Published in J. Environ. Qual. 33:1943-1944 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA
SHORT COMMUNICATIONS
Adsorption of Trace Metals on Glass Fiber Filters
Mark Fuhrmann* and
Jeffrey P. Fitts
Environmental Sciences Department, Building 830, Brookhaven National Laboratory, Upton, NY 11973
* Corresponding author (fuhrmann{at}bnl.gov).
Received for publication February 6, 2004.
 |
ABSTRACT
|
|---|
Filters, containing glass-fiber (GF) filter material, are commonly used as the primary filter or as the prefilter in sampling natural waters and laboratory experiments with high concentrations of suspended solids. We observed that GF filter material removed substantial quantities of trace metals from solutions of low ionic strength at near neutral and slightly acidic pH. The GF material sorbed essentially all Pb and Ag from 5-mL aliquots of solutions containing 0.054 and 0.093 mM, respectively. Somewhat less Ni was sorbed from a 0.099 mM solution. This material retained about 43 µmol of Ag per gram of GF material (4600 µg/g). The Ag and Ni sorption was highest at low KNO3 concentrations (as background electrolyte) and decreased to a constant concentration of sorbed metal at approximately 10 mM KNO3. Glass-fiber filter material should only be used with careful testing for the elements of concern under conditions that closely match expected environmental or experimental conditions.
Abbreviations: GF, glass fiber • SFCA, surfactant-free cellulose acetate
WHILE CONDUCTING method blank tracer tests in association with sorption experiments for a grout system, we observed that syringe filters, containing GF filter material, removed substantial quantities of cations from solutions of low ionic strength at near neutral and slightly acidic pH. Syringe filters, including those with GF prefilters, are often used in environmental research and in sampling natural waters, with the assumption that they are essentially inert. While passive capillary samples using GF material have been reported to release ions such as sodium and silica (Goyne et al., 2000), we are not aware of any reports regarding significant adsorption of trace metals onto GF material. An informal survey of environmental sampling and research groups and manufacturers indicated that there was no recognition of significant sorption onto GF material.
A set of salt solutions of toxic metals (Ag, Pb, Ni, Se, Cr) and radionuclide tracers (137Cs, 125I) were prepared in distilled water for use in a series of sorption isotherm experiments. Their concentrations and pHs are given in Table 1, and their ionic strengths ranged from 0.09 to 0.3 mM. Following our standard lab practice, a set of method blank solutions (containing the element of interest) were prepared and treated in the same manner as the actual samples, but without adding the solids. Both blanks and isotherm samples were passed through syringe filters and analyzed by inductively coupled plasma–optical emission spectroscopy (Ag, Pb, Ni, Se, Cr), liquid scintillation counting (125I), and gamma-ray spectroscopy using a high-purity Ge detector (137Cs). Normally we use surfactant-free cellulose acetate (SFCA) syringe filters. Because these experiments contained several grams of solids, SFCA filters were used with built-in GF prefilters.
Analyses of the method blanks revealed significantly lower concentrations than expected. Subsequent tests using different syringe filter materials identified the GF portion of the filter as the cause of the discrepancy. The following filters were tested:- Cole-Parmer SFCA + GF (0.45 µm; Cole-Parmer, Vernon Hills, IL)
- Whatman Puradisc GF/F (0.7 µm; Whatman, Maidstone, UK)
- Nalge SFCA (0.45 µm; Nalge Nunc International, Rochester, NY)
- Cole-Parmer SFCA (0.20 µm)
Table 1 shows the solution concentrations before and after being passed through the various filter media. The "no filter" column is the starting, or reference, concentration. With the exception of Ni, all of the concentrations of SFCA-only filters are within analytical error (±3.4% for metals and ±9.8% for radionuclides at 2
) of the "no filter" values. Uptake only occurred for cations, that is, all elements tested except Cr(VI), I, and Se. Filters containing the GF material sorbed essentially all Pb and Ag from 5-mL aliquots of solutions containing 0.054 and 0.093 mM, respectively. Somewhat less Ni was sorbed from a 0.099 mM solution. In all cases, much less metal, if any, was sorbed on the SFCA material.
A sorption experiment was conducted with five Ag(NO3) concentrations ranging from 0.011 to 2.32 mM to compare Ag concentrations in solutions after passing through SFCA filters (0.45 µm) and SFCA + GF filters (0.45 µm). Five-milliliter aliquots were passed through previously unused filters. No background salt was added. Figure 1 shows that the SFCA + GF filter retained all Ag from the two lowest solution concentrations and a significant fraction from the higher concentration solutions. The figure indicates the breakthrough concentration under these conditions, for the GF filter, was about 0.44 mM for a 5-mL sample, or 2.22 µmol (240 µg) of Ag. These filters contain about 0.052 g of GF material and therefore can sorb about 43 µmol of Ag per gram of GF material (4600 µg/g).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 1. Concentration of Ag solution passed through glass-fiber (GF) material compared with surfactant-free cellulose acetate (SFCA) material. Error bars are smaller than the symbols. A 1:1 line is provided for reference.
|
|
To determine the relationship between sorption on the GF material and concentration of a background salt, several experiments were conducted in which the starting concentrations of Ag(NO3) at 0.191 mM and NiCl2 at 0.083 mM were held constant while the concentration of KNO3 was varied from 0 to 50 mM. Figure 2 shows how Ag and Ni sorption on the GF material was highest at low salt concentrations and decreased with increasing KNO3 concentration until reaching a constant concentration of sorbed metal at approximately 10 mM KNO3.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2. Effect of increasing KNO3 concentration on Ag and Ni sorption on glass-fiber (GF) material. Error bars are shown for Ag; for Ni, they are smaller than the symbol.
|
|
It is important to note that cation sorption on GF material was greatest under solution conditions commonly encountered in natural waters. At very high pH (12.5) with relatively high concentrations of other ions (solutions from our grout experiments), there was little if any sorption on the filter material. For samples at low ionic strength and near-neutral pH (i.e., natural) the GF filter material sorbs significant quantities of cations from solution. Under conditions of low particle loading, it may be possible to precondition the filter by passing some of the target solution through the filter before collecting the sample. Our results, however, show that GF materials should only be used with careful testing for the elements of concern under conditions that closely match expected environmental conditions.
|
ACKNOWLEDGMENTS
|
|---|
This work was funded by U.S. Department of Energy, West Valley Demonstration Project, Dan Westcott, project manager. The authors thank John Heiser and Jay Adams for their help and observations in the lab.
|
REFERENCES
|
|---|
- Goyne, K.W., R.L. Day, and J. Chorover. 2000. Artifacts caused by collection of soil solutions with passive capillary samplers. Soil Sci. Soc. Am. J. 64:1330–1336.[Abstract/Free Full Text]
Related articles in JEQ:
- This Issue in Journal of Environmental Quality
JEQ 2004 33: 1589-1599.
[Full Text]
TOP
ABSTRACT
REFERENCES
