DIY Reference Solution for Alkalinity tests that use pH Titration

[Jamie7907]

We can try a modification of the last recipe using sodium bicarbonate instead of sodium carbonate, but it is hard to ensure sodium bicarbonate is pure sodium bicarbonate without added water or loss of CO2:

Dissolve 2.41 g sodium bicarbonate in 497.6 g of water.

Take out 100 g and add to 500 g of water. Conc ~ 0.482 grams sodium bicarbonate in 600 grams total

Take out 89.5 g and add to 210.5 g of water. Conc ~ 0.0719 grams in 300 grams total, ~0.2397 grams/L, ~ 2.853 mM ( 7.99 dKH)

Would ACS or reagent grade address the issue of using sodium bicarbonate? If so how would it need to be stored; ie will it draw moisture over time.

 

[Randy Holmes-Farley]

Would ACS or reagent grade address the issue of using sodium bicarbonate? If so how would it need to be stored; ie will it draw moisture over time.

Not necessarily. Any grade can lose CO2 and water to the air, and may have done so to some extent before you got it. .

 

[kblazewicz]

Hi Randy,

I know this is an old thread but this is where Google brought me when I searched for DIY KH reference solution so I decided to add my 2 cents here.

I find your recipe a bit too hard to follow without specialized equipment, so I came up with my own which requires fairly simple tools, that is:

• two 250 ml beakers or other vessels with 250 ml volume mark you trust (volumetric flask would be best)
• scale with 0.01 g precision (you can get one really cheap)
• a 5 ml syringe (or 1 ml for better precision)

The recipe:

1. Fill the beaker up to about 3/4 of the 250 ml mark with pure water.
2. Add 3.00 g of NaHCO3 to the water, stir until everything dissolves.
3. Fill the beaker up to the 250 ml mark.
4. Move 5 ml of solution to clean beaker using syringe.
5. Fill the beaker with pure water to 250 ml of total volume.

This should yield 250 ml of approx (see below) 8.01 dKH solution.

You can estimate error by multiplying uncertainty of each measurement (salt purity, beaker, scale and syringe). For example assuming 99% purity of sodium bicarbonate and 1% errors on all measurements total error is approx 4% (1 - 0.99^4).

With 4% error you could expect actual KH in range from 7.69 to 8.33, so more precise measurements are preferable.

I've wrote a piece of Python code which can be used to verify this method. I think it should be readable to non programmers as well.

## NaHCO3 weight in grams
m_nahco3 = 3.00
# m_nahco3 = 2.41
## solution volume in mililiters
V = 250
# V = 497.6
## dilution factor
df = 1/50
# df = 1/6 * 89.5 / (89.5 + 210.5)

## Note: NaHCO3 solubility in water is ~10 g/100 ml at 20 deg C

M_h = 1.008  # g/mol
M_c = 12.011
M_o = 15.999
M_na = 22.990
M_ca = 40.078
M_hco3 = M_h + M_c + 3 * M_o
M_cao = M_ca + M_o
M_nahco3 = M_na + M_hco3

n_nahco3 = m_nahco3 / M_nahco3  # mol
m_hco3 = n_nahco3 * M_hco3      # g

## According to definition of german degree of hardnes,
## concentration should be in milligrams per litre of water (solution).
p_hco3 = m_hco3 / V * 1_000_000  # mg/l

## after dilution concentration is reduced
p_hco3_d = p_hco3 * df

## 1 deg dH is defined as 10 mg/l CaO (https://en.wikipedia.org/wiki/DGH)
## 1 mole of CaO binds 1 mole of Ca
## 1 mole of Ca binds 2 moles of HCO3 (Ca(HCO3)2)
## so 1 mole of CaO is equivalent to 2 moles of HCO3
alk = M_cao / (2 * M_hco3) / 10 * p_hco3_d

I have tested the code on numbers from your recipe and it almost matches (I got 8.04, you got 8.01). Difference is that you used ppm as mg/kg of solution and I have used mg/l. Definition of german degree I found used mg/l.

 

[kblazewicz]

Also if you want to find other proportions you can use simplified formula:

alk = 33377 * m_nahco3 / V * df

Where mass and volume are in grams and mililiters, or kilograms and liters (both will work).

Equation can of course be converted. For example to find amount of NaHCO3 to use with 500 ml solution and 1:100 dilution ratio:

m_nahco3 = 8.0 / (33377 / 500 * 0.01)  # 11.98 g

 

[randyadammartin]

Alternatively, here's a better method using the scale you have.

EDITED DUE TO A MISTAKE

Dissolve 1.52 g in 498.5 g of water.
Take out 100 g and add to 500 g of water. Conc ~ 0.304 grams in 600 grams total,

Take out 89.5 g and add to 210.5 g of water. Conc ~ 0.0453 grams in 300 grams total, ~0.1512 grams/L, ~ 1.426 mM (8.01 dKH)

Hey @Randy Holmes-Farley ,

That’s my name as well. I followed the above instruction verbatim. Unfortunately upon testing using both a Hanna alk checker and a kh director they failed. They were out of range of both instruments of measure.

I used rodi water for the water to mix the sodium carbonate with. Should I have instead used salt water from the tank? My thoughts were if I did that I would be measuring the alakinity if the salt and not the reference fluid.

What did I do wrong ?

 

[Randy Holmes-Farley]

Hey @Randy Holmes-Farley ,

That’s my name as well. I followed the above instruction verbatim. Unfortunately upon testing using both a Hanna alk checker and a kh director they failed. They were out of range of both instruments of measure.

I used rodi water for the water to mix the sodium carbonate with. Should I have instead used salt water from the tank? My thoughts were if I did that I would be measuring the alakinity if the salt and not the reference fluid.

What did I do wrong ?

The Hanna is not an alkalinity titration like a normal kit. I think they add a specific amount of acid and look for the resulting pH by color change. . The answer may be inaccurate in fresh water using that method.

The KH director may be similar (not sure how it does its titration).

 

[fr3n0z]

Sodium carbonate has a molecular weight of 106 grams per mole.

An 8 dKH standard will contain 2.88 meq/L of sodium carbonate, and since each carbonate carries two units of alkalinity, we want half that, or 1.44 mM sodium carbonate, or 1.44 millimoles per liter.

1.44 millimoles is 1.44 millimoles x 106 mg/mmole = 152.6 milligrams per liter.

So dissolving 152.5 milligrams in a liter (1.00 kg) of water will give an 8 dKH standard.

If that is not able to be weighed out accurately (or the 1 L measured accurately), you can make a higher concentration and dilute it down, but that too potentially introduces errors, in the volume measurements.

I might also suggest baking the sodium carbonate first to be sure it has no moisture in it and it is all present as carbonate and very little as bicarbonate. 1 h at 400 deg C should be sufficient.

the water, should be distilled, de-ionised or also RODI could be good?

 

[Randy Holmes-Farley]

the water, should be distilled, de-ionised or also RODI could be good?

Any of those should be ok if it is a couple of ppm tds or less.

 

[fr3n0z]

Any of those should be ok if it is a couple of ppm tds or less.

thx !!

 

[Randy Holmes-Farley]

You're welcome.

Happy Reefing.

 

[vahegan]

The Hanna is not an alkalinity titration like a normal kit. I think they add a specific amount of acid and look for the resulting pH by color change. . The answer may be inaccurate in fresh water using that method.

The KH director may be similar (not sure how it does its titration).

I think that KH Director adds acid until the pH of the sample drops to 4.5
Not sure why the suggested reference solutions should not work with it. GHL says that, when using their own dKH7.5 calibration solution you need to run the test at least three times in a row to get the proper results - maybe the reason for failure was insufficient number of tests?

I have fixanal vials for making 0.1N solutions for both sodium carbonate and bicarbonate. I have calculated that I need to take 14.3 ml of sodium carbonate 0.1N solution, or 28.6 ml of sodium bicarbonate 0.1N solution, and make up to 1 liter in a calibrated flask, to obtain 8 dKH standard solution. I believe that the second option would be better as inaccuracy of 0.1N solution sampling is 0.28 dKH/ml, whereas it is twice that much for sodium carbonate.

I am planning to make this solution and test my KH Director accuracy, I will report back when I run the tests. I am seeing up to 0.5 dKH variation in measured KH values during the day, which I think is a bit high - I suspect that inaccuracies might be caused by poor design of their dosing heads (rotation from the motor shaft is transferred to the rolleres solely by friction, and if there is accumulated dirt/grease between the rollers and the shaft, they slip), but would like to first try to calibrate the device with a proper alkalinity standard, before looking into anything else.

 

[Randy Holmes-Farley]

I think that KH Director adds acid until the pH of the sample drops to 4.5
Not sure why the suggested reference solutions should not work with it. GHL says that, when using their own dKH7.5 calibration solution you need to run the test at least three times in a row to get the proper results - maybe the reason for failure was insufficient number of tests?

I have fixanal vials for making 0.1N solutions for both sodium carbonate and bicarbonate. I have calculated that I need to take 14.3 ml of sodium carbonate 0.1N solution, or 28.6 ml of sodium bicarbonate 0.1N solution, and make up to 1 liter in a calibrated flask, to obtain 8 dKH standard solution. I believe that the second option would be better as inaccuracy of 0.1N solution sampling is 0.28 dKH/ml, whereas it is twice that much for sodium carbonate.

I am planning to make this solution and test my KH Director accuracy, I will report back when I run the tests. I am seeing up to 0.5 dKH variation in measured KH values during the day, which I think is a bit high - I suspect that inaccuracies might be caused by poor design of their dosing heads (rotation from the motor shaft is transferred to the rolleres solely by friction, and if there is accumulated dirt/grease between the rollers and the shaft, they slip), but would like to first try to calibrate the device with a proper alkalinity standard, before looking into anything else.

That all sounds reasonable.