Anal Bioanal Chem (2002) 372 : 795–800
DOI 10.1007/s00216-002-1263-5
O R I G I N A L PA P E R
Ivan Švancara · Božidar Ogorevc · Milko Nović ·
Karel Vytřas
Simple and rapid determination of iodide in table salt
by stripping potentiometry at a carbon-paste electrode
Received: 16 July 2001 / Revised: 29 January 2002 / Accepted: 13 February 2002 / Published online: 4 April 2002
© Springer-Verlag 2002
Abstract A simple and rapid procedure, utilising con-
stant-current stripping analysis (CCSA) at a carbon-paste
electrode containing tricresyl phosphate as a pasting liq-
uid (TCP-CPE), has been developed for the determination
of iodide in table salt. Because of a synergistic accumula-
tion mechanism based on ion-pairing and extraction of io-
dide in combination with electrolytic pretreatment of the
TCP-CPE, the method is selective for iodide and enables
direct determination of iodide in samples of table salt con-
taining anticaking agents such as K4[Fe(CN)6] (food addi-
tive “E 536”) or MgO. The iodide content (calculated
as KI) can be determined in a concentration range of 2 to
100 mg kg–1 salt, with a detection limit (S/N=3) of 1 mg kg–1,
and a recovery from 90 to 115%. The proposed method
has been used to determine iodide in several types of arti-
ficially iodised table salt and in one sample of natural sea
salt. The results obtained agreed well with those obtained
by use of three independent reference methods (titration,
spectrophotometry, and ICP–MS) used to validate the
CCSA method, indicating that the developed method is
applicable as a routine procedure for rapid testing in salt
production process control and in the analysis of marketed
table salts.
Keywords Constant-current stripping analysis ·
Carbon-paste electrode · Iodide · Table salts
Introduction
Table (food grade), feeding and industrial salts are often
artificially enriched with minor constituents such as se-
B. Ogorevc · M. Nović
Analytical Chemistry Laboratory, National Institute of Chemistry,
P.O. Box 3430, 1001 Ljubljana, Slovenia
I. Švancara (✉) · K. Vytřas
Department of Analytical Chemistry,
Faculty of Chemical Technology, University of Pardubice,
nám. Čs. legií 565, 532 10 Pardubice, Czech Republic
e-mail: Ivan.Svancara@upce.cz
lected trace elements and special additives [1]. A typical
example is iodine, because this bioessential element often
occurs in natural salt (rock and sea salts, brine from salt
springs) at a very low concentrations. In commercial salt
used for human and animal nutrition, iodine is usually
present as added iodide or iodate. For example, while the
recommended minimum iodide content of iodised salts is
5 mg KI kg–1 salt [1], the regulation for commercial prod-
ucts in many countries requires an average KI content of
25 mg kg–1 (min. 20 and max. 30 mg kg–1). Potassium
hexacyanoferrate(II), labeled as food additive “E 536” is
a special additive allowed to be added to table salt (max.
10 mg kg–1) as a drying agent to prevent undesirable cak-
ing and aggregation of the salt during its storage. Similar
effects can be achieved by adding other compounds such
as MgO and MgCO3 [1, 2]. The additive content of salt,
including iodide, is strictly regulated and thus must be
controlled at both production and public health institution
sites. This is particularly important, because of the poten-
tial harm to health and the environment caused by high
concentrations of these minor constituents. Reliable, sim-
ple, and rapid methods are therefore needed for routine
monitoring of regulated additives, including iodide, in salt.
Several methods are used for determination of (total)
iodine and iodide or iodate in table salt, e.g. classical titra-
tion (still the official method) [3, 4], traditional [5] and
modern spectrophotometry [6], flow injection [7], ion
chromatography [8], and ICP–MS [9]. These methods do,
however, have disadvantages – either they are not suffi-
ciently sensitive and/or are time-consuming (e.g. classical
titrimetric and spectrophotometric methods) or they re-
quire specialised and expensive equipment (ICP–MS, ion
chromatography).
Electrochemical stripping methods have been shown to
be a promising alternative for the measurement of iodide
[10, 11, 12, 13, 14]. These methods are based on:
1. accumulation of iodide (or chemically pre-reduced io-
date) at a working electrode modified with suitable
ion-pairing agents, e.g. protonated organic esters [15]
or cationic surfactants of the tetraalkylammonium salt
type [13], as a first step; then
796
2. electrolytic oxidation of the iodide moiety in the ion-
associate to elemental iodine which is then reduced
back to iodide during the stripping step, giving the an-
alytical signal [12].
The stripping step can be performed either by employing
a cathodic voltammetric scan [10, 11, 12, 13, 14] or by us-
ing a stripping potentiometric detection technique, e.g.
constant-current stripping analysis (CCSA), as shown in
our recent preliminary study [16]. CCSA is a relatively
new technique particularly promising for practical use in
analysis of samples with complex matrices [17].
This article describes a method for direct determina-
tion of iodide in table salt, in the presence of anticaking
agents; it uses CCSA in combination with a special car-
bon-paste electrode based on liquid tricresyl phosphate as
binder [15]. In acidic media lipophilic molecules of tricre-
syl phosphate are readily protonated, whereupon they act
as an effective “ion-pairing trap” for iodide anions. The
practical exploitation of this principle then provides a
simple, flexible, and reliable procedure for the determina-
tion of iodide in table salts, as is demonstrated in the fol-
lowing sections.
Experimental
Apparatus
All electrochemical measurements were performed with an Auto-
lab (Eco Chemie, Utrecht, The Netherlands) modular electrochem-
ical system equipped with PGSTAT10 and ECD modules and con-
trolled by GPES software (Eco Chemie). A conventional voltam-
metric cell was used in combination with a three-electrode config-
uration where a tricresyl phosphate-based carbon-paste electrode
(TCP-CPE), a platinum wire, and an Ag/AgCl (KCl satd.) elec-
trode served as the working, auxiliary and reference electrodes, re-
spectively. During the preconcentration step solutions were stirred
at approximately 100 rpm. All experiments were performed at
room temperature (23±1 °C).
Chemicals and solutions
All chemicals were of analytical reagent grade, except for sodium
chloride, which was of Suprapur quality (Merck). Stock solutions
of NaCl (2 mol L–1, supporting electrolyte) and 0.01 mol L–1 stan-
dard solutions of KI, K4[Fe(CN)6] and K3[Fe(CN)6] were diluted
as needed, daily. Testing reagent containing magnesium oxide was
an approximately 10% (m/v) suspension of MgO in water. Before
potentiometric stripping measurements all solutions were purged
with pure nitrogen. De-ionised and additionally purified (Milli-Q,
Millipore) water was used throughout this work.
Working electrode (TCP-CPE)
Carbon paste was made by thorough hand-mixing of 0.5 g RW-B
spectroscopic graphite powder (Ringsdorff Werke, Germany) with
0.2 mL tricresyl phosphate (mixture of isomers, Fluka). The ho-
mogenised carbon paste was packed into a laboratory-made elec-
trode body driven with a piston for extrusion of the paste through
a hole in the end (diameter 2 mm).
Samples
Model salt solutions
To a 3% NaCl stock solution acidified with 65% HNO3 to pH 1,
appropriate amounts of KI and K4[Fe(CN)6] were added as re-
quired.
Table salt samples
Four commercially marketed table salts were purchased in a local
food store in Ljubljana (group A samples). Other samples of table
salt (group B sample) were obtained from a company producing
table salt; they were collected from the production line at different
stages of addition of KI and ferrocyanide and of salt homogenisa-
tion by the mixing of one type of salt mixture. From each package
or collection of a salt, a portion of approximately 50 g salt was sep-
arated, divided into four parts and used as the analytical samples.
Each sample (3 g) was dissolved in 50 mL water, acidified with
1 mL 65% HNO3, and diluted with water to 100 mL.
Procedure for constant-current stripping analysis
After each stripping run the surface of the electrode was renewed
by extruding ca. 0.5 mm carbon paste and smoothing the surface
with a wet filter paper. The TCP-CPE was stored in a beaker with
the tip immersed in de-ionised water.
The TCP-CPE was first conditioned at –1.5 V relative to the
Ag/AgCl electrode for 15 s. Deposition was performed at +0.8 V
for 30 s unless specified otherwise, and, after an equilibration pe-
riod of 10 s, the stripping step was run from +0.8 to –0.4 V by ap-
plying a stripping current of –5 µA. Peak areas from recordings of
dt/dE against E served as the analytical signals. The results of io-
dide determinations were calculated either as the recovery (in %)
or by means of the Dean–Dixon statistical method for small sets of
data [18].
Reference determinations
Three reference methods for the determination of iodide (total io-
dine) were used for validation of the proposed CCSA method.
Titration
This official standard analytical method [3, 4] is based on chemi-
cal oxidation of iodide in the sample by use of bromine water and
subsequent titration of the released iodine with thiosulfate stan-
dardised against dichromate.
ICP–MS
The aqueous stock solutions of samples of different table salt were
prepared by dissolving approximately 100 mg salt in 100 mL de-
ionised water. Quantification was performed by use of a standard
addition method employing a standard solution containing 1.01 g L–1
iodide. The measurements were performed with a Hewlett–Packard
(Waldbronn, Germany) HP4500 element-selective inductively-
coupled plasma mass spectrometer (ICP–MS) equipped with a
Babington nebuliser and operated under normal plasma conditions
(RF power 1350 W, reflected power <2 W; gas flows (L min–1):
plasma 14.8, auxiliary 0.93, carrier 1.30; cooling temperature
2 °C).
Spectrophotometry
Measurements with a standard laboratory spectrophotometer were
performed by a method based on the reaction of As(III) with
Ce(IV) [4].

Results and discussion
Development and characterisation
of the CCSA method using a TCP-CPE
Initial measurements confirmed that most experimental con-
ditions originally optimised for stripping voltammetry [12]
can be adopted in CCSA mode [17]. In this work, therefore,
experimental conditions such as carbon-paste composition,
supporting-electrolyte constituents, deposition potential, and
both stripping limits were selected accordingly.
By following these preliminary results, the stripping
current for reduction of iodine was optimised using a
model solution containing 0.5 mol L–1 NaCl+0.1 mol L–1
HNO3+4×10–6 mol L–1 I–. The choice of the iodide con-
centration corresponded to 22 mg KI kg–1 salt, i.e. to a
value being approximately in the middle of the range
10–30 mg KI. The stripping current was found to be nearly
constant within a wide range from –10 to –0.01 µA; a
value of –5 µA was selected as optimum because of the
most favourable signal-to-noise characteristics. Although
dissolved oxygen does not usually interfere in classical
stripping potentiometric measurements [17], in iodide de-
termination using the TCP-CPE oxygen was found to be a
competitive agent reducible by the applied cathodic strip-
ping current [19]. Removal of oxygen by purging the so-
lutions with nitrogen for at least 5 min was, therefore,
found to be necessary.
Because tricresyl phosphate is a more volatile binder
compared with common pasting liquids, it is recom-
mended that the TCP-CPE tip be kept immersed in water
when the electrode is not in use [15, 20]. It was, however,
found in this work that when the TCP-CPE was stored for
longer than 6 h the surface layer of the paste gradually be-
came saturated with water, which resulted in a very high
background. In these circumstances mechanical removal
of approximately 5 mm carbon paste was necessary to ob-
tain a new layer with optimum behaviour. This could be
achieved simply by use of a piston-driven type of elec-
trode [20]; this was used in this work. To avoid possible
memory effects during experimentation some paste was
always removed after each measurement.
Pretreatment of the carbon-paste surface by exposure
to extremely high potentials is occasionally used to im-
prove the performance of CPEs [20, 21]. This prompted
us to examine the effect of electrolytic activation of the
TCP-CPE when used for iodide measurement. As is evi-
dent from Fig. 1, application of high positive potential be-
fore measurement in a model solution containing 4 µmol L–1
iodide resulted in a distorted stripping record (curves
a and b). In contrast, cathodisation of the TCP-CPE re-
sulted in an increased peak size, with the optimum effect
being observed on application of –1.5 V (curve e), in ac-
cordance with previous experience [20]. Although the ef-
fect of carbon-paste pre-anodisation has already been sat-
isfactory explained [21, 22], the effect of application of
high negative potentials is not yet clear. A possible reason
for the enhanced response to iodide is that pre-cathodisa-
797
Fig. 1 The effect of pretreatment (electrolytic activation) of the
TCP-CPE on CCSA performance in the determination of iodide.
Activation time, 15 s; activation potential: (a) +1.7 V relative to
Ag/AgCl, (b) +1.0 V, (c) unactivated, (d) –1.0 V, (e) –1.5 V. The
testing medium was 0.5 mol L–1 NaCl+0.1 mol L–1 HNO3+4×
10–6 mol L–1 I–, the deposition potential +0.8 V relative to
Ag/AgCl, the deposition time 30 s, the equilibration time 10 s, the
stripping limits +0.8 to –0.4 V, and the stripping current –5 µA
tion of the TCP-CPE surface might produce negatively
charged sites which selectively repel the chloride ions
present in large excess in the highly saline electrolytes
used, and might thus favour ion-pair formation between
iodide and tricresyl phosphate. Such an assumption is in
agreement with an observation that the lower the concen-
tration of iodide measured, the more pronounced the en-
hancement induced by pre-cathodisation of the TCP-CPE
(not shown). Activation at –1.5 V also improved the re-
producibility of CCSA measurements.
Study and optimisation of analytical conditions
The analytical conditions used for CCSA measurement of
iodide by use of the TCP-CPE were studied and optimised
by use of a model solution containing 4×10–6 mol L–1
I–+0.5 mol L–1 NaCl+0.1 mol L–1 HNO3 and simulating a
salt sample (with an iodide content of ca 22 mg KI kg–1
salt). Because of penetration of iodide into the carbon
paste, by extraction, and because of the subsequent re-dif-
fusion of the extracted species from the carbon paste [20,
23], the reductive stripping of the released iodine has re-
sulted in broad peaks that had to be evaluated as peak area
[12, 16].
Peak area was found linearly dependent on iodide con-
centration in the range 4.0×10–7 to 2.0×10–5 mol L–1 (i.e.
from 2 to 100 mg KI kg–1 salt when re-calculated to the
amount in a sample of salt; see Experimental). The limit of
detection estimated by use of the S/N=3 criterion was ap-
proximately 1.5×10–7 mol L–1 I– (for a deposition time of
5 min). The reproducibility of CCSA measurements at a
renewed and activated TCP-CPE and for a concentration
798

iodide and hexacyanoferrate(II) corresponding to 22.1 mg

KI and 21.1 mg K4Fe(CN)6.3H2O per kg–1 dry NaCl. The
results (mean 23.4±2.6 mg KI kg–1 salt, recovery
90–115%, n=8) were satisfactory, considering the direct,
simple, and rapid (ca 10 min per analysis) determination
of iodide in the presence of the specified anticaking agent.
Real table salt samples analysed in this work were
sorted into two groups, A and B. The results from analysis
of these salt samples for iodide using CCSA with a
TCP-CPE, together with the results from reference deter-
minations, are summarised in Tables 1 and 2. Group A
comprises four commercial (domestic and imported) table
salts purchased in a local food-store, of which samples A1
(artificially iodised) and A3 (containing natural iodine)
were sea salt whereas A2 and A4 were artificially iodised
rock table salt with a label-declared content of E 536

Fig. 2 Comparison of the behaviour of [Fe(CN)6]4– and Table 1 Results from analysis of some commercial table salts
[Fe(CN)6]3– with that of iodide at the TCP-CPE. A. (a) blank me- (group A samples) for iodide (CCSA) and total iodine (titrimetry
dium, (b) a+1.5×10–6 mol L–1 [Fe(CN)6]4–, (c) a+1.5×10–5 mol L–1 and ICP–MS); CCSA: constant-current stripping analysis with
[Fe(CN)6]4–; B. (a) blank, (b) a+1.5×10–6 mol L–1 [Fe(CN)6]3–, (c) TCP-CPE (the proposed method); Titration: standard method [3,
a+1.5×10–5 mol L–1 [Fe(CN)6]3–, (d) a+4×10–6 mol L–1 I-; other 4]; ICP–MS: for details, see Experimental. All results and label-
conditions as for Fig. 1 declared data are given in mg KI kg–1 salt; results are given as
means from three replicate analyses. Sample specifications are
given in the text
of 4×10–6 mol L–1 I– was found to be satisfactory; the rela-
Sample Declared content Found
tive standard deviation (r.s.d.) was approximately 5%
(n=7). KI K4[Fe(CN)6] CCSA Titration ICP–MS
Hexacyanoferrate(II) as a possible component of table
or feeding salt might be a potentially concurrent ion-pair- A1 20–30 – 25.7±3.1 20.6a 24.1b
A2 “Added”c “Added” 12.6±3.3 22.1 25.4
ing species in the CCSA determination of iodide. Because
A3 – – 2.6±1.9 0.8 1.2
hexacyanoferrate(II) is readily converted into hexacyano-
A4a 22 “Added” 34.2±5.1 – –
ferrate(III), and vice versa [24], both anions were consid-
A4b 22 “Added” 18.0±2.1 – –
ered as potential interferents in this work. Figure 2 dis-
A4c 22 “Added” 8.1±2.3 – –
plays the CCSA behaviour at the TCP-CPE in the pres- A4d 22 “Added” 25.1±3.9 18.1d 25.6
ence of different concentrations of either [FeII(CN)6]4– or
[FeIII(CN)6]3– ions (A, B; curves b, c) compared with that ar.s.d. ca 5% at 20 mg kg–1, 25% at 1 mg kg–1 (the limit of the
in the presence of iodide ions (curve d). The concentra- method)
br.s.d. ca 5% over the whole concentration range
tions of both species in model solutions were chosen to cotherwise unspecified
correspond to the maximum admissible potassium hexa- dan unclear solution was obtained after dissolution of the table salt

cyanoferrate(II) content of table salt (20 mg kg–1) [1] in water

(curve b) and to a 10-fold excess relative to a typical con-
centration of iodide in salt (curve c). As is evident from Table 2 Results from analysis of some salt mixtures collected
the figure, neither hexacyanoferrate at either concentra- from a table salt production line (group B samples) for iodide
tion levels affected the baseline region where the peak of (CCSA) and total iodine (spectrophotometry); CCSA: the pro-
interest would appear. Finally, Fig. 2 also illustrates typi- posed method using a TCP-CPE; SP: standard spectrophotometric
method [4]. All results and producer-declared data are given in mg
cally distorted potentiograms, including the baseline it- KI or K4[Fe(CN)6] kg–1 salt; results are given as means from three
self, registered at a higher sensitivity and obtained in the replicate analyses. Sample specifications are given in the text
highly saline media used in this work (curves a–c).
The other anticaking agent tested, magnesium oxide, Sample Declared content Found
was also found to be inactive; even though MgO was in- KI K4[Fe(CN)6] CCSA SP
troduced to the sample in the form of a rather concen-
trated suspension (10% w/w), no changes in the iodide B1 20–30 1–10 20.2±2.7 25.6
peak were observed (not shown). B2 10–25 0–5 14.6±1.5 15.9
B3 None 10–20 – –
B4a – 10 20.2±2.7 25.2
Sample analysis B4b – 10 20.3±0.9 25.4
B4c – 10 23.9±4.0 23.3
The proposed method was first tested by analysis of
B4d – 10 23.9±4.0 22.5
model solutions, simulating table salt samples, containing
 799

(potassium hexacyanoferrate(II), Table 1). Sample A4 was

appropriately divided into three sub-samples to control its
homogeneity. These fractions (A4a–A4c) were analysed
by CCSA only whereas their mixture (A4d) was analysed
by CSA and the two reference methods. The analyses of
group A samples were performed to demonstrate the use-
fulness of employing the proposed CCSA method as a
fast, reliable, and simple test method applicable, for exam-
ple, in public health control to check the regulated quality
of marketed table salt. Group B (Table 2) contained four
artificially iodised and homogenised salt mixtures (all with
added anticaking agent E 536), sampled from a production
line at different phases before and after addition of iodide
and E 536 by use of a spray technology. In this group, one Fig. 3 Typical stripping potentiograms obtained by analysis of ar-
sample (B4) was again divided into four fractions tificially iodised table salt by use of CCSA with the TCP-CPE. (a)
(B4a–B4d) to test the homogeneity of iodide distribution. sample A1, (b) and (c) standard additions of iodide (two aliquots).
Analyses of group B samples were performed in an effort Experimental conditions as for Fig. 1
to demonstrate the possibilities of the method in the direct
control of salt homogenisation achieved by its mixing.
As is apparent from Tables 1 and 2, the results obtained
by CCSA with the TCP-CPE agree well with those ob-
tained by use of the reference methods. The results also
correspond to the KI content declared by the manufactur-
ers of the table salt analysed. This fully confirms the suit-
ability of the proposed method for determination of iodide
in food-grade salt in terms of reliability, accuracy, and sat-
isfactory precision. Among the results summarised in
Table 1, only the value for sample A2 obtained with
CCSA differs significantly from those obtained with both
reference methods. This might indicate either that this salt
was not sufficiently homogeneous or, more probably, that
the salt was iodised with potassium iodate. In such cir-
cumstances the simple version of the CCSA method gave
different results because of its inability to detect pentava- Fig. 4 Detection of iodide at the low microgram level in natural
sea salt by use of CCSA with the TCP-CPE electrode. (a) blank,
lent iodine, in contrast with both titration and ICP–MS, (b) sample A3; deposition time: 5 min; dashed lines indicate the
which furnish results for total iodine, i.e. even for iodate. manner of evaluation of peak area; other conditions as for Fig. 1
By coupling the CCSA method – used herein for specific
detection of iodide – with a simple pre-reduction step it
would, however, be possible to adapt the whole potentio- these sub-samples were confirmed by reference analyses
metric stripping procedure for the determination of both of a mixture of all three fractions (A4d, Table 1).
I– and IO3–. Reduction with hydrazinium salt [12] can be The iodide content was quantified by the method of
performed directly in the sample solution and quantitative standard addition with at least two aliquots. This is illus-
transformation of iodate to iodide is complete within trated in Fig. 3, which shows typical potentiograms ob-
5 min. The procedure with the TCP-CPE modified in this tained by analysis of a real table salt. It can be seen that
way has already been successfully applied in both ca- reduction peaks of iodine at the TCP-CPE were well-de-
thodic stripping voltammetric [12, 25] and CCSA [26] fined even in the presence of real salt matrix. Here also
modes to analyse table salts containing iodate. Analysis of the whole experimental procedure using CCSA was per-
sample A3 suggests that the method employing CCSA formed typically within 10 min. (For comparison, the
and the TCP-CPE has some promise even for salt samples standard titration and spectrophotometric procedures took
with a natural iodide content. approximately one hour per sample whereas ICP–MS
Analysis of the fractionated salt samples furnished an measurement required approximately 5 min). With regard
interesting insight into the homogeneity of the salts to other typical characteristics of the technique used,
analysed in this work. Whereas test sample B4 of the salt CCSA could be applied with a short deposition time of
mixture collected from the production line was found to 30 s, except for sample A3 (natural sea salt) when the pre-
be appropriately homogeneous (Table 2, B4a–B4d), analy- concentration had to be prolonged to 5 min to obtain a
sis of a commercial iodised rock salt revealed the iodide sufficient response (Fig. 4).
content to be highly diverse over the individual fractions, It is worth mentioning that the reliability of the CCSA
and so the homogeneity of this table salt was poor method was confirmed by measurements performed over
(A4a–A4c, Table 1). The results obtained by CCSA of a period of several months (Table 1 compared with Table 2),
800
i.e. each time employing a new TCP-CPE, because the
lifetime of this type of electrode is 3 weeks, maximum
[15].
Conclusions
A novel rapid and simple method based on CCSA at the
TCP-CPE is presented for the determination of iodide in
table salt in the presence of some anticaking agents. Be-
cause of the excellent selectivity of the TCP-CPE for io-
dide, the procedure proposed enables direct analysis of
table and related salts without sample pretreatment. The
method utilises a unique accumulation mechanism at the
TCP-CPE based on both ion-pairing and extraction, in ef-
fective combination with computer-controlled constant-
current stripping analysis [16, 27]. In addition, the CCSA
procedure combined with pre-reduction is fully applicable
for analysis of table salts containing iodate [26].
The iodide determination is rapid, and easy to perform,
because quite simple measurement operations and elec-
trode pretreatment and regeneration steps are involved.
The method combining CCSA with the TCP-CPE seems
to be an interesting alternative to the classical (official)
titration and spectrophotometric procedures. Compared
with these the CCSA method is simpler, analyses can be
performed much more rapidly, and the whole procedure
seems to be attractive economically because of the spar-
ing use of chemicals and reagents. Compared with ICP–
MS this method requires significantly less expensive
equipment and levels of accuracy and reliability are com-
parable.
Analyses of real salt samples validated by three inde-
pendent reference methods have demonstrated that the
method based on CCSA and the TCP-CPE is particularly
suitable for rapid testing (screening) of the iodide level
and its uniform distribution in iodised marketed table salt,
in public health control service, in salt mixtures (after
component additions and homogenisation), or in the con-
trol of salt-production processes.
Acknowledgements This work was supported by the Ministry of
Education, Science, and Sport of the Republic of Slovenia
(P1–0509–0104, 3311–01–838008) and by the Ministry of Educa-
tion, Youths, and Sports of the Czech Republic (Kontakt ME 241).
The authors wish to thank Dr I. Grabec Svegl and Dr B. Divjak
(NIC) for conducting standard titrimetric and ICP–MS analysis.
References
1. Belitz HD, Grosch W (1994) Food Chemistry. Springer, Berlin,
pp 718–719
2. EC Directive No. 95/2/EC on food additives other than colours
and sweeteners (OJ L 061, 18.03.95, p 1; amended, OJ L 086,
28.03.97, p 4; OJ L 295, 04.11.98, p 18)
3. Cunniff P (1995) (ed) Official Methods of Analysis, 16th edn.
AOAC International, AOAC Official Method 925.56, Arling-
ton (USA), Vol. I, p 31
4. Ettel W, et al. (1997) (eds) Schweiz Lebensmittelbuch Kapitel
37, Section B 5–5.1, 5.2, 5.3
5. Sandell EB, Kolthoff IM (1937) Mikrochim Acta I:9
6. Fischer PW, L’Abbe MR, Giroux A (1986) J Assoc Off Anal
Chem 69:687
7. Sun JY, Chen XG, Hu ZD (1997) Fresenius J Anal Chem
357:1002
8. Hu W, Haddad PR, Hasebe K, Tanaka K, Tong P, Khoo C
(1999) Anal Chem 71:1617
9. Larsen EH, Ludwigsen MB (1997) J Anal At Spectrom 12:435
10. Lu GH, Yang M, Zheng QF, Wang AL, Jin ZX (1995) Electro-
analysis 7:591
11. Kwak MK, Park DS, Won MS, Shim YB (1996) Electroanaly-
sis 8:680
12. Svancara I, Konvalina J, Schachl K, Kalcher K, Vytras K (1998)
Electroanalysis 10:435
13. Svancara I, Cermakova I, Goessler W, Kalcher K, Vytras K (1999)
Sci Pap Univ Pardubice, Ser A 5:95
14. Walcarius A, Lefevre G, Rapin JP, Renaudin G, Francois M
(2001) Electroanalysis 13:313
15. Svancara I, Vytras K (1993) Anal Chim Acta 273:195
16. Vytras K, Konvalina J (1998) Electroanalysis 10:787
17. Ostapczuk P (1993) Anal Chim Acta 273:35
18. Dean RB, Dixon WJ (1951) Anal Chem 23:636
19. Jagner D (1982) Analyst 107:593
20. Svancara I, Schachl K (1999) Chem Listy 93:490
21. Rice M, Galus Z, Adams RN (1983) J Electroanal Chem 143:
89
22. Ravinchandran K, Baldwin RP (1984) Anal Chem 56:1744
23. Wang J, Desmukh BK, Bonakdar M (1985) J Electroanal Chem
194:339
24. Kalcher K (1986) Analyst 111:625
25. Konvalina J, Svancara I, Vytras K, Kalcher K (1997) Sci Pap
Univ Pardubice, Ser A, 3:153
26. Konvalina J (2001) PhD Thesis, University of Pardubice, Par-
dubice
27. Konvalina J, Vytras K (2001) Chem Listy 95: 344