1 Field of the Invention 2 The present invention relates to the solvent fractionation 3 of a polymer. More specifically, the invention relates to the 4 fractionation of a copolymer with a single solvent at different temperatures.
6 Background 7 Crystalline copolymers, such as linear, low density poly-8 ethylene (LLDPE) and ethylene vinyl acetate (EVA), are known to have 9 molecular weight distributions and composition distributions. The properties of copolymers having similar average compositions can vary 11 considerably depending upon the compositional distribution of the 12 copolymer. For example, in co-pending Canadian application 13 serial number 553J796 filed 12/8/87, it 14 was established that the molecular weight distribution for Exxon LL3001 linear low density polyethylene resin was narrower than the 16 molecular weight distribution of another commercially available 17 linear low density polyethylene. Compositional distributions are 18 known to have a strong effect on the physical properties of co-19 polymers, e.g., heat sealing, tear strength, and FDA extraction limits.
21 The molecular weight and compositional distributions of 22 copolymers have been determined by solvent fractionation of the 23 crystalline copolymer and analysis of the fractions for molecular 24 weight and composition. Apparatus for solvent fractionating of copolymers are known in the art. Such apparatus typically operate by 26 dissolving the copolymer in a hot solvent and allowing the solvent to 27 cool to ambient temperature within a packed column. The copolymer 28 crystallizes on the column packing as the polymer cools. It is known 29 in the art to cool the polymer to slightly above ambient temperature, about 30C, before starting fractionation with pure solvent.
31 Fractionation of the crystallized polymer is typically 32 conducted with a single solvent that is passed through the column at 33 increasing column temperatures beginning at the lowest temperature 34 obtained during cooling of the copolymer. Less soluble fractions of the copolymer are dissolved and removed from the column as the column 36 temperature rises such that the solvent effluent from the column 37 contain1ng dissolved copolymer can be collected in consecutive - 2 - l 3 3 5 0 7 2 1 fractions which are identified by a starting and ending column 2 temperature. These fractions are then analyzed for molecular weight 3 distribution and monomer composition by conventional means.
4 Summary of the Invention The present invention is a method for determining the 6 compositional distributions of a crystalline copolymer comprising the 7 steps of dissolving the copolymer in a solvent at elevated 8 temperature, filling a column with the solvent containing the g dissolved copolymer, crystallizing the copolymer within the column by reducing the temperature of the column to below ambient temperature, 11 e.g. to temperatures at least about 6C or lower, initially passing 12 pure solvent through the column at a constant temperature below 13 ambient to determine the amount of uncrystallized copolymer, and then 14 at increasing temperature to gradually dissolve the copolymer, continuously measuring the concentration of the copolymer in the 16 solvent exiting the column. The method is preferably conducted with 17 a fractionation temperature range of from 0 to 120C and with 18 tetrachloroethylene as the solvent. The effluent from the 19 fractionation column is preferably continuously analyzed by an IR
detector to determine the concentration of polymer in the solvent.
21 Brief Description of the Drawings 22 Figure l is a schematic diagram of the sample preparation 23 mode for an apparatus which can determine the solubility distribution 24 curve of a crystalline polymer in accordance with the method of the present invention;
26 Figure 2 shows the apparatus of Figure l in the solvent 27 fractionation mode;
28 Figure 3 is an exemplary solubility distribution curve which 29 was generated by the method of the present invention; and B0 Figure 4 is another exemplary solubility distribution curve 31 generated by the method of the present invention.
32 Detailed Descri~tion of a Preferred Embodiment 33 The method of the present invention has been practiced with 34 an apparatus that automatically determines the solubility distribution curve of a crystalline copolymer. Referring to Figure 36 l, a steel column is packed with glass beads (approximately 35 mesh) 37 and immersed in an oil bath lO. A column having an inside diameter 38 of 35 mm and a length of 400 mm has worked well. The temperature of - 3 - l 335072 1 the oil bath can be programmed over any desirable temperature range, 2 e.g. from below ambient, preferably about or below 6C, or about or 3 below -12C, although even lower temperatures may be employed, to a 4 temperature up to about 150C. Solvent used in the fractionation may be prevented from boiling by operating the apparatus at about 3 6 atmospheres pressure. A weighed amount of sample, usually about 1.6 7 grams, is placed in a sample preparation chamber 20, th~t is then 8 sealed and repeatedly evacuated and filled with argon. A metered 9 volume of solvent 30, preferably tetrachloroethylene, is then pumped through a three-way valve 32 into the sample preparation chamber 20 11 where it is stirred and heated to obtain a solution of about l 12 percent concentration. A metered volume of this solution, usually 13 about lO0 cc, is then pumped into the packed column 2 which has been 14 thermostated at a high temperature of usually at least about 120C.
The polymer solution sample is subsequently crystal!ized by 16 cooling the polymer in the column 2 at a programmed rate of 5C per 17 hour to below ambient temperature in order to increase the proportion 18 of the polymer crystallized, e.g., to below or about 6C. However, 19 some polymers may require lower crystallization temperatures, e.g.
about or below -6C or -12C. The column 2 is then maintained at 21 this low temperature for at least an hour. Thereafter, the elution 22 of fractionation stage shown in Figure 2 is started by pumping pure 23 solvent 30 through the column 2 at the rate of 6 cc per minute.
24 Effluent from the column 2 passes through a reheater 40 where it is heated to 120C before passing through an IR detector 50 which is 26 used to measure the absorbance of the effluent stream. The infrared 27 absorbance of the polymer carbon-hydrogen stretching bands at about 28 2960 cm~l serves as a continuous measure of the relative 29 concentration of polymer in the effluent. After passing through the infrared detector 50 the temperature of the effluent is reduced to 31 about 110C and the pressure is reduced to l atmosphere before 32 passing the stream into an automatic fraction collection 60.
33 In the elution stage, the pure solvent is initially pumped 34 through the column 2 at the low initial temperature for one hour.
This serves to flush polymer that has not crystallized during the 36 crystallization stage out of the column 2 so that the percent of 37 uncrystallized polymer can be determined from the infrared trace.
38 The temperature is then programmed upward at 10C per hour to 100C
_ 4 _ 1 335072 1 and at 20C per hour from lO0C to 120C.
2 The compositions of fractions obtained from the various 3 polymers are determined by infrared spectroscopy. The IR
4 compositions are obtained from the intensity of the l378 cm~l methyl band, the thickness of the sample, and a calibration curve 6 based on samples whose compositions can be determined independently 7 by Cl3 NMR. No corrections for polymer end groups is usually made 8 in obtaining compositions from infrared data.
g Example l An example of a solubility distribution obtained from a 11 typical LLDPE sample, Dow 2045 [poly(ethylene-co-octene), 0.9l8 g/cc, 12 melt index l.0] obtained using the procedure of the invention 13 described above is shown in Figure 3. A composition scale obtained 14 by analysis of fractions from poly(ethylene-co-butene), poly(ethylene-co-hexene) and poly(ethylene-co-octene) eluted at 16 various temperatures is also shown in Figure 3.
17 As seen in Figure 3, the solubility versus temperature 18 distributions for the polymer has a peak for the 6C first-hour 19 elution temperature. This initial peak represents the fraction of total polymer that is not crystallizable at the lowest temperature of 21 the experiment (about 6C). The typical run length for a sample is 22 about 40 hours.
23 Example 2 24 Another example of a solubility distribution and composition distribution is shown in Fiqure 4. The polymer sample was Mitsui 26 Petrochemical Company Tafmoer A4085TM poly(ethylene- co-butene, density 27 ~0.88 g/cc, melt index lØ The curve was obtained according to 28 the invention using the procedures described above. This example 29 illustrates the importance of elution at sub-ambient temperature since only a small portion of this sample would have been resolved by 31 elution only at 30C and above as done in the prior art.
32 The apparatus and method described above provide a plot of 33 concentration of polymer versus elution temperature. The fractions 34 of solvent containing dissolved copolymer can then be analyzed by conventional means for molecular weight and composition to establish 36 the molecular weight distribution and composition distribution of 37 monomer. Alternatively to determining the monomer composition for 38 each fraction, a calibration curve relating elution temperature to _ 5 _ 1 3 3 5 0 7 2 1 monomer composition can be generally derived for specific comonomers.
2 The foregoing description is illustrative and explanatory of 3 the invention and is not intended to limit the invention to the 4 specific embodiment that is described.