Document 996vvYv6dZxrLaqg9JDk3xa96

iHter-Office Memo TENNECO CHEMICALS, INC. < a? To H. B. Carr At Burlington Date November 4, 1969 From F. w. Kanzler At Burlington Subject Technical Memorandum Infra Red Analysis of Bound Acetate in Vinyl Chloride/Vinyl Acetate Copolymers Copy to H. H. Duhamel E. Farber J. c. Fisher A. P. Gallmayer fi.E. J. Hourihan L. Kanyok D. Johnson C. W. Johnston M. K. Rosen M. C. Spalding P. R. Searito W. D. Wesely M. W. Williams Burlington P. E. SUMMARY 1. An infra red method has been developed for the analysis of bound acetate in vinyl chloride/vinyl acetate copolymers. 2. The analysis is accomplished by scanning a dissolved copolymer sample in a sealed cell from 5.55y (1800 en_l) to 6.06 < (1650 cm"l) on the Perkin-Elmer 700 infra red spectrophotometer. The absorption of the carbonyl group at 5.75^* (1740 cm"^) is utilized as the analytical point for comparing the'absorbances obtained for samples to that of known concentration standards. B.A.C.M. #200 is attached to provide a detailed outline of the analysis. % 3. Correlation studies with the A & E Laboratory indicate excellent agreement (Table 1). A "t test" of significance at 95% confidence performed on this data showed that the results from the two laboratories are statistically similar. * 4. The return time for this analysis is approximately 2 hours. RECOMMENDATIONS 1. Correlation studies with the A & E Laboratory should be continued at a periodic frequency. 2. The method should be introduced as a grading method for copolymer resins. COLORITE 0X6313 2 November 4, 1969 3. The data should be utilized to establish copolymer plant process capability and to assure consistency between lots. 4. A program should be initiated to study in-lot variations of the bound acetate level. DISCUSSION 1. Selection of the Analytical Point Several analytical points for quantitative estimation of bound acetate in vinyl chloride/vinyl acetate copolymers are discussed in (1.). Attempts were made to calculate the ratio of the bands at 7.0/` (1430 cm"l) and 7.5/>. (1335 cm-*) versus the band at 7.3/ (1370 cm-*) for differ ent concentrations. The accuracy was found to be very poor since the levels of the bound acetate could only be estimated. This method also had the disadvantage of re quiring the casting of a thin film (ca. 0.001") from solu tion which is not a practical step in a routine laboratory analysis. A method described by Weinberger, Kagarise (2) was attempted next. Since this method also employed a cast film it was dismissed for the above mentioned reasons To eliminate the requirement of thin film casting a sealed cell was constructed from sodium chloride crystals and spacers to obtain an approximate thickness of 0.1 mm. The exact thickness of the cell was determined from the interference pattern of the empty cell to be 0.112 mm. Considering the excellent resolution and the freedom from interfering absorbances, the band at 5.75y* (1740 cm"*) was selected as the analytical point. Standards of known bound acetate levels were Vised for calibration purposes. 2. Operating Parameters Sample size: 4.0 gm of copolymer Solvent: Tetrahydrofuran, reagent grade, 100 ml Scan: 5.55/ (1800 cm"^) to 6.06/(1650 cm" Analytical point: 5.75/* (1740 cm"*) Cell: Sodium Chloride crystals, 0.112 mm thickness 3. Calibration Initially a standard copolymer provided by the A & E Lab oratory with a known level of bound acetate was used for COLORXTE 0X6314 3 November 4, 1969 calibration purposes. Since this level was in the 10% bound acetate range and the pen speed on the Perkin-Elmer 700 is not adjustable, a multi level standardization was decided upon to compensate for nonlinearity of the instru ment response. Due to lack of availability of a high level bound acetate standard, a blend of a vinyl chloride homppolymer and a vinyl acetate homopolyroer was prepared. The blends were made up in levels of 5%, 10%, 15% poly vinyl acetate in polyvinyl chloride. The standards were individually prepared for each series of samples run. Initial correlation data with the A & E Laboratory showed good agreement in the levels of about 9.5% to about 12.0% bound acetate (Table 2). However, almost all high level (13% to 15%) data showed a considerable amount of dis agreement between laboratories. (Memo to H. B. Carr from P. W. Kanzler, dated February 17, 1969). Subsequent investigations of various possible sources of error led to the following steps: a. Pen Speed As mentioned above, it was determined that the constant pen speed of the P & E 700 resulted in an overshoot on low absorbances and undershoot on high absorbances. For this reason, calibration standards that cover the entire range of targeted bound acetate are required for meaningful analyses. b. Base line For the calculation of the bound acetate levels, the absorbance of a sample is compared with the absorbance of a standard. This absorbance is read at 5.75/-1 (1740 cnT^) and is measured as the distance from the maximum transmittance to the minimum transmittance of the carbonyl band. To find a reproducible point at maximum transmittance, a base line is used. Initially, in Burlington, the base'line was drawn from the maximum transmittance at 5.62/> (1780 cirT^) to the maximum trans mittance at 5.95yu (1680 cm-^) . A study to compare the above type of base line with a base line drawn as in Flemington from a point at about 5.62(1780,cm"^) para llel to the transmittance lines on the chart paper was made. The results shown in Table 3 show a better agree ment with the Flemington Laboratory when using the latter base line. C. Instrument Correlation A set of two samples was run on both the Flemington and Burlington infra red spectrophotometer. Table 4 COLORITE 016315 -4- November 4, 1969 shows the results of this study. As can be seen from the data, the position on the chart where the peak is started has a great influence on the final result. Both instruments show good agreement, when the methods which are presently used in the two laboratories for the determination of the bound acetate level are com pared. d. Change of Sample to Solvent Ratio To overcome the nonlinearity of the instrument response it was decided to alter the sample size and/or the solvent content of the sample to be analyzed. This was done to achieve the same absorbance for all bound acetate levels. Inaccurate results and nonreproduci bility led to the dismissal of this step. In the future, a variable thickness cell may be tried to allow for change in absorbance. 4. Planned Correlation Two copolymer samples of a 382 and a 315 type copolymer will be run on a round robin basis to find the correct level of bound acetate. These samples will serve as standards and will allow for an elimination of the blends. LITERATURE 1. J. Haslam, H. A. Willis, Identification and Analysis of Plastics. 2. L. A. Weinberger, R. E. Kargarise, O.T.S. Bulletin No. PB 111438. FWK/rjl Attachment _( 1 (AAa F. V. Kanzler COLORITE 016316 \ TABLE 1 Bound Acetate Analysis Sample ID Resin Type 609066 619150 639025 619155 719008 719017 659055 Pre. 659055 Lot 649040 659062 699019 Pre. 314 315 382 315 315-2 315-2 385 385 384 385 389 % Bound Acetate Flem. Burl. 14.5 14.4 15.0 14.8 9.7 10.2 15.8 15.1 12.7 12.5 14.2 14.0 14.8 14.4 14.5 14.2 14.0 13.8 13.8 13.8 10.8 11.1 Specification Range 13.5-15.3 13.5-15.3 9.2-10.6 13.5-15.3 13.5-15.3 13.5-15.3 13.1-14.9 13.1-14.9 13.1-14.9 13.1-14.9 10.5-12.0 Analyst: J. E. Ravelli COLORITE 016317 'X Sample ID 648845 658050 658055 698035 698036 698040 698041 TABLE 2 Bound Acetate Analysis Correlation Date November, 1968 Resin Type 384 385 385 389 389 389 389 % Bound Acetate Flem. Burl. Spec. Range 13.0, 13.9 13.1, 13.6 12.9, 13.3 14.6, 14.5 13.1-14.9 14.5, 14.0 13.1-14.9 14.0, 13.6 13.1-14.9 10.3 10.5, 10.0 10.5-12.0 XX 8 11.7, 11.7 10.5-12.0 10.5 9.9 10.5-12.0 9.9, 11.0 11.1, 10.3 10.5-12.0 Analyst: F. W. Kanzler COLORITE 016318 \ TABLE 3 Bound Acetate Analysis Base line Comparison Sample ID 719003 719004 719005 719006 719007 619160 659061 619170 639030 Resin Tvoe 315-2 315-2 315-2 315-2 315-2 315 385 315 382 Flem. % Bound Acetate Burl. Burl. (Flem. Base line) Spec. Range 13.3 13.2 12.8 12.4 13.1 12.7 13.5-15.3 13.5-15.3 13.3 13.5 12.7 12.6 12.8 12.9 13.5-15.3 13.5-15.3 12.9 12.8 12.8 13.5-15.3 14.5 13.1 14.9 10.1 14.2 12.5 15.4 10.2 14.5 13.1 15.7 10.3 13.5-15.3 13.1-14.9 13.5-15.3 9.2-10.6 Analyst: J. E. Ravelli COLORITE 016319 Samnle ID 659055 619170 619202 (Std.) TABLE 4 Bound Acetate Analysis Base line and Peak Location Study* Orig. Analysis Flem. Burl. Resin Spec. Tvne Ranee 14.5 13.5 385 13.114.9 14.9 15.2, 15.4 315 -- 315 13.515.3 13.515.3 Flemlngton Infra Red Spectronhototneter Normal Flem. Peak Start Peak Start Analysis at 75% T at 92% T (Peak Start at 100% Tl 1969 10/16 10/24 10/24/69 10/24/69 Burlington Infra Red Spectrophotometer' Normal Burl. Flem. Base Peak Start (Peak Start line Tech at 90% T at 75% T) nique (Peak St. at 75%T 1969 10/16 10/24 10/16 10/24 10/24/69 13.9 13.9 13.3 14.1 14.1, 13.9, 13.8 13.8 14.4, 13.9 14.4, 14.1 13.7 15.1 15.7 14.4 15.7 15.2 15.4 15.6 15.7 14.8, 14.9 14.3 13.6 14.4 - 14.0 - 14.2 14.3 * The peak location study on the Burlington infra red spectrophotometer was only run on two types of peak start transmittance ranges. Since the 90% T peak start showed lower results, the Flemington type analysis at 100% T was not performed in Burlington.