The Reef Builders’ Big ICP Test Review, by Sanjay Joshi

Initially introduced to the hobby market by Triton, ICP tests have been around for several years now and have become a staple in the hobby. The allure of these tests has been the relatively cheap ability to get a reading of the various major and trace elements in the aquarium water.

Trace elements have been typically difficult to measure with home aquarium test kits. The ICP-based tests are reported to measure trace elements to levels of parts per billion (ppb). This has been used to promote “new” methods of reef keeping (eg. Triton Method, Moonshiners Method) where water changes are not necessary, and the results of the tests can be then used to restore the low or high values of the elements tested to bring them within suggested seawater values.

The concept is attractive, especially since we are beginning to understand the importance of some trace elements and need an easy way to measure them. I believe that this is a useful capability. However, it is also important to understand the differences and issues related to this kind of testing to allow for informed decisions about their reef based on the test results.

The intent of this article is to compare various ICP methods with respect to several criteria, as listed below. 

1. Samples needed
2. Cost and speed of results
3. Presentation of the results
4. Comparison of the results 

The actual quality of the test result is not the goal here as this is not easy to establish due to the lack of verified traceable standards for seawater. I think each of the ICP providers should take the onus to convince us that their tests are accurate with low levels of uncertainty and expected variability.

Method

5 ICP tests that were currently available in my inventory were used.  

1. Reef-labs (https://reefs-lab.com)
2. Fauna Marin (https://lab.faunamarin.de/en/home)
3. ICP-Analysis (https://www.icp-analysis.com/)
4. Triton (https://www.triton-lab.de/en/aquaria-administration)
5. Oceamo (https://en.oceamo.com/)

Roughly 2 gallons of water were removed from my 500G reef aquarium at 12:38 PM on Feb. 28, 2023. This water was then used to collect samples for all the ICP tests, minimizing any variability in the water. The samples were all collected within a span of 20 minutes. The sample identification data was entered into the respective websites, and all were dropped off at the US post office at the same time on Feb. 28, 2023.

What is ICP Spectroscopy? A simple explanation.

ICP (inductively coupled plasma) spectroscopy is an analytical method used to analyze chemical samples. Simplistically, the sample to be tested is introduced into a high-temperature plasma generated in the machine where the sample is ionized into positively charged elements and electrons. The detection of the elements is based on using the light emitted by the source ICP-OES (Optical Emission Spectroscopy) or by separation of the elements by their mass-to-charge ratio, ICP-MS (Mass Spectroscopy.)

The ICP-OES separates the light (photons) generated by the different elements using their wavelength. The count of photons at the wavelength is proportional to the amount of each element. This method has a high dynamic range (range based on the smallest to the largest measurement) and elements can be detected at sub-ppm levels up to a couple of 1000 ppm for up to 50-60 elements simultaneously, in less than a minute. 

The ICP-MS separates the element ions by the ratio of the mass to charge. ICP-MS has a higher dynamic range and hence can achieve measurements at the parts per billion levels at the lower end. ICP-MS tends to be more sensitive and typically requires a clean room to operate, along with extra care in the handling of the samples. Method development is also more complex, compared to the OES. Hence it tends to have lower throughput compared to ICP-OES and might be more expensive than ICP-OES.

In both these methods, the total of each element is reported independently of the source. So, if the source of the element is several different compounds, these are all combined in the result.  Only single elements are detected, not their composition. It cannot distinguish between dissolved inorganic phosphates and dissolved organic phosphates. Hence, these tests cannot detect PO4 for example. It will report just the value of P, and any value of PO4 reported is a calculated value.  

In both methods, the measurements are eventually converted to concentrations of the elements and reported in quantities of ppm (parts per million, mg/L), or ppb (parts per billion micrograms/L. The basic conversion is:

1 ppm = 1000 ppb

1 ppb = .001 ppm 

The companies providing these tests, provide the results for each element tested in the sample along with some indication as to which elements are high, low, or acceptable. This categorization is based on some range determined by each company. Some of them provide the range that they use and recommend, others may just provide a categorization of the value.

User’s Perspective

Initially ignoring the actual test results and assuming that all the tests attempt to provide reasonable results, other factors that may play into a user’s selection, three such factors evaluated here are – samples required, cost and speed of results, and presentation of results. Since the last factor on the actual accuracy of the results cannot be easily verified and is outside the scope here, the results of these tests are presented to allow the user to make comparisons and show the encountered variability in the results.

1. Type of ICP and Samples Required

The table below shows the types of ICP analysis performed by each company. Two of the companies use ICP-MS and the other 3 use ICP-OES for analysis. At the time of writing, (February 2023,) Oceamo and Fauna Marin also provide an analysis of the RO/DI water.

All these require logging into the respective websites and setting up an account. This allows you to record the samples being submitted by entering the code affixed to each sample. This way the sample can be traced to you. A good strategy is to take pictures of the codes provided to label each sample. In case the sample gets misplaced at the company you will have a record of it.

2. Shipping and Time to Results

The tests were all mailed through USPS on Feb. 28^th 2023.  The costs of these tests are based on the best costs I found on the internet. (better deals and sales may exist). As seen in the table below, the costs of the test are fairly close to each other with the cheapest around $40 and the most expensive around $49. Just comparing the cost is not enough, as some of these also include the cost of shipping, effectively reducing the total cost of the test. My shipping costs were within $0.55 of each other, using the cheapest shipping options available. Some of the companies provided acknowledgment of receipt of the test. Reef-labs provides continuous updates on the tracking of the package. Acknowledgment of the receipt is an important attribute. (Editor’s note, Triton Test and ICP Analysis now include shipping. The test was carried out in February 2023. We can update the Cost and Shipping of the featured tests as new pricing becomes available.)

The days to result is an important criteria for consideration. If you are using these tests to make corrections to the parameters, the longer the time it takes to get the results, the longer the lag in your control system. You will be making corrections based on old data and may need to keep track of changes made during the delay in the test results. If you are just using the data as a check, then the days to result may not be as critical. 

3. Presentation of Results

ICP-Analysis

Type of Test: ICP-MS

Number of Elements Tested: 61

Results are presented online in the form of an image with certain elements colored in red and yellow (see Figure 1). There is no color key provided, but I am guessing that the yellow color indicates values that are low, and red indicates values that are considered high with respect to some baseline values. Clicking on the image in the element box reveals additional information about each element, and target values, and provides the users with a baseline of preset values to choose as the reference point (see Figure 2)

Figure 1.  ICP Analysis Partial Image showing the online display of the results

Figure 2:  ICP Analysis  Screen image of the additional information revealed after clicking the element box for Fe

As seen in Figure 2, it provides the user with a choice of several baselines to choose from for comparison and use to model your reef parameters. Additionally for products that require dosing to bump up the levels based on your selected baseline, it provides options of various products that can be used to add the element. Clicking further on one of these options then provides the user with the quantity of that product to be added to bring the level in line with your chosen target waters (see Figure 3). Elemental notes are provided for the element. For elements that are high, the suggested solution is to do a water change or stop dosing and let it naturally decline over time. It seems that this chart also has the ability to plot the data from several prior tests to show the trends graphically. Since I only have one test, there was no data available to test this feature.

Figure 3. ICP Analysis – The calculated dosage of ATI Iron required to bring the levels to 20 ppb

The results can also be downloaded as a pdf file. in the form of a pdf image and pdf list. Downloading as a pdf image resulted in the red color not showing in the pdf image.  (see Figure 4)

Figure 4.  ICP-Analysis Partial image of the downloaded pdf image file (showing the missing red color)

The results can also be downloaded as a pdf list. A sample of what is contained in the list is shown in Figure 5. Note that there is no indication of any high or low markers in the list file, this is due to the fact that these numbers rely on a preselected reference.

Figure 5.  ICP-Analysis Partial image of the data contained in the pdf list file. There is no option to download the data as a .csv file that can be opened in Excel. 

Triton

Type of Test: ICP-OES

Number of Elements Tested: 42

Results are presented and viewable on the website. Data is available in 3 forms – online viewable data, downloadable in the form of pdf and CSV files. It also allows the capability to share the data through the use of a shareable URL. 

The online viewable data shows the values for each of the 42 elements tested. The suggested setpoints and setpoint ranges for the values are also provided. How these setpoint values are established is not exactly clear. A visual aid is provided to visualize the values in relation to acceptable ranges determined by Triton. In cases where a setpoint range is provided it is easy to interpret the green color as the extremes of the range, however, how the yellow and red are determined is again not clear.  For example in Figure 6, picking Bromide as an example, it has a setpoint value of 62mg/L with a measured value of 68 mg/L. Why this would be considered as red is not clear. The graphical presentation of the data is a good idea but I feel this implementation could be improved to provide some more useful information.

Figure 6.  shows the user view of the online viewable data for the Triton ICP.

Downloading the files does not show the same view as shown on the web. The data provided is the same except the visual color scale diagram is replaced by warning colors, red, yellow, and green as seen in Figure 7.

Figure 7. Partial view of the data in the Triton pdf file.

The .csv provides the same data as the pdf file and can be easily downloaded.  This will allow you to keep track of your tests within Excel and use it to plot your own charts.

The website also provides the capability of visualizing the data from the past test in a single graph. Multiple elements can be selected and data plotted for the previous and current tests, as seen in Figure 8.

Figure 8. Triton visualization of previous tests for the elements.

Reef-Labs ICP

Type of Test: ICP-OES

Number of Elements Tested: 36

The test results are presented online only, with a URL that can be used to share the results with others. In addition to the standard elemental test results, additional information is provided and relates to the summary of observations as seen in figure 9. I have found this to be useful.  Important warnings and notable changes since the last test are features that allow you to quickly home in on key elements of the data.

Figure 9. Reef-labs ICP Important warnings and notable changes since last test

The results also provide several ratios of elements that Reef-Labs have deemed important. I am not sure how these ratios are established and why these are the only important ratios. These ratios if they are of interest to the user can be extracted from the elemental results for the individual elements. However, it is convenient to see them calculated. See Figure 10.

Figure 10.  Reef-Labs Important Ratios and Calculated Values

Additional calculated values on the salinity and phosphate, both of which are calculated from the test results are also provided. The phosphate is simply a conversion of all the Phosphorous to phosphate obtained by multiplying the P measured by 3. How the salinity is calculated is not clear.  

The elemental data is provided by grouping the various elements into three categories – Essential Macro Elements, Beneficial Minor and Trace Elements, Contaminants, and Toxins. The results from one such grouping are shown in Figure 11. I particularly like the information that compares it to the last test. I don’t have to pull up past tests or graphs to see the change. For each element, it also provides information on the detection limit, optimal range, and importance. Using the suggested optimal range, the elements are colored based on the key provided.

Figure 11. Reef-labs partial test data for individual elements

Fauna Marin 

Type of Test: ICP-OES and Additional Physical-Chemical and Photometric Tests

Number of Elements Tested: 38 + 18 (RO/DI)

In addition to testing the aquarium water, the Fauna Marin ICP test also tests the input freshwater used to mix the salt and used for auto top-off at no additional cost. This is a good feature and it allows for some additional problem-solving capabilities and helps track the potential source of elements. 

The test results are presented online and in the form of a pdf file. The online results are presented in a very useful and easy-to-use format. It provides a quick overview of the analysis showing the elements of interest that need attention. An overview of how much of the elements are needed for dosing is also provided (Figure 12). However, since the source of this additive is not specified, it is assumed that it is referring to the Fauna Marin line of additives.

Figure 12.  Fauna Marin’s view of the initial information

The test data is grouped into 6 categories – Physical-Chemical based values, Macro Elements, Macro Nutrients, Physiologically relevant Elements, Other trace elements and potential pollutants, and Relative values between elements and halogens that are computed from the test data (Figure 13).

Figure 13.  Partial view of the Fauna Marin grouping

This view also allows the users to pick specific elements and probe the details for each element.  A partial view of this data is shown in Figure 14. This view of the data provides the user with visual as well as text-based information with explanations along with links to further detailed information from the Fauna Marin developed knowledge base. I highly recommend that the hobbyist probe this database to further their knowledge of the elements and their potential role in coral husbandry. Additionally, it also provides a historical graph of the change in the values over time using past test information. This is a very useful feature especially if you are a regular user of this ICP test. This view also provides tips on what to do, to bring the values in line with their suggested values.

For each group of elements, the data is also presented as a graph that helps in visualizing the elements that are above the recommended relevance line. (figure 14). This graph visually shows that in this test the Li and La values are high.

Figure 14. Fauna Marin ICP results – Trace element group along with the relevance line

The results also provide calculated relationships from the tested values that Fauna Marin has determined to be important indicators of aquarium water quality. In their view, it is not only the individual elemental values that are important, but the relational values with respect to other elements are also important. Figure 15 shows how this result is presented online. It would be helpful if this also included a graph of the historical values.

Figure 15. Fauna Marin ICP partial view of the Relevance Ratios.

The Physical-chemical basic values are also provided. These are not based on the ICP analysis but provide useful information that cannot be measured by the ICP test. Information such as Electrical conductivity, Density, Salinity, pH, Carbonate Hardness, CO2, Acid Binding capacity, color, and smell are included.  

The macronutrients are also measured, and results are provided. The total phosphate is calculated based on the phosphorous measured. Ortho phosphates are measured using photometric means. Nitrate measurements provided, are most likely based on photometric tests.  

In addition to providing access to information online, the data can also be downloaded as a pdf file. The data contained in the pdf file is basically in text format with a few visual graphs related to each grouping of the elements. The suggested reference range and target values for each element are also included. Figure 16, shows a partial image of the pdf file that shows both the textual data as well as the graphical view.

Figure 16. Fauna Marin ICP partial view of the pdf data report

In the Fauna Marin test it also provides an analysis of the Reverse Osmosis water used to make up the salt water and auto top off. 18 different elements are tested. This can provide information on the quality of the water as well over time can help with monitoring the performance of your RO/DI system to provide clues on when it should be replaced. The data included in the test for Osmosis water is shown in figure 17.

Figure 17. Fauna Marin ICP analysis of RO/DI water

OCEAMO

Type of Test: ICP-MS 

Number of Elements Tested: Aquarium water (44), RO/DI (3)

The test results are presented online only in the form of downloadable pdf file. The capability to visualize the change in elements over time is also provided online. The test data is grouped into several categories similar to the others. The categories used are Main Parameters, Main Elements, Trace Elements, Pollutants, Nutrients, and Osmose. The measured values for each element are presented along with a suggested target value and range as seen in Figure 18. A color-coded rating of each element is provided and gives a quick visual focus to the elements where action is recommended. The recommended action is to input these values in the Moonshiner’s calculator (https://www.reefmoonshiners.com/) to determine the dose of elements required based on the Moonshiner products.

Figure 18. Oceamo ICP a partial view of the results in pdf file

The Oceamo ICP also tests the RO/DI water for three elements and reports it in the same format. (Figure 19)

Figure 19. Oceamo ICP Results of test of RO/DI water

The test also reports on the phosphate and nitrate values and is included in the Nutrients grouping. The phosphate values are reported as being based on photometric methods. Interestingly, no phosphorous measurements are included in the individual element list.  (Figure 20)

Figure 20.  Oceamo ICP results in the Nutrient grouping.

4. Comparison of the ICP Results

The table below provides a complete comparative analysis of the ICP results from the five companies.

The nutrient values as provided – these values are NOT determined by ICP testing.

Discussion

The data clearly shows that there is a fair amount of variability in the results from the different vendors. There is variability in how long it takes to get results, the quality of presentation of the results, the quality of the information provided online and in the pdf files, availability of data in CSV formats, the number of elements tested, and the type of test methods used. The user should evaluate these to help in making a final decision on which test to use.  

As for the variability in the actual values of elements tested using identical water, this was to be expected. There have been prior comparisons of ICP data reported by several others. One such analysis is here (https://fritzaquatics.com/assets/files/uploads/ICP_TES.pdf), several others can be found on reef forums. 

The objective here was not to establish which test is best or most accurate but to show the range of variability one can expect in the results. I will leave it up to the companies to establish and provide data to convince the user about the accuracy of the tests in light of the variability seen in the data. Understanding this variability between ICP testing vendors is critical to the home reef keeper who may be thinking they are getting rock solid and actionable testing results. 

Given this variability, how useful are these tests to reef keepers? Do they provide useful information? Should they be used to tweak the levels of trace elements? Who’s ICP should you trust? What is the risk-reward from these tests?

How useful are these tests to reefkeepers? This depends on what you are using the tests for. If the objective is to detect potentially high levels of some potentially toxic element, then these tests may provide useful information in spite of the variability between vendors.  

For example, past single ICP tests of my tank water showed elevated levels of Nickel. The Ni levels as tested by the 5 tests range from 21 ppb to 38.4 ppb. The range for Nickel is 2-5 ppb according to Oceanmo, 3-6 ppb according to Fauna Marin, and a setpoint of 5 ppb according to Triton (again these ranges are determined by each provider).

So just eyeballing the numbers the Ni values are about 7-8 times these ranges. So it might be safe to say that the Ni values in my aquarium are high (though if you only have a single test result to consider, unlike the 5 in this article, your confidence in that number expressing actionable information may be less). You could then look for a potential cause of why this value is high, maybe an exposed magnet, some exposed rusting steel, etc. If you find it, and it ends up lowering the Ni values, then you found the source and fixed the problem by using the ICP result, avoiding any further issues in the future.

But, what if you were not really able to find the source? Or, what if you do nothing, and the next ICP test shows ‘normal’ range results for Nickel? The question to ask may be, is this high value (remember it is still in parts per billion) a real matter of concern? Are there known toxicity levels for Ni? So while we can potentially determine that levels are “high” we really have no idea if this is really a concern. 

Similar situations are encountered with Lithium. Every ICP test I have seen in the US tests high for Lithium. Yet, we do not see any issues that can be traced to high Lithium levels. So yes, these tests may be able to point to higher levels of some trace elements that are considered toxic, but we lack the knowledge about what are the levels where it impacts the corals. Given the lack of this knowledge, a safe bet is to work within the range of natural sea water levels as a basis for what we try to replicate in our systems.

The recommended numbers are determined internally by each vendor, and they differ from vendor to vendor. While there is no agreement on ‘standard acceptable ranges’ there is a reasonable overlap of the values provided by the different vendors, so you can play it “safe” by choosing values towards the lower end of the acceptable range, while not panicking about values that are beyond the values seen in the suggested ranges. So far there are only a few studies on the toxicity levels of individual elements. Further, we really cannot point to one high element and attribute any problems, current or future anticipated problems to it. Hence it is not a surprise that there are enough anecdotal reports from ICP users showing normal ICP test results while having issues with the tanks. Vice versa there are situations where the high values may have no impact.

These tests seem to provide some useful information and there are no hobbyist test kits that can easily test for all these elements at reasonable levels of cost. However, the usefulness of these tests in my view is limited to the extremes and can provide useful information for data that far exceeds what the hobbyist considers normal ranges.

Should hobby ICP test results be used to tweak the values of individual elements at finer resolution? There is a school of reefkeeping that believes in no water changes and that individual trace elements should be tweaked to fall within some accepted range. Or, that individual trace elements should be added and maintained at some specified range.

While in theory, this is achievable, it relies on 2 main tenants. One, the ICP tests are accurate and timely, and two, the selected range has a rational basis and operation outside of this range is tantamount to creating issues in the corals. At this point, there is no foundational basis to establish which test is accurate and how accurate the results are. Secondly, there is no scientific basis that establishes limits that these trace elements can take without causing issues. Neither is there any information that correlates levels of each element to color or success in reef keeping. Remember that as a group, we don’t agree on what levels major constituents like Nitrate and Phosphate are detrimental to our animals, and we have been playing with tweaking those numbers for decades. There is a lot of subjective anecdotal data that in itself should be evaluated with skepticism.

Whose ICP should you trust? At this point, it’s mostly a matter of personal preference or practicality (how quickly you can get the results back, ease of interpreting results, ability to download results to a spreadsheet, etc). The companies have done little to inspire trust to the general reef user, and they don’t seem to be working together to establish baselines so we are all on the same page. I have contacted some individual companies and spent time talking about their testing, and the methods used to establish confidence in the data (usually considered proprietary information) and yet there is significant variability in the data across companies, which makes it obvious that large issues like the Nickel example above seem actionable, but finer tweaks less trustable. Past tests in my aquarium have indicated high levels of  Ni, Iodine and since I have been using Lanthanum Chloride in my aquarium to reduce phosphates, it is interesting to compare the tests with respect to these 3 elements.

Clearly, this demonstrates the variability with the current generations of ICP and the challenges with using the results to agonize over making changes based on these results. According to four of the five tests, the Iodine values are very high (and they still have a big spread), and according to one, they are in the acceptable range.  
Overall, I still feel that the ICP results are still useful since they provide additional data and another tool that has the potential to be useful in improving the health of the aquarium. Don’t take the numbers as absolutes, evaluate trends, review values that stick out of typical norms, and don’t panic if the trace element numbers are being reported as out of line. The science on what levels are dangerous to coral health is still not established and there is no direct correlation to these values actually being beneficial or harmful.

There is some work being done to create a database of toxicity values as discussed in Ouédraogo, DY., Mell, H., Perceval, O. et al. What are the toxicity thresholds of chemical pollutants for tropical reef-building corals? A systematic review. Environ Evid 12, 4 (2023). https://doi.org/10.1186/s13750-023-00298-y. There are scientific papers that show that some trace elements are part of the biological process and useful, but there is nothing that shows what levels these need to be in the water to be useful.

About the author

Sanjay Joshi is a Professor of Industrial and Manufacturing Engineering at Penn State University and has been on the reefing scene since 1992. He runs a 500-gallon tank at the university as well as a 500-gallon SPS dominant reef tank at home. He has published many articles in magazines over the years including Marine Fish and Reef Annual, Aquarium Frontiers, Aquarium Fish, and Advanced Aquarist. He has been an invited speaker at several marine aquarium society meetings in the US and Europe, including headlining our very own ReefStock. He also received the MASNA award for his contributions to the marine aquarium hobby in 2006.