What Restricts Direct Compression of the Botanical Powder in Multi-Station Presses?

In pilot-scale oral solid dosage production, direct compression of the dried Dijin Hecao powder is constrained less by the mechanical compression force applied on a rotary press than by the residual moisture equilibrium of the botanical matrix under uncontrolled room relative humidity. Batches conditioned at RH ≥60% for more than 4 h have shown picking on 10 mm round concave tooling, with average ejection force rising from below 0.4 kN to above 0.9 kN during consecutive compression cycles. The root cause is the hygroscopicity of the powder's tannin fraction, which lowers the softening point of the compressed material and increases adhesion to steel tooling surfaces. A production-scale remedy is to pre-dry the API in a fluid-bed dryer at 45–50°C until loss on drying by USP <921> Karl Fischer or infrared balance is ≤5.0%, then compress within an environment maintained at 35–40% RH.

For a direct-compressed tablet core, a starting addition ratio for the API is 10–25% w/w, with the remainder composed of microcrystalline cellulose at 30–50% w/w, spray-dried lactose monohydrate at 20–40% w/w, croscarmellose sodium at 2.0–5.0% w/w, and magnesium stearate at 0.5–1.0% w/w. The exact API fraction is corrected against the supplier certificate of analysis for total tannin or flavonoid marker content so that the finished tablet does not drift outside the intended label claim. Blend uniformity is assessed by stratified sampling with acceptance criteria derived from USP <905> Uniformity of Dosage Units; tablet friability is controlled at ≤1.0% after 100 revolutions in the apparatus described in USP <1216>, and disintegration is verified in 0.1 N hydrochloric acid at 37±2°C per USP <701> with complete disintegration within 30 min.

The downstream process for the tablet route comprises screening the pre-dried API and excipients through a 1.0 mm sieve, blending in a bin blender at 15 rpm for 15 min, and compressing on a 16-station rotary tablet press at average hardness 8–14 kN. If the tablet is film-coated, a hydroxypropyl methylcellulose coating at 3.0–4.0% weight gain is applied after core compression. The terminal product is a nonsterile oral tablet for companion-animal veterinary pharmacies, distributed in bulk for repackaging or in final blister formats where the market authorization permits it.

Membrane fouling, pyrogen burden, and the sterile filling path for injectable solutions

Injectable-grade processing of this powder begins with a depyrogenation and clarification step that is not required for oral forms; the as-supplied oral-grade API is not injected directly. For an injectable solution, the powder is normally re-extracted or redissolved, filtered through a depth-media filter, and spray-dried or freeze-dried before sterile formulation to reduce endotoxin, water-insoluble fiber, and high-molecular-weight polyphenol aggregates. The final sterile solution typically contains 1.0–5.0% w/v of the processed extract, a co-solvent such as propylene glycol or PEG 300 at 20–40% v/v, and a citrate or phosphate buffer to maintain pH 4.5–6.5. The addition ratio is set on a mass-per-volume basis: a 2.0% w/v batch requires 20 g of processed API per liter of final solution, subject to adjustment by marker assay.

The principal downstream process risk is not terminal sterilization chemistry but membrane fouling during aseptic filtration. Tannins and residual polysaccharides in the extract can occlude 0.22 µm PVDF final membranes when the transmembrane pressure exceeds 0.8 bar; a staged filtration train with a 0.45 µm polyethersulfone prefilter and a 0.22 µm final membrane is therefore used, with maximum transmembrane pressure held below 0.8 bar. Terminal steam sterilization at 121°C for 15 min can cause color increase, polymerization, and precipitant formation in some batches; aseptic filtration followed by filling under ISO 14644-1:2015 Class 5 conditions is therefore used. Published data for extract-specific membrane fouling under commercial batch sizes is limited; filter capacity should be confirmed by a Vmax test on each batch before scale-up.

The injectable finished product is tested for sterility per USP <71>, bacterial endotoxins per USP <85> with an acceptance criterion of not more than 0.5 EU/mg of extract, and particulate matter per USP <788>. Residual solvents from any extraction step are controlled under USP <467> or VICH GL18, and elemental impurities are controlled under USP <233>. The terminal product is a sterile injectable solution for cattle, sheep, or swine, typically presented in 50 mL or 100 mL vials; a multidose presentation may include benzyl alcohol at 1.5% v/v if approved in the target market, but the preservative must be excluded from intrathecal or intra-articular routes.

Broiler and layer production systems that still maintain withdrawal-period constraints frequently evaluate water-soluble botanical powders as adjunctive products for periods of intestinal challenge, but the downstream formulation is not simply a dilution of the raw milled herb. The API is incorporated into a concentrated water-soluble powder at 20–40% w/w, with the balance composed of glucose monohydrate or lactose monohydrate carrier, sodium citrate dihydrate at 5.0–10.0% w/w, and anhydrous citric acid sufficient to bring a 1 g/L solution to pH 4.5–5.5. At farm level, the concentrated powder is typically metered at 0.5–1.0 kg per 1000 L of drinking water, although the actual intake varies with water consumption rate, ambient temperature, and age of the flock.

In manufacture, the raw API and carrier are de-agglomerated through a 0.8 mm conical mill, blended in a ribbon blender for 20 min, and filled into laminated foil sachets under a nitrogen flush if the residual moisture exceeds 5.0%. The finished powder is checked for reconstitution behavior: a 1 g sample in 1 L of hard water at 25°C should disperse without visible sediment within 3 min. If the source water contains iron or calcium above 200 mg/L total hardness, the formula may require an additional chelating acidifier because tannin fractions can complex with divalent cations and form insoluble precipitates.

Microbial limits follow USP <61> and USP <62>, with absence of Salmonella in 25 g and yeast/mold count below 100 CFU/g. Homogeneity is tested on 10 stratified samples by UV-Vis marker content with RSD ≤5.0%. The terminal product is a water-soluble powder for in-line drinking water proportioner systems in broiler and layer houses, intended for short-duration administration under veterinary oversight.

Weaned piglet premix stability demands a pelleting temperature below the degradation threshold

Feed mill trials with this API consistently identify the conditioner temperature as the parameter with the most pronounced effect on post-pellet marker recovery. Steam conditioning at 75–85°C for 20–40 s is normally required to gelatinize starch and achieve pellet durability above 90% on a Holmen durability tester, but temperatures above 85°C can reduce extractable marker compound recovery by 12–18% compared with the unprocessed mix. The reduction is attributed to binding of tannin fractions to denatured feed proteins and, in some batches, to Maillard-type reactions with reducing sugars. The process control window for feeds containing this API is therefore narrowed: conditioning at 70–75°C with 16–17% moisture, pellet die diameter 2.5–3.5 mm, and post-pellet cooling to within 5°C of ambient.

In a medicated premix, the API is typically incorporated at 2.0–5.0% w/w of the premix, with a carrier such as calcium carbonate or wheat middlings. When diluted at 10–20 kg premix per tonne of finished feed, the resulting final feed contains 500–1500 g of Dijin Hecao powder per tonne; exact inclusion is adjusted to the approved representative species and batch-specific marker assay. The desired coefficient of variation for the active marker across 10 sampled points is ≤10%.

Manufacturing compliance in relevant EU markets falls under Regulation (EU) 2019/4 on medicated feed, with homogeneity and carry-over control demonstrated in manufacture; sampling may follow ISO 6497:2005 for animal feeding stuffs. The terminal product is a pelleted medicated feed for weaned piglets during post-weaning enteric challenge periods, shipped in 25 kg or bulk bags with moisture controlled to ≤13.0% in finished pellets.

Because capsule fill precision in small-batch veterinary compounding is governed by powder flow and density rather than compression behavior, the API requires a different particle-size specification than tablet manufacturing. The raw powder is often milled to a d50 of 80–120 µm and blended into size 1 or size 0 hard gelatin capsules at 40–70% w/w API, corresponding to a total fill weight of 200–500 mg. The remaining excipients are microcrystalline cellulose at 20–50% w/w, colloidal silicon dioxide at 0.5–1.5% w/w, and sodium starch glycolate at 2.0–4.0% w/w if rapid disintegration is required. The capsule fill is calculated from the supplier COA marker assay rather than from raw powder weight alone.

The compounding process consists of a pre-mix of the API with colloidal silicon dioxide for 5 min, screening through a 0.5 mm sieve, blending with microcrystalline cellulose in a V-blender at 25 rpm for 15 min, and filling on a semi-automatic capsule machine with tamping pins adjusted to give a tapped density of 0.45–0.65 g/mL. Fill weight is checked every 15 min, with rejection limits set at ±5.0% of target; capsules are polished and visually inspected before packaging.

Nonsterile capsules must meet microbial limits per USP <61> and USP <62>; water activity measured with a dew-point instrument is controlled below 0.60 to prevent mold growth and gelatin cross-linking. The terminal product is a compounded capsule used in veterinary pharmacies or contract manufacturing for companion animals and exotic species, where dose flexibility and small batch size override the higher production cost.

When high-humidity barn top-dressing makes granulation mandatory

Under high ambient moisture in open-barn environments, the free-flowing API powder tends to cake within 24–48 h if packaged in polyethylene-lined kraft bags without desiccant. Granulation is therefore the preferred downstream route when the product is intended for on-farm top-dressing of feed. The powder is wet-granulated in a high-shear mixer with an aqueous binder solution of povidone K30 at 2.0–5.0% w/w dry basis; the wet-mass target moisture is 12–14% loss on drying before drying. Drying is performed in a fluid-bed dryer with inlet air temperature 55–60°C and product temperature not exceeding 40°C to preserve the tannin fraction. The dried granules are sized through a 1.5 mm oscillating sieve, and the 0.5–1.5 mm fraction is retained for packaging; fines below 0.5 mm are recycled into the next wet granulation batch at no more than 20% of the total mass.

The granule formula contains 30–50% w/w Dijin Hecao powder, corn starch at 20–40% w/w, microcrystalline cellulose at 10–20% w/w, povidone K30 at 2.0–5.0% w/w, and colloidal silicon dioxide at 0.5–1.0% w/w. At use, the granules are top-dressed at a rate of 20–50 g per animal per day depending on body weight and prescriber instruction; exact dosing must be established under veterinary supervision and is influenced by marker content and target species.

Water content in the finished granules is controlled by USP <921> to ≤6.0%; granule friability is monitored by a sieving test after 10 min in a Retsch AS 200 at 1.5 mm amplitude. The povidone K30 supplier specification for residual peroxide should be reviewed because peroxide can oxidize polyphenol markers and reduce batch potency. The terminal product is an oral granule for top-dressing on feed for ruminants, horses, or swine in large-animal practice, packaged in sealed polyethylene terephthalate jars with desiccant.