Chlorine output from SWG

[chem geek]

Mark,

Great experimenting (again)! waterbear also asked about water shielding UV but in the range where most of the destruction occurs, mostly for hypochlorite ion around 270-320 nm (see this link) and the low absorption of water in this same range (see this link), it would seem from your experimental results that the protection from "shielding" is hugely significant and is a predominant factor and this shielding is from CYA and possibly also from the chlorinated CYA species. Deep water looks blue because most light except blue is absorbed and some of the blue gets scattered back.

Now it is possible that the other components of pool water, such as the carbonates, provide some shielding, but this link showed a modest shielding effect in waters with lower Dissolved Organic Carbon (DOC) of 0.53 ppm Carbon and that's organics, not carbonates (the water would already have far higher than 0.5 ppm carbon as carbonate just from dissolved carbon dioxide). The implied absorption coefficient is from 0.006 to 0.01 while the pure water number is around 0.0001 or 100 times less. So various organics in the water can clearly have a shielding affect (CYA itself is an organic, after all, but a very powerful one for shielding UV).

By the way, your experiment with different pH for SWG production is interesting and wasn't what I was asking for, but I'm glad you did it since that is also good to know and is a little unexpected. There is a variation in the reactions with pH, but the relatively high (for electrochemistry) voltages in the SWG would dwarf such differences and also explain why my original theory of CYA protection in the cell is not the gating item. Diffusion might be a more important factor. I suspect that there might be a difference in SWG efficiency with flow rate -- faster speed being more efficient, somewhat similar to what is seen with solar panels though the mechanism is different. (Yes, I know, I should just shut up with these experiment ideas...).

What I was really asking regarding pH was the chlorine lifetime where I believe that the chlorine will last longer at lower pH than at higher pH, all else equal, due to the difference in hypochlorous acid vs. hypochlorite ion concentrations (the former seems to be more resistant to breakdown from UV than the latter, based on the link I gave above).

If you could measure the chlorine drop in the spa and the bucket a little sooner -- before the bucket chlorine went too low too quickly, that would be better. Also (yes, another experiment), if you have a clean bucket you can use for tap water without CYA in it (though you could add a small amount of baking soda and calcium chloride to simulate a pool environment) and compare that with a similar sized bucket with your pool water that has CYA in it, then that would tell us the protection from CYA due to creating different chemical species as opposed to the shielding effect (it would actually be a little of a mix since the bucket has depth, but I can sort that out if I know the depth of the bucket).

Just so you know, I'm going on vacation for the next two weeks so may not be able to work on any calculations until I get back. I think we've got a good handle on what experiments we are looking for, however, and already have a strong indication that the shielding effect is a strong one second only to the species effect (i.e. the disinfecting chlorine concentration being much, much smaller in the presence of CYA). With no CYA, the half-life of chlorine (from one of the links above) seems to be around 10 minutes near the surface (in Florida direct noontime sun).

Richard

 

[cliff_s]

I did measure some time ago the chlorine levels at the various depths after the water sitting in sunlight for 6 hours.
I regret not recording exactly the readings, but from memory the chlorine level did increase with depth.
I always make it a practice to let the filter pump run at least 20 min. before taking any readings. This is to get a
average reading. Another practice is to always run the filter pump with the SWG running whenever the pool
is in use. This is to compensate for the bathers, also skims the junk off the surface of the pool.

I normally have about 45 ppm CYA and do find almost a 50% loss of chlorine during daylight hours. To me this loss
is a compromise between chlorine effectiveness and CYA level. To me the object is to increase the chlorine
effectiveness and not just to make some presupposed the test kit reading.

As far as the Oxychlorine generation the article was about using these for wound disinfection and healing.
According to the article the Oxychlorine would not attack bacteria that humans normally have because of the
the size of the bacteria, but was very effective on the smaller viruses and bacteria. I expect it is a subject
for further study.

I have a observed a increase in chlorine output at lower pH levels. I would suspect the the pool water conductivity
varies with pH. Since most of the SWG units use a regulated power supply, just measuring current to the cell
might not mean very much. The units that do have a built in salt level indicator are probably just using cell current
to derive the salt reading. Some units do adjust the voltage levels with temperature. I have not seen any measurements
on chlorine production verses water flow velocity. On most pools the cell has all the filter water plumbed through it
so the water velocity is quite high. This requires a large pump(1 hp+) to be run many hours, with electricity constantly
climbing in cost becomes a factor. I would be curious as to the amount of chlorine generated and various water flow
levels. It would seem as though a small pump(about 1/4hp) could make a large savings of electricity if it could generate
the same amount of chlorine. I have thought of trying this, but I have not found any small circulating pumps that are self
priming. In fact I have one(not self-priming) sitting on the shelf in the garage.

The ORP systems do measure water conductivity to obtain their readings. I would suspect the sodium being
a conductive metal would tend to increase conductivity. I haven't measured ORP myself so I really don't know what effect the
various parameters have on the ORP readings. This has been on my list to do for some time.

My suspicions are that with a salt pool the chlorine levels don't need to be as high as recommended to provide algae
prevention and proper sanitation. I am not convinced that conventional water parameters are necessary and what parameters
are ideal.

As far as calcium levels, I lived in Oregon for many years and the water there has very little calcium and the plaster pools
I know of held up just fine. In that area the swimming season was about 3 months max, so most concrete pools are drained during
the winter, this is to prevent freezing.

Calsaway is not cheap, the average pool is $400. Although if this was done a fill time and the calcium reduced it would sure
pay for itself in the saving on tile cleaning. Since Pool services want about $5.00 a linear ft. to glass blast the tile, plus a water refill, the price
is reasonable.

As various people do the tests we all learn more and more of what happens during chlorine generation by electrolysis
and how to increase the efficiency of our systems.


Cliff s

 

[chem geek]

Cliff,

Thanks for the info. It's nice to know we're on the right track to sorting this out (re: chlorine loss vs. depth). I have some comments to a couple of things you said as follows:

The ORP systems do measure water conductivity to obtain their readings. I would suspect the sodium being
a conductive metal would tend to increase conductivity. I haven't measured ORP myself so I really don't know what effect the
various parameters have on the ORP readings. This has been on my list to do for some time.

The ORP systems don't measure conductivity, but rather measure the oxidation reduction potential of the water (relative to a reference standard that's around 250 mV if memory serves correctly -- 0 mV is the hydrogen ion to gas equation at unimolarity and one atmosphere pressure). In chlorine pools the ORP is overwhelmingly from the hypochlorous acid (HOCl) so the ORP reading can be used as a proxy for that level, though it is affected by pH more than the HOCl concentration is. Also, different ORP sensors get different results, but for a given sensor one can generally calibrate it to an FC level (at a fixed CYA level) and use the ORP to try and keep FC constant. Depending on the sophistication of the system, this can work (if the system doesn't react too quickly and "overshoot", for example). There's more about ORP at this post of mine that I copied over from the Pool Forum.

My suspicions are that with a salt pool the chlorine levels don't need to be as high as recommended to provide algae
prevention and proper sanitation. I am not convinced that conventional water parameters are necessary and what parameters are ideal.

If by salt pool you mean the SWG itself, I agree. I don't think the salt level itself is doing much to inhibit algae, but the SWG does seem to super-chlorinate with high chlorine levels and low pH near the chlorine generating plate and though only a small fraction of the water is exposed to that, with multiple turnovers this does seem to allow for lower chlorine levels in most SWG pools. There have been some exceptions reported, but generally it's true. It probably depends on whether biofilms form on pool surfaces since clearly the super-chlorination in the SWG has no effect on algae that is not free-floating. Probably regular brushing in an SWG pool would ensure that lower chlorine levels could be used.

I have a observed a increase in chlorine output at lower pH levels. I would suspect the the pool water conductivity
varies with pH.

Actually, the conductivity does not vary much at all with pH (at least from 7.0 to 8.0). The bulk of the conductivity is due to the sodium and chloride ions that make up the bulk of the TDS. However, the hydrogen gas production is pH dependent and is more likely at lower pH (since it's hydrogen ions that get converted to hydrogen gas), but since the voltages in the SWG are so high I didn't think this small difference in electrochemical voltage would make a difference. It may have more to do with the rate of diffusion of the hydrogen ion at different concentrations since that would affect rates more, but I really don't know.

Richard

 

[chem geek]

Mark,

I'm back from vacation and did have some time (on planes) to put together a spreadsheet with various extinction/absorption coefficient data, solar spectrum in the UV range, etc. With this I was able to "predict" a 35 minute half-life of chlorine at a pH of 7.5 near the water's surface (i.e. no extinction effects). This isn't that close to the 10 minutes that several studies have shown and the solar spectrum I used was in the same Florida area as one of those studies, but it's close enough to try modeling what goes on with different pool depths at different chlorine levels. I'll have to guess what the wavelength dependency is for CYA and will have to estimate the extinction coefficient based on your data you've presented so far.

Richard

 

[mas985]

Richard,

I was able to repeat several of the measurements and here are a few conclusions.

Without a doubt, chlorine production is NOT dependent on CYA. I did tests at CYA of 45,60,80,100 and all had the same chlorine production within my measurement capabilities. Production seemed to have a dependency on PH, about of 15% decline from PH 7.5 to 8.0.

Extinction is heavily dependent on CYA levels. However, it was difficult to get consistent results for these tests because they were heavily dependent on weather and organics in the spa. I could not prevent things from getting in there so I am not sure how much of the chlorine was lost to the sun vs other. This was probably less of an issue at CYA of 45 ppm since most of the loss was due to the sun and I could easily test in 24 hours.

Anyway at CYA of 45 ppm, I lost 50% of the chlorine each day and those test seem to be somewhat consistent. More so than CYA of 80 ppm. The bucket next to the spa showed about the same reduction over the day. The first time I did this the bucket chlorine disapeared completely during the day but I think it must have been dirty or had something in it since I have repeated this test twice and the bucket and spa seemed to lose about the same chlorine at the same time. I did a side by side test, two test kits, and the bucket chlorine was a hair less than the spa but not enough to quantify. Also, I did the test at a PH of 8.0 without any difference in extinction results although production did drop 15%.

For 80 ppm, the first test was 15% lost each day and my second test showed a drop of 50% over four days or about the same 15% per day. The bucket test was again too close to call although I think the bucket was slightly less than the spa. I ran another test recently at 80 ppm but the chlorine dropped from 5.0 to 1.5 over three days which is more than 15% per day but the spa was very dirty over those three days. So I would throw these results out.

I got a FAS-DPD test kit so I am going to try it at much higher CL levels so that organics will not be of much concern and to get a little more precise on my measurements so I can see a decline in 24 hours for CYA 80 ppm. I think I finally have a good process down for both production and extinction although the later is still dependent on weather and organics.

One other thing I noticed was over the course of each test the PH of the spa would rise substantially while the bucket PH stayed about the same. It is a plaster pool so my first thought was that was accounting for the PH rise. What are your thoughts?

So I am thinking that CYA may have more retention value then what has been quoted in literature. It would be nice to prove it though with some theory.

 

[chem geek]

Mark,

So at 45 ppm CYA you lose 50% of the chlorine while at 80 ppm you lose 15% over the same period. The rate is -LN(fraction remaining)/time so at 45 ppm it's 0.69 per unit time while at 80 ppm it's 0.16 per unit time. That's a factor of 4.3 in improvement with a CYA ratio of 80/45 = 1.8 so more than makes up for the fact that you need to keep a higher chlorine level at the higher CYA level.

Now the thing that does not make sense is why the bucket is losing chlorine at the same rate as the deeper spa. Some of that might be due to the shape differences with the sun not exposing the bucket as much, but I doubt that's the reason. If extinction, or CYA absorption of UV rays, were what was happening, then the deeper spa should have had longer lasting chlorine than the bucket. If the protection is independent of depth, then something else is going on and that's very strange. If the disinfecting chlorine level were lower than calculated, then that would explain it, but that's inconsistent with other data we have (such as Ben's table and scientific studies and equilibrium constants). If the CYA protection were very strong over short distances (i.e. its molar extinction coefficient was very high so only the first few inches of depth matters), then that would explain what you are seeing but would also lead to far longer lasting chlorine than seen.

As for the pH rise, if you used your SWG to maintain the chlorine level in your pool, then your pool's pH would rise and the mixing of the pool water with the spa water when you "reset" back to 45 ppm CYA would cause the spa's pH to rise. That's my best guess. Yes, your plaster can cause the pH to rise, but only if it's curing, but I thought your pool isn't that new, is it (i.e. less than a year)?

Richard

 

[mas985]

The pool is close to two years old so I wouldn't expect it to still be curing but I have heard of long term PH rise due to plaster pools.

The PH rise in the spa was during the time I was measuring chlorine loss so the water was just sitting. No circulation or anything else. The only thing was maybe a bit of wind.

 

[chem geek]

Well, the pH rise could just be due to normal outgassing though usually it's not that fast unless there is a source of aeration so having the pump off with still water should normally be more stable. It also depends on the TA and pH. When you tested at lower pH, that would have had more of a tendency to rise, but that doesn't explain what happened at normal (7.5) pH.

I figured out a way that could explain the data if I break one of the other assumptions that the industry had with regard to the chlorine half-life chart vs. CYA. If the chlorine attached to CYA is less stable than implied in the industry graph but CYA has a higher absorption itself, then that could explain what you see. If this is true, then a quite shallow surface should lose chlorine more rapidly in the sun -- say, a Pyrex glass dish. Perhaps the bulk of the "shielding" occurs in the first foot so that is why the bucket didn't show a much larger loss. At least this is something else to try, but unfortunately adds some other variables such as a high surface area to volume ratio so losses could also occur from outgassing of chlorine itself (but that would also occur without sun) though that's normally not very much except at higher temperatures (such as in a spa).

Richard