Comparative analysis on homogeneity of Pb and Cd in
epoxy molding compounds
Kyun-Gmee LEE1, Yong-Keun SON1, Jin-Sook LEE2, Tai-Min NOH3, Hee-Soo LEE3
1. Department of Chemistry, Sungkyunkwan University, Gyeonggi-do, Korea;
2. Department of Chemistry, Yonsei University, Seoul, Korea;
3. School of Materials Science and Engineering, Pusan National University, Busan 609-735, Korea
Received 21 April 2010; accepted 10 September 2010
Abstract: Reference materials for quantitative determination of regulated heavy metals, such as Pb and Cd in electronic components, were designed and investigated in terms of stability and homogeneity. Reference materials with two concentration levels of heavy metals were prepared by spiking Pb and Cd compounds to epoxy molding compounds made by mixing silica powders and epoxy resin. The concentration changes of the reference materials during stability test for 1 a were not observed. In the homogeneity assessment, the as-prepared reference materials were studied by using three different analytical tools, inductively coupled plasma atomic emission spectrometry (ICP-AES), X-ray fluoroescence spectrometry (XRF) and laser ablation ICP mass. The results show different homogeneities by the characteristics of analytical tools and the materials.
Key words: epoxy molding compound; Pb; Cd; reference materials; homogeneity
1 Introduction
In recent years the hazardousness of heavy metal in various places has become an important factor for the electrical and electronic industries. EU directive[1] has banned uses of four metals (Pb, Cd, Hg and Cr6+) and brominated flame retardants such as polybrominated biphenyls (PBBs) and polybrominated diphenyl ethers (PBDEs) for electrical and electronic devices and products since July 2006. To control and monitor restricted substances, an international standard testing method (IEC 62321) was released in 2008[2].
Also to validate the analytical results, many types of reference materials (RMs) have been developed[3], but such RMs for composite materials are few because it is difficult to define and generate homogeneity for composite materials, while it is also challenging to access the homogeneities of the materials. The homogeneity is an important factor in the preparation of RMs. To extract a usable RM all analysts participating in using the materials will obtain the same property values from the materials. This implies that replicate subsamples, within and between lots of RMs, show no significant difference within the boundaries of the precision of the testing methods.
For the purpose of extracting such RMs, the testing methods to define homogeneity for composite materials were used. Generally, inductively coupled plasma atomic emission spectrometry (ICP-AES) and X-ray fluoroescence spectrometry (XRF), which are considered to be the most popular analytical methods, were used to analyze heavy metals. In this work, more precise analytical tools than ICP-AES or XRF for the purpose of reaching the homogeneity for RMs were strived to be developed. For example, laser ablation ICP mass is also currently used as another option for an analytical tool for extracting the homogeneity because it gives micro sampling and shows micro heterogeneity of the samples. Moreover, ICP mass provides the highest sensitivity for heavy metals among the inorganic testing instruments.
Nowadays, different types of plastic epoxy molding compounds (CRMs) containing heavy metals, such as VDA001-004[4] and BCR 680-681 (IRMM, Belgium)[5] for polyethylene (PE) containing heavy metals, JASC 0601-0602[6] for polyester or ABS as a matrix, JSAC 0611-0615[7] for polyesters containing Pb, Cd and Cr, and JSAC 0621-0625[8] for polyesters containing Hg, are available, where there are also two types of CRMs for glass and powder types of paints available. However, no CRMs are reported for Pb and Cd in EMCs, which is commonly found in electrical and electronic components to determine the amount in quantitative ways. Epoxy molding compounds are known as a kind of material composed of a composite material and used in representative semiconductor packaging. They are composed of silica, epoxy plastics and other additives. In this work, the preparation of RMs and the homogeneity of Pb and Cd were investigated through analytical tools. Also the comparative methods were performed to investigate the effect of the composite material on the result of the each analytical tool.
2 Experimental
The composition of EMCs as a starting material is listed in Table 1 with the total mass of 1 kg. The additives were of reagent grade. Pb was Pb(C18H35O2)2 from WAKO Pure Chemical Industries, Ltd. Cd was in CdO from KANTO Chemical Co., Inc.
Table 1 Components of candidate EMC CRMs
EMCs were prepared by mixing with blended filler and coupling agent (No.1+No.2) and blended epoxy resin, hardener, catalyst, and wax (No.3+No.4+No.5+ No.6) twice for 1 min in a stainless-steel mixer. Pb(C18H35O2)2 and CdO were weighed and spiked in the prepared EMCs and mixed 10 times for 1 min. The EMCs with hazardous metals passed through a kneader machine at 121 °C and then cooled down. The cooled EMC CRMs were crushed into powder less than 3 mm and post-cured at 175 °C for 4 h. Finally, the EMC CRMs had the particle size of less than 1 mm after the second crushing. 40 g homogeneously-made EMC CRMs powders were put into a brown bottle.
The homogeneity, stability and certified values are very important factors in the candidate CRM. They were investigated by non-destructive and destructive methods. For a non-destructive method that reduces the time-consuming for sample pretreatment, prepared EMC CRMs powders were pressed in a form of a disk with 40 mm in diameter and 3 mm in thickness using uni-axial pressure at 100 MPa. The homogeneity of prepared disks was examined by checking the intensity of Pb and Cd using wavelength dispersive X-ray fluorescence (WD-XRF, Rigaku, ZSX 100E) and inductively coupled plasma-mass spectrometer (ICP-MS, Perkin-Elmer, Elan DRC) laser ablation. For the destructive analytic method, test samples of 0.5 g at high level and 1.0 g at low level within 1 mg error range were transferred into a polytetrafluoro ethylene (PTFE, Teflon) beaker. H2SO4 and HNO3 solutions were added to the beaker for the decomposition of organic compounds, and HF solution was added to the beaker for the decomposition and the removal of sample matrix Si. The test procedure is shown in Fig.1. Concentrations of Pb and Cd were measured by inductively coupled plasma-atomic emission spectrometer (ICP-AES, PERKIN-ELMER Optima 4300). The inductively coupled plasma-mass spectrometer (ICP-MS, Perkin-Elmer Elan DRCⅡ) was used to determine the certified values of candidate EMC CRMs.
Fig.1 Flow chart of chemical analysis for candidate EMC CRMs
3 Results and discussion
3.1 Homogeneity
Extracting and defining the homogeneity is one of the most important factors to prepare CRMs for EMC. Firstly, the homogeneity tests should enable the estimation of the variation of average concentrations between bottles. Secondly, it should establish the inhomogeneity within a bottle to establish the minimum sample intake, thus a stability test and determination of certified value can progress only for the homogeneously- prepared CRMs.
X-ray is widely irradiated to a sample surface, which is 20 mm in diameter while the surface irradiated by laser is approximately 4. It was measured with three times intensities of Pb and Cd repeatedly under conditions of 25 samples with 40 mm in diameter and 3 mm in thickness. The results obtained from both homogeneity studies are shown in Table 2. Relative standard deviations for high level are 3.78% and 4.82% for Pb, whereas 5.23% and 2.84% for Cd. The relative standard deviations for low level are 1.66% and 3.57% for Pb, while 8.63% and 9.29% for Cd. As a result, there is no significant difference in the relative standard deviations found. It is found that the EMC CRMs are homogenously made in terms of measuring mechanism without considering the surface roughness or the particle size that affects the measurement.
Table 2 Homogeneity intensities of Pb and Cd by XRF and ICP-MS laser ablation (Unit : cps)
The between-bottles and within-bottle homogeneity tests were performed according to the layout of Guide 35[9]. Twenty bottles were selected from the batch and the destructive analysis was carried out three times for each sample. Although Pb is precipitated into PbSO4, it can be all dissolved in 15% HNO3 solution. The analytical results are given in Table 3 and Fig.2. Relative standard deviations of two EMC CRMs are ranged from 0.78 % to 1.43 % for Pb and Cd. The analysis of variance test, ANOVA, was applied to the analytical results for a between-bottle and within-bottle homogeneity evaluation of the batch. F value is 1.42 for Pb and 0.88 for Cd at high level, and 1.25 for Pb and 1.11 for Cd at low level. These results are smaller than 1.85, which is F critical value (Fc). Since F values of two EMC CRMs are smaller than F critical value (Fc), the sample can be demonstrated to be homogeneous in between-bottles. From the results of analysis of variance, the standard deviations for repeatability of within-bottle (Sr) and between-bottles (Sbb) were calculated. The Sr and Sbb values are ranged from 0.02 to 0.48 mg/kg for Pb and Cd, respectively. These values exhibit that the repeatability standard deviation of the measurement method is con- siderably smaller than the combined standard uncertainty from the determination of the property value. As a consequence, within-bottle and between-bottles homogeneity of candidate EMC CRMs should be established.
Table 3 Homogeneity results of candidate EMC CRMs by ICP-AES: ANOVA calculations
Fig.2 Graphical illustrations of homogeneity for candidate EMC CRMs by ICP-AES: (a) Low level; (b) High level
3.2 Stability
Since candidate CRMs with a certified value should not change during storage and transport, a short term and long-term safety tests must be carried out. In an early stage, they were stored at room temperature, and after 60 d stored below 25 °C. The test method was exactly the same as the evaluation of between-bottles and within-bottle homogeneity tests, and its stability for 1 a was studied during the research. The tendency of concentration change with time is listed in Table 4. Since the F-table can be used for evaluating the significance of regression. It becomes significant for p<0.05 at 95% confidence level[10].
Table 4 Test results on stability of candidate EMC CRMs by ICP-AES: F-test calculation of regression analysis
F-test and linear regression analysis by software (Microsoft, Excel) was applied to the analytical results of the candidate EMC CRMs. The p-values and uncertainty of slope (sb1) were calculated. The p-values were ranged from 0.15 to 0.95 for Pb and Cd of two difference level for EMC CRMs. Since the p-values are bigger than 0.05 at 95% confidence level, regression is insignificant. The uncertainty of slope (sb1) due to long-term stability is very small. Therefore, the candidate EMC CRMs are no trend of concentration change. The good stabilities during the research (1 a) were obtained.
3.3 Certification by isotope dilution mass spectrometry (IDMS)
The property values of candidate EMC CRMs were certified by IDMS[11-13] according to ISO Guide 35. The isotopes, 99.97% 206Pb and 96.31% 111Cd were used, 112Cd/111Cd and 208Pb/206Pb mass spectrometric measurements were measured by ICP-MS.
IDMS results were calculated by Eq.(1):
(1)
where c is the concentration of analyte in sample; ms is the mass of the spike; m is the mass of the sample; K is the ratio of the natural atomic mass and the atomic mass of 111Cd-enriched material; A is the natural abundance of 112Cd (the reference isotope); B is the natural abundance of 111Cd (the spike isotope); As is the natural abundance of 112Cd in the 111Cd -enriched spike; Bs is the natural abundance of 111Cd in the 111Cd -enriched spike and R is the measured mass ratio of 112Cd to 111Cd.
The combined standard uncertainty associated with the property value can be computed by Eq.(2):
(2)
3.4 Certified values and uncertainties
The final uncertainty (UCRM) of a certified value includes the uncertainty of the sample measurement (uchar), the uncertainty of between-bottle homogeneity (ubb) , and the uncertainty of long-term stability (ults):
(3)
According to Eq.(3), expanded uncertainties can be computed.
The uncertainty values and expanded uncertainties are summarized in Table 5. Since the degree of freedom is bigger than 10, coverage factor k=2 was used in these results at 95% level of confidence[9]. The results were obtained to be 3%-5%. Consequently, the certified value of EMC CRMs has a shelf life time of 1 a, showing (188.0±2.5) mg/kg for Pb and (74.5±1.4) mg/kg for Cd at the high level, while (23.4±0.5) mg/kg for Pb and (17.3±0.4) mg/kg for Cd at the low level.
Table 5 Certified values and uncertainties of EMC CRMs by ICP-MS
4 Conclusions
1) The overall results demonstrate that the CRMs developed are very useful for the quantitative determination of Pb and Cd in epoxy molding compounds.
2) CRMs prepared in the form of powder show good homogeneity and stability characteristics.
3) The certified values of the EMC CRMs have a shelf life time of 1 a and the uncertainties of the 95% confidence interval are ranged from 3% to 5% for the two concentration levels.
References
[1] Directive 2002/95/EC, Directive on the restriction of the use of certain hazardous substances in electrical and electronic equipment [S]. 2003.
[2] IEC62321 Edition 1.0, Electrotechnical products—Determination of levels of six regulated substances (lead, mercury, cadmium, hexavalent chrome, polybrominated biphenyls, polybrominated diphenyl ethers) [S]. 2008.
[3] IEC62321, Ed.1, Procedures for the determination of levels of six regulated substances (lead, mercury, cadmium, hexavalent chrome, polybrominated biphenyls, polybrominated diphenyl ethers) CDV(Committee Draft for Vote) [S]. 2006.
[4] PAUWELS J, LAMBERTY A, de BIEVRE P, GROBECKER K H, BAUSPIES C. Certified reference materials for the determination of cadmium in polyethylene [J]. Fresenius J Anal Chem, 1994, 349: 409-411.
[5] LAMBERTY A, van BORM W, QUEVAUVILLER P. Collaborative study to improve the quality control of trace element determinations in polymers [J]. Fresenius J Anal Chem, 2001, 370: 811-818.
[6] NAKANO K, NAKAMURA T, NAKAI I, KAWASE A, IMAI M, HASEGAWA M, ISHIBASHI Y, INAMOTO I, SUDOU K, KOZAKI M, TURUTA A, ONO A, KAKITA K, SAKATA M. Preparation and certification of the new certified reference materials, plastics (JSAC 0601-1, JSAC 0602-1) for determination of hazardous metal constituents [J]. Bunseki Kagaku, 2006, 55(10): 799-806.
[7] NAKANO K, NAKAMURA T, NAKAI I, KAWASE A, IMAI M, HASEGAWA M, ISHIBASHI Y, INAMOTO I, SUDOU K, KOZAKI M, TURUTA A, HOMMA H, ONO A, KAKITA K, SAKATA M. Plastic certified reference materials JSAC 0611-0615 for determination of hazardous constituents using X-ray fluorescent analysis [J]. Bunseki Kagaku, 2006, 55(7): 501-507.
[8] NAKANO K, NAKAMURA T, NAKAI I, KAWASE A, IMAI M, HASEGAWA M, ISHIBASHI Y, INAMOTO I, SUDOU K, KOZAKI M, TURUTA A, ONO A, KAKITA K, SAKATA M. Preparation and certification of the new reference materials; plastics(disk form, JSAC 0621-0625) for determination of mercury using X-ray fluorescent analysis [J]. The Japan Society for Analytical Chemistry, 2006, 22: 1265-1268.
[9] ISO Guide 35. Reference materials—General and statistical principles for certification [S]. 2006.
[10] ISO Guide 98—3. Uncertainty of measurement: Part 3. Guide to the expression of uncertainty in measurement [S]. 1995.
[11] FASSETT J D, PAULSEN P. Isotope dilution mass spectrometry for accurate elemental analysis [J]. Anal Chem, 1989, 61: 643.
[12] MOODT J R, EPSTEIN M S. Definitive measurement methods [J]. Spectrochimica Acta: Part B, 1991, 46: 1571.
[13] PAPADAKIS I, TAYLOR P D P, DE BIEVRE P. SI-traceable values for cadmium and lead concentration in the candidate reference material, MURST-ISS A1 Antarctic sediment by combination of ICP-MS with isotope dilution [J]. Analytical Chimica Acta, 1997, 346: 17-22.
(Edited by CHEN Wei-ping)
Foundation item: Project(2010-0008-276) supported by NCRC (National Core Research Center) program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology
Corresponding author: Hee-Soo LEE; Tel: +82-51-510-225; E-mail: heesoo@pusan.ac.kr
