The granular API is transferred from foil-laminated PE bags into a 600 L bin blender with a fill volume held below 65% of rated capacity. Prior to loading, the headspace relative humidity is recorded and the bed is purged with compressed air complying with ISO 8573-1:2010 class 2.4.2. The granule fraction is passed through a 0.8 mm conical sieve at 1,200 rpm to break soft agglomerates. Sieve analysis is performed per Ph. Eur. 2.9.38 using a stack of 45 μm, 125 μm, 250 μm, 500 μm, and 800 μm screens; the mass fraction retained on the 800 μm screen is held below 5.0% before blending. For direct compression feasibility, the API particle size measured by dry dispersion laser diffraction at 0.5 bar should yield a D90 below 300 μm. When the measured D90 exceeds 300 μm, wet granulation is selected. In the direct compression arm, a preblend is prepared with microcrystalline cellulose at 25–45 wt%, lactose monohydrate at 20–35 wt%, croscarmellose sodium at 2–5 wt%, and silicon dioxide at 0.5–1.0 wt%. The API loading is adjusted to the label claim; for dose strengths at or below 50 mg, a 3:1 preblend of API and lactose is first mixed for 12 min at 15 rpm in a V-blender. Blend uniformity is verified by withdrawing 10 thief samples through the discharge gate at 3 sampling depths; acceptance requires an RSD below 5.0% and a mean assay from 90.0% to 110.0% of label claim. Wet granulation is used when the API fraction exceeds 35 wt% because tablet hardness variability in direct compression rises above 15%. A binder solution of povidone K30 at 2–8 wt% in purified water is sprayed at 5–10 g/min per kg dry mass into a high-shear mixer with main impeller speed of 180–240 rpm and chopper speed of 1,500–2,000 rpm. The wet mass is discharged through a 1.2 mm screen and dried in a fluid-bed dryer with inlet air temperature of 55–65°C until loss on drying reaches 1.5–2.5%. Dried granules are dry-screened through a 0.8 mm screen and lubricated with magnesium stearate at 0.25–0.75 wt% for 3–5 min. Compression is performed on a 16-station rotary tablet press; pre-compression force is set at 5–8 kN and main compression force adjusted within 45–75 kN to achieve tablet hardness of 70–110 N. Friability is measured per Ph. Eur. 2.9.7; a result below 0.8% is required before film coating. Aqueous film coating is applied to a weight gain of 2.5–4.0% using a side-vented pan at inlet air temperature of 60–70°C and bed temperature of 38–45°C. Dissolution testing is developed per Ph. Eur. 2.9.3 using apparatus II at 50 rpm in 900 mL of phosphate buffer pH 6.8; acceptance criteria are fixed during method validation. Finished tablets are packed in PVC/Alu blisters after equilibrium at 25°C/40% RH to prevent capping and edge chipping. The terminal product is a compressed tablet for oral administration in cattle, swine, and equine species.

If roll compaction is selected as an alternative to wet granulation, the powder is fed to a roller compactor with roll pressure of 30–50 bar, roll speed of 5–15 rpm, and dry screen size of 0.8–1.25 mm. Ribbon density is monitored on-line; target ribbon solid fraction is 0.55–0.75. Recompression is performed with 5–10% extragranular lubricant and disintegrant to restore compactibility. Punch sticking is a known failure mode when residual moisture exceeds 2.5%; in that case, the granulator outlet air is adjusted to lower the product dew point and the punch tips are conditioned with a food-grade anti-stick coating before the batch is restarted. The in-process moisture specification is therefore tighter than the storage specification: the granulate is held at 1.5–2.5%, while finished tablets may equilibrate at 25°C/40% RH after film coating.

What Limits Sterile Filtration Throughput in Large-Volume Parenterals?

In veterinary parenteral manufacturing, the granular API is first screened for aqueous solubility at 20–25°C using water for injection; a saturation solubility below 10 mg/mL at 20°C triggers co-solvent screening with propylene glycol, glycerin, or polyethylene glycol 400 in stepwise ratios from 5% to 40% v/v. The solution pH is adjusted with 0.1 M hydrochloric acid or 0.1 M sodium hydroxide to a target range of 5.5–7.5 unless forced-degradation data indicate hydrolytic instability outside this window. Terminal sterilization is evaluated at 121°C for 15 min; if assay loss exceeds 2.0% or total impurities exceed the VICH GL18 threshold after a scouting autoclave run, aseptic filtration through a 0.22 μm polyethersulfone membrane is selected. Published forced-degradation data for Qishen Granules at 121°C may be limited in initial development; therefore, a scouting terminal-sterilization run at 116°C for 20 min and a second run at 121°C for 15 min are executed before the sterilization mode is fixed. Integrity testing is performed before and after filtration by bubble point and diffusive flow methods in accordance with the filter manufacturer’s validated limits; minimum bubble point for a 0.22 μm PES membrane is typically not below 3.2 bar when wet with water. The filtered solution is filled in an ISO Class 5 laminar airflow zone with EU GMP Annex 1 compliant monitoring; vial size ranges from 10 mL to 100 mL Type I glass. Particulate matter is tested per USP <788>; for containers larger than 100 mL, the light obscuration acceptance limit is 25 particles/mL at ≥10 μm and 3 particles/mL at ≥25 μm. Residual oxygen is controlled by nitrogen overlaying if the API is oxidation-sensitive. Batch sterilization records include cold-point mapping from 12 thermocouples distributed across the chamber; the minimum F₀ value at the slowest-to-heat point is held above 15 min when moist heat 121°C is used. The terminal dosage form is a sterile injection for parenteral administration in cattle, horses, and swine; label storage at 15–25°C is assigned unless photostability testing per VICH GL5 demonstrates a need for amber vials.

Filter capacity is a function of membrane area, solution viscosity, and particulate load. For a 0.22 μm polyethersulfone cartridge with an effective filtration area of 0.6 m², water permeability is measured at 20°C under a differential pressure of 0.5 bar; a permeability below 15 L/min indicates premature fouling or incomplete dissolution. The maximum cumulative filtration time is capped at 4 h before aseptic filling to prevent microbial proliferation in the holding vessel. Bioburden in the bulk solution is sampled before sterilizing-grade filtration per Ph. Eur. 2.6.12 and must be below 10 CFU/100 mL. When bioburden exceeds this limit, a 0.45 μm prefilter is inserted upstream of the 0.22 μm membrane. Container-closure compatibility is evaluated with bromobutyl stoppers and aluminum caps; extractables screening is performed at 25°C and 40°C for 3 months using headspace gas chromatography and liquid chromatography with mass spectrometric detection. The batch record includes a line clearance after each filling campaign to document that no residual product or glass fragments remain in the filling tunnel.

For companion animal capsule filling, the dried granule fraction with a bulk density between 0.45 g/cm³ and 0.65 g/cm³ and a Carr index below 25 is transferred to a dosator-type capsule machine operating at 10,000–25,000 capsules/h. The granulate is prepared by roller compaction or wet granulation to reduce electrostatic fines; a 60-mesh screen is used to remove particles below 250 μm. Capsule fill weight is checked at startup and every 30 min using a sample of 20 capsules; acceptance value must not exceed 15 per Ph. Eur. 2.9.5. Low-dose capsules use an API fraction from 5–30 wt%, lactose monohydrate at 35–55 wt%, microcrystalline cellulose at 20–40 wt%, crospovidone at 1–3 wt%, and colloidal silicon dioxide at 0.5–1.0 wt%. The mixture is blended in a 100 L double-cone blender at 8–12 rpm for 20 min; the intensifier bar is run for the final 2 min at 1,000 rpm. Bulk moisture is held at 2.0–3.5% because a moisture content below 2.0% increases electrostatic adhesion to capsule shell interiors and a moisture content above 3.5% can produce shell softening. Capsule size is selected by tapped density per USP <616>; for a 0.21 mL volume, size 2 capsule is filled; for 0.68 mL, size 0. The filled capsules are dedusted, weight-sorted, and packaged in HDPE bottles with desiccant. The finished product is a hard gelatin capsule for canine and feline oral dosing, with a tamper-evident seal; opened bottle storage at 25°C/60% RH for 21 days is evaluated before package qualification. Because published moisture sorption data for Qishen Granules in this specific configuration may be limited, a dynamic vapor sorption scan from 0% to 90% RH at 25°C is included in early formulation development.

When Qishen Granules Are Dispersed in Acidic Drinking Water Matrices

Water-dispersible granules for mass medication are produced by fluid-bed top-spray granulation from the granular API, dextrose, citric acid, sodium bicarbonate, and a low-foaming wetting agent. The target bulk density is 0.50–0.75 g/cm³ and the dispersibility test requires 95% of the labeled dose to pass through a 250 μm sieve after 3 min of stirring in water at 25°C. The formulation pH is buffered to 5.0–6.5 after complete dissolution at the maximum use rate of 1 g/L or 2 g/L depending on the species dose schedule. If the granular API contains moisture-sensitive fractions, the granulation dryer outlet humidity is maintained below 10% RH by adjusting the fluid-bed inlet air to 70–80°C and product temperature to 35–45°C. Loss on drying is held below 1.5% and the finished granules are filled into triple-layer foil pouches with a residual air volume below 2%. At the farm, the product is added to the drinking water line through a proportioner calibrated at 1:100 or 1:200; the resulting stock solution is mixed with a circulation pump providing 0.5–1.0 m/s flow velocity to prevent settling. Chlorinated water with free chlorine above 0.2 ppm is neutralized with sodium thiosulfate at 0.1–0.3 g/L before addition because free chlorine can degrade oxidizable constituents. Hard water with total alkalinity above 200 mg/L as calcium carbonate may require citric acid adjustment to prevent insoluble carbonate complexes and pH drift above 7.0 in the drinking line. The terminal product is a water-soluble granule used in poultry and swine operations; batch records include a field trial protocol in 6 farms with water sample collection at 0, 2, 6, and 24 h after addition. EU Regulation 2019/4 applies when the product is administered via drinking water as a medicated feed additive; the label must include the withdrawal period assigned by the competent authority.

Feed Premix Homogeneity and Carryover Control

During premix production, the granular API is first pre-blended with rice hulls or ground limestone as a carrier in a 100 kg pilot-scale ribbon mixer with a working volume of 60–70% of geometric capacity. The preblend ratio is typically 1:10 API-to-carrier; subsequent dilution uses a 2,000 L double-shaft paddle mixer to produce a 2–10 wt% active premix. Homogeneity is verified by collecting 20 samples from the mixer using a systematic sampling probe at 6 points across the bed and analyzing each sample by a stability-indicating HPLC method. The coefficient of variation must not exceed 5.0% and the mean assay must fall within 95.0–105.0% of the calculated premix activity. Carryover control is executed after each batch: the mixer is purged with ground limestone at 2.0% of working volume for 8 min; the wash burden is discarded or used only in non-medicated feed if the assay confirms 0.1% or less of the previous premix concentration. Dust extraction is routed through a cyclone with a baghouse filter rated at 0.5 μm particulate emission limit; operators handle the API under local exhaust ventilation with a minimum capture velocity of 0.4 m/s. The finished premix is filled into 25 kg antistatic bags with a safety-inlet valve. At the feed mill, the premix is metered into final feed at inclusion rates of 0.1–2.0 kg/tonne; the completed feed passes through a conditioner at 65–85°C for 30–90 s before pelleting at die temperature not exceeding 85°C if the API is heat-sensitive. Where heat lability has not been excluded, published data for this specific formulation should be generated; lower pelleting temperatures below 65°C are maintained until stability is confirmed. The terminal product is a medicated feed for pigs and poultry; EU Regulation 2019/4 mandates batch-specific carryover documentation and cross-contamination risk assessment, while FDA 21 CFR 558 applies to US feed use.

A recirculating high-shear rotor-stator is used to prepare an oral drench from the granular API, purified water, xanthan gum at 0.2–0.5 wt%, sodium carboxymethylcellulose at 0.2–0.8 wt%, and a preservative system of sodium benzoate 0.1–0.2 wt% with potassium sorbate 0.1–0.2 wt%. The pH is adjusted to 4.5–6.0 with 0.1 M citric acid; the targeted viscosity at 25°C is 100–500 mPa·s measured by a rotational viscometer at 60 rpm per Ph. Eur. 2.2.10. The suspension or solution is homogenized for 15–20 min at 8,000–10,000 rpm until the volume mean particle size, measured by laser diffraction, is below 150 μm for suspension-type drenches. The liquid is filled into 100 mL, 500 mL, or 1 L high-density polyethylene bottles with a dosing cup. Extractable and leachable screening follows Ph. Eur. 3.2.2 for plastic containers. Preservative efficacy is tested per Ph. Eur. 5.1.3; the product must meet criterion A for bacteria and fungi. Microbiological quality is assessed by Ph. Eur. 5.1.4; total aerobic microbial count must not exceed 10² CFU/g and total yeast and mold count must not exceed 10¹ CFU/g unless justified. The terminal product is an oral drench for calves, foals, and lambs; bottle stability studies at 25°C/40% RH, 30°C/35% RH, and 40°C/25% RH are run for 24 months with retention samples at 0, 3, 6, 9, 12, 18, and 24 months. Sedimentation or creaming is assessed by visual inspection after 7 days of static storage; redispersibility must occur within 30 s of hand shaking to be acceptable for a field-use oral dosing product.