I'll help W5WSS out here. He's not the most articulate sometimes when he posts so let me give it a try:
So, if copper becomes contaminated and the new surface of the wire becomes some compound of copper, doesn't that mean that the new conductive surface is just below the copper compound? Do copper compounds act like a Faraday cage and keep the RF from reaching the new surface of the copper?
First question:Yes. Second:No. Faraday cage is not applicable here and I don't believe he suggested it was.
For example, you say corrosion impedes RF allot(sp). How much? 1%? 50%? 90%? Without providing some actual figures your words are nothing more than a guess that something is happening.
Depends on the level and type of corrosion. Corrosion reduces the cross sectional area of copper conductors, thereby reducing conductivity. Corrosion also increases the surface area of the conductive copper that remains. This corroded surface area is very large (roughness is very high),depending on the corrosion mechanism, relative to new copper. Increased surface roughness/area alone has been shown to increase the resistance to RF as much as 25% at 10 MHz in just one study that did not manage to reproduce the high surface area due to corrosion by just sulfate alone. That was on a log scale so 30 MHz would be a wee bit higher percentagewise. You can find various electrical resistivity measurement devices described on the internet that measure corrosion rates in the chemical industry based on these concepts including resistance vs corrosion rate values for the probes that are used as well as the study I just mentioned, conducted for the US Government.
For any given diameter of conductor, there is a frequency where the skin effect makes it into a poor conductor. You are saying that if part of the skin of a wire is no longer a good conductor that it limits the ability of that wire to carry RF. Yet, you say that if you put a perfect insulator around it (plastic insulation), that is ok.
How does that work?
I disagree that skin effect has that effect anywhere near the HF frequencies we are talking about and there is no "skin of the wire" but instead a layer of non-conductive corrosion. Skin effect is an introduction of eddy currents at higher frequencies that increases the resistance to flow and is frequency dependent. We don't want to confuse the two. The answer to your second question, is that the non-corroded wire remains intact and maintains its integrity without increased resistance.
The point I believe that WSS was making, is that it may be better to use a corrosion resistant insulated ground radial system to start with, when putting down 30-40 radials, than using bare copper that when buried is subject to attack by chemicals that exist naturally in most soils, bacterial action such as sulfate reducing bacteria, alkalinity, and oxygen corrosion cells that may be present. Unlike oxide corrosion that most hams are familiar with and that provides a protective coating after a while, these corrosion mechanisms are the gifts that keep on giving (more corrosion). Its not a sure bet that resistance of bare copper wire radials will be maintained for a long time at desirable levels. As it pertains to commercial broadcast systems, there are several instances that are mentioned, on the internet, where commercial interests replaced their grounding systems or a few radials due to corrosion. You may look to NIST for reports about WWV replacement of ground radials for that reason.
Others will disagree with these statements which are debated here on eham ad nauseum, so eventually you have to make those decisions yourself. The use of insulated ground radials for amateur use is my position and I believe it is WSS's position as well as SteppIR, DX Engineering, and others (available on the net too).