USING YOUR PHOSPHOROUS HANNA CHECKER HI-736 TO TEST FOR LOW LEVES OF NITRATES
Over the last few weeks I have continued my quest to find a better way to measure low levels of nitrates without having to do color interpretation. Back in October I posted my first efforts. You can find it here: https://www.reef2reef.com/threads/diy-nitrate-test-meter.330892/#post-4114495
From the feedback on this posting I proceeded with the work. Randy provided some insights based on some information from HACH on using a standard salt solution for the calibration (also included in the previous post). Based on my tests the calibration does need to be done using the seawater calibration procedure as seen by Fig 1
FIG 1
This would necessarily mean that the standard solutions would need to be prepared at or near the chloride level of my display tank. That being said I prepared all of my standard solutions in accordingly. I prepared calibration standards at .25ppm, .5ppm, 1ppm and 2ppm using the Hach Nitrate-Nitrogen Standard Solutions (converted from NO3 – N to NO3) I chose these levels because they presented the most difficulty in measuring with my current colorimetric method.
I started the measurements using the Hanna HI-96702 Copper High Range Portable Spectrometer at the .25ppm level. The results were that at this level the instrument was unable to detect the presence of NO3. I tested 5 samples and got mostly “0” or “.01”. It was slightly better at the .5ppm level but the data scatter was too large. So this method would not work….Enter the Phosphorous Checker!!
I noticed when I was doing my monthly check on the Phosphorous checker using the Range Certified Standard Kit that the 100ppb sample was a very light pink…just like the low level (.25,.5ppm) NO3 samples. I wondered if this checker might be able to detect these low levels of Nitrate. I ran 5 tests of the .25ppm using the Phosphorous checker and the results were encouraging (See Table 1)
Table 1
.25ppm NO3 Test Data
With these encouraging results I proceeded to run all 4 levels. I ran 5 separate tests at each level (20 tests) and charted the date. (See Table 2)
Table 2
Meter Readings on HI-736 Checker
Using the data averages I then created a regression equation using Excel that best fit the data. (See Fig 2)
Fig. 2
Regression Equation from Excel
With the correlation coefficient (R2 ) being so high this indicated this might be a winner!. I further refined the equation using a program called MyCurveFit, which is also useful in getting other valuable statistical information that tells you about the quality of the fit and how accurate your predictions might be (F-Test, p-test, and Standard Error of the Estimates) …(Here is a link to the website if you are interested https://www.mycurvefit.com/ ) After refining the equation and looking at the regression analysis data it appears this will work.
Here is the Equation I am using. Y= -.000003258494 x X2 +.01296094 x X +.0534317
Y= the Outcome---The NO3 in PPM X= The HI-736 measurement reading
I am not sure if this equation will work with another HI-736 without creating new calibration curve because I have not tested it.
I have been using this method now for about two months.(See Table 3) I addition I have data from outside testing services (AWT and ATI). I continue to run the NYOS test in parallel with this method. If the level on the NYOS is below 1ppm I have to estimate the value. I also cross check with the Red Sea color wheel to see if I am in the right range…The color differentiation with the Red Sea wheel at the lower levels is still a bit difficult for me to distinguish but it is a good marker to see if my meter readings are in the range.
The procedure is quite simple:
1) Run the Red Sea Nitrate Pro Test according to the procedure.
2) After 7 minutes of the 9 minute reaction time (2minutes remaining) I place a cuvette of the unreacted test water into the Phosphorous Checker and measure this as a blank (C1).
3) I then empty the cuvette and wait for the reaction time to complete.
4) When it is completed I decant off 10mL of the reacted solution and fill the cuvette.
5) I pace the cuvette into the Hanna Phosphorous Checker (HI-736) and press the button (C2). I do not hold the button in to activate the 3 min timer…I just press the button and get the reading.
The short 7 min video below shows the test procedure. The video does not detail the Red Sea Test but only shows the procedure.
Data Results can be seen in Table 3
Table 3
Reference Measurements compared to HI-736 testing results
*These are visually estimated values using NYOS test results and the Red Sea Pro Color Wheel
From my perspective the data looks promising. The errors from the estimated reference measurement are small, most 12% or less with a few (3) exceptions (See Table 3). With my current data set the Standard Error of the Estimate is about .07-.12 depending on the level in PPM. I am going to continue to gather additional data from my own testing as well as outside sources. This will help to refine the regression equation as well as my testing method to increase the accuracy of the prediction. My goal would be to have the Standard Error of the Estimate to be 10% or less of the level tested…. 0.025 for the .25 ppm, 0.05 for the .5ppm etc.
It is important to note that the reference measurements (with the exception of the external tests and knowns) are based on visual assessment using the NYOS color card and the RED SEA PRO Nitrate Wheel. This being the case the value is left to my interpretation. I do the best I can to differentiate between levels that are not specifically on the color chart… EXAMPLE..Looks greater than .25 but less than .5...Looks closer to .5 than .25…must be a .3 or .4 so I take my best guess. I do this first before I do the calculation so as not to influence the value of the reference by the results of the calculation. None the less as you can see the data lines up very well.
I have much more confidence in the known samples that I create for standards and have seen some very good results from these. I only have one outside data point at this time with their value being at .4 ppm and mine at .57 ppm. After reading the posts on R2R on Aqua Medic testing as well as the a Reefs Magazine article by Rich Ross & Dr Maupin on Triton Lab ICP-OES Testing (link here http://packedhead.net/2015/triton-lab-icp-oes-testing-of-a-certified-artificial-saltwater-standard/ ) , I don’t have a high level of confidence that external testing is all that solid of a reference point…In my opinion. I have yet to evaluate the ATI data results and would appreciate any feedback from the community on your experience.
Well that about sums it up. I will continue on my quest and will provide future updates as they are available. For anyone out there who would like to see if they can replicate these results I would be happy to provide any additional details you might need.
FUTURE WORK
· I will also be working on an Iodine test following a similar procedure.
· Refine regression equation for Hanna HI-736 /Red Sea Procedure
· Explore other Hanna Checkers to see what kind of results they produce
· 2 other tests that might be worth exploring…Potassium & Silica
Rick
Last edited: