We get questions all the time from people who ask if they need a more advanced RO/DI system for well water versus city water or even if they need an RO/DI system to begin with. Up until now, the right answer has not really been established because it all depends on your source water. Without knowing what is in the tap water, it is difficult to say which exact RO/DI system is optimal for your application.


Bulk Reef Supply RO/DI Systems


When it comes to RO/DI water filtration you can choose specific filter cartridges to help you effectively target certain contaminants based on the levels of these particular contaminants in your source water. In order to determine the best type of filters and configuration for your particular RO/DI system and source water, you first need to know what exactly is in your source water.


Water Treatment Plant


Our curious minds turned to the mail-in ICP test kits to see if this lab grade water analysis can help hobbyists choose the best RO/DI filter combination for their particular source water.


The Experiment

We collected 7 different tap water samples from both city municipalities and well water sources around the greater Twin Cities area. We chose the budget friendly Coralvue ICP Water Analysis test kit and sent off a sample to be tested from each of the different freshwater sources.


ICP Water Analysis


The ICP reports we received back gave us a snapshot of 40 different elements found in these tap water sources. We compared those ICP results to really see how much the different sources vary in terms of contaminants.


Before we get into the results, let’s look at the the 7 different water sample sources. This can be broken down into two groups, city municipality and well water, all of which were taken within a 40 mile radius around the Minneapolis Metro area. City water sources are filtered, treated and disinfected before being delivered and the well water sources are untreated coming directly from a well or aquifer.


  • City Sample #1 - BRS tap water treated directly from the Mississippi River
  • City Sample #2 - Treated groundwater from a local aquifer within 5 miles of sample #1
  • City Sample #3 - Treated groundwater collected 20 miles north of sample #1
  • City Sample #4 - Treated groundwater collected 20 miles south of sample #1
  • Well Water #5 - Untreated groundwater collected within 5 miles of sample #1 near a local lake
  • Well Water #6 - Untreated groundwater collected within just a few miles of well-water sample #5 on a different shoreline of the same lake.
  • Well Water #7 - Untreated groundwater collected near the Mississippi river, 20 miles north of sample #1

water sample map


Some interesting things to look at through our testing was how much the levels of contaminants varied between the samples. In some cases these sources are sharing the same aquifer or water source but produce very different ICP reports. All the more reason to have your particular tap water checked before deciding what filters are most appropriate for your RO/DI system. The contaminants you need to target could easily be different than your local fish store or even your reefing buddy down the road.


It needs to be said that although ICP-Analysis testing can provide us with a clearer picture into some of the elements that make up our drinking water, they are definitely not designed nor should be reasonably interpreted as the most accurate testing method available for drinking water.


EPA Website


There are entire government agencies in place whose primary role is specifically designed around analyzing the public’s drinking water and if you want to dive even further into the breadth of what those testing methods are, you can absolutely find more information on the EPA’s website. All this means is we need to keep reasonable expectations for the accuracy and depth of the ICP results.


With that in mind, we are not going to decipher or speculate each of the results, what they mean and why those levels exist. Rather we want to draw upon the similarities or notable differences between our city and well water samples in order to find out if there’s a better way of choosing a more adequate RO/DI filtration approach for your specific water source or if you even need one to begin with.


Undetectable contaminants eliminated from study


The Results

The ICP Water Analysis kits we used for today’s experiment tests for 40 different elements, so we slimed down the list by first removing anything that was not registered or came in with a measurement below 1 part per billion.


Elements found in natural seawater


We also skipped over the elements that are present or in some cases even desirable in seawater. After pulling out those two groups of test results, we are left with 17 elements to evaluate.


The first element is Aluminum where we can see that each of the 7 water samples registered very small amounts, none of which were above 3 parts per billion.

Arsenic tested at super low levels below 2-parts per billion across the board with 3 having zero.

Two samples (both in well water and city water) had Boron levels up into the 6 parts per billion range but still pretty low overall.

All of the tests came back with Barium being presenand the BRS city sample #1 was the lowest registering at just 2 parts per billion.

With Bromine we see a wide range of results from 0 to 10 up to as much as 40 parts per million, all within a 40 mile radius of each other. These varying contaminant levels confirms that you cannot trust or expect your tap water chemistry to match other sources in your area.

Chromium registered in 3 of the 7 water samples, however, all of those tested under 2 parts per billion so really not much at all.

Copper is one of those elements that perks our ears as reef hobbyists and all 7 water samples had measurable levels of copper. It’s tough to say whether these copper levels were derived directly from the water sources themselves, if the type of plumbing contributed…or maybe something different altogether. The important thing to understand is that copper is present and the ICP tests detected varying levels from source to source, even when the sources were only 5 miles apart coming from the same aquifer.

With Iodnine we see all 7 samples tested for differing levels from 0.12 parts per million at the lowest and at the highest slightly above 1.5 parts per million.

There were very faint amounts of Lithium in nearly all of the water samples with the exception of City #4 and none of them tested above 2 parts per billion.

Phosphorus is another one reefers watch closely and is a big part of why we use RO/DI systems to begin with. All of the samples showed measurable levels ranging from very minor amounts that are not really a concern all the way up to 1.6 parts per million.

Rubidium tested under 0.25 parts per million for each sample with 3 of the samples coming back at zero.

Sulfur was detected in all 7 samples ranging from 3 to 15 parts per million across the board.

Silicon is one of the major contaminants to avoid, just like copper and phosphorus. The scary part is all of the water samples registered measurable levels of Silicon ranging from 5.5 parts per million up through 13.5 parts per million. Silicon and silicates are some of the trickiest elements to remove from source water since the most effective filtration approach is directly tied to the specific type of DI resin.

Thallium was present in each water sample with nothing higher than 4 parts per billion.

Uranium was also registered in all of the samples but in relatively small amounts.

Only 2 out of 7 water samples registered as having levels of Tungsten.

Zinc was detected in 3 water samples where Well and City Sample #2 were shown to have levels at 0.35 and 0.23 parts per million respectively..




Reef Fantasy or Reef Certainty Scale


Could a $30 dollar ICP-analysis help decide if we need a more or less aggressive RODI filtration approach for making ultra-pure water for our saltwater tanks?

We voted for a 7/10 on our Reef Certainty Scale because in almost all of the water samples we tested there were some obvious elements present which we actively try to avoid or remove from our tanks. Elements like copper, phosphorus and silicon, all of which can be removed with Reverse Osmosis or aggressive DI filtration.


What the ICP tests don’t capture are those organic compounds that may be present in our tap water coming from volatile organics, pesticides or herbicides which can also be removed with RO/DI filtration but will require more advanced testing to detect.


Mounting add-on DI filter cartridge to RO/DI system


Conclusion

The ICP test can definitively tell you that an RO/DI system is going to be needed. With 99% of successful reefers using an RO/DI system, that was sort of a given.


Since many of the elements evaluated with ICP kits can easily be removed with an RO membrane, you can choose either a 75 GPD or 100 GPD membrane because of the 96% - 99% rejection rates. Even still, some of the ICP tested elements can and will break through the RO membrane altogether. The ICP test results can then help you determine the best DI configuration to target these elements.


Phosphorus, Silicon, and copper ICP test results


For example, if you are testing for higher levels of silicone, phosphorus or copper you will benefit from specific DI filters to ensure these elements do not make it into your aquarium. Something like a multiple stage single bed DI resin approach in which the anion and cation resins are separated and utilized individually followed by a third and final mixed bed “catch-all” stage. To employ those resins correctly, you might consider and add-on like the Triple DI Saver upgrade kit.


The ICP kit will not tell you if your better suited for a higher quality sediment filter or whether the single carbon-block filter will suffice. This is why simply always choosing the best option for sediment and carbon filters is a good practice. The cost difference will be minimal but the benefits will be measurable.


To learn more about the various DI resin configurations and what the available filters can do, check out our entire Reverse Osmosis Filters video library.