WEF Discussion Forums
Laboratory Management and Technical Issues
COD Testing using HACH TNTPlus
Our lab presently tests the COD of our effluent samples using a HACH DR/890 Colorimeter.
We have found that the results for the same sample can vary even if the glass vial is inserted the same way everytime.
I have recently come across a different way to test for COD, using HACH's TNTplus vials and different spectrophotometer. With this new system, the vial is rotated in the meter with 10 readings taken, then averaging the readings for a more accurate result.
Has anyone had experience testing COD using TNTplus? I don't know the cost of running these tests yet versus the method we use presently, however, I can assume it is more expensive. If someone has had experience or have heard some feedback on this, can you please provide me with some feedback. I am trying to decide if this is something we would should purchase.
When you say you have found a variance in the sample, are you referring to the same tube inserted in the spectrometer reads differently or that the same sample yields different results when digested multiple times? What magnitude of variance are you talking about? Are you getting pre-made vials with the reagents in them or are you filling the vials yourself? It sounds to me like somehow the digested solution is being disturbed and particulates are causing fluctuations in the reading. I'm not familiar with the TNT Plus system, but I'm not sure how a regulator would look upon the 10 measurement averaging system. The method does not call for anything similar to that in making the reading.
We have some variance in our readings for the same sample in the same vial. We use the HACH vials that are already made up with reagents. I think the varience is caused by the variations in the glass. I can put in a vial and get a reading of 400 mg/L then take it out and put it in again and get a reading of 480 mg/L. Sometimes the variation is greater than that.
This test is not a permit test for us, so I am not overly concerned about the accuracy as we use the test mainly to anticipate our BOD results.
From my understanding, the TNTplus vials will eliminate the variability I am experiencing because the vial spins in the spectrophotometer taking 10 readings, eliminating any outliers, and averaging the results.
I was mainly curious if anyone has used this technology yet and if it is worth switching over to.
We use Hach vials with the Hach 4000 or 890 meters. Maybe your cell holder is worn and does not seat uniformly. We see maybe 20 mg/L difference at 800 mg/L COD. Are you mixing the tubes, as cooling, to eliminate the water condensate layer? Difference in glass should not make any difference reading the same tube if you insert it the same each time.
I also use the DR890 and see the same phenomenon you do. When I read about the TNT method and that it spins the vials during the readings it confirmed my thought that the glass vials are not uniformly made and light transmittance is affected by the imperfections. Otherwise, why spin the vial? Don't buy the new TNT supplies. Here is what I do with my DR890: Take 4 readings per vial with a quarter turn spin of the vial between readings and average the 4 results. This accomplishes the same thing as the TNT equipment.
We've had the same problem on our DR2100. I don't know how your DR890 works but I think it could be the same issue.
This was what I found at the time:
When inserting the vial into the spec, the orientation of it in terms of around the circumference of the vial caused the numbers to deviate sometimes +/-15 mg/L COD. We deal with low range COD 0-150mg/L for wastewater analysis so it was a big deal. I ran a small study where I designated the circular HACH label on the vial as 0 degrees and then took readings while manually rotating around an imaginary 8 point compass and watched the readings fluctuate until it went back to the original value at 0 degrees.
When I told HACH of my findings, they suggested that I try their new DR2800 spec with TNTplus. Their reasoning was that the old spectrophotometers had outdated optics systems and these new ones would be much better. They gave us a loaner instrument for a couple of weeks where I repeated the same rotation experiment and sure enough, after rotating the 8 point compass, the COD deviations was only +/-1 mg/L, much better precision. I believe that the old systems' optics were so weak that the minor imperfections in vial glass thickness greatly affected the absorbance readings since cell-wall thickness is a big variable in Beer's Law.
With regards to TNTplus... if you are only analysing COD, it isn't worth the extra 4-digit expense. I believe that you've confused TNTplus for something else. The TNTplus on the DR2800 did not auto-rotate the vial and average the numbers. It is really just a barcode scanner built into the spec so that when you insert TNTplus COD vials, the instrument automatically knows that you are analysing a COD sample, and configures the spec to do such. So you pay for the barcode feature, then are forced to buy barcoded vials, all to save you the time of selecting COD analyses in the main menu. TNTplus is more for people who analyse for many different parameters on the same spec at the same time.
We ended up purchasing the DR2700 model, which is just the DR2800 minus the TNTplus feature.
I hope that helped.