Contents lists available at ScienceDirect

Bioresource Technology Reports

journal homepage: www.journals.elsevier.com/bioresource-technology-reports

นิมี น
ไอ อวน ทองแดง จาก ส ล ล โดย ฟาง ข้าวสาลี ดัดแปลง โผ ล้ เอทิลีน
ดูดซับ
การ .

Adsorption of copper ion from solution by polyethylenimine modified wheat

. .

straw
Jiajia Dong, Yangyang Du, Rusheng Duyu, Yu Shang, Shusheng Zhang, Runping Han
⁎

College of Chemistry and Molecular Engineering, Zhengzhou University, No 100 of Kexue Road, Zhengzhou 450001, China

A RT ICLE INFO ABSTRACT

Keywords: Wheat straw (WS) was modified by polyethylenimine (PEI) to enhance its adsorption capacity toward copper
Copper ion ions from solution through epichlorohydrin (ECH) as grafting agent and PEI-WS was the product. Adsorption
Adsorption property of PEI-WS toward Cu2+ was performed from solution in batch mode and fixed-bed column mode. The
Wheat straw results showed that adsorption quantity became larger with the rise of solution pH, initial copper concentration
Regeneration
while the co-existing ion had little effect. Adsorption quantity was to 48.6 mg/g from experiments at 293 K.
Equilibrium experimental data were better fitted by Langmuir model while kinetic processes were better fitted
by Elovich model. In column mode, effects of flow rate, bed depth and initial Cu2+ concentration on break-
through curves were studied and Yan model was suitable to fit the results. Cu-loaded adsorbent was efficiently
regenerated by 0.1 mol/L HCl solution. It is implied that PEI-WS be used to remove metal ions from aqueous
solution.

1. Introduction modification, the adsorption capacity can be enhanced (Krishnani,

2015; Lin et al., 2018; Wu et al., 2018; Xu et al., 2010; Zhou et al.,
Heavy metal ions are toxic and widely existed in environments and 2015).
can cause any harm to humans and the environment. As heavy metal A large number of wheat straw (WS) in China is generated annually
cannot be biodegradable and thermal degradation, it can long-term stay in wheat harvest-time. So it is very cheap and easily obtained. There are
in the environment, and accumulate to reach harmful levels (Fu and lignin, cellulose and hemicelluloses containing some functional groups
Wang, 2011). Copper is one of essential elements. Drinking water with which are helpful to adsorption. WS can be directly used to adsorb
copper exceeded permissible quantity for longer time could cause heavy metal ions or dye from solution (Wu et al., 2009). But adsorption
kidney or liver damage and harm of copper is concealment and irre- capacity is relatively lower. In order to improve its adsorption quantity,
versible. So treatment of wastewater containing copper ion is important some chemical agents can be used to modify the surface of WS. In our
in the field of environmental protection. study, adsorption quantity toward copper ions and methylene blue is
Adsorption technique is very important to remove the refractory significantly enhanced after citric acid modification (Han et al., 2010a).
pollutants from solution as there is high removal efficiency, adjustable Amino modification may be a promising way to enhance the ad-
working condition, good stability and selectivity (Anastopoulos et al., sorption capacity of metal ions and anionic pollutants as the amino part
2018; Singh et al., 2018; Wang and Chen, 2009; Wu et al., 2017). Ac- has a powerful coordinated functional group or easily protonized in
tivated carbon is common used with higher adsorption capacity. But the acidic conditions (Ong et al., 2007; Wang et al., 2011). Poly-
cost is higher and this limits its wide use in remediation field (Crini, ethylenimine (PEI) with better water soluble property is one polymeric
2006). So other materials from agriculture or industry as alternatives amine with a more number of eNH2 and eNH groups, which are ad-
can be used in adsorption field. Agricultural by-products with cellulose, vantage of removal of anionic pollutants or some heavy metal ions. PEI-
lignin, etc. are low-cost, abundant and easily biodegradable which can modified biomass can efficiently adsorb anionic pollutants, such as
be selected as adsorbents to bind or remove pollutants. Several reviews phenol red, Cr(VI) anions, direct dyes (Sajab et al., 2013; Deng and
have been published about natural or modified by-products as ad- Ting, 2005a; Sadaf and Bhatti, 2014). PEI can be used as coagulant to
sorbent (Bhatnagar and Sillanpaa, 2010; Wan Ngah and Hanafiah, remove nickel or copper ions from wastewater (Wang et al., 2013).
2008; Ahluwalia and Goyal, 2007; Xu et al., 2016). Through chemical Moreover, PEI-modified biomass can effectively adsorb cationic metal

⁎
Corresponding author.
E-mail address: [email protected] (R. Han).

https://doi.org/10.1016/j.biteb.2019.02.011
Received 14 December 2018; Received in revised form 11 February 2019; Accepted 15 February 2019
J. Dong, et al.

ions as there are many amino groups on surface of materials (Deng and Adsorption quantity of PEI-WS was calculated using the following
Ting, 2005b). So PEI modified WS may be an appropriate adsorbent to equation:

qe =
bind copper ions.
V (C0 Ce )
Modification of biomass can be operated in organic solvent
m (1)
(N,N‑dimethylformamide) and thionyl dichloride (SOCl2) or epi-
chlorohydrin (ECH) as bridged agent to graft amino group. The ad- where Ce and C0 (mg/L) are the equilibrium and initial concentration of
sorption capacity was significantly improved toward anionic ions (Xu copper solution.
et al., 2010; Xu et al., 2016). But the modification can be performed in Cu-loaded PEI-WS was obtained at C0 = 50 mg/L, solution pH = 6
NaOH solution and ECH is as bridged agent to graft amino group. PEI at contact time 6 h at 303 K. The regeneration was performed using
can be physically adsorbed by biomass and glutaraldehyde is used as 10 mL 0.1 mol/L HCl solution (30 min). The recovered PEI-WS was used
cross-linking agent (Deng and Ting, 2005a, 2005b; Shang et al., 2016). to adsorb as before. The cycle of desorbing and reuse was repeated
In this study, the PEI-WS was obtained by grafting method using ECH in three times. The efficiency of regeneration was calculated as following
NaOH solution. There is not organic solvent and the process is relatively expression:

= ◊ 100%
green, cleaner and sustainable. The adsorption property toward copper qn
ion was presented in batch mode and column mode and adsorption qe (2)
models were applied. The reusability of PEI-WS was also studied.
where qn was adsorption quantity of recovered adsorbent and qe was
2. Materials and methods the adsorbent quantity of fresh PEI-WS.

2.1. Materials and reagents 2.4. Column adsorption

Polyethylenimine (molecular weight 600) and epichlorohydrin The column mode was performed in fixed-bed column for con-
(ECH) were used to modify WS. All the reagents (HCl, NaOH, NaCl, tinuous adsorption. The glass column (with 1 cm in inner diameter,
Na2SO4, CuSO4) used in this experiments are analytical grade. WS was 25 cm in height) was packed with some amount of PEI-WS. Cu2+ so-
collected in a suburb of Zhengzhou City, China. The solution of Cu2+ lution with constant concentration was pumped by a peristaltic pump
(500 mg/L) was prepared by dissolved CuSO4 in distilled water. This (at various flow rates) from top of column to down mode. The effluent
solution was diluted to obtain desired concentration. from bottom column was collected at proper intervals. The experi-
mental conditions were following: (1) flow rate of solution was at 3.5,
2.2. Preparation of PEI-WS 6.5, and 9 mL/min, keeping constant of 50 mg/L Cu2+ concentration
and 6.0 cm bed depth of adsorbent (0.58 g adsorbent), respectively; (2)
Collected WS was washed with tap water to remove dirty substances initial Cu2+ concentration was at 10, 30, and 50 mg/L with flow rate
and then soaked in distilled water for one whole day, then washed with 9 mL/min and 6.0 cm bed depth of PEI-WS; (3) bed depth of adsorbent
distilled water several times and dried at 80 °C in heating oven for 15 h. was fixed at 6 cm (0.58 g), 7.5 cm (0.79 g) and 10 cm (1.01 g) with in-
WS was milled to small particles and sieved to size from 40 to 60 mesh. itial 50 mg/L Cu2+ concentration and 9 mL/min flow rate. The bed
The final WS was obtained. volumes were 4.7, 5.9, 7.8 mL for bed depth 6, 7.5 and 10 cm, re-
The PEI modified wheat straw (PEI-WS) were as follows. 9.0 g NaOH spectively. Cu-loaded PEI-WS in column was also regenerated by elu-
and 150 mL water were added in a 300 mL Erlenmeyer flask and al- tion of 0.1 mol/L HCl solution (adsorption condition: 50 mg/L Cu2+,
lowed to cool to room temperature. 15 mL ethanol (95%), 22.5 mL ECH 6.0 cm bed depth and 9 mL/min flow rate).
were added in the flask and mixed well. 1.5 g WS were added in solu- The adsorption behavior of Cu2+ onto PEI-WS in column mode is
tion and stirred in the oscillator at 303 K for 5 h. The mixtures were illustrated by breakthrough curve, and Ct/C0 (C0 and Ct are the Cu2+
filtrated and the obtained solid were putted in a 100 mL flask con- concentration in influent and effluent) versus effluent volume or time t
taining 40 mL 30 g/L PEI. Then the mixtures were stirred for 4 h at for a column height. Breakthrough time is related to Ct/C0 = 0.05 and
303 K. The solids obtained from filtration was washed several times to half time is related to Ct/C0 = 0.5.
remove excess PEI and ECH. The desired production was dried in the
oven at 80 °C and PEI-WS was obtained. Finally PEI-WS was preserved 2.5. Analytical method
in the desiccators for use.
The process of preparation can be expressed as following: Copper ions in solution are determined by flame atomic absorption

Cell–abb. of cellulose existed in surface of WS. spectrometer (FAAS) at 324.7 nm (Perkin–Elmer, AAnalyst 300). The
measurement was double repeated and the average results are shown.
2.3. Batch adsorption Solution pH was measured by potentiometry using pH glass electrode
and saturated calomel electrode.
The process of adsorption in batch mode was as following: A certain
amount of PEI-WS (16 mg) was put in 50 mL flasks containing 20 mL 3. Result and discussion
copper solution in desired concentration. The mixture was shaken at for
different time at 293 K, 303 K, 313 K with 100 rpm. Effect of solution Elemental analysis showed that there were 3.20% N of PEI-WS and
pH on Cu2+ adsorption was performed by adjusting the solution pH 0.12 N of WS. FTIR analysis (figure not presented) showed that middle
from 1.5 to 6. Effect of common salts was also evaluated. After ad- strong peak at 1730 cm−1 (stretching vibration of C]O from eCOOH,
sorption, the mixtures were filtered, and the Cu2+ concentration in eCOOR) disappeared after modification. This may be related to a large
filtrate was measured. number of the PEI molecules on the surface of WS, which screen the
J. Dong, et al.

40
50
35

30 40

25
qe/(mg/g)

30

qe/(mg/g)
20 NaCl
NaNO3
15 20 Ca( NO3) 2

10
10
5

0 0
2 3 4 5 6 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16
pH
C/(mol/L)
Fig. 1. Effect of pH on copper ion adsorption (C0 = 50 mg/L, T = 303 K). 2+
Fig. 2. Effect of common ions on Cu adsorption (C0 = 50 mg/L, pH = 6.0,
T = 303 K).
absorption of C]O existed on surface, or the peak was shifted to lower
wavenumber as form of eCOOH and eCOOR were to eCOO– after wastewater and the co-existing ions can have a strong influence on
NaOH treatment. Peak at 2919 cm−1 and peak at 1425 cm−1 became adsorption quantity.
larger, which were due to stretching vibration of CeH and bending Different concentrations of NaCl, NaNO3, Ca(NO3)2 are chosen to
vibration of CeH, respectively (Shang et al., 2016). The absorption at study the effect and results are presented in Fig. 2. It was observed from
3420 cm−1 from OeH and NeH became stronger and wider. So it was Fig. 2 that co-existing ions with various concentrations had little effect
implied that the PEI be successfully grafted onto surface of WS. on values of qe. As the adsorption major mechanism is the coordinate
action between copper ion and nitrogen atom and there is no interac-
3.1. Adsorption in batch mode tion force between the Ca2+ or Na+ and the adsorbent, the ionic
strengthen has little effect on copper adsorption. So it was inferred that
3.1.1. Effect of solution pH on Cu2+ adsorption PEI-WS can be used to remove Cu2+ in real wastewater which often
Solution pH can play important role during adsorption process as it contain some amount of common salts.
can affect form of adsorbate and property of adsorbent surface (such as If interaction between adsorbents and ionic adsorbates is major
dissociation of functional groups and surface charge). Fig. 1 illustrates ionic exchange or electrostatic attraction, there is disadvantage of ad-
the effect of solution pH on values of qe. As copper ion will be hydro- sorption with co-existed common salts, such as positive Cd2+ and Pb2+
lysis in higher solution pH, pH value is range from 1.4 to 6.5. As shown adsorption onto modified ligno-cellulosic material (Huang et al., 2015),
in Fig. 1, the adsorption capability of copper ion strongly depends on Cu2+ and Pb2+ adsorption on beer yeast (Han et al., 2006).
pH. The PEI-WS adsorption capability increased sharply from 1.27 mg/
g to 36.67 mg/g as solution pH changed from 1.4 to 6.0. The reason
may be that the groups contained nitrogen atoms were protonized 3.1.3. Adsorption isotherm of Cu2+
which lead to a decrease of lone pair electrons and ultimately lead to The equilibrium of copper ion adsorption onto values of qe is per-
reduce the adsorption capability. As protonation of amine group occurs formed at 293, 303 and 313 K, respectively. The experimental results
on the surface, the surface of PEI-WS should be positively charged at are presented in Fig. 3. It was shown from Fig. 3 that the values of qe
lower pH. Therefore, the electrostatic actions between PEI-WS and increased with the increase in Cu2+ concentration. This phenomenon
Cu2+. The increase in solution pH lead to the electrical repulsion force
being weaker and Cu2+ ions may be adsorbed on surface of PEI-WS. 60
Moreover, large number of hydrogen ion in acid solution may compete
293 K
with copper ions for the active site. So it can be referred that the me-
50 303 K
chanism of binding copper may be coordinate bond between copper ion 313 K
and lone pair electrons from the nitrogen atom. The adsorption cap-
ability remained unchanged between pH 5.0–6.0. In this study pH 6 was 40
selected in next study.
There is not in favor of positive metal ions adsorption in low pH
qe(mg/g)

value. Similar results were observed in other studies about heavy metal 30
ions adsorption by modified biomass (Han et al., 2006; Huang et al.,
2015). 20
_ _ _ Langmuir model fitted curves
At same conditions (solution pH 6.0, C0 = 50 mg/L), adsorption
Freundlcih model fitted curves
quantity of WS was only 10.0 mg/g. So adsorption quantity was sig- ...
nificantly enhanced after PEI modification. So the modification was 10
valuable. The reason is that amine groups from PEI are introduced into
surface of WS and there is high affinity of amine groups toward copper 0 10 20 30 40 50 60 70
ions.
Ce/(mg/g)
3.1.2. Effect of co-existing ion on Cu2+ adsorption
In general, there were some co-existing ions or common salts in real Fig. 3. Adsorption isotherm of copper ion on PEI-WS (t = 180 min).
J. Dong, et al.

Table 1 45
Fitted results by Langmuir model and Freundlich model.
40 293 K
Langmuir model
303 K
T/K qe(exp)/(mg/g) KL/(1/mg) qm(theo)/(mg/g) R2 SS
313 K
35

293 46.6 0.115 ± 0.038 48.3 ± 3.6 0.887 16.5

303 50.2 0.175 ± 0.052 49.4 ± 3.1 0.901 17.0 30

qt/(mg/g)
313 54.4 0.161 ± 0.028 56.6 ± 2.6 0.963 8.14
25

Freundlich model
20
T/K qe(exp)/(mg/g) KF 1/n R2 SS
_ _ _ Pseudo-second-order kinetic model fitted curves
15
293 48.6 12.7 ± 0.6 0.313 ± 0.013 0.992 1.10 . . . Elovich model fitted curves
303 52.2 13.9 ± 0.9 0.309 ± 0.019 0.983 2.90
313 56.4 15.2 ± 0.8 0.321. ± 0.015 0.988 2.47 10
0 20 40 60 80 100 120 140 160 180 200
Note: SS = ∑ (q ‐ qc)2, q and qc are the experimental value and calculated value
according the model, respectively.
t/min
2+
was probably because the increase in Cu concentration led to an Fig. 4. Kinetic process at different temperature (C0 = 50 mg/L, m = 0.008 g,
increased mass transfer of copper ion to the surface of adsorbent. It was pH = 6.5).
also found that the adsorption quantity became larger at higher tem-
perature, which indicated the adsorption process was endothermic. To further study the mechanism of copper ion adsorption on PEI-
It is helpful and useful to appropriately correlate to equilibrium WS, the pseudo-second-order kinetic model and Elovich model were
data. Two common adsorption isotherms, Langmuir model and used to fit the experimental data.
Freundlich model are adopted to fit the experimental data. The expression of pseudo-second-order kinetic model is following
Langmuir model is expressed as following equation (Song et al., (Zhang et al., 2014):
2011; Wang et al., 2018):

qe = qt =
k2 qe2 t
1 + KL ce 1 + k2 qe t
qm KL ce
(3) (5)

Freundlich model is presented as following equation (Song et al., where k2 is rate coefficient, qt(mg/g) is adsorption quantity at contact
2011): time t and qe (mg/g) is adsorption quantity at adsorption equilibrium.
qe = KF c1/ n (4) The expression of Elovich equation is following (Ho et al., 2000; Lin
et al., 2018):

qt = A + B ln t
where qm (mg/g) and KL (L/mg) are maximum adsorption capability
and a constant related to adsorption energy; KF and 1/n is Freundlich (6)
parameters related to the adsorption capability of adsorbent and the
intensity of adsorption. where A and B are both Elovich constants.
Method of nonlinear regressive analysis is used to obtain the para- Parameters of kinetic models are listed in Table 2 using nonlinear
meters of models according to less sum of difference square and the regressive analysis. It was seen that Elovich model was better to de-
relative parameters of both models, determined coefficients (R2) and scribe the adsorption kinetic behavior with higher values of R2 (be-
errors (SS) are all listed in Table 1. The fitted isotherm curves are also tween 0.927 and 0.985). Furthermore, the fitted curves from Elovich
showed in Fig. 3. equation were closer to experimental curves. The low value of R2 ob-
It was found that Freundlich model was relatively better fitting tained from the pseudo-second-order kinetic model showed that it was
degree with higher values of R2 (0.937 to 0.983) and lower values of SS. not suitable to predict the adsorption kinetic behavior. But the pre-
Moreover, the curves from models were closer to the curves from ex- dicted values of qe(theo) from pseudo-second-model were closer to ex-
periments than those from Langmuir model, respectively. So it could be periment data qe(exp), thus the model can predict the adsorption quan-
concluded that the adsorption be one heterogeneous multilayer process. tity.
The values of 1/n (0.1 < 1/n < 0.5) showed the good adsorption
capability (Aksu, 2002). The values of KF became larger at higher Table 2
temperature, which also indicated that it was favor of copper adsorp- Fitted results by Pseudo–second–order model and Elovich model.
tion at higher temperature. Pseudo–second–order kinetic model
Adsorption quantity of PEI-WS was higher than that of WS, other
agricultural by-products and commercial AC, but less than ionic ex- T/K qe(exp)/(mg/g) qe(theo)/(mg/g) k2 × 10−3 R2 SS
change resin (Han et al., 2010a). As there is little effect of common salts
293 32.5 32.7 ± 1.7 2.52 ± 0.66 0.891 6.21
on adsorption quantity and process of preparation is simple, PEI-WS is 303 36.0 36.2 ± 1.8 1.98 ± 0.50 0.905 7.03
used for removal of Cu2+ from solution. 313 39.9 41.2 ± 2.5 1.49 ± 0.43 0.888 11.4

3.1.4. The kinetic adsorption of Cu2+

Effect of contact time is also important and adsorption kinetics Elovich equation

evidently controls process at solid–solution interface. Fig. 4 shows the T/K qe(exp)/(mg/g) A B R2 SS
results of kinetic adsorption. It was found that adsorption was rapid at
initial stage and then became slow down until it nearly reached equi- 293 32.5 3.47 ± 1.30 5.47 ± 0.34 0.970 1.69
librium. It was also seen that values of qt became larger at higher 303 36.0 2.33 ± 1.45 6.25 ± 0.38 0.971 2.10
313 39.9 1.26 ± 1.19 7.23 ± 0.57 0.952 4.83
temperature.
J. Dong, et al.

Table 3
Results of column adsorption from experiments and Yan model.
Z/cm v/(mL/min) C0/(mg/L) a b R2 qe(cal)/(mg/g) qe(exp)/(mg/g) t0.05/min t0.5/min SS × 10−3

6.0 6.5 50 1.80 ± 0.20 182.4 ± 11.8 0.954 15.7 ± 2.1 28.4 10 25 4.68
7.5 6.5 50 1.56 ± 0.16 291.0 ± 20.6 0.958 18.4 ± 1.8 29.0 14 38 5.15
10 6.5 50 2.48 ± 0.30 485.0 ± 27.3 0.963 24.1 ± 1.4 29.4 37 66 5.45
6.0 3.5 50 2.05 ± 0.23 258.9 ± 15.8 0.959 22.3 ± 1.4 28.3 30 61 5.38
6.0 9.0 50 1.39 ± 0.12 153.7 ± 10.0 0.974 13.3 ± 0.9 30.3 5 15 2.60
6.0 6.5 30 1.84 ± 0.12 419.8 ± 16.8 0.984 21.7 ± 0.9 25.7 21 64 2.03
6.0 6.5 10 2.57 ± 0.21 676.4 ± 20.8 0.983 11.7 ± 0.4 13.1 45 99 1.91

3.1.5. Regeneration of Cu2+-loaded PEI-WS the increase in t0.05 and t0.5. There was broadened mass transfer zone at
Desorption is important in adsorption study as it is helpful to find higher bed depth.
possibility methods for recycling the adsorbent (Bhatti et al., 2018; There were not significantly different about values of qe(exp) except
Rong et al., 2018; Sadaf et al., 2014). The stripping agent of 10 mL C0 = 10 mg/L. The values of qe(exp) changed little with difference of bed
0.1 mol/L HCl can be desorb Cu2+ from Cu2+-loaded PEI-WS (36.2 mg depth and flow rate in experimental conditions and became lower with
Cu was loaded on per gram PEI-WS). The phenomenon could be at- the decrease in influent Cu2+ concentration.
tributed to that the H+ could destroy the complex bond between the Experimental results from Fig. 5 can be analyzed by column model.
copper ion and nitrogen. Then copper ion was released and the activate Thomas model is common used, but there is error to fit the whole
site was vacated. But there were also other interaction between Cu2+ breakthrough curves. To solve this problem, Yan model is applied as
and PEI-WS as the regeneration efficiency was not over 70%. Third- following expression (Liu et al., 2018; Yan et al., 2001):

=1
cycles showed that the regeneration efficiency slowly decreased, but
1 + (vt /b) a
Ct 1
still over 60%.
C0 (8)

3.2. Adsorption in column mode where both a and b are parameters of Yan model.
The parameters of Yan model and values of R2, SS were also listed in
3.2.1. Breakthrough curves of Cu2+ onto PEI-WS Table 3. The curves predicted by Yan model were also presented in
Adsorptive property can be further performed by column mode as Fig. 5.
this performance is continuous. The breakthrough curves of Cu2+ It was found from Table 3 that values of R2 were higher while values
binding onto PEI-WS are illustrated in Fig. 5(a,b,c), respectively. of SS were lower at all conditions. Moreover, it was observed from
The adsorption uptake (qe(exp)) is calculated as following expression: Fig. 5 that the predicted breakthrough curves were reasonably closer to
the experimental curves, respectively. These results implied that Yan
qe(exp) = =
vA v t = ttotal
model can be applied to predict copper adsorption in column mode. But
t=0
(C0 Ct ) dt
1000m 1000 (7) values of q0 (bC0/m) from Yan model was lower than those from ex-
periments at same condition. So it was not to predict the adsorption
where v, ttotal and A are volumetric flow rate (mL/min), total flow time
quantity. Similar results were also found that Yan model was better to
(min), and the area under the breakthrough curve, respectively; m is the
describe breakthrough curves in other studies (Song et al., 2011; Su
dry weight of PEI-MS (g).
et al., 2013; Yan et al., 2001; Wu et al., 2010).
Values of qe(exp), breakthrough time (t0.05) and half breakthrough
time (t0.5) (at time of Ct/C0 = 0.5) are calculated and listed in Table 3.
The shapes of all curves were “S” type. 3.2.2. Desorb Cu2+ and reuse of PEI-WS in column mode
According to information of Fig. 5(a), the breakthrough curves be- Spent adsorbent or adsorbate loaded onto PEI-WS in column mode
came flatter as the flow rate became slower. Values of t0.05 were 5, 10 can also be regenerated or desorbed in column mode and this make
and 30 min at flow rate 9.0, 6.5 and 3.5 mL/min, respectively while adsorbent more economical (Han et al., 2010b). Cu-loaded PEI-WS
values of t0.5 were 15, 25 and 61 min with the decrease in flow rate (30.3 mg Cu loaded on per gram PEI-WS) was also regenerated using
from 9.0 mL/min to 3.5 mL/min (shown in Table 3). The reason may be 0.1 mol/L solution and the desorbed curve was presented in Fig. 6(a).
that effects of diffusion were higher as the sufficient residence time of Cu2+ was fast desorbed because the process nearly completed in <
Cu2+ was relative long in the column at lower flow rate. Thus lower 10 min. There was lower concentration of peak for the second re-
flow rate was favorable to remove metal ions (Akzu and Gonen, 2004). generation. So HCl solution was effective in the elution of Cu2+ from
It was seen from Fig. 5(b) that more flatter breakthrough curve was spent adsorbate. Similar results were also found in the study of Cu2+
also found at lower initial Cu2+ concentration. Both values of t0.05 and adsorption on WS and citric acid modified WS (Li et al., 2013).
t0.5 decreased with the rise of adsorbate concentration. The driving The recycles of PEI-WS were illustrated in Fig. 6(b). These results
force of Cu2+ adsorption onto PEI-WS is due to the gradient of ad- confirmed that PEI-WS was used repeatedly with some losing their
sorbate concentration. Low concentration differences with lower adsorption capacity. So spent PEI-WS can be recovered for reuse with
driving force are not favor of adsorption. Steeper breakthrough curve some advantage.
showed short zone of mass transfer at higher initial concentrations (Su
et al., 2013). 4. Conclusion
The breakthrough curves became flatter with the increase in bed
depth. As the bed height became higher, adsorbate had relatively en- The adsorption capability of WS toward Cu2+ was enhanced after
ough time to contact with the active sited of adsorbent and there was PEI modification. There was little effect of common salts on adsorption
higher removal efficiency. Furthermore, the higher depth resulted in quantity and the main mechanism between copper ion and PEI-WS was
J. Dong, et al.

1200
1.0 a
a
1000
0.8
800
0.6

Ct/(mg/L)
first
3.5 mL/min 600 second
Ct/C0

0.4 6.5 mL/min

9.0 mL/min 400

0.2
_ _ _ Yan model fitted curves 200

0.0
0

0 50 100 150 200 0 20 40 60

t/min t/min

1.0 1.0
b
b
0.8
0.8

0.6
0.6
Ct/C0

0.4 initial
Ct/C0

0.4 first
10 mg/L second
30 mg/L 0.2
0.2
50 mg/L
0.0
_ _ _ Yan model fitted curves
0.0
0 50 100 150 200
0 50 100 150 200
t/min
t/min
Fig. 6. Desorbing curves using 0.1 mol/L (a) and recycles for Cu2+ adsorption
(b).

1.0
c shape of breakthrough curves and Yan model can predict the process.
PEI-WS is good adsorbent to remove copper ions from solution.
0.8
Conflict of interest
0.6
There is no conflict of interest to declare.
Ct/C0

6.0 cm
0.4 7.5 cm
10.0 cm Acknowledgements

0.2 This work was supported by the National Natural Science

Foundation of China for undergraduate cultivation in basic science
_ _ _ Yan model fitted curves
(J1210060).
0.0

References
0 50 100 150 200

Ahluwalia, S.S., Goyal, D., 2007. Microbial and plant derived biomass for removal of
t/min heavy metals from wastewater. Bioresour. Technol. 98, 2243–2257.
Aksu, Z., 2002. Determination of the equilibrium, kinetic and thermodynamic parameters
Fig. 5. Breakthrough curves of Cu2+ adsorption at different conditions. (a. of the batch biosorption of nickel(II) ions onto Chlorella vulgaris. Process Biochem. 38,
various flow rate; b. various Cu2+ concentration; b. various bed depth). 89–99.
Akzu, Z., Gonen, F., 2004. Biosorption of phenol by immobilized activated sludge in a
continuous packed bed: prediction of breakthrough curves. Process Biochem. 39,
599–613.
complexation. Freundlich model and Elovich model were better to Anastopoulos, I., Mittal, A., Usman, M., Mittal, J., Yu, G.H., Núñez-Delgado, A., Kornaros,
M., 2018. A review on halloysite-based adsorbents to remove pollutants in water and
predict equilibrium adsorption and kinetic behavior, respectively. The wastewater. J. Mol. Liq. 269, 855–868.
Cu2+ loaded or spent PEI-WS was regenerated by 0.1 mol/L HCl solu- Bhatnagar, A., Sillanpaa, M., 2010. Utilization of agro-industrial and municipal waste
tion. Flow rate, bed depth, initial Cu2+ concentration can influence the materials as potential adsorbents for water treatment-a review. Chem. Eng. J. 157,
J. Dong, et al.

277–296. Shang, Y., Zhang, J.H., Wang, X., Zhang, R.D., Xiao, W., Zhang, S.S., Han, R.P., 2016. Use
Bhatti, H.N., Hayat, J., Iqbal, M., Noreena, S., Nawaz, S., 2018. Biocomposite application of polyethylenemine modified wheat straw for adsorption of Congo red from solution
for the phosphate ions removal in aqueous medium. J. Mater. Res. Technol. 7, in batch mode. Desalin. Water Treat. 57, 8872–8883.
300–307. Singh, N.B., Nagpal, G., Agrawal, S., Rachna, 2018. Water purification by using ad-
Crini, G., 2006. Non-conventional low-cost adsorbents for dye removal: a review. sorbents: a review. Environ. Technol. Innov. 11, 187–240.
Bioresour. Technol. 97, 1061–1085. Song, J.Y., Zou, W.H., Bian, Y.Y., Su, F.Y., Han, R.P., 2011. Adsorption characteristics of
Deng, S.B., Ting, Y.P., 2005a. Polyethylenimine-modified fungal biomass as a high-ca- methylene blue by peanut husk in batch and column modes. Desalination 265,
pacity biosorbent for Cr(VI) anions: sorption capacity and uptake mechanisms. 119–125.
Environ. Sci. Technol. 39, 8490–8496. Su, Y.Y., Zhao, B.L., Xiao, W., Han, R.P., 2013. Adsorption behavior of light green anionic
Deng, S.B., Ting, Y.P., 2005b. Characterization of PEI-modified biomass and biosorption dye using cationic surfactant modified wheat straw in batch and column mode.
of Cu(II), Pb(II) and Ni(II). Water Res. 39, 2167–2177. Environ. Sci. Pollut. Res. 220, 5558–5568.
Fu, F.L., Wang, Q., 2011. Removal of heavy metal ions from wastewaters: a review. J. Wan Ngah, W.S., Hanafiah, M.A.K.M., 2008. Removal of heavy metal ions from waste-
Environ. Manag. 92, 407–418. water by chemically modified plant wastes as adsorbents: a review. Bioresour.
Han, R.P., Li, H.Q., Li, Y.H., Zhang, J.H., Xiao, H.J., Shi, J., 2006. Biosorption of copper Technol. 99, 3935–3948.
and lead ions by waste beer yeast. J. Hazard. Mater. 137, 1569–1576. Wang, J.L., Chen, C., 2009. Biosorbents for heavy metals removal and their future.
Han, R.P., Zhang, L.J., Song, C., Zhang, M.M., Zhu, H.M., Zhang, L.J., 2010a. Biotechnol. Adv. 27, 195–226.
Characterization of modified wheat straw, kinetic and equilibrium study about Wang, Z.W., Han, P., Jiao, Y.B., He, X.T., Dou, C.C., Han, R.P., 2011. Adsorption of Congo
copper ion and methylene blue adsorption in batch mode. Carbohydr. Polym. 79, red using ethylenediamine modified wheat straw. Desalin. Water Treat. 30, 195–206.
1140–1149. Wang, G., Chang, Q., Han, X.T., Zhang, M.Y., 2013. Removal of Cr(VI) from aqueous
Han, R.P., Wang, Y., Sun, Q., Wang, L.L., Song, J.Y., He, X.T., Dou, C.C., 2010b. Malachite solution by flocculant with the capacity of reduction and chelation. J. Hazard. Mater.
green adsorption onto natural zeolite and reuse by microwave irradiation. J. Hazard. 248-249, 115–121.
Mater. 175, 1056–1061. Wang, W.X., Huang, G.H., An, C.J., Zha, S., Che, X.J., Zhang, P., 2018. Adsorption of
Ho, Y.S., Ng, J.C.Y., McKay, G., 2000. Kinetics of pollutant sorption by biosorbents: re- anionic azo dyes from aqueous solution on cationic gemini surfactant-modified flax
view. Sep. Purif. Methods 29, 189–232. shives: Synchrotron infrared, optimization and modeling studies. J. Clean. Prod. 172,
Huang, Y.Q., Yang, C.P., Sun, Z.C., Zeng, G.M., He, H.J., 2015. Removal of cadmium and 1986–1997.
lead from aqueous solutions using nitrilotriacetic acid anhydride modified ligno- Wu, Y.J., Zhang, L.J., Gao, C.L., Ma, J.Y., Ma, X.H., Han, R.P., 2009. Adsorption of copper
cellulosic material. RSC Adv. 5 11475–1148. ions and methylene blue in single and binary system on wheat straw. J. Chem. Eng.
Krishnani, K.K., 2015. Lignocellulosic wheat straw-derived ion-exchange adsorbent for Data 54, 3229–3234.
heavy metals removal. Appl. Biochem. Biotechnol. 178, 670–686. Wu, X.L., Wang, Y., Liu, J.L., Ma, J.Y., Han, R.P., 2010. Study of Malachite Green ad-
Li, Y.K., Zhao, B.L., Zhang, L.J., Han, R.P., 2013. Biosorption of copper ion by natural and sorption onto natural zeolite in fixed-bed column. Desalin. Water Treat. 20, 228–233.
modified wheat straw in fixed-bed column. Desalin. Water Treat. 51, 5735–5745. Wu, S.H., He, H.J., Inthapanya, X., Yang, C.P., Lu, L., Zeng, G.M., Han, Z.F., 2017. Role of
Lin, C., Luo, W.J., Luo, T.T., Zhou, Q., Li, H.F., Jing, L.R., 2018. A study on adsorption of biochar on composting of organic wastes and remediation of contaminated soils-a
Cr (VI) by modified rice straw: Characteristics, performances and mechanism. J. review. Environ. Sci. Pollut. Res. 24, 16560–16577.
Clean. Prod. 196, 626–634. Wu, M.J., Liu, H.Y., Guo, J.Y., Yang, C.P., 2018. Enhanced enzymatic hydrolysis of wheat
Liu, M.Y., Li, X.Y., Du, Y.Y., Han, R.P., 2018. Adsorption of methyl blue from solution straw by two-step pretreatment combining alkalization and adsorption. Appl.
using walnut shell and reuse in a secondary adsorption for Congo red. Bioresour. Microbiol. Biotechnol. 102, 9831–9842.
Technol. Res. https://doi.org/10.1016/j.biteb.2018.11.006. Xu, X., Gao, B.Y., Yue, Q.Y., Zhong, Q.Q., 2010. Preparation and utilization of wheat
Ong, S.T., Lee, C.K., Zainal, Z., 2007. Removal of basic and reactive dyes using ethyle- straw bearing amine groups for the sorption of acid and reactive dyes from aqueous
nediamine modified rice hull. Bioresour. Technol. 98, 2792–2799. solutions. J. Hazard. Mater. 182, 1–9.
Rong, Y.C., Li, H., Xiao, L.H., Wang, Q., Hu, Y.Y., Zhang, S.S., Han, R.P., 2018. Adsorption Xu, X., Gao, B.Y., Jin, B., Yue, Q.Y., 2016. Removal of anionic pollutants from liquids by
of malachite green dye from solution by magnetic activated carbon in batch mode. biomass materials: a review. J. Mol. Liq. 215, 565–595.
Desalin. Water Treat. 106, 273–284. Yan, G., Viraraghavan, T., Chen, M., 2001. A new model for heavy metal removal in a
Sadaf, S., Bhatti, H.N., 2014. Batch and fixed bed column studies for the removal of biosorption column. Adsorpt. Sci. Technol. 19, 25–43.
indosol yellow BG dye by peanut husk. J. Taiwan Inst. Chem. Eng. 45, 541–553. Zhang, R.D., Zhang, J.H., Zhang, X.N., Dou, C.C., Han, R.P., 2014. Adsorption of Congo
Sadaf, S., Bhatti, H.N., Arif, M., Amin, M., Nazar, F., 2014. Adsorptive removal of direct red from aqueous solutions using cationic surfactant modified wheat straw in batch
dyes by PEI-treated peanut husk biomass: Box–Behnken experimental design. Chem. mode: kinetic and equilibrium study. J. Taiwan Inst. Chem. Eng. 45, 2578–2583.
Ecol. 30, 52–65. Zhou, T., Lu, W.Z., Liu, L.F., Zhu, H.M., Jiao, Y.B., Zhang, S.S., Han, R.P., 2015. Effective
Sajab, M.S., Chia, C.H., Zakaria, S., Khiew, P.S., 2013. Cationic and anionic modifications adsorption of light green anionic dye from solution by CPB modified peanut in
of oil palm empty fruit bunch fibers for the removal of dyes from aqueous solutions. column mode. J. Mol. Liq. 211, 909–914.
Bioresour. Technol. 128, 571–577.