DIY Alkatronic reagent

[iLMaRiO]

Little bit OT: titration has to reach pH 4.3 or 4.5?

 

[Dennis Cartier]

Little bit OT: titration has to reach pH 4.3 or 4.5?

According to Randy's article, titrate to a pH of 4.5

A DIY Alkalinity Test: By Randy Holmes-Farley - REEFEDITION

Alkalinity is one of the most important measurements that a reef aquarist can make. It can become rapidly depleted in many aquaria, requiring frequent measurement in order to maintain stable levels.

www.reefedition.com

 

[Randy Holmes-Farley]

According to Randy's article, titrate to a pH of 4.5

A DIY Alkalinity Test: By Randy Holmes-Farley - REEFEDITION

Alkalinity is one of the most important measurements that a reef aquarist can make. It can become rapidly depleted in many aquaria, requiring frequent measurement in order to maintain stable levels.

www.reefedition.com

There's some debate in the literature, or at least different mentioned values. Jim Welsh and I discussed this here once. It should hardly matter in terms of the alk difference, but it's not clear why. Maybe slightly different pH scales, different assumptions of the actual alk being measured (that does alter the pH endpoint), and things like the assumed pKa for carbonic acid under different conditions all attainable in seawater.

 

[Randy Holmes-Farley]

I did take a stab at a dilution of the HCL assuming it was 10N, but flubbed the dilution and accidentally used 100 uL instead of 1000 uL, and got 111 ppm, 6.21 dKH. When I noticed my mistake I added 900 uL (of the HCL) to the volume, and got 66 ppm, 3.69 dKH. My gut says that the HCL is probably just under 10N, as the extra 900 uL added to the 100 ml can't account for the difference.

I didn't quite understand the last sentence. The method overall seems fine, if you accept the Hanna as accurate.

 

[Dennis Cartier]

I didn't quite understand the last sentence. The method overall seems fine, if you accept the Hanna as accurate.

The Hanna appears to be as accurate as my Alkatronic. When I compare the two, the Hanna reads 0.06 dKH higher than the Alkatronic. I am happy with that.

 

[Randy Holmes-Farley]

The Hanna appears to be as accurate as my Alkatronic. When I compare the two, the Hanna reads 0.06 dKH higher than the Alkatronic. I am happy with that.

Sounds good!

 

[JimWelsh]

There's some debate in the literature, or at least different mentioned values. Jim Welsh and I discussed this here once. It should hardly matter in terms of the alk difference, but it's not clear why. Maybe slightly different pH scales, different assumptions of the actual alk being measured (that does alter the pH endpoint), and things like the assumed pKa for carbonic acid under different conditions all attainable in seawater.

Over the last few years, I've done numerous alk titrations using 100 mL of sample and 0.01N HCl as the titrant. I'm using a just-calibrated pH probe, and am taking data points throughout the titration, and determining the endpoint via the inflection point of the third-order polynomial of the curve. For a typical titration, where the sample is between 8 and 9 dKH, it takes between 29 and 32 mL of titrant, so the sample is being diluted by the titrant in a way that is not insignificant, which does affect the pH of the endpoint. The pH of the inflection point is typically right at 4.41, with a standard deviation of around 0.01 to 0.02 pH units.

 

[iLMaRiO]

There's some debate in the literature, or at least different mentioned values. Jim Welsh and I discussed this here once. It should hardly matter in terms of the alk difference, but it's not clear why. Maybe slightly different pH scales, different assumptions of the actual alk being measured (that does alter the pH endpoint), and things like the assumed pKa for carbonic acid under different conditions all attainable in seawater.

so, when titrating with a DIY system, if with 10ml of HCl i reach pH 4.6 and with 10.2ml i reach ph4.1 (as example) should i take the 10ml value because its closest to ph4.5 than the 10.2 value?

 

[iLMaRiO]

last question: in the formula [volume of acid added added (mL)/Volume of tank sample (mL)] x 280, what is 280 and does It change based on the normality of the used HCl ? (0.1, 0.01, 0.02, ...)

 

[Randy Holmes-Farley]

so, when titrating with a DIY system, if with 10ml of HCl i reach pH 4.6 and with 10.2ml i reach ph4.1 (as example) should i take the 10ml value because its closest to ph4.5 than the 10.2 value?

I'd assume the real value is about 10.1 ml. But the error from 10.0 vs 10.1 or 10.1 vs 10.2 is only 1%. Your other aspects of the experiment may not even be that accurate (multiple volume measurements, pH calibration, etc.)

 

[Randy Holmes-Farley]

last question: in the formula [volume of acid added added (mL)/Volume of tank sample (mL)] x 280, what is 280 and does It change based on the normality of the used HCl ? (0.1, 0.01, 0.02, ...)

Certainly the normality of the acid alters the calculations. If the acid is not 0.1 N (where the calculation came from) but is 0.2 N, the alkalintiy is twice as much.

The 280 comes from the calculations below. In equation 1, there is a factor of 100, which comes as follows:

1. The ratio of moles of acid added acid to tank water in L to reach the endpoint is the alkalinity in equivalents per L.
2. We want the value in meq/L, so we multiply the value by 1000 (1 eq = 1000 meq)
3. We are using 0,1 N HCl so the moles (equivalents) of acid added is 1/10th of the volume in L.
4. Combing the 1000 x and 1/0th X, we get a factor of 100 when using mL of 0.1 N HCl

Equation 2 is in dKH, and 1 meq/l = 2.8 dKH, so we have to multiply the meq/L result in equation 1 by 2.8 to convert dKH.

Step 4: Calculation of the Alkalinity

Take the volume of acid you added when the pH was closest to 4.5, and use one of the equations below to determine the alkalinity (depending on which units you want to get the answer in).

Equation 1: Alkalinity (meq/L) = [volume of acid added added (mL)/Volume of tank sample (mL)] x 100

Equation 2: Alkalinity (dKH) = [volume of acid added added (mL)/Volume of tank sample (mL)] x 280
For example, if you added 20 ml of 0.1 N acid to get to the endpoint using 500 mL of aquarium water, the alkalinity would be 11 dKH (which is (20/500)x280 = 11 dKH).

I have included two tables at the end of the article (Table 1 (in dKH) and Table 2 (in meq/L)) for those who worry that they may be using the equations wrong, but you can get more precise calculations from the equations when the volume added doesn’t exactly match a table entry.

 

[iLMaRiO]

I'd assume the real value is about 10.1 ml. But the error from 10.0 vs 10.1 or 10.1 vs 10.2 is only 1%. Your other aspects of the experiment may not even be that accurate (multiple volume measurements, pH calibration, etc.)

it was just an example, to understand if i can use the nearest value to pH 4.5 (even if higher than pH 4.5 like 4.6) or i have to use a lower value like pH 4.1 if more distant to 4.5 then 4.6

 

[iLMaRiO]

Certainly the normality of the acid alters the calculations. If the acid is not 0.1 N (where the calculation came from) but is 0.2 N, the alkalintiy is twice as much.

The 280 comes from the calculations below. In equation 1, there is a factor of 100, which comes as follows:

1. The ratio of moles of acid added acid to tank water in L to reach the endpoint is the alkalinity in equivalents per L.
2. We want the value in meq/L, so we multiply the value by 1000 (1 eq = 1000 meq)
3. We are using 0,1 N HCl so the moles (equivalents) of acid added is 1/10th of the volume in L.
4. Combing the 1000 x and 1/0th X, we get a factor of 100 when using mL of 0.1 N HCl

Equation 2 is in dKH, and 1 meq/l = 2.8 dKH, so we have to multiply the meq/L result in equation 1 by 2.8 to convert dKH.

Step 4: Calculation of the Alkalinity

Take the volume of acid you added when the pH was closest to 4.5, and use one of the equations below to determine the alkalinity (depending on which units you want to get the answer in).

Equation 1: Alkalinity (meq/L) = [volume of acid added added (mL)/Volume of tank sample (mL)] x 100

Equation 2: Alkalinity (dKH) = [volume of acid added added (mL)/Volume of tank sample (mL)] x 280
For example, if you added 20 ml of 0.1 N acid to get to the endpoint using 500 mL of aquarium water, the alkalinity would be 11 dKH (which is (20/500)x280 = 11 dKH).

I have included two tables at the end of the article (Table 1 (in dKH) and Table 2 (in meq/L)) for those who worry that they may be using the equations wrong, but you can get more precise calculations from the equations when the volume added doesn’t exactly match a table entry.

clear, thank you.
so, if using a 0.02 HCl, i have to multiply the meq/l for the HCl mole (ie 0.02 or divide for 50, that's the same) then multiply 1000:

1 meq/l * 0.02 / 1000

 

[Dennis Cartier]

Let us know how it works out!

OK, success! I have managed to calculate a dilution for the muriatic acid that I have, to normalize it to 0.1N. Better yet, I wrote it all down as I was doing it instead of trying to remember after I was done, lol.

Firstly, for a control for comparison, I had a partially used bottle of 0.1N Sodium Hydroxide (NaOH) on the shelf. It is expired by a couple of years, and only 40% of the bottle remained, but I figured it was better than testing against NSW or tank water.

I calculated that 19 ml of RODI & 1 ml of NaOH should be 14 dKH. I mixed it up and tested it an got 249 ppm, 13.94 dKH.

That was close enough for me to correct using the not fresh NaOH. I also forgot that I should have used RO instead of RODI. My RO water comes direct from my fridge and does not get exposed to the air, but my RODI is from a Rubbermaid and probably has some CO2 in it.

Then I created a mixture using the NaOH and the 0.1N Sulfuric acid solution that I have on hand for making Alkatronic reagent. This is what I am trying to replace. I mixed 90 ml RODI, 7.5 ml NaOH, 2.5 ml 0.1N H2SO4 and tested the mixture. I got 249 ppm, 13.94 dKH. This confirmed my planned method of testing the HCL dilutions to come.

Then I mixed up a dilution of the HCL on the assumption it was 10N. I did 59.4 ml RODI and 600 uL HCL to hopefully get me to 0.1N HCL.

I tested the 0.1N HCL candidate by mixing 7.5 ml of the NaOH with 2.5 ml of the candidate. Then I took 1 ml of the mixture along with 9 ml RODI and got 241 ppm, 13.49 dKH.

I thought that was a decent start, but I had a feeling that I had messed up during the creation of the candidate. I was pipette'ing out the 59 ml of RODI by hand, and part way through I thought I might have did a couple of bad transfers (amateur pipette user, lol).

In any case, I calculated out the correction factor and a new amount of HCL to use ( 580 uL) and mixed up a new candidate. This time I used a 50 ml volumetric flask, so I only had to do one 9 ml pipette correctly. Though I still had to do the small < 1ml pipette operations, but those are actually easy compared to larger volume ones for some reason.

I created a new mixture of NaOH and the candidate, and tested it. It came in at 258 ppm, 14.44 dKH. When I calculated the correction needed to get me to the correct 249 ppm, I came out to the original amount of 600 uL.

So by switching to the flask, it appeared that I had managed to mix up a weak, but correctly mixed, dilution.

Finally, I mixed up another candidate using the original ratios and tested it. It was 249 ppm, 13.94 dKH

The muriatic acid I have on hand (Home Hardware brand) requires 1:99 dilution to make a 0.1N solution.

Does this sound reasonable Randy?

This is the muriatic acid that I have.

 

[iLMaRiO]

Which dKH precision do you have using a 0.1 hcl and with how much water sample ? In example, reef factory uses a 0.02 hcl. Any advantage (other than using much less reagent with the 0.1) ?

 

[dwest]

OK, success! I have managed to calculate a dilution for the muriatic acid that I have, to normalize it to 0.1N. Better yet, I wrote it all down as I was doing it instead of trying to remember after I was done, lol.

Firstly, for a control for comparison, I had a partially used bottle of 0.1N Sodium Hydroxide (NaOH) on the shelf. It is expired by a couple of years, and only 40% of the bottle remained, but I figured it was better than testing against NSW or tank water.

I calculated that 19 ml of RODI & 1 ml of NaOH should be 14 dKH. I mixed it up and tested it an got 249 ppm, 13.94 dKH.

That was close enough for me to correct using the not fresh NaOH. I also forgot that I should have used RO instead of RODI. My RO water comes direct from my fridge and does not get exposed to the air, but my RODI is from a Rubbermaid and probably has some CO2 in it.

Then I created a mixture using the NaOH and the 0.1N Sulfuric acid solution that I have on hand for making Alkatronic reagent. This is what I am trying to replace. I mixed 90 ml RODI, 7.5 ml NaOH, 2.5 ml 0.1N H2SO4 and tested the mixture. I got 249 ppm, 13.94 dKH. This confirmed my planned method of testing the HCL dilutions to come.

Then I mixed up a dilution of the HCL on the assumption it was 10N. I did 59.4 ml RODI and 600 uL HCL to hopefully get me to 0.1N HCL.

I tested the 0.1N HCL candidate by mixing 7.5 ml of the NaOH with 2.5 ml of the candidate. Then I took 1 ml of the mixture along with 9 ml RODI and got 241 ppm, 13.49 dKH.

I thought that was a decent start, but I had a feeling that I had messed up during the creation of the candidate. I was pipette'ing out the 59 ml of RODI by hand, and part way through I thought I might have did a couple of bad transfers (amateur pipette user, lol).

In any case, I calculated out the correction factor and a new amount of HCL to use ( 580 uL) and mixed up a new candidate. This time I used a 50 ml volumetric flask, so I only had to do one 9 ml pipette correctly. Though I still had to do the small < 1ml pipette operations, but those are actually easy compared to larger volume ones for some reason.

I created a new mixture of NaOH and the candidate, and tested it. It came in at 258 ppm, 14.44 dKH. When I calculated the correction needed to get me to the correct 249 ppm, I came out to the original amount of 600 uL.

So by switching to the flask, it appeared that I had managed to mix up a weak, but correctly mixed, dilution.

Finally, I mixed up another candidate using the original ratios and tested it. It was 249 ppm, 13.94 dKH

The muriatic acid I have on hand (Home Hardware brand) requires 99:1 dilution to make a 0.1N solution.

Does this sound reasonable Randy?

This is the muriatic acid that I have.

I taught AP chemistry as part of my second career. We used NaOH to standardize Muriatic acid from Home Depot (that was about 1/100 of the price the school bought it for). We always got about 10M which agrees with your findings.

 

[Dennis Cartier]

Which dKH precision do you have using a 0.1 hcl and with how much water sample ? In example, reef factory uses a 0.02 hcl. Any advantage (other than using much less reagent with the 0.1) ?

I am not sure if this question was to me or Randy. If you are wondering about me mixing up a 0.1N solution for my Alkatronic, the Alkatronic also uses 0.02N reagent (like reef factory).

The undiluted reagent Focustronic sells is 0.1N, and you dilute it 1:4 for actual use in the device.

 

[iLMaRiO]

I am not sure if this question was to me or Randy. If you are wondering about me mixing up a 0.1N solution for my Alkatronic, the Alkatronic also uses 0.02N reagent (like reef factory).

The undiluted reagent Focustronic sells is 0.1N, and you dilute it 1:4 for actual use in the device.

It was for you.

moving from 0.02 to 0.1 does reduce the precision?
you would use a lot less amount of reagent, but what about the dkh ?

 

[Dennis Cartier]

It was for you.

moving from 0.02 to 0.1 does reduce the precision?
you would use a lot less amount of reagent, but what about the dkh ?

Hmm, as I am attempting to match the OEM reagent that the Alkatronic requires, I can't really say what the effect of using a 0.1N standard would be over the 0.02N standard that they, and other OEM's have settled on using.

Putting on my DIY hacker hat, I would guess that the reason they have settled on using 0.02N is because they have 5 times the volume of reagent to work with over 0.1N, and that volume falls within the acceptable range of dosing precision that the Kamoer stepper head that they use to dose the reagent can achieve. The resulting precision is quite good in my experience.

I can give you an example of their precision. I use my Alkatronic readings to modulate the effluent flow through my CaRx. I recently had an issue where the process that was doing the modulation lost contact with my Alkatronic for a couple of days (unexpected side effect of some firmware updates in my networking equipment). In the Alkatronic screen below, you can see where the contact was lost before Jan 29th, and continued through part of Jan 30th. On those days, my analog backup fail safe kicked in and turned off the CaRx effluent completely on the 29th, and I re-established the connection part way through the 30th.

The exponential back off that I use to modulate overshoot of my alk target (8.6), works fairly well to prevent serious alk spikes, but it tends to be fairly heavy handed, causing the effluent flow rate to be backed off too much and cause the flow rate to ping pong between too much and too little to maintain a steady alk rate during these overshoots.

On the Feb 1st (in the graph) I decided to add another gentler damping function in addition to the existing exponential version. The plan was for the new damping to be able to gradual remove flow when the target is exceeded, and to gradual unwind the damping and add flow when the alk is below the alk target. The caveat is that my new gentle damping would only work well if the Alkatronic can provide alk readings precise enough for the process to function effectively.

On Feb 2nd in the graph below, you can see new damping function start to impact the alk readings, making them less saw tooth than they would have been in the past after a substantial disruption in the effluent flow rate. By Feb 3rd, the damping function has taken up the excess in the flow rate, and is now sync'd up and effectively bending the alk reading down, back towards the target of 8.6. Each of the alk readings is taken 6 hours apart, and the Alkatronic is able to provide readings precise enough, using the .02N reagent, to achieve the desired control, and allow the damping function to flat line the chart.

So in my view, the Alkatronic is quite precise, while working with the volumes that the .02N standard entails during a test.

 

[Randy Holmes-Farley]

It was for you.

moving from 0.02 to 0.1 does reduce the precision?
you would use a lot less amount of reagent, but what about the dkh ?

That's a complicated question that may depend on the details (such as which is more limiting/variable, the known acidity of the acid used, the volume measurement of the titrant, the accuracy of the pH measurement, or the human or machine interpretation of when the endpoint is reached.

Diluting the acid impacts all of these in different ways, except accuracy of the pH measurement. The acidity becomes less well known by diluting more, the volume measurement may become better measured, and the endpoint may be harder to discern.