This invention relates to a new process for production of isobutene polymers and co-polymers, and to the devices thereof.

In particular, the invention relates to the production of high MW isobutene-isoprene copolymers, known as Butyl Rubber or "IIR".

IIR is known for about 50 years (U.S. 2356128 and U.S. 2356130). It finds application as a low gas diffusion and oxidation resistant rubber. IIR is commer cially produced in the presence of cationic catalysts, also known ad Friedel-Craft catalysts, at extremely low temperatures (e.g. 173 K ^δ ), in a reaction medium of methylchloride. Unlike to its monomers, IIR is not solu ble in methylchloride, so that a polymer slurry is for- med. Details on this technology are available in the lj__ terature (C. Whitby Synthetic Rubber, Chapt.24 J. Wiley & Sons and Kirk Othmer Encyclopedia of Chem. Technol., Butyl Rubber) .

Even if the present IIR processes are stil^. va lid, they show important drawbacks. These drawbacks can be summarized as follows: i) the energy consumption of the process is very high ii) the plant investment cost is also very high iϋ) the quality constancy control is difficult

In more detail : i) The high energy consumption is due to a series of cooling cycles for supplying the process with the necessary frigories at 173 K° . As an order of magni

tude the energy cost of cooling cycles can be as high as 75% of the total energy requirement. ii) The Investment cost of a IIR plant Is also very high. Beside the cost of the mentioned cooling cycles, "which can be as high as 35% of the total inve- st en , the polymerization reactors are very expensive. ^' In fact their design is very complex and they are built with expensive criogenic materials. The investment cost is exacerbated by the fact that every 2-30 reactors in operation a supplementary stand-by reactor is needed, since the Internal heat exchangers fowling imposes a rather frequent reactor shut-down for cleaning. iii) The IIR quality depends on its MW, and in turn MW depends on polymerization temperature. The a- bove mentioned reactor fowling is conducive to a tempera ture Increase inside the reactor. Consequently, the ope_ rating life of the reactor lasts for a few ten hours. This fact has an impact on IIR quality constancy.

In order to overcome the aforementioned draw¬ backs, several solutions have been proposed and paten ted. However these solutions were not solving the basic problems of IIR process, or were not pratically applica ble.

Among the relevant proposed solutions, the use of new catalyst systems, called by some researches "syncatalysts" has been put forward about 20 years.ago.

The syncatalysts allowed in principle to obtain high MW IIR even at a temperature as high as 223K ⁰ , vis a vis the 173K° of the conventional processes. However, the pratical use of syncatalysts has been ruled out by the worsening of the already critical problem of reactor

fowling .

More recently a new IIR process (U.S. 4714747) has been patented in which syncatalysts are used without meeting heat exchange problems due to reactor fowling. According to U.S. 4714747, monomers and reac- tion medium are fed into a double screw, self cleaning extruder. Inside this extruder the reactant fluids fol¬ lowed a peculiar fluidodynamics known as "plug flow" . The boiling of reaction medium inside the extruder was avoiding the need of a heat exchange through the reactor walls. However, as point out in U.S. 4714747, the double screw extruder was compacting the polymer in a "viscous and sticky mass" which hampered the vapor separation and, consequently, the reactor,temperature control.

In conclusion, also this solution was difficult to put in pratice.

All the aforementioned drawbacks are overcome by the process and by the devices of this invention, in which are contemporarily present all the following critical characteristics: 1) The polymerization is performed at tempera tures between 208 and 288K°, in the presence of synca¬ talysts, whose two (or more) components are fed directly and separately to the polymerization reactor.

2) The polymerization heat is removed by al- lowing the reaction medium to boil inside the reactor, and maintaining the reactor content as a low viscosity polymer slurry.

3) The polymerization reactor is designed in order to keep the reactor content as a low viscosity slurry. A coalescence of polymer particles in bigger

and bigger crumbs is avoided, which is conducive to the "viscous and sticky mass" shown in U.S. 4714747. In the reactor of the present invention, the vapors produced by the exothermic polymerization reaction can easily escape from the polymer slurry, owing to its low.visc£ sity, and collect In the reactor dome.

4) The polymerization reactor is designed in order to fulfill other important duties : one is that of homogenizing the liquid phase in every point of the reactor condensed phase, in order to maintain a con¬ stant composition of reactants. The second is that of conveying the polymer phase across the reactor, toward the polymer slurry outlet. The third is that of concen¬ trating the polymer slurry discharged by the reactor". The last two duties, which are somewhat recalling cer¬ tain dairy processes, have been dubbed with the word "skimming".

5) The skimmed slurry flowing from the reactor outlet in concentrated further on in a vertical extruder connected with the reactor. This vertical extruder, re¬ ferred to as "discharge screw", conveys and presses the solid polymer contained in the slurry upward and "squeez¬ es" the liquid downward (i.e. backward) toward the reactor. In this way, valuable catalyst components, so- luble in the reaction medium, are re-fluxed to the reac tor for a further utilization.

The details of the process of the present in¬ vention and of the devices for its pratical application are dealt with herebelow. Polymerization reactor

The reactor is a vertical cylindrical vessel,

equipped with a special stirrer. In fig. 2 a vertical section is shown. The reactor wall is indicated by 1; 2 and 3 are respectively the reactor ceiling and bot¬ tom, 4 is the stirrer shaft, 5 is a standing shaft fa- stened to the reactor bottom, 6 is one of the vertical paddles, moved by shaft 4 to which they are connected by the cross bar 7. The rotating bars 8 intermesh with standing crossbars 9. 10 is an extruder barrel fastened to the reactor, to which it is connected via slit 12. This extruder, equipped with its screw 11, is hereinafter referred to as "discharge screw".

An horizontal reactor section is also shown in fig. 2. The section is along a standing bar 9. In this figure, three paddles 6, >one bar 9 and the slit 12 are shown. The three paddles 6 have a peculiar shape and orientation. When they rotate according to the direction indicated by an arrow, the crearance between a point on the reactor wall and the paddle decreases till to a very small clearance. The reactor is equipped with on inlet for the monomers/polymerization medium mixture, and of two inlets for feeding the catalyst components. The reactor is also equipped with a vapor outlet, placed in the reactor ceiling. The following description of the reactor opera tion in a continuous, steady state operation will better explain the function of every reactor fitting.

The reactor operates at constant level of con¬ densed phase (which includes liquid monomers and reac- tion medium and polymer particles).

IIR particles are continuously formed in the

reactor, and are immediately conveyed by the stirrer across the reactor, toward its peripheric part, where the slurry discharge 12 is placed, with a continuous skimming action. The rather small clearance among moving and standing surfaces of the stirrer/reactor system limits the surfaces fowling, but does not hamper the liquid phase to flow across the clearances. In the way, the liquid phase is homogenized throughout all the reactor. The polymer skimming is performed in the fol¬ lowing way: the polymer slurry present in the peri- pherical part of the reactor is continuously intaken into the rather large clearance between the paddle leading edge and the reactor wall. The liquid phase. can easily flow through the clearance between the pad¬ dle rear edge and the reactor wall. On the contrary the polymer particles are "catched" owing to their si¬ ze. Therefore the slurry concentration in the space between the paddle and the reactor wall increases dur ing the paddles motion. The rear paddle edge squeezes the polymer slurry against the wall for a phenomenon of vlsco-dynamic pressurization (Z. Tadmor, C. Gogos Prin¬ ciples of Polymer processing Whiley-Interscience Chap. 10) , which increases the solid content of the slurry, whereas the liquid is squeezed out. Whenever a paddle 6 passes In front of slit 12, the slurry is pumped through it, toward the discharge screw 11, which car ries it up. Therefore, whenever a paddle 6 passes in front of slit 12, the slurry contained in the volume between paddle and wall is cleared out, and is ready to be re-filled by another amount of slurry, coming

from the axial part of the reactor. A step carved in the rear edge of paddles 6 decreases the gap between paddle and wall and increases the skimming efficiency. At a certain stirrer rotation speed, the polymer discharged by the reactor in unit time increa ses proportionally to the polymer concentration.

At a certain polymer concentration, the poly¬ mer discharged by the reactor in unit time increases proportionally to the stirrer rotation speed. Consequently, the above reactor has a built- in capacity of controlling the polymer concentration inside the reactor. Once a certain value of polymer output has been selected, by feeding the catalyst at a certain rate, the polymer concentration inside the reactor cannot overcome an upper threshold limit, which is imposed by the stirrer rotation speed. In this way, any excessive polymer content inside the reactor is avoided, with the related risks of reactor "blocking" or of excessive mechanical stress applied to the stirrer. The process

Details on the process of the present inven¬ tion are given herebelow, with reference to fig. 1. The polymerization reactor is indicated by number 1. The reactor outlet 12 joins the reactor to the discharge screw 11, which in turn is connected via throttle valve 13, to the double screw devolatilizer 14. Monomers and reaction medium make-up are fed via lines 15, A, B and C to heat exchanger 16, where they are cooled down to polymerization temperature, and there¬ from to reactor, via line 17. The two catalyst compo-

nents are separately fed to the reactor,, via lines 18A and B. As soon as the catalyst components come in con tact with the monomers the polymerization takes place. Polymer particles are produced around the active cen- ters. It has been observed with surprise that a quick withdrawal of polymer particles from the reactor (abo¬ ve indicated as skimming) does not influences neither the polymerization rate nor the polymer yield. The po¬ lymerization is performed at temperatures between 208 and 288K°, at pressures correspondent to the vapor pressure of the reacting system, in general between 0,1 and 4 bar. In this way the reactor content Is allowed to boil. The vapors are collected in the reactor dome and then, via line 19 to the knockout drum 20, and, via line 21, sent to compressor 22. The compressed va pors are cooled and condensed in heat exchanger 23. The condensate is flashed in flash drum 24, via throttle valve 25, which drops the pressure to the same level as that of the reactor 1. In flash drum 24 a part of the condensate fed to throttle valve 25 is vaporized and recycled to knock-out drum 20, via line 26, and therefrom to compressor 22, via line 21, together with the vapors coming from the reactor, via line 19. A part of the condensate fed to throttle val ve 25 Is collected in flash drum 24 as a liquid, at a temperature close to that of reactor 1. This liquid is sent to the reactor under level control, via line 27. In a steady state, the vapors produced by reactor 1 are almost completely recycled to it as liquids, at a temperature close to that of the reactor. Summing up, the polymerization heat Is removed from the reactor

as latent vaporization heat.

Any non condensable in heat exchanger 23 is purged via an outlet placed in the coldest part of heat exchanger 23 and sent, via line 28, to a vacuum pump (not shown in the figure).

The slurry produced by the reactor is con¬ centrated, as already said, in the discharge screw 11, fed through throttle valve 13 to mechanical devolati- lizer 14, where the residual liquids are removed, by combined thermal and mechanical action, as vapors via line 29. For the sake of exampli ication the polymer concentration at the reactor outlet can be high as 50% wt and at the discharge screw outlet as high as.70% wt. The produced IIR shows a, high MW. Number average MN's between 100000 and 500000 are obtained by the process as above described. The catalysts

Suitable catalysts for the process of the present invention are the syncatalysts, which have been extensively patented BE 663319, BE 663320, GB Pat. Nos. 1362295, 1407414, 1407415, 1407416, 1407417, 1407418, 1407419, 1407420, 1409337; The composition and the characteristics of said catalysts are broadly described in the mentioned patents and in several pu blication on this subject. (M. Baccaredda, M. Bruzzo- ne et al Chim. e Ind. (Milano) 55,109 (1973). A cha¬ racteristic of syncatalysts is the fact that the ac¬ tive centers are obtained by interaction of, at least, two components, in a stoichiometric ratio not lower/ higher than three decades, unlike conventional Friedel- Craft catalysts, in which the co-catalyst, or co-ini-

tlator, is present in a very small amount, sometime non measurable amount. Several syncatalysts are made of just two components-. The first one belongs to the class of Aluminium trialkyls or of Aluminium dialky_l monohalides, as for ex. AlEt3, AlEt2Cl, AlEt2Br, AliBu2Cl etc.

Typical examples of second components of syncatalysts are : halogens, interhalogenic compounds and organic halides which contain at least a halogen atom susceptible to be exchanged with the first comp£ nent of the catalyst system in the process conditions. C12, Br2, Iodine chloride, tert-butylchloride, tert- butylbromide, chloranil are non-limitative examples of second component of syncatalysts. The ratio between the two catalyst components is in general indicated by the atomic ratio Al/X, where X is the halogen present in the second component of the catalyst mixture. The Al/X ratio is in general higher than 1 and can be as high as 1000, the preferred ratio ranging from 10 to 300.

The consumption of second component, expres _^ sed in mol per mol of polymerized monomer, is in ge- neral ranging from 1x10 and 50x10 EXAMPLE

The flow sheet of the polymerization run of this example is show in fig. 1. The reactants amount, expressed in kg/h, and their temperature, expressed in C°, in steady state conditions are shown in the attached Table. In this Table also the cross- reference numbers with fig. 1 are indicated.

The reactor of the experiment was the one

described in fig. 2. The reactor was insulated with polystyrene foam. A certain quantity of pure Methyl¬ chloride (MeCl) was fed as a liquid, via line 15 C, to the exchanger 16, where it was cooled to -45°C, and therefrom to reactor 1, via line 17, up to a selected level. Isobutene and isopropene were then fed, via lines 15 A and B; in the quantity shown in the Table, and the two components of the catalyst, via lines 18 A and B. The two components were Aluminu diethyl- monochloride (DEAC)

TABLE

Uni ts : kg/h

and tert-Butylchloride (TBC). Upon feeding of the two catalyst components, a milky slurry of polymer was immediately formed. The reactor temperature an _d pres sure showed a trend to increase. The pressure in¬ crease fostered the intervention of the pressure control placed on the knock-out drum 20, and compressor 22 began to suck the reactor vapors, via line 19 an _d knock-out drum 20. From this moment onward the pres _¬ sure in the reactor remained constant at the value of

0.34 bar, corresponding to the equilibrium temperatu¬ re of -45C° in the reactor. The other ^s controls shown in fig. 1 were put into operation. The vapors compres_ sed by compressor 22 were condensed in heat exchanger 23 and the condensate was expanded in vessel 24, kept at the same pressure of the reactor. The condensate was partically vaporized, with a consequent tempera¬ ture drop, till to a value close to that of the reactor. The cold liquid collected in vessel 24 was sent to the reactor under level control, via line 27, whereas the vapors returned to the compressor via line 26 and knock-out drum 24. In this way the vapors produced by the reactor were quantitatively recyded to the same as liquids, at a temperature close to its temper ature. The polymer concentration in the reactor in- creased till to a steady value, at which the formed polymer was quantitatively removed by discharge screw 11. The polymer concentration at throttle valve 13 was about 60%. The polymer was fed, through valve 13 in desolventizer 14, where by combined thermal and mechanical action, the residual liquids ware va¬ porized and recovered via line 29. The IIR free from volatiles was collected as a rubbery extrudate on a conveyor belt 30. The polymer output was about 12,5 kg/h. The formation of polymer in the reactor and its removal by discharge screw 11 were decrea¬ sing the monomers concentration in the reaction me¬ dium and the polymer mass contained in the reactor. The first effect was detected by a decrease of mono- ers concentration in the vapors at the reactor out

let (line 19), controlled by an analyzer which foste red an increase of isobutene feed. In turn, this in crease fostered an increase of isoprene feed, through a controller of feed ratio. The second effect was de¬ tected by a decrease of liquid level in the reactor, which fostered the intervention of the reactor le¬ vel controller, which increased the MeCl feed. All fresh monomers and MeCl were cooled to -45 C° in heat exchanger 16.

According to the above flow sheet, the po- lymerization heat was removed as latent heat of vap£ rization, and the vapor output from the reactor was dependent on catalyst concentration inside the reactor. Consequently, the feed of one component of the cata¬ lyst system was controlled by the vapors output of the reactor. The feed of the second component of the catalyst was kept proportional to the first one by a controller of feed ratio.