The Tarnished Silver Skin That Slowly Robs a Cavity of the Sharpness It Was Built For

A new cavity resonator is a beautiful thing on the bench. Freshly silver-plated, it rings with a quality factor that lets a filter pass exactly what it should and reject everything else, and the builder admires the steep skirts on the analyzer with satisfaction. Months later, sometimes only weeks in a humid or industrial environment, the same cavity has gone subtly soft. The skirts are less steep, the insertion loss has crept up, the resonance is a touch broader. Nothing was touched, nothing was knocked, no solder joint failed. What happened lives at the surface, in a film of tarnish too thin to see clearly, growing on the silver that was supposed to be the lowest-loss skin available. The slow oxidation and sulfidation of a silver plating is one of the quietest ways a high-performance resonator degrades, and because it hides on a surface the current never lets the builder forget, it deserves a clear accounting.

The puzzle for many builders is why a surface film microns thick or less should matter at all to a cavity whose walls are solid metal millimeters thick. The answer is the skin effect, the single fact that governs everything about cavity loss. Radio-frequency current does not flow through the bulk of a conductor. It crowds into an extraordinarily thin layer at the surface, and that layer is exactly where tarnish forms. The current and the corrosion occupy the same few microns, so a film that would be electrically irrelevant at DC sits squarely in the path of every ampere the cavity carries.

Why all the current rides in a skin thinner than the tarnish

The skin effect concentrates alternating current near the surface of a conductor, with the current density falling off exponentially into the metal. The characteristic depth at which the current has fallen to about 37 percent of its surface value is the skin depth, given by

delta = sqrt(2 rho / (omega mu))

where rho is the resistivity, omega is the angular frequency, and mu is the permeability. Equivalently, in practical form,

delta = 1 / sqrt(pi f mu * sigma)

with sigma the conductivity. The defining feature is that skin depth shrinks as the square root of frequency, so the higher the frequency, the thinner the conducting layer.

Put numbers to silver at a few frequencies to feel the scale. Silver has a conductivity of about 6.3 times ten to the seventh siemens per meter. At 100 MHz the skin depth works out to roughly

delta = 1 / sqrt(pi 1e8 4pi1e-7 * 6.3e7) = about 6.3 micrometers

At 1 GHz it falls to about 2 micrometers, and at 10 GHz to about 0.64 micrometers. So at microwave frequencies essentially all the current flows within a layer one or two microns thick. A tarnish film of comparable thickness is therefore not a trivial surface blemish, it is a substantial fraction of the entire conducting cross-section, and its much higher resistivity forces the current to struggle through poor material exactly where the current density is highest.

How surface resistance sets the quality factor

The quantity that ties surface condition to cavity performance is the surface resistance, the effective resistance per square of the conducting skin. It follows from the skin depth as

Rs = 1 / (sigma delta) = sqrt(pi f * mu / sigma)

and it rises as the square root of frequency. The unloaded quality factor of a cavity is the ratio of energy stored to energy lost per cycle, and the loss is dominated by the surface resistance of the walls. To good approximation the quality factor is

Q = G / Rs

where G is a geometric factor set by the cavity shape and the mode, independent of the wall material. Everything about the material enters through Rs. This relation is the crux: anything that raises the surface resistance lowers the quality factor in direct proportion. A film that doubles the effective surface resistance halves the quality factor.

This is also why silver is the plating of choice in the first place. Of all common metals silver has the lowest resistivity, slightly lower than copper, so it yields the lowest surface resistance and the highest quality factor. But the advantage is smaller than intuition suggests, because the resistance depends on the square root of resistivity through the skin depth, not on resistivity directly. The practical observation from resonator builders bears this out: replacing copper walls with silver typically lowers the surface resistance by only about four percent, improving the quality factor by a similar four percent. Silver is chosen not for a dramatic conductivity edge but because that modest edge is free once you are plating anyway, and crucially because of what happens to the alternative when it corrodes.

The semiconducting film that makes corrosion worse than dilution

Here the story turns from simple dilution to something nastier. If tarnish were merely a thin layer of slightly worse conductor, the loss penalty would be modest and easy to estimate. The real problem is the electrical nature of the corrosion products. Silver sulfide, the dark tarnish that forms when silver meets trace hydrogen sulfide in the air, is not an insulator and not a good conductor either. It is a semiconductor. Copper's tarnish behaves similarly, and resonator builders note explicitly that the tarnish on copper can be semiconducting, which is one of the genuine long-term arguments for silver despite the small conductivity difference.

A semiconducting film in the skin-depth region is far more damaging than its thickness alone would predict, because its resistivity is orders of magnitude higher than the metal beneath it, and the current is forced to flow partly through this lossy material. The loss is concentrated precisely where the field is strongest, at the surface, so a semiconducting tarnish film raises the effective surface resistance disproportionately. Worse, semiconducting films can introduce loss mechanisms beyond simple resistance, including the dielectric-like losses associated with the film acting as a thin lossy layer between the field and the good conductor. The upshot is that a tarnish film does not merely add a little series resistance, it can degrade the quality factor by a fraction much larger than the film's thickness ratio would suggest.

A numerical estimate of the quality factor a tarnish film steals

Walk through the arithmetic for a concrete case. Take a cavity resonating at 1 GHz with clean silver walls giving an unloaded quality factor of 12000, a typical value for a well-made copper-or-silver cavity at that frequency. The skin depth in silver at 1 GHz is about 2 micrometers, and the surface resistance is about 8.2 milliohms per square.

Now suppose a silver sulfide film of 0.5 micrometers grows on the surface, a quarter of the skin depth. The film's resistivity is far higher than silver's, perhaps a thousand times or more, so within that 0.5 micrometer layer the current meets dramatically worse conductor. A simple way to bound the effect is to treat the tarnished surface as adding an extra series resistance proportional to the fraction of the skin occupied by the bad material weighted by its resistivity ratio. Even a conservative model in which the film raises the effective surface resistance by 30 percent gives a new surface resistance of

Rs_new = 8.2 * 1.30 = 10.7 milliohms per square

and since the quality factor is inversely proportional to surface resistance,

Q_new = 12000 * (8.2 / 10.7) = 9200

The quality factor has fallen from 12000 to 9200, a loss of nearly a quarter, from a tarnish film a quarter-micron thick that the eye registers only as a slight darkening. Translate that into filter performance and the insertion loss of a multi-pole filter, which scales inversely with quality factor, rises noticeably, while the skirts soften. The cavity has not been damaged in any mechanical sense, yet it now performs as a markedly lesser instrument, all because a film thinner than a wavelength of light grew where the current cannot avoid it.

How fast the film grows and how to catch it

The degradation has a time course, and understanding it tells the builder how long a cavity will hold its performance. Tarnish growth on silver is driven by the partial pressure of corrosive gases, chiefly hydrogen sulfide, and proceeds by a diffusion-limited law once an initial film has formed. The film thickness x grows not linearly but as the square root of time, because the reacting species must diffuse through the film already present to reach fresh silver beneath:

x = sqrt(k * t)

where k is a rate constant that rises steeply with humidity and the concentration of sulfur-bearing gases. This square-root law has a practical consequence: tarnish grows fast at first and then slows, so a cavity loses quality factor most rapidly in its early life and then more gradually. A film that reaches 0.2 micrometers in the first month, in the same environment, reaches not 0.8 but only about 0.4 micrometers by the fourth month, since quadrupling the time only doubles the thickness.

Put the rate to a forecast. If a cavity in a benign indoor environment grows tarnish at a rate constant such that the film reaches 0.1 micrometers in 30 days, then the thickness at later times follows from the square-root law:

x(90 days) = 0.1 sqrt(90/30) = 0.1 1.73 = 0.17 micrometers
x(365 days) = 0.1 sqrt(365/30) = 0.1 3.49 = 0.35 micrometers

so after a year the film approaches the quarter-micron level that the earlier calculation showed costs a quarter of the quality factor. In an industrial or coastal atmosphere with far higher sulfur loading the rate constant can be ten or a hundred times larger, collapsing that timeline from a year to weeks.

Catching the degradation is a matter of periodic measurement rather than visual inspection, because the eye sees gross discoloration long after the electrical penalty has begun. The diagnostic is the unloaded quality factor itself, measured from the resonance bandwidth. The quality factor relates to the resonant frequency and the half-power bandwidth by

Q = f0 / BW

so the builder sweeps the cavity, reads the center frequency and the 3 dB bandwidth, and computes the quality factor. A clean baseline taken when the cavity is new, compared against periodic remeasurement, reveals the slow climb of the bandwidth long before any filter specification is violated. A 10 percent broadening of the bandwidth is a 10 percent loss of quality factor, an unambiguous early warning that the surface is tarnishing, and one that no amount of looking at the metal would reveal as reliably.

The remedies and the reason gold sometimes wins

The defenses against this slow degradation follow directly from the mechanism. Since the trouble is a reactive surface meeting a corrosive atmosphere, one approach is to seal the surface. A thin lacquer over the silver keeps the sulfur-bearing air away and is cheap to apply, and resonator builders note it as a practical alternative to fighting the tarnish chemically. The drawback is that any coating in the skin-depth region is itself a thin dielectric layer carrying field, so the lacquer must be thin and low-loss enough not to undo the benefit it provides.

The more durable answer used in demanding work is a noble overplate. Gold does not tarnish, so a thin gold flash over the silver preserves the low surface resistance indefinitely. The laboratory practice for high-quality-factor cavities makes this explicit: cavities are built of oxygen-free copper, plated first with silver and then with gold specifically to prevent oxidation of the metal surfaces and thereby minimize additional resistive losses. The silver provides the low resistance, the gold provides the permanence, and the combination defeats the slow tarnish that bare silver cannot escape. A more exotic route protects the surface with a passivating layer such as graphene, which has been shown to prevent the active growth of the oxide layer and block impurity deposition while adding negligible loss of its own.

The deeper lesson for any builder of resonant structures is that the surface is the circuit. At radio frequencies the bulk metal is almost a bystander; the few microns at the surface carry the current, store the loss, and set the quality factor. A resonator is only as good as the condition of that skin, and the skin is exactly the part exposed to the air. Treating the plated surface as a permanent given is the mistake. It is a living surface, slowly reacting with its environment, and the sharpness designed into a cavity on the bench survives only as long as that surface stays clean. The builder who understands the skin effect understands why a film too thin to photograph can quietly steal a quarter of the quality factor, and why the best cavities are the ones whose surfaces were protected before the air ever reached them.