Explainers · 2026-07-03 · ~3,900 words

Patreon for jewelry making creators: silver alloy metallurgy, firescale mechanics, solder flux chemistry, AWG wire gauge selection, liver of sulfur patina chemistry, and the Apple Tax

The SEO guide to jewelry making Patreon covers tier structure, solder sequence rationale, bezel height, wire gauge relationships, and CAD wall thickness — the what of a working metalsmithing Patreon. This post covers the why at the chemistry and metallurgy level: why copper in sterling causes firescale through internal oxide diffusion below the silver surface; why argentium’s germanium content interrupts that mechanism; how work hardening accumulates in an FCC crystal lattice and what annealing does to undo it; what the boric acid and borax flux layer is doing on the metal surface at the molecular level; how pickle removes copper oxides through an acid dissolution reaction and what happens when a steel tool contacts the pickle bath; how firescale forms below the surface rather than on it, and why that makes it invisible until the polish reveals it; how AWG gauge numbers correspond to actual wire diameters and why that matters for documentation; and how liver of sulfur creates color through thin-film interference from a growing Ag⊂2;S layer rather than bulk pigmentation. Jewelry audiences are among the most iOS-concentrated in craft content — the Apple Tax exposure for Instagram-primary jewelry creators is among the highest of any craft category.

Silver alloy metallurgy and work-hardening mechanics

Sterling, fine silver, and argentium: what the alloying elements actually do

Sterling silver is defined as 92.5% silver and 7.5% copper (denoted Ag 925 in hallmarking). The copper is a deliberate alloying addition: pure silver (99.9% Ag, fine silver) is far too soft for most structural jewelry applications. The slip planes in the face-centered cubic (FCC) silver lattice allow dislocations to move with little resistance, producing a metal that deforms easily under minimal force. Adding 7.5% copper atoms into the FCC silver lattice creates solid solution strengthening: the copper atoms, which are slightly smaller than silver atoms, sit in the lattice positions normally occupied by silver atoms and create local lattice strain fields around themselves. These strain fields interact with dislocations, impeding their movement and raising the stress required to deform the metal. The result is a measurably harder, springier alloy that holds its shape through the forming, setting, and burnishing operations that would distort fine silver.

Fine silver (99.9% Ag) has no copper and therefore no copper-related chemistry: no firescale, no copper oxides in the pickle bath, no copper flash risk. It is brighter white than sterling after finishing because the silver surface reflects light without copper oxide interference. It is the alloy of choice for fine granulation work (where individual silver spheres are fused to a base sheet without solder through a diffusion bonding mechanism), for PMC (precious metal clay, which fires down to near-fine-silver purity), and for any application where freedom from firescale is worth the reduction in mechanical hardness.

Argentium silver (93.5% Ag + 6.5% Ge in the most common formulation; argentium 935) substitutes germanium for part of the copper alloying fraction. Germanium is a metalloid in Group 14 of the periodic table, two rows below carbon. In the argentium alloy, germanium occupies FCC lattice positions alongside the silver atoms, contributing to solid solution strengthening as copper does, but with a different set of surface chemistry properties. The key difference: germanium has a substantially higher affinity for oxygen than copper at the temperatures used in jewelry soldering. When argentium is heated, germanium atoms that have diffused to the surface oxidize preferentially to form germanium dioxide (GeO⊂2;), a thin, transparent oxide layer. This GeO⊂2; layer forms a self-limiting barrier at the alloy surface before copper can oxidize in appreciable quantity. Because copper oxidation is what drives firescale formation, the germanium surface barrier suppresses firescale dramatically. Argentium also supports fusion welding — joining two argentium pieces by heating the contact zone to melting temperature without added solder, allowing the silver to bond by diffusion at the join interface — a technique that is difficult with sterling because the copper concentration at the join area varies and sterling’s firescale tendency complicates the clean surface contact that fusion welding requires.

Work hardening in FCC silver: dislocation mechanics

Silver and copper are both FCC metals. In the FCC crystal structure, twelve nearest-neighbor atoms surround each atom in a close-packed arrangement, and plastic deformation proceeds by dislocation movement on the close-packed {111} planes in the <110> directions — the slip systems in the FCC lattice. When a dislocation line moves through the crystal on its slip plane, it shifts the crystal above the slip plane relative to the crystal below by one Burgers vector, allowing plastic deformation to occur at a stress far below what would be required if all atoms in a plane moved simultaneously.

Cold deformation — any bending, drawing, coiling, or hammering done at room temperature without subsequent annealing — does not merely move pre-existing dislocations through the crystal. It generates new dislocations through multiplication mechanisms. A Frank-Read source is a segment of dislocation line pinned at both ends by point defects or intersecting dislocations; when applied stress forces the segment to bow outward, it curves around its anchor points, eventually wrapping around and reconnecting behind itself, producing a complete dislocation loop that expands outward under stress and leaves the original pinned segment ready to repeat the process. Each stress cycle generates more dislocation loops, increasing dislocation density per unit volume.

As dislocation density increases, dislocations from different slip systems intersect each other. Dislocations on intersecting slip planes interact through their elastic strain fields and can become physically entangled at intersection points: a dislocation cannot pass through another dislocation without both cutting through each other and leaving behind jogs (kinks out of the slip plane) that require additional energy to move. These jogs and entanglements progressively impede dislocation movement, requiring greater applied stress to continue deforming the metal. This is work hardening: the metal becomes harder and stiffer with each manipulation because its dislocation density has increased to the point where further dislocation movement is mechanically obstructed.

Wire temper designations and annealing mechanics

Wire is sold in three standard temper designations that name positions on the work-hardening spectrum. Dead-soft wire is freshly annealed: it has been heated to the recrystallization temperature and held there long enough for new, low-dislocation grains to form and grow through the metal. Dislocation density is at or near minimum. Dead-soft wire bends easily in any direction without springback, is forgiving of errors and re-bends, and cannot hold coil tension without slipping. It is the starting point for most wire-wrapping work. Half-hard wire has undergone partial work hardening, either through the wire drawing process at the manufacturer or through deliberate work by the jeweler. It has measurable springback after bending, holds its formed shape without additional work hardening, and is the preferred temper for structural frames in wire wrapping and for prong stock that needs to hold a bent shape. Full-hard wire is at or near maximum achievable work hardening for that alloy and gauge — it is stiff, springy, and resistant to bending. It is used for spring components and structural elements that need to hold form under repeated flexion.

Annealing mechanics: for sterling silver, recrystallization nucleation occurs at high-energy grain boundaries (where dislocation density is highest and the stored elastic strain energy is greatest) at approximately 580–650°C. At this temperature range, thermal energy is sufficient to allow short-range atomic diffusion within the metal, enabling new grain nuclei to form at boundaries and then grow by consuming the surrounding deformed, high-dislocation material. The growing new grains have low dislocation density and erase the work-hardening history of the deformed regions they consume. Above approximately 900°C, grain coarsening begins: already-recrystallized grains grow larger by consuming neighboring grains, driven by the reduction in total grain boundary area and surface energy. Overly coarse grain structure reduces toughness and can make the metal surface appear textured or “orange peel” after rolling or forming. The annealing target for sterling is to reach 580–650°C (visible as a dull, barely visible red heat in dim conditions — just before first visible glow or at the threshold of first visible glow) and hold briefly, then quench in water after the piece has cooled to black heat. Quenching at black heat (not immediately on removal from flame) removes the surface copper oxides from the quench water before they diffuse further into the metal surface.

Solder alloy composition and flux chemistry

Silver solder grade compositions and flow temperatures

Silver solders are silver-copper-zinc ternary alloys (with some formulations also containing tin, cadmium-free substitutes, or small additions of other elements). The silver content determines the flow temperature: higher silver content raises the solidus temperature; lower silver content and higher zinc fraction lowers it. The grade system — hard, medium, easy, extra-easy, and IT — names positions in the flow temperature range.

Hard solder (IT solder) contains approximately 82% Ag plus copper and zinc, with a flow temperature of approximately 785°C (1445°F). IT solder is used for the first join on a piece because no subsequent soldering operation in the workflow will reach 785°C — all subsequent joins will be made at lower solder grades, which flow at lower temperatures, leaving the hard solder join solid. Medium solder contains approximately 70% Ag plus copper and zinc, flowing at approximately 720°C (1328°F). It is used for second joins in a sequence and for any join that must be made at a temperature safely below the hard solder flow point. Easy solder contains approximately 65% Ag, flowing at approximately 700°C (1292°F). It is used for third joins, for delicate settings where lower heat minimizes thermal damage to adjacent elements, and for any situation where the proximity to prior joins requires the lowest possible flow temperature. Extra-easy solder flows at approximately 665°C and is used for the final joins in complex pieces where the near-proximity of multiple prior joins makes anything higher risk a cascade reflow. The color match between solder and base metal varies with silver percentage: hard solder at 82% Ag is a closer color match to sterling than easy solder at 65% Ag, which appears slightly yellower due to the higher copper and zinc fraction.

Flux mechanism: the molten glass layer and its failure modes

Flux serves one functional purpose: to maintain an oxygen-free environment at the metal surface during the heating cycle so that copper in the sterling alloy cannot oxidize and so that any existing surface oxides are chemically reduced before solder flows. Without flux, the copper in sterling begins oxidizing at approximately 400°C, forming a thin black CuO layer at the surface. Solder cannot wet or flow across an oxidized surface; the oxide layer acts as a physical barrier between the liquid solder and the base metal. Flux prevents oxide formation during the heating phase and promotes oxide dissolution where thin oxide layers exist before the flux was applied.

The standard flux for silver soldering is a boric acid and borax mixture. Boric acid (H⊂3;BO⊂3;) is a mild acid that reacts with metal oxide surfaces, partially dissolving them. Borax (sodium tetraborate, Na⊂2;B⊂4;O⊂7;) melts at approximately 740°C to form a glassy liquid (the same chemistry exploited in forge welding flux). Together, mixed as a paste or dissolved in water with the addition of a small amount of dish soap to improve adhesion, they produce a flux that behaves in a defined sequence: water evaporation (100°C), boric acid melting and initial surface reaction (~170°C), borax softening and beginning to flow (~500°C), full borax melt and liquid glass layer formation (~740°C). At and above 740°C, the flux is a liquid borosilicate-like glass layer across the metal surface, excluding atmospheric oxygen from reaching the metal below.

Flux failure diagnostics are specific and teachable. Flux burns brown and disappears before solder flows: the metal was overheated past the flux’s upper service temperature, consuming the flux before solder could be introduced. The solution is to heat more slowly and introduce solder as soon as the flux has fully melted and the metal is at the solder’s flow temperature. Flux bubbles violently and displaces from the surface: water in the flux was not fully evaporated before the heat was increased. The boiling water steam literally erupted through the flux coating, carrying it off the surface. The solution is to hold the torch farther from the metal during initial heating to drive water off gently before increasing temperature. Flux lumps and beads on the surface rather than flowing: the flux was applied too thickly, the outer surface dried and crusted before the interior had evaporated, and the crust prevents water from escaping evenly. Thin the flux with more water or apply more sparingly. Flux for temperatures above 800°C: fluoride-added fluxes extend the service temperature range. Fluoride ions more aggressively attack metal oxide surfaces and the flux mixture melts at a higher temperature than standard borax/boric acid combinations. They are required when using hard (IT) solder, which flows at 785°C, because standard borax flux has already partially burned away at this temperature if heating was extended.

Pickle chemistry: sodium bisulfate, copper oxide removal, and galvanic copper flash

Pickle is an acid bath used after every solder join to remove flux residue and copper oxides from the metal surface. The standard modern pickle is sodium bisulfate (NaHSO⊂4;) dissolved in water — sold as pH Down, pool acid granules, or branded jewelry pickle. In water, sodium bisulfate dissociates: NaHSO⊂4; → Na⊃+; + H⊃+; + SO⊂4;⊃2;−. The free hydrogen ion (H⊃+;) and the bisulfate ion (HSO⊂4;−) create an acidic solution with pH typically in the range of 1.5–2.5 at working concentration.

The acid reacts with copper oxides by converting them to soluble copper sulfate: CuO + H⊂2;SO⊂4; → CuSO⊂4; + H⊂2;O. Copper(II) sulfate (CuSO⊂4;) is soluble in water and turns the pickle solution progressively bluer as it accumulates over multiple sessions. The surface black (CuO) and some of the surface cuprous oxide come off in the pickle, leaving the silver surface bright and clean. The flux residue — glassy borax that has cooled to a solid — is also attacked by the acid and dissolves into solution.

Pickle solution temperature materially affects reaction rate. Warm pickle at 50–60°C acts significantly faster than cold pickle — the elevated temperature increases the reaction kinetics of acid attack on copper oxide surfaces. A piece that requires five minutes in cold pickle may be clean in ninety seconds in warm pickle. However, warm pickle exhausts sooner: the faster reaction consumes the available acid capacity more quickly, and the pickle may need replacing or refreshing more often in studios that run warm. Document per session: the pickle solution temperature at the start of the session (thermometer reading), the approximate time for a freshly soldered piece to emerge clean, and the visual appearance of the solution (pale yellow is fresh; blue-green indicates significant copper accumulation; dark blue-green indicates exhausted or near-exhausted solution).

The galvanic copper flash hazard: steel tweezers, binding wire clips, or any iron or steel tool that contacts the pickle solution creates an electrochemical cell. The steel (iron) is the anode in the galvanic couple and dissolves: Fe → Fe⊃2;+ + 2e−. The copper ions already dissolved in the pickle solution are reduced at the cathode — which is any silver or copper metal in contact with the solution through the same electrically conductive path: Cu⊃2;+ + 2e− → Cu. The reduced copper metal deposits on the silver piece as a pink-orange copper flash. This copper-flashed surface must be removed before any further soldering: solder flows over the copper flash rather than the silver, creating a join with copper contamination at the interface that will discolor over time and may fail mechanically. Use only copper, titanium, or plastic tweezers and tools in the pickle. If a piece receives copper flash, clean it by returning it to fresh pickle (without steel contact), or by light abrasion with fine abrasive paper or fiberglass scratch brush, and rinse thoroughly before soldering.

Firescale formation and prevention documentation

What firescale actually is and why pickle cannot remove it

Firescale is one of the most misunderstood phenomena in silversmithing because it is frequently confused with surface oxidation. Surface oxidation — the black copper oxide (CuO) layer that forms on the exterior of heated sterling — is visible immediately after heating, is removed in pickle, and leaves no lasting trace. Firescale is something entirely different: it is an internal cuprous oxide (Cu⊂2;O) layer formed below the silver surface, not on it.

Here is the mechanism. When sterling is heated above approximately 400°C, copper at the surface and throughout the bulk of the alloy begins to oxidize. At the outer surface, CuO (black copper oxide, in which copper is in the +2 oxidation state) forms first. But simultaneously, oxygen from the atmosphere diffuses inward into the metal below the surface. Silver has a higher oxygen permeability than copper at high temperatures: oxygen atoms dissolve in the silver lattice and diffuse inward through it. Where oxygen diffusing inward meets copper atoms in the sterling lattice, Cu⊂2;O (cuprous oxide, in which copper is in the +1 oxidation state) precipitates within the metal. Cu⊂2;O is thermodynamically more stable than CuO at the oxygen partial pressures and temperatures found in the interior of the metal, which is why the sub-surface oxide forms as Cu⊂2;O rather than CuO.

The result is a two-layer oxide structure: an outer CuO layer (visible as black) on the metal surface, and a sub-surface Cu⊂2;O layer a few microns below the silver-rich outer surface. Pickle removes the outer CuO layer readily — the acid dissolves it and the metal emerges bright silver. But the sub-surface Cu⊂2;O layer is protected by a thin layer of silver-rich metal above it. Pickle cannot reach it. The silver emerges from pickle looking clean. It looks clean after light polishing. Only after polishing through the silver-rich outer layer and exposing the Cu⊂2;O beneath does the firescale become visible: a blue-grey, purple-grey, or brownish stain that was not apparent before polishing and cannot be removed by further pickling. The only remedies are abrasive removal (polishing or sanding past the depth of the firescale layer, which removes significant metal and changes dimensions), electrostripping (an electrochemical process that removes the firescale layer selectively), or prevention.

Firescale prevention workflow and per-session documentation

Firescale prevention works by limiting oxygen access to the silver surface during heating. The two main prevention methods are the denatured alcohol flux coat and atmosphere management in the flame.

The boric acid/denatured alcohol method: dissolve boric acid powder in denatured alcohol (approximately 1 tablespoon per cup of alcohol) to make a thin, fluid solution. Dip the entire piece in this solution before any heating begins — including before the first annealing, not just before soldering. The piece is then carefully ignited: the alcohol burns off in a brief, controlled flame that leaves behind a thin white coat of pure boric acid glass across the entire metal surface. This boric acid glass coat melts during subsequent heating and forms a thin protective layer that reduces atmospheric oxygen access to the copper in the sterling surface. It does not eliminate firescale formation entirely, but it substantially reduces it.

Flame atmosphere management: a reducing flame (fuel-rich, with excess unburned fuel in the flame envelope) contains very little free oxygen. The unburned fuel — propane or butane in the torch flame — consumes the available oxygen in the combustion zone, leaving the flame outer cone oxygen-depleted. Working in the reducing zone of the flame limits oxygen available at the metal surface and slows copper oxidation kinetics. An oxidizing flame (air-rich, with excess oxygen relative to fuel) accelerates copper oxidation and worsens firescale formation. The correct torch setup for sterling work is a slightly reducing flame: full fuel, less air than maximum, producing a soft inner cone and a slightly fuzzy outer envelope. Document per session: whether the boric acid coat was applied, the flame type used during heating (reducing, neutral, oxidizing), the estimated peak temperature reached, and the post-pickle surface appearance assessed before polishing. The systematic comparison of these variables across sessions determines which combination of prevention strategies minimizes firescale on the specific alloys and shapes in use.

AWG wire gauge selection at the wire-structure level

AWG numbering and actual millimeter diameters

AWG — American Wire Gauge — is an inverse-logarithmic scale in which higher gauge numbers correspond to thinner wire. The scale was historically defined by the number of wire drawing passes required to reduce a starting rod to the target diameter: more drawing passes produce a thinner wire and a higher gauge number. The practical result is that each step in the AWG sequence corresponds to a roughly fixed percentage reduction in diameter, producing a logarithmic diameter scale. The actual wire diameters at gauges commonly used in jewelry making:

18 AWG: 1.024 mm diameter. Heavy structural wire — bangles, heavy cuffs, structural frames in large wire-wrapped pendants, prong stock where substantial rigidity is needed. This gauge work-hardens slowly relative to thinner gauges because its large cross-section area contains a large total number of atoms and can accumulate more total dislocations before reaching critical density. 20 AWG: 0.812 mm diameter. The standard frame wire for most wire-wrapped cabochon pendants and smaller structural applications. It provides meaningful rigidity without being prohibitively stiff to form by hand. 22 AWG: 0.644 mm diameter. Medium frame wire, useful for smaller pendants and finer structures where 20 AWG would feel visually heavy. 24 AWG: 0.511 mm diameter. Light frame or coarser binding wire. Often used as a structural element in lightweight wire crochet or in fine wire-wrapped structures. 26 AWG: 0.405 mm diameter. The standard binding wire when 20 AWG is the frame gauge (six AWG steps above frame). Fine enough to coil tightly and invisibly between frame wires; substantial enough to hold coil tension without breaking during wrapping. 28 AWG: 0.321 mm diameter. Very fine binding, used with finer frame gauges or in applications where minimal wire visibility is the design intent. Prone to kinking and breaking if overworked; extremely sensitive to work hardening.

Frame-vs-binding gauge relationship and why it is a starting rule, not a fixed law

The conventional rule that binding wire should be six AWG steps finer than frame wire (20 AWG frame → 26 AWG binding; 22 AWG frame → 28 AWG binding) is a rule of thumb derived from visual balance: the binding wire should be fine enough to visually recede relative to the frame wires while being substantial enough to hold coil tension without distorting. It is not a physical law, and the actual visual outcome depends on coil spacing.

With the same 20 AWG frame and 26 AWG binding wire, tight-coiled binding (coils touching each other with no gap) creates a completely different visual density than spaced coils (coils separated by two wire-widths). The tighter the coil spacing, the more visual weight the binding wire carries, and the more it competes with the frame wires for visual dominance. A jeweler who makes tightly packed coils on a 20 AWG frame may find that 28 AWG binding provides better visual balance than 26 AWG, while a jeweler who uses widely spaced coils on the same frame may find that 24 AWG binding reads better. Document not just the gauge relationship but the coil spacing for each project, so that the documentation captures the actual design variables rather than just the material selection.

Work hardening accumulation rate by gauge and manipulation count documentation

The rate at which wire work-hardens per manipulation depends on the gauge because thinner wire has a smaller total cross-sectional area. With a smaller cross-section, each bending operation imposes a higher strain on a smaller total volume of material, generating more dislocations per unit volume per manipulation than the same bending operation on a thicker wire. A 20 AWG wire subjected to twelve 90-degree bends at a 1mm bending radius may still manipulate without kinking; the same number and severity of bends on 26 AWG wire may produce kinking and brittle behavior after eight bends because the smaller cross-section reached critical dislocation density sooner.

This creates a project-specific rather than universal annealing schedule: the 20 AWG wire from one manufacturer that can withstand twelve manipulations before needing annealing may require annealing after eight manipulations on the next project if the bending radii are tighter, because the higher strain per bend generates more dislocations per operation. The documentation that distinguishes useful process records from generic notes is the manipulation count at each anneal: note the number of bends or form operations performed since the last anneal or since the wire was removed from the fresh spool; note the observation of stiffness, springback, or the beginning of kink formation that triggered the annealing decision; note the gauge and the bending radius category (tight <1mm, medium 1–3mm, open >3mm). Across multiple projects using the same wire gauge and stock, this documentation reveals the manipulation threshold for that specific material from that specific supplier.

Liver of sulfur patina chemistry

Polysulfide composition and the silver sulfide reaction

Liver of sulfur is not a single chemical compound. It is potassium polysulfide — a mixture of several potassium sulfide species: K⊂2;S⊂2;, K⊂2;S⊂3;, K⊂2;S⊂4;, and K⊂2;S⊂5;, in proportions that vary between products and that change as the material ages and degrades. The “x” in K⊂2;Sx refers to the number of sulfur atoms in the polysulfide chain; longer chains (higher x) are generally more reactive. Solid chunk liver of sulfur degrades more slowly than gel or liquid formulations because exposure to air and moisture is limited by the chunk form; solution made from fresh chunks is typically more reactive than aged pre-dissolved solution.

In warm water, the polysulfide ions dissociate and the sulfide species react with the silver surface. The fundamental reaction: 2Ag + K⊂2;S → Ag⊂2;S + 2K⊃+;. Ag⊂2;S is silver sulfide, specifically the mineral acanthite (the thermodynamically stable silver sulfide polymorph at room temperature). Acanthite in bulk is black — it absorbs light across the visible spectrum with high efficiency. However, the color observed during the early stages of liver of sulfur patination is not Ag⊂2;S in bulk: it is a very thin Ag⊂2;S film over a reflective silver substrate, and the color of a thin film over a reflective surface is determined by thin-film optical interference, not by the bulk absorption properties of the film material.

Thin-film interference and the color progression

Thin-film optical interference occurs when light reflects from two surfaces: the top surface of the thin film and the substrate (silver metal) surface beneath the film. The two reflections travel different path lengths depending on the film thickness; when they recombine, they interfere constructively at wavelengths whose path-length difference is a half-integer multiple of their wavelength, and destructively at other wavelengths. The reflected color shifts as the film grows thicker, cycling through the visible spectrum as the constructive interference wavelength shifts.

The color progression in liver of sulfur patination of silver, from thinnest to thickest Ag⊂2;S layer: pale gold (film approximately 50–80 nm thick) is the first visible color, appearing in seconds to tens of seconds at moderate LOS concentration and temperature; bronze-brown (approximately 100–150 nm) follows as the film thickens; dark brown (approximately 150–200 nm) appears as the interference peak shifts further; iridescent blue-purple (approximately 200–250 nm) is a particularly sought-after stage where the interference produces a striking optical effect; and eventually near-black where the Ag⊂2;S layer is thick enough that bulk absorption dominates and the thin-film interference colors are washed out by the strong absorption of the thick sulfide layer.

Each color stage can be targeted by controlling the three reaction variables: water temperature, LOS concentration, and immersion time. A baseline for documentation: approximately 1/4 teaspoon of liver of sulfur chunks dissolved in 500 ml of warm water (60°C / 140°F). At this concentration and temperature, pale gold typically appears in 5–15 seconds of immersion, bronze-brown at 30–60 seconds, dark brown at 1–2 minutes, and iridescent stages at 2–4 minutes. Near-black requires continued immersion of 5–10 minutes or more at this concentration. Higher temperatures and higher concentrations accelerate the progression; lower temperatures and dilute solutions slow it. The iridescent stage is highly time-sensitive: moving from iridescent to near-black can happen in seconds at high concentration, making the iridescent stages easiest to capture in dilute, cooler solution where the reaction is slower.

Neutralization, selective polish, and differential patina on mixed-metal content

Removing the piece from the LOS solution does not stop the sulfidation reaction immediately: traces of polysulfide solution on the metal surface continue reacting after removal from the bath. Neutralization arrests the reaction by introducing an alkaline medium: transfer the piece to a baking soda solution (approximately 1 tablespoon of sodium bicarbonate, NaHCO⊂3;, per 500 ml of water) and allow it to soak briefly. The bicarbonate reacts with any remaining polysulfide ions and raises the surface pH, halting further Ag⊂2;S deposition. Rinse thoroughly with clean water after neutralization.

Selective polish after LOS patination is the step that produces the standard antiqued appearance: Ag⊂2;S is removed from raised surfaces and high points while being retained in recessed areas. 0000 steel wool applied with light pressure to raised surfaces burnishes away the soft Ag⊂2;S layer, exposing the bright silver beneath. Because the Ag⊂2;S in recessed areas is protected by geometry from the steel wool, it remains as a dark patina that accentuates texture and depth. The contrast between bright raised surfaces and dark recesses is the visual effect that LOS patination provides. Pro-Polish pads or polishing cloths can be used for lighter selective polishing on raised surfaces that cannot tolerate the abrasiveness of steel wool. Fine abrasive papers at 400–600 grit are useful for flat surfaces where precise selective removal is needed.

Copper within sterling silver produces a different color progression in LOS than the silver-rich surface produces. Copper sulfides include Cu⊂2;S (chalcocite, black) and CuS (covellite, indigo-black), and the copper sulfidation reaction proceeds at a different rate than the silver sulfidation reaction. In mixed-metal or copper-heavy areas of a sterling piece — such as areas where the copper has preferentially concentrated during heat work or where a copper inlay is used — the patina develops more rapidly and in a slightly different color tone, producing variation in the patina appearance. This variation is not a defect; it is an inherent property of the sterling alloy’s heterogeneous surface composition after heat work.

Argentium silver, because of the thin germanium oxide layer that inhibits firescale, also develops a more muted patina response to liver of sulfur than sterling silver. The GeO⊂2; surface layer partially inhibits sulfide ion access to the silver surface, slowing the Ag⊂2;S formation rate. Argentium LOS patination at the same concentration and temperature as sterling produces colors that develop more slowly and tend toward more muted, cooler tones rather than the warm gold and bronze tones that appear rapidly on fresh sterling. Document LOS patina results on argentium and sterling separately in process notes, noting the temperature and concentration used and the time required to reach each color stage, so that the documentation reflects the actual alloy-specific behavior rather than a generic LOS procedure.

Apple Tax for jewelry making Patreon creators

Jewelry making creators face iOS rates that are among the highest across all craft creator categories. The concentration of jewelry content on Instagram — a platform consumed almost entirely on mobile — and the visual, platform-native character of jewelry process content produces iOS subscription rates far above the craft category average. A metalsmithing creator whose audience has built primarily on YouTube sits at 50–65% iOS; an Instagram-primary jewelry creator with a secondary YouTube presence sits at 75–85% iOS. TikTok jewelry making content also produces 75–85% iOS. Metalsmithing with explicit educational content — tutorials, technique deep-dives — sits slightly lower at 55–65% iOS because educational content attracts a more desktop-research-oriented audience that is more likely to convert through a browser than through the iOS Patreon app.

The Apple Tax beginning November 1, 2026: Patreon applies Apple’s 30% in-app purchase fee to all subscriptions processed through the iOS Patreon app. At 75% iOS, a jewelry creator loses $0.225 of every dollar generated by iOS subscribers before Patreon’s platform fee is applied. These are not small rounding errors at any meaningful MRR level.

These are among the highest Apple Tax exposures of any craft creator category, driven by the Instagram concentration in jewelry audiences. The dollar amounts represent revenue already earned from patrons paying full subscription prices — the same patrons, the same amounts, with a different fraction arriving in the creator’s account depending on how the patron subscribed.

The fix: enable Patreon’s web-only billing toggle in Creator Settings before October 31, 2026. Update the Instagram bio link first, because Instagram is the source of the highest iOS traffic concentration in jewelry creator audiences. Update TikTok bio links, YouTube description links, and the YouTube channel About page. All links should point to the Patreon web URL rather than the app. Patrons who subscribe through a browser’s web checkout do not generate iOS-billed subscriptions and are unaffected by the Apple IAP fee. From Safari on iPhone, click each updated link and verify that the result is a web payment dialog, not an Apple IAP prompt, before November 1.

Use the KeepTier Apple Tax Calculator to run your specific MRR and iOS rate and see the monthly and annual exposure. KeepTier is a web-only membership page for creators who want 100% of their tier revenue minus only Stripe fees — no iOS IAP pathway and no platform percentage. Plans from $9/month.

Patreon for jewelry making creators — SEO guide (tiers, solder sequence, wire gauge basics, stone setting, Apple Tax table) · Patreon for blacksmithing creators · How the Apple Tax works · All explainers

Frequently asked questions

Related: Patreon for jewelry making creators (SEO guide) · Patreon for blacksmithing creators · Patreon for woodworking creators · How the Apple Tax works · All explainers