The Ultimate Salt Test

[Jeeperz]

Maybe @Randy Holmes-Farley can chime in on this

I mean the part about IO having no standards to their salts. I do believe it's him that uses IO purple but it could have been someone else high up on the experience list that recommend io purple as it's supposedly consistent

 

[Privateye]

Weird question, but do you take requests? I know of a couple of anecdotal ICP tests that found higher levels of a certain contaminant in a specific salt recently. One that was not tested here. The manufacturer denied these claims, but the person (a Ph.D) tested their source water and 3 different brands of salt and only found the contaminant in this brand. Their organisms were dying in correlation. I don't want to throw shade on the manufacturer without further testing though, so PM me if you're interested.

 

[rtparty]

One question - some of those salts have different mixing instructions - did you follow them all - or mix all the same way( to my reading you mixed them all the same way). Additionally alkalinity is not measured by ICP - Curious if you did an independent measure - at your house - and what this showed (alkalinity wise) second did you measure the salinity of each product independently/. Not sure that dosing per weight is the way to do it (but rather according to manufacturers instructions)?

I followed any special instructions each salt had. If they didn’t have any instructions, I followed my own protocol outline in post 1.

ATI measures carbonate hardness with their ICP test. It was routinely .5-1dkh lower than what I was getting with Salifert and Hanna (when used.)

I tested alkalinity, calcium, and magnesium with Salifert kits at home when I collected the ICP samples. I have all these numbers in my notes that I will update the OP with at a later date.

I have a chart from BRS where they tested how many gallons each bucket makes compared to what the manufacturer states you’ll get. So X amount of grams per gallon of water. I followed that chart and found it spot on. Salinity was tested with a calibrated refractometer and a Milwaukee refractometer.

All this got me within range and well within the scope of our hobby grade kits. Most manufacturers do not give weight but instead volume and volume is the absolute worst way to mix up salt and additives for our tanks. A half cup doesn’t equal a half cup but 755 grams always equals 755 grams. If that makes sense

 

[rtparty]

Weird question, but do you take requests? I know of a couple of anecdotal ICP tests that found higher levels of a certain contaminant in a specific salt recently. One that was not tested here. The manufacturer denied these claims, but the person (a Ph.D) tested their source water and 3 different brands of salt and only found the contaminant in this brand. Their organisms were dying in correlation. I don't want to throw shade on the manufacturer without further testing though, so PM me if you're interested.

When it comes to contamination and impurities that we don’t want, I’m far from an expert on the subject. ICP-OES testing doesn’t have a low enough range to see some of these impurities. ICP-MS is better for this but also much more expensive.

At $45 per test and all the salt used, this was over $1000 already.

I’ve got at least 5 more salts I’d like to get added to the spreadsheet but need to find the time and funds.

 

[MnFish1]

I followed any special instructions each salt had. If they didn’t have any instructions, I followed my own protocol outline in post 1.

ATI measures carbonate hardness with their ICP test. It was routinely .5-1dkh lower than what I was getting with Salifert and Hanna (when used.)

I tested alkalinity, calcium, and magnesium with Salifert kits at home when I collected the ICP samples. I have all these numbers in my notes that I will update the OP with at a later date.

I have a chart from BRS where they tested how many gallons each bucket makes compared to what the manufacturer states you’ll get. So X amount of grams per gallon of water. I followed that chart and found it spot on. Salinity was tested with a calibrated refractometer and a Milwaukee refractometer.

All this got me within range and well within the scope of our hobby grade kits. Most manufacturers do not give weight but instead volume and volume is the absolute worst way to mix up salt and additives for our tanks. A half cup doesn’t equal a half cup but 755 grams always equals 755 grams. If that makes sense

Actually - ATI - just measures alkalinity in the sample using a test - not ICP (that was my point). Now perhaps their test is more accurate than, perhaps a Hanna checker. I was just pointing out that it wasn't an ICP measurement.

Thanks for the rest of the information!!

 

[rtparty]

Actually - ATI - just measures alkalinity in the sample using a test - not ICP (that was my point). Now perhaps their test is more accurate than, perhaps a Hanna checker. I was just pointing out that it wasn't an ICP measurement.

Thanks for the rest of the information!!

I never said it was an ICP measurement. Just pointed out their results were consistently lower than mine. I inquired with Randy why that was.

Our home kits test total alkalinity but ATI only tests carbonate hardness. There is a difference between the two. That’s my only point.

 

[Randy Holmes-Farley]

I mean the part about IO having no standards to their salts. I do believe it's him that uses IO purple but it could have been someone else high up on the experience list that recommend io purple as it's supposedly consistent

Not sure what you mean.

IO doesn't publish their specifications, but they certainly have them.

 

[Randy Holmes-Farley]

Our home kits test total alkalinity but ATI only tests carbonate hardness. There is a difference between the two. That’s my only point.

Do they?

ATI says they do it by titration, which gives total alkalinity. The only way to get to carbonate alkalinity is if they subtract out contributions like borate. Do you know they do that? It's not a trivial thing to do. They would have to measure boron by ICP, then use a pH and pKa calculation to determine how much is borate and how much is boric acid.

Choose the Right ICP Testing for Your Reefing Needs: a Comparison - ATI North America

ICP testing is one of the best ways to get a closer look at what is going on in your reef, but not all ICP testing is the same.

www.atinorthamerica.com

Alkalinity Testing​

ATI uses a titrator to provide users with the most accurate Alkalinity readings possible.

 

[rtparty]

Do they?

ATI says they do it by titration, which gives total alkalinity. The only way to get to carbonate alkalinity is if they subtract out contributions like borate. Do you know they do that? It's not a trivial thing to do. They would have to measure boron by ICP, then use a pH and pKa calculation to determine how much is borate and how much is boric acid.

Choose the Right ICP Testing for Your Reefing Needs: a Comparison - ATI North America

ICP testing is one of the best ways to get a closer look at what is going on in your reef, but not all ICP testing is the same.

www.atinorthamerica.com

Alkalinity Testing​

ATI uses a titrator to provide users with the most accurate Alkalinity readings possible.

When I inquired about it I was told they measure carbonate hardness and not total alkalinity. I can inquire again and see what they say.

.5-1dkh isn’t end of the world though and well within my own testing errors

 

[rtparty]

Not sure what you mean.

IO doesn't publish their specifications, but they certainly have them.

Any company using food grade or pharmaceutical grade ingredients is going to market that. It’s a selling point and something to hang their hat on.

 

[taricha]

This is excellent. Thanks a bunch!
You said you did hobby kit measurements too - but I can't find them in the spreadsheet. (haven't watched your video yet)

Your data produced a large number of distinguishing features between the salts that are bigger than ICP variation. Fancy way of saying most of what's in the table is real differences in salt mixes.

Zinc, Vanadium, Cobalt, and Chromium differences are small enough and likely less than the ICP uncertainty.

 

[rtparty]

This is excellent. Thanks a bunch!
You said you did hobby kit measurements too - but I can't find them in the spreadsheet. (haven't watched your video yet)

Your data produced a large number of distinguishing features between the salts that are bigger than ICP variation. Fancy way of saying most of what's in the table is real differences in salt mixes.

Zinc, Vanadium, Cobalt, and Chromium differences are small enough and likely less than the ICP uncertainty.

I did at home testing of the big 3. I have all those results in my notes but haven’t added those yet. It’s very short hand and raw data that I added as I tested and noticed things. I did write down exactly how much salt I used of each salt mix so those who want to weigh will have a good idea.

I am hoping to add more to the first post tonight as time allows

 

[Jeeperz]

Not sure what you mean.

IO doesn't publish their specifications, but they certainly have them.

The op was implying io has no standards to their salts, like no qc or manufacturing standards or something is what I understood. But I'm pretty sure public aquarium use io and they would definitely want a salt with standards. Maybe I'm misunderstanding what was meant by no standards

 

[rtparty]

The op was implying io has no standards to their salts, like no qc or manufacturing standards or something is what I understood. But I'm pretty sure public aquarium use io and they would definitely want a salt with standards. Maybe I'm misunderstanding what was meant by no standards

They don’t state food grade or pharmaceutical grade ingredients from what I could find. That’s what I mean by that. They obviously have to have some QC but I don’t think it should shock anyone if their numbers aren’t super consistent (and I think this can be said for a lot of salts on the market. Not just IO.)

I actually talked a lot about IO Purple in the video. Since the live with Deven I’ve had a few people reach out to me and learned some great things about the manufacturing of salt for our hobby.

 

[taricha]

the IO vs RC anomalous result is interesting.

Something isn't right with your IO tests. No way is regular IO dkh 11 and reef crystals at dkh 9.4

I address this in the video but the results matched my own kits.

Your home test kits confirm the ICP is not a blown test.

It says that it's worth checking how our salt actually mixes up - even for very widely used salts like IO and RC.

So it's not anomalous tests, it's anomalous bag of salt.

 

[taricha]

I did at home testing of the big 3. I have all those results in my notes but haven’t added those yet.

Great! @Rick Mathew @Dan_P And myself will be very interested to see these.

 

[Jay Hemdal]

I wanted to get this thread up and going before the holidays soak up all my time. I did a live YouTube with Deven from ReefDudes going over things but it is about 1.5 hours long. I will write up a more detailed explanation here in this first post soon.

I am a hobbyist first and foremost. I have no formal training on exact testing methods or anything like that. I did the best I could and think my method was good enough for our use. I am not taricha

TESTING METHOD

All salts were purchased randomly so no cherry picking from a company. Not a sinlge company knew I was doing this. I used a BRS 7 stage RODI unit that had all new filters and tested 0's across the board on my own tank's recent ICP. I then monitored every RODI result with each test sent in to see how the filters were doing. We saw all 0's the whole time. 15 gallons of water was added to a grey Brute trash can and each salt was WEIGHED. I would heat the RO water up to 74F before adding salt UNLESS the salt specifically stated otherwise (Red Sea Coral Pro for example.) Salt was added slowly to not cause precipitation.

All salt was mixed for 24 hours (except Red Sea Pro) before the ICP was collected and I ran my own home testing on Salifert test kits that were brand new for this test. Why did I choose Salifert? Becasue Amazon had them to me next day when I realized all my Red Sea kits had expired or were super close.


YouTube Video:


Google Sheets Link:




TLDR version: I sent off brand new batches of saltwater to ATI for ICP testing. These are the results. No bias. No sponsorship. All paid out of my own pocket and on my own time.

I think you need to adjust the PSU to be equivalent for all three samples in order to make an easier comparison. If one mix was 5% higher in a given element, but was also 5% higher in PSU, wouldn't that be a wash?

Testing sea salts is a tough thing to do. Here is an update of a test that I had done some years ago:

Sea Salt testing - 2022 update

Evaluating Synthetic Sea Salts C. Jay Hemdal 2016 Revised January 12, 2022 This is an update of my 2016 article that examines some testing methods to evaluate synthetic seawater, some of which can be done at home. A short history and list of...

www.reef2reef.com

Jay

 

[rtparty]

I think you need to adjust the PSU to be equivalent for all three samples in order to make an easier comparison. If one mix was 5% higher in a given element, but was also 5% higher in PSU, wouldn't that be a wash?

Testing sea salts is a tough thing to do. Here is an update of a test that I had done some years ago:

Sea Salt testing - 2022 update

Evaluating Synthetic Sea Salts C. Jay Hemdal 2016 Revised January 12, 2022 This is an update of my 2016 article that examines some testing methods to evaluate synthetic seawater, some of which can be done at home. A short history and list of...

www.reef2reef.com

Jay

I wasn’t testing the salts against each other in any way. Simply mixing and and sending off for ICP. I knew there would be margins of error at play.

I think ICP testing like I’ve done is only a small part in “what makes a good salt” and should not be THE determining factor in why a hobbyist chooses a salt over another.

My own priorities and grading scale are very likely not the norm.

 

[taricha]

based on Jay's suggestion....

I think you need to adjust the PSU to be equivalent for all three samples in order to make an easier comparison.

I wasn’t testing the salts against each other in any way. Simply mixing and and sending off for ICP.

And in spite of rtparty's not intending it as a comparison, I did a little math for those who DO want to use the data for comparisons....
This sheet below has the data with all test values normalized to 35 salinity as Jay suggested. It also has the % variations of each one from the average (mean, unless outliers - then I used the median). And for the statistically inclined there's also included the # of standard deviations of the measured value above or below the average.

The color coding is where I went back and looked at ICP test notes from work that @Dan_P @Rick Mathew and myself did.
I color-coded those values that were further above (orange) or below (blue) average than typical variation in the ICP results for replicate samples in that element.

For example, no salts had Na or Cl variations larger than the +-5% or so that was typical in ICP tests of replicate samples for those. (expected - since the salinity was corrected to 35)
Some had Mg variations in excess of the upper single digit variations we saw with Mg. Sr measurements of replicates could be off by +-25% so only one salt was further from the average than that.
edit: darker blue or orange if the difference from average was more than about double they typical ICP variations we saw.

enjoy. Happy Thanksgiving!

 

[Jay Hemdal]

based on Jay's suggestion....




And in spite of rtparty's not intending it as a comparison, I did a little math for those who DO want to use the data for comparisons....
This sheet below has the data with all test values normalized to 35 salinity as Jay suggested. It also has the % variations of each one from the average (mean, unless outliers - then I used the median). And for the statistically inclined there's also included the # of standard deviations of the measured value above or below the average.

The color coding is where I went back and looked at ICP test notes from work that @Dan_P @Rick Mathew and myself did.
I color-coded those values that were further above (orange) or below (blue) average than typical variation in the ICP results for replicate samples in that element.

For example, no salts had Na or Cl variations larger than the +-5% or so that was typical in ICP tests of replicate samples for those. (expected - since the salinity was corrected to 35)
Some had Mg variations in excess of the upper single digit variations we saw with Mg. Sr measurements of replicates could be off by +-25% so only one salt was further from the average than that.
edit: darker blue or orange if the difference from average was more than about double they typical ICP variations we saw.

enjoy. Happy Thanksgiving!

I should add, I used the caveat of "I think...." in regards to adjusting the salinities of the samples before making comparisons. Is that a valid conclusion? Then, can the adjustment be made mathematically, or would you need to start with samples that were all diluted to the same salinity? Math is not my forte (grin).

Jay