Newest Achievements In Pervaporation Biology Essay


The literature overview presents the newest achievements in pervaporation and comparison different types of pervaporation membranes in separation water and ethanol mixture. Also this literature review includes a brief theoretical basis of membrane processes and describes recent membrane market.

After fermentation in the production of bioethanol, the next step is dehydration of aqueous mixture 5 - 12 wt% of ethanol, which is called 'broth' or 'beer' [1]. To separate this ethanol-water mixture in bioethanol production manufactories is usually used two stages:

The first stage is conventional distillation, which is involved dehydration ethanol from 10 wt% to 92.4wt% ethanol. However further distillation, up than 92.4 wt%, became impossible because of approaching azeotrope point (at 95.6wt%), which makes simple distillation nonsensical as the azeotrope boiling temperature lower than boiling temperature of pure ethanol [1]. Breaking azeotrope is one of the main problems in bioethanol industry, by David P. Brayant [2] to purify bioethanol from aqueous mixture by one stage distillation is possible until 95 %. The simple diagram of producing 95 wt% bioethanol from biomass is presented in figure 2.1.

Lady using a tablet
Lady using a tablet


Essay Writers

Lady Using Tablet

Get your grade
or your money back

using our Essay Writing Service!

Essay Writing Service

2.1. The simple scheme of producing 95 wt% aqueous bioethanol mixture.

The second stage (producing anhydrous bioethanol) is basically more power-intensive process. To produce anhydrous alcohol it is necessary to add extra modification into distillation process and it usually consumes 40% - 60% of whole plant energy consumption [2]. Therefore at this step of dehydration are used more sophisticated apparatus and methods than at the first step to reduce energy consumption and produce pure bioethanol. Basically, to separate water-ethanol mixture at azeotrope point (see figure 2.2) is used cyclohexane but the disadvantage of this method is cyclohexane will never release from alcohol. This drawback of cyclohexane as a breaker of azeotrope mixtures makes it unsuitable to produce alcohol of high purity [3]. Other the most widely used methods are azeotropic distillation (AD), different types of extraction distillation and pervaporation or other membrane processes [1].

Figure 2.2. Vapour-liquid equilibrium curve

To purify ethanol and make it cheaper in this thesis will be advised to use membrane separation process, which is so-called membrane pervaporation process.

2.1.2 Membrane market.

Membranes are one of the fast developing scientific fields in chemical industry. They have successfully been applied in dairy manufactures and food industry. Today membranes are widely used in water treatment and in medicine to help people, who are suffering with kidney disease [4].

In all industries where there are distillation, dehydration and evaporation, membranes could substitute or modify classical methods thereby reduce energy consumption. By a survey in 1986 the annual energy use in the USA, which was spent on dehydration, distillation and evaporation, formed 4.2 quads (1 quad approximately equal to 2,93Ã-1011 Kilowatt-hours (kWh)). This survey noted that by introduction of membranes into petrol, chemical and food industries and using them in the above-mentioned processes, it is possible to reduce expended energy more than 0.8 quads [5].

Today the sale of membrane is a very fast developing market, annual sales in 1998 amounted more than US$ 4.4 billion dollars [4]. The biggest consumers are: the USA around 45% of all sales in 1996 and 29% Europe and the Middle East. However the Asian and South American markets are rising rapidly and in the nearest future will present big part of membrane consumers [4]. Also it was noted that further grow of this market will continue and form 10%/year [4].

2.1.3 Types of membrane and membrane processes.

Generally all membrane processes could be explained like separation of a flow into two streams one of them is permeate and another one is retentate, a schematic diagram of membrane process is showed in fig. 2.3. Even it could be seen very simple process, the most important thing in a membrane processes is control of permeation and separation through a membrane [6]. Therefore the main goal of engineers and scientist is to achieve the best separation of a mixture, which will be economically profitable.

Figure 2.3. The basic circuit of the membrane process.

At the recent time, there are six main types of membrane processes: microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), reverse osmosis (RO), pervaporation (PV) and electrodialysis.

The first four of them microfiltration, ultrofiltration, nanofiltration and hyperfiltration (reverse osmosis) are developed and widely used by companies around the world. Microfiltration, ultrafiltration and nanofiltration differ from each other by the diameter of pores in the membrane, hence all three methods are based on filtration technique, which is consisted in separation dissolved or suspended solutes from a flow, by using porous membrane with different diameters. The forth type, reverse osmosis, uses membranes, which virtually do not have pores, because diameter of pores 3 - 5 angstrom (1 angstrom equal to 10-10 meters), thus the driving force of the reverse osmosis is differences in solubility and mobility of solutes [5, 6]. Basically electrodialysis was designed to obtain fresh water by desalination of sea salt water by using as a drive force electrical potential, today this process is widely used in the food industry, for example to deionise cheese whey [6]. However to separate azeotropic ethanol-water mixture all these types of membrane cannot be used, hence they will not be discussed in this literature review.

Lady using a tablet
Lady using a tablet


Writing Services

Lady Using Tablet

Always on Time

Marked to Standard

Order Now

Although pervaporation is still developing process, the first time when was published the term of pervaporation was in 1917 in the Journal of the American Chemical Society [7]. Nowadays it has huge potential in dehydration of ethanol-water mixture and breaking azeotrope, the latter one makes big energy consumptions by using distillation processes. Basically pervaporation based on separation a liquid mixture by one component selective membrane, where the driving force is differential pressure between upstream flow (retantate side) and downstream flow (permeate side). The advantage of pervaporation over conventional distillation is ability to separate close bowling mixture or azeotrope mixtures. In the pervaporation process a membrane represents the third phase, which selectively passes water molecules through it, hence boiling point of azeotropic mixture does not influence on separation.

Pervaporation process.

Basically principle scheme of pervaporation seems like other membrane processes (see figure 2.4), however difference from the latter ones is that one component of water-organic or organic-organic mixture preferentially extracted through a nonporous permeate selective membrane by partial vaporization. In the pervaporation process a feed is in liquid phase contacts with membrane's active layer, while on the opposite side of the membrane there is a vacuum, which makes driving force through membrane mass. As soon as molecules of permeate are approached the opposite side of the membrane, permeate evaporates with the vacuum help [8]. Whole process could be described in tree steps:

Preferential sorption of a required component by selective membrane

Diffusion of the fastest component through the membrane

Desorption of permeate and vaporization by vacuum.

Figure 2.4. The basic circuit of the membrane process.

There are two main parameters which describe pervaporation process. The first parameter, which shows permeation ratio, can be defined by the following equation


Where, - permeation flux (), - the mass of the permeate (), - the area of the membrane in contact with the feed mixture (), - the permeation time ().

The second parameter shows selectivity of a membrane and efficiency of separation, it can be also found by equation 2-2 [9]:


where, - dimensionless separation factor, and - the mass fractions of water and ethanol in the permeate side, respectively, and - the mass fractions of water and ethanol in the feed side, respectively.

As higher a separation factor of a membrane, so better separation could be achieved in a pervaporation process, hence superselectivity can be at [3]. Basically, in the pervaporation process there is inverse proportionality between flux ratio (J) and separation factor (α), so-called 'trade off' phenomenon, however, some membranes show exception of regularity [10, 11]. By R.Y.M. Huang et al. [12] trade-off phenomenon is not common for absolutely all types membrane but, basically, membranes without strong polar interaction show this regularity.

Variation of the flux and separation factor to produce the best membrane makes very hard to find good membrane with excellent pervaporation properties. Therefore scientists have introduced new equation (see equation 2-3) for comparison different membranes, which is so-called 'pervaporation separation factor' (PSI) [13].


Although a lot of scientists are using this equation to examine different types of membrane, by Peter D. Chapman this equation has one disadvantage that the same PSI could be achieved either by membrane with law separation factor and high permeation rate or membrane with high separation factor and law permeation rate. Therefore, in some cases the highest PSI does not indicate a membrane with the best pervaporation properties.

Experimental data of temperature dependence of total permeation flux generally exhibits an Arrhenius relationship [13]:


where is permeation flux, R - gas constant, T - operating temperature, and are the pre-exponential factor and the apparent activation energy of permeation, respectively, at the same time the latter described by the following equation [12]:


where and are the activation energy of diffusion and the enthalpy of sorption if the permeant in the membrane, respectively.

Another important parameter, which can make affect on pervaporation properties, is swelling degree, usually it is very hard to predict the swelling of membrane at a different process conditions. Swelling degree is very important as swelling of a membrane exert changes in flux and separation factor, the equation to determine the swelling degree is 2-5.

Lady using a tablet
Lady using a tablet

This Essay is

a Student's Work

Lady Using Tablet

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Examples of our work


Membranes' over swelling lead to increasing permeation flux of both components, therefore separation factor of membrane could be decreased dramatically. Although that over swelling is not good for membrane, un-swollen active layer and/or too much dry underneath support layer cause membrane distortion and as the result membrane lose own separation properties [14]. Therefore it is very important to control swelling degree of a membrane by adding crosslinking element into membrane, hence polymer became more packed. By other words crosslinking reduce molecular chains' movement and free volumes in a membrane.

Membranes for pervaporation process.

For the last two decades there were written a lot of scientific articles and books about pervaporation membranes. Different types of studies have investigated the next following parameters, which required membranes to be used in industry [15]:

Economical profitability of producing membranes;

Damage and leaking resistance;

Improving efficiency of separation and reducing energy waste;

Increasing of permeation through a membrane;

Be easy to clean.

Satisfaction to all these parameters makes a membrane high effective in pervaporation process. However, to produce a good membrane, which will content to all above requirements, it depends to a large extent on raw material for a membrane. Classification of membranes by type of materials.

In pervaporation process there are three types of membrane which are differed from each other by type of materials. The first type is organic membranes usually they are made from polymers; they became widely used after developing polymer production industry. The second type is inorganic membranes, they are much expensive than polymers membranes, but have better properties. The third type is inorganic-organic membranes, which are so-called hybrid membranes (or composite membrane). Basically hybrid membranes seem like polymer base with small inorganic particles inside the mass or inorganic porous base with polymer selective layer on top of it [3, 15].

Ι. Inorganic membranes.

Two main inorganic materials zeolite and ceramic [16] provide membrane with high permeability (flux ratio) and separation factor, because of these properties they are widely used to produce inorganic membranes. As a result the pervaporation properties of inorganic membranes are quite good to use them in industry and make commercial membranes. However, the disadvantages of inorganic membranes are high price of feedstock for this type of membrane and physical weakness.

In own article about pervaporation alcohol-water and organic aqueous mixtures D. Shah et al. [16] tested commercial NaA-zeolite membranes and in detail investigated mechanism of pervaporation through NaA-zeolite membranes. The pervaporation process was under following temperature conditions 40oC, 60oC and 70oC and feed alcohol concentration varied from 0 wt% to 100 wt%. Therefore it was found that during the pervaporation of ethanol-water mixture (from 0 wt% to 70 wt%) at 60oC the permeation flux almost did not vary, by authors' opinion it is because of high hydrophilicity of the zeolite membranes. In spite of the high water affinity of NaA-zeolite membranes, flux reduced after increasing alcohol concentration up to 100 wt%. D. Shah et al. supposed that raised solvent concentration reduces water activity and as a result chemical potential reduces as well. In addition authors explained zeolite structure, sorption mechanism and movement through the membranes by dipole moment and molecules interaction. By D. Shah et al. molecular size is not complete for the best separation through zeolite membranes and as example of this theory authors compared two different mixtures one of them is ethanol-water and another one is dimethylformamide (DMF)-water. Despite the fact that molecules of DMF bigger than ethanol, molecules of dimethylformamide crossed through the membranes faster than ethanol's. Scientists described this behaviour by high polar activity of dimethylformamide molecules, which makes a contention to water molecules, hence water flux reduced. Thus, separation factor of the NaA-zeolite membranes for dimethylformamide-water mixture is lower than for ethanol-water. To support own assumption scientists compared water sorption during the pervaporation of the same concentration ethanol-water and DMF-water mixtures, hence it was found that water sorption in the latter mixture 30% lesser than ethanol-water mixture. Another interesting point is that inside of zeolite membranes there are non-zeolitic pores, which are similar to polymer membranes' free volumes, but by D. Shah et al. these pores are not properly studied. In addition scientists compared these NaA-zeolite membranes with PVA membrane, so it was found that flux and separation characteristics of PVA membranes are much less than NaA-zeolite membranes (table 1).

Table 1

Comparison of selected solvent/water PV results for zeolite and PVA membranes [16].



Feed solvent (wt%)

Temperature (oC)

Separation factor

Permeate (solvent) concentration (wt%)

Total flux (kg/m2h)















Another study about the zeolite membrane was recently written by Y. Hasegava et al. [17], where scientists in their article about NaA-type zeolite membranes investigated 'the influence of the acid' on membrane performance. All changes in separation of water-ethanol mixture were monitoring in a real-time by a mass spectrometer. In own experiments Y. Hasegawa et al. used acetic acid as an addition to the ethanol-water mixture, the value of acid varied from 0.1 ml - 1.0 ml. Therefore it was observed that acetic acid destroys NaA-zeolite membranes and as a result the fluxes of ethanol and water through the membranes increased significantly but separation factor dramatically dropped. Whole 100 minute experiment was conditionally divided into three steps, where at the first step flux decreased immediately, then on the next step slightly raised and at the last step membrane was not able to separate mixtures, as separation factor sharply dropped. After the experiment the membrane was totally destroyed, hence it lost its separation ability.

Robert W. Van Gemert and F. Petrus Cuperus [18] investigated performance of ceramic membrane by pervaporation separation following mixtures: methanol-water, ethanol-water, and 2-propanol-water. The main aim of their research was to create the first satisfying dense membrane, which will be sufficient for industrial applications. By authors these new ceramic membranes, which were based on γ-Alumina with deposited silica by sol-gel technique showed readiness to separate different types of mixtures. The membranes had been testing for a period of 3 months and during this time flux and separation factor did not change significantly.

ΙΙ. Organic membranes.

During the literature research, it was found that main materials to produce organic membranes are: poly(vinyl)alcohol, chitosan, different types of alginates, polyimides and polyamids. All of them have very good water selective properties, which are very important for ethanol-water separation. However the main problem of these materials and membranes, which were made from them, is over swelling. In this case different scientists tried to modify membrane structure by adding some special composites, which can restrain the swelling.

Jing Ma et al. [14] created and examined chitosan/polyacrylonitrile membranes with added mussel adhesive-mimetic molecule, carbopol. By Jing Ma et al. carbopol increased pervaporation properties of chitosan/polyacrylonitrile membranes by increasing separation factor. It was interesting to notice that to prepare this membrane was used layer-by-layer technique, hence, majority of carbopol molecules were located in intermediate layer between chitosan and polyacrylonitrile. However with increasing carbopol content in the membrane flux decreases, by the authors' opinion the reason of flux decrease is influence of carbopol on thickness of intermediate layer. Another interesting fact that is these membranes showed 'anti-trade-off' behaviour but this result was achieved while pervaporation of 90 wt% ethanol-water mixture. Though membranes broke usual 'trade-off', with increasing water content in the feed side swelling degree enhanced and caused reduction of separation factor. Thus, with increasing water content membranes proved general behaviour, by other words with increasing flux separation factor decreases. But the most interesting fact that during 150 hours of pervaporation membrane's properties reduced slightly and it was found any mechanical defects, hence these membranes showed good long-term pervaporation abilities.

P. Kanti et al. [19] tested crosslinked chitosan and sodium alginate membrane by pervaporation of ethanol-water mixtures. In contrast to the research of Jing Ma et al., where chitosan and alginate presented independent layers, in this study membranes were made by blending deacetylated chitosan with sodium alginate and further this solution was crosslinked by addition glutaraldehyde to reduce excessive swelling. P. Kanti et al. investigated water feed concentration influence on membrane pervaporation properties, hence it has been found that with increasing water content in the feed from 5 wt% to 45 wt%, the flux increased from 0.55 kg/m2h to 2.12 kg/m2h, while the selectivity sharply dropped from 436.3 to 16.6. Also the authors examined influence of membrane thickness to pervaporation conditions, thus, it was mentioned that significantly dropped from 0.22 kg/m2h to 0.07 kg/m2h, while thickness increased from 25 μm to 190 μm. However selectivity increased from 436.3 to 2118.5 with an increase at the same thickness. The last point, which P. Kanti et al. described in the article is influence of permeate pressure on membrane characteristics; hence, it was found that with increasing permeate pressure pervaporation flux and selectivity decreased.

R. Jiraratananon et al. [13] investigated chitosan/hydroxylethylcellulose (CS/HEC) composite hydrophilic membranes based on porous cellulose acetate. They found that porous cellulose acetate improved chitosan/hydroxylethylcellulose membranes performance during pervaporation of ethanol-water mixture. As a result, for 90 wt% and 95 wt% ethanol aqueous mixture these membranes showed higher flux and separation factor than dense chitosan/hydroxylethylcellulose at the same alcohol concentrations and temperature 60oC. For example for dense CS/HEC membrane at 90 wt% ethanol-water mixture and temperature 60oC flux and separation factor were 112 g/m2h and 10491. However, the same parameters for CS/HEC on top of porous acetate cellulose showed 424 g/m2h - of the membrane and 5469 - separation factor. To compare these two types of membranes R. Jiraratananon et al. used pervaporation separation index (PSI), which is showed in equation 2-3. According to this equation dense CS/HEC has PSI equal to 1,178,278 but CH/HEC-AC's PSI is 2,332,188. However, by R. Jiraratananon et al. other chitosan membranes, which were made by other scientists, showed much lower PSI. Also authors described influence of temperature, permeate pressure and water feed concentration on separation conditions, hence it was found that these CS/HEC membrane based on porous cellulose acetate, showed general 'trade-off'. Although these membranes showed usual 'trade-off' behaviour, separation factor and flux was very high at high alcohol concentration and 60oC process temperature.

Tcrosslinking = 80 min Jp = 700 α ≈ 315, hence according to equation 2-2 PSI = 219800

Tcrosslinking = 85 min Jp ≈ 630 α = 400, hence according to equation 2-2 PSI = 251370

The author also mentioned information about influence of temperature and ethanol feed content but did not take into account swelling degree. In the second part of this research [20] authors continued investigation of properties chitosan/hydroxylethylcellulose (CS/HEC) composite membranes, the main point of this research was studying of mass transport through these membranes and also affects of process conditions on membranes pervaporation properties. Scientists by mathematical equation proved that water molecules can be selectively extracted through membranes; however ethanol has negative resistance, by R. Jiraratananon et al. it means that ethanol molecules' '...transport of ethanol was in opposite direction to water' [20].

R.Y.M. Huang et al. [12] also investigated properties of new alginate/chitosan membranes based on top of polyvinylidene fluoride (PVDF). Alginate membranes show strong hydrophilic properties, however there are two main problems of alginate membranes they are water solubility and mechanical weakness. Therefore, scientists investigated possibility of membrane improvement by adding other materials. The over water solubility of alginate layer was prevented by ionic crosslinking. Mechanical properties are ameliorated by using chitosan layer and make membranes by casting method layer after layer. The experiment was going under 50oC and concentration of ethanol in the feed varied from 50 wt% - 95 wt%. During the study there were investigated different modifications of alginate/chitosan membranes and their ability to separate ethanol-water and isopropanol-water mixtures under different temperature conditions. In own research R.Y.M. Huang et al. examined swelling degree of the membranes and then after swelling experiments scientists compared sodium alginate membrane with alginate/chitosan. Therefore, it was found that swelling degree of latter one was lower than sodium alginate membrane at high alcohol concentrations, by authors' opinion, better water suppress due to strong hydrogen bonding of alginate/chitosan membrane.

R.Y.M. Huang et al. [21] examined sodium alginate membranes, which were prepared for dehydration of ethanol-water and isopropanol-water mixtures, by pervaporation process. The authors investigated that alginate membrane, which was crosslinked by Ca2+ ions, exhibited the best relationship between flux and separation factor compare to alginate membranes with another divalent and trivalent ions like Zn2+, Mn2+, Co2+, Fe2+ and Al3+. R.Y.M. Huang et al. noticed that the highest separation factor was achieved for 90 wt% ethanol-water and isopropanol-water with the same concentration at 50. However, authors compared calcium alginate membrane with another one, which was crosslinked with Na+ ions. To investigate the best membrane researchers varied two operating parameters the first is water content in the feed from 5 wt% to 30 wt% and the second is temperature of the process from 40oC to 70oC. Sodium alginate membrane showed higher pervaporation performance than calcium except the pervaporation of 95 wt% ethanol-water and isopropanol-water mixtures mixture. It is very interesting to notice that for sodium alginate membrane with increasing in water content in the feed, concentration of permeate water increases. By R.Y.M. Huang et al. this behaviour of the membrane cannot be described by usual swelling theory, because with increasing swelling degree flux of ethanol would increase as a result separation factor would decrease. Therefore, authors supposed that this phenomenon could be explained by strong communication between water molecules in the feed and sodium acetate and hydroxyl groups in the alginate membrane. Another interesting fact that these Ca2+ and Na+ alginate membranes showed "anti-trade-off" behaviour, where with increasing in temperature separation and flux increase, by R.Y.M. Huang et al. it could be explained by relaxation behaviour of a polymer membrane.

Recently, Wulin Qiu et al. [9] tested to different membranes, which were made from two types of commercial polyimide, the first one was Matrimid and the second was Torlon. The aim of this research was to see how these polyimide membranes separate ethanol-water mixture by varying temperature and feed alcohol concentration. It is very interesting to notice that Wulin Qiu et al. tested polyimide membranes for a long term period of time at different alcohol concentrations, hence, these membranes showed increase in flux and decrease in separation factor or so-called general 'trade-off'. At first time both types of polyimide membranes showed very low separation factor, for instance, membranes which were made from Matrimid at 85 wt% ethanol-water mixture these membranes' separation factor was around three. Low separation factor of these polyimide membranes could be explained by plasticization process of the membranes, by John D. Wind et al. [22], basically, glassy polymeric membranes' plasticization prolongs chain relaxation, hence pervaporation flux increases and separation factor decreases. Therefore, to improve pervaporation properties and reduce plasticization of polyimide membranes, it was decide to annealing hollow fibre membranes and as a result separation performance increased significantly. However, temperature and time of annealing process is very important, for example the best temperature and time conditions for Matrimid membrane is 260oC and 5 hours, respectively. At these conditions this hollow fibber membrane achieved the best separation factor, which accordingly equal to around 200. There were three different temperatures of pervaporation process 30oC, 40oC and 75oC and it was found that at 30oC the separation factor was the highest, where after further increasing of temperature the separation factor decreased. By Wulin Qiu et al. decrease in separation factor was explained like with temperature increase chains movement and free volume of membrane also increase, hence, water and ethanol fluxes increased. To investigate membrane stability scientists made four day pervaporation test for Matrimid and Torlon membranes with 95 wt% and 15 wt% ethanol-water mixtures. As a result it was found that

Matrimid membranes showed dramatic drop in separation factor at 95 wt% alcohol- water mixture from 240 at the beginning of the experiment to 10 after three days, authors explained this radical change due to fast plasticization of the membranes. However, during the pervaporation of 15 wt% ethanol-water mixture membrane showed more stable separation performances.

Torlon membrane showed better separation factor at higher ethanol concentration but also like Matrimid reduced separation factor dramatically in three days. Reduction in separation factor in this case also was explained by plasticization of polymer.

During this research it was found that the Matrimid polyimide membranes were plasticized by ethanol but Torlon membranes showed high suppression the membrane swelling. Despite the fact that in this study explained hydrophilicity and hydrophobicity of membrane and also temperature influence, these membranes need further investigation, for example influence of crosslinking on membranes properties.

Another article about pervaporation of alcohol-water mixture through polyimide membrane was written by Jeong-Hoon Kim et al. [23], where scientists dehydrated ethanol-water mixture through polyimide membrane. In this research Jeong-Hoon Kim et al. used different diamines as monomers for polyimide membranes, such as hexamethylene diamine (HXDA), ethylene diamine (EDA) and m-phenylene diamine (m-PDA). Also polyimide membranes were based on polysulfone porous ultrafiltration membrane, which was as a support layer, but further experiments included membranes without polysulfone support. During pervaporation test scientists investigated influence of imidization temperature on properties of membranes. Therefore, it was found that polyimide membranes, which were made from EDA and HXDA, had higher flux and separation factor than m-PDA. Authors supposed that better flux and separation factor could be explained by physical structure rather than chemical structure. In case of EDA and HSDA, EDA showed higher flux and lower separation factor than HXDA, this situation was clarified by high hydrophilicity of EDA than HXDA. In addition, author mentioned that with increasing imidization temperature of polyimide membranes higher than 150oC, flux and separation factor decreased, for EDA and HXDA separation factor dramatically dropped.

2.2.3. Hybrid membranes.

Nowadays hybrid membranes became more popular, as this kind of membranes has relatively lower price than inorganic membranes, while has better pervaporation and mechanical properties than polymer membranes.

Tadashi Uragami et al. [24] investigated membranes, which were made from organic and inorganic materials. Organic part of the membrane was presented by chitosan (q-Chito) and inorganic part of the membrane was presented by tetraethoxysilane (TEOS). During this research Tadashi Uragami et al. found that q-Chito membranes with TEOS showed higher pervaporation properties than pure q-Chito membranes. By the authors better water/ethanol selectivity was achieved by cross linked structure of the membrane, therefore swelling degree of the organic inorganic membrane decreased. However it was noticed that, with increasing in TOES content more than 45 mol%, water/ethanol selectivity of the membranes decreased significantly, hence selectivity was lower than q-Chito membranes. This membranes' behaviour was described by changing in structure of the membranes, where at the concentration of TEOS less than 45% structure was homogenise and at higher concentration of TEOS it was changed to heterogeneous.

Another work, where were studied organic-inorganic hybrid membranes was written by Dong Yang et al. [25]. Dong Yang et al. explored the influence of titanium oxide on an alginate membrane properties, such as flux and separation factor. It is interesting to note, that under the same conditions CS/TiO2 membranes made by situ sol-gel technology showed higher flux and separation factor than CS/TiO2 hybrid membranes made by blending method. The optimal concentration of titanium oxide is 6 wt%, at which membranes showed average permeation flux and separation factor 0.340 kg/m2h and 196 for 90 wt% ethanol-water mixture. Just increasing the content of titanium oxide by more than 6 wt% increases the swelling of the membrane, thereby reducing the separation factor and increasing permeation flux. Also the authors mentioned information about influence of the pervaporation temperature on permeation flux and separation factor, hence with increasing in temperature flux increases as well, however, separation factor decreases with rising in temperature. By Dong Yang et al. with rising temperature, chain movement increases, therefore free volume widens and as a result availability of passing bigger radius molecules of ethanol. Another interesting fact that was noticed by authors is that diffusivity of water and ethanol through membrane, where it was found that with temperature rising diffusion coefficient of ethanol grows faster than water's.

According to the aim of this study producing of membrane is not necessary. Therefore in this research will be used commercial organic membrane based on poly(vinyl)alcohol and polyimide, which was made by Heggemann company. Nowadays studies in polymer and membrane industry allow us to produce good organic membranes with acceptable pervaporation properties. This organic membrane has two advantages, which will be important for manufactory use, like ability to operate at temperature up to 150oC and pressure 5 Barr and relatively cheaper price. High water selective poly(vinyl)alcohol layer was crosslinked by polyimide to avoid an overswelling of the membrane. Classification of membranes by feed mixtures.

To choose membrane for pervaporation process it is very important to take into account a compound of a feed and components content. In practise, there are two types of existed mixtures, which can be of interest for membrane pervaporation process: organic-water and organic-organic. However, aqueous organic mixtures are divided into two subgroups, where difference between the first subgroup and the second is occurred in varied concentration of organic part. Therefore, scientists assign free types of membrane: hydrophilic, hydrophobic and organophilic [8].

Ι. Hydrophilic membranes.

This type of membranes was found for dehydration of aqueous organic solution. Hydrophilic materials like poly(vinyl)alcohol or chitosan have high flux and separation factor but excessive hydrophilicity of this materials leads to its swelling. Despite the fact that with the swelling increases the flux of membrane, The main problem of all hydrophilic polymer membranes such as chitosan or poly(vinyl)alcohol [19, 21] is high swelling degree, which causes increasing velocity of chain movement and growing free volume in membrane. The disadvantage of excessive swelling is fast increasing of the flux of all components and as a result reduction of separation factor. In order to prevent excessive swelling of a membrane is used the method of cross-linking [10, 11, 19, 21, 24, 25]. Also different kinds of crosslinking agents provide hydrophilic membranes by more mechanical strength and better flux [8].

By P.A. Peters et al. [11], who examined organic-inorganic hybrid membranes based on poly(vinyl)alcohol and γ-Al2O3, some pervaporation membranes are '...contradicting the flux-selectivity paradigm'. In this article P.A. Peters et al. compared flux ratio and separation factor of each membrane by variation process conditions and types of alcohols, such as ethanol, 1-propanol, 2-prpopanol and 1-butanol. Thus, it was been found that during pervaporation of ethanol and 1-propanol aqueous mixtures at the same conditions membrane showed typical trade-off proportionality between the permeation rate and separation factor. It means with increasing temperature of the process the flux ration was increased and the separation factor was decreased. However, during of purification of 2-propanol-water and 1-butanol-water mixtures membranes showed anti-trade-off paradigm, where both the permeation rate and the selectivity of membranes were increased. The same phenomenon was observed by Qiu Gen Zhang et al. [10], who tested hybrid membranes based on top of the γ-amino propyltriethoxysilane (APTEOS) and poly(vinyl)alcohol (PVA). By the authors the novel PVA/APTEOS membrane showed increasing separation factor with increasing temperature of a process. The key point of this research was finding an optimal concentration of APTEOS to improve flux rate and separation factor, hence the best APTEOS content was consisted 5 wt% also it was mentioned that selectivity was decreased significantly, while concentration of APTEOS exceeded 5wt%. By Qiu Gen Zhang et al. sharply drop of separation factor could be explained by increasing in hydrophobicity of a membrane. In addition the authors wrote that permeation and separation of these PVA/APTEOS membranes depends on water feed concentration. With increasing water content in the feed was observed the usual trade-off.

Young Moo Lee et al. [26] examined crosslinked hydrophylic chitosan membranes and compared with poly(vinyl)alcohol membranes. Chitosan membranes were made by casting solution method based on top of a porous polyethersulfone ultrafiltration membrane. Crosslinking element was sulphuric acid, hence acid residue SO4-2 and NH3+ of chitosan molecule made strong ionic connection. In this article scientists figure out an optimal crosslinking time for chitosan membranes. It was found that ethanol-water mixture required much longer crosslinking time than isopropanol-water mixture. Young Moo Lee et al. described a reason of different crosslinking time by molar volumes of solutes, where the molar volume of isopropanol and ethanol constitute 76 cm3/mol and 41 cm3/mol, respectively. The result of this research was that chitosan membranes showed better pervaporation properties and less temperature-dependency than poly(vinyl)alcohol. Also it is very interesting to notice that at 80oC and 90oC, poly(vinyl)alcohol membranes lager changes in free volume than chitosan membranes.

ΙΙ. Hydrophobic membranes in pervaporation membrane processes.

In addition to the ongoing processes of pervaporation by hydrophilic membranes, there are some studies of pervaporation processes, which are used hydrophobic membranes. This type of membranes is widely used to extract organic component from aqueous mixture. In bioethanol industry hydrophobic membranes are widely used at the stage of fermentation to extract alcohol from fermentation 'broth' [27-29]. Membrane for this process is made from hydrophobic materials, which are organic selective. It is not common process but developing hydrophobic polymer membranes cause new opportunities in front of fermentation/pervaporation process.

Leland M Vane [27] in own review about product recovery from fermentation broth, author identified and discussed possible changes in the process to make it more competitive performance in comparison with distillation. He observed the following parameters of pervaporation, which are very important for adequate comparison pervaporation with other types of biofuels recovery systems. The first parameter is energy efficiency theoretically membrane process should be less energy-consuming process. The cost of pervaporation system and integration of pervaporation unit with fermentation process are very important as to be competitive it is very important to be economically profitable. By the Leland M Vane using pervaporation process for extracting ethanol from fermentation broth has several problems to be used instead of usual distillation processes. The first reason is energy efficiency, where the main problems are commercial hydrophobic membranes does not have high separation factor. The second reason is membranes still has low flux, hence it is required to use bigger membrane area but to reduce the capital cost it is very important to improve permeability of the membrane. The next reason is importance of using long run experiments with membranes to investigate stability of a membrane unit. Also author noticed that to use pervaporation membrane it is very important to use micro or ultra filtration as in fermentation broth there are solid particles like enzymes and microorganisms, which could reduce efficiency of the pervaporation membranes and foul up whole rig.

ΙΙΙ. Organophilic membranes in pervaporation process.

Organophilic membranes are developed and made to extract particular types of organic elements from organic-organic mixture. These membranes preferentially sorb indispensable kinds of molecules, while other molecules are rejected by membrane surface. The principle of separation the same as other types of pervaporation processes, where diffusion through a membrane going by vacuum on a permeate side. The common example of organic-organic separation is extraction of aromatic elements from aliphatics.

In the thesis will be investigated ethanol water separation, which is can be separated by two ways the first one is using hydrophilic membrane and another one is separation through hydrophobic membrane. However according to the searched materials there are not many good hydrophobic membranes, which has satisfactory flux ratio and selectivity. Big achievements in studying of hydrophilic membranes lead to use this type of membrane.

2.3. Pervaporation laboratory plant design.

During the literature search it was noticed that there are not a lot of narrow information about plant design, it is due to a commercial secretness of pervaporation plant. However each research, which was found, included simple plant scheme and operation conditions of a process.

The design of the pervaporation rig or plant does not vary significantly, as schematically they are very similar to each other. The main principium of pervaporation is separation of feed flow into two flows the first one is a renetate and other one is permeate. On the permeate side is required a vacuum pressure, to maintain a diffusion of permeate through membrane mass. The pressure on the permeate side should be maintained low than 10-20Torr (1Torr ≈1 mmHg) [16, 23]. Then due to evaporation of diffused component, there should be existed condenser which has to be enough effective to condense whole amount of vaporised permeate, basically as a condenser is used a cold trap. The retentate side usually recycled to a feed tank, hence it seems like batch operation system. The feed mixture should be heated by temperature controller and monitored by thermocouples. The temperature of the feed depends on membrane pervaporation properties, for example for ethanol-water mixture the temperature varies from 40 oC to 80oC. The principle scheme of pervaporation plant is shown in figure 2.5 [9, 16, 19, 21, 23, 30].

Figure 2.5 Principle scheme of pervaporation.

2.3.1. Membranes geometry.

Design of the pervaporation plant obviously depends on membranes geometry which divides into three main groups: planar, which is also so-called plate-and-frame, spiral wound and capillary [15, 31].

Ι. Planar membranes in membrane industry.

Basically the simplest type of membrane, by Robert H. Perry 'it is a very much like a filter press' [15]. Although schematically planar membranes' conceptually is very simple, their sealed construction requires a lot of sealants. A planar membrane is shown in figure 2.6a [15].

ΙΙ. Spiral wound membranes.

Another inexpensive membrane module is called spiral wound membrane. This type of membranes consist of permeate pipe, which has collection holes, and sheets, which are rounded of central pipe. Feed comes into a membrane from one of cylindrical membrane and leaves it through an opposite side, while permeate goes into permeate pipe and then collected from it. By Daniel Bernal, spiral wound membranes require competitively less sealed materials than planar membranes but they are very hard to clean. The spiral wound membrane is shown on the figure 2.4b [15].

ΙΙΙ. Hollow fibber membranes.

This type of membranes consist a lot of small capillaries, hence in some sources it is also called like capillary membranes. Usually a diameter of each capillary for pervaporation process varies from 250 µm up to 6 mm [15] and by Robert H. Perry 'there is no obvious limits for the future' for self-supported hollow fibber membranes (see figure 2.6c) [15].

a) b)


Figure 2.5. Different types of pervaporation polymer membranes classified by membrane geometry. (a) Planar membrane; (b) Hollow fibber membrane; (c) Spiral membrane [15].

In this research will be used commercial polymer hydrophilic membrane with hollow fibber geometry based on poly(vinyl)alcohol and polyimide. Crosslinking structure of a membrane will suppress swelling of membrane and hollow fibber geometry will allows easy installation and operation. In addition in this research there will not be a recycle loop, which is used to return permeate to a feed tank. Thus, by other words the pervaporation plant will represent a continuous operation system instead of usual batch system, hence all experiences will be required less time and will be higher efficient.

1. Huang, H.J., et al., A review of separation technologies in current and future biorefineries. Separation and Purification Technology, 2008. 62(1): p. 1-21.

2. Bryant, D.P. Bio-energy: A brief overview. in Application of Bioenergy Technologies. March 1996. Rotorua, New Zealand.

3. Chapman, P.D., et al., Membranes for the dehydration of solvents by pervaporation. Journal of Membrane Science, 2008. 318(1-2): p. 5-37.

4. S. P. Nunes and K. V. Peinemann, Membrane technology in the chemical industry. 2006, New York, the USA.

5. R.W. Baker, et al., Membrane separation systems recent developments and future directions. 1991, Park Ridge, N.J.

6. Najafpour, G.D., Membrane Separation Processes, in Biochemical Engineering and Biotechnology. 2007, Elsevier: Amsterdam. p. 351-389.

7. Huang, R.Y.M., Pervaporation membrane separation processes. 1991: Elsevier in Amsterdam, New York.

8. Li, N., Advanced membrane technology and applications. 2008: Wiley-Interscience.

9. Qiu, W., et al., Dehydration of ethanol-water mixtures using asymmetric hollow fiber membranes from commercial polyimides. Journal of Membrane Science, 2009. 327(1-2): p. 96-103.

10. Zhang, Q.G., et al., Anti-trade-off in dehydration of ethanol by novel PVA/APTEOS hybrid membranes. Journal of Membrane Science, 2007. 287(2): p. 237-245.

11. Peters, T.A., et al., Ceramic-supported thin PVA pervaporation membranes combining high flux and high selectivity; contradicting the flux-selectivity paradigm. Journal of Membrane Science, 2006. 276(1-2): p. 42-50.

12. Huang, R.Y.M., R. Pal, and G.Y. Moon, Pervaporation dehydration of aqueous ethanol and isopropanol mixtures through alginate/chitosan two ply composite membranes supported by poly(vinylidene fluoride) porous membrane. Journal of Membrane Science, 2000. 167(2): p. 275-289.

13. Jiraratananon, R., et al., Pervaporation dehydration of ethanol-water mixtures with chitosan/hydroxyethylcellulose (CS/HEC) composite membranes I. Effect of operating conditions. Journal of Membrane Science, 2002. 195(2): p. 143-151.

14. Ma, J., et al., Mussel-inspired fabrication of structurally stable chitosan/polyacrylonitrile composite membrane for pervaporation dehydration. Journal of Membrane Science, 2010. 348(1-2): p. 150-159.

15. Perry, R. and D. Green, Perry's chemical engineers' handbook. 1997: McGraw-Hill New York.

16. Shah, D., et al., Pervaporation of alcohol-water and dimethylformamide-water mixtures using hydrophilic zeolite NaA membranes: Mechanisms and experimental results. Journal of Membrane Science, 2000. 179(1-2): p. 185-205.

17. Hasegawa, Y., et al., Influence of acid on the permeation properties of NaA-type zeolite membranes. Journal of Membrane Science, 2010. 349(1-2): p. 189-194.

18. Van Gemert, R.W. and F. Petrus Cuperus, Newly developed ceramic membranes for dehydration and separation of organic mixtures by pervaporation. Journal of Membrane Science, 1995. 105(3): p. 287-291.

19. Kanti, P., et al., Dehydration of ethanol through blend membranes of chitosan and sodium alginate by pervaporation. Separation and Purification Technology, 2004. 40(3): p. 259-266.

20. Jiraratananon, R., A. Chanachai, and R.Y.M. Huang, Pervaporation dehydration of ethanol-water mixtures with chitosan/hydroxyethylcellulose (CS/HEC) composite membranes: II. Analysis of mass transport. Journal of Membrane Science, 2002. 199(1-2): p. 211-222.

21. Huang, R.Y.M., R. Pal, and G.Y. Moon, Characteristics of sodium alginate membranes for the pervaporation dehydration of ethanol-water and isopropanol-water mixtures. Journal of Membrane Science, 1999. 160(1): p. 101-113.

22. Wind, J., et al., Relaxation dynamics of CO2 diffusion, sorption, and polymer swelling for plasticized polyimide membranes. Macromolecules, 2003. 36(17): p. 6442-6448.

23. Kim, J.H., K.H. Lee, and S.Y. Kim, Pervaporation separation of water from ethanol through polyimide composite membranes. Journal of Membrane Science, 2000. 169(1): p. 81-93.

24. Uragami, T., et al., Dehydration of an ethanol/water azeotrope by novel organic - Inorganic hybrid membranes based on quaternized chitosan and tetraethoxysilane. Biomacromolecules, 2004. 5(4): p. 1567-1574.

25. Yang, D., et al., Chitosan/TiO<sub>2</sub> nanocomposite pervaporation membranes for ethanol dehydration. Chemical Engineering Science, 2009. 64(13): p. 3130-3137.

26. Lee, Y.M., S.Y. Nam, and D.J. Woo, Pervaporation of ionically surface crosslinked chitosan composite membranes for water-alcohol mixtures. Journal of Membrane Science, 1997. 133(1): p. 103-110.

27. Vane, L.M., A review of pervaporation for product recovery from biomass fermentation processes. Journal of Chemical Technology and Biotechnology, 2005. 80(6): p. 603-629.

28. O'Brien, D.J., L.H. Roth, and A.J. McAloon, Ethanol production by continuous fermentation-pervaporation: A preliminary economic analysis. Journal of Membrane Science, 2000. 166(1): p. 105-111.

29. Ikegami, T., et al., Bioethanol production by a coupled fermentation/pervaporation process using silicalite membranes coated with silicone rubbers.

30. Lipnizki, F., R.W. Field, and P.-K. Ten, Pervaporation-based hybrid process: a review of process design, applications and economics. Journal of Membrane Science, 1999. 153(2): p. 183-210.

31. Bernal, D., Optimization of Small Scale Ethanol Production, in DEPARTMENT OF CHEMICAL AND PETROLEUM ENGINEERING. 2009, UNIVERSITY OF CALGARY: Calgary, Alberta, Canada.