Explainers · 2026-07-11

Patreon for mead making creators: honey sugar chemistry, yeast osmotic stress and HOG pathway, YAN deficiency and TOSNA 3.0 nutrient protocols, Ehrlich pathway fusel alcohols, volatile acidity control, SO₂ and sorbate stabilization, bentonite and Sparkolloid fining chemistry, and the Apple Tax

Meadmaking Patreon retention is built on the scientific documentation layer that YouTube tutorial videos cannot carry: the honey fermentation biochemistry that explains why high-gravity must creates such a punishing lag phase, the specific enzymatic pathway by which yeast converts branched-chain amino acids to fusel alcohols under nutrient stress, why volatile acidity threshold is a legal concept with biochemical roots in Acetobacter metabolism, and exactly how SO₂ molecular fraction pH-dependence determines whether your stabilization is effective. Meadmaking audiences are steadily iOS across YouTube and Instagram fermentation content. Apple Tax exposure begins November 1, 2026.

Honey sugar chemistry: fructose, glucose, water activity, and the osmotic challenge at inoculation

Honey is not simply “liquid sugar.” Its sugar composition, water activity, pH, and enzymatic content differ from sucrose in ways that matter at every stage of mead fermentation—from yeast osmotic stress at inoculation through flavor development during aging. The cheesemaker’s parallel is understanding why different milk proteins behave differently from generic “dairy protein”; for the meadmaker, honey composition is the prerequisite for understanding everything that follows in the fermentation.

Raw honey from Apis mellifera is approximately 38–40% fructose, 31–35% glucose, 17–20% water, 1–2% sucrose, and 2–7% other oligosaccharides and higher carbohydrates (maltose, turanose, erlose, kojibiose). The fructose-to-glucose ratio varies by floral source: clover honey is approximately 1.1:1 fructose-to-glucose; tupelo honey is approximately 1.5:1 (unusually high fructose, which is why tupelo mead is exceptionally smooth and slow to granulate); buckwheat honey has a 1.0:1 ratio; acacia (black locust) honey is approximately 1.4:1 fructose-dominant. This ratio matters for fermentation because Saccharomyces cerevisiae transports glucose and fructose through different hexose transporter isoforms (Hxt1p-Hxt17p family, with different Km values for each sugar), and ferments glucose preferentially at early fermentation when both are present. In high-fructose honey must, glucose is consumed earlier and fructose accumulates proportionally in the later fermentation stages, which can affect the apparent attenuation rate and residual sweetness trajectory.

The sucrose present in honey is not dietary sucrose added to the honey—it is sucrose from nectar that was partially hydrolyzed during processing by the bee’s own enzymes. Worker bees secrete invertase (β-fructofuranosidase, EC 3.2.1.26) and glucose oxidase (EC 1.1.3.4) into the nectar during processing and evaporation in the hive. Invertase hydrolyzes sucrose to fructose + glucose; glucose oxidase converts glucose to gluconolactone + H₂O₂, with gluconolactone hydrolyzing to gluconic acid. The H₂O₂ acts as a mild antimicrobial in fresh honey; the gluconic acid (pKa 3.86) contributes to honey’s pH of 3.5–4.5. Ripe honey is fully inverted: residual sucrose reflects incomplete processing time or high-sucrose nectar sources. The meadmaker’s honey will complete this inversion naturally at fermentation pH—yeast’s own invertase (Suc2p) rapidly hydrolyzes any remaining sucrose in the must.

Water activity is the critical physical property that explains both why honey is shelf-stable and why it creates osmotic stress in yeast. Water activity (a𝑤) is the effective concentration of free water in a solution, scaled from 0 (absolutely dry) to 1.0 (pure water). Raw honey at 17–20% water has a𝑤 of 0.55–0.60—far below the minimum for yeast growth (0.80–0.85 for S. cerevisiae) and even below the minimum for Zygosaccharomyces rouxii (0.62), the osmotolerant yeast that causes honey fermentation in improperly stored honey. Honey at these water activities is self-preserving precisely because no organism can grow in it. When honey is diluted to mead must at Brix 22–30 (specific gravity 1.090–1.130), the water activity rises to 0.92–0.97—well within the range permissive for yeast growth but still osmotically demanding compared to 10–14 Brix grape juice must (a𝑤 0.97–0.99). The osmotic gradient at must inoculation determines the severity and duration of the HOG pathway stress response described in the next section.

pH of honey must before and during fermentation is another compositionally driven variable. The 3.5–4.5 pH of honey is maintained partly by gluconic acid and partly by organic acids absorbed from pollen and nectar sources (malic, citric, succinic acids). When honey is diluted to must, pH typically rises to 3.7–4.2 depending on the water source and dilution ratio. Unlike wine must (pH 3.0–3.8 from tartaric and malic acid in grapes), mead must has minimal buffering capacity, so it is more susceptible to pH drift during fermentation. Lactic acid bacterial contamination drops pH rapidly in mead precisely because there is little buffering to resist the added acid. Pre-adjusting mead must to pH 3.7–3.9 before inoculation using food-grade malic acid or tartaric acid provides a modest buffering improvement and establishes conditions hostile to many bacterial contaminants while remaining within the yeast’s optimal growth range.

Yeast osmotic stress: the HOG MAP kinase pathway and glycerol accumulation

The extended lag phase that puzzles and panics new meadmakers—12, 18, sometimes 24 hours without visible fermentation activity—is not a fermentation failure. It is the predictable, biochemically determined response of Saccharomyces cerevisiae to the osmotic shock of being transferred from the rehydration medium (a dilute aqueous solution) into high-Brix honey must. The molecular mechanism of this response is the High Osmolarity Glycerol (HOG) MAP kinase pathway, the most thoroughly characterized eukaryotic osmostress signaling system, and understanding it provides the quantitative basis for predicting and managing lag phase length.

The HOG pathway operates through two parallel osmosensor branches that converge on a single MAP kinase. The first branch is the Sln1p histidine kinase branch. Sln1p is a transmembrane receptor with a histidine kinase domain facing the cytoplasm. At low osmolarity (the conditions in the rehydration medium), Sln1p is constitutively active as a kinase, autophosphorylating at His576 and relaying the phosphate through a two-component phosphorelay: Sln1p-His576-P → Sln1p-Asp1144 → Ypd1p-His64 → Ssk1p-Asp554. When Ssk1p is phosphorylated, it is inactive and cannot activate the downstream MAP kinase kinase kinases Ssk2p and Ssk22p. In other words, at low osmolarity, Sln1p activity holds the HOG pathway in the “off” state. At high osmolarity, the plasma membrane tension change inactivates Sln1p kinase activity; Ssk1p becomes dephosphorylated and active; Ssk2p and Ssk22p are activated by phosphorylation of their activation loops.

The second branch uses Sho1p, a four-transmembrane domain protein that recruits the adaptor Ste50p and MAP kinase kinase kinase Ste11p to the plasma membrane at high osmolarity, activating Ste11p via the small GTPase Cdc42p. Ste11p then directly phosphorylates and activates Pbs2p (MAP kinase kinase). Both branches therefore converge on Pbs2p, which is the sole MAP kinase kinase in the HOG pathway. Active Pbs2p phosphorylates Hog1p MAP kinase at the conserved TGY dual-phosphorylation motif (Thr174 and Tyr176 in yeast Hog1p). Activated Hog1p then executes the osmotic stress response through multiple parallel effector mechanisms.

The fastest effector response of Hog1p is direct phosphorylation of the Fps1p aquaglyceroporin channel. Fps1p is a plasma membrane channel that, under normal growth conditions, allows passive equilibration of glycerol across the plasma membrane—when the cell produces glycerol as a fermentation byproduct, Fps1p exports it to maintain the intracellular glycerol concentration near zero. Phosphorylation of Fps1p at Thr231 and Thr101 by Hog1p causes channel closure within minutes of osmotic shock, trapping intracellular glycerol. Simultaneously, Hog1p upregulates transcription of GPD1 (glycerol-3-phosphate dehydrogenase 1, which reduces dihydroxyacetone phosphate to glycerol-3-phosphate) and GPP2 (glycerol-3-phosphatase, which removes the phosphate to produce free glycerol). The combined effect is rapid intracellular glycerol accumulation: glycerol is the compatible solute that yeast uses to restore turgor by matching intracellular osmolarity to the high-osmolarity external environment without disrupting protein structure or enzymatic activity. Glycerol is a small, uncharged, highly hydroxylated molecule that does not significantly affect protein folding at concentrations that would be toxic for most salts.

The practical duration of this lag phase is determined by the initial Brix of the must and the health of the pitched yeast. At Brix 22 (typical 12% ABV traditional mead target), lag phase is 4–8 hours in well-pitched, Go-Ferm-rehydrated yeast; at Brix 30 (high-gravity sack mead targeting 16–18% ABV), lag phase can extend to 18–36 hours even with optimal nutrient preparation, because the HOG pathway must generate enough intracellular glycerol against a much steeper osmotic gradient. Staged honey addition—starting the fermentation at Brix 18–22 and adding honey in increments as fermentation proceeds, a practice called “step feeding”—avoids the worst of the osmotic shock by never subjecting the yeast to must above Brix 22 at any single inoculation event. The HOG pathway still activates at each honey addition, but from a much smaller baseline osmolarity difference.

The glycerol produced during the HOG response is not lost from the mead; it remains dissolved and contributes positively to mouthfeel. Glycerol at 10–20 g/L (typical in well-fermented mead) adds sweetness perception (detection threshold approximately 5 g/L), body, and viscosity that distinguishes well-fermented mead from thin, watery products. The HOG pathway’s glycerol response is therefore both a stress response and an inadvertent quality contributor—high-gravity meads that experience more osmotic stress and thus produce more glycerol tend to have fuller body at the same residual sweetness level.

YAN deficiency: nitrogen fractions in honey, hydrogen sulfide, and the TOSNA 3.0 protocol

Yeast assimilable nitrogen (YAN) is the term for the nitrogen compounds in the fermentation substrate that yeast can actually absorb and use for biosynthesis. YAN encompasses three fractions: free amino nitrogen (FAN), which is the alpha-amino nitrogen from individual amino acids and small peptides; free ammonia (ammonium ion, NH₄⁺); and urea. The sum of these three fractions in mg/L as nitrogen determines whether yeast have sufficient nitrogen to complete fermentation without stress-related defects.

Honey is fundamentally nitrogen-poor. The dominant nitrogen compound in honey is proline, which constitutes 25–50% of total amino acids by mass in most honey types. Proline is poorly assimilated by S. cerevisiae during anaerobic fermentation: proline catabolism requires proline oxidase (Put1p), which is a mitochondrial enzyme that requires oxygen and NAD⁺ as an electron acceptor. Under the anaerobic conditions of active fermentation, Put1p activity is suppressed by both oxygen limitation and glucose repression via catabolite repression of the PUT gene regulon. Proline in honey must is therefore largely unavailable to yeast as a nitrogen source during fermentation, even though it dominates the amino acid profile. Effective YAN from honey is typically 25–100 mg/L from the non-proline amino acids (glutamic acid, alanine, phenylalanine, tyrosine, leucine) plus trace ammonia; total effective YAN rarely exceeds 100 mg/L without addition.

The minimum YAN for clean fermentation depends on must gravity. At 10–14 Brix (wine or low-gravity mead), 150 mg/L is generally sufficient. At 22–26 Brix (traditional mead), 250–300 mg/L YAN is the recommended minimum. At 28–30 Brix (high-gravity sack mead), 350–450 mg/L YAN is required. The scaling relationship exists because higher-gravity fermentations produce more biomass (more yeast cell divisions) and subject yeast to greater stress (osmotic, ethanol), which increases nitrogen demand per unit of sugar fermented. When YAN is insufficient at any must gravity, the first symptom is hydrogen sulfide production.

Hydrogen sulfide is produced when sulfite reductase (EC 1.8.1.2, the enzyme that reduces sulfite to sulfide as part of the cysteine/methionine biosynthesis pathway) cannot access adequate organic nitrogen precursors downstream. The mechanism: yeast assimilates sulfate from the must into adenosine-5-phosphosulfate (APS) and then 3-phosphoadenosine-5-phosphosulfate (PAPS), then reduces sulfite to sulfide. Sulfide is normally immediately incorporated into O-acetylserine by cystathionine γ-synthase to begin cysteine biosynthesis. When serine or the entire cysteine biosynthesis pathway is limiting from nitrogen deficiency, sulfide cannot be immediately captured and is released from the cell as volatile H₂S (rotten egg odor, detection threshold 0.1–10 ppb). H₂S production in mead is thus a specific diagnostic indicator of YAN deficiency, not of general fermentation problems, and the correct response is a Fermaid-O addition—not degassing, temperature changes, or yeast re-pitching.

TOSNA 3.0 addresses YAN deficiency through a combination of high-quality rehydration and staggered organic nitrogen additions. The protocol begins before the yeast ever touches the must. Go-Ferm Protect Evolution (a specific commercial formulation by Scott Laboratories containing autolyzed yeast extracts rich in cell membrane sterols, unsaturated fatty acids, vitamins, and zinc) is dissolved in water warmed to 43°C at 20 g per gallon of must to be fermented. Active dry yeast (typically EC-1118, 71B, D47, or K1-V1116 depending on target flavor profile) is added to the warm Go-Ferm suspension at a rate of 25 g per 5 gallons and rehydrated for 20 minutes with occasional stirring. The 43°C temperature is specifically chosen: it is above the minimum growth temperature of most bacterial contaminants (eliminating the risk of the Go-Ferm suspension becoming a growth medium during rehydration), and it is below the temperatures that cause yeast viability loss (generally above 50°C for most active dry yeast strains). During the 20-minute rehydration, yeast cells that are in the process of reconstituting their plasma membrane from the dehydrated state have transiently permeable membranes that allow Go-Ferm’s ergosterol and oleic acid components to be absorbed directly. Ergosterol (a critical membrane sterol in yeast that functions analogously to cholesterol in mammalian cells, regulating membrane fluidity and ethanol tolerance) and oleic acid (an unsaturated fatty acid that maintains membrane fluidity at fermentation temperatures) are synthesized de novo by yeast but require oxygen for their biosynthetic pathways. Once anaerobic fermentation is established, yeast can no longer synthesize sterols or unsaturated fatty acids; the supply absorbed during rehydration must last the entire fermentation. Yeast rehydrated in Go-Ferm enter the fermentation with a larger initial sterol/USFA reserve, which translates to greater ethanol tolerance and fermentation completion rate.

After pitching the rehydrated yeast into must (temperature-adjusted to within 10°C of the must temperature to avoid thermal shock), Fermaid-O additions are made at 24-hour, 48-hour, and 72-hour intervals, then at the 1/3 sugar depletion point. Each addition supplies approximately 0.9–1.25 g/L Fermaid-O (providing approximately 25–40 mg/L YAN per addition), for a total of 100–150 mg/L YAN from additions on top of the honey’s native 25–100 mg/L. The staggered timing matches nitrogen supply to cell growth rate: at 24 hours, the HOG-mediated lag phase is ending and cells are entering exponential growth, requiring nitrogen for structural proteins; at 48–72 hours, biomass is maximal and protein synthesis rates are highest; at 1/3 depletion, the remaining nitrogen addition supports the final cell population through the slowing fermentation. Adding all nitrogen at pitch (front-loading) creates a brief nitrogen surplus that drives excess amino acid catabolism via the Ehrlich pathway, producing more fusel alcohols; the staggered approach keeps nitrogen just above deficiency threshold throughout, minimizing both H₂S (too little) and fusel alcohol (too much) production.

Ehrlich pathway: fusel alcohol biosynthesis from branched-chain amino acid catabolism

Fusel alcohols—isoamyl alcohol, isobutanol, n-propanol, 2-phenylethanol, and others—are produced by yeast through the Ehrlich pathway, a set of three enzymatic reactions that catabolize excess amino acids during fermentation. Understanding this pathway provides meadmakers with the molecular basis for both the causes of fusel production (too much nitrogen from amino acid sources, high temperature, nitrogen starvation) and the specific fermentation conditions that minimize them.

The Ehrlich pathway begins with transamination: the amino group is transferred from an amino acid to alpha-ketoglutarate by a transaminase (branched-chain amino acid transaminase Bat1p/Bat2p for leucine, isoleucine, valine; aromatic amino acid transaminase Aro8p/Aro9p for phenylalanine, tyrosine, tryptophan; threonine aldolase Cha1p for threonine). The products are the corresponding alpha-keto acid (also called 2-oxoacid) and glutamate. The second step is decarboxylation: the alpha-keto acid is decarboxylated by one of the pyruvate decarboxylase family enzymes (Pdc1p, Pdc5p, Pdc6p, Aro10p) to the corresponding aldehyde plus CO₂. The third step is reduction: the aldehyde is reduced to the alcohol by alcohol dehydrogenase (Adh1p–Adh5p) using NADH, regenerating NAD⁺ for use in glycolysis—the same role that ethanol fermentation plays.

The specific fusel alcohol produced depends on which amino acid enters the pathway: leucine (2-amino-4-methylpentanoic acid) → alpha-ketoisocaproate (4-methyl-2-oxopentanoate) → 3-methylbutanal → isoamyl alcohol (3-methyl-1-butanol). Isoamyl alcohol has a banana/fusel odor threshold of 50–65 mg/L in mead; concentrations above 150 mg/L produce the harsh, nail-polish-remover character associated with poorly made mead and cheap wine. Valine (2-amino-3-methylbutanoic acid) → alpha-ketoisovalerate (3-methyl-2-oxobutanoate) → 2-methylpropanal → isobutanol (2-methyl-1-propanol). Isobutanol threshold approximately 200 mg/L; harsh, solvent-like at higher concentrations. Phenylalanine → phenylpyruvate → phenylacetaldehyde → 2-phenylethanol. 2-Phenylethanol is a qualitatively valuable fusel alcohol—its threshold is 10–75 mg/L in mead and its character is rose, honey, floral, which at sub-threshold concentrations contributes positively to mead aroma complexity. 2-Phenylethanol production is maximized by using yeast strains with high Aro10p activity (such as 71B, which is selected partly for its high 2-phenylethanol production in fruit wines and mead) and is minimized by suppressing the pathway through low-temperature fermentation.

Fusel alcohol production is highest under three conditions: excess amino acid nitrogen (causing high Ehrlich pathway flux as yeast catabolize excess amino acids to eliminate the nitrogen burden), nitrogen starvation (when yeast must catabolize their own stored amino acids to access nitrogen for biosynthesis), and high fermentation temperature (elevated temperature increases all enzymatic reaction rates including transaminase and decarboxylase activities). The TOSNA approach minimizes the first two: staggered Fermaid-O additions keep nitrogen just sufficient without excess, and the organic form of nitrogen in Fermaid-O (amino acids from autolyzed yeast, which yeast transport via specific amino acid permeases and incorporate directly into proteins) is preferentially used for biosynthesis rather than Ehrlich pathway catabolism. Low fermentation temperature (15–18°C for traditional and fruit meads, versus 22–27°C for quick traditional) reduces fusel production at the cost of longer fermentation duration.

Volatile acidity: Acetobacter contamination, threshold limits, and prevention

Volatile acidity (VA) is the measurement of acetic acid and other short-chain volatile acids (formic, propionic, butyric in trace amounts) in a finished fermented beverage. Acetic acid constitutes over 95% of VA in mead and wine. VA is measured as g/L acetic acid by steam distillation followed by acid-base titration (Cash still method), or estimated by enzymatic assay. In mead, the typical threshold for detectable vinegary character is 0.9–1.2 g/L, though some judges and consumers detect it at 0.6 g/L. Competition standards follow BJCP and AHA guidelines: VA above 1.0 g/L in mead is a fault; above 1.5 g/L is typically a disqualifying fault.

VA is produced primarily by Acetobacter species (Acetobacter aceti, A. pasteurianus) in the presence of oxygen. Acetobacter are obligate aerobic bacteria that oxidize ethanol to acetic acid via the membrane-bound alcohol dehydrogenase (ADH, EC 1.1.99.8) and aldehyde dehydrogenase (ALDH, EC 1.2.99.7) enzymes: ethanol → acetaldehyde → acetic acid. The oxygen requirement is absolute—Acetobacter cannot produce VA in a fully anaerobic environment. VA formation in mead is therefore a contamination and oxygenation management problem: the mead must never be exposed to air once fermentation is established and ethanol is present. The combination of ethanol (substrate) + oxygen (electron acceptor) + Acetobacter (enzyme source) produces VA; remove any one of the three and VA production stops. Prevention strategies in order of importance: maintain an airlock that excludes oxygen throughout fermentation and aging; fill vessels as full as possible during bulk aging to minimize headspace; use SO₂ at effective levels (below pH 3.5, effective molecular SO₂ is more achievable) to inhibit Acetobacter growth; rack under inert gas (argon or CO₂) rather than exposing to air during transfers; monitor VA by titration or test strips every 30 days during bulk aging.

Saccharomyces cerevisiae also produces acetic acid as a minor fermentation byproduct through pyruvate decarboxylase → acetaldehyde → acetaldehyde dehydrogenase → acetate, and through the acetyl-CoA pathway. Yeast-derived VA is typically 0.1–0.3 g/L in clean mead fermentations and is not a quality problem at these concentrations. VA above 0.4–0.5 g/L usually indicates microbial contamination rather than yeast metabolism, though high-gravity or stressed fermentations can produce 0.4–0.5 g/L from yeast alone.

SO₂ and sorbate stabilization: molecular fraction pH-dependence and the geranium off-flavor

Sulfur dioxide (SO₂) is the primary antimicrobial and antioxidant agent used in mead stabilization before packaging. Its effective antimicrobial form is molecular SO₂ (H₂SO₃, the undissociated species), not the total SO₂ added. The distribution between molecular SO₂ and its ionic forms (bisulfite HSO₃⁻ and sulfite SO₃²⁻) depends entirely on pH through the Henderson-Hasselbalch relationship. At the pKa values for SO₂ in aqueous solution (pKa1 = 1.81 for H₂SO₃ ↔ H⁺ + HSO₃⁻; pKa2 = 6.91 for HSO₃⁻ ↔ H⁺ + SO₃²⁻), the fractions at mead pH are: at pH 3.2, approximately 0.81% molecular SO₂; at pH 3.5, approximately 0.45%; at pH 3.8, approximately 0.23%; at pH 4.0, approximately 0.14%. The antimicrobial effective concentration of molecular SO₂ is approximately 0.8 mg/L free molecular SO₂. To achieve 0.8 mg/L molecular SO₂ at pH 3.5, total free SO₂ must be 0.8 / 0.0045 = 178 mg/L. At pH 3.8, the required total free SO₂ is 0.8 / 0.0023 = 348 mg/L. At pH 4.0: 0.8 / 0.0014 = 571 mg/L. Above approximately 350 mg/L total SO₂, the mead develops a pungent sulfur dioxide odor detectable by tasters, and above 450 mg/L it is likely undrinkable. This means that mead at pH 4.0 cannot be effectively stabilized with SO₂ alone—the required SO₂ level is above the sensory tolerance threshold. Acidifying mead to pH 3.4–3.6 before adding SO₂ allows effective stabilization at total free SO₂ of 40–75 mg/L (a level undetectable to most tasters), which is why commercial meaderies that produce shelf-stable bottled mead adjust pH downward if necessary before sulfiting.

Potassium sorbate (E202) is used as a second antimicrobial agent specifically to prevent yeast re-fermentation in back-sweetened meads. Sorbic acid inhibits yeast by interfering with the plasma membrane proton gradient and suppressing the uptake of yeast assimilable nitrogen. It does not inhibit yeast at the ion concentrations available (25–250 mg/L sorbic acid equivalent) at wine/mead pH—the inhibition is specifically against the cell energetics of active growth. Sorbate does not sterilize; it prevents refermentation in still-alive yeast. It is ineffective against bacteria, particularly against lactic acid bacteria (LAB). The most critical practical limitation of sorbate is the geranium off-flavor that develops when sorbate reacts with lactic acid bacteria. The mechanism: Oenococcus oeni, Lactobacillus, and Pediococcus species convert sorbic acid via decarboxylation to 1,3-pentadiene, then further metabolize it to 2-ethoxyhexa-3,5-diene (geraniol-like compound) and related structures that produce a powerful geranium flower or pelargonium leaf odor at threshold concentrations of 0.01–0.1 ppb. A mead that smells like geranium flowers after back-sweetening + sorbation is a mead that was contaminated with LAB before sorbate addition. The prevention is absolute: SO₂ at effective molecular levels must be established before sorbate addition, to kill LAB before they encounter the sorbate substrate. Adding sorbate to a mead that contains active LAB reliably produces the geranium defect and cannot be remediated.

Fining chemistry: bentonite cation exchange and Sparkolloid electrostatic bridging

Mead clarity is a quality marker and a practical requirement for long-term stability. Colloidally turbid mead contains suspended particles (yeast cells, yeast cell wall fragments, protein-tannin complexes, pectin from fruit additions, and in raw-honey meads, pollen grains) that can sediment unpredictably during bottle aging, creating haze or sediment that consumers may interpret as contamination. Fining agents work through two distinct physicochemical mechanisms—cation exchange adsorption and electrostatic bridging—and selecting the correct fining agent requires knowing which mechanism removes which type of particle.

Bentonite is a montmorillonite clay mineral (a phyllosilicate with the formula (Na,Ca)₀₁₂(Al,Mg)₂Si₄O₁₀(OH)₂·nH₂O) that carries a permanent negative surface charge at the pH range relevant to mead (pH 3–4). The negative surface charge arises from isomorphous substitution of Al³⁺ for Si⁴⁺ in the tetrahedral silicate layer, and Mg²⁺ for Al³⁺ in the octahedral layer, creating a permanent charge deficit that is compensated by exchangeable cations (Na⁺ or Ca²⁺, depending on the bentonite type) in the interlayer space. Bentonite fines positively charged particles by cation exchange: positively charged particles (proteins at pH below their isoelectric point—most wine/mead proteins are positively charged at pH 3–4 since their pI is typically 4–6) are attracted to and bind to the negatively charged bentonite surface, forming flocs that sediment under gravity or centrifugal force. The bentonite surface provides binding sites for thousands of protein molecules per platelet; each bentonite floc aggregates many platelets covered with adsorbed protein, and the large floc mass sinks rapidly through the mead. Bentonite is the standard fining agent for heat stability testing: adding bentonite and then heat-shocking (80°C/15 minutes) the bentonite-treated mead predicts whether residual heat-unstable proteins will cause haze after bottling, because bentonite removes the same proteins that heat-shock would denature-aggregate.

Sparkolloid is a commercial fining agent (Scott Laboratories) composed of polysaccharide (principally β-glucan derived from diatomaceous earth processing) complexed with a diatomaceous earth carrier. It carries a net positive charge at mead pH. Its mechanism is electrostatic bridging rather than cation exchange: Sparkolloid particles (positively charged) attract and bind to negatively charged particles suspended in the mead (including yeast cell wall fragments, which are negatively charged at mead pH because their surface mannoproteins are negatively charged; certain pollen particles; and large protein aggregates with net negative charge at pH above their isoelectric point). Sparkolloid is often used after bentonite fining as a “polish” step: bentonite removes positively charged protein haze; Sparkolloid removes remaining negatively charged particles. Together, they target the full charge spectrum of suspended particles. Sparkolloid is prepared by dissolving in boiling water (to fully hydrate and disperse the polysaccharide) and added while still hot to the mead; the positively charged Sparkolloid particles are distributed throughout the mead and electrostatically attract and bind negatively charged particles on contact.

Chitosan (from Aspergillus niger cell walls, not crustaceans, in the commercial preparation used for vegan-compatible clarification) is a positively charged polysaccharide that works similarly to Sparkolloid through electrostatic attraction to negatively charged particles. Kieselsol (colloidal silicon dioxide, SiO₂, negatively charged at mead pH) is used as the first step in a two-stage kieselsol-chitosan fining: kieselsol adsorbs positively charged protein particles; chitosan is then added to bridge the kieselsol-protein flocs into larger aggregates that sediment faster. Isinglass (collagen derived from fish swim bladders, positively charged at wine/mead pH) is a traditional fining agent with high specificity for yeast cell wall fragments and works through both electrostatic attraction and physical aggregation. Each fining agent has different particle selectivity, settling rate, and potential for over-fining (stripping too much protein and reducing body), so understanding the mechanism determines when each agent is appropriate and at what dose.

iOS rates and Apple Tax

Meadmaking and craft fermentation creator iOS rates are high across the primary audience platforms. YouTube meadmaking content—TOSNA protocol walkthroughs, nutrient addition logs, fermentation kinetics temperature curves, fining trials side-by-side, bottling and carbonation tutorials, honey varietal flavor comparisons, competition entry preparation—tracks at 62–74% iOS, reflecting a home fermentation audience that watches tutorial content on tablet in the brewing space and researches protocols on phone. Instagram meadery content—finished mead clarity documentation in backlit bottles, honey varietal sourcing photography, competition medal announcements, bottling label aesthetics—tracks at 70–82% iOS; the craft beverage aesthetics and artisan food community is heavily iOS-concentrated. Pinterest mead boards—recipe pins, honey sourcing guides, equipment setup photography—track at 70–80% iOS.

Beginning November 1, 2026, Apple charges Patreon 30% on every subscription payment processed through the iOS app.

At $200/month with 68% iOS: approximately $40.80/month ($489.60/year) in Apple fees. At $350/month with 74% iOS: approximately $77.70/month ($932.40/year). At $500/month with 80% iOS: approximately $120/month ($1,440/year). Enable Patreon’s web-only billing toggle before October 31, 2026 and update all subscription CTAs—Instagram bio link, YouTube About page URL, Reddit profile link—to the direct Patreon web URL. Verify with a test subscription from Safari on an iPhone before November 1.

KeepTier is a self-hosted membership page for creators who want 100% of their tier revenue and zero Apple Tax. Plans from $9/month.


Patreon for mead making creators · Patreon for cheesemaking creators · Patreon for sourdough creators · Eight Patreon alternatives compared