EP0614466A1

EP0614466A1 - Method for the preparation of an olefin polymerization catalyst consisting of particles of equal size

Info

Publication number: EP0614466A1
Application number: EP92924733A
Authority: EP
Inventors: Thomas Garoff; Jukka Koskinen
Original assignee: Borealis Holding AS; Neste Oyj
Current assignee: Borealis Holding AS
Priority date: 1991-11-29
Filing date: 1992-11-27
Publication date: 1994-09-14
Also published as: FI92327C; FI92327B; FI915630A0; FI915630A; NO941970D0; NO941970L; JPH07501568A; WO1993011164A1

Abstract

Selon cette invention, on prépare une composition de procatalyseur particulaire destinée à la polymérisation d'oléfines en faisant réagir les particules porteuses composées d'un complexe MgCl2-C2H5OH avec du TiCl4. Lorsque ce procédé est appliqué à une échelle pilote, il a été remarqué que la composition de procatalyseur obtenue produisait un polymère oléfinique, dont la partie composée d'un matériau finement divisé était trop importante et pouvait en provoquer la détérioration. Ceci était dû au fait que les particules de la composition de procatalyseur s'étaient décomposées en un matériau finement divisé au cours de la synthèse, la structure de ce matériau étant alors reproduite, au cours de la polymérisation, dans la structure du polymère par le phénomène dit de répétition. Selon l'invention, il a été remarqué, de façon imprévue, que le fractionnement des particules porteuses n'était pas le résultat d'une contrainte mécanique, par exemple, mais était dû au fait que, lorsque les particules porteuses réagissent avec le TiCl4, une libération importante de chlorure d'hydrogène est initialement provoquée par la réaction qui fractionne les particules porteuses. Celles-ci peuvent, selon l'invention, demeurer intactes au cours de la réaction dans la mesure où l'on répartit la libération du chlorure d'hydrogène, dans la réaction ou à partir de la réaction, à une vitesse molaire ne dépassant pas cinq fois la valeur de la vitesse molaire moyenne de la libération. L'on effectue ceci, entre autres, en ajoutant les réactifs avec une lenteur inhabituelle, à des températures et des vitesses de chauffage plus basses, en intensifiant l'agitation, en utilisant une surpression et/ou en amenant un gaz inerte diluant dans le mélange de réaction.According to this invention, a particulate procatalyst composition intended for the polymerization of olefins is prepared by reacting the carrier particles composed of a MgCl2-C2H5OH complex with TiCl4. When this process is applied on a pilot scale, it was noticed that the procatalyst composition obtained produced an olefinic polymer, the part of which composed of a finely divided material was too large and could cause its deterioration. This was because the particles of the procatalyst composition had broken down into a finely divided material during the synthesis, the structure of this material then being reproduced, during the polymerization, in the structure of the polymer by the phenomenon known as repetition. According to the invention, it was unexpectedly noticed that the splitting of the carrier particles was not the result of mechanical stress, for example, but was due to the fact that when the carrier particles react with the TiCl4 , a large release of hydrogen chloride is initially caused by the reaction which splits the carrier particles. These can, according to the invention, remain intact during the reaction insofar as the release of hydrogen chloride is distributed, in the reaction or from the reaction, at a molar rate not exceeding five times the value of the average molar rate of release. This is done, among other things, by adding the reactants with unusual slowness, at lower temperatures and heating rates, by intensifying the agitation, by using overpressure and/or by passing an inert diluent gas into the reaction mixture.

Description

Method for the preparation of an olefin polymerization catalyst consisting of particles of equal size

The invention relates to a method for the preparation of a particulate procatalyst composition intended for the poly¬ merization of olefins by reacting the carrier particles consisting of a MgCl₂-C₂H₅OH-complex with TiCl₄.

For the polymerization of olefins, generally, the Ziegler- Natta-catalyst system, which consists of a so-called pro¬ catalyst and a co-catalyst, is used. The procatalyst is based on a compound of a transition metal belonging to any of the groups IVA-VIII (Hubbard) of the periodical table of the elements and the co-catalyst is based on an or- ganometallic compound of a metal belonging to any of the groups IA-III(A) (Hubbard) of the periodical table of the elements. The catalyst system can also comprise a carrier, on which the transition metal compound is layered, and electron donor compounds improving and modifying the cata¬ lytic properties.

When Ziegler-Natta-procatalysts intended for the polymeri¬ zation of olefins and particularly of propene are prepa- red, carrier particles consisting of solid magnesium chlo¬ ride and ethanol are reacted with titanium tetrachloride. As a result a magnesium chloride is created, the crystal structure of which is strongly modified, i.e. becoming in practice amorphous and which can coordinate with titanium tetrachloride. Thus, the titanium tetrachloride, which has been combined with the amorphous magnesium chloride, then forms the active centers of the catalyst. As byproducts of this process hydrogen chloride and ethoxy titanium tri¬ chloride (1, 2) are formed:

MgCl₂.EtOH + TiCl₄ = MgCl₂.+ TiCl₃.0Et + HC1 (1) 10MgCl₂. + TiCl₄ = 10MgCl.TiCl₄ (2)

In laboratory conditions polyolefins have been obtained using the procatalysts prepared in this manner. The acti- vity of the catalyst has been good. The particle size, particle size distribution and other properties of the polyolef ns have been satisfactory.

When going over from laboratory scale to production in larger scale it has, however, been noticed that the par¬ ticle size distribution of the polymer product is every¬ thing but satisfactory. The product contains, namely, com¬ pared with a product prepared with a laboratory procata- lyst very great amounts of finely-divided substance or material, the diameter of which is less than 1mm. This can be seen among other things in fig. 1, where the portion of the finely-divided material becomes multiplied when going over from a laboratory-prepared procatalyst to a procata- lyst prepared in pilot-scale. A so-called "replica"-phe¬ nomenon prevails in polymerization using this kind of a procatalyst. This means that the morphology of the polymer particles created corresponds to the morphology of the procatalyst used in the preparation of the polymer, it could right off be assumed that the creation of the fine¬ ly-divided material in the polymer was a result of the decomposition of the procatalyst particles intended for the polymerization in connection with the pilot-scale preparatio .

In the preparation of the procatalysts it was unexpectedly noticed that there prevailed a clear correlation between the maximum value of the releasing velocity of the hydro¬ gen chloride created in the reaction (1) and the portion of the finely-divided substance. This correlation is pre¬ sented in fig. 2. The figure shows that the portion of the finely- ivided polymer increases when the momentary hyd¬ rogen chloride discharge strengthens. In the same connec¬ tion it was noticed that the hydrogen chloride discharge strengthened when the ethanol portion of the magnesium chloride-ethanol complex increased. From this an idea came out that the scale-up problem of the catalyst preparation attached with the creation of the finely-divided polymer would result from a too heavy release of the hydrogen chloride.

For this reason a system, in which there was present mag¬ nesium chloride diethanol complex, titanium tetrachloride and hydrogen chloride created in a reaction between these components, a system, in which there was present only titanium tetrachloride and hydrogen chloride, and a sys- tem, in which there was present magnesium dichloride, titanium tetrachloride and hydrogen chloride, were compa¬ red with each other.

When the first-mentioned system was heated from the tempe- rature -20°C to the temperature of +110°C a heavy dischar¬ ge of hydrogen chloride took place between +10°C and +20°C, after which the formation of the hydrogen chloride became normalized corresponding to the normal behaviour of a dissolved gas at the temperature of above 40°C. This phenomenon can be seen in figures 3 and 4, figure 3 of which presents the amount of the hydrogen chloride re¬ leased from the system (expressed as the consumption of NaOH) as a function of the temperature and the reaction time and figure 4 presents the releasing velocity of hyd- rogen chloride of the system as a function of the tempera¬ ture.

In the second- and third-mentioned systems, of which mag¬ nesium chloride ethanol complex was missing no discharge of the above-mentioned type was observed.

On the basis of the preliminary tests it could thus be noted that the creation of the detrimental finely-divided polyolefin when a procatalyst prepared in pilot-scale was used, which has been obtained from a reaction between the MgCl₂-C₂H₅0H complex particles and TiCl₄, is essentially a result thereof that the reaction mentioned and presented in the reaction equation (1) produces on one time so much hydrogen chloride that when it is gasified at once it blows the carrier particles into finely-divided material, which then repeats in the polyolefin to be created.

The aim of the invention is to provide a method for the preparation of a particulate procatalyst composition inte- ded for the olefin polymerization. The procatalyst compo¬ sition should be created using a method giving a usable particle size distribution, from which the finely-divided material is essentially missing. Moreover, the particles must be suitable for use as to their shape, chemical com¬ position, activity and stereospecifity. In this invention it is also tried to see to it that the polymer prepared by the procatalyst composition does not contain finely-divi¬ ded material to any considerable amount. An aim is simul¬ taneously a polyolefin, the particle shape, isotacticity and crystallinity, melt viscosity and bulk density of which are suitable for use.

This aim taken has now been reached by a new method for the preparation of the particulate procatalyst intended for the polymerization of olefins, which is mainly charac¬ terized by what is mentioned in the characterizing clau- se of claim 1. It has thus been realized that the carrier particles consisting of MgCl₂-C₂H₅OH-complex expressly de¬ compose as a result of a hydrogen chloride discharge, when they are activated with titanium tetrachloride. (Previously it was believed that the decomposition among other things is a result of the mechanical strain caused by the mixing). This phenomenon occurs above all when the preparation of the procatalyst takes place in larger, e.g. in pilot-scale. Due to the decomposition of the carrier particles both the carrier and thus also the procatalyst composition contains a great deal of finely-divided ma¬ terial. The particle size distribution of the procatalyst composition then repeats in the polymerization in the particle size distribution, whereby even the polymer comp¬ rises a great deal of finely-divided material.

The above-mentioned realization that the great amount of the finely-divided material in the polymer product is a result of the heavy discharge of the hydrogen chloride gas in the procatalyst synthesis, enables such a method for the preparation of a procatalyst, which gives a procata¬ lyst composition and also a polyolefin containing very little finely- ivided material. In the method the carrier particles are maintained undamaged during the reaction by distributing the release of the created hydrogen chloride in the reaction or from the reaction mixture to a molar velocity which does not exceed the fivefold molar mean velocity value. When scanning again the figures 3 and 4 it is noticed that without the special measures according to the invention the maximal velocity of the discharge of the hydrogen chloride rises to a value corresponding to a NaOH-consumption of about 16ml/°C, the mean discharge velocity being of an order corresponding to the NaOH-con¬ sumption of about 2ml/°C. These valves corresponds to an experimental arrangement where 0.1 mol of MgCl₂.3EtOH is reacted with 300 ml of TiCl₄. In the experiment concerned the hydrogen chloride discharge of a system was measured, the complex of which contained three ethanol molecules as per one magnesium chloride molecule, whereby the discharge peak of the hydrogen chloride thus was about eight times as high as the mean discharge velocity. Be it mentioned that when the complex contains 4.5 moles ethanol as per magnesium chloride mole the maximal discharge corresponded to even a NaOH-consumption of 90ml/°C, the mean volume velocity of the hydrogen chloride release being cf the order corresponding to a NaOH-consumption of l-5ml/°C i.e. the discharge was above 20-fold compared with the mean value of the release velocity.

Figure 4 also shows that the release of the hydrogen chloride should be restricted so that not more than about 5% could be discharged in a minute, preferably about 2% at the most of the total amount of hydrogen chloride release stoichiometrically in the reaction. This is another way to express the prevention of a quantitatively heavy discharge of hydrogen chloride.

It is, within certain limits, advantageous to minimize the momentary discharge of hydrogen chloride, so that the breaking of the carrier particles would be avoided as efficiently as possible. According to one embodiment of the invention the release of the hydrogen chloride created in the reaction between the carrier particles and the titanium tetrachloride is disributed to a molar velocity that does not exceed the threefold, preferably double, value of the molar mean velocity of the release.

The invention is based on the idea that the reaction bet¬ ween the carrier particles and the TiCl₄ is maintained in such a way that a sudden discharge of the hydrogen chlori¬ de is prevented during the activation. The attenuation of the sudden discharge i.e. the distribution of the hydrogen chloride release can be carried out by any suitable chemi¬ cal and/or physical means. The prevention of sudden chemi¬ cal discharges is in the technical chemistry a general measure and the present invention is not related to cer- tain embodiments of this general measure only, but ex¬ pressly to the use of this measure to prevent the forma¬ tion of finely-divided material in a procatalyst.

The release of the hydrogen chloride can be distributed by selecting a correct adding order and adding velocity of the reagents. The adding order preferably is such that the TiCl₄ is added -to the carrier particles and not vice ver¬ sa. Moreover, the carrier particles have preferably been suspended into an inert reaction medium. According to one embodiment of the invention TiCl₄ is controllably added to a liquid suspension of carrier particles consisting of MgCl₂-C₂H₅OH-complex during 0.5 to 3 hours, preferably about 1.0 hour. As the medium of liquid suspension in regard to reagents inert hydrocarbons are usually used, the amount of which can be regulated so as to decrease the discharge tendency of the hydrogen chloride.

The controlled, slow adding of TiCl₄ reduces the discharge risk of the hydrogen chloride by producing hydrogen chlo¬ ride more slowly (see the reaction formula (1) ). Hydrogen chloride has then time to diffund out from the carrier particles before oversaturation of hydrogen chloride in the solution, leading to gas formation in the pores takes place. Moreover, at least some stages in the reactions between a MgCl₂-C₂H₅0H-complex and TiCl₄ apparently are exothermic, whereby the slow adding of TiCl₄ enables a slow removal of heat from the carrier particles and thus prevents the carrier particles from heating locally above the discharge temperature of the hydrogen chloride.

The aforesaid also shows that TiCl₄ shall preferably be added in so low a temperature that no fast reaction nor a discharge of hydrogen chloride will take place. According to one embodiment TiCl₄ is added at the temperature -30 to -10°C and preferably at the temperature -25 to -15°C.

Due to the thermodynamic and/or physical balancing systems in the pores of the carrier particles and the TiCl₄-solu- tion hydrogen chloride is released during a long time after the adding of the TiCl₄. Hereby it is advantageous to raise the temperature during the adding of the TiCl₄ and/or after it with a controlled slow velocity about 5 to 20°C/h, more advantageously with a velocity about 5 to 15^cC/h. It is worth continuing this slow increase of the temperature up to a temperature of +40°C and especially above it, when the internal donor has been added to the reaction mixture of the carrier particles and the TiCl₄.

The release of the hydrogen chloride created in the reac- tion can also be distributed by using a strong agitation during the reaction between the carrier particles and the TiCl₄. Although the strength of the agitation is dependent of many factors such as other means for distributing the release of the hydrogen chlorid, it can be said that the amount of agitation exceeds in this invention the conven¬ tional agitation values and causes an essential distribu¬ tion of the release of the hydrogen chloride. As in the preparation of procatalyst compositions the agitation has often been accused for the breaking of the carrier parti¬ cles, it is very surprising that the adding of agitation will help keeping the carrier particles undamaged. The realization to add agitation to reach the aims of this invention must be seen against this background. Be it mentioned that in the reactor used in connection with this invention the starting speed of rotation of the mixer was raised when going over to the method according to this invention from the value 15rpm to the value 30rpm.

Another way to increase the material and heat transfer distributing the release of the hydrogen chloride is to alter the viscosity of the liquid reaction mixture. This can take place, as mentioned above, by selecting a suitab¬ le dispersion medium such as C₅-C₈-hydrocarbon, but it can also take place by adding to the reaction mixture extra liquid reducing the viscosity.

A further way to distribute the release of the hydrogen chloride created in the reaction is to use an overpressure during the reaction. The overpressure has both a thermo¬ dynamic and a physical influence on the release of the hydrogen chloride. The thermodynamic influence is based on the prevention of the evaporation of the hydrogen chlori¬ de, which reflects back to the distribution reaction op- tionally creating hydrogen chloride and slowing it down. The physical influence is based on the fact that a higher pressure raises the gasifying temperature of the hydrogen chloride and prevents the gasification at a certain tempe¬ rature. According to one embodiment overpressure is used during the reaction which is in the order of 1 to 100 bars, preferably about 1 to 50 bars and most preferably 1 to 5 bars.

The release of the hydrogen chloride created in the reac¬ tion can also be distributed by leading to the reaction mixture or above it nitrogen gas. The use of an inert gas flow influences principally physically so that the hydro¬ gen chloride content reduced by the environment accelera¬ tes the removal of the hydrogen chloride created in the reaction from the pores, whereby the saturation point of the hydrogen chloride corresponding to the gasification point will never be reached. It is, however, worth while using the nitrogen or other inert gas carefully for ther- modynamically its influence is the opposite i.e. by remo¬ ving rapidly hydrogen chloride from the pores of the car¬ rier particles the possible distribution reaction creating hydrogen chloride will be accelerated and will produce abundantly of heat causing discharging of hydrogen chlori¬ de.

It is also possible to add to the reaction mixture a li- quid dissolving the non-gasifying product of the reaction between the complex particles and the TiCl₄. As was pre¬ sented above in the reaction formula (1), as a result of the reaction between a MgCl₂-C₂H₅OH-complex and TiCl₄ there is created in addition to the activated magnesium chloride and hydrogen chloride ethoxytitanium trichloride, the presence of which in the same solution with hydrogen chlo¬ ride decreases the hydrogen chloride absorption capacity of the solution and thus also the gasification temperature of the hydrogen chloride. By adding a substance dissolving ethoxytitanium trichloride the releasing temperature of the hydrogen chloride can be raised and its release dist¬ ributed according to the aim of the invention. The pro- ducts of the reaction between the complex and TiCl₄, such as ethoxytitanium chloride and hydrogen chloride, can also be removed by changing during the reaction the excess amount of TiCl₄. As appears from the preliminary tests, the composition of the MgCl₂-C₂H₅0H-complex has a decisive effect on how rapidly the hydrogen chloride created in the reaction between the complex and TiCl₄ is released. The preliminary results and the reaction formula (1) show that the discharge velocity increases when the C₂H₅OH-content increases. Simultaneously, however, MgCl₂'s ability to become activated with TiCl₄ also increases, so an optimum must be found for the ethanol content of the complex, which theoretically can vary between 1 and 6, which opti¬ mum is dependent on how efficiently the release of the hydrogen chloride created in the reaction can be distribu¬ ted by other means. According to one embodiment of the invention the molar ratio between the MgCl₂ and C₂H₅0H of the MgCl₂-C₂H₅OH-complex used as the material of the car¬ rier particles is between 2.5 to 3.5 and preferably bet- ween about 2.7 to 3.3. As too large ethanol contents al¬ most always cause hydrogen chloride discharges breaking carrier particles, it can generally be said that the C₂H₅OH-content must be below about 60% of the weight of the procatalyst carrier material.

As the too large carrier particles are broken more easily and due to their deep pore structures accentuate the fac¬ tors leading to a rapid discharge of the hydrogen chlori¬ de, it is advantageous to use carrier particles, the size of which is below about 150μm.

The success of the invention is also influenced by the water content, mechanical strength, particle morphology and particle size distribution of the carrier particles. Therefore it has been noticed that such carrier particles consisting of MgCl₂-C₂H₅0H-complex are particularly suit¬ able for the method according to the invention that have been prepared by a spray-crystallization method. According to one preferred embodiment the complex melt, the composi¬ tion of which is a MgCl₂- 3.5C₂H₅OH and the temperature about 120-130°C, is sprayed through a distributing nozzle into a chamber in which the temperature is about 30 to

50°C, after which the stiffly flowing drops are conducted through another zone, the temperature of which is about 10°C lower and where the particles finally solidify. Par¬ ticles prepared by this kind of a method are mechanically much stronger than e.g. the particles prepared by spray- evaporation and are thus particularly suitable for the method according to the invention, in which the discharge tendency of the hydrogen chloride sets high mechanical requirements for the particles. Simultaneously, particles are created giving a pore volume in the procatalyst, which is suitable for the method according to the invention.

The release of the hydrogen chloride can also be accelera¬ ted by the presence of an internal donor in the reaction mixture of carrier particles and TiCl₄. This is due to the fact that the reactions of the donor taking place in con¬ nection with the adding usually are exothermic. If the donor is added at a stage or in a temperature range where a discharge of the hydrogen chloride is possible, it may, in fact, lead to an uncontrolled release of the hydrogen chloride. This can happen e.g. when the reaction mixture after the adding of the TiCl₄ is heated to a temperature of about +20°C and and the donor is added to the reaction mixture. According to one embodiment it is advantageous to distribute the release of the hydrogen chloride created in the reaction in connection with the adding of the internal donor by not adding the donor into the reaction mixture until about 0.5 to 6 hours, preferably about 2 to 5 hours, after the adding of the TiCl₄. It is also important to add internal donor into the reaction mixture control1ably during about 0.5 to 1.5 hours. To be absolutely sure the heating temperature must not be increased until about 1 to 3 hours after the adding of the donor, whereby the adding of the donor does not cause strong hydrogen chloride dis¬ charges.

The above-mentioned measures for the distribution of the release of the hydrogen chloride created in the reaction can, according to the invention, be used alone or then combined in any possible way. According to one embodiment of the invention the measures are used in the manner shown in figure 5.

In figure 5 a temperature profile according to one embodi¬ ment of the invention has been presented when procatalyst compositions are prepared with a reaction between MgCl₂- C₂H₅OH-complex and TiCl₄. In the figure:

A is a slow increase of temperature after the first adding of TiCl₄,

B is the delay of the reaction after the adding of the donor, C is a more rapid increase of temperature for the comple¬ tion of the first reaction between the carrier and the TiCl₄,

D is the completion of the first reaction between the carrier and the TiCl₄, E is the second treatment of the carrier with TiCl₄, F is the washing and G is the drying.

According to the invention it is advantageous to carry out one or several of the above-mentioned measures for the distribution of the release of the hydrogen chloride crea¬ ted in the reaction in the points I, II, and III marked in the figure. Hereby, it is natural to carry out the measu¬ res connected with the adding of the TiCl₄ and the heating taking place in connection with it in point I and the corresponding measures connected with the adding of the donor in points II and III. In other respects the measures can be optimized, taking into consideration the whole pro¬ cess, for each reactor and the desired product.

As even the smallest impurity can disturb the reaction and cause a surprising hydrogen chloride discharge, the subs¬ tances must be as pure as possible and above all dry. Special attention must be paid to the medium and to the inert gas, so that no harmful sidereactions can occur. To keep the reaction environment as clean as possible it is advantageous to use a multi-function reactor, by which the activation, washing and drying stages can be carried out without removing the procatalyst or its intermediary pro¬ duct from the reactor. This kind of a reactor and method have been presented in the Finnish Patent No. 83329, which is hereby attached to the application as a reference.

In the following a few comparative examples and embodi¬ ments are presented for illustrating the invention.

Starting materials

The TiCl₄ used in the synthesis was liquid and non-aque¬ ous.

The MgCl₂-C₂H₅OH-complex carrier was prepared from complex melt, to the composition of which MgCl₂*3.5C₂H₅OH was ad¬ justed. Melt was sprayed at the temperature of about 120 to 130°C through a nozzle structure distributing it, the temperature of the reception zone of which was about 30 to 50°C. From the reception zone the melt drops were conduct¬ ed through a slightly colder zone for the final hardening of the particles. Moreover, the the particles were sieved so as to have the particle size below about 150μm. The composition of the final particles was about MgCl₂ ^• 3C₂H₅OH.

The morphology of the carrier particles was splendid. The particle size distribution was between 1.50 to 3.4 (span) and the particles did not contain finely-divided material. The microscope pictures of the carrier particles have been presented in connection with the microscope pictures of the procatalysts and the polymers in the handling of the results.

Comparative example 1

A multi-function reactor furnished with a sieve bottom, the volume of which was 1.5m³, was cooled to the tempera¬ ture of -20°C. Then the inert hydrogen carbon solvent (Neste product LIAV), TiCl₄ and the carrier, MgCl₂-2.7C₂H₅0H, were added into the reactor, in the order mentioned. The amount of the carrier was 44kg and the ratio TiCl₄ (mole)/carrier C₂H₅0H (mole) was 10. The ratio LIAV (kg)/carrier (kg) was 4.5 and the ratio TiCl₄ (mo¬ le)/Mg (mole) was 30. The agitation velocity was 15rpm.

After the adding of the reagent the temperature was slowly with the velocity of 0.22°C/min. raised to the temperature +20°C. The agitation velocity still was 15rpm.

Five hours after the adding of the reagent di-isobutylph- thalate(DIBP)-donor was added at the temperature +20°C so that the ratio donor (mole)/Mg (mole) was 0.15. Then the temperature was raised with the mean velocity of about 0.1°C/min. to the value 110"C. The agitation velocity still was 15rpm. The temperature and the agitation were maintained for about 1 hour, after which the activation residues with the excess amount of TiCl₄ were removed by rinsing through the sieve bottom of the reactor.

The second treatment with TiCl₄ was carried out by adding the reagent into a purified solid intermediary product. The temperature still was 110°C, the agitation velocity 15rpm and the reaction time 2 hours. The ratio TiCl₄ (mo- le)/Mg (mole) being also now 30.

Finally, the product was washed four times with a hydrogen carbon solvent (LIAV) so that the ratio LIAV (kg)/carrier (kg) was 9. During the washing the agitation velocity was about the same as above. After this the product was dried with a N₂-gas flow at the temperature 70°C without agita¬ tion.

In connection with this comparative example it was noticed that when carrier was added above the TiCl₄ the mixture began to boil already at the temperature of -20°C. The particle size distribution of the catalyst product was such that 64.7% by weight of it had the diameter below 20μm. D₅₀ was 17μm and the width of the particle size distribution i.e. the span figure was 3.1. The portion of the finely-divided material in the polymer produced with the catalyst was 70% by weight (d<lmm) and the bulk densi¬ ty was 0.43g/ml.

Comparative examples 2 to 5

A multi-function reactor furnished with a sieve bottom, the volume of which was 1.5m³, was cooled to the tempera- ture of -20°C. Then the inert hydrogen carbon solvent

(Neste product LIAV), 24kg of the carrier, MgCl₂- 3.0C₂H₅OH and TiCl₄ were added into the reactor, in the order men¬ tioned. The ratio TiCl₄ (mole)/carrier C₂H₅OH (mole) was 10, the ratio LIAV (kg)/carrier (kg) was between 4.5 to 9 and the ratio TiCl₄ (mole)/Mg (mole) was 30. The agitation velocity was 15rpm.

After the adding of the reagent the temperature was slowly with the velocity of 0.22°C/min. raised to the temperature +20°C. The agitation velocity still was 15rpm. 3 to 5 hours after the adding of the reagent di-isobutylphthala- te(DIBP)-donor was added at the temperature +20°C so that the ratio donor (mole)/Mg (mole) was 0.15.

Then the temperature was raised with the mean velocity of about l°C/min. to the value 110°C. The agitation velocity still was 15rpm. The temperature and the agitation were maintained in about 1 hour, after which the activation residues with the TiCl₄ excess amounts were removed by rinsing through the sieve bottom of the reactor.

The other treatment with TiCl₄ was carried out by adding the reagent into a purified solid intermediary product. The temperature still was 110°C, the agitation velocity 15rpm and the reaction time about two hours. The ratio TiCl₄ (mole)/Mg (mole) also now 30.

In connection with these comparative examples it was noti¬ ced that during the heating gas was produced. The particle size distribution of the catalyst product was such that 11 to 28% by weight of it had the diameter below 20μm. D₅₀ was between 30 and 62μm and the width of the particle size distribution i.e. the span figure was between 1.46 and 2.96. The rest of the properties are presented in connec- tion with the results.

Comparative examples 1 to 4

A multi-function reactor furnished with a sieve bottom, the volume of which was 1.5m³, was cooled to the tempera¬ ture of -20°C. Then the inert hydrogen carbon solvent (Neste product LIAV), carrier, MgCl₂'3.0C₂H₅OH, and TiCl₄ were added into the reactor, in the order mentioned. The amount of the carrier varied between 24 to 29kg and the ratio TiCl₄ (mole)/carrier C₂H₅OH (mole) was 10. The ratio LIAV (kg)/carrier (kg) was 9.0 and the ratio TiCl₄ (mo- le)/Mg (mole) was 30.

Deviating from the comparative example the starting velo¬ city of the agitation was 30rpm, the overpressure of the reactor was 22.5 bars and through the reactor was conduct- ed dry N₂-gas with the flowing velocity of 5kg/hour until the maximum temperature was reached.

After the adding of the reagents the temperature was slow¬ ly raised with the velocity of 0.22°C/min. to the tempera- ture +20°C. At this stage the agitation had been reduced from the original value of 30rpm to the usual value of 15rpm in order to decrease the mechanical strain caused by the agitation.

Three hours after the adding of the reagents di-isobu- tylphthalate(DIBP)-donor was added at the temperature of +20°C so that the ratio donor (mole)/Mg (mole) was 0.15.

Then the temperature was raised with the mean velocity of about l°C/min. to the value 110°C, whereby the overpressu¬ re and the N₂-flow were removed. The agitation velocity still was 15rpm. The temperature and the agitation were maintained in about 1 hour, after which the activation residues with the TiCl₄ excess amounts were removed by rinsing through the sieve bottom of the reactor.

The other treatment with TiCl₄ was carried out by adding the reagent into a purified solid intermediary product. The temperature still was 110°C, the agitation velocity 15rpm and the reaction time about two hours. The ratio TiCl₄ (mole)/Mg (mole) was also now 30. Finally, the product was washed three times with a hydro¬ gen carbon solvent (LIAV) so that the ratio LIAV (kg)/car¬ rier (kg) was 9. During the washing the agitation velocity was about the same as above. After this the product was dried with a N₂-gas flow at the temperature 70°C without agitation.

In connection with these comparative examples no boiling was noticed during the heating. The particle size distri- bution of the catalysts was such that 21 to 22% by weight of it had the diameter below 20μm. D₅₀ was 44 to 62μm and the width of the particle size distribution i.e. the span figure was 2.25 to 2.59 (except in the fourth pilot-test). The rest of the properties are presented in connection with the handling of the results.

Test polymerization

All the procatalysts of the comparative examples and the embodiments prepared in laboratory or pilot-scale were tested under standard polymerization conditions. A 2-liter bench reactor was used. 20 to 30mg of procatalyst was used in each test polymerization run. To this amount 620μml of triethyl aluminium co-catalyst and 200μml of 25% heptane solution of an internal cyklohexylmethyl methoxysilane donor were added. As the intermediary medium 900ml of hep¬ tane was used. The polymerizations were carried out at the temperature of +70°C and in the propene monomer pressure of 10 bars. The partial pressure of hydrogen during the polymerization was 0.2 bars, llmmol H₂ was added to adjust the molecular weight. The polymerization was continued for three hours. Thereafter the activity of the procatalyst was measured on the basis of the polymerization yield. The solvent portion of the polymer was measured by dissolving a certain polymer amount into the solvent and measuring the evaporation residue of the pure solution. The bulk density and the particle size distribution of all the polymer samples were determined. In connection with the particle size distribution measurements the total amount of the finely-divided material was estimated. Here- by all polymer particles, the diameter of which was smal¬ ler than 1mm was defined as finely-divided material. The isotacticity was measured by means of heptane elution and the isotacticity index was measured by using the results of the evaporation residue measurements. The melt index was measured at the temperature of 230°C by using a 2.16kg weight.

Results

The above-mentioned test runs have been carried out in pilot-scale. The original tests were carried out in labo¬ ratory. Hereby, it was observed that in pilot-scale there occurred problems that did not occur in laboratory scale. In laboratory tests the release of the hydrogen chloride gas was uniform in all the tests, whereas in the pilot- scale in the comparative tests a heavy discharge of hydro¬ gen chloride gas took place as a difference from the embo¬ diment tests, in which the release of the hydrogen chlori¬ de could be distributed by the measures according to the invention.

The release of hydrogen chloride was followed in the embo¬ diments by registering how much the exhaust of the reactor could open the escape valve connected with the reactor and regulating the pressure of the reactor. In figure 6 the opening of this kind of a valve is presented as a function of the temperature and the time. Higher up in the figure there is presented the gradient of the temperature of the reactor i.e. the profile and lower down the opening of the valve i.e. the escaping velocity of the hydrogen chloride gas as a function of the time. The escaping velocity of the hydrogen chloride is relatively even and does not show any peaks like those shown in figure 3 during the heating.

The titanium content, the donor content, the particle size distribution, the catalyst yield, the catalyst activity, the polymer-isotacticity, the polymer melt index and the polymer-bulk density of the procatalysts of the embodi¬ ments prepared in laboratory and in pilot-scale were nor¬ mal for catalysts of corresponding type.

The titanium content varied between 2.4 and 3.6% by weight in laboratory catalysts and between 2.4 and 4.5% by weight in pilot-procatalysts. The donor content varied between 15.9 and 19.2% by weight in laboratory-procatalysts and between 9.7 and 15.4% by weight in pilot-procatalysts (in embodiment 1 the pilot-scale donor feed was unsuccessful).

The particle size distribution of the procatalysts appear from table 1. Only in the last i.e. fourth pilot-scale preparation a small procatalyst decomposition could be noticed, which can be seen as a broad distribution, the span-figure of which was 3.41.

The procatalyst yield was satisfactory varying between 74 and 99% in pilot-run and between 82 and 92% in laboratory. This shows that finely-divided material has not e.g. been rinsed off to any in any greater extent in the pilot-run compared to the laboratory-run.

The activity of the pilot-procatalysts was at its best 15.8kgPP/g cat, which is a good value and of the same order as in the procatalysts prepared in laboratories.

The polymer-isotacticity varied in laboratory-procatalysts between 98.9 and 99.3% (the index between 98.3 to 98.9) and in the pilot-procatalysts between 96.8 and 97.5% (the index between 93.3 and 98.1) being at a satisfactory le- vel .

The polymer melt indices varied in laboratory procatalysts between 4.3 and 9.6 and in pilot-procatalysts between 5.0 and 7.4 (the MI of the fourth run was exceptionally 19.4) corresponding the melt index of a normal polypropene.

The polymer bulk densities varied in laboratory procata¬ lysts between 0.41 and 0.47g/ml and in pilot-procatalysts between 0.40 and 0.44g/ml.

Special attention was paid to the particle size distribu¬ tion of the polymer material given by the procatalysts prepared by the method according to the invention. The results have been presented in table 2. A graphical pre¬ sentation can be found in figure 7.

In table 2 (see page 20) only the particle size distribu¬ tion of the polymers obtained with the laboratory- and pilot-procatalysts according to the embodiments 1 to 4, whereas in figure 8 polymers prepared with the laboratory- and pilot-procatalysts according to the comparative exam¬ ples and the polymers prepared with the laboratory- and pilot-procatalysts according to the embodiments have been compared. The figure clearly shows that the comparative examples and the embodiments give the same kind of results when a procatalyst synthesis carried out in laboratory scale is concerned.

When going over to pilot-scale a remarkable improvement can be achieved with the measures according to the inven¬ tion compared with the procatalysts according to the com¬ parative examples. In pilot-scale the portion of the fine¬ ly-divided material of the polymer can be reduced to at least about the fourth and at the utmost to even less than a tenth of the results of the comparative tests. In the pilot comparative example 1, where the adding order of the reagents was different (see the directions above) as much as 70% by weight was obtained as the portion of the fine¬ ly-divided substance, which value is much higher than those of the pilot embodiments.

In figures 9 to 12 electron microscope pictures of the carrier of the embodiment 1 (fig. 8) , the laboratory-pro- catalyst (fig. 9) , the pilot-procatalyst (fig. 10) , the polymer obtained with a laboratory-procatalyst (fig. 11) and the polymer obtained with a pilot procatalyst (fig. 12) are shown. These figures relating to the embodiment 1 do visually represent well the morphology of the product, proceeding from a spray-crystallized carrier to a f nished polypropene product. When figures 8, 10 and 12 are compa¬ red, it can be seen that the portion of the finely-divided material in the polymer or the procatalyst is not much higher than in the carrier. When figures 10 and 12 are compared with figures 9 and 11 it can be seen that both the procatalyst and the polymer are a little more agglome¬ rated in the pilot-scale synthesis than in the laboratory- scale synthesis. This tendency was not, however, as strong as in the embodiments 2 to 4.

SUBSTITUTE SH Table 1 The particle size distribution of the procata¬ lysts prepared in pilot- and laboratory-scale

Table 2 Polymer particle size distributions given by the catalysts in pilot- and laboratory-scale

Embodiment Procatalyst

1 Pilot

Lab. 2 Pilot

Lab.

3 Pilot Lab.

4 Pilot Lab. In the enclosed figures 1 to 12 the following is present¬ ed:

Fig. 1: Formation of the finely-divided substance in PP- polymer when comparing pilot-catalyst to laboratory cata¬ lyst,

Fig. 2: Correlation between the amount of the finely-di¬ vided fraction of the polymer and the maximum releasing velocity (ml/min) of the HCl created in the catalyst syn¬ thesis,

Fig. 3: Release of HCl when heating a system containing a MgCl₂-C₂H₅0H-complex, HCl and TiCl₄,

Fig. 4: The rate of HCl gas release as a function of the temperature, when the system contains a MgCl₂-C₂H₅0H-comp- lex, TiCl₄, and HCl,

Fig. 5: A preferred use of the measures according to the invention in a procatalyst process; (A) slow increase of temperature after a TiCl₄-addition, (B) delay of the reac- tion during the adding of the donor and after it, (C) increase of temperature for the completion of the reacti¬ on, (D) first titanization, (E) second titanization, (F) washing, (G) drying,

Fig. 6: Releasing velocity of the hydrogen chloride (lower curve) as a function of the temperature (upper curve) and time,

Fig. 7: Total amount of finely-divided substance (d<l mm) in polymers prepared with laboratory- and pilot-catalysts of the comparative examples (old) and the embodiments _, Fig. 8: carrier of embodiment 1, length unit lOOμm,

Fig. 9: laboratory procatalyst of embodiment 1 (25μm),

Fig. 10: pilot procatalyst of embodiment 1 (25μm),

Fig. 11: polymer obtained with laboratory procatalyst of embodiment 1,

Fig. 12: polymer obtained with pilot procatalyst of embo¬ diment 1.

Claims

1. Method for the preparation of a particulate pro¬ catalyst composition intended for the polymerization of olefins by reacting the carrier particles consisting of a MgCl₂-C₂H₅OH-complex with TiCl₄, whereby hydrogen chloride is formed, characterized in that the carrier particles are maintained undamaged during the reaction by distributing the release in the reaction or from the reaction of the hydrogen chloride created in the reaction to a molar velo¬ city not exceeding the fivefold value of the molar mean velocity of the release.

2. Method according to claim 1, characterized in that the release of the hydrogen chloride created in the reaction is distributed to a volume velocity not exceeding the threefold, preferably double, value of the mean volume velocity of the release.

3. Method according to claim 1 or 2, characterized in that the release of the hydrogen chloride from the carrier complex or the reaction mixture is restricted so that not more than about 5%, preferably less than about

2%, is discharged in a minute from the stoichiometrically released hydrogen chloride.

4. Method according to claim 1, 2, or 3, characteri- zed in that the release of the hydrogen chloride is dis¬ tributed by controllably adding TiCl₄ into the liquid sus¬ pension of the carrier particles during 0.5 to 3.0 hours, preferably during 1.0 hour.

5. Method according to claim 4, characterized in that the TiCl₄ is added at the temperature of -30...-10°C, preferably at the temperature of -25...-15°C.

6. Method according to claim 5, characterized in that the temperature is raised during the adding of the TiCl₄ or after it at the velocity of about 5-20°C/h, pre¬ ferably at the velocity of about 10°C/h, between about -20...+40°C.

7. Method according to any of the preceding claims, characterized in that the release of the hydrogen chloride created in the reaction is distributed by using strong agitation during the reaction.

8. Method according to any of the preceding claims, characterized in that the release of the created hydrogen chloride is distributed by using in the reaction mixture a liquid reducing its viscosity.

9. Method according to any of the preceding claims, characterized in that the release of the hydrogen chloride created in the reaction is distributed by using during the reaction an overpressure, which preferably is of the order of 1 to 100 bars, more preferably about 1 to 50 bars, and most preferably abnout 1 to 5 bars.

10. Method according to any of the preceding claims, characterized in that the release of the hydrogen chloride created in the reaction is distributed by conducting to the reaction mixture or above it nitrogen gas.

11. Method according to any of the preceding claims, characterized in that the release of the hydrogen chloride created in the reaction is distributed by using in the reaction mixture a liquid dissolving the product of a reaction between TiCl₄ and a MgCl₂-C₂H₅0H-complex.

12. Method according to claim 11, characterized in that the liquid dissolving the product of a reaction bet¬ ween TiCl₄ and a MgCl₂-C₂H₅OH-complex is an excess amount of TiCl₄ and that TiCl₄ is exchanged during the reaction to new TiCl₄ for the removal of HCl as dissolved in it.

13. Method according to any of the preceding claims, characterized in that a MgCl₂-C₂H₅OH-complex is used, the molar ratio of which between MgCl₂ and C₂H₅OH being 2.5 to 3.5, preferably 2.7 to 3.3.

14. Method according to any of the preceding claims 3 to 13, characterized in that a MgCl₂-C₂H₅OH-complex is used, the C₂H₅0H-content of which is below 60% by weight.

15. Method according to any of the preceding claims, characterized in that carrier particles are used, the size of which is below about 150μm.

16. Method according to any of the preceding claims, characterized in that the carrier particles consisting of MgCl₂-C₂H₅0H-complex have been prepared by spray-crystal- lization.

17. Method according to any of the preceding claims, characterized in that the release of the hydrogen chloride created in the reaction is distributed in connection with the adding of an internal donor by not adding the donor into the reaction mixture until about 0.5 to 6 hours, preferably about 2 to 5 hours, after the adding of the TiCl₄.

18. Method according to claim 15, characterized in that the internal donor is controllably added into the reaction mixture during about 0.5 to 1.5 hours.

19. Method according to claim 15 or 16, characterized in that heating velocity is not raised until about 1 to 3 hours after the adding of the donor and the raising of the temperature is stopped at about 110 to 120°C.

20. Method according to any of the preceding claims, characterized in that essentially the following temperatu¬ re profile is used:

in which A is a slow increase of the temperature after the first TiCl₄-addition,

B is a delay of the reaction after the adding of the donor,

C is a more rapid increase of temperature for the completion of the first reaction between the carrier and the TiCl₄,

D is the completion of the first reaction between the carrier and the TiCl₄,

E is the second treatment of the carrier with TiCl₄, F is the washing and

G is the drying, whereby the measures mentioned in the foregoing claims for the distribution of the release of the hydrogen chloride created in the reaction are carried out in points I, II, and III.

21. Method according to claim 18, characterized in that the measure according to claims 3, 4, and/or 5 is carried out in point I, and the measure according to claims 15, 16, and/or 17 is carried out in points II and III.

22. Method according to any of the preceding claims, characterized in that a multi-function reactor furnished with a sieve bottom is used, by which the activation, washing and drying stages can be carried out without re¬ moving the procatalyst or the intermediary product leading to it from the reactor.

AMENDED CLAIMS

[received by the International Bureau on 27 April 1993 (27.04.93); original claims 1,4 and 11 replaced by amended claim 1; claims 2,3 and 12 - 22 renumbered as claims 3,4 and 9-19; claims 4-7 amended and renumbered 5-8 other claims unchanged (5 pages)]

1. Method for the preparation of a particulate pro¬ catalyst composition intended for the polymerization of olefins by reacting carrier particles consisting of a 5 MgCl₂-C₂H₅OH-complex with TiCl₄, whereby hydrogen chloride is formed, characterized in that one or several steps are used, in which:

(a) the TiCl₄ is controllably added to a liquid suspension of the carrier particles during 0.5 - 3.0 hours; 0 (b) the TiCl₄ is added at the temperature of -30...-10°C;

(c) the temperature is raised during and after the adding of the TiCl₄ at a rate of about 5...20°C/h to a range of about -20...+40°C;

(d) vigorous stirring is used during the reaction; 5 (e) a viscosity lowering liquid is used in the reaction composition;

(f) over-pressure being of the order of 1...100 bars is used during the reaction; g) nitrogen gas is conducted into the reaction composition or above it, and

(h) liquid dissolving the product formed in the reaction between the carrier particles and the TiCl₄ is used in the reaction composition, so that the release of hydrogen chloride per minute is restricted to not more than 5% of the total hydrogen chlo¬ ride amount formed in the reaction in order to maintain the carrier particles undamaged.

2. Method according to claim l, characterized in that the release of the hydrogen chloride created in the reaction or from the reaction is levelled to a molar velo¬ city not exceeding the fivefold value of the mean volume velocity of the release.

3. Method according to claim l or 2, characterized in that the release of the hydrogen chloride created in the reaction is levelled to a volume velocity not excee¬ ding the threefold, preferably double, value of the mean 4. Method according to claim 1, 2, or 3, characteri¬ zed in that the release of the hydrogen chloride from the carrier complex or the reaction mixture is restricted so that not more than about 2% is discharged per minute from the total hydrogen chloride amount formed in the reaction.

5. Method according to claim 1, 2, 3, or 4, charac¬ terized in that the release of the hydrogen chloride is distributed by controllably adding TiCl₄ into the liquid suspension of the carrier particles during about 1 hour.

6. Method according to any of the preceding claims, characterized in that the TiCl₄ is added at the temperatu¬ re of 25...-15°C.

7. Method according to any of the preceding claims, characterized in that the temperature is raised during the adding of the TiCl₄ or after it at the velocity of about 5-15°C/h between about -20...+40°C.

8. Method according to any of the preceding claims, characterized in that the release of the hydrogen chloride formed in the reaction is levelled by using during the reaction an overpressure, which preferably is of the order of 1 to 50 bars, preferably about 1 to 5 bars.

9. Method according to claim 8, characterized in that the liquid dissolving the product of a reaction bet¬ ween TiCl₄ and a MgCl₂-C₂H₅0H-complex is an excess amount of TiCl₄ and that for the removal of HCl dissolved in it the TiCl₄ is exchanged during the reaction to new TiCl₄.

10. Method according to any of the preceding claims, characterized in that a MgCl₂-C₂H₅0H-complex is used, the molar ratio of which between MgCl₂ and C₂H₅0H being 2.5 to 3.5, preferably 2.7 to 3.3.

11. Method according to any of the preceding claims, characterized in that a MgCl₂-C₂H₅0H-complex is used, the C₂H₅OH-content of which is below 60% by weight.

12. Method according to any of the preceding claims, characterized in that carrier particles are used, the size of which is below about 150μm.

13. Method according to any of the preceding claims, characterized in that the carrier particles consisting of MgCl₂-C₂H₅0H-complex have been prepared by spray-crystal- lization.

14. Method according to any of the preceding claims, characterized in that the release of the hydrogen chloride created in the reaction is levelled in connection with the adding of an internal donor by not adding the donor into the reaction mixture until about 0.5 to 6 hours, prefera¬ bly about 2 to 5 hours, after the adding of the TiCl₄.

15. Method according to claim 12, characterized in that the internal donor is controllably added into the reaction mixture during about 0.5 to 1.5 hours.

16. Method according to claim 12 or 13, characterized in that heating velocity is not raised until about 1 to 3 hours after the adding of the donor and the raising of the temperature is stopped at about 110 to 120°C.

17. Method according to any of the preceding claims, characterized in that essentially the following temperatu¬ re profile is used:

B is a delay of the reaction after the adding of the donor,

D is the completion of the first reaction between the carrier and the TiCl₄,

E is the second treatment of the carrier with TiCl₄, F is the washing and

18. Method according to claim 17, characterized in that the measure according to claims la), lb), lc) , 5, 6, and/or 7 is carried out in point I, and the measure accor- ding to claims 14, 15, and/or 16 is carried out in points II and III.

19. Method according to any of the preceding claims, characterized in that a multi-function reactor furnished with a sieve bottom is used, by which the activation, washing and drying stages can be carried out without re¬ moving the procatalyst or the intermediary product leading to it from the reactor.