This invention relates to new methacrylate-alkylene-methacrylate (MAM) tri- block or multi-block (preferably penta-block) copolymers.

MAM block copolymers having alkyl methacrylate end blocks, and gels made therefrom, are described in our co-pending International Patent Application No PCT/GB96/01381 (RK509). The new block copolymers according to the present invention have properties not contemplated by that earlier development.

The present invention provides a rnethacrylate-alkylene-methacrylate (MAM) triblock or multi-block (preferably penta-block) copolymer of the general structure

GM-(E)-A-(E)-GM

wherein GM represents a glycidyl methacrylate (GMA) homo- or co-polymer end block; (E) represents one or more optional intermediate blocks of alkyl methacrylate homo- or co¬ polymer or styrene homo- or co-polymer; and A represents a hydrogenated or unhydrogenated polyalkylene block. The polyglycidylmethacrylate end blocks provide the new MAM copolymers with reactive capabilities not previously available, which may be used for cross-linking or grafting reactions in subsequent use of the new materials.

One preferred form of the new MAM copolymers has the triblock copolymer structure GM-A-GM.

In another preferred form, the intermediate block E is an alkyl methacrylate (M'), preferably methylmethacrylate (MMA), homopolymer block, the MAM copolymer having the structure GM-M'-A-M'-GM. Alternatively, the M' block may be a methacrylate copolymer mono-block (that is, a single block of random or other copolymer including the methacrylate units, as distinct from a di-block copolymer) of two or more

alkylmethacrylates, or a copolymer mono-block of at least one alkylmethacrylate and another monomer other than a glycidyl methacrylate.

In a further preferred form of the new MAM copolymers, the intermediate block E is a styrene (S) homopolymer block, the MAM copolymer having the structure GM-S-A-S- GM. In this context, styrene is understood to include lower alkyl derivatives, notably alpha-methyl styrene, although unsubstituted styrene is preferred. Copolymer mono-blocks of styrene with .another monomer, for example butadiene, may be useful as alternatives to the preferred styrene homopolymer blocks.

Other multi-block forms of the new MAM copolymers may be useful, for example those wherein the intermediate block E comprises two different M' blocks (M' 1 and M'2), or an M' block and a styrene block, between the glycidyl methacrylate end blocks GM and the alkylene mid-block A, as illustrated by the structures GM-M'2-M' 1-A-M' l-M'2-GM and GM-M'-S-A-S-M'-GM. Each M' block could be either a homopolymer block or a co¬ polymer mono-block.

In all embodiments of the present invention, GM may be a co-polymer mono- block comprising glycidyl methacrylate and another monomer, preferably an alkyl methacrylate, especially methyl methacrylate (MM A). The preferred polyalkylene blocks A comprise polyisoprene, polybutadiene (PBD), and especially their hydrogenated forms poly(ethylene/propylene), poly(ethylene butylene), and copolymer mixtures thereof. The number average molecular weight Mn of the triblock copolymers for some purposes is preferably within the range 40,000 - 300,000, the methacrylate blocks preferably having Mn within the range 6000 - 70,000, and the alkylene mid-blocks preferably having Mn within the range 30,000 - 160,000. However, these or other molecular weights will be selected to suit the desired end use of the polymers, for example for making gels.

The invention includes synthesis of an MAM copolymer as hereinbefore described, comprising (i) polymerisation of an alkylene monomer (preferably butadiene or isoprene) in a substantially apolar solvent (preferably cyclohexane and/or toluene),

preferably with addition of a more polar solvent (preferably diethyl ether), to form a difunctional living polyalkylene block, followed by (ii) anionic polymerisation, in the presence of that polyalkylene block, of glycidyl methacrylate at a polymerisation temperature lower than -40 C, preferably lower than -60°C. When intermediate blocks (E) are to be included, the synthesis will include, between steps (i) and (ii), an intermediate step (iii) comprising anionic polymerisation of an alkyl (preferably methyl) methacrylate or styrene to form the intermediate block E. Step (iii) may be repeated sequentially with different monomers when two or more intermediate blocks E are desired. All steps may use mixtures of monomers to give random copolymer blocks.

For some monomers, notably higher alkyl methacrylates, it may be possible to carry out the anionic polymerisation at temperatures higher than -40 C, for example up to +40 C in the case of isobornyl methacrylate, as described and claimed in our co-pending British Patent Applications No.s 9608748.1 and 9612602.4 (RK549). The synthesis may in other respects correspond to those described in the aforementioned PCT/GB96/01381 (RK509), the disclosure of which is incorporated herein by reference. In all cases, it is preferred that the anionic polymerisation is effected in the presence of a polar solvent, preferably comprising tetrahydrofuran and/or diethyl ether, preferably in a mixture with substantially apolar solvent, preferably toluene and/or cyclohexane.

The new block copolymers of the MAM type have been successfully synthesised by using the di-adduct of tert-butyllithium (t-BuLi) to meta-diisopropenylbenzene (m-DIB) as a difunctional initiator. The alkylene midblock A has been synthesised in a cyclohexane/diethylether (100/6, v/v) mixture at room temperature, whereas the methacrylate outer blocks have been synthesised in a mixture of cyclohexane/diethylether/THF (100/6/150, v/v/v) at -78°C. Block copolymers of a very narrow molecular weight distribution (1.10) have been analysed by differential scanning calorimetry (DSC), transmission electron microscopy (TEM) and tensile testing. These materials are phase-separated and can exhibit tensile strength up to 22 MPa together with very high elongation at break (1500%).

Various aspects of the present invention will now be illustrated in more detail by way of example, using materials and methods generally as follows.

Materials; GMA and MMA (Aldrich) were first refluxed over CaH ₂ under a nitrogen atmosphere, then distilled under reduced pressure and stored under nitrogen at - 20 C. Just before polymerisation, GMA was added at -78 C to a 50/50 (v/v) mixture of diisobutyl aluminium hydride (DIBAH: 0.1N in toluene) and triethylaluminium (TEA: 0.1 N in toluene), whereas MMA was added at room temperature to a similar TEA solution, until a persistent yellowish green colour was observed in the two cases. These monomers were then distilled under reduced pressure. LiCl (99.99% purity, Aldrich) was dried overnight at 130°C and dissolved in dry THF (0.2N solution). Cyclohexane and diethylether were dried over CaH ₂ for 24h, and THF was refluxed over the deep purple sodium-benzophenone complex. After separation, all these solvents were further distilled from polystyryllithium under reduced pressure just before use. Tert-butyllithium (t-BuLi) (Aldrich, 1.3M in cyclohexane) was diluted with cyclohexane into a 0.2N solution and the final concentration determined by double titration. Meta-diisopropenyl benzene (m-DIB, Aldrich) was first dried over CaH ₂ for 24 h and then over fluorenyllithium before use. 1,1- diphenylethylene (DPE, Aldrich) was dried over sec-BuLi and distilled from diphenylmethyllithium before use. Butadiene was dried over n-BuLi.

Polymerisation: Block co-polymerization of butadiene (BD) and GMA was carried out in a previously flamed glass reactor equipped with a magnetic stirrer under an inert atmosphere. Solvent, initiator and monomers were transferred into the reactor with a syringe and/or stainless steel capillaries. Butadiene was first polymerised in a cyclohexane/diethylether mixture (100/6, v/v) at room temperature over night, using a diadduct of m-DIB and two equivalents of t-BuLi (deep red color) as a difunctional initiator previously prepared in cylcohexane at 50 C for 2 hours. When the polymerisation of butadiene was complete, an aliquot of the polymer solution was picked out and deactivated. The polymer formed was recovered by precipitation into methanol and analysed by size exclusion chromatopgrahpy. The PBD dianions were then end-capped with diphenylethylene (DPE) at room temperature for 30 minutes. Finally THF containing

LiCl (ratio LiCl/living sites=5) was added to the reactor at 0°C, with formation of a cyclohexane/THF (40/60, v/v) mixture (deep red color), to which MMA and/or GMA was finally added and polymerised at -78 C. Block copolymers were recovered by precipitation in methanol and dried at room temperature for 2 days in vacuum.

Film Preparation; Block copolymers were mixed with lwt% Irganox 1010 (Trade Mark, Ciba-Geigy Corp.) hindered phenol antioxidant, and dissolved in toluene. This solution (8wt%) was poured into a Petri dish and the solvent was allowed to evaporate slowly over 3 to 4 days at room temperature. The resulting films were dried to constant weight in a vacuum oven at 40 C. They were elastomeric, transparent and colourless with a smooth surface.

Analysis: Molecular weight and molecular weight distribution were measured by size exclusion chromatography (SEC) with a Waters GPC 501 apparatus equipped with linear styragel columns. THF was the eluent (flow rate of 1 ml/min) and polystyrene (PS) standards were used for calibration with the following viscosimetric relationships : [ηM.SόxlO^ M ⁰ - ⁷¹⁴ for PS in THF; [η]=4.57xl0 ^"4 M ⁰⁶⁹³ for PBD in THF.

H NMR Spectra were recorded with a Brucker AM-400 spectrometer, using CDC1 ₃ as a solvent. Content of the PBD 1 ,2 units was calculated by H NMR from the relative intensity of the signals at 4.9 ppm (CH=: 1,2 double bond) and at 5.4 ppm (CH=: 1,2 plus 1,4 double bond). Composition of the copolymers was calculated by H NMR from the integration of the signal for the 1,2 units of PBD and the signal at 3.5 ppm for the O-CH ₃ group of the MMA units or the signals at 2.64, 2.84, 3.23 ppm for the epoxy-ring protons of the GMA units.

Differential scanning calorimetry (DSC) was carried out with a DuPont 900 instrument, calibrated with indium. The heating rate was 20°C/min, and glass transiti temperature was reported as the inflection point of the heat capacity jump.

Toluene cast films were microtomed into 70 mm thick sections and exposed to a 1% aqueous solution of Os0 for 30 minutes. These samples were analysed with a transmission electron microscope (model Philips CM- 12; accelerating voltage of 100 kv).

Tensile measurements were conducted with an Adamel Lhomargy tensile tester. Testing samples (microdumbells) cut from solution cast films were extended at 200 mm min at room temperature. Reported data are the average of three measurements.

Synthesis and characterisation; Block copolymers with PBD as the midblock and GMA homopolymer, or copolymers of GMA and MMA, as the outer blocks were synthesised as follows with the di-adduct of m-DIB and two t-BuLi equivalents as a difunctional initiator.

GMA homopolymer as the outer blocks: In the aforementioned PCT/GB96/01381, the diadduct of t-BuLi onto m-DIB is used as the initiator for the synthesis of MAM triblock copolymers by sequential addition of butadiene and MMA, respectively. In substitution of GMA for the MMA, the first problem to be solved was the purification of commercial GMA (Aldrich) contaminated by various impurities: inhibitors, methacrylic acid and epichlorohydrin or glycidol. In contrast to MMA, which can be purified by careful distillation from triethylaluminium (TEA), GMA was very prone to polymerisation in the presence of TEA at room temperature. When GMA is purified by distillation in the presence of small amounts of polystyryl lithium, reproducibility of synthesis of GMA- BD- GMA triblock copolymer with a mono-modal molecular weight distribution is quite a problem. Although t-butyl methacrylate has been efficiently purified by a combination of diisobutyl aluminium hydride (DIB AH) with TEA, when this purification recipe is used on GMA at room temperature, a very exothermic reaction is observed with polymerisation of GMA. However, when the mixture DIBAH/TEA (50/50, v/v) is added to GMA previously cooled down to -78°C, GMA can be distilled under vacuum without significant polymerisation, and this method, which is another aspect of the present invention, has been successfully used for preparation of GMA-BD-GMA triblock copolymer. Thus, addition of the purified GMA to PBD dianions end-capped with DPE

has led to a copolymer with a narrow molecular weight distribution (1.10), as shown in Table 1. The reaction medium forms a gel as result of the GMA polymerisation, which gel dissolves when treated with an acid and warmed up to room temperature. Typical SEC traces for both the PBD midblock and the triblock copolymer show a very narrow molecular weight distribution (1.10), which suggests that the polybutadienyl dianions end- capped by DPE quantitatively initiate the GMA polymerisation. The copolymerization yield is quantitative.

The typical Η NMR spectrum of a poly(GMA-b-BD-b-GMA) triblock shows: absence of resonance characteristic of the methacrylic unsaturation, in agreement with the GMA polymerisation through the carbon-carbon double bond; the two resonances of the α-methyl protons of GMA units at 0.94 (syndiotactic triads) and 1.10 ppm (heterotactic triads) allow the PGMA tacticity to be calculated; a large resonance at ca. 2.10 ppm, which is typical of the methylene protons in the PBD and PGMA main chain; three characteristic resonances for the oxirane protons are observed at 2.64, 2.84 and 3.23 ppm; the two resonances at 3.83 and 4.08 ppm have to be assigned to the methylene protons in the PGMA side chain; three resonances peaks are observed of the protons of the double bonds in the PBD chain at 4.95 (-CH^FL,, 1,2 units), 5.38 (-CH=CH-, 1,4 units) and 5.56 (- CH=CH ₂ , 1,2 units) ppm, respectively. The relative intensity of these peaks is consistent with the expected molecular structure. The final triblock copolymer is soluble in common solvents, such as THF and toluene. Within the limits of the experimental accuracy, each GMA monomeric unit bears an epoxide, indicating that the methacrylic unsaturation is selectively involved in the anionic polymerisation of GMA, whereas the oxirane substituent is kept unchanged.

GMA and MMA copolymers as the outer blocks: The nucleophilicity of most methacrylic esters is similar enough for block and random copolymers of GMA and MMA to be synthesised. Table 1 shows that GMA-MMA-BD-MMA-GMA pentablock (sample B) can be synthesised under the same experimental conditions as above, by the sequential addition of MMA and GMA, respectively, to the polybutadienyl dianions end-capped by DPE. The deep red color of the reaction medium disappears upon the addition of MMA,

which indicates a fast initiation of the MMA polymerisation. Thirty minutes later, the second monomer GMA is added. A very narrow molecular weight distribution is observed for both the PBD midblock and the pentablock copolymer (1.15), in agreement with the quantitative cross-addition of MMA to the polybutadienyl dianions end-capped by DPE and of GMA to the living PMMA anions.

A mixture of MMA and GMA monomers has also been added to the living PBD dianions end-capped by DPE to make (GMA-co-MMA)-BD-(MMA-co-GMA) triblock co¬ polymer (sample C). Although the anionic polymerisation of a mixture of two monomers does not necessarily result in a random copolymer, it has been shown that both the MMA anions+GMA and the GMA anions+MMA cross-additions are efficient, which confirms a quite comparable electro-affinity. Both the PBD midblock and the copolymer are of a narrow molecular weight distribution (1.10). The Η NMR spectrum of this copolymer shows the expected resonances for the methoxy protons of the MMA units at 3.56 ppm and the oxirane protons at 2.64, 2.84 and 2.32 ppm, which confirms the GMA and MMA copolymerization. The chemical composition calculated from this H NMR spectrum is in good agreement with the weight ratio of the monomers added to the reaction medium (Table 1).

An MMA-BD-MMA (hereafter MBM) triblock copolymer (sample D, Table 1) of a weight composition similar to the previously discussed copolymers (A, B, C) has been similarly synthesised for sake of comparison.

Differential Scanning Calorimetry (DSC): DSC is a very useful technique to detect phase separation in binary blends and block copolymers of two immiscible components, provided that the two glass transition temperatures (Tg) are sufficiently different from each other and the weight composition is far from the extreme values. Indeed, the observation of Tg of the outer block in thermoplastic elastomers is usually a problem because this block is the minor component. This situation prevails for the block copolymers mentioned in Table 1, since DSC of toluene cast films shows a very diffuse and ill-defined Tg for the outer block and a well-defined Tg for the polybutadiene block at

-58°C. This observation is consistent with data previously reported for MBM copolymers. Thus the soft PBD block component is completely phase separated from the blocks of PGMA, poly(GMA-co-MMA), PGMA-PMMA sequential blocks, and PMMA in the block copolymers reported in Table 1.

Morphology and properties: The phase morphology was analysed by transmission electron microscopy (TEM) and confirmed the phase separation seen by DSC, the thermoplastic outer block, ie. PGMA (sample A), poly(MMA-b-GMA) (sample B), and poly(MMA-co-GMA) (sample C), forming hard micro-domains dispersed in a continuous rubbery phase. The hard phase appears to be spherical, with an average diameter of ca.l0~15nm for the block copolymers of samples A, B and C.

The phase separation has also been confirmed by stress-strain measurements, which show a stress-strain behaviour typical of a cross-linked rubber, ie, low initial modulus, high elongation at break and high ultimate tensile strength. This observation is indicative of phase separation and efficiency of the PGMA micro-domains in restricting the flow of the soft polybutadiene segments. Table 2 compares the tensile properties of the block copolymer samples. The MGM copolymer sample D is much harder than the three PGMA based block polymers A, B and C of a comparable outer block content, since the initial modulus of MBM copolymer is ca. ten times higher and the elongation at break is smaller (Table 2). The three copolymers containing PGMA based outer blocks have similar tensile properties. Nevertheless, the pentablock copolymer has somewhat higher initial modulus and ultimate tensile strength, which might be the signature of the PMMA component, known to impart hardness to the triblock D. The permanent set is not very different for all samples.

Table 1. Block co-polymers with a central PBD block and GMA (or MMA) based outer blocks.

Sample Hard PBD Composition (%f Copolymer Syndio

Block Mn Mn Mw/ 1,2 PBD PMMA PGMA Mw/Mn ^b (%) cal ^a sec b Mn ^b (%) ^c

A PGMA 42 52 1.10 44 80(78) / 20(22) 1.10 75

B MMA-b- 42 49 1.10 42 77(78) 1(1) 22(21) 1.15 / GMA

C MMA-co- 42 46 1.10 44 83(80) 7(8) 10(12) 1.10 / GMA

D PMMA 70 85 1.10 42 80(78) 20(22) / 1.10 78 ^a calculat ed from amoun t of monomer and initiator usec ; ^D SEC based on the universal calibration method; c 1 H NMR (theoretical value in brackets).

Table 2. Thermal and mechanical properties of block co-polymers with a PBD mid-block and PGMA/PMMA outer blocks.

Sample Tg° (°C) Initial Modulus Tensile Elongation at Permanent Set (MPa) ^b Strength Break (%) at Break (%) ^c

(MPa)

A -58 0.5 16 1500 22

B -58 0.8 22 1400 20

C -56 0.5 13 1400 25

D -60 6.0 22 1150 20 a ' T D-XS-IC/-I 1 heating rate 20°C/min; ratio of stress to strain at small deformation; ratio of the irreversible deformation at break to the initial length of sample.