High ph of RO water

[ludlumjohn]

Hello,

New to this site but have been keeping fish and plants for more than 50 years, yes I’m old 🤪

Currently running 2 planted tanks with C02 and doing PPS auto dosing. Tanks are only about a month old. The startup went very well, absolutely no algae issues other than some minor diatom issues after the 1st couple weeks.

I have available both RO and well water available and I trying to get the right mix of both to end up with about a 7.5 ph.

I set aside two glasses of water from the well and RO and had the following readings straight from the tap.

RO 6.2 ph and a tds of 10
Well 7.07 ph and tds of 220

After airating for a few hours I end up with the following
RO 7.95
Well 8.64
Any ideas on why my RO water would go so high. I want to reach equilibrium but curious if the airation is throwing it off and I should just let it settle.

The 2 tanks that are running now I did a 75 RO and 25 well. Tanks read about 7.0 in the morning and down to about 6.1 after c02. I would prefer to start with 7.5 and go down to 6.5 after c02. When I set these two tanks up I just did my normal mix of the 75/25 but was surprised how low is is. This is after a water change (2-3 days. I was thinking I would end up with a 50/50 mix but confused by my RO reading and my tanks actual reading. Hoping someone can make some sense of this for me.

Thanks in advance

 

[Yugang]

entirely pointless exercise

It raises the question under what conditions the measurements can be trusted in our aquarium, and if these measurements can be used for practical applications (like CO2 concentrations as we discuss in this thread).

 

[dw1305]

Hi all,

It raises the question under what conditions the measurements can be trusted in our aquarium,

You can trust pH measurement in alkaline, heavily carbonate buffered, water, water with an <"excess of proton (H+) acceptors">, you know it will be ~ pH 8 and that is what the electronic meter will confirm for you. Same applies to water with an excess of proton donors, the pH meter will tell you it's acidic.

and if these measurements can be used for practical applications .....

The problems come in <"water with few solutes">, unfortunately the kind of <"water many of us have in our tanks">. I've used the <"state of the snail shells and conductivity"> as proxies for likely pH levels, it is crude estimation, but not <"actively misleading">.

... practical applications (like CO2 concentrations as we discuss in this thread).

I'm not a CO2 user, but if I was I'd use a <"drop checker for the same reasons">, not perfect, but no moving parts and no electronics.

Cheers Darrel

 

[Andy Pierce]

How would one measure pH in RO water? Can the theory be experimentally verified?

One low-tech way is to take a water sample and add some salt (NaCl or KCl) and the measure the pH of the salted sample. For most purposes this is probably "good enough" to get the job done - the salt doesn't change the pH* and the added ionic strength makes the pH meter behave better.

* There are some complicated chemistry reasons why this isn't completely perfect - the AI can explain it much better than me if you're really interested.

 

[dw1305]

Hi all,

One low-tech way is to take a water sample and add some salt (NaCl or KCl) and the measure the pH of the salted sample. For most purposes this is probably "good enough" to get the job done - the salt doesn't change the pH* and the added ionic strength makes the pH meter behave better.

We use the <"neutral salt"> method with low ionic strength water samples, and it definitely shortens the time the pH meter takes to stabilise.

Cheers Darrel

 

[hax47]

Probably this is a bad idea but since the topic came up, I asked Claude 3.7 about the effect of phosphate. Interestingly in the very first go, Claude made the exact same mistake as me about converting ppm CaCO3 to eq/L. We fixed that and I inquired about effects of phosphate at some typical aquarium concentrations. An interesting point that came up relates to whether you measure the TA of your water and then add KH2PO4, or whether you have first added the KH2PO4 and then measured the TA of your water. Here's what Claude came up with (absolutely no guarantees on any of it although it seems plausible).

This is an interesting table. I think it shows that at measurable carbonate hardness values, the effect of the phosphate is negligible. I have the same feeling about other buffers as well.

Wenn sich CO₂ auflöst, steigt die DIC , aber auch pH- und KH-Wert ändern sich .

So, it's perfectly possible to have 0 dKH water and have significant levels of CO2 dissolved in it. That might be counter-intuitive, but the experiment is easy enough to do with an alkalinity titration kit and CO2-injected vs. degassed water.
Here's another way of looking at it: TA = 2[CO3-2] + [HCO3-] + [H2BO3-] + 2[HBO3-2] + 3[BO3-3] + [OH-] + [organic/inorganic H+ acceptors] - [H+]. When you add CO2 gas it converts to HCO3- and H+ in equal amounts since CO2(g) + H2O ⇌ H2CO3 ⇌ H⁺ + HCO3⁻ so the overall TA doesn't change because the amount of bicarbonate produced from the CO2 is exactly balanced by the amount of H+ produced from the CO2.

[TA] = [HCO₃⁻] + 2[CO₃²⁻] + [OH⁻] – [H⁺]​

This is the solution to a complicated process that confuses people, as it has confused me as well for a while. If we think that KH (total alkalinity) equals HCO3-, we could conclude that when CO2 is dissolved in water and bicarbonate is formed, the KH rises. However, it does not, since an equal amount of protons are produced, and, according to the above equation, the alkalinity will not change.
I have created a calculator, that @Marcel G was kind to translate to English and host: CO2/KH/pH calculator, one can set a certain KH value and check how the KH and the bicarbonate change when the CO2 slider is moved (KH does not). I must disclose though, that it is not surprising that there is no change in the alkalinity, since the calculations are based on the assumption that alkalinity does not change upon CO2 concentration changes... This is in agreement with what @Andy Pierce already said, but also with what is explained in "Aquatic Chemistry Concepts, Second Edition" by James F. Pankow (a good read for those who want to dig into water chemistry a little bit deeper).

My calculators are online, if you want to compare my and your data I would send you a link. Maybe you used PH calculation with a formula. Since the equations are unsolvable by formulas we used iteration very precisly for calculation PHs. But in this case the PH is gven, so if you apply Henderson equations you can calculate it as I did.

My calculator is also iteration-based. The problem is with dynamic calculations (i.e. what happens when we raise the CO2 in water with certain parameters), that if we add CO2, part of that CO2 is converted to HCO3-, thus in a Henderson-Hasselbach equation not only the denominator (CO2) changes but also the numerator (HCO3-) and the pH. And the pH change determines how much CO2 will be converted. So the pH determines the HCO3-/CO2 ratio, but the HCO3-/CO2 ratio determines the pH, and we are in a situation where we cannot calculate pH without HCO3-, and we cannot calculate the HCO3- without knowing the pH, even if we know the CO2. For a certain CO2/HCO3 combination you can easily calculate all the other values (pH, CO3--, DIC), but once you change the CO2 in the same setup, everything will change to a new equilibrium. Therefore, because alkalinity remains stable, one could iterate all the other values after CO2 changes until the total alkalinity change is minimized/zero to get the new equilibrium (this is what I did). I guess this concept might be different in @Ahnungsloser's calculator.

To summarize, once we add CO2 into RO water, HCO3- will increase (one could use my simulator to get some values), but the alkalinity/KH will not. The iteration does not always converge in my calculator, which is why there are no values with certain CO2/KH values.

The main reason I created this calculator was a common concept among aquarists (at least in Hungary), that we need to use more CO2 in more alkaline water to maintain the desired dissolved CO2 concentration. To my surprise, alkaline water will indeed require significantly more CO2 to initially elevate the concentration to 30 ppm from let's say 3 ppm, but this extra CO2 seems to be "buffered"/stored in the form of bicarbonate which will be released after turning off the CO2 dosing, slowing down the CO2 drop a little bit. The higher the alkalinity and smaller the CO2 elevation the more percent of the initially injected CO2 will be temporarily stored in the form of HCO3-. One can also check some calculated values in the calculator on the "CO2 storage" tab.

 

[dw1305]

Hi all,

..... The problem is with dynamic calculations (i.e. what happens when we raise the CO2 in water with certain parameters), that if we add CO2, part of that CO2 is converted to HCO3-, thus in a Henderson-Hasselbach equation not only the denominator (CO2) changes but also the numerator (HCO3-) and the pH. And the pH change determines how much CO2 will be converted. So the pH determines the HCO3-/CO2 ratio, but the HCO3-/CO2 ratio determines the pH, and we are in a situation where we cannot calculate pH without HCO3-, and we cannot calculate the HCO3- without knowing the pH, even if we know the CO2. For a certain CO2/HCO3 combination you can easily calculate all the other values (pH, CO3--, DIC), but once you change the CO2 in the same setup, everything will change to a new equilibrium

I think that is the problem with a lot of the scientific literature, which quotes CO2 values for rivers etc. <"https://www.sciencedirect.com/science/article/abs/pii/S0380133022000211">*, they aren't measured values, but values extrapolated from a chart of alkalinity and pH.

*Elizabeth C. Minor, Gabriella Brinkley, (2022) "Alkalinity, pH, and pCO2 in the Laurentian Great Lakes: An initial view of seasonal and inter-annual trends," Journal of Great Lakes Research, 48:2, pg 502-511.

cheers Darrel

 

[hax47]

Hi all,

I think that is the problem with a lot of the scientific literature, which quotes CO2 values for rivers etc. <"https://www.sciencedirect.com/science/article/abs/pii/S0380133022000211">*, they aren't measured values, but values extrapolated from a chart of alkalinity and pH.

*Elizabeth C. Minor, Gabriella Brinkley, (2022) "Alkalinity, pH, and pCO2 in the Laurentian Great Lakes: An initial view of seasonal and inter-annual trends," Journal of Great Lakes Research, 48:2, pg 502-511.

cheers Darrel

I think that is less of a problem, once we have pH and alkalinity, we can confidently calculate the bicarbonate/carbonate and the CO2 from these. That is, only if we assume that there are no other buffers in significant concentration besides the bicarbonate buffer, which I think is usually a good assumption and we have good enough methods to accurately measure, and we know the constants for the given temperature (not trivial). There could be problems though with direct measurements as well, when we measure CO2 directly in the headspace in tubes with water samples (with relatively large headspace), because of the CO2-buffering capacity of the water that I mentioned above:

Technical note: CO2 is not like CH4 – limits of and corrections to the headspace method to analyse pCO2 in fresh water

Abstract. Headspace analysis of CO2 frequently has been used to quantify the concentration of CO2 in fresh water. According to basic chemical theory, not considering chemical equilibration of the carbonate system in the sample vials will result in a systematic error. By analysing the potential...

bg.copernicus.org

Auf welchen KH-Wert in der Tabelle beziehen Sie sich? Auf den KH-Wert vor der CO₂-Lösung aus einer externen Quelle oder auf den KH-Wert danach ?

Ihr Wasser hat zunächst einen bestimmten KH -Wert , einen bestimmten pH -Wert und einen bestimmten DIC -Wert (gelöster anorganischer Kohlenstoff) . Wenn CO₂ hinzugefügt wird – beispielsweise über einen Ausströmer –, nimmt das Wasser nur dann CO₂ auf, wenn der Partialdruck in der Atmosphäre höher ist als im Wasser. Nehmen wir an, das ist der Fall.

Wenn sich CO₂ auflöst, steigt die DIC , aber auch pH- und KH-Wert ändern sich .

Some uncertainty arises though when we have a known pH/KH and we start to dissolve CO2 and we want to predict the new CO2 levels based on pH, since some of the CO2 turns into HCO3-, affecting the calculations. However, if we calculate with constant bicarbonate (what we usually do), the difference in calculation is negligible at usual alkalinity values. That is why the usual formula of 1 pH drop = 10x CO2 concentration increase works.

Bigger might be the impact of this CO2/HCO3- transformation if we do not raise the CO2 level to a certain concentration (or pH to a certain value), but instead we add a certain amount of CO2 into alkaline water. Like when someone adds a single dose of CO2 a day (soda water for example). In this case, the CO2 concentration can not be calculated as a simple dilution, especially at high alkalinity and small CO2 changes, but one needs to take into account the carbonate equilibrium shift, as described in the paper above.

 

[dw1305]

Hi all,

I think that is less of a problem, once we have pH and alkalinity, we can confidently calculate the bicarbonate/carbonate and the CO2 from these. That is, only if we assume that there are no other buffers in significant concentration besides the bicarbonate buffer, which I think is usually a good assumption and we have good enough methods to accurately measure, and we know the constants for the given temperature (not trivial).

That makes sense, I just wish my chemistry was better.

There could be problems though with direct measurements as well, when we measure CO2 directly in the headspace in tubes with water samples (with relatively large headspace), because of the CO2-buffering capacity of the water that I mentioned above:

Thank-you <"Technical note: CO2 is not like CH4 – limits of and corrections to the headspace method to analyse pCO2 in fresh water"> that is very useful.

cheers Darrel

 

[Marcel G]

One low-tech way is to take a water sample and add some salt (NaCl or KCl) and the measure the pH of the salted sample. For most purposes this is probably "good enough" to get the job done - the salt doesn't change the pH and the added ionic strength makes the pH meter behave better ...

Thanks so much for the enlightenment! I didn't know that and I will surely make good use of it in my experiments (as I very often work with low conductivity water). In case someone wants to practice this as well, I am attaching some practical calculations for preparing the required samples:

1) How to prepare saturated (= 3M) KCl solution?

weight (g) = (molarity * molecular weight * volume) / 1000
Calculation: 3M * 74.55 g/mol * 100 ml / 1000 = 3 * 74.55 * 100 / 1000 = 22.365 g
Result: ~22.4 g KCl dissolved in 100 ml water gives 3M KCl solution.

Now, according to this article, adding KCl at 0.01M may shift pH by +0.02 pH. So, if we want the amount of KCl added to raise the pH by a maximum of 0.02 ...

2) How much saturated KCl solution do we need to add to the sample?

volume (ml) = final molarity * 100 / initial molarity
Calculation: 0.01 * 100 / 3 = 0.33 ml
Result: By the above procedure (i.e. adding ~0.3 ml of saturated KCl solution to 100 ml of sample) the resulting molarity of KCl in the sample will be 0.01M, which will result in an increase in the measured pH of only 0.02.

You can now measure the pH in a sample of pure [= low conductivity] water.

 

[dw1305]

Hi all,
I've always used sodium chloride (NaCl) as my <"neutral salt">, but I can see that there are actually advantages to using potassium chloride (KCl).

Result: ~22.4 g KCl dissolved in 100 ml water gives 3M KCl solution.

You can use that as the storage solution for the pH electrodes - <"pH Electrode Storage Solution, Reagecon - pH and Electrochemistry, Probes and Electrodes">. You don't need to buy it made up, you can make it up from the salt (food grade KCl is ~£15 / kg) and it doesn't have to be an accurate 3 mol. solution for electrode storage.

2) How much saturated KCl solution do we need to add to the sample?

volume (ml) = final molarity * 100 / initial molarity
Calculation: 0.01 * 100 / 3 = 0.33 ml
Result: By the above procedure (i.e. adding ~0.3 ml of saturated KCl solution to 100 ml of sample) the resulting molarity of KCl in the sample will be 0.01M,

and that is also one of the calibration solutions (1413 microS) for conductivity meters <"Conductivity Standard Solution, 1413 µS/cm, KCl, 500 mL">.

In this case you do need to make the solution up accurately <"TDS pen recommendations">, you can still use the salt (KCl) and DI water.

cheers Darrel