Dicerium tricarbonate

Additional information on environmental fate and behaviour

Administrative data

Endpoint:

additional information on environmental fate and behaviour

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: This publication fully describes the experimental conditions and results but there is no refence to the international guideline nor the GLP references.

Data source

Reference

Reference Type:

publication

Title:

NEW INSIGHTS INTO CERIUM ANOMALIES IN ORGANIC RICH ALKALINE WATERS

Author:

Pourret O.

Year:

2008

Bibliographic source:

Chemical Geology, Volume 251, Issues 1-4, Pages 120-127

Materials and methods

Test guideline

Qualifier:

no guideline followed

Principles of method if other than guideline:

The cerium anomaly mechanism, which is the phenomenon whereby Cerium (Ce) concentration is either depleted or enriched relative to the other rare earth elements (REEs), was experimentally elucidated by studying the competition between carbonate and humic acids for complexing rare earth elements (REE)

GLP compliance:

no

Test material

Test material information

Constituent 1

Reference substance name:

lanthanides nitrate

IUPAC Name:

lanthanides nitrate

Details on test material:

All chemicals used were of analytical grade, and all the experimental solutions were prepared with doubly deionized water. Synthetic REE solutions were prepared from nitrate REE standards (10 mg L-1, Accu Trace™ Reference Standard).

Results and discussion

Any other information on results incl. tables

 

The proportion of Rare Earth Element (REE) complexed to carbonates as a function of pH for various alkalinity conditions (10-2 mol/L ; 5*10-3 mol/L and 10-3 mol/L) and the proportion of REE complexed to Humic Acid as a function of the same parameters are presented by the author in the figure 1. The figure clearly shows the positve or negative anomaly of the cerium in comparaison of the other 13 rare earth:
- 92 % or more of cerium is complexed to the humic acid for each experimental condition
- by contrast only less than 8 % (for alkalinity of 10-2 and 5*10-3 mol/L) and only 0.5 % (for alkalinity of 10-3 mol/L) are present uin the test under a carbonate form.

The cerium anomaly (Ce/Ce*) as a function pf pH, carbonate concentration or ionic strenght are also showed on graphics  

presented in the figure 2

The role of HA competing with carbonate on Ce anomaly development in alkaline-rich waters was investigated using a standard batch equilibration technique. 100 mL of solution were prepared with 50 μg L-1 of each REE (e.g., 360 nmol L-1 La to 286 nmol L-1 Lu), 5 mg L-1 of HA and various concentrations of NaHCO3 (from 10-3 to 10-2 mol L-1) in a 10-3 mol L-1 NaCl solution. Prior to addition of NaHCO3, the pH of the solution was 4.0. The initial hydroxide concentration was thus low (ca. 10-10 mol L-1), leading to negligible concentration of LnOH2+. After addition of NaHCO3, the pH increased to 5.0 and was then adjusted to the needed pH in the range of 6.6 to 10.6 by adding NaOH (diluted from a stock at 4 mol L-1). Under these experimental conditions, inorganic speciation calculations have shown that hydroxide and free species are only present at low concentrations (i.e., <0.2% and <3.8% of the inorganic fraction for La after the addition of NaOH to bring the pH to the range of 6.6 to 10.6, respectively. The pH was measured with a combined Radiometer Red Rod electrode. The electrode was calibrated with standard solutions (WTW™ pH 4.0, 7.0 and 10.0). The accuracy of the pH measurement was ±0.05 pH units. The redox potential was measured with a Mettler combinated Pt electrode (Eh varies between 350 and 450 mV). Ionic strength was thus different between samples, varying from 1.33 x 10-3 mol L-1 to 2.58 x 10-3 mol L-1, 2.88 x 10-3 mol L-1 to 8.83 x 10-3 mol L-1 and 5.15 x 10-3 mol L-1 to 1.62 x 10-2 mol L-1, respectively for the three increasing alkalinity conditions (i.e., 10-3 mol L-1, 5 x 10-3 mol L-1 and 10-2 mol L-1). Experimental solutions were stirred for 48 h (equilibrium time determined from preliminary kinetic experiments) to allow equilibration and partitioning of REE between the aqueous solution and the humate suspension. Aliquots of 10 mL were sampled twice: (i) at the beginning of the experiment; and (ii) after 48 h, at equilibrium state. REE complexed by the HA were separated from the remaining inorganic REE by ultrafiltration. Ultrafiltrations were carried out by centrifugating the 10 mL solution samples through 15 mL centrifugal tubes equipped with permeable membranes of 5 kDa pore size (Millipore Amicon Ultra-15). All centrifugal filtering devices used were washed and rinsed with 0.1 mol L-1 HCl and MilliQ water twice before use in order to minimize contamination. Centrifugations were performed using a Jouan G4.12 centrifuge with swinging bucket rotor at 3000 g for 30 minutes. This allowed the REE-HA complexes to be quantitatively separated from inorganic REE species, notably the carbonate complexes. The selectivity of the 5 kDa membrane with respect to the REE-HA complexes was verified by monitoring the Dissolved Organic Carbon (DOC) contents of the ultrafiltrates.

 

Results showed that the latter were systematically lower or equal to blank values (below 0.1 mg L-1). Possible adsorption of inorganic REE species onto ultrafiltration membranes or onto cell walls was also monitored. The lack of REE adsorption onto the ultrafiltration membranes or the walls of the cell devices used was checked in two ways. First, inorganic REE solutions of given REE concentration were ultrafiltered several times. Results showed that between 98.91% (for Ho) and 99.98% (for Yb) of the REE present in solution was recovered in the ultrafiltrates, demonstrating that none of the REE was adsorbed either, on the membrane or, on the walls of the cell devices. Secondly, in order to check that no retention of REE or carbonate occurs inside the membrane during ultrafiltration, mass balance calculations were performed. The initial concentration of each element was compared with the sum of each element concentration in the ultrafiltrate and the retentate. The element concentration sum measured from the ultrafiltrate and the retentate differed from the initial content by less than 2% for REE and by less than 5% for dissolved carbonate and DOC. Moreover, in order to ensure that no precipitation occurred (e.g., as cerianite), samples were filtered through 0.2 μm cellulose acetate membranes (Sartorius) before ultrafiltration. No difference was found (within the uncertainty of the measurements) between the 0.2 μm filtrate and the raw sample for REE, HA and carbonate.

 

The amount of REE complexed with HA corresponds to the difference between the initial REE concentration and the remaining REE concentration into the <5 kDa ultrafiltrates. REE concentrations were determined at Rennes University using an Agilent TechnologiesTM HP4500 ICP-MS instrument. Quantitative analyses were performed using a conventional external calibration procedure. Three external standard solutions with REE concentrations similar to the analyzed samples were prepared from a multi-REE standard solution (Accu TraceTM Reference, 10 mg L-1, USA). Indium was added to all samples as an internal standard at a concentration of 0.87 μmol L-1 (100 μg L-1) to correct for instrumental drift and possible matrix effects. Indium was also added to the external standard solutions. Calibration curves were calculated from measured REE/indium intensity ratios. The instrumental error on REE analysis in our laboratory as established from repeated analyses of multi-REE standard solution (Accu TraceTM Reference, USA) and of the SLRS-4 water standard is below ±2%.

 

Chemical blanks of individual REE were all lower than the detection limit (1 ng L-1), which is negligible since they are three to four orders of magnitude lower than the concentrations measured in the synthetic solutions used for the complexation experiments. Organic sample aliquots were all immediately digested with sub-boiled nitric acid (HNO3 14 mol L-1) at 100°C and then resolubilized in 0.4 mol L-1 HNO3 after complete evaporation in order to avoid any matrix effect due to organic matter during mass analysis. DOC concentrations were determined at Rennes University using a Shimadzu 5000 TOC analyzer. The accuracy of DOC concentration measurements is estimated at ±5% as determined by repeated analyses of freshly prepared standard solutions (potassium biphtalate). Carbonate concentrations were determined by potentiometric titrations (HCl 0.1 mol L-1), with Gran method analysis. The uncertainty was better than 5%.

The role of HA and carbonate competition on Ce oxidation and Ce anomaly development is examined by considering the REE patterns of the organic and inorganic fractions as a function of pH. Inorganic REE speciation can be modelled using WHAM 6 (Tipping, 1998) in which well accepted infinite dilution (25°C) stability constants for REE carbonate complexes are included (Luo and Byrne, 2004). REE were shown to be mostly carried as carbonate complexes. REE carbonate complexation is described by the following equations:

Ln3+ + CO3 2- ⇋ LnCO3 + (1)

Ln3+ + 2 CO3 2- ⇋ Ln(CO3)2 - (2)

whereas humic complexation is described by the following equation:

Ln3+ + HA- ⇋ LnHA2+ (3)

As shown by several studies (e.g., Luo and Byrne, 2004), LnCO3+ and Ln(CO3)2- concentrations strongly depend on pH and carbonate content. LnCO3+ concentrations decrease with pH increase above ca. 7, whereas Ln(CO3)2 - concentrations increase with increasing pH. Ln(CO3)2- and LnCO3+ concentrations thus increase with carbonate alkalinity (Fig. 1). Moreover, no ternary surface complex between metal (Ln3+), ligand (CO32-) and surface (HA) was considered.

Negative Ce anomalies are clearly developed at high pH in the ultrafiltrates, which is mirrored by a positive Ce anomaly in the HA colloids fraction (Fig. 1). The progressive development of a Ce anomaly is accompanied by increased fractionation of the REE patterns.

Ultrafiltrates become progressively enriched in heavy REE (HREE) with increasing pH and alkalinity, whereas the fraction corresponding to HA complexes show progressive Light REE (LREE) enrichment. The progressive LREE enrichment in the HA fraction and thus the corresponding progressive HREE enrichment in the ultrafiltrates correspond to the competitive reaction between

mono- (Eq. [1]) and di- (Eq. [2]) carbonato-complexation reactions. This result is consistent with the difference in the complexation constants of LREE- and HREE-carbonate complexes (Luo and Byrne, 2004). The influence of redox processes on Ce/REE fractionation can be evaluated by calculating the magnitude of Ce anomalies as follows:

Ce/Ce*= Ce/(Pr+(Pr-Nd))

Equation 4 was selected to avoid any La anomaly interference during Ce/Ce* calculation. In this study, relative values of Ce, Pr and Nd are used rather than normalized Ce, Pr and Nd values, because in our experimental study we can normalize to initial concentrations. Ce anomaly values greater than or less than one represent preferential enrichment or removal of Ce, respectively. The development of a negative Ce anomaly is observed for pH values greater than 8.2, 8.6 and 8.7, depending on alkalinity values (Fig. 2). The magnitude of the negative Ce anomaly in solution increases with pH from 1.00 to 0.05 (Fig. 2a). The development of a

negative Ce anomaly in solution is observed for various carbonate concentrations. Negative Ce anomalies develop for CO3

2- concentrations above 10-4 mol L-1 (Fig. 2b). Moreover, the development of the negative Ce anomaly is not related to ionic strength (Fig. 2c); Ce anomaly increases whereas ionic strength remains in the same order of magnitude (i.e., between 1.33 x 10-

3 mol L-1 and 2.58 x 10-3 mol L-1, 2.88 x 10-3 mol L-1 and 8.83 x 10-3 mol L-1 and 5.15 x 10-3 mol L-1 and 1.62 x 10-2 mol L-1, respectively for the three increasing alkalinity conditions).

Complementary positive Ce anomalies develop in the HA fraction. However, positive Ce anomalies are only significant (up to 1.22) for the more competing condition (i.e., for alkalinity of 10-2 mol L-1).

.

Applicant's summary and conclusion

Conclusions:

The role of HA competing with carbonate on Ce anomaly development in alkaline-rich waters was investigated using a standard batch equilibration technique. This experimental method combines an ultrafiltration technique and Inductively Coupled Plasma Mass Spectrometry. The role of HA and carbonate competition on Ce oxidation and Ce anomaly development was examined by considering the REE patterns of the organic and inorganic fractions as a function of pH. A negative Ce anomaly is developed in the REE fraction bound to carbonate at pH above 8.2, 8.6 and 8.7 at alkalinities of 10-3 mol L-1, 5 x 10-3 mol L-1 and 10-2 mol L-1, respectively, whereas a positive Ce anomaly is developed in the organic fraction (i.e., >5 kDa). Partitioning is observed between the organic phase (LnHA) and inorganic phases (Ln-carbonate). In the inorganic phase, REE patterns display HREE enrichment typical of seawater. These experiments shed more light, not only on a new way to develop a Ce anomaly, but also on the understanding of cerium anomaly cycle in natural waters at alkaline pH. Indeed, these results suggest a new mechanism for cerium anomaly development. In the presence of carbonate, cerium is readily oxidized to Ce(IV), which is easily removed from solution by preferential adsorption to HA. Humic substances take up Ce(IV) from the "truly" dissolved part of solution (i.e., <5 kDa), and a negative cerium anomaly thus develops in the inorganic fraction. A complementary positive Ce anomaly develops in the organic fraction. The preferential Ce(IV) sorption is masked. Therefore, Ce anomalies can not reliably be used as a proxy of redox conditions in unfiltered samples of organic-rich waters or in precipitates formed in equilibrium with organic-rich waters. Overall, the results of this study suggest that further considerations about organic matter should be taken into account especially at alkaline pH in organic-rich water by performing ultrafiltration on these samples.

Executive summary:

This study demonstrates the development of a negative cerium anomaly in the inorganic fraction of a carbonate-HA suspension at alkaline pH and high [CO3 2-], whereas the development of a positive cerium anomaly is observed in the humic phase. Oxidation of Ce(III) to Ce(IV) leads to the development of cerium anomalies compared with the adjacent REE, which are strictly trivalent.

The author suggests the following chemical equation to explain the behavior of the cerium carbonate in environment:

4 Ce(III)(CO3)2- + 12 (CO3)2- + O2(g) + 4 H+ ⇋ 4 (Ce(IV)(CO3)5)6- + 2 H2O

In order to check if such a mechanism (i.e., Ce(III) oxidation) could occur in the experimental conditions encountered in the study, two pe-pH diagrams for the system Ce-C-O-H (Fig. 3a and 3b) were built using this reaction for a range of ionic strength (I) relevant to the experimental conditions. Ce does not form CeO2 but Ce(IV)(CO3)5 6-, under the conditions prevailing in the experiments (illustrated by black arrows on Fig. 3). The author assumed that a(Ce(III)(CO3)2-)/a(Ce(IV)(CO3)5 6-) is greater than a(Ce(III)(CO3)2-)/a(Ce(IV)O2), Ce(IV)(CO3)5 6- will not be broken even if the CeO2 boundary is crosses with a further increase in pH. As illustrated by Möller and Bau (1993), stabilization in solution of pentacarbonato-Ce(IV)-complex led to the development of a positive Ce anomaly in alkaline carbonate-rich solutions.