Phosphate testing

[JonasRoman]

I've ran into this a few times, my Hanna & Salifert results are always higher than what I get back on ICP. I'm no chemist but i've always had a suspicion that nutrients are getting consumed by the bacteria in the water on the 2 week ride to testing.

Doesn't matter if po4 is consumed as icp measure P, not po4, and thus is not sensitive to consumption of po4.

 

[BuffaloLake]

I have a 310 gallon system running the full Triton method. I have a Trident NP on the system, backed up by Hanna checkers. I had been getting wildly different results on the readings. So I recorded the Trident results, then ran 3 Hanna tests, then took a sample for an ICP. These are my readings from that test:
Trident NP 0.22
Hanna ULR Phosphate average 0.39
Triton ICP 0.07

Nitrate was also wildly different between Hanna and Trident. But I didn’t see nitrate listed on the ICP

Kinda at a loss on what I can trust.

ICP measures P not PO4. What was measured or reported?

Phosphorus came back 25.00

Phosphate at 0.077

Per Triton ICP

 

[BuffaloLake]

I have a 310 gallon system running the full Triton method. I have a Trident NP on the system, backed up by Hanna checkers. I had been getting wildly different results on the readings. So I recorded the Trident results, then ran 3 Hanna tests, then took a sample for an ICP. These are my readings from that test:
Trident NP 0.22
Hanna ULR Phosphate average 0.39
Triton ICP 0.07

Nitrate was also wildly different between Hanna and Trident. But I didn’t see nitrate listed on the ICP

Kinda at a loss on what I can trust.

ICP measures P not PO4. What was measured or reported?

 

[JonasRoman]

Phosphorus came back 25.00

Phosphate at 0.077

Per Triton ICP

PO4 is not measured by triton. It's just calculated from 25 ppb Phosphorus.

 

[Randy Holmes-Farley Verified Community Expert]

No Randy, Triton do not measure true po4 with photometry but P with icp and then calculate po4, so it shouldn't be time sensitive as they measure just P atoms, which can't dissappear/consumed.

If it gets into organisms or precipitates on the container walls, those would not make it to the icp plasma.

 

[EnterName]

If it gets into organisms or precipitates on the container walls, those would not make it to the icp plasma.

Never thought about that, but it makes sense... However, is there enough surface area on these plastic vials to actually reduce phosphate meaningfully?

... another good reason why these kits should probably include sterile filters.

 

[Randy Holmes-Farley Verified Community Expert]

Never thought about that, but it makes sense... However, is there enough surface area on these plastic vials to actually reduce phosphate meaningfully?

... another good reason why these kits should probably include sterile filters

I’ve never tried to quantitatively determine an answer to that.

 

[Dan_P]

Phosphorus came back 25.00

Phosphate at 0.077

Per Triton ICP

Thanks for the clarification.

 

[Dan_P]

I’ve never tried to quantitatively determine an answer to that.

You could monitor the PO4 concentration of a sterile sample of aquarium water in a plastic sample tube (15 mL centrifuge tube). To increase the surface area, add pieces of cut up plastic tube. I tried this several years ago but don’t recall what I observed. We had this discussion before when there were more experimenters online. This is where an AI agent would be helpful, recalling past conversations.

 

[EnterName]

Would be interesting to know if the ICP labs acidify their tubes prior to testing to avoid the precipitation issue and maybe even the potential biofilm growth on the tube walls.

Interestingly the effect on metals seems to be high enough that there are US regulations in place. US rules say a water sample may be shipped without acid preservation, but acid must be added at least 24 hours before analysis to dissolve any metals that adsorb to the container walls. (These regulations will not apply to aquarium samples, but it shows that this is a concern):

• https://www.deq.nc.gov/metals-epa-2008-checklist-pdf/open
• https://www.health.mn.gov/communities/environment/mnelap/docs/samplehold.pdf

 

[JonasRoman]

If it gets into organisms or precipitates on the container walls, those would not make it to the icp plasma.

Really? Sure on that?
Precipitation on glass?, think that's very unlikely to be a significant reason be honest.

It's converting to organic phosphorus(bacterial growth, convertingto organism) may happen but that should be released as P atoms in the 10000K heated plasma.

Filtering water may help as that can prevent bacteria to come into sample. But if you do not do that I have showed with comparison between photometric measure (thus true po4 measure) and icp calculated that icp calculated is almost always some false high as organic P will be released in heated plasma.

 

[Clarkjw2002]

Really? Sure on that?
Precipitation on glass?, think that's very unlikely to be a significant reason be honest.

Maybe not precipitation but I would imagine the surface area would be a colonization zone for the bacteria consuming the PO4.

It's converting to organic phosphorus(bacterial growth, convertingto organism) may happen but that should be released as P atoms in the 10000K heated plasma.

This assumes that the said bacteria that colonized the surface and consumed the PO4 are washed into suspension before processing.

No idea what the protocols are for these tests so I may be completely off base here. Either way, I would just pick one test, standardize my process, and stick with it.

I use the Hanna and have always wondered how much variability the powder adds to my results. I pre-dump it onto folded filter paper, then use a funnel to add to the sample but I'm pretty sure I never get 100% of the powder into the sample.

 

[JonasRoman]

Maybe not precipitation but I would imagine the surface area would be a colonization zone for the bacteria consuming the PO4.

This assumes that the said bacteria that colonized the surface and consumed the PO4 are washed into suspension before processing.

No idea what the protocols are for these tests so I may be completely off base here. Either way, I would just pick one test, standardize my process, and stick with it.

I use the Hanna and have always wondered how much variability the powder adds to my results. I pre-dump it onto folded filter paper, then use a funnel to add to the sample but I'm pretty sure I never get 100% of the powder into the sample.

To be honest I do not believe in significant precipitation that makes the po4 stay in the sample vial. Precipitates will also release all P in 10000K and on vial surface there will be no significant precipitation.
And all po4 that have been transformed as organic P like inside bacterial cells, algue or whatever will also be 100 % released in the 10000K plasma. So icp oes is not sensitive to transport or time in sample.

So back to thread: its very unlikely an icp is that much off, especially not in that direction, so main hypothesis must be its your manual tests and Trident NP that is off. BUT To be sure, measure with a ref fluid. If you after thar find the manual tests ARE correct, then we have just a seldom exception where something went wrong in that particular ICP test.

I have sent in many icp tests and always compare the icp calculated po4 with true photometric measurements. If you filter the water with 0
22û then no bacteria will come into sample. During those circumstances the calculated icp po4 is close to true photometric po4.
With no filtering the difference can be more, like 100% difference between true po4 and icp calculated as in those cases bacteria will consume true po4 but icp calculated will ofcourse still be the same as icp measure p atoms.

 

[Aqua Ultraviolet]

I have a 310 gallon system running the full Triton method. I have a Trident NP on the system, backed up by Hanna checkers. I had been getting wildly different results on the readings. So I recorded the Trident results, then ran 3 Hanna tests, then took a sample for an ICP. These are my readings from that test:
Trident NP 0.22
Hanna ULR Phosphate average 0.39
Triton ICP 0.07

Nitrate was also wildly different between Hanna and Trident. But I didn’t see nitrate listed on the ICP

Kinda at a loss on what I can trust.

I would not try to make those three numbers agree exactly. Different methods can measure different forms of phosphate and have different accuracy ranges, especially at low concentrations. Your three Hanna readings are useful for establishing a trend, while ICP is better treated as an independent reference point rather than a day to day control test.

For nitrate, I would pick one reliable test and use it consistently rather than chasing differences between instruments. Run the same sample several times, follow the timing and reagent instructions carefully, and compare results over several days. For a Triton system, consistency and trends are usually more useful than trying to hit one exact number from different testing methods.

 

[Randy Holmes-Farley Verified Community Expert]

Really? Sure on that?
Precipitation on glass?, think that's very unlikely to be a significant reason be honest.

It's converting to organic phosphorus(bacterial growth, convertingto organism) may happen but that should be released as P atoms in the 10000K heated plasma.

Filtering water may help as that can prevent bacteria to come into sample. But if you do not do that I have showed with comparison between photometric measure (thus true po4 measure) and icp calculated that icp calculated is almost always some false high as organic P will be released in heated plasma.

let’s see what an icp systems expert believes is important.

@Christoph

 

[BuffaloLake]

Maybe not precipitation but I would imagine the surface area would be a colonization zone for the bacteria consuming the PO4.

This assumes that the said bacteria that colonized the surface and consumed the PO4 are washed into suspension before processing.

No idea what the protocols are for these tests so I may be completely off base here. Either way, I would just pick one test, standardize my process, and stick with it.

I use the Hanna and have always wondered how much variability the powder adds to my results. I pre-dump it onto folded filter paper, then use a funnel to add to the sample but I'm pretty sure I never get 100% of the powder into the sample.

To be honest I do not believe in significant precipitation that makes the po4 stay in the sample vial. Precipitates will also release all P in 10000K and on vial surface there will be no significant precipitation.
And all po4 that have been transformed as organic P like inside bacterial cells, algue or whatever will also be 100 % released in the 10000K plasma. So icp oes is not sensitive to transport or time in sample.

So back to thread: its very unlikely an icp is that much off, especially not in that direction, so main hypothesis must be its your manual tests and Trident NP that is off. BUT To be sure, measure with a ref fluid. If you after thar find the manual tests ARE correct, then we have just a seldom exception where something went wrong in that particular ICP test.

I have sent in many icp tests and always compare the icp calculated po4 with true photometric measurements. If you filter the water with 0
22û then no bacteria will come into sample. During those circumstances the calculated icp po4 is close to true photometric po4.
With no filtering the difference can be more, like 100% difference between true po4 and icp calculated as in those cases bacteria will consume true po4 but icp calculated will ofcourse still be the same as icp measure p atoms.

I wonder if it was just an anomaly with the ICP. Previous ICP results showed Phosphate much higher, and closer to the manual tests. I have been using a Trident NP, backed up by the Hanna checkers. After a recent reagent change in the Trident, its readings were off. Nitrate reading high, phosphate low compared to the Hanna. Since I was due for an ICP anyway, I used the opportunity to calibrate the Trident. I pulled a water sample, ran the Trident on it, ran three Hanna tests, then filled the ICP vials. Afterwards I calibrated the Trident using the Hanna values, and waited on the ICP. I was just shocked at how different the ICP values were.