Silica Gel For Beer

SGB-30 and SGB-75 are amorphous silicon dioxide beer stabilization agents supplied under food-additive classification E 551 according to Commission Regulation (EU) No 231/2012 and FDA 21 CFR 172.480. SGB-30 is a low-moisture xerogel with a laser-diffraction median particle size d50 of 12 µm to 18 µm under ISO 13320:2020, a BET specific surface area of 300 m²/g to 400 m²/g under ISO 9277:2010, and a nitrogen total pore volume of 0.7 cm³/g to 1.1 cm³/g under ISO 15901-2:2006. SGB-75 is the corresponding hydrogel grade with moisture 60% to 65% by ISO 787-2, median particle size 15 µm to 25 µm, and a BET surface area of 500 m²/g to 800 m²/g after drying. The grades are differentiated from desiccant silica gel by controlled mesopore geometry, reduced iron and arsenic content, and low coarse-particle residue; their principal function is selective adsorption of haze-active proteins in finished beer, not moisture uptake.

How Does Mesopore Geometry Influence Haze-Protein Selectivity in Wort?

Haze-protein adsorption is governed by steric access to internal silanol surfaces. Pore diameters below 5 nm exclude high-molecular-mass protein fractions but admit low-molecular-mass peptides and free amino acids; pore diameters above 50 nm allow diffusion but reduce capillary contact area per unit volume. The SGB range is controlled to an average mesopore diameter of 20 nm to 30 nm, which corresponds to high affinity for the cross-linked polyphenol-protein aggregates that form colloidal chill haze at 0 °C to 4 °C. Adsorption is pH-dependent: at beer pH 4.0 to 4.6, the silica surface is negatively charged and the haze-active protein fraction carries positive domains, permitting electrostatic attraction in addition to hydrogen bonding. Competitive adsorption by low-molecular-mass peptides is partially reversible; prolonging contact beyond 24 h does not increase adsorption capacity and may release weakly bound peptide fractions back into the beer due to surface re-equilibration. The xerogel grade SGB-30 exhibits faster hydration and longer dry-storage stability, whereas the hydrogel grade SGB-75 disperses with lower dust formation but requires sealed packaging to prevent syneresis.

In production-scale operation, silica gel is prepared as a 5% to 15% w/w slurry in deaerated filtered water at 2 °C to 6 °C and metered into beer after yeast removal and before maturation or cold storage. Dosing rate is adjusted against laboratory chill-haze testing; supplier technical bulletins cite a starting window of 30 g/hL to 80 g/hL for heavily malted brews and 20 g/hL to 50 g/hL for blended high-adjunct beers, but site validation is required because wort protein load varies with malt modification, mashing intensity, and boiling regime. Contact time in a plug-flow line reactor should be between 10 min and 20 min at flow rates up to 300 hL/h; in 1000 hL horizontal maturation tanks, contact occurs over 24 h to 72 h with periodic carbon dioxide rousing. Filtration pressure rise after stabilization is limited by ensuring that the d10 particle size remains above 8 µm in laser diffraction analysis, because attrition from centrifugal transfer pumps can generate fines below 5 µm that block trap filters and DE filter precoat layers. Rotary-lobe or progressive-cavity metering pumps are preferred over diaphragm and centrifugal transfer for slurry dosing; observed production faults include air locks in diaphragm heads and incomplete dispersion in laminar-flow sections.

Xerogel and Hydrogel Specification Envelope for Low-Iron Beer Contact Use

The following table lists lot-acceptance parameters for SGB-30 and SGB-75 based on compendial methods. Values are supplier specification intervals, not measured performance in a single beer matrix.

The reduced iron specification is critical because free iron catalyzes Fenton-type oxidation and can shift the beer redox potential, accelerating staling aldehyde formation. Arsenic and lead limits mirror the food-additive purity envelope. Compendial identity tests differentiate beer-grade silica gel from industrial desiccant grades by the absence of cobalt chloride or other moisture indicators, and by the requirement that water-soluble substances remain below 1.0% after extraction under ISO 787-3.

Before dose confirmation at the filter, forced-chill and alcohol-chill tests are run on unfiltered and laboratory-dosed beer. The standard alcohol-chill test incubates a dosed sample at 0 °C for 24 h followed by turbidity measurement at 90° scatter. A dose-response curve is generated at 0 g/hL, 20 g/hL, 40 g/hL, 60 g/hL, and 80 g/hL; the optimum is selected where additional silica gel produces less than 0.2 EBC turbidity reduction per 10 g/hL increment. Foam stability measurements under NIBEM or Rudin apparatus are performed in parallel because overdosing can co-adsorb foam-positive proteins. Silica gel does not reduce microbiological challenge; downstream cold-sterile filtration remains necessary. Published data for specific foam-loss thresholds in dry-hopped unfiltered beers is limited, so pilot-scale confirmation on the actual bright beer tank is required before full production deployment.

When Silica Gel Is Substituted for PVPP in Cold-Aging Stabilization Sequences

Substitution of PVPP by silica gel is appropriate when forced-aging tests show a proteinaceous chill haze that is reversible at 20 °C and when polyphenol haze is absent or controlled by malt selection. In this configuration, the silica gel is dosed into the cold transfer line at 2 °C to 4 °C immediately after bright beer filter precoat and before a 10 min reaction coil. The downstream filter must be sized for the added solids load: a xerogel dose of 50 g/hL adds approximately 1.5 kg/h of dry solids on a 300 hL/h line, which is acceptable for candle filters with 1.0 m²/hL/h specific filtration area but can blind plate-and-frame sheet filters if precoat body feed is below 80 g/m². When replacing PVPP, the operator should monitor residual polyphenol content by photometric methods and retain a split-stream PVPP option for high-tannin seasonal malt loads. Published data for full replacement of PVPP in heavily hopped dry-hopped beers is limited; in such cases, silica gel is normally combined with 10 g/hL to 20 g/hL PVPP rather than used as a single stabilizer.

Silica gel is not interchangeable with polyvinylpolypyrrolidone, silica sol, bentonite, or tannic acid because each targets a different haze-forming species. PVPP removes polyphenols rather than proteins; its use at 10 g/hL to 40 g/hL is complementary but does not eliminate the chill-haze protein load. Silica sol is a liquid protein adsorbent with similar chemistry but smaller effective contact mass and can raise silicon dioxide residuals if not flocculated before filtration. Bentonite removes both protein and foam-positive polypeptide fractions, and overdosing at more than 40 g/hL produces filter-cake swelling and beer loss. Tannic acid precipitates protein but adds a polyphenol load that may require subsequent PVPP treatment. The comparative matrix below is limited to the stabilization mechanism and dose envelope; it does not quantify filtration resistance, which must be measured on the target filter type.

Restricting Iron, Arsenic, and Moisture Uptake Boundaries in Storage and Slurry Systems

Once delivered to the cellar, silica gel must be stored in closed containers below 60% relative humidity. Pre-drying of partially opened xerogel packaging is required if storage time exceeds 48 h without nitrogen blanketing, because moisture uptake above 5% reduces flowability and can cause bridging in silo screw feeders. The hydrogel grade must not be stored below 0 °C because ice crystal growth fractures the gel structure and shifts the particle-size distribution toward fines. Avoid simultaneous metering of bentonite slurry and silica gel slurry in the same dosing leg; charge neutralization between the anionic silica surface and cationic bentonite surfaces can produce dense agglomerates that settle in low-flow transfer lines and do not redisperse under normal cellar turbulence. The product is also incompatible with oxygen-scavenging systems that rely on sulfite addition at the same injection point due to localized pH depression. Silicon dioxide is not intended to remain in the finished beer as a dispersed phase; it is removed in the filtration sequence, and residual silicon dioxide must be verified against national food law and the filter manufacturer’s validated retention rating.