BACKGROUND OF THE INVENTION

It is known to form polyketones by the polymerization of carbon monoxide and one or more olefins by a vapor phase polymerization process. For example, U.S. Patent No.

4,778,876 of M. J. Doyle et al. describes such a process in which the monomers are contacted with the catalyst compositions in the substantial absence of a liquid, non- polymerizable diluent. Any small quantity of diluent that might be present is vaporized and present in exclusively the gaseous state during the polymerization (Col. 3, lines 35- 40) . This teaching of having any added alcohol diluent present exclusively or substantially in the gaseous condition during the polymerization reaction is substantially echoed by other, more recent patent references: European Patent

Publication No. 248,483 (see Col. 3, lines 8-13 and Claim 2). European Patent Publication No. 443,587 (see Col. 3, lines 3- 8 and Claim 2); and European Patent Publication No. 506,168 (see Col. 5, lines 29-35 and Claims 8-9) . Netherlands Patent Application No. 9101114, filed June

27, 1991, and corresponding European Patent Publication No. 520,584 advocate a gas phase polymerization process for the manufacture of polyketones in which water is added to the polymerization reaction environment to enhance the polymerization activity of the polymerization process. The amount of water should range from more than one mole per grair. atom of Group VIII metal in the catalyst, preferably more than 10 moles, for example, from about 5,000 to about 50,000 moles per mole of Group VIII metal. The upper limit of water added is set by the requirement that gas should form the continuous phase in the reactor. The European patent citation exemplifies the add-on of 1 ml and 7 ml water, respectively, to 8 g of porous carbon monoxide/ethene

copolymer catalyst support. The present investigators have determined that a carbon monoxide/ethene copolymer becomes saturated at far higher levels of water add-on than 7 ml, namely, at about 40-43 g. The productivity using 7 ml of water (an add-on of about 88%) was inferior to that observed when a far lower add-on (14% for 1 ml of water) was employed.

SUMMARY OF THE PRESENT INVENTION

The present invention relates to a gas phase process for catalytically polymerizing carbon monoxide and at least one olefin to produce a polyketone. The particular improvement in the process of the present invention is conducting the polymerization in the presence of a catalyst support which is substantially saturated with a liquid, non-polymerizable diluent which increases the amount of polyketone that is formed in the process. In one preferred embodiment of the process, a ketone, such as acetone, is used with the addition of hydrogen to yield a product of reduced molecular weight, increased bulk density, and increased thermal stability.

DETAILED DESCRIPTION OF THE PRESENT INVENTION

The aforementioned U.S. Patent No. 4,778,876 is incorporated herein by reference for its generalized teaching of how polyketones can be formed by vapor phase polymerization processes in regard to the types of monomers which can be used and the types of catalysts which can be utilized. As indicated before, polymers of carbon monoxide and an olefin, such as ethylene, propylene or combinations of ethylene and propylene is an area of known expertise. Such monomers can be polymerized into high molecular weight linear alternating polymers of carbon monoxide and the olefin or olefins that are chosen by using a catalyst which comprises a compound of a Group VIII metal, such as palladium, cobalt and

nickel, an anion of a non-hydrohalogenic acid having a pK _a of less than about six and an appropriate polydentate ligand, such as a bidentate ligand of the type shown in the aforementioned U.S. Patent No. 4,778,876. In the vapor phase process shown in U.S. Patent No.

4,778,876, it is taught that the monomers need to be contacted with the catalyst composition in the substantial absence of a liquid non-polymerizable diluent such as a monohydric a cohol having from one to four carbon atoms. In accordance with the present invention it has been unexpectedly found that the amount of polyketone product that can be formed in a vapor phase polymerization process of the general type set forth in U.S. Patent No. 4,778,876 can be dramatically increased by utilizing a supported catalyst with the support being substantially saturated with a liquid, non- polymerizable diluent. In general terms, the amount that is allowed to remain and saturate the support should not be so large as to produce a visually identifiable amount of free liquid so as to qualify the polymerization reaction as a solution or suspension polymerization reaction. In other words, if one compares the entire amount of diluent in the reaction system to the amount adsorbed by the support, one would realize that, preferably, essentially all the liquid diluent is adsorbed on the support rather than being present as free, non-adsorbed water. The preferred way to insure saturation is to continue to add the diluent to the support until traces of liquid diluent just begin to be visually observable within the reactor. If the amount of diluent which is allowed to remain and saturate the support is expressed in terms of the amount of catalyst, it could range from about 3,000 equivalents of diluent for every equivalent of catalyst up to about 20,000 equivalents of diluent per equivalent of catalyst. The amount of diluent can range from about 50% to about 200% or more ( e.g., up to about 400%) by

weight of support or substrate, for example 100% to about 200% by weight of the support, so that the amount which remains results in an increase in the amount of polyketone polymer formed per gram of metal in the catalyst system which is used.

The type of solvent which is to be allowed to remain in the amounts disclosed herein should be those which are capable of coordinating to a metal. Especially preferred are those solvents containing an oxygen or nitrogen atom (or both) so as to be capable of coordinating to the metal in the catalyst via the unbonded electrons on the oxygen and/or nitrogen atom(s) , for example, therein. Examples of suitable. organic solvents include acetonitrile, such ketones as acetone, such carboxylic acids as acetic acid, and such alcohols as ethanol. Compatible mixtures of these can be used. Solvents containing phosphorus or sulfur atoms are also contemplated for use.

The catalyst to be used will be in contact with a high surface area support. For example, an inorganic support such as silica can be used as long as the potential contaminating effects of the support can be tolerated. An alternative support is a high surface area polyketone (e.g., carbon monoxide-ethylene formed by solution or suspension polymerization) . As earlier mentioned, the general types of monomers and catalyst described in U.S. Patent No. 4,778,876 can be used in accordance with the present invention.

Since the present process is a high activity polymerization procedure, vigorous stirring is needed to prevent setting up or clumping up of polymer particle product within the reactor.

In one embodiment of the present invention, it has been found that hydrogen gas can be added to the polymerization medium to produce a more thermally stable product having a higher bulk density and a reduced molecular weight. Hydrogen gas will produce a polyketone product having a much higher

proportion of less reactive alkyl end groups and a correspondingly lower amount of more reactive vinyl, ester, alkoxy, or acid end groups. For example, it has been found that a tapped bulk density of 0.66 g/ml was obtained when acetone was the diluent and 17 bar of hydrogen was added for a total pressure of about 82 bar. Moreover, such a product exhibited good thermal stability and had a C ₂ /C* value of 0.47 using the procedure shown in U.S. Patent No. 5,049,630, which is incorporated herein by reference in regard to its description of such procedure. When the hydrogen addition was 10 bar for a total pressure of about 82 bar, the bulk density was 0.63 g/ml. Use of the process when no hydrogen addition was made for a total pressure of about 82 bar resulted in a significantly less thermally stable product with a C-j/Cj value of 0 and a lower bulk density of 0.37 g/ml.

The product resulting from the instant process with a ketone diluent in which hydrogen addition is preferably used, is deemed to be novel in various respects. For one thing, it can have a bulk density of higher than about 0.60 g/ml, which is above the 0.55 g/ml value shown in U.S. Patent No.

4,914,183 (Example 4). The use of hydrogen will decrease end vinyl unsaturation that might be present in the polyketone formed with use of the ketone solvent (e.g. , acetone) thereby giving an essentially alkyl group-end capped product. Such a product will be substantially free of the more reactive vinyl, ester, alkoxy, or acid end groups.

The present invention is further illustrated by the Examples which follow.

COMPARATIVE EXAMPLE 1

This Example illustrates the inferior results that were obtained when an essentially vapor phase polymerization of carbon monoxide and ethylene was carried out in the substantial absence of a liquid non-polymerizable diluent. The catalyst preparation that was employed in this Example was a modification of that used in European Patent Publication No. 301,664, Example 3. The catalyst for the reaction was prepared as follows: 0.11 g palladium acetate, 0.22 g toluenesulfonic acid, and 0.22 g of l,3-bis(diphenyl- phosphino) propane were placed in a flask with a stir bar. To this was added 20 ml of dry acetone and 14 ml of dry acetonitrile. The mixture was then stirred for ten minutes at room temperature and was then warmed to 50°C for five minutes. This resulted in formation of a clear yellow solution. The solution was cooled to 30°C, and the solvent was removed under vacuum. The product was a yellow solid which was then dissolved in 50 ml of dry acetone for spraying on a high surface area polymer substrate (i.e., carbon monoxide-ethylene formed by solution polymerization) .

The polymer substrate was placed in a round bottom flask under nitrogen and was wetted with enough acetone so that not all of the acetone was absorbed by the polymer, i.e., there was a small amount of free acetone visible on the walls of the flask. The polymer sample (weighing 171 g) required approximately 300 ml of acetone to reach this point.

The solution of catalyst was then sprayed onto the wet polymer substrate while the substrate flask was vigorously shaken to ensure even dispersion of the catalyst. The polymer substrate with catalyst coated on it was then dried under vacuum to remove substantially all liquid solvent. The dry powder was then placed in an autoclave which was then flushed with nitrogen and pressurized to 82 bar with a 1:1

weight ratio of carbon monoxide:ethylene at a temperature of 65°C.

The data given below shows the sequential growth of a sample of polymer which was formed through such a polymerization process and which was then used as the polymer substrate.

COMPARATIVE EXAMPLE 2

The following table shows the results of a similar sequential reaction to that described in the previous Example in which the polymer substrate was formed through a solution process and was then coated with catalyst. Additionally, the catalyst charge was three times as large as in Example 1 (i.e., 0.33 g of palladium acetate, 0.66 g p-toluenesulfonic acid, and 0.66 g of l,3-bis(diphenylphosphino) propane), and the reaction was performed for a much longer time:

LVΝ

Weight Weight Time 100°C, Productivity in fq) out (g) (hours) m-Cresol (q polymer/g Pd/Hr)

125 2600 137 6.6 114

EXAMPLE 3

This illustrates the superior results in regard to polymer productivity when the solvent-impregnated support containing the catalyst does not have the liquid non- polymerizable diluent substantially removed before the vapor phase polymerization.

The following table shows the results of a similar reaction to that shown in the previous Example in which the polymer substrate was formed through a solution process then coated with catalyst. There was no evaporation of the acetone solution used to coat the catalyst onto the substrate. There was no free acetone in solution form, and the coated substrate was handled as if it were simply a solid even though residual solvent remained as defined in Comparative Example 1. Additionally, the catalyst charge was three times as large as Example 1 (i.e., 0.33 g palladium acetate, 0.66 g p-toluenesulfonic acid, and 0.66 g of 1,3- bis(diphenylphosphino) propane) . Finally, 4 bar of H ₂ were added to reduce the molecular weight of the product, increase the bulk density, and potentially improve the thermal stability of the polyketone product. Improved thermal stability can be of importance, for example, if the polyketone product is to be melt spun into fibers. Use of hydrogen as a molecular weight regulator is not deemed to materially affect productivity.

Weight Weight Time Productivity in (a) out (a) (hours) LVN (a polvmer/ Pd/Hr)

100 5500 17 1 2000

EXAMPLE 4

This Example shows the same general procedure employed in Example 3, without hydrogen addition, using a mixture of methanol and acetic acid as the inert liquid diluent. In this Example, 125 g of substrate, 0.5 millimole of catalyst, with the reaction of carbon monoxide and ethylene (1:1 weight ratio) being conducted at 1200 psig at 65"C were employed. Two hundred grams of methanol were used to initially saturate the catalyst support followed by the addition of 15 ml of acetic acid. The polymer product 3700 g with an LVN (100°C, m-cresol) of 6.0.

The foregoing Examples are being provided for illustrative purposes only and should not be construed in a limiting sense. The scope of protection which is sought is set forth in the claims which follow.