Kraft White Liquor Charge and Eop Extraction Caustic Demand

In continuous and displacement batch kraft fibre lines, the white liquor charge is normally expressed as effective alkali (EA) on oven-dry wood, and the residual effective alkali leaving the blowline or extraction screen becomes a direct boundary condition for downstream Eop caustic demand. White liquor EA is determined according to TAPPI T 624 cm-11 or an equivalent mill titration, with the effective alkali calculation converted from NaOH and Na₂S concentrations after sulfidity is fixed on a Na₂O basis. A softwood line operating at 16–20% EA on wood and 30–45% sulfidity typically leaves 8–14 g/L residual effective alkali as NaOH at the blowline at H-factors of 1200–1800. Hardwood lines using 14–18% EA and H-factors of 400–800 may leave 6–12 g/L residual effective alkali. These residual levels are not simply excess chemical; they maintain the spent cooking liquor pH above the precipitation threshold of dissolved lignin and hydroxy acids. When post-washing carryover drops below 6 g/L as NaOH on a softwood grade, elevated kappa variability is frequently observed under ISO 302:2015, and the precipitated organic acid load subsequently raises Eop caustic requirement because the alkali-soluble material must be re-ionised in extraction. The mass transfer boundary is also controlled by the liquor-to-wood ratio: continuous digesters typically run at 3.0–4.5:1, while displacement batch systems may operate at 3.5–5.0:1. Lower liquor-to-wood ratios concentrate alkali and accelerate delignification, but they reduce hydraulic dilution and can create local dead zones in the extraction screens; the resulting uneven residual EA profile is carried to the oxygen stage and then into D0/Eop as a fluctuating acid load.

Where the fibre line includes a two-stage oxygen delignification system operating at 80–100 °C and 0.4–0.8 MPa oxygen partial pressure, the residual alkali from brownstock washing is often used as part of the oxygen-stage chemical charge. Oxygen delignification reduces the kappa entering the bleach plant from a typical softwood range of 28–32 to 12–18 under ISO 302:2015, thereby removing a substantial portion of the aromatic structures that would otherwise consume Eop caustic after D0. However, oxygen-stage caustic addition must be balanced against carryover: an oxygen-stage NaOH charge of 1.5–3.0% on oven-dry pulp and 0.1–0.5% MgSO₄ are common operating windows, but if the inlet pulp consistency to the medium-consistency pump is below 10% or the twin-roll press discharge is above 32%, local pH inhomogeneities generate partially oxidised fragments that are recalcitrant in extraction. The oxygen reactor should be equipped with gas-separation cyclones and top-scraper units to minimise channelling; when channelling is present, dissolved oxygen is unevenly transferred to the pulp fraction, and the measured kappa reduction under ISO 302:2015 may be within target while the total acidic group content and Eop caustic demand remain higher than expected. In such cases, published data for this specific configuration is limited, and laboratory conductometric titration of total acidic groups is required before adjusting the Eop NaOH setpoint.

When Residual Effective Alkali Falls Below the Oxygen-Stage Buffering Threshold

The process conflict is that reducing white liquor EA to save recovery loading can push the brownstock residual alkali below the oxygen delignification buffering threshold, commonly observed in mill audits at 6–8 g/L as NaOH. At residual EA below 6 g/L, the oxygen-stage pH decays rapidly during the first 15–20 min of retention, and the concentration of undissociated organic acids increases. These acids are carried into D0 and then into Eop as weakly ionised material, requiring a disproportionate increase in Eop NaOH to reach a final extraction pH of 10.5–11.5. In a D0(Eop)D1 sequence, the D0 stage typically operates at a kappa factor of 0.15–0.25 per incoming kappa unit, and the D0 washer discharge consistency is maintained at 10–14% before the Eop MC pump. Acid carryover from D0 at pH 2.0–3.0 is a direct caustic demand, but its magnitude is smaller than the acid generated by oxidative ring opening of residual lignin and hexenuronic acid when oxygen and peroxide are applied. Mill control systems therefore use a feedforward term from the online kappa analyser calibrated against ISO 302:2015, a feedback term from the Eop discharge pH probe, and an alkalinity measurement of the press filtrate. The NaOH charge to Eop for softwood pulps after oxygen delignification generally falls in the range of 0.8–2.5% on oven-dry pulp, while hardwood pulps with elevated hexenuronic acid may require 1.0–2.0% even when the measured kappa is lower. The presence of dissolved carbonate in filtrate from the recausticizing area can add 0.05–0.20% NaOH-equivalent to the Eop charge if the filtrate is used as wash water and is not purged; this dead load is quantified by TAPPI T 624 cm-11 and should be subtracted in the caustic balance.

The risk boundary is narrow: if Eop pH is allowed to rise above 11.5, alkaline peroxide decomposition accelerates, and the peroxide charge is consumed by hydroperoxide anion disproportionation rather than by chromophore oxidation. This overshoot is especially acute when Eop caustic is added at a concentration above 5% NaOH at the medium-consistency mixer suction, because localised high-pH zones cause oxidative degradation of cellulose and reduce pulp viscosity measured by ISO 5351:2010. Conversely, if Eop pH falls below 10.5, the extraction of chlorinated and oxidised lignin fragments is incomplete, so the final D1 brightness may be depressed by 1–3 ISO units and the residual peroxide in the stage is not fully utilised. The caustic control loop must therefore be constrained with an anti-reset windup on the pH controller and with a minimum and maximum NaOH charge clamp of 0.5% and 3.5% on oven-dry pulp unless a mill-specific titration justifies a wider window.

Why Does Hexenuronic Acid Inflate Eop Caustic Demand Independent of Kappa?

The decoupling of kappa and Eop caustic requirement is most pronounced on hardwood lines running low H-factors and low sulfidity. Hexenuronic acid is counted as kappa under ISO 302:2015 but is not an aromatic lignin structure, and it contributes carboxylic acid functionality after oxidative ring opening in D0 and Eop. Hardwood kraft pulps produced from birch, eucalyptus, and acacia can contain hexenuronic acid levels sufficient to inflate the measured kappa by 2–8 units. Two pulps with the same Eop feed kappa may differ in caustic demand by 0.3–0.8% NaOH on oven-dry pulp if their hexenuronic acid contents differ by 20–50 mmol/kg. The practical consequence is that a kappa-only feedforward strategy will bias the Eop caustic setpoint low on high-HexA hardwood pulps and high on low-HexA softwood pulps. Where mill-specific selective hydrolysis and UV data are not available, published data for this specific configuration is limited, and the Eop NaOH setpoint should be verified by a staged titration curve of the actual washed D0 pulp rather than by relying solely on ISO 302:2015 kappa.

Transition Metal Management and Chelant–Alkali Interactions

Transition metal carryover from brownstock and oxygen stages affects Eop caustic demand through peroxide decomposition and the formation of additional acid groups from radical side reactions. The Eop stage is commonly operated at 70–90 °C with 0.05–0.15% MgSO₄ on oven-dry pulp to precipitate manganese and iron; when the total manganese and iron concentration in the washed D0 pulp is above 20–50 mg/kg, a chelant such as DTPA or EDTA is typically applied at 0.1–0.3% on oven-dry pulp. This addition must be checked against the Eop pH target because an excessive chelant charge can mobilise metals and increase the peroxide decomposition rate, particularly when the pH is above 11.5. Magnesium overdose above 0.20% can blind washer screens and contribute to carbonate scaling, while underdosing below 0.05% leaves transition metals free to catalyse hydrogen peroxide disproportionation. The resulting oxygen gas formation in the Eop tower may create pressure fluctuations and reduce the actual peroxide residual; in such cases, the caustic controller may incorrectly call for more NaOH because the pH drops from generated acidic oxidation products, creating an alkali–peroxide–metal loop that consumes additional chemical. Published data for this specific configuration is limited, and the metal management strategy should be adjusted only after acid digestion and inductively coupled plasma optical emission spectrometry of the actual pulp stream.

At the medium-consistency pump inlet, the Eop caustic setpoint is interpreted with a pH sensor loop that includes a retractable holder, automatic washing, and a dead-time compensator set to 60–120 s because the Eop tower discharge is not homogeneous. The NaOH is normally injected as a 2–5% solution into the suction side of the MC pump or the dilution header to avoid localised alkali shock. If the injection point is moved to the tower top, the alkali distribution is limited by channeling and the extracted pulp may show brightness variation of 1–2 ISO units under ISO 2470-1:2016 after D1. The control system should also integrate the D0 washer discharge consistency signal from a rotary consistency transmitter so that the caustic charge per dry tonne of pulp remains constant when the consistency changes by ±1%. In a two-stage Eop tower operated at 0.2–0.6 MPa oxygen partial pressure and 45–90 min retention, the gas-liquid mass transfer rate is sensitive to the pulp suspension consistency; maintaining the inlet consistency at 10–14% ensures that the peroxide and oxygen are dispersed before the alkali-sensitive pH zone develops. When the D0 washer is bypassed or the dilution factor is disturbed, the Eop caustic demand fluctuates by more than 0.3% NaOH on pulp in less than 30 min, which exceeds the response capability of a conventional feedback loop; therefore a feedforward compensation based on washer discharge consistency and kappa under ISO 302:2015 is required.

Interpreting Carryover Alkalinity from Oxygen Delignification Washers

Pressure diffuser and twin-roll press washing after oxygen delignification separate the dissolved organic acids from the pulp, but the measured filtrate alkalinity does not correspond directly to the NaOH available for Eop extraction. The filtrate contains sodium carbonate, sodium bicarbonate, and sodium salts of organic acids; the buffering capacity of these species means that a filtrate pH of 10.0–11.0 may overstate the hydroxide carryover entering D0. Wash liquor displacement ratios of 2.0–3.5 m³/t pulp and a dilution factor of 1.0–2.0 m³/t are typical in oxygen-stage washing, but if the dilution factor is above 2.5 m³/t, the recovery load increases without improving Eop caustic efficiency because the additional wash liquor becomes contaminated with recovery-cycle carbonate. The correct credit for carryover alkalinity is obtained by titrating the press filtrate with acid to pH 8.3 and 4.5, using a two-endpoint titration consistent with TAPPI T 624 cm-11 for carbonate and bicarbonate distribution. Mills that reuse causticizing area filtrate as wash water without carbonate purge add 0.05–0.20% NaOH-equivalent to the Eop charge as dead load; this amount is not available for lignin extraction because it is already neutralised under the acidic D0 condition.

Sodium carbonate dead load in white liquor and washer filtrate is measured as the difference between effective alkali and total titratable alkali under TAPPI T 624 cm-11; a low causticizing efficiency can leave 15–25 g/L Na₂CO₃ in white liquor, which does not delignify in cooking but contributes to the buffering of the oxygen-stage and Eop filtrate. When this carbonate-rich filtrate is used as Eop wash water after the D0 stage, it is acidified and releases CO₂, providing an apparent alkalinity that is not hydroxide and cannot saponify chlorinated lignin fragments. The resulting caustic demand is therefore underestimated if the DCS credits total alkalinity rather than effective hydroxide; the correction is typically 0.1–0.3% NaOH on oven-dry pulp for a closed-cycle mill with high dead load. Published data for this specific closed-cycle configuration is limited, and the correction should be derived from the measured carbonate concentration in the exact filtrate stream feeding the Eop dilution header.

The analytical methods used to close the white liquor and Eop caustic balance are listed in the following compliance table.

The operational boundary for Eop caustic control is defined by three pH-related thresholds: the extraction floor at 10.5, the peroxide stability ceiling at 11.5, and the localised alkali shock threshold associated with NaOH concentration above 5% at the injection point. Carbon steel piping at pH 11.0 and 80 °C is outside the acceptable corrosion envelope unless a protective oxide film is maintained; duplex stainless steel or lined pipe is specified when chloride in the bleach plant filtrate exceeds 200 mg/L. Amine-based reducing agents and sulphite-based carryover should not be combined with Eop peroxide because they can consume the oxidant before chromophore oxidation occurs; published data for this specific additive combination is limited. The caustic demand cannot be reduced by simply increasing the Eop temperature, since above 90 °C the peroxide half-life shortens and the pH response loop becomes unstable, while below 70 °C the extraction of oxidised lignin is too slow to meet the stage retention target.