Folie 1

1
Fundamental Principles and Selected Applications
of the Isotope Dilution Technique for Elemental
Trace and Elemental Species Analysis by ICP-MS
Klaus G. Heumann Institute of Inorganic Chemistry
and Analytical Chemistry Johannes
Gutenberg-University Mainz, Germany
2
Historical view of IDMS
External calibration of ICP-MS measurements
Increasing number of elemental trace ICP-IDMS
analyses
After 1990
The first elemental species analyses by
HPLC/ICP-IDMS appeared (Ebdon et al., Heumann et
al.)
1994
From Ref. 5
3
The principle of isotope dilution mass
spectrometry (IDMS)

? Only the isotope ratio of the isotope diluted
sample (R) must be measured and not an absolute
amount of the analyte !

? Isotope ratio measurement is independent on
Matrix effects
Signal drifts of the instrument
? Sample loss does not affect the result after
the isotope dilution step has taken place
(determination of recovery not necessary)
4
The principle of isotope dilution mass
spectrometry (IDMS)

R (NS x hS2 NSp x hSp2) / (NS x hS1 NSp x
hSp1) R isotope ratio of isotope diluted
sample (to be measured) NS, Sp number of
sample and spike atoms (NS unknown) h1,2 isotope
abundance of reference and spike isotope ()
5
Schematic representation of the isotope dilution
equation for lead determination (poly-isotopic
element)
R(208Pb/206Pb) (NSxhS208 NSpxhSp208) /
(NSxhS206 NSpxhSp206)
6
Schematic representation of the isotope dilution
equation for iodine determination (mono-isotopic
element and radioactive spike)
R(129I/127I) NSpx100 / (NSxhS127
NSpxhSp127)
7
Transformation of the fundamental IDMS equation
into sample concentration values
Blanks are not considered in equation (3)
8
Many other forms of IDMS equations exist in the
literature Example Equations from ref. 3
However All equations are based on the
fundamental equation (1) for R
9
Summary of experimental conditions and
limitations for IDMS using ICP-MS
? Only the isotope ratio of the isotope diluted
sample must be measured and not an absolute
amount of the analyte !

? Isotope ratio measurement is independent on
Matrix effects
Signal drifts of the instrument
? Sample loss does not affect the result after
the isotope dilution step has taken place
(knowledge of recovery not necessary)
? One point internal calibration is
time-effective
? Total equilibration between spike and analyte
must be carried out for elemental trace analysis
(total decomposition of solid samples)
? Element must have at least two stable or
long-lived radioactive isotopes
? ICP-MS has a high risk of wrong results by
spectrometric interferences
10
Summary of experimental conditions and
limitations for IDMS using ICP-MS
? Only the isotope ratio of the isotope diluted
sample must be measured and not an absolute
amount of the analyte !

? Isotope ratio measurement is independent on
Matrix effects
Signal drifts of the instrument
? Sample loss does not affect the result after
the isotope dilution step has taken place
(knowledge of recovery not necessary)
? One point internal calibration is
time-effective
? Total equilibration between spike and analyte
must be carried out for elemental trace analysis
(total decomposition of solid samples)
? Element must have at least two stable or
long-lived radioactive isotopes
? ICP-MS has a high risk of wrong results by
spectrometric interferences
11
Comparison of Mo determinations by ICP-MS with
external calibration and ICP-IDMS in solutions
with increasing content of dissolved organic
carbon (DOC)
(L. Rottmann and K.G. Heumann)
10
ICP-MS with external calibration (non-matrix
matched)
8
6
4
ICP-IDMS
2
0
-2
0
20
40
60
80
100
120
DOC concentration in mg/L
12
Summary of experimental conditions and
limitations for IDMS using ICP-MS
? Only the isotope ratio of the isotope diluted
sample must be measured and not an absolute
amount of the analyte !

? Isotope ratio measurement is independent on
Matrix effects
Signal drifts of the instrument
? Sample loss does not affect the result after
the isotope dilution step has taken place
(knowledge of recovery not necessary)
? One point internal calibration is
time-effective
? Total equilibration between spike and analyte
must be carried out for elemental trace analysis
(total decomposition of solid samples)
? Element must have at least two stable or
long-lived radioactive isotopes
? ICP-MS has a high risk of wrong results by
spectrometric interferences
13
Sample preparation for the determination of
silicon traces in organic samples by ICP-IDMS
? Loss of analyte (SiF4) has no effect on the
result ? Closed digestion system avoids
contamination
14
Comparison of results for silicon trace
determinations in organic samples by ICP-IDMS
with an interlaboratory study
Sample ICP-IDMS Interlaboratory study
Serum 5.1 1.1 5.3 3.3
Pork liver 5.3 1.0 14.1 5.2
(contamination)
Spinach (powder) 333 11 176 189 (loss
of substance)
SD of the mean of means from 14 labs
15
Summary of experimental conditions and
limitations for IDMS using ICP-MS
? Only the isotope ratio of the isotope diluted
sample must be measured and not an absolute
amount of the analyte !

? Isotope ratio measurement is independent on
Matrix effects
Signal drifts of the instrument
? Sample loss does not affect the result after
the isotope dilution step has taken place
(knowledge of recovery not necessary)
? One point internal calibration is
time-effective
? Total equilibration between spike and analyte
must be carried out for elemental trace analysis
(total decomposition of solid samples)
? Element must have at least two stable or
long-lived radioactive isotopes
? ICP-MS has a high risk of wrong results by
spectrometric interferences
16
Comparison between standard addition method and
IDMS
On the contrary IDMS is an one-point
calibration method
17
Summary of experimental conditions and
limitations for IDMS using ICP-MS
? Only the isotope ratio of the isotope diluted
sample must be measured and not an absolute
amount of the analyte !

? Isotope ratio measurement is independent on
Matrix effects
Signal drifts of the instrument
? Sample loss does not affect the result after
the isotope dilution step has taken place
(knowledge of recovery not necessary)
? One point internal calibration is
time-effective
? Total equilibration between spike and analyte
must be carried out for elemental trace analysis
(total decomposition of solid samples)
? Element must have at least two stable or
long-lived radioactive isotopes
? ICP-MS has a high risk of wrong results by
spectrometric interferences
18
Elements accessible to ICP-IDMS technique
Second stable isotope for spiking available
Li
Long-lived radioactive isotope available
Be
19
Elements accessible to ICP-IDMS technique and
availability of isotope-enriched elements
? Most elements consist of more than 1 stable
isotope or long-lived radioactive isotopes are
available (e.g. 238U/235U, 129I) important
exceptions P, Co, As
? Isotope-enriched elements are commercially
available for nearly all poly-isotopic elements
? Certified spike solutions of 12 elements are
available from Merck, Darmstadt, since 1999
20
Optimization of spike amount and costs per
analysis
Amount of spike should be comparable with the
amount of analyte in the sample (optimization of
error multiplication factor fR) atomic ratio
(sample)/(spike) usually in the range 0.1 - 10
SDNS2 SDNSp2 fR2 x SDR2 SD Standard deviation
Optimum R-value Ropt (h2/h1)S x (h2/h1)Sp ½
Necessary spike amount per trace analysis is
usually lt 1 µg 1 µg of an isotope-enriched
compound costs 0.1 - 100 Cents
21
Major problems in receiving accurate results by
LA/ICP-MS
? Lack of reliable reference materials for
matrix-matched calibration
? Availability of suitable internal standard
elements
? Varying laser ablation rates
? Signal variations caused by instrumental drifts
? Possible fractionation processes during LA
22
Preparation of powdered samples for LA/ICP-IDMS
Isotope diluted sample
PVA substrate
Preconditions for LA/ICP-IDMS ? quantitative
and homogeneous adsorption of the spike at the
powdered sample ? ablation process must be
independent on the chemical form of sample and
spike isotopes
23
Correlation between trace element concentrations
determined by LA-ICP-IDMS and certified values of
seven reference materials
Cr
R2 0.993
IDMS data within not within the
uncertainty of certified values
28 results of 30 agree well within their standard
deviations with the certified values
24
Correlation between trace element determination
by LA-ICP-MS with non-matrix matched
calibration and certified values of two reference
materials
IDMS data within not within the
uncertainty of certified values
Most results do not fit the certified values !
25
Application of ICP-IDMS for elemental speciation
using hyphenated techniques
On-line coupling of HPLC and GC with ICP-MS is
simple, CE-coupling little more complicated but
possible
Complete separation of species necessary
? No isotope exchange between different elemental
species is allowed
Species-specific or species-unspecific spiking
mode can be applied
Lack of other accurate methods (validation
problem)
26
HPLC/ICP-IDMS system for species-specific and
species-unspecific spiking mode

27
The species-specific spiking mode

? Elemental species must be well defined by
composition and structure, e.g. CrO42-, IO3-,
MeHg
? Best spiking mode if isotope-labeled species is
available (spiking prior to separation)
? Isotope-labeled species must usually be
synthesized
? The isotopic composition is constant over the
whole chromatographic peak (real-time
determination)
? Isotope exchange between different species must
be avoided (in contrast to trace analysis)
28
How complicated is synthesis of isotope-labeled
species ?
? Relatively simple for most inorganic species
like iodate

HNO3/HClO3
Na129I
Na129IO3
? Not too complicated for organometallic species
like MeHg
201HgCl2 Me-Co Me201Hg
Me-Co Methylcobalamin
? Usually too complicated or impossible for
large biomolecules
29
Two different spike solutions are commercially
available at the moment
1. Isotope-labeled 202HgMe spike from
Institute for Reference Materials and
Measurements (IRMM), Geel/Belgium

• 119SnBu3 / 119SnBu22 / 119SnBu3 (mixed
  spike)
• from ISC Science, Gijón/Spain

30
Chromatogram of a multi-species determination by
species-specific GC/ICP-IDMS (mussel
tissue, CRM 477) (N. Poperechna and K.G. Heumann)
4000
20000
200Hg / 208Pb
Reference isotopes
120Sn
2000
10000
0
0
Intensity (cps)
4000
20000
202Hg / 206Pb
Spike isotopes
119Sn
2000
10000
0
0
Retention time (min)
31
Results of multi-species determination by
species-specific GC/ICP-IDMS in mussel
tissue (CRM 477) and tuna fish (CRM 463)
Excellent agreement between GC/ICP-IDMS and
certified/indicative values
32
HPLC/ICP-IDMS system for species-specific and
species-unspecific spiking mode

33
The species-unspecific spiking mode
? Necessary for all elemental species where
exact composition and structure is not known,
e.g. metal complexes of humic substances,
proteins etc...
? Addition of spike must be carried out after
complete separation of different species
(post-column spiking)
? Spike may exist in any chemical form and
isotope exchange between different
species is avoided
? Signal intensity of the measured element must
be independent on the species form Found for
normal nebulizer systems like cross-flow, but not
for an ultrasonic nebulizer with membrane
desolvator
? The isotope ratio varies over the whole
chromatographic peak However, real-time
concentrations are available
34
Conversion of an isotope ratio chromatogram for a
copper species into a mass flow chromatogram
35
Fractionation of heavy metal species with humic
substances by their molecular size using
SEC/ICP-IDMS (J. Vogl and K.G. Heumann)
Determination of heavy metal complexes in waste
water samples from two different sewage disposal
plants
? In both samples the distribution of individual
heavy metals shows a typical fingerprint for this
type of water samples
? Zn prefers a high molecular fraction, Cu
interacts with most HS fractions and Mo only with
a small fraction of medium molecular size
36
Determination of metallothionein isoforms by
sulfur analysis with CE/ICP-IDMS using a sector
field instrument at mass resolution 4000 (G.
Marx, K.G. Heumann, D. Schaumlöffel and A. Prange)
S/s)
37
For volatile elemental species GC/ICP-IDMS can
be used in the species-specific and
species-unspecific spiking mode
38
GC/ICP-IDMS for determination of MeHg
Ethylation of mercury species by NaBEt4
39
Characterization of a Me201Hg spike solution by
GC/ICP-MS
Spike isotope ratio 201Hg/202Hg 6.53 (natural
0.44)
A species-pure monomethylmercury spike was
obtained
40
MeHg analysis in seawater by GC/ICP-IDMS using a
Me201Hg spike
Transformation of MeHg into Hgo
Transformation can only be identified by
isotope-labeled species
41
Determination of MeHg in a river water sample
spiked with Me201Hg prior to alkylation by
NaBEt4 and NaBPr4
201Hg/202Hg 1.69 3.80.1 pg/mL
201Hg/202Hg 1.61 3.60.1 pg/mL
MeHg
Hgo
Hgo
Hg2
Species transformation during ethylation, not
with propylation
However, in both cases identical results are
obtained !
42
What can we learn from species-specific
GC/ICP-IDMS of methylmercury ?
? The use of isotope-labeled species identifies
species transformations
? Even if species transformation takes place,
accurate quantification is possible by
species-specific ICP-IDMS (if total mixture
between sample and spike species has taken place
prior to transformation)
? Species-specific ICP-IDMS can best be used
for validation of analytical methods for
elemental species !
43
Determination of MeHg in environmental and
biological samples by species-specific GC/ICP-IDMS
CRM 580 (sediment)
CRM 463 (tuna fish)
Sediment sample Me201Hg in diluted HNO3
Biolog. sample Me201Hg in TMAH
Microwave extraction (5 min)
Buffering at pH 4.8 and ethylation by NaBEt4
In-situ extraction of MeEtHg by nonane (5 min)
Injection of nonane extract into GC 201Hg/202Hg
isotope ratio measurement in separated mercury
species (6 min)
Certified value (70.3 3.4) ng g-1 GC/ICP-IDMS
(72.6 1.3) ng g-1
Certified value (2.83 0.15) µg
g-1 GC/ICP-IDMS (2.91 0.07) µg g-1
44
Multiple species-specific spiking to study
degradation and formation of species in the
environment or during sample preparation
Examples are known for Cr, Hg and Sn speciation
Example Butyltin compounds (ref.
5) Triple-spike technique was used with TBT
enriched in 117Sn, DBT in 118Sn and MBT in 119Sn
by measuring three isotope ratios for each of the
species 120/117Sn,
120/118Sn, 120/119Sn

• In principle 12 interconversion reactions are
  possible
• A complex mathematical system of equations allows
  (see ref. 5)
• Identification of relevant interconversions
• Determination of all Sn species
• Accuracy of quantitative results are doubtful

45
Isotope exchange between different species must
be avoided
Is this precondition always fulfilled ?
Example Determination of volatile halogenated
hydrocarbons by GC/ICP-IDMS
46
Iodine isotope chromatogram by GC/ICP-MS of
natural iodinated hydrocarbons spiked with
129I-labeled ethyl iodide
CH2I2
Intensity (cps)
Retention time (min)
Isotope exchange takes place with all iodinated
species
Species-specific ICP-IDMS not possible !
47
81Br and 79Br chromatograms of natural brominated
hydrocarbons spiked with 79Br-labeled ethyl
bromide
7000
7000
81Br chromatogram
79Br chromatogram
C2H5Br
6000
6000
CHBr
3
CHBr
5000
5000
3
CH
BrCl
2
CH
BrCl
4000
4000
2
C
H
Br
Br Intensity (cps)
Br Intensity (cps)
4
9
C
H
Br
4
9
3000
3000
C
H
Br
2
5
2000
2000
81
79
1000
1000
0
0
0
5
10
15
20
25
30
35
Retention time (min)
No isotope exchange occurs between C2H579Br spike
and other brominated hydrocarbons
Species-specific ICP-IDMS is possible !
Bond strength C - I lt C - Br lt C - Cl lt C
- F
48
Is species-unspecific spike determination of
elemental species necessary even if volatile
compounds are usually well defined ?
Advantage of species-unspecific spiking mode

? Will avoid synthesis of all different spike
compounds
? Quantification of unknown elemental species
possible
? Possible isotope exchanges between different
species are avoided (post-column spiking)
Disadvantage
? No loss of analyte allowed until post-column
spiking
49
Determination of MeHg and inorganic Hg2 in
biological materials by species-unspecific
ETV/ICP-IDMS (I. Gelaude, F. Vanhaecke and R.
Dams)
Fractionated evaporation of MeHg and inorg. Hg
by a two-step ETV temperature program
Isotope diluton of all Hg species by
isotope-enriched 200Hg
Permeation tube guarantees continuous spike flow
into the time-delayed analytes evaporated by ETV
50
Determination of MeHg and inorganic Hg2 in the
certified reference material TORT-2 (lobster) by
species-unspecific ETV/ICP-IDMS
(I. Gelaude, F. Vanhaecke and R. Dams)
51
Summary Elemental trace analysis by ICP-IDMS
? The relatively simple sample treatment,
one-point calibration, independence on matrix
effects and the multi-element capability fulfils
all conditions for a powerful and accurate
analysis
? It is typically a method where precision and
accuracy plays an important role at trace level
concentrations
? ICP-IDMS in connection with direct sampling of
powdered and liquid samples by laser ablation is
also possible
? Isotope-enriched standard solutions are now
available for various elements
? IDMS analyses are not possible for some
important mono-isotopic elements (e.g. P, Co, As)
? Spectrometric interferences are the most
critical problem
? If spectrometric interferences are under
control ICP-IDMS has a great potential to be used
as routine method
52
Summary Elemental species analysis by ICP-IDMS in
connection with hyphenated techniques
? Elemental speciation can be carried out by
species-specific and species-unspecific ICP-IDMS
coupled with HPLC, GC and CE (ETV)
? Online coupling of ICP-MS with GC, HPLC and
ETV is technically easy but little more
difficult for CE
? Isotope-labeled species must usually be
synthesized for species-specific IDMS but the
first commercial spike compounds are now available
? Species-specific ICP-IDMS is an ideal tool for
method validation
? Hyphenated techniques in combination with IDMS
are the only possibility to obtain real-time
concentrations of elemental species
? Because of a lack of alternatives GC/ICP-IDMS
and HPLC/ICP-IDMS are the most powerful methods
today for quantitative elemental speciation and
therefore also suitable for routine analysis
53
Literature

• K.G. Heumann (1988) Isotope Dilution Mass
  Spectrometry, in Inorganic Mass Spectrometry
  (Eds. F. Adams, R. Gijbels, R. van Grieken),
  Wiley, New York, pp. 301-376.
• D.H. Smith (2000) Isotope Dilution Mass
  Spectrometry, in Inorganic Mass Spectrometry
  Fundamentals and Applications (Eds. C.M.
  Barshick, D.C. Duckworth, D.H. Smith), Dekker,
  New York, pp. 223-240.
• M. Sargent, R. Harte, C. Harrington (2002)
  Guidelines for Achieving High Accuracy in Isotope
  Dilution Mass Spectrometry (IDMS), Royal Society
  of Chemistry, Cambridge.
• K.G. Heumann (2003) Calibration in Elemental
  Speciation, in Handbook of Elemental Speciation
  Techniques and Methodology, Wiley, Chichester,
  pp. 547-562.
• P. Rodriguez-González, J.M. Marchante-Gayón, J.I.
  Garcia Alonso, A. Sanz-Medel (2005) Isotope
  Dilution Analysis for Elemental Speciation A
  Tutorial Review, Spectrochim. Acta B 60, 151-207.

54
Many thanks for your attention !