Professional Documents
Culture Documents
Chitosan
Chitosan
Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol
National Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China
Key Laboratory of Green Process and Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China
University of Chinese Academy of Sciences, Beijing 100049, China
a r t i c l e
i n f o
Article history:
Received 22 September 2015
Received in revised form
27 November 2015
Accepted 24 December 2015
Available online 29 December 2015
Keywords:
Electrospun chitosan nanober membrane
Chemical ltration
Chromium (VI) adsorption
Breakthrough curves
Model tting
a b s t r a c t
Chitosan nanobers (average diameter of 75 nm) were electrospun on polyester (PET) scrim to form
composite nanober membranes with controlled pore size. The membranes were then stacked as a membrane bed for chemical ltration of Cr (VI) of 15 mg/L. The performance of the bed with respect to loading
capacity at breakthrough, bed saturation and utilization efciency were carefully investigated. The results
showed that while these three parameters were dependent on pH, ow rate, ow distribution and packed
pattern of the membrane, the latter two were less affected by feed Cr (VI) concentration and bed length.
The maximum bed loading capacity for 1 mg/L Cr (VI) ltration at breakthrough was found to be 16.5 mgchromium/g-chitosan, higher than the static adsorption capacity of 11.0 mg-chromium/g-chitosan using
nanober mats, indicating the membranes better potential for dynamic adsorption. The minimum bed
length required to avoid breakthrough was determined to be three layers of stacked membranes with
nanober deposition density of 1 g/m2 by applying bed depth service time (BDST) model.
2015 Elsevier Ltd. All rights reserved.
1. Introduction
Electrospinning is a simple and versatile method for preparing
nanobers in the form of either a loose web, or a dense membrane, depending on the bers deposition density (Ahmed, Lalia,
& Hashaikeh, 2015; Bhardwaj & Kundu, 2010; Doshi & Reneker,
1995; Reneker & Yarin, 2008; Rutledge & Fridrikh, 2007; Teo &
Ramakrishna, 2006). Electrospun nanobrous membranes exhibit
extraordinary properties, such as small ber diameter, large surface area and high porosity, as well as easy pore size tailorability.
These properties make such membranes more attractive than common nonwovens or even casted membranes for a wide variety of
mechanical ltration and chemical ltrationa ltration process
involving interface reaction between the porous material surface
and the target solute (Aliabadi, Irani, Ismaeili, Piri, & Parnian, 2013;
Deng et al., 2011; Li, Li, Cao, & Yang, 2015; Ma, Hsiao, & Chu,
2013; Niu et al., 2013). Adsorptive ltration is one of the examples. Adsorptive electrospun nanober membranes have drawn
Corresponding author at: National Key Laboratory of Biochemical Engineering, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190,
China.
E-mail addresses: lileilyq@outlook.com (L. Li), yxli@ipe.ac.cn (Y. Li),
cfyang@ipe.ac.cn (C. Yang).
http://dx.doi.org/10.1016/j.carbpol.2015.12.067
0144-8617/ 2015 Elsevier Ltd. All rights reserved.
300
of 25.5 g/m2 was obtained from Fiberweb, Inc. All other chemicals
were analytical grade and used without purication.
2.2. Electrospinning and cross-linking process
An electrospinning solution containing 5 wt% chitosan was prepared by dissolving chitosan in 90 wt% acetic acid solution (Li, Li,
et al., 2015). The apparatus used for electrospinning was purchased
from Ucalery Co., China. The solution was ejected from the syringes
using a feed rate of 0.1 mm/min, an applied voltage of 23 kV, and
a tip-to-collector distance of 6.8 cm. The bers 75 nm in diameter were collected on the spunbond PET substrate continuously.
Composite membranes with different nanober basis weight were
obtained by controlling the electrospinning time. After electrospinning, the composite membranes were placed in a vacuum oven
at 60 C for 24 h to remove the remaining solvent. They were then
transferred to a desiccator containing 20 mL glutaraldehyde liquid
for cross-linking by glutaraldehyde vapor at room temperature for
12 h. The cross-linked nanober membranes were dried in a vacuum oven at 60 C for 12 h to remove the remaining cross-linker.
2.3. Dynamic adsorption experiments
The set-up for ltration experiments is shown in Fig. S1 in Supplementary Material le. It is composed of a peristaltic pump, a
pressure gauge and a cylindrical membrane holder. The effective
diameter of the holder is 34.7 mm. Membrane discs 47 mm in diameter were layered together to form a stacked membrane bed as
shown in Fig. 1. The bed was then placed in the holder and sealed.
Water containing Cr (VI) was pumped through the membrane just
once for continuous adsorption. The feed concentration of Cr (VI)
was controlled at 15 mg/L. pH of the solutions was adjusted using
diluted H2 SO4 and NaOH. Flow rate was 13 ml/min. Efuent samples were collected at different times for Cr (VI) concentration
measurement.
The amount of Cr (VI) adsorbed on the membrane was
determined by comparing the difference of feed and efuent concentrations of Cr (VI). The ratio of the efuent concentration and
feed concentration was plotted against time to obtain the breakthrough curves. The adsorption capacity of the membrane was
determined using the following equation:
t
q=
(C0 Ct )Qdt
(1)
2. Experimental
qtb
=
qts
tb
0ts
0
(C0 Ct )Qdt
(C0 Ct )Qdt
100%
(2)
where qtb and qts are the amounts of adsorbed Cr (VI) onto unit mass
of the membrane at breakthrough time and saturation time (mg/g).
For an ideal membrane bed, the bed utilization efciency should be
close to 100%, and the breakthrough curve is typically a step function. The non-ideal situation always gives a breakthrough curve
with a prole of changing slope, which is commonly encountered
in practice. Nonetheless, higher bed efciency is always desired
C
= exp[kAB C0 t (kAB N0 Z/v)]
C0
(3)
Doseresponse model :
C
=1
C0
(4)
1+
C0 V
q0 m
a
301
Fig. 1. Illustration of the stacked membranes and the individual part structure of the composite membrane.
302
Fig. 2. SEM images and corresponding pore size of the composite membranes with different basis weight of chitosan nanobers (a, substrate; b, 0.18 g/m2 ; c, 0.5 g/m2 ; d,
1.0 g/m2 ; e, 1.5 g/m2 ; f, 2.0 g/m2 ).
303
Fig. 3. Effect of feed concentration on dynamic adsorption [(a) breakthrough as a function of time at different feed concentration, solid lines are prediction using doseresponse
model, inset shows the experimental data and prediction lines using AdamsBohart model for initial breakthrough where C/C0 < 0.15, (b) breakthrough as a function of mass
of Cr (VI) fed to the bed, (c) adsorption capacity as a function of Cr (VI) feed concentration and (d) bed utilization as a function of feed concentration. Experiment conditions:
pH = 6, 15 layers of 1 g/m2 chitosan nanober membrane, ow rate = 2 ml/min].
Fig. 4. Static adsorption kinetics using cross-linked chitosan nanober mats [experimental conditions: pH = 6, nanober dosage is 11, 21 and 44 mg/L for 1, 2 and 5 mg/L
Cr (VI) respectively].
304
Fig. 5. Effect of ow rate on dynamic adsorption and the predicted breakthrough curves using doseresponse model [experiment conditions: pH = 6, C0 = 1 mg/L, 10 layers
of 1 g/m2 chitosan nanober membrane].
N0 Z
1
ln
C0 v
kB C0
C
Cb
(5)
v
kB N0
ln
C
Cb
(6)
305
Fig. 6. Effect of bed length on dynamic adsorption (ac) and BDST model plot (d) [experiment conditions: pH = 6, C0 = 1 mg/L, ow rate = 2 ml/min].
the value of N0 , which indicates that the bed length has a minor
effect on column adsorption capacity. kB and Z0 indicate the capability of the bed to resist breakthroughlarge kB and small Z0 imply
better bed performance. By calculation, the value of the critical bed
length was determined to be 0.057 cm, equivalent to a bed length
of a 3-layer membrane. The relatively large value of kB and small
value of Z0 of the present system are strong indication that a short
bed length will be sufcient to avoid breakthrough (Elwakeel et al.,
2014; Li et al., 2016; Vijayaraghavan et al., 2005). This further indicates that nanober membrane bed is highly efcient for removing
Cr (VI) ions.
The AdamsBohart and doseresponse models again present
a good interpretation of Cr (VI) sorption results. The trend of the
values of kAB , N0 and q0 is similar to those studied at other experimental conditions as shown in Table S1.
than the other three beds, which leads to a net increase of 237%,
100% and 28% bed volumes at 10% breakthrough compared with
the other three beds respectively. It translates to a loading capacity
of 16.5 mg-chromium/g-chitosan. At 80% saturation, the loading is
44.9 mg-chromium/g-chitosan. The bed utilization is also increased
from 23% to 37%. The trend is better shown in Fig. 7b, where the
nanober deposition density is expressed as the corresponding
pore size, against which the bed adsorption capacity and bed
utilization efciency are plotted. Smaller pore size gives rise to
high capacity and higher bed utilization. This result has practical
implications for more efcient brous membrane design using
electrospinning technology, simply because it offers the exibility
of depositing controllable amounts of functional nanobers to
maximize their utilization as adsorptive membranes.
306
Fig. 7. Effect of ber deposition density (a) and corresponding pore size (b) on dynamic adsorption of Cr (VI) [experiment conditions: pH = 6, C0 = 1 mg/L, ow rate = 2 ml/min].
Fig. 8. Effect of ow distribution on breakthrough curves [experiment conditions: pH = 6, C0 = 1 mg/L, ow rate = 2 ml/min].
seen from the table, even though the number of the entire bed volumes at breakthrough for chitosan nanober/PET membrane is less
than that of Thiol-eggshell and PGMAFe membranes because of the
presence of thick PET substrate that does not adsorb, the number of
adsorptive chitosan nanober volumes is much larger. In addition,
the electrospun chitosan nanober membrane has an advantage
over the other membranes in effective area and bed length, which
is of signicance for practical applications.
4. Conclusions
Chitosan nanober membranes with different ber deposition
density on polyester scrim were fabricated by electrospinning.
Nanober membranes used as a packed bed exhibited fairly good
dynamic adsorption capacity for Cr (VI) ions. While the performance of the nanober membranes including adsorption capacity
and bed utilization efciency was pH, ow rate and membrane
packing pattern dependent, it was less sensitive to feed concentration and bed length change. However, bed adsorption capacity
and utilization can be greatly improved by increasing nanober
deposition density of the composite membrane. The best loading and bed utilization could be established by using 10 layers
of 2 g/m2 nanober membranes to treat 1 mg/L Cr (VI) at a ow
rate of 2 ml/min. The maximum adsorption capacity obtained for
1 mg/L Cr (VI) using the nanober membranes at 10% breakthrough
is 16.5 mg-chromium/g-chitosan, exceeding the saturation capacity of static adsorption of 11.0 mg-chromium/g-chitosan using
chitosan nanober mats. The result indicated that the nanober
membranes designed in this work were effective for chemical ltration of Cr (VI) in a single-pass ow as adsorptive membranes,
a promising outcome for applying such membranes for water
decontamination. The bed breakthrough data obtained at different feed Cr (VI) concentration, ow rate and bed length were well
307
Hardick, O., Dods, S., Stevens, B., & Bracewell, D. G. (2013). Nanober adsorbents
for high productivity downstream processing. Biotechnology and
Bioengineering, 110(4), 11191128.
Ho, Y. S., & McKay, G. (2000). The kinetics of sorption of divalent metal ions onto
sphagnum moss peat. Water Research, 34(3), 735742.
Horzum, N., Boyaci, E., Eroglu, A. E., Shahwan, T., & Demir, M. M. (2010). Sorption
efciency of chitosan nanobers toward metal ions at low concentrations.
Biomacromolecules, 11(12), 33013308.
Li, H., Wang, X., Cao, L., Zhang, X., & Yang, C. (2015). Gold-recovery PVDF
membrane functionalized with thiosemicarbazide. Chemical Engineering
Journal, 280, 399408.
Li, L., Li, Y. X., Cao, L. X., & Yang, C. F. (2015). Enhanced chromium (VI) adsorption
using nanosized chitosan bers tailored by electrospinning. Carbohydrate
Polymers, 125, 206213.
Li, Y., Li, L., Cao, L., & Yang, C. (2016). Promoting dynamic adsorption of Pb2+ in a
single pass ow using brous nano-TiO2 /cellulose membranes. Chemical
Engineering Journal, 283, 11451153.
Ma, H. Y., Hsiao, B. S., & Chu, B. (2013). Electrospun nanobrous membrane for
heavy metal ion adsorption. Current Organic Chemistry, 17(13), 13611370.
Mahadevaiah, N., Venkataramani, B., & Prakash, B. S. J. (2008). Interaction of
chromate on surfactant modied montmorillonite: Breakthrough curve study
in xed bed columns. Industrial & Engineering Chemistry Research, 47(6),
17551759.
Niu, J. F., Xu, J. J., Dai, Y. R., Xu, J. L., Guo, H. Y., Sun, K., et al. (2013). Immobilization
of horseradish peroxidase by electrospun brous membranes for adsorption
and degradation of pentachlorophenol in water. Journal of Hazardous Materials,
246, 119125.
Reneker, D. H., & Yarin, A. L. (2008). Electrospinning jets and polymer nanobers.
Polymer, 49(10), 23872425.
Roper, D. K., & Lightfoot, E. N. (1995). Separation of biomolecules using adsorptive
membranes. Journal of Chromatography A, 702(12), 326.
Rutledge, G. C., & Fridrikh, S. V. (2007). Formation of bers by electrospinning.
Advanced Drug Delivery Reviews, 59(14), 13841391.
Sarkar, K., Gomez, C., Zambrano, S., Ramirez, M., de Hoyos, E., Vasquez, H., et al.
(2010). Electrospinning to forcespinning (TM). Materials Today, 13(11), 1214.
Senthilkumar, R., Vijayaraghavan, K., Thilakavathi, M., Iyer, P. V. R., & Velan, M.
(2006). Seaweeds for the remediation of wastewaters contaminated with
zinc(II) ions. Journal of Hazardous Materials, 136(3), 791799.
Sridhar, P. (1996). Design of afnity membrane bioseparations. Chemical
Engineering & Technology, 19(5), 398404.
Suen, S. Y., & Etzel, M. R. (1992). A mathematical analysis of afnity membrane
bioseparations. Chemical Engineering Science, 47(6), 13551364.
Teo, W. E., & Ramakrishna, S. (2006). A review on electrospinning design and
nanobre assemblies. Nanotechnology, 17(14), R89R106.
Vijayaraghavan, K., Jegan, J., Palanivelu, K., & Velan, M. (2005). Batch and column
removal of copper from aqueous solution using a brown marine alga Turbinaria
ornata. Chemical Engineering Journal, 106(2), 177184.
Yan, G. Y., Viraraghavan, T., & Chen, M. (2001). A new model for heavy metal
removal in a biosorption column. Adsorption Science & Technology, 19(1),
2543.
Yang, C. F. (2012). Aerosol ltration application using brous media An industrial
perspective. Chinese Journal of Chemical Engineering, 20(1), 19.
Yoon, K., Hsiao, B. S., & Chu, B. (2008). Functional nanobers for environmental
applications. Journal of Materials Chemistry, 18(44), 5326.