Introduction to Membrane Science and Technology

A membrane is a selective barrier that permits the separation of certain species in a fluid by
combination of sieving and sorption diffusion mechanism. Separation is achieved by
selectively passing (permeating) one or more components of a stream through the
membrane while retarding the passage of one or more other components. See the Figure.
Membranes can selectively separate components over a wide range of particle sizes and
molecular weights, from macromolecular materials such as starch and protein to
monovalent ions. Membranes have gained an important place in chemical technology and
are used in a broad range of applications.

The key properties determining membrane performance are high selectivity and fluxes,
good mechanical, chemical and thermal stability under operating conditions, low fouling
tendencies and good compatibility with the operating environment; and cost effective and
defect-free production.

Although the major uses of membranes are in the production of potable water and
separation of industrial gases, they can be used for many other important applications such
as filtration of particulate matter from liquid suspensions, air or industrial flue gas and the
dehydration of ethanol azeotropes.

More specialised applications include ion separation in electrochemical processes, dialysis of

blood and urine, artificial lungs, controlled release of therapeutic drugs, membrane-based
sensors, etc.

Membrane processes are characterized by the fact that a feed stream is divided into 2
streams: retentate and permeate. The most general process can be depicted by the
following Figure:

1
The retentate is that part of the feed that does not pass through the membrane, while the
permeate is that part of the feed that does pass through the membrane. The optional
"sweep" is a gas or liquid that is used to help remove the permeate. The component(s) of
interest in membrane separation is known as the solute. The solute can be retained on the
membrane and removed in the retentate or passed through the membrane in the permeate.

It is important to note that there are 3 different mechanisms by which membrane can
perform separations:

By having holes or pores which are of such a size that certain species can pass through
and others cannot. This mechanism is called size exclusion.

By selective retardation by the pores when the pore diameters are close to molecular
sizes. This mechanism is called pore flow.

By dissolution into the membrane, migration by molecular diffusion across the

membrane, and re-emergence from the other side. This is called solution diffusion.

Industrial Applications of Membrane Separation Processes

 Reverse Osmosis
o Desalination of brackish water

o Treatment of wastewater to remove impurities

o Treatment of surface and ground water

o Concentration of foodstuff

o Removal of alcohol from beer and wine

2
 Dialysis

o Separation of nickel sulfate from sulphuric acid

o Hemodialysis - removal of waste metabolites, excess body water, and

restoration of electrolyte balance in blood

 Electrodialysis

o Production of table salt from seawater

o Concentration of brines from reverse osmosis

o Treatment of wastewater from electroplating

o Demineralisation of cheese whey

o Production of ultrapure water for semiconductor industry

 Microfiltration

o Sterilisation of drugs

o Clarification of biological stabilisation of beverages

o Purifications of antibiotics

o Separation of mammalian cells from a liquid

 Ultrafiltration

o Pre-concentration of milk before making cheese

o Clarification of fruit juice

o Recovery of vaccines and antibiotics from fermentation broth

o Color removal from Kraft black liquor in paper-making

 Pervaporation

o Dehydration of ethanol-water azeotropes

o Removal of water from organic solvents

o Removal of organics from water

 Gas Permeation

3
 o Separation of CO2 or H2 from methane and other hydrocarbons

o Adjustment of the H2/CO ratio in synthesis gas

o Separation of air into nitrogen- and oxygen-enriched streams

o Recovery of helium

o Recovery of methane from biogas

 Liquid Membranes

o Recovery of zinc from wastewater

o Recovery of nickel from electroplating solutions

Advantages and Disadvantages, Challenges of Membrane Separations

Membrane processes have a number of pluses and minuses compared to alternative means
of performing separations.

The advantages include:

 Because membrane processes can separate at the molecular scale up to a scale at

which particles can actually be seen, this implies that a very large number of
separation needs might actually be met by membrane processes.
 Membrane processes generally do not require a phase change to make a separation
(with the exception of pervaporation). As a result, energy requirements will be low
unless a great deal of energy needs to be expended to increase the pressure of a
feed stream in order to drive the permeating component(s) across the membrane.

 Membrane processes present basically a very simple flowsheet. There are no moving
parts (except for pumps or compressors), no complex control schemes, and little
ancillary equipment compared to many other processes. As such, they can offer a
simple, east-to-operate, low maintenance process option.

 Membranes can be produced with extremely high selectivities for the components
to be separated. In general, the values of these selectivities are much higher than
typical values for relative volatility for distillation operations.

 Because of the fact that a very large number of polymers and inorganic media can be
used as membranes, there can be a great deal of control over separation selectivities.

 Membrane processes are able to recover minor but valuable components from a
main stream without substantial energy costs.

 Membrane processes are potentially better for the environment since the
membrane approach require the use of relatively simple and non-harmful materials.

4
The disadvantages include:

 Membrane processes seldom produce 2 pure products, that is, one of the 2 streams
is almost always contaminated with a minor amount of a second component. In
some cases, a product can only be concentrated as a retentate because of osmotic
pressure problems. In other cases the permeate stream can contain significant
amount of materials which one is trying to concentrate in the retentate because the
membrane selectivity is not infinite.
 Membrane processes cannot be easily staged compared to processes such as
distillation, and most often membrane processes have only one or sometimes two or
three stages. This means that the membrane being used for a given separation must
have much higher selectivities than would be necessary for relative volatilities in
distillation. Thus the trade-off is often high selectivity/few stages for membrane
processes versus low selectivity/many stages for other processes.

 Membranes can have chemical incompatibilities with process solutions. This is

especially the case in typical chemical industry solutions which can contain high
concentrations of various organic compounds. Against such solutions, many polymer-
based membranes (which comprise the majority of membrane materials used today),
can dissolve, or swell, or weaken to the extent that their lifetimes become
unacceptably short or their selectivities become unacceptably low.

 Membrane modules often cannot operate at much above room temperature. This is
again related to the fact that most membranes are polymer-based, and that a large
fraction of these polymers do not maintain their physical integrity at much above 100
o
C. This temperature limitation means that membrane processes in a number of
cases cannot be made compatible with chemical processes conditions very easily.

 Membrane processes often do not scale up very well to accept massive stream sizes.
Membrane processes typically consist of a number of membrane modules in parallel,
which must be replicated over and over to scale to larger feed rates.

 Membrane processes can be saddled with major problems of fouling of the

membranes while processing some type of feed streams. This fouling, especially if it
is difficult to remove, can greatly restrict the permeation rate through the
membranes and make them essentially unsuitable for such applications.

Challenges of Membrane Processes

The following questions generally need to be considered when considering a membrane

option:

 What is the most appropriate niche for membranes in the real, large-scale operating
plant?
 How effective would the membranes be in comparison with "traditional" separation
methods such as distillation, absorption, etc?

Classification of Synthetic Membranes

5
A membrane can be natural or synthetic, thick or thin, its structure can be homogeneous or
heterogeneous, transport across membrane can be active or passive, passive transport can
be driven by various means (e.g. pressure, concentration, electrical difference), neutral or
charged. As such, membranes can be classified according to different viewpoints. The first
classification is by nature, i.e. biological or synthetic membranes. This is the clearest
distinction possible. Synthetic membranes can be subdivided into organic (polymeric or
liquid) and inorganic (e.g. ceramic, metal) membranes.

Another means of classifying membranes is by morphology or structure - for the case of

solid synthetic membranes, the 2 types of membrane structures are the symmetric and
asymmetric (anisotropic) membranes.

The principal types of polymeric membranes are classified as shown in the Figure below:

The Figure shows schematically the different types of membranes. This classification is,
however, rather arbitrary, and many structures fit more than one of the above-mentioned
classes. For example, a membrane may be microporous, asymmetric in structure, and have
electrical charges.

6
The IUPAC classification for pores sizes are as follows:

The following membranes are briefly discussed:

Isotropic Membranes

 Microporous Membranes
 Non-Porous, Dense Membranes

 Electrically-Charged Membranes

Anisotropic (Asymmetric) Membranes

 Loeb-Sourirajan Membranes
 Thin-Film (Composite) Membranes
7
  Liquid Membranes

Other Membranes

Isotropic Membranes

Isotropic membranes have a uniform composition structure throughout, and they can be
porous or dense. The resistances to mass transfer in these membranes are determined by
the total membrane thickness. A decrease in membrane thickness results in an increased
permeation rate.

Microporous Membranes

The simplest form of microporous membrane is a polymer film with cylindrical pores or
capillaries. However, more commonly microporous membranes have a more open and
random structure, with interconnected pores. They are very similar in structure and
function to conventional filters. However, in contrast with conventional filters, these pores
are extremely small, on the order of 0.01 to 10 micrometer in diameter.

The separation of particles is mainly the function of molecular size and membrane pore size
distribution.

All particles larger than the largest pores are completely rejected by the membrane.

 Particles smaller than the largest pores, but larger than the smallest pores are
partially rejected, according to the pore size distribution of the membrane.
 Particles much smaller than the smallest pores will pass through the membrane.

In general, only molecules that differ considerably in size can be effectively separated by
microporous membrane, e.g. in ultrafiltration or microfiltration. These membranes are also
as the substrate (support) layer in composite membranes, and as a support matrix for liquid
membranes.

Non-Porous, Dense Membranes (Homogeneous Membranes)

This type of membranes consists of a dense film through which permeants are transported
by diffusion under the driving force of a pressure, concentration, or electrical potential
gradient.

The separation of various components of a mixture is related directly to their relative

transport rates within the membrane, which are determined by their diffusivity and
solubility in the membrane material. Thus, this type of membranes can separate permeants
of similar size if their concentrations in the membrane material (i.e. their solubility) differ
significantly.

Dense membranes have the disadvantage of low flux unless they can be made extremely
thin. For this reason, dense membrane properties are incorporated into the top "skin" layers
of asymmetric membranes.

8
Most gas separation, Pervaporation, and reverse osmosis processes use dense membrane to
perform the separation.

Electrically-Charged Membranes

These types of membranes are also referred to as ion-exchange membranes. They can be
dense or microporous, but most commonly are very finely microporous, with the pore walls
carrying fixed positively or negatively charged ions.

A membrane fixed with positively-charged ions is called an anion-exchange membrane

because it binds anions (negatively charged ions) in the surrounding fluid. The reverse is true
for a cation-exchange membrane.

Separation is achieved mainly by exclusion of ions of the same charge as the fixed ions on
the membrane structure, and is affected by the charge and concentration of ions in the
solution. This type of membranes is used for processing electrolyte solutions in
electrodialysis.

Anisotropic (asymmetric) Membranes

These membranes consist of a number of layers, each with different structures and
permeabilities. A typical anisotropic membrane has a relatively dense, extremely thin
surface layer (i.e. the "skin", also called the permselective layer) supported on an open,
much thicker porous substructure.

The separation properties and permeation rates are determined exclusively by the surface
layer; and the substructure functions as mechanical support, with virtually no separating
function. The resistance to mass transfer is determined largely or completely by the thin
surface layer.

The membrane can be made thick enough to withstand the compressive forces used in
separation. The thin film is always on the high-pressure side of the membrane, that is, the
feed side, since in this way maximum use of the support layer is made in stabilising the thin
film. These membranes had the advantage of higher fluxes, and almost all commercial
processes use such membranes.

Loeb-Sourirajan Membrane

Membranes made by the Loeb-Sourirajan process consist of a single membrane material,

but the porosity and pore size change in different layers of the membrane. Anisotropic
membranes made by other techniques and used on a large scale often consist of layers of
different materials which serve different functions.

Interfacial (Thin-film) Composite Membrane

9
This membrane consists of a thin dense film of highly cross-linked polymer formed on the
surface of a thicker microporous support. The dense polymer layer is extremely thin, on the
order of 0.1 mm or less, so membrane permeability is high. Because it is highly cross-linked,
its selectivity is also high.

Interfacial composite membranes are widely used in reverse osmosis and nanofiltration.

Solution-coated Composite Membrane

This type of membrane is formed by solution coating a thin (0.5 - 2.0 micrometer) selective
layer on a suitable microporous support.

Because the selective layer of the composite membrane is often very thin and delicate, a
sealing layer, also formed from a highly permeable material, is frequently used to protect
such membranes. See the Figure below.

The gutter layer acts as a conduit to transport material to the support membrane pores.

Liquid Membranes

Liquid membranes have become increasingly significant in the context of facilitated

transport, which utilizes "carriers" to selectively transport components such as metal ions at
a relatively high rate across the membrane interface.

Generally, formation of a thin fluid film is not a problem. Difficulty is encountered, however,
in maintaining and controlling this film and its properties during a mass separation process.
Reinforcement is necessary to avoid break-up of the film. Liquid membranes are used on a
10
pilot-plant scale for selective removal of heavy-metal ions and organic solvents from
industrial waste streams. They have also been used for the separation of oxygen and
nitrogen.

Liquid Membranes are discussed in greater detail in later Section.

Other Membranes

The application of polymer membranes is generally limited to temperature below about 200
o
C and to the separation of mixtures that are chemically inert. For high temperature
operation and/or with chemically active mixtures, membranes made of inorganic materials
(e.g. ceramic, metal) can be used. These inorganic membranes can be isotropic or
anisotropic.

Inorganic membranes can withstand very high temperatures up to 4,000 oC. The high
temperature resistance makes these membranes very attractive for gas separation
operation, especially in combination with a chemical reaction where the membrane is used
as a catalyst as well as a selective barrier to remove one or more of the components which
has been formed.

The chemical stability of existing polymeric materials is limited with respect to pH and
organic liquids. The chemical stability of inorganic materials is superior and is especially
suitable for application in harsh environments. Another important factor is the ease of
cleaning for application under high fouling conditions. For inorganic membranes all kinds of
cleaning agents can be used, allowing strong acid and alkali treatment.

Lastly, the lifetime of inorganic membranes is generally greater than that or organic
polymeric membranes.

Membrane Materials

Almost all industrial membrane processes are made from natural or synthetic polymers.
These membranes are sometimes known as organic membranes. Natural polymers include
wool, rubber (polyisoprene) and cellulose. Examples of synthetic polymers include
polyamide, polystyrene and polytetrafluoroethylene (Teflon).

Click here for more information on polymeric membranes.

Membranes can also be made from other non-polymeric materials. Such membranes
include inorganic membranes (for example metal, ceramic, carbon and zeolites) and liquid
membranes.

In addition, recent developments had led to the introduction of the so-called Hybrid
Membranes (or Mixed Matrix Membranes, where both organic and inorganic components
are used).

Another variation in membranes application is the Bipolar Membranes (BPM). where

membranes of different ionic charge are "sandwiched" together. Click here for more
information.
11
Polymeric Membranes

Synthetic polymers are produced by polymerisation of a monomer (condensation or

addition) or by the co-polymerisation of 2 different monomers. The resulting polymer is
categorised as having:

(1) a long, linear chain, such as linear polyethylene

(2) a branched chain, such as polybutadiene

(3) a three-dimensional highly cross-linked structure, such as phenol-formaldehyde

(4) moderately cross-linked structure, such as butyl rubber

The linear-chain polymers soften with an increase in temperature and are often soluble in
organic solvents, and are referred to as thermoplastic polymers.

At the other extreme, high cross-linked polymers do not soften appreciably, are almost
insoluble in most organic solvents, and are referred to as thermosetting polymers.

Examples of polymers used in membrane fabrication are shown in the Figures below.

12
The Table below showed typical applications of various polymeric materials in selected
membrane processes.

13
Inorganic Membranes

Inorganic membranes refer to membranes made of materials such as ceramic, carbon, silica,
zeolite, various oxides (alumina, titania, zirconia) and metals such as palladium, silver and
their alloys. Inorganic membranes can be classified into 2 major categories based on its
structure: porous inorganic membranes and dense (non-porous) inorganic membranes.
Microporous inorganic membranes have 2 different structures: symmetric and asymmetric;
and include both amorphous and crystalline membranes.

Microporous inorganic membranes can be obtained by coating of a porous support with a

colloidal solution, called sol. The sol can consist of either dense spherical particles (colloids
of oxides such as Al2O3, SiO2 or ZrO2) or polymeric macromolecules.

The Table showed the advantages and disadvantages of inorganic membranes in comparison
with polymeric membranes. It can be seen that although inorganic membranes are more
expensive than organic polymeric membranes, they possess advantage of: temperature
stability, resistance towards solvents, well-defined stable pore structure, and the possibility
for sterilization.

14
Application of dense inorganic membranes is primarily for highly selective separation of
gases such as hydrogen and oxygen. However, dense membranes have limited industrial
application due to their low permeability compared to porous inorganic membranes.
Therefore, today's commercial inorganic membrane market is dominated by porous
membranes.

Metal Membrane

Dense, metal membranes are being considered for the separation of hydrogen from gas
mixtures. Palladium (Pd) and its alloys are the dominant material used, due to its high
solubility and permeability for hydrogen. Palladium, however, is expensive. One alternative is
to coat a thin layer of palladium on a tantalum or vanadium support film. Alternative to
palladium and less expensive are tantalum and vanadium, which are also quite permeable to
hydrogen.

Recent focus is on supported thin metallic membranes with thickness ranging from
submicron to a few ten microns. The advantages include reduced material costs, improved
mechanical strength and possibly higher flux. The main developments have been the
production of composite palladium membranes for use in catalytic membrane reactors
(CMRs). This development is based on the concept of process intensification, one important
aspect if which is the potential for combining the reaction and separation stages of a process
in one unit. One such application is the CMR. Apart from the benefit inherent in cost
reduction of plant and maintenance, there is also the potential attainment of higher
conversions and product yields.

The composite palladium membrane used in the CMR is composed of a thin layer of Pd, or
Pd alloy, deposited onto a porous substrate, such as a ceramic or stainless steel.

The composite palladium membrane is placed adjacent to a catalyst bed and effects the
selective removal of hydrogen from the catalytic reaction source. Another application is the
use of these membranes to control the feed rate during partial oxidation reactions (e.g.
addition of hydrogen).

15
A major problem associated with metal membranes is the surface poisoning effects (e.g. by
a carbon-containing source) which can be more significant for thin metal membranes.

Ceramic Membrane

These membranes are made from aluminum, titanium or silica oxides. They have the
advantages of being chemically inert and stable at high temperatures. This stability makes
ceramic microfiltration and ultrafiltration membranes particularly suitable for food,
biotechnology and pharmaceutical applications in which membranes require repeated
steam sterilization and chemical cleaning. Ceramic membranes have also been proposed for
gas separations.

An example application of recent development is in the production and processing of syngas

(synthetic gas - a mixture of hydrogen and carbon monoxide). The key part of the process
involves the separation of oxygen from air in the form of ions to oxidize the methane.

A schematic representation of the process is given in the Figure.

Oxygen feeding from air is split at the perovskite-type membrane surface and is transported
as O2-. The advantage of the membrane-based process is that the production of syngas takes
place in a single-step operation occurring on one of the membrane sides, This process
eliminates the need for a separate oxygen production plant, and might lead to significantly
lower energy and capital costs.

Some remaining problems include:

 Difficulties in proper sealing of the membranes in modules operating at high

temperature
 Extremely high sensitivity of membranes to temperature gradient, leading to
membrane cracking

 Chemical instability of some perovskite-type materials in the high temperature

reduction conditions

16
Carbon Membranes

Carbon molecular sieve (CMS) membranes have been identified as very promising
candidates for gas separation, both in terms of separation properties and stability. Carbon
molecular sieves are porous solids that contain constricted apertures that approach the
molecular dimensions of diffusing gas molecules. As such, molecules with only slight
differences in size can be effectively separated through molecular sieving.

Carbon membranes can be divided into 2 categories: supported and unsupported.

Unsupported membranes have 3 different configurations: flat (film), hollow fiber and
capillary while supported membranes consisted of 2 configurations: flat and tube.

The Figure showed a comparison between carbon asymmetric membrane and polymeric
asymmetric membrane where the main difference is in the skin layer.

CMS membranes can be obtained by pyrolysis of many thermosetting polymers such as

poly(vinylidene chloride) or PVDC, poly(furfural alcohol) or PFA, cellulose triacetate,
polyacrylonitrile or PAN, and phenol formaldehyde.

Zeolite Membranes

Zeolites are microporous crystalline alumina-silicate with a uniform pore size. Zeolites are
used as catalysts or adsorbents in a form of micron or submicron-sized crystallites
embedded in millimeter-sized granules. The zeolite type prepared most often as a
membrane is MFI.

Main problem - relatively low gas fluxes compared to other inorganic membranes …. Due to
the fact that relatively thick zeolite layers are necessary to get a pinhole-free and crack-free
zeolite layer. Overcome: use thin layer supported on others.

Other problem: thermal effect of zeolites. The zeolite layer can exhibit negative thermal
expansion, i.e. in the high temperature region the zeolite layer shrinks …. But the support
17
continuously expands, resulting in thermal stress problems for the attachment of the zeolite
layer to the support, as well as for the connection of the individual micro-crystals within the
zeolite layer.

Hybrid Organic-Inorganic Membranes

Compared to certain inorganic membranes, polymeric membranes have relatively low

separation performance. However, the processing flexibility and low cost of polymers still
makes them highly attractive for many industrial applications. Most of the existing industrial
polymers have been applied to membrane preparation. However, despite many attempts at
maximizing membrane performance by varying the molecular structure, polymers still
exhibit a trade-off between productivity and selectivity, which is demarcated by a so-called
"upper bound" as shown in the Figure.

As shown in the plot, there is a strong inverse relationship between flux (permeability) and
selectivity. The straight line shown links the most permeable polymers at a particular
selectivity. It is known as the "upper bound", beyond which no better material is known. The
position of the upper bound has moved significantly between 1980 and 1990, reflecting the
difference the first air-separation membrane materials and the current much-improved
membrane materials. While research in membrane materials in still on-going, the movement
of the upper bound has slowed since 1990.

Hybrid organic-inorganic materials has been developed as an alternative to limitations of

conventional polymeric or inorganic materials.

Mixed Matrix Membranes

18
Mixed matrix membranes (MMM) essentially comprise molecular sieving materials (such as
zeolite or CMS - carbon molecular sieves) embedded (dispersed) in a polymer matrix. Mixed
matrix membranes have been proposed for gas separation applications, such as N 2
production from air as an alternative to the conventional polymeric membranes. A
schematic diagram is shown in the Figure.

Zeolites and CMS are size- and shape-selective, thus allowing molecular sieving
discrimination of gas molecules: smaller-size gas penetrants can diffuse at a much higher
rates than larger-sized penetrants. These materials thus offer very attractive permeation
properties with permeabilities and selectivities significantly higher than polymeric materials.
These materials, however, has the disadvantage of higher cost, and more difficult processing
requirements. The inherent rigidity and brittleness of these materials create difficulties in
forming continuous, defect-free membranes and impractical for large-scale industrial
applications. Polymeric material processing, on the other hand, is well established. However,
separation by polymeric materials suffered from O2 permeability and O2/N2 selectivity
limitations.

The mixed matrix membrane combines the advantages of each medium: high separation
capabilities of the molecular sieving materials with the desirable mechanical properties and
economical processing capabilities of the polymers.

Material selection for both matrix and sieve phases is the key aspect of the development of
mixed matrix membranes. Polymer matrix selection determine minimum membrane
performance: the polymer selected must provide industrially acceptable performance. For
example, silicone rubber only provides an O2/N2 selectivity of 2.1 and is therefore not
attractive. Similarly, the choice of molecular sieving material is important. For example,
zeolite 13X with an aperture of 10 Å is not suitable for O 2 and N2 molecules with lengths of
3.75 and 4.07 Å respectively. On the other hand, zeolite 4A with an effective aperture of 3.8
Å is suitable for O2/N2 separation.

19
Some examples of commercially available MMM are silicalite-cellulose acetate and
Matrimid®-CMS MMMs.

Example Application

An example application of hybrid inorganic membrane is the controlled vapor-phase

oxidation as shown in the Figure. In a typical process, vapor phase hydrocarbons are
selectively oxidized to produce epoxides, anhydrides, carboxylic acids and other products.
The reactions are exothermic, and the maximum allowable temperature is usually limiting.

By using a hybrid inorganic membrane, the amount of oxygen supplied into the reaction
zone can be stoichiometrically controlled, enabling a reduction in the reactor's maximum
temperature and thus decreasing the rate of sequential reactions that convert desirable
products into undesirable by-products. Also, any propensity for a temperature runaway
reaction is minimized at the lower reaction temperature.

Bipolar Membranes (BPM)

Principles of Bipolar Membranes

The conventional method for generating H+ and OH- ions from water uses electrolysis.
Electrolysis also generates O2 and H2 and the production of these gases consumes about half
of the electrical energy of the process. In contrast, special ion-exchange membranes such as
the bipolar membranes are capable of splitting water directly into H + and OH- ions without
generating O2 and H2.

In its simplest form a bipolar membrane is a cation-exchange membrane laminated together

with an anion-exchange membrane, through an intermediate layer (the "junction" layer).

20
The intermediate layer is the most important part of the membrane. The principle of
"splitting" water by using a bipolar membrane is shown in the Figure.

Water diffuses from across the cation- and anion-exchange layer to junction layer where it
dissociates into H+ and OH- ions. The H+ ions then migrates across the cation-exchange layer
towards the negative electrode while the OH- ions migrate across the anion-exchange layer
to the positive electrode.

Any other anions (X-) are excluded from the junction layer by the cation-exchange layer and
any other anions (M+) are excluded from the junction layer by the anion-exchange layer.

The 2 main factors that determine the behavior of bipolar membranes are the structure of
the bipolar junction between the 2 ion-exchange layers, and the nature of charged groups
which are attached to the polymeric matrix.

Example Application of Bipolar Membranes

Although the use of bipolar membranes is an emerging technology, several applications of

this technology are already used on an industrial scale. An example application is the
treatment of concentrated salt solutions such as Na2SO4 to produce H2SO4 and NaOH.

A cell system consisting of an anion-exchange membrane (AEM), a bipolar membrane

(BPM), and a cation-exchange membrane (CEM) as a repeating unit is shown in the Figure.

21
Such configuration is often referred to as a 3-compartment cell. This elementary cell is
repeated and placed between 2 electrodes.

The Na2SO4 solution flows between the cation-exchange and anion-exchange membranes
(the salt channel), while water flows between the ion-exchange membranes and the bipolar
membranes, i.e. in the H2SO4 and NaOH channels.

When direct current is applied, Na + will be attracted to the cathode while SO 42- will be
attracted to the anode. Na+ will flow across the CEM into the NaOH channel while SO 42- will
flow across the AEM into H 2SO4 channel. In the bipolar membrane, water will dissociate to
form the equivalent number of H+ and OH- ions. The H+ ions will permeate through the
cation-exchange side of the BPM and form H2SO4 with the SO42- ions in the H2SO4 channel.

Similarly, the OH- ions will permeate through the anion-exchange side of the BPM and form
NaOH with Na+ ions in the NaOH channel. The final result is the production of NaOH and
H2SO4 from Na2SO4 solution at a significantly lower cost than by other methods.

In a similar way, the use of bipolar membrane process offers an interesting alternative to the
conventional membrane electrolysis for the production of NaOH and HCl from NaCl using
the same cell as shown above. The existing method of electrolysis produces chlorine, which
is an environmental concern. However, a disadvantage in using bipolar membranes is that
the concentration of HCl and NaOH in practical situations is limited to about 20%, while
using conventional process this concentration reaches 30%.

Liquid Membranes

A liquid membrane (LM) is literally a membrane made of liquid. It consists of a liquid phase
(e.g. a thin oil film) existing either in supported or unsupported form that serves as a
membrane barrier between two phases of aqueous solutions or gas mixtures.

22
One of the benefits of using a liquid membrane is that LMs are highly selective, and, with
the use of carriers for the transport mechanism, specific molecular recognition can be
achieved. LMs are relatively high in efficiency, and as such, are being looked into for
industrial applications. The major problem restricting widespread application is stability:
liquid membranes require stability in order to be effective, and if they are pushed out of the
pores or ruptured in some way due to pressure differentials or turbulence, then they just do
not work.

There are 2 basic types of liquid membranes - the Emulsion Liquid Membrane (ELM) and the
Immobilized Liquid Membrane (ILM), also called a Supported Liquid Membrane.

An ELM can be visualized as consisting of a "bubble within a bubble". The inner most bubble
is the receiving phase, and the outer bubble is the separation "skin" containing the carriers.
Anything outside the bubble is the source phase. In an ELM set-up there would be huge
numbers of these bubbles, such as shown schematically in the Figure.

The disadvantages of ELM concerns the formation of the emulsion:

 Anything effecting emulsion stability must be controlled, i.e. ionic strengths, pH, etc.
 If, for any reason, the membrane does not remain intact during operation, the
separation achieved to that point is destroyed.

 In order to recover the receiving phase, and in order to replenish the carrier phase,
you have to break down the emulsion. This is a difficult task, since in order to make
the emulsion stable, you have to work against the ease of breaking it back down.

An ILM is much simpler to visualize. It is made of some kind of rigid polymer membrane,
with lots of microscopic pores in it which are filled with organic liquid. In the liquid are the
carriers that perform the required separation. What happens then is that the ILM takes
things from one side of the rigid membrane (the source phase) and carries it to the other
side (the receiving phase) through this liquid phase. A schematic of an ILM is shown in the
Figure.

23
In such a system, instability is caused by the removal of carrier or organic liquid in the pores
of that supporting membrane. There are two possible ways for this to occur:

 carrier or solvent evaporation

 a large pressure differential across the membrane that effectively pushes the fluid
out

Commonly employed supported liquid membrane modules include hollow fiber and spiral
wound.

MEMBRANE PROPERTIES

Various membrane properties can be classified as follows:

 Physical - porosity, pore size and pore distribution, thickness, tortousity, thermal
stability, etc.
 Separation - permeate flow rate or permeation flux, permeability, selectivity.

 Surface and electrochemical - streaming potential, zeta potential (or electrokinetic

potential), membrane potential (voltage potential difference), surface charge density,
wettability (hydrophilic or hydrophobic), etc

 Electrical - impedance measurements to determine the membrane conductance and

capacitance

Separation properties of membranes depend not only on their physical properties, but also
on their surface charge and electrical properties. Many of these properties are inter-related,
and vary from membrane to membrane, as well as the environment in which the membrane
is exposed, e.g. feed conditions (pH, temperature, concentration of solutes, various
properties of solutes, etc). Suffice to say, each membrane-solute system is unique, and must
be evaluated experimentally.

Membrane Physical Properties

Some of the more common parameters used to characterise microporous membranes

include the following:

Pore Diameter

24
Pore diameter is probably the most important property characterising a microporous
membrane. Although microporous membranes are usually characterised by a single pore
diameter, note that most membranes actually contain a range of pore sizes (i.e. there exist a
pore size distribution).

Porosity

Porosity is the fraction of total membrane volume that is porous. Typical microporous
membranes have average porosities from 0.3 to 0.7.

Tortousity

Membrane tortousity (t) is the length of the average pore compared to the membrane
thickness. Simple cylindrical pores at right angles to the membrane surface have a tortousity
of 1.0. Since pores usually take a more meandering path through the membrane, typical
tortousities are in the range of 1.5 to 2.5.

Thermal Stability of Polymeric Membranes

At low temperatures, typically below 100 oC, polymers can be classified as glassy or
crystalline. A glassy polymer is one that is brittle and glassy in appearance and lacks any
crystalline structure (i.e. they are amorphous). A crystalline polymer is one that is brittle,
hard, and stiff, with a crystalline structure.

If the temperature of a glassy polymer is increased, a point may be reached where the
polymer becomes rubbery. The temperature is known as the glass-transition temperature,
Tg. If the temperature of a crystalline polymer is increased, a point known as the melting
temperature Tm, is reached where the polymer becomes a melt. (Note that a thermosetting
polymer never melts)

The Table showed the Tg and Tm values for several natural and synthetic polymers from
which membranes have been fabricated. Most polymers have both amorphous and
crystalline regions, that is, a certain degree of crystallinity that varies from 5 to 90%, making
it possible for some polymers to have both Tg and Tm.

Membrane Separation Properties

25
Permeation Flux

The membrane permeation flux is defined as the volume flowing through the membrane per
unit area per unit time. The SI unit used is m3/m2.s although other are often used as well.

For the case of transport of gases and vapors, the volume is strongly dependent on pressure
and temperature. As such, gas fluxes are often given in terms of a "standard condition"
which is defined as 0 oC and 1 atmosphere (1.0013 bar).

Permeability Coefficient

The permeability coefficient, P (or simply the permeability) is defined as the transport flux of
material through the membrane per unit driving force per unit membrane thickness. It's
value must be experimentally determined. The barrer is the commonly used unit for gas
separation and it is defined as:

The term cm3 @STP / cm2.s refers to the volumetric trans-membrane flux of the diffusing
species in terms of standard conditions of 0oC and 1 atm, the term cm refers to the
membrane thickness, and cm-Hg refers to the trans-membrane partial pressure driving force
for the diffusing species.

Other commonly used units include: kmol.m.m-2.s-1.kPa-1, or m3.m.m-2.s-1.kPa-1, or kg. m.m-2.s-

1
.kPa-1. Note here that the driving force is the pressure difference across the membrane.

In general, permeability of a polymer for a gas mixture increases with decreasing size and
increasing solubility (or condensability) of the gas. The relative permeability of a gas is given
below in order of decreasing gas permeability as:

H2 > He > H2S > CO2 > O2 > Ar > CO > CH4 > N2

The permeability of component-i is a product of 2 terms:

where Ki is the sorption (or partition) coefficient and Di is the permeate diffusion
coefficient.

The sorption coefficient is an equilibrium term linking the concentration of a permeating

component in a fluid phase with its concentration in the membrane polymer phase. It
accounts for the solubility of the component in the membrane.

The diffusion coefficient is a kinetic term that reflects the effect of the surrounding
environment on the molecular motion of the permeating component. It accounts for the
diffusion of the component through the membrane.

26
Permeabilities (in Barrer) of several pure gases in widely used polymers are shown in the
Table.

Permeance

The permeance PM is defined as the ratio of the permeability coefficient (P) to the
membrane thickness (L). The permeance for a given component diffusing through a
membrane of a given thickness is analogous to a mass transfer coefficient.

Membrane Selectivity

In gas separation, the membrane selectivity is used to compare the separating capacity of a
membrane for 2 (or more) species. The membrane selectivity, a (also known as the
permselectivity) for one component (A) over another component (B) is given by the ratio of
their permeabilities:

Replacing for PA and PB, and re-arrange, we have:

27
The ratio DA/DB is the ratio of the diffusion coefficients of the 2 gases and can be viewed as
the mobility selectivity, reflecting the different sizes of the 2 molecules. The ratio K A/KB is
the ratio of the sorption coefficients of the 2 gases and can be viewed as the sorption or
solubility selectivity, reflecting the relative condensabilities of the 2 gases.

In all polymer materials, the diffusion coefficient decreases with increasing molecular size,
because large molecules interact with more segments of the polymer chain than small
molecules do. Hence the mobility selectivity always favour the passage of small molecules
over large ones.

However, the magnitude of the mobility selectivity depends greatly on whether the
membrane material is above or below its glass transition temperature, T g. The polymer is
glassy below Tg and its structure is tough and rigid. The polymer is rubbery above T g and
there is flexibility in its structure.

Diffusion coefficients in glassy polymers decrease much more rapidly with increasing
permeate size than diffusion coefficients in rubbers.

The sorption coefficient of gases and vapours increase with increasing condensability of the
permeant. This dependence on condensability means that the sorption coefficient also
increases also increases with molecular size. Thus sorption selectivity favours large, more
condensable molecules such as hydrocarbon vapour, over permanent gases such as O 2 and
N2. However, the difference between the sorption coefficients and permeabilities in rubber
and glassy polymers is far less marked than the difference in diffusion coefficients.

Thus, the balance between the mobility term and the sorption selectivity term is different
for glassy and rubbery polymers.

In glassy polymers the mobility term is usually dominant, thus small molecules permeate
preferentially. When used to segregate organic vapour from N 2, glassy polymer preferentially
permeates nitrogen. In rubbery polymers the sorption term is usually dominant, so larger
molecules permeate preferentially. When used to separate organic vapour from N 2, rubbery
polymer preferentially permeates the organic vapour.

Selectivity obtained from ratio of pure gas permeabilities is called the ideal membrane
selectivity or the ideal permselectivity. This is an intrinsic property of the membrane
material.

If a membrane had holes or pores which were all exactly the same diameter, then those
molecules or particles whose diameters were smaller than the pore diameter would pass
through the membrane, and those molecules or particles with larger diameters than the
pore diameter would be totally rejected. Such a membrane would show an infinite
selectivity.

28
The selectivities of actual membranes are less than infinite due to several factors. First,
seldom will all pores in a membrane be exactly the same size. Thus smaller pores might
exclude one component while larger pores permit it to pass. In such a case, the selectivity
would be a function of the relative populations of various pore sizes.

Second, molecules may be able to deform to some extent and may actually enter pores
slightly smaller than their original diameter.

Third, molecules of one type may adsorb on the walls of the pores and reduce the effective
diameters of these pores. In this case a pore's effective diameter might vary with the
feed/retentate and permeate compositions, depending on how much of the adsorbing
component is in each of the streams.

It is also important to note that practical gas separation processes are performed with gas
mixtures rather than pure gases. If the gases in a mixture do not interact strongly with the
membrane material, the pure gas intrinsic selectivity and the gas mixture selectivity will be
equal. This is usually the case for mixtures of O 2 and N2. Gas mixtures are also usually non-
ideal, especially under high pressure, and thus the actual selectivity may be quite different
from the ideal value. In many cases, such as the mixture of CO 2 and CH4, the CO2 is
sufficiently sorbed by the membrane that permeability of CH 4 is affected. The selectivity for
such a gas mixture will deviate from the calculated selectivity from pure gas measurements.

Nonetheless, selectivities are usually reported under ideal conditions because pure gas
permeabilities are more frequently available.

Separation Factor

Similarly, in liquid separation, the selectivity for component A over B in a binary liquid
mixture may be expressed by the separation factor as:

Membrane Surface Characteristics

Membranes acquire a characteristic electric surface charge when contacted with an aqueous
solution. The membrane charge interacts with ions electrostatically and affects the
separation efficiency of ions through the partition of ions from the bulk solution into the
membrane. Such interaction can have either a positive or a negative effect on the filtration
process. Depending on the sign of membrane charge and charge of ions, molecules or
particles; attraction or repulsion may occur: for example, opposite signs of charge can create
attractive forces, leading to fouling and flux decline.

The membrane surface charges can be characterized by membrane potential measurements

and zeta potential measurements.

The "Double Layer" and Zeta Potential

The development of a nett charge at a membrane surface (or any particles) in contact with
an aqueous solution affects the distribution of ions in the neighboring interfacial region,
resulting in an increased concentration of counter-ions (i.e. ions of charge opposite to that of

29
the surface) in the solution close to the surface. This leads to the development of the
"electrical double layer" to maintain electrical neutrality of the system.

The surface of a membrane can become charged due to the dissociation of specific ionizable
groups on the base polymer, depending on the pH and concentration of the feed solution.

The Figure showed a schematic of a typical ion distribution near a negatively-charged

membrane surface. Ions in the solution are attracted and absorbed to the surface, forming
the so-called Stern layer. The ions in this layer are adsorbed relatively strongly to the
surface. The extent of ion adsorption is determined by electrical and other interactions
between the individual ions and the membrane surface. The Stern surface (also referred to
as the outer Helmholtz plane) refers to the plane drawn through the center of the adsorbed
ions.

Next to this layer of ions is an outer, or diffuse layer (also known as the Gouy-Chapman
layer). The Stern layer and the diffusion layer together constitute the so-called double layer.

The surface of shear (or shear plane or slipping plane) defines a region at which the fluid
becomes "mobile" and shows elastic behavior. It is an imaginary surface that is located
outside the Stern layer, but for practical reasons, is assumed to be identical to the Stern
surface. The electrical potential at the surface of shear is defined as the zeta potential.

The potential at the Stern layer that is the actual potential influencing the behaviour of the
charged species. However, as this potential cannot be measured directly, the zeta potential is
often considered as an adequate substitute. Therefore, the zeta potential is the value that is
typically used to characterize the electrical properties of the surface.

30
From the Stern layer, and into the diffusion layer, the potential decays as the distance from
the surface increases until, at sufficient distance, it reaches the bulk solution value,
conventionally taken to be zero. See The Figure.

If an electric field (potential) is applied parallel to the surface, forces will be exerted on the
double layer. The part of the double layer closest to the surface (i.e. the Stern layer) will
remain stationary while the remainder (i.e. the diffuse layer) will move under the influence
of the electric field.

When a layer of macromolecules, whether a polyelectrolyte or an uncharged polymer, is

adsorbed on the surface of the membrane, this can alter the zeta potential simply because it
shifts the location of the shear plane further from the actual surface.

The zeta potential is therefore a function of the surface charge of the membrane, any
adsorbed layer at the interface, plus the nature of the surrounding medium in which the
membrane is exposed (e.g. ionic species, pH, etc). From zeta potential measurements, the
surface charge density can be calculated.

In practice, zeta potentials are usually negative, but they can lie anywhere in the range from
-100 to +100 mV. Changes in membrane zeta potential could be used to investigate the
behavior of cake deposition and fouling during the filtration of colloids. The adsorption
capacity of a membrane decreases with decreasing (more negative) zeta potential.

Electrical Properties of Membranes

The surface and subsurface structure of microporous membranes are key factors in
determining their performance. A simple microporous membrane can be represented using
an electrical analogy of a resistor and capacitor in a parallel circuit, as shown in the Figure.

31
The resistance of the resistor is R (ohm, W) and the capacitance of the capacitor is C (Farad,
F).

For a composite membrane, which consists of a thin-skin layer and a much thicker sublayer
(i.e. the supporting matrix), an extension of the above analogy results in 2 sets of RC system,
as shown in the Figure, each with its own resistance and capacitance.

Often, it is more convenient to use the term conductance G instead of resistance. The
conductance is simply the reciprocal of resistance, i.e.

The conductance G describes the ability of the membrane to conduct electrical charges
while the capacitance C describes the ability of the membrane to store electrical charges.

For a membrane slab of cross-sectional area A and thickness x, these properties are given
by:

and

where the constants s and e are the electrical conductivity and dielectric permittivity of the
membrane respectively.

Note that the conductivity is related to resistivity (r) as its reciprocal (s = 1/r). Another
frequently encountered term is the reactance (X), which is the reciprocal of the capacitance.

Impedance (Z) is an expression used to describe the opposition that an electronic circuit
(such as the parallel resistor-capacitor shown above) offers to the passage of alternating
and/or direct electric current. Impedance is a vector (2-dimensional) quantity consisting of 2
independent scalars (1-dimensional): the conductance (G) and the capacitance (C).

An alternative way of expressing the impedance is in terms of the admittance (Y), which is
the reciprocal of impedance.

Most commonly, impedance and admittance are expressed in complex number form as:

Membrane Characterization

Membrane processes can cover a wide range of separation problems with a specific
membrane being required for every problem. Membranes may differ significantly in their
structure and consequently in their functionality. To know what membrane to use in a
particular separation process, different membranes must be characterized in terms of
structure and mass transport properties. Because very different membranes are used,
different techniques are required for characterization.

Membrane characterization is a very important part of membrane research and

development because the design of membrane processes and systems depends on reliable
data relating to membrane properties.

The following sections briefly discuss the following:

32
  Characterization of Porous Membranes
 Characterization of Dense, Homogeneous Membranes

 Characterization of Charged Membranes

Characterization of Porous Membranes

Porous micro- or ultrafiltration membranes are generally characterized in terms of their

trans-membrane flux, pore size, pore size distribution, and molecular mass cut-off.

Electron Microscopy

The structure of porous membranes can be determined by electron microscopy. Scanning

Electron Microscopy (SEM) gives an especially clear picture of membrane structure and
requires minimum sample preparation; however, resolution is limited to about 50-nm.
Higher resolution can be obtained with Transmission Electron Microscopy (TEM), but sample
preparation is significantly more complex and the structure is not nearly as clear as that
obtained by scanning electron microscopy.

For a basic understanding of how an SEM works, visit the Internet at

http://www.mos.org/sln/sem/seminfo.html

Bubble-Point Test

The "bubble-point" test is a simple method for determining the maximum pore size of a
membrane. One side of the membrane is filled with liquid such as water, although i-propanol
is often used as the standard liquid. If the other side of the membrane is exposed to air at a
certain pressure, air bubbles will penetrate through the pores of the membrane when the
radius of the air bubble is equal to the radius of the pore. The pressure needed to penetrate
a pore is inversely proportional to pore size. Thus, penetration occurs first through the
largest pores. When the pressure is increased further, pores with smaller diameters are also
penetrated.

The basic principle of the bubble point method is shown in the Figure.

33
Because the surface tension between water and air is rather high, high pressure is required
for the determination of small pores. In practice, pore sizes between 0.1 and 10 mm are
determined by hydrostatic pressures of 1500 - 15 kPa. The main application of the bubble-
point test is to determine pinholes and leaks in micro- and ultrafiltration membranes and
modules.

Filtration Tests

In filtration tests, trans-membrane flux and membrane solute retention are determined. To
avoid the influence of concentration polarization or any other boundary layer phenomenon,
the trans-membrane flux is generally measured as a function of applied hydrostatic pressure
with ultra-pure water.

Membrane flux often decreases with time during the filtration test due to compaction of the
membrane structure under pressure. A compaction factor has been defined, which is
determined from the slope of the curve obtained when the trans-membrane flux at constant
pressure is plotted versus time on a semi-logarithmic scale.

Unfortunately, the fluxes measured with ultra-pure water, at least in ultra- and
microfiltration, often bear little relation to those obtained with solutions containing
macromolecules or suspended particles.

Molecular Weight Cut-Off

Many membranes, especially those used in ultrafiltration, are also characterized in terms of
their molecular mass cut-off. Here, the rejection of a given membrane for various molecular
mass components is determined in filtration tests. Rejection is plotted versus molecular
34
mass and the resulting curve describes the membrane's solute retention properties.
Components with a diameter smaller than the smallest pores pass through the membrane
unhindered; all components with a diameter smaller than the largest pores but larger than
the smallest pores are rejected in proportion to the pore size distribution of the membrane.

The retention R is a dimensionless number expressed simply as:

where cF and cP are the solute concentration in the feed and permeate respectively. The
value of R varies between 100% (complete retention of the solute) and 0% (solute and
solvent pass through the membrane freely).

Some commercial membranes have a very narrow pore size distribution and thus a sharp
molecular mass cut-off. Others have a wider pore size distribution and therefore a more
diffuse cut-off.

Characterization of Dense, Homogeneous Membranes

Dense membranes are normally used to perform separations on a molecular level. As such,
the chemical nature and morphology of the polymeric membrane and the extent of
interaction between the polymer and the permeate are important factors to consider.
Characterization methods used with porous membranes need to be supplemented with
other procedures. One of the simplest method of characterizing these membranes is to
determine the membrane permeability towards gases and liquids. Various physical methods
can be used to characterize the parameters that affect the permeability, such as the glass
transition temperature and polymer crystallinity (degree, size and shape).

Characterization of Charged Membranes

The methods used to determine the structural properties of ion-exchange membranes are
similar to those used for characterizing other polymer membranes. These methods are also
supplemented by various electrokinetic parameters such as the zeta potential and surface
charge density.

Because ion-exchange membranes are used mainly in separation processes with electrical
potential driving forces, their permeability and selectivity are determined under
experimental conditions closely related to their practical use, and their properties are
expressed in terms commonly used in electrochemistry (e.g. electrical resistance, ion-
transfer numbers, charge densities).

Electrochemical Impedance Spectroscopy (EIS)

Electrochemical impedance spectroscopy is a relatively new non-destructive technique for

characterizing membranes in terms of its electrical properties. The method involves the
measurement of the electrical impedance of a membrane system over a wide range of
frequencies (1 - 10,000 Hz).

Impedance measurements are made by injecting an alternating current (ac) of known

angular frequency (w) and small amplitude (IO) into the system. As a result, an electrical
potential difference (voltage) is developed across the membrane. The amplitude (VO) and
35
phase difference (F) of the developed voltage are then measured. This is achieved through
comparison with a complex electrical circuits within the impedance spectrometer.

The EIS method is useful in determining the electrical properties of heterogeneous

membrane system (e.g. those composed of a thin skin on a porous support) as it permits the
evaluation of contribution of each layer separately.

The basic set-up is shown schematically in Figure x-xx for a simple 1-membrane system.

A membrane separates the container into 2 chambers, and both are filled with an electrolyte
(e.g. potassium chloride solution). The electrodes are immersed as shown. Current is
injected via the electrodes.

To determine the impedance requires the measurement of the potential difference (and
phase) of the potential developed across the "system", which in this case includes not only
the membrane but also the solution in contact with the membrane (including the unstirred
layer of solution at the membrane surface), plus the electrode-solution interface
impedances which are in series with it.

A further complication arises if the potential developed is measured using the same
electrodes used to inject the current, for then we must also include the impedance of the
double layer which is present between the electrodes and the aqueous solution. The
impedance of the electrode-solution double layer is strongly dependent on the frequency,
especially at low frequencies; and its effect can be very large. The use of 2-electrode
impedance measurement method is best confined to high frequencies.

This problem can be eliminated through the use of 4-electrodes system for impedance
measurement because it allows only the electrical potential developed between the voltage
electrodes to be measured. The potentials developed at the current-injecting electrodes
need not be considered. Another advantage of the 4-electrode system is that the voltage
electrodes may be placed arbitrarily close to the membrane. A typical 4-eletrodes set-up is
shown in Figure x-xx. Now the membrane "system" being considered consists of the
membrane itself, plus the unstirred layers on both sides of the membrane, and the
electrolyte between the 2 voltage electrodes.

NOTE:
In the 4-electrode system, the voltage electrodes are connected to a measuring device with
very high input impedance. This will ensure that the current flowing in these voltage
electrodes is very small, thus eliminating any significant electrode-solution impedance
effects at these electrodes.

The Figure showed the equivalent electrical circuit for the membrane system. Note that
we've used the conductance G in place of the resistance R for the membrane.

Subscript "M" is introduced to denote electrical properties of the membrane: CM and GM.
Also, the unstirred layers and the "slices" of electrolyte between the voltage electrodes are
lumped together and is represented as a conductance GElectrolyte, one on each side of the
membrane

36
The magnitude and phase of the impedance of the membrane system to be determined are
given by:

and

The conductance and capacitance can be obtained from:

and

From the equations for G and C, the membrane thickness (x) can be estimated.

The impedance of the membrane system is dependent on the frequency used. The
measured capacitance and conductance of the membrane varies with the frequency, with
the variation occurs most notably around the frequency given by:

NOTE:
To determine the electrical properties of the membrane (CM and GM) from the measured
values of the system, GElectrolyte must be determined accurately and subtracted from the
measured values. GElectrolyte can simply be measured by making identical measurements
without a membrane present.

37