Downsides to salt pools

[chem geek]

I found the following graph or similar variations of it in several documents talking about corrosion.

Though it is hard to read, first note that this is about corrosion of iron (raw steel) and NOT stainless steel. Also, the scale on the X-axis says "Sodium Chloride, g/l" but that is wrong as the scale is actually Sodium Chloride g/100ml (or possibly %). The saturation level of Sodium Chloride is about 36g/100ml (which is 26.5% by weight or 265,000 ppm). The interesting thing to note is that the relative corrosion rate of iron increases to a peak around 1g/100ml (9,900 ppm) and then declines so that sea water at 3.5% (35,000 ppm) is actually less corrosive even though it has more salt. This is because the amount of dissolved oxygen in high salinity waters starts to decrease and dissolved oxygen is a larger factor in corrosion than salinity.

In fact, for iron, it would be inaccurate to say that sea water is much more corrosive than a salt pool just because it has over 10 times the salinity. The rate of corrosion for iron would actually be about the same. But for iron, almost any amount of salt is corrosive and iron still corrodes in even near-salt-free (though still wet with dissolved oxygen) environments, so you can only take this so far, but it is still interesting. [EDIT] By the way, cast iron corrosion isn't always fatal as some large commercial pool pumps use cast iron and though they do show corrosion, the rust (iron oxide) does seem to partially protect the pump from disintegrating -- it probably still corrodes, but is relatively slow given the thickness of the cast iron. Also, the relative corrosion rate only increases by 50% (factor of 1.5) going from a salt level of around 100 ppm to the peak at around 1g/100ml (9900 ppm) so this does not appear to be a huge difference, at least for iron. [END-EDIT]

The corrosion of stainless steel operates with a different mechanism since stainless steel normally has a passivity layer that protects it. This layer is thought to be (a possibly hydrated) very thin layer of chromium oxide (which is why stainless steel has chromium in it) which forms from the combination of the chromium in the stainless steel with oxygen. So there appears to be two competing reactions that occur -- oxidizers (such as hypochlorous acid) break down the passivity layer while oxygen combines with the chromium to form a new layer, but is inhibited in doing so by the presence of chloride ions (which apparently compete to form chromium chlorides). With two competing mechanisms, corrosion would appear to have something of a "threshold" where these rates were equal. Nevertheless, it would be inaccurate to say that corrosion does not occur below the threshold -- it still does, but at a slower rate (and would be less visible due to the "healing" of the stainless steel forming a new passivity layer).

So the study referred to in this post (above) suggested that corrosion from salinity had a threshold around 6000 ppm. There was no indication (with these electrolytic cell tests) as to the chlorine levels nor to how long the test lasted nor what materials were being tested for corrosion. It may be that the threshold is somewhere near 6000 ppm, but it could also be that at 3000 ppm the stainless steel corrodes or at least gets thinner over 2-3 years compared to a pool with 500 ppm salt that may have the stainless steel last 10-20 years.

Corrosion of grout, stone, cement and other coping and hardscape surfaces is a totally different matter. This appears to occur from repeated wetting and evaporation to concentrate salt. Some processes appear to dissolve or pit the stone or to crack it while others are more mechanical with forces having to do with crystallization. It seems that these processes may be more linear and not have the "threshold" effect as shown by stainless steel (or other metals with passivity layers). As has been reported by some users seeing corrosion of limestone in non-salt pools, the quality of the stone and its maintenance (sealing, rinsing) is a big factor in its ability to resist corrosion. So even non-SWG pools that have a minimum of 300 ppm salt and usually more may still be corrosive to grout/stone/cement, but that with one-fifth to one-tenth the amount of salt, it may take 5-10 times longer for the equivalent corrosion to occur. I don't have definitive data about this so this is clearly speculative.

So for me, at least, I am not convinced that salt pools are not more corrosive than non-salt pools to the degree that perhaps recommendations should be made regarding higher quality materials to be used (Type 316L stainless steel; cupro-nickel heat exchangers; sealed stone or concrete that is more resistant to salt). Also, for all pools, the chlorine levels need to be watched and even indoor pools may need to have some CYA (10-20 ppm). That is not to say that there aren't pools with SWG systems out there (using CYA) that show no signs of (common Type 304) stainless steel corrosion in the first year or two -- it's the longer term (when using less salt-resistant materials) that I'm still not sure about. The jury is still out, I need more data, this does not compute, does not compute, compute...

Richard

[EDIT]
P.S.

I communicated with a pool installer from Australia who indicated that they see very little corrosion in their pools (SWG or non-SWG), but nearly all of their installations use the marine-grade 316 stainless steel, the paving is limestone or clay that is fired and that the supplier warrants for use in salt pools or it is coated to make it salt/sulfate resistant. They normally install heat pumps with titanium coils, rather than gas heaters (and such copper and cupro-nickel heaters do show corrosion). They also typically see higher TDS due to higher salt and sulfates in their tap water so even non-SWG pools have high salt levels, especially after seasons of chlorinating liquid usage.
[END-EDIT]