Direct compression of Banlangen Danggui Powder veterinary grade API on high-speed rotary tablet presses is rarely feasible without dry granulation because the co-processed extract exhibits hygroscopic bridging and poor flow at ambient relative humidity above 45% RH. The native powder typically requires milling to a particle size D90 ≤150 µm, followed by loss-on-drying control in a vacuum tray dryer at 50°C and -0.08 MPa until residual moisture reaches 2.5–3.5% w/w. A production-suitable tablet formulation uses microcrystalline cellulose PH102 at 20–30% w/w, spray-dried lactose monohydrate at 15–25% w/w, crospovidone at 2–4% w/w, sodium starch glycolate at 2–4% w/w, colloidal silicon dioxide at 0.5–1.0% w/w, and magnesium stearate at 0.50–0.75% w/w. Roller compaction is executed at 18–25 kN/cm roll pressure with a 1.0 mm ribbon screen, and fines below <75 µm are kept below 25% of total granulate mass to avoid feed-frame segregation. On a 16-station rotary press, compression force is maintained between 8–15 kN, producing uncoated tablets with hardness 60–100 N, friability ≤1.0% per USP <1216>, disintegration ≤30 min per Ph. Eur. 2.9.1, and content uniformity acceptance value AV ≤15.0 per USP <905>. Magnesium stearate above 0.75% w/w is avoided because hydrophobic coating of crospovidone particles delays disintegration. Tablet film coating, where required for swine or poultry administration, uses a hydroxypropyl methylcellulose-based aqueous system with inlet temperature 60–65°C and pan speed 4–8 rpm in a 48-inch perforated coating pan. The finished uncoated or film-coated tablet is intended as an oral solid dosage form packed in PVC/aluminum blister cavities with desiccant, because moisture uptake above 4.0% w/w increases sticking and darkening of ferulic acid-containing extract surfaces.

For batches above 100 kg, the roller compaction step is often split into two campaigns to limit heat accumulation and preserve ligustilide-related fractions. When the milled extract contains more than 10% w/w of fines below <75 µm, bulk density drops below 0.40 g/cm³ and the powder bridges over the press hopper. In that case, precompression with a roll force at the lower bound of 18 kN/cm and dry screening through 1.25 mm produces granules with improved die fill. Operators on GMP veterinary solid-dose lines observe capping when granules are overwarmed above 28°C before compression because the binder phase softens and reduces interparticulate contact. The acceptable granule temperature window is therefore controlled at 22–28°C. Tablet weight variation on rotary presses is monitored by automatic checkweigher with ±3% alarm limits. The terminal tablet product must also pass heavy metal limits generally applied to veterinary herbal extracts, including lead ≤5 mg/kg, cadmium ≤1 mg/kg, and arsenic ≤2 mg/kg, verified by inductively coupled plasma-mass spectrometry.

When Sterile Filtration Is Preferred to Terminal Steam Sterilization of Lignan-Containing Injectables

Injectable dosage forms derived from Banlangen Danggui Powder require a fully depyrogenated and sterilized process stream because the extract contains water-soluble polysaccharides, ferulic acid conjugates, and small amounts of ligustilide-type lactones that are thermally sensitive. The bulk solution is produced from a purified aqueous extract or reconstituted spray-dried intermediate in water for injection, adjusted with citrate or phosphate buffer to pH 5.0–6.5, and made isotonic with sodium chloride 0.80–0.95% w/v. Clarification is performed through a 0.45 µm nylon or polyethersulfone cartridge followed by a 0.22 µm polyvinylidene fluoride membrane filter. For 10 mL and 20 mL multi-dose or single-dose vials, terminal steam sterilization at 121°C for 15 min is generally replaced by aseptic filtration because forced degradation data for ferulic acid and ligustilide show marked assay loss under alkaline steam exposure. The filtered solution is held at 2–8°C for not more than 8 h before filling under nitrogen overlay with residual dissolved oxygen ≤2 ppm. Antioxidant protection uses sodium metabisulfite at 0.05–0.20% w/v, but only after compatibility with the extract is confirmed; some polysaccharide-rich batches show visible haze when bisulfite exceeds 0.20% w/v. Container closure selection uses bromobutyl rubber stoppers with fluoropolymer-coated contact surfaces, validated for extractables under 25°C and 60% RH after 6 months. The finished injectable solution is tested for bacterial endotoxin with a limit of ≤0.5 EU/mL by Ph. Eur. 2.6.14, sterility by membrane filtration, and particulate matter by USP <788>; for containers up to 100 mL, counts must be ≤6000 particles at ≥10 µm and ≤600 particles at ≥25 µm per container. If a market authorization specifically demands terminal sterilization, published kinetics for this particular Banlangen Danggui mixture are limited, so a forced degradation study must be conducted on each batch to justify any claimed shelf life.

The aseptic filling line for small-volume veterinary parenterals typically operates at 150–250 vials/min with 100% checkweighing and vacuum leak detection. However, the extract contributes weak chromophores, so automated visual inspection is calibrated against a positive control containing 20 µm placebo particles. Terminal product pH is checked before and after filling because carbon dioxide absorption can shift unbuffered solutions by 0.3–0.5 pH units. If the solution is intended for intravenous administration in cattle or swine, the endotoxin specification is tightened to ≤0.25 EU/mL, and a 1.2 µm prefilter is introduced to reduce bioburden before the final sterilizing filter. The compatibility of the Banlangen Danggui solution with plastic administration sets should not be assumed; PVC lines may adsorb ligustilide fractions over 4 h contact time, so validation with the intended tubing material is required.

Capsule filling of the dry extract is governed less by biochemical compatibility than by powder mechanics: Hausner ratio, Carr index, and angle of repose determine whether a dosator-type capsule machine can maintain fill weight below ±3% variation. Native Banlangen Danggui extract often has a Hausner ratio above 1.35 and a Carr index above 30%, which is unsuitable for 120,000 capsules/h production. Dry granulation by roller compaction or slugging brings the Hausner ratio to ≤1.25 and the Carr index to ≤25%. Capsule formulations for swine or canine administration use size 0 or 00 hard gelatin or hypromellose capsules, with fill weight 250–500 mg. The granulate includes crospovidone 2–4% w/w, sodium starch glycolate 2–3% w/w, colloidal silicon dioxide 0.5–1.0% w/w, and magnesium stearate 0.25–0.50% w/w. Loader trays are kept at 20–25°C and 35–40% RH; exposure to >60% RH for more than 4 h causes gelatin shell softening and extract caking at the capsule shoulder. Disintegration of filled capsules is measured by USP <2040> with a limit of ≤30 min in water at 37°C. Dissolution testing, where required for batch release, uses USP Apparatus 2 at 50 rpm in 900 mL water; if ferulic acid is used as marker, a typical herbal extract capsule specification sets Q ≥75% in 45 min, but published data for this specific Banlangen Danggui formulation is limited and should be verified by pilot dissolution curves. The filled capsules are packed in HDPE bottles with silica gel desiccant and finished moisture is controlled at ≤4.0% w/w.

On capsule lines equipped with tamping pin stations, the number of tamping pins and pin compression force are adjusted when the granulate bulk density falls outside 0.45–0.60 g/cm³. Excessive fines below <75 µm cause dust accumulation around the closing rings, while particles above 850 µm increase cap-lock failure. The production-scale remedy is not longer mixing but re-screening through 1.0 mm with fines recycle limited to 20%. If hypromellose capsules are used, the shell moisture is conditioned to 11–16% w/w, because lower moisture leads to brittleness during separation on high-speed filling machines. The terminal capsule product must comply with the same heavy metal and microbial limits as other oral veterinary solid dosage forms, and batch records should document residual solvent from any granulating step if ethanol or isopropanol was used.

Particle Size Distribution, Sachet Fill Variance, and Granulation Endpoints for Oral Powders

Oral powders and granules made from Banlangen Danggui Powder are intended for direct mixing into feed or drinking water, so the critical processing targets are dust suppression, rapid wetting, and particle-size uniformity across sachet filling lines. A standard oral powder formulation blends the spray-dried extract with dextrose monohydrate or lactose monohydrate at 40–60% w/w, maize starch at 10–20% w/w, and colloidal silicon dioxide at 0.5–1.0% w/w. When a granulated product is required, povidone K30 is dissolved in purified water to 3–5% w/v and sprayed onto the blend in a top-spray fluid-bed granulator with inlet temperature 60–70°C, product temperature 35–45°C, and exhaust humidity 10–20% RH. Drying continues until loss on drying reaches 2.0–3.5% w/w, then the granules are sieved through 500 µm and 150 µm screens to retain 60–80% of mass in the target fraction. Bulk density of the finished granule is generally 0.35–0.55 g/cm³, with tapped density 0.50–0.75 g/cm³. Stick-pack filling machines at 60–120 packs/min maintain fill weight variation below ±2% when the powder flow function coefficient is above 4. The finished oral powder is packed in laminated aluminum sachets with moisture permeability below 0.5 g/m²/day; residual moisture above 4.5% w/w accelerates acid-catalyzed hydrolysis of extract conjugates and increases microbial risk.

Wet granulation is not always preferable. For feed-directed powders with a high proportion of water-soluble polysaccharides, overdrying below 1.5% w/w creates brittle granules that dust during sachet compaction. Conversely, product temperature above 45°C during fluid-bed drying may degrade volatile constituents such as ligustilide and produce brown specks. Operators observe that spray nozzle blockage becomes frequent when the binder solution is prepared at concentrations above 5% w/v because the extract fines adsorb povidone rapidly. The appropriate binder ratio is therefore confirmed by a granulation yield check: granules between 150 µm and 850 µm should constitute at least 70% of the batch before lubrication. If the value falls below 60%, the batch is reworked by dry sieving rather than adding more binder. Uniformity of dosage units for sachets is evaluated by weight variation with an acceptance value AV ≤15.0, and microbial limits follow compendial oral veterinary criteria, typically total aerobic microbial count ≤10⁴ CFU/g and total combined yeasts and moulds ≤10² CFU/g.

Why does stratification override mixer coefficient of variation in finished premix shipments?

Premix manufacturing for Banlangen Danggui Powder at 5%–20% w/w active concentration requires carrier selection that matches the bulk density and median particle size of the extract, or post-mix segregation will dominate even when ribbon blender samples show a coefficient of variation below 5%. Calcium carbonate has high density near 2.7 g/cm³ and tends to sink relative to herbal extract particles, while lightweight rice hull powder segregates upward. A more stable carrier system uses corn starch or ground corncob adjusted to a median particle size 250–400 µm and bulk density within ±0.15 g/cm³ of the extract granulate. Ribbon mixing is performed at 60–70% of vessel capacity for 15–20 min, with sampling at 10 points after discharge. The acceptance criterion for blend uniformity is CV ≤5.0% by active marker assay. However, in 25 kg multiwall paper bags, vibration during road transport can resegregate particles even when the mixer passed. Therefore, finished premix is also tested after a drop test or simulated transit vibration; if marker concentration in the top third of the bag differs by more than 10% from the bottom third, the carrier ratio or particle-size window is adjusted.

The finished premix is added to final feed at inclusion rates commonly between 0.1% w/w and 1.0% w/w, requiring the active extract to be uniformly distributed across a 2,000 kg feed batch. Because Banlangen Danggui Powder contains hygroscopic components, the premix is not combined with moist feed pellets above 14% moisture unless immediate consumption is intended, since local moisture around particle contact points can concentrate extract and form adherent spots. In mash feed, mixing in a horizontal paddle mixer for 8–10 min after adding the premix is generally sufficient to restore uniformity below 5%, but prolonged mixing above 15 min can reintroduce electrostatic segregation of fine extract particles. The terminal medicated feed or premix is packaged in sealed bags with desiccant when storage exceeds 30 days at 25°C and 60% RH. For regulatory compliance, the premix dossier should include a homogeneity study specifying the sampling plan, assay method for a marker such as ferulic acid, and acceptance limit.

For multi-dose oral solutions and drench formulations, the dominant stability risk is pH-dependent hydrolysis of ferulic acid conjugates and phase separation of ligustilide-rich fractions. The Banlangen Danggui extract is dissolved or dispersed in purified water at 5–15% w/v, with propylene glycol 5–15% v/v, glycerin 5–10% v/v, and polysorbate 20 0.5–2.0% v/v as solubilizer. The solution pH is buffered to 4.5–6.5 with citrate or phosphate buffers; above pH 6.5, ligustilide-type lactones degrade faster under ambient light. Sodium benzoate is effective at 0.10–0.20% w/v when pH is maintained below 5.0, while potassium sorbate at 0.15% w/v is preferred for pH 5.0–5.5. Multi-dose bottles are filled under nitrogen with headspace oxygen ≤2%, and the preservative assay must remain within 90–110% of label claim through the in-use period. The solution is passed through a 1.0 µm prefilter and then a 0.45 µm clarifying filter before filling; filtration pressure above 1.0 bar often signals polysaccharide fouling and requires filter change. Terminally, the oral solution is packed in amber polyethylene terephthalate or high-density polyethylene bottles with tamper-evident closures and stored at 15–25°C, protected from light.

Multi-dose containers for poultry drinking water systems may require additional physical stability data because temperature fluctuations between 5°C and 35°C in agricultural buildings can cause precipitation of extract components if polysorbate 20 is below 0.5% v/v. The use of high-shear mixing during solution preparation above 1,200 rpm is avoided because it increases air entrainment and oxidative stress on the marker compounds. Instead, a low-shear magnetic or propeller stirrer at 300–500 rpm for 20–30 min is used until visual clarity and filterability are achieved. Published data for this specific Banlangen Danggui oral solution under extreme field storage is limited; therefore, a pilot stability study at 30°C/65% RH and 40°C/75% RH is required before assigning a shelf life.