E621 Solubility Requirements for Premix Uniformity in Spray Dried Soup Bases

Monosodium glutamate monohydrate, designated INS 621 in the Codex General Standard for Food Additives and permitted under Annex II of Regulation (EC) No 1333/2008, is incorporated into spray-dried soup bases at dry-solids levels commonly between 2.0 wt% and 12.0 wt% when normalised to a moisture-free basis. Published reference data for the pure compound indicate equilibrium solubility in water at 25 °C between 60 g and 75 g per 100 g water, depending on the hydration state and crystallisation route; the corresponding aqueous solution has a pH of approximately 6.7–7.2 at a concentration of 5 wt%. These values place E621 among the most water-soluble ingredients in a typical soup-base premix, yet the practical issue in spray drying is not whether the compound can dissolve, but whether it dissolves rapidly enough under high-shear mixing to avoid residual crystalline material that persists through transfer lines and into the atomizer feed. Industrial premix preparation typically uses a jacketed conical-bottom vessel with a rotor-stator homogenizer operating at tip speeds of 15–25 m/s; in this equipment, crystalline E621 with a volume mean diameter of 150–250 µm may require between 5 min and 20 min for complete disappearance, with the longer times applying when the premix contains high concentrations of sodium chloride or when the water addition is below 55 wt%. Undissolved E621 has two consequences in a spray-dried soup powder: it produces localised glutamate assay variation in the dry mix, and it introduces dense angular particles that alter the bulk flow and sachet filling behaviour. The entire premix pathway must therefore be designed around the compound’s solubility kinetics, not its equilibrium solubility, to achieve a final powder with a water activity below 0.40 measured by ISO 18787:2017 and a particle size distribution span below 1.8 measured by laser diffraction per ISO 13320:2020.

How Does E621 Solubility Influence Wet Premix Homogeneity in Spray Drying?

In spray-dried soup bases, premix uniformity is a function of both the bulk-water dissolution of crystalline E621 and the subsequent distribution of glutamate within the continuous aqueous phase before atomization. When E621 is charged as dry crystals into a slurry containing 10–20 wt% maltodextrin DE 10–20, 5–15 wt% sodium chloride, and 5–10 wt% sucrose on a dry-solids basis, the apparent dissolution rate is significantly lower than in pure water because the dissolved salts reduce water activity and raise solution viscosity. At a Brookfield viscosity above 400–600 mPa·s measured by ISO 3219:2008 at 25 °C, turbulent mixing in an agitated vessel can be damped, leaving a stagnant heel at the bottom discharge port; this heel, composed of salt and E621 crystals, intermittently releases concentrated slugs of glutamate into the high-pressure pump, producing fluctuations in feed solids of ±1.5–3.0 wt% and corresponding shifts in finished-powder density. Production-scale mixing vessels with impeller diameter-to-tank diameter ratios below 0.35 show this fault more frequently than vessels with high-shear recirculation loops. The corrective sequence is to dissolve sodium chloride and sugars first, raise the batch temperature to 45–50 °C, and then add E621 while maintaining recirculation turnover times of 3–7 min; at these temperatures, dissolution is rapid, and the solution can be transferred through a 150–250 µm in-line filter before the feed tank. Holding the completed wet mix above 70–75 °C for longer than 20–30 min in the presence of reducing sugars from tomato, onion, or yeast extract can initiate non-enzymatic browning that reduces available glutamic acid and shifts colour by 2 ΔE or more as measured by a reflectance spectrophotometer; the process alarm is therefore set at 55 °C, with a critical control band of 45–50 °C, where dissolution speed and browning risk are balanced. These conditions are transferred to the spray-dryer feed tank with a residence time not exceeding 45 min, because even fully dissolved E621 can re-crystallise as a fine solid at product contact surfaces if the line cools below 30 °C during hold-up.

Atomizer Feed Pressure, Viscosity, and Dissolved E621

Once E621 is fully dissolved, the solution viscosity and surface tension at the atomizer determine the droplet size distribution and therefore the distribution of E621 within the dried powder. Rotary atomizers on a production-scale spray dryer of the GEA Niro FSD 4.0 class operate at wheel speeds of 9,000–12,000 rpm and generate droplets with a Sauter mean diameter of 35–70 µm from feed solids of 45–50 wt%; these conditions produce a particle shell that traps dissolved E621 in the amorphous maltodextrin phase, so crystalline re-crystallisation is suppressed and the glutamate is uniformly distributed. Pressure nozzle systems at 120–180 bar create a finer droplet size but are more sensitive to viscosity; if any undissolved E621 reaches the nozzle insert, the crystals accelerate abrasive wear and distort the atomization pattern, leading to coarse agglomerates with localised glutamate concentration. Dissolved E621 increases feed viscosity by roughly 2–8% compared with an E621-free premix at the same solids loading; this viscosity shift is operationally minor but must be considered when selecting a positive-displacement pump and when setting the low-level alarm for the feed tank. The drying air inlet temperature of 180–220 °C and outlet temperature of 80–95 °C reduce powder moisture to 2.5–4.0 wt%; at this moisture level, a maltodextrin-E621 matrix remains below the stickiness boundary, and the final powder can be discharged into a fluid bed conditioner with an outlet air temperature of 35–45 °C. Adding E621 as a dry blend to already-dried powder after this step is not equivalent to wet-mix dissolution: the compound then exists as discrete primary crystals and will not become uniformly distributed unless a separate dry-blending stage of at least 10–15 min in a ribbon blender at 60–70% fill is used.

Dry blending of spray-dried E621-containing bases with post-dried starches, vegetable powders, and spice particulates introduces a separate solubility-related defect mechanism: the pre-dissolved E621 in the glassy matrix is more hygroscopic than crystalline E621, and pockets of exposed fine material can absorb moisture, collapse, and form liquid bridges. In a ribbon blender with a usable capacity of 2,000 kg and a fill level of 60–70%, rough handling of the powder generates fines; if ambient relative humidity exceeds 55–65%, the powder can rise in water activity from 0.25 to 0.50 within 24 h when the packaging area is not conditioned to 40% RH or lower. This moisture uptake is not only a caking risk; it changes the ionic mobility of the dry matrix and can cause local segregation of E621 because collapsed regions bind to adjacent non-hygroscopic particles and resist later mixing. Production facilities therefore discharge the spray-dried base through a fluid bed dryer at an outlet air temperature of 35–45 °C and pack the finished blend in multi-layer barrier bags with a water vapour transmission rate below 5 g/m²/day at 38 °C and 90% RH measured by ASTM F1249. Analysis of dry blend uniformity uses a sampling thief with 20 mm slots inserted at 10 fixed points; the acceptance criterion is a coefficient of variation below 5.0% for glutamic acid. If the coefficient of variation exceeds 5.0%, the blender is re-run for 10–15 min and sampling is repeated, but reprocessing is not permitted if the water activity has reached 0.60 because the powder then enters a plasticising regime where further mechanical action can increase stickiness rather than improve mixing.

If Potassium Chloride Partially Replaces E621 in Reduced-Sodium Formulations

In low-sodium spray-dried soup bases, E621 levels may be partially replaced by potassium chloride or L-lysine hydrochloride at substitution ratios of 20–40% on a molar sodium basis to achieve a sodium reduction target of 25–30%. The solubility profile of this mixed salt system is less forgiving than E621 alone: potassium chloride has a water solubility of approximately 34.2 g per 100 g water at 20 °C, and its dissolution rate in mixed NaCl-E621-KCl brines is slower; thus a simple drop-in replacement without adjusting premix temperature or water addition can produce undissolved KCl crystals in the slurry, which are less dense but more abrasive than E621 crystals and can accumulate in the recirculation loop. In a premix with 35 wt% total solids, replacing 30% of the E621 with KCl raises the ionic strength and may reduce the apparent solubility of E621 due to common ion effects; published data for this specific configuration is limited, but plant-level turbidity monitoring typically shows persistence of suspended particles for 15–25 min longer than an E621-only baseline. The critical control point is therefore not equilibrium solubility but the wash-in sequence: E621 should be added after full hydration of salt and maltodextrin, followed by KCl only after the E621 solution has cleared and the temperature has reached 45–50 °C. Deviating from this sequence in a batch with a high proportion of sugar alcohols or soy sauce powder can produce a sediment of K⁺/glutamate complexes that reduces final soup pH buffering and changes the solubility of the spray-dried base upon reconstitution; the dry product may show white specks and a glutamic acid CV above 7.0%, failing internal uniformity criteria of 5.0%. In this case, the wet premix must be reworked through a colloid mill or a high-shear mixer before drying, but rework is limited to one cycle because repeated shear introduces excess air and destabilises the emulsion fraction.

A Solubility-Adjusted Sampling Protocol Must Address Salt Bridging and Fines Migration

Verification of solubility-adjusted uniformity requires multi-point sampling after blending and after pack-off, because E621 fines can migrate during transfer and packing due to particle size differences. A dry soup base with a mean particle size of 180–250 µm and a tapped bulk density of 650–750 g/L may segregate if the concentration of fine particles below 75 µm exceeds 15 wt% of the total batch; this segregation is measurable as a shift in glutamic acid content between the top and bottom of a filled hopper. The sampling plan therefore samples 10 points from the mixer and 10 points from the filling line, with the requirement that no individual point deviates from the target glutamic acid content by more than 10% and the coefficient of variation remains below 5.0%. Hot-water extraction of the powder followed by HPLC with UV detection after derivatization provides a method quantification limit of 0.1 mg/g and a linear range up to 100 mg/g; the method is validated for repeatability at a relative standard deviation below 2.0% across duplicate analyses. The same sample aliquot can be used for water activity per ISO 18787:2017 and for Karl Fischer moisture per ASTM E203, so that a high glutamic acid coefficient of variation can be correlated with a local water activity above 0.40 or a moisture above 4.0 wt%. When the CV fails, the investigation should first check the pre-spray-dryer wet mix, because dry blending cannot fully repair a solubility failure that has already produced uneven E621 distribution in the primary powder.

Evaluating Dissolution Rate Limits Under Vacuum Mixing

Vacuum mixing is sometimes specified when air entrainment in high-solid soup slurries leads to oxidative degradation of flavour components or excessive foaming in the spray-dryer feed tank. Under a vacuum of −0.4 bar to −0.6 bar, the dissolution rate of crystalline E621 can be slightly reduced because the collapse of gas bubbles removes microturbulence and lowers the local dispersion energy at the crystal surface; in a batch with 45 wt% total solids, the time for complete disappearance can extend by 3–8 min compared with atmospheric mixing at the same impeller speed. The vacuum mixing protocol should therefore delay the start of vacuum application until after E621 has dissolved, or reduce the E621 particle size to below 100 µm before charging. Process experience in this setting also shows that a drop in temperature of 5 °C during vacuum cooling can reduce the apparent saturation point enough to cause a fine re-crystallisation haze in the feed line; for this reason, the vacuum blend tank is jacketed at 50 °C and the vacuum level is released slowly over 10 min before transfer. The final powder from a vacuum-mixed batch must still meet the same dry-powder acceptance criteria: a glutamic acid coefficient of variation below 5.0%, a particle span below 1.8, and a water activity below 0.40. In case of failure, the wet premix can be sampled through an in-line turbidity sensor at 850 nm to distinguish between undissolved E621, starch granule swelling, and air bubbles; the turbidity baseline after complete dissolution should return to below 200 NTU in a well-mixed aqueous soup premix.

Reconstitution behaviour of the packaged dry soup base in 95 °C water provides the final check of solubility-adjusted premix uniformity. A 25 g serving of spray-dried product with an E621 content of approximately 1.8 g per serving should disperse completely within 60 s with gentle spoon stirring; if agglomerates remain after 90 s, the likely cause is not the E621 solubility itself but the presence of hydrophobic fat particles or overcooked starch fractions introduced during drying. In reconstituted product, glutamic acid can be measured by a hot-water extraction followed by the same HPLC procedure; the acceptable recovery is 90–110% of the calculated wet-mix contribution, with a repeatability relative standard deviation below 2.0% for duplicate analyses. In high-temperature retort soup bases that are not spray dried, E621 solubility becomes less critical because normal canning temperatures exceed 100 °C and complete dissolution occurs during thermal processing; however, for dry bouillon cubes, granules, and powdered soup bases, the controlled dissolution and re-solidification of E621 within the spray-dried matrix remains the primary determinant of final visual and gustatory uniformity. Batch acceptance for dry soup bases is therefore set at a water activity below 0.40, a particle span below 1.8, and a glutamic acid coefficient of variation below 5.0% measured across the entire pack-off shift.