Hello,

What do you use as a blank sample. I guessed that your calibration zero value is higher than your sample value. That is why You got negative value. Please, check it and note here.

Hi. Yes, you need to check your blank(s) and calibration.

Hello,

Yes your problem is juste your blank(s) and calibratiion. But if you have concentrations traces

I agree with all the previous comments, which provide good suggestions. One thing to consider is your calibration range, if your calibration range is much higher than the samples you're analyzing, then the accuracy at the lower end of the calibration range may decrease, to the point where samples that are just low concentrations read off the curve as negative concentrations.

As the others stated, blank subtraction may also be an issue. Make sure that you are using ultra-pure reagents and that your "blank" actually is blank.

This probably goes without saying, but make sure that you're using the most sensitive lines for your analysis that don't have any interferences to be concerned about.

If you're doing any sort of sample dilution, consider backing off of the dilution factor some if matrix effects aren't a concern.

Finally, while this may not be an option if you don't have one of these just laying around, you may consider switching to ICP-MS if your concentrations are still too low for ICP-OES.

Anyway, I hope some of these suggestions are helpful!

Thank you sir all of you for your valuable suggestions

Sreenivasulu,

If your samples were from a marine environment, presumably they were sea water or brackish water. That is, they have a high salt content. Your standards should have the same high salt content. In common parlance, your samples and standards need to be matrix matched: precisely so for ICP-MS, approximately so for ICP-AES.

I have worked quite a bit with the Optima 7000 and found that quantification limits are typically ~0.4 ng/g for Mn and ~ 3ng/g for Fe. A few tests with Ni pointed to values similar to Fe and I would expect the same or worse for Co. These values are comparable to concentrations measured in sea waters not strongly affected by anthropogenic input. If the Optima 8000 has similar sensitivity as the Optima 7000 you could probably just about confirm that the elements you're interested in are present in your test solution (that is if you were to measure your seawater undiluted).

I guess your lab manager would not be very fond of this idea, though; I would not recommend measuring solutions with cation concentrations much higher than 100 µg/g. Such high concentrations tend to cause all sorts of problems (memory effects, viscosity, deposition of salts on the torch, mirrors and in capillaries ...).

Diluting your solutions won't be an option, because then you'll certainly end up somewhere in the baseline noise with your analyte elements. So I think you'd have to look into measuring on a well-maintained Quadrupole or Sector field ICP-MS.

All of the answers given above are true but you should, ideally, also be using matrix matched standards, where the matrix is that of standard seawater. Just match the major cations and anions, not the trace elements.

You could also try a standard addition method. This would avoid the requirement to matrix-match your standards.

I agree with all partially.

Practically its is so obvious getting negative values for marine sample analysis especially in trace amounts.

We generally are compelled to do the following and we must be extra cautious at each of the steps we take during our analysis starting from standards preparation to final unknown calculations.

First of all salt and other componets in marine water definitely matters. Hence all the other basic parameters to be carried out for the marine sample whose concentration is to be determined. 

Although the standards are to be prepared using distilled water, we have to carry parallelly another set of (atleast 2) standards with aged and suitable filtered marine water. Then compare whther we get good values with marine standards also. 

If yes then proceed and IF NOT then redo it. Again if the same errors get repeated then note the blank value with pure distilled water with and also with reagents etc. 

After measuring the absorbance values of all the standards, blanks, samples    justify yourself where it went wrong. More or less this type of exercise is a must for each set of marine samples.

Reasons are plenty and its a different chemistry and dont bother yourself much.

This is a must what ever be the sophisticated instruments you are using.

Keep doing and with practice you can understand what to do and what not do...

Contact me for further clarifications if any....

Good luck.

What do your calibration curves look like? 

Water samples should be preserved with dilute HCL or another suitable acid, [NOT H2SO4] and kept refrigerated until analysis.

Sample acidification prevents metals sticking to the walls of the sample container, stops all biological activity and breaks apart colloids, homogenizing the sample. 

This practice could help your analysis as well as acid washing your sample containers prior to sampling. It is also worth noting that seawater samples experience significant amounts of spectral interferences so matrix modification will be necessary.

Dear Sreen

I agree about the fact that you need to take care of your calibration values, as normally the concentrations of trace metals in water are so small. Therefore, you calibration concentrations should be small at the scale of nM or uM,

I have a question for you please, did you use CRM? if so, how did you interpret the numbers obtained in the ICP OES report compared to those given in the certificate of the CRM's?

Dear Sreenivasula,

You don't say what elements you were seeking, nor whether they were likely below detection limits for AES, so its hard to help.

Did you matrix-match your samples and standards? That is, were your standards made up in artificial seawater made from solutions free of the trace elements targeted?

Finally, what is the 2sd of the blanks that you measured? Maybe your elements are below detection and you simply have essentially zero trace elements in your seawaters (ie. below detection) plus a slight bias in those non-detects (from matrix mis-match) that puts them all a bit lower and so all negative. Remember, noise averages to zero, so analysis of a series of samples that are genuinely below detection  should actually return negative concentrations for half of them. Add a negative bias of about the detection limit and almost all will be negative.

Analysis ain't easy.

John McArthur

Hi,

Most earlier comments are valid to improve your calibration and you need to confirm your detection limits to asses if your negative values have any real meaning. You should do a standard addition to assess the concentration of your target analytes, this way you will account for matrix effects.

Negative value of Trace elements, it is surprising. Check the standard solutions and calibrate the instrument perfectly and repeat the analysis.