Anal Bioanal Chem (2002) 373 : 9397

DOI 10.1007/s00216-002-1290-2

O R I G I N A L PA P E R

Ari Visnen Reijo Suontamo Janne Silvonen

Jukka Rintala

Ultrasound-assisted extraction in the determination of arsenic,

cadmium, copper, lead, and silver in contaminated soil samples
by inductively coupled plasma atomic emission spectrometry
Received: 10 December 2001 / Revised: 11 March 2002 / Accepted: 11 March 2002 / Published online: 11 April 2002
Springer-Verlag 2002

Abstract An extraction method was developed for the

determination of toxic elements in contaminated soil samples by inductively coupled plasma atomic emission spectrometry (ICP-AES). The determination of arsenic, cadmium, lead, and silver in ultrasound-assisted extracts of
SRM 2710 and SRM 2711 by ICP-AES was carried out
with high accuracy and precision (RSD<3.7%). The certified concentrations of the SRMs were obtained for arsenic, cadmium, lead, and silver by using an ultrasoundassisted extraction method with a digestion solution of
(1+1)-diluted aqua regia. The determination of copper in
SRMs by the ultrasound-assisted extraction method and
analysis by ICP-AES failed to obtain the certified concentrations at the 95% level of confidence using (2 s) as
confidence limits of the mean. However, the same results
were observed with the use of the microwave digestion
method and reflux, which is the ISO 11466 standard
method. The analysis of the SRMs showed that the ultrasound-assisted extraction method is highly comparable
with the other methods used for such purposes. The major
advantages of the ultrasound-assisted extraction method
compared to the microwave and reflux methods are the
high treatment rate (50 samples simultaneously in nine
minutes) and low reagent usage, the main benefit of
which are the low chloride and nitrate concentrations in
the extracts.
Keywords Ultrasound-assisted extraction ICP-AES
Elemental analyses Toxic metals Contaminated soil

A. Visnen () R. Suontamo J. Silvonen

University of Jyvskyl, Department of Chemistry,
P.O. Box 35, 40351, Jyvskyl, Finland
e-mail: avaisane@cc.jyu.fi
J. Rintala
University of Jyvskyl,
Department of Biological and Environmental Science,
P.O. Box 35, 40351, Jyvskyl, Finland

Introduction
Metal contamination is a persistent problem at many contaminated sites. The main sources of this contamination
are metal smelting and processing, secondary metal production, lead and nickel cadmium battery manufacturing,
pigment and chemical manufacturing, and disposal of
wastes. The metal most commonly found at contaminated
sites is lead [1]. Metals such as arsenic, cadmium,
chromium, copper, mercury, and zinc are also playing a
significant role in soil contamination [2]. It is well known
that these metals can pose a significant risk to human
health and to the ecosystem [3].
Areas contaminated by metals can be extended through
the mobility of toxic metals particularly arsenic and antimony, in pore or ground water [4]. The mobility of toxic
metals can increase the risk of their accumulation in
plants and animals, which in turn impairs the safety of
agricultural soils [5, 6]. The mobility of metals and their
bioavailability in soils have been studied by sequential extraction procedures [7, 8, 9]. The sequential extraction
procedure of Tessiers et al. [10], which includes five-step
fractionation from exchangeable metals to total metal
content, is probably the most useful method for this purpose. In the Tessiers method, the total metal analysis of
soil samples is performed by digestion with a mixture of
concentrated HF and HClO4. However, the recovery of total metal content is one of the most critical steps in the
whole procedure.
At present, only a few standard methods exist for the
analysis of toxic metals in soils [11]. The analytical methods normally applied are based on the methods recommended for the analysis of sediments and agricultural
soils. Extraction of metals from soil samples is usually
based on digestion with concentrated or (1+1)-diluted
acids such as nitric acid, hydrofluoric acid, hydrochloric
acid, perchloric acid, or aqua regia (ISO 11466 standard
method) [11, 12]. Digestion is usually carried out by the
microwave digestion procedure, but the digestion under
reflux or in an autoclave can also be used [12, 13, 14].

94
Table 1 Calibration data of
the determination of samples
by ICP-AES

aLimit

of detection (a+3sa).
of quantification
(a+10sa)

bLimit

Element

Wavelength
(nm)

LODa
(mg L1)

LOQb
(mg L1)

Calibration range
(mg L1)

Ag
As
Cd
Cu
Cu
Pb

328.068
189.042
214.438
324.754
324.754
283.306

0.9997
1.0000
0.9998
1.0000
0.9997
1.0000

0.020
0.200
0.020
0.060
0.700
0.400

0.060
0.600
0.060
0.180
2.200
1.300

0.10.9
2.026.0
0.10.9
0.18.1
1.033.0
5.065.0

Microwave digestion procedures have been reviewed

by Lamble and Hill [15]. It is well documented that the
microwave digestion method is suitable for the digestion
of various sample matrices without the loss of analyte elements. The main reason for the interest shown in the microwave digestion method is the reduced digestion time,
without reduction in effectiveness, compared to reflux or
digestion in an autoclave. The microwave digestion of soil
samples can also be carried out with hydrofluoric acid and
perchloric acid, which cannot suitably be applied under reflux. It should be noted that hydrofluoric acid cannot be
used in conjunction with the inductively coupled plasma
atomic emission spectrometry (ICP-AES) technique because of corrosion of the quartz plasma torch [16].
Slurry sampling has been used in the analysis of metals
in several kinds of samples by the electrothermal atomic
absorption spectrometry (ETAAS) technique [17]. The
method is based on that of ultrasound-assisted homogenization with a few milligrams (~10 mg) of sample. Ultrasound-assisted homogenization is typically carried out
with ETAAS autosampler cups with a volume of about
2 mL [18, 19, 20].
The ultrasound-assisted extraction method described
above can effectively be used with ETAAS analysis. In
this study, an ultrasound-assisted extraction method for
the determination of metals in soil samples by atomic absorption techniques and inductively coupled plasma (ICP)
techniques with either atomic emission spectrometry (AES)
or with mass spectrometry (MS) was developed. The total
extraction of metals can be carried out by means of an ultrasonic water bath. In this method, sample weight can be
increased up to 500 mg and plastic screw-top bottles with
stoppers can be used as digestion vessels. The optimization of the whole extraction procedure was carried out by
the analysis of two contaminated soil standard reference
materials SRMs. The advantages of this method are, firstly,
the speed of extraction that can be as high as 50 samples
in nine minutes and, secondly, that extraction can also be
carried out in an ultrasonic water bath, which is a standard
laboratory item, and therefore the acquisition of special
instruments is not required.

with the instrument default parameters (viewing height 15 mm,

nebulizer flow 1.000 L min1, plasma power 1000 W, auxiliary gas
flow 1.0 L min1, and plasma gas flow 15 L min1). The wavelengths used are shown in Table 1. An automatic background correction was used at each wavelength.
Reagents and samples
All the reagents used were of analytical reagent-grade, and only
high purity ELGA water was used. Nitric acid (16 mol L1) (65%,
p.a.), and hydrochloric acid (12 mol L1) (3638%, p.a.) were supplied by Riedel-de-Han. The standard stock solutions of Ag, As,
Cd, Cu, and Pb (1000 mg L1) for the ICP-AES measurements
were supplied by Merck. The working concentration ranges used
are shown in Table 1.
Two contaminated soil standard reference materials, SRM
2710 and SRM 2711, both certified by the National Institute of
Standards and Technology (NIST) [21, 22] were analyzed.
Procedure
The aim of this study was to develop an effective and rapid method
of extraction with high sample treatment capacity for the evaluation of metal contaminated areas. One highly contaminated SRM
2710 [21] and one moderately contaminated SRM 2711 [22] were
analyzed. An ultrasound-assisted extraction method was developed and compared to two universally used digestion methods, including the ISO 11466 standard method [11].
Ultrasound-assisted extraction
A SRM sample of about 500 mg was accurately weighed into a
50 mL plastic screw-top bottle with a stopper and wide neck (graduated bottle, supplied by Kartell), and 10 mL of digestion solution
of (1+1)-diluted aqua regia was added. The bottle was closed and
placed into a 400 W, 40 kHz, Model FS400b ultrasonic water bath
supplied by Decon. The ultrasound-assisted extraction procedure
was carried out at a temperature of about 40 C. The optimized
sonication procedure lasts 9 min, divided into three equal steps.
The optimization procedure is described below in the results and
discussion section. After the first 3 min sonication step, the sample
bottle was shaken by hand to prevent sedimentation, after which
the sample solution was placed into an ultrasonic water bath for
the second 3 min sonication step. This procedure was repeated
once before the final 3 min sonication step. After cooling, the sample solution was filtered (Whatman No. 41 filter paper) into a
50 mL volumetric flask. The residue was washed three times with
a few milliliters of water and diluted with water to a volume of
50 mL. The sample solution was transferred into a plastic screwtop bottle for storage.

Experimental
Instrumentation

Reflux

All measurements were performed with a Perkin-Elmer Model

2000 ICP-AES. The measurements of the elements were made

A SRM sample of about 1000 mg was accurately weighed into a

round-bottomed 100 mL flask and 30 mL of digestion solution of

95
Table 2 The microwave digestion programme for SRMs
Stage

Power
(W)

Ramp
timea
(min)

Pressure
(Pa105)

Temperature (C)

Hold
time
(min)

1
2

600
1200

5
10

5.0
10.0

175
210

3.0
10.0

aRamp

time of the pressure needed

(1+1)-diluted HNO3 or concentrated agua regia (ISO 11466 standard method) was added. After 2 h, the mixture was gently heated
at reflux for 4 h and then allowed to cool to room temperature. After cooling, it was filtered through a Whatman No. 41 filter paper
into a 100 mL volumetric flask. The residue was washed with
15 mL of water introduced in 3 portions, and the filtrate was diluted to 100 mL with water. The sample solution was transferred
into a plastic screw-top bottle for storage.
Microwave digestion
A SRM sample of about 300 mg was accurately weighed into a
PTFE pressure vessel and 10 mL of digestion solution of conc.
HNO3 was added. The closed vessel was placed in a microwaveassisted sample digestion system (MARS 5 supplied by CEM Corporation). The microwave digestion program [15] is presented in
Table 2. After cooling, the sample solution was filtered (Schleicher
& Schuell 589 filter paper) into a 50 mL volumetric flask. The
residue was washed with water and diluted to a volume of 50 mL
with water. The sample solution was transferred into a plastic
screw-top bottle for storage.

Results and discussion

Calibration
All the concentration measurements were carried out using a six-point calibration. Single element calibration
curves for the elements (Ag, As, Cd, Cu, and Pb) were
used. The determination of each element was performed
by taking the most sensitive emission lines (Table 1) to
obtain maximum sensitivity. The calibration data (correlation coefficient of regression (r), limit of detection
(LOD), and limit of quantification (LOQ)) (Table 1) were
calculated by Excel. The calibration line is expressed in
the usual way as y=bx+a, where b is the slope of the line
and a is the intercept. The LOD and LOQ values were calculated from the parameters of the calibration line (slope
b and its standard deviation sb, intercept a and its standard
deviation sa) by substituting a+3sa or a+10sa respectively
into the calibration line y=bx+a. Exceedingly high values
were obtained for the regression correlation coefficients,
as shown in Table 1.
Optimization of the ultrasound-assisted
extraction procedure
The ultrasound-assisted extraction procedure was optimized by the analysis of two replicate samples of about
250 or 500 mg of two contaminated soil SRMs. The opti-

Fig. 1ac Effect of sonication time on a As, b Cu, and c Pb concentrations of ultrasound-assisted extracts [(1+1)- or (3+1)-diluted
aqua regia] of SRM 2710, Cert. the certified concentration of Pb in
SRM 2710 and S sonication time divided into two or three equal
steps

mization was carried out by the determination of the As,

Cu, and Pb concentrations by ICP-AES. These elements
were selected because they are all relevant to our series of
studies of soil contaminated by lead pellets. Digestion solutions of 10 mL of (1+1)- or (3+1)-diluted and conc.
HNO3 or aqua regia were tested. Sonication time was
415 min with and without dividing into two or three
steps.
The optimization process showed that the concentrated
aqua regia digestion solution could not be used with the
plastic digestion vessels owing to the high formation of
active chlorine gas. Extraction with HNO3 was carried out
successfully in the case of the Pb concentrations, but not
in the case of the As and Cu concentrations. Pb was effec-

96

Fig. 2 Effect of a sonication time, divided into two or three equal

steps, on Cu and Pb concentrations of ultrasound-assisted extracts
[(1+1) or (3+1)-diluted aqua regia] of SRM 2711, Cert. the certified concentrations of Pb and Cu in SRM 2711

tively extracted from the soil samples by the ultrasoundassisted extraction method with the three digestion solutions when the sonication time was divided into steps including shaking by hand between the steps. The extraction of Pb was carried out with great success when (1+1)diluted aqua regia was used with three sonication steps for
SRM 2710 as can be seen in Fig. 1. In the case of SRM
2710, slightly higher concentrations of As and Cu were
obtained when (1+1)-diluted aqua regia was used compared (3+1)-diluted aqua regia. As significant differences
were not found between the sample weights of about 250
and 500 mg, 500 mg weight was used to increase the concentrations to be determined.
The extracted concentrations of all three elements in the
SRM 2710 were affected by the sonication time and the division of the sonication time into steps, as can be seen in
Fig. 1. The certified concentrations of Pb were obtained for
both SRMs (Figs. 1 and 2) with (1+1)-diluted aqua regia as
the digestion solution and the sonication time of 9 min divided into three equal steps. The sonication time of 9 min
with steps was evaluated to be the best for the extraction of
As and 6 or 9 min for Cu in the case of SRM 2710.
According to the optimization procedure of SRM
2710, the different sonication times were tested with steps
only in the case of SRM 2711. The As concentrations in
SRM 2711 were too low to be determined by ICP-AES.
Table 3 Determined element
concentrations (mg g1) with
2 s of three replicate samples
of SRM 2710 and 2711

Ultrasound aqua
regia (1+1)

The extraction of Cu was not affected by the sonication

time when (3+1)-diluted aqua regia was used as the digestion solution (Fig. 2). Extension of the sonication time
from 4 to 9 min increased the determined Cu concentrations significantly when (1+1)-diluted agua regia was
used as the digestion solution. The extracted Pb concentrations of SRM 2711 showed no linear relationship with
the sonication time when (3+1)-diluted aqua regia (Fig. 2)
was used as the digestion solution.
According to the optimization process, the ultrasoundassisted extraction procedure with 10 mL of (1+1)-diluted
aqua regia as the digestion solution and a sonication time
of 9 min divided into three equal steps were valid for the
determination of Ag, As, Cd, Cu, and Pb in the soil samples by ICP-AES. The use of (1+1)-diluted aqua regia as
the digestion solution will yield nitrate concentrations below 0.5 mol L1 (2%), which is in perfect agreement with
the criteria for successful analysis by ETAAS.

Analysis
Two contaminated soil standard reference materials, SRM
2710 and SRM 2711, were analyzed. The ultrasound-assisted extraction method was compared to two universally
used digestion methods. The elements to be determined,
Ag, As, Cu, and Pb, were selected because they are all
present in areas contaminated by lead shotgun pellets. Sb
was also of interest, but it results in too low concentrations in SRMs, as do Ag and As in SRM 2711 to be determined by ICP-AES. Cd was selected in a group of detectable elements owing to its very high toxicity.
The concentrations (mean2 s) of the five elements
(Ag, As, Cd, Cu, and Pb) in SRMs extracted by the three
different methods and determined by ICP-AES are shown
in Table 3. We used mean concentrations with the standard deviation multiplier 2 (2 s) instead of s as given in
the certificates of analysis. According to the normal distribution, approximately 95% of the determined values lie
within 2 s of the mean [23]. However, it was found that
the determination of three replicate samples is precise
(RSD<3.7%) enough (Table 3) for the analyses of metal
contaminated soils, which significantly reduces the need
for sample pre-treatment.

Microwave conc.
HNO3

Reflux conc.
aqua regia

Reflux HNO3
(1+1)

Certified

SRM 2710
Ag
0.0360.002
As
0.6030.028
Cd
0.02210.0012
Cu
2.720.06
Pb
5.490.30

0.02880.0008
0.6040.026
0.02280.0022
2.890.04
5.570.14

0.03330.0012
0.6020.052
0.02320.0030
2.760.12
5.350.24

0.03070.0010
0.6030.022
0.02430.0014
2.660.04
5.280.24

0.03530.0015
0.6260.038
0.02180.0002
2.950.13
5.530.08

SRM 2711
Cd
0.0420.002
Cu
0.1050.008
Pb
1.170.03

0.0390.004
0.1100.004
1.030.06

0.0430.003
0.1190.002
1.160.06

0.0420.002
0.1160.006
1.150.04

0.04170.0003
0.1140.002
1.1620.031

97

The determination of these elements in SRM 2710 and

SRM 2711 by ICP-AES was carried out with great success as can be seen in Table 3. The certified concentrations of Ag, As, Cd, and Pb in SRM 2710 were obtained
with the ultrasound-assisted extraction method according
to the mean2 s. The determination of Cu in SRMs failed
to achieve the certified concentrations at the 95% level of
confidence using 2 s as the confidence limits of the
mean. It should be noted that in the case of SRM 2710, the
determined Cu concentrations were also too low in extractions performed by the standard method and by reflux
with (1+1)-diluted HNO3 as the digestion solution. The
microwave-assisted extraction method resulted in concentrations that were too low in Ag in SRM 2710 and Pb in
SRM 2711 according to 2 s as the confidence limits of
the mean. In general, the ultrasound-assisted extraction
method with (1+1)-diluted aqua regia as the digestion solution and the standard method (reflux with aqua regia)
resulted in almost equal element concentrations.

Conclusion
An extraction method for the determination of Ag, As,
Cd, Cu, and Pb in contaminated soil samples by ICP-AES
was developed. The ultrasound-assisted extraction method
with (1+1)-diluted aqua regia as the digestion solution and
33 min sonication time was found to be useful for the
analysis of contaminated soil samples. This extraction
method was compared to microwave digestion and reflux
with digestion solutions of (1+1)-diluted HNO3 and aqua
regia, which is the standard method [11]. The analysis of
the SRMs showed the effectiveness of this extraction
method for these purposes. The certified element concentrations of the SRMs were obtained with high accuracy
and precision, as can be seen in Table 3. It should be noted
that highly volatile elements such as As could also be extracted by the ultrasound-assisted extraction method without loss of analyte. The only exception was Cu in which
the certified concentrations in SRM 2710 were not obtained by any of these methods. The analytical results for
element concentrations obtained with the different extraction methods are highly comparable, and none of these
methods are intrinsically superior. The advantages of the
ultrasound-assisted extraction method are: 1) speed of extraction, 2) high sample capacity and 3) low nitrate and
chloride concentrations. The fact that 50 samples could be
treated at the same time in nine minutes is significant
when this method is compared to the conventional methods of digestion. The standard method ISO 11466 is an effective one, but far too difficult to be used in continuous
laboratory routines. About 816 samples can be treated in
1070 minutes by the microwave digestion method,
which is a significantly inferior to the ultrasound-assisted
extraction method. Chloride and nitrate concentrations are
also significantly lower in the ultrasound-assisted extraction method than in the other methods. Chloride concentrations should be as low as possible when samples are to
be analyzed by the ETAAS technique because of the pos-

sible interferences of chlorides on analyte elements and

the corrosion of graphite tubes [24, 25]. However, the ultrasound-assisted extraction method with (1+1)-diluted
aqua regia as the digestion solution is probably also a useful method for the determination of elements by ETAAS.
Future studies should investigate the use of the ultrasound-assisted extraction method in the determination of
arsenic and cadmium in complicated sample matrices by
ETAAS.

References
1. Smith LA (1995) Remedial options for metals-contaminated
sites. Lewis Publishers, Boca Raton
2. Heikkinen P (2000) Retention and migration of harmful substances in the soil. Report of investigation, Geological survey
of Finland, Espoo
3. Berman E (1980) Toxic metals and their analysis. Heyden &
Son, Cambridge
4. Hammel W, Steubing L, Debus R (1998) Ecotoxicol Environ
Saf 40:173176
5. Fytianos K, Katsianis G, Triantafyllou P, Zachariadis G (2001)
Bull Environ Contam Toxicol 67:423430
6. Syakalima M, Choongo K, Chilonda P, Ahmadu B, Mwase M,
Onuma M, Sugimoto C, Tsubota T, Fukushi H, Yoshida M, Itagaki T, Yasuda J, Nakazato Y (2001) Bull Environ Contam
Toxicol 67:438445
7. Ahnstrom ZA, Parker DR (1999) Soil Sci Soc Am J 63:1650
1658
8. Murphy AP, Coudert M, Barker J (2000) J Environ Monit 2:
621627
9. Tack FMG, Verloo MG (1995) Intern J Environ Anal Chem
59:225238
10. Tessier A, Campbell PGC, Bisson M (1979) Anal Chem 51:
844851
11. ISO 11466 (1995) Soil quality; Extraction of trace elements
soluble in aqua regia
12. Karstensen KH (1996) Nordic guidelines for chemical analysis
of contaminated soil samples. Nordtest report 329, Espoo
13. Binstock DA (2000) Anal Lett 33:33973406
14. Visnen A, Suontamo R, Rintala J (2001) J Anal At Spectrom
16:533537
15. Lamble KJ, Hill SJ (1998) Analyst 123:103133R
16. Visnen A, Matilainen R, Tummavuori J (2000) Fresenius J
Anal Chem 367:755760
17. Cal-Prieto M, Felipe-Sotelo M, Carlosena A, Andrade J,
Lpez-Mahi P, Muniategui S, Prada D (2002) Talanta 56:1
51
18. Capelo JL, Lavilla I, Bendicho C (1998) J Anal At Spectrom
13:12851290
19. Lavilla I, Capelo JL, Bendicho C (1999) Fresenius J Anal
Chem 363:283288
20. Amoedo L, Capelo JL, Lavilla I, Bendicho C (1999) J Anal At
Spectrom 14:12211226
21. Gills TE (1997) Certificate of analysis, standard reference material 2710. National Institute of Standards and Technology,
Gaithersburg
22. Gills TE (1993) Certificate of analysis, standard reference material 2711. National Institute of Standards and Technology,
Gaithersburg
23. Miller JN, Miller JC (2000) Statistics and chemometrics for analytical chemistry. Pearson Education Limited, Dorchester
24. Lajunen LHJ (1992) Spectrochemical analysis by atomic absorption and emission. The Royal Society of Chemistry, Cambridge
25. Schlemmer G, Radziuk B (1999) Analytical graphite furnace
atomic absorption spectrometry. Birkhuser Verlag, Berlin