The veterinary-grade API of Buzhong Yiqi Powder is a spray-dried or vacuum-dried extract of the classical multi-herb formula, not a single-entity synthetic compound. It is handled as an amorphous, hygroscopic, brown-to-tan powder with a characteristic bitter-aromatic odor and a complex chemical matrix that includes astragaloside IV, calycosin, ferulic acid, liquiritin, atractylenolide, and water-soluble polysaccharides. The product is supplied in several vendor-specific grade classes differentiated by extraction ratio, particle-size distribution, residual moisture, microbial burden, and endotoxin budget. Typical grade identifiers follow a suffix convention: BZYQ-V-T for tablet/capsule/powder conversion, BZYQ-V-G for granulation, BZYQ-V-I for injection-grade processing, and BZYQ-V-P for premix/feed blending. Because no global harmonized monographs exist for botanical veterinary APIs, the pharmacopoeial reference is jurisdiction-specific; manufacturers relying on the Chinese Veterinary Pharmacopoeia must apply the current Buzhong Yiqi San monograph while validating impurities and marker ratios for the specific extraction site.

Identity is confirmed by botanical microscopy and high-performance liquid chromatography with diode-array and evaporative light scattering detection. A fingerprint similarity index of not less than 0.95 against a certified reference extract is a common vendor acceptance criterion, but it is not a substitute for species identification of each herbal component. Seasonal variation in Astragalus root and Angelica sinensis influences the ratio of calycosin to ferulic acid; batch-to-batch marker ratios may shift by ±15% unless blending is actively controlled. Published data for this specific configuration is limited; purchasing specifications should therefore include a defined marker ratio window rather than relying only on a single-component assay.

Compendial Status and Grade Coding for Veterinary Dosage-Form Conversion

Because the same extract cannot be used interchangeably across all dosage forms, the material is separated into grades. The main process drivers are particle size and extract ratio. Tablet and capsule grades can tolerate larger particles but require low residual moisture to avoid punch sticking. Granule grades require faster wetting and moderate cohesiveness. Injection-grade material must satisfy endotoxin and particulate limits that are rarely relevant for oral powders. Premix-grade powder is optimized for flow and dust suppression rather than dissolution. The table below lists representative vendor-documented target ranges for each grade; these are not compendial specifications and must be confirmed for the registered formulation.

Particle-size data are determined by laser diffraction according to ISO 13320:2020. Bulk density is measured by USP <616> Method I, moisture by USP <921> or Ph. Eur. 2.2.32, and heavy metals by ICP-MS with limits aligned to ICH Q3D where applicable.

Tablet and capsule processing of the BZYQ-V-T grade is dominated by poor flow and high hygroscopicity. The powder shows a Hausner ratio above 1.35 and a Carr index above 30% when measured according to USP <1174>, indicating cohesive flow. Direct compression is therefore not a robust process unless the product is co-processed with a highly flowable filler. Roller compaction at roll force 8–12 kN/cm and screen size 1.0 mm is preferred to dry granulation. If wet granulation is required, a hydroalcoholic binder containing 60–70% v/v ethanol reduces stickiness and prevents polysaccharide gelation. Drying temperature is kept below 60 °C to preserve astragaloside IV and prevent darkening. The granulated material should reach loss on drying ≤3.0% before compression. Punch sticking can still occur if ambient relative humidity exceeds 60%; in that case pre-drying at 50 °C for 2 h is applied to the granules. Formulators should avoid strongly alkaline wet granulation because saponin esters are hydrolyzed above pH 8.0.

Capsule filling of this botanical powder requires low compaction settings on dosator machines to prevent plug formation. A hydrophobic lubricant such as magnesium stearate is limited to 0.5% w/w to avoid delaying disintegration and to prevent marker recovery loss due to overwrapping of particles.

What Sterility-Assurance Controls Are Required for the Injectable Grade?

Injectable processing of the BZYQ-V-I grade assumes the dry extract is not sterile by default. It is a low-bioburden, low-endotoxin extract intended for further aseptic processing. Bacterial endotoxin testing using kinetic chromogenic LAL per Ph. Eur. 2.6.14 or USP <85> is applied to the reconstituted solution and, when required, to the dry powder using a validated extraction procedure. Endotoxin limits are set by the formula K/M, where K is the threshold endotoxin dose per kg and M is the maximum dose administered per kg. Because veterinary dose volumes vary widely across species, a single fixed limit is not meaningful; however, vendor documentation for the injection grade commonly targets <0.50 EU/mg as a starting point.

Aqueous decoction extracts carry Gram-negative bacterial lipopolysaccharides that are not destroyed by spray drying. Depyrogenation is therefore a necessary step for injectable use. The dry extract is reconstituted and passed through cation-exchange or affinity depyrogenation media before sterile filtration. Ultrafiltration with a 10 kDa–30 kDa molecular weight cutoff is used to remove lipopolysaccharide aggregates and high-molecular-weight pectin fractions that would otherwise foul the sterilizing filter. Because polysaccharide molecular weight distribution varies with harvest season, filterability must be confirmed for each batch rather than assumed from supplier data. Published data for this specific configuration is limited, and filter validation should follow ISO 13408-1:2011 for aseptic processing and PDA Technical Report 26 for filter validation.

Insoluble particulate control is the second bottleneck. Reconstituted injection-grade material contains starch fragments, tannin-protein complexes, and crystalline calcium oxalate needles if pH is not controlled. The solution is filtered through a 0.45 µm polyethersulfone prefilter and then a 0.22 µm polyethersulfone sterilizing filter under nitrogen pressure not exceeding 1.0 bar. At flux rates above 200 L/m²/h, membrane fouling accelerates when total polysaccharide content exceeds 5.0% w/w. A pH between 5.5 and 6.5 minimizes haze; alkaline conditions increase precipitation and should be avoided. After aseptic filling, the final injection is steam-sterilized only if stability data support terminal sterilization; many botanical formulations are sterilized by filtration because saponin degradation and color change are accelerated at 121 °C for 15 min.

If Wet Granulation Is Selected for Oral Granules and Solutions

For oral granules and solutions, the BZYQ-V-G grade is produced with a slightly lower moisture limit and finer D90 than the tablet grade because oral granules require rapid dispersion. The powder is not fully water-soluble. It forms a polydisperse suspension in which dissolved saponins and polysaccharides coexist with insoluble fiber particles. When preparing an oral solution or drench, the powder should first be dispersed in cold purified water at 25 °C under high-shear mixing at 1,000–2,000 rpm. Heating above 60 °C before hydration is complete can gel the polysaccharide fraction and produce difficult-to-disperse lumps. pH is maintained between 4.5 and 6.5 with 0.1 M hydrochloric acid or citrate buffer. Above pH 8.0, astragaloside saponins hydrolyze, and the formulation should not be combined with alkaline chemical additives.

For oral suspensions, a suspending agent is required because gravitational settling follows Stokes behavior modified by particle flocculation. A dispersion at 1.0% w/v without suspending agent shows visible sedimentation within 30–60 min; with a microcrystalline cellulose/carboxymethylcellulose co-structured system at 0.2–0.3% w/w, the suspension can be held without hard packing for 24 h. These are vendor-documented performance targets rather than compendial requirements. Incompatibility with cationic preservatives such as benzalkonium chloride should be evaluated because anionic polysaccharide fractions can form precipitates at concentrations above 0.01% w/w.

For granulation, fluid-bed processing is preferred over pan granulation because it reduces localized overwetting. Inlet air temperature is kept at 50–60 °C; product temperature is maintained below 40 °C. The binder is a hydroalcoholic solution of povidone at 3–5% w/w dry basis. Final granules are dried to loss on drying ≤3.0% and sized through a 1.0 mm screen. If the granules are to be filled into sachets, desiccant is added when the package is not an aluminum foil laminate, because the extract equilibrates to ambient moisture within 24 h at relative humidity above 60%.

Premix Blending Segregation Occurs When Particle Density and Shape Are Mismatched

Medicated feed premix operations expose the BZYQ-V-P grade to high-dust and segregation-prone conditions because spray-dried particles are hollow spheres with low tapped density. Bulk density measured by USP <616> Method I is typically 0.45–0.65 g/mL; after 1,250 taps, tapped density may be 10–15% higher. In tumble blenders, this densification creates a flow-induced density gradient: the fine API fraction migrates to the bottom and to vessel walls, reducing recovered potency in the first 10% of discharge after blending. To limit segregation, the premix carrier should have a similar particle-size distribution and density. A granular lactose or wheat middlings carrier with D50 between 150 µm and 250 µm is preferred. Colloidal silicon dioxide is included at 0.5–1.0% w/w to coat the extract particles and reduce electrostatic dust adhesion.

Mixing time is critical. Excessive blending beyond 20 min in a double-cone blender at 50% fill volume can increase demixing rather than improve uniformity. Production-scale ribbon blenders with 0.5 m³ working capacity are typically operated at 70–80% loading and 50 rpm. Blend uniformity samples are assayed by HPLC for astragaloside IV; acceptance limits of 90–110% label claim with a relative standard deviation not more than 5.0% are commonly applied, following the general principles of USP <905> or Ph. Eur. 2.9.40.

A direct comparison with single-entity veterinary APIs clarifies the specification and processing differences. Synthetic single-molecule actives such as tylosin phosphate or amoxicillin trihydrate are defined by a single potency assay, a crystalline or defined amorphous form, and predictable solubility. The herbal extract is defined by multiple marker compounds and fingerprint similarity, with batch-to-batch variability controlled by herb source and extraction ratio.

Finished dosage-form manufacturers should require batch-specific certificates that include microbial limits per USP <61> and USP <62>, absence of specified pathogens, residual solvent analysis per USP <467>, and elemental impurities by ICP-MS according to ICH Q3D. Because the material contains naturally occurring mineral content, heavy metal results should be read against the appropriate veterinary product limits, not human pharmacopoeial limits. Stored material should be kept in double polyethylene-lined drums with desiccant, at or below 25 °C and 60% RH, and used within the vendor-declared retest period. Any change in extraction ratio or particle size class requires revalidation of blend uniformity, dissolution or dispersion behavior, and, for the injectable grade, filterability and endotoxin recovery.