Membranes in chemical and bioprocess engineering

Membrane – a selective barrier between two phases, the term “selective” being inherent to a membrane or a
membrane process.

A membrane can be:

• Thick or thin
• Structure can be homogeneous or heterogeneous
• Transport can be passive or active
• Active transport can be driven by a pressure, concentration or temperature difference
• Natural or synthetic
• Neutral or charged

Two phases system separated by a membrane

• Phase 1 – is usually considered as a feed or upstream side phase

• Phase 2 – is considered as a permeate
• Separation is achieved because the membrane has the ability to
transport one component from the feed mixture more readily
than any other component(s).
• The efficiency of it is determined by flow through the membrane
(sometimes denoted as a flux or permeation rate:
𝑉𝑜𝑙𝑢𝑚𝑒 𝑓𝑙𝑜𝑤𝑖𝑛𝑔 𝑡ℎ𝑟𝑜𝑢𝑔ℎ 𝑡ℎ𝑒 𝑚𝑒𝑚𝑏𝑟𝑎𝑛𝑒 𝑚3
[ 2 ]
𝑢𝑛𝑖𝑡 𝑜𝑓 𝑎𝑟𝑒𝑎 ∙ 𝑡𝑖𝑚𝑒 𝑚 ∙ℎ
• This is considered as a volume flux but it can be converted also into a mass flux or mole flux

“Dead-end” filtration is when feed is applied perpendicular to the membrane surface.

Particles smaller than the effective pore size pass through as filtrate, and particles that
are larger build up as a cake layer or the membrane surface.

“Cross flow” filtration is when the flow is applied tangentially across the membrane
surface. As feed flows across the membrane surface, filtrate passes through it while
concentrate accumulates at the opposite end of the membrane. The tangential flow of
the membrane creates a shearing effect on the surface of the membrane, which in
turn reduces fouling.

In pressure driven membrane processes, flux is related to pressure

𝑃−∆𝜋
as follows: 𝐽 = (𝜂𝑅
𝑡𝑜𝑡 )
𝑃 – pressure exerted on the membrane,
𝜂 – feed viscosity,
𝑅𝑡𝑜𝑡 – the sum of resistances
Flux behavior in membrane processes

• In an ideal scenario, such as in the case of a feed solution

comprised solely of pure water, we would expect that
pressure and flux to follow a direct and linear
relationship
• In this case, the total resistance is equal to only the
resistance exerted by the membrane, which would
remain constant, as would the feed viscosity is at
constant temperature
• However, with the addition solutes to the feed stream,
the total resistance increases
• This can be attributed to a number of factors, such as concentration polarization and fouling via pore blockage
or cake layer formation
• For this reason, flux behavior in pressure driven membrane processes typically decreases over time (the figure
on the right)

Concentration polarization and fouling

• In membrane technology, concentration polarization refers to

the emergence of concentration gradients at a membrane/solution
interface resulted from selective transfer of some particles through
the membrane under the effect of transmembrane driving forces

• Generally, the cause of concentration polarization is the ability

of a membrane to transport some particles more readily than the
other(s)
• The retained particles are concentrated at the upstream
membrane surface while the concentration of transported species
decreases
• Fouling is the accumulation of unwanted material on solid
surfaces

Advantages of cross-flow vs dead-end process

• Because cross-flow removes build up

from the surface of the membrane, the
permeate flux does not drop as fast
when compared to dead-end filtration
(a)
• Cross-flow technology also provides the
benefit of an improved membrane
lifespan by helping to prevent
irreversible fouling (b)
The performance of a given membrane is determined by selectivity

This parameter is generally expressed by one of parameters:

• Retention factor (R)

• Separation factor (S)

For dilute aqueous mixtures:

𝑐𝑟 − 𝑐𝑝
𝑅=
𝑐𝑟 ∙ 100%
Where: 𝑐𝑟 is the solute concentration in the retentate and 𝑐𝑝 is the solute concentration in the permeate

Membrane selectivity towards gas mixture and mixtures of organic liquids is usually expressed in terms of separation
factor (S):
𝑦𝐴
𝑦𝐵
𝑆 = 𝑥 ∙ 100%
𝐴
𝑥𝐵
Where: 𝑦𝐴 and 𝑦𝐵 are the concentrations of components A and B in the permeate and 𝑥𝐴 and 𝑥𝐵 are concentrations
of them in the feed.

Molecular property Separation processes

Filtration, microfiltration, ultrafiltration, dialysis, gas separation, gel
Size
permeation chromatography
Vapor pressure Distillation, membrane distillation
Freezing point Crystallization
Extraction, adsorption, absorption, hyper filtration, gas separation,
Affinity
pervaporation, affinity chromatography
Charge Ion exchange, electrodialysis, electrophoresis
Density Centrifugation
Chemical nature Complexation, liquid membranes

Basic classification of membranes

Synthetic: An artificial membrane, or synthetic

membrane, is a synthetically created membrane which
is usually intended for separation purposes in
laboratory or in industry:

• Liquid membranes
• Polimeric membranes
• Ceramic 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.

The inorganic membrane possesses:

• high thermal and chemical stability,
• inertness to microbiological degradation,
• ease of cleaning after fouling compared to organic counterparts.
However, inorganic membranes tend to have higher capital costs due to specific thickness requirements needed to
withstand pressure drop differences.

• The thickness of symmetric memebranes

range roughly from 10 to 200 µm
• The resistance to mass transfer being
determined by the total membrane thicknes
• A decrease in membrane thickness results in
an increased permeation rate.
• Suported by a porous sublayer with thicknes
of about 50 to 150 µm

• These membranes combine the high selectivity of a dense

membrane with the high permeation rate of very thin
membrane

Driving force can be gradients in pressure, concentration, electrical

potential or temperature.
 Microfiltration Pervaporation Gas separation
Ultrafiltration
Nanofiltration

Reserved osmosis
Dialysis
Electrodialysis

Membrane distillation
Membrane extraction

• When particles of diameter is higher than 100 nm (and have to be retained) the hydrodynamic resistance of applied
membrane is low and small driving forces (low hydrostatic pressures) are sufficient to obtain high fluxes – the
membrane process is called MICROFILTARTION
• To separate macromolecules from aqueous solution, the membrane structure must be more dense and hence its
hydrodynamic resistance also increases. The applied pressure is greater than in microfiltration and this separation
is called ULTRAFILTRATION
• To separate very low molecular weight components a very dense membrane is used and it results in very high
hydrodynamic resistance. This process is called NANOFILTRATION or REVERSE OSMISIS
• Electrodialysis is a membrane process in which the driving force for (ions) transport is supplied by an electrical
potential difference. This process can use only when charged molecules are present. A typical (and logical) feature
of this process is that ionic or charged membranes are necessary
• Gas separation and pervaporation where dense membranes can be used – transport takes place via diffusion
Pervaporation is the only one membrane process where a phase transition occurs with the feed being a liquid and
the permeate a vapor.

• Two compensating phase transitions occur in membrane distillation. In this case, two aqueous solutions at different
temperatures are separated by microporous hydrophobic membrane and because of a difference in partial
pressure (i.e. temperature difference) vapor transport takes place through the pores of the membrane from the
hot the cold side.
• The solutions may not wet the membrane. Vaporization occurs at the high temperature side while the vapor
condenses at the low temperature side
• When a concentration difference is applied across a homogeneous membrane, the process is called dialysis.
Transport takes place by diffusion and separation is obtained through differences in diffusion rates because of
differences in molecular weight
Pressure-driven membrane processes and their separation characteristics

• Microfiltration is designed to retain particles in the

“micron” range – suspended particles in the range
from 0.10 µm to about 5 µm (larger particles better
separated using conventional cake filtration methods)
• Ultrafiltration can be looked at as a method for
simultaneously purifying, concentrating, and
fractionating macromolecules or fine colloidal
suspensions
• Microfiltration is used mainly as a clarification
technique, separating suspended particles from
dissolved substances

The transport trough a membrane occurs because of high

pressure value in front of membrane surface:

• Membranes are usually classified according to the size

of the separated components, and thus particle sizes in
MF applications are specified in microns
• With UF membranes, it is customary to refer to the
''molecular weight cut-off" (MWCO) instead of pore
size (in MF)
• Molecular weight alone does not determine the size
(e.g. of a protein)
• Many manufacturers use dextrans (rather than
proteins) to characterize MWCO
• UF membranes are characterized by studying the
relative permeabilities of proteins and polyethylene
glycols, which were characterized in terms of their
molecular weights.
Microfiltration

• Microfiltration is the membrane process which is closely resembled conventional filtration

• The pore sizes of microfiltration membranes range from 10 to 0.05 µm, making the process suitable for retaining
suspension and emulsions
• The volume flow through these microfiltration membranes can be described by Darcy’s law, the flux through the
membrane being directly proportional to the applied pressure: 𝐽 = 𝐾 ∙ ∆𝑃, where the permeability constant K
contains structural factors such as the porosity and pore size (pore size distribution). Furthermore, the viscosity of
the permeating liquid is also included in this constant.

• In order to optimise microfiltration membranes, it is essential to ensure that the structural parameters are such
that the surface porosity is as high as possible with the pore size distribution as narrow as possible

• The convective flow described by these equations only involves membrane-related parameters and (none apply to
the solutes).

• The main problem encountered when MF is applied is flux decline

• This is caused by concentration polarization and fouling

• To reduce fouling as much as possible it is important that careful control the mode of process operation

• Adsorption phenomena may also an important role in fouling and hence it is important to the select an appropriate
membrane material

Industrial applications of MF:

• Cold sterilization of beverages and pharmaceuticals

• Cell harvesting
• Clarification of fruit juice, wine, beer
• Metal recovery as colloidal oxides or hydroxides
• Waste-water treatment
• Continuous fermentation
• Separation of oil – water emulsion

Ultrafiltration

• Ultrafiltration is a membrane process whose nature lies between nanofiltration and microfiltration
• The pore sizes of the membranes used range from 0.05µm to 1nm
• UF is typically used to retain macromolecules from a solution, the lower limit being solutes with molecular weights
of few thousand Daltons
• UF membranes (like MF ones) can be considered as porous membranes where rejection is determined mainly by
the size and shape of the solutes relative to the pore size in the membrane
and where the transport of solvents is directly proportional to the applied
pressure: 𝐽 = 𝐾 ∙ ∆𝑃
• The permeability constant K includes all kinds of structural factors.
• The value of this constant K for UF membranes is much smaller than for
microfiltration membranes.
• An important difference (between MF and UF) is that UF membranes have an
asymmetric structure where the hydrodynamic resistance is mainly
determined within a small part of the total membrane thickness, whereas in
microfiltration the whole of the membrane thickness contributes towards the
hydrodynamic resistance.
Industrial applications of UF:

• Dairy (milk separation, whey proteins separation, cheese making)

• Food (potato starch and proteins recovery)
• Metallurgy (oil – water emulsions, elecropaint recovery)
• Textile (indigo recovery)
• Pharmaceutical (enzymes, antibiotics, pyrogens recovery)
• Water treatment (pure water recovery from waste water)

How MF and UF are similar?

They are both separation processes that use membranes for filtration

• Both MF and UF utilize membranes to filter out contaminating substances

• Membrane filtration can remove various particulates, bacteria, viruses, organic material, improving odour, taste,
colour, etc., and are used for various process separation, purification, and concentration processes.

They are both used prior to RO to reduce fouling

• MF/UF has been an integral part of pre-treating makeup water, as the technologies can help eliminate several
costly steps, like clarification and sand filtration

They can come in the same configurations

• Both MF and UF can be assembled into a system in similar ways. Depending on the facility’s needs, the MF/UF
system will generally include some type of:
1. Cartridge filter; to catch the large particles prior to MF/UF
2. Membrane module; whether MF or UF to separate out unwanted particles
3. Pressure pump; to help keep the liquid flowing through the process
4. Water supply; whether from a raw water source or other
5. Holding tanks; to store permeate
• MF/UF membranes can also be constructed similarly, separately from their overall system configuration.
Depending on the manufacturer and purpose for the membrane, both MF and UF membrane configurations can
include: Hollow fiber, Plate and frame, Spiral-wound, Tubular
• They can be made of the same materials

Pore size (MF) vs. MWCO (UF)

MWCO – Molecular weight cut-off or MWCO refers to the lowest molecular weight solute (in Daltons) in which 90% of
the solute is retained by the membrane or the molecular weight of the molecule (e.g. globular protein) that is 90%
retained by the membrane

PORE SIZE – Pore size is generally the distance between two opposite walls of the pore (diameter of cylindrical pores,
width of slip-shaped pores)

Fouling and concentration polarization phenomenon

• During an actual separation, the membrane performance (or better a system performance) can change very much
with time, and often a typical flux-time behavior may be observed: the flux trough a membrane decreases over
time
• Flux decline can be caused by several factors, such as concentration polarization, adsorption, gel layer formation
and plugging the pores
• All these factors induce additional resistances on the feed side to the transport across the membrane
• The extent of these phenomena is strongly dependent on the types of membrane process and feed solution
employed
𝐷𝑟𝑖𝑣𝑖𝑛𝑔 𝑓𝑜𝑟𝑐𝑒 ∆𝑃
• The flux through a membrane can be written as: 𝑓𝑙𝑢𝑥 = =
(𝑉𝑖𝑠𝑐𝑜𝑠𝑖𝑡𝑦∗𝑇𝑜𝑡𝑎𝑙 𝑟𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒) (𝜂∗𝑅𝑡𝑜𝑡
 • The various resistances contribute with
different extent to the total resistance (Rtot)
• Flux decline has a negative influence on the
economics of a given membrane process operation,
and for this reason measures must be taken to reduce
its incidence
• Because the membrane has an ability to
transport one component more readily than other
components, or in some cases completely retain the
solutes, there will be an accumulation of retained
molecules near the membrane surface
• This results in a highly concentrated near the membrane and this layer exerts a resistance towards mass transfer,
i.e. the concentration polarization resistance Rcp
• Polarization phenomena always occur and are inherent to membrane separation process
• The concentration of the accumulated solute molecules may become so high that a gel layer can be formed which
exert the gel layer resistance Rg
• This mainly happens when the solution contains proteins
• With porous membranes it is possible for some solutes to penetrate into the membrane and block pores, leading
to the block – pore resistance Rp
• Finally, a resistance can arise due to adsorption phenomena, i.e. the resistance Ra
• Adsorption can take place upon the membrane surface as well as within the pores themselves
• Under a pressure and during MF/UF separation near
the membrane surface a boundary layer is formed
where the concentration increases and reaches a
maximum value at the membrane surface (cm)
• The convective flow of solutes towards the
membrane may be written as J*C
• If the solute „p” is not completely retained by
membrane, there will be a solute flow through the
membrane equal to J*Cp
• The accumulation of the solute „p” at the membrane
surface leads to diffusive back flow towards the bulk
of the feed
• The solute concentration at the membrane surface attains a high value and a maximum concentration – the gel
concentration (cg)
• The gel concentration depends on the size, shape, chemical structure and degree
of solvation but is independent of the bulk concentration
• Continuous flux decline is a result of membrane fouling ((ir)revisable deposition
of retained particles, colloids, emulsions, suspensions, macromolecules, salts,
etc. on or in membranes)
• Fouling occurs mainly in MF/UF (porous membranes which are implicitly
susceptible to fouling)
• Fouling depends on physical and chemical parameter such as concentration,
temperature, pH, ionic strength and specific integrations (hydrogen bonding,
dipole-dipole interactions)
• The initial decline of flux is caused by concentration polarisation, further by
fouling mainly
Methods to reduce flux decline

• Pre-treatment of feed solution

Pre-treatment methods employed include: heat treatment, pH adjustment,

addition of complexing agent (EDTA etc.), chlorination, adsorption onto active
carbon, chemical clarification, pre-microfiltration or pre-ultrafiltration

• Membrane properties

─ Narrow pore size distribution can reduce a fouling

─ Using a hydrophilic rather than hydrophobic membranes can also help (e.g.
proteins adsorb more strongly at hydrophobic surfaces and are less readily
removed than at hydrophilic surfaces)

─ Using a (negatively) charged membranes (e.g. in the presence of (negatively)

charged colloids in the feed)

• Module and process conditions

─ Fouling phenomena diminish as concentration polarisation decreases. Concentration polarisation can be reduced
by increasing the mass transfer coefficient (HIGH FLOW VELOCITIES)

─ Using various kinds of turbulences promoters

─ Fluidised bed system

─ Rotary module system

─ Chemical cleaning - This is the most important method

for reducing fouling, with a number of chemicals being
used separately or in combination.

─ Mechanical cleaning - This system can only applied in

tubular system using oversized sponge balls

─ Hydraulic cleaning

This methods include back-flushing (only applicable to MF and UF membranes), alternate pressurising and
depressurising and by changing the flow direction at given frequency

• After a given period of time, the feed pressure is released and the
direction of the permeate reversed from the permeate side in
order to remove the fouling layer within the membrane or the
membrane surface
Nanofiltration

• Nanofiltration (NF) is used when low molecular weight solutes

such as inorganic salts or small organic molecules such as glucose
have to be separated from the solvent
• The difference between ultrafiltration and nanofiltration lies in the
size of the solute
• Denser membranes (than in UF) are required with a much higher
hydrodynamic resistance
• NF membranes can be considered as being intermediate between
open porous membranes (MF/UF) and dense nonporous
membranes – reversed osmosis
• Because of their higher membrane resistance a much higher
pressure must be applied to force the same amount of solvent
trough the membrane
• The solutions containing the low molecular weight solutes have a
much higher osmotic pressure than the macromolecular solutions
used in UF. The osmotic pressure of seawater, for example, is about
25 bar
• The membrane is permeable to the solvent (water in case of salt
solution) but not to the solute (salt)
• In order to allow water to pass through the membrane, the
applied pressure must be higher than the osmotic pressure
• Water flows from the dilute solution (pure water) to the
concentrated solution if the applied pressure is smaller than the
osmotic pressure. When the applied pressure is higher than the
osmotic pressure water flows from the concentrated solution to
the dilute solution.
• The effective water flow can be represented by equation (if it is
assumed that no solute permeates through the membrane): 𝐽𝑤 = 𝐴 ∗ (∆𝑃 − ∆𝜋)
• In practice, the membrane may be a little permeable to low molecular solutes and hence the real osmotic pressure
difference across the membrane is not Δπ but σΔπ, where σ is the reflection coefficient of the membrane towards
that particular solute
When R<100% , then σ<1 : 𝐽𝑤 = 𝐴 ∗ (∆𝑃 − 𝜎∆𝜋)
• The value of water permeability “A”, which is a function of the distribution coefficient (solubility) and the diffusivity
𝐷𝑤 ∗𝑐𝑤 ∗𝑉𝑤
(with the lowest value being observed for denser membranes), is a constant for a given membrane: 𝐴 =
𝑅∗𝑇∗𝛥𝑥
• The solute flux can be described by: 𝐽𝑠 = 𝐵 ∗ ∆𝐶𝑠
• Where B is the solute permeability coefficient and Δcs the solute concentration difference across the membrane
𝐷𝑠 −𝐾𝑠
(Δcs=cf-cp). The B value is lowest for high rejection membranes: 𝐵 = 𝛥𝑥
K – permeability constant
• Solute permeability coefficient “B” is a function of the diffusivity
and the distribution coefficient (solubility).
• The pressure used in NF ranges form 20 to 100 bar (much higher
than in UF)
• In contrast to UF and MF, the choice of material directly influences
the separation efficiency trough the constants “A” and “B”
• The constant “A” must be as high as possible whereas the constant
“B” must be as low as possible to obtain an efficient separation
• The membrane (material) must have a high affinity for the solvent
(mostly water) and a low affinity for the solute
• This implies that the choice of material is very important because it determines the intrinsic membrane properties
• The difference to UF/MF, where the dimensions of pores in the material determine the separation properties
Membranes for nanofiltration

• The flux through the membrane is as important as its selectivity towards various kinds of solute
• The flux through the membrane can be improved by reducing thickness
• The flux is approximately inversely proportional to the membrane thickness and for this reason most NF
membranes have an asymmetric structure with thin dense top layer (thickness ≤ 1µm) supported by a porous
sublayer (thickness ≈50 – 150µm)
• the resistance towards transport being determined mainly by the dense top layer
• Two different types of membrane with an asymmetric structure can be distinguish:
1. Integral – asymmetric membranes
2. Composite membranes

Composite membranes have a thin dense homogeneous

polymer skin formed over a microporous support film. While
integrally skinned membranes are made in a one-step
procedure, composite membranes are made in two or more
steps.

Integral – asymmetric membranes

• In these membranes, both top layer and the sublayer consist of the same material
• They are prepared by phase inversion techniques
• An important class of asymmetric NF membrane prepared by this techniques are the cellulose esters, especially
cellulose diacetate and triacetate
• They are suitable for desalination because of their high permeability towards water in combination with a (very)
low solubility of salt
• Their stability against chemicals, temperature and bacteria is very poor
• Typical operations conditions of such membranes are over the pH range 5 to 7 at temperature below 30°C thus
avoiding hydrolysis of the polymer
• The extent of this hydrolysis decreases as the degree of acetylation increases (cellulose diacetate is less resistant
than cellulose triacetate)
• Other materials that have been used frequently for NF membranes are aromatic polyamides – these materials
also show a high selectivity towards salts but their water flux is lower (than cellulosic materials)
• Polyamides can be used over a wider pH range, approximately 5 -9
• The main drawback of polyamides (or of polymers with an amide group –NH-CO in general) is their susceptibility
against free chlorine Cl2 which causes degradation of the amide group
• Asymmetric membrane as well as symmetric membranes have been prepared from these polymers by melt or dry
spinning to obtain hollow fibbers with small dimensions (outside diameters of such hollow fibbers <100µm)
• The third class of material that have been used are the polybenzimiadazoles and polimides
Aplication of NF

• NF can be used in principle for wide range of applications, which can be roughly classified as solvent purification
(where the permeate is a product) and solute concentration (where the feed is the product)
• Most of applications are in the purification of water, mainly desalination of brackish and especially seawater to
produce potable water
• NF is used as a concentration step particularly in the food industry (concentration of fruit juice, sugar coffee), the
galvanic industry (concentration of waste streams) and the diary industry (concentration of milk prior to cheese
manufacture)
• An example of NF application in agave
juice concentration and agave fructants
production:

Summary of nanofiltration

Main applications:

• Desalination of brackish and seawater

• Production of ultrapure water (electronic industry)
• Concentration of food juice and sugars (food industry), and the
concentration of milk (food industry)

Reverse osmosis

• Reverse osmosis (RO) – is a water purification process that

uses a partially permeable membrane to remove ions,
unwanted molecules and larger particles from drinking
water.
• A high pressure pump is used to increase the pressure on
the salt side of the RO and force the water to flow across
the semi-permeable membrane, leaving most part
(approximately 95%-99%) of the dissolves salts behind the
stream
• Reverse osmosis is capable of removing dissolved salts
(ions), particles, colloids, organics, bacteria and pyrogens
from the water
• RO membranes do not remove gases like CO2 or O2. These gases are not highly ionized (charged) while they’re in
solution and have a very low molecular weight
• The water is electrically neutral, it means that the sum of cations equals the sum of anions when expressed as
equivalents. This also happens with RO permeate, for each cation that passes through the membrane, an anion
must pass through as well.

Sorption – diffusion mechanism. Transport through nonporous membranes.

• Basically, the transport of a gas, vapour or liquid trough a dense, nonporous membrane can be described in terms
of solution – diffusion mechanism, i.e.:
• PERMEABILITY (P) = SOLUBILITY (S) x DIFFUSIVITY (D)
• Solubility is a thermodynamic parameter and gives a measure of the amount of penetrant sorbed by the membrane
under equilibrium conditions
• The diffusivity is a kinetic parameter which indicates how fast a penetrant is transported through the membrane
• Diffusivity is dependent on geometry of the penetrant
• The diffusion coefficient is concentration-dependent with
interacting systems
• Even large (organic) molecules having the ability to swell the
polymer can have large diffusion coefficient

Comparison of NF and RO – how RO and NF are similar?

• Both RO and NF benefit from pre-treatment

With the smallest pores of all filtration membranes, RO and NF membranes can foul quickly if larger particles aren’t
removed by upstream filtration technologies, such as media filtration or MF/UF.

Pre-treating process streams to remove these

problematic particulates can also reduce the
amount of energy needed to maintain adequate
pressure in a RO/NF system.

When process streams are sufficiently pre-treated,

RO/NF membranes can have a long service life and
require relatively little maintenance.

• Both RO and NF require pressure to function

While RO and NF are fairly efficient purification technologies, both demand energy to move water across their
membranes. This is because their fine pores result in a high concentration of salts and other compounds on the
retentate side of the membrane, and as a result, enough pressure must be applied that the water is able to overcome
the osmotic pressure that causes water to resist flowing through the membrane.

• Both are used in place of conventional treatment technologies

Conventional water treatment trains typically consist of several unit processes, including: coagulation, flocculation,
sedimentation, clarification, ion exchange, and filtration.

While conventional treatment systems are effective in removing dissolved solids, they do so through a complex series
of steps that often demand a large footprint and investment in an assortment of specialized equipment and chemical
agents.

As RO/NF membrane technologies have become more efficient and affordable, they are increasingly being adopted as
more compact, efficient, and environmentally-friendly alternatives to conventional treatment trains.

Comparison of NF and RO – how RO and NF are different?

• While RO and NF are very similar, they can be distinguished based on the size of particulates that each is able to
remove
• Comparatively, RO and NF are capable of removing finer contaminates than MF and UF, with applications including
the removal of hardness, nitrates, sulfates, total dissolved solids (TDS), heavy metals, radionuclides, and organic
macromolecules from process and waste streams.
Reverse osmosis

• RO is the finest of all membrane filtration systems, with extremely small pores capable of removing particles as
small as 0.1 nm
• RO is used chiefly for desalination, as in potable water generation from seawater or brackish water sources
• RO membranes are very efficient at removing all ions, large and small

Nanofiltration

• NF delivers slightly coarser filtration than RO, with

the ability to remove particles as small as 0.002 to
0.005 μm in diameter.
• NF removes harmful contaminates, such as pesticide
compounds and organic macromolecules, while
retaining minerals that RO would otherwise remove.
• Nanofiltration membranes are capable of removing
larger divalent ions such as calcium sulfate, while allowing smaller monovalent ions such as sodium chloride to
pass through.

Obtaining a drinking water in the World Desalination of (brackish) seawater

• Desalination is one of the most important applications of membrane process

• The most appropriate technique for this purpose is RO or NF
• High performanceRO membrane exhibit a slat rejection >99%
• This single – stage RO system can give a product purity of about 300 ppm of salt
• To improve a quality further, a two-stage (or multi-stage) system is often used
• Although seawater is a relatively clean feed stream, pre-treatment is necessary to reduce fouling and to avoid a
membrane damage
• Flocculation agents such as iron chloride or polyelectrolytes are added in order to remove suspended solids, but
scaling can be a very severe problem

Scaling

• Scaling is the precipitation of salts which arises because of tis

solubility products have been exceeded.
• The precipitation of calcium salts (CaSO4, CaCO3) or silica (SiO2) in
particular at the membrane surface can cause a problem in the case
of seawater
• To reduce scaling, the pH adjusted by the addition of acid (calcium,
barium, magnesium salts will not precipitate at low pH values and
silica at high pH value)
• Chlorine is then added to remove bacteria and algae; with membrane
materials which are not resistant to free chlorine (e.g. polyamides), a
treatment with sodium hydrogen sulphite (NaHSO3) is necessary to
remove the chlorine

Desalination

• the saline feed-water is drawn from oceanic or underground sources

• The product water of the desalination process is generally water with less than 500 mg dissolved solids, which is
suitable for most domestic, industrial, and agricultural uses
• A by-product of desalination is brine
• Brine is a concentrated salt solution (with more than 35 000 mg dissolved solids) that must be disposed of, generally
by discharge into deep saline aquifers or surface waters with a higher salt content
• Brine can also be diluted with treated effluent and disposed of by spraying on golf courses and/or other open space
areas
• The major energy requirement is for the initial
pressurization of the feed=water. For brackish
water desalination the operating pressures
range from 15 to 25 bar, and for seawater
desalination from 55 to even 70 bar
• RO/NF system to desalination consists of four
major components/processes:
1. pre-treatment,
2. pressurization
3. membrane separation
4. post-treatment stabilization

• Pretreatment (1):

The incoming feed-water is pre-treated to be compatible with the membranes by removing suspended solids,
adjusting the pH, and adding an inhibitor to control scaling caused by constituents such as calcium sulphate.

• Pressurization (2):

The pump raises the pressure of the pre-treated feed-water to an operating pressure appropriate for the membrane
and the salinity of the feed-water.

• Separation (3):

The membranes inhibit the passage of dissolved salts while permitting the desalinated product water to pass through.

Applying feed-water to the membrane assembly results in a freshwater product stream and a concentrated brine
reject stream.

Because no membrane is perfect in its rejection of dissolved salts, a small percentage of salt passes through the
membrane and remains in the product water.

RO/NF membranes come in a variety of configurations. Two of the most popular are spiral wound and hollow fine
fiber membranes – they are generally made of cellulose acetate, aromatic polyamides.

• Stabilization (4):

The product water from the membrane assembly usually requires pH adjustment and degasification before being
transferred to the distribution system for use as drinking water. The product passes through an aeration column in
which the pH is elevated from a value of approximately 5 to a value close to 7. In many cases, this water is discharged
to a storage cistern for later use.

RO home system

• When water first enters an RO system, it goes through pre-filtration.

Pre-filtration typically includes a carbon filter and a sediment filter to
remove sediment and chlorine that could clog or damage the RO
membrane.
• Next, water goes through the RO membrane where dissolved particles,
even too small to be seen with an electron microscope, are removed.
• After filtration, water flows to the storage tank, where it is held until
needed.
• RO system continues to filter water until the storage tank is full and
then shuts off
• Once you turn on your drinking water faucet, water comes out of the
storage tank through another post-filter to polish drinking water before it gets to your faucet.
Concentration difference as a driving force

• In many processes, including those in nature, transport

proceeds via diffusion rather than convection
• Substances diffuse spontaneously from places with higher
chemical potential to those where the chemical potential is
lower

• Processes which makes use of a concentration difference as

the driving force are gas separation, pervaporation, and
liquid membrane processes

• Dialysis is also membrane process where the concentration difference is a driving force but based on a different
mechanism

• Whereas a MF, UF and NF are more or less similar processes, dialysis, gas separation, pervaporation and separation
use liquids membranes differ quite considerably from each other

• In gas separation and pervaporation to improve an effectiveness of process the pressure differences is applied as
well (to obtain a higher volume of permeate)

• The basic feature that they have in

common is the use of nonporous
asymmetric (composite) membrane

• It was discussed (on second lecture) that

the permeability of a gas through an
elastomeric and glassy material may differ
by more than five orders of magnitude,
despite both materials being nonporous

Permeability

• This difference arises from large differences in segmental motion which is

very restricted in the glassy state
• The presence of crystallites can further reduce the mobility
• A factor that enhances segmental mobility, or chain mobility in general, is
the presence of low molecular penetrants
• The concentration of penetrant inside the polymeric membrane is
determined mainly by the affinity between the penetrant and the polymer

Gas separation

• Gas separation is possible even with the two extreme types of membrane considered, i.e. porous and nonporous
• The transport mechanisms through these two types of membrane are completely different

Permeability of gasses

• In gas separation there is hardly any interaction between the gas molecules and the
membrane material and the gas concentration in the membrane is very low
• The gas molecules must diffuse trough a rigid membrane structure with the state of the polymer being hardly
effected by their presence. In contrast, with liquid penetrants the solubility in the membrane may be appreciably
higher with results in an enhanced chain mobility.
Gas separation in porous membranes

• When gas transport takes place by viscous

flow (as in case of MF membranes, for
example), no separation is achieved because
the mean free path of the gas molecules is
very small relative to the pore diameter
• By decreasing the pore diameter of the pores
in the membrane the mean free path of the
gas molecules may become greater than the
pore diameter

Gas separation through nonporous membranes

• Gas separation through nonporous

membranes depends on differences in the
permeability of various gases through a given
membrane
• Henry’s law is manly applicable to amorphous
elastomeric polymers

Diffusivity (D)

• The other factor affecting on permeability is a diffusivity

• It depends mainly on two factors:
1. Molecular size of the gaseous penetrant – it is reflected in the diffusion coefficient (i.e. the smaller its size the
higher the diffusion coefficient)
2. The choice of polymer
𝑘∗𝑇
• The thermodynamic diffusion coefficient can be expressed: 𝐷𝑇 =
𝑓
• Where “f” is the frictional coefficient and Stokes’ law demonstrates that the frictional coefficient is related to the
size of the diffusing molecule by: f=6*π*η*r
• The diffusion coefficient is inversely proportional to the molecular size
• The diffusion coefficient also depends strongly on the nature of the polymer

Solubility (S)

• The constant developed experimentally (expressed as a pressure)

• Is connected with given components in given temperature

Aspects of gas separation

• To describe the fundamentals of gas separation, however, other factors relating to the nature of the polymer (i.e.
chemical structure) need to be consider
• Two parameters are important in this context:
1. The glass transition temperature
2. The crystallinity
• The glass transition temperature determines whether a polymer is in the glassy state
• In general, permeability through a rubbery material (elastomer) is much higher relative to glassy polymers because
of the higher mobility of the chain segments
• In contrast, the selectivity of glassy polymers is higher
The typical schematic drawing of a gas separation process

• The real separation factor also depends on the

pressure ratio across the membrane
• The separation efficiency is a maximum when the
pressure ratio is high (or pf/pp => 0)
• The selectivity decreases as the pressure ratio
decreases
• The high efficiency of separation can be established
either by applying a high pressure on the feed side and/or maintaining a low pressure on the permeate side

Applications of gas separation

• CO2/CH4

This kind of separation arises in many applications: the purification of CH4 from landfill drainage gas, the purification
of CH4 from natural gas and recovery of CO2 in enhanced oil recovery

• H2 or He from other gases

Hydrogen and helium have relatively small molecular size compared to other gases and exhibit high selectivity ratios
in glassy polymers. Application can be found in the recovery of H2 from purge gas streams ammonia synthesis,
petroleum refineries and methanol synthesis

• H2S/CH4

Besides CO2 in natural gas H2S is often present in appreciable

concentrations. The concentration of this very toxic, highly
corrosive gas has to be reduced to less than 0.2%

• O2/N2

Separation can be effected to obtain both oxygen-enriched air

and nitrogen-enriched air. Nitrogen-enriched air (95-99.9%)
can be used as an inert gas in the blanketing of fuel tanks

• H2O from gases

All kinds of gases can be dried by the removal of water

• SO2 from smoke gas

Desulphurisation of smoke gas

Pervaporation

• This is a membrane process in which a liquid is maintained

at the atmospheric pressure on the feed or upstream side
of the membrane and where the permeate is removed as a
vapour because of a low vapour pressure existing on the
permeate or downstream side
• This low (partial) vapour pressure can be achieved by
employing a carrier gas or using vacuum pump
• The (partial) downstream pressure must
be lower than the saturation pressure at
least
• Pervaporation is a complex process in
which mass and heat transfer occurs
• The membrane acts as a barrier layer
between a liquid and a vapour phase
implying that a phase transition occurs in
going from feed to the permeate
• The heat vaporisation must be supplied
• Transport can be described by means of a solution-diffusion mechanism where the selectivity is determined by
selective sorption and/or selective diffusion
• In fact, the same type of membrane or membrane material can be used both gas separation or pervaporation
• Essentially, the pervaporation process involves a sequence of three steps:
─ Selective sorption into the membrane on the feed side
─ Selective diffusion through the membrane
─ Desorption into a vapour phase on the permeate side
• However, the affinity of a liquid towards a polymer is generally much higher than that of a gas in a polymer so that
the solubility is much higher
• This effect can be noticed in the case of organic vapour which exhibit much higher permeabilities than gases such
a nitrogen
• In gas separation the selectivity towards a mixture can be estimated from the ratio of the permeability coefficients
of the pure gases
• However, with liquid mixtures the separation characteristics are far different from those a pure liquid because of
coupling phenomena and thermodynamic integrations

Methods of PV process

• The installation where the permeate is in vacuum condition (a)

• Inert gas application (elution of ingredients passing through a membrane) (b)

• PV is very useful in the area where distillation is impossible – around

azeotropic point (for ethanol – 95,6% of ethanol)

Membranes for pervaporation

• For PV the nonporous membranes are required

preferably with an anisotropic morphology, an
asymmetric structure possessing a dense top layer and
open porous sublayer (asymmetric and composite
membranes)
• The requirements for the substructure are in fact the same as for gas separation membranes:
• An open substructure to minimise resistance to vapour transport an to avoid capillary condensation
• A high surface porosity with a narrow pore size distribution
• Pressure loss on the permeate side results in an increase in activity and hence in a decrease in flux
• When the pores (in support) are too small, the pressure loss maybe so high that even capillary condensation may
occur
• If the pores in the support layer are too large it is difficult to apply a thin selective layer directly upon the support
– it is a very important that the surface porosity should be high

Applications of PV

• This process is used mainly to separate (or better remove) a small amount of liquid from a liquid mixture
• When highly selective membranes are used, only the heat of vaporisation of the almost pure permeate has to be
supplied
• This separation becomes very attractive when the liquid mixture exhibits an azeotropic composition (where the
liquid and vapour have the same composition)
• ‘Ordinary’ distillation can not be used to separate such mixtures
• Mixtures of an organic solvent with water exhibit an azeotrope in the composition region of the pure organic
solvent. This is very advantageous to use pervaporation to dehydrate these types of mixture
• They are also used in other areas such as food and pharmaceutical industries to concentrate heat-sensitive
products, for remove volatile organic contaminants from waste water or in analytical applications to enrich a given
component for quantitative detection

Applications – Aqueous mixtures

Two main classes can be distinguished as: either a small amount of water has to be removed from an organic solvent
(dehydration) or a small amount of organic solvent has to be removed from water

• Dehydration

- Removal of water from alcohols or other organic solvents. Even traces of water can be removed

• Removal of organic components from water

- Alcohol from fermentation broth

- Volatile organic contaminations (aromatic, chlorinated hydrocarbons) from waste water

Applications – non-aqueous mixtures

A further sub-calssification can again be made:

• Polar/nonpolar

- Alcohols/aromatics (methanol/toluene)

- Alcohols/aliphatics (ethanol/hexane)

• Aromatics/aliphatics

- Cyclohexane/benzene

- Hexane/toluene

• Saturated/unsaturated

- Butane/butane

• isomers

- C-8 isomers (o-xylene, m-xylene, p-xylene,

styrene, ethylbenzene)