Liquid propionic acid is metered into compound feed at 0.1–0.5% by mass, with the undissociated acid fraction controlled by the pKa of 4.87 at 25 °C. In a 3 m³ horizontal paddle mixer, direct spray injection is typically completed within the final 30–60 s of the batch cycle to limit contact between free acid and 316L stainless steel parts. The closed-cup flash point of liquid propionic acid is near 54 °C, the boiling point is 141 °C, and vapour pressure is approximately 0.4 kPa at 20 °C; these properties force vapour extraction, PTFE-sealed pumps, and inert-gas line flushing. The operating problem described by buffer salts and coated forms improving propionate handling in feed mills arises because the liquid is corrosive, volatile, and difficult to retain through standard steam conditioning at 70–85 °C.

When liquid propionic acid is sprayed onto a premix containing calcium carbonate, the local pH drop releases carbon dioxide and converts limestone to calcium propionate at the droplet surface. The resulting paste can build up on mixer shafts, reduce batch uniformity, and transfer propionate activity into the pellet mill. Buffered salt solutions avoid the lowest pH excursion because they contain little free acid; however, the same equilibrium that reduces corrosivity also reduces the fraction of undissociated propionic acid available for antimicrobial action. At pH 5.5, the undissociated fraction is approximately 19%; at pH 6.0, it is approximately 7%. This pH dependence must be considered when buffer salts are added to feeds with high limestone or dicalcium phosphate buffering.

What Changes When Ammonium Propionate Replaces Liquid Acid in a Batch Mixer?

Ammonium propionate contains 80.2% propionate anion by mass, while calcium propionate and sodium propionate contain 78.5% and 76.1% respectively. If a batch card specifies 0.3% liquid propionic acid, the equivalent propionate anion addition is 0.296% because liquid propionic acid is 98.6% propionate anion by mass. Replacing that input with ammonium propionate requires 0.37% of the salt, not 0.3%. A liquid acid pump calibrated with a mass flow meter at 0.3% cannot be transferred directly to a dry salt feeder without recalculating active anion and adjusting the feeder setpoint.

On a 5 t/h line, an aqueous ammonium propionate solution at 50% solids delivered by a positive displacement pump with EPDM seals reduces free acid vapour and iron propionate residue formation on 304 stainless steel. The dosing solution pH rises from <2 for liquid acid to 6.5–7.5 for the buffered salt. This pH shift lowers corrosivity, but the solution viscosity increases at low storage temperatures; jacketed lines or storage at 15–25 °C are required when ambient temperatures fall below 10 °C. In high-buffering feeds, the salt may not depress the feed surface pH below the pKa, so preservation can fail if the dose is not raised or the formulation is not acidified in the mixer with a separate organic acid.

Dry calcium propionate and sodium propionate are handled as crystalline solids. Calcium propionate has a water solubility near 49 g/100 mL at 20 °C; sodium propionate is more soluble, near 100 g/100 mL at 20 °C. In a 2 t ribbon mixer, pre-blending the salt with 5–10% of the mineral carrier before feed addition avoids localized wetting from moist grains and reduces lump formation. Transfer lines should be short and vertical with 60° or steeper chutes; pneumatic conveying into a scale hopper can generate static dust that adheres to the hopper walls and should be controlled by dust extraction meeting 21 CFR 507 animal food CGMP requirements and EU feed hygiene regulation 183/2005.

When Storage Dew Point Exceeds 15 °C, Dry Propionate Caking Accelerates

Dry propionate powders absorb atmospheric moisture when the storage silo wall temperature drops below the air dew point. In a coastal feed mill at 20 °C and 70% relative humidity, condensation on uninsulated silo walls creates localized caking that later releases as slabs into the discharge auger. Published data for the moisture uptake rate of commercial propionate powders under continuous mill storage are limited; however, the processing boundary is established by silo cone half-angles of at least 70°, fluidizing pads supplied with dried air at 1–2 bar, and rotary valves with 0.2–0.5 mm clearance to reduce bridging. Loss-in-weight feeders holding ±0.5% dose accuracy require refill intervals below 15 min; longer intervals allow flow rate drift and trigger alarms in automated batching systems.

Moisture caking also alters particle size distribution. If sieving is performed according to ISO 2591-1, the mass fraction retained above 1.0 mm should remain below 5% after 24 h storage; otherwise, the rotary valve may bind and the metering screw can overload. The addition of 0.2–0.5% precipitated silica or calcium stearate reduces interparticle cohesion, but this dilutes the propionate activity and must be accounted for in the batching records.

Dose Control and Flow Defects in Coated Propionate Premixes

Coated propionate particles are typically produced by spray congealing or fluid-bed encapsulation. A hydrogenated lipid melt at 10–20 °C above its melting point is atomized into a chilled chamber, or propionate crystals are coated in a Wurster column. The resulting micro-prills often have a mass median diameter of 300–800 µm; particles above 1.0 mm segregate in premix bins, while fines below 100 µm create dust and expose core material. A narrow particle size distribution with a coefficient of variation below 20% supports uniform feeding; however, published correlations between particle size distribution and preservation efficacy for specific proprietary coatings are limited. The dosing line for coated forms should include a mass flow hopper, a vibratory feeder with 0–100 kg/h range, and a rotary airlock vented to a dust collector.

Surface oil on coated particles can smear on feeder screws and produce feed rate drift. Extraction of the surface lipid by petroleum ether and analysis of free oil content should be part of incoming QA; if free oil exceeds 2% of particle mass, the feeder screw should be inspected after each shift. Coated calcium propionate at 0.4% by mass delivers 0.31% propionate anion when the core salt is 78.5% propionate anion.

Active Propionate Equivalent and Handling Matrix

Table 1 compares the propionate anion contributions and handling profiles of the primary forms used in compound feed production. The values are calculated from formula weights and are used to adjust batching setpoints when switching between liquid, salt, and coated salt forms.

Propionic acid is cleared as a feed antimicrobial under 21 CFR 184.1081, calcium propionate under 21 CFR 184.1221, and sodium propionate under 21 CFR 184.1784. In the EU, feed hygiene is regulated under 183/2005, and the substance identity uses EC 201-176-3. Worker exposure limits for propionic acid vapour are commonly set at 10 ppm as an 8-hour time-weighted average; local exhaust ventilation is required when liquid acid is sprayed in open mixers. Coated and buffered forms reduce vapour exposure, but dust and particle handling require separate controls under 21 CFR 507 and the EU animal feed additive framework.

Thermal Degradation Pathways in Coated Propionate During Pellet Die Passage

Pellet conditioning at 70–85 °C with 30–90 s residence time creates the main thermal boundary for coated propionate. A coating based on fully hydrogenated palm oil with a melt onset near 58 °C may release during conditioning at 80 °C, while fully hydrogenated rapeseed oil with a melt onset near 70 °C survives moderate conditioning but can fail at 85 °C. Differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) map the coating melt profile; a sharp endotherm with an onset below the conditioner temperature indicates that the protective layer will soften before pellet die entry.

The particle Biot number determines whether the core reaches the lipid melt onset during short residence. For particles with a small Biot number, internal temperature gradients are limited; however, published data for the heat transfer kinetics of specific coated propionates in steam conditioning are limited. Mill trials should collect samples from the conditioner discharge, pellet die, and cooler discharge, with propionic acid recovery measured by steam distillation or ion chromatography and compared to the propionate anion input. The uncoated control is expected to show greater loss across the pellet die when free acid vapour is stripped by steam; direct comparative recovery data in commercial mills are limited.

In a 5 t/h line producing 3 mm broiler crumbles, the decision between liquid acid, buffer salts, and coated forms is driven by available post-pelleting application equipment. If post-pelleting liquid acid spraying is not installed, coated calcium propionate or ammonium propionate buffer salt can be incorporated in the main mixer to reduce handling exposure and improve retention through conditioning. The batching software should contain separate raw material records for each propionate source because active anion content, moisture, and particle size distribution differ. When switching from liquid acid at 0.2% to coated calcium propionate, the batch weight must be recalculated to 0.25% to deliver the same propionate anion as 0.2% liquid propionic acid if the coated salt contains 78.5% propionate anion. Mills that verify the delivered propionate anion by ion chromatography or steam distillation avoid under-dosing caused by salt or coating fraction differences.