Prior to dry blending, the residual moisture of the Shuanghuang Baidu Granules Veterinary Grade API is determined by loss-on-drying at 105 °C to constant mass; processing specifications set the upper endpoint at 5.0% w/w because higher values are associated with picking and sticking on rotary tablet tooling during pilot runs using a 16-station rotary press equipped with B-tooling. Milled API is incorporated at 30–40% w/w, microcrystalline cellulose PH102 at 40–50% w/w, crospovidone at 2.0–4.0% w/w, colloidal silicon dioxide at 0.5–1.0% w/w, and magnesium stearate at 0.5–1.0% w/w. The API granulate is first passed through a 0.8 mm conical mill screen at 1,200 rpm, then combined with intragranular excipients in a high-shear mixer bowl of 150 L capacity with impeller speed 250 rpm and chopper speed 1,500 rpm. A 5% w/v PVP K30 binder solution in purified water is added at a rate of 2.0–3.0% w/w of dry granulate mass; wet massing continues for 3–5 min, after which the wet mass is dried in a fluid-bed dryer with inlet air at 55–65 °C until loss-on-drying falls to 1.5–2.5% w/w. Dried granules are passed through a 1.0 mm oscillating granulator and lubricated in a bin blender for 3 min after sieving. Compression force is set between 8 kN and 14 kN to reach tablet hardness of 60–100 N; friability is tested according to USP <1216> with a limit of not more than 1.0%, disintegration per USP <701> with a limit of 15 min, and dosage-unit uniformity per USP <905> with an acceptance value not higher than 15.0. The terminal product is a round tablet of 0.5 g target fill weight and 12 mm diameter, packed in HDPE bottles with a desiccant when the bulk stability data show a moisture gain above 0.3% w/w at 25 °C / 60% RH. For veterinary tablets, dissolution testing may be replaced by disintegration if the veterinary pharmacopoeial monograph allows; where dissolution is required, a paddle apparatus at 75 rpm in 900 mL of 0.1 M hydrochloric acid is used according to USP <711>. Capping has been observed when the granulate fraction below 75 µm exceeds 25% w/w; in such batches, dry binder addition or compression force reduction should be evaluated under the process validation plan.

What Limits Aseptic Filtration Throughput for the Injectable Fraction?

Because the supplied API granulate is a multi-component botanical extract, aqueous injection concentrates are pre-dispersed in Water for Injection at 40 °C and clarified by a 0.45 µm polyvinylidene fluoride membrane cartridge before terminal 0.22 µm polyethersulfone sterilizing filtration. Filter throughput is limited by the total suspended solids and bioburden of the bulk solution; during pilot batches the prefilter differential pressure should not exceed 1.0 bar and the terminal filter differential pressure should remain below 0.8 bar. The filtered solution is adjusted to pH 5.0–7.0 with 0.1 M hydrochloric acid or 0.1 M sodium hydroxide, and osmolality is adjusted to 280–320 mOsm/kg with sodium chloride unless a hypertonic formulation is required. The active concentration is set by the labelled marker content, expressed as mass per mL of bulk solution; because marker content of the API may vary among batches, the mass balance is based on the certificate of analysis rather than a fixed API weight. The solution is filled aseptically into 10 mL Type I glass vials under Grade A laminar airflow, closed with bromobutyl rubber stoppers, and sealed with aluminium flip-off caps. Terminal sterilisation is not assumed to be compatible; published data for the thermal stability of this specific API under moist-heat conditions are limited, so aseptic processing is selected unless the marketing authorisation holder provides forced degradation data at 121 °C for 15 min. Depyrogenation of vials is performed in a hot air tunnel at 250 °C for not less than 30 min; stoppers are washed, siliconised, and sterilised by steam. Sterility is tested per USP <71> / Ph.Eur. 2.6.1, bacterial endotoxins per USP <85>, particulate matter per USP <788> where the fill volume exceeds 100 mL, and pH per USP <791>.

Before capsule filling, the milled API is blended with spray-dried lactose monohydrate and pregelatinised starch to improve flow into a dosator or tamping pin capsule filling machine. The target fill formulation consists of 20–35% w/w API, 60–70% w/w lactose monohydrate, 5–10% w/w pregelatinised starch, 0.5–1.0% w/w colloidal silicon dioxide, and 0.5–1.0% w/w magnesium stearate. The blend is screened through a 20-mesh sieve and mixed in a bin blender for 15 min at 25 rpm. Fill weights between 150 mg and 500 mg are accommodated in size 1 or size 0 hard gelatin capsules; the choice depends on bulk density, which is maintained between 0.45 g/mL and 0.60 g/mL by adjusting the spray-dried lactose grade. Capsule fill weight variation is controlled per USP <905>, and disintegration per USP <701> with a limit of 15 min. Because hard gelatin capsules become brittle when the fill moisture is below 2.0% w/w and tacky above 4.0% w/w, the blend moisture is specified at 2.5–3.5% w/w before filling; HPMC capsules are substituted when bulk stability data show shell pellicle formation or crosslinking after storage at 40 °C / 75% RH for 6 months. The finished capsules are packed in PVC/PVdC-aluminium blisters or HDPE bottles with a desiccant. Low-shear blending below 15 min has produced fill weight drift in dosator machines, while over-blending above 45 min has generated electrostatic adherence to the dosator pins and should be avoided unless forced degassing is performed.

When a Finished Feed Dilution Drops Below 1 kg/T, Mixing Variance Becomes the Release Criterion

Geometric dilution of the API granulate into a carrier phase consisting of ground corn cob, calcium carbonate, and light mineral oil is used for medicated feed applications. The API is typically incorporated at 5–10% w/w in the premix, with carrier phase at 85–90% w/w and mineral oil at 0.5–1.5% w/w. A first-stage 1:10 manual preblend is passed through a 10-mesh screen before introduction into a 500 kg double-ribbon blender with a usable volume of 400 kg; blending continues for 20 min at 12 rpm. Ten sampling points are defined across the blender, and the active marker content is measured by a validated HPLC method; the coefficient of variation for blend homogeneity should not exceed 5.0% RSD. When the final feed inclusion rate drops below 1 kg/T, the variance in finished feed sampling becomes the primary release risk, and a separate finished feed homogeneity study is required per ISO 6497 for sampling and ISO 6498 for sample preparation. Carry-over testing after cleanout should include swab samples from the ribbon shaft, discharge gate, and elevator boot; regional limits for carry-over are applied to protect non-target species. The finished premix is packed in 25 kg multiwall paper bags with an inner polyethylene liner and closed by heat sealing. For food-producing animals, the withdrawal period and residue depletion profile must be obtained from the marketing authorisation holder; published data for this specific API in minor species are limited and should not be extrapolated without a targeted residue study.

When flock or herd medication is delivered through drinking water, the API is blended with anhydrous glucose or lactose monohydrate to produce a water-soluble oral powder. The API fraction is selected between 10% w/w and 50% w/w based on the labelled dose and drinking water volume; the carrier fraction ranges from 49% w/w to 89% w/w, and anhydrous citric acid is added at 0.5–1.0% w/w to maintain acid pH during reconstitution. The powder is blended in a V-type mixer for 20 min at 20 rpm, passed through a 40-mesh screen, and then discharged into aluminium foil laminated sachets or HDPE jars with a desiccant. Reconstitution performance is evaluated by adding the powder to potable water at 1–2 g/L and stirring for 2 min; undissolved material retained on a 75 µm sieve should not exceed 0.5% w/w of the nominal fill weight unless the product is explicitly a suspension. The powder moisture content is controlled at ≤4.0% w/w by loss-on-drying per USP <731>, and water activity is kept below 0.60 to reduce microbial growth. Non-sterile microbial limits are applied per USP <61> and USP <62>, with absence of Salmonella and Escherichia coli required for the bulk powder. The terminal product is labeled with the reconstituted solution shelf life, which is typically limited to 24 h at 25 °C unless preservative efficacy is demonstrated according to USP <51>. Caking during storage at 30 °C / 65% RH has been traced to residual moisture above 4.5% w/w and can be reduced by lowering the carrier moisture prior to blending and sealing sachets under 35% RH.

Fluid-Bed Regranulation Limits and Loss-on-Drying Endpoints

For final oral granule products intended for direct administration or reconstitution, the supplied API is regranulated by top-spray fluid-bed processing to adjust particle size, flow, and wetting kinetics. The batch composition uses 40–60% w/w API, 30–50% w/w mannitol or sucrose granules, 2–5% w/w PVP K30 binder, and 0.5–2% w/w palatability aids. The binder solution is sprayed from a top-spray nozzle with an atomisation air pressure of 2.0–3.0 bar onto the fluidised bed; inlet air temperature is 55–65 °C, product temperature is held at 30–35 °C, and spray rate is adjusted to keep the product below the over-wet endpoint. The granulation endpoint is defined by loss-on-drying at 1.5–2.5% w/w; in-process particle size is checked by sieve analysis with a target D50 of 150–300 µm, D10 above 75 µm, and D90 below 600 µm. Over-granulation above 600 µm reduces reconstitution speed, while excessive fines below 75 µm create dust and segregation during sachet filling. The dried granules are filled into sachets of 1 g to 5 g target fill weight on a volumetric auger filler; fill weight variation is controlled per USP <905>. Reconstitution time is tested in 500 mL of water at 25 °C with stirring at 100 rpm; complete dispersion should occur within 120 s for a freely dispersible granule. Moisture is confirmed by Karl Fischer titration, not only loss-on-drying, because bound water in the botanical fraction can produce misleading endpoint values. Terminal packaging uses aluminium foil laminate sachets to keep moisture gain below 0.3% w/w during 6 months at 25 °C / 60% RH. In feed-room conditions where humidity exceeds 60% RH, pre-drying of the API at 50 °C before weighing is required to prevent crust formation at the spray nozzle.

Because young ruminants and swine often reject solid feed during acute disease episodes, a ready-to-use oral drench solution is prepared by dissolving the API in purified water with a co-solvent system. The vehicle consists of propylene glycol at 5–15% v/v or glycerol at 5–10% v/v, sodium benzoate at 0.1–0.2% w/v as preservative, citric acid to pH 4.0–5.5, and purified water to volume. The active concentration is expressed as mg of marker compound per mL; pilot batches range from 5 mg/mL to 20 mg/mL total marker, depending on target species and dose volume, but the exact value must be aligned with the authorised veterinary label. The solution is mixed in a 1,000 L high-shear mixer for 30 min, then passed through a 10 µm polypropylene filter before filling into 1 L amber PET bottles with tamper-evident closures. Amber packaging is used because the botanical marker fraction undergoes ultraviolet-accelerated degradation under ICH Q1B conditions; when clear PET is required for market presentation, the label must include light-protection storage and the stability program must include direct light exposure. The finished solution is tested for pH per USP <791>, preservative content by HPLC, and non-sterile microbial limits per USP <61> and USP <62>. Precipitation during storage at 2–8 °C has been observed when the co-solvent fraction drops below 5% v/v; freeze-thaw cycling of 3 cycles from -5 °C to 25 °C is used to confirm physical stability. The product is administered as a single oral drench with a dosing pump calibrated to deliver the labelled volume; the zero-dose period and withdrawal period for food-producing species must be sourced from the marketing authorisation holder and printed on the outer carton.