First Year Engineering is coming soon

We're still writing and reviewing the notes for every subject, and we'll open it when every subject is ready. Meanwhile you can practise aptitude, ask Aria about any subject, or see the branches that are ready.

Desalination: reverse osmosis and electrodialysis

Desalination methods based on reverse osmosis and electrodialysis: principles, laws, formulas, design ideas, numericals, and exam-focused notes.

Drafted with Aria, reviewed by the AiCanCode.org team · Updated . How we write and check our notes. Spotted an error? Use Give Feedback at the bottom of the page.

Why it matters

Engineers use desalination to produce fresh water from seawater and brackish water for drinking, industry (boilers, cooling towers), and irrigation. Reverse osmosis (RO) and electrodialysis (ED) are two leading technologies: RO is widely used for seawater and brackish water; ED is effective for brackish water and specific ion removals.

Key ideas

  • Semi-permeable membrane: a thin barrier that allows solvent (water) to pass but restricts solute (salts/ions) transport.
  • Osmosis: the spontaneous flow of solvent across a semi-permeable membrane from a region of lower solute concentration to higher solute concentration.
  • Osmotic pressure (π): the minimum pressure that must be applied to the solution side to stop osmosis across a semi-permeable membrane at a given temperature. For dilute solutions, it follows van’t Hoff’s law.
  • Reverse osmosis (RO): a pressure-driven membrane process where applied hydraulic pressure greater than the osmotic pressure forces water to flow from the saline side to the pure-water side, rejecting most dissolved salts.
  • Permeate: the water that passes through the membrane (product water).
  • Retentate (brine): the concentrated saline stream that does not pass through the membrane.
  • Salt rejection (R): the fraction of salt retained by the membrane, typically expressed as a percentage.
  • Water flux (Jw): volume of water passing through unit membrane area per unit time.
  • Concentration polarization: build-up of solute concentration near the membrane surface on the feed side, reducing performance and increasing scaling risk.
  • Electrodialysis (ED): an electric-field-driven membrane process that uses selective ion-exchange membranes to move ions out of water, creating alternating dilute and concentrate streams.
  • Ion-exchange membranes: polymer membranes with fixed charges that selectively allow counter-ions to pass.
  • Cation-exchange membrane (CEM): a membrane with fixed negative groups that permits cations (e.g., Na⁺, Ca²⁺) to pass but blocks anions.
  • Anion-exchange membrane (AEM): a membrane with fixed positive groups that permits anions (e.g., Cl⁻, SO₄²⁻) to pass but blocks cations.
  • Transport number (t⁺, t⁻): fraction of current carried by cations (t⁺) or anions (t⁻) in an electrolyte; in membranes, reflects ion selectivity.
  • Current efficiency (ηI): fraction of applied electrical current that effectively transports salt ions across membranes (0–1).

Exact statements and derivations commonly examined

  1. Van’t Hoff’s law for osmotic pressure (dilute solutions) Statement: For an ideal dilute solution, the osmotic pressure is given by π = i C R T, where i is the van’t Hoff factor, C is molar concentration, R is the universal gas constant, and T is absolute temperature. Assumptions: (a) Solute behaves ideally and is dilute; (b) solvent activity approximates mole fraction; (c) membrane is perfectly semi-permeable. Derivation sketch (university style):
  • Consider solvent chemical potential equality at equilibrium with applied pressure π: μw(pure, P) = μw(solution, P + π).
  • Using partial molar relations and assuming ideality for dilute solution, μw(solution) ≈ μw° + RT ln xw, with xw ≈ 1 − nsolute/n total.
  • Linearizing ln xw for dilute solutions and equating pressure-dependent terms (via dμw = Vw dP), we obtain π Vw ≈ RT (nsolute/nw). Using V ≈ nw Vw for dilute solutions gives π ≈ (nsolute/V) RT = C RT. For electrolytes, include dissociation via van’t Hoff factor i, hence π = i C R T.
  • Units: if C in mol·m⁻³ and R = 8.314 J·mol⁻¹·K⁻¹, π in Pa.
  1. RO water and solute flux (solution–diffusion model) Statement: In the solution–diffusion model, steady-state water flux is Jw = A (ΔP − Δπ), and solute flux is Js = B (ΔC), where A and B are membrane transport coefficients. Assumptions: (a) Solvent and solute dissolve and diffuse independently in the membrane; (b) no coupling other than osmotic terms; (c) isothermal, steady state; (d) negligible convection in membrane. Explanation: ΔP is transmembrane hydraulic pressure difference; Δπ is osmotic pressure difference; ΔC is the solute concentration difference across membrane. Net driving force for water is hydraulic minus osmotic; for solute, concentration gradient through membrane controls Js. Salt rejection definition: R = 1 − (Cp/Cf), where Cf is feed concentration at the membrane interface (or bulk if polarization negligible), Cp is permeate concentration. Effect of concentration polarization: increases Cf at the wall to Cw > Cbulk, effectively raising local Δπ and reducing Jw; described empirically by film theory: Cw/Cb = exp(Jw/k), where k is the mass-transfer coefficient.

  2. Minimum work of separation and practical pressure in RO Statement: The theoretical minimum specific work to separate water from saline feed at recovery depends on the change in mixing free energy; in practice, applied pressure must exceed the average osmotic pressure of the retentate to maintain positive flux: ΔP > Δπavg. Assumptions: isothermal, reversible limit for minimum work; practical plants add allowances for losses and polarization.

  3. Electrodialysis ion transport and Faraday’s law Statement (Faraday’s first law): The amount of substance (in moles) transported electrochemically is proportional to the total charge passed: n = (I t)/(z F), where I is current, t is time, z is ionic charge number, and F is Faraday’s constant. Assumptions: 100% current efficiency and selective transport of targeted ions; in practice, use ηI to adjust: neffective = ηI (I t)/(z F). ED stack principle: With alternating CEM and AEM between an anode and a cathode, cations migrate toward the cathode passing CEM but blocked by AEM, while anions migrate toward the anode passing AEM but blocked by CEM. Thus adjacent compartments alternately become diluate (losing ions) and concentrate (gaining ions). Limiting current density: Above a critical current density, ion concentration at the membrane surface approaches zero, leading to water splitting and sharp voltage rise; operation is kept below this limit for efficiency and to avoid scaling.

What to draw (diagram descriptions)

  • RO module schematic: Draw a rectangular cross-section showing: Feed inlet on left to a spiral-wound element; indicate semi-permeable membrane layers; arrows for water permeating to a central permeate tube; brine exiting at right. Label: Feed Cf, Permeate Cp, Brine Cb, Applied pressure ΔP, Membrane area A, Direction of flow. Add a small boundary layer on feed side labeled "concentration polarization".
  • ED stack: Draw two electrodes at extreme left (anode +) and right (cathode −). Between them, alternating vertical membranes: CEM (− fixed charge) and AEM (+ fixed charge). Indicate compartments between membranes with flow channels labeled "Diluate" and "Concentrate" alternately. Show cation (Na⁺) arrows moving toward cathode through CEM and blocked by AEM; anion (Cl⁻) arrows moving toward anode through AEM and blocked by CEM. Label power supply, I, and direction of ion migration.

Design/operation notes (concise)

  • RO: Pretreatment (filtration, antiscalant, pH control) to prevent fouling/scaling; recovery ratio θ = Qp/Qf; energy recovery devices for seawater RO; staging and arrays to manage pressure and recovery.
  • ED: Best for brackish water (e.g., 500–5000 mg/L TDS), selective demineralization (e.g., nitrate removal), and where variable salinity demands modular control; polarity reversal (EDR) reduces scaling.

Formulas

π = i · C · R · T

  • π: osmotic pressure (Pa); i: van’t Hoff factor (dimensionless); C: molar concentration (mol·m⁻³); R: gas constant (8.314 J·mol⁻¹·K⁻¹); T: absolute temperature (K). Applies to ideal dilute solutions.

Jw = A · (ΔP − Δπ)

  • Jw: water flux (m³·m⁻²·s⁻¹ = m·s⁻¹); A: water permeability (m·s⁻¹·Pa⁻¹); ΔP: hydraulic pressure difference across membrane (Pa); Δπ: osmotic pressure difference (Pa). Applies to RO under solution–diffusion model.

Js = B · ΔC

  • Js: solute flux (mol·m⁻²·s⁻¹); B: solute permeability (m·s⁻¹); ΔC: solute concentration difference across membrane (mol·m⁻³). Applies to RO solute transport under solution–diffusion model.

R = 1 − (Cp/Cf)

  • R: salt rejection (fraction or %); Cp: permeate concentration (same units as Cf); Cf: feed concentration at membrane surface. Applies to RO; if polarization is negligible, Cf ≈ Cfeed bulk.

Cw/Cb = exp(Jw/k)

  • Cw: solute concentration at membrane wall; Cb: bulk feed concentration; Jw: water flux (m·s⁻¹); k: mass-transfer coefficient (m·s⁻¹). Film theory for concentration polarization.

θ = Qp/Qf

  • θ: recovery ratio (dimensionless); Qp: permeate flow (m³·s⁻¹); Qf: feed flow (m³·s⁻¹). Applies to RO system balance.

n = (I · t)/(z · F)

  • n: moles transported (mol); I: current (A); t: time (s); z: ionic charge number (dimensionless); F: Faraday constant (96485 C·mol⁻¹). Applies to ED ideal transport; multiply by current efficiency ηI when not ideal.

m = n · M

  • m: mass transported (kg); n: moles (mol); M: molar mass (kg·mol⁻¹). Use to convert in ED calculations.

E = V · I · t

  • E: electrical energy (J); V: cell voltage (V); I: current (A); t: time (s). Applies to ED electrical energy use; divide by volume treated for specific energy.

1 bar = 1 × 10⁵ Pa

  • Useful conversion between practical RO pressure units and SI.

Worked examples

Example 1: RO — Osmotic pressure and required applied pressure for seawater; estimate flux Given: Seawater approximated as NaCl solution with TDS = 35 g·L⁻¹; assume all as NaCl (M = 58.44 g·mol⁻¹), complete dissociation i = 2; Temperature T = 298 K; membrane water permeability A = 3.0 × 10⁻¹² m·s⁻¹·Pa⁻¹; applied pressure ΔP = 60 bar.

  1. Convert concentration to molarity:
  • C (mol·L⁻¹) = (35 g·L⁻¹) / (58.44 g·mol⁻¹) = 0.599 mol·L⁻¹ (3 s.f.).
  • Convert to SI concentration: C = 0.599 mol·L⁻¹ × 1000 L·m⁻³ = 599 mol·m⁻³.
  1. Osmotic pressure by van’t Hoff:
  • Formula: π = i · C · R · T.
  • Numbers: π = 2 × 599 mol·m⁻³ × 8.314 J·mol⁻¹·K⁻¹ × 298 K = 2 × 599 × 8.314 × 298 Pa.
  • Compute: 8.314 × 298 = 2477.6; 599 × 2477.6 = 1.485 × 10⁶; multiply by 2 → π ≈ 2.97 × 10⁶ Pa.
  • Convert to bar: π = (2.97 × 10⁶ Pa) / (1 × 10⁵ Pa·bar⁻¹) ≈ 29.7 bar.
  1. Net driving pressure and flux:
  • Given ΔP = 60 bar; assume permeate osmotic pressure is small compared to feed (first estimate), so Δπ ≈ 29.7 bar.
  • Net: ΔP − Δπ ≈ 60 − 29.7 = 30.3 bar = 3.03 × 10⁶ Pa.
  • Flux formula: Jw = A (ΔP − Δπ) = 3.0 × 10⁻¹² m·s⁻¹·Pa⁻¹ × 3.03 × 10⁶ Pa = 9.09 × 10⁻⁶ m·s⁻¹.
  • Convert to common unit: 1 m·s⁻¹ = 3600 m·h⁻¹; Jw = 9.09 × 10⁻⁶ × 3600 = 0.0327 m·h⁻¹ = 32.7 L·m⁻²·h⁻¹ (LMH).

Answer: Osmotic pressure ≈ 29.7 bar; at 60 bar and given A, estimated flux ≈ 33 LMH.

Note: Actual flux will be lower due to concentration polarization and nonzero permeate salinity; design often uses staged arrays and energy recovery for seawater.

Example 2: ED — Time to reduce NaCl concentration in a batch diluate Given: Batch volume Vd = 0.100 m³ (100 L) of NaCl solution initially at 2000 mg·L⁻¹; target 500 mg·L⁻¹; ED stack operated at current I = 10.0 A; current efficiency ηI = 0.85; assume NaCl transports as Na⁺ (z = 1) and Cl⁻ (z = 1), overall salt equivalents per mole = 1 (for NaCl); Faraday constant F = 96485 C·mol⁻¹; molar mass of NaCl M = 58.44 g·mol⁻¹. Assume concentrate side and volumes are such that the removed salt is fully accepted (batch view on diluate only). Temperature effects ignored.

  1. Mass of salt to remove:
  • Initial mass = 2000 mg·L⁻¹ × 100 L = 200,000 mg = 200 g = 0.200 kg.
  • Target mass = 500 mg·L⁻¹ × 100 L = 50,000 mg = 50 g = 0.050 kg.
  • To remove: Δm = 0.200 − 0.050 = 0.150 kg.
  1. Convert mass to moles:
  • n(NaCl) = m/M = 0.150 kg / (0.05844 kg·mol⁻¹) = 2.565 mol (3 s.f.).
  1. Charge required at given efficiency:
  • From Faraday’s law (for salt equivalents): n = ηI · (I t)/(F). For NaCl, z = 1 equivalent per mole of salt.
  • Solve for t: t = n F/(ηI I) = 2.565 × 96485 C·mol⁻¹ / (0.85 × 10.0 A).
  • Numerics: numerator = 247,800 C; denominator = 8.5 A → t ≈ 29,153 s.
  • Convert to hours: t = 29,153 s / 3600 s·h⁻¹ ≈ 8.10 h.

Answer: Required ED operating time to reach 500 mg·L⁻¹ is approximately 8.1 h at 10 A and 85% current efficiency.

Note: In continuous ED, flow rates and stack cell-pairs also determine concentration change; this simplified batch calculation illustrates Faraday’s law application.

Example 3: Theory — Compare RO and ED for brackish versus seawater Model answer outline:

  • Principle: RO is pressure-driven, moving water through a semi-permeable membrane while rejecting salts; ED is electrically driven, moving ions through selective membranes, producing alternating dilute and concentrate streams.
  • Feed salinity applicability: RO is suitable across a wide salinity range, especially seawater (≈ 35,000 mg·L⁻¹ TDS), because applied pressure can exceed high osmotic pressures. ED is more energy-efficient at low to moderate salinities (e.g., 500–5000 mg·L⁻¹) because electrical energy scales roughly with removed ions (charge), not with hydraulic pressure.
  • Energy use trend: RO energy rises with osmotic pressure (thus salinity and recovery); ED energy rises with salt removal (coulombs passed). Therefore, RO dominates seawater desalination; ED is preferred for low TDS adjustments and selective ion removal (e.g., nitrate).
  • Water recovery: RO recovery is limited by scaling/osmotic constraints (typical seawater single-pass 35–50% with energy recovery; brackish 50–85%). ED can achieve high water recovery because water remains mostly in the diluate, but voltage and limiting current set practical limits.
  • Fouling/scaling: RO membranes are prone to particulate fouling, biofouling, and scaling; requires pretreatment and cleaning-in-place (CIP). ED experiences scaling at membranes/spacers, mitigated by polarity reversal (EDR) and pH control.
  • Product quality: RO permeate has very low TDS and removes non-ionic organics and pathogens to a significant extent; ED removes ions but passes uncharged species and organics, so post-treatment may be needed if such contaminants are present.
  • Maintenance: RO needs high-pressure pumps and energy recovery devices; ED needs DC power supply, control of current density, and regular reversal in EDR.

Common mistakes

  • Using ΔP < π in RO flux calculation: forgetting that applied pressure must exceed the osmotic pressure difference to obtain positive flux.
  • Confusing salt rejection with recovery: R is about quality (salt left in retentate), while recovery θ is about quantity of water produced.
  • Ignoring concentration polarization: calculating Jw from bulk concentration rather than wall concentration, overestimating flux.
  • Misusing Faraday’s law in ED: forgetting current efficiency (ηI) or ion valence z, leading to wrong time or energy estimates.
  • Thinking ED removes uncharged contaminants: ED mainly removes ions; uncharged organics are not driven by the electric field.
  • Mixing units of pressure: using bar in the formula with R in J·mol⁻¹·K⁻¹ without converting to Pa causes inconsistent results.

For your exam

  • Short notes: concentration polarization and its control; RO pretreatment and fouling; ED limiting current density; EDR (electrodialysis reversal) principle.
  • Derivations: van’t Hoff equation for π from chemical potentials; Jw = A(ΔP − Δπ) and its physical meaning; Faraday’s law application to ion removal in ED.
  • Numericals: calculate π for given salinity and T; estimate RO flux for given A, ΔP, Δπ; compute ED time/current to reach a target TDS; convert energy units and pressures correctly.
  • Diagrams: labeled RO spiral-wound module schematic; ED stack with alternating AEM/CEM and ion directions; show electrodes, ion paths, and diluate/concentrate compartments.
  • Comparative essays: RO vs ED suitability by salinity, energy trends, product quality, and maintenance; include typical ranges (e.g., seawater vs brackish) and practical constraints (scaling, limiting current).
  • Full marks: correct formulas with units, stepwise calculations, clear assumptions, neat labeled diagrams, and a brief practical note (e.g., pretreatment or efficiency factors).

Quick check

  1. Write the formula for osmotic pressure by van’t Hoff and state the SI units for each symbol.
  2. In RO, why must ΔP exceed Δπ, and what happens if it does not?
  3. An ED unit runs at 5.0 A for 2.0 h with ηI = 0.80 removing NaCl. How many moles of NaCl are transferred? (Assume z = 1 for NaCl equivalents.)

Answers: (1) π = i C R T; π in Pa, i dimensionless, C mol·m⁻³, R 8.314 J·mol⁻¹·K⁻¹, T K. (2) Because net driving force for water is ΔP − Δπ; if ΔP ≤ Δπ, water flux is zero or reverse (no desalination). (3) n = ηI I t/(z F) = 0.80 × 5.0 A × (2.0 × 3600 s)/(1 × 96485 C·mol⁻¹) ≈ 0.298 mol.

Finished this topic? Mark it so your progress, study plan and readiness keep up.

Stuck on something here?