You are on page 1of 124

CEEN 572

Environmental Pilot Lab

Water Treatment
Membrane Processes
- Introduction -
Membrane Separation
Membrane

Feed
Permeate

Particle or Solvent
Solute Molecule
Definition of Membrane Process
 In a membrane separation process, a feed consisting
of a mixture of two or more components is partially
separated by means of a semipermeable barrier
through which one or more species move faster than
the other species
 In water and wastewater treatment applications,
membrane processes are used as a solid/liquid
separation process. In this case, water is more
readily transported through the membrane than
solids (both suspended and dissolved)
Classification of
Membrane Operations
 Driving force
 Mechanism of separation
 Membrane structure
 Phases in contact
Classification of
Membrane Operations
 Pressure-driven membrane operations
 Permeation operations
 Dialysis operations
Pressure-driven
Operations
Microfiltration (MF)
Ultrafiltration (UF)
Nanofiltration (NF)
Reverse Osmosis (RO)
Pressure-driven Membrane Processes
Pressure-driven Membrane
Processes
RO
 The solvent is transferred through a dense
membrane tailored to retain salts and low-
molecular-weight solutes
 To produce “pure” water from saline solution, feed
pressure must exceed the osmotic pressure of the
feed solution
 In order to obtain economically viable flows, at least
twice the osmotic pressure must be exerted as
hydraulic pressure (e.g., 50-80 bars (700-1,100 psi)
for seawater)
Pressure-driven Membrane
Processes
NF
 Sometimes referred to as low-pressure RO or
membrane softening process
 Lies between RO and UF in terms of selectivity of the
membrane
 Designed to remove multivalent ions but can remove
sodium and chloride fairly well
 Looser NF membranes are more like UF and tighter
NF membranes more closely resemble RO
 Recently has been employed for organic control
 Typical operating pressure: 5-14 bar (70-210 psi)
Pressure-driven Membrane
Processes
UF
 Considered as a clarification and disinfection operation
 Membrane is porous and rejects most macromolecules,
microorganisms, and all types of particles
 Osmotic pressure effects are negligible
 Typical operating pressure: 0.5-5 bar (7-70 psi)

MF
 Major difference between MF and UF is pore size –
0.05-5 micron for MF
 Primary application is particulate removal (clarification)
 Typical pressures like UF
Selection of Membrane Processes

‘loose’ and ‘tight’


Membranes in Treatment of
Drinking Water
 The application of specific pressure-driven
membrane process is highly dependent on the
characteristics and quality of the source water

◼ Surface water: MF, UF, NF


◼ Groundwater (fresh): MF, UF
◼ Groundwater (brackish): MF/UF pretreatment, NF, RO
◼ Seawater: MF/UF pretreatment, NF, RO
Membranes in Treatment of
Wastewater
 The application of specific pressure-driven
membrane process in wastewater treatment is
highly dependent on the characteristics/quality of
the source water and the pretreatment process/es
used
◼ Raw wastewater: MF/UF, MBR, FO (not mainstream…yet)
◼ Effluent: MF/UF pretreatment, NF, RO
Membrane Technologies and their
Traditional Counterparts

Membrane Constituents Removed Comparable traditional Water


Separation Treatment Method
Technology
MF Bacteria and large colloids; Ozonation-UV, chlorination, sand
precipitates and coagulates filtration, bioreactors, coagulation-
sedimentation
UF All of the above + viruses, high MW Sand filter, bioreactor, activated
proteins, organics carbon
NF All of the above + divalent ions, large Lime-soda softening, ion exchange
monovalent ions, color, odor
RO All of the above + monovalent ions Distillation, evaporation, ion
exchange
ED/EDR Dissolved ionic salts Ion exchange
Target Solutes
 MF: Microbes (protozoa and bacteria)
Turbidity (particles and colloids)
 UF: Same as MF + viruses, “some” NOM
 NF: Same as UF + NOM, SOCs (e.g., Atrazine),
Divalent cations (Ca2+, Mg2+, Zn2+, Cd2+, etc.),
Polyvalent anions (SO42-, PO43-, AsO43-, CrO42-, etc.)
 RO: Same as NF + simple ions (TDS, NO3-, ClO4-)
 MF + Coagulant: viruses, NOM (also fouling reduction)
 UF + PAC: SOCs, NOM (also fouling reduction)
 Submerged MF and UF: Fe and Mn (aeration),
NOM (with coagulant), SOCs (with PAC)
Ranges of Pressure and Flux
Membrane Pore Size Pressure Flux
Class or MWCO psi kPa gfd LMH
(μm or Dalton) (L/m2·hr)
(gal/ft2·day)
MF 0.1 – 0.5 mm 1.4 - 14 10-100 60 – 120 100 – 200
UF 1 – 100 kD 7.0 - 70 50-500 30 – 60 50 – 100
NF 100 – 500 D 100 - 400 700-2,800 15 – 30 20 – 50
RO n/a 200 - 1,000 1400-7,000 15 – 30 20 – 50
MF 0.2 mm -1.4 -10
(Immersed)
UF 0.04 mm -7.0 -50 20 - 50 35 – 85
(Immersed)

conversions
1 atm = 101.3 kPa (kN/m2) = 14.7 psi gfd = LMH x 1.7
1 kPa = 0.145 psi or 1 psi = 6.90 kPa
1 psi = 0.068 atm
Ranges of Energy Consumption
Membrane Recovery Pressure Energy consumption
Class kWh per
psi kPa 1,000 gal m3
MF 94-98 15 100 0.1 0.4
UF 70-80 75 525 0.8 3.0
NF 80-85 125 875 1.4 5.3
LPRO 70-85 225 1,575 2.7 10.2
RO 70-85 400 2,800 4.8 18.2
ED 75-85 2.5 9.5
Permeation
Operations
Gas Permeation (GP)
Gas Diffusion
Pervaporation (PV)
Membrane Stripping (MS)
Membrane Distillation (MD)
Engineered Osmosis (EO)
Gas Separation
 Industrial Applications of GP:
 Separation of H2 from CH4, (H2 permeation rate
through dense membrane is very high)
 Adjustment of H2-to-CO ratio in synthesis of gas
 O2 enrichment of air
 Removal of CO2
 Drying of natural gas and air
 Removal of He and organic
solvents from air
Gas Separation
N2

VENT

AIR
Pervaporation (PV)
 Liquid/vapor separation – liquid partially vaporized through a
dense membrane
 Solvent dissolves in the polymeric membrane, diffuses, and
evaporates on the permeate side
 Rate of transfer of a constituent depends on its solubility in
the membrane
 Activity difference maintained by partial vacuum on the
permeate side
 Separation of solvents
 PV is used to separate dissolved volatile
organic compounds (VOCs) from
aqueous solutions
Pervaporation (PV)
 PV membrane separation systems are used in food
processing, gas separation, and water treatment.
Examples of applications include:
◼ Recovery of flavor compounds from food industry process
streams
◼ Recovery of ethanol from fermentation
and food industry process streams
◼ Removal of organic contaminants
from waste streams
◼ C2H4/C3H8
Membrane Stripping and
Membrane Distillation (MD)
 Separation of volatile constituents from solution:
◼ Removal of volatile solutes or volatile solvent
 Microporous hydrophobic membrane
◼ Polypropylene, PTFE, PVDF, nylon
 Heated aqueous feed solution is brought into
contact with feed side of the membrane
 The hydrophobic nature of membrane prevents
penetration of aqueous solution into the pores
 Lower vapor pressure on the permeate side of the
membrane induces evaporation through the pores
 Pores remain dry throughout the process !
Membrane Stripping and Membrane
Distillation – Basic Configurations

Feed Clean
Solution Water
Air Gap
Feed Sweep Feed
Solution Gas Direct Contact MD (DCMD) Solution

Sweeping Gas MD Air Gap MD


= Membrane Stripping

Feed Vacuum
Solution

Vacuum MD (VMD)
Engineered Osmosis
 Forward osmosis (applications: wastewater treatment,
pretreatment, desalination, concentration)

Brine
Reconcentration Draw
Feed
Solution

DP=Dp

Brine Feed

Osmosis Forward Osmosis (FO)


“engineered osmosis”
UFO-MBR System
Wastewater Feed
Stream
Diluted DS

DS
Anoxi Tank
c
Tank RO
Concentrated Permeate
DS Tank

UF Permeate and
Backwash Tank
Dialysis
Operations
Dialysis (DIA)
Donnan Dialysis
Electrodialysis (ED)
Dialysis Operations
 The solute is transferred through the membrane, not
the solvent
 The driving force is activity or an electrical potential
difference

 Dialysis (DIA)
◼ The driving force is a transmembrane concentration
difference
◼ Selective passage of ions and low-molecular-weight
solutes and rejection of larger colloidal and high-
molecular-weight solutes
◼ Main application is hemodialysis
Electrodialysis (ED)
 Operation by which ions diffuse through ion-
exchange membrane under the influence of electric
potential
 Ion exchange membranes
◼ AEM: quaternary amines
◼ CEM: carboxylates, sulfonates
 Plate and frame stacks
 Efficient for desalination
of brackish water
Electrodialysis Reversal (EDR)
 EDR is a variant of ED
 The same membranes are used to provide a continuous self-
cleaning electrodialysis process which uses periodic reversal
of the DC polarity to allow systems to run at higher recovery
rates
 Polarity reversal causes the concentrating and diluting flow
streams to switch after every cycle
 Any fouling or scaling constituents are removed when the
process reverses, sending fresh product water through
compartments previously filled with concentrated waste
streams
 EDR systems operate with higher concentrations in the brine
or concentrate streams with less flow to waste
http://www.gewater.com/images/multimedia/edr/index_flash.html
Electrodialysis Reversal (EDR)
 Applications for EDR technology include municipal
drinking water, industrial process water, and
wastewater reuse projects
 Municipalities find EDR successful for the removal of
radium, arsenic, and perchlorate as well as
desalination of well and surface waters
Other Classifications
Separating Mechanisms
 Separation based on difference in size (sieving)
◼ MF, UF, DIA
 Separation based on difference in solubility and
diffusivity of material in the membrane (solution-
diffusion mechanism)
◼ GP, PV, RO, FO
 Separation based on difference in charges of the
species to be separated (electrochemical effects)
◼ ED, EDR
Rejection Capabilities
(pressure-driven processes)
 RO membranes are typically characterized by manufacturers
in terms of NaCl rejection, e.g., 96% or 99.9% NaCl rejection
 NF membranes may be characterized in terms of NaCl or
MgSO4 rejection or they may be characterized in terms of
molecular weight cut-off (MWCO)*, e.g., 98% MgSO4 and 80%
NaCl rejection
 UF membranes are typically characterized using MWCO,
e.g., 13,000 or 80,000 MWCO
 MF membranes are typically characterized by pore size, e.g.,
0.1 or 1 µm

* MWCO is determined by fitting rejection data of acromolecules (e.g.,


dextrans or proteins)
Porosity
 Porous membranes (MF, UF, NF, DIA)
◼ Macroporous: > 50 nm
◼ Mesoporous: 2 – 50 nm
◼ Microporous: < 2 nm
 Nanoporous membranes
◼ Dense media
◼ Diffusion of species takes place in the free volume present
between the macromolecules chains of the membrane
material
 IX membranes
◼ Specific type of nanoporous membranes
Morphology
Symmetric

Cylindrical Porous Porous Homogeneous


(resistance to mass transfer is determined by total membrane thickness)

Asymmetric – Single Material


top layer

Porous Porous with Top Layer

Asymmetric – Composite
dense skin layer (0.1 to 0.5 µm)
porous membrane (50 to 150 µm)

(resistance to mass transfer determined by skin layer thickness)


Structure of Membranes
Geometry / Packaging
 Flat-sheet membranes (spiral wound, plate-and-
frame)

 Tubular membranes (shell-and-tube, immersed)


◼ Tubes
◼ Capillaries
◼ Hollow fibers
Geometry / Packaging
Spiral Wound Module
Single Element vs. Bank or Array

Roga Module 1
ca. 1964
Spiral Wound Module Installation
Hollow Fiber Membrane Single
Fiber (left) vs. Module (right)
Hollow Fiber Membrane
Module (left) vs. bank (right)
Submerged Membrane – MF/UF
 Uses
◼ Surface water treatment
◼ Pretreatment for RO
◼ Membrane bio reactors (MBR)
◼ Filtration for non-potable reuse (add MF after secondary
WW treatment and produce water for irrigation)
 Operation
◼ Membranes are immersed in basin of feed water
◼ Operate under suction
 Advantages
◼ Operate at lower pressures than pressurized systems
◼ Less fouling potential - good for wastewater treatment
◼ Membrane cleaning and fixing
Submerged Membranes

Feed Water Filtrate


To RO

Air Bubble
Scouring of MF/UF
Membrane Hollow Fiber
Surface Basin
Membranes
System Configuration
Mix Fill and
React (1 hr)

BR1

MT
BR2 React
Draw (1 hr)

SEPTIC TANK Permeate Tank


Immersed Membrane (GE / Zenon)
Cassette vs. System
Immersed
Membrane
Backwash
(USFilter)
Submerged MBR - Zenon
Submerged Membrane Designs
Flow Configuration
Cross Flow vs. Dead End Filtration
Cross Flow Operation
Feed Concentrate

Permeate

Dead End Operation


Feed

Permeate
Cross Flow Operation

Feed vt Concentrate
vd

Permeate

 Feed flow is parallel to membrane surface


 Retained particles are scoured
 Have concentrate stream
 Preferable for concentrated solutions to control
thickness of deposit on membrane (fouling)
Dead End Operation
Feed

Permeate

 Feed flow is perpendicular to membrane surface


 Retained particles form a cake layer on surface
 No concentrate stream
 Preferable for dilute solutions due to lower energy
requirements (pumping)
Comparison of Cross-flow
Membrane Configurations

Operating
Packing Membrane Fouling
Cost Pressure Cleanability
Density Types Resistance
Capacity

Traditional
Spiral-Wound Low High High Many Fair Fair

Hollow Fiber UF-High


UF-Low UF-Good UF-Good
Low RO-Very Few
RO-High RO-Poor RO-Poor
High
Tubular UF-
High Low Few Very Good Very Good
Moderate

Plate & Frame


High Moderate High Many Fair Fair

Adapted from "Select Engineering Principles of Crossflow Membrane Technology" Osmonics Inc. Technical Paper, P/N 56821
Membrane Materials
 Polymeric membranes:  Ceramic membranes:
◼ Polysulfone ◼ Aalumina
◼ Polyethersulfone ◼ Titania
◼ Polyphenylsulfone ◼ Zirconia
◼ Polyvinylidene Fluoride (PVDF) ◼ ATZ mix
◼ Polypropylene (PP)
◼ Polyethylene (PE) ➢ chemical, mechanical and
◼ Cellulose and Cellulose acetates thermal stability
(CA) ➢ ability of steam sterilization
◼ Polyamide (PA) and back flushing
◼ Polyacrylonitrile (PAN) ➢ high abrasion resistance
◼ Polytetrafluoroethylene (PTFE) ➢ high fluxes
◼ Polycarbonate (PC) ➢ durable
◼ Polymethylmethacrylate ➢ bacteria resistance
(PMMA) ➢ possibility of regeneration
➢ dry storage after cleaning
Membrane Properties
 Pure water permeability (PWP)
 Pore size
 Molecular Weight Cut-Off (MWCO)
 Hydrophobicity/hydrophilicity
 Surface/pore charge
 Surface roughness
 Chemical stability / chlorine tolerance
Principles of Mass Transport and
Rejection in Pressure-Driven
Membrane Processes

macromolecules particles ions


microbes
Overview
 RO, NF, UF, and MF have many similarities
(geometry, flow configuration, material…)
 Principals of rejection differ substantially
◼ In RO, function of relative affinity of solute and solvent to
the membrane
◼ In MF, mainly due to physical sieving
Membrane Performance
 The performance of a membrane is determined by
mainly two parameters, flux and rejection:
◼ Flux (J), or permeation rate, is the volume flowing through
the membrane per unit area per time (Q/A)
◼ Rejection (R), refers to a local relationship between
upstream and downstream concentrations
 Another important parameter is recovery (r), which
is defined as the amount of material collected as a
useful product divided by the total amount of the
material entering the process: in membrane
separations, the useful product is most often the
permeate water
Water and Solute Flux
 Water flux (Jw), or permeation rate, is the volume flowing through the
membrane per unit area per time (Q/A)
Qwater Vwater
Jw = =
Amembrane Amembrane  time
◼ In membrane processes it is a function of driving force, membrane
properties, and feed quality

 Specific permeate flux is the water flux calculated above normalized to


the applied driving force
Qp
Jsp = [gpd/ft 2 ·psi (gfd/psi), LMH/bar]
A×P

Note: in MF/UF, DP = net applied pressure (NAP)


Water and Solute Flux
 Solute Flux is the mass of solute flowing through the
membrane per unit of area per time

m solute molsolute
Js = =
Amembrane Amembrane  time

 In membrane processes it is a function of driving


force (concentration), membrane properties, and
solute/particle properties
Water Recovery Rate (r)
 The ratio of the useful product (permeate) flow rate and the
flow rate of feed to the process
 Qp 
r =   * 100%
 Qf 

 Global recovery rate:

◼ Where Qp is the product (permeate) flow rate and


◼ Qf is the feed flow rate

 In membrane processes, because of the modularity and


various configurations, it is important to distinguish between
membrane/module recovery and system recovery
Material Balance in Membrane
Separation

Feed Concentrate
A, B
Qf, Cf, Pf Qc, Cc, Pc

Permeate
Qp, Cp, Pp
Material Balance in Membrane
Separation
 Mass balance for water flow
Qf = Qc + Qp

 Mass balance for solute flux


QfCf = QcCc + QpCp

 Product recovery
r = (Qp/Qf)·100%
Example of Process Recovery
 Assuming each membrane (or each stage) operated
at 20% recovery, what is the total system recovery

R1 gal R2 gal
100 gal R3 gal

P1 gal P2 gal P3 gal


Example of Process Recovery
 Repeat the exercise with 50% recovery per stage,
what is the production rate of the third stage?

R1 gal R2 gal
100 gal R3 gal

P1 gal P2 gal P3 gal


Staging
 In RO and NF operations, membranes are often
staged

tapered design compensates for loss of feed volume through


system
Recirculation
 In MF and UF, some concentrate is often recirculated
to the inlet

◼ Flexible (can control degree of recirculation)


◼ Economics (tradeoff between power for recirculation
pump and additional recovery)
Material Balance in Membrane
Separation
❑ Mass balance for water flow
Qf = Q c + Qp

❑ Mass balance for solute flux


QfCf = QcCc + QpCp

❑ Product recovery
r = (Qp/Qf)·100%

❑ Global Rejection
 Cf − Cp   Cp 
R =    100% = 1 −   100%
 Cf   Cf 
Rejection (R)
 Location-specific ratio of product concentration and
feed concentration
 Global rejection…  cp 
R = 1 −  * 100%

 cf 

where cp is the solute concentration in the permeate and cf is the


solute concentration in the feed

 cp 
 Global system rejection… Rmass = 1 − r  * 100%
 cf 

◼ May yield different value as function of time


◼ Variability of feed, permeate {Cp=f(R)}, membrane condition…
Rejection (R)
 Local rejection due to change in bulk feed concentration in the
flow channel
 cp 
R = 1 −  * 100%
 cwall 

cwall ≥ cbulk ≥ cfeed


 If we know the permeate flux and mass transfer coefficient
(we will talk about it later), the concentration at the
membrane surface can be predicted by calculating a
polarization factor (PF)… cwall = PF · cbulk

 Apparent rejection calculated based on bulk concentration:


 c 
Rapparent = 1 − p  * 100%
 cbulk 
Temperature Effects
 Higher transmembrane pressure in the winter, or
 More membranes in the winter to prevent
fouling/cleaning
 Water demand difference between summer and
winter may offset loss in membrane productivity
 Membrane fouling

http://watertreatmentguide.com/temperature_correction.htm
Temperature Effects
 Change in temperature may result in a wide range of
effects that go beyond the viscosity of the permeate
alone
 Different ways to model effects of temperature:
◼ Arrhenius equation: JT = J20 exp (s/T)
◼ J20 = permeate flux at reference temperature of 20 °C
◼ s = empirical constant, membrane specific
◼ T = temperature
JT
 For MF and UF: Flux20 = FluxT (μT/μ20); or = 1.03(T −25 )
J 25

◼ Temperature Corrected Flux (TCF)


Temperature Effects
 Primary effect due to influence on viscosity
 For temperatures in the range of 0-35 °C
◼ μ (centipoise) = 1.777 – 0.052 T + 6.25x10-4 T2
◼ centipoises = Pa-sec x 1,000
◼ μ (Pa-sec) = 3.797x10-11 T4 – 9.963*10-9 T3 + 1.029x10-6 T2 – 5.589x10-5 T + 1.783x10-3

y = 3.7967E-11x 4 - 9.9632E-09x 3 + 1.0291E-06x 2 - 5.5887E-05x +


2.0E-03 1.7828E-03
1.8E-03 R2 = 9.9987E-01

1.6E-03
1.4E-03
1.2E-03
1.0E-03
8.0E-04
6.0E-04
4.0E-04
2.0E-04
0.0E+00
0 20 40 60 80 100
Membrane Fouling
 Deposition
◼ Silt and suspended solids
 Scaling
◼ Inorganic deposits formed due to
concentration of sparingly soluble
salts beyond the chemical solubility
limit
 Biofouling
◼ Microbiological growth entering or
within element
 Organic fouling
◼ Interactions of natural or synthetic
organics
Scaling SEM
Scaling SEM
Silt Density Index (SDI)
 Empirical test of filterability
 Measures the tendency of a raw water to foul a
membrane
◼ Use 0.45 μm filter in a dead-end filtration cell
◼ ti – time required to filter a fixed volume of raw water
through a clean membrane (~500 ml)
◼ tf – time required to filter the same volume after the
membrane has been used for a defined length of time
◼ Standard conditions: 47 mm filter, 2 bar (30 psi)
transmembrane pressure, total time (tt) of 900 sec
100(1 − t i t f )
SDI =
tt
CEEN 470
Water and Wastewater Unit Operations

Water Treatment
Membrane Processes
Microfiltration and Ultrafiltration
Overview
 Initial use of deep filtration microfilters…disposable,
not sustainable…
 MF membranes provide removal by retention of
contaminants on the membrane surface
 Lowest pressure membrane process
 Pore size of 0.05 – 5 micron
 Cake filtration provides additional removal
capabilities …smaller particles than pore size can be
removed
Current Status
 MF and UF generally accepted as being capable of
meeting filtration requirements for drinking water
production
◼ Turbidity removal / disinfection
 MF can resolve the conflict between need to provide
primary disinfection and DBP formation
 LT2ESWTR identified membranes as treatment
technique for higher level removal of
cryptosporidium
 Substantial diversification of membrane processes
and configurations
Filtration Spectrum
Treatment Capabilities
 Removal of particulate matter
◼ Turbidity
◼ Particles
◼ Microbial control
 Removal of organic and inorganic species when feed
water is pretreated (coagulation, adsorption)
◼ DOC/DBP precursors
◼ color / taste / odor
◼ Pesticides
◼ Iron / manganese (aeration / chemical oxidation)
◼ Arsenic
Treatment Capabilities
Parameter Pretreatment needed for substantial removal
MF UF
Particulate/microbial Turbidity None None
Protozoa None None
Bacteria None None
Viruses Coagulation None
Organic TOC Coagulation / PAC Coagulation / PAC
DBP precursor Coagulation / PAC Coagulation / PAC
Color Coagulation / PAC Coagulation / PAC
T&O Coagulation / PAC Coagulation / PAC
Pesticides PAC PAC
Inorganic Iron & manganese Oxidation Oxidation
Arsenic Coagulation Coagulation
Hydrogen sulfide Oxidation Oxidation
Modes of Application
Turbidity Removal
Particle Removal
Water Permeation Across
Clean MF/UF Membranes
 Pure water transport through clean porous
membrane is:
◼ Directly proportional to transmembrane pressure (ΔP)
◼ Inversely proportional to viscosity (μ)
 Modeled using modified form of Darcy’s Law:

Qtotal DP
J= =
A mRm

 Rm ≡ hydraulic resistance of the clean membrane to


water permeation (units?)
Example: Membrane resistance
 An MF membrane is tested in the lab by filtering
clean, deionized water and the flux is found to be
2,000 LMH (L/m2-hr) at 20 C and 0.7 bar.
Calculate the membrane resistance coefficient.
Water Permeation Across
Clean MF/UF Membranes
 Absolute transmembrane pressure vs. pressure
gradient
 Typical units of water flux
◼ gfd (gal/ft2-day)
◼ LMH (l/m2-hr)

 LMH x 1.7 = gfd


Flow Through a Cylindrical Pore
(Poiseuille’s Law)

p r 4 DP
Q1 pore =
8m Dz

 ΔP/Δz is pressure gradient


 In real membranes pores are not perfectly cylindrical

◼ Dimensionless tortuosity factor (t) is often introduced
Flow Through Membrane Pores
p r 4 DP
Qtotal = A pore
8mt Dz

pore = #pores/A

p r DP 4
 r DP 8tDz
2
J=  pore =  Rm =
8mt Dz 8mt Dz  r2
Significant Parameters
 Pore size has the highest effect on resistance to
water flow
 Pore size distribution
 Specific flux for membrane comparison
◼ Calculated based on area on feed side

PFi + PFo
DP = − Pp
2
Reduction in Membrane Productivity
 Flux Decline Mechanisms
◼ Fouling
◼ Concentration polarization
◼ Resistance in Series

http://books.google.com/books?id=SqHUSd3vk1oC&pg=PA160&lpg=PA160&dq=temperature+correction+factor+ultrafiltration&source=web&ots=21o8VFucPS&sig=hpvbPyECYwEiLkX4BZSiX
Reductions in Permeate Flux
Crossflow Filtration
Pressure

Feed Concentrate

Membrane

Permeate
Reduction in Permeate Flux over
Time
 Reversible vs. irreversible fouling
Increase in Transmembrane
Pressure Over Time
 Reversible vs. irreversible fouling
Classwork: UF Recovery
 Assuming the following operating scenario:
◼ UF treatment plant with 48 HydraCap60 membrane
elements
◼ 20 min operation with average productivity of 55 gfd
◼ backwash with product water for 45 sec uses 1500 gal of
permeate

 What additional information is needed?


 What is the water recovery rate?
 What is the backwashing flux?

http://www.membranes.com/pdf/HYDRAcap.pdf
High-Pressure
Membrane Technologies:
Nanofiltration and Reverse Osmosis
Membrane Separation

Nanofiltration (NF)
“Membrane Softening”

Reverse Osmosis (RO)


“Hyperfiltration”
RO and NF Membrane Applications
 Desalting/total dissolved solids
◼ 1967 first spiral-wound configurations (General Atomics) for
brackish water desalination
 Disinfection by-product precursors
◼ Leading to TTHMs and HAA formation
 Hardness, color, and turbidity
◼ After introducing NF membranes in 1986 (Filmtec Corp.)
 Inorganic chemicals
◼ For heavy metal, nitrate and fluoride removal
RO and NF Membrane Applications
 Synthetic organic chemicals
◼ Pesticide removal since the 1990s
 Pathogens
◼ Very effective, but usually no driver for implementation
 Indirect Potable Reuse
◼ since the 1980s using reclaimed water for direct injection
Colligative Properties of
Ionic Solutions
 Properties of solution that depend on the number of
solute molecules present, but not on the nature of
the solute (number of particles in a given volume of
solvent and not the mass of the particles)
 Osmotic pressure, vapor pressure, freezing point
depression, and boiling point elevation are examples
of colligative properties
 Osmotic pressure and vapor pressure are two
properties of solution that play a major role in
various membrane processes
Ionic vs. Covalent
Electrolyte vs. Nonelectrolyte
 Substance when dissolve can break up to ions or stay
intact (i.e., NaCl vs. sugar)

Type % Ionization Solubility


Electrolyte: conduct electricity
Strong electrolyte 100% Very soluble
Weak electrolyte < 100% Slight-to-very soluble
Nonelectrolyte: does not conduct electricity
No conduction 0% Insoluble or soluble
Solute molecule

Osmosis Solvent molecule


Semipermeable membrane

 Semipermeable membrane: permits passage of some


components of a solution, example: cell membranes
and cellophane
 Osmosis: the movement of a solvent from low solute
concentration to high solute concentration
 There is movement in both directions across a
semipermeable membrane (!)
 As solvent moves across the membrane, the fluid
levels in the arms of a U-tube becomes uneven
 Eventually the pressure difference between the arms
stops osmosis
Experiment in Osmosis
Examples of Osmosis
 Cucumber placed in NaCl solution loses water to
shrivel up and become a pickle
 Limp carrot placed in water becomes firm because
water enters via osmosis
 Salty food causes retention of water and swelling of
tissues (edema)
 Water moves into plants through osmosis
 Salt added to meat or sugar to fruit prevents
bacterial infection (a bacterium placed on the salt
will lose water through osmosis and die)
Osmotic Pressure
 Osmotic pressure, π, is
the pressure required
to stop osmosis

◼ π = osmotic pressure
◼ M = molarity (mol/L)
◼ R = Ideal Gas Constant =
0.0821 L·atm/mol·K
◼ T = temperature (K)
Osmotic Pressure
Applied pressure
needed to stop
osmosis
STEP 1

Pure Solvent

Solution

Semipermeable
Membrane
Osmotic Pressure
 Osmotic Pressure is a VERY sensitive measure of
Molarity
 Seawater contains 34 g NaCl per liter
M = 34 g/L / 58.5 g/mol = 0.582 M

◼ π (@ 25 °C) = (0.582 mol/L)·2·(0.0821 L·atm/mol·K )(298 K)


= 28.4 atm

◼ 1 atm supports column of water 10.34 m high


◼ (28.4 atm)(10.34 m/atm) = 294 m
◼ (294 m)(3.28 ft/m) ~ 963 feet
Van’t Hoff Factor (i)
 Factor which accounts for deviation of colligative
properties due to electrolytic nature of solutes
 Van’t Hoff Factor (i) can be measured by
measured colligativ e property (from experiment )
i=
Expected value if solute were nonelectro lyte

◼ Typical nonelectrolyte (i.e., urea, sucrose, glucose): i = 1


◼ Electrolytic solute:
Electrolyte Ideal i Measured i
NaCl 2 1.9
HIO3 2 1.7
MgCl2 3 2.7
AlCl3 4 3.2
Discrepancy between Real and
Ideal Van’t Hoff Factor i
 Consider: AlCl3 → Al3+ + 3 Clˉ
◼ Ideally i = 4 because four ‘particles’ in solution, but
actually…
◼ Real i = 3.2
 The reason measured value (3.2) is lower than ideal:
strong electrolytic attraction forces between
oppositely charged ions causes some of the ions to
be held together through ion-pair
 Van’t Hoff factor follows ideal value best when the
concentration is very low. Large deviation from Van’t
Hoff factor with very high concentrations
Exercise
 A mixture of solid NaCl and solid sucrose, C12H22O11
has an unknown composition. When 15.0 g of the
mixture is dissolved in enough water to make 500 mL
of solution, the solution exhibits an osmotic pressure
of 6.41 atm at 25 °C.
◼ Determine the mass percentage of NaCl in the mixture.
◼ What mass of NaCl and sucrose must be added so that the
osmotic pressure is 10 atm at 25 °C ?
Reverse Osmosis / Nanofiltration
Pressure greater
than Δπsoln

STEP 2

Pure Solvent

Solution

Semipermeable
Membrane
Mass Transfer and Permeate Flux
 Water (permeate) Flux
◼ Jw = Qp/A = Aw(ΔP – Δπ) [gfd, LMH]
◼ Js = BDCs
 Kw = water mass transfer coefficient
 ΔP = transmembrane pressure differential

  = reflection coefficient (unitless)


 Δπ = transmembrane osmotic pressure differential

 A = effective membrane area

 B = salt permeability

 Specific Permeate Flux


◼ Jw = Qp/A*P [gfd/psi, LMH/kPa]
◼ Note: ΔP-Δp = net applied pressure (NAP)
The Reflection Coefficient, σ
 1<σ<0
 σ = 0, non-selective
solute
pore

water

 σ = 1, semipermeable
solute
pore
water
The Reflection Coefficient, σ
 1 < σ < 0, partial selective

solute pore

water
Mass Transfer and Permeate Flux
Mass transfer limited

Permeate flux independent of


transmembrane pressure
Film Layer Model
D B cwall
J= ln
Permeate flux

dcp c bulk

k’: mass transfer coefficient


Linton-Sherwood Correlation:
k ldh
Sh = = 0.023(Re)0.83 (Sc )0.33
Clean Water Flux DB

Transmembrane pressure
Concentration Polarization
and Membrane Scaling
Boundary Scale
Layer

J C
m
C s J = Qp/Ah
dc
D K = 1.62(UD2/2bL), U=Qt/Av
dx
Cb
Cp

Bulk Product

if cm>cs then precipitation/scaling occurs


Mass Transfer and Permeate Flux
 Temperature Effects
◼ JT = J20 exp (s/T)
 J20 = permeate flux at reference temperature of 20 °C
 s = empirical constant, membrane specific

 T = Temperature

 Temperature Corrected Flux (TCF)

JT (T − 25 )
= 1.03
J 25
Temperature Effects
22
20
Water Flux (L/m2-hr)

18

16

14

12

10

8 Feed 20 degC - Permeate 20 degC


Feed 25 degC - Permeate 20 degC
6 Feed 30 degC - Permeate 20 degC
4 No Temp Control
Feed 30 degC - Permeate deg30C
2
25 35 45 55 65 75
NaCl Conc. (g/L)
Influence of Operating
Parameters on RO
 Important implication of SD
model is that rejection
increases with feed pressure
◼ Water flux  (DP - Dp)
◼ Salt flux  dcs/dz
◼ Increasing feed pressure doesn’t
change salt flux!
 Flux and recovery inversely
related due to reduction in
driving force from feed to
retentate in the module
Impact on Permeate Flux
NF-90 NF-270 NF-4040 TFC-S

1
0.9
Normailzed Permeate Flux (J/Jo)

0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 50 100 150 200 250
Runtime (hours)

You might also like