66 HASLAM, HAMILTON AND JEFFS : THE DETERMINATION OF [Vol, 83 The Determination of Poly(Ethy1 Esters) in Methyl Methacrylate Copolymers BY J. HASLAM, J. B. HAMCLTON AND A. R. JEFFS (Imperial Chemical Industries Ltd., Plastics Division, FVelwyn Garden City, Herts.) A method has been devised for the determinakion of poly(ethy1 esters), e.g., poly(ethy1 acrylate), in methyl met hacrylate copolymers. The alkoxyl groups in the sample are converted to the corresponding iodides, which are then determined by gas - liquid chi-omatographic test. THE determination of small proportions of poly(ethy1 esters) in methyl methacrylate copoly- mers is an interesting analytical problem. Its solution appeared to depend on a quantitative determination of the total alkoxyl groups present in the polymer, followed by an effective separation of the individual alkyl iodides formed.Our original efforts were directed to finding a simple process by which, for example, the methoxyl group of poly(methy1 methacrylate) could be converted to methyl iodide in very high yield. Such a process has been devised and is essentially based on the method for determining alkoxyl groups in cellulose ethers worked out by Easterbrook and one of us (J. B. H.).l The method that we adopted differs from that of Easterbrook and Hamilton only in the following respects- (a) nitrogen was used in conjunction with a mercury lute and a flowmeter as the sweeping-out gas, instead of carbon dioxide, ( b ) the spiral scrubber contained 4 ml of 25 per cent. w/v sodium acetate solution, and (c) the bromine solution absorber was modified so that a second bromine solution absorber could be fitted in series, as a precautionary measure.Preliminary experiments showed that the phenol used in Easterbrook and Hamilton’s method to solubilise the cellulose ether was also necessary when working with polymers. Experiments carried out without phenol gave only two- thirds of the theoretical yield of alkyl iodide. When this modified method was applied to known samples (about 20 mg) of disintegrated poly(methy1 methacrylate) sheet, samples of pcily(methy1 methacrylate) moulding granules and a sample of poly(ethy1 acrylate), the results shown in Table I were obtained. From these results it was seen that the recovery of iodides was sufficiently satisfactory for a ratio method to be used for the determination of the relative proportions of the two iodides. It was now necessary to distinguish between the iodides in a quantitative manner.February, 19581 POLY(ETHYL ESTERS) IN METHYL METHACRYLATE COPOLYMERS TABLE I 67 DETERMINATION OF ALKOXYL CONTENT BY A MODIFIED METHOD Theoretical methoxyl content of poly(methy1 methacrylate) = 31.03 per cent.w/w Theoretical ethoxyl content of poly(ethy1 acrylate) = 45-06 per cent. w/w Sample Alkoxyl content, Conversion, 01 01 70 70 Sample A [disintegrated unplasticised poly- Sample B [disintegrated unplasticised poly- Sample C [poly(methyl methacrylate) mould- ing granules] . . .. .. . . . . 30.32, 30.27 (-OCH,) 97.8, 97.7 Sample D [poly(methyl methacrylate) mould- inggranules] , . .... . . .. 29.98, 29.94 (-OCH,) 96-7, 96.6 Sample E [poly(ethyl acrylate)] . . . . 44.09, 44.13 (-OCH,CH,) 97.8, 97.9 (methyl methacrylate) sheet] . . . . 30.45, 30.56 (-OCH,) 98.2, 98-6 (methyl methacrylate) sheet] . . . . 30.23, 30.21 (-OCH,) 97.5, 97.5 Many methods of separating mixed alkyl iodides derived from alkoxyl groups have been suggested, and a review of these was included in a paper on the subject by Gunnar Gran.2 However, gas - liquid chromatography seemed to offer great advantages over ordinary chemical methods. Indeed, a paper entitled “Selective Microdetermination of Alkoxy Groups by Gas - Liquid Chromatography” by Martin and Vertalier3 was read at the Congress on Analytical Chemistry, Lisbon, September, 1956. Cotton-wool pads . B 10 , G lass-rod spiral Fig.1. n-Heptane trap with an extension for an absorber containing buffered bromine soh tion In their paper these authors were concerned with monomeric substances and distilled the iodides liberated by the Zeisel method into an absorber containing 0.2 ml of chloroform, which in turn contained 5 per cent. w/v of methylene dichloride. The latter acted as an internal standard in their subsequent gas - liquid chromatographic procedure. For methoxyl and ethoxyl groups, Martin and Vertalier found it was possible to ascertain the ratio in which the groups occurred in a single molecule (molar ratio) and to determine them at the same time in an approximate manner (with an error of 5 to 10 per cent.) by working on 20 mg of substance. This knowledge permitted the exact determination of each group when this.information was supplemented by an iodimetric determination of the total alkoxyl groups.68 HASLAM, HAMILTON AND JEFFS THE DETERMINATION OF [Vol. 83 Since copolymers may contain the polymerised esters in any ratio, it seemed important to devise an absorber for retaining the liberated iodides quantitatively. The absorber shown in Fig. 1 was found tolbe satisfactory and was used in all our work. This contained 0-5 ml of solvent and was attached to the spiral scrubber of the apparatus used by Easterbrook and Hamilton. A drying agent was included, since water interfered with the gas-liquid chromatographic procedure. I t was necessary that the solvent employed in the absorber should not interfere with the separation of the two iodides in the subsequent chromatographic test and several solvents were tried, the trap being kept at 0" C and at -80" C (i.e., the temperature of a solid carbon dioxide - methanol mixture). Eventually, pure synthetic n-heptane was used at -80" C and proved ideal for the purpose.Its efficiency was evaluated by connecting a second absorber containing buffered bromine solution in series with the n-heptane absorber to retain any iodides escaping absorption. Determination of the iodides in this buffered bromine solution indicated the proportion of the alkoxyl groups converted to iodide that were retained in solution by the n-heptane absorber. The results, which are based on the mean values for alkoxyl content given in Table I, were 99.4 and 99.2 per cent. of methyl iodide retained by the n-heptane absorber for Sample A and 99.9 and 99.9 per cent.of ethyl iodide for Sample E. Further preliminary experiments showed that, if n-heptane containing a small amount of methylene dichloride as internal standard was employed as absorbing solvent, loss of methylene dichloride (about 5 to 10 per cent.) resulted with the passage of nitrogen through the absorber even at -80" C. This was overcome by absorbing the alkyl iodides initially in 0.500 ml of n-heptane; after absorption was complete 0.500 ml of n-heptane, containing internal standard, was then added to the contents of the absorber. In fact, we found it advantageous to use two internal standards in the n-heptane-methylene dichloride for comparison with the larger methyl iodide peak and a smaller amount of ethylidene dichloride for comparison with the smaller ethyl iodide peak.Our full method for the determination of poly(ethy1 acrylate) in copolymers with poly(methy1 methacrylate) is described below. APPARATUS- and the absorber shown in Fig. 1. METHOD The reactionJEask, spiral scrubber and glass spoon as used by Easterbrook and Hamilton, Conventional gas - liquid chromatographic aj!$aratus. Agla micrometer-syringe pipette. Hydriodic acid, sp.gr. 1.17-The M.A.R. reagent, which is supplied in 6-in1 ampoules. Phenol-Analyt ical-reagent grade. n-Heptane-The pure synthetically prepared material. Sodium acetate solution-A 25 per cent. w/v solution in water. Methylene dichloride. E thylidene dichloride. Methyl iodide. E thy1 iodide. The column used consisted of a U-tube having a total length of 6 feet and a nominal bore of 0.25 inches, packed with a 30 per cent, w/w mixture of dinonyl sebacate on Celite 545 that had been graded as described by Martin and James.* REAGENTS- PROCEDURE FOR CALIBRATING THE GAS - LIQUID CHROMATOGRAPHIC APPARATUS- The conditions were as follows- Column temperature, 75" c; Inlet pressure, Exit pressure, Nitrogen flow-rate, Bridge current, 140 mA; Katharometer temperature, 650 mm o f mercury; 150 mm of mercury; 3.0 litres per hour ; room temperature (22" C).Add 0.500 ml of methylene dichloride and 0.250 ml of ethylidene dichloride accurately to n-heptane contained in a 100-ml calibrated flask by means of an Agla micrometer-syringe pipette. Dilute the contents of the flask to the mark with n-heptane and shake thoroughly.February, 19581 POLY (ETHYL ESTERS) IN METHYL METHACRYLATE COPOLYMERS 69 1 ml of this n-heptane solution contains 0405 ml of methylene dichloride and To 10.0-ml aliquots of the n-heptane solution add various known volumes of methyl and ethyl iodides accurately by micrometer-syringe pipette (see Table 11).Subject 2 to 5 drops of each of these solutions to gas - liquid chromatographic test under the above-mentioned conditions (the actual amount depending on the iodide content). From the chromatograms determine the ratios of (a) peak height of methyl iodide to the peak height of 0.005 ml of methylene dichloride and (b) peak height of ethyl iodide to peak height of 0.0025ml of ethylidene dichloride, and construct calibration curves, with the weights of iodide as ordinates and these ratios as abscissae.Our results are summarised in Table I1 and the calibration curves are shown in Fig. 2. 0.0025 ml of ethylidene dichloride. Ratio of peak height of iodide to peak height of corresponding internal standard Fig. 2. Calibration curves: curve A, methyl iodide ; curve B, ethyl iodide TABLE I1 RESULTS OBTAINED IN THE CONSTRUCTION OF THE CALIBRATION CURVES d2izg for methyl iodide 2.279 for ethyl iodide 1.933 Standard Volume of methyl iodide No. added, ml 1 0.170 3 3 0.100 4 0.080 5 0.050 6 0.020 - Weight of methyl iodide, mg per ml of solution 38.63 - 22.79 18.23 11.40 4.56 Volume of ethyl iodide added, ml 0.170 0.010 0-030 0.060 0.100 - Weight of ethyl iodide, mg per ml of solution Ratio (a) Ratio (b) - 5-79 - 32.86 - 7.96 1.93 3-33 0-42 5.79 2-62 1.40 11.58 1-66 2-87 19-33 0.67 4.68 PROCEDURE- Clean the reaction flask, spiral scrubber and the n-heptane absorber assembly with chromic - sulphuric acid mixture and wash them thoroughly with distilled water.Dry the apparatus by washing it with acetone and allowing residual acetone to evaporate. Place 2-5 g of phenol and the contents of a 6-ml ampoule of hydriodic acid in the reaction flask. Put 4ml of 25 per cent. w/v sodium acetate solution in the spiral scrubber and place the stopper in the open end. Connect the spiral scrubber to the reaction flask with suitable tension springs. With water flowing through the condenser and a nitrogen flow of 6 ml per minute passing through the apparatus, heat the contents of the reaction flask to boiling and maintain gently boiling for 30 minutes.This serves to condition the apparatus for the test. Withdraw the burner and allow the reaction flask to cool. Measure accurately 0.500 ml of n-heptane into the absorber by micrometer-syringe pipette and fit the absorber assembly to the spiral scrubber by means of tension springs. Weigh out accurately about 20 mg of copolymer into the glass spoon and lower the spoon plus sample carefully down through the condenser into the reaction flask. Reconnect the spiral scrubber and when assembled submerge the n-heptane absorber in solid carbon dioxide - methanol mixture.70 pol. 83 Heat the contents of the reaction flask to boiling and maintain boiling for 1 hour. Disconnect the absorber assembly.Add 0.500ml of a rt-heptane solution containing 1 per cent. v/v of methylene dichloride and 0.5 per cent. v/v oE ethylidene dichloride (made up accurately by micrometer-syringe pipette) to the contents of the absorber by micrometer-syringe pipette and mix the solutions. Hence the absorber now contains 0405 ml of methylene dichloride, 04025 ml of ethylidene dichloride, n-heptane up to 1 ml and the evolved iodides. Subject 1 to 6 drops of this heptane solution to chromatographic test under the conditions of calibration. With copolymers having a low poly(ethy1 acrylate) content, it is necessary to run two chromatograms to find the ratios (a) and ( b ) , Le., 1 drop of solution to obtain ratio (a) and 5 or 6 drops of solution to find ratio (b) (see Fig.3). This would be unnecessary with a gas - liquid chromatograph fitted with an attenuator. Calculate the ratios (a) and (b) and from the calibration curves find the weights of each iodide evolved from the polymer. Calculate the percentage of poly(ethy1 acrylate) in the copolymer from the yield of the two iodides. HASLAM, HAMILTON AND JEFFS : THE DETERMINATION OF 4 f /B / c /D D Sample I" -- Time Chromatograms: (a), 6 or 7 drops of trap solution; and ( b ) , 1 drop of trap solu- tion. A, methyl iodide; 13, methylene dichlor- ide; C, ethylidene dichloride; D, eth:yl iodide ; E, n-heptane Fig. 3. RESULTS A 19-9-mg sample of a copolymer known to consist of 91 per cent. of poly(methy1 meth- acrylate) and 9 per cent. of poly(ethy1 acrylate.) was subjected to the procedure described above. The peak heights of methyl iodide, methylene dichloride, ethylidene dichlclride and ethyl iodide were measured and the ratios (a) and (b) were calculated. By means of the calibration curve it was found th.at the trap therefore contained- The chromatograms obtained are shown in Fig.3. These ratios were 3.75 and 0.66, respectively. 25.6 mg of methyl iodide = 18.05 mg of poly(niethy1 methacrylate) and 2.8 mg of ethyl iodide = 1-80 mg of poly(ethy1 acrylate). (NOTE-Methyl methacrylate and ethyl acrylate are isomeric and have a molecular weight of 100.1 1 .)February, 19581 POLY (ETHYL ESTERS) IN METHYL METHACRYLATE COPOLYMERS :. Recovery (in terms of original polymer) = 19.85 mg. 71 :. Poly(ethy1 acrylate) in copolymer = - x 100 19.85 = 9.07 per cent.Other results found for this copolymer were 9.04 and 9-09 per cent. Similarly for a copolymer known to consist of 97 per cent. of poly(methy1 methacrylate) and 3 per cent. of poly(ethy1 acrylate), the results were 2-96, 3-15 and 3.09 per cent. When our control samples, i.e., Samples A and E in Table I, were subjected to the complete pro- cedure as described above, Sample A [poly(methyl methacrylate)] yieIded no trace of ethyl iodide and Sample E [poly (ethyl acrylate)] yielded no trace of methyl iodide. Incidentally, experiments in the presence of dibutyl phthalate indicate that this plasticiser does not interfere with the test. The accuracy of the gas - liquid chromatographic method, with close control, is generally accepted as being within 2 per cent. of the percentage determined and within these limits we believe our test to be quite satisfactory. As far as we are aware, this is the first time that the determination of alkoxyl groups in polymers has been recorded, and we consider that the principles of our test may prove useful in the general determination of the proportions of mixed alkoxyl groups in monomeric substances. The n-heptane trap, which has proved so efficient for absorbing methyl and ethyl iodides, may (with some modification) have other useful applications, e.g., for absorbing substances from the gas stream in gas - liquid chromatography for examination by ultra-violet and infra-red methods. REFERENCES 1. 2. 3. 4. Easterbrook, W. C., and Hamilton, J. B., Analyst, 1953, 78, 551. Gran, G., Suensk Papperstidning, 1954, 57, 702. Martin, F., and Vertalier, S., Presented at the XVth International Congress on Pure and Applied Martin, A. J. P., and James, A. T., Biochem. J., 1952, 50, 679. Chemistry (Analytical Chemistry), Lisbon, September 8th to 16th, 1956. Received August 13th, 1967