Title of Invention	A PROCESS FOR POLYMERISING OLEFINS
Abstract	The invention discloses a process for polymerising olefins in two liquid full loop reactors connected in series wherein different molecular weight fractions are produced in the presence of a Ziegler-Natta catalyst system, characterised in that the Ziegler-Natta catalyst has a particle size distribution d50 of less than 20µm and greater than 8µm.

Full Text

The present invention concerns the use of a catalyst component having controlled grain size to prepare polyolefins and to prevent or reduce defects in products made from these polyolefins. The invention concerns an olefin polymerisation process using a Ziegler Natta-type catalyst. Olefin polymerisation processes are generally known. Further, it is well known that polymers of olefins can be prepared by olefin polymerisation in a hydrocarbon diluent or in monomers acting as diluents. On an industrial scale, one reactor type, which may be applied in such processes, is a turbulent flow reactor such as a continuous pipe reactor in the form of a loop. However, other types of reactors such as stirred reactors may be used. Polymerisation is carried out in a loop reactor in a circulating turbulent flow. A so -called loop reactor is well known and is described in the Encyclopaedia of Chemical Technology, 3rd edition, vol. 16 page 390. This can produce LLDPE and HDPE resins in the same type of equipment. The loop reactors may be connected in parallel or in series. In this regard, in a double loop reactor where the two reactors are connected in series, a high molecular weight fraction may be produced in the first loop reactor and a low molecular weight fraction may be produced in the second loop reactor. In this way, a bimodal polymer or a polymer having a broad molecular weight distribution is made. In a double loop reactor where the two reactors are connected in parallel, either a monomodal or a bimodal product is made. EP0649860, the contents of which are incorporated herein by reference, describes a process for producing polyethylene in two liquid full loop reactors, connected in series. The ethylene is injected with the comonomer in the firs t loop reactor as well as the catalytic system (i.e. the catalyst precontacted with the activating agent). Suitable comonomers, which can be used, include alpha -olefins with from 3 to 10 atoms of carbon, preferably 1 -hexene. Polymerisation is done at a t emperature of between 50 and 120°C, preferably between 60 and 110°C, and at a pressure between 1 and 100 bars, preferably between 30 and 50 bars. The flow of ethylene polymer obtained in the first reactor is transferred into the second reactor by means of one or more settling legs of the first reactor, for example by using two settling legs (each being filled independently with the suspension coming from the reactor, the solids being concentrated by gravity settling and discharge). In any olefin polymerisation process, the polyolefin is produced in the reactor in the presence of an olefin polymerisation catalyst. Such catalysts generally may be classified into three groups: metallocene -type catalysts, chromium-type catalysts and Ziegler -Natta-type catalysts. Typically, the catalyst is used in particulate form. The polyolefin is produced as a resin/ powder (often referred to as "fluff") with a hard catalyst particle at the core of each grain of the powder. The "fluff" is removed from the reactor and must be extruded before it is sold. Typically, an extruder works by melting and homogenising the "fluff" and then forcing it through holes before cutting to form pellets. The pellets then may be transformed by subjection to further processing in applications such as pipe making, fibre making, and blow-moulding. In processes where the catalyst used in the olefin polymerisation is a Ziegler-Natta-type catalyst, the present inventors have become aware that defects sometimes are visible in the finished product. In p articular, the present inventors have become aware that dots or specks and/or rough patches sometimes are visible on the surface of a pipe made from pellets of a bimodal product that were produced using a Ziegler-Natta or metallocene catalyst. Such defects can make the pipe weaker and can affect the free flow of liquid through the pipe. The present inventors have recognised that these defects likely occur because of homogenisation problems in the extruder. As such, attempts have been made to try to prevent or reduce these defects. In particular, efforts have been concentrated on optimising the specific energy delivered to the extruder. However, this has not proved an entirely satisfactory solution to the problem. In view of the above, there remains a need to provide a new olefin polymerisation process for making polyolefin resin and, subsequently polyolefin pellets, wherein defects in any product that is produced from the polyolefin pellets are prevented or reduced. Accordingly, the present invention disci oses a process for polymerising olefin that is carried out in the presence of a Ziegler-Natta catalyst system in two liquid full loop reactors connected in series wherein different molecular weight fractions are produced, characterised in that the Ziegler-Natta catalyst has a particle size distribution d 50 of less than 20um and greater than 5um. The present invention also discloses a polyolefin obtainable by the process according to the present invention. The present invention further discloses the use of a polyolefin formed by the process according to the present invention for making pipes, particularly for making a water or gas (such as liquid propane or natural gas) pipe distribution network. The present invention yet further discloses a pipe and a distribution pipe network prepared from the polyolefins of the present invention. Clearly, in the process according to the present invention, the catalyst necessarily is in particulate form. The catalyst's particle size distribution (PSD) d 50 is measured by laser diffraction analysis on a Malvern type analyser (Malvern 2000S) after having put the catalyst in suspension in cyclohexane. The particle size distribution d50 is defined as the particle size for which fifty percent by volume of the particles has a size lower than the d50. The fluff particle size distribution (PSD) d 50 is measured according to ASTM method D 1921-89 and is defined as the particle size for which fifty percent by weight of the particles has a size lower than the d50. The particle si ze distribution of the catalyst used in the present process is lower than previously used in olefin polymerisation processes. It was expected that by using a smaller catalyst size a correspondingly smaller resin particle would be obtained. In this regard, it has been thought previously in this art that a smaller resin particle size would be disadvantageous because it in fact was thought that it would be more difficult to homogenise the resulting polyolefin because smaller resin grains would not flow as ea sily through the extruder. It was also expected that smaller resin particles would result in a lower settling efficiency. By using a smaller catalyst particle size, the size of the grains in the produced polyolefin resin ("fluff") also is smaller. This has been found to have the benefit of improved and easier homogenisation during extrusion because the "fluff" is able to be made fully molten. Other advantages of a smaller resin particle size also have been found unexpectedly. These include: Improved unit throughput in the process. The smaller resin particles pack together more efficiently in the settling legs. This means that less diluent and more solid resin product is removed per unit volume. This effect may be attributed to improved "fluff" bulk density (BD) which gives higher settled "fluff" mass in the reactor legs (Bulk density is measured following the method of standard test ASTM D 1895). - Comonomer (for example hexene) incorporation is improved because a lower amount of comonomer is necessary to obtain an equivalent density. - Lower pump power consumption to circulate the solids in the reactor. - Resin product can remain in the reactor for longer. This is because higher solids levels can be maintained in the reactor with the same power consumption on the circulation pump. A higher fluff residence time enables an improvement in the catalyst yield. In other words, more product is produced per kg of catalyst. One way of indirectly measuring the "fluff" bulk density is by obtaining the ratio of diluent to monomer injected into the reactor. This is measured at the reactor feeds. As such, where the diluent is isobutane and the polymer is polyethylene, a ratio isobutane:ethylene can be obtained as an indication of the fluff settling efficiency. This rati o substantially is not effected by decreasing the d 50 of the catalyst. This is because the lower d 50 of the "fluff" is compensated by the higher settled "fluff" mass in the reactor legs (due to an improved "fluff" bulk density). Also, it has been found t hat the level of fines present in the "fluff" is not disadvantageously higher when using a catalyst with a lower d50. The present inventors have found that when using a smaller catalyst particle size, the size of the produced resin particle is not as small as expected. This is because the resin particle size in fact depends on two factors. These are: (1) the catalyst particle size and (2) the catalyst productivity. Unexpectedly, it has been found that productivity is increased when the catalyst particle is made smaller. As such, a smaller difference in resin particle size is observed than was expected (because it was expected that productivity would remain the same). To exemplify the increase in productivity, it can be reported that, previously, 1g of cat alyst (particle size 23µm) produced 5,000-10,000g of resin. It has now been found that lg of catalyst (particle size 13µm) produces 20,000g of resin. As such, the present invention also provides a process wherein the catalyst has an improved productivity. Increased productivity means a lower catalyst cost per unit resin. Preferably, the catalyst has a d50 of less than 15µm. Preferably, the catalyst has a d 50 of greater than 8µm. Most preferably, the catalyst has a d 50 of about 13µm. Desirably, the polyolefin resin made by the present process has a particle size distribution of less than 500µm, more desirably about 400µm. This can be compared with previously known processes where the polyolefin may have had a particle size distribution of greater than 600um. Generally, Ziegler-Natta type catalysts usable in the present process comprise a transition metal compound (compound A) of Group IV-VIII (mainly Ti or V) supported on a carrier. Such catalysts are well known in the art. Examples of Ziegler-Natta catalysts are TiCl4, TiCl3, VCl4, VOCl3- An MgCl2 support is preferred. Preferably, the Ziegler -Natta catalyst comprises from 10 to 18 % Mg by weight and from 3 to 10 % Ti by weight. More preferably, the Ziegler -Natta catalyst comprises about 13% Mg by weight and about 7% Ti by weight. It will be understood in the present process that, where necessary, an activating agent will be needed to activate the catalyst. Suitable activating agents, where needed, are well known in this art. Suitable activating agent s include organometallic or hydride compounds of Group I to III, for example organo-aluminium compounds such as those of general formula AlR3, Rµ2AlCl or Rµµ3Al2Cl3 in which R, Rµ, and Ruu each independently is a hydrocarbon radical, preferably an alkyl group with 1 to 16 carbon atoms, preferably from 2 to 12 carbon atoms. Suitable activating agents include, for example Et3Al, Et2AlCl, and (i-Bu)3Al. One preferred activating agent is triisobutylaluminium. Typically the polymerisation process will be carried out in a hydrocarbon diluent. One suitable diluent is isobutane. A separate catalyst diluent may be required. Suitable catalyst diluents w ill be known to those skilled in this art. Preferably, the present process is used to make an ethylene or propylene, homopolymer or copolymer. Also, preferably the present process is used to make a polymer (for example polyethylene) having a broad mole cular weight distribution, such as a bimodal polymer such as bimodal polyethylene. The molecular weight distribution (MWD) can be fully described by means of the graph obtained by gel permeation chromatography. However, the molecular weight distribution can be generally described by a figure which represents the ratio between the mean molecular weight by weight and the mean molecular weight by number (the polydispersisity index). Depending on the applications, the molecular weight distribution required may vary from 10 to 30, preferably of from 12 to 24. More preferably, the process is used to make a polymer (for example polyethylene) having a typical molecular weight distribution of greater than 15. Preferably, hydrogen is used to control the molecular weight of the polymer. A higher pressure of hydrogen leads to a lower mean molecular weight. Any suitable reactors can be used,, for example one or more loop reactors and/or one or more continuously stirred reactor. Preferably the present process is carr led out in a two-reactor system wherein at least one reactor is a loop reactor. Preferably, the process for producing polyethylene is carried out in two liquid full loop reactors (a "double loop" reactor) as referred to above. In this regard, it is prefe rable that the present process is carried out in accordance with the process of EP 0649860, as described above. Preferably a low concentration of hydrogen is maintained in the first reactor, for instance between 0 and 0.1% by volume, and a high concentrat ion of hydrogen in the second reactor, for instance between 0.5 and 2.4% by volume. Where a double loop reactor is used, preferably the polymers formed in the first reactor have an HLMI (high load melt index according to ASTM standard D1238, 190°C/21.6 kg) between 0.01 and 5 grams per 10' preferably between 0.1 and 2 grams per 10'. Preferably, the final polymers have an HLMI higher than 5 grams per 10' . The properties of the resin are however selected according to the end use of the polymer. Typical prop erties of resins suitable respectively for pipe, blow-moulding or film applications are summarised in Table 1 MI2, MI5 and HLMI are measured following the methods of standard test ASTM D 1238 at a temperature of 190 °C and under a load respectively of 2.16 kg, 5 kg and 21.6 kg. A preferred reaction temperature range in the present process may be said to be from 60 to 120, more preferably from 75 to 100°C. A preferred applied pressure range may be said to be from 30 to 55 bars, more preferably from 40 to 50 bars. The reactor pressure to some extent controls the quantity of slurry that is taken out of the reactor. One embodiment of a "double loop" reactor process can be described as follows: -The process is a continuous process. A monomer (e.g. ethylene polymerises in a liquid diluent (e.g. isobutane) in the presence of a comonomer (e.g. hexene), hydrogen, catalyst, activating agent and anti-fouling agent. The slurry is maintained in circulation by an axial pump consisting in a reactor essentially of vertical jacketed pipe sections connected by trough elbows. The polymerisation heat is extracted by a water -cooling jacket. The reactor line includes two reactors that can be used in parallel or in series. The approximate volume of the reactors may be about 100m3. -The product (e.g. polyethylene) is taken out of the reactor with some diluent through settling legs and discontinuous discharge valves. A small fraction of the total circulating flow is withdrawn. It is moved to a polymer degassing section in which the solid content is increased. -While being depressurised, the slurry is transferred through heated flash lines to a flash tank. In the flash tank, the product and diluent are separated. The degassing is completed in a purge column. -The powder product is transported under nitrogen to fluff silos and extruded into pellets along with some specific additives. A pellet treatment unit comprising silos and hot and cool air flow allows the removal of residual components from the pellets. The pellets then are directed to homogenisation silos before final storage. -The gas coming out from the flash tank and from the purge column is treated in a distillation section. This allows the separate recovery of diluent, monomer and comonomer. -This embodiment of the double loop reactor process in fact would be usable with chromium type, Ziegler -Natta type or indeed metallocene catalysts. Each catalyst type would have a specific injection system. Examples and Experiments An evaluation was carried out in order to compare a catalyst having a particle size distribution d50 of 23µm with a catalyst having a particle size distribution d50 of 13µm. The evaluation was carried out in a double loop reactor and was divided into four time frames as explained below: I. Polyethylene production using the standard 23µm catalyst in a "double loop" reactor. II. Polyethylene production using 13um catalyst. III. Polyethylene production using 13um catalyst at maximal reactor throughput. IV. Polyethylene production using standard 23um catalyst. List of accompanying figures. Figure 1 represents the melt index HLMI expressed in dg/min as a function of time expressed in days f or the 13 urn particles in reactor 1 at the beginning of the evaluation. Figure 2 represents the melt index MI5 expressed in dg/min as a function of time expressed in days in reactor 2 for the 13 µm particles at the beginning of the evaluation. Figure 3 represents the reactor rate expressed as the ratio of high molecular weight fraction to total production as a function of time expressed in days for the 13 urn particles at the beginning of the evaluation. Figure 4 represents the particle size distribution (PSD) expressed in wt% as a function of particle size expressed in microns. Figure 5 shows a comparison between the 13 microns and 23 microns fluff PSD. Figure 6 shows the fluff's d50 expressed in microns and the span evolution expressed as the ratio (d 90-d10)/d50 as a function of time expressed as the date of experiment. Figure 7 represents the weight percent of fluff after respectively the 1000 microns/ 63 microns and bottom sieves as a function of time expressed as the date of experiment. Figure 8 is a plot of the ratio C2/iC4 in the feed vs bulk, density expressed in g/cm3 of the polymer fluff in reactor 1, Figure 9 is a plot of the ratio C2/iC4 of the feed vs bulk density expressed in g/cm3 of the polymer fluff in reactor 2. Figure 10 represents the solid content expressed in weight percent of fluff as a function of time expressed as the date of experiment. Figure 11 represents the pump power consumption expressed in kw per weight percent of fluff as a function of time expressed as the date of experiment. Figure 12 shows the productivity expressed in g/g as a function of solid residence time expressed in hours. Experimental Detail The conditions have been adjusted to obtain equivalent Reactor 1 melt index and reactor ratio for the 13um and 23pm catalysts. (See Figure 1, Figure 2 & Figure 3). Influence of fluff morphology on reactor throughput The particle size distribution (PSD) of 13 -µm and 23-µm catalysts are given in Figure 4. The polyethylene fluff PSD was lower when the 13µm catalyst was us ed (See Figure 5) . The d50 is reduced from about 625 µm down to 400 µm. The higher span [ (d90 - d10)/d50] is due to the lower d50, the PSD broadness being about the same (See Figure 6). Figure 7 shows that the 13um catalyst generates a smaller quantity of large fluff particles (about 1% instead of 6 - 8 % using 23 µm catalyst) but produces the same quantity of fines (bottom and 63 µm sieves) • This fluff morphology should be advantageous, when the product is used in pipe - making, for the improvement of the inner aspect of the pipe product. Despite the lower fluff d50 coming from the 13µm catalyst, the reactor throughput was maintained due to a significant improvement of the bulk density observed in Reactor 1 and Reactor 2 (see Figure 8 and Figure 9) . This higher bulk density had an important impact on the settling efficiency. It was observed that: - Owing to the lower pump power consumption for the lower fluff particle size using 13 -µm catalyst, it was possible to raise the solids content in Reactor 1 and, in this way, increase the solids residence time (See Figure 10 and Figure 11) and thus the productivity. The 13um catalyst has the same hydrogen response as the 23µm catalyst whereas the comonomer incorporation of the 13µm catalyst is better than that of the 23µm catalyst. Taking into account the solids residence time and the ethylene off-gas, the activity of 13-µm catalyst is 20 - 30% higher to the activity of 23 -µm catalyst (See Figure 12). - The molecular weight distribution is the same for both catalyst grain sizes. - Despite the lower fluff d50 (400 µm instead of 625 µm), the reactor throughput was not penalised thanks to the improvement of fluff bulk density (+0.04) generating good settling and high solids in both reactors. The lower fluff d50 is due to a lower amount of big particles {1000 µm), the amount of fines ( being the same or even lower. This is interesting in terms of settling efficiency and pipe aspect. WE CLAIM: 1. A process for polymerising olefins in two liquid full loop reactors connected in series wherein different molecular weight fractions are produced in the presence of a Ziegler-Natta catalyst system, characterised in that the Ziegler-Natta catalyst has a particle size distribution d50 of less than 20µm and greater than 8µm. 2. A process according to claim 1, wherein the Ziegler- Natta catalyst has a d50 of less than 15µm. 3. A process according to any one of the preceding claims, wherein the Ziegler-Natta catalyst has a d50 of about 13µm. 4. A process according to any one of the preceding claims, wherein the present process is used to form a polyethylene. 5. A process according to claim 4, wherein the process is used to make a polyethylene having a typical molecular weight distribution of greater than 15. 6. A process according to any one of the preceding claims wherein the ethylene is injected with a comonomer in the first liquid full loop reactor and the comonomer comprises an alpha-olefin having from 3 to 10 atoms of carbon. 7. A process according to any one of the preceding claims, wherein the formed polyolefin is extruded to form pellets. The invention discloses a process for polymerising olefins in two liquid full loop reactors connected in series wherein different molecular weight fractions are produced in the presence of a Ziegler-Natta catalyst system, characterised in that the Ziegler-Natta catalyst has a particle size distribution d50 of less than 20µm and greater than 8µm.

Documents:

02055-kolnp-2006 abstract.pdf

02055-kolnp-2006 assignment.pdf

02055-kolnp-2006 claims.pdf

02055-kolnp-2006 correspondence others.pdf

02055-kolnp-2006 description(complete).pdf

02055-kolnp-2006 drawings.pdf

02055-kolnp-2006 form-1.pdf

02055-kolnp-2006 form-3.pdf

02055-kolnp-2006 form-5.pdf

02055-kolnp-2006 international publication.pdf

02055-kolnp-2006 international search authority report.pdf

2055-KOLNP-2006-ABSTRACT 1.1.pdf

2055-KOLNP-2006-AMANDED CLAIMS.pdf

2055-KOLNP-2006-ASSIGNMENT.pdf

2055-KOLNP-2006-CANCELLED PAGES 1.1.pdf

2055-KOLNP-2006-CORRESPONDENCE 1.1.pdf

2055-KOLNP-2006-CORRESPONDENCE 1.2.pdf

2055-KOLNP-2006-CORRESPONDENCE.pdf

2055-KOLNP-2006-DESCRIPTION (COMPLETE) 1.1.pdf

2055-KOLNP-2006-DRAWINGS 1.1.pdf

2055-KOLNP-2006-EXAMINATION REPORT REPLY RECIEVED.pdf

2055-KOLNP-2006-EXAMINATION REPORT.pdf

2055-KOLNP-2006-FORM 1.1.1.pdf

2055-KOLNP-2006-FORM 18.1.pdf

2055-kolnp-2006-form 18.pdf

2055-KOLNP-2006-FORM 2.pdf

2055-KOLNP-2006-FORM 3.1.1.pdf

2055-KOLNP-2006-FORM 3.pdf

2055-KOLNP-2006-FORM 5.pdf

2055-KOLNP-2006-GPA.pdf

2055-KOLNP-2006-GRANTED-ABSTRACT.pdf

2055-KOLNP-2006-GRANTED-CLAIMS.pdf

2055-KOLNP-2006-GRANTED-DESCRIPTION (COMPLETE).pdf

2055-KOLNP-2006-GRANTED-DRAWINGS.pdf

2055-KOLNP-2006-GRANTED-FORM 1.pdf

2055-KOLNP-2006-GRANTED-FORM 2.pdf

2055-KOLNP-2006-GRANTED-SPECIFICATION.pdf

2055-KOLNP-2006-OTHERS(U.S., EP PATENT APPLN., COPY OF IPRP ON THE PCT APPLN.).pdf

2055-KOLNP-2006-OTHERS.pdf

2055-KOLNP-2006-OTHERS1.1.pdf

2055-KOLNP-2006-PCT PRIORITY DOCUMENT NOTIFICATION.pdf

2055-KOLNP-2006-PETITION UNDER RULE 137.pdf

2055-KOLNP-2006-REPLY TO EXAMINATION REPORT.pdf

2055-KOLNP-2006-TRANSLATED COPY OF PRIORITY DOCUMENT.pdf

2055-KOLNP-2006.pdf