Industrial sodium hydroxide for saponification is supplied as 50% w/w aqueous solution, anhydrous micropearl, or flake. The membrane-cell solution grade typically contains NaOH ≥50.0% w/w, NaCl ≤0.01% w/w, Na₂CO₃ ≤0.05% w/w, and Fe ≤2 mg/kg; diaphragm-cell material may contain NaCl up to 1.0% w/w and is therefore less suitable for high-clarity toilet soap without additional electrolyte management. Sodium hydroxide is assigned CAS 1310-73-2, molecular weight 40.00 g/mol, and under CLP Regulation (EC) No 1272/2008 is classified as Skin Corr. 1A with H314 when concentration is ≥5% w/w, Skin Corr. 1B with H314 at 2–5% w/w, and Eye Irrit. 2 with H319 at 0.5–2% w/w. The raw material specification must include carbonate, chloride, sulphate, iron, and mercury limits because these species partition into the saponified mass and affect odour, colour, and phase stability. For high-transparency glycerine soap, iron above 2 mg/kg promotes rancidity and darkening; for industrial laundry soap, chloride tolerance is broader but still bounded by the required electrolyte balance in the neat soap phase.

How Does Sodium Hydroxide Drive Saponification Stoichiometry in Mixed Triglyceride Systems?

The saponification of a triglyceride proceeds with a fixed stoichiometric demand of 3 mol sodium hydroxide per 1 mol triglyceride, producing 3 mol sodium carboxylate and 1 mol glycerol. For triolein, molecular weight 885.45 g/mol, the theoretical sodium hydroxide demand is 120.00 g NaOH per 885.45 g triolein, equivalent to 135.5 g/kg oil. Because feedstocks are mixtures of triglycerides with different fatty acid distributions, plant dosing is controlled through the saponification value determined by ASTM D5558. The conversion from milligrams KOH per gram oil to grams NaOH per kilogram oil uses the factor 0.713, derived from the quotient 40.00/56.106. For a palm oil lot with a saponification value of 195 mg KOH/g, the stoichiometric NaOH demand is 139.0 kg per 1000 kg oil. Toilet soap formulations are commonly calculated with a lye discount of 5–8 mass %, so the actual charge for the same lot would be 127.9–132.1 kg NaOH per 1000 kg oil. The discount is not inert; it preserves a controlled unreacted oil fraction that reduces free caustic irritation and modifies bar hardness.

Once the caustic soda solution is charged to the oil phase, the system transitions from a two-phase oil-in-water emulsion through a high-viscosity gel stage and then into a smooth neat soap phase. The gel stage is the most constrained processing window in batch saponification; for a blended tallow-coconut charge, the kettle temperature is typically held at 80–90 °C, with a narrow control band of ±5 °C during the trace-to-gel transition. Below 75 °C, the reaction rate decays and the mass may remain as a poorly saponified emulsion with free oil streaks; above 95 °C, foaming and rapid water vapour evolution can carry soap over the kettle headspace. Agitation must be reduced from initial emulsification tip speeds of 3–6 m/s to 1–2 m/s during gel phase to avoid air entrainment and to prevent motor overload as the apparent viscosity reaches 20,000–50,000 mPa·s in weakly mixed zones. A recirculation loop with a positive-displacement pump is often used to maintain turnover through a scraped-surface heat exchanger because stagnant gel on the vessel wall forms an insulating layer that shifts the effective reaction temperature outside the control band.

In continuous saponification, the same phase transitions are managed in a cascade of 2–4 stirred reactors rather than a single vessel. The first reactor receives the metered oil and caustic streams and operates in the emulsion-formation regime; a high-shear rotor-stator mixer with a tip speed of 10–20 m/s is employed only near the injection point to form a high-surface-area dispersion. The subsequent reactors provide residence time of 30–60 min at 85–95 °C with gentle anchor agitation, allowing the saponification to reach 97–99% conversion before drying. Published data for the exact kinetic parameters of mixed triglyceride feeds is limited because saponification rate depends on fatty acid chain length, degree of unsaturation, and the water-to-oil ratio; lauric oils such as coconut and palm kernel saponify measurably faster than stearic-rich tallow, and plant schedules are therefore adjusted by feedstock-specific residence time trials using production-scale vessels.

Controlled Caustic Dilution and Exotherm Management

Membrane-cell sodium hydroxide is commonly purchased as 50% w/w solution and diluted in a dedicated 316L stainless steel or lined carbon steel mixing tank to 30% w/w for kettle saponification. The dilution of 600 kg 50% w/w NaOH with 400 kg demineralised water yields 1000 kg 30% w/w NaOH solution, and the heat of dilution is removed through a half-pipe cooling jacket supplied with water at 15–20 °C. On production lines, the mixing sequence is fixed: the full water charge is added first, the agitator is started, and caustic soda is introduced through a dip pipe below the liquid surface at a controlled rate. Reverse addition of water into concentrated caustic can create local boiling at the liquid interface, splatter of corrosive liquid, and stress-corrosion risk in stainless steel. Temperature is monitored with a PT100 element in the recirculation line and interlocked with the caustic dosing pump; if the tank temperature exceeds 40 °C, the addition stops and cooling continues until the temperature falls below 35 °C. The final concentration is confirmed by density measurement at 20 °C with a value near 1.328 g/cm³ for 30% w/w NaOH, or by automatic titration against a known acid standard.

Concentrated 50% w/w NaOH freezes at approximately 12 °C; dilution to 30% w/w depresses the freezing point to approximately 1 °C, which still requires heat tracing in unheated outdoor transfer lines. Storage tanks for the concentrated solution are specified with external steam or electric tracing maintaining 20–30 °C and with insulation to prevent local crystallisation in nozzles and instrument legs. Pumps are magnetically driven or sealless centrifugal units in 316L stainless steel, with PTFE-lined hoses for transfer; elastomeric seals are selected from EPDM or FFKM because natural rubber and nitrile degrade rapidly in concentrated caustic.

Because Saponification Is a Biphasic Reaction, Mixing Dictates Yield

Batch saponification without adequate dispersion produces a heterogeneous mass in which the local caustic concentration is high at the oil-water interface but starved in the bulk oil phase. The result is an unacceptable product containing both free alkali pockets and unsaponified fat domains, even when the overall stoichiometry is correct. To prevent this, high-shear mixing is applied during the first 10–20 min after addition of the caustic solution; rotor-stator units with a power input of 0.5–1.5 kW/m³ are used in 10,000–50,000 L kettles to reduce oil droplet size below 100 µm before the gel stage begins. As conversion rises and the soap phase forms, the droplet deformation and coalescence behaviour changes because the soap acts as an emulsifier and the continuous phase viscosity increases; the mixer is then switched to an anchor agitator operating at 10–30 rpm to maintain turnover without shearing the soap structure. Field experience on production kettles shows that overmixing the neat soap phase introduces air, lowers bulk density by 5–15%, and generates microporosity in the finished bar; undermixing leaves 1–3% unreacted oil at the vessel wall and bottom dished head.

If Feedstock Saponification Value Drifts by ±10 mg KOH/g, What Are the Caustic Dosing Consequences?

At a production scale of 10,000 kg oil, an error of +10 mg KOH/g in saponification value changes the theoretical caustic soda charge by 71.3 kg NaOH (0.7129 kg NaOH per 1000 kg oil per 10 mg KOH/g). If the dosing system does not correct for the actual lot analysis, an oil with a higher saponification value than the target will leave excess unreacted oil after saponification, producing a soft, odour-unstable bar with elevated unsaponifiable matter. Conversely, an oil with a lower saponification value will receive an overdosed caustic charge, and the finished soap will contain free alkali above the skin-safety threshold. A control strategy based on ASTM D5558 titre and near-infrared process spectroscopy updates the caustic solution flow meter set point before the oil lot is charged. The flow meter is calibrated with the actual density and concentration of the caustic solution, and the control system uses a mass-flow ratio rather than a volumetric ratio to avoid density error from temperature drift. In plants without inline analytical control, a lot-to-lot saponification value range of ±5 mg KOH/g is absorbed by adjusting the lye discount within the 5–8% window, but broader variation requires reblending of oil stocks or a holding tank campaign change.

The electrolyte balance in the kettle is controlled by the chloride and carbonate content of the caustic soda as well as by added sodium chloride. In the spent lye, chloride levels of 5–10% w/w are typical after salting-out, but an excess of chloride in the starting caustic, above 0.5% w/w in diaphragm-grade material, can promote premature curd formation and reduce glycerine solubility in the aqueous phase. Sodium carbonate forms through reaction with atmospheric carbon dioxide in storage vents and appears in the neat soap as a fine precipitate that scatters light and reduces clarity in transparent soap. Carbonate in caustic soda is limited by blanketing storage tanks with nitrogen or by using a sealed tank vent with a drying tower. Sulphate and iron are controlled because iron catalyses oxidative rancidity of unsaturated fatty acid residues, and sulphate above 50 mg/kg can produce visible specks in high-glycerine translucent soap. The use of membrane-cell caustic soda is therefore preferred when the product is a superfatted toilet soap with a low free caustic specification; diaphragm-cell material is reserved for industrial laundry soap lines where the process includes a brine wash and the final free caustic limit is generally higher.

Free Alkali Limits in Finished Soap and the Analytical Methods That Enforce Them

Finished toilet soap is routinely controlled for free caustic alkali as NaOH according to ISO 684:1974, with a typical release limit of 0.05–0.10 mass % as NaOH. The limit is not simply a skin-irritation threshold; free caustic alkali above 0.10 mass % accelerates soap oxidation, causes bar sweating during storage at 30–35 °C and 75–85% RH, and destabilises fragrance components. Total alkali, determined by ISO 685:1975, includes carbonate and bicarbonate and is used for process control and regulatory labelling. The sampling procedure follows ASTM D460, with composite samples taken from the plodded billet after milling and before final extrusion. In process troubleshooting, a sample of neat soap is dissolved in neutralised ethanol and titrated potentiometrically with 0.1 mol/L hydrochloric acid; the titration curve shows a first inflection for free hydroxide, a second for carbonate, and a third for carboxylate, allowing separation of the alkaline species. A 1% aqueous solution of a well-formulated toilet soap has a pH range of 9.5–10.5, but pH alone is not a reliable release parameter because the buffering action of soap ions masks small changes in free caustic content.

After saponification, the soap mass is treated with a calculated amount of sodium chloride to reduce the water content and separate the glycerol-bearing spent lye. A typical spent lye from a tallow-coconut kettle contains 8–12% w/w glycerine, 6–10% w/w sodium chloride, and 0.05–0.3% w/w free NaOH; the exact values depend on the water-to-oil ratio, the salt addition, and the final kettle temperature. The free caustic in the spent lye is neutralised with hydrochloric acid before glycerine recovery to avoid caramelisation during evaporation under vacuum. The recovered glycerine is concentrated in multiple-effect evaporators at 60–70 °C to prevent decomposition, and the salt is separated in a settler or centrifuge. Caustic soda lost in the spent lye represents a direct yield loss; for a 10,000 kg oil batch, a spent lye containing 0.20% w/w free NaOH in 8,000 kg aqueous phase contains 16 kg NaOH, which is 1.15% of the stoichiometric charge for a palm oil formulation at 139.0 kg per 1000 kg oil. Recovery of this residual caustic is frequently not cost-effective in small kettles, but continuous operations may reuse a portion of the filtered waste lye as saponification liquor.

Occupational exposure control for sodium hydroxide in a soap saponification plant is governed by the OSHA permissible exposure limit of 2 mg/m³ as an 8-hour time-weighted average under 29 CFR 1910.1000 Table Z-1 and by the NIOSH recommended ceiling limit of 2 mg/m³. Engineering controls include local exhaust ventilation at open kettles, enclosed transfer lines, and splash guards at sampling points. Operators wear chemical protective gloves tested according to EN 374-1:2016, safety goggles meeting EN 166, and a full-face shield when opening vessels. Emergency showers and eyewash stations are installed within 10 s travel distance as specified by ANSI Z358.1. Because sodium hydroxide reacts violently with aluminium, zinc, galvanised steel, magnesium, and acids, transfer lines and fittings must be constructed of 316L stainless steel, PTFE, polypropylene, or high-density polyethylene. Stress-corrosion cracking of stainless steel is a known failure mode when concentrated caustic is held above 60 °C in welded zones with residual tensile stress; post-weld stress relief and lower-temperature storage reduce this risk.

The safety data sheet for the caustic soda grade used in saponification must identify the specific concentration limits for skin corrosion and eye irritation under CLP, state the 2 mg/m³ occupational exposure ceiling, and provide REACH registration data under EC 1907/2006. A process safety review for a soap plant using 50% w/w sodium hydroxide includes line-break procedures, isolation of dilution tanks during maintenance, and verification that the emergency shower network complies with ANSI Z358.1. These control layers define the operational boundary within which the stoichiometric and mixing requirements described above can be executed without producing a corrosive release or an off-specification soap batch.