Swimming Pool Chemistry

[Helios_Hyperion]

Hi All,

First off, if there are posts that I have missed, i.e. questions of mine that have been answered, please excuse me, I am trying to formulate my problem / plan / scumbag brain as best I can, and just delving into posts can confuse me a bit if I don't know how the other persons mind works

So, currently I am busy with a thesis in which I am trying to combine all aspects of a swimming pool into one mathematical model...

The more I researched the topic, the more complex and in depth it became (as is the way of our world), but I have found so many conflicting ideas and advice, that I am currently at a loss: I think the easiest would be to link my introduction as it is now, and then you will have a better understanding of what I am trying to do (please ask questions). (please see attachment)

----------------------------------------------------------------------------------------------------------------------------

1. Current Place in Space and Time

I have reached the point where the graphs done by chem_geek here: http://www.troublefreepool.com/pool-water-chemistry-t628.html#p4366 are starting to become applicable, but I have no idea how chem_geek set them up...

In the graphs, and as I understand, factors like pH, Temp, and TDS in these graphs are a constant, but we all know that these factors are constantly changing and evolving on a /sec basis. @chem_geek: I have studied quite a few articles out there, but I have yet to find something that gives me a clear definition / idea of what is exactly going on... on what "facts" / literature did you base your calculations? I say "facts", since as far as I understand, a lot of the papers are based on experimental data, but no one has actually sat down and tried to define a proper model for these reactions.

----------------------------------------------------------------------------------------------------------------------------

2. Choices, choices, choices... can't it just be the red pill?

Then there is the matter of all the products to produce disinfectant (HClO) out there:

• Cl₂ (which of course is not used anymore, but a lot of literature still use this as the base for chlorination)
• NaOCL (for salt pools?)
• Ca(OCL)₂ (this will obviously influence Calcium Hardness, L.S.I., but Ca²⁺ is needed to combat grout deterioration?)
• (C-NClO)₃

Since the thesis is only the first step (my Masters will be in Chemical Eng, and then the model will be expanded), I made a choice in using (C-NClO)₃ since this will form the disinfectant HClO as well as add CYA (C-NOH) to the system in order to absorb U.V. .

1. (C-NClO)₃ + H₂O -> 3HClO + (C-NOH)₃

Regarding U.V. absorption factors, I still have not been able to find any "proper" literature on the rate of decomposition that the 290 - 350 nm (please correct me if I am ever mistaken) bandwidth radiation has on the HOCL and OCL^-, or even the C-NOH for that matter.

----------------------------------------------------------------------------------------------------------------------------

3. Apparently matter can exist in 4 states, if given the correct conditions: http://www.sciencealert.com/scientists-have-discovered-a-new-state-of-matter-the-jahn-teller-effect

The next part I have arrived at, is the formation of chloramines as the HClO reacts with nitrogen based compounds; such as the blood, sweat and tears of the average frustrated pool owner:

1. NH₃ + HClO -> NH₂Cl + H₂O
2. NH₂Cl + HClO -> NHCl₂ + H₂O
3. NHCl₂ + HClO -> NCl₃ + H₂O

The literature I came across, but don't yet fully understand, concludes that the formation of these chloramines (especially the unwelcome NCl₃) is closely related to pH. This makes sense yes, since HClO is more readily available at a lower pH, but what confuses me is this graph: http://www.lenntech.com/processes/disinfection/chemical/disinfectants-chloramines.htm, originally taken from (Palin, 1950). As I understand graphs, there should be no NCl₃ at the desired pH of 7.6... but in all the articles, they refer to the formation of NCl₃ as being a clear and present danger... maybe Chemical graphs are different from mathematical graphs

----------------------------------------------------------------------------------------------------------------------------

Right, so now maybe to just summarise my questions:

1. How were the HClO vs. TFC graphs constructed?
   1. What about Temp, TDS, pH changes?
2. Any information on the energy absorption characteristics of (C-NOH)₃?
3. How do I calculate the state of the chloramines? How does Shocking influence this?
4. Anything else that someone can think of that I am missing

----------------------------------------------------------------------------------------------------------------------------

A word or two or twenty:

Regarding Q. 2: I have found a reliable model to predict the incidence angle of radiation from the Sun. It has been published by NREL: Solar Position Algorithm for Solar Radiation Applications. Since I know the angle of incidence, I can integrate over the bandwidth in question, which then gives me a very clear picture of the exact amount of radiation introduced to the pool. So I am sure I can calculate the rate of reaction from this... I just need the missing link.

Regarding Q. 3: I understand that it forces a reverse reaction which in turn breaks down the chloramines, but does this then escape the system as gas? Or are my chloramines locked in the system? In other words, won't they just reform?

Well, I think this might be a good start... obviously there are an insane amount of variables, but I always say: "Ever wondered how to many possibilities there are to a 64-bit password? No Problem, only Solution."

Regards,
Helios_Hyperion

- - - Updated - - -

Don't know if it attached my attachment

 

[chem geek]

Most of the graphs that I have done that relate to chlorine and Cyanuric Acid have used the Pool Equations spreadsheet I wrote based on the equilibrium constants in the O'Brien paper from 1974 plus other information from other sources. Note that I have the temperature dependence of the chlorine/CYA relationship turned off by default. That's near line 230 that says "Use Temp. Dependent Cl-CYA" in columns B and C where you can change "FALSE" to "TRUE". I generally keep that off because it's more conservative and because it's based on one paper from Wojtowicz where the data might not be quite right. His info is generally OK, but sometimes is different than what we've seen and it's peer reviewed through JSPSI and not through a larger journal.

As for the effects of different chlorine chemicals on pool water chemistry and for other useful information see Certified Pool Operator (CPO) training -- What is not taught where you will see how much CYA is added by stabilized chlorine (Trichlor and Dichlor), how much CH is added by Cal-Hypo, how much salt is added by each, and how much the TA is dropped by each after accounting for chlorine usage/consumption. There are other relevant posts in the Pool Water Chemistry thread you linked to including this one that shows how with CYA in the water the pH doesn't affect the active chlorine (hypochlorous acid) level as much as when there is no CYA in the water. Basically, CYA is a hypochlorous acid buffer. There is also this one that shows why chlorine usage/consumption is an acidic process so the net pH effect of using a hypochlorite source of chlorine is pH neutral except for the excess lye. That same post also shows why saltwater chlorine generators are effectively the same as using hypochlorite. There is this one that derives the rough rule-of-thumb that the FC/CYA ratio gives the equivalent FC with no CYA in terms of having the same active chlorine level near a pH of 7.5.

As for UV absorption, there is very detailed info on the extinction coefficients for HOCl and OCl^- in Figure 3 in the paper Photolysis of aqueous free chlorine species (HOCl and OCl^-) with 254 nm ultraviolet light. The most relevant part of the spectrum is from around 300 nm (below that sun irradiance is too low due to UV absorption in the atmosphere) to 370 nm (above that the extinction coefficients are too low). I have a spreadsheet that uses these extinction coefficients along with the UV solar spectrum at the Earth's surface to calculate half-life in both shallow and deep water and it correctly calculates the expected half-life near the surface at pH 7.5 of 35 minutes. For hypochlorous acid it is 2 hours and 10 minutes while for hypochlorite ion it is 20 minutes. Note that with no CYA in the water, there is a shielding effect of UV in lower depths from chlorine breakdown since they absorb a significant number of photons. That is not so much the case when CYA is in the water, at least in terms of the unbound chlorine, so any shielding effect would have to come from either CYA or Cl-CYA species.

As for CYA and Cl-CYA absorption, that is where there is missing data for extinction coefficients in this sunlight UV region. The O'Brien paper lists the UV absorption for various CYA and chlorine bound to CYA chemical species in Figure 14.3 (205-230 nm) and Table 14.II (196-230 nm). This paper shows CYA absorption spectra from 220-240 nm (it's actually more likely cyanurate ions absorption and is consistent with the O'Brien paper) while this paper shows molar absorptivity for CYA species from just below 190 to 250 nm and is again largely consistent with the O'Brien paper. This paper in Figure 4 gives CYA absorption from roughly 210 nm to 260 nm and does so using a log scale for a large range. The latter paper implies no absorption above 250 or 260 nm but does not explicitly test in the range for sunlight UV (300-370 nm) nor does it test the chlorine bound to CYA species (i.e. the chlorinated isocyanurates).

As for chloramines, you have both inorganic chloramines as well as organic chloramines. For the former, there are several models including the latest ones from Jafvert & Valentine and these are described in the thread Chloramines and FC/CYA where I link to a spreadsheet I made for the various inorganic chloramine breakpoint chlorination models but also some speculative models for organic chloramine oxidation based on papers from Blatchley and others. As for explaining nitrogen trichloride at near-neutral pH, the reason is twofold. First is that those graphs people made are wrong in that while the percentage of nitrogen trichloride will be low, it will not be as low as they show. Second, it takes much lower concentrations of nitrogen trichloride to be irritating where the level of detection is at around 20 ppb (0.02 ppm). The models used for these graphs were older and they also didn't take into account the primary source of nitrogen trichloride production which is from urea so from organic chloramines such as chlorourea. Note that urea is by far the largest nitrogenous chemical in sweat and urine (see Table 4.1 in this link).

I suggest you do more than just put together existing literature to make a model. You should fill in missing pieces yourself and publish them such as the UV spectral absorption curves (extinction coefficients) for CYA species (you can separate them by selective pH) and for Cl-CYA species (you can separate them by selective pH and FC/CYA ratios) basically similar to what was done in the O'Brien paper, but focussing on the UV absorption range for sunlight -- say from 290-400 nm to be broader than what is needed. That would be a major positive contribution to science and would not be hard to do, though a bit tedious given the number of chemical species so pH and FC/CYA combinations.

The chloramines vary in volatility and you can find Henry Constants in this link. Basically, aqueous chlorine gas (molecular chlorine) is the most volatile (0.095 M/atm) followed by nitrogen trichloride (0.01 M/atm) then dichloramine (29 M/atm) then monochloramine (chloramide) (94 M/atm) then hypochlorous acid (930 M/atm). However, the amount of aqueous chlorine gas at normal pool pH is so small that it's greater volatility is not relevant and it is hypochlorous acid that is outgassed more than chlorine gas, though both are very low when CYA is present because the unbound chlorine amounts are relatively low (roughly 0.1 ppm FC equivalent). The only time you'll effectively see some chlorine gas volatility and significant hypochlorous acid volatility is around Trichlor pucks if there is no circulation. You also get noticeable hypochlorous acid volatility and hypochlorite ion volatility from bleach and chlorinating liquid.

 

[espejo]

Don't forget the flux capacitor and it requirements of 1.21 gigawatts.

Just kidding. You both are way over my head.

 

[Helios_Hyperion]

Hi Chem_Geek,

First off, thank you for your reply. I have been working through this, and currently have a question regarding this part:

As for UV absorption, there is very detailed info on the extinction coefficients for HOCl and OCl^- in Figure 3 in the paper Photolysis of aqueous free chlorine species (HOCl and OCl^-) with 254 nm ultraviolet light. The most relevant part of the spectrum is from around 300 nm (below that sun irradiance is too low due to UV absorption in the atmosphere) to 370 nm (above that the extinction coefficients are too low). I have a spreadsheet that uses these extinction coefficients along with the UV solar spectrum at the Earth's surface to calculate half-life in both shallow and deep water and it correctly calculates the expected half-life near the surface at pH 7.5 of 35 minutes. For hypochlorous acid it is 2 hours and 10 minutes while for hypochlorite ion it is 20 minutes. Note that with no CYA in the water, there is a shielding effect of UV in lower depths from chlorine breakdown since they absorb a significant number of photons. That is not so much the case when CYA is in the water, at least in terms of the unbound chlorine, so any shielding effect would have to come from either CYA or Cl-CYA species.

Since the study they did was at pH 5 and 10, and they mention pg. 280, Results and Discussions, paragraph 1, that the results might be different at different pH because of the combined concentration of HClO and OCl-, does your spreadsheet compensate for this? Or is the difference negligible?

Also, it would seem that the absorption rates are quite low for HClO in your proposed range of 300-370nm... me not being a chemist (YET!) would then surmise that the effect of U.V. is not actually that bad? (although I know this to be wrong)... but then again, since this is the first absorption coefficient graph I have seen, I have no idea as too how (bad) these coefficients are...

 

[chem geek]

Yes my spreadsheet uses the actual absorption spectra from their Figure 3 to calculate the half-life or loss rate of HOCl separate from OCl^- (that's the HOCl-Extinction.xls spreadsheet). So the Pool Equations spreadsheet that calculates HOCl and OCl^- can be used to figure out how much of each species is present even when there is CYA in the water. What they are talking about in the study at the two pH's was determining the quantum yield, but for the FC concentrations we are talking about the quantum yield for both HOCl (Figure 4) and OCl^- (Figure 5) is essentially 1 (technically 0.9 for OCl^-) meaning you can consider an absorption (extinction coefficient) to be a single degradation (i.e. one photon absorption leads to one molecular split).

The absorption rate is not "low" for HOCl though it is certainly lower than for OCl^-. The half-life of HOCl near the surface of the water or when the concentration is low (as is the case when CYA is present) is 2 hours and 10 minutes. So even if you had low pH with mostly HOCl you would still lose 87.5% of the chlorine in 6-1/2 hours. Of course the short 20 minute half-life of OCl^- means that at neutral to high pH the loss rate is even higher. So with no CYA in the water there is no mystery about the loss rate. At pH 7.5 the half-life of the combination is 35 minutes but this is near the surface of the water and the loss is so great that the absorbed photons near the surface partially shield depths (i.e. they use up photons) so the net result at normal pool depth is a half-life of about an hour.

What becomes unexplained is what happens when CYA is in the water because if you run through the calculations given the small unbound chlorine concentrations the chlorine loss in sunlight should be negligible for say 10% FC/CYA ratios where the unbound chlorine is around 0.1 ppm FC so the loss would be 0.05 ppm per hour so for 8 hours of equivalent noontime sun that would be a loss of only 0.4 ppm FC. Actual losses are higher so there may be some losses from chlorine bound to CYA and furthermore higher CYA levels seem to lower this loss rate. This is an area rich and fertile for new research as it has never been explored or explained.