Fc/cya Ratio

[DickP3]

Hi chem geek

My background is chemistry and biology and I'm interested in the information that led to the ratio of 4 ppm FC to 20 ppm CYA. I have tried searching this board for it but can't seem to find it. Can you point me to it?

Thanks

[chem geek]

Hi chem geek

My background is chemistry and biology and I'm interested in the information that led to the ratio of 4 ppm FC to 20 ppm CYA. I have tried searching this board for it but can't seem to find it. Can you point me to it?

Thanks

Start with this thread including the PoolEquations.zip link to a ZIPped spreadsheet near the end of the first post. The original 1973/1974 paper with the equilibrium constants of the chlorinated cyanurates is not online and I had to go to a university archives to find it, but the same data from it (with a minor exception for some pKa constants of CYA) may be found in this EPA PDF file on document page 12 (PDF page 18). If you find any errors in the spreadsheet, please let me know. It's been quite stable for quite some time so I think any errors have been flushed out of it, but you never know. Most people don't know enough about the chemistry to be able to validate it, but you can always take the end results of the calculations and see that they are consistent with the equilibrium constants (adjusted for ionic strength) which is what I've done to try and find any errors (that, plus some debugging checks I have).

For disinfection rates, see this post where you will see the sources for the 30-50 CT values for the bacteria that causes "hot tub itch", Pseudomonas aeruginosa, though realize that I've seen another older source with a far lower CT value (< 0.2) for this same bacteria.

You should understand that the pool/spa industry (mostly chemical manufacturers) says that such chemistry cannot be directly applied to "real pools" and they refer to a study they did to prove that. I discuss it in this thread where I link to that study and do some analysis and give my (different) interpretation of the results. Again, if you see any flaws in the reasoning or have a different interpretation of the data, please let me know. I only want to find the truth, to the extent it can be known.

The bottom line is that the 4 ppm FC with 20 ppm CYA is most likely to be overly conservative, but I'd rather start there and see how things go for people before adjusting the recommendation upwards. My gut feel is that the "right" balance between appropriate disinfection but using the lowest chlorine level possible (without additional slower-acting disinfectants) is somewhere in the 30-50 ppm neighborhood though it's possible that up to 100 ppm CYA is OK, but then at the higher CYA there are other issues (even if slower disinfection were OK) such as too slow a breakpoint reaction (for chlorine oxidizing ammonia). I have a spreadsheet that models the breakpoint reaction (now using the Jafvert & Valentine 1992 model), but I haven't posted that anywhere as I wasn't happy with it until I paid for this latest model ($25 to get an online paper describing the model -- science isn't free unless you're near a university with the right journals). I've also seen papers talk about how urea is slower to oxidize (by chlorine) than ammonia, but I can't find any definitive rate data on that.

Richard

[DickP3]

Hi chem geek

My background is chemistry and biology and I'm interested in the information that led to the ratio of 4 ppm FC to 20 ppm CYA. I have tried searching this board for it but can't seem to find it. Can you point me to it?

Thanks

Start with this thread including the PoolEquations.zip link to a ZIPped spreadsheet near the end of the first post. <snip>

Richard

WOW!!!! Be careful what you ask for. It's going to take me some time to work through this (assuming that I can). You evidently have spent an awful lot of time on pool and spa chemistry.

One question: Is there a way to print posts from troublefreepool.com? Everything I tried missed the graphs or some of the text.

Thanks for your efforts. We all benefit from them.

[chem geek]

One question: Is there a way to print posts from troublefreepool.com? Everything I tried missed the graphs or some of the text.

I just tried printing the Pool Water Chemistry and it seems OK except it skips to the next page sometimes (i.e. inserts an extra blank page), but I didn't see any missing graphs or text. Perhaps it's the specific browser you are using. I'm using Firefox on Mac OS X. I did need to scale down (in the Page Setup in the print driver) in order for it to fit the page, but I generally have to do that with any web pages that are wide.

[DickP3]

One question: Is there a way to print posts from troublefreepool.com? Everything I tried missed the graphs or some of the text.

I just tried printing the Pool Water Chemistry and it seems OK except it skips to the next page sometimes (i.e. inserts an extra blank page), but I didn't see any missing graphs or text. Perhaps it's the specific browser you are using. I'm using Firefox on Mac OS X. I did need to scale down (in the Page Setup in the print driver) in order for it to fit the page, but I generally have to do that with any web pages that are wide.

I'm using Internet Explorer with Windows XP so I downloaded Firefox but I'm getting the same results - missing text and graphs. (I know - I should get a Mac) I guess I'll just save it to my disk and read it on the screen. Hard to make notes on it that way, but at least you've done the hard part putting it together. Thanks again.

[DickP3]

Hi chem geek

My background is chemistry and biology and I'm interested in the information that led to the ratio of 4 ppm FC to 20 ppm CYA. I have tried searching this board for it but can't seem to find it. Can you point me to it?

Thanks

Start with this thread including the PoolEquations.zip link to a ZIPped spreadsheet near the end of the first post. The original 1973/1974 paper with the equilibrium constants of the chlorinated cyanurates is not online and I had to go to a university archives to find it, but the same data from it (with a minor exception for some pKa constants of CYA) may be found in this EPA PDF file on document page 12 (PDF page 18). If you find any errors in the spreadsheet, please let me know. It's been quite stable for quite some time so I think any errors have been flushed out of it, but you never know. Most people don't know enough about the chemistry to be able to validate it, but you can always take the end results of the calculations and see that they are consistent with the equilibrium constants (adjusted for ionic strength) which is what I've done to try and find any errors (that, plus some debugging checks I have).

For disinfection rates, see this post where you will see the sources for the 30-50 CT values for the bacteria that causes "hot tub itch", Pseudomonas aeruginosa, though realize that I've seen another older source with a far lower CT value (< 0.2) for this same bacteria.

You should understand that the pool/spa industry (mostly chemical manufacturers) says that such chemistry cannot be directly applied to "real pools" and they refer to a study they did to prove that. I discuss it in this thread where I link to that study and do some analysis and give my (different) interpretation of the results. Again, if you see any flaws in the reasoning or have a different interpretation of the data, please let me know. I only want to find the truth, to the extent it can be known.

The bottom line is that the 4 ppm FC with 20 ppm CYA is most likely to be overly conservative, but I'd rather start there and see how things go for people before adjusting the recommendation upwards. My gut feel is that the "right" balance between appropriate disinfection but using the lowest chlorine level possible (without additional slower-acting disinfectants) is somewhere in the 30-50 ppm neighborhood though it's possible that up to 100 ppm CYA is OK, but then at the higher CYA there are other issues (even if slower disinfection were OK) such as too slow a breakpoint reaction (for chlorine oxidizing ammonia). I have a spreadsheet that models the breakpoint reaction (now using the Jafvert & Valentine 1992 model), but I haven't posted that anywhere as I wasn't happy with it until I paid for this latest model ($25 to get an online paper describing the model -- science isn't free unless you're near a university with the right journals). I've also seen papers talk about how urea is slower to oxidize (by chlorine) than ammonia, but I can't find any definitive rate data on that.

Richard

Hi Richard

I’m working through the postings you referred me to. Pool and spa chemistry involves a complicated set of reactions and it looks like you’ve done a great job ferreting out the details needed to understand them.

I’m interested in the contents of the “HOCl Table.” It looks like this chart along with the knowledge that 0.011 ppm HOCl approximates the US and WHO minimum standard for disinfecting allows an estimation of how “safe” a given set of spa conditions are. Am I correct? Were the values in the table calculated using the spreadsheet? (I’m afraid that my inorganic chemistry is rusty by 40+ years and so I’m not going to be much help validating it).

How was the US and WHO standard for disinfecting derived?

In the “disinfection rates” post (viewtopic.php?t=873), what disinfectant is used to calculate CT? Is it FC or HOCl?

Thanks

[chem geek]

Hi Richard

I’m working through the postings you referred me to. Pool and spa chemistry involves a complicated set of reactions and it looks like you’ve done a great job ferreting out the details needed to understand them.

I’m interested in the contents of the “HOCl Table.” It looks like this chart along with the knowledge that 0.011 ppm HOCl approximates the US and WHO minimum standard for disinfecting allows an estimation of how “safe” a given set of spa conditions are. Am I correct? Were the values in the table calculated using the spreadsheet? (I’m afraid that my inorganic chemistry is rusty by 40+ years and so I’m not going to be much help validating it).

How was the US and WHO standard for disinfecting derived?

In the “disinfection rates” post (viewtopic.php?t=873), what disinfectant is used to calculate CT? Is it FC or HOCl?

Thanks

The contents of the HOCl Table came from the PoolEquations spreadsheet (a link to the ZIP file is near the end of the first post in the pool water chemistry link). I just put in different amounts of FC and CYA into the spreadsheet and copied them to a separate spreadsheet (that I did not upload online) to make the graph. The core calculations are all in the PoolEquations spreadsheet and it is equilibrium chemistry, but as you point out there are lots of simultaneous equations (equilibria).

The ORP value of 650 mV that the U.S. and WHO use as a minimum disinfection standard is something arrived at empirically, mostly promoted by the ORP sensor industry. Since I have found that there is a rather large discrepancy in ORP measurements from different sensors, it's not such a great "absolute" standard after all. Nevertheless, it's better than looking at FC alone. ORP more or less measures the HOCl (it technically measures total oxidation potential, but in water with chlorine the HOCl is the dominant oxidizer), but it is also more affected by pH (beyond the affect of pH on HOCl/OCl^- equilibrium itself) and temperature so it's really only a proxy for HOCl at a constant pH. I don't have a single source of where the actual 650 mV "standard" came from, but it appears that there were field studies where pools below that ORP showed vastly more bacterial counts than pools above that level. Since the ORP sensors vary, the 0.011 ppm FC isn't an absolute standard and I've put several different ORP estimates into the spreadsheet near row 370. This post was in the thread on pool water chemistry I linked to earlier and it has links to a Commercial Spa Study where you can see ORP vs. bacterial counts.

[EDIT]

The problem is that so little chlorine is needed to kill most bacteria that looking at HOCl or FC/CYA or even FC alone appears to predict sanitation in real pool studies. There are many, many scientific peer-reviewed studies that show significant reductions in disinfection (increases in CT time) when CYA is present, but these studies vary in the linearity of the effect once CYA is present. This PDF file has a graph of kill time vs. CYA level that is pretty typical with a huge drop (very fast kill) at no CYA and a more linear slope as more CYA is added, but not always the doubling time with doubling CYA that would be predicted from HOCl concentration. This PDF file shows a similar trend though with less proportionality. This PDF file shows almost no effect of CYA on kill time against algae, but I have serious problems with this study's methodology (that I won't get into here) and it's completely inconsistent with what we see with pools from hundreds of users on multiple pool forums. This link gives an abstract that essentially says the breakpoint reaction is slowed down when there is CYA with chlorine and it corresponds to the HOCl concentration. This link shows that virus kill rates are slower with CYA by factors that correspond to HOCl conentration (when I use my spreadsheet to calculate it). This link shows a direct correlation between HOCl concentration (not FC, when CYA is present) and inactivation time for protozoan cysts. This PDF file is a study the SWG industry uses partly to justify salt not being corrosive at the normal 3000 ppm levels used today (I don't think that's the real full conclusion -- less corrosive, yes, but I wouldn't say "not" corrosive), but it also shows a reduction in corrosion rates from chlorine when CYA is present. And of course, there is the lowering effect on ORP when CYA is present. There are also direct HOCl sensors (using selective membranes) and there are amperometric sensors that more directly measure HOCl though I've never seen a pool study looking at HOCl levels (rather than FC or ORP).

[END-EDIT]

If you look at the Pinellas pool study I linked to, you will see how there is lots of bacteria at no chlorine and very low chlorine levels and it drops off rapidly with just a little chlorine. They didn't measure ORP, but I've received studies from some water quality people in various counties across the country, but that's their data and I can't publish that online (though some have been presenting it themselves).

As for the disinfectant used for all CT tables, NONE of them use CYA so it is certainly not FC in the presence of CYA that is used. For some tables it is FC with no CYA and such tables often have a pH dependence. Other more rare tables use HOCl so do not have the pH dependence, but most I've seen are the former even if they don't show columns of pH. That is, for most CT tables, it's FC with no CYA and you need to see the assumption of what pH is being used. If it's a low pH, then the table is effectively for HOCl since most FC will be HOCl; if the table is at 7.5 pH, then roughly half of the FC will be HOCl. [EDIT] If you look at Appendix B, CT Tables, in this PDF file from the EPA, you will see in Table B-1 how the inactivation time for Giardia as a function of pH tracks the HOCl concentration (you can use my spreadsheet to calculate the HOCl concentration vs. pH with no CYA or use traditional industry graphs). In other words, HOCl is the disinfectant and hypochlorite ion, OCl^- plays a minor role that can usually be neglected. [END-EDIT]

I wouldn't sweat being super-accurate about the HOCl vs. FC distinction since there is quite a variability in CT values as well. It's best to just look at it as an order-of-magnitude guideline.

Richard

[DickP3]

Hi Richard

I’m working through the postings you referred me to. Pool and spa chemistry involves a complicated set of reactions and it looks like you’ve done a great job ferreting out the details needed to understand them.

I’m interested in the contents of the “HOCl Table.” It looks like this chart along with the knowledge that 0.011 ppm HOCl approximates the US and WHO minimum standard for disinfecting allows an estimation of how “safe” a given set of spa conditions are. Am I correct? Were the values in the table calculated using the spreadsheet? (I’m afraid that my inorganic chemistry is rusty by 40+ years and so I’m not going to be much help validating it).

How was the US and WHO standard for disinfecting derived?

In the “disinfection rates” post (viewtopic.php?t=873), what disinfectant is used to calculate CT? Is it FC or HOCl?

Thanks

The contents of the HOCl Table came from the PoolEquations spreadsheet (a link to the ZIP file is near the end of the first post in the pool water chemistry link). I just put in different amounts of FC and CYA into the spreadsheet and copied them to a separate spreadsheet (that I did not upload online) to make the graph. The core calculations are all in the PoolEquations spreadsheet and it is equilibrium chemistry, but as you point out there are lots of simultaneous equations (equilibria).

The ORP value of 650 mV that the U.S. and WHO use as a minimum disinfection standard is something arrived at empirically, mostly promoted by the ORP sensor industry. Since I have found that there is a rather large discrepancy in ORP measurements from different sensors, it's not such a great "absolute" standard after all. Nevertheless, it's better than looking at FC alone. ORP more or less measures the HOCl (it technically measures total oxidation potential, but in water with chlorine the HOCl is the dominant oxidizer), but it is also more affected by pH (beyond the affect of pH on HOCl/OCl^- equilibrium itself) and temperature so it's really only a proxy for HOCl at a constant pH. I don't have a single source of where the actual 650 mV "standard" came from, but it appears that there were field studies where pools below that ORP showed vastly more bacterial counts than pools above that level. Since the ORP sensors vary, the 0.011 ppm FC isn't an absolute standard and I've put several different ORP estimates into the spreadsheet near row 370. This post was in the thread on pool water chemistry I linked to earlier and it has links to a Commercial Spa Study where you can see ORP vs. bacterial counts.

[EDIT]

The problem is that so little chlorine is needed to kill most bacteria that looking at HOCl or FC/CYA or even FC alone appears to predict sanitation. There are many, many studies that show significant reductions in disinfection (increases in CT time) when CYA is present, but these studies vary in the linearity of the effect once CYA is present. This PDF file has a graph of kill time vs. CYA level that is pretty typical with a huge drop (very fast kill) at no CYA and a more linear slope as more CYA is added, but not always the doubling time with doubling CYA that would be predicted from HOCl concentration. This PDF file shows a similar trend though with less proportionality. This PDF file shows almost no effect of CYA on kill time against algae, but I have serious problems with this study's methodology (that I won't get into here) and it's completely inconsistent with what we see with pools from hundreds of users on multiple pool forums. This link gives an abstract that essentially says the breakpoint reaction is slowed down when there is CYA with chlorine and it corresponds to the HOCl concentration. This link shows that virus kill rates are slower with CYA by factors that correspond to HOCl conentration (when I use my spreadsheet to calculate it). This link shows a direct correlation between HOCl concentration (not FC, when CYA is present) and inactivation time for protozoan cysts. This PDF file is a study the SWG industry uses partly to justify salt not being corrosive at the normal 3000 ppm levels used today (I don't think that's the real full conclusion -- less corrosive, yes, but I wouldn't say "not" corrosive), but it also shows a reduction in corrosion rates from chlorine when CYA is present. And of course, there is the lowering effect on ORP when CYA is present. There are also direct HOCl sensors (using selective membranes) and there are amperometric sensors that more directly measure HOCl though I've never seen a pool study looking at HOCl levels (rather than FC or ORP).

[END-EDIT]

If you look at the Pinellas pool study I linked to, you will see how there is lots of bacteria at no chlorine and very low chlorine levels and it drops off rapidly with just a little chlorine. They didn't measure ORP, but I've received studies from some water quality people in various counties across the country, but that's their data and I can't publish that online (though some have been presenting it themselves).

As for the disinfectant used for all CT tables, NONE of them use CYA so it is certainly not FC in the presence of CYA that is used. For some tables it is FC with no CYA and such tables often have a pH dependence. Other more rare tables use HOCl so do not have the pH dependence, but most I've seen are the former even if they don't show columns of pH. That is, for most CT tables, it's FC with no CYA and you need to see the assumption of what pH is being used. If it's a low pH, then the table is effectively for HOCl since most FC will be HOCl; if the table is at 7.5 pH, then roughly half of the FC will be HOCl.

I wouldn't sweat being super-accurate about the HOCl vs. FC distinction since there is quite a variability in CT values as well. It's best to just look at it as an order-of-magnitude guideline.

Richard

Hi Richard

I’m trying to understand document http://www.epa.gov/etv/pubs/02_vs_oxi2b.pdf and then apply it to the discussion at http://www.troublefreepool.com/viewtopic.php?t=873 relating to P. Aeruginosa. In the EPA document, they seem to be using FC as a measure of disinfectant present. The calculated CT(4-log) of 50 was based on the actual hydraulic retention time (which was 34.1 minutes). The average FC in the finished water was 1.5 mg/L. So, 1.5 * 34.1 yields a CT(4-log) of 51, right. All at zero ppm CYA. The treated water pH was 7.4 and the alkalinity was 17 ppm. Plugging these numbers into the spreadsheet showed that the HOCl was 0.79 ppm.

Since 0.79 ppm HOCl is pretty strong, the discussion in the above document suggested a CT(50%) instead of CT(99.99%). The calculated CT(50%) is 3.76. The discussion above says that “at 15 minutes generation time this means a hypochlorous acid level of 0.25.” However, it looks to me like in the EPA document, the CT is based on concentration of FC in ppm (not hypochlorus acid) and time in minutes. So, a contact time of 15 minutes gives a FC of 0.25 ppm. From the spreadsheet, an FC of 0.25 at 0 CYA gives an HOCl concentration of 0.13 ppm or 5 ppm FC at 20 ppm CYA.

Since my conclusions don’t agree with those in the above discussion, I think I must have missed something. Can you please point it out to me?

Thanks

[chem geek]

The first paragraph in your analysis seems correct.

For the second paragraph, I need to explain why I used a CT(50%) value instead of the CT(99.99%) value in the EPA submission document. They were doing water treatment where the contact times are relatively short so for them they want to virtually eliminate pathogens from the water which is why you almost always see at least CT(99%) if not higher reductions of pathogen counts. This is usually for drinking water purposes. This is different than swimming pools and hot tubs where the contact time is much longer. It is true that to prevent transmission from person to person one wants to look at thorough kill times; so a CT(99%) or higher would be appropriate for things like fecal bacteria/viruses/protozoan cysts, but my take on the hot tub itch bacteria is that it is normally present in the environment and on skin so what is really desired in this case is to not provide a water environment where it grows to much larger numbers that can then be a more serious problem (via open wounds or breathed into the lungs). So that's why I focussed on CT(50%) compared to the typical generation (doubling) rate of bacteria.

To get from CT(99.99%) to CT(50%) one uses the formula CT(50%) = CT(n-log) * 0.30103 / n so this is 50 * 0.30103 / 4 = 3.76 as you indicated. To find the "C" chlorine concentration we need to use a "T" time in minutes and I used 15 minutes as the short time for the bacterial generation rate so 3.76 / 15 = 0.25 and it is true that this would be an FC value at a pH of 7.4 to be consistent with the original CT value that was used (which was not what I did -- I incorrectly assumed HOCl). And you are right that you get an equivalent HOCl if you use 5 ppm FC at 20 ppm CYA. There is nothing wrong with your calculation. You can't compare the 0.79 ppm from the first paragraph with the 0.13 ppm in the second paragraph since the first number was with a contact time of 34.1 minutes and a 99.99% kill rate while the second number assumes 15 minutes and a 50% kill rate. Their relationship is 0.79 * (3.76 / 50) * (34.1 / 15) = 0.13 (rounding .135 down).

I got to 4 ppm at 20 ppm CYA by being rough and not assuming their initial CT used FC so your methodology was better than mine. However, there are a LOT of assumptions made along the way including actual bacterial generation time and the fact that the real CT value would likely be lower at higher temperature (most CT values get lower at higher temperature because the chemical reactions get faster at a faster rate than pathogens can reproduce since reproduction involves movement of larger-scale organelles that is much slower than chemical reactions). So, bottom line, there is nothing wrong with your calculation, but a number of 4 or 5 ppm FC with 20 ppm CYA is rather roughly determined anyway and is probably overly conservative, especially since another source gives a far lower CT value for hot tub itch bacteria.

Since some of the rashes people have been getting go away rather quickly when not exposed to the water, it's most likely these are not due to bacteria. Only a couple of cases early on were confirmed through analysis with doctors (though I don't think they took actual swabs with growth and analysis under a microscope) -- one was hot tub itch and another was hot tub lung. The rest seem to be skin sensitivity. In at least one case this seems to be from excessive use of MPS (in the first week). In most of the others, they occur after one or more months of Dichlor use so I'm leaning more towards an explanation that involves monochloramine buildup, partly due to the slower breakpoint at higher CYA levels and partly due to some cases probably not using enough chlorine. But this is, of course, just speculation.

Nevertheless, to achieve reasonably fast breakpoint of ammonia/urea, one needs to keep the CYA level lower (actually, the FC/CYA ratio higher). The FC level should really be whatever is needed to not run out of chlorine. This usually is around 4 ppm FC per use, but higher bather loads would need more chlorine unless supplemented by MPS. At 4 ppm FC and 20 ppm CYA, breakpoint of ammonia is 50% complete after around 40 minutes and 90% complete after almost 2 hours. At 4 ppm FC and 100 ppm CYA, it is 50% complete after 3 hours and 90% complete after 9 hours so roughly 4.5 times slower. In practice, it's worse than this (i.e. all times longer) since these numbers are for a small amount of ammonia that consumes less than 1/10th of the FC. If I assume an amount of ammonia that is half the chlorine equivalent (i.e. in similar units), then breakpoint for 4 ppm FC and 20 ppm CYA is 50% complete in almost 100 minutes and 90% complete after 7-1/2 hours. The breakpoint for 4 ppm FC and 100 ppm CYA is 50% complete in a little over 7 hours and 90% complete after a little more than 30 hours. So there is still at least a factor of 4 difference, but since the times are so long there can be a buildup of monochloramine from day to day. I didn't post the breakpoint spreadsheet online yet as I only just recently got a good model for it (Jafvert & Valentine 1992), but if you're interested then E-mail me your E-mail and I'll send it to you. [EDIT] NOTE: The Jafvert & Valentine model is at 77F and does not give a temperature dependence. Clearly, the reactions are faster at 104F spa temperatures and could be around twice as fast but the factor of 4-5 slowdown caused by 100 ppm CYA vs. 20 ppm CYA would still occur. [END-EDIT]

Richard