Pervaporation dehydration of isopropanol-water mixtures using chitosan

zeolite-a membranes

Mohd. Ghazali Hj. Mohd. Nawawi1, Le T. Ngoc Tram2

Department of Chemical Engineering,

Faculty of Chemical & Natural Resources Engineering
Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia
1
Tel: +60 7 5535593, Fax: +60 7 5581463, E-mail: [email protected]
2
Tel: +84 71 834267, Fax: +84 71 831151, E-mail: [email protected]

Abstract

Pervaporation (PV) dehydration of isopropanol (IPA) using modified chitosan

membranes was studied. The membranes were prepared from chitosan and were modified
by using zeolite-A. Pre-weighed amount of zeolite-A was added into acidic solvent and
stirred to produce homogeneous solution. Chitosan flakes were then added into the
solution and stirred overnight. Various ratios of zeolite-A and chitosan from 1:20 to 1:2
were used to produce the modified chitosan membranes. The chitosan-zeolite A solution
was casted on glass plate and dried at room temperature. The dried membranes were
treated with alkaline solution and were thoroughly washed in deionized water. The
hydrophilicity of the membranes was studied through swelling test. The test was carried
out in a 90 wt% IPA-water mixture. The result showed that the degree of swelling
decreased with the increase of the amount of zeolite-A. The mechanical properties of
membranes were also tested for tensile strength and elongation at break. Then the
membranes were investigated for the PV separation of IPA-water mixtures at the
permeate pressure of 720 mmHg under vacuum. The feed concentration was varied from
0 to 95 wt% IPA and the feed temperature was varied from 30 to 70°C. The results
showed that the ratio 1:8 of zeolite-A and chitosan produced the best combination to
modify the membrane for the separation of water-IPA mixtures. The structural
morphologies of the chitosan filled zeolite-A membranes with ratio 1:8 and 1:2 wt
zeolite-A/wt chitosan was studied under Scanning Electron Microscopy (SEM). The
results showed that the membranes were dense, no pores were visible. The addition of the
zeolite did not alter the structure of the membranes.

Keywords: Pervaporation, dehydration, membrane, chitosan, zeolite A, isopropanol

________________________________________________________________________
1. Introduction

Membranes have been used in separation technique for a long time. Almost all
membranes are prepared from synthetic materials. Natural polymers are not so many like
synthetic but more biocompatable and biodegradable. Besides cellulose and its
derivatives, the next most common natural polymer is chitin and chitosan. Chitosan have
been used not only in pharmaceutics, ophthalmology, cosmeitcs, agriculture, food
processing [1-5] but also in membrane separation. Chitosan is insoluble in water but
chitosan membrane shows high hydrophilicity. Many researchs investigated separation
alcohol-water mixtures using pervaporation system through chitosan membranes and
modified chitosan membranes by blending, crosslinking or complex [6-12].

Recently zeolite filled membranes were studied. Yong et al. [13] studied zeolite
filled polyimide membrane containing 2,4,6-triaminopyrimide for gas separation. Boom
et al. [14] studied rubbery polymer EPDM, Viton and Estane with zeolite X and Fajitsu.
The addition of hydrophilic zeolite X and hydrophobic zeolite silicate led to an increase
in methanol flux and a decrease in toluene flux through membranes. Gao and Li [15] used
composite hydrophilic membranes polyvinyl alcohol PVA with zeolite KA, NaA, CaA,
NaX. With KA zeolites, the flux rised with increasing zeolite content and the separation
factor is maintained upon addition of zeolite until 11.1% and then decreased. The flux
increased with decreasing zeolite particle size and the separation factor unchanged at low
zeolite content and decreased at high zeolite content. The addition zeolites to polymer
improved the flux of membranes. With larger alcohol molecules like IPA, the separation
factors of the majority of A-type zeolite filled membranes were improved markedly.
Nawawi et al. [16] used chitosan with zeolite 13X. The total permeation flux improved
but separation factor decreased compared to the homogeneous chitosan membranes.
Pervaporation Separation Index gave the highest value at 0.1 wt.% zeolite. In this
research, chitosan filled zeolite A membranes were studied to determine the effect of
zeolite A concentration to dehydrate IPA.

2. Materials and experiments

2.1.Materials
Chitosan powders were supplied by Chitochem company (Malaysia) with the
average molecular weight MW= 50,000-100,000. Molecular sieves, 4Å with the average
particle size ~5 µm and the pore diameter approximately 4Å, was purchased from Sigma-
Aldrich (M). Acetic acid and sodium hydroxide were supplied from Merck (Germany);
ethanol from Fluka (Switzerland) and isopropanol from J.T.Baker (USA). Deionized
water from the laboratory was also used in this study.

2.2. Membranes preparation

Zeolite A was added in acetic acid 10 wt.% to produce homogeneous solution.

The concentration of zeolite-A in the solution was varied from 0.1 wt.% to 1 wt.% of the
solvent (equivalent the ratio of zeolite A:chitosan =1:20 to 1:2). Pre-weighed chitosan
was stirred in this solution overnight. The casting solution consisted of 2 wt.% chitosan.
After the impurities and undissolved chitosan were removed, the solution was casted onto
glass plates and evaporated at room temperature. The membranes were peeled off from
the plates. The membranes were treated with the alkaline solution. Then membranes were
washed with deionized distilled water to remove trace of alkaline completely and dried at
room temperature.

2.3. Membrane characterization

All types of chitosan membranes were tested for their tensile strength, elongation
and Young’s modulus at the dry states by using INSTRON 5567.

The hydrophilicity of membranes was studied through the swelling test. The
weight of the dry chitosan membranes was measured. The membranes were then
immersed in 90 wt.% IPA at room temperature for 3 days. The membranes were picked
out and weighed after the surface liquid was quickly removed with tissue papers. Then
the membranes were put in the sample tube to desorb the liquid absorbed using the
vacuum pump. The desorption apparatus was shown in Figure 1.

Control Valve
Vacuum pump

Cold Trap
Sample tube

Swollen membranes

Figure 1: The Desorption Apparatus To Determine The Composition Of The Mixture Absorbed
In Swollen Membranes

The liquid sorbed from the membrane was condensed in the cold trap with liquid
nitrogen. The liquid collected in the cold trap was analyzed to determine its composition.
The degree of swelling is calculated from:

The weight of swollen membrane Ws − Wd

DS (%) = = x100 (1)
The weight of dry membrane Wd

Ws : the weight of swollen membrane

Wd : the weight of dry membrane
2.4. Pervaporation

The membrane performance was investigated through the effect of zeolite

concentration and the feed temperature to dehydrate IPA solution. The effect of zeolite
concentration was carried out from 0.1 to 1 wt.% in the solvent (or the ratio zeolite
A:chitosan = 1:20 to 1:2). The feed concentration was 90 wt.% IPA. The membrane
thickness was 25 µm. The permeate pressure was kept at 720 mmHg under vacuum. The
membrane effective area was 52.81 cm2. The feed temperature was studied from room
temperature to 70°C. The pervaporation apparatus has been described elsewhere [16].
The permeation flux was calculated from the amount of permeate collected in a given
period time:
Q
J= (2)
A
where Q: the weight of permeate in a given period (g/h)
A: the effective membrane area (m2)

The permeation composition was determined by using Refractormeter. The

permeation flux of each component was defined from the total permeate and the
permeation composition. The separation factor α was calculated from:
Y /Y
α = W IPA (3)
X W /X IPA

where Y, X : the weight fraction of water and IPA in the permeate side and
weight fraction of water and IPA in the feed mixture

And the pervaporation separation index PSI was calculated from:

PSI = J.(α-1) (4)

3. Results and discussion

3.1. The membrane characterization

Mechanical testing gives a good indication of the membrane physical properties.

Figure 2 shows the tensile strength, elongation and Young’s modulus of all types of
membranes. As the increase of zeolite A, the tensile strength and Young’s modulus also
increase but the percentage of elongation declines. It is suggested that the interaction
between chitosan and zeolite molecules is stronger than the interaction among the
chitosan molecules. Therefore the higher the zeolite added in the membrane, the stronger
the modified membranes resist the deformation. The polymer chain of chitosan has
plasticity while zeolite A has the crystalline structure. The addition of zeolite A made
chitosan membranes break their interactions among the chitosan molecules, which
reduced the plasticity. So the increase of zeolite content made the chitosan membranes
more brittle. In general, the more the zeolite content in the chitosan membranes, the
stronger but more brittle the membranes.
 80 4400

70
4000
60
Tensile strength (MPa

Modulus (MPa)
Tensile strength 3600
Elongation (%)

50

40 3200

30 Young's Modulus 2800

20 Elongation
2400
10

0 2000
HOMO 0.10% 0.25% 0.50% 0.75% 1.00%

Zeolite A content (wt.%)

Figure 2: The Tensile Strength, Elongation and Young’s Modulus

3.2.The hydrophilicity of membranes

The degree of swelling and the sorption selectivity was shown in Figure 3. The
degree of swelling decreases with the increase of the zeolite. When the zeolite
concentration increases, the chitosan content will decrease. The chitosan membranes have
the free volume segments while zeolite adsorbed through its pore size. As suggested
above, the interactions between the the chitosan and zeolite molecules are tight and the
chitosan content in the membranes decreases, thus leaving less free volume segments
than the unmodified chitosan. Therefore the modified membranes are less swollen. The
sorption selectivity started to diminish at zeolite content 0.25 wt.%. At the low content of
zeolite, the IPA molecule saturated with the IPA and water molecules in the pores. So the
sorption selectivity in the chitosan zeolite membranes is the same as in the unmodified
membranes. The sorption selectivity decreases significantly at 0.5 wt.% concentration of
zeolite. When the zeolite content increased, both water and IPA sorbed much more in the
zeolite pores. The addition of zeolite made the membranes less hydrophilic and selective.
 12 6000

11
5000
10
Degree of swelling (%

Sorption selectivity
4000
9

8 3000

7
2000
6
1000
5

4 0
0 0.1 0.25 0.5 0.75 1

Zeolite A content (wt.%)

Figure 3: The Degree of Swelling and Sorption Selectivity

3.3. The effect of zeolite A concentration

The PV separation of isopropanol-water was carried out at 90 wt.% IPA in the

feed. The total permeation flux and separation factor were shown in Figure 4. The total
flux increases slightly with the addition of zeolite concentration up to 0.5 wt.%. Actually
the diffusion through chitosan zeolite membranes took placed in the polymer chain of
chitosan and the zeolite pores. The interactions between the chitosan and zeolite
molecules prevent IPA molecules to pass through. So the separation factors at 0.1 and
0.25 wt.% zeolite are higher than the unmodified membrane. However the zeolite pores
also affected the permeate flux and separation factor. Therefore the permeate flux
increases while the separation factor decreases with the increase of zeolite.

When the zeolite content is higher than 0.5 wt.%, the permeate flux diminishes
significantly while the separation factor only increases at 1 wt.% zeolite. For the tensile
strength, chitosan zeolite membranes showed the significant increase started at 0.5 wt.%
zeolite. The links between chitosan and zeolite molecules are stronger and tighter. This
prevents water and IPA to pass through the membrane. So the permeate flux reduces and
the separation increases at 1 wt.% of zeolite A.
 Overall performance is determined by Pervaporation Separation Index (PSI),
which combines the total permeation flux and the selectivity. Figure 5 shows the PSI
versus the zeolite concentration. PSI of modified chitosan membranes improved only at
low zeolite content.

400 90

350 80

70
The total permeate flux (g/m2.h)

300

60
250

Separation factor
50
200
40
150
30

100
20

50 10

0 0
10

25

50

75

00
o
om

0,

0,

0,

0,

1,
H

Zeolite A content (wt.%)

Figure 4: The Total Permeation Flux and Separation Factor versus Zeolite Concentration of all
types of membranes
 21000

Pervaporation Separation Index

16000

11000

6000

1000
0 0,25 0,5 0,75 1
Zeolite A content (wt.%)

Figure 5: Pervaporation Separation Index

3.4. The effect of feed temperature

The effect of feed temperature was carried out from room temperature to 70°C.
The feed concentration was 90 wt.% IPA. The effect of feed temperature of membranes
with 0.25 wt.% zeolite was chosen to study. The total permeation flux of unmodified
membrane at room temperature was assumed to be 1. The normalized flux, which is the
ratio of the permeate flux to the unmodified membrane at room temperature, and
separation factor were shown in Figure 6. When the feed temperature increases, the
permeation flux of modified membrane increases much higher than that of unmodified
membrane. All the separation factors decrease with the increase of the feed temperature
but the chitosan zeolite membranes decline faster than homogeneous ones. Only at low
temperature both 2 types of the membranes give the higher selectivity. Both the increase
of the permeate flux and the decrease of the separation factor at high temperature is
probably caused by the thermal motion of the feed components and the polymer chains in
the membranes. As the temperature raises, the thermal agitation increases and create
more free volume in the polymer matrix. Therefore the permeation flux increases very
fast and more molecules of alcohol and water can diffuse through the membranes.
However both the permeation flux and the separation factor of chitosan zeolite
membranes are higher than homogeneous membranes except at 70°C. The pervaporation
activation energy of homogeneous chitosan membranes is 4.47 kcal/mol for water and
8.43 kcal/mol for IPA while that of modified membranes is 6.95 kcal/mol for water and
13.88 kcal/mol for IPA.

8 70
7 60
6 50

Separation factor
The normalized flu

5
40
4
30
3
20
2
1 10

0 0
Room 40 50 60 70
temperature Feed temperature (C degree)
Unmodified Times of flux Separation factor Modified

Figure 6: The Times of Permeation Flux Compared with The Unmodified at room temperature
and Separation Factor of Both Unmodified and Modified Membranes versus Feed Temperature

4. Conclusion

The conclusions that can be withdrawn from this study:

1. As the zeolite content increases, the tensile strength as well as Young’s

modulus also increase but the elongation decreases. The modified membranes become
stronger but more brittle than the unmodified membranes.

2. The chitosan zeolite membranes with 0.25 wt.% in the chitosan solution
gives the best performance. The membranes improve both the permeation flux and the
selectivity at low zeolite concentration. The overall PSI of chitosan zeolite membranes is
always higher than the homogeneous membranes and gives the better performance at low
feed temperature.
References

[1] Chandy, T. and Sharma, C.P. (1989). “Bioactive molecules immobilized to

liposome modified Albumin-blended chitosan membranes-Antithrombotic and
Permeability Properties”. Journal of Colloid and Interface Science. USA: Academic
Press. 130(2). 331-340

[2] Shah, K.R. and Mowbray, S.L. (1992). “Compatible blends containing chitosan”.
EP 0,486,294

[3] Wendel, H. and Lang, G. (1988). “Cosmetic composition based upon chitosan and
ampholytic copolymerizates, as well as new chitosan/polyampholyte salts”. DE
3,644,097

[4] Hadwiger, L.A. (1990). “Method for treating cereal crop seed with chitosan to
enhance yield, root growth and stem strength”. US 4,978,381

[5] Imeri, A. and Knorr, D. (1988). “Effects of chitosan on yield and compositional
data of carrot and apple juice”. Journal of Food Science. Institute of Food Technologists.
53. 1707-1709

[6] Mohd. Ghazali Mohd. Nawawi (1997). “Pervaporation dehydration of

isopropanol-water systems using chitosan membranes”. Ph.D Thesis. Canada: University
of Waterloo.

[7] Feng, X. and Huang, R.Y.M (1996). “Pervaporation with chitosan membrane”.
Journal of Membrane Science. The Netherlands: Elsevier. 116. 67-76

[8] Mohd. Ghazali Mohd. Nawawi and Huang, R.Y.M. (1997). “Pervaporation
dehydration of iso-propanol with chitosan membranes”. Journal of Membrane Science.
The Netherlands: Elsevier. 124. 53-62

[9] Huang, R.Y.M, Pal, R. and Moon, G.Y. (1999). “Crosslinked chitosan composite
membrane for pervaporation dehydration of alcohol mixtures and enhancement of
structural stability of chitosan/polysulfone composite membranes”. Journal of Membrane
Science. The Netherlands: Elsevier. 160(1). 17-30

[10] Shich, J.J and Huang, R.Y.M. (1998). “Chitosan/N_methylol nylon 6 blend
membranes for the pervaporation separation of ethanol-water mixtures”. Journal of
Membrane Science. The Netherlands: Elsevier. 148(2). 243-255

[11] Huang, R.Y.M, Moon, G.Y. and Pal, R. (2001). “Chitosan/anionic surfactant
complex membranes for the pervaporation separation of methanol/MTBE and
characterization of the polymer/surfactant system”. Journal of Membrane Science. The
Netherlands: Elsevier. 184(1). 1-15
[12] Lee, Y.M (1993). “Modified chitosan membranes for pervaporation”.
Desalination. The Netherlands: Elsevier. 90. 277-290

[13] Yong, H.H., Park, H.C., Kang, Y.S., Won, J. and Kim, W.N. (2001). “Zeolite-
filled polyimide membrane containing 2,4,6-triaminopyrimidine”. Journal of Membrane
Science. The Netherlands: Elsevier. 188. 151-163

[14] Boom, J.P., Pünt, I.G.M., Zwijnenberg, H., Boer, R., Bargeman, D., Smolders,
C.A and Strathman, H. (1998). “Transport through zeolite filled polymeric
membranes”.Journal of Membrane Science. The Netherlands: Elsevier. 138. 237-258

[15] Gao, Z., Yue, Y. and Li, W. (1996). “Application of zeolite-filled pervaporation
membrane”. Zeolites. New York: Elsevier. 16. 70-74

[16] Mohd. Ghazali Mohd. Nawawi, Hashim Hassan and Abdul Rahim Jalil. (2003).
Modification of Chitosan-Based Membranes Using Zeolite for Pervaporation Separation
of Isopropanol-Water Mixtures. Proceedings of International Conference On Chemical
and Bioprocess Engineering. 87-90