ICP-MS measurement of Seawater Certified Reference Material

Christoph

Active Member
Review score
+0 /0 /-0
View Badges
Joined
May 23, 2017
Messages
170
Reaction score
339
Review score
+0 /0 /-0
Location
Vienna, Austria
Dear Reef2Reefers!

I have seen a lot of skepticism regarding seawater laboratory analysis in this forum, especially when in comes to reported values for trace elements. I must say, that all skepticism is very well justified, since there is many potential pitfalls in ICP analysis: Sampling, sample handling in the laboratory, laboratory contamination, interferences, wrong interpretation of data, and many more. Those points even increase in importance when very low detection limits are crucial.

Disclaimer: I can speak here only for the procedures in our lab (Oceamo). I hope this post does not have too much of an advertisment character (@Randy Holmes-Farley, please remove it if this should be the case)

For transparency reasons and to build some confidence in trace element analysis id like to share some data with you. We have started our ICP-MS service late last year. ICP-MS is way more sensitive, so we can detect and quantify trace elements and pollutants at much lower concentration compared to ICP-OES - but ICP-MS also comes with many challenges the analytical chemist must handle to provide reliable results.

To control the quality of our data we are using the seawater certified reference material ERM-CA403 supplied by the European Comission. This is seawater, that has very clearly determined levels of certain trace elements. Im Posting the certificate of analysis here:

ERM1.PNG


erm2.PNG


This reference material is included in all our ICP-MS sequences, to check if our measurements are correct.

That you get a feeling for the level of accuracy and uncertainty you can expect, i am posting our results and the certified value for the more difficult elements here, and will also supply the calibration plots that show linearity and our limit of detection.

Lets start with Nickel:

Concentration in ERM-CA403: 1,04 µg/l (uncertainty 0,16 µg/l)
Oceamo ICP-MS measurement 1: 1,15 µg/l
Oceamo ICP-MS measurement 2: 1,17 µg/l

Ni.PNG

Cobalt:

Concentration in ERM-CA403: 0,074 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,063 µg/l
Oceamo ICP-MS measurement 2: 0,068 µg/l

Co.PNG

Copper:

Concentration in ERM-CA403: 0,87 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 0,754 µg/l
Oceamo ICP-MS measurement 2: 0,871 µg/l
Cu.PNG


Cadmium:

Concentration in ERM-CA403: 0,094 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,094 µg/l
Oceamo ICP-MS measurement 2: 0,099 µg/l

Cd.PNG

Arsenic:

Concentration in ERM-CA403: 1,90 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 1,86 µg/l
Oceamo ICP-MS measurement 2: 1,63 µg/l

As.PNG

Selenium:

Concentration in ERM-CA403: 0,06-0,094 µg/l
Oceamo ICP-MS measurement 1: 0,083 µg/l
Oceamo ICP-MS measurement 2: 0,092 µg/l

Se.PNG

Lead:

Concentration in ERM-CA403: 0,098 µg/l (uncertainty 0,010 µg/l)
Oceamo ICP-MS measurement 1: 0,114 µg/l
Oceamo ICP-MS measurement 2: 0,097 µg/l

Pb.PNG


I hope you found this information useful, and that it contributes to lift the curtains from the blackbox "seawater laboratory".

By this data you can also see, why i have decided to expand towards ICP-MS: The levels measured here are all within the relevant concentration range of those trace elements in seawater (and thus also in our aquaria). - But are at the same time below the realistic limit of detection for ICP-OES, and thus would appear on the analysis report as "non detectable". ICP-OES (which we also frequently use) is a great tool for many elements that appear in higher concentration ranges, and to (in many cases) detect elevated levels - but fails to reliably diagnose (too) low levels for many essential trace elements. This is where ICP-MS fills the void.

If youre having any questions i am happy to help!
Best regards,
Christoph
 
Corals.com

Woodyman

Valuable Member
Review score
+0 /0 /-0
View Badges
Joined
May 25, 2021
Messages
2,146
Reaction score
2,678
Review score
+0 /0 /-0
Location
USA
Dear Reef2Reefers!

I have seen a lot of skepticism regarding seawater laboratory analysis in this forum, especially when in comes to reported values for trace elements. I must say, that all skepticism is very well justified, since there is many potential pitfalls in ICP analysis: Sampling, sample handling in the laboratory, laboratory contamination, interferences, wrong interpretation of data, and many more. Those points even increase in importance when very low detection limits are crucial.

Disclaimer: I can speak here only for the procedures in our lab (Oceamo). I hope this post does not have too much of an advertisment character (@Randy Holmes-Farley, please remove it if this should be the case)

For transparency reasons and to build some confidence in trace element analysis id like to share some data with you. We have started our ICP-MS service late last year. ICP-MS is way more sensitive, so we can detect and quantify trace elements and pollutants at much lower concentration compared to ICP-OES - but ICP-MS also comes with many challenges the analytical chemist must handle to provide reliable results.

To control the quality of our data we are using the seawater certified reference material ERM-CA403 supplied by the European Comission. This is seawater, that has very clearly determined levels of certain trace elements. Im Posting the certificate of analysis here:

ERM1.PNG


erm2.PNG


This reference material is included in all our ICP-MS sequences, to check if our measurements are correct.

That you get a feeling for the level of accuracy and uncertainty you can expect, i am posting our results and the certified value for the more difficult elements here, and will also supply the calibration plots that show linearity and our limit of detection.

Lets start with Nickel:

Concentration in ERM-CA403: 1,04 µg/l (uncertainty 0,16 µg/l)
Oceamo ICP-MS measurement 1: 1,15 µg/l
Oceamo ICP-MS measurement 2: 1,17 µg/l

Ni.PNG

Cobalt:

Concentration in ERM-CA403: 0,074 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,063 µg/l
Oceamo ICP-MS measurement 2: 0,068 µg/l

Co.PNG

Copper:

Concentration in ERM-CA403: 0,87 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 0,754 µg/l
Oceamo ICP-MS measurement 2: 0,871 µg/l
Cu.PNG


Cadmium:

Concentration in ERM-CA403: 0,094 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,094 µg/l
Oceamo ICP-MS measurement 2: 0,099 µg/l

Cd.PNG

Arsenic:

Concentration in ERM-CA403: 1,90 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 1,86 µg/l
Oceamo ICP-MS measurement 2: 1,63 µg/l

As.PNG

Selenium:

Concentration in ERM-CA403: 0,06-0,094 µg/l
Oceamo ICP-MS measurement 1: 0,083 µg/l
Oceamo ICP-MS measurement 2: 0,092 µg/l

Se.PNG

Lead:

Concentration in ERM-CA403: 0,098 µg/l (uncertainty 0,010 µg/l)
Oceamo ICP-MS measurement 1: 0,114 µg/l
Oceamo ICP-MS measurement 2: 0,097 µg/l

Pb.PNG


I hope you found this information useful, and that it contributes to lift the curtains from the blackbox "seawater laboratory".

By this data you can also see, why i have decided to expand towards ICP-MS: The levels measured here are all within the relevant concentration range of those trace elements in seawater (and thus also in our aquaria). - But are at the same time below the realistic limit of detection for ICP-OES, and thus would appear on the analysis report as "non detectable". ICP-OES (which we also frequently use) is a great tool for many elements that appear in higher concentration ranges, and to (in many cases) detect elevated levels - but fails to reliably diagnose (too) low levels for many essential trace elements. This is where ICP-MS fills the void.

If youre having any questions i am happy to help!
Best regards,
Christoph

Thank you for the data.

Being in quality this is nice to know you run reference standards with your analysis. It would be nice to see other companies provide the same in the future!

Are you guys certified in any capacity? ISO, etc.?
 

Rick Mathew

Valuable Member
Review score
+0 /0 /-0
View Badges
Joined
Mar 21, 2016
Messages
1,257
Reaction score
4,250
Review score
+0 /0 /-0
Location
North Central Florida
Dear Reef2Reefers!

I have seen a lot of skepticism regarding seawater laboratory analysis in this forum, especially when in comes to reported values for trace elements. I must say, that all skepticism is very well justified, since there is many potential pitfalls in ICP analysis: Sampling, sample handling in the laboratory, laboratory contamination, interferences, wrong interpretation of data, and many more. Those points even increase in importance when very low detection limits are crucial.

Disclaimer: I can speak here only for the procedures in our lab (Oceamo). I hope this post does not have too much of an advertisment character (@Randy Holmes-Farley, please remove it if this should be the case)

For transparency reasons and to build some confidence in trace element analysis id like to share some data with you. We have started our ICP-MS service late last year. ICP-MS is way more sensitive, so we can detect and quantify trace elements and pollutants at much lower concentration compared to ICP-OES - but ICP-MS also comes with many challenges the analytical chemist must handle to provide reliable results.

To control the quality of our data we are using the seawater certified reference material ERM-CA403 supplied by the European Comission. This is seawater, that has very clearly determined levels of certain trace elements. Im Posting the certificate of analysis here:

ERM1.PNG


erm2.PNG


This reference material is included in all our ICP-MS sequences, to check if our measurements are correct.

That you get a feeling for the level of accuracy and uncertainty you can expect, i am posting our results and the certified value for the more difficult elements here, and will also supply the calibration plots that show linearity and our limit of detection.

Lets start with Nickel:

Concentration in ERM-CA403: 1,04 µg/l (uncertainty 0,16 µg/l)
Oceamo ICP-MS measurement 1: 1,15 µg/l
Oceamo ICP-MS measurement 2: 1,17 µg/l

Ni.PNG

Cobalt:

Concentration in ERM-CA403: 0,074 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,063 µg/l
Oceamo ICP-MS measurement 2: 0,068 µg/l

Co.PNG

Copper:

Concentration in ERM-CA403: 0,87 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 0,754 µg/l
Oceamo ICP-MS measurement 2: 0,871 µg/l
Cu.PNG


Cadmium:

Concentration in ERM-CA403: 0,094 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,094 µg/l
Oceamo ICP-MS measurement 2: 0,099 µg/l

Cd.PNG

Arsenic:

Concentration in ERM-CA403: 1,90 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 1,86 µg/l
Oceamo ICP-MS measurement 2: 1,63 µg/l

As.PNG

Selenium:

Concentration in ERM-CA403: 0,06-0,094 µg/l
Oceamo ICP-MS measurement 1: 0,083 µg/l
Oceamo ICP-MS measurement 2: 0,092 µg/l

Se.PNG

Lead:

Concentration in ERM-CA403: 0,098 µg/l (uncertainty 0,010 µg/l)
Oceamo ICP-MS measurement 1: 0,114 µg/l
Oceamo ICP-MS measurement 2: 0,097 µg/l

Pb.PNG


I hope you found this information useful, and that it contributes to lift the curtains from the blackbox "seawater laboratory".

By this data you can also see, why i have decided to expand towards ICP-MS: The levels measured here are all within the relevant concentration range of those trace elements in seawater (and thus also in our aquaria). - But are at the same time below the realistic limit of detection for ICP-OES, and thus would appear on the analysis report as "non detectable". ICP-OES (which we also frequently use) is a great tool for many elements that appear in higher concentration ranges, and to (in many cases) detect elevated levels - but fails to reliably diagnose (too) low levels for many essential trace elements. This is where ICP-MS fills the void.

If youre having any questions i am happy to help!
Best regards,
Christoph

Christoph: Thank you for posting this it is very helpful in understanding ICP measurement data.

Rick
 
Coral Mania

taricha

2500 Club Member
Review score
+0 /0 /-0
View Badges
Joined
May 22, 2016
Messages
4,193
Reaction score
6,404
Review score
+0 /0 /-0
Thanks Christoph, I love nice straight well-behaved calibration curves.

So is it correct to infer that when I send a sample to Oceamo, that an ICP-OES would be done to cover many of the basics, and an ICP-MS would be run for particular low concentration elements of interest?
 

JimWelsh

Valuable Member
Review score
+0 /0 /-0
View Badges
Joined
Nov 5, 2011
Messages
1,472
Reaction score
1,566
Review score
+0 /0 /-0
Location
Angwin, CA
Dear Reef2Reefers!

I have seen a lot of skepticism regarding seawater laboratory analysis in this forum, especially when in comes to reported values for trace elements. I must say, that all skepticism is very well justified, since there is many potential pitfalls in ICP analysis: Sampling, sample handling in the laboratory, laboratory contamination, interferences, wrong interpretation of data, and many more. Those points even increase in importance when very low detection limits are crucial.

Disclaimer: I can speak here only for the procedures in our lab (Oceamo). I hope this post does not have too much of an advertisment character (@Randy Holmes-Farley, please remove it if this should be the case)

For transparency reasons and to build some confidence in trace element analysis id like to share some data with you. We have started our ICP-MS service late last year. ICP-MS is way more sensitive, so we can detect and quantify trace elements and pollutants at much lower concentration compared to ICP-OES - but ICP-MS also comes with many challenges the analytical chemist must handle to provide reliable results.

To control the quality of our data we are using the seawater certified reference material ERM-CA403 supplied by the European Comission. This is seawater, that has very clearly determined levels of certain trace elements. Im Posting the certificate of analysis here:

ERM1.PNG


erm2.PNG


This reference material is included in all our ICP-MS sequences, to check if our measurements are correct.

That you get a feeling for the level of accuracy and uncertainty you can expect, i am posting our results and the certified value for the more difficult elements here, and will also supply the calibration plots that show linearity and our limit of detection.

Lets start with Nickel:

Concentration in ERM-CA403: 1,04 µg/l (uncertainty 0,16 µg/l)
Oceamo ICP-MS measurement 1: 1,15 µg/l
Oceamo ICP-MS measurement 2: 1,17 µg/l

Ni.PNG

Cobalt:

Concentration in ERM-CA403: 0,074 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,063 µg/l
Oceamo ICP-MS measurement 2: 0,068 µg/l

Co.PNG

Copper:

Concentration in ERM-CA403: 0,87 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 0,754 µg/l
Oceamo ICP-MS measurement 2: 0,871 µg/l
Cu.PNG


Cadmium:

Concentration in ERM-CA403: 0,094 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,094 µg/l
Oceamo ICP-MS measurement 2: 0,099 µg/l

Cd.PNG

Arsenic:

Concentration in ERM-CA403: 1,90 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 1,86 µg/l
Oceamo ICP-MS measurement 2: 1,63 µg/l

As.PNG

Selenium:

Concentration in ERM-CA403: 0,06-0,094 µg/l
Oceamo ICP-MS measurement 1: 0,083 µg/l
Oceamo ICP-MS measurement 2: 0,092 µg/l

Se.PNG

Lead:

Concentration in ERM-CA403: 0,098 µg/l (uncertainty 0,010 µg/l)
Oceamo ICP-MS measurement 1: 0,114 µg/l
Oceamo ICP-MS measurement 2: 0,097 µg/l

Pb.PNG


I hope you found this information useful, and that it contributes to lift the curtains from the blackbox "seawater laboratory".

By this data you can also see, why i have decided to expand towards ICP-MS: The levels measured here are all within the relevant concentration range of those trace elements in seawater (and thus also in our aquaria). - But are at the same time below the realistic limit of detection for ICP-OES, and thus would appear on the analysis report as "non detectable". ICP-OES (which we also frequently use) is a great tool for many elements that appear in higher concentration ranges, and to (in many cases) detect elevated levels - but fails to reliably diagnose (too) low levels for many essential trace elements. This is where ICP-MS fills the void.

If youre having any questions i am happy to help!
Best regards,
Christoph
I would have higher confidence in the slope/intercept and R2 values of the calibration curves if they included more evenly distributed standards across the range of interest, without the high-leverage outlier data point that magically cleans up the statistics.

For example, when you create a calibration curve that consists of the following standards and y values (entirely arbitrary numbers I just made up to illustrate my point):

X (std)Y
0​
0.4​
1​
1.4​
2​
1.8​
3​
2.9​
4​
3.8​
5​
5.2​

The calibration curve would look like this, with the slope, intercept, and R2 value as shown in this chart:
1644543428694.png


But, if I add just one high-leverage point, the chart tells an entirely different story:
X (std)Y
0​
0.4​
1​
1.4​
2​
1.8​
3​
2.9​
4​
3.8​
5​
5.2​
100​
99.9​

1644543549991.png


Wow! The slope, intercept, and R2 values have all gotten soooo much better! Woot! That means the data in the low range will be better now, right? (</sarcasm>)

I'll link the following document, which better express some of my concerns, especially in section 2.3.1: Preparation of Calibration Curves - A Guide to Best Practice
 

Woodyman

Valuable Member
Review score
+0 /0 /-0
View Badges
Joined
May 25, 2021
Messages
2,146
Reaction score
2,678
Review score
+0 /0 /-0
Location
USA
I would have higher confidence in the slope/intercept and R2 values of the calibration curves if they included more evenly distributed standards across the range of interest, without the high-leverage outlier data point that magically cleans up the statistics.

For example, when you create a calibration curve that consists of the following standards and y values (entirely arbitrary numbers I just made up to illustrate my point):

X (std)Y
0​
0.4​
1​
1.4​
2​
1.8​
3​
2.9​
4​
3.8​
5​
5.2​

The calibration curve would look like this, with the slope, intercept, and R2 value as shown in this chart:
1644543428694.png


But, if I add just one high-leverage point, the chart tells an entirely different story:
X (std)Y
0​
0.4​
1​
1.4​
2​
1.8​
3​
2.9​
4​
3.8​
5​
5.2​
100​
99.9​

1644543549991.png


Wow! The slope, intercept, and R2 values have all gotten soooo much better! Woot! That means the data in the low range will be better now, right? (</sarcasm>)

I'll link the following document, which better express some of my concerns, especially in section 2.3.1: Preparation of Calibration Curves - A Guide to Best Practice

Agreed, but compared to what I've seen from the other companies they have done a good job providing this extra data, and explaining it.

Would be nice to see more, but this is a welcome step in the right direction.
 
OP
Christoph

Christoph

Active Member
Review score
+0 /0 /-0
View Badges
Joined
May 23, 2017
Messages
170
Reaction score
339
Review score
+0 /0 /-0
Location
Vienna, Austria
Hi everyone!

Thanks for the comments!

Thank you for the data.

Being in quality this is nice to know you run reference standards with your analysis. It would be nice to see other companies provide the same in the future!

Are you guys certified in any capacity? ISO, etc.?

No, we dont have an ISO certification. I personally do think that the certification doesn not necessarily lead to better analyses, but regulates many things in the lab (which is of course also a good thing). The certification comes with significant additional cost (especially in regards to personnel), and i would need to pass on this cost to the end customer. Since it is required to keep the price as low as possible for the hobby sector (while keeping the quality of data high), we have decided against a certification.

Where to buy your tests?

Our partner in the US is Aquabiomics, in the UK Oceamo is available via Reefquest. In Europe you can get testkits directly from us or in many reef shops.

Thanks Christoph, I love nice straight well-behaved calibration curves.

So is it correct to infer that when I send a sample to Oceamo, that an ICP-OES would be done to cover many of the basics, and an ICP-MS would be run for particular low concentration elements of interest?

Taricha, when somebody chooses to get the ICP-MS upgrade all "ICP-Parameters" are measured using the MS (no ICP-OES used at all). The machine has a dynamic range spanning over 10-11 orders of magnitude, so we can also measure macroelements (K, Ca, Mg and even Na) without issues, when optimized tune parameters are used.

@JimWelsh : Your post is unfortunately too long to quote correctly :). I do see your point, and i am always open to criticizm. However, since we need to span a wide concentration range (from very low to also elevated levels), it is feasible (and common practice in ICP MS analysis) to have calibration points not linearly distributed but at orders of magnitude. The shown elements are in their optimal range at very low concentrations, but also elevated levels occur in aquaria - thus we have the higher calibration point up there at 30 µg/l. By our data from the certified reference material you can see, that the calibration is working also well at very low concentration levels. Its the beauty of ICP-MS that linearity is excellent over many orders of magnitude.

Best regards,
Christoph
 

Mehboob97

New Member
Review score
+0 /0 /-0
View Badges
Joined
Feb 24, 2022
Messages
3
Reaction score
5
Review score
+0 /0 /-0
Location
Lahore, Punjab, Pakistan
Dear Reef2Reefers!

I have seen a lot of skepticism regarding seawater laboratory analysis in this forum, especially when in comes to reported values for trace elements. I must say, that all skepticism is very well justified, since there is many potential pitfalls in ICP analysis: Sampling, sample handling in the laboratory, laboratory contamination, interferences, wrong interpretation of data, and many more. Those points even increase in importance when very low detection limits are crucial.

Disclaimer: I can speak here only for the procedures in our lab (Oceamo). I hope this post does not have too much of an advertisment character (@Randy Holmes-Farley, please remove it if this should be the case)

For transparency reasons and to build some confidence in trace element analysis id like to share some data with you. We have started our ICP-MS service late last year. ICP-MS is way more sensitive, so we can detect and quantify trace elements and pollutants at much lower concentration compared to ICP-OES - but ICP-MS also comes with many challenges the analytical chemist must handle to provide reliable results.

To control the quality of our data we are using the seawater certified reference material ERM-CA403 supplied by the European Comission. This is seawater, that has very clearly determined levels of certain trace elements. Im Posting the certificate of analysis here:

ERM1.PNG


erm2.PNG


This reference material is included in all our ICP-MS sequences, to check if our measurements are correct.

That you get a feeling for the level of accuracy and uncertainty you can expect, i am posting our results and the certified value for the more difficult elements here, and will also supply the calibration plots that show linearity and our limit of detection.

Lets start with Nickel:

Concentration in ERM-CA403: 1,04 µg/l (uncertainty 0,16 µg/l)
Oceamo ICP-MS measurement 1: 1,15 µg/l
Oceamo ICP-MS measurement 2: 1,17 µg/l

Ni.PNG

Cobalt:

Concentration in ERM-CA403: 0,074 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,063 µg/l
Oceamo ICP-MS measurement 2: 0,068 µg/l

Co.PNG

Copper:

Concentration in ERM-CA403: 0,87 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 0,754 µg/l
Oceamo ICP-MS measurement 2: 0,871 µg/l
Cu.PNG


Cadmium:

Concentration in ERM-CA403: 0,094 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,094 µg/l
Oceamo ICP-MS measurement 2: 0,099 µg/l

Cd.PNG

Arsenic:

Concentration in ERM-CA403: 1,90 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 1,86 µg/l
Oceamo ICP-MS measurement 2: 1,63 µg/l

As.PNG

Selenium:

Concentration in ERM-CA403: 0,06-0,094 µg/l
Oceamo ICP-MS measurement 1: 0,083 µg/l
Oceamo ICP-MS measurement 2: 0,092 µg/l

Se.PNG

Lead:

Concentration in ERM-CA403: 0,098 µg/l (uncertainty 0,010 µg/l)
Oceamo ICP-MS measurement 1: 0,114 µg/l
Oceamo ICP-MS measurement 2: 0,097 µg/l

Pb.PNG


I hope you found this information useful, and that it contributes to lift the curtains from the blackbox "seawater laboratory".

By this data you can also see, why i have decided to expand towards ICP-MS: The levels measured here are all within the relevant concentration range of those trace elements in seawater (and thus also in our aquaria). - But are at the same time below the realistic limit of detection for ICP-OES, and thus would appear on the analysis report as "non detectable". ICP-OES (which we also frequently use) is a great tool for many elements that appear in higher concentration ranges, and to (in many cases) detect elevated levels - but fails to reliably diagnose (too) low levels for many essential trace elements. This is where ICP-MS fills the void.

If youre having any questions i am happy to help!
Best regards,
Christoph
Appreciate your efforts to share this information.
 
REEFTIDE

reeFnerD#38401

New Member
Review score
+0 /0 /-0
View Badges
Joined
Mar 11, 2022
Messages
5
Reaction score
1
Review score
+0 /0 /-0
Location
Columbia TN
Dear Reef2Reefers!

I have seen a lot of skepticism regarding seawater laboratory analysis in this forum, especially when in comes to reported values for trace elements. I must say, that all skepticism is very well justified, since there is many potential pitfalls in ICP analysis: Sampling, sample handling in the laboratory, laboratory contamination, interferences, wrong interpretation of data, and many more. Those points even increase in importance when very low detection limits are crucial.

Disclaimer: I can speak here only for the procedures in our lab (Oceamo). I hope this post does not have too much of an advertisment character (@Randy Holmes-Farley, please remove it if this should be the case)

For transparency reasons and to build some confidence in trace element analysis id like to share some data with you. We have started our ICP-MS service late last year. ICP-MS is way more sensitive, so we can detect and quantify trace elements and pollutants at much lower concentration compared to ICP-OES - but ICP-MS also comes with many challenges the analytical chemist must handle to provide reliable results.

To control the quality of our data we are using the seawater certified reference material ERM-CA403 supplied by the European Comission. This is seawater, that has very clearly determined levels of certain trace elements. Im Posting the certificate of analysis here:

ERM1.PNG


erm2.PNG


This reference material is included in all our ICP-MS sequences, to check if our measurements are correct.

That you get a feeling for the level of accuracy and uncertainty you can expect, i am posting our results and the certified value for the more difficult elements here, and will also supply the calibration plots that show linearity and our limit of detection.

Lets start with Nickel:

Concentration in ERM-CA403: 1,04 µg/l (uncertainty 0,16 µg/l)
Oceamo ICP-MS measurement 1: 1,15 µg/l
Oceamo ICP-MS measurement 2: 1,17 µg/l

Ni.PNG

Cobalt:

Concentration in ERM-CA403: 0,074 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,063 µg/l
Oceamo ICP-MS measurement 2: 0,068 µg/l

Co.PNG

Copper:

Concentration in ERM-CA403: 0,87 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 0,754 µg/l
Oceamo ICP-MS measurement 2: 0,871 µg/l
Cu.PNG


Cadmium:

Concentration in ERM-CA403: 0,094 µg/l (uncertainty 0,011 µg/l)
Oceamo ICP-MS measurement 1: 0,094 µg/l
Oceamo ICP-MS measurement 2: 0,099 µg/l

Cd.PNG

Arsenic:

Concentration in ERM-CA403: 1,90 µg/l (uncertainty 0,13 µg/l)
Oceamo ICP-MS measurement 1: 1,86 µg/l
Oceamo ICP-MS measurement 2: 1,63 µg/l

As.PNG

Selenium:

Concentration in ERM-CA403: 0,06-0,094 µg/l
Oceamo ICP-MS measurement 1: 0,083 µg/l
Oceamo ICP-MS measurement 2: 0,092 µg/l

Se.PNG

Lead:

Concentration in ERM-CA403: 0,098 µg/l (uncertainty 0,010 µg/l)
Oceamo ICP-MS measurement 1: 0,114 µg/l
Oceamo ICP-MS measurement 2: 0,097 µg/l

Pb.PNG


I hope you found this information useful, and that it contributes to lift the curtains from the blackbox "seawater laboratory".

By this data you can also see, why i have decided to expand towards ICP-MS: The levels measured here are all within the relevant concentration range of those trace elements in seawater (and thus also in our aquaria). - But are at the same time below the realistic limit of detection for ICP-OES, and thus would appear on the analysis report as "non detectable". ICP-OES (which we also frequently use) is a great tool for many elements that appear in higher concentration ranges, and to (in many cases) detect elevated levels - but fails to reliably diagnose (too) low levels for many essential trace elements. This is where ICP-MS fills the void.

If youre having any questions i am happy to help!
Best regards,
Christoph
This sort of transparency certainly provides a little more comfort with your services.
 

Reefahholic

Reef Junkie T.V.
Review score
+2 /0 /-0
View Badges
Joined
Sep 5, 2014
Messages
5,169
Reaction score
4,351
Review score
+2 /0 /-0
Location
Houston, TX
Finally here. Been waiting a long time for this.


These tests can be ordered through their partner Aquabiomics:



The upgrade can be directly purchased on their webpage here:


Or the ReefQuest website here:


Attention: The upgrade can only be used currently in connection with an Oceamo Seawater analysis. Once purchased we will inform Dr Christoph Denk of your standard ICP number requiring the upgraded MS-ICP analysis


The portal is here:

 

Reefahholic

Reef Junkie T.V.
Review score
+2 /0 /-0
View Badges
Joined
Sep 5, 2014
Messages
5,169
Reaction score
4,351
Review score
+2 /0 /-0
Location
Houston, TX
@Christoph can you help me fully understand the difference between Phosphorus and Phosphate?

When our systems are depleted of one or both elements… what is the best product to supplement?

Should these be addressed individually?

Are we really dosing Phosphorus or Phosphate back to the system?

Lastly, what is the importance of listing both elements on the ICP?

Thank you sir!
 

beesnreefs

Active Member
Review score
+0 /0 /-0
View Badges
Joined
Sep 23, 2020
Messages
278
Reaction score
236
Review score
+0 /0 /-0
Location
Firestone
@Christoph this is great stuff! Really appreciate your transparency. Certainly contributes to higher confidence in the method and using Oceamo for my tank testing.

Have you run any qualification testing such as accuracy across the curve and precision (repeatability and intermediate precision)? If so I’d love to see the results.
 

Randy Holmes-Farley

Reef Chemist
Review score
+0 /0 /-0
View Badges
Joined
Sep 5, 2014
Messages
51,596
Reaction score
43,639
Review score
+0 /0 /-0
Location
Arlington, Massachusetts, United States
@Christoph can you help me fully understand the difference between Phosphorus and Phosphate?

When our systems are depleted of one or both elements… what is the best product to supplement?

Should these be addressed individually?

Are we really dosing Phosphorus or Phosphate back to the system?

Lastly, what is the importance of listing both elements on the ICP?

Thank you sir!

I can answer that. Icp detects atoms like phosphorus, regardless if the chemical form.

Phosphate is one form phosphorus can take, consisting of oxygen, phosphorus and, in most situations in seawater, hydrogen.

In most scenarios in a reef tank, inorganic phosphate is higher in concentration than other forms of phosphorus, such as organic phosphates.

you can dose organic or inorganic phosphate, if you want to dose, but must folks elect to dose inorganic phosphate, such as sodium phosphate.
 
BRS

Randy Holmes-Farley

Reef Chemist
Review score
+0 /0 /-0
View Badges
Joined
Sep 5, 2014
Messages
51,596
Reaction score
43,639
Review score
+0 /0 /-0
Location
Arlington, Massachusetts, United States
I would also add that I think most icp companies are not measuring inorganic phosphate, even if they seem to report it. They are just taking the phosphorus value from icp and multiply by 3.06 to get phosphate.
 

taricha

2500 Club Member
Review score
+0 /0 /-0
View Badges
Joined
May 22, 2016
Messages
4,193
Reaction score
6,404
Review score
+0 /0 /-0
interestingly, Oceamo claims a separate chemical test is done for PO4.
their "10 good reasons for Oceamo"
"7. Additional photometric analysis of phosphate. This gives you a very reliable phosphate value."
 

Randy Holmes-Farley

Reef Chemist
Review score
+0 /0 /-0
View Badges
Joined
Sep 5, 2014
Messages
51,596
Reaction score
43,639
Review score
+0 /0 /-0
Location
Arlington, Massachusetts, United States
interestingly, Oceamo claims a separate chemical test is done for PO4.
their "10 good reasons for Oceamo"
"7. Additional photometric analysis of phosphate. This gives you a very reliable phosphate value."

Interesting. I didn’t know that. Maybe Christoph can chime in with a general comment on what fraction of the P is detectable in typical reefs as inorganic phosphate, and hence how much must be other forms.
 

Woodyman

Valuable Member
Review score
+0 /0 /-0
View Badges
Joined
May 25, 2021
Messages
2,146
Reaction score
2,678
Review score
+0 /0 /-0
Location
USA
interestingly, Oceamo claims a separate chemical test is done for PO4.
their "10 good reasons for Oceamo"
"7. Additional photometric analysis of phosphate. This gives you a very reliable phosphate value."

@Christoph

Are you using IC (ion chromatography) for your phosphate analysis?

or just UV spectrophotometers?

DIfferent photometric ways/methods. Further insight would be nice!
 
Coral Mania

Did you start off in the reef hobby buying used equipment? Do you still use pre-owned equipment? (Choose all that apply)

  • Yes, I did start off with some used equipment. It made getting everything going easier!

    Votes: 136 40.5%
  • Yes, I did start off with some used equipment, but it was a mistake I won't make again.

    Votes: 17 5.1%
  • I still buy used equipment! It's a helpful way to save money in this hobby!

    Votes: 105 31.3%
  • Nope! Everything I used to get started was new! And I'm glad I started with all new stuff!

    Votes: 81 24.1%
  • Nope! Everything I used to get started was new! But, I should've looked around for used options!

    Votes: 33 9.8%
  • I still only buy new equipment! Lots of problems avoided that way!

    Votes: 43 12.8%
  • Other... (Post in the thread)

    Votes: 13 3.9%
Top