T. Lifvergren1*, P. Suèr1, U. Wievegg2
1 Man–Technology–Environment Research Centre, Örebro University, SE-701 82 Örebro,
Sweden
2 SAKAB, P.O. Box 904, SE-692 29 Kumla, Sweden
*Corresponding author. Tel.: +46 19 30 36 05, fax +46 19 30 34 63
e-mail:
thomas.lifvergren@nat.oru.se
Two microwave
assisted digestion procedures for determination of aqua regia soluble mercury in solid soil samples have been tested:
Closed vessel and open focused microwave assisted digestion. The two methods
were compared with a standardised conventional (DIN) heating method. The microwave digestion procedures were as
efficient as the standardised method. The lowest deviation from the mean was
observed for the open focused microwave system. Significant time was gained by
using microwave digestions. The standardised method took over 14 hours, while
the closed vessel procedure and the open focused procedure only took 1 hour and
10 minutes, respectively. No losses of volatile mercury species were observed
during aqua regia digestion in the open focused microwave system.
Knowledge of speciation and mobility of the relevant elements, as well as of their total concentration, is essential for assessment of the environmental impact of metal pollution in soils, sediments and sludges. Analytical procedures are generally fast and accurate. However, the sample preparation (digestion, desorption/dissolution) is slow and often the weakest link in terms of precision in the analysis of environmental samples. There are several standardised methods for determination of the total contents of metals in environmental samples based on digestion in concentrated acids. The leaching acid can be selected to desorb or dissolve only secondary metal species (adsorbed, precipitated/co-precipitated), not affecting the mineral matrix (e.g. silicates or crystalline oxides).
Microwave radiation causes dipole rotation and ion-transmission effects
in molecules, which can lead to vibrations and thereby a mobilization of trace
metals that are bound to a solid surface without degradation of the solid
matrix (Kingston and Jassie, 1988). Thus, microwave assisted procedures have the
potential to reduce leaching times, as well as enhance the separation of the
associated metals from the solid soil/sediment matrix without inflicting a
complete chemical degradation of the sample. Two types of microwave procedures
are currently in use: open focused microwave and closed vessel microwave digestions.
The open focused microwave systems operate at atmospheric pressure. The
temperature that can be obtained is limited by the boiling point of the
extraction liquid used. Microwaves are focused on each sample. The radiation
intensity may be adjusted for achievement of the desired digestion temperature.
This leads to the possibility for mild extractions, where selected metal
species can be mobilized by the digestion procedure (Schmitt et al.,
1996/1997). Closed vessel microwave digestion operates at
elevated pressures, which allows higher digestion temperatures. The microwave
radiation is non-focused. The vessels are rotated to improve the
reproducibility. Increased boiling temperatures compensate the loss in
vibration effects on the aggregates due to non-focused radiation. Radiation
intensity is controlled through pressure gauges in the vessels. The lack of
temperature control and the indirect control on the microwaves make this
technology unsuitable for the purpose of metal speciation.
It is claimed that open microwave procedures are unsuitable for analysis
of trace metals in solid samples (Bulska et
al., 1995; Zhou et al., 1996). The main objection is that the open system
would be less efficient than the closed vessel system due to the lower
digestion temperature. For an element that can be transferred to volatile
species like mercury the loss due to volatilization is often stressed.
This article describes a comparison of three digestion practices using aqua regia: conventional heating, closed
vessel microwave digestion and open focused microwave digestion. First, methods
for microwave assisted digestions of solid samples had to be developed. This
was achieved by working with a reference material until satisfactory mobilization
(recovery) of mercury was obtained. The microwave digestions were compared with a
standardised conventional heating procedure. A mercury contaminated soil from a
chlor-alkali industrial site in southern Sweden was provided for the test.
Reference
material
A sewage sludge of industrial origin, from the
European communities bureau of reference, (BCR 146R), was used as a reference.
With a total mercury content of 8.62 mg/kg ± 0.33
and an aqua regia soluble fraction of
8.39 mg/kg ±0.25 (Quevauviller
et al., 1996) was
used as a reference material.
Soil:
A soil from a mercury contaminated chlor-alkali
industrial site in south west of Sweden was used in the studies. The
homogenised soil contained 5.2 % water and 0.5% organic matter and had a pH of
7.5 (water suspension).
Mercury addition
Elemental mercury and methyl mercury
(pro analysi quality; from
Kebo, Sweden) were used for spiking.
Conventional heating
A standardized procedure was followed (for details see DIN 38 414, 1983). Aqua
regia (HCl:HNO3 = 3:1) was added to the solid sample (28 ml per
3 g) in a glass vessel. The mixture was heated according to the program given
in Table 1. The aqua regia was
quantitatively collected. All samples were stored in a refrigerator until
mercury analysis.
Closed vessel microwave assisted heating
Aqua
regia was added to the solid sample (10 ml per 0.3
g) in a teflon vessel. The vessels were placed in a closed microwave oven
(CEM-MDS 2000) and heated under pressure according to the program given in
Table 1.
Open focused
microwave assisted heating
Aqua
regia was added to the solid sample (12 ml per 0.3
g) in a 250 ml glass vessel and which was capped with a reflux condenser. The aqua regia was automatically added to
the vessels in the open system (Prolabo Microdigest 3.1). The program in Table 1 was followed.
Two types of CVAAS systems
were used for analysis. The BCR reference
sample was analyzed with a Perkin Elmer-FIMS (Flow Injection Mercury System),
using tin(II)chloride as reducing agent and argon (flow rate 20ml/min) as
stripping gas. The soil samples were analyzed with a Perkin Elmer 3030 equipped
with MHS-10 (Mercury Hydride System) system, using sodium
tetraborhydrate as reducing agent.
As can be seen in table 2, all digestion procedures extract similar amounts of mercury when the mercury polluted soil was used as sample. The amount of mercury extracted when the certified BCR 146R sludge was used as sample is also provided in table 2. However, in the latter case, the recovery of mercury WAS only 94,6% and 96,8% for the open respectively closed vessel system compared to the certified value. The reason for a not total recovery is probably not caused by lower extraction efficiency in the microwave procedures. This was supported by the fact, that 8,01 ± 0.25mg Hg/kg was extracted during a triplicate digestion made on the BCR sludge following the DIN method (certified value 8.39±0.25mg Hg/kg). Therefore, it is likely to expect that the observed divergence has its origin in the analysis or in the sample it selves.
Nevertheless, if the two microwave procedures are compared (see table 2), a little higher dispersal in the results can be seen for the closed vessel system in both tests. This could be due to the non-focused microwave radiation used in the closed system, which causes differences in the radiation uptake between different samples. It is thereby not unlikely that a divergence in the extraction efficiency appear. In the soil test, homogenisation artefacts in the soil can not be excluded and could thereby be the explanation.
The open microwave system was further tested by analysis of methyl mercury and metallic mercury following the procedures of Table 1. Recoveries in the range 87-105% and 94-108 % for methyl mercury and metallic mercury, respectively, were observed. The spreading reflects the uncertainty introduced by the large dilution of the sample that was required prior to analysis. There were no indications of mercury losses due to the releases of volatile species during the dissolution. Evidently, the oxidising conditions during leaching with aqua regia stabilises mercury as soluble Hg(II), and thereby preventing volatilisation.. Tseng and co-workers (1998) have previously used open microwave system for quantitative leaching of methyl mercury from certified solid materials with high recovery.
Table 1: Digestion procedures
|
|
Standardised-DIN |
Closed system |
Open system |
Weight of
sample, g
|
3 |
0.3 |
0.3 |
|
Volume of aqua regia, ml |
28 |
10 |
12 |
|
Microwave power, W |
- |
630/12 samples |
80a/1 sample |
|
Program: |
|
|
|
|
- Preheating time , min |
720 |
- |
5 |
|
- Boiling/heating, min |
120 |
3+10+30b |
4 |
|
- Cooling, min |
-c |
15 |
-c |
Total time, h
|
14 |
1 |
0.15 |
a
40% of total 200 W/sample
b
3 min at 60 psi, 10 min at 110 psi, 30 min at 180 psi
c
No cooling time is needed, the liquid could be transferred to volumetric flasks
immediately
Table 2: Mercury analysis on aqua regia digested soil samples and reference material (mean
values and 95% confidence interval)
|
|
Soil sample |
Reference material |
Standardised –DIN, 12 replicates
|
80,5 ± 3.0 |
8.01 + 0.25a |
|
Closed system, 20 replicates |
84,6 ± 4.3 |
8.12 ± 0.82 |
|
Open system, 20 replicates |
82.4 ± 1.6 |
7.94 ± 0.17 |
|
Certified value, reference material |
- |
8.39 ± 0.25 |
aThe result is only based on three samples.
This study has shown that
the microwave digestion procedures are as efficient in mercury extraction as
the standardised method. The lowest deviation from the mean values was observed
for the open focused microwave digestion. The standardised method takes over 14
hours. The closed and open microwave procedures, however, only require 1 hour
and 10 minutes, respectively.
Bulska E., Kandler W., Paslawski P. and Hulanicki A. (1995) Atomic absorption spectrometric determination of mercury in soil standard reference material following microwave sample pretreatment. Mikrochimica Acta 119, 137-146.
DIN 38 414 (1983) Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung, Schlamm und Sedimente (Gruppe S), Aufschluss mit Königswasser zur nachfolgenden Bestimmung des Säurelöslichen Anteils von Metallen (S 7) (ed. N. W. N. i. D. D. I. f. N. e.V.), pp. 1-6. Beuth Verlag GmbH, Berlin.
Kingston H. M. and Jassie L. B. (1988) Monitoring and predicting parameters. In Introduction to Microwave Sample Preparation, pp. 112-153. American Chemical Society, Washington DC.
Quevauviller P., Muntau H., Fortunati U. and Vercoutere K. (1996) The certification of the total contents (mass fractions) of Cd, Co, Cr, Cu, Hg, Mn, Ni, Pb and Zn and the aqua regia soluble contents (mass fractions) of Cd, Co, Cr, Cu, Hg, Mn, Ni, Pb and Zn in a sewage sludge from industrial origin. European Commission.
Schmitt V. O., de-Diego A., Cosnier A., Tseng C. M., Moreau J. and Donard O. F. X. (1996/1997) Open focused microwave-assisted sample preparation procedures: fundamentals and application to the speciation of tin and mercury in environmental samples. Spectroscopy 13, 99-111.
Tseng C. M., Schmitt V. O., de Diego A. and Donard O. F. X. Open focused microwave assisted sample preparation: A rapid and simple solution for mercury species determination in environmental solid samples. Submitted to: American Environmental Laboratory.
Zhou C. Y., Wong M. K., Koh L. L. and Wee Y. C. (1996) Comparison of acid mixtures in high pressure microwave digestion methods for the determination of the total mercury in sediments by cold-vapor atomic absorption. Analytical Sciences 12, 471-476.