Pharmaceutical unit operations: granulation, blending and tablet compression

In short

Granulation, blending and compression are the three core solid-dosage unit operations that convert bulk API and excipient powders into tablets of uniform dose, strength and dissolution. Granulation fixes poor flow and poor compressibility by enlarging particles; blending distributes a small mass of API through a large mass of excipient; compression consolidates the mix into a compact under controlled force. India makes roughly a fifth of the world's generic medicines by volume and tablets are the dominant dosage form, so these operations are the day-to-day work of production, F&D and QA staff at Sun Pharma, Cipla, Dr Reddy's, Lupin, Zydus, Aurobindo, Torrent, Mankind, Alkem and Intas. The topic is examined in B.Pharm/B.Tech pharmaceutical engineering papers, in GPAT and in GATE (particle technology, mixing, size enlargement), and is governed in India by Schedule M of the Drugs and Cosmetics Rules, 1945 and the Indian Pharmacopoeia 2026 (10th edition).

Explained

Why granulate: powder properties that decide everything

• Flow must be good enough to fill a die in a few tens of milliseconds; fine cohesive powders bridge and arch, giving weight variation.
• Compressibility (ability to reduce volume under load) and compactibility (ability to form a strong compact) are different properties; a powder can have one without the other.
• Measured by bulk and tapped density, Carr's compressibility index, Hausner ratio, angle of repose and flow through an orifice; IP/USP give a common flow scale (Carr ≤10 excellent, 11–15 good, 16–20 fair, 21–25 passable, >25 poor to very poor).
• Particle size distribution is measured by sieve analysis (IS sieves/ASTM mesh), laser diffraction or image analysis; the granule size band is usually set so that fines are 10–25% to fill voids without causing weight variation.
• Granulation also prevents segregation of a fine API from coarse excipients, reduces dust (operator exposure, OEB-based containment), improves wettability of hydrophobic drugs and allows higher dose loading.

Granulation methods and mechanisms

• Wet granulation: powders are agglomerated with a liquid binder, then dried and sized. Suits low-dose, poorly compressible, hydrophobic and heat-stable drugs; most common route in India.
  • Binders: PVP K-30, HPMC E5/E15, starch paste, pregelatinised starch; solvents: purified water, IPA or hydro-alcoholic for moisture-sensitive drugs.
  • Mechanism (Rumpf/Ennis view): wetting and nucleation → consolidation and coalescence (growth) → attrition and breakage. Liquid-bridge states progress pendular → funicular → capillary → droplet (over-wetted, forms a paste).
  • Growth regime depends on the dimensionless Stokes deformation number and pore saturation; too little liquid gives fines, too much gives over-granulation and hard, slow-dissolving granules.
• Dry granulation: densification without liquid or heat, for moisture- or heat-labile drugs.
  • Slugging: compress into large slugs on a heavy-duty press, mill and sieve. Slow, dusty, largely superseded.
  • Roller compaction: powder is fed by a screw between two counter-rotating rolls, forms a ribbon, then milled to granules. Continuous, no drying step; key variables are roll force (kN/cm), roll gap, roll speed and feed-screw speed. Limitation: loss of compactibility (work hardening) — the material is compacted twice, so tablets are weaker.
• Direct compression: blend, lubricate, compress — no granulation. Needs directly compressible excipients (microcrystalline cellulose, spray-dried lactose, dibasic calcium phosphate, mannitol DC grades). Cheapest and fastest, but limited to good-flowing APIs at low to moderate dose and prone to segregation and content-uniformity failure.
• Other size-enlargement routes: melt/thermoplastic granulation (PEG, stearic acid; no solvent), extrusion–spheronisation (pellets), spray drying/fluid-bed layering, steam granulation, foam granulation, and continuous twin-screw wet granulation (now used on continuous manufacturing lines).

Equipment and end-point control

• Rapid mixer granulator (RMG, high-shear): bowl with a bottom impeller (mixing and wet massing) and side chopper (breaks lumps); binder added by pump and spray nozzle. Short cycle (10–20 min), dense granules, needs separate drying. Indian suppliers: Gansons, Shakti Pharmatech, Prism, Saan.
• Fluid-bed granulator (FBG, top-spray): powder is fluidised by heated air while binder is sprayed from above; granulation and drying in one bowl. Gives porous, light, fast-dissolving granules; bulk density is low. Critical parameters: inlet air temperature and volume, atomisation air pressure, spray rate, product and exhaust temperature, filter-bag shaking interval.
• Low-shear mixers: planetary mixer, sigma blade mixer — slow, used for small batches and ointment-like wet masses.
• End-point determination (wet massing): impeller power consumption or motor torque profile (classical and most used), probe-based methods, and PAT tools such as NIR moisture, acoustic emission and FBRM chord-length measurement. ICH Q8/Q9/Q10 and QbD expect end-point to be defined by a measured attribute, not by time alone.
• Drying: fluid-bed dryer (fast, uniform, 30–60 min) or tray dryer in a hot-air oven (slow, 6–12 h, risk of migration of soluble dye/drug to granule surface). End-point is loss on drying (LOD), typically 1–3% w/w depending on the formulation; measured by IR moisture balance or Karl Fischer for water content.
• Sizing/milling: multi-mill, oscillating granulator, cone mill (comil) with defined screen aperture and impeller speed; wet milling before drying and dry milling after.

Blending, lubrication and segregation

• Mixing mechanisms: convection (bulk movement of groups of particles), diffusion (random motion of individual particles at the interface) and shear (slip planes within the bed). Tumbling blenders work mainly by convection and diffusion; high-shear mixers add shear.
• Blender types:
  • Tumbling: V-blender (twin shell), double cone, octagonal, bin/IBC blender. Charge to 50–70% of volume; typical 10–20 rpm; critical speed must not be approached or the bed centrifuges.
  • Convective: ribbon blender, paddle, Nauta conical screw (gentle, good for large ratio differences), high-shear plough-share blenders.
  • Bin blending (IBC on a tumbling frame) is preferred in modern plants: closed transfer, no product change-over cleaning of the blender shell, less segregation on discharge.
• Ordered (interactive) mixing: fine API adheres to the surface of coarse carrier particles — used for very low-dose drugs and in dry powder inhalers.
• Segregation (de-mixing) mechanisms: sifting/percolation (fines fall through coarse — the commonest), trajectory (different stopping distances), fluidisation/elutriation (fines carried by air), agglomeration and vibration-driven rise of large particles. Controlled by matching size and density, granulating, minimising free-fall during transfer and avoiding long chute drops.
• Lubrication: magnesium stearate 0.25–1.0% w/w is the usual boundary lubricant; alternatives are sodium stearyl fumarate, stearic acid, glyceryl behenate, hydrogenated vegetable oil. Glidants (colloidal silicon dioxide, talc) improve flow; anti-adherents (talc, magnesium stearate) prevent sticking.
• Over-lubrication: magnesium stearate shears into a hydrophobic film over granule surfaces, reducing tablet hardness and slowing disintegration and dissolution. Hence it is added last, through a 40/60 mesh sieve, blended for a short defined time (typically 3–5 min) at low speed; lubrication is specified as blender revolutions, not minutes, for scale-independence.
• Blend uniformity (BU): stratified sampling with a sample thief, usually 10 locations × 1–3 dosage-unit masses. Common internal acceptance: each location 90.0–110.0% of label claim and RSD ≤ 5.0% (regulatory expectation is through content uniformity of the finished dosage unit per the pharmacopoeia; BU limits are company specifications).
• Scale of scrutiny: the sample mass at which uniformity is judged must equal one dose; a blend "uniform" on 1 kg samples can fail at 500 mg.

Tablet compression: machine, stages and science

• Machines: single-punch (eccentric) press — one die, lab/R&D scale, output a few thousand tablets per hour, compression from the upper punch only. Rotary press — multi-station turret, compression from both punches, output tens of thousands to over 300,000 tablets/hour; single-rotary (one feed and one compression point per revolution) and double-rotary/double-sided (two of each, so twice the output).
• Parts: hopper, force feeder (paddle feeder) or gravity feed frame, dies and upper/lower punches in the turret, weight-adjustment (lower) cam, pre-compression and main compression rollers, ejection cam, take-off blade and scrape-off, tablet chute, de-duster and metal detector downstream.
• Cycle of events: die filling → weight adjustment (excess scraped off at the weight cam) → pre-compression (deaeration, light consolidation) → main compression → decompression → ejection → take-off.
• Consolidation and bonding: particle rearrangement → elastic deformation (recoverable) → plastic deformation (MCC, starch) or brittle fracture (lactose, dicalcium phosphate) → interparticulate bonding by van der Waals forces, mechanical interlocking and solid bridges.
• Key machine variables: compression force (kN), pre-compression force, turret speed (rpm), dwell time (contact time of the punch head flat with the roller), punch penetration, feeder speed, fill depth. Faster turret = shorter dwell = less time for plastic flow and air escape, hence capping and lower hardness in strain-rate-sensitive materials such as MCC.
• Tooling: TSM (American) and EU (ISO) standards; B tooling (19 mm barrel, up to about 16 mm tablet) and D tooling (25.4 mm barrel, larger tablets); flat, flat-bevelled-edge, shallow/standard/deep concave faces, embossing and break-lines. Materials: hardened tool steel with chrome or titanium nitride coating.
• Instrumentation and control: strain-gauge load cells on compression rollers; the press controls tablet weight by compression force feedback (force is the measured surrogate for fill weight), rejects out-of-limit tablets automatically, and may use NIR/PAT for real-time release testing on continuous lines.
• Compression defects, cause and remedy:
  • Capping/lamination: entrapped air and elastic recovery; remedy — reduce speed, raise pre-compression, taper dies, reduce fines and over-dried granules, add plastic binder.
  • Sticking/picking: excess moisture, insufficient lubricant, worn or poorly polished/engraved punches; remedy — dry further, increase lubricant, polish or chrome-plate punches, redesign embossing.
  • Chipping/cracking: too dry granules, worn die edges, deep concave tooling.
  • Mottling: uneven colour from migration of soluble dye during tray drying or poor dye dispersion; use lakes and fluid-bed drying.
  • Weight variation: poor flow, inconsistent feeder speed, worn lower punches, mixed punch lengths.
  • Double impression: free rotation of punches without keys on engraved tooling.
  • Hardness variation / binding in die: uneven force distribution, insufficient die-wall lubrication.
• In-process checks during compression (every 30 min typically): appearance, weight of 20 tablets and individual weights, thickness, hardness (crushing strength, N), friability, disintegration time, and periodic content uniformity and dissolution by QC.

Indian standards, codes and regulatory frame

• Drugs and Cosmetics Act, 1940 and Rules, 1945; manufacture is licensed by the State Licensing Authority (Forms 25/28 for allopathic drugs) under Schedule M — Good Manufacturing Practices and requirements of premises, plant and equipment.
• The revised Schedule M (notified 28 December 2023) aligns Indian GMP with WHO-GMP and adds a Pharmaceutical Quality System, product quality review, qualification and validation, computerised system validation and stability requirements. It applied to larger manufacturers from 2024 and to small manufacturers (turnover up to about ₹250 crore) after an extension to 31 December 2025, so it now governs all licensed units.
• Indian Pharmacopoeia 2026 (10th edition), published by the Indian Pharmacopoeia Commission, gives the compendial tests: uniformity of mass (weight), uniformity of content, disintegration, dissolution and friability. Typical IP limits — uncoated tablets disintegrate within 15 min, film-coated within 30 min, dispersible within 3 min, enteric-coated resist 2 h in 0.1 M HCl then disintegrate within 60 min in buffer; friability loss not more than 1.0% w/w.
• Uniformity of mass (IP/BP style): average mass 250 mg or more — ±5%; more than 80 mg and less than 250 mg — ±7.5%; 80 mg or less — ±10%. Not more than 2 of 20 tablets may exceed the limit and none may exceed twice it.
• ICH Q8 (Pharmaceutical Development/QbD), Q9 (Quality Risk Management), Q10 (Quality System), Q7 (API GMP) are followed by export-oriented plants; US FDA 21 CFR 211 and EU GMP Annex 15 apply for regulated markets.
• Worker safety, dust explosion prevention and occupational exposure limits in the powder area now fall under the Occupational Safety, Health and Working Conditions Code, 2020 (in force 21 November 2025, replacing the Factories Act, 1948); solvent granulation areas need flameproof/ATEX-rated equipment and earthing.
• Equipment qualification, cleaning validation and process validation of these very operations are the subject of separate notes; here only the in-process control linkage matters.

Key points

• Granulation - size enlargement to improve flow, compressibility, dose uniformity and dust control.
• Wet granulation - binder liquid, massing, drying, sizing; best for low-dose, hydrophobic, poorly compressible drugs.
• Dry granulation - slugging or roller compaction; for moisture- and heat-sensitive drugs, but loses compactibility by double compaction.
• Direct compression - cheapest route, needs DC-grade excipients, risk of segregation and content-uniformity failure.
• RMG - high shear, dense granules, 10–20 min cycle; FBG - granulates and dries in one bowl, porous light granules.
• End-point of wet massing - impeller power/torque profile or PAT, never time alone.
• LOD target typically 1–3% w/w; over-dried granules cap, under-dried granules stick.
• Carr's index and Hausner ratio - flow indicators from bulk and tapped density; angle of repose below about 30° is excellent flow.
• Mixing mechanisms - convection, diffusion, shear; tumbling blenders filled to 50–70%.
• Segregation - mainly sifting/percolation of fines; cured by granulating and matching particle size.
• Lubricant blending is specified in blender revolutions and kept short; over-blending magnesium stearate softens tablets and slows dissolution.
• Scale of scrutiny - uniformity must be judged on a one-dose sample mass.
• Rotary press - compression from both punches; double-rotary doubles output over single-rotary at the same rpm and station count.
• Pre-compression - deaerates the powder bed and prevents capping and lamination.
• Dwell time falls as turret speed rises; strain-rate-sensitive materials lose strength.
• Heckel plot - slope K gives yield pressure Py = 1/K; low Py means plastic, high Py means brittle/elastic.
• Capping and lamination - entrapped air plus elastic recovery; the classic compression defect.
• IP 2026 and revised Schedule M are the governing Indian compendium and GMP requirement for tablet manufacture.

Formulas and examples

Example 1 — flow characterisation of a granule batch. A granule sample of 60.0 g occupies 150.0 mL untapped and 115.4 mL after tapping.

• ρb = 60.0/150.0 = 0.400 g/mL; ρt = 60.0/115.4 = 0.520 g/mL.
• CI = 100(0.520 − 0.400)/0.520 = 100 × 0.1200/0.520 = 23.1%.
• HR = 0.520/0.400 = 1.30.
• Both fall in the "passable" flow band (CI 21–25, HR 1.26–1.34). Verdict: usable on a slow press, but add a glidant or re-granulate before running a high-speed double-rotary.

Example 2 — drying mass balance and water removed. 50.0 kg of wet granules at 25.0% w/w LOD are dried to 2.0% LOD in a fluid-bed dryer.

• Dry solids = 50.0 × (1 − 0.25) = 37.5 kg (unchanged by drying).
• Final batch mass = 37.5/(1 − 0.02) = 37.5/0.98 = 38.27 kg.
• Water removed = 50.0 − 38.27 = 11.73 kg.
• At an evaporation rate of 12 kg/h the drying time is about 11.73/12 ≈ 59 min, consistent with a typical FBD cycle.

Example 3 — tablet weight, batch yield and press time. Granules assay 48.0% w/w of API; the label dose is 250 mg; batch size is 200 kg of lubricated blend; the press is a 45-station double-rotary at 40 rpm.

• Target tablet weight = 250/0.480 = 520.8 mg, set as 521 mg.
• Weight limits (average ≥ 250 mg, so ±5%) = 494.9 to 547.1 mg; none may fall outside ±10% (468.9–573.1 mg).
• Theoretical tablets = 200,000 g / 0.521 g = 383,877 tablets; at 97% yield ≈ 372,400 tablets.
• Output = 45 × 40 × 60 × 2 = 216,000 tablets/hour → compression time ≈ 383,877/216,000 ≈ 1.78 h (1 h 47 min).

Example 4 — dwell time and compaction pressure. Turret pitch-circle diameter 410 mm, 40 rpm, punch head flat length 6.0 mm, round flat punch 10.0 mm diameter, main compression force 20.0 kN.

• v = π × 0.410 × 40/60 = π × 0.2733 = 0.859 m/s.
• td = 0.0060/0.859 = 7.0 × 10⁻³ s ≈ 7.0 ms.
• A = (π/4) × 10.0² = 78.54 mm²; P = 20,000/78.54 = 254.6 MPa.
• If the turret is sped up to 60 rpm, v = 1.288 m/s and td = 4.7 ms — a 33% loss of dwell, the usual cause of capping appearing only at high speed.

Example 5 — tensile strength, elastic recovery and Heckel slope. A 10.0 mm tablet, 4.00 mm thick, breaks at 80 N; thickness in the die at maximum load was 3.72 mm.

• σt = 2 × 80/(π × 10.0 × 4.00) = 160/125.66 = 1.27 MPa (below the ~1.7 MPa often taken as robust for coating and packing, so raise compression force or binder level).
• ER = 100 (4.00 − 3.72)/3.72 = 100 × 0.28/3.72 = 7.5%.
• Heckel data: at 50 MPa, ln[1/(1−D)] = 1.20; at 150 MPa, = 2.45 → K = (2.45 − 1.20)/(150 − 50) = 0.0125 MPa⁻¹, Py = 1/0.0125 = 80 MPa — a plastically deforming material (MCC-like); brittle fillers give Py of several hundred MPa.

Where it is used

• Oral solid dosage plants of Sun Pharma, Cipla, Dr Reddy's, Lupin, Zydus Lifesciences, Torrent, Alkem, Mankind, Aurobindo, Intas and Hetero — Baddi, Sikkim, Ahmedabad, Hyderabad, Goa and Indore clusters.
• Equipment and tooling manufacture and service: ACG (tooling and presses), Cadmach and Karnavati/Rimek (Ahmedabad), Gansons, Prism Pharma Machinery, Shakti Pharmatech, Saan Engineers — design, erection, FAT/SAT and troubleshooting roles.
• Job roles for freshers: Production Officer/Executive (granulation or compression section), F&D/Formulation Development trainee, QA in-process (IPQA), QC analyst (dissolution, uniformity), validation engineer, technology transfer officer.
• Nutraceuticals, veterinary and agrochemical tablets, detergent and catalyst pellets use the same granulation and compaction physics; roller compaction is standard in fertiliser and detergent granule plants.
• Continuous manufacturing lines (twin-screw wet granulation + continuous drying + continuous direct compression) are being installed by Indian export units, with NIR-based PAT and real-time release.
• Generic export dossiers (ANDA/EU): granulation end-point, blend uniformity and compression-force data form the process-understanding section of the filing under ICH Q8.

Interview questions

• Q: Why do we granulate instead of compressing the powder directly? A: To improve flow for uniform die filling, improve compactibility, prevent segregation of a fine API, reduce dust and improve wettability of hydrophobic drugs.
• Q: When would you choose roller compaction over wet granulation? A: When the drug is moisture- or heat-labile, when the dose is high and a solvent step is undesirable, or when a continuous, lower-energy process is wanted; accepting that compactibility will be lower.
• Q: What is the difference between compressibility and compactibility? A: Compressibility is the ability of a powder to reduce in volume under pressure; compactibility is its ability to form a mechanically strong compact. A powder can densify well and still give weak tablets.
• Q: How do you determine the end-point of wet massing in an RMG? A: By the impeller power consumption or motor torque profile, confirmed by granule appearance, LOD and sieve analysis; modern plants use NIR or acoustic PAT. Time alone is not a valid end-point.
• Q: Why is magnesium stearate added last and blended for only a few minutes? A: It shears into a hydrophobic film over the granules; prolonged blending reduces tablet hardness and retards disintegration and dissolution.
• Q: Why is lubrication specified in blender revolutions rather than minutes? A: Revolutions represent the actual shear work done and transfer across blender sizes and speeds, whereas minutes do not scale.
• Q: What causes capping and how is it fixed? A: Air entrapment and elastic recovery exceeding bond strength; fixed by increasing pre-compression force, reducing turret speed, using tapered dies, controlling fines and moisture, and adding a plastic binder.
• Q: What is dwell time and why does it matter? A: The time the punch head flat is in contact with the compression roller at peak load; a longer dwell allows plastic flow and air escape, so strain-rate-sensitive materials need low turret speed.
• Q: What does the Heckel plot tell you? A: Its linear slope K gives the yield pressure Py = 1/K; a low Py indicates plastic deformation, a high Py indicates brittle fracture or elastic behaviour, which guides excipient and speed selection.
• Q: How does a rotary press control tablet weight? A: Fill depth is set by the lower weight cam, and the measured compression force at each station is used as a surrogate for fill weight; the control system corrects fill depth and rejects out-of-limit tablets.
• Q: What is "scale of scrutiny" in blend sampling? A: The sample size at which uniformity is assessed; it must be about one dosage-unit mass (usually 1–3× tablet weight), because a blend uniform on large samples may be non-uniform at dose level.
• Q: A 300 mg tablet is sampled: 2 of 20 deviate by 6% from the mean. Pass or fail on uniformity of mass? A: It passes — for an average mass of 250 mg or more the limit is ±5%, not more than 2 units may exceed it, and none may exceed ±10%.
• Q: Tablets stick to the upper punch during a run. What do you check? A: Granule moisture (LOD), lubricant level and distribution, punch face condition and polish, embossing design, and press room humidity and temperature.

Common mistakes

• Treating LOD (wet basis) and moisture content (dry basis) as the same number in drying calculations.
• Quoting a granulation end-point as a fixed massing time instead of a measured power/torque or moisture signal.
• Over-blending the lubricant, then blaming the disintegrant when dissolution fails.
• Forgetting the factor f = 2 for a double-rotary press, halving the calculated output; or using turret rpm as tablets per minute.
• Sampling the blend from a convenient spot instead of stratified locations and at the wrong scale of scrutiny.
• Assuming higher compression force always gives a better tablet — beyond a point it causes capping, longer disintegration and slower dissolution.
• Confusing hardness (crushing force in N or kp) with tensile strength (MPa), and comparing tablets of different size or shape by force alone.
• Reporting granule or tablet weights without stating whether the basis is theoretical or actual yield.

Related topics

Tablets and capsules · Pharmaceutical process scale-up and technology transfer · Process validation and equipment qualification: DQ, IQ, OQ and PQ · Good Manufacturing Practice (GMP) · Cleanroom and HVAC design for pharma plants · Particle technology: size reduction, screening and mixing