microalgae cells, because it can improve the surface the surface functional groups of pure Fe3O4, Fe3O4-chitosan and its after
accessibility of microalgae cells suspensions, and carry out the floccu­ harvesting were detected by FTIR technology, and the results are shown
lation process with more surface-active sites [30,31]. in Fig. 2d. The peak at 574 cm− 1 corresponded to the Fe–O group in the
The XRD patterns of Fe3O4 and Fe3O4-chitosan were shown in images of pure Fe3O4 [33]. In the case of Fe3O4-chitosan, a similar Fe–O
Fig. 2b, respectively. According to the analysis (JCPDS database file, No. group (574 cm− 1) also appeared, demonstrating that the Fe3O4 was

Fig. 1. The SEM images of Fe3O4 (a), Fe3O4-chitosan nanocomposites (b) and its after harvesting (c).

4
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

Fig. 2. The physical/chemical characteristics of modified samples, (a) Specific surface area, (b) XRD patterns, (c) VSM and (d) FTIR analysis.

successfully encapsulated on chitosan [34]. The peak at 3415 cm− 1 was 400.20 eV were attributed to –NH and C(O)–NH, respectively [24]
attributed to the stretching vibration of O–H, and the peak at 2925 and (Fig. 3c). Fig. 3d showed the binding energies of Fe2p peaks at 724.2 eV
2858 cm− 1 were assigned to the stretching vibration of C–H of the and 710.9 eV referring to the 2p1/2 and 2p3/2, respectively [37]. The
polymer, respectively [35]. The results were like the study by [32]. The above results showed that the modification of Fe3O4 by biopolymer
peak of 1653 cm− 1 corresponded to the stretching and bending vibration chitosan was successfully. These additional functional groups can pro­
of N–H skeleton, and the 1424 cm− 1 was assigned to the stretching vide more adsorption sites for the capture of microalgae cells and further
vibration of C–O of primary alcoholic groups [35]. The peak of C–O improve the harvesting capacity.
bending vibration and N–H bending vibration shifted from 1370 cm− 1
to 1382 cm− 1, from 1660 cm− 1 to 1720 cm− 1, respectively [24]. The
reduction of peaks intensity and the displacement of peaks location 3.2. Assessment of magnetic harvesting procedures
suggested that the microalgae cells were in contacted with these func­
tional groups. 3.2.1. Effects of various ratios and dosages of Fe3O4-chitosan on
The XPS survey of Fe3O4, Fe3O4-chitosan and chitosan are displayed microalgae harvesting
in Fig. S3. It could be seen that there is no Fe peak on the image of single Fig. 4a showed the correlation between the harvest efficiency and the
chitosan. After modification, Fe3O4-chitosan mainly included that C1s, dosage of Fe3O4-chitosan at different ratio of chitosan/Fe3O4 (α =
N1s, O1s, and Fe2p, indicating that Fe3O4 has been loaded onto the 0.5,1,2 and 3). When the ratios were 0.5, 1, 2 and 3, the harvesting ef­
surface of chitosan. The specific functional groups of Fe3O4-chitosan ficiency of magnetic chitosan on microalgae cells suspension reached
were qualitatively analyzed by high-resolution XPS, and the results are 66.36, 68.23, 80.00 and 97.81%, respectively. It could be observed that
shown in Fig. 3a–d. There are two peaks in high resolution spectrum of the harvested efficiency of Fe3O4-chitosan for microalgae cells was
C1s, positioned at 287.1 and 285.4 eV, which assigned to C–O (alkoxy) positively correlated with the ratios of chitosan/Fe3O4, and the higher
and C– – C bonds in aromatic ring, respectively (Fig. 3a) [36]. The two the proportion of chitosan/Fe3O4, the higher the harvested efficiency.
peaks located at 531.5 and 533.10 eV corresponded to H–O and COOH When the ratio of flocculant was 3, the collecting capacity reached
(carboxyl), respectively (Fig. 3b) [30]. Two peaks placed in 402.5 and 97.81%, which might be because more chitosan particles could provide
positive charge to the microalgae cells and enhance their charge

5
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

Fig. 3. The high-resolution spectrum of C1s (a), O1s (b), N1s (c) and Fe2p (d), respectively.

Fig. 4. The effect of different ratio and dosage (a) and pH (b) and on microalgae harvesting by Fe3O4-chitosan.

6
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

neutralization ability. Notably, the zeta potential of Fe3O4-chitosan (α = harvesting process [38,39]. High harvesting cost has always been one of
0.5, 1, 2 and 3) were 15.33, 15.60, 19.74 and 23.80 mV at pH 3.0, the bottlenecks that restricting the application of microalgae cells and
respectively, and the positive charge increases with increase of the the rapid development of the industry. The use of low dosage flocculant
amount of chitosan. Thus, in subsequent harvesting experiment, the could be saved the cost of collecting microalgae process when it is used
ratio of 3 was selected as flocculant to collect microalgae cells. in commercial production on a large scale.
The dosage of flocculants plays an important role in harvesting
process of microalgae cells process, since both excessive or insufficient 3.2.2. Effects of settle time on microalgae harvesting
dosage will affect the harvesting cost and efficiency of microalgae cells To save the cost of harvesting time, the effect of magnetic flocs
for bioproducts production on a large scale. Hence, the selection of precipitation time on magnetic separation was studied. As can be seen
appropriate amount of flocculant is a necessary condition for microalgae from Fig. 5c, with the sedimentation time increasing from 0.5 min to 10
harvesting process. Taking the Fe3O4-chitosan (α = 3) as an example, min, the harvesting efficiency first increased and then stabilized at
when the dosage of Fe3O4-chitosan increased from 0.1 to 0.7 g/L, the 95.49%, indicating that the magnetic flocs were closely connected and
harvest efficiency ranged from 45.27 to 97.81 and then to 94.00 %, kept in good condition, and the harvesting efficiency would not decrease
which exhibited a tendency of increasing at first and then decreasing. with the increase of time. This flocculant could be connected microalgae
However, when the content of Fe3O4-chitosan increased to more than cells more tightly through strong bonds on the surface of microalgae. A
0.5 g/L, the harvesting efficiency of Fe3O4-chitosan decreased slightly, study by [40] reported a magnetic separation time of 27 min for
which might be due to the destabilization of microalgae flocs caused by Chlorella vulgaris with a magnetic flocculant. As seen from Fig. 5c, when
the electrostatic repulsion of –NH3 on the chitosan, resulting in insta­ the settle time was only 30 s, the collected capacity reached 92.56%. The
bility between microalgae cells suspension and bioflocculant system magnetic flocs were able to aggregated together and have high magnetic
[17]. This results are similar to the research of [28]. When the ratio of and weight, thus the Fe3O4-chitosan-microalgae cells flocs could be
chitosan/Fe3O4 was 3, and the dosage was 0.5 g/L, the harvesting per­ achieved magnetic separation in a short time. It is concluded that when
formance of Fe3O4-chitosan could be achieved 97.81%. Consequently, the precipitation time was 3 min, the harvesting efficiency of Fe3O4-
Fe3O4-chitosan (α = 3, 0.5 g/L) will be selected for the subsequent chitosan for microalgae cells was higher, which saved the time for the
microalgae harvesting experiment in this study. Notably, the amount of harvesting process, and thus reduced the flocculant cost.
flocculant was closely related to the harvesting efficiency and the cost of

Fig. 5. The fitting results of Langmuir (a) and Freundlich (b) isotherm models for microalgae harvesting by Fe3O4-chitosan. The effect of settling time on microalgae
harvesting (c).

7
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

3.2.3. Effects of pH on microalgae harvesting changed from 5 to 12, the surface of both magnetic chitosan and
The pH value influences the interaction between the flocculant and microalgae cells are negatively, but the harvesting efficiency increased
the microalgae cells by affecting the zeta potential value. The metabolic slowly, indicating that the electrostatic interactions is not the only
activity of microalgae is highly pH dependent, and recent studies have mechanism, and there are other flocculation mechanisms between
confirmed that Chlorella vulgaris preferred a relatively alkaline envi­ flocculant and algae cells. Based on the above analysis, it can be proved
ronment [41]. Thus, this chapter studied the influence of pH value on that pH has greatly influence on the harvesting process of Fe3O4-chito­
the harvesting process, hoping to find the best pH to obtain a higher san and algae cells. The relatively low pH was favorable for the col­
harvest efficiency [42], as shown in Fig. 4b. The trapping efficiency of lecting process. Thus, the subsequent harvesting experiment of Chlorella
Fe3O4-chitosan fluctuated wildly depending on the different pH values. vulgaris was carried out at low pH 3.0.
When the pH was 3, the Fe3O4-chitosan had the strongest harvesting
efficiency ability on microalgae cells. The zeta potential of microalgae 3.2.4. Adsorption isotherms and possible harvesting mechanism
cells suspension and Fe3O4-chitosan was − 8.14 mV and + 23.8 mV, Langmuir and Freundlich isotherm model were used to further
respectively, which might be attributed to the charge neutralization. evaluate the interaction between Fe3O4-chitosan and microalgae cells,
Because the –NH3 is easily protonated under acidic conditions, the the fit results are shown Fig. 5a–b, respectively. The fitting parameters
surface of Fe3O4-chitosan has a positive charge, so it can effectively (Qm, KF, 1/n, R2, χ2 and RMSE) were calculated from the Eqs. (4), (5),
capture the microalgae cells through charge neutralization [43,44]. (6) and (7), respectively. These values are summarized in the insets of
Consequently, the charge neutralization was existed between micro­ Fig. 4a–b. Based on the correlation factors (R2), the fit results of
algae cells and Fe3O4-chitosan. With the increase of pH from 2 to 5, the Freundlich model (R2 = 0.98) were better than that of the Langmuir
harvesting amount obviously decreased from 95.22% to 68.04%. Com­ model (R2 = 0.65), the RMSE of the Freundlich model (RMSE = 2.97)
bined with the zeta potential analysis, the values of Fe3O4-chitosan was lower than the Langmuir model (RMSE = 6.95). This demonstrated
decreased from +37.13 mV to − 30.63 mV with increasing of pH, while that the harvesting process of microalgae by Fe3O4-chitosan composites
the microalgae cells maintained a negative charge when the pH was flocculant was a multilayer adsorption process on heterogeneous sites
2–12. Therefore, the harvesting efficiency decreased, which could be [23]. The maximum collecting capacity that calculated from the
caused by the electrostatic repulsion. Specially, when the pH values Freundlich model, which described by the KL was 73.45 g/g. The high

Fig. 6. The images of light microscope photos of free Chlorella vulgaris cells (a), Fe3O4-chitosan (b), Fe3O4-chitosan-microalgal cells after harvesting (c). The effect of
cycle number on microalgae harvesting (d).

8
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

collecting capacity was related to the SSA, because the large SSA could and adsorption bridging.
be provided more adsorption sites for microalgae cells. Moreover, the
value of 1/n > 1, suggesting that the adsorption process of microalgae 3.2.5. Modeling
cells was a complex process. A study by [40] investigated that the According to Eq. (9), the harvesting efficiency of magnetic flocculant
maximum monolayer adsorption capacity of Fe3O4 NPs-I obtained was on Chlorella vulgaris was evaluated. The simulation results are compared
3.49 mg/g. Significantly, the pure Fe3O4 nanoparticles also have a with the original harvest data as shown in Fig. 7a. The simulation results
certain adsorption ability to microalgae cells, and the adsorption ca­ are in good agreement with the original data, and the correlation coef­
pacity was enhanced after binding with chitosan [45]. ficient (R2) is about 0.8624. The simulation coefficient obtained by
There are four main types of flocculation mechanisms, namely, MATLAB software is shown in Eq. (10).
charge neutralization, sweep flocculation, bridging and double layer
Y = 73.465 + 53.467x1 − 7.7315x2 + 121.6289x3 − 48.9180x21 + 1.8088x22
compression. According to the above analysis of pH and zeta potential
results, the harvesting efficiency of Fe3O4-chitosan composites − 88.0723x23 − 9.8753x1 x2 + 75.2511x1 x3 − 19.6919x2 x3 (R2 = 0.8624)
decreased from 95.22% to 47.01% at pH 2–6, and then gradually (10)
increased 68.04% at pH 12. When the value of pH was 2–12, the zeta
Based on Eq. (10), b2 was negative, indicating that the moderate pH
potential of microalgae cells suspension was negative in the range of
is beneficial to the harvesting process of microalgae cells by magnetic
− 3.14 mV- − 25.47 mV. When the value of pH was 2–10, the zeta po­
chitosan, while b1 and b3 are negative values, demonstrating that the
tential of Fe3O4-chitosan composites was positive charge, while negative
greater the concentration of microalgae cells and the higher dosage of
at pH 11–12. As reported, charge neutralization and double layer
magnetic chitosan are not conducive to the microalgae harvesting. From
compression are reportedly triggered when two particles have opposite
the above analysis, it can be further confirmed that microalgae with
charges [23]. Hence, these two harvesting mechanisms could explain
higher concentration requires a lower pH and a lower flocculant dosage.
the high harvesting efficiency under acidic conditions. When the pH
It is worth noting that b1 and b2 remain a low value, while b3 is much
values were 11–12, the zeta potential of Fe3O4-chitosan gradually be­
higher than b1 and b2, which further confirmed that the magnetic floc­
comes negative. However, the harvesting efficiency gradually increased,
culant content on microalgae harvesting process has a greater impact on
which suggested that the harvesting mechanism included not only
the harvest process of microalgae than that of pH and microalgae con­
electrostatic interactions. It was reported that the principle of sweeping
centration. In addition, the interaction coefficient of b11, b33, b12 and b23
flocculation depends on the formation of metal hydroxides. Obviously,
was negative, indicating that the microalgae concentration, pH, and the
sweeping flocculation was not a possible harvesting mechanism as it has
dosage of flocculant had positive effects on each other. However, b22 and
no effect in either acidic or neutral solutions [46].
b13 were positive, indicating that the simultaneous interaction between
Fig. 6a–c displayed the images of light microscope photos of Fe3O4-
pH, microalgae concentration and the dosage of flocculant had no po­
chitosan, microalgal cells, and Fe3O4-chitosan-microalgal cells flocs,
sition effect on microalgae harvesting performance. In conclusion, the
respectively. Microalgae cells were observed to be loaded on the surface
similarity between the theoretical data and experimental data and the
of Fe3O4-chitosan and to be gathered to form magnetic flocs. Conse­
reasonable fitting coefficient are shown in Fig. 7a, which proved that the
quently, adsorption bridging was a possible mechanism for magnetic
second-order polynomial model of magnetic chitosan for Chlorella vul­
flocculant to capture microalgae cells. Hence, in addition to the har­
garis could effectively analyze the experimental results and establish the
vesting mechanism of charge neutralization and double layer compres­
external operating impact factors-activity relationship between mag­
sion, Fe3O4-chitosan could be adsorbed and bridged microalgae cells.
netic chitosan and external factors.
This result was in accordance with the results of Suparmaniam et al.,
who found that charge neutralization and sweeping were the floccula­
3.2.6. Reusability of Fe3O4-chitosan and medium culture
tion mechanisms of shell waste-derived bioflocculants for Chlorella vul­
To investigate the reusability of Fe3O4-chitosan, the reuse experi­
garis harvesting [47]. Based on the above analysis results, it was
ments were carried out, as shown in Fig. 6d. With the increase of the
concluded that the harvesting mechanism of Chlorella vulgaris by Fe3O4-
number of cycles, the harvesting efficiency decreased gradually. After
chitosan involved in charge neutralization, double layer compression
four desorption experiments, the collection harvesting remained at

Fig. 7. The original and simulative harvesting efficiency data based on Eq. (10) by using Fe3O4-chitosan with various operating factors(a); Microalgae growth in the
100 mL reused medium after harvesting, 50 mL medium culture (50 mL reused medium + 50 mL BG11) and 100 mL BG11, respectively (b).

9
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

85.10%. The Fe3O4-chitosan could be preserved their magnetic prop­ in Table S1. The chemicals agents used in preparation process of Fe3O4-
erties (7.275 emu/g), which was suitable for the microalgae biomass and chitosan were analytically pure. When the concentration of Fe3O4-chi­
culture medium separation after harvesting process. Cost is an important tosan was 0.5 g/L, the harvesting efficiency reached more than 97%, and
factor in collecting microalgae on a large scale. The recovery and utili­ the harvest dry weight was 1.05 g/L. The cost per ton of Fe3O4-chitosan
zation of flocculant could be reduced the cost of harvesting process and was approximately ~7452US$, thus the harvested cost per ton of dry
improved the economic benefit of magnetic chitosan. Consequently, the weight microalgae biomass was about ~3548.64 US$. Table 1 compared
Fe3O4-chitosan could be used as a flocculant for the microalgae har­ the cost of Chlorella vulgaris harvesting by different flocculants. The cost
vesting with low cost and excellent reusability. However, it was worth of the Fe3O4-chitosan (~3548.64US$) proposed in this study is lower
noting that the recovery efficiency of magnetic flocculant was not high, than that of chitosan (~7280US$), aluminum sulphate/chitosan
because the magnetic flocs were tightly connected. (~4920 US$) and ferric chloride/chitosan (~7925US$), indicating that
The large portion of the microalgae cultivation is assigned to water the magnetic harvesting is expected to harvest Chlorella vulgaris effec­
due to the low concentration of microalgae cells. In the microalgae tively with low cost and energy consumption. Cost of Fe3O4-chitosan
cultivation on a large scale, the possible advantages and disadvantages applied in this work were relative higher than that of GO-Fe3O4/PDDA
of culture medium reuse have been controversial. Recent studies have (~2652US$), Al2(SO4)3 (~28US$), Flopam (~157US$) and Tanfloc
amply demonstrated the possibilities and benefits of water recycling and (~38US$). The Al2(SO4)3 flocculant were less expensive, costing US$28
the challenges associated with long-term reuse of water resources. per ton of harvesting biomass. However, the inorganic salts are not
Therefore, it is very necessary to reduce the cost of microalgae cultiva­ recommended for microalgae harvesting on a large scale due to the re­
tion and harvesting through water recycling. Considering the high effi­ sidual metal pollution in the biomass. On the other hand, some poly­
ciency of Fe3O4-chitosan in the recovery of microalgae cells and mers, likes Flopam and Tanfloc, were much more expensive than salts.
reusability of culture medium, the subsequent after harvesting process However, dispersion of polyacrylamide oligomers might be harmed to
was recycled and re-cultivated. To demonstrate the biocompatibility of the natural environment and health of human beings. However, mag­
magnetic chitosan, the supernatant after harvesting by Fe3O4-chitosan netic chitosan is more simple, rapid, and efficient than other chitosan-
as culture medium to re-cultivation the microalgae cells. In the reuse of based flocculants. Recycling culture medium and bioflocculant could
culture medium, 100 mL reused medium, 50 mL reused medium + 50 be reduced the cost of harvesting process, the economic feasibility could
mL fresh BG11, and 100 mL fresh BG11 were used as the culture medium be improved by further decreasing the preparation process costs and
for re-cultivation, respectively, as shown in Fig. 7b. The experimental energy consumption. Please note that it is expected that the existing
results showed that Chlorella vulgaris could grow continuously in 50 mL bottlenecks and difficulties will be overcome and the considerable costs
reused medium + 50 mL fresh BG11 circulating water system without will be obtained through further technical improvements and the
decline. The results indicated that the Fe3O4-chitosan had no influence establishment of a market system. It should emphasize that the cost
on the growth of microalgae cells and the culture medium after har­ feasibility analysis is not to give the best preparation method, but to
vesting process could be reused for microalgae re-cultivation. This work provide technical and data reference for the practical application of
was similar to the study by [28]. Therefore, Fe3O4-chitosan could be harvesting microalgae.
considered as a biocompatible flocculant for microalgae harvesting,
which was beneficial to the microalgae re-cultivation and cost saving. 4. Conclusions

3.3. Cost feasibility analysis A magnetic flocculant (Fe3O4-chitosan) was successfully prepared for
the rapid harvesting of Chlorella vulgaris. When the pH was 3.0, the mass
The harvesting process accounts for 20–30% to the total cost of the ratio of chitosan/Fe3O4 was 3:1, and the dosage was 0.5 g/L, the har­
microalgae biomass production [20,48]. In order to investigate the vesting efficiency was over 97% within 3 min. This process was
economic feasibility of collecting Chlorella vulgaris with Fe3O4-chitosan accompanied by a multilayer adsorption on heterogeneous sites. The
composites, the cost feasibility analysis was carried out with the dosage mechanisms of rapid separation between Fe3O4-chitosan composites and
of Fe3O4-chitosan flocculant, the preparation technology, harvesting microalgae cells were mainly double layer compression, adsorption
performance and biomass production as the main parameters. Notably, bridging and charge neutralization. The second-order polynomial model
considering the high harvesting efficiency and the four cycles of mag­ could better simulate the process of microalgae harvesting by Fe3O4-
netic flocculant as well as the reuse of water resources after harvesting chitosan (R2 = 0.8624), and construct the constitutive relationships
process, we converted the cost of raw flocculants to 30%. Moreover, the between Fe3O4-chitosan and microalgae cell concentration, pH, and
costs associated with the harvesting energies also accounts for 30% of flocculant dose. The recultivation of supernatant and recyclability of
the total costs [49]. The harvesting costs were expressed in US$ per Fe3O4-chitosan saved the cost of microalgae harvesting. The cost feasi­
metric ton of dry weight biomass. bility analysis pointed out the flocculant cost per ton of biomass har­
The specific calculation process of the total harvesting cost is shown vested (3548.64 US$) were lower than that of other chitosan-based

Table 1
Cost analysis of harvesting Chlorella vulgaris using different flocculants.
Flocculant (Dosage Harvesting Biomass Flocculants needed per ton of Flocculant cost (US$ Flocculant cost per ton of References
g/L) efficiency (%) harvested (g/L) biomass harvested (ton) per ton -1) biomass harvested (US$)

Fe3O4-chitosan 97 1.05 0.4762 7452 3548.64 In this

(0.5) study
Chitosan – – – 20,000–50,000 7280 [21]
Aluminum 80% – – – 4920 [21]
sulphate/chitosan
Ferric chloride/ 80% – – – 7925 [21]
chitosan
GO-Fe3O4/PDDA 95.35 0.191 0.3665 7237 2652 [50]
(0.07)
Al2(SO4)3 (0.200) 85 0.213 0.094 300 28 [51]
Flopam (0.005) 98 0.255 0.020 8000 157 [49]
Tanfloc (0.005) 100 260 0.019 2000 38 [49]

10
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

flocculants. Therefore, the Fe3O4-chitosan is an efficient, low-cost bio­ [11] S. Li, P. Wang, C. Zhang, X. Zhou, Z. Yin, T. Hu, D. Hu, C. Liu, L. Zhu, Influence of
polystyrene microplastics on the growth, photosynthetic efficiency, and
flocculant for application in microalgae harvesting, which should be
aggregation of freshwater microalgae Chlamydomonas reinhardtii, Sci Total
continuously explored and developed in future microalgae harvesting Environ 714 (2020) 136767.
studies. [12] J. Yang, M. Xu, X. Zhang, Q. Hu, M. Sommerfeld, Y. Chen, Life-cycle analysis on
biodiesel production from microalgae: water footprint and nutrients balance,
Bioresour. Technol. 102 (2011) 159–165.
CRediT authorship contribution statement [13] Y. Li, Y.F. Chen, P. Chen, M. Min, W. Zhou, B. Martinez, J. Zhu, R. Ruan,
Characterization of a microalga Chlorella sp. well adapted to highly concentrated
Zhihong Yin: Conceptualization, Methodology, Investigation, Data municipal wastewater for nutrient removal and biodiesel production, Bioresour.
Technol. 102 (2011) 5138–5144.
curation, Writing – original draft, Writing – review & editing. Lingbo [14] R. Chu, S. Li, L. Zhu, Z. Yin, D. Hu, C. Liu, F. Mo, A review on co-cultivation of
Zhang: Resources, Visualization, Supervision, Funding acquisition. Dan microalgae with filamentous fungi: Efficient harvesting, wastewater treatment and
Hu: Resources, Visualization, Supervision, Funding acquisition. biofuel production, Renew. Sust. Energ. Rev. 139 (2021) 110689.
[15] X. Zang, H. Zhang, Q. Liu, L. Li, L. Li, X. Zhang, Harvesting of Microcystis flos-
Shuangxi Li: Validation, Data curation, Visualization, Project admin­ aquae using chitosan coagulation: Influence of proton-active functional groups
istration, Funding acquisition. Ruoyu Chu: Validation, Data curation, originating from extracellular and intracellular organic matter, Water Res. 185
Visualization, Project administration, Funding acquisition. Chenchen (2020) 116272.
[16] L. Christenson, R. Sims, Production and harvesting of microalgae for wastewater
Liu: Formal analysis, Supervision. Yuanfei Lv: Formal analysis, Super­ treatment, biofuels, and bioproducts, Biotechnol. Adv. 29 (2011) 686–702.
vision. Jianfeng Bao: Project administration. Mingdeng Xiang: Re­ [17] R. Dineshkumar, A. Paul, M. Gangopadhyay, N.D.P. Singh, R. Sen, Smart and
sources, Visualization, Supervision, Funding acquisition. Liandong Reusable Biopolymer Nanocomposite for Simultaneous Microalgal Biomass
Harvesting and Disruption: Integrated Downstream Processing for a Sustainable
Zhu: Validation, Writing – review & editing, Visualization, Supervision, Biorefinery, ACS Sustain. Chem. Eng. 5 (2016) 852–861.
Funding acquisition. [18] T.H. Ang, K. Kiatkittipong, W. Kiatkittipong, S.C. Chua, J.W. Lim, P.L. Show, M.J.
K. Bashir, Y.C. Ho, Insight on Extraction and Characterisation of Biopolymers as the
Green Coagulants for Microalgae Harvesting, Water 12 (2020) 1388.
Declaration of Competing Interest [19] Y. Huang, C. Wei, Q. Liao, A. Xia, X. Zhu, X. Zhu, Biodegradable branched cationic
starch with high C/N ratio for Chlorella vulgaris cells concentration: Regulating
The authors declare that they have no known competing financial microalgae flocculation performance by pH, Bioresour. Technol. 276 (2019)
133–139.
interests or personal relationships that could have appeared to influence [20] Z. Yin, D. Hu, X. Li, Y. Zhong, L. Zhu, Shell-derived chitosan as a green flocculant to
the work reported in this paper. harvest microalgae for biofuel production, Biofuel Bioprod. Bior. 15 (2021)
637–645.
[21] H.P. Vu, L.N. Nguyen, G. Lesage, L.D. Nghiem, Synergistic effect of dual
Acknowledgments flocculation between inorganic salts and chitosan on harvesting microalgae
Chlorella vulgaris, Environ. Technol. Inno. 17 (2020).
This work was supported by the National Key R&D program of China [22] A.L. Ahmad, N.H.M. Yasin, C.J.C. Derek, J.K. Lim, Optimization of microalgae
coagulation process using chitosan, Chem. Eng. J. 173 (2011) 879–882.
(2019YFC1803405, 2019YFD1101300) and the Start-up Grant (1305/ [23] S.K. Wang, F. Wang, Y.R. Hu, A.R. Stiles, C. Guo, C.Z. Liu, Magnetic flocculant for
600460004) from the Wuhan University in China. high efficiency harvesting of microalgal cells, ACS Appl. Mater. Interfaces 6 (2014)
109–115.
[24] T. Wang, W.L. Yang, Y. Hong, Y.L. Hou, Magnetic nanoparticles grafted with
Appendix A. Supplementary material amino-riched dendrimer as magnetic flocculant for efficient harvesting of
oleaginous microalgae, Chem. Eng. J. 297 (2016) 304–314.
Supplementary data to this article can be found online at https://doi. [25] K. Gerulová, A. Bartošová, L. Blinová, K. Bártová, M. Dománková, Z. Garaiová,
M. Palcut, Magnetic Fe3O4-polyethyleneimine nanocomposites for efficient
org/10.1016/j.seppur.2021.119679. harvesting of Chlorella zofingiensis, Chlorella vulgaris, Chlorella sorokiniana,
Chlorella ellipsoidea and Botryococcus braunii, Algal Res. 33 (2018) 165–172.
References [26] Z. Lu, S. Loftus, J. Sha, W. Wang, M.S. Park, X. Zhang, Z.I. Johnson, Q. Hu, Water
reuse for sustainable microalgae cultivation: Current knowledge and future
directions, Resour. Conserv. Recy. 161 (2020) 104975.
[1] R.A. Voloshin, M.V. Rodionova, S.K. Zharmukhamedov, T. Nejat Veziroglu, S.
[27] S. Li, R. Chu, D. Hu, Z. Yin, F. Mo, T. Hu, C. Liu, L. Zhu, Combined effects of 17β-
I. Allakhverdiev, Review: Biofuel production from plant and algal biomass, Int. J.
estradiol and copper on growth, biochemical characteristics and pollutant
Hydrogen Energ. 41 (2016) 17257–17273.
removals of freshwater microalgae Scenedesmus dimorphus, Sci. Total Environ.
[2] S.S. Rosli, J.W. Lim, K. Jumbri, M.K. Lam, Y. Uemura, C.D. Ho, W.N. Tan, C.
730 (2020) 138597.
K. Cheng, W.N.A. Kadir, Modeling to enhance attached microalgal biomass growth
[28] K. Lee, S.Y. Lee, J.G. Na, S.G. Jeon, R. Praveenkumar, D.M. Kim, W.S. Chang, Y.
onto fluidized beds packed in nutrients-rich wastewater whilst simultaneously
K. Oh, Magnetophoretic harvesting of oleaginous Chlorella sp. by using
biofixing CO2 into lipid for biodiesel production, Energ. Convers. Manage. 185
biocompatible chitosan/magnetic nanoparticle composites, Bioresour. Technol.
(2019) 1–10.
149 (2013) 575–578.
[3] L. Zhu, Z. Wang, J. Takala, E. Hiltunen, L. Qin, Z. Xu, X. Qin, Z. Yuan, Scale-up
[29] P. Hu, S. Zhuang, S. Shen, Y. Yang, H. Yang, Dewaterability of sewage sludge
potential of cultivating Chlorella zofingiensis in piggery wastewater for biodiesel
conditioned with a graft cationic starch-based flocculant: Role of structural
production, Bioresour. Technol. 137 (2013) 318–325.
characteristics of flocculant, Water Res. 189 (2021) 116578.
[4] L. Zhu, S. Li, T. Hu, Y.K. Nugroho, Z. Yin, D. Hu, R. Chu, F. Mo, C. Liu, E. Hiltunen,
[30] Z. Yin, Y. Liu, S. Liu, L. Jiang, X. Tan, G. Zeng, M. Li, S. Liu, S. Tian, Y. Fang,
Effects of nitrogen source heterogeneity on nutrient removal and biodiesel
Activated magnetic biochar by one-step synthesis: Enhanced adsorption and
production of mono-and mix-cultured microalgae, Energ. Convers. Manage. 201
coadsorption for 17β-estradiol and copper, Sci. Total Environ. 639 (2018)
(2019) 112144.
1530–1542.
[5] N. Kumar, C. Banerjee, N. Kumar, S. Jagadevan, A novel non-starch based cationic
[31] Z. Yin, Y. Liu, X. Tan, L. Jiang, G. Zeng, S. Liu, S. Tian, S. Liu, N. Liu, M. Li,
polymer as flocculant for harvesting microalgae, Bioresour. Technol. 271 (2019)
Adsorption of 17β-estradiol by a novel attapulgite/biochar nanocomposite:
383–390.
Characteristics and influencing factors, Process. Saf. Environ. 121 (2019) 155–164.
[6] Z. Yin, L. Zhu, S. Li, T. Hu, R. Chu, F. Mo, D. Hu, C. Liu, B. Li, A comprehensive
[32] W. Mxie, J. Wang, Immobilized lipase on magnetic chitosan microspheres for
review on cultivation and harvesting of microalgae for biodiesel production:
transesterification of soybean oil, Biomass Bioenerg. 36 (2012) 373–380.
Environmental pollution control and future directions, Bioresour. Technol. 301
[33] M. Barekati-Goudarzi, M. Reza Mehrnia, F. Pourasgharian Roudsari, D. Boldor,
(2020) 122804.
Rapid separation of microalga Chlorella vulgaris using magnetic chitosan: Process
[7] C.N. Ogbonna, E.G. Nwoba, Bio-based flocculants for sustainable harvesting of
optimization using response surface methodology, Particul. Sci. Technol. 34 (2015)
microalgae for biofuel production. A review, Renew. Sust. Energ. Rev. 139 (2021)
165–172.
110690.
[34] S.H.A. Hassan, M. Koutb, N.A. Nafady, E.A. Hassan, Potentiality of
[8] S. Hena, L. Gutierrez, J.P. Croue, Removal of pharmaceutical and personal care
Neopestalotiopsis clavispora ASU1 in biosorption of cadmium and zinc,
products (PPCPs) from wastewater using microalgae: A review, J. Hazard. Mater.
Chemosphere 202 (2018) 750–756.
403 (2021) 124041.
[35] R. Ahmad, A. Mirza, Facile one pot green synthesis of Chitosan-Iron oxide (CS-
[9] S.R. Chia, K.W. Chew, P.L. Show, A. Xia, S.H. Ho, J.W. Lim, Spirulina platensis
Fe2O3) nanocomposite: Removal of Pb(II) and Cd(II) from synthetic and industrial
based biorefinery for the production of value-added products for food and
wastewater, J. Clean Prod. 186 (2018) 342–352.
pharmaceutical applications, Bioresour. Technol. 289 (2019) 121721.
[36] O. Akhavan, E. Ghaderi, H. Emamy, F. Akhavan, Genotoxicity of graphene
[10] A.A. Mohd-Sahib, J.W. Lim, M.K. Lam, Y. Uemura, M.H. Isa, C.D. Ho, S.R.M. Kutty,
nanoribbons in human mesenchymal stem cells, Carbon 54 (2013) 419–431.
C.Y. Wong, S.S. Rosli, Lipid for biodiesel production from attached growth
[37] S. Yuan, J. Gu, Y. Zheng, W. Jiang, B. Liang, S.O. Pehkonen, Purification of phenol-
Chlorella vulgaris biomass cultivating in fluidized bed bioreactor packed with
contaminated water by adsorption with quaternized poly(dimethylaminopropyl
polyurethane foam material, Bioresour. Technol. 239 (2017) 127–136.

11
Z. Yin et al. Separation and Purification Technology 279 (2021) 119679

methacrylamide)-grafted PVBC microspheres, J. Mater. Chem. A 3 (2015) Life-cycle assessment for screening species and harvesting methods, Algal Res. 54
4620–4636. (2021) 102226.
[38] U. Suparmaniam, M.K. Lam, Y. Uemura, J.W. Lim, K.T. Lee, S.H. Shuit, Insights [45] Y.R. Hu, C. Guo, X. Ling, F. Wang, S.K. Wang, Z.M. Hu, C.Z. Liu, A magnetic
into the microalgae cultivation technology and harvesting process for biofuel separator for efficient microalgae harvesting, Bioresour. Technol. 158 (2014)
production: A review, Renew. Sust. Energ. Rev. 115 (2019) 109361. 388–391.
[39] B. Behera, P. Balasubramanian, Natural plant extracts as an economical and [46] D. Gheraoed, B. Ghernaout, Sweep flocculation as a second form of charge
ecofriendly alternative for harvesting microalgae, Bioresour. Technol. 283 (2019) neutralization—A review, Desalin. Water Treat. 44 (2012) 15–28.
45–52. [47] U. Suparmaniam, M.K. Lam, Y. Uemura, S.H. Shuit, J.W. Lim, P.L. Show, K.T. Lee,
[40] A.A. Markeb, J. Llimos Turet, I. Ferrer, P. Blanquez, A. Alonso, A. Sanchez, Y. Matsumura, P.T.K. Le, Flocculation of Chlorella vulgaris by shell waste-derived
J. Moral-Vico, X. Font, The use of magnetic iron oxide based nanoparticles to bioflocculants for biodiesel production: Process optimization, characterization and
improve microalgae harvesting in real wastewater, Water Res. 159 (2019) kinetic studies, Sci. Total Environ. 702 (2020) 134995.
490–500. [48] Z. Yin, R. Chu, L. Zhu, S. Li, F. Mo, D. Hu, C. Liu, Application of chitosan-based
[41] J. González Camejo, A. Robles, A. Seco, J. Ferrer, M.V. Ruanoa, On-line monitoring flocculants to harvest microalgal biomass for biofuel production: A review, Renew.
of photosynthetic activity based on pH data to assess microalgae cultivation, Sust. Energ. Rev. 145 (2021) 111159.
J. Environ. Manage. 276 (2020) 111343. [49] F. Roselet, D. Vandamme, M. Roselet, K. Muylaert, P.C. Abreu, Screening of
[42] V. Okoro, U. Azimov, J. Munoz, H.H. Hernandez, A.N. Phan, Microalgae commercial natural and synthetic cationic polymers for flocculation of freshwater
cultivation and harvesting: Growth performance and use of flocculants-A review, and marine microalgae and effects of molecular weight and charge density, Algal
Renew. Sust. Energ. Rev. 115 (2019) 109364. Res. 10 (2015) 183–188.
[43] S. Acosta Ferreira, O.S. Castillo, J.T. Madera Santana, D.A. Mendoza Garcia, C. [50] P.R. Liu, H.L. Zhang, T. Wang, W.L. Yang, Y. Hong, Y.L. Hou, Functional graphene-
A. Nunez Colin, C. Grijalva Verdugo, A.G. Villa Lerma, A.T. Morales Vargas, J. based magnetic nanocomposites as magnetic flocculant for efficient harvesting of
R. Rodriguez Nunez, Production and physicochemical characterization of chitosan oleaginous microalgae, Algal Res. 19 (2016) 86–95.
for the harvesting of wild microalgae consortia, Biotechnol. Rep. 28 (2020) 00554. [51] D. Vandamme, I. Foubert, I. Fraeye, K. Muylaert, Influence of organic matter
[44] Q. Wang, K. Oshita, M. Takaoka, Flocculation properties of eight microalgae generated by Chlorella vulgaris on five different modes of flocculation, Bioresour.
induced by aluminum chloride, chitosan, amphoteric polyacrylamide, and alkaline: Technol. 124 (2012) 508–511.

12