In mechanically agitated 30 m³ jacketed 316L fermenters fitted with three-stage Rushton turbines and a side-entry powder induction manifold, ACP-1920 is dry-fed as a free-flowing powder to suppress protein-induced foam in thermostable α-amylase and neutral protease fermentations using Bacillus subtilis and Bacillus licheniformis strains. The charge media for these batches combines 12–18 wt% corn starch hydrolysate at DE 18–22, 3–8 wt% corn steep liquor, and 1–2 wt% soybean meal with pH adjusted to 6.8–7.2 and sterilised at 121 °C for 30 min. Foam generation initiates at 6–8 h post-inoculation, when extracellular protein and glucan oligomers raise the apparent viscosity to 35–120 mPa·s at a shear rate of 100 s⁻¹ and the broth shifts from Newtonian to shear-thinning behaviour. The powder is metered through a weight-loss feeder at 0.02–0.08 g/L active polydimethylsiloxane, drawn into a high-velocity eductor using 0.45 µm filtered compressed air, and discharged below the liquid surface at the radial transition between the lower and middle Rushton impellers. Recirculating-column testing according to ASTM E2407-04(2015) is used as a release method for foam knockdown; silicone defoamers at 0.02–0.08 g/L active polydimethylsiloxane typically reduce foam half-life by 80–90 % in protein-stabilised foam systems, though batch-specific ACP-1920 data should be verified against the actual corn steep liquor lot. Foam height is measured by capacitance probes mounted at the vessel headspace, and the feed rate is trimmed against the resulting signal because batch-to-batch variation in corn steep liquor saponin content causes a ±15 % shift in the required dose. Over-addition above 0.15 g/L active polydimethylsiloxane is known to reduce the volumetric oxygen transfer coefficient \(k_L a\) in sparged protein broths by 15–40 % relative to defoamer-free controls; this effect is amplified when the fermenter is operated at a low airflow of 0.5 vvm and high cell density above 60 g/L dry cell weight. The operational window for oxygen-sensitive enzyme production is therefore maintained at 0.02–0.08 g/L, with a narrower trim range of ±0.02 g/L used once the dissolved oxygen controller begins to command additional agitation. Residual polydimethylsiloxane entering the downstream clarification train must comply with FDA 21 CFR 173.340 as a defoaming agent and, where the enzyme is placed on the EU market, with Regulation (EC) No 1333/2008 Annex II for dimethylpolysiloxane carry-over in food enzyme preparations. The fermentation broth is cooled to 8–12 °C and clarified on a rotary vacuum filter precoated with diatomaceous earth; the filtrate is then concentrated on 10 kDa polyethersulfone spiral-wound cassettes at a transmembrane pressure of 1.2–1.5 bar and a crossflow velocity of 1.0–1.5 m/s. Silicone antifoam residues above 20 mg/kg in the filtered broth are associated with a measurable increase in membrane resistance because hydrophobic polydimethylsiloxane adsorbs onto the polyethersulfone surface and reduces clean-water flux recovery after caustic–hypochlorite cleaning. Terminal spray-dried products include bacterial neutral protease and α-amylase preparations tailored for bakery and brewing process aid use.

What Operational Windows Govern Defoamer Addition in Molasses-Based Yeast Propagation?

Because saponins in cane molasses produce a heat-stable foam that persists through 121 °C sterilisation and enters the downstream stripping column, yeast producers commonly apply ACP-1920 as a dry-powder defoamer in the molasses dilution tank at 10–25 mg/L active polydimethylsiloxane. The diluted molasses feed typically holds 12–18 °Bx dissolved solids, with pH adjusted to 4.8–5.5 and temperature controlled at 28–32 °C; aeration in the first propagation stage is 0.5–1.0 vvm, increasing to 1.5 vvm in the final stage. ACP-1920 is introduced into the molasses dilution tank through a high-shear mixer operating at 1,500–3,000 rpm before the broth is transferred to the propagation vessel. The powder feed must be located upstream of the plate heat exchanger because the thermal gradient across the exchanger plates can cause localised foam collapse and uneven distribution of the silicone within the preheated molasses stream. Foam height is monitored by a radar sensor in the vessel headspace, and the powder is metered through a twin-screw feeder to maintain the 10–25 mg/L active polydimethylsiloxane range. If the dose exceeds 35 mg/L, the dissolved oxygen concentration at constant agitation and sparge rises less than expected because published studies on polydimethylsiloxane in aerobic fermentation report oxygen transfer coefficient losses of 15–40 % at high defoamer loads; batch-specific data for ACP-1920 are limited. The oxygen limitation is particularly critical during the Crabtree-positive phase of Saccharomyces cerevisiae growth, when ethanol formation is favoured by glucose pulses above 0.5 g/L and dissolved oxygen falls below 10 % saturation. After yeast separation in a disc-stack centrifuge operating at 9,000 × g, the spent medium contains residual silicone; if the plant operates a side-stream ultrafiltration unit to recover yeast extract or treat process water, the membrane flux on 10 kDa regenerated cellulose is more stable than on polyethersulfone because the hydrophilic regenerated cellulose surface adsorbs less polydimethylsiloxane. Terminal products include baker’s yeast cream at 18–22 wt% dry matter, compressed yeast, and instant dry yeast. In bioethanol operations, the stillage from the distillation column is sent to a decanter centrifuge; silicone carryover into distillers grains must be below the applicable feed ingredient limit, and published data for this specific configuration are limited.

At the point where ACP-1920 is applied to filamentous Aspergillus niger broths for citric acid production, the culture has already developed pseudowalls and extended mycelia that trap air and create a stable foam column. In a 200 m³ bubble column or stirred tank with a non-agitated air-lift section, the defoamer is pulse-fed at 20–50 mg/L active polydimethylsiloxane when the foam height exceeds 1.0 m above the liquid level. The broth exhibits apparent viscosities of 30–150 mPa·s at 50 s⁻¹ and contains high concentrations of citric acid, with harvest pH in the range 2.0–3.5. ACP-1920 must disperse at low pH where the mineral carrier may dissolve and leave free silicone; poor dispersion at this stage can generate hydrophobic agglomerates that attach to the mycelial mat and reduce filterability. Overdosing above 80 mg/L active polydimethylsiloxane is linked to downstream ion-exchange fouling because hydrophobic polydimethylsiloxane can adsorb to strong acid cation resin beads used in the citric acid recovery sequence. The clarification train includes a rotary vacuum filter precoat of rice hull ash or diatomaceous earth, followed by crossflow ultrafiltration on 100 kDa polysulfone membranes at 50–60 °C, a transmembrane pressure of 2.0–3.0 bar, and a crossflow velocity of 4–6 m/s in tubular modules. Residual polydimethylsiloxane in the clarified citric acid broth should be controlled below 10 mg/kg silicon to avoid haze in citric acid crystals; silicon may be quantified by ICP-OES according to ISO 11885:2007. Terminal products include citric acid monohydrate, trisodium citrate, and anhydrous citric acid for beverage, confectionery, and preservative applications.

Centrifugation and Tangential-Flow Filtration Boundaries in Probiotic Broth Defoaming

A pronounced reduction in cell adhesion in simulated gastrointestinal assays has been observed when overdosed defoamers accumulate in Lactobacillus and Bifidobacterium concentrates; therefore ACP-1920 is limited to 5–15 mg/L active polydimethylsiloxane after the exponential phase begins and the headspace foam exceeds 300 mm. The fermentation medium contains 2.0–4.0 g/L Tween 80, 5.0–10.0 g/L yeast extract, 20–30 g/L lactose, and buffer salts; pH is controlled at 6.5–6.8 and temperature at 37 °C. Because Tween 80 and yeast extract together generate a persistent interfacial film, the silicone powder works primarily by spreading on the air–liquid interface and displacing the protein–surfactant complex that stabilises the foam. Excessive silicone can alter the cell surface hydrophobicity of probiotic strains and reduce adhesion to mucus in simulated gastrointestinal assays, though published data for this specific ACP-1920 configuration are limited. The broth is cooled to 4–8 °C and separated in a disk-stack centrifuge operating at 9,000 × g; the cell concentrate is washed by diafiltration on 0.1 µm ceramic crossflow membranes at a transmembrane pressure of 1.0–1.8 bar and a crossflow velocity of 5–7 m/s. Permeate flux declines sharply if the silicone dose exceeds 20 mg/L, because polydimethylsiloxane deposits on the ceramic lumen and is not fully removed by standard caustic–chlorine cleaning at 80 °C; acid cleaning with 1 % nitric acid followed by 0.5 % hydrogen peroxide is required to restore 90 % of clean-water flux. Terminal powders are lyophilised at a shelf temperature ramp from −35 °C to +20 °C and typically contain 10¹¹ CFU/g viable cells. The finished product must comply with the relevant carry-over limits for dimethylpolysiloxane under Regulation (EC) No 1333/2008 and, where applicable, the microbiological criteria designated in the importing jurisdiction.

When Acid Whey from Fermented Dairy Processes Enters 10 kDa Polyethersulfone Systems

Acid whey derived from lactic fermentation, such as the byproduct of strained yogurt production, contains pH 4.5–4.7, 0.6–0.9 wt% lactic acid, 5–8 wt% total solids, and denatured whey proteins that stabilise foam during pump transfer and air incorporation. ACP-1920 is added in-line at 10–20 mg/L active polydimethylsiloxane before a centrifugal pump that feeds the ultrafiltration skid; the point of addition is upstream of an in-line static mixer with 6–8 elements to disperse the powder without causing excessive shear. The ultrafiltration system uses spiral-wound polyethersulfone membranes with a molecular weight cut-off of 10 kDa; normal operating conditions are a transmembrane pressure of 1.0–1.5 bar, a feed temperature of 45–50 °C, and a crossflow velocity of 0.6–1.0 m/s in the feed channel. If the silicone dose exceeds 25 mg/L active polydimethylsiloxane, the silicone can form a secondary hydrophobic film on the membrane surface, and membrane surface adsorption of polydimethylsiloxane is documented for polyethersulfone; at silicone residual above 20 mg/kg, clean-water flux recovery may fall by 15–25 % relative to the pre-exposure baseline, though site-specific pilot data are required. This effect is more pronounced on polyethersulfone than on hydrophilic regenerated cellulose, and it can be detected by a rise in feed pressure at constant retentate valve position. Cleaning with 0.5 % caustic at 50 °C for 30 min followed by acid rinse restores flux, but if the defoamer is allowed to age in the retentate loop for more than 4 h, a two-stage cleaning with 0.1 % sodium hypochlorite at 40 °C is required. The retentate is concentrated to 18–22 wt% total solids and then spray-dried to produce whey protein concentrate 34–80 wt% protein; residual polydimethylsiloxane must comply with the applicable E900 limits in the final food category, and silicon may be quantified by ICP-OES according to ISO 11885:2007. The primary operational boundary is the combination of low pH and high calcium content; acid whey can destabilise the silicone dispersion if the powder is added directly to the retentate side rather than the feed side, because carrier dissolution is less effective in high-viscosity retentate.

L-Glutamic acid recoveries above 80 % from Corynebacterium glutamicum fermentation require a foam-control strategy that does not compromise downstream isoelectric precipitation or ion-exchange resin capacity. In a 50–150 m³ stirred fermenter fed with glucose or tapioca starch hydrolysate and ammonium sulfate, the broth foams rapidly during the late-exponential phase due to phospholipids and glycolipids released by the cell membrane. ACP-1920 is fed at 15–30 mg/L active polydimethylsiloxane through a stainless steel powder induction hopper equipped with a vibratory bridge breaker and a dehumidified air purge; the hopper is installed in a room conditioned to ≤50 % relative humidity to prevent the carrier from absorbing moisture and clogging the eductor. Batch-to-batch variation in the phospholipid content of the corn-based carbon source changes the dose by ±10 %; the control system uses a foam capacitance probe and a timed pulse feeder that adds the powder in 5 g increments per 100 m³ working volume. Fermentation is run at 32–34 °C, pH 7.0–7.5, and a dissolved oxygen setpoint of 20 % saturation; addition above 50 mg/L active polydimethylsiloxane can lower the \(k_L a\) enough to require a 10–15 % increase in airflow to maintain the same dissolved oxygen, which in turn increases evaporative water loss and broth osmotic strength. After the fermentation is terminated at 36–48 h, the broth is clarified on a rotary drum filter and ultrafiltered through 10 kDa spiral-wound membranes; residual antifoam above 15 mg/kg silicon in the ultrafiltered liquor is associated with reduced anion resin capacity for glutamate binding. The final crystallised monosodium glutamate must meet the heavy metal and impurity specifications of the food-grade market; its defoamer carry-over is regulated under FDA 21 CFR 173.340 and EU Regulation (EC) No 1333/2008.