Algorithm for application of internal standard on ICP-MS transient signals Stefanova, V., Kmetov, V., Plovdiv/BG UNIVERSITY OF PLOVDIV Paissii Hilendarski,

Algorithm for Algorithm for application of internal application of internal

standard on ICP-MS standard on ICP-MS transient signalstransient signals

Stefanova, V., Kmetov, V., Stefanova, V., Kmetov, V., Plovdiv/BGPlovdiv/BG

UNIVERSITY OF PLOVDIV Paissii Hilendarski,UNIVERSITY OF PLOVDIV Paissii Hilendarski, Department of Analytical ChemistryDepartment of Analytical Chemistry, , 24 Tsar Assen St., BG 4000 Plovdiv, BULGARIA24 Tsar Assen St., BG 4000 Plovdiv, BULGARIA

S U M A R R YS U M A R R Y Matrix effects in ICP-MS are commonly corrected applying the method of internal standardisation (IS). In the case of discrete sampling both plasma source and mass filter work at non-equilibrium conditions because of variable matrix load. Due to the fact that the sample injection pass is a time-depending processes overlapped by a sequential mass scanning, the matrix correction efficiency depends on the time distance between analyte and IS measurements. Well-known IS applications to transient mode refer to the ratio of both analyte and appointed internal standard peaks estimated as maximum or area. The suggested IS algorithm is based on the calculation ANALYTE / IS RATIO for every scan sweep (ESIS) resulting a new converted function - ESISratio vs time. Demanded merit of the converted function is the dynamic variable matrix influence estimation.Transient signals from three systems coupled to ICP-MS Perkin Elmer SCIEX ELAN 500 were studied: Flow Injection (FI), Injection into an air carrier (Dry Injection) and Air Segmented Discrete Introduction (ASDI). Pulse or continuos IS introduction were evaluated as well. The ESISratio was applied for 10 elements determination in VODKA by FI-ICP-MS. The sample 0.18 ml was injected into IS (Rh, Re) containing water carrier. Better signal / noise than the conventional approach was found for the converted ESISratio function. The calculated ESISratio was stable into the peak FWHM for pulse IS introduction. Up to two times reduction of signal / noise was obtained by ESISratio applied to ASDI-ICP-MS analysis of SEAWATER. The achieved matrix correction efficiency in NaCl solutions (0.1 to 3%) was in the range 105-93% .12 elements were analysed in RM - MOSS PLANT by Dry Injection (0,18 ml) with a good agreement to the reference values. Up to 6 times better Determination limits were obtained by the ESISratio function after selecting of appropriate plateau region.

Three discrete sample introduction systems have been tested: ASDI, Dry Injection and Flow Injection. First two systems gave plateau shaped profiles due to the removed sample/carrier inter-diffusion and allow pulse introduction of IS only. IS introduction by FI could be pulse or continuous.

water

water

air

air

sample

sample

sample

air

air water

water

ASDI

01000020000300004000050000

0 4 8 12 16 20 24 28 32 36 40 44

DryInj

0

10000

20000

30000

40000

50000

0 4 8 12 16 20 24 28 32 36 40 44

FI

IS

01000020000300004000050000

0 4 8 12 16 20 24 28 32 36 40 44

Air Segmented Discrete introduction mode (ASDI) [1]

Dry Injection mode (DryInj) [2,3]

Flow Injection mode (FI) [4-8]

ICP-QMS performs a sequential multi-element measurement by a number scans [5]. In

the case of small sample volume injected the sample load is changed during every

scan sweep. The total injection time should be distributed adequately optimizing Dwell

time and number of isotopes.

0 5 10 15 20 25 30 34 39

27 Al

63 Cu

88 Sr

0.E+00

1.E+04

2.E+04

3.E+04

4.E+04

5.E+04

6.E+04

Ion

/s

time (s)

Fig.2 Single and multielement registration of transient signal by FI- ICP-QMS.

Scan sweep

The new IS algorithm is based on analyte / IS ratio calculation for every single sweep of the

mass scan (Every Sweep Internal Standard). This is the shortest time cycle for both analyte

and IS measurements. The variable sample load could be estimated more precise if analyte

and corresponding IS are measured as close as possible.

ESISratio application for FI-ICP-MS (Sciex-Perkin Elmer-Elan 500) introduction of ESISratio application for FI-ICP-MS (Sciex-Perkin Elmer-Elan 500) introduction of

20% ethanol by continuous Internal standard in both sample injection and carrier.20% ethanol by continuous Internal standard in both sample injection and carrier.

A) Average value and standard deviation of every point of

the transient signal 55Mn (n=5)

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

0 5 10 15 20 25 30 34 39 44time (s)

Ion

/s

AVREAGE SD

B) Average value and standard deviation of theESISratio Mn/Rh for every signal point

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 5 10 15 20 25 30 34 39 44time (sec)

Ra

tio

Mn

/Rh

AVREAGE SD

FI is preferable for “difficult”organic

solvents [6], because of the well

known deterioration of ICP ionisation

conditions. The strong signal

suppression injecting the alcohol-

water mixtures reflects in warping of

signal profile and repeatability

worsening.

ESISratio application leads to both

improvement of the transient peak

profile and lowering SD values for

every point of the peak.

As a result a new well defined

transient peak (ESISratio versus time)

was obtained and the common

treatment procedures as PEAK

MAXIMUM or AREA could be

adequately applied on it.

Fig 3. FI introduction of 0.18ml 50 ppb Mn and IS-Rh in 20 % ethanol

Fig. 5. RSD of 5 injections Vodka + 50 ppb Standard using different IS algorithms

0

2

4

6

8

10

Al/Rh Cr/Rh Mn/Rh Cu/Rh As/Rh Zn/Rh Sr/Rh Cd/Rh Ba/Rh Pb/Re

RS

D%

Peak Maximum Peak Area

ESIS FWHM (10 points) pulse IS Area ESIS continuous IS

ESISratio application for FI-ICP-MS pulse IS introduction in sample injectionESISratio application for FI-ICP-MS pulse IS introduction in sample injection..

ESISratio was calculated for sample

injection peaks of 0,18 ml Vodka

spiked with 50 ppb Cd and Rh (IS) [7].

The region of stable ratio was

observed into the FWHM zone of the

sample peak [5]. The sum of ratios in

FWHM zone was used as quantitative

variance for further calibration and

analysis. At the optimized conditions

the constant ratio zone includes 10

points per peak.

The IS application using conventional

peak maximum and area calculation

were compared to ESISratio for both

cases pulse and continuous IS

introduction. The best obtained

repeatability shows calculation of the

peak area after ESIS application

(continuous IS introduction)

Fig.4. FI introduction 5 injections of 0.18 ml Vodka + 50 ppb Cd and IS - Rh

0

20000

40000

60000

80000

100000

1 inj 2 inj 3 inj 4 inj 5 inj

Ion

/s

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Rat

io (

Cd

/Rh

)

Rh

Cd

Cd/Rh

Fig.6. Ga and As in sea water 5 injections 11 s (0,180 ml) ASDI ICP-MS

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

1 14 28 1 14 28 1 14 28 1 14 28 1 14 28time (s)

Ion

/s

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

Ra

tio

As

/Ga

Ga As(multiplied by 5) As/Ga

ESISratio application for ASDI-ICP-MS introduction of Sea WaterESISratio application for ASDI-ICP-MS introduction of Sea Water..

Sea water is a classical example for a strong complex matrix influence. High salt content and concomitant easy ionisable elements make the ICP-MS analysis problematic. A sample from Black Sea coast was preliminary purified by extraction of trace elements and spiked with 50 ppb standard. 0,18 ml portions ware introduced directly by ASDI system. The registered transient signal has a plateau shaped profile with pseudo steady state region with a similar sensitivity as continuous introduction. The stability of calculated ESISratio (As/Ga) for 5 injections corresponds to the zone of the signal plateau Fig.6.

ESISratio statistical evaluation of the stability zone for Sea WaterESISratio statistical evaluation of the stability zone for Sea Water.

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50RS

D %

Fig.7 ESISratio RSD of 5 ASDI injections (0.18 ml) of sea water spiked with 50 ppb standard

Peak Max 2.73 3.42 2.91 3.01 3.31

Integration total 1.95 2.13 2.46 2.94 2.93

Integration FWHM 1.80 2.05 1.49 1.65 2.39

Integration Plateau 1.75 1.99 1.62 1.48 1.71

Co/Ga Cu/Ga As/Ga Cd/Rh Pb/Rh

The appropriate ESISratio stability

zone was estimate comparing the

ratio RSD of 5 injections sea water

using three different time intervals:

• total injection 7,2-31.3 s

• injection FWHM 8,8-25.6 s

• injection plateau 9,6-17.6 s

The obtained RSD slightly

decrease in the order :

TOTAL INJECTION

> FWHM

> SIGNAL PLATEAU.

ESISratio matrix correction efficiency for ASDI-ICP-MS analysis ofESISratio matrix correction efficiency for ASDI-ICP-MS analysis ofhigh salt content solutions. high salt content solutions.

Fig.8 Matrix supression correction by ESISratio. ASDI introduction of Pb (50 ppb) in

the presence of NaCl

0

20

40

60

80

100

120

0 0.5 1 1.5 2 2.5 3 NaCl % in the injected solution

sig

na

l %

ch

an

ge

Pb Pb/Rh

A very strong suppression of

ion signal for 208Pb up to 80 %

in comparison with the water

standards was corrected using

Pb / Rh ESISratio. The obtained

ratio change was in the range

105-93% for the tested matrix

loaded solutions.

The cleaning effect of washing

carrier solvent prevents the

nebulisation system blocking.

Fig.9. Dry injection (0,18 ml) signal for 57Fe, 74Ge and claculated ESISratio for 5 replicates of moss sample

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

0 13 26 38 51 64 77 89 102 115 128 140 153 166 179time (s)

Ion

/s

0

1

2

3

4

5

6

7

8

9

10

Ra

tio

Fe

/Ge

Fe 58 Ge 74 Fe/Ge

ESISratio application for Dry Injection-ICP-MS of Moss sample solutionsESISratio application for Dry Injection-ICP-MS of Moss sample solutions..

Dry Injection is the most critical sampling mode concerning the signal noise [2,3]. The “dry plasma” ionization conditions are dramatically changed due to the fast sample load. The introduction of cleaning injection between the samples is necessary to avoid the memory effect. The generated by the peristaltic pump pulsation is seen in the plateau region. ESISratio could smooth because the pulsation frequency is much lower than the ESIS registration.

ESISratio statistical evaluation of the stability zone for ESISratio statistical evaluation of the stability zone for Dry Injection-ICP-MS analysis of Moss sampleDry Injection-ICP-MS analysis of Moss sample.

The determination of the ESISratio stability zone was estimated by RSD comparison of four different sample zones ( 5 injections of moss solution): I) total injection; ii) FWHM; iii) signal plateau; iv) part of the signal plateau.

As a zone with maximum ESISratio stability appeared to be the first part of the plateau region 15 sweeps (from 14-23 s). Further decrease of the integration zone to 10 sweeps reflects in repeatability worsening as well. The selected final ESISratio region covers about 70% of the plateau zone and 50 % of the total analyte signal. Because of the strong signal pulsation it was difficult to estimate FWHM of the peak.

The optimized conditions were applied in Dry injection- ICP-MS Moss Reference material analysis. The injected sample was with dilution factor DF=100 and volume 0,18ml. Nine isotopes per injection at 60ms dwell time were measured. Two set parameters were created for elements with low and high masses respectively. 74Ge and 166Er were chosen as appropriate internal standards [9]. ESISratio application was carried out by integration of 15 sweeps. The raw data were transferred to the external PC and all calculations were made using MS Excel 97.

Limits of determination obtained by ESISratio algorithm were close to common peak area calculation and 2 to 6 times better than peak maximum (Fig. 10).

The results were compared to the moss reference material certified by interlaboratory test - 16 labs from the Nordic and Baltic region countries. A good coincidence was obtained

Referent value

Recovery

LOD (k=10) mg/kg

C mg/kg

RSD %C

mg/kg %As 0.08 1.06 5.0 0.98 108Ba 0.04 18 1.0 17.6 105Ca 48 66 2095 3.2 1910 110Cd 0.02 0.44 7.1 0.46 97Cr 1 <1 - 1.0 -Cu 0.23 66 2.5 68.7 96Fe 57 25 243 4.9 262 93Mn 0.07 368 3.6 342 108Pb 0.3 6.71 1.2 6.4 105Sr 0.02 5.26 1.2 5.3 99U 0.026 0.023 12 0.02 110V 0.1 1.47 3.5 1.4 103Zn 8.7 36.6 3.3 36.0 102

Ele-ment

Dry Injection ESIS (15 sweeps)

Moss Referent Material Analysis ResultsFig 10. LOD ratio for Peak Maximum and Area to ESISratio from Dry Injection Moss analysis

0 2 4 6

As

Ba

Ca

Cd

Cu

Fe

Mn

Pb

Sr

U

V

Zn

Bi

LOD ratio

Area IS/ESIS

Max IS/ESIS

ConclusionsConclusions ESISratio algorithm for transient signals generated by 3 injection systems coupled to ICP-MS Perkin Elmer Sciex ELAN 500 was tested.

• Flow InjectionFlow Injection was tested by both continuous and pulse IS introduction.ESISratio with continuous IS converts the transient peak signal / time into ratio / time. The converted peak has better profile and lower noise due to the dynamic matrix effect correction for every scan sweep. The ESISratio peak could be estimated as common area or maximum. ESISratio with pulse IS shows a zone of stable ratio corresponding of FWHM.For most of the elements tested signal/noise ratio decrease in the order Peak Maximum > Peak Area > ESIS pulse > ESIS continuous. The best results for FI -ICP-MS analysis of Vodka were obtained by continuous IS introduction and calculation of converted Peak Area.

•ASDI-ICP-MS ASDI-ICP-MS was tested introducing sea water. The ESISratio stability zone was determined to be in the signal plateau. The repeatability improvement up to 2 times in comparison with Peak Maximum was achieved.The ESISratio algorithm was effective in strong matrix suppression correction. The obtained change of the ESISratio Pb/Rh was in the range 105-93% for NaCl 0.1-3%

•Dry Injection -ICP-MSDry Injection -ICP-MS was applied for Moss sample analysis. The registered transient signals were burdened with strong pulsation generated by peristaltic pump. The ESISratio algorithm allowed to reduce significantly the periodic noise improving the figures of merit. The ESISratio stability zone selection for Dry injection was found to be the most critical in comparison to other introduction systems studied. For 0,18 ml injection and 9 isotopes at 60ms dwell time it was found that the ESISratio stability zone is ~70% of the signal plateau region which corresponds to 15 sweeps.

The obtained recovery of analytical results for Moss RM were in the range 93-110%. The ESIS application improves LOD in comparison with conventional transient signal peak maximum and area IS by averaged factors up to 6 and 3 respectively.

Acknowledgements:Acknowledgements:the authors would like to express their special thanks to:the authors would like to express their special thanks to:

• German Chemical Society, EWCoPS Organazing Committee and TEAM Ltd Bulgaria

for thegrant in conference participation and finacial support given to Veselin Kmetov

• Prof. L. Yurukova from the Institute of Botany at BG Academy of Science for Moss RM supplied

REFERENCESREFERENCES

1. Stefanova V., Kmetov V., Futekov L., JAAS, 12, 1271, (1997)

2. Houk, R.S., Thompson, J.J., Biomed.Mass.Spectrom., 10, 107 (1983)

3. Beauchemin D., Analyst,118, 815 (1993)

4. McClenathan D. M., Ray S. J., Hieftje G. M., JAAS, 16, 987 (2001)

5. Denoyer E.R., Atom. Spec, 15, 1, 7 (1994)

6. Dressler V.L., Pozebon D., Curtius A.J., Anal.Chim.Acta, 379, 1-2, 175 (1998)

7. Stefanova V., Kmetov V., Futekov L., Georgieva D., Analytical Laboratory, 6, 1, 43 (1997)

8.Benkhedda K., Infante H.G., Ivanova E., Adams F.C., JAAS, 15, 10, 1349 (2000)

9.Vanhaecke F., Vavhoe H., Dams R., Vandecasteele C., Talanta, 39, 7, 737 (1992)

Fig.6. Ga and As in sea water 5 injections 11 s (0,18 ml) ASDI ICP-MS

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

1 5 9 13 17 21 25 29 33 37time (s)

Ion

/s

0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

ratio

As/

Ga

Ga

As(signal*5)

As/Ga 1 inj

As/Ga 2 inj

As/Ga 3 inj

As/Ga 4 inj

As/Ga 5 inj

ESIS application for ASDI-ICP-MS introduction of Sea WaterESIS application for ASDI-ICP-MS introduction of Sea Water..

Sea water is a classical example for a strong complex matrix influence. High salt content and concomitant easy ionisable elements make the ICP-MS analysis problematic.

A sample from coast region of Black Sea was purified by extraction of trace elements and introduced directly by ASDI system. Standard solution (50 ppb) was added to the sea sample. The removed sample diffusion leads to similar sensitivity as continuous introduction[1]. The registered transient signal has a plateau shaped profile with pseudo steady state region. The stability of calculated ESIS ratio (As/Ga) for 5 injections corresponds to the zone of the signal plateau Fig.6.