Method to Minimize the Transition Time From One Polymer Grade to Another Polymer Grade in a Polymerization Plant

Abstract

The present invention relates to a method for reducing the transition time and/or the polymer waste being out of the specification during the change from a first polymer grade to a second polymer grade in a continuous polymerization process conducted in a polymerization plant, wherein the polymerization plant comprises at least one back-mixing reactor, at least one monomer and at least one processing agent selected from the group consisting of catalysts, co-catalysts, polymerization initiators, comonomers, chain-transfer agents, branching agents, solvents and arbitrary combinations of two or more of the aforementioned agents are added before and/or during the polymerization process into the at least one back-mixing reactor, wherein the concentration of one or more of the at least one added processing agent in the feed introduced into the at least one back-mixing reactor is varied as a function of time from a first value associated with the first polymer grade to a final value associated with the second polymer grade, —the first polymer grade and the second polymer grade are one or more parameters related to the molecular weight of the polymer to be produced and/or one or more parameters related to the composition of the polymer to be produced and/or one or more parameters related to the structure of the polymer to be produced and/or one or more parameters related to the amount of the polymer to be produced, wherein during the variation of the concentration of the at least one added processing agent from the first value to the final value one or more intermediate values are adjusted, wherein at least one, preferably the majority and more preferably all of the one or more intermediate values are closer to the final value than to the first value, wherein the absolute difference between the at least one, preferably each of the majority and more preferably each of all of the one or more intermediate values from the first value is greater than the absolute difference between the final and the first value, wherein the one or more intermediate values are maintained for a time which is calculated on the basis of only residence time in the at least one back-mixing reactor and steady-state correlations between input and output of the reactor and/or of the polymerization plant, and wherein the method is performed without performing dynamic modelling.

Claims

1. A method for reducing the transition time and/or the polymer waste being out of the specification during the change from a first polymer grade to a second polymer grade in a continuous polymerization process conducted in a polymerization plant, wherein the polymerization plant comprises at least one back-mixing reactor, at least one monomer and at least one processing agent selected from the group consisting of catalysts, co-catalysts, polymerization initiators, co-monomers, chain-transfer agents, branching agents, solvents and arbitrary combinations of two or more of the aforementioned agents are added before and/or during the polymerization process into the at least one back-mixing reactor, wherein the concentration of one or more of the at least one added processing agent in the feed introduced into the at least one back-mixing reactor is varied as a function of time from a first value associated with the first polymer grade to a final value associated with the second polymer grade, the first polymer grade and the second polymer grade are one or more parameters related to the molecular weight of the polymer to be produced and/or one or more parameters related to the composition of the polymer to be produced and/or one or more parameters related to the structure of the polymer to be produced and/or one or more parameters related to the amount of the polymer to be produced, wherein during the variation of the concentration of the at least one added processing agent from the first value to the final value one or more intermediate values are adjusted, wherein at least one of the one or more intermediate values is closer to the final value than to the first value, wherein the absolute difference between the at least one of the one or more intermediate values from the first value is greater than the absolute difference between the final and the first value, wherein the one or more intermediate values are maintained for a time which is calculated on the basis of only residence time in the at least one back-mixing reactor and steady-state correlations between input and output of the reactor and/or of the polymerization plant, and wherein the method is performed without performing dynamic modelling.

2. The method in accordance with claim 1, wherein the method is performed without the use of a controller.

3. The method in accordance with claim 1, wherein during the variation of the concentration of the at least one added processing agent, the concentration of the at least one added processing agent is decreased from the first value c.sub.1 to one or more intermediate values α.Math.c.sub.2 and then increased from the last intermediate value α.Math.c.sub.2 to the final value c.sub.2, wherein the one or more intermediate values α.Math.c.sub.2 are lower than the final value so that α<1, and wherein the intermediate values are maintained for a time $\Delta .Math..Math.t\le \tau .Math..Math.\ln \left(\frac{1-\alpha .Math..Math.c_2.Math.\text{/}.Math.c_1}{\left(1+\mathrm{.Math.}-\alpha \right).Math.c_2.Math.\text{/}.Math.c_1}\right)$ wherein: c.sub.1 is the first concentration of the at least one added processing agent, c.sub.2 is the final concentration of the at least one added processing agent, α is the minimum factor, by which the minimum intermediate value is lower than the final concentration c.sub.2, τ is a time period being at least the average residence time of the reaction mixture in the back-mixing reactor and ε is the relative tolerance on the final concentration of the at least one added processing agent, which ensures that the steady state second polymer grade is so that it is considered by the user as being within the specification, i.e. has not more than an acceptable difference to the target second polymer grade.

4. The method in accordance with claim 1, wherein during the variation of the concentration of the at least one added processing agent, the concentration of the at least one added processing agent is increased from the first value c.sub.1 to one or more intermediate values α.Math.c.sub.2 and then decreased from the last intermediate value α.Math.c.sub.2 to the final value c.sub.2, wherein the one or more intermediate values α.Math.c.sub.2 are higher than the final value so that α>1, and wherein the intermediate values are maintained for a time $\Delta .Math..Math.t\le \tau .Math..Math.\ln \left(\frac{1-\alpha .Math..Math.c_2.Math.\text{/}.Math.c_1}{\left(1-\mathrm{.Math.}-\alpha \right).Math.c_2.Math.\text{/}.Math.c_1}\right)$ wherein: c.sub.1 is the first concentration of the at least one added processing agent, c.sub.2 is the final concentration of the at least one added processing agent, α is the maximum factor, by which the maximum intermediate value is higher than the final concentration c.sub.2, τ is a time period being at least the average residence time of the reaction mixture in the back-mixing reactor and ε is the relative tolerance on the final concentration of the at least one added processing agent, which ensures that the steady state second polymer grade is so that it is considered by the user as being within the specification, i.e. has not more than an acceptable difference to the target second polymer grade.

5. The method in accordance with claim 4, that ε is less than 0.2, preferably less than 0.1 and more preferably less than 0.05.

6. The method in accordance with claim 4, τ is at most ten times the average residence time in the back-mixing reactor, more preferably at most five times the average residence time in the back mixing reactor, even more preferably at most three times the average residence time in the back mixing reactor and most preferably equal to the average residence time in the back mixing reactor.

7. The method in accordance with claim 1, wherein the first polymer grade and the second polymer grade are each selected from the group consisting of molecular weight of the polymer to be produced, polydispersity of the polymer to be produced, melt flow index of the polymer to be produced, density of the polymer to be produced, viscosity of the polymer to be produced, degree of branching of the polymer to be produced, solid concentration of the polymer to be produced, stereochemical arrangement of the monomers in the polymer to be produced and arbitrary combinations of two or more of the aforementioned grades.

8. The method in accordance with claim 7, wherein the first polymer grade and the second polymer grade are each the number average molecular weight or the weight average molecular weight of the polymer to be produced.

9. The method in accordance with claim 1, wherein the at least one back-mixing reactor is a reactor selected from the group consisting of loop reactors, continuous stirred tank reactors and arbitrary combinations of two or more of the aforementioned reactors.

10. The method in accordance with claim 1, wherein a premixer is installed before the back-mixing reactor in order to homogenize the feed streams to the back-mixing reactor.

11. The method in accordance with claim 1, wherein at least one monomer selected from the group consisting of lactide, L-lactide, D-lactide, meso-lactide, combinations of L-lactide, D-lactide and meso-lactide, cyclic monomers, styrene, vinyl acetate, acrylates, methacrylates, such as methyl-methacrylate, ethylene, propylene, butylene, a combination of an isocyanate compound and a polyol, amides, ε-caprolactone, glycolide and mixtures of one or more of the aforementioned monomers is used.

12. The method in accordance with claim 1, wherein at least one catalyst is used, wherein the catalyst is at least one organometallic compound comprising a metal selected from the group consisting of magnesium, titanium, zinc, aluminum, indium, yttrium, tin, lead, antimony, bismuth and any combination of two or more of the aforementioned metals, wherein the at least one organometallic compound preferably comprises as organic residue a residue selected from the group consisting of alkyl groups, aryl groups, halides, oxides, alkanoates, alkoxides and any combination of two or more of the aforementioned groups, and/or wherein the catalyst is an azo- and/or diazo-compound, preferably 4-pyrrolidinopyridine, 1,8-diazabicycloundecene, 1,5,7-triazabicyclo[4.4.0]dec-5-ene, 7-methyl-1,5,7-triazobicyclo[4.4.0]dec-5-ene or 2-tert-butylimino-2-diethylamino-1,3-dimethylperhydro-1,3,2-diazaphosphorine.

13. The method in accordance with claim 1, wherein at least one initiator is used, wherein the initiator is at least one compound comprising at least one carboxyl group and/or hydroxyl group and preferably at least one compound selected from the group consisting of water, alcohols, lactic acid, oligomers of a cyclic ester, polymers of a cyclic ester and any combination of two or more of the aforementioned substances.

14. The method in accordance with claim 1, wherein at least one polymerization initiator is used, wherein the polymerization initiator is selected from the group consisting of peroxides, hydroperoxides, azo-compounds, disulfides, tetrazenes and combinations of two or more of the aforementioned compounds and preferably a peroxide in combination with ferrous (Fe.sup.2+) ions, a peroxide in combination with other metallic ions, namely Cr.sup.2+, V.sup.2+, Ti.sup.3+, Co.sup.2+, Cu.sup.2+, or a peroxide in combination with an amine.

15. The method in accordance with claim 1, wherein i) the at least one monomer is selected from the group consisting of lactide, L-lactide, D-lactide, meso-lactide and mixtures of one or more of the aforementioned monomers, ii) the first polymer grade and the second polymer grade are the number average molecular weight or the weight average molecular weight of the polylactic acid polymer to be produced and iii) as processing aid at least one catalyst and at least one initiator are added, wherein the concentration of the at least one initiator is varied as a function of time from a first value associated with the molecular weight of the polylactic acid polymer to a final value associated with the molecular weight of the polylactic acid polymer to be produced.

16. The method in accordance with claim 1, wherein i) the at least one monomer is selected from the group consisting of styrene, vinyl acetate, acrylates, methacrylates, such as methyl-methacrylate, ethylene, propylene, butylene and mixtures of one or more of the aforementioned monomers, ii) the first polymer grade and the second polymer grade are the number average molecular weight or the weight average molecular weight of the polymer to be produced and iii) as processing aid at least one catalyst and at least one polymerization initiator or one chain transfer agent are added, wherein the concentration of the at least one polymerization initiator or chain transfer agent is varied as a function of time from a first value associated with the molecular weight of the polymer to a final value associated with the molecular weight of the polymer to be produced.

17. The method in accordance with claim 3, wherein that ε is less than 0.2, preferably less than 0.1 and more preferably less than 0.05.

18. The method in accordance with claim 3, wherein τ is at most ten times the average residence time in the back-mixing reactor, more preferably at most five times the average residence time in the back mixing reactor, even more preferably at most three times the average residence time in the back mixing reactor and most preferably equal to the average residence time in the back mixing reactor.

Description

[0071] Specific embodiments in accordance with the present invention are now described with reference to the appended drawings.

[0072] FIG. 1 is a schematic drawing showing the change of the concentration of an initiator in a continuous stirred tank reactor (top figure, bold line) in dependency of a change of the concentration of the initiator in the feed introduced into the continuous stirred tank reactor (top figure, thin line) as well as a schematic drawing showing the corresponding change of a property, P, such as the melt flow index of the polymer product at the outlet of the continuous back-mixing reactor (bottom figure, bold line) performed in accordance with a prior art method.

[0073] FIG. 2 is a schematic drawing showing the change of the concentration of an initiator in a continuous stirred tank reactor (top figure, bold line) in dependency of a change of the concentration of the initiator in the feed introduced into the continuous stirred tank reactor (top figure, thin line) as well as a schematic drawing showing the corresponding change of a property, P, such as the melt flow index of the polymer product at the outlet of the continuous back-mixing reactor (bottom figure, bold line) performed in accordance with the present invention in comparison to the change of the property of polymer product at the outlet of the continuous stirred tank reactor as shown in FIG. 1 (dashed bold line).

[0074] FIG. 3 is a schematic drawing showing the change of the concentration of an initiator in the feed introduced into the continuous stirred tank reactor in accordance with another embodiment of the present invention.

[0075] FIG. 4 is a schematic drawing showing a polymerization apparatus suitable for performing the method in accordance with the present invention, as used in example 1.

[0076] FIG. 5 shows the evolution of the weight average molecular weight for samples collected at the outlet of a polymerization apparatus vs. time from a first polymer grade (Mw(t)Mw1=1) to a second polymer grade (below the horizontal line) of example 1 and comparative example 1. The data shown represent the evolution obtained with the standard step change method (Δ) known in the prior art and with the method in accordance with the present invention ().

[0077] FIG. 1 shows the change of the concentration of an initiator in a continuous stirred tank reactor (CSTR) (top figure, bold line) in dependency of a change of the concentration of the initiator in the feed introduced into the CSTR (top figure, thin line) performed in accordance with a prior art method. As shown by the thin line, the concentration of an initiator in the feed introduced into the CSTR is stepwise decreased at the time t=0 from a first value c.sub.1 to a final value c.sub.2. Apart from the initiator, the feed includes lactide as monomer and catalyst, namely tin octoate octoate. Due to the residence time of the reaction mixture in the CSTR, the concentration of the initiator within the CSTR only slowly changes, as shown by the bold line, and reaches the final concentration c.sub.2 only with a significant delay. The time period between t=0 and the time, when the concentration of the initiator within the CSTR reaches c.sub.2, is the transition time of the initiator in the CSTR. Correspondingly, the property P of the polymer produced in the CSTR also changes with time. The time period needed for the polymer property P at the outlet of the reactor to change from the starting value to the final steady state value is the transition time of the polymer property P. In more complex polymerization plants, in which additional reactors and equipments are present downstream the CSTR reactor, the polymer property P is preferably measured at the plant outlet rather than at the CSTR outlet.

[0078] In accordance with the present invention, the transition time is significantly reduced by intentionally and accurately changing the concentration of the initiator in the feed introduced into the CSTR firstly to an intermediate value α.Math.c.sub.2 being lower than the intended final concentration c.sub.2, wherein the intermediate value α.Math.c.sub.2 is maintained for a time

[00004]$\Delta .Math..Math.t\le \tau .Math..Math.\ln \left(\frac{1-\alpha .Math..Math.c_2.Math.\text{/}.Math.c_1}{\left(1+\mathrm{.Math.}-\alpha \right).Math.c_2.Math.\text{/}.Math.c_1}\right)$

wherein:
c.sub.1, c.sub.2, α, ε and τ are as defined above.

[0079] Due to this higher change of the concentration of the initiator as that from the first to the final concentration value, the concentration change of the initiator is accelerated in the polymerization reactor and as a consequence thereof also the change of the polymer grade from the first grade, characterized by the polymer property P1, to the intended one, characterized by the polymer property P2, is accelerated. As shown in FIG. 2, on account of the intermediate undershoot concentration of the initiator (FIG. 2, top), the final polymer grade property in the polymerization reactor is reached, as shown by the bold line in FIG. 2, bottom, earlier than in the case of a stepwise concentration decrease (cf. FIG. 2, dot-dashed bold line), which leads to a significant reduction in the transition time.

[0080] FIG. 3 shows a more complex curve for the change of the concentration of an initiator in the feed introduced into the CSTR in accordance with another embodiment of the present invention, in which the concentration of the initiator in the feed introduced into the CSTR is not only reduced to one intermediate concentration corresponding to α.Math.c.sub.2 as shown in FIG. 2, but to more intermediate concentrations c.sub.2. In such a case, the time of over/undershoot in the feed, Δt, is calculated from the first instant, when the feed concentration exits the (1±ε).Math.c.sub.1-range, with ε being the tolerance in the feed to obtain at steady state the polymer property P1, to the first next instant, after the intermediate value α.Math.c.sub.2, when the feed concentration enters the (1±ε) c.sub.2-range, with a being the tolerance in the feed to obtain at steady state the polymer property P2. The corresponding a is calculated using the maximum (minimum) value reached by the feed concentration during the transition from a lower (higher) to a higher (lower) set point.

[0081] Subsequently, the present patent application is illustrated by means of non-limiting examples.

EXAMPLE 1

[0082] FIG. 4 shows a schematic drawing (not to scale) of a polymerization apparatus for the continuous production of polylactic acid from the corresponding cyclic diester monomer (actide) according to a preferred embodiment.

[0083] The actide monomer feed 1 is mixed with a stream of the processing agent, such as but not limited to, polymerization catalyst and/or initiator 2 into a premixer unit 3. The so-obtained premixed phase 4 is then pumped to the loop reactor 6. A fraction of the loop outlet stream 8 is then pumped back and fed to the inlet of the loop reactor together with the feed stream 4. The remaining fraction of the loop outlet stream 9 is pumped to a plug flow reactor 10, where the conversion further increases up to the targeted final value. At the outlet of the plug flow reactor, the reacted stream containing mainly polymer 11, is pumped to a final unit 12, where the manufacturing process is completed. The final unit 12 can comprise one or more subunits selected from but not limited to one or more devolatilization steps, one or more units for mixing and/or blend additives and/or other polymers in order to improve the mechanical, rheological and/or thermal properties, finishing and/or pelletization units, drying and/or crystallization units, before the final product is collected at the outlet of the polymerization apparatus 13. The types and amount of subunits present in 12 depends on the manufacturing needs.

[0084] In another embodiment the polymerization apparatus may contain separate inlet streams for the catalyst and the initiator to the premixer unit.

[0085] In another embodiment of the polymerization apparatus, the premixer unit can be absent and the feed streams to the reactor can be mixed in a separate unit before feeding them to the loop reactor.

[0086] In another embodiment of the polymerization apparatus, the lactide and the other chemical agents are fed separately directly into the loop reactor.

[0087] In the preferred polymerization apparatus embodiment shown in FIG. 4, the average residence time in the premixer unit 3, defined as volume of premixer unit divided by the flowrate feed to the premixer unit, is less than ½, more preferably less than ⅕, and even most preferably less than 1/10 than the average residence time in the loop reactor 6, defined as the ratio between the loop reactor volume and the total feed flowrate (1+2, or 4). But not limited to the preferred conditions, the present invention can be applied also to cases where larger premixers are used, e.g. with residence time larger than ½ or comparable to the residence time in the loop.

[0088] The concentration of the initiator feed to the loop reactor 4 is calculated as the ratio of the flow rates fed to the premixer as:

C.sub.1=F.sub.l/(F.sub.l+F.sub.cat+F.sub.Lactide)

wherein C.sub.1 is the initiator concentration in the feed to the loop reactor used to reach the first steady state conditions,
F.sub.l is the flowrate of (pure) initiator in stream 1,
F.sub.cat is the flowrate of (pure) catalyst in the stream 1 and F.sub.Lactide is the flowrate of lactide 2.

[0089] It has to be noted that the equation reported to calculate the concentration of initiator fed to the reactor can be calculated with the same formula also when, according to another embodiment above, the different streams of monomer and agents are fed through separate streams with their respective flow rates, directly into the loop reactor, and then mixed together with the stream circulating in the loop reactor 5. The lactide and the agents can be fed as separated streams to different points of the loop reactor too.

[0090] In any case, the concentration of a given chemical agent fed to the continuously mixing reactor is calculated according to standard definitions as the concentration of the agent in the whole amount of material actually entering the continuously mixing reactor volume (6), independently from the possibly present premixing steps and from the number of streams and agents which may constitute the overall feed before entering the continuously mixing reactor volume.

[0091] In this example, liquid lactide was pumped continuously to the premixer unit of a polymerization apparatus as depicted in FIG. 4 at a constant flowrate of F.sub.lactide=25 kg/h.

[0092] At the inlet of the premixer, a controlled amount of catalyst (tin octoate) and of initiator (ethyl-hexanol) was fed via the streamline 1.

[0093] The streams of lactide and catalyst and initiator were kept constant until the plant operated under steady state conditions and the molecular weight of the polylactic acid collected at the outlet of the apparatus 13, Mw.sub.1, corresponding to the specification of a first polymer grade within the accepted tolerance, was constant in time.

[0094] The apparatus was operated under steady state conditions until the desired amount of the first polymer grade was produced.

[0095] Then, to switch the production from this first polymer grade characterized by a first weight average molecular weight Mw.sub.1, to a second polymer grade characterized by a second molecular weight Mw.sub.2=0.67×Mw.sub.1 with tolerance of 20% (ε=0.2), the feed flowrate of initiator to the polymerization plant was first changed to an intermediate value αC.sub.2, with α=1.333 and after a time Δt it was then decreased back to the final value C.sub.2 associated with the production of the second polymer grade characterized by the second molecular weight Mw.sub.2.

[0096] Because, as known in prior art, in some cases it can be assumed as a reasonable estimate that the molecular weight is inversely proportional to the amount of initiator used, the flowrate of initiator was changed such that the second final concentration value of the initiator in the feed was C.sub.2=1.5×C.sub.1.

[0097] According to the present invention, the intermediate concentration value had to be maintained for a time calculated:

[00005]$\Delta .Math..Math.t\le \tau .Math..Math.\ln \left(\frac{1-\alpha .Math..Math.c_2.Math.\text{/}.Math.c_1}{\left(1-\mathrm{.Math.}-\alpha \right).Math.c_2.Math.\text{/}.Math.c_1}\right)$

which, in the present example using τ equal to five times the residence time in the loop reactor gives:

[00006]$\Delta .Math..Math.t\le \tau .Math..Math.\ln \left(\frac{1-1.333.Math.1.5}{\left(1-0.2-1.333\right).Math.1.5}\right)=0.223.Math.\tau$

[0098] Accordingly, Δt=0.184τ was used, τ equal to five times the residence time in the loop reactor.

[0099] FIG. 5 reports the measured weight average molecular weights vs. time as collected at the outlet of the polymerization plant 13.

[0100] To allow a fair comparison of the two sets of data, the time scale was normalized by the average residence time in the loop and the value of time=0 was assigned to the instant when the initiator concentration was firstly changed from the first value c.sub.1 to a second value c.sub.2 (step-change) or to a second value αc.sub.2 (new method), respectively.

[0101] The horizontal line in the figure represents the limit below which the second polymer molecular weight grade was considered within specification for a specific application. This limit was calculated as 20% higher than the targeted molecular weight, as reported above.

[0102] The experimental results evidence that the with the new procedure the molecular weight as a function of time Mw(t) decreases much more rapidly to the new steady state than with the standard procedure.

[0103] This way, the transition from a first polymer grade with a first molecular weight to a second polymer grade with a second molecular weight can be performed faster and the amount of off spec material produced during the shorter transition time results remarkably reduced.

COMPARATIVE EXAMPLE 1

[0104] Liquid lactide was pumped continuously to the premixer unit of a polymerization apparatus (as depicted in FIG. 4) at a constant flowrate of F.sub.lactide=25 kg/h.

[0105] At the inlet of the premixer, a controlled amount of catalyst (tin octoate) and of initiator (ethyl-hexanol) was fed via the streamline 1.

[0106] The streams of lactide and catalyst and initiator were kept constant until the plant operated under steady state conditions and the molecular weight of the polylactic acid collected at the outlet of the apparatus 13, Mw.sub.1, corresponding to the specification of a first polymer grade within the accepted tolerance, was constant in time.

[0107] The apparatus was operated under steady state conditions until the desired amount of the first polymer grade was produced.

[0108] Then, to switch the production from this first polymer grade characterized by the first weight average molecular weight Mw.sub.1, to a second polymer grade characterized by a second weight average molecular weight, Mw.sub.2, the feed flowrate of the initiator to the polymerization plant was changed to such an extent that its concentration in the feed to the loop reactor changed stepwise from the first value C.sub.1 associated to the first molecular weight Mw.sub.1, to a second value C.sub.2, associated with a second molecular weight, Mw.sub.2=0.65×Mw.sub.1. Because, as known in prior art, in some cases it can be assumed as a reasonable estimate that the molecular weight is inversely proportional to the amount of initiator used, the flowrate of initiator was changed such that the second concentration value of the initiator in the feed was C.sub.2=1.54×C.sub.1.

[0109] The time evolution of the molecular weight at the outlet of the polymerization plant 13 is shown in FIG. 5.