Measuring Organic Matter. A Brief Study

[EnterName]

Another day, another experiment:
To make the calibration curve I decided to use 4 different dilutions of KMnO₄:

• 390ppm "100%"
• 292.5ppm "75%"
• 195ppm "50%"
• 97.5ppm "25%"

The sample preparation went as usual:

Group A (tank water):

• Sample A₁: 10mL tank water (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 390ppm KMnO₄ = 780µg KMnO₄
• Sample A₂: 10mL tank water (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 292.5ppm KMnO₄ = 585µg KMnO₄
• Sample A₃: 10mL tank water (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 195ppm KMnO₄ = 390µg KMnO₄
• Sample A₄: 10mL tank water (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 97.5ppm KMnO₄ = 195µg KMnO₄

Group B (freshly mixed 35ppt nyos pure saltwater):

• Sample B₁: 10mL fresh saltwater (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 390ppm KMnO₄ = 780µg KMnO₄
• Sample B₂: 10mL fresh saltwater (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 292.5ppm KMnO₄ = 585µg KMnO₄
• Sample B₃: 10mL fresh saltwater (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 195ppm KMnO₄ = 390µg KMnO₄
• Sample B₄: 10mL fresh saltwater (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 97.5ppm KMnO₄ = 195µg KMnO₄

Results:

• Samples A₁, A₂, A₃, A₄: 1.92, 1.39, 0.79, 0.18
• Samples B₁, B₂, B₃, B₄: 2.30, 1.71, 1.05, 0.43

​

Linear Regression Group A:
y = 0.005969x - 0.3850
P Value: 0.0005
R Square: 0.9990

Linear Regression Group B:
y = 0.006431x - 0.1950
P Value: 0.0002
R Square: 0.9996

Now what @Dan_P ? :D

 

[Dan_P]

Another day, another experiment:
To make the calibration curve I decided to use 4 different dilutions of KMnO₄:

• 390ppm "100%"
• 292.5ppm "75%"
• 195ppm "50%"
• 97.5ppm "25%"

The sample preparation went as usual:

Group A (tank water):

• Sample A₁: 10mL tank water (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 390ppm KMnO₄ = 780µg KMnO₄
• Sample A₂: 10mL tank water (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 292.5ppm KMnO₄ = 585µg KMnO₄
• Sample A₃: 10mL tank water (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 195ppm KMnO₄ = 390µg KMnO₄
• Sample A₄: 10mL tank water (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 97.5ppm KMnO₄ = 195µg KMnO₄

Group B (freshly mixed 35ppt nyos pure saltwater):

• Sample B₁: 10mL saltwater (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 390ppm KMnO₄ = 780µg KMnO₄
• Sample B₂: 10mL saltwater (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 292.5ppm KMnO₄ = 585µg KMnO₄
• Sample B₃: 10mL saltwater (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 195ppm KMnO₄ = 390µg KMnO₄
• Sample B₄: 10mL saltwater (unfiltered), 2mL ⋅ 2% NaOH, 2mL ⋅ 97.5ppm KMnO₄ = 195µg KMnO₄

Results:

• Samples A₁, A₂, A₃, A₄: 1.92, 1.39, 0.79, 0.18
• Samples B₁, B₂, B₃, B₄: 2.30, 1.71, 1.05, 0.43

​

Linear Regression Group A:
y = 0.005969x - 0.3850
P Value: 0.0005
R Square: 0.9990

Linear Regression Group B:
y = 0.006431x - 0.1950
P Value: 0.0002
R Square: 0.9996

Now what @Dan_P ? :D

Were these heated or unheated samples? While waiting for your answer I will think about your results :-)

 

[EnterName]

Were these heated or unheated samples? While waiting for your answer I will think about your results :-)

Heated,

all samples have gone through the full process:

1. take sample,
2. add NaOH,
3. shake,
4. add KMnO₄,
5. shake,
6. heat to 100°C for a while,
7. let cool,
8. centrifuge,
9. zero HannaChecker,
10. measure sample

 

[Dan_P]

Heated,

all samples have gone through the full process:

1. take sample,
2. add NaOH,
3. shake,
4. add KMnO₄,
5. shake,
6. heat to 100°C for a while,
7. let cool,
8. centrifuge,
9. zero HannaChecker,
10. measure sample

The heating step might be consistently executed by placing the tube in a hot water bath at 100 C for ten minutes. Ten vs thirty minutes did not make much difference pick a time and stay with it.

Some observations on your data.

The blank curve (fresh saltwater, orange line).
The Hanna Checker response is linear.The negative intercept indicates there is some loss of permanganate from just heating and/or reacting with something in the fresh salter mix. The color intensity of a non-heated sample should be higher by this amount (roughly). Even though you have this baseline, it’s a good idea to run a fresh saltwater blank in every run to compare the test sample color intensity against in case of baseline drift.

Aquarium water curve (blue line)

As expected, these samples lost color intensity consistent with an elevated DOC. The loss is relatively independent of the permanganate concentration. I don’t understand why the difference with the baseline Is smaller than your previous test.

 

[EnterName]

Heated,

all samples have gone through the full process:

1. take sample,
2. add NaOH,
3. shake,
4. add KMnO₄,
5. shake,
6. heat to 100°C for a while,
7. let cool,
8. centrifuge,
9. zero HannaChecker,
10. measure sample

The heating step might be consistently executed by placing the tube in a hot water bath at 100 C for ten minutes. Ten vs thirty minutes did not make much difference pick a time and stay with it.

Some observations on your data.

The blank curve (fresh saltwater, orange line).
The Hanna Checker response is linear.The negative intercept indicates there is some loss of permanganate from just heating and/or reacting with something in the fresh salter mix. The color intensity of a non-heated sample should be higher by this amount (roughly). Even though you have this baseline, it’s a good idea to run a fresh saltwater blank in every run to compare the test sample color intensity against in case of baseline drift.

Aquarium water curve (blue line)

As expected, these samples lost color intensity consistent with an elevated DOC. The loss is relatively independent of the permanganate concentration. I don’t understand why the difference with the baseline Is smaller than your previous test.

That's a very fair point. The heating step was something I should have done more consistently between experiments. I will improve that.

Comparison of Experiments:
For unheated freshly mixed saltwater samples (2mL 390ppm KMnO₄, no NaOH) I got an average reading of 3.02. Today including the heating step, I got 2.30. That's a larger difference than freshly mixed water vs tank water. So the heating step really seems to significantly impact the results (or it's the NaOH which I left out for the unheated sample).

Experiment 3 on sunday was equivalent to sample A₁ in today's experiment. I got 1.90 instead of today's 1.92. That's within margin of error. I would ignore the results of earlier experiments as the sample size changed a bit. This might happen again once the pipettes arrive and I can finally consistently dose 2 - 10mL volumes without relying on syringe markings and eyesight.

Next Steps:
If I understood you correctly, you suggest I should repeat the procedure from today, but with 3 groups instead of 2:

Group A: unheated, freshly mixed (I will try to include the NaOH this time to be sure... I hope I can get rid of the cloudyness through the centrifuge. I could also do a 4th group unheated, without NaOH if that's of interest)
Group B: heated, freshly mixed
Group C: heated, tank water

 

[Dan_P]

That's a very fair point. The heating step was something I should have done more consistently between experiments. I will improve that.

Comparison of Experiments:
For unheated freshly mixed saltwater samples (2mL 390ppm KMnO₄, no NaOH) I got an average reading of 3.02. Today including the heating step, I got 2.30. That's a larger difference than freshly mixed water vs tank water. So the heating step really seems to significantly impact the results (or it's the NaOH which I left out for the unheated sample).

Experiment 3 on sunday was equivalent to sample A₁ in today's experiment. I got 1.90 instead of today's 1.92. That's within margin of error. I would ignore the results of earlier experiments as the sample size changed a bit. This might happen again once the pipettes arrive and I can finally consistently dose 2 - 10mL volumes without relying on syringe markings and eyesight.

Next Steps:
If I understood you correctly, you suggest I should repeat the procedure from today, but with 3 groups instead of 2:

Group A: unheated, freshly mixed (I will try to include the NaOH this time to be sure... I hope I can get rid of the cloudyness through the centrifuge. I could also do a 4th group unheated, without NaOH if that's of interest)
Group B: heated, freshly mixed
Group C: heated, tank water

Group A looks good.

Did you make a nonheated mixture of fresh saltwater, permanganate and replace the sodium hydroxide with water? That mixture tells us how high pH changes the permanganate concentration. You only need one concentration of permanganate.

 

[EnterName]

That's a very fair point. The heating step was something I should have done more consistently between experiments. I will improve that.

Comparison of Experiments:
For unheated freshly mixed saltwater samples (2mL 390ppm KMnO₄, no NaOH) I got an average reading of 3.02. Today including the heating step, I got 2.30. That's a larger difference than freshly mixed water vs tank water. So the heating step really seems to significantly impact the results (or it's the NaOH which I left out for the unheated sample).

Experiment 3 on sunday was equivalent to sample A₁ in today's experiment. I got 1.90 instead of today's 1.92. That's within margin of error. I would ignore the results of earlier experiments as the sample size changed a bit. This might happen again once the pipettes arrive and I can finally consistently dose 2 - 10mL volumes without relying on syringe markings and eyesight.

Next Steps:
If I understood you correctly, you suggest I should repeat the procedure from today, but with 3 groups instead of 2:

Group A: unheated, freshly mixed (I will try to include the NaOH this time to be sure... I hope I can get rid of the cloudyness through the centrifuge. I could also do a 4th group unheated, without NaOH if that's of interest)
Group B: heated, freshly mixed
Group C: heated, tank water

Group A looks good.

Did you make a nonheated mixture of fresh saltwater, permanganate and replace the sodium hydroxide with water? That mixture tells us how high pH changes the permanganate concentration. You only need one concentration of permanganate.

Unfortunately I didn't prepare such samples in a comparable way yet, but I will do so :)

I would suggest 3 samples:
- 10mL freshly mixed saltwater + 2mL 2% NaOH + 2mL 390ppm KMnO₄
- 10mL freshly mixed saltwater + 2mL RO/DI + 2mL 390ppm KMnO₄
- 12mL freshly mixed saltwater + 2mL 390ppm KMnO₄

I can prepare all of them in heated and unheated versions and will report back once I have the results :)

 

[EnterName]

Experiment Parameters:
Group A: 10mL 35ppt freshly mixed saltwater + 2mL 2% NaOH + 2mL 390ppm KMnO₄
Group B: 10mL 35ppt freshly mixed saltwater + 2mL RO/DI + 2mL 390ppm KMnO₄
Group C: 12mL 35ppt freshly mixed saltwater + 2mL 390ppm KMnO₄

Results (Heated | Unheated):
Group A: 2.35 | 2.49
Group B: 2.40 | 2.57
Group C: 2.47 | 2.57

Notes:
Yesterday the heated sample of Group A showed 2.30 instead of 2.35, but 0.05 is the standard error with my current methodology, so results seem relatively consistent.

I did a similar run as Group B & C just without adding RO/DI or more saltwater to compensate for the missing NaOH on sunday (Experiment 2). The results ranged from 2.84 to 3.14 and averaged to 3.02. The sample size differed by 14% due to the additional 2mL RO/DI or saltwater that were added this time to compensate for the missing NaOH.
This means the readings of 2.57 today are quite consistent with the readings from sunday, if we simply assume a 14% stronger dilution of the found 3.02 average:

3.02 ⋅ (100% - 14%) ≈ 2.59​

 

[Dan_P]

Experiment Parameters:
Group A: 10mL 35ppt freshly mixed saltwater + 2mL 2% NaOH + 2mL 390ppm KMnO₄
Group B: 10mL 35ppt freshly mixed saltwater + 2mL RO/DI + 2mL 390ppm KMnO₄
Group C: 12mL 35ppt freshly mixed saltwater + 2mL 390ppm KMnO₄

Results (Heated | Unheated):
Group A: 2.35 | 2.49
Group B: 2.40 | 2.57
Group C: 2.47 | 2.57

Notes:
Yesterday the heated sample of Group A showed 2.30 instead of 2.35, but 0.05 is the standard error with my current methodology, so results seem relatively consistent.

I did a similar run as Group B & C just without adding RO/DI or more saltwater to compensate for the missing NaOH on sunday (Experiment 2). The results ranged from 2.84 to 3.14 and averaged to 3.02. The sample size differed by 14% due to the additional 2mL RO/DI or saltwater that were added this time to compensate for the missing NaOH.
This means the readings of 2.57 today are quite consistent with the readings from sunday, if we simply assume a 14% stronger dilution of the found 3.02 average:

3.02 ⋅ (100% - 14%) ≈ 2.59​

Now that you have this method under control, what’s next?

 

[EnterName]

Experiment Parameters:
Group A: 10mL 35ppt freshly mixed saltwater + 2mL 2% NaOH + 2mL 390ppm KMnO₄
Group B: 10mL 35ppt freshly mixed saltwater + 2mL RO/DI + 2mL 390ppm KMnO₄
Group C: 12mL 35ppt freshly mixed saltwater + 2mL 390ppm KMnO₄

Results (Heated | Unheated):
Group A: 2.35 | 2.49
Group B: 2.40 | 2.57
Group C: 2.47 | 2.57

Notes:
Yesterday the heated sample of Group A showed 2.30 instead of 2.35, but 0.05 is the standard error with my current methodology, so results seem relatively consistent.

I did a similar run as Group B & C just without adding RO/DI or more saltwater to compensate for the missing NaOH on sunday (Experiment 2). The results ranged from 2.84 to 3.14 and averaged to 3.02. The sample size differed by 14% due to the additional 2mL RO/DI or saltwater that were added this time to compensate for the missing NaOH.
This means the readings of 2.57 today are quite consistent with the readings from sunday, if we simply assume a 14% stronger dilution of the found 3.02 average:

3.02 ⋅ (100% - 14%) ≈ 2.59​

Now that you have this method under control, what’s next?

You tell me, you are the expert, it's your experiment :D

BOD5 test?

I feel like I'm not fully understanding how to translate the readings into a meaningful unit.

Sure, I know what "0 organics" look like (simply by measuring freshly mixed saltwater), and I can track changes in my tank over time. But I don't know at which point I should try to reduce organics and at which point increasing them would be better.

 

[Dan_P]

You tell me, you are the expert, it's your experiment :D

BOD5 test?

I feel like I'm not fully understanding how to translate the readings into a meaningful unit.

Sure, I know what "0 organics" look like (simply by measuring freshly mixed saltwater), and I can track changes in my tank over time. But I don't know at which point I should try to reduce organics and at which point increasing them would be better.

Your last paragraph seems to summarize where the hobby is today. There are occasional “flare ups” of interest in the subject, lately it was the link between coral health and “organics” in the water. The problem is there are no useful links for hobbyists, such as, at what level of DOC do I need to act? There is an opinion that elevated levels of “organics” are a cause or at least associated with cyanobacteria growth. The popular remedy to reduce “organics” is to pour bottles of unknown bacteria into aquarium which seem to work half the time :-)

I am following the level COD and BOD5 (among other observations and measurements) in two identical small aquaria as they develop over time. To tamp down any expectations of major discoveries, I consider this study as an exercise in learning to observe.

 

[areefer01]

Your last paragraph seems to summarize where the hobby is today. There are occasional “flare ups” of interest in the subject, lately it was the link between coral health and “organics” in the water. The problem is there are no useful links for hobbyists, such as, at what level of DOC do I need to act?

This is how I feel about the tests be it organic, icp, and dna. Too many variables be it the hobbyist installation, methods, husbandry, and lack of related standards, reference points (oceanic reef, locations, data points) to compare too.

We have maybe 1 quarter of that data and that is difficult to do any meaningful analysis in my opinion. We get a number of 33 and have no real understanding if that is good or bad. Do I run with my arms waving in the air and scream like Gertie meeting E.T. for the first time or just roll with it?

Then there is the whole thing of what numbers are we even using in the hobby. Seems to me that we sort of pick and choose what to use from what numbers we have from oceanic studies be it alk, ca, etc.

BTW - shoot me a pm if you are still doing the oceanmo organic data thing - my data just came in although it is not exciting in any way. Or if you are sharing the data I can also add it - do not mean to make you do work - sorry about that.

 

[Dan_P]

This is how I feel about the tests be it organic, icp, and dna. Too many variables be it the hobbyist installation, methods, husbandry, and lack of related standards, reference points (oceanic reef, locations, data points) to compare too.

We have maybe 1 quarter of that data and that is difficult to do any meaningful analysis in my opinion. We get a number of 33 and have no real understanding if that is good or bad. Do I run with my arms waving in the air and scream like Gertie meeting E.T. for the first time or just roll with it?

Then there is the whole thing of what numbers are we even using in the hobby. Seems to me that we sort of pick and choose what to use from what numbers we have from oceanic studies be it alk, ca, etc.

BTW - shoot me a pm if you are still doing the oceanmo organic data thing - my data just came in although it is not exciting in any way. Or if you are sharing the data I can also add it - do not mean to make you do work - sorry about that.

I am interested in your organo-ms data. I am still compiling and comparing over in Lasse thread on the subject of this test.

 

[EnterName]

We have maybe 1 quarter of that data and that is difficult to do any meaningful analysis in my opinion. We get a number of 33 and have no real understanding if that is good or bad. Do I run with my arms waving in the air and scream like Gertie meeting E.T. for the first time or just roll with it?

I've started performing the KMnO₄ test Dan proposed regularly to see if there is some trend over time. If we can motivate more people to do the same, we could learn a bit more about the readings we get.

The thing is: Not everyone has access to a centrifuge. But after performing the test many times I believe the centrifuge might not be necessary if you use long test tubes. The precipitate gathers at the top and bottom of the sample and it should be possible to carefully siphon out the clear part with a syringe. This would make the test a bit more accessible to everyone who can get their hands on a KMnO₄ solution.

I am interested in your organo-ms data.

I second this, as the data is also relevant for designing amino acid studies.