Get documents about "Operational speciation of thallium in environmental solid samples by
Document Sample
Operational speciation of thallium in environmental solid samples by
electrothermal atomic absorption spectrometry according to the BCR
sequential extraction scheme
´
Marıa Villar, Fausto Alava, Isela Lavilla and Carlos Bendicho*
´ ´ ´ ´
Departamento de Quımica Analıtica y Alimentaria, Area de Quımica Analıtica,
´
Facultad de Ciencias (Quımica), Universidad de Vigo, As Lagoas-Marcosende s/n,
36200 Vigo, Spain. E-mail: bendicho@uvigo.es
Received 20th July 2001, Accepted 18th October 2001
First published as an Advance Article on the web 19th November 2001
Electrothermal atomic absorption spectrometry was applied to the determination of extractable Tl in soil,
sediment and fly ash certified reference materials (certified only in total Tl content) following the application of
the BCR sequential extraction scheme. Strong matrix interference due to the presence of chloride arising from
the hydroxylamine hydrochloride reagent used for leaching the reducible phase was observed. Thermal
programmes were optimised for atomisation of Tl from aqueous standards, matrix-matched standards with
extractants and real extracts obtained after treatment using the above method. In order to efficiently stabilise Tl
for its determination in the acid-soluble and oxidisable fractions, 8 mg Pd were needed as a matrix modifier.
The depressive interference caused by chloride required the use of 8 mg Pd z 8 mg ascorbic acid for
determination of Tl in the reducible and residual phases. Application of stabilised temperature platform
conditions in conjunction with a transverse-heated atomiser and longitudinal Zeeman background correction
allowed calibration to be performed with aqueous standards in the acid soluble and oxidisable phases, whilst
the standard additions technique was mandatory in the reducible and residual phases in order to achieve
accurate results. Low extractability of Tl was observed in most of the analysed samples, thus indicating a low
mobility for this metal. Between-batch precision values were lower than 5%. The LOD of Tl when determined
in BCR extracts was 0.05 mg kg21.
Introduction proposed by the BCR (Community Bureau of Reference),
now the Standards, Measurement and Testing programme, has
Sensitive and accurate methods for determination of Tl are become very popular during recent years, perhaps as a result of
increasingly in demand as a consequence of the intense toxic both the diminished number of stages and also matrix effects as
properties displayed by this metal, comparable to those of Pb compared with earlier SES.18 Additionally, this scheme has
and Hg. Tl is a rare element, which occurs in the earth’s crust in been collaboratively tested and, hence, its validation is more
an estimated abundance of 0.1 to 0.5 mg kg21 and is mainly advanced in comparison with other schemes.
associated with K and Rb. The main sources of pollution So far, the BCR SES has been applied to partitioning of Cu,
nowadays come from anthropogenic emissions from refineries, Cr, Ni, Pb, Cd and Zn in soils,19 sediments,20–23 sewage
coal-fired power stations, metal smelters and the cement sludge24 and industrially contaminated ground,25 but no
industry.1 methods have been reported for thallium.
Several methods have been proposed for determination of Tl The strong interferences with Tl caused by the matrix,
in environmental and biological samples, such as anodic occurring with the use of widespread analytical techniques
stripping voltammetry,2–5 X-ray fluorescence spectrometry,6 (e.g., ETAAS), and the low concentrations of this metal in non-
spectrophotometry,7 neutron activation analysis,8 fluorimetry,9 polluted sites, may account for the small number of applica-
laser-induced fluorescence,10 atomic emission spectrometry,11 tions published in the environmental field.
electrothermal and flame atomic absorption spectrometry,12–15 To the best of our knowledge only one paper, by Sager,26 has
and inductively coupled plasma mass spectrometry.16 dealt with the sequential leaching of thallium in sediments.
Determination of potentially toxic metals in environmental Difficulties arising from the application of SES with 5–6 stages,
solid samples (e.g., sediment, soil) is an important issue in the and non-optimised ETAAS conditions (e.g., use of uncoated
monitoring of environmental pollution. Although the total graphite tubes and wall atomisation) are evidenced by the need
metal concentration in such samples may provide relevant for separation methods prior to thallium determination and
information about the degree of contamination, it has been calibration with the standard additions technique.
recognised that knowledge of the metal fraction associated with The chloride interference on the thallium signal using
significant phases is an essential requirement for assessing the ETAAS is well documented.27,28 This interference can be
mobility and bioavailability of the metal, which is directly brought under control by using a reduced Pd modifier and the
related to toxicological effects. application of the STPF concept, preferably in a transverse-
In order to discover the distribution patterns of trace metals heated atomiser with integrated platform and Zeeman-effect
among specific solid phases, sequential extraction schemes background correction.29 In this work, the operational
(SES) are commonly used.17 In these schemes, each significant speciation of thallium using sequential extraction is reported.
solid phase is selectively leached by the appropriate chemical Main efforts have been made towards optimisation of the
extractant (i.e., operationally-defined speciation). Among the furnace conditions in the presence of reagents associated with
variety of SES available for metal partitioning, the one the BCR SES.
1424 J. Anal. At. Spectrom., 2001, 16, 1424–1428 DOI: 10.1039/b106546g
This journal is # The Royal Society of Chemistry 2001
Table 1 Thermal program for Tl 50 mL polyethylene centrifuge tubes. Following each treat-
ment, the supernatant liquid was removed after centrifugation
Temperature/ Hold Gas flow rate/
at 2500 rev min21 for 5 min and made up to volume. The solid
Stage uC time/s Ramp/s mL min21
residue was washed with 10 mL of ultrapure water, the
Drying 1 110 30 1 250 washings being discarded, and the extracts stored in stoppered
Drying 2 130 30 15 250 polyethylene vessels. Under optimised furnace conditions, a
Pyrolysisa 1000 20 10 250 10 mL volume of extract was injected into the furnace along
Atomisationb 1600 5 0 0 with 8 mg of Pd (acid soluble and oxidisable fractions) or 8 mg
Cleaning 2450 3 1 250
a
Pd z 8 mg ascorbic acid (reducible fraction).
Pyrolysis stage was eliminated for the determination of Tl in the Method development was carried out using matrix-matched
residual fraction. bRead was set up in the atomisation stage.
standards with extractant solutions or extracts spiked with
known amounts of thallium so that matrix effects could be
evaluated. The following certified reference materials (CRMs)
Experimental were employed for testing the BCR SES: NIST SRM 1633b
Reagents (Fly Ash) and NIST SRM 2711 (Montana Soil) from the
National Institute of Standards and Technology (USA); CRM
All reagents were of analytical-reagent grade and the solutions GBW7405 (Soil), CRM GBW7406 (Soil), CRM GBW7302
were prepared with ultrapure water from a Milli-Q purifier (Sediment) and GBW7311 (Sediment) from the National
system (Millipore, Molsheim, France). A thallium stock Research Center for Certified Reference Materials, China.
solution (1000 mg L21) was prepared by dissolving the All these materials are only certified for total Tl content, but
appropriate amount of TlNO3 (Merck, Darmstadt, Germany). can be useful for assessing the Tl extractability.
A 2000 mg L21 Pd solution, used as matrix modifier, was
prepared from the nitrate salt in 3% v/v HNO3. A 2000 mg L21
Microwave-assisted digestion for determination of non-
ascorbic acid solution was used when required in order to
extractable Tl
obtain reduced Pd. The following reagents were used for
application of the BCR SES: acetic acid (Panreac, Barcelona, Microwave-assisted digestion was used for sample dissolution
Spain), hydroxylamine hydrochloride (Panreac), ammonium so that Tl in the aluminosilicate phase could be released. For
acetate (Merck), and 33% v/v hydrogen peroxide (Merck). this purpose, the solid residue remaining after sequential
Nitric acid (65% m/m, Merck), 48% m/m HF (Panreac) and extraction was placed in a Teflon PFA reactor. Afterwards,
37% m/m HCl (Merck) were used for microwave-assisted 5 mL of 65% m/m HNO3 and 2 mL of 48% m/m HF were
digestion of environmental CRMs. added. Once the reactor was sealed, it was inserted into a
carousel, and the digestion was allowed to proceed for a pre-
Apparatus selected programme, which consisted of the following four
stages with the power, maximum pressure to be reached and
¨
A PerkinElmer (Uberlingen, Germany) Model 4110 ZL atomic digestion time: stage 1 (300 W; 20 psi; 2 min time); stage 2
absorption spectrometer, a PerkinElmer AS-72 autosampler (300 W; 40 psi; 2 min time); stage 3 (300 W, 80 psi; 2 min time);
and a transverse-heated graphite atomiser (THGA) were used and stage 4 (300 W; 140 psi; 25 min time). Once the sample was
for thallium determination. An electrodeless discharge lamp of dissolved, the reactor was allowed to cool to room temperature
Tl (EDL System II, PerkinElmer) was used as the radiation before it was opened. The resulting solution was heated to
source. Longitudinal Zeeman-effect was employed for back- almost dryness so that the HF excess could be removed. A
ground correction. The instrumental parameters and thermal 10 mL volume of digest was injected into the furnace along with
programmes for Tl are shown in Table 1. Integrated absorb- a Pd–ascorbic acid matrix modifier (8 mg Pd z 8 mg ascorbic
ance was measured. acid). The pyrolysis stage was omitted for determination of
An Alresa (Barcelona, Spain) centrifuge (10–15 mL capacity total or residual Tl after microwave-assisted digestion by
tubes) was used for separation of the solid residue from the ETAAS.
extraction liquid after each stage of the BCR SES. A Metrohm
pH-meter equipped with a combined glass electrode was used
for pH adjustments. Results and discussion
A CEM MDS-2000 closed-vessel microwave digestion Optimisation of thermal programmes for determination of
system (630 ¡ 50 W full power) (Matthews, NC, USA) equip- extractable Tl
ped with pressure control, a sample carousel and advanced
composite vessels was used. Optimisation of furnace conditions for determination of
thallium in each fraction of the BCR SES was mainly focused
on the study of pyrolysis and atomisation temperatures and
BCR sequential extraction scheme
amount of matrix modifier. Integrated absorbance was used for
Operating conditions used in the BCR SES were described in signal quantification.
detail elsewhere18 and are summarised in Table 2. Sequential Preliminary experiments for atomisation of 1 ng of Tl with-
extraction was carried out in triplicate on 0.5 g of each CRM in out matrix modifier showed that thallium was lost at very low
Table 2 Reagents and operating conditions employed for the original sequential extraction scheme proposed by the European Community of
Reference (BCR)
Stage Operationally-defined phase Reagent Stirring time and temperature
1 Acid soluble (e.g., carbonates) 20 mL HOAc 0.11 mol dm23 16 h at 25 uC
2 Reducible (e.g., Fe–Mn oxides) 20 mL NH2OH–HCl 0.1 mol dm23 (pH ~ 2) 16 h at 25 uC
3 Oxidisable (e.g., organic matter) 5 mL H2O2 30% m/v (evaporation) 1 h at 25 uC
5 mL H2O2 30% m/v (evaporation) 1 h at 25 uC
25 mL NH4OAc 1 mol dm23 16 h at 25 uC
4 Residual HNO3 z HF
J. Anal. At. Spectrom., 2001, 16, 1424–1428 1425
Fig. 1 Pyrolysis and atomisation curves for Tl in an aqueous standard
with Pd matrix modifier (a, pyrolysis curve; A, atomisation curve) and Fig. 3 Pyrolysis and atomisation curves for Tl in spiked extractants,
without matrix modifier (b, pyrolysis curve; B, atomisation curve). A corresponding to the BCR sequential extraction scheme. Acetic acid
100 ng mL21 Tl standard solution was employed together with a 5 mg (stage 1): with Pd matrix modifier (a, pyrolysis curve; A, atomisation
mass of Pd (as nitrate). curve) and without matrix modifier (d, pyrolysis curve; D, atomisation
curve). Hydroxylamine hydrochloride (stage 2): with matrix modifier
(c, pyrolysis curve; C, atomisation curve) and without matrix modifier
pyrolysis temperatures (e.g., 300 uC). The appearance tempera- (e, pyrolysis curve; E, atomisation curve). Ammonium acetate (stage 3):
ture of thallium was about 1100 uC. Addition of 5 mg of Pd with Pd matrix modifier (b, pyrolysis curve; B, atomisation curve) and
delayed significantly the atomisation of Tl. Tl was efficiently without matrix modifier (f, pyrolysis curve; F, atomisation curve).
Spiked extracts containing a 100 ng mL21 Tl concentration were
stabilised with addition of Pd as matrix modifier; losses being
employed along with a 5 mg mass of Pd as matrix modifier.
observed from 1000 uC. The optimum atomisation temperature
was 1600 uC (Fig. 1). These values are in good agreement with
Likewise, a depressive effect was observed when thallium was
those reported in the literature.29 When studying the influence
atomised from a 1 mol dm23 ammonium acetate solution,
of the Pd mass on integrated absorbance of thallium, a mass of
which was overcome by adding 2 mg Pd as matrix modifier.
at least 2 mg of Pd was needed for good thermal stabilisation.
Under these conditions, optimal pyrolysis and atomisation
No further effects of the Pd mass on integrated absorbance of
temperatures can be set up as above.
thallium were observed in the range 2–14 mg (Fig. 2). Thermal
In order to check for matrix effects on thallium atomised
programmes were optimised for atomisation of thallium from
from real samples, the BCR SES was applied to NIST SRM
0.11 mol dm23, 0.1 mol dm23 hydroxylamine hydrochloride
2711 (Montana Soil). In the first stage, optimal pyrolysis and
and 1 mol dm23 ammonium acetate spiked with 100 ng mL21
atomisation temperatures were identified to be the same as
of thallium. The behaviour of thallium atomised from the
those found for thallium when atomised from an aqueous
acetic acid solution using matrix modifier was similar to that
standard (Fig. 4). Nevertheless, an increased Pd mass was
found with aqueous standards. Optimal pyrolysis and atomisa-
required to reach maximum integrated absorbance. This can be
tion temperatures were very close to those initially obtained
ascribed to matrix co-extraction that can influence thallium
(Fig. 3), and the Pd mass needed for stabilisation was at least
stabilisation. As can be seen in Fig. 5, the optimum range of Pd
2 mg (Fig. 2). No matrix effects were observed for determi-
modifier mass was 6–8 mg for stabilisation of Tl when atomised
nation of thallium in the first stage of the BCR SES. On the
from the acid-soluble and oxidisable fractions.
contrary, a depressive effect was observed when thallium was
Matrix effects were even more severe when thallium was
atomised from the hydroxylamine hydrochloride when no
determined in spiked extracts (reducible fraction) obtained from
modifier was employed, even using very low pyrolysis tem-
the above material. In this case, the depressive interference due
peratures (i.e., 300 uC); integrated absorbance was about 25%
to chloride ions could be promoted by the presence of alkaline
lower as compared with that measured for a thallium aqueous
chlorides.29 Reduced Pd has been recommended in other
standard solution. As expected, this effect can be attributed to
studies on thallium to minimise this interference. Integrated
the formation of TlCl, which is very volatile and leaves the
furnace without being atomised. This matrix effect could be absorbance increased significantly when the ascorbic acid mass
overcome with a 2 mg Pd mass. A Pd mass greater than 10 mg in the matrix modifier mixture was increased, a plateau being
caused the integrated absorbance to decrease when Tl was reached for a 4 mg ascorbic acid mass. In order to ensure the
atomised from hydroxylamine hydrochloride solutions (Fig. 2). formation of reduced Pd, 8 mg of ascorbic acid was chosen for
Fig. 4 Pyrolysis and atomisation curves for Tl in spiked extracts
obtained from NIST SRM 2711 (Montana Soil). Acetic acid (stage 1):
Fig. 2 Influence of the matrix modifier mass (Pd) on the integrated b, pyrolysis curve; B, atomisation curve. Hydroxylamine hydrochloride
absorbance of Tl for an aqueous standard and matrix-matched (stage 2): c, pyrolysis curve; C, atomisation curve. Ammonium acetate
standards with extractants: , aqueous standard; &, 0.11 mol dm23 (stage 3): a, pyrolysis curve; A, atomisation curve. Spiked extracts
*
acetic acid; #, 0.1 mol dm23 hydroxylamine hydrochloride; +, containing a 100 ng mL21 Tl concentration were employed along with
1 mol dm23 ammonium acetate. Optimisation curves are shown for a a 5 mg Pd mass (stages 1 and 3) and 8 mg Pd z 8 mg ascorbic acid (stage
100 ng mL21 Tl concentration. 2) as matrix modifier.
1426 J. Anal. At. Spectrom., 2001, 16, 1424–1428
produced by HCl at a concentration as low as 0.05% v/v.
Consequently, addition of the latter acid to the acid mixture
was omitted so as to avoid complete signal removal. Preli-
minary experiments showed that the use of 8 mg Pd z 8 mg
ascorbic acid as matrix modifier for determination of Tl in
matrix-matched standards with different acids gave a recovery
higher than 95%, thus indicating a good performance of the
matrix modifier employed above for stabilisation of Tl.
Nevertheless, no Tl atomisation signal was observed when
the latter matrix modifier was used for stabilisation of Tl from
spiked digests obtained after microwave-assisted digestion of
NIST SRM 2711 (Montana Soil). Although not completely
removed, matrix effects could be kept under control when the
Fig. 5 Influence of the matrix modifier mass (Pd or Pd z ascorbic acid) pyrolysis stage was omitted but the matrix modifier was
on the integrated absorbance of Tl for spiked extracts obtained from retained. There is evidence in the literature about the suitability
NIST SRM 2711 (Montana Soil): +, 0.11 mol dm23 acetic acid; #,
0.1 mol dm23 hydroxylamine hydrochloride (Pd as the only matrix
of this approach for determination of Tl by ETAAS in NaCl
modifier); à, 0.1 mol dm23 hydroxylamine hydrochloride (8 mg solutions.27
Pd z ascorbic acid; variable amounts of ascorbic acid in mg are Validation results of the microwave-assisted digestion
shown in the abscissa of the graph); &, 1 mol dm23 ammonium method for determination of total Tl content in Soil, Sediment
acetate. A 100 ng mL21 Tl concentration was spiked in all cases. and Fly Ash by ETAAS are shown in Table 3. In all cases, an
accurate Tl content was found following calibration with the
further studies. Other strategies attempted for elimination of standard additions method, as can be confirmed by applying
chloride interference on Tl, such as the use of reduced Pd the t-test for comparison of the experimental and certified
following a thermal treatment at 1000 uC, were unsuccessful. values.
With suitable masses of Pd (8 mg; acid soluble and oxidisable
fractions) or Pd z ascorbic acid (8 mg z 8 mg; reducible
Analytical figures of merit
fraction), the optimal pyrolysis and atomisation temperatures
were 1000 and 1600 uC, respectively. The equation of the calibration curve of Tl is:
A recovery study was carried out using spiked extracts
obtained from NIST SRM 2711 (Montana Soil) and the y ~ 0.0827x z (3.8 6 1023) (1)
optimal furnace conditions described above. Tl recoveries, esti-
mated by interpolating the corresponding integrated absor- where y is integrated absorbance (s) and x is Tl mass (ng); the
bance in a calibration curve made from aqueous standards of regression coefficient was 0.999. The calibration curve was
Tl, were 93 ¡ 3%, 74 ¡ 1% and 103 ¡ 0.5% for the first, linear at least up to 100 ng mL21 (10 mL injection volume) and
second and third stage of the BCR SES, respectively. the characteristic mass (m0) was 53.6 ng, which is in excellent
agreement with that recommended by the manufacturer
(53 pg). The LOD, defined as 3s/m (s, standard deviation and
Optimisation of furnace conditions for determination of residual
m, slope of the calibration graph) was 0.05 mg kg21 whilst the
Tl
LOQ, defined as 10s/m was 0.17 mg kg21. The precision was
For determination of the total Tl content in the residual established from within-batch and between-batch measure-
fraction, the solid residue must be completely dissolved. In this ments carried out with different CRMs. RSDs were usually
work, an acid digestion procedure using microwave heating lower than 5%. Additionally, the F-test (a ~ 0.05) showed that
was assessed. Since most samples had large amounts of non-significant differences between both sets of precision
siliceous components, addition of HF was mandatory. Tl values occurred, thereby demonstrating a good repeatability
determination by ETAAS depends critically on matrix effects for the extraction procedure.
such as those caused by chlorides. The potential interference
caused by several acids that could be used for acid digestion
Partitioning of Tl in certified reference materials
was studied. Fig. 6 shows the integrated absorbance of a
100 ng mL21 solution in the presence of various acid con- The BCR SES method was applied to the partitioning of Tl in
centrations (e.g., HNO3, HF and HCl). HNO3 and HF gave several certified reference materials using the optimised
rise to a depressive interference when their concentrations were conditions established above. The samples and their total
greater than 2% v/v. As expected, more severe interference was certified Tl concentrations were: NIST SRM 1633b Fly Ash
(5.9 mg kg21, indicative value); NIST SRM 2711 Montana Soil
(2.47 ¡ 0.15 mg kg21); GBW 7405 (1.6 ¡ 0.4 mg kg21) and
GBW 7406 (2.4 ¡ 0.6 mg kg21) soils; and GBW 7302
Table 3 Analytical results for determination of total Tl contents in
environmental CRMs
Found Certified
contenta/ contenta/ texp RSD
CRM mg kg21 mg kg21 (tcrit ~ 4.3) (%)b
NIST SRM 2711 2.39 ¡ 0.11 2.47 ¡ 0.15 1.38 4.2
NIST SRM 1633b 6.02 ¡ 0.095 5.9c 2.47 1.4
GBW07405 1.53 ¡ 0.1 1.6 ¡ 0.4 1.22 4.9
GBW07406 2.53 ¡ 0.16 2.4 ¡ 0.6 1.15 4.5
Fig. 6 Influence of the acid concentration on integrated absorbance for GBW07302 2.06 ¡ 0.12 1.9 ¡ 0.5 2.66 4.1
different acids used in microwave-assisted digestion of the residual GBW07311 3.16 ¡ 0.15 2.9 ¡ 0.5 3.34 3.5
fraction: #, HNO3; ', HF; , HCl (5 mg Pd was used as matrix
*
modifier); and &, HCl (8 mg Pd z 8 mg ascorbic acid mixture was used
a
Average content ¡ confidence interval (P ~ 0.95; n ~ 3). bBetween-
batch precision values are given. cIndicative value.
as matrix modifier).
J. Anal. At. Spectrom., 2001, 16, 1424–1428 1427
Table 4 Analytical results (mg kg21)a for Tl following application of BCR SES to environmental CRMs
Fraction SRM 2711 GBW7405 SRM 1633b GBW7406
1 (Acid soluble) vLOQb vLOQb 1.1 ¡ 0.052 vLOQb
2 (Reducible) vLOQb vLOQb 0.5 ¡ 0.021 vLOQb
3 (Oxidisable) 0.26 ¡ 0.01 vLOQb 0.65 ¡ 0.032 0.19 ¡ 0.006
S1–3 0.26 – 2.25 0.19
4 (Residual) 2.10 ¡ 0.19 1.4 ¡ 0.08 4.2 ¡ 0.3 2.0 ¡ 0.15
S1–4 2.36 1.4 6.45 2.2
Certifiedc 2.47 ¡ 0.15 1.6 ¡ 0.4 5.9d 2.4 ¡ 0.6
Recovery (%)e 96 88 109 92
a
Average value ¡ standard deviation (n ~ 3). bTl content is below the LOQ (~10s/m), where s is the standard deviation of the blank, and m is
the slope of the corresponding calibration graph. cAverage value ¡ confidence interval (n ~ 3). dIndicative value. eRecovery was calculated as
the following ratio: [S1–4 content/certified content] 6 100.
(1.9 ¡0.5 mg kg21) and GBW 7311 (2.9 ¡ 0.5 mg kg21) References
sediments. Calibration with aqueous standards was feasible
for quantification of Tl in the acid-soluble and oxidisable 1 F. H. Kemper and H. P. Bertram, in Metals and their Compounds
in the Environment, ed. E. Merian, VCH, Weinheim, 1991,
fractions using optimised furnace conditions. On the other pp. 1227–1241
hand, the standard additions technique was needed for 2 A. Ciszewski, W. Wasiak and W. Ciszewska, Anal. Chim. Acta,
accurate quantification of Tl in the reducible and residual 1997, 343, 225.
fractions. 3 I. Svancara, P. Ostapezuk, J. Arunachalam, H. Emons and
The extraction results (Table 4) are reported as mean values K. Vytras, Electroanalysis, 1997, 9, 26.
and standard deviations(s) corresponding to three separate 4 D. Boehrer and G. Schwedt, Fresenius’ J. Anal. Chem., 1998, 362,
224.
sequential extractions. A low extractability of Tl was observed
5 T. Lu and I. Sun, Electroanalysis, 1998, 10, 1052.
in most CRMs. Tl was only detected in the oxidisable fraction 6 D. Mihajlovic and T. Stafilov, X-Ray Spectrom., 1998, 27, 397.
for soils. The only CRM with extractable Tl in all fractions was 7 D. Mihajlovic and T. Stafilov, Fresenius’ J. Anal. Chem., 1996, 356,
the Fly Ash, where the overall extraction efficiency was close to 371.
40%. No extraction was observed for the sediment CRMs. 8 J. Kucera, M. Vobecky, L. Soukal, D. Zakoucky and D. Venos,
According to the extraction results shown, a low mobility is to J. Radioanal. Nucl. Chem., 1997, 217, 131.
´ ´ ´
9 T. Perez-Ruiz, C. Martınez-Lozano, V. Tomas and R. Casajus, ´
be expected for Tl from environmental samples such as those
Analyst, 1996, 121, 813.
tried in this work. 10 V. Cheam, J. Lechner and R. Desrosoiers, Fresenius’ J. Anal.
This method developed for determination of extractable Tl Chem., 1998, 360, 81.
in environmental solid samples could not be validated against a 11 A. Kuznetsova, N. Chumakova and L. Petrov, J. Anal. Chem.,
material with certified extractable contents for this metal. So 1999, 54, 186.
far, the materials available, such as BCR CRM 601 and BCR 12 T. Asami, C. Mizui, T. Shimada and M. Kubota, Fresenius’
J. Anal. Chem., 1996, 356, 348.
CRM 701 sediments, provide certificate or indicative values for
13 E. Ivanova, X. Yan, W. van Mol and F. Adams, Analyst, 1997,
extractable contents only for Cd, Cu, Cr, Ni, Pb and Zn. 122, 667.
14 T. Stafilov and K. Cundeva, Talanta, 1998, 46, 1321.
´ ´ ˜
15 I. Lopez-Garcıa, E. Navarro, P. Vinas and M. Hernandez- ´
´
Cordoba, Fresenius’ J. Anal. Chem., 1997, 357, 642.
Conclusions ´ ´
16 O. Mestek, R. Koplik, H. Fingerova and M. Suchanek, J. Anal.
At. Spectrom., 2000, 15, 403.
Partitioning patterns of Tl have been established for several 17 A. M. Ure and C. M. Davidson, Chemical Speciation in the
environmental CRMs (Soil, Fly Ash and Sediment) following Environment, Blackie, Glasgow, 1995, pp. 201–271.
the BCR sequential extraction scheme. Furnace conditions 18 A. M. Ure, Ph. Quevauviller, H. Muntau and B. Griepinck, Int.
have been shown to be critical so that accurate determination J. Environ. Anal. Chem., 1993, 51, 135.
of Tl in the extracts or digests is achieved. Tl suffers from 19 M. D. Ho and G. J. Evans, Anal. Commun., 1997, 34, 363.
20 Z. Mester, C. Cremisini, E. Ghiara and R. Morabito, Anal. Chim.
strong interferences from hydroxylamine hydrochloride and
Acta, 1998, 359, 133.
acids. The use of reduced Pd as matrix modifier along with the 21 A. U. Belazi, C. M. Davidson, G. E. Keating, D. Littlejohn and
omission of the pyrolysis stage and avoidance of hydrochloride M. McCartney, J. Anal. At. Spectrom., 1995, 10, 233.
acid in the digestion mixture are recommended. Aqueous 22 J. Userno, M. Gamero, J. Morillo and I. Gracia, Environ. Int.,
standard calibration is feasible for Tl quantification in the acid- 1998, 24, 487.
soluble and oxidisable fractions. On the other hand, the 23 B. Marin, M. Valladon, M. Polve and A. Monaco, Anal. Chim.
Acta, 1997, 342, 91.
standard additions technique is mandatory for the reducible
´
24 B. Perez-Cid, I. Lavilla and C. Bendicho, Analyst, 1996, 121, 1479.
and residual fractions. 25 C. M. Davidson, A. L. Duncan, D. Littlejohn, A. M. Ure and
L. M. Garden, Anal. Chim. Acta, 1998, 363, 45.
26 M. Sager, Mikrochim. Acta, 1992, 106, 241.
27 B. Welz, G. Schlemmer and J. R. Mudakavi, Anal. Chem., 1988,
Acknowledgements 60, 2567.
28 H. Qiao, T. M. Mahmood and K. W. Jackson, Spectrochim. Acta,
˜
This work was financially supported by the Ministerio Espanol Part B, 1993, 48, 1495.
´
de Ciencia y Tecnologıa within the framework of project PB98- 29 B. Welz and M. Sperling, in Atomic Absorption Spectrometry,
1081. Wiley-VCH Verlag GmbH, Weinheim, 1999, pp. 561, 606.
1428 J. Anal. At. Spectrom., 2001, 16, 1424–1428
Related docs
