Results of Proficiency Test Styrene October 2007

Transcription

1 Results of Proficiency Test Styrene October 27 Organised by: Authors: Correctors: Report: Institute for Interlaboratory Studies (iis) Spijkenisse, the Netherlands Ing. R.J. Starink Ing. S. Lagendijk & dr. R.G. Visser iis7c1x November 27

2 Institute for Interlaboratory Studies Spijkenisse, November 27 CONTENTS 1 INTRODUCTION SET UP ACCREDITATION PROTOCOL SAMPLES STABILITY OF THE SAMPLES ANALYSES RESULTS STATISTICS GRAPHICS Z-SCORES EVALUATION EVALUATION PER TEST PERFORMANCE EVALUATION FOR THE GROUP OF LABORATORIES COMPARISON OF THE PROFICIENCY TEST OF OCTOBER 27 WITH PREVIOUS PT S Appendices: 1. Data and statistical results Reported details Aldehydes determination Reported response factor and peaks for non-aromatics determination List of participants Abbreviations and literature page 2 of 44 Styrene: iis7c1x

3 Spijkenisse, November 27 Institute for Interlaboratory Studies 1 INTRODUCTION Since 1995, the Institute for Interlaboratory Studies organised a proficiency test for the analysis of Styrene. As part of the annual proficiency test program of 27/28, the Institute decided to continue this proficiency test on Styrene. In this international interlaboratory study 34 laboratories in 2 different countries have participated. See appendix 4 for a list of participants in alphabetical country order. In this report the results of the proficiency test are presented and discussed. 2 SET UP The Institute for Interlaboratory Studies (i.i.s.) in Spijkenisse, The Netherlands, was the organiser of this proficiency test. It was decided to send 2 different samples, one sample regular Styrene (1*5 ml, labelled #758) and one sample Styrene spiked with several components (1*25 ml, labelled #759). Participants were requested to report rounded and unrounded results. The unrounded results were preferably used for statistical evaluation. 2.1 ACCREDITATION The Institute for Interlaboratory Studies in Spijkenisse, the Netherlands, is accredited in agreement with ISO guide 43 and ILAC-G13:2, (R7), since January 2, by the Dutch Accreditation Council (Raad voor Accreditatie). This ensures 1% confidentially of participant s data. Feedback from the participants on the reported data is encouraged and customer s satisfaction is measured on regular basis by sending out questionnaires. 2.2 PROTOCOL The protocol followed in the organisation of this proficiency test was the one as described for proficiency testing in the report i.i.s. Interlaboratory Studies: Protocol for the Organisation, Statistics and Evaluation (i.i.s.-protocol, version 3.) of November SAMPLES The necessary bulk material was obtained from a local Styrene producer. The first batch, approx. 3 litre bulk sample was homogenised in a precleaned drum and divided over 47 amber glass bottles of 5 ml (labelled #758). The second batch of approx. 14 litre bulk sample was spiked with several components (see table 1). The batch was homogenised and divided over 48 amber glass bottles of 25 ml (labelled #759). Styrene: iis7c1x page 3 of 44

4 Institute for Interlaboratory Studies Spijkenisse, November 27 Component Weight in gram Added Concentration in mg/kg o-xylene m-xylene p-xylene Ethylbenzene n-propylbenzene m-ethyltoluene p-ethyltoluene Table 1: intake table of components added to 12.3 kg The homogeneity of the sub samples #758 was checked by determination of Peroxide in accordance with ASTM D234:3 and by determination of Aldehydes in accordance with ASTM D2119:3 on 4 stratified random selected samples. Aldehydes in mg/kg Peroxide in mg/kg sample # sample # sample # sample # Table 2: homogeneity test of subsamples #758 The homogeneity of the sub samples #759 was checked by determination of n- Propylbenzene in accordance with ASTM D5135:7 on 4 stratified random selected samples. n-propylbenzene in mg/kg sample # sample # sample # sample # Table 3: homogeneity test of subsamples #759 From the results in table 2 and 3 the repeatabilities were calculated by multiplication of the standard deviations by 2.8 and compared with the repeatability requirements of the standards: Aldehydes in mg/kg Peroxides in mg/kg n-propylbenzene in mg/kg r sample # r sample # Reference method ASTM D2119:3 ASTM D234:3 ASTM D5135:7 r(reference method) Table 4: repeatabilities of subsamples #758 The repeatabilities of the results for Aldehydes, Peroxide and n-propylbenzene were in agreement with the requirements as mentioned in the respective standards. Therefore, homogeneity of the sub samples #758 and #759 were assumed. page 4 of 44 Styrene: iis7c1x

5 Spijkenisse, November 27 Institute for Interlaboratory Studies To each of the participating laboratories, 1 bottle of 5 ml (labelled #758) and 1 bottle of 25 ml (labelled #759) was sent on September 12, STABILITY OF THE SAMPLES In order to be sure that the material, which was used in this proficiency test, was stable for the valid period, the stability of the material, packed in the brown glass bottles, was checked prior to use. 2.5 ANALYSES The participants were asked to determine on sample #758: Aldehydes as benzaldehyde, Colour Pt/Co, Inhibitor (according the 2 and/or 26 version), Density, Peroxide as H 2 O 2, Polymer, Sulphur, Water, Purity and the Impurities: Benzene, Ethylbenzene, m+p- Xylenes, Cumene, o-xylene, n-propylbenzene, m+p-ethyltoluenes, alpha-methylstyrene, Phenylacetylene, Benzaldehyde and Nonaromatics. On sample #759: Purity and the Impurities; Benzene, Ethylbenzene, m+p-xylenes, Cumene, o-xylene, n-propylbenzene, m+p-ethyltoluenes, alpha-methylstyrene, Phenylacetylene, Benzaldehyde and Nonaromatics. To get comparable results a detailed report form, on which the units and the standard methods were printed, and a detailed questionnaire about the details of the Aldehyde determination was sent together with each set of samples. Also a letter of instructions and a MSDS were added to the package. 3 RESULTS During four weeks after sample despatch the results of the individual laboratories were received. The original reported results were tabulated per determination in appendix 1 of this report. The laboratories are presented by their code numbers. Directly after deadline, a reminder fax was sent to those laboratories that had not yet reported results. Shortly after the deadline the available results were screened for suspect data. A result was called suspect in case the Huber Elimination Rule (a robust outlier test) found it to be an outlier. The laboratories that produced these suspect data were asked to check the results. Additional or corrected results were used for data analysis and original results were placed under 'Remarks' in the result tables in appendix STATISTICS Statistical calculations were performed as described in the report i.i.s. Interlaboratory Studies: Protocol for the Organisation, Statistics and Evaluation of November 23 (i.i.s.- protocol, version 3.). Styrene: iis7c1x page 5 of 44

6 Institute for Interlaboratory Studies Spijkenisse, November 27 For the statistical evaluation the unrounded (when available) figures were used instead of the rounded results. Results reported as '< ' or '> ' were not used in the statistical evaluation. First the normality of the distribution of the various data sets per determination was checked by means of the Lilliefors-test. After removal of outliers this check was repeated. In case a data set did not have a normal distribution, the (results of the) statistical evaluation should be used with due care. In accordance to ISO 5725 (1986 and 1994) the original results per determination were submitted subsequently to Dixon and Grubbs outlier tests. Outliers were marked by D(.1) for the Dixon test and by G(.1) or DG(.1) for the Grubbs test. Stragglers were marked by D(.5) for the Dixon test and by G(.5) or DG(.5) for the Grubbs test. Both outliers and stragglers were not included in the calculations of the averages and the standard deviations. Finally, the reproducibilities were calculated from the standard deviations by multiplying these with a factor of GRAPHICS In order to visualise the results against the reproducibilities from literature, Gauss plots were made, using the sorted data for each determination (see appendix 1). On the Y-axis the analytical results were plotted. The corresponding laboratory numbers were under the X-axis. A straight line presented the average of the reported data. Two striped lines presented the reproducibility limits of the selected standard, calculated as mean ± target reproducibility, parallel to the average line. Outliers and other data, which were excluded from the calculations, were represented as a cross. Accepted data were represented as a triangle. Secondly also two-sample Youden plots were made (see appendix 1). On the X- axis the average result of sample 758 of a certain laboratory is plotted against the average result of sample 759 of the same laboratory on the Y-as. Therefore, the results of each participating laboratory are presented by one point in the graph. Outliers and other data, which were excluded form the calculations, are presented in a cross. All other data are presented as a triangle. The means of the results are presented by the dotted lines. The intersection of these lines is the target value, where the participant s triangle should be positioned if both results were accurate. Furthermore, Graphs were made. This is a method for producing a smooth density approximation to a set of data that avoids some problems associated with histograms (see appendix 5; nr.12 and 13). 3.3 Z-SCORES To evaluate the performance of the participating laboratories the z-scores were calculated. As it was decided to evaluate the performance of the participants in this proficiency test (PT) against the literature requirements, e.g. ASTM reproducibilities, the z-scores were calculated using a target standard deviation. This result was an evaluation independent of the spread of this interlaboratory study. The target standard deviation was calculated from page 6 of 44 Styrene: iis7c1x

7 Spijkenisse, November 27 Institute for Interlaboratory Studies the literature reproducibility by division with 2.8. The z-scores were calculated in accordance with: z (target) = (result - average of PT) / target standard deviation The z (target) scores for the average results of the samples #758 and #759 were listed in appendix 1. Absolute values for z<2 were very common and absolute values for z>3 were very rare. Therefore the usual interpretation of z-scores was as follows: z < 1 good 1 < z < 2 satisfactory 2 < z < 3 questionable 3 < z unsatisfactory 4 EVALUATION In this proficiency test major problems were encountered with the despatch of the samples to the laboratories in: Brazil, India, Mexico, P.R. of China and Russia. Due to custom clearance problems, most laboratories in these countries did receive the samples near or after the final reporting date. One laboratory did not receive the samples at all. Not all participants were able to report results for all the requested tests. Finally, 33 participants did send in 553 numerical results. Observed were 26 outlying results, which is 4.7%. In proficiency studies outlier percentages of 3 % % are quite normal. 4.1 EVALUATION PER TEST In this section the results were discussed per test. The mentioned standard reproducibilities for the tested impurities were extrapolated out of the reproducibility values, which are mentioned in the standard ASTM D 5135:7 (table 3 and 4). Only for the Density not a normal distribution was found. Therefore, the statistical evaluation for this determination should be used with care. Aldehydes: It is difficult to draw significant conclusions, as the data seems bimodally divided. This was also noticed in previous round robins. Therefore the participants were requested to send in some critical details of the Aldehydes determination (see appendix 2). From the details it was observed that many titration results are rounded much too far (to.1 ml) to give reliable Aldehyde results. The spread in the volume for the second titration step of the sample is very large (. 1.7 ml). Volume step 1 in ml Volume step 2 in ml Volume blank in ml Normality NaOH Intake in ml Density in kg/l Aldehyde result mg/kg Table 5: example of rounding effect of volume on Aldehyde result Styrene: iis7c1x page 7 of 44

8 Institute for Interlaboratory Studies Spijkenisse, November 27 Furthermore, ASTM D 2119 is not clear in some critical steps, which might easily lead to significant differences in the results. For instant; during the standing period of 1 hour before titration step 1 and titration step 2, it is not clear for how long and at what speed the sample should be shaken. CO 2 -absorption during this hour may influence the result significantly. Colour: Inhibitor: Density: Peroxides: Polymer: Sulphur: Water: Purity: This determination is not problematic. No results were outside the reproducibility limits and the calculated reproducibility is in full agreement with the requirements of ASTM D 129:5. This determination was very problematic for two individual laboratories. The calculated reproducibility of the results found with ASTM D459 version 2 is, after excluding of the statistical outliers, in agreement with the requirements of the standard. No results were outside the reproducibility limits. The calculated reproducibility of the results found with ASTM D459 version 26 is, after excluding of the statistical outliers, almost in agreement with the requirements of the standard. Only two results were outside the reproducibility limits. The results found according ASTM D459: vs the results found according ASTM D459:6 differ significantly. This determination is not problematic. No results were outside the reproducibility limits and the calculated reproducibility is, after rejection of the statistical outliers, in good agreement with the requirements of ASTM D452:2e1. This determination was very problematic. Four results were outside the reproducibility limits and the calculated reproducibility is not at all in agreement with the requirements of ASTM D234:3. It is hard to draw any significant conclusions as the assigned value is near or below the application range of the test method. It is hard to draw any significant conclusions as the assigned value is near or below the application range of the test methods. This determination was very problematic. Five results were outside the reproducibility limits and the calculated reproducibility is, even after rejection of the statistical outliers, not at all in agreement with the requirements of ASTM E164:5. This determination is not problematic. None of the results were outside the reproducibility limits. The calculated reproducibilities of both samples #758 and #759 are in full agreement with the requirements of ASTM D 5135:7. page 8 of 44 Styrene: iis7c1x

9 Spijkenisse, November 27 Institute for Interlaboratory Studies It was remarkable that some participants reported GC or in house while the available ASTM D 5135:7 let the user free in choice of GC, detector, column and internal standard to be used. Only the use of an internal standard for quantification is mandatory. Benzene: m+p-xylenes: Ethylbenzene: Cumene: o-xylene: For sample #758 and #759, too few numerical results were reported for meaningful conclusions. This determination is problematic. For sample #759, five results were outside the reproducibility limits. The calculated reproducibility was, after rejection of the statistical outlier, not at all in agreement with the requirements of ASTM D5135:7. Unfortunately, for sample #758 no numerical results were reported for meaningful conclusions. The average recovery of m+p-xylenes (calculated from the difference between the value for the original sample #758 and the spiked sample #759) and the theoretical increment of 12.5 mg/kg is good (97%). The individual recoveries vary over a small range (62 15%). No analytical problems were observed. In total only one result was outside the reproducibility limits and the calculated reproducibilities were, after rejection of the statistical outlier, both in good agreement with the requirements of ASTM D 5135:7. The average recovery of Ethylbenzene (calculated from the difference between the assigned value for the original sample #758 and the spiked sample #759) and the theoretical increment of 17.6 mg/kg is good (93%). The individual recoveries vary over a large range (34 119%). For samples #758 and #759, too few numerical results were reported for meaningful conclusions. This determination is not problematic. For sample #759, no results were outside the reproducibility limits and the calculated reproducibility is in agreement with the requirements of ASTM D5135:7. Unfortunately, for sample #758 no numerical results were reported for meaningful conclusions. The average recovery of o-xylene (calculated from the difference between the value for the original sample #758 and the spiked sample #759) and the theoretical increment of 45.6 mg/kg is good (97%). The individual recoveries vary over a rather small range (88 113%). n-propylbenzene: This determination is problematic. For sample #759, only one result was outside the reproducibility limits. The calculated reproducibility is, not in agreement with the requirements of ASTM D5135:7. Unfortunately, for sample #758 no numerical results were reported for meaningful conclusions. The average recovery of n-propylbenzene (calculated from the difference between the value for the original sample #758 and the spiked sample Styrene: iis7c1x page 9 of 44

10 Institute for Interlaboratory Studies Spijkenisse, November 27 #759) and the theoretical increment of 29.4 mg/kg is good (98%). The individual recoveries vary over an acceptable small range (79 119%). m+p-ethyltoluenes: This determination is problematic. For sample #759, one result was outside the reproducibility limits. The calculated reproducibility is, after rejection of the statistical outlier, not in agreement with the requirements of ASTM D5135:7. Unfortunately, for sample #758 too few numerical results were reported for meaningful conclusions. The average recovery of m+p-ethyltoluenes (calculated from the difference between the value for the original sample #758 and the spiked sample #759) and the theoretical increment of 48. mg/kg is unsatisfactory (59%). The individual recoveries vary over a large range (29 78%). α-methylstyrene: This determination not problematic. In total four results were outside the reproducibility limits and the calculated reproducibilities are both, after rejection of the statistical outliers, in full agreement with the requirements of ASTM D 5135:7. The concentrations found for both samples differ not significantly, which was expected as this component was not spiked. The fact that the ratio R/r is smaller than 3, suggests that random and systematic errors are present. Phenylacetylene: This determination is not problematic. In total three results were outside the reproducibility limits and the calculated reproducibilities are both, after rejection of the statistical outliers, in full agreement with the estimated requirements calculated using the strict Horwitz equation. The concentrations found for both samples differ not significantly, which was expected as this component was not spiked. The fact that the ratio R/r is smaller than 3, suggests that random and systematic errors are present. Benzaldehyde: Nonaromatics: This determination is problematic for two individual laboratories. In total three results were outside the reproducibility limits. The calculated reproducibilities are both, after rejection of the statistical outliers, in agreement with the estimated requirements calculated using the strict Horwitz equation. The concentrations found for both samples differ not significantly, which was expected as this component was not spiked. The fact that the ratio R/r is smaller than 3, suggests that random and systematic errors are present. For this determination some analytical problems were observed. All participants that did send us a chromatogram agreed on the location of the Nonaromatics in the chromatogram (all peaks before the Ethylbenzene-peak except the Benzene and Toluene peak). However, some participants detected less then 5 peaks, other participants even more then 1 peaks (see appendix 3). So obviously the results of this determination depend strongly on the column, the GC and the (integration) threshold. page 1 of 44 Styrene: iis7c1x

11 Spijkenisse, November 27 Institute for Interlaboratory Studies 4.2 PERFORMANCE EVALUATION FOR THE GROUP OF LABORATORIES A comparison has been made between the reproducibility as declared by the relevant standard and the reproducibility as found for the group of participating laboratories. The average results per sample, calculated reproducibilities and reproducibilities, derived from literature standards (in casu ASTM standards) are compared in the next tables. Parameter unit n average 2.8 *sd R R (lit) Aldehydes mg/kg Colour Inhibitor (Old version) mg/kg Inhibitor (New version) mg/kg Density kg/l Peroxides as H 2 O 2 mg/kg Polymer mg/kg Sulphur mg/kg unknown Water mg/kg Purity %M/M Benzene mg/kg unknown Ethylbenzene mg/kg m+p-xylenes mg/kg Cumene mg/kg o-xylene mg/kg n-propylbenzene mg/kg m+p-ethyltoluenes mg/kg α-methylstyrene mg/kg Phenylacetylene mg/kg Benzaldehyde mg/kg Nonaromatics mg/kg unknown Table 6: reproducibilities of sample #758 Parameter unit n average 2.8 *sd R R (lit) Purity %M/M Benzene mg/kg unknown Ethylbenzene mg/kg m+p-xylenes mg/kg Cumene mg/kg o-xylene mg/kg n-propylbenzene mg/kg m+p-ethyltoluenes mg/kg a-methylstyrene mg/kg Phenylacetylene mg/kg Benzaldehyde mg/kg Nonaromatics mg/kg unknown Table 7: reproducibilities of sample #759 Styrene: iis7c1x page 11 of 44

12 Institute for Interlaboratory Studies Spijkenisse, November 27 Without further statistical calculations it can be concluded that for many tests there is a reasonable compliance of the group of participating laboratories with the relevant standards. The tests that are problematic have been discussed in paragraph 4.1. For the GC components Youden-plots are made. In these plots the results for the components of sample #758 and #759 are plotted against each other. The results for Ethylbenzene, a-methylstyrene, Benzaldehyde and Nonaromatics are found in a circular cloud, which points at random errors dominating over systematic errors. The results for Purity and Phenylacetylene are found in a line, which points at dominating systematic errors. See appendix 1 for the plots. 4.3 COMPARISON OF THE PROFICIENCY TEST OF OCTOBER 27 WITH PREVIOUS PT S October 27 October 26 October 25 October 24 Number of reporting labs Number of Results reported 553 1) 675 1) 566 1) 325 Statistical outliers Percentage outliers 4.7 % 4.7 % 3.2 % 4.6 % Table 8: comparison with previous proficiency tests 1) Two samples were sent out In proficiency tests, outlier percentages of 3% - 7.5% are quite normal. The performance of the determinations of the proficiency tests was compared against the requirements of the respective standards. The conclusions are given the following table: page 12 of 44 Styrene: iis7c1x

13 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination October 27 October 26 October 25 October 24 Aldehydes Colour +/ Inhibitor (Old version) Inhibitor (New version) n.d. n.d. Density n.d. n.d. Peroxides as H 2 O Polymer +/- ++ +/- +/- Sulphur n.e. +/- n.e. - Water Purity Ethylbenzene m+p-xylenes Cumene n.e o-xylene n-propylbenzene m+p-ethyltoluenes alpha-methylstyrene n.e. Phenylacetylene n.e. Benzaldehyde +/ n.d. Table 9: comparison determinations against the standard The performance of the determinations against the requirements of the respective standards is listed in the above table. The following performance categories were used: ++: group performed much better than the standard + : group performed better than the standard +/-: group performance equals the standard - : group performed worse than the standard -- : group performed much worse than the standard n.e.: not evaluated n.d.: not determined Styrene: iis7c1x page 13 of 44

14 Institute for Interlaboratory Studies Spijkenisse, November 27 APPENDIX 1 Determination of Aldehydes as benzaldehyde on sample #758; results in mg/kg lab method value mark Z(targ) remarks 15 D D D D D D D D D D D D D D D D D D D D D In house D normality OK n 23 outliers mean (n) st.dev. (n) R(calc.) 15.7 R(D2119:3) page 14 of 44 Styrene: iis7c1x

15 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of Colour Pt/Co on sample #758; lab method value mark Z(targ) remarks 15 D D D D D D D D D D D D D129 < D129 < D D D D D D D D D D D D D D D D normality OK n 28 outliers mean (n) 7.5 st.dev. (n) 2.47 R(calc.) 6.9 R(D129:5) Styrene: iis7c1x page 15 of 44

16 Institute for Interlaboratory Studies Spijkenisse, November 27 Determination of Inhibitor (D 459:) (p-tbc) on sample #758; results in mg/kg lab method value mark Z(targ) remarks D D D D D D in house D D D C -.36 First reported D D D in house 8.3 G(.5) in house D D D459-8 G(.5) normality OK n 16 outliers 2 mean (n) 6.9 st.dev. (n).595 R(calc.) 1.66 R(D459:) page 16 of 44 Styrene: iis7c1x

17 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of Inhibitor (D 459:6) (p-tbc) on sample #758; results in mg/kg lab method value mark Z(targ) remarks 15 D D D D D D D D D D D G(.1) D D D D D G(.5) D D normality OK n 16 outliers 2 mean (n) 6.7 st.dev. (n).568 R(calc.) 1.59 R(D459:6) Styrene: iis7c1x page 17 of 44

18 Institute for Interlaboratory Studies Spijkenisse, November 27 Determination of Density at 2 o C on sample #758; results in kg/l lab method value mark Z(targ) remarks 15 D DG(.5) D D D D D D D D D D D D D D D D D D DG(.5) D D D D D D D D normality not OK n 25 outliers 2 mean (n).963 st.dev. (n).75 R(calc.).21 R(D452:2e1) page 18 of 44 Styrene: iis7c1x

19 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of Peroxides as H 2 O 2 on sample #758; results in mg/kg lab method value mark Z(targ) remarks 15 D D D D D D D D D D D D D D D D D D D D D D D normality OK n 23 outliers mean (n) 28.7 st.dev. (n) R(calc.) R(D234:3) Styrene: iis7c1x page 19 of 44

20 Institute for Interlaboratory Studies Spijkenisse, November 27 Determination of Polymers on sample #758; results in mg/kg lab method value mark Z(targ) remarks 15 D2121 < D2121 < D2121 < D D2121 < D2121 < D D D2121 < D2121 < D2121 < D2121 < D2121 < D2121 < D2121 < D2121 < D D D D2121 < D2121 < D2121 < D D2121 < D D D D normality OK n 11 outliers mean (n).81 st.dev. (n).348 R(calc.).97 R(D2121:7) page 2 of 44 Styrene: iis7c1x

21 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of Sulphur on sample #758; results in mg/kg lab method value mark Z(targ) remarks 15 D5453 < D5453 < D3961 < D5453 < D5453 < D D5453 < D5453 < IP373 < D5453 < in house < D5453 <.5 C First reported D D5453 < D5453 < D D normality OK n 4 outliers mean (n).18 st.dev. (n).143 R(calc.).4 R(D5453:6) unknown Styrene: iis7c1x page 21 of 44

22 Institute for Interlaboratory Studies Spijkenisse, November 27 Determination of Water on sample #758; results in mg/kg lab method value mark Z(targ) remarks 15 E E E E G(.5) E E E E E E E E DG(.5) E E E E E E E E E E E DG(.5) E E G(.5) E E normality OK n 23 outliers 4 mean (n) st.dev. (n) R(calc.) R(E164:5) page 22 of 44 Styrene: iis7c1x

23 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of Purity on samples #758 and #759; results in %M/M lab method #758 mark Z(targ) remarks #759 mark Z(targ) remarks 15 D D D D D D D GC/MSD C 1.93 First reported C 1.83 First reported D D D D D D D D D D D D in house D D D D in house D D normality OK OK n outliers mean (n) st.dev. (n) R(calc.) R(D5135:7) Styrene: iis7c1x page 23 of 44

24 Institute for Interlaboratory Studies Spijkenisse, November # # #758 #759 page 24 of 44 Styrene: iis7c1x

25 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of Benzene on samples #758 and #759; results in mg/kg lab method #758 mark Z(targ) remarks #759 mark Z(targ) remarks 15 D5135 < < D5135 < < D5135 < < in house < < D4534 < < D4534 < < < < GC/MDS < < in house < < in house < < in house < < D5135 < < D5135 < < in house < L < in house < < D5135 < < D D5135 < < in house in house < < in house < < < < in house in house < < in house normality not OK not OK n 5 5 outliers mean (n) st.dev. (n) R(calc.) R(lit) unknown unknown Styrene: iis7c1x page 25 of 44

26 Institute for Interlaboratory Studies Spijkenisse, November 27 Determination of m+p-xylenes on samples #758 and #759; results in mg/kg lab method #758 mark Z(targ) remarks #759 mark Z(targ) Recov. % remarks 15 D5135 < D5135 < D5135 < D5135 < D5135 < D5135 < D5135 < GC/MDS < C First reported D D5135 < D5135 < D5135 < D5135 L D5135 < D5135 < D5135 < D5135 < D5135 < D5135 < D5135 < G(.1) D5135 < in house < D5135 < D normality n.a. OK n 23 outliers 1 Spike: mean (n) n.a (= / 12.5 * 1) st.dev. (n) n.a R(calc.) n.a R(D5135:7) n.a page 26 of 44 Styrene: iis7c1x

27 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of Ethylbenzene on samples #758 and #759; results in mg/kg lab method #758 mark Z(targ) remarks #759 mark Z(targ) Recov % remarks 15 D D D D D D D GC/MDS D D D D D D G(.1) D D D D D D in house D D D D in house D D normality OK OK n outliers 1 Spike: mean (n) (= / 17.6 * 1 ) st.dev. (n) R(calc.) R(D5135:7) Styrene: iis7c1x page 27 of 44

28 Institute for Interlaboratory Studies Spijkenisse, November # # #758 #759 page 28 of 44 Styrene: iis7c1x

29 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of Cumene on samples #758 and #759; results in mg/kg lab method #758 mark Z(targ) remarks #759 mark Z(targ) remarks 15 D5135 < < D5135 < < D5135 < < D5135 < < D5135 < < D5135 < < D5135 < ex False positive result? 343 GC/MDS < ex False positive result? D5135 < < D5135 < < D5135 < < D5135 < < D5135 L L D5135 < < D5135 < < D5135 < < D5135 < < in house D5135 < < D5135 < < D5135 < < D5135 < < in house < < D5135 < < D normality n.a. n.a. n 2 outliers mean (n) n.a. n.a. st.dev. (n) n.a. n.a. R(calc.) n.a. n.a. R(D5135:7) n.a. n.a. Styrene: iis7c1x page 29 of 44

30 Institute for Interlaboratory Studies Spijkenisse, November 27 Determination of o-xylene on samples #758 and #759; results in mg/kg lab method #758 mark Z(targ) remarks #759 mark Z(targ) Recov % remarks 15 D5135 < D5135 < D5135 < D5135 < D5135 < D5135 < D5135 < < False negative? 343 GC/MDS < G(.1) D D5135 < D5135 < D5135 < D5135 < D5135 L D5135 < D5135 < D5135 < D5135 < in house D5135 < D5135 < D5135 < D5135 < in house < D5135 < D normality n.a. OK n 23 outliers 1 Spike: mean (n) n.a % (= / 45.6 * 1) st.dev. (n) n.a R(calc.) n.a R(D5135:7) n.a page 3 of 44 Styrene: iis7c1x

31 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of n-propylbenzene on samples #758 and #759; results in mg/kg lab method #758 mark Z(targ) remarks #759 mark Z(targ) Recov % remarks 15 D5135 < D5135 < D5135 < D C First rep C First reported D5135 < D5135 < D5135 < GC/MDS < D5135 < D5135 < D5135 < D5135 L D5135 < D5135 < D5135 < D in house D5135 < D5135 < D5135 < < D5135 < in house < D5135 ex normality n.a. OK n 3 23 outliers Spike: mean (n) n.a ( = / 29.4 * 1 ) st.dev. (n) n.a R(calc.) n.a. 1.1 R(D5135:7) n.a Styrene: iis7c1x page 31 of 44

32 Institute for Interlaboratory Studies Spijkenisse, November 27 Determination of m+p-ethyltoluenes on samples #758 and #759; results in mg/kg lab method #758 mark Z(targ) remarks #759 mark Z(targ) Recov % remarks 15 D5135 < D5135 < D5135 < D5135 < D5135 < D5135 < D5135 < < False negative? 343 GC/MDS < W Withdrawn f.r D D5135 < D5135 L D5135 < D5135 < D5135 < G(.5) D5135 < in house D5135 < D5135 < D5135 < D in house < < False negative? D5135 ex normality OK OK n 3 17 outliers 1 Spike: mean (n) n.a % (= 28.1 / 48. * 1) st.dev. (n) n.a R(calc.) n.a R(D5135:7) n.a page 32 of 44 Styrene: iis7c1x

33 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of alpha-methylstyrene on samples #758 and #759; results in mg/kg lab method #758 mark Z(targ) remarks #759 mark Z(targ) remarks 15 D D D D D D D GC/MDS D D D D D G(.5) G(.1) D D D D D D G(.5) D C.49 First rep C.22 First rep in house D D D G(.5) G(.5) D in house D D normality OK OK n outliers 3 2 mean (n) st.dev. (n) R(calc.) R(D5135:7) Youden Plot Normality OK n 24 / 27 outliers 3 mean (n) st.dev. (n) r (calc) R(calc.) 26.6 R/r ratio Styrene: iis7c1x page 33 of 44

34 Institute for Interlaboratory Studies Spijkenisse, November # # #758 #759 page 34 of 44 Styrene: iis7c1x

35 Spijkenisse, November 27 Institute for Interlaboratory Studies Determination of Phenylacetylene on samples #758 and #759; results in mg/kg lab method #758 mark Z(targ) remarks #759 mark Z(targ) Remarks 15 D5135 < < False negative 171 D D D5135 < < False negative 323 D D D DG(.5) G(.5) GC/MDS < < False negative 347 D D < < False negative D D5135 < < False negative 865 D D D DG(.5) in house D D5135 < < False negative 1429 D D in house D normality OK OK n outliers 2 1 mean (n) st.dev. (n) R(calc.) R(Horwitz) Youden Plot Normality OK n 13 / 15 outliers 2 mean (n) st.dev. (n) r (calc) 1.48 R(calc.) 3.58 R/r ratio Styrene: iis7c1x page 35 of 44