Please , show the matrix constituents in final sample solutions and after that can discussed the problem " line selection". Please, note the type of background correction procedure.

Thank you for your attention!

I use ICP-AES for quantative analysis of KTP crystals(KxNa(1-x)TiOPO4). I dissolve them as discribed in the article "Application of inductively coupled plasma atomic emission spectrometry in analytical control of single crystals of potassium titanylphosphate: spectral interferences, line selection, detection limits and trace element determination" (N. Daskalova, S. Velichkov, P. Slavova, E. Ivanova, L. Aleksieva). Thus there are Ti,P,W(approximately 1% of crystall), K, Na, H2SO4 in the investigated solution. Also we use standart additions to calibrate so in solutions with additions there are a lot of elements from standart: Al , As, Ba, Be, Bi, B, Ca, Cd, Ce, Co, Cr, Cs, Cu, Dy, Er, Eu, Ga, Gd, Ho, In, Fe, La, Pb, Li, Lu, Mg, Mn, Nd, Ni, P, K, Pr, Re, Rb, Sm, Sc, Se, Na, Sr, Tb, Tl, Th, Tm, U, V, Yb, Y, Zn (we have to use ICP-MS standarts). I correct background using software approximation or flat background substraction.

The crystal could be contaminated by traces of Na from the starting substances, from the furnace materials, utensils and ambient air during the synthesis. We use reagents of highest grade were used (Suprapure, Merck) and tri-distilled water from a quarz aqpparatus. Because of this we determine Na as impurities and Na , K or Rb in structural analogues of KTP. Please, see the following our publications : N. Daskalova, L. Aleksieva, G. Gentsheva, S. Velichkov, Analytical control of the preparation of single crystal materials by inductively coupled plasma atomic emission spectrometry, Spectrochim. Acta Part B 57 (2002) 755-768.

Calibration procedure

Standard addition method is not appropriate for calibration in ICP-OES, because of the high level of spectral interferences – line and background interferences.

The reference solutions for determination of Na and K were prepared on the basis of a blank containing the acids for digestion (H2SO4 -142 mg ml-1, H2O2 – approximately 0.005 mg ml-1). HF was removed. This is in accordance with our dissolution procedure.

The reference solutions for determination of Na and K were prepared on the basis of a matrix blank containing : acids for digestion (H2SO4 -142 mg ml-1, H2O2 – approximately 0.005 mg ml-1). HF was removed. To acid blank were added Ti and P (from mono standard solutions of Ti and P). The concentration of Ti and P are the same, as well as in final sample solution (without K an Na).

Therefore matrix blank contains: H2SO4 -142 mg ml-1, H2O2 + Ti + P

The reference solution were prepared on the basis of a matrix blank + Na and K depending on the necessary concentration range for Na and K.

The following signals were measured at the peak wavelength a :

1. Na I 588.995 nm or K I 766.023 nm from a given reference solution: Xprominent line + Xmatrix blank

2. Matrix blank Xmatrix blank

3. Net line signal = Xprominent line + Xmatrix blank - Xmatrix blank

Our experience show that only the background correction with matrix blank in peak wavelength a ensures the accuracy of results.

With king regards, N. Daskalova

Thank you very much! I really appreciate your help!

Concerning lines of K and Na, do you often use just one line for determination of elements? Or it's just coincidence (other lines seem to be "bad")?

I can\t find the line K 766.023 nm in spectrometer software( VARIAN 720-es) nor in Boumans tables. I've just found the line K 766.491 nm. It is the same, isn't it?

Please, use K 766.491 nm. You can use different prominent lines of K and Na in depending on the sensitivity of line and concentration of these elements in final sample solutions.In all cases you have to applied background correction with matrix blank .

Thank you! I hope your answers will help us to solve our problems with analysis of KTP crystals.