- Email: [email protected]
UDMH mixtures
Separation and Purification Technology 47 (2006) 173–178
Short communication
Preparation of NaA zeolite membranes for separation of water/UDMH mixtures Mansoor Kazemimoghadam, Toraj Mohammadi ∗ Research Lab for Separation Processes, Faculty of Chemical Engineering, Iran University of Science and Technology, Narmak, Tehran, Iran Received 2 February 2005; received in revised form 13 June 2005; accepted 23 June 2005
Abstract Water/unsymmetrical dimethylhydrazine (UDMH) mixtures were separated at ambient temperature and atmospheric pressure by pervaporation (PV) using NaA zeolite membranes. Effects of crystallization time and temperature on the membrane performance were studied. These membranes showed very high selectivity of water for water/UDMH mixtures. Separation factor as high as 52 000 was obtained for UDMH feed concentration of 5%. Total mass flux was also obtained up to 3.95 kg/m2 h. The results confirm the superior performance of the zeolite membranes for dehydration of water/UDMH mixtures due to their uniform nanopores. © 2005 Elsevier B.V. All rights reserved. Keywords: NaA; Pervaporation; Zeolite membrane; UDMH; Dehydration
1. Introduction Polymeric membranes are not generally suitable for applications involving harsh chemicals like UDMH due to membrane chemical instability. However, a recent development of chemical-and-temperature resistant hydrophilic ceramic membranes has made it possible to overcome the limitations of hydrophilic polymeric membranes [1]. PV is an economical separation technique compared with conventional separation methods such as distillation especially in processes involving azeotropes, isomers and (removal or recovery of) trace substances. Due to its high separation factor and flux rate, PV results in energy cost saving and safe operation [2–4]. UDMH is an important liquid propellant, however, it also has many other applications such as an oxygen scavenger for boiler-feed water, a starting material for drug and dye intermediates, a catalyst for polymerization reactions, etc. [5,6]. UDMH is produced by the rashing process which involves reaction of ammonia with sodium hypochloride to give chloramines which in turn are reacted with dimethylamine to ∗
Corresponding author. Tel.: +98 21 789 6621; fax: +98 21 7896 620. E-mail address: [email protected] (T. Mohammadi).
1383-5866/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.seppur.2005.06.013
obtain UDMH. Its reaction liquor contains low concentrations of UDMH (<2%) besides impurities such as sodium chloride (8%) and water (>90%). It was reported that UDMH and water makes an azeotrope (1–5 wt.%). Thus, it is important to find a commercial separation process for dilute UDMH/water mixtures. The conventional purification methods, which involve flash vaporization followed by extractive distillation (at 20% UDMH, aqueous azeotrope), are highly energy-intensive and about 210 kg of steam is required to obtain 1 kg of UDMH. In addition, this method is very hazardous since UDMH is highly flammable and forms explosive mixtures with air. Moreover, UDMH is very corrosive and its vapor is extremely toxic and carcinogenic. Therefore, search for a safer and more economical alternative technology is a challenging problem. PV has all the requirements for completely replacing extractive distillation for separation of the azeotrope. This can be combined with simple distillation, as a hybrid process, for enrichment of UDMH to high purity levels [7,8]. Dehydration of hydrazine and monomethylhydrazine (MMH) using ethyl cellulose membranes were previously carried out [9]. However, ethyl cellulose membranes could not be used to dehydrate UDMH because they were degraded rapidly. Also, hydrophobic polymers such as polydimethyl-
174
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
siloxane (PDMS) could not withstand the highly alkaline (pH = 13–14) medium whereas membranes made of polyethylene and polypropylene gave negligible flux so that UDMH could not be selectively extracted from its dilute reaction liquor. Problem of chemical compatibility is also encountered in the case of hydrophilic polymers such as polyvinyl alcohol and polyacrylic acid [10]. Chitosan, a derivative of the naturally abundant biopolymer chittin, is fully stable in aqueous UDMH and hence can be selected for its dehydration, keeping in minds its highly hydrophilic nature and good mechanical strength. The promising potential of chitosan as a PV membrane was studied for dehydration of alcohols such as ethanol and isopropanol. This polymer was recently used to form highly selective and permeable blend membranes with polyvinyl alcohol, sodium alginate, etc. [11–16]. In this study, zeolite NaA membranes were fabricated and then used to separate water/UDMH mixtures. Zeolite NaA layers were coated on external surface of porous tubular mullite supports using hydrothermal method. These membranes were successfully used for dehydration of water/UDMH mixtures.
2. Experimental In ceramic membranes, thin dense layers are usually deposited over porous supports. The porous supports provide mechanical strength for the thin selective layers. Porous supports can be made from alumina, cordierite, mullite, silica, spinel, zirconia, other refractory oxides and various oxide mixtures, carbon, sintered metals and silicon carbide [17,18]. In this research, mullite supports were prepared from kaolin clay. Kaolin clay is thermally converted to mullite via high temperature calcination. The mullitization reaction takes place when kaolin clay is utilized as the sole source of silica and alumina. The reaction can be represented by the following equation where the approximate chemical formula for kaolin (without the water of hydration) is given as Al2 O3 ·2SiO2 and the formula for mullite is given as 3Al2 O3 ·2SiO2 : 3(Al2 O3 ·2SiO2 ) → 3Al2 O3 ·2SiO2 + 4SiO2 The term represented by 4SiO2 is the free silica generated as a result of the conversion. This free silica was leached and then porous mullite bodies were prepared. Mullite has several distinct advantages over other bodies such as alumina. Since these bodies are heated to high temperatures to achieve the mullite conversion reaction, strong inter-crystalline bonds between mullite crystals, are formed and this results in excellent strength and attrition. Leaching time depends on several factors including: • • • •
the quantity of free silica to be removed, the porosity of the body prior to leaching, concentration of the leaching solution and temperature [19].
Table 1 Analysis of the kaolin clay Component SiO2 TiO2 Al2 O3 Fe2 O3 K2 O Na2 O LOI Total
Percent 51.9 0.1 34.1 1.4 0.8 0.1 11.6 100
Phases Kaolinite Illite Quartz Feldspar
Percent 79 8 10 3
100
The kaolin material used in this study (SL-KAD grade) was supplied from WBB cooperation. The analysis of the kaolin is listed in Table 1. UDMH (98%) and sodium hydroxide (Merck) were also used in all experiments. Cylindrical shaped (tubular) bodies (i.d.: 10 mm, o.d.: 14 mm and L: 15 cm) were conveniently made by extruding a mixture of about 67–75% kaolin clay and 33–25% distilled water using an extruder. Suitable calcination temperatures and periods are those at which the clay converts to mullite and free silica. Good results were achieved by calcining for about 3 h at temperatures of about 1250 ◦ C [20]. Free silica was removed from the calcined bodies by leaching with strong alkali solutions. Removal of this free silica causes mesoporous tubular supports to be made with very high porosity. Free silica removal was carried out with aqueous solutions containing 20% by weight NaOH at a temperature of 80 ◦ C for 5 h. Supports were washed with 2 l of water for 12 h at a temperature of 80 ◦ C in order to remove NaOH. Porosity of the support before leaching is 24.3% while after treatment it increases to 49%. Flux of the support before and after free silica removal at 1 bar and 20 ◦ C are 6 and 10 kg/m2 h, respectively. The porosity of support was measured by water absorption method [21]. Thin zeolite NaA membrane layers were grown hydrothermally over the external surface of the porous supports. Synthesis solution was prepared by mixing aluminate and silicate solutions. NaOH (4.87 g) was dissolved in 76 ml of distilled water. The solution was divided into two equal volumes and kept in polypropylene bottles. Aluminate solution was prepared by adding 6.23 g sodium aluminate (Aldrich, 50–56% Al2 O3 ) to one part of the NaOH solution. It was mixed until cleared. Silicate solution was prepared by adding 16.57 g sodium silicate (Merck, 25–28% SiO2 ) to another part of the NaOH solution. Silicate solution was then poured into aluminate solution and mixed until a thick homogenized gel was formed. Composition of the homogeneous solution of zeolite NaA is represented by the following molar ratio: 1.926 SiO2 :Al2 O3 :3.165 Na2 O:128 H2 O [21,22]. Two ends of the supports were closed with rubber caps to avoid any precipitation of the zeolite crystals on internal surface of the supports during membrane synthesis. The seeded supports were placed vertically in a Teflon autoclave. The
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
175
Fig. 1. PV setup (a) dead end and (b) cross-flow (1: feed container and PV cell, 2: liquid nitrogen trap, 3: permeate container, 4: three stage vacuum pump, 5: centrifuge pump and 6: tank feed).
solution was carefully poured in the autoclave and then the autoclave was sealed. Crystallization was carried out in an oven at temperatures of 70, 90, 100, 110 and 130 ◦ C for 1, 2.5, 3, 4 and 5.5 h. The samples were then taken and the synthesized membranes were washed several times with distilled water. The samples were then dried at room temperature for 12 h in air. Some samples were coated two and three times to study effect of number of coating. The zeolite membranes were used for dehydration of aqueous UDMH. Dilute UDMH mixtures (2 and 5 wt.%) were used and experiments were carried out at room temperature (25 ◦ C) within a period of 30–60 min. Permeate concentrations were measured using GC (TCD detector, Varian 3400, carrier gas: hydrogen, column is polyethylene glycol, sample size: 5 micron, column and detector temperatures: 120–150 ◦ C, detector flow rate: 15 ml/min, carrier flow: 5 ml/min, column pressure: 1.6 kPa, GC input pressure: 20 kPa). Performance of PV was evaluated using values of total flux (kg/m2 h) and separation factor (dimensionless) [20]. A typical experimental setup was employed as presented in Fig. 1. To make sure the process was steady state, the permeate was collected after 20 min. Membrane surface area (A) was 0.0044 m2 and permeation
time (t) was 0.5 h. Flux and separation factor were calculated using the following equations: j=
m , At
α=
Y (1 − X) X(1 − Y )
where m is the permeate weight (kg) and X and Y are the weight fractions of UDMH in feed and permeate, respectively. It must be mentioned that the PV experiments were carried out in two modes: dead end and cross-flow. In the former, the process is batch and the feed is kept over the membrane surface, while in the latter, the process is continuous and the feed is passed over the membrane surface (Fig. 1). The mullite, cristobalite and SiO2 phase identification was performed by X-ray diffractometry (Philips PW1710, Philips Co., The Netherlands) with Cu K␣ radiation. XRD patterns were presented in Fig. 2. Morphology of the support and the membrane was examined by Scanning Electron Microscopy (JEM-1200 or JEM-5600LV equipped with an Oxford ISIS300 X-ray disperse spectroscopy (EDS)). SEM photographs were presented in Fig. 3.
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
176
Fig. 2. XRD patterns of (a) the support and (b) the membrane.
3. Results and discussion To study effects of crystallization time and temperature on NaA zeolite membrane performance, the membranes were synthesized at different temperatures (70, 90, 100, 110, and 130 ◦ C) for different times (1, 2.5, 3, 4, and 5.5 h). It must be also mentioned that three samples were prepared for each condition. The results were presented on average and the maximum deviation was less than 3%. The membranes were evaluated in the PV setup as shown in Fig. 1. As mentioned, the synthesis procedure was performed using different times. As seen in Table 2, increasing crystallization time decreases flux (samples 6 and 7). However, there is no change in separation factor. This may be due to the fact that at a longer crystallization time a thicker membrane layer is formed. This cases flux to decrease. This shows
that these membranes behave very high selectivity. It must be mentioned that 20 400 is the highest measurable value using the GC at 2 wt.% UDMH concentration. The results show that short crystallization time (1 h) is not enough to make an effective zeolite layer on the support (sample 5). Also, long crystallization time (5.5 h) causes NaA zeolite to transform to other zeolites such as NaX [3]. As a result, this sample (8) shows poor selectivity. The crystallization time in a range of 2.5–4 h was found to be very effective for making the NaA zeolite layer. The synthesis procedure was also performed using different temperatures. As seen in Table 2, increasing crystallization temperature increases flux (samples 1–3). Also, it can be observed that there is no change in separation factor. This may be due to the fact that at higher crystallization temperature, a thinner layer is formed. It is because, at higher temperatures,
Table 2 Flux and separation factor of NaA zeolite membranes (dead end) Sample
Number of coating
Crystallization time (h)
Crystallization temperature (◦ C)
Concentration of UDMH in feed (wt.%)
Flux (kg/m2 h)
Separation factor
1 2 3 4 5 6 7 8 9 10 11 12
1 1 1 1 1 1 1 1 1 1 2 3
3 3 3 3 1 2.5 4 5.5 3 3 3 3
70 90 110 130 100 100 100 100 110 100 100 100
2 2 2 2 2 2 2 2 5 5 5 5
0.0654 0.0708 0.0871 0.4685 1.1400 0.6200 0.2900 0.3600 0.0940 0.4437 0.3840 0.3243
20400 20400 20400 20400 41 20400 20400 2 52600 52600 52600 52600
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
177
Table 3 Flux of cross-flow measurements (sample 12) Run
Concentration of UDMH in feed (wt.%)
P (bar)
Q (l/min)
T (◦ C)
Flux (kg/m2 h)
1 2 3 4 5
5 5 5 5 5
2 2 2 2 2
3 1.5 0.5 3 3
55 55 55 25 40
3.95 3.61 3.30 2.33 3.44
the support to be formed after a two or three stage synthesis. Therefore, quality of NaA zeolite membrane layer may be improved. The synthesized layers by a multi stage procedure are substantially thicker than the corresponding layers by a single stage. As a result, flux is less for the three stage sample (12) as shown in Table 2. The experiments were carried out using two different concentrations. The results show that increasing UDMH concentration in feed increases separation factor and flux as shown in Table 2 (samples 3 and 9). It must also be mentioned that 52000 is the highest measurable value using the GC at wt.% UDMH concentration. Cross-flow PV experiments were carried out and the results were presented in Table 3. As seen, increasing flow rate increases flux (runs 1–3). This can be due to the fact that higher flow rates diminish effect of boundary layer. Also, increasing temperature increases flux (runs 1, 4 and 5). This can be due to enhancement of adsorption and diffusion of water to and through the membrane. The results are comparable with (or even better than) those of other simple water/organic mixtures in the literature [14,15]. Fig. 2 shows XRD patterns of the mullite support (a) and the NaA zeolite membrane (b). The XRD pattern of NaA zeolite membrane confirms that zeolite NaA crystals were formed. Fig. 3 shows SEM photographs of the mullite support (a), the NaA zeolite membrane (surface) (b) and the NaA zeolite membrane (cross-section) (c). Porous structure of the support and thin layer of the membrane can be easily observed. The results confirm that NaA zeolite membranes synthesized at 130 ◦ C for 3 h via a single stage process can be recommended for dehydration of aqueous UDMH dilute mixtures. Fig. 3. SEM micrograph of (a) the support, (b) the membrane (surface) and (c) the membrane (cross-section).
4. Conclusion NaA zeolite crystals are smaller. This shows that these membranes behave very high selectivity. The results show that at a high temperature of 130 ◦ C, zeolite NaA is also well formed much thinner (sample 4). The crystallization temperature in a range of 70–130 ◦ C was found to be very effective for making the NaA zeolite layer. Synthesis procedure was repeated where fresh synthesis gel was added periodically to the cooled and cleaned samples (samples 10–12). This causes more NaA zeolite crystals on
NaA zeolite membrane was firstly used for dehydration of aqueous UDMH mixtures. Zeolite NaA membranes were synthesized on the porous mullite tubes by hydrothermal method. The mullite supports were made by extruding kaolin clay. Zeolite membranes showed much higher fluxes and selectivities than commercially available polymeric membranes due to their uniform nanopores. These membranes showed very good membrane performance for separation of
178
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
UDMH/water mixtures. It is expected that even significantly higher fluxes, with similar selectivities, can be achieved at higher temperatures. Since zeolite NaA membranes can withstand high temperatures and harsh environments (pH > 12), dehydration of UDMH mixtures with high pH can be performed. It is expected that PV using these membranes can be a highly interesting tool for industry, provided they can be produced cheap at a large scale. PV through a NaA zeolite membrane is an effective technique to separate UDMH from water. Separation factors as high as 52 000 was obtained at wt.% UDMH concentration.
Acknowledgments Authors would like to appreciate Industrial Development and Renovation Organization (IDRO) of Iran for the support.
References [1] J. Jafar, M. Budd, Separation of alcohol/water mixtures by pervaporation through zeolite NaA membranes, Microp. Mater. 12 (1997) 305–311. [2] Q. Liu, R. Noble, L. Falconer, H. Funke, Organic/water separation by pervaporation with a zeolite membrane, J. Membr. Sci. 117 (1996) 163–174. [3] D. Shah, K. Kissick, A. Ghorpade, R. Hannah, D. Bhattacharyya, Pervaporation of alcohol–water and dimethylformamide–water mixtures using hydrophilic zeolite NaA membranes: mechanisms and experimental results, J. Membr. Sci. 179 (2000) 185–205. [4] A. Verkerk, P. van Male, M.A.G. Vorstman, J.T.F. Keurentjes, Properties of high flux ceramic pervaporation membranes for dehydration of alcohol/water mixtures, Sep. Purif. Technol. 22–23 (2001) 689–695. [5] R. Ravindra, S. Sridhar, A.A. Khan, A.K. Rao, Pervaporation of water, hydrazine and monomethylhydrazine using ethylcellulose membranes, Polymer 41 (2000) 2795–2806. [6] S. Sridhar, R. Ravindra, A.A. Khan, Recovery of monomethylhydrazine liquid propellant by pervaporation technique, Ind. Eng. Chem. Res. 39 (2001) 2485-L2490. [7] Li. Xin-Gui, I. Kresse, Z.K.J. Zhou, J. Springer, Effect of temperature and pressure on gas transport in ethyl cellulose membrane, Polymer 42 (2001) 6801–6810.
[8] R. Ravindra, K.R. Krovvidi, A.A. Khan, A.K. Rao, D.S.C. studies of states of water, hydrazine and hydrazine hydrate in ethyl cellulose membrane, Polymer 40 (1999) 1159– 1165. [9] R. Ravindra, A. Kameswara, A. Khan, A qualitative evaluation of water and monomethyl hydrazine in ethyl cellulose membrane, J. Appl. Polym. Sci. 72 (1999) 689–700. [10] S. Sridhar, G. Susheela, G.J. Reddy, A.A. Khan, Cross-linked chitosan membranes: characterization and study of dimethylhydrazine dehydration by pervaporation, Polym. Int. 50 (2001) 1156–1165. [11] R. Ravindra, A.K. Rao, A.A. Khan, Processing of liquid propellant reaction liquors by pervaporation, J. Appl. Polym. Sci. 72 (1999) 141–149. [12] R. Ravindra, S. Sridhar, A.A. Khan, Separation studies of hydrazine from aqueous solutions by pervaporation, J. Polym. Sci., Part B: Polym. Phys. 37 (1999) 1969–1980. [13] R. Ravindra, R. Krovvidi, A. Khan, Solubility parameter of chitin and chitosan, Carbohydr. Polym. 36 (1998) 121–127. [14] M. Kondo, M. Komori, H. Kita, K.I. Okamoto, Tubular-type pervaporation module with zeolite NaA membrane, J. Membr. Sci. 133 (1997) 133–141. [15] Y. Morigami, M. Kondo, J. Abe, H. Kita, K.I. Okamoto, The first large-scale pervaporation plant using tubular-type module with zeolite NaA membrane, Sep. Purif. Technol. 25 (2001) 251– 260. [16] T. Gallego-Lizon, Y.S. Ho, L.F.D. Santos, Comparative study of commercially available polymeric and microporous silica membranes for the dehydration of IPA/water mixtures by pervaporation/vapour permeation, Desalination 149 (2002) 3–8. [17] T. Gallego-Lizon, E. Edwards, G. Lobiundo, L.F.D. Santos, Dehydration of water/t-butanol mixtures by pervaporation: comparative study of commercially available polymeric, microporous silica and zeolite membranes, J. Membr. Sci. 197 (2002) 309– 319. [18] W.C. Wong, L.T.Y. Au, C.T. Ariso, K.L. Yeung, Effects of synthesis parameters on the zeolite membrane growth, J. Membr. Sci. 191 (2001) 143–163. [19] T. Mohammadi, A. Pak, Effect of calcination temperature of kaolin as a support for zeolite membranes, Sep. Purif. Technol. 30 (2003) 241–249. [20] M. Kazemimoghadam, A. Pak, T. Mohammadi, Dehydration of water/1-1-dimethylhydrazine mixtures by zeolite membranes, Microp. Mesop. Mater. 70 (2004) 127–134. [21] T. Mohammadi, A. Pak, Making zeolite NaA membrane from kaolin by electrophoresis, Microp. Mesop. Mater. 56 (2002) 81–88. [22] H. Robson, R.W. Thompson, K.C. Franklin, Verified Synthesis of Zeolitic Materials, Elsevier Publisher, 2001.
Short communication
Preparation of NaA zeolite membranes for separation of water/UDMH mixtures Mansoor Kazemimoghadam, Toraj Mohammadi ∗ Research Lab for Separation Processes, Faculty of Chemical Engineering, Iran University of Science and Technology, Narmak, Tehran, Iran Received 2 February 2005; received in revised form 13 June 2005; accepted 23 June 2005
Abstract Water/unsymmetrical dimethylhydrazine (UDMH) mixtures were separated at ambient temperature and atmospheric pressure by pervaporation (PV) using NaA zeolite membranes. Effects of crystallization time and temperature on the membrane performance were studied. These membranes showed very high selectivity of water for water/UDMH mixtures. Separation factor as high as 52 000 was obtained for UDMH feed concentration of 5%. Total mass flux was also obtained up to 3.95 kg/m2 h. The results confirm the superior performance of the zeolite membranes for dehydration of water/UDMH mixtures due to their uniform nanopores. © 2005 Elsevier B.V. All rights reserved. Keywords: NaA; Pervaporation; Zeolite membrane; UDMH; Dehydration
1. Introduction Polymeric membranes are not generally suitable for applications involving harsh chemicals like UDMH due to membrane chemical instability. However, a recent development of chemical-and-temperature resistant hydrophilic ceramic membranes has made it possible to overcome the limitations of hydrophilic polymeric membranes [1]. PV is an economical separation technique compared with conventional separation methods such as distillation especially in processes involving azeotropes, isomers and (removal or recovery of) trace substances. Due to its high separation factor and flux rate, PV results in energy cost saving and safe operation [2–4]. UDMH is an important liquid propellant, however, it also has many other applications such as an oxygen scavenger for boiler-feed water, a starting material for drug and dye intermediates, a catalyst for polymerization reactions, etc. [5,6]. UDMH is produced by the rashing process which involves reaction of ammonia with sodium hypochloride to give chloramines which in turn are reacted with dimethylamine to ∗
Corresponding author. Tel.: +98 21 789 6621; fax: +98 21 7896 620. E-mail address: [email protected] (T. Mohammadi).
1383-5866/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.seppur.2005.06.013
obtain UDMH. Its reaction liquor contains low concentrations of UDMH (<2%) besides impurities such as sodium chloride (8%) and water (>90%). It was reported that UDMH and water makes an azeotrope (1–5 wt.%). Thus, it is important to find a commercial separation process for dilute UDMH/water mixtures. The conventional purification methods, which involve flash vaporization followed by extractive distillation (at 20% UDMH, aqueous azeotrope), are highly energy-intensive and about 210 kg of steam is required to obtain 1 kg of UDMH. In addition, this method is very hazardous since UDMH is highly flammable and forms explosive mixtures with air. Moreover, UDMH is very corrosive and its vapor is extremely toxic and carcinogenic. Therefore, search for a safer and more economical alternative technology is a challenging problem. PV has all the requirements for completely replacing extractive distillation for separation of the azeotrope. This can be combined with simple distillation, as a hybrid process, for enrichment of UDMH to high purity levels [7,8]. Dehydration of hydrazine and monomethylhydrazine (MMH) using ethyl cellulose membranes were previously carried out [9]. However, ethyl cellulose membranes could not be used to dehydrate UDMH because they were degraded rapidly. Also, hydrophobic polymers such as polydimethyl-
174
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
siloxane (PDMS) could not withstand the highly alkaline (pH = 13–14) medium whereas membranes made of polyethylene and polypropylene gave negligible flux so that UDMH could not be selectively extracted from its dilute reaction liquor. Problem of chemical compatibility is also encountered in the case of hydrophilic polymers such as polyvinyl alcohol and polyacrylic acid [10]. Chitosan, a derivative of the naturally abundant biopolymer chittin, is fully stable in aqueous UDMH and hence can be selected for its dehydration, keeping in minds its highly hydrophilic nature and good mechanical strength. The promising potential of chitosan as a PV membrane was studied for dehydration of alcohols such as ethanol and isopropanol. This polymer was recently used to form highly selective and permeable blend membranes with polyvinyl alcohol, sodium alginate, etc. [11–16]. In this study, zeolite NaA membranes were fabricated and then used to separate water/UDMH mixtures. Zeolite NaA layers were coated on external surface of porous tubular mullite supports using hydrothermal method. These membranes were successfully used for dehydration of water/UDMH mixtures.
2. Experimental In ceramic membranes, thin dense layers are usually deposited over porous supports. The porous supports provide mechanical strength for the thin selective layers. Porous supports can be made from alumina, cordierite, mullite, silica, spinel, zirconia, other refractory oxides and various oxide mixtures, carbon, sintered metals and silicon carbide [17,18]. In this research, mullite supports were prepared from kaolin clay. Kaolin clay is thermally converted to mullite via high temperature calcination. The mullitization reaction takes place when kaolin clay is utilized as the sole source of silica and alumina. The reaction can be represented by the following equation where the approximate chemical formula for kaolin (without the water of hydration) is given as Al2 O3 ·2SiO2 and the formula for mullite is given as 3Al2 O3 ·2SiO2 : 3(Al2 O3 ·2SiO2 ) → 3Al2 O3 ·2SiO2 + 4SiO2 The term represented by 4SiO2 is the free silica generated as a result of the conversion. This free silica was leached and then porous mullite bodies were prepared. Mullite has several distinct advantages over other bodies such as alumina. Since these bodies are heated to high temperatures to achieve the mullite conversion reaction, strong inter-crystalline bonds between mullite crystals, are formed and this results in excellent strength and attrition. Leaching time depends on several factors including: • • • •
the quantity of free silica to be removed, the porosity of the body prior to leaching, concentration of the leaching solution and temperature [19].
Table 1 Analysis of the kaolin clay Component SiO2 TiO2 Al2 O3 Fe2 O3 K2 O Na2 O LOI Total
Percent 51.9 0.1 34.1 1.4 0.8 0.1 11.6 100
Phases Kaolinite Illite Quartz Feldspar
Percent 79 8 10 3
100
The kaolin material used in this study (SL-KAD grade) was supplied from WBB cooperation. The analysis of the kaolin is listed in Table 1. UDMH (98%) and sodium hydroxide (Merck) were also used in all experiments. Cylindrical shaped (tubular) bodies (i.d.: 10 mm, o.d.: 14 mm and L: 15 cm) were conveniently made by extruding a mixture of about 67–75% kaolin clay and 33–25% distilled water using an extruder. Suitable calcination temperatures and periods are those at which the clay converts to mullite and free silica. Good results were achieved by calcining for about 3 h at temperatures of about 1250 ◦ C [20]. Free silica was removed from the calcined bodies by leaching with strong alkali solutions. Removal of this free silica causes mesoporous tubular supports to be made with very high porosity. Free silica removal was carried out with aqueous solutions containing 20% by weight NaOH at a temperature of 80 ◦ C for 5 h. Supports were washed with 2 l of water for 12 h at a temperature of 80 ◦ C in order to remove NaOH. Porosity of the support before leaching is 24.3% while after treatment it increases to 49%. Flux of the support before and after free silica removal at 1 bar and 20 ◦ C are 6 and 10 kg/m2 h, respectively. The porosity of support was measured by water absorption method [21]. Thin zeolite NaA membrane layers were grown hydrothermally over the external surface of the porous supports. Synthesis solution was prepared by mixing aluminate and silicate solutions. NaOH (4.87 g) was dissolved in 76 ml of distilled water. The solution was divided into two equal volumes and kept in polypropylene bottles. Aluminate solution was prepared by adding 6.23 g sodium aluminate (Aldrich, 50–56% Al2 O3 ) to one part of the NaOH solution. It was mixed until cleared. Silicate solution was prepared by adding 16.57 g sodium silicate (Merck, 25–28% SiO2 ) to another part of the NaOH solution. Silicate solution was then poured into aluminate solution and mixed until a thick homogenized gel was formed. Composition of the homogeneous solution of zeolite NaA is represented by the following molar ratio: 1.926 SiO2 :Al2 O3 :3.165 Na2 O:128 H2 O [21,22]. Two ends of the supports were closed with rubber caps to avoid any precipitation of the zeolite crystals on internal surface of the supports during membrane synthesis. The seeded supports were placed vertically in a Teflon autoclave. The
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
175
Fig. 1. PV setup (a) dead end and (b) cross-flow (1: feed container and PV cell, 2: liquid nitrogen trap, 3: permeate container, 4: three stage vacuum pump, 5: centrifuge pump and 6: tank feed).
solution was carefully poured in the autoclave and then the autoclave was sealed. Crystallization was carried out in an oven at temperatures of 70, 90, 100, 110 and 130 ◦ C for 1, 2.5, 3, 4 and 5.5 h. The samples were then taken and the synthesized membranes were washed several times with distilled water. The samples were then dried at room temperature for 12 h in air. Some samples were coated two and three times to study effect of number of coating. The zeolite membranes were used for dehydration of aqueous UDMH. Dilute UDMH mixtures (2 and 5 wt.%) were used and experiments were carried out at room temperature (25 ◦ C) within a period of 30–60 min. Permeate concentrations were measured using GC (TCD detector, Varian 3400, carrier gas: hydrogen, column is polyethylene glycol, sample size: 5 micron, column and detector temperatures: 120–150 ◦ C, detector flow rate: 15 ml/min, carrier flow: 5 ml/min, column pressure: 1.6 kPa, GC input pressure: 20 kPa). Performance of PV was evaluated using values of total flux (kg/m2 h) and separation factor (dimensionless) [20]. A typical experimental setup was employed as presented in Fig. 1. To make sure the process was steady state, the permeate was collected after 20 min. Membrane surface area (A) was 0.0044 m2 and permeation
time (t) was 0.5 h. Flux and separation factor were calculated using the following equations: j=
m , At
α=
Y (1 − X) X(1 − Y )
where m is the permeate weight (kg) and X and Y are the weight fractions of UDMH in feed and permeate, respectively. It must be mentioned that the PV experiments were carried out in two modes: dead end and cross-flow. In the former, the process is batch and the feed is kept over the membrane surface, while in the latter, the process is continuous and the feed is passed over the membrane surface (Fig. 1). The mullite, cristobalite and SiO2 phase identification was performed by X-ray diffractometry (Philips PW1710, Philips Co., The Netherlands) with Cu K␣ radiation. XRD patterns were presented in Fig. 2. Morphology of the support and the membrane was examined by Scanning Electron Microscopy (JEM-1200 or JEM-5600LV equipped with an Oxford ISIS300 X-ray disperse spectroscopy (EDS)). SEM photographs were presented in Fig. 3.
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
176
Fig. 2. XRD patterns of (a) the support and (b) the membrane.
3. Results and discussion To study effects of crystallization time and temperature on NaA zeolite membrane performance, the membranes were synthesized at different temperatures (70, 90, 100, 110, and 130 ◦ C) for different times (1, 2.5, 3, 4, and 5.5 h). It must be also mentioned that three samples were prepared for each condition. The results were presented on average and the maximum deviation was less than 3%. The membranes were evaluated in the PV setup as shown in Fig. 1. As mentioned, the synthesis procedure was performed using different times. As seen in Table 2, increasing crystallization time decreases flux (samples 6 and 7). However, there is no change in separation factor. This may be due to the fact that at a longer crystallization time a thicker membrane layer is formed. This cases flux to decrease. This shows
that these membranes behave very high selectivity. It must be mentioned that 20 400 is the highest measurable value using the GC at 2 wt.% UDMH concentration. The results show that short crystallization time (1 h) is not enough to make an effective zeolite layer on the support (sample 5). Also, long crystallization time (5.5 h) causes NaA zeolite to transform to other zeolites such as NaX [3]. As a result, this sample (8) shows poor selectivity. The crystallization time in a range of 2.5–4 h was found to be very effective for making the NaA zeolite layer. The synthesis procedure was also performed using different temperatures. As seen in Table 2, increasing crystallization temperature increases flux (samples 1–3). Also, it can be observed that there is no change in separation factor. This may be due to the fact that at higher crystallization temperature, a thinner layer is formed. It is because, at higher temperatures,
Table 2 Flux and separation factor of NaA zeolite membranes (dead end) Sample
Number of coating
Crystallization time (h)
Crystallization temperature (◦ C)
Concentration of UDMH in feed (wt.%)
Flux (kg/m2 h)
Separation factor
1 2 3 4 5 6 7 8 9 10 11 12
1 1 1 1 1 1 1 1 1 1 2 3
3 3 3 3 1 2.5 4 5.5 3 3 3 3
70 90 110 130 100 100 100 100 110 100 100 100
2 2 2 2 2 2 2 2 5 5 5 5
0.0654 0.0708 0.0871 0.4685 1.1400 0.6200 0.2900 0.3600 0.0940 0.4437 0.3840 0.3243
20400 20400 20400 20400 41 20400 20400 2 52600 52600 52600 52600
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
177
Table 3 Flux of cross-flow measurements (sample 12) Run
Concentration of UDMH in feed (wt.%)
P (bar)
Q (l/min)
T (◦ C)
Flux (kg/m2 h)
1 2 3 4 5
5 5 5 5 5
2 2 2 2 2
3 1.5 0.5 3 3
55 55 55 25 40
3.95 3.61 3.30 2.33 3.44
the support to be formed after a two or three stage synthesis. Therefore, quality of NaA zeolite membrane layer may be improved. The synthesized layers by a multi stage procedure are substantially thicker than the corresponding layers by a single stage. As a result, flux is less for the three stage sample (12) as shown in Table 2. The experiments were carried out using two different concentrations. The results show that increasing UDMH concentration in feed increases separation factor and flux as shown in Table 2 (samples 3 and 9). It must also be mentioned that 52000 is the highest measurable value using the GC at wt.% UDMH concentration. Cross-flow PV experiments were carried out and the results were presented in Table 3. As seen, increasing flow rate increases flux (runs 1–3). This can be due to the fact that higher flow rates diminish effect of boundary layer. Also, increasing temperature increases flux (runs 1, 4 and 5). This can be due to enhancement of adsorption and diffusion of water to and through the membrane. The results are comparable with (or even better than) those of other simple water/organic mixtures in the literature [14,15]. Fig. 2 shows XRD patterns of the mullite support (a) and the NaA zeolite membrane (b). The XRD pattern of NaA zeolite membrane confirms that zeolite NaA crystals were formed. Fig. 3 shows SEM photographs of the mullite support (a), the NaA zeolite membrane (surface) (b) and the NaA zeolite membrane (cross-section) (c). Porous structure of the support and thin layer of the membrane can be easily observed. The results confirm that NaA zeolite membranes synthesized at 130 ◦ C for 3 h via a single stage process can be recommended for dehydration of aqueous UDMH dilute mixtures. Fig. 3. SEM micrograph of (a) the support, (b) the membrane (surface) and (c) the membrane (cross-section).
4. Conclusion NaA zeolite crystals are smaller. This shows that these membranes behave very high selectivity. The results show that at a high temperature of 130 ◦ C, zeolite NaA is also well formed much thinner (sample 4). The crystallization temperature in a range of 70–130 ◦ C was found to be very effective for making the NaA zeolite layer. Synthesis procedure was repeated where fresh synthesis gel was added periodically to the cooled and cleaned samples (samples 10–12). This causes more NaA zeolite crystals on
NaA zeolite membrane was firstly used for dehydration of aqueous UDMH mixtures. Zeolite NaA membranes were synthesized on the porous mullite tubes by hydrothermal method. The mullite supports were made by extruding kaolin clay. Zeolite membranes showed much higher fluxes and selectivities than commercially available polymeric membranes due to their uniform nanopores. These membranes showed very good membrane performance for separation of
178
M. Kazemimoghadam, T. Mohammadi / Separation and Purification Technology 47 (2006) 173–178
UDMH/water mixtures. It is expected that even significantly higher fluxes, with similar selectivities, can be achieved at higher temperatures. Since zeolite NaA membranes can withstand high temperatures and harsh environments (pH > 12), dehydration of UDMH mixtures with high pH can be performed. It is expected that PV using these membranes can be a highly interesting tool for industry, provided they can be produced cheap at a large scale. PV through a NaA zeolite membrane is an effective technique to separate UDMH from water. Separation factors as high as 52 000 was obtained at wt.% UDMH concentration.
Acknowledgments Authors would like to appreciate Industrial Development and Renovation Organization (IDRO) of Iran for the support.
References [1] J. Jafar, M. Budd, Separation of alcohol/water mixtures by pervaporation through zeolite NaA membranes, Microp. Mater. 12 (1997) 305–311. [2] Q. Liu, R. Noble, L. Falconer, H. Funke, Organic/water separation by pervaporation with a zeolite membrane, J. Membr. Sci. 117 (1996) 163–174. [3] D. Shah, K. Kissick, A. Ghorpade, R. Hannah, D. Bhattacharyya, Pervaporation of alcohol–water and dimethylformamide–water mixtures using hydrophilic zeolite NaA membranes: mechanisms and experimental results, J. Membr. Sci. 179 (2000) 185–205. [4] A. Verkerk, P. van Male, M.A.G. Vorstman, J.T.F. Keurentjes, Properties of high flux ceramic pervaporation membranes for dehydration of alcohol/water mixtures, Sep. Purif. Technol. 22–23 (2001) 689–695. [5] R. Ravindra, S. Sridhar, A.A. Khan, A.K. Rao, Pervaporation of water, hydrazine and monomethylhydrazine using ethylcellulose membranes, Polymer 41 (2000) 2795–2806. [6] S. Sridhar, R. Ravindra, A.A. Khan, Recovery of monomethylhydrazine liquid propellant by pervaporation technique, Ind. Eng. Chem. Res. 39 (2001) 2485-L2490. [7] Li. Xin-Gui, I. Kresse, Z.K.J. Zhou, J. Springer, Effect of temperature and pressure on gas transport in ethyl cellulose membrane, Polymer 42 (2001) 6801–6810.
[8] R. Ravindra, K.R. Krovvidi, A.A. Khan, A.K. Rao, D.S.C. studies of states of water, hydrazine and hydrazine hydrate in ethyl cellulose membrane, Polymer 40 (1999) 1159– 1165. [9] R. Ravindra, A. Kameswara, A. Khan, A qualitative evaluation of water and monomethyl hydrazine in ethyl cellulose membrane, J. Appl. Polym. Sci. 72 (1999) 689–700. [10] S. Sridhar, G. Susheela, G.J. Reddy, A.A. Khan, Cross-linked chitosan membranes: characterization and study of dimethylhydrazine dehydration by pervaporation, Polym. Int. 50 (2001) 1156–1165. [11] R. Ravindra, A.K. Rao, A.A. Khan, Processing of liquid propellant reaction liquors by pervaporation, J. Appl. Polym. Sci. 72 (1999) 141–149. [12] R. Ravindra, S. Sridhar, A.A. Khan, Separation studies of hydrazine from aqueous solutions by pervaporation, J. Polym. Sci., Part B: Polym. Phys. 37 (1999) 1969–1980. [13] R. Ravindra, R. Krovvidi, A. Khan, Solubility parameter of chitin and chitosan, Carbohydr. Polym. 36 (1998) 121–127. [14] M. Kondo, M. Komori, H. Kita, K.I. Okamoto, Tubular-type pervaporation module with zeolite NaA membrane, J. Membr. Sci. 133 (1997) 133–141. [15] Y. Morigami, M. Kondo, J. Abe, H. Kita, K.I. Okamoto, The first large-scale pervaporation plant using tubular-type module with zeolite NaA membrane, Sep. Purif. Technol. 25 (2001) 251– 260. [16] T. Gallego-Lizon, Y.S. Ho, L.F.D. Santos, Comparative study of commercially available polymeric and microporous silica membranes for the dehydration of IPA/water mixtures by pervaporation/vapour permeation, Desalination 149 (2002) 3–8. [17] T. Gallego-Lizon, E. Edwards, G. Lobiundo, L.F.D. Santos, Dehydration of water/t-butanol mixtures by pervaporation: comparative study of commercially available polymeric, microporous silica and zeolite membranes, J. Membr. Sci. 197 (2002) 309– 319. [18] W.C. Wong, L.T.Y. Au, C.T. Ariso, K.L. Yeung, Effects of synthesis parameters on the zeolite membrane growth, J. Membr. Sci. 191 (2001) 143–163. [19] T. Mohammadi, A. Pak, Effect of calcination temperature of kaolin as a support for zeolite membranes, Sep. Purif. Technol. 30 (2003) 241–249. [20] M. Kazemimoghadam, A. Pak, T. Mohammadi, Dehydration of water/1-1-dimethylhydrazine mixtures by zeolite membranes, Microp. Mesop. Mater. 70 (2004) 127–134. [21] T. Mohammadi, A. Pak, Making zeolite NaA membrane from kaolin by electrophoresis, Microp. Mesop. Mater. 56 (2002) 81–88. [22] H. Robson, R.W. Thompson, K.C. Franklin, Verified Synthesis of Zeolitic Materials, Elsevier Publisher, 2001.