Ionic Liquids for SimultaneousPreconcentration of SomeLanthanoids Using DispersiveLiquid-Liquid MicroextractionTechnique in Uranium DioxidePowderM O H A M M A D H . M A L L A H , † , ‡

F A R Z A N E H S H E M I R A N I , * , † A N DM O H A M M A D G . M A R A G H E H ‡

Department of Analytical Chemistry, Faculty of Chemistry,Group of Science, University of Tehran, P.O. Box 14155-64555,Tehran, Iran, and Jaber Ibn Hayan Research Laboratories,Nuclear Fuel Cycle Research School, Nuclear Science &Technology Research Institute, Atomic Energy Organization ofIran, End of North Karegar Ave. P.O. Box 14395-836, Tehran, Iran

Received October 29, 2008. Revised manuscript receivedDecember 23, 2008. Accepted January 14, 2009.

Ionic liquids in a dispersive liquid-liquid microextractiontechnique were used for determination of lanthanoids such assamarium, europium, gadolinium, and dysprosium in uraniumdioxide powder. In this process, an appropriate mixture ofextraction solvent and disperser is rapidly injected into anaqueous sample containing samarium, europium, gadolinium,and dysprosium ions complexes with 1-hydroxy-2, 5-pyrro-lidinedione, and consequently a cloudy solution is formed. Itconsists of fine droplets of extraction solvent which are dispersedentirely into the aqueous phase. After centrifugation of thissolution, the whole enriched phase was determined by inductivelycoupled plasma optical emission spectrometry. In the presentwork, the preconcentration factor, limit of detection, andrelative standard deviation were investigated for samarium,europium, gadolinium, and dysprosium in uranium dioxide powder.

Introduction

The continued use of organic solvents as liquid media forchemical reaction, extraction, and formulation is a majorconcern in today’s chemical processing industry. The per-ceived deleterious effects of these materials on human health,safety, and the environment combined with their volatilityand flammability has to increase pressure for minimizingtheir use from both a public relations and a cost perspective.Ionic liquids as new solvents are recently interesting for manyresearchers. Some of the properties which make the ionicliquids attractive as media for various applications are thewide liquid range, nonvolatility (negligible vapor pressure),nonflammable nature, lower reactivity, and strong ability todissolve a large variety of organic and inorganic substancesincluding even polymer materials in high concentration.Many of these properties have made ionic liquids a nature-

friendly “green solvent” that can be applied in liquid-liquidsolvent extraction processes for the separation of metal ions(1-6).

In recent years, the lanthanoids, often called rare earthelements (REE), have widely been used in high technologyapplications, such as in superconductors, lasers, and otherproducts in industry. The amounts of the used lanthanoidshave increased considerably in the modern societies. Con-sequently, the emission of lanthanoids into the environmenthas also been increasing. These circumstances have resultedin an increase in our exposure to lanthanoids and an increasein our dietary intake of lanthanoids (7). Under thesecircumstances, the concentration of lanthanides may beincreasing even in natural waters; hence, monitoring tech-niques for lanthanoids such as samarium (Sm), europium(Eu), gadolinium (Gd), and dysprosium (Dy) are required forenvironmental protection. Furthermore, the analysis andmeasurement of trace lanthanoids has been increased inorder to elucidate their biological functions and toxicities(8, 9). The quantification of these lanthanoids is extremelydifficult because they are a group of closely related elementshaving similar physical and chemical properties. Neutronactivation analysis (NAA), inductively coupled plasma massspectrometry (ICP-MS), X-ray fluorescence (XRF), spectro-photometers with arsenazo III and II, and inductively coupledplasma optical emission spectrometers (ICP-OES) are tech-niques allowing the individual determination of these lan-thanoids with good sensitivity by applying a sample prepa-ration method as a prior step to its determination (10-13).Recently, these methods have been extended to include noveland powerful supercritical fluid (SCF) extraction, SCFchromatographic, and new combined dispersive liquid-liquidmicroextraction (DLLME) method with ICP-OES. In thismethod an appropriate mixture of extraction and dispersersolvent is injected into the aqueous sample, which forms acloudy solution. The cloudy state results from the formationof fine droplets of the extraction solvent which are dispersedin the sample solution. The cloudy solution can be centri-fuged, and fine droplets are sedimented at the bottom of aconical test tube. Determination of analytes in the remainingphase can be performed by instrumental techniques. In thisextraction method, every component in the solution, directlyor indirectly after derivatization, interacts with the finedroplets of the extraction solvent and is consequentlyextracted from the initial solution and concentrated to a smallvolume in the remained phase. The performance of DLLMEwas illustrated with the determination of polycyclic aromatichydrocarbons (PAHS), organphosphorus pesticides, chlo-robenzenes, trihalomethanes, chlorophenols, and metals ionsin aqueous samples (14-21).

The authors intended to eliminate extraction organicsolvent in DLLME to enhance the level of environmentalprotection and equivalent boron content calculation of Sm,Eu, Gd, and Dy (Figure 1) using ICP-OES. Therefore, in thismethod ionic liquids were used instead of the organic solvent,using the 1-hydroxy-2, 5-pyrrolidinedione (HYD) ligandwhich showed a significant increase in the preconcentrationfactor of Sm, Gd, and Dy ions, not Eu, in comparison withorganic solvents.

Experimental SectionIonic liquids such as 1-butyl-3-methylimidazolium hexafluo-rophosphate (1-B) and 1-hexyl-3-methylimidazolium hexaflu-orophosphate (1-H), methanol (for spectroscopy), HNO3

(65%, suprapur), uranium dioxide, sodium dodecyl sulfate

* Corresponding author fax: +982188221113; e-mail: shemiran@khayam.ut.ac.ir.

† University of Tehran.‡ Atomic Energy Organization of Iran.

Environ. Sci. Technol. 2009, 43, 1947–1951

10.1021/es8030566 CCC: $40.75 2009 American Chemical Society VOL. 43, NO. 6, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1947

Published on Web 02/12/2009

(SDS), lanthanoids (in oxide form) and 1-hydroxy-2, 5-pyr-rolidinedione (HYD) (analytical grade), di-natriumtetrabo-rate-10-hydrate, Cadmium-acetate-2-hydrate were pur-chased from Merck Co. (Germany). The stock solutions ofsamarium, europium, gadolinium, and dysprosium wereprepared at a concentration of 5 × 10-2 mol/L of ions bydissolving an appropriate amount of Sm, Gd, and Dy oxidesin 20 mL of 12.8 mol/L HNO3 solution, and the solutionswere then evaporated carefully at low temperature (not above60 °C) to dry. The residues were dissolved in 26.4 mL of 7.6mol/L nitric acid solution and diluted to 200 mL. All solutionswere prepared using double-distilled water.

An aqueous ammonium solution (1% w/w) was used foradjusting the pH of a 10-3 mmol/mL solution of 1-hydroxy-2, 5-pyrolidinedione salt. All vessels used for trace analysiswere kept in a 1 mol/L HNO3 solution at least 24 h andsubsequently washed twice with double-distilled water beforeusing.

A simultaneous inductively coupled plasma optical emis-sion spectrometer (Optima 2100 DV) equipped with aminitorch and equipped with a segmented-array charge-coupled device detector and peristaltic pump was used. Theoperating conditions and analytical wavelength are sum-marized in Table 1. The pH values were measured with aSchoct pH-meter (CG 841) supplied with a glass-combinedelectrode. Phase separation was assisted using a centrifuge (Mistral 1000, MSB 100/CE 1.4). Meanwhile, a furnace

(EHRET) was used for drying the precipitate phase.An 80 mL solution containing Sm, Eu, Gd, and Dy in equal

amounts at a concentration of 5 × 10-5 mol/L ion, and 1mol/L HNO3, the pH being maintained by an aqueousammonium solution (0.1mol/L) and SDS 0.02 w/v, was addedto a 100 mL screw-cap glass test tube with a conical bottom.Eight milliliters of methanol as disperser solvent, whichcontains 600 µL of ionic liquid as extraction solvent (Ex) and1 mL solution of HYD 1 mmol/L as chelating agent, wasrapidly injected into a sample solution. A cloudy solutionwas formed in the test tube. In this step, Sm, Eu, Gd, and Dyions reacted with chelating reagent and were extracted intothe fine droplets of the ionic liquid. The mixture was thencentrifuged (Cen) for 6 min at 3500 rpm. After this process,the dispersed fine droplets of ionic liquid were precipitatedat the bottom of the conical test tube (0.5 mL). This precipitate

FIGURE 1. Chemical structure of 1-hydroxy-2, 5-pyrrolidinedione(HYD).

TABLE 1. Instrumental Parameters of ICP-OES

parameters values

RF generator power (W) 1100plasma gas flow rate (L min-1) 15auxiliary gas flow rate (L min-1) 0.8nebulizer pressure (kPa) 200torch mode (minitorch) axialpump flow rate (mL min-1) 1wavelength of samarium (nm) 359.160wavelength of europium (nm) 381.967wavelength of gadolinium (nm) 376.838wavelength of dysprosium (nm) 353.170

FIGURE 2. Relationship between preconcentration factor andpH (Cen: 3500 rpm, methanol: 8 mL, Ex: 1-hexyl-3-methylimid-azolium hexafluorophosphate (600 µL), t: 6 min, HYD: 1 mL, T:160 ((5) °C, surfactant: 0.02% w/v SDS).

FIGURE 3. Relationship between preconcentration factor andextraction solvent type (Cen: 3500 rpm, methanol: 8 mL, Ex: 600µL, t: 6 min, pH: 9.5, HYD: 1 mL, T: 160 ((5) °C, surfactant: 0.02%w/v SDS).

FIGURE 4. Relationship between preconcentration factor andextraction solvent volume (Cen: 3500 rpm, methanol:8 mL, Ex:1-hexyl-3-methylimidazolium hexafluorophosphate, t: 6 min, pH:9.5, HYD: 1 mL, T: 160 ((5) °C, surfactant: 0.02% w/v SDS).

FIGURE 5. Relationship between preconcentration factor andligand volume of HYD with 10-3 mol L-1 (Cen: 3500 rpm, methanol:8 mL, Ex: 1-hexyl-3-methylimidazolium hexafluorophos-phate (600 µL), t: 6 min, pH: 9.5, T: 160 ((5) °C, surfactant: 0.02% w/v SDS).

1948 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 6, 2009

phase which is usually mixed with drops of dispersive solventor an excess amount of HYD was dried by a furnace in 160((5) °C. Then it was diluted to 0.5 mL by adding solution 1mol/L HNO3. The sample was introduced into the inductivelycoupled plasma optical emission spectrometer by a peristalticpump. This procedure was repeated for uranium dioxideusing the same method.

Results and DiscussionPreconcentration factor was calculated using the followingequation:

PF)Cpp/C0

where PF, Cpp, and C0 are respectively preconcentration factor,concentration of the analyte in the precipitate phase, andinitial concentration of the analyte in the aqueous sample.They were determined by ICP-OES. The analyte concentrationin the precipitate phase was defined from the direct calibra-tion curve.

This study considered the application of ionic liquids ina dispersive liquid-liquid microextraction technique with ahigher preconcentration factor rather than with organicsolvents. In addition to the same effects, some other factorswere reported in a previous paper, such as the type ofdispersive solvent and volume, the extraction time, and thecentrifuge speed (22). Nevertheless, there are some otherfactors which can affect the simultaneous preconcentrationof samarium, europium, gadolinium, and dysprosium in thedispersive liquid-liquid microextraction technique whenionic liquids solvents are used as an extraction solvent anda new ligand of HYD. Therefore, the operating conditionsshould be optimized. For this reason, the one-variable-at-a-time optimization was used.

Figure 2 shows the effect of pH on the extraction of Sm,Eu, Gd, and Dy complexes. It can be seen that the precon-centration factor increased with the increase of pH up to 9.5.At lower pH values, the ligand is protonated. The precon-centration of metal ions by dispersive liquid-liquid mi-croextraction involved a prior formation of a complex withsufficient hydrophobicity to be extracted into the smallvolume of the precipitate phase; thus, for obtaining thedesired preconcentration, pH played a unique role inmetal-ligand formation and subsequent extraction of Sm,Eu, Gd, and Dy ions. This parameter was fixed by aqueousammonium solution (0.1 mol/L).

The type of extraction solvent used in DLLME is anessential consideration for efficient simultaneous precon-centration. Generally, there are two requirements duringselection of extraction solvent. First, it should have higherdensity than water. Second, it should have a low solubilityin water and be nonvolatile to prevent solvent loss during

extraction. For this purpose, the performance of two kindsof ionic liquid (1-butyl-3-methylimidazolium hexafluoro-phosphate and 1-hexyl-3-methylimidazolium hexafluoro-phosphate) was compared to an organic solvent (carbontetrachloride). A series of sample solutions was studied using8 mL of methanol containing 1 mL of HYD and differenttypes and volumes of the extraction solvent to achieve 0.5mL of the precipitate phase. The solubility of the extractionsolvents in water was different. Therefore, for obtaining 0.5mL of the precipitate phase at the bottom of the test tube,it was necessary to add an excess amount to account thesolubility. According to Figures 3 and 4, 600 µL of 1-hexyl-3-methylimidazolium hexafluorophosphate showed the bestresult. It can be found that the ionic liquids with a longerchain improved the preconcentration factor in the DLLMEtechnique. It also depends on the cation type in the ionicliquids.

The effect of the chelating agent volume (10-3 mol/L) ofHYD on preconcentration factor of Sm, Eu, Gd, and Dy ionswas also studied. As can be seen in Figure 5, the precon-centration factor of Sm, Eu, Gd, and Dy ions increased upto a known volume of HYD and then decreased. Therefore,a volume of 1 mL was chosen for the subsequent experiments.

After the centrifugation step in DLLME, a precipitate ofionic liquid solvents was observed on the test tube wallswhich caused a loss of repeatability and reduced precisionin the measurements. To solve this problem, a surfactantwas used because this material surrounds ionic liquid finedroplets and reduces its adhesiveness. The preconcentrationfactors of Sm, Eu, Gd, and Dy ions as a function of theconcentration of three kinds of the surfactants, that is, TritonX-114, Triton X-11, and SDS, in the range of 0-0.09% (w/v)were investigated. The results showed that SDS was the mosteffective surfactant. As can be seen in Figure 6, the precon-

FIGURE 6. Relationship between preconcentration factor andsurfactant concentration (Cen: 3500 rpm, methanol: 8 mL, Ex:1-hexyl-3-methylimidazolium hexafluorophosphate (600 µL), t: 6min, pH: 9.5, HYD: 1 mL, T: 160 ((5) °C).

FIGURE 7. Relationship between preconcentration factor andsalt concentration (Cen:3500 rpm, methanol: 8 mL, Ex: 1-hexyl-3-methylimidazolium hexafluorophosphate (600 µL), t: 6 min, pH:9.5, HYD: 1 mL, T: 160 ((5) °C, surfactant: 0.02% w/v SDS).

FIGURE 8. Relationship between preconcentration factor anddrying temperatures of the samples (Cen: 3500 rpm, methanol: 8mL, Ex: 1-hexyl-3-methylimidazolium hexafluorophosphate (600µL), t: 6 min, pH: 9.5, HYD: 1 mL, surfactant: 0.02% w/v SDS).

VOL. 43, NO. 6, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1949

centration factor slightly increased with an increase in theconcentration of the surfactant and reached a maximum ata concentration of 0.02% (w/v) SDS due to the enhancementof extraction and then remained almost constant over theselected range. This value was employed as the optimumconcentration for the subsequent experiments.

Salt addition is frequently used to adjust the ionic strength.It improves the extraction efficiency and reduces the limitof detection. The effect of the ionic strength on the pre-concentration factor was tested in the concentration rangeof 0-0.5 mol/L NaNO3. As observed in Figure 7, the saltaddition did not have a significant effect on the precon-centration factor of Sm, Eu, Gd, and Dy ions. On this basis,all the extraction experiments were carried out without saltaddition.

According to Figure 8, the preconcentration factor in-creased with an increase in the drying temperature of thesamples and reached a constant value which was selectedas drying temperature of the samples (up to 150 °C). At thistemperature, the volume of ionic liquid solvent in samplesdecreased and its adverse effects on the instrument mea-surement was also eliminated.

Also, the effect of various ions on the preconcentrationfactor of Sm, Eu, Gd, and Dy ions has been investigated.Among the ions tested B3+ and Cd2+ were important in thatthey did not show interference at the concentration 50 timeshigher than that of the Sm, Eu, Gd, and Dy concentration.

A calibration curve was obtained by simultaneous pre-concentration of 80 mL of sample standard solutionscontaining known amounts of Sm, Eu, Gd, and Dy ions underoptimum conditions. The preconcentration factor is definedas the ratio of the calibration curves slopes with or withoutpreconcentration. To confirm the optimum method, theprocess evaluation was carried out by establishing the basicanalytical requirements of the performance which wasappropriate for quantitative determination of Sm, Eu, Gd,and Dy in aqueous samples. The limit of detection (LOD)was calculated as the lowest concentration required forproducing a signal which is three times higher than thestandard deviation of a matrix blank signal. Eight replicatedanalyses of the aqueous sample were previously chosenwithout any Sm, Eu, Gd, and Dy, which were used forestimating the matrix blank signal standard deviation. Inaddition, eight replicated measurements of a sample con-taining of 200 µg L-1, 20 µg L-1, 100 µg L-1, and 50 µg L-1 Sm,Eu, Gd, and Dy were used for the calculation of relativestandard deviation (RSD).

The optimum conditions of extraction are listed in Table2. Furthermore, the obtained values of preconcentrationfactor, LOD, and RSD for Sm, Eu, Gd, and Dy ions by ionicliquid solvent of 1-hexyl-3-methylimidazolium hexafluoro-phosphate with HYD reagent are presented in Table 3.

This study presents the DLLME technique that resultsunder the best operating conditions. ICP-OES results showthat the best operating condition for DLLME was suitable forthe simultaneous preconcentration of Sm, Eu, Gd, and Dywith ionic liquid solvents using a HYD reagent. This resultsuggests that the hexafluorophosphate anions with a partialion exchange mechanism in ionic liquid solvents play a keyrole in the electrical neutralization of the lanthanoid complex

in DLLME. This is the first report to successfully demonstrateapplication of ionic liquids in a DLLME method for deter-mination of Sm, Eu, Gd, and Dy in uranium dioxide powders.Therefore, it can be referred as a new advanced techniquefor practical determination of equivalent boron contents ofnuclear materials.

Literature Cited(1) Gordon, C. M.; Holbrey, J. D.; Kennedy, A. R.; Seddon, K. R.

Ionic liquids crystals: hexafluorophosphate salts. J. Mater. Chem.1998, 8, 2627–2636.

(2) Armstrong, D. W.; He, L. F.; Liu, Y. S. Examination of ionic liquidsand their interaction with molecules, when used as stationaryphases in gas chromatography. Anal. Chem. 1999, 71, 3873–3876.

(3) Sun, J.; MacFarlane, D. R.; Forsyth, M. A new family of ionicliquids based on the 1-alkyl-2-methyl pyrrolinium cation.Electrochim. Acta 2003, 48 (12), 1707–1711.

(4) Holbrey, J. D.; Seddon, K. R. Ionic liquid. Clean Technol. Environ.Policy 1999, 1, 223–236.

(5) Liu, F.; Jiang, G. B.; Chi, Y. G.; Cai, Y. Q.; Hu, J. T.; Zhou, Q. X.Use of ionic liquids for liquid-phase microextraction of poly-cyclic aromatic hydrocarbons. Anal. Chem. 2003, 75, 5870–5876.

(6) Binnemans, K. Lanthanides and actinides in ionic liquids. Chem.Rev. 2007, 107, 2592–2614.

(7) Vela, N. P.; Olson, L. K.; Caruso, J. A. Elemental speciation withplasma mass spectrometry. Anal. Chem. 1993, 65 (13), 585A–597A.

(8) Vandecasteele, C.; Block, C. B. Modern methods for trace elementdetermination; Wiley: Chichester, 1993.

(9) Sawatari, H.; Toda, T.; Saizuka, T.; Kimata, C.; Itoh, A.; Haraguchi,H. Multielement determination of rare earth elements in coastalseawater by inductively coupled plasma mass spectrometry afterpreconcentration using chelating resin. Bull. Chem. Soc. Jpn.1995, 68, 3065–3070.

(10) Doherty, W. An internal standardization procedure for thedetermination of yttrium and the rare-earth elements ingeological materials by inductively coupled plasma massspectrometry. J. Spectrochim. Acta, Part B 1989, 44, 263–280.

(11) Djingova, R.; Ivanova, J. Determination of rare earth elementsin soils and sediments by inductively coupled plasma atomicemission spectrometry after cation-exchange separation. Ta-lanta 2002, 57 (5), 821–829.

(12) Smet, T.; Roelandts, I. Radiochemical neutron activation traceelement analysis of two USGS ultrabasic rock reference samples:peridotite PCC-I and dunite DTS-I. Geostand. Newsl. 1978, 2,61–70.

(13) Robinson, P.; Higgins, N. C.; Jenner, G. A. Determination ofrare-earth elements, yttrium and scandium in rocks by an ionexchange-X-ray fluorescence technique. Chem. Geol. 1986, 55,121–137.

(14) Rezaee, M.; Assadi, Y.; Milani Hosseini, M. R.; Aghaee, E.; Ahmadi,F.; Berijani, S. Determination of organic compounds in waterusing dispersive liquid-liquid microextraction. J. Chromatogr.,A 2006, 1116, 1–9.

(15) Berijani, S.; Assadi, Y.; Anbia, M.; Milani Hosseni, M. R.; Aghaee,E. Dispersive liquid-liquid microextraction combined with gas

TABLE 2. Optimum Operating Conditions

operatingconditions

type and amount ofextraction solvent

type and amount ofextraction solvent extraction time

volume of thechelating reagent

(HYD)centrifuge

speeddrying temperature

of the samples

results

1-hexyl-3-methylimidazoliumhexafluorophosphate(600 ( 0.001 µL) Methanol (8 ( 0.001 mL) 6 min 1 ( 0.001 mL 3500 rpm 160 ((5) °C

TABLE 3. Preconcentration Factor, LOD, and RSD

target ions PF LOD (µg L-1) RSD (%)

Sm 84.04 1.29 2Eu 19.34 0.66 0.8Gd 86.04 0.59 2Dy 67.92 0.34 1.5

1950 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 6, 2009

chormatography-flame photometric detection very simple, rapidand sensitive method for the determination of organophos-phorus pesticides in water. J. Chromatogr., A 2006, 1123, 1–9.

(16) Rahnama Kozani, R.; Assadi, Y.; Shemirani, F.; Milani Hosseni,M. R.; Jamali, M. R. Part-per-trillion determination of chlo-robenzenes in water using dispersive liquid-liquid microex-traction combined gas chromatography-electron capture de-tection. Talanta 2007, 72, 387–393.

(17) Rahnama Kozani, R.; Assadi, Y.; Shemirani, F.; Milani Hosseni,M. R.; Jamali, M. R. Determination of trihalomethanes indrinking water by dispersive liquid-liquid microextraction thengas chromatography with electron-capture detection. Chro-matographia 2007, 66, 81–86.

(18) Fattahi, N.; Assadi, Y.; Milani Hosseni, M. R.; Jahromi, E. Z.Determination of chlorophenols in water samples using si-multaneous dispersive liquid-liquid microextraction and de-rivatization followed by gas chromatography-electron-capturedetection. J. Chromatogr., A 2007, 1157, 23–29.

(19) Ahmadi, F.; Assadi, Y.; Hosseni, M. M. R.; Rezaee, M. Deter-mination of organophosphorus pesticides in water samples bysingle drop microextraction and gas chromatography-flamephotometric detector. J. Chromatogr., A 2006, 1101, 307–312.

(20) Jahromi, E. Z.; Bidari, A.; Assadi, Y.; Milani Hosseni, M. R.; Jamali,M. R. Dispersive liquid-liquid microextraction combined withgraphite furnace atomic absorption spectrometry. Ultra tracedetermination of cadmium in water samples. Anal. Chim. Acta2007, 585, 305–311.

(21) Shokoufi, N.; Shemirani, F.; Assadi, Y. Fiber optic-linear arraydetection spectrophotometry in combination with dispersiveliquid-liquid microextraction for simultaneous preconcentra-tion and determination of palladium and cobalt. Anal. Chim.Acta 2007, 597, 349–356.

(22) Mallah, M. H.; Shemirani, F.; Ghannadi Maragheh, M. Use ofdispersive liquid-liquid microextraction for simultaneous pre-concentration of samarium, europium, gadolinium and dys-prosium.J.Radioanal.Nucl.Chem.2008,doi:10.1007/s10967-007-7220-1.

ES8030566

VOL. 43, NO. 6, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 1951