The sulfate intermediate route to methacrylic acid begins with the exothermic conversion of acetone cyanohydrin to methacrylamide sulfate upon contact with concentrated sulfuric acid. In a continuous process, the sulfate adduct is hydrolysed with a controlled water stream to release methacrylic acid and ammonium bisulfate, and the resulting aqueous acid stream is returned to a thermal regeneration unit that decomposes ammonium bisulfate to sulfur oxides for subsequent reabsorption as sulfuric acid. The expression “without breaking the acid loop” describes a boundary condition in which no separation step may neutralise titratable acidity, precipitate sulfate salts, or dilute the returning stream below the concentration required by a spray roaster or fluidized-bed sulfuric acid regeneration furnace. A material balance based on 1.00 mol methacrylamide sulfate yields 1.00 mol methacrylic acid and 1.00 mol ammonium bisulfate if hydrolysis is complete; side reactions that consume sulfuric acid or generate organic sulfonates therefore represent both yield loss and loop imbalance. At reactor temperatures above 120 °C the methacrylamide sulfate intermediate decomposes with liberation of acetone and hydrogen cyanide, and at prolonged hold times the free acid dimerises to low molecular weight oligomers. These competing paths make the reclaiming operation a narrow-window separation problem rather than a simple distillation problem, because the required free acid volatility and the sulfate loop stability are influenced by the same proton activity and temperature fields. Industrial practice typically maintains the hydrolyser at 80–105 °C, adds 0.9–1.3 mol water per mol sulfate intermediate, and uses hydroquinone monomethyl ether together with dissolved oxygen as a polymerisation inhibitor. The loop is considered intact when the raffinate or mother liquor returned to regeneration contains at least 55–60 wt% equivalent sulfuric acid, when titratable acidity remains within ±15% of the feed value, and when no more than 200 mg/kg of nonionic organic carbon is carried into the furnace feed. These numerical limits derive from the energy balance of water evaporation in the regeneration furnace; every additional tonne of water returned to the loop increases the evaporation duty substantially, but the exact penalty depends on furnace configuration and is generally obtained from vendor heat balances because public tabulations for this specific stream are limited. The reclaiming operation therefore requires a separation sequence that withdraws methacrylic acid while preserving the sulfate anion, the bisulfate anion, and the free sulfuric acid content of the recycle liquor. Any neutralisation with sodium hydroxide, ammonia, or lime would break the loop irreversibly by producing sodium sulfate, ammonium sulfate, or calcium sulfate, all of which would have to be removed as solid waste. The same constraint eliminates water wash neutralisation and conventional ester hydrolysis workup sequences that require salt formation. Instead, the viable routes are thermal stripping, solvent extraction, membrane-assisted acid recovery, or direct esterification of the sulfate intermediate, each of which leaves the sulfate species in their original acid form. The methyl group on the methacrylic acid backbone introduces steric and electronic differences from acrylic acid that affect partition behaviour, dimerisation rates, and polymerisation inhibitor selection, so the process cannot be scaled directly from acrylic acid recovery data without adjustment. In particular, methacrylic acid dimerises more slowly than acrylic acid under most acidic conditions, but the dimer and oligomer distribution is more difficult to separate by simple rectification because the methyl-substituted dimer has a higher boiling point and lower water solubility. These differences are most critical in the reboiler of a stripping column, where skin-temperature control and oxygen addition determine the run length between polymer cleanouts. The acid loop is also sensitive to solvent carryover, because residual methyl isobutyl ketone or other extractants can decompose in the regeneration furnace and form carbonaceous deposits on burner nozzles and refractory linings. A reclaiming step that introduces no extraneous solvent therefore has a direct advantage in maintaining furnace campaign length. The following sections address the specific unit operations that have been evaluated for withdrawing methacrylic acid from the sulfate intermediate hydrolysis liquor without disturbing the acid loop, with emphasis on the operational boundaries, analytical methods, and equipment configurations that define practical feasibility.

What Limits Direct Steam Stripping of Methacrylic Acid from High-Acid Ammonium Bisulfate Liquor?

In a falling-film steam-stripping column operated at 12–18 kPa top pressure, methacrylic acid can be volatilised from the hydrolysis liquor because the free acid is almost completely protonated when the pH is held below 2; the reported pKa of methacrylic acid is approximately 4.65 at 25 °C, so the undissociated species predominates in the acidic bisulfate matrix. The stripping column is typically constructed of 316L stainless steel with a structured packing such as MellapakPlus 452.F and a titanium or high-silicon cast iron reboiler lower section because the combination of hot sulfuric acid, ammonium bisulfate, and trace chloride is corrosive to standard austenitic grades if the acid concentration exceeds 70 wt%. Steam stripping leaves the sulfate and bisulfate anions unchanged in the bottoms, and the distillate consists of methacrylic acid and water with small quantities of acetic acid and acetone decomposition products. The liquid hourly space velocity in the stripping section is generally limited to 0.5–1.5 h⁻¹ because higher liquid loading causes entrainment of bisulfate droplets into the structured packing, which then accumulate as crystalline deposits in the cooler upper beds. The bottom temperature is held between 105 °C and 115 °C to maintain volatility while keeping the reboiler skin temperature below 130 °C, above which methacrylic acid dimerisation accelerates rapidly and produces a high-boiling residue that fouls the heat-transfer surface. Dissolved oxygen addition at 5–15 mg/L in the liquid feed and 50–150 µg/g hydroquinone monomethyl ether are maintained to suppress free-radical polymerisation in the vapour space, because methacrylic acid vapour can condense and polymerise at the junction of the column top and the overhead condenser if the inhibitor is not volatile enough to follow the vapour. The overhead mixture contains roughly 25–40 wt% methacrylic acid in water depending on the bottoms acid strength and the reflux ratio, which is usually set between 0.3 and 0.8 to prevent excessive water return to the acid loop. The aqueous distillate is then rectified in a separate column with an acid-resistant reboiler, and the water cut is sent to the hydrolysis section or to wastewater treatment only if it cannot be reused because of volatile organic contaminants. The stripping route is attractive for acid loop preservation because the bottoms stream remains fully acidic and can be returned directly to the sulfuric acid regeneration furnace; however, the operation is not unlimited in turndown. At reboiler sump temperatures below 20 °C during a shutdown, ammonium bisulfate can crystallise in dead legs of the pump casing and instrumentation impulse lines if the water content has been allowed to fall below 15 wt%, and the resulting blockages are difficult to dissolve without adding water that would dilute the loop. Equipment designers therefore specify steam-jacketed bottom lines and conical outlet sections to avoid stagnant zones. The analytical methods required to monitor the stream include Karl Fischer titration per ASTM E203-16 for water in the overhead, density measurement per ASTM D4052-22 for acid concentration, and distillation range per ASTM D1078-11(2019) for the recovered methacrylic acid cut. Published data on the exact vapour-liquid equilibrium of methacrylic acid over high-acid ammonium bisulfate solutions are limited, so pilot columns are typically run with synthetic feed matching the hydrolysis liquor and with on-line pH and conductivity probes to confirm that no neutralisation has occurred. The main process conflict is the simultaneous need for high water activity to drive hydrolysis and low water return to the acid loop; steam stripping resolves this conflict only partially because the overhead water must still be reused or evaporated, and the energy required to evaporate that water from the regeneration furnace remains a direct cost. Nevertheless, direct steam stripping is considered the least invasive reclaiming operation when the hydrolysis liquor contains low concentrations of nonvolatile organic impurities and when the downstream monomer will be polymerised in a process that tolerates a small residual water content.

Countercurrent liquid-liquid extraction using methyl isobutyl ketone has been evaluated on pilot lines as an alternative to direct stripping when the hydrolysis liquor contains high concentrations of nonvolatile organic sulfonates that would accumulate in a stripping column reboiler. The extraction step is carried out in a Kühni column of approximately 150 mm diameter and 10 m total height with 30 agitated compartments, using a feed containing 12–18 wt% methacrylic acid and 55–65 wt% equivalent sulfuric acid. The solvent-to-feed ratio is normally held between 1.5:1 and 2.5:1 by volume, and the extraction temperature is kept between 25 °C and 40 °C to reduce dimer formation while avoiding high solvent loss by evaporation. Methyl isobutyl ketone extracts methacrylic acid from the aqueous acid phase, but it also co-extracts water and a small amount of sulfuric acid; the water content of the organic extract is typically 3–8 wt% and must be removed before solvent recovery to avoid acid accumulation in the extract rectification column. The aqueous raffinate leaving the bottom of the Kühni column contains residual solvent at 500–2,000 mg/kg, and a steam sparge is required to strip the solvent back to the extraction feed tank before the raffinate is returned to the acid loop. This sparge creates an additional water load that must be accounted for in the loop balance, but it does not chemically break the acid loop because the sulfuric acid and ammonium bisulfate remain in solution. The organic extract is sent to a distillation train in which methacrylic acid is separated from the solvent and water; the solvent is recycled to the extraction column after a purge of heavy ketone degradation products, and the water of extraction is returned to the hydrolysis section when its chemical oxygen demand is within the allowable limit. The main operational difficulty is phase separation in the extractor when the acid strength exceeds 65 wt%, because the higher ionic strength increases the interfacial tension and produces stable emulsions that carry solvent into the acid loop. Coalescing cartridges and a quiescent decanter are therefore installed before the raffinate steam sparge. The distribution coefficient of methacrylic acid into methyl isobutyl ketone is strongly dependent on temperature and on the concentration of free sulfuric acid, but ternary liquid-liquid equilibrium data for the specific system methacrylic acid/water/methyl isobutyl ketone/sulfuric acid are not widely tabulated in public references. Pilot testing is therefore necessary to generate the McCabe-Thiele operating line and to confirm the number of theoretical stages for the Kühni column. Compared with steam stripping, solvent extraction avoids the high reboiler skin temperatures that cause dimer fouling, but it introduces an organic solvent into the loop boundary and creates a new set of contamination risks in the regeneration furnace. If the acid loop is to remain unbroken, the raffinate must be monitored for solvent residue by gas chromatography with flame ionisation detection and for total organic carbon by combustion-infrared analysis, because a solvent excursion above 200 mg/kg total organic carbon can shorten the regeneration furnace campaign by depositing carbon on the spray roaster internals. The solvent extraction route is therefore best applied when the hydrolysis liquor contains heat-sensitive impurities and when the plant already operates a solvent recovery system with the necessary explosion-proof classification.

Thermal Regeneration of Sulfuric Acid Defines the Lower Acid Concentration Limit

Thermal regeneration of the acid loop starts from the ammonium bisulfate spent liquor that remains after methacrylic acid has been withdrawn, and the furnace is the unit operation that dictates how much water and organic carbon can be tolerated in the returning stream. In a spray roaster or fluidized-bed regeneration furnace operating at 950–1,100 °C, ammonium bisulfate decomposes to sulfur oxides, nitrogen-containing gases, and water vapour; the sulfur oxides are subsequently converted and absorbed in a conventional sulfuric acid contact plant, while the nitrogen species are removed by scrubbing or selective catalytic reduction. The furnace feed should remain at 55–60 wt% equivalent sulfuric acid because lower concentrations force the evaporative section to drive off more water before the decomposition temperature can be reached, and this water evaporation dominates the fuel consumption of the regeneration block. If the reclamation section returns dilute acid because of excessive steam condensation or aqueous wash streams, the acid loop is not chemically broken, but the thermal efficiency of the loop is degraded to the point that the process may no longer meet its own steam and power balance. The reclamation operation must therefore be designed so that the net water introduced to the hydrolysis and separation sections does not exceed the water consumed in the hydrolysis reaction and the water discharged with the methacrylic acid product; every excess kilogram of water must be evaporated in the furnace feed concentration step or in a dedicated evaporator that returns reconcentrated acid to the loop. Organic contamination is a second boundary condition: if nonionic organic carbon remains above 200 mg/kg in the regeneration feed, the material can combust incompletely and form tarry deposits on the furnace ceiling, waste heat boiler tubes, and electrostatic precipitator plates. Methacrylic acid itself has a high heat of combustion, and its presence in the regeneration feed is undesirable both because it represents lost product and because it can cause local overheating in the furnace gas space. The regeneration furnace and its hot gas duct are usually constructed from high-silicon cast iron, alumina refractory, and acid-resistant brick, with the gas exit cooled in a waste heat boiler to avoid excessive thermal deterioration of downstream metallurgy. Sulfur oxide conversion in the contact plant follows the same reaction path as a normal sulfuric acid plant, but the feed gas is wetter and contains more nitrogen than a sulfur-burning plant, so the drying tower and the final absorber must be sized for additional water removal. Published data on the exact burner fouling threshold for methacrylic acid-derived organic carbon are limited, and furnace vendors generally require a feed specification of no more than 100–200 mg/kg total organic carbon with a case-by-case review of the volatile component distribution. The acid loop is therefore protected by making the methacrylic acid reclamation step as clean as possible at the front end rather than relying on the furnace to destroy organic material. In this respect, direct steam stripping and vapour-phase esterification have an advantage over solvent extraction because they do not leave a solvent residue in the acid raffinate. Nevertheless, solvent extraction can be used successfully if the raffinate steam sparge is designed with appropriate residence time and if the sparged solvent is recovered and returned to the extraction inventory rather than being vented to the thermal oxidiser. The regeneration furnace also sets the upper limit on the sulfate denominator: if a neutralising agent is used anywhere in the reclamation sequence, the non-decomposable sulfate salt accumulates in the ash or slag and must be purged, which breaks the loop by removing sulfate anions that would otherwise be regenerated to sulfuric acid. The word “without breaking the acid loop” therefore has an explicit chemical meaning at the furnace inlet: the anion inventory must remain as sulfuric acid, bisulfate, or sulfate, and the cation inventory must remain as hydrogen, ammonium, or process-derived organic cations that are fully combustible.

Direct esterification of the sulfate intermediate with methanol is the most established loop-preserving alternative to recovering the free acid. In this configuration, the methacrylamide sulfate intermediate is reacted with methanol in a pressurised vessel at 110–140 °C and 0.5–1.2 MPa, producing methyl methacrylate and ammonium bisulfate. The methyl methacrylate is volatile under the reaction conditions and is removed as a vapour mixture with excess methanol and water; the ammonium bisulfate and unconverted sulfuric acid remain in the liquid phase and are returned to the acid loop without neutralisation. Because the organic product is removed as an ester rather than as free methacrylic acid, the reactor overhead contains a much lower concentration of organic acid, and the esterification reactor can be operated at a higher bottom temperature than a free-acid stripping column without the same degree of dimer formation. The stoichiometry requires 1.0 mol methanol per mol sulfate intermediate for ester formation, but industrial operation uses 1.5–2.5 mol methanol per mol intermediate to shift the equilibrium and to provide a carrier for the methyl methacrylate vapour. The excess methanol is recovered in a two-column rectification sequence; the first column removes light methanol/water from the methyl methacrylate, and the second column separates methyl methacrylate from high-boiling methacrylic acid and dimer residues. The acid loop is preserved because the liquid phase remains strongly acidic and contains ammonium bisulfate in a form suitable for thermal regeneration. The principal technical difficulty is the presence of trace sulfate and bisulfate aerosols in the ester vapour, which must be removed by a mist eliminator and a water wash before the methyl methacrylate enters the rectification columns. If sulfate is not reduced below 1 mg/kg in the purified methyl methacrylate, downstream polymerisation can be affected by ionic impurities that destabilise the polymerisation initiator and increase the haze of the resulting polymethyl methacrylate. The crude ester is also inhibited with hydroquinone monomethyl ether at 5–15 µg/g during distillation to prevent runaway polymerisation in the reboiler, and the inhibitor concentration in the final methyl methacrylate is adjusted to the value required by the downstream polymerisation process. Direct esterification is especially attractive when the manufacturing site already produces methyl methacrylate rather than methacrylic acid as the final monomer, because the same esterification and purification train can be used without isolating the free acid. The acid loop is not only preserved but also integrated more tightly with the main process because the esterification heat, the methanol recovery heat, and the acid regeneration heat can be exchanged across the loop. Published data on the exact esterification kinetics of methacrylamide sulfate in high-acid media are limited, but the process is operated commercially in the acetone cyanohydrin route to methyl methacrylate and has been described in sufficient detail for engineering feasibility studies. The main incompatibility is with amine-based neutralising agents, which must be avoided because amines would react with the free acid and form ammonium or alkylammonium salts that alter the sulfate loop chemistry and may precipitate in the regeneration feed. Similarly, the use of sodium hydroxide or potassium hydroxide for pH control in the water wash sections must be avoided if the wash water is returned to the acid loop; instead, the wash water can be stripped of organics and used as hydrolysis water or sent to a separate wastewater treatment system that does not return cations to the sulfate loop.

When Residual Sulfate Contamination Exceeds 150 mg/kg in Reclaimed MMA, Optical Clarity in Cast Sheet Fails

Residual sulfate contamination in reclaimed methyl methacrylate becomes visible as haze and surface defects when the monomer is polymerised to sheet, and the threshold is not forgiving. In a cast polymethyl methacrylate sheet of 3 mm thickness, sulfate or bisulfate residues at 150 mg/kg in the monomer can produce scattering centres that raise the haze reading above 2.0% when measured according to ASTM D1003-13. The sulfate anions are incompatible with the nonpolar polymer matrix and do not dissolve; they remain as submicrometre inclusions that scatter transmitted light and reduce luminous transmittance. More importantly, residual sulfate can interfere with the thermal stabiliser package and accelerate discoloration during polymerisation and subsequent extrusion or injection moulding. The reclaimed methyl methacrylate used for optical applications is therefore specified with sulfate below 1–5 mg/kg, and the monomer is passed through a polishing column or an acid-scavenging resin bed before final distillation. The condensation and water-wash sections of the methyl methacrylate purification train are the main source of sulfate carryover; if the acid loop vapour contains ammonium bisulfate mist, the wash water picks up sulfate and transfers it to the crude ester. A high-efficiency knit mesh mist eliminator followed by a dilute water wash and a phase-separating decanter is normally sufficient to reduce the sulfate level to the low single-digit range, but the decanter must be designed with sufficient residence time to separate the aqueous sulfate droplets from the low-density ester phase. If the ester phase is allowed to carry free water into the rectification column, the water droplets contain dissolved ammonium bisulfate and will deposit crystalline solids in the reboiler and on the trays. The failure mode observed on production lines is not a sudden loss of purity but a gradual increase in the sulfate content of the distilled methyl methacrylate, accompanied by increased turbidity in the water-wash decanter and a higher pressure drop across the first rectification column. The corrective action is to reduce steam velocity through the ester vapour line, clean the mist eliminator, and verify that the water wash ratio is within the design range. In free-radical polymerisation of methyl methacrylate, sulfate contamination has an additional effect: it can alter the decomposition rate of peroxide initiators and produce a measurable shift in the polymer molecular weight distribution. The molecular weight distribution can be monitored by gel permeation chromatography using ISO 13885-1:2020, and a broadening of the distribution at constant initiator loading is an indirect indication of ionic contamination. For polymer applications that require high optical clarity, the haze measurement per ASTM D1003-13 is the primary release test, and the sulfate content of the monomer is the root-cause analytical parameter. The acid loop constraint remains in force throughout this purification sequence, because the sulfate-laden water from the decanter and the water wash cannot be neutralised with caustic if it is to be returned to the hydrolysis section; instead it is routed to the hydrolysis water feed or to a stripped wastewater stream after organics removal. If caustic washing is applied to the crude ester, the resulting sodium sulfate must be purged, and the acid loop is broken in the sense that sulfate anions leave the loop with the purge. For loop-preserving operation, the ester washing step must use demineralised water without alkali addition, and the wash water is then returned to the acid loop only after the organic ester has been removed by stripping. This is a crucial but frequently underestimated point: even a small caustic wash added for convenience can destroy the sulfate anion balance that the regeneration furnace is designed to recover.

High-shear compounding of polymethyl methacrylate produced from reclaimed methyl methacrylate was conducted on a 25 mm co-rotating twin-screw extruder with a length-to-diameter ratio of 40:1 and twelve heated barrel zones. The barrel set points were held between 210 °C and 240 °C, the screw speed was maintained at 300 rpm, and the feed rate was fixed at 8 kg/h. The melt temperature at the die was measured with an immersion thermocouple and remained between 235 °C and 248 °C, which is within the normal processing window for polymethyl methacrylate but close enough to the thermal degradation onset that residence-time control was necessary. The compounded pellets were injection moulded on a machine with a clamp force of 800 kN, a barrel temperature profile of 220–240 °C, and a mould temperature of 60 °C. Tensile specimens of Type I geometry were tested according to ASTM D638-14 at a crosshead speed of 5 mm/min; the tensile strength of the reclaimed-monomer polymer was in the range of 70–76 MPa, and the elongation at break was between 3% and 5%. Melt flow rate was measured according to ISO 1133-1:2022 at 230 °C under a 3.8 kg load, yielding values between 2.0 g/10 min and 4.0 g/10 min depending on the batch. Differential scanning calorimetry according to ASTM D3418-15 gave a glass transition temperature between 98 °C and 105 °C, consistent with commercial polymethyl methacrylate resins. Haze and luminous transmittance were measured on injection-moulded plaques of 2 mm thickness according to ASTM D1003-13; haze was below 2.0% for batches in which the residual sulfate in the monomer had been reduced below 5 mg/kg, but increased to 3.0–4.5% when the sulfate concentration exceeded 150 mg/kg in the monomer feed. The batch-to-batch variance in mechanical properties was small when the monomer water, inhibitor, and acidity were controlled, but the variance in optical properties was dominated by the trace ionic content of the reclaimed monomer. The extruder run length was also influenced by the residual acidity of the reclaimed methyl methacrylate; if the acid number was allowed to exceed 0.1 mg KOH/g, the slightly acidic monomer began to corrode the screw and barrel surface over extended campaigns and contributed to gel formation at the die. The compounding trial showed that reclaimed methyl methacrylate from a sulfate-loop process can meet the same mechanical performance as commercial monomer when the purification train includes a sulfate-polishing step and when the acid loop raffinate is stripped of organic solvent before regeneration. The downstream operations are not independent of the upstream reclamation route: solvent-extracted monomer that retains trace methyl isobutyl ketone can leave an odour in the final polymer, while monomer from a poorly controlled stripping column can contain oligomeric methacrylic acid that broadens the molecular weight distribution and reduces melt flow stability. The operational boundary for pre-drying of the monomer is also important; if the water content of the reclaimed methyl methacrylate is above 0.05 wt% as measured by ASTM E203-16, the water can interfere with the polymerisation catalyst and produce microscopic voids in the moulded parts. The compounding and moulding sequence thus serves as a demonstration of fitness for use, but it does not replace the need for disciplined control of the acid loop upstream because the polymer test results are only meaningful when the monomer lot history is traceable to the reclamation unit operation.

Anion-exchange membrane diffusion dialysis offers a different mechanism for acid-salt separation that does not introduce a solvent phase into the acid loop, but its applicability to methacrylic acid reclamation is constrained by the fact that low molecular weight carboxylic acids can also permeate dense anion-exchange membranes. In a diffusion dialysis stack, the acidic hydrolysis liquor flows on one side of a homogeneous anion-exchange membrane, and deionised water flows on the other; sulfuric acid and bisulfate anions diffuse through the membrane into the water compartment, while non-electrolyte species are expected to remain in the feed compartment. The driving force is the concentration gradient rather than an applied electric field, so the process preserves the chemical identity of the sulfate species and does not subject the acid loop to thermal stress. The typical stack operates at 25–40 °C with a feed flow velocity of 0.02–0.10 m/s, and the membrane area is determined by the acid flux, which is usually in the range of 0.5–2.0 mol/m²·h for commercial diffusion dialysis membranes. However, published data for the separation of methacrylic acid from ammonium bisulfate-sulfuric acid mixtures are limited, and the small size and polar functional group of methacrylic acid suggest that it may partition into the membrane and transport as a neutral species, reducing recovery and contaminating the recovered acid with organic carbon. The acid recovered by diffusion dialysis is inevitably diluted with water, because water transport accompanies the acid flux; the recovered acid concentration is typically 10–25 wt% lower than the feed acid concentration. For an acid loop that must return to the furnace at 55–60 wt% equivalent sulfuric acid, this dilution imposes an additional evaporation load and may require a reconcentration evaporator between the membrane stack and the regeneration unit. Diffusion dialysis is therefore more suitable as a partial acid recovery step than as a primary methacrylic acid reclamation step. It can be used to remove a fraction of free sulfuric acid from the hydrolysis liquor before extraction or stripping, thereby reducing the acid strength and improving the partition behaviour of methacrylic acid in a downstream solvent, while the recovered acid is returned to the loop after reconcentration. The membrane stack must be protected from suspended solids and from organic fouling; the feed should be filtered through a 10 µm cartridge filter and maintained below 40 °C to prevent the membrane from losing exchange capacity. The acid loop is not chemically broken, but the water management of the loop becomes more complex. If the diluted recovered acid is not reconcentrated, the furnace energy demand increases sharply, and the overall plant steam balance may be impossible to close without importing external heat. For this reason, diffusion dialysis has not displaced thermal stripping or direct esterification as the preferred route for withdrawing methacrylic acid from the sulfate intermediate process, but it remains an option for debottlenecking an existing plant when the regeneration furnace has spare evaporation capacity and when the wastewater system can accept the water that leaves with the organic product.