Available online at www.sciencedirect.com
Procedia Chemistry 7 (2012) 703 – 708
ATALANTE 2012 International Conference on Nuclear Chemistry for Sustainable Fuel Cycles
Development and validation of methods for the analysis of reprocessing solvents: role of CETAMA Working Group 24 C. Riviera*, B. Camèsa, C. Lamourouxb, F. Casanovab, M. Daubizitc, C. Neulata, M. Organistad, P. Pochona, D. Roudila a
CEA/DEN/DRCP, Centre de Marcoule, BP 17171, 30207 Bagnols-sur-Ceze cedex, France b CEA/DEN/DPC, Centre de Saclay, 91191 Gif sur Yvette, France c AREVA NC, BP 170, 30206 Bagnols-sur-Ceze cedex, France d AREVA MINES/SEPA, BP 71, 87250 Bessines, France
Abstract An interlaboratory comparison has been organized by CETAMA Working Group 24 “Organic analysis” for the validation of a malonamide analysis method, DMDOHEMA, by Gas Chromatography coupled with a Flame Ionization Detector (GC-FID). This compound is studied as a new extraction solvent in nuclear reprocessing processes. Most of the results for DMDOHEMA showed an agreement between the laboratory values and the reference value better than 10%. Repeatability and reproducibility of the method were evaluated by robust statistics and detection limits were also estimated from laboratory data. The description of the method has been published in the CETAMA ANASOR collection book. ©2012 2012The TheAuthors. Authors.Published Publishedby byElsevier ElsevierB.V. B.V.Selection and /or peer-review under responsibility of the Chairman of the © Selection and/or peer-review under responsibility ofunder the Chairman of thelicense. ATALANTE 2012 Program ATALANTA 2012 Program Committee Open access CC BY-NC-ND Keywords: DMDOHEMA ; reprocessing solvent ; GC-FID ; method validation ; interlaboratory comparison
1. Introduction Since 1993 at CEA, the Working Group 24 “Analysis of organic compounds” of CETAMA (Commission for establishment of analytical methods) has been developing methods for the analysis of organic compounds used in * Corresponding author. Tel.: +33-4-66796665; fax: +33-4-66796980. E-mail address:
[email protected].
1876-6196 © 2012 The Authors. Published by Elsevier B.V. Selection and /or peer-review under responsibility of the Chairman of the ATALANTA 2012 Program Committee Open access under CC BY-NC-ND license. doi:10.1016/j.proche.2012.10.107
704
C. Rivier et al. / Procedia Chemistry 7 (2012) 703 – 708
the nuclear spent fuel reprocessing industry. Until today, most of the compounds studied were TriButylPhosphate and its main degradation products and 13 different analytical methods have been published in the CETAMA ANASOR collection book. In 2008, the Working Group decided to extend the scope of the ANASOR collection book to extraction solvents used in new liquid-liquid partitioning processes for separating the minor actinides from spent fuel. One of the reference molecule is N,N’-DiMethyl-N,N’-DiOctyl-Hexyl-Ethoxy-Malonamide (DMDOHEMA). The degradation of this molecule under hydrolysis and radiolysis has been studied in detail by the Laboratory of Physico-Chemistry of Processes (LPCP) at CEA Marcoule using GC-FTIR and GC-MS techniques [1, 2]. The main degradation products identified in the organic solution after radiolysis or hydrolysis in presence of nitric acid aqueous phase were an amidic-acid (1), a monoamide (3), diamides (DMOHEMA (4) with a loss of octyl group and MDOHEMA (5) with a loss of methyl group), carboxylic acids and amines such as MethylOctyl Amine (2) (MOA) (Fig. 1). O Amidic Acid (1) (MOCHOBA)
DMOHEMA (4) O N
3
CH3
C8 H17
5 Alcohol
O
C H OC H 2 4 6 13
N
6
C8 H17
7
4
C6H13
3
O2 O 1 N
O
O
OH
CH N
C2H4
CH3
CH 8 17 N
O H
C8 H17
O
O 8
H N
MethylOctyl Amine (2) (MOA)
CH3
9
C8 H17 N CH3
N C2H4
O
H
CH 8 17 N
O
O C6H13
R
OH
MDOHEMA (5) Carboxylic acids (R = C3-C6)
C H OC H 3 6 6 13
CH
3 Monoamide (3) MOHOBA
Fig. 1. Simplified scheme for radiolytic or hydrolytic degradation of DMDOHEMA [1]
The degradation products modify the extraction and/or the hydrodynamic performances of the process. It is therefore necessary to accurately follow the composition of the organic phase in order to guarantee that the extraction process performances are maintained. The analytical methods used have to be fully validated and their performances have to be characterized. The objective was to develop and validate a method for the analysis, of DMDOHEMA, MOHOBA and MOCHOBA by gas chromatography coupled with a Flame Ionization Detector (GC-FID). A first protocol was proposed and an interlaboratory comparison, piloted by LPCP, was organized in 2009 for the validation of this new ANASOR method. Significant biases, till 35%, between the reference value and the laboratories values were observed and difficulties, due to the derivatization of the amidic acid, MOCHOBA, were encountered by the participants. These unsatisfactory results lead to the organization of a second interlaboratory comparison in 2011 whose results are presented in this article.
C. Rivier et al. / Procedia Chemistry 7 (2012) 703 – 708
2. Organization of the comparison 2.1. Preparation of the solutions The preparation of the solutions was done by LPCP by gravimetric dilution of pure compounds in ethyl acetate. DMDBHDEMA (N,N-DiMethyl N,N-DiButyl HexaDecylEthoxyMAlonamide), used as an internal standard, was added to the solutions by LPCP. The derivatization of MOCHOBA in each solution was also done by LPCP before the shipment. Calibration solutions were supplied by LPCP to the participants. To guarantee the stability of the compounds, the solutions were shipped in cooled containers. 2.2. Description of the approach used for the data processing The data were processed according to the requirements of NF ISO 13528 [3] and NF ISO 5725-5 [4] standards. These standards are based on robust statistics which allow the data to be analyzed in such a way that it is not required to eliminate the results which could be considered as outliers. The different steps of the data processing are: x Determination of the assigned value and of its uncertainty x Determination of method repeatability, reproducibility and trueness 2.3. Determination of the assigned values The assigned values are based on the gravimetric dilution of high purity compounds (> 99%) in ethyl acetate. The uncertainties of the assigned values were calculated according to the requirements of the guide to the expression of uncertainty in measurement [5]. They take into account the uncertainties on weighings, purity of the compounds and density of the solution. Table 1. Assigned values Concentration of the analyte in the unknown solution (mol/l) DMDOHEMA -2
1.939.10 r 0.028.10
MOHOBA -2
MOCHOBA
-3
3.756.10 r 0.054.10
-3
4.742.10-3 r 0.068.10-3
The uncertainties are given with a coverage factor k = 2. 2.4. Determination of method repeatability, reproducibility and trueness The repeatability standard deviation, sr, is calculated from algorithm S of NF ISO 5725-5 standard. This algorithm is applied to within-laboratory standard deviations and yields a robust pooled value of the standard deviations to which it is applied. The reproducibility standard deviation, sR, is calculated from the following equation:
sR
s d2
n 1 u s 2 n
r
705
706
C. Rivier et al. / Procedia Chemistry 7 (2012) 703 – 708
Where sd is calculated from algorithm A of NF ISO 5725-5 standard and n is the number of replicates for each determination (3 replicates were asked for each determination). To detect a potential bias of the method, the following test is applied:
x * X 1,25 u sd 2 u 2
En
X
p
Where x* is the robust mean calculated from algorithm A of NF ISO 5725-5, X is the assigned value, p is the number of laboratories and uX is the standard uncertainty of the assigned value. If En value is outside [-2 ; +2] interval, the method bias is considered significant. The source of the bias has to be identified before applying the method. 3. Results and discussion 3.1. DMDOHEMA, MOHOBA and MOCHOBA raw results The results of the participating laboratories are represented in Figure 2.
0.025 0.02 0.015 0.01 0.005 Lab 6
Lab 5
Lab 4
Lab 3
Lab 2
Lab 1
Concentration in unknown solution (mol/l)
DMDOHEMA 0.03
0.005
0.004 0.0035 0.003 0.0025
0.0055 0.005 0.0045 0.004 0.0035 0.003 0.0025 0.002 0.0015 Lab 6
Lab 5
Lab 4
Lab 3
Lab 2
Lab 6
Lab 5
Lab 4
Lab 3
Lab 2
Lab 1
0.002
Concentration in unknown solution (mol/l)
MOCHOBA
0.0045
Lab 1
Concentration in unknown solution (mol/l)
MOHOBA
Fig. 2. Interlaboratory comparison results
The black lines represent the assigned values and their associated uncertainties. The grey lines represent the robust means calculated according to algorithm A of NF ISO 5725-5 standard and their associated uncertainties. All uncertainties represented in the figure are expanded uncertainties calculated with a coverage factor k = 2.
707
C. Rivier et al. / Procedia Chemistry 7 (2012) 703 – 708
3.2. Determination of the characteristics of the method Due to technical difficulties reported from Lab 5 and 6 (same laboratory having reported 2 results), the results from this laboratory were not taken into account for the calculation of method repeatability, reproducibility and trueness. The detection limits for each compound have been estimated from the calibration data provided by the laboratories. The synthesis of the method performance characteristics is given in Table 2. Table 2. Method performance characteristics DMDOHEMA -2
MOHOBA -2
-3
4.742.10-3 r 0.068.10-3
Assigned value (mol/l)
1.939.10 r 0.028.10
Robust mean (mol/l)
1.87.10-2 r 0.12.10-2
3.75.10-3 r 0.25.10-3
4.653.10-3 r 0.093.10-3
Repeatability relative standard deviation (%)
0.7
1.2
1.1
Reproducibility relative standard deviation (%)
5.1
5.0
1.6
En number for trueness test
-1.1
-0.1
-1.5
Detection limit (mol/l)
-5
2.10
3.756.10 r 0.054.10
MOCHOBA -3
1.10
-5
1.10-5
The reproducibility relative standard deviations are around 2% for MOCHOBA and around 5% for DMDOHEMA and MOHOBA. It must be noted that these values do not take into account the spread associated with the preparation of the solutions (calibration solutions were provided by LPCP and solutions were derivatized by LPCP prior to the shipment). These reproducibility standard deviation values represent therefore a minimized value of the method reproducibility. Moreover the limited number of data used for the calculations has to be considered. No significant bias has been observed during the comparison (En numbers between [-2 ; +2]). Therefore, it is possible to use the reproducibility standard deviation values to estimate the uncertainties associated with the analysis of DMDOHEMA, MOHOBA and MOCHOBA [6] [7]. These uncertainties are around 4% for MOCHOBA and 10% for DMDOHEMA and MOHOBA (for a coverage factor k = 2). These values are minimized as they do not take into account the preparation of the calibration solutions and the preparation steps before the analysis (especially the derivatization process for MOCHOBA). The detection limits are indicative and are only based on calibration curve data. In case of a more complex analytical process (addition of preparation steps for example) or in case of a change in the calibration range, these detection limits have to be evaluated again. The evaluation of the detection limit in real sample matrices is recommended. Lab 1 and Lab 2 used on-column injection mode while the other laboratories used a split/splitless injector. Lab 3 and Lab 4 results demonstrate that the use of a split/plitless injector is possible for the analysis of DMDOHEMA, MOHOBA and MOCHOBA. However this injector, working at temperatures between 200°C and 270°C, requires to optimize the injection conditions in order to guarantee that no degradation of the compounds happens. These temperature effects could explain the deviations observed for Lab 5 and Lab 6 results. For these reasons, it is recommended to use an on-column injector which allows a progressive increase of the temperature from ambient temperature to 270°C and limits the degradation of the compounds sensitive to temperature.
708
C. Rivier et al. / Procedia Chemistry 7 (2012) 703 – 708
4. Conclusion For more than 50 years, CETAMA has been playing a major role in the nuclear field by providing interlaboratory comparisons for method validation. Since 1993, 13 different methods for organic compounds analysis have been published by CETAMA Working Group 24 in the ANASOR collection book. The development of new extraction processes for minor actinide separation requires the use of fully validated analytical methods to survey and maintain the process performances. The scope of CETAMA WG 24, which first focused on PUREX process and therefore on TriButylPhosphate analysis, was recently extended to the analysis of “new” extractants studied for the spent fuel reprocessing processes under development at CEA. Among these new extractants, DMDOHEMA has a specific place as it could be used in different processes depending on the strategy chosen for spent fuel reprocessing. Therefore, CETAMA WG 24 decided to focus on this compound and its main degradation products and proposed in 2011 an interlaboratory comparison for the validation of DMDOHEMA analysis method by GC-FID. The performance characteristics of the method (repeatability, reproducibility, trueness, detection limit) were evaluated from the interlaboratory comparison results and the measurements uncertainties could be estimated. This new method was recently added to the ANASOR book which can be supplied to each CETAMA member laboratory. In the future, CETAMA WG 24 will continue to satisfy the nuclear research and industrial needs by validating new analytical methods for organic molecules of new processes.
References [1] Camès B, Bisel I, Hill C, Rudloff D, Saucerotte B. Diamex solvent behavior under continuous degradation and regeneration operations, Nuclear Energy and the Environment, ACS symposium series 2010, vol. 1046, Chapter 21, pp 255-269 [2] Berthon L, Camès B, Rapport scientifique 1999; CEA Report: CEA-R-5892, Direction du cycle du combustible, 2000. pp 206-211. [3] NF ISO 13528:2005, Statistical methods for use in proficiency testing by interlaboratory comparisons [4] NF ISO 5725-5:1998, Accuracy (trueness and precision) of measurements methods and results – Part 5 : Alternative methods for the determination of the precision of a standard measurement method [5] JCGM 100:2008, Evaluation of measurement Data –Guide to expression of uncertainty in measrement, Joint Comittee for guides in metrology [6] NF ISO 13528:2010, Guidance for the use of repeatability, reproducibility and trueness estimates in measurement uncertainty estimation [7] Désenfant M, Priel M, Rivier C, Evaluation des incertitudes des résultats d’analyse, Techniques de l’ingénieur 2006, P 105, pp 1-17
