- Email: [email protected]
Microwave-induced combustion of high purity nuclear flexible graphite for the determination of potentially embrittling elements using atomic spectrometric techniques
Microchemical Journal 124 (2016) 321–325
Contents lists available at ScienceDirect
Microchemical Journal journal homepage: www.elsevier.com/locate/microc
Microwave-induced combustion of high purity nuclear flexible graphite for the determination of potentially embrittling elements using atomic spectrometric techniques Michele Stefani Peters Enders, Juliana Pinheiro de Souza, Paula Balestrin, Paola de Azevedo Mello, Fabio Andrei Duarte, Edson Irineu Muller ⁎ Departamento de Química, Universidade Federal de Santa Maria, 97105-900 Santa Maria, RS, Brazil
a r t i c l e
i n f o
Article history: Received 22 June 2015 Received in revised form 31 August 2015 Accepted 17 September 2015 Available online 25 September 2015 Keywords: High purity nuclear flexible graphite Embrittling elements Microwave-induced combustion Inductively coupled plasma optical emission spectrometry Digestion Elemental impurities
a b s t r a c t Microwave-induced combustion was evaluated for the digestion of high purity nuclear flexible graphite for further determination of potentially embrittling elements (Ag, As, Bi, Cd, Ga, Hg, In, Pb, Sb, Sn and Zn) using atomic spectrometric techniques. Flexible graphite is obtained by exfoliation process of conventional graphite using oxidant agents and subsequent fast heating at high temperatures. Thus, high chemical inertness for specific purposes is attained which causes inertness also to conventional digestion methods. Microwave-assisted digestion method using maximum temperature and pressure of 275 °C and 180 bar (UltraWave™ system) respectively, and dry ashing were also evaluated for flexible graphite digestion. However, insoluble residues in final digests and analyte losses were observed for those methods, respectively. For microwave-induced combustion method, the use of cellulose pellet (300 mg) as combustion aid allowed the efficient oxidation of 100 mg of flexible graphite. In order to assure the quantitative recovery of all analytes after microwave-induced combustion two absorbing solutions should be used: 4 mol L−1 HNO3 for Ag, As, Cd, Ga, Hg, In, Pb, and Zn and inversed aqua regia for Bi, Sb and Sn. As flexible graphite (or similar matrix) is not available as certified reference material, accuracy was evaluated using coal (NIST 1632c and BCR 40). For all elements, except for In (not informed in certified reference materials), significant differences were not observed by comparing the results obtained by microwave-induced combustion and certified values (t test, 95% confidence level). Limits of detection using inductively coupled plasma optical emission spectrometry and chemical vapor generation atomic absorption spectrometry (only for Hg) were lower than 12 mg kg−1 and in compliance with the recommendation of General Electric for nonmetallic materials (the limit for each element is 200 mg kg−1 and the sum of all embrittling elements should be lower than 500 mg kg−1). Microwave-induced combustion method was suitable for quality control of high purity nuclear flexible graphite. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Flexible graphite (FG) is a flexible sheet produced from natural graphite by intercalation with sulfuric or nitric acid, followed by exfoliation by fast heating at high temperature (up to 1000 °C). Exfoliation process increases the resilience, impermeability to fluids and chemical inertness of FG in comparison to graphite used as raw material. Due to its specific characteristics, FG has been used as a non-metallic material in water-reactor nuclear power plant (WRNPP) applications [1–4]. The degradation of different materials used in WRNPP has been a frequent problem. One of these problems is the embrittlement process caused by metals in WRNPP operational conditions. Embrittlement is a loss of ductility of a material, making it brittle and can be caused by diffusion of atoms of metal, either solid or liquid into the WRNPP materials. ⁎ Corresponding author. E-mail address: [email protected] (E.I. Muller).
http://dx.doi.org/10.1016/j.microc.2015.09.015 0026-265X/© 2015 Elsevier B.V. All rights reserved.
In this sense, elements such as Ag, As, Bi, Cd, Ga, Hg, In, Pb, Sb, Sn and Zn can promote embrittling process because the temperatures used in WRNPP are close to the temperatures that these elements can promote the embrittling effect [5–8]. In order to avoid problems in WRNPP the purity of FG materials regarding the potentially embrittling elements should be controlled. The contamination of embrittling elements present in FG can occur due to the presence of trace elements in raw materials (e.g. graphite and intercalate reagents) used during the production of FG. According to the recommendation of General Electric for nonmetallic materials, the maximum concentration of each potentially embrittling metal should be lower than 200 mg kg−1 and the sum of all of these elements should be lower than 500 mg kg−1 [8]. Despite the requirements for the determination of embrittling elements in this kind of material, works related to the determination of these elements in FG are scarce in literature, probably due to the refractory characteristics of this material to the conventional wet digestion methods. According to the literature, FG
322
M.S.P. Enders et al. / Microchemical Journal 124 (2016) 321–325
material is more refractory to oxidation process than common graphite probably due to the exfoliation process that increases the chemical inertness of this material [1,2]. Thus, conventional wet digestion methods, even using high temperature and pressure and combining strong oxidizing acids, can be ineffective. Particularly in these cases and for organic refractory matrices, combustion methods are preferred allowing to obtain final digests with relatively low residual carbon content (RCC). Among the main methods used for the digestion of refractory organic samples include dry ashing and microwave-induced combustion (MIC). Dry ashing is a very simple method that allows the oxidation of relatively large sample mass of organic matrix (up to 10 g) with subsequent dissolution of resultant residues using diluted acids [9]. It is well reported in literature that losses of volatile elements (such as As, Cd, Pb and Hg) were significant for dry ashing digestion as well as contamination caused by the laboratory environment. In this sense, the heating program for dry ashing should be evaluated for each sample and additives can be added to the sample in order to minimize analyte losses [9,10]. MIC has been used for the digestion of several organic refractory matrices providing final digests with residual carbon content (RCC) lower than 1% [10–17]. Recently, MIC was used for the digestion of common graphite allowing the digestion of sample mass up to 400 mg [16]. In order to increase the sample mass or assure the efficient digestion of refractory organic samples or analyte release from matrix using MIC, some attempts have been carried out mixing easily combustible materials (e.g. cellulose) with these samples to provide enough energy to aid the combustion process [18]. Additionally, it is important to point out that the choice of absorbing solution for the determination of metals and metalloids by MIC depends on the characteristics and solubility of each element. Usually, quantitative recoveries for elements have been obtained using diluted acids (e.g. for metals) or alkaline solutions (e.g. for halogens). On the other hand, concentrated acids (HNO3 and HCl) or a mixture of these acids is required in order to assure the quantitative recoveries for some applications [19]. In this work, MIC was evaluated for FG sample digestion for subsequent determination of potentially embrittling elements (Ag, As, Bi, Cd, Ga, Hg, In, Pb, Sb, Sn and Zn) using atomic spectrometric techniques. Some parameters of MIC method such as sample mass, use of cellulose as combustion aid, type and concentration of absorbing solution were evaluated to assure the quantitative recoveries of embrittling elements. Additionally, dry ashing and microwave-assisted digestion (MAD) were also used for FG digestion. Finally, accuracy of MIC was evaluated using certified reference materials (CRMs) of coal (NIST 1632c and BCR 40) once a CRM for FG is not commercially available. 2. Experimental 2.1. Reagents, solution and samples Purified water (resistivity of 18.2 MΩ cm) was obtained using a Milli-Q system (Millipore, Bedford, USA) and it was used to prepare all solutions. Reagents were of analytical grade and acquired from Merck (Darmstadt, Germany). Nitric acid 65% has been purified by sub-boiling distillation system (Milestone, Sorisole, Italy). MIC digestions were carried out under oxygen pressure with purity better than 99.6% (White Martins, São Paulo, Brazil). Plasma-based instruments were operated using argon with purity higher than 99.998% (White Martins, São Paulo, Brazil). External calibration for the determination of Ag, As, Bi, Cd, Ga, Hg, In, Pb, Sb, Sn and Zn was carried out using suitable dilution of 1000 mg L−1 standard solutions (Spex CertiPrep, Metuchen, USA). Four samples of flexible graphite sheet (named as FG I, FG II, FG III and FG IV) donated by two private companies were used in this work. Accuracy of MIC method was evaluated using CRMs with high carbon content matrices — NIST 1632c (National Institute of Standard and Technology, trace elements in coal) and BCR 40 (Community Bureau of Reference, trace elements in coal).
For recovery experiments, spiked samples were prepared by adding a known amount of a standard solution (the same that was used for calibration) containing all potentially embrittling elements, over the sample pellet before the MIC and dry ashing digestions. FG sample “IV” was arbitrarily chosen for these experiments and the results obtained by the analyte spike were used to calculate the recovery of all analytes. 2.2. Instrumentation FG sheet samples were milled in a cryogenic mill (Spex CertiPrep, model 6750, Metuchen, USA) and this procedure was repeated until the whole sample showed particle size lower than 80 μm. An analytical balance (model AY 220, Shimadzu, São Paulo, Brazil) was used to weigh samples and reagents. All embrittling elements (exception of Hg) were determined using an inductively coupled plasma optical emission spectrometer (Ciros CCD model, Spectro, Kleve, Germany) equipped with a Scott-type double pass spray chamber, a cross-flow nebulizer, a torch and a quartz injector. The operational conditions were summarized in Table 1. Determination of Hg was carried out using a homemade flow-injection (FI) system for chemical vapor generation (CVG) with detection by atomic absorption spectrometry (AAS). It consisted of a peristaltic pump (Ismatec, Zurich, Switzerland), an injection valve fitted with 100 μL sample loop and a U type gas–liquid separator. Mercury determination was carried out using an atomic absorption spectrometer (Vario 6 model, Analytik Jena, Jena, Germany) equipped with a heated quartz cell furnace (HS5, 100 mm length, 10 mm i.d.). A mercury hollow cathode lamp operated at 4.0 mA was used as a radiation source and a deuterium lamp was used for background correction. The wavelength was set at 253.7 nm and the spectral band pass at 0.5 nm. All measurements were performed in integrated absorbance (peak area) [20]. Some embrittling elements were also determined in CRMs using inductively coupled plasma mass spectrometry (ICP-MS) because their concentrations were lower than the LOD provided by ICP-OES (Table 1). Determination was carried out using inductively coupled plasma mass spectrometer (Elan DRC II model, Perkin Elmer). Digestion of FG samples using MIC was performed using a Multiwave 3000 microwave sample preparation system (Anton Table 1 Operational parameters for analyte determination by ICP-OES and ICP-MS. Parameter
ICP-OES
ICP-MS
RF power wavelength (W) Plasma gas flow rate (L min−1) Auxiliary gas flow rate (L min−1) Nebulizer gas flow rate (L min−1) Spray chamber Nebulizer Wavelength (nm)
1600 14 1.00 0.80 Scott-type double pass Cross-flow Ag 328.068 As 188.979 Bi 223.061 Cd 228.802 Ga 417.206 In 230.606 Pb 220.353 Sb 206.836 Sn 189.927 Zn 206.200 – – – – – – –
1400 15 1.20 1.09 Cyclonic Concentric –
Dwell time (ms) Sweeps/reading Readings/replicate Replicates Data collection mode Sampler and skimmer cones Isotopes (m/z)
50 5 3 3 Peak hopping Pt 107 Ag 209 Bi 111 Cd 71 Ga 121 Sb 118 Sn
M.S.P. Enders et al. / Microchemical Journal 124 (2016) 321–325
Paar, Graz, Austria) equipped with up to eight pressurized quartz vessels. Maximum pressure and internal volume allowed by quartz vessels were 80 bar and 80 mL, respectively. Dry ashing was performed in a muffle furnace (LF0912 model, Jung, Brazil) using platinum crucibles. Microwave-assisted digestion was also carried out using a UltraWave™ system (Milestone, Sorisole, Italy) equipped with five quartz vessel (total volume of 40 mL). The microwave cavity (1 L) is covered with polytetrafluoroethylene (PTFE) vessel and before digestion was sealed and pressurized with 40 bar of argon 99.996% (White Martins, São Paulo, Brazil). All digestions using UltraWave™ system were carried out using maximum power, temperature and pressure of 1500 W, 275 °C and 180 bar, respectively [21].
323
3. Results and discussion Potentially embrittling elements in high purity FG arise mainly from raw material used for its production. In this sense, the determination of these elements requires that the organic matrix should be oxidized in order to bring the analytes into a suitable solution for subsequent atomic spectrometric techniques. In this study, the main efforts were devoted to MIC method optimization. However, some attempts were also performed to evaluate dry ashing and MAD for FG samples. It is important to mention that MAD did not assure the complete digestion of FG samples and insoluble residues were observed in final digests. In this sense, the analyte quantification was not performed and the results for MAD were not discussed in this section. 3.1. Dry ashing digestion
2.3. Dry ashing Sample masses up to 1 g of milled FG were directly weighed in to the platinum crucibles and positioned inside the muffle furnace and heated at 600, 800 or 1000 °C, during 4 h. Heating ramp of 10 °C min−1 was used until reaching the final temperature (600, 800 or 1000 °C). Some experiments were also performed using a previous oxidation of FG sample with concentrated HNO3 (5 mL) heated in a hot plate (95 °C, during 1 h). The residue was dissolved with 5 mL of 7 mol L−1 HNO3 under heating in a water bath for 30 min and diluted up to 25 mL with water. In order to minimize analyte losses, samples were initially wetted with concentrated H2SO4 (5 mL) and burnt in a Bunsen burner. Subsequently, the remaining residues were digested using a muffle furnace at 1000 °C. The efficiency of digestion using dry ashing was evaluated by the characteristics of final residues and by weighing of the platinum crucibles after heating program. Recovery experiments using analyte spike with standard solutions were performed for all experiments using dry ashing method.
2.4. Microwave-assisted digestion (MAD) Sample mass up to 100 mg was weighed and added to the quartz vessels. Concentrated HNO3 (6 mL) was used as oxidant agent. All digestions were performed using maximum temperature and pressure available in the microwave system (275 °C and 180 bar, respectively). The heating program was as follows: i) ramp of 10 min up to 1500 W; ii) 1500 W for 40 min and iii) 0 W for 20 min (cooling).
2.5. Microwave-induced combustion (MIC) For MIC, 100 mg of samples were pressed as pellets (8 mm diameter) using a hydraulic press set at 3 tons. Additionally, pellets of microcrystalline cellulose (100, 200 or 300 mg) were also produced and used together with the sample pellets in order to aid the combustion of FG samples. FG and cellulose pellets were placed on a small disk of filter paper on the quartz holder. Filter paper was wetted with 50 μL of 6 mol L− 1 NH4NO3 and the quartz holder was positioned inside the quartz vessel filled with 6 mL of absorbing solution. The absorbing solutions of HNO3 (2, 4, 7 and 14 mol L−1) or inversed aqua regia (4.5 mL of concentrated HNO3 and 1.5 mL of concentrated HCl) were evaluated. Quartz vessels were closed, placed in the rotor and pressurized with 20 bar of oxygen. Rotor was capped and placed inside the microwave oven and the following heating program was started: i) 1400 W for 10 min (ignition and reflux steps) and ii) 0 W for 20 min (for cooling). Final solutions were diluted up to 25 mL with water and analyzed by ICP-OES and CVG-AAS. Decontamination of quartz vessels and holders was performed using concentrated HNO3 with the same microwave heating program used for combustion of samples. Recovery tests were carried out using spike of all analytes.
Initially, the digestion of FG samples was evaluated using oxidation of organic sample matrix in the presence of oxygen atmosphere inside the muffle furnace at high temperature. The temperatures, generally recommended for dry ashing of common organic matrices (such as leaves, meat and flour) are about 500 °C [9]. However, the oxidation of graphite at temperatures between 400 and 500 °C is very small (lower than 1%) [22]. According to the literature the amount of oxidized carbon of graphite increased greatly from 500 to 800 °C (with air flow of 20 mL min−1) [22]. In this sense, the oxidation of FG sample was carried out using temperatures of 600, 800 and 1000 °C which are higher than those recommended for common organic matrices due to the refractory characteristics of FG matrix. Significant oxidation was not observed for the FG samples heated using temperature up to 600 °C and FG matrix maintained its original appearance. Moreover, the mass of platinum crucibles containing the sample after the dry ashing program did not decrease showing that there were no losses of sample mass. On the other hand, an oxidation of about 50% of sample mass was observed after dry ashing at 800 °C. In order to assure the complete oxidation of graphite matrix, dry ashing was carried out at 1000 °C. In this temperature the sample mass loss was higher than 95% and the remaining residue presented as white residue indicating the complete oxidation of FG matrix. In order to decrease the temperature of dry ashing, experiments were performed using a previous oxidation of FG sample with concentrated HNO3 in hot plate heating (95 °C). However, complete oxidation of FG sample was only observed for dry ashing at 1000 °C even using previous oxidation with HNO3. Although practically complete oxidation of FG, recovery tests were performed to verify the losses of analyte during digestion by dry ashing. In this work, recoveries obtained for all analytes were lower than 80%. Additionally, Ag, As, Cd, Hg, Pb and Zn presented recovery values lower than 10% probably due to the high temperature required for complete oxidation of FG matrix (1000 °C). To overcome the problem related with losses of analytes, experiments were carried out adding concentrated sulfuric acid before the dry ashing digestion. However, quantitative recoveries for all potentially embrittling elements were not observed even using sulfuric acid as additive to minimize losses of some analytes. In this sense, dry ashing was not suitable for FG sample digestion and subsequent determination of potentially embrittling elements. In order to assure the accuracy of results, MIC was evaluated and optimized for FG sample digestion. 3.2. Microwave-induced combustion In order to achieve the limits of detection required to comply the recommendation of General Electric for nonmetallic materials, sample mass of FG that can be efficiently digested by MIC should be increased. Some initial experiments showed that the use of easily combustible materials (e.g. cellulose) combined with FG sample can provide enough energy to aid the combustion of refractory samples.
324
M.S.P. Enders et al. / Microchemical Journal 124 (2016) 321–325
14 mol L-1
7 mol L-1
2 mol L-1
4 mol L-1
Inversed aqua regia
110 100 90
Recovery (%)
80 70 60 50 40 30 20 10 0 Ag
As
Cd
Ga
Hg
In
Pb
Zn
Bi
Sb
Sn
Fig. 1. Influence of concentration and type of absorbing solution using MIC for subsequent determination of embrittling elements by atomic spectrometric techniques (error bars are the standard deviation, n = 5).
3.2.1. Evaluation of cellulose mass Initially, pellets of FG sample (up to 100 mg) were placed together with pellets of cellulose (100, 200 or 300 mg) on the quartz holder. Efficient digestion of FG sample (without residues on quartz holder) was only observed for the combination of sample pellets (100 mg) and cellulose (300 mg). In this condition, the final pressure of the system was 60 bar that is equivalent to 75% of maximum pressure allowed by the manufacturer. For subsequent experiments pellets of 300 mg of cellulose were used as aid for the digestion of 100 mg of FG sample.
6 mL of 4 mol L− 1 HNO3 as absorbing solution. Additionally, even using concentrated HNO3, the recoveries for Bi, Sb and Sn were lower than 80%. These low recoveries can be explained by the formation of species that are not soluble in HNO3, requiring the use of inversed aqua regia [23]. In this respect, the use of aqua regia was evaluated as absorbing solution providing quantitative recoveries (better than 95%) for Bi, Sb and Sn (Fig. 1). In this way, a protocol for quality control can be proposed where, each FG sample should be digested by MIC method using: i) 4 mol L−1 HNO3 as absorbing solution for Ag, As, Cd, Ga, Hg, In, Pb, and Zn and ii) inversed aqua regia for Bi, Sb and Sn.
3.2.2. Evaluation absorbing solution The choice of absorbing solution for MIC depends on the characteristics of analyte. Diluted HNO3 solutions are preferred for metals and metalloids in order to minimize interferences on determination by plasma-based techniques and assure the complete analyte absorption. However, for some elements (such as Sn and Sb) inversed aqua regia should be used to assure the quantitative recovery of these analytes. In this work, HNO3 solutions (2, 4, 7 and 14 mol L−1) were evaluated as absorbing solution for all embrittling elements. On the other hand, inversed aqua regia (4.5 mL of concentrated HNO3 and 1.5 mL of concentrated HCl) was used to absorb only the analytes that showed low recovery values (Sn, Sb and Bi). In Fig. 1, the recoveries for all analytes using these solutions are shown. For all elements, with the exception of Bi, Sb and Sn, the recovery was quantitative (better than 95%) using
3.3. Decomposition of FG samples using MIC method Four FG samples were digested using the optimized MIC method with cellulose for further determination of embrittling elements by ICP-OES and CVG-AAS. The maximum concentration of each embrittling element should be lower than 200 mg kg− 1 and the sum of these analytes cannot exceed 500 mg kg−1. According to the results shown in Table 2, all analyzed samples are in agreement with the recommendation of General Electric for nonmetallic materials. FG sample I presented the highest sum of embrittling elements when compared to other samples. The limit of detection obtained for all embrittling element was lower than 12 mg kg−1 and the sum of all LODs was lower than 37 mg kg−1, that are suitable for FG quality control analysis. The relative standard deviations of the results obtained by MIC digestion
Table 2 Concentration of embrittling elements in FG samples and CRMs determined by ICP-OES, ICP-MS and CVG-AAS after decomposition using MIC (mg kg−1, n = 5, mean ± standard deviation). Element
MIC FG I
Ag As Bi Cd Ga Hg In Pb Sb Sn Zn a *
b0.5 6.4 ± 0.9 b7.0 2.6 ± 0.3 b8.0 b0.1 b12 4.3 ± 0.5 b2.5 b3.0 21 ± 2
Certified values FG II b0.5 b2.0 b7.0 b0.20 b8.0 b0.1 b12 3.2 ± 0.3 b2.5 b3.0 10 ± 1
Determination carried out using ICP-MS. Informed value.
FG III b0.5 b2.0 b7.0 b0.20 b8.0 b0.1 b12 b1.0 b2.5 b3.0 b0.30
FG IV b0.5 b2.0 b7.0 b0.2 b8.0 b0.1 b12 b1.0 b2.5 b3.0 b0.30
BCR 40 b0.5 13 ± 2 b7.0 0.12 ± 0.01a b8.0 0.38 ± 0.08 b12 25 ± 2 b2.5 b3.0 32 ± 3
NIST 1632 c a
0.11 ± 0.01 6.5 ± 1.0 0.09 ± 0.01a 0.07 ± 0.01a 3.0 ± 0.4a b0.1 b12 3.6 ± 0.3 0.45 ± 0.06a 1.2 ± 0.1a 11 ± 1
BCR 40
NIST 1632 c
n.d. 13.2 ± 1.1 n.d. 0.11 ± 0.02 n.d. 0.35 ± 0.06 n.d. 24.2 ± 1.7 n.d. n.d. 30.2 ± 1.9
0.1* 6.18 ± 0.27 0.1* 0.072 ± 0.007 3* 0.0938 ± 0.0037 n.d. 3.79 ± 0.07 0.461 ± 0.029 1* 12.1 ± 1.3
M.S.P. Enders et al. / Microchemical Journal 124 (2016) 321–325
and ICP-OES determination were lower than 15%. As FG matrix is not available as certified reference material, accuracy was evaluated using certified reference material of coal (BCR 40 and NIST 1632c) that presents high carbon content. Additionally, it was not possible to evaluate accuracy for In because this element is not informed in CRMs used in this work. It is also important to point out that some elements (Ag, Bi, Cd, Ga, Sb and Sn) were also determined by inductively coupled plasma mass spectrometry (ICP-MS) in CRM digests in order to achieve the LOD required to evaluate the accuracy by comparison of certified reference values and the results obtained using the proposed digestion method. In this sense, significant difference (t test, 95% confidence level) was not observed between certified reference values and the results obtained using MIC and atomic spectrometric techniques. 4. Conclusion Among the digestion methods evaluated for FG digestion only MIC was considered suitable for the digestion of up to 100 mg of sample. For MAD and dry ashing, insoluble residues in final digest and losses of analytes were observed, respectively. Cellulose should be used as combustion aid for MIC to assure the complete oxidation of FG matrix that is very refractory even using MIC. Two independent absorbing solutions (4 mol L−1 HNO3 and inversed aqua regia) were required to allow the quantitative recovery of all embrittling element for MIC digestion. LODs obtained by MIC and determination by ICP-OES and FI-CVG-AAS were satisfactory for quantification of all potentially embrittling elements and are in agreement to the recommendation of General Electric for nonmetallic materials. Acknowledgments The authors are grateful to CNPq (304424/2012-9), FAPERGS (19402551/13-4) and CAPES (23038.002513/2014-37) for supporting this study. References [1] M. Cai, D. Thorpe, D.H. Adamson, H.C. Schniepp, Methods of graphite exfoliation, J. Mater. Chem. 22 (2012) (24992-25002.2). [2] D.D.L. Chung, Review exfoliation of graphite, J. Mater. Sci. 22 (1987) 4190–4198. [3] R. Chugh, D.D.L. Chung, Flexible graphite as a heating element, Carbon 40 (2002) 2285–2289. [4] X. Luo, D.D.L. Chung, Flexible graphite under repeated compression studied by electrical resistance measurements, Carbon 39 (2001) 985–990. [5] S.P. Lynch, Metal-induced embrittlement of materials, Mater. Charact. 28 (1992) 279–289. [6] C.F. Old, Liquid metal embrittlement of nuclear materials, J. Nucl. Mater. 92 (1980) 2–25.
325
[7] H.C. Furtado, I. May, High temperature degradation in power plants and refineries, Mater. Res. 7 (2004) 103–110. [8] General Electric nuclear non-metallic material specification D50YP12 (rev. 2), Material Specification for Non-metallic Impurity Limits in Engineering Materials Exposed to Boiling Water Reactor1983. [9] M. Hoenig, Dry ashing, in: Z. Mester, R. Sturgeon (Eds.), Sample Preparation for Trace Element Analysis, Elsevier 2003, pp. 235–254. [10] J.S. Barin, B. Tischer, R.S. Picoloto, F.G. Antes, F.E.B. Silva, F.R. Paula, E.M.M. Flores, Determination of toxic elements in tricyclic active pharmaceutical ingredients by ICP-MS: a critical study of digestion methods, J. Anal. At. Spectrom. 29 (2014) 352–358. [11] E.M.M. Flores, E.I. Muller, F.A. Duarte, P. Grinberg, R.E. Sturgeon, Determination of trace elements in fluoropolymers after microwave-induced combustion, Anal. Chem. 85 (2013) 374–380. [12] A.L.H. Muller, C.C. Muller, F. Lyra, P.A. Mello, M.F. Mesko, E.I. Muller, E.M.M. Flores, Determination of toxic elements in nuts by inductively coupled plasma mass spectrometry after microwave-induced combustion, Food Anal. Methods 6 (2013) 258–264. [13] J.V. Maciel, C.L. Knorr, E.M.M. Flores, E.I. Muller, M.F. Mesko, E.G. Primel, F.A. Duarte, Feasibility of microwave-induced combustion for trace element determination in Engraulis anchoita by ICP-MS, Food Chem. 145 (2014) 927–931. [14] J.S.F. Pereira, P.A. Mello, D.P. Moraes, F.A. Duarte, V.L. Dressler, G. Knapp, E.M.M. Flores, Chlorine and sulfur determination in extra-heavy crude oil by inductively coupled plasma optical emission spectrometry after microwave-induced combustion, Spectrochim. Acta B 64 (2009) 554–558. [15] J.S.F. Pereira, D.P. Moraes, F.G. Antes, L.O. Diehl, M.F.P. Santos, R.C.L. Guimarães, T.C.O. Fonseca, V.L. Dressler, E.M.M. Flores, Determination of metals and metalloids in light and heavy crude oil by ICP-MS after digestion by microwave-induced combustion, Microchem. J. 96 (2010) 4–11. [16] S.M. Cruz, L. Schmidt, F.M. Dalla Nora, M.F. Pedrotti, C.A. Bizzi, J.S. Barin, E.M.M. Flores, Microwave-induced combustion method for the determination of trace and ultratrace element impurities in graphite samples by ICP-OES and ICP-MS, Microchem. J. 123 (2015) 28–32. [17] T.S. Nunes, C.C. Muller, P. Balestrin, A.L.H. Muller, M.F. Mesko, P.A. Mello, E.I. Muller, Determination of chlorine and sulfur in high purity flexible graphite using ion chromatography (IC) and inductively coupled plasma optical emission spectrometry (ICP OES) after pyrohydrolysis sample preparation, Anal. Methods 7 (2015) 2129–2134. [18] R.S. Picoloto, H. Wiltsche, G. Knapp, P.A. Mello, J.S. Barin, E.M.M. Flores, Determination of inorganic pollutants in soil after volatilization using microwave-induced combustion, Spectrochim. Acta B 86 (2013) 123–130. [19] J.S. Barin, E.M.M. Flores, M.F. Mesko, P.A. Mello, J.S.F. Pereira, Microwave-induced combustion, in: E.M.M. Flores (Ed.), Microwave-assisted Sample Preparation for Trace Element Determination, Elsevier 2014, pp. 143–173. [20] L.E. Kaercher, F. Goldschmidt, J.N.G. Paniz, E.M.M. Flores, V.L. Dressler, Determination of inorganic and total mercury by vapor generation atomic absorption spectrometry using different temperatures of the measurement cell, Spectrochim. Acta B 60 (2005) 705–710. [21] J.A. Nobrega, C. Pirola, L.L. Fialho, G. Rota, C.E.K.M.A. Campos Jordao, F. Pollo, Microwave-assisted digestion of organic samples: how simple can it become? Talanta 98 (2012) 272–276. [22] L. Xiaowei, R. Jean-Charles, Y. Suyuan, Effect of temperature on graphite oxidation behavior, Nucl. Eng. Des. 227 (2004) 273–280. [23] P.J. Walter, S. Chalk, H.M.S. Kingston, Overview of microwave-assisted sample preparation, in: H.M.S. Kingston, S.J. Haswell (Eds.), Microwave-enhanced Chemistry. Fundamentals, Sample Preparation and Applications, American Chemical Society, Washington 1997, pp. 55–216.
Contents lists available at ScienceDirect
Microchemical Journal journal homepage: www.elsevier.com/locate/microc
Microwave-induced combustion of high purity nuclear flexible graphite for the determination of potentially embrittling elements using atomic spectrometric techniques Michele Stefani Peters Enders, Juliana Pinheiro de Souza, Paula Balestrin, Paola de Azevedo Mello, Fabio Andrei Duarte, Edson Irineu Muller ⁎ Departamento de Química, Universidade Federal de Santa Maria, 97105-900 Santa Maria, RS, Brazil
a r t i c l e
i n f o
Article history: Received 22 June 2015 Received in revised form 31 August 2015 Accepted 17 September 2015 Available online 25 September 2015 Keywords: High purity nuclear flexible graphite Embrittling elements Microwave-induced combustion Inductively coupled plasma optical emission spectrometry Digestion Elemental impurities
a b s t r a c t Microwave-induced combustion was evaluated for the digestion of high purity nuclear flexible graphite for further determination of potentially embrittling elements (Ag, As, Bi, Cd, Ga, Hg, In, Pb, Sb, Sn and Zn) using atomic spectrometric techniques. Flexible graphite is obtained by exfoliation process of conventional graphite using oxidant agents and subsequent fast heating at high temperatures. Thus, high chemical inertness for specific purposes is attained which causes inertness also to conventional digestion methods. Microwave-assisted digestion method using maximum temperature and pressure of 275 °C and 180 bar (UltraWave™ system) respectively, and dry ashing were also evaluated for flexible graphite digestion. However, insoluble residues in final digests and analyte losses were observed for those methods, respectively. For microwave-induced combustion method, the use of cellulose pellet (300 mg) as combustion aid allowed the efficient oxidation of 100 mg of flexible graphite. In order to assure the quantitative recovery of all analytes after microwave-induced combustion two absorbing solutions should be used: 4 mol L−1 HNO3 for Ag, As, Cd, Ga, Hg, In, Pb, and Zn and inversed aqua regia for Bi, Sb and Sn. As flexible graphite (or similar matrix) is not available as certified reference material, accuracy was evaluated using coal (NIST 1632c and BCR 40). For all elements, except for In (not informed in certified reference materials), significant differences were not observed by comparing the results obtained by microwave-induced combustion and certified values (t test, 95% confidence level). Limits of detection using inductively coupled plasma optical emission spectrometry and chemical vapor generation atomic absorption spectrometry (only for Hg) were lower than 12 mg kg−1 and in compliance with the recommendation of General Electric for nonmetallic materials (the limit for each element is 200 mg kg−1 and the sum of all embrittling elements should be lower than 500 mg kg−1). Microwave-induced combustion method was suitable for quality control of high purity nuclear flexible graphite. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Flexible graphite (FG) is a flexible sheet produced from natural graphite by intercalation with sulfuric or nitric acid, followed by exfoliation by fast heating at high temperature (up to 1000 °C). Exfoliation process increases the resilience, impermeability to fluids and chemical inertness of FG in comparison to graphite used as raw material. Due to its specific characteristics, FG has been used as a non-metallic material in water-reactor nuclear power plant (WRNPP) applications [1–4]. The degradation of different materials used in WRNPP has been a frequent problem. One of these problems is the embrittlement process caused by metals in WRNPP operational conditions. Embrittlement is a loss of ductility of a material, making it brittle and can be caused by diffusion of atoms of metal, either solid or liquid into the WRNPP materials. ⁎ Corresponding author. E-mail address: [email protected] (E.I. Muller).
http://dx.doi.org/10.1016/j.microc.2015.09.015 0026-265X/© 2015 Elsevier B.V. All rights reserved.
In this sense, elements such as Ag, As, Bi, Cd, Ga, Hg, In, Pb, Sb, Sn and Zn can promote embrittling process because the temperatures used in WRNPP are close to the temperatures that these elements can promote the embrittling effect [5–8]. In order to avoid problems in WRNPP the purity of FG materials regarding the potentially embrittling elements should be controlled. The contamination of embrittling elements present in FG can occur due to the presence of trace elements in raw materials (e.g. graphite and intercalate reagents) used during the production of FG. According to the recommendation of General Electric for nonmetallic materials, the maximum concentration of each potentially embrittling metal should be lower than 200 mg kg−1 and the sum of all of these elements should be lower than 500 mg kg−1 [8]. Despite the requirements for the determination of embrittling elements in this kind of material, works related to the determination of these elements in FG are scarce in literature, probably due to the refractory characteristics of this material to the conventional wet digestion methods. According to the literature, FG
322
M.S.P. Enders et al. / Microchemical Journal 124 (2016) 321–325
material is more refractory to oxidation process than common graphite probably due to the exfoliation process that increases the chemical inertness of this material [1,2]. Thus, conventional wet digestion methods, even using high temperature and pressure and combining strong oxidizing acids, can be ineffective. Particularly in these cases and for organic refractory matrices, combustion methods are preferred allowing to obtain final digests with relatively low residual carbon content (RCC). Among the main methods used for the digestion of refractory organic samples include dry ashing and microwave-induced combustion (MIC). Dry ashing is a very simple method that allows the oxidation of relatively large sample mass of organic matrix (up to 10 g) with subsequent dissolution of resultant residues using diluted acids [9]. It is well reported in literature that losses of volatile elements (such as As, Cd, Pb and Hg) were significant for dry ashing digestion as well as contamination caused by the laboratory environment. In this sense, the heating program for dry ashing should be evaluated for each sample and additives can be added to the sample in order to minimize analyte losses [9,10]. MIC has been used for the digestion of several organic refractory matrices providing final digests with residual carbon content (RCC) lower than 1% [10–17]. Recently, MIC was used for the digestion of common graphite allowing the digestion of sample mass up to 400 mg [16]. In order to increase the sample mass or assure the efficient digestion of refractory organic samples or analyte release from matrix using MIC, some attempts have been carried out mixing easily combustible materials (e.g. cellulose) with these samples to provide enough energy to aid the combustion process [18]. Additionally, it is important to point out that the choice of absorbing solution for the determination of metals and metalloids by MIC depends on the characteristics and solubility of each element. Usually, quantitative recoveries for elements have been obtained using diluted acids (e.g. for metals) or alkaline solutions (e.g. for halogens). On the other hand, concentrated acids (HNO3 and HCl) or a mixture of these acids is required in order to assure the quantitative recoveries for some applications [19]. In this work, MIC was evaluated for FG sample digestion for subsequent determination of potentially embrittling elements (Ag, As, Bi, Cd, Ga, Hg, In, Pb, Sb, Sn and Zn) using atomic spectrometric techniques. Some parameters of MIC method such as sample mass, use of cellulose as combustion aid, type and concentration of absorbing solution were evaluated to assure the quantitative recoveries of embrittling elements. Additionally, dry ashing and microwave-assisted digestion (MAD) were also used for FG digestion. Finally, accuracy of MIC was evaluated using certified reference materials (CRMs) of coal (NIST 1632c and BCR 40) once a CRM for FG is not commercially available. 2. Experimental 2.1. Reagents, solution and samples Purified water (resistivity of 18.2 MΩ cm) was obtained using a Milli-Q system (Millipore, Bedford, USA) and it was used to prepare all solutions. Reagents were of analytical grade and acquired from Merck (Darmstadt, Germany). Nitric acid 65% has been purified by sub-boiling distillation system (Milestone, Sorisole, Italy). MIC digestions were carried out under oxygen pressure with purity better than 99.6% (White Martins, São Paulo, Brazil). Plasma-based instruments were operated using argon with purity higher than 99.998% (White Martins, São Paulo, Brazil). External calibration for the determination of Ag, As, Bi, Cd, Ga, Hg, In, Pb, Sb, Sn and Zn was carried out using suitable dilution of 1000 mg L−1 standard solutions (Spex CertiPrep, Metuchen, USA). Four samples of flexible graphite sheet (named as FG I, FG II, FG III and FG IV) donated by two private companies were used in this work. Accuracy of MIC method was evaluated using CRMs with high carbon content matrices — NIST 1632c (National Institute of Standard and Technology, trace elements in coal) and BCR 40 (Community Bureau of Reference, trace elements in coal).
For recovery experiments, spiked samples were prepared by adding a known amount of a standard solution (the same that was used for calibration) containing all potentially embrittling elements, over the sample pellet before the MIC and dry ashing digestions. FG sample “IV” was arbitrarily chosen for these experiments and the results obtained by the analyte spike were used to calculate the recovery of all analytes. 2.2. Instrumentation FG sheet samples were milled in a cryogenic mill (Spex CertiPrep, model 6750, Metuchen, USA) and this procedure was repeated until the whole sample showed particle size lower than 80 μm. An analytical balance (model AY 220, Shimadzu, São Paulo, Brazil) was used to weigh samples and reagents. All embrittling elements (exception of Hg) were determined using an inductively coupled plasma optical emission spectrometer (Ciros CCD model, Spectro, Kleve, Germany) equipped with a Scott-type double pass spray chamber, a cross-flow nebulizer, a torch and a quartz injector. The operational conditions were summarized in Table 1. Determination of Hg was carried out using a homemade flow-injection (FI) system for chemical vapor generation (CVG) with detection by atomic absorption spectrometry (AAS). It consisted of a peristaltic pump (Ismatec, Zurich, Switzerland), an injection valve fitted with 100 μL sample loop and a U type gas–liquid separator. Mercury determination was carried out using an atomic absorption spectrometer (Vario 6 model, Analytik Jena, Jena, Germany) equipped with a heated quartz cell furnace (HS5, 100 mm length, 10 mm i.d.). A mercury hollow cathode lamp operated at 4.0 mA was used as a radiation source and a deuterium lamp was used for background correction. The wavelength was set at 253.7 nm and the spectral band pass at 0.5 nm. All measurements were performed in integrated absorbance (peak area) [20]. Some embrittling elements were also determined in CRMs using inductively coupled plasma mass spectrometry (ICP-MS) because their concentrations were lower than the LOD provided by ICP-OES (Table 1). Determination was carried out using inductively coupled plasma mass spectrometer (Elan DRC II model, Perkin Elmer). Digestion of FG samples using MIC was performed using a Multiwave 3000 microwave sample preparation system (Anton Table 1 Operational parameters for analyte determination by ICP-OES and ICP-MS. Parameter
ICP-OES
ICP-MS
RF power wavelength (W) Plasma gas flow rate (L min−1) Auxiliary gas flow rate (L min−1) Nebulizer gas flow rate (L min−1) Spray chamber Nebulizer Wavelength (nm)
1600 14 1.00 0.80 Scott-type double pass Cross-flow Ag 328.068 As 188.979 Bi 223.061 Cd 228.802 Ga 417.206 In 230.606 Pb 220.353 Sb 206.836 Sn 189.927 Zn 206.200 – – – – – – –
1400 15 1.20 1.09 Cyclonic Concentric –
Dwell time (ms) Sweeps/reading Readings/replicate Replicates Data collection mode Sampler and skimmer cones Isotopes (m/z)
50 5 3 3 Peak hopping Pt 107 Ag 209 Bi 111 Cd 71 Ga 121 Sb 118 Sn
M.S.P. Enders et al. / Microchemical Journal 124 (2016) 321–325
Paar, Graz, Austria) equipped with up to eight pressurized quartz vessels. Maximum pressure and internal volume allowed by quartz vessels were 80 bar and 80 mL, respectively. Dry ashing was performed in a muffle furnace (LF0912 model, Jung, Brazil) using platinum crucibles. Microwave-assisted digestion was also carried out using a UltraWave™ system (Milestone, Sorisole, Italy) equipped with five quartz vessel (total volume of 40 mL). The microwave cavity (1 L) is covered with polytetrafluoroethylene (PTFE) vessel and before digestion was sealed and pressurized with 40 bar of argon 99.996% (White Martins, São Paulo, Brazil). All digestions using UltraWave™ system were carried out using maximum power, temperature and pressure of 1500 W, 275 °C and 180 bar, respectively [21].
323
3. Results and discussion Potentially embrittling elements in high purity FG arise mainly from raw material used for its production. In this sense, the determination of these elements requires that the organic matrix should be oxidized in order to bring the analytes into a suitable solution for subsequent atomic spectrometric techniques. In this study, the main efforts were devoted to MIC method optimization. However, some attempts were also performed to evaluate dry ashing and MAD for FG samples. It is important to mention that MAD did not assure the complete digestion of FG samples and insoluble residues were observed in final digests. In this sense, the analyte quantification was not performed and the results for MAD were not discussed in this section. 3.1. Dry ashing digestion
2.3. Dry ashing Sample masses up to 1 g of milled FG were directly weighed in to the platinum crucibles and positioned inside the muffle furnace and heated at 600, 800 or 1000 °C, during 4 h. Heating ramp of 10 °C min−1 was used until reaching the final temperature (600, 800 or 1000 °C). Some experiments were also performed using a previous oxidation of FG sample with concentrated HNO3 (5 mL) heated in a hot plate (95 °C, during 1 h). The residue was dissolved with 5 mL of 7 mol L−1 HNO3 under heating in a water bath for 30 min and diluted up to 25 mL with water. In order to minimize analyte losses, samples were initially wetted with concentrated H2SO4 (5 mL) and burnt in a Bunsen burner. Subsequently, the remaining residues were digested using a muffle furnace at 1000 °C. The efficiency of digestion using dry ashing was evaluated by the characteristics of final residues and by weighing of the platinum crucibles after heating program. Recovery experiments using analyte spike with standard solutions were performed for all experiments using dry ashing method.
2.4. Microwave-assisted digestion (MAD) Sample mass up to 100 mg was weighed and added to the quartz vessels. Concentrated HNO3 (6 mL) was used as oxidant agent. All digestions were performed using maximum temperature and pressure available in the microwave system (275 °C and 180 bar, respectively). The heating program was as follows: i) ramp of 10 min up to 1500 W; ii) 1500 W for 40 min and iii) 0 W for 20 min (cooling).
2.5. Microwave-induced combustion (MIC) For MIC, 100 mg of samples were pressed as pellets (8 mm diameter) using a hydraulic press set at 3 tons. Additionally, pellets of microcrystalline cellulose (100, 200 or 300 mg) were also produced and used together with the sample pellets in order to aid the combustion of FG samples. FG and cellulose pellets were placed on a small disk of filter paper on the quartz holder. Filter paper was wetted with 50 μL of 6 mol L− 1 NH4NO3 and the quartz holder was positioned inside the quartz vessel filled with 6 mL of absorbing solution. The absorbing solutions of HNO3 (2, 4, 7 and 14 mol L−1) or inversed aqua regia (4.5 mL of concentrated HNO3 and 1.5 mL of concentrated HCl) were evaluated. Quartz vessels were closed, placed in the rotor and pressurized with 20 bar of oxygen. Rotor was capped and placed inside the microwave oven and the following heating program was started: i) 1400 W for 10 min (ignition and reflux steps) and ii) 0 W for 20 min (for cooling). Final solutions were diluted up to 25 mL with water and analyzed by ICP-OES and CVG-AAS. Decontamination of quartz vessels and holders was performed using concentrated HNO3 with the same microwave heating program used for combustion of samples. Recovery tests were carried out using spike of all analytes.
Initially, the digestion of FG samples was evaluated using oxidation of organic sample matrix in the presence of oxygen atmosphere inside the muffle furnace at high temperature. The temperatures, generally recommended for dry ashing of common organic matrices (such as leaves, meat and flour) are about 500 °C [9]. However, the oxidation of graphite at temperatures between 400 and 500 °C is very small (lower than 1%) [22]. According to the literature the amount of oxidized carbon of graphite increased greatly from 500 to 800 °C (with air flow of 20 mL min−1) [22]. In this sense, the oxidation of FG sample was carried out using temperatures of 600, 800 and 1000 °C which are higher than those recommended for common organic matrices due to the refractory characteristics of FG matrix. Significant oxidation was not observed for the FG samples heated using temperature up to 600 °C and FG matrix maintained its original appearance. Moreover, the mass of platinum crucibles containing the sample after the dry ashing program did not decrease showing that there were no losses of sample mass. On the other hand, an oxidation of about 50% of sample mass was observed after dry ashing at 800 °C. In order to assure the complete oxidation of graphite matrix, dry ashing was carried out at 1000 °C. In this temperature the sample mass loss was higher than 95% and the remaining residue presented as white residue indicating the complete oxidation of FG matrix. In order to decrease the temperature of dry ashing, experiments were performed using a previous oxidation of FG sample with concentrated HNO3 in hot plate heating (95 °C). However, complete oxidation of FG sample was only observed for dry ashing at 1000 °C even using previous oxidation with HNO3. Although practically complete oxidation of FG, recovery tests were performed to verify the losses of analyte during digestion by dry ashing. In this work, recoveries obtained for all analytes were lower than 80%. Additionally, Ag, As, Cd, Hg, Pb and Zn presented recovery values lower than 10% probably due to the high temperature required for complete oxidation of FG matrix (1000 °C). To overcome the problem related with losses of analytes, experiments were carried out adding concentrated sulfuric acid before the dry ashing digestion. However, quantitative recoveries for all potentially embrittling elements were not observed even using sulfuric acid as additive to minimize losses of some analytes. In this sense, dry ashing was not suitable for FG sample digestion and subsequent determination of potentially embrittling elements. In order to assure the accuracy of results, MIC was evaluated and optimized for FG sample digestion. 3.2. Microwave-induced combustion In order to achieve the limits of detection required to comply the recommendation of General Electric for nonmetallic materials, sample mass of FG that can be efficiently digested by MIC should be increased. Some initial experiments showed that the use of easily combustible materials (e.g. cellulose) combined with FG sample can provide enough energy to aid the combustion of refractory samples.
324
M.S.P. Enders et al. / Microchemical Journal 124 (2016) 321–325
14 mol L-1
7 mol L-1
2 mol L-1
4 mol L-1
Inversed aqua regia
110 100 90
Recovery (%)
80 70 60 50 40 30 20 10 0 Ag
As
Cd
Ga
Hg
In
Pb
Zn
Bi
Sb
Sn
Fig. 1. Influence of concentration and type of absorbing solution using MIC for subsequent determination of embrittling elements by atomic spectrometric techniques (error bars are the standard deviation, n = 5).
3.2.1. Evaluation of cellulose mass Initially, pellets of FG sample (up to 100 mg) were placed together with pellets of cellulose (100, 200 or 300 mg) on the quartz holder. Efficient digestion of FG sample (without residues on quartz holder) was only observed for the combination of sample pellets (100 mg) and cellulose (300 mg). In this condition, the final pressure of the system was 60 bar that is equivalent to 75% of maximum pressure allowed by the manufacturer. For subsequent experiments pellets of 300 mg of cellulose were used as aid for the digestion of 100 mg of FG sample.
6 mL of 4 mol L− 1 HNO3 as absorbing solution. Additionally, even using concentrated HNO3, the recoveries for Bi, Sb and Sn were lower than 80%. These low recoveries can be explained by the formation of species that are not soluble in HNO3, requiring the use of inversed aqua regia [23]. In this respect, the use of aqua regia was evaluated as absorbing solution providing quantitative recoveries (better than 95%) for Bi, Sb and Sn (Fig. 1). In this way, a protocol for quality control can be proposed where, each FG sample should be digested by MIC method using: i) 4 mol L−1 HNO3 as absorbing solution for Ag, As, Cd, Ga, Hg, In, Pb, and Zn and ii) inversed aqua regia for Bi, Sb and Sn.
3.2.2. Evaluation absorbing solution The choice of absorbing solution for MIC depends on the characteristics of analyte. Diluted HNO3 solutions are preferred for metals and metalloids in order to minimize interferences on determination by plasma-based techniques and assure the complete analyte absorption. However, for some elements (such as Sn and Sb) inversed aqua regia should be used to assure the quantitative recovery of these analytes. In this work, HNO3 solutions (2, 4, 7 and 14 mol L−1) were evaluated as absorbing solution for all embrittling elements. On the other hand, inversed aqua regia (4.5 mL of concentrated HNO3 and 1.5 mL of concentrated HCl) was used to absorb only the analytes that showed low recovery values (Sn, Sb and Bi). In Fig. 1, the recoveries for all analytes using these solutions are shown. For all elements, with the exception of Bi, Sb and Sn, the recovery was quantitative (better than 95%) using
3.3. Decomposition of FG samples using MIC method Four FG samples were digested using the optimized MIC method with cellulose for further determination of embrittling elements by ICP-OES and CVG-AAS. The maximum concentration of each embrittling element should be lower than 200 mg kg− 1 and the sum of these analytes cannot exceed 500 mg kg−1. According to the results shown in Table 2, all analyzed samples are in agreement with the recommendation of General Electric for nonmetallic materials. FG sample I presented the highest sum of embrittling elements when compared to other samples. The limit of detection obtained for all embrittling element was lower than 12 mg kg−1 and the sum of all LODs was lower than 37 mg kg−1, that are suitable for FG quality control analysis. The relative standard deviations of the results obtained by MIC digestion
Table 2 Concentration of embrittling elements in FG samples and CRMs determined by ICP-OES, ICP-MS and CVG-AAS after decomposition using MIC (mg kg−1, n = 5, mean ± standard deviation). Element
MIC FG I
Ag As Bi Cd Ga Hg In Pb Sb Sn Zn a *
b0.5 6.4 ± 0.9 b7.0 2.6 ± 0.3 b8.0 b0.1 b12 4.3 ± 0.5 b2.5 b3.0 21 ± 2
Certified values FG II b0.5 b2.0 b7.0 b0.20 b8.0 b0.1 b12 3.2 ± 0.3 b2.5 b3.0 10 ± 1
Determination carried out using ICP-MS. Informed value.
FG III b0.5 b2.0 b7.0 b0.20 b8.0 b0.1 b12 b1.0 b2.5 b3.0 b0.30
FG IV b0.5 b2.0 b7.0 b0.2 b8.0 b0.1 b12 b1.0 b2.5 b3.0 b0.30
BCR 40 b0.5 13 ± 2 b7.0 0.12 ± 0.01a b8.0 0.38 ± 0.08 b12 25 ± 2 b2.5 b3.0 32 ± 3
NIST 1632 c a
0.11 ± 0.01 6.5 ± 1.0 0.09 ± 0.01a 0.07 ± 0.01a 3.0 ± 0.4a b0.1 b12 3.6 ± 0.3 0.45 ± 0.06a 1.2 ± 0.1a 11 ± 1
BCR 40
NIST 1632 c
n.d. 13.2 ± 1.1 n.d. 0.11 ± 0.02 n.d. 0.35 ± 0.06 n.d. 24.2 ± 1.7 n.d. n.d. 30.2 ± 1.9
0.1* 6.18 ± 0.27 0.1* 0.072 ± 0.007 3* 0.0938 ± 0.0037 n.d. 3.79 ± 0.07 0.461 ± 0.029 1* 12.1 ± 1.3
M.S.P. Enders et al. / Microchemical Journal 124 (2016) 321–325
and ICP-OES determination were lower than 15%. As FG matrix is not available as certified reference material, accuracy was evaluated using certified reference material of coal (BCR 40 and NIST 1632c) that presents high carbon content. Additionally, it was not possible to evaluate accuracy for In because this element is not informed in CRMs used in this work. It is also important to point out that some elements (Ag, Bi, Cd, Ga, Sb and Sn) were also determined by inductively coupled plasma mass spectrometry (ICP-MS) in CRM digests in order to achieve the LOD required to evaluate the accuracy by comparison of certified reference values and the results obtained using the proposed digestion method. In this sense, significant difference (t test, 95% confidence level) was not observed between certified reference values and the results obtained using MIC and atomic spectrometric techniques. 4. Conclusion Among the digestion methods evaluated for FG digestion only MIC was considered suitable for the digestion of up to 100 mg of sample. For MAD and dry ashing, insoluble residues in final digest and losses of analytes were observed, respectively. Cellulose should be used as combustion aid for MIC to assure the complete oxidation of FG matrix that is very refractory even using MIC. Two independent absorbing solutions (4 mol L−1 HNO3 and inversed aqua regia) were required to allow the quantitative recovery of all embrittling element for MIC digestion. LODs obtained by MIC and determination by ICP-OES and FI-CVG-AAS were satisfactory for quantification of all potentially embrittling elements and are in agreement to the recommendation of General Electric for nonmetallic materials. Acknowledgments The authors are grateful to CNPq (304424/2012-9), FAPERGS (19402551/13-4) and CAPES (23038.002513/2014-37) for supporting this study. References [1] M. Cai, D. Thorpe, D.H. Adamson, H.C. Schniepp, Methods of graphite exfoliation, J. Mater. Chem. 22 (2012) (24992-25002.2). [2] D.D.L. Chung, Review exfoliation of graphite, J. Mater. Sci. 22 (1987) 4190–4198. [3] R. Chugh, D.D.L. Chung, Flexible graphite as a heating element, Carbon 40 (2002) 2285–2289. [4] X. Luo, D.D.L. Chung, Flexible graphite under repeated compression studied by electrical resistance measurements, Carbon 39 (2001) 985–990. [5] S.P. Lynch, Metal-induced embrittlement of materials, Mater. Charact. 28 (1992) 279–289. [6] C.F. Old, Liquid metal embrittlement of nuclear materials, J. Nucl. Mater. 92 (1980) 2–25.
325
[7] H.C. Furtado, I. May, High temperature degradation in power plants and refineries, Mater. Res. 7 (2004) 103–110. [8] General Electric nuclear non-metallic material specification D50YP12 (rev. 2), Material Specification for Non-metallic Impurity Limits in Engineering Materials Exposed to Boiling Water Reactor1983. [9] M. Hoenig, Dry ashing, in: Z. Mester, R. Sturgeon (Eds.), Sample Preparation for Trace Element Analysis, Elsevier 2003, pp. 235–254. [10] J.S. Barin, B. Tischer, R.S. Picoloto, F.G. Antes, F.E.B. Silva, F.R. Paula, E.M.M. Flores, Determination of toxic elements in tricyclic active pharmaceutical ingredients by ICP-MS: a critical study of digestion methods, J. Anal. At. Spectrom. 29 (2014) 352–358. [11] E.M.M. Flores, E.I. Muller, F.A. Duarte, P. Grinberg, R.E. Sturgeon, Determination of trace elements in fluoropolymers after microwave-induced combustion, Anal. Chem. 85 (2013) 374–380. [12] A.L.H. Muller, C.C. Muller, F. Lyra, P.A. Mello, M.F. Mesko, E.I. Muller, E.M.M. Flores, Determination of toxic elements in nuts by inductively coupled plasma mass spectrometry after microwave-induced combustion, Food Anal. Methods 6 (2013) 258–264. [13] J.V. Maciel, C.L. Knorr, E.M.M. Flores, E.I. Muller, M.F. Mesko, E.G. Primel, F.A. Duarte, Feasibility of microwave-induced combustion for trace element determination in Engraulis anchoita by ICP-MS, Food Chem. 145 (2014) 927–931. [14] J.S.F. Pereira, P.A. Mello, D.P. Moraes, F.A. Duarte, V.L. Dressler, G. Knapp, E.M.M. Flores, Chlorine and sulfur determination in extra-heavy crude oil by inductively coupled plasma optical emission spectrometry after microwave-induced combustion, Spectrochim. Acta B 64 (2009) 554–558. [15] J.S.F. Pereira, D.P. Moraes, F.G. Antes, L.O. Diehl, M.F.P. Santos, R.C.L. Guimarães, T.C.O. Fonseca, V.L. Dressler, E.M.M. Flores, Determination of metals and metalloids in light and heavy crude oil by ICP-MS after digestion by microwave-induced combustion, Microchem. J. 96 (2010) 4–11. [16] S.M. Cruz, L. Schmidt, F.M. Dalla Nora, M.F. Pedrotti, C.A. Bizzi, J.S. Barin, E.M.M. Flores, Microwave-induced combustion method for the determination of trace and ultratrace element impurities in graphite samples by ICP-OES and ICP-MS, Microchem. J. 123 (2015) 28–32. [17] T.S. Nunes, C.C. Muller, P. Balestrin, A.L.H. Muller, M.F. Mesko, P.A. Mello, E.I. Muller, Determination of chlorine and sulfur in high purity flexible graphite using ion chromatography (IC) and inductively coupled plasma optical emission spectrometry (ICP OES) after pyrohydrolysis sample preparation, Anal. Methods 7 (2015) 2129–2134. [18] R.S. Picoloto, H. Wiltsche, G. Knapp, P.A. Mello, J.S. Barin, E.M.M. Flores, Determination of inorganic pollutants in soil after volatilization using microwave-induced combustion, Spectrochim. Acta B 86 (2013) 123–130. [19] J.S. Barin, E.M.M. Flores, M.F. Mesko, P.A. Mello, J.S.F. Pereira, Microwave-induced combustion, in: E.M.M. Flores (Ed.), Microwave-assisted Sample Preparation for Trace Element Determination, Elsevier 2014, pp. 143–173. [20] L.E. Kaercher, F. Goldschmidt, J.N.G. Paniz, E.M.M. Flores, V.L. Dressler, Determination of inorganic and total mercury by vapor generation atomic absorption spectrometry using different temperatures of the measurement cell, Spectrochim. Acta B 60 (2005) 705–710. [21] J.A. Nobrega, C. Pirola, L.L. Fialho, G. Rota, C.E.K.M.A. Campos Jordao, F. Pollo, Microwave-assisted digestion of organic samples: how simple can it become? Talanta 98 (2012) 272–276. [22] L. Xiaowei, R. Jean-Charles, Y. Suyuan, Effect of temperature on graphite oxidation behavior, Nucl. Eng. Des. 227 (2004) 273–280. [23] P.J. Walter, S. Chalk, H.M.S. Kingston, Overview of microwave-assisted sample preparation, in: H.M.S. Kingston, S.J. Haswell (Eds.), Microwave-enhanced Chemistry. Fundamentals, Sample Preparation and Applications, American Chemical Society, Washington 1997, pp. 55–216.