WO2001062803A2 - Supported catalysts for use in atom transfer radical polymerization and continuous processes for atom transfer radical polymerization - Google Patents

Abstract

A series of support systems for atom transfer radical polymerization of various monomers is disclosed. The complexes of a copper bromide and a multidentate amine physically adsorbed onto silica gel, alumina gel or other polymeric particles mediate living polymerization of non- or low polar monomers, including methacrylates, acrylates and styrenic monomers. The polymers have controlled molecular weights and narrow molecular weight distributions. The catalyst/ligand complex may be easily recovered for subsequent uses. The recycled catalysts have high activities and mediated living polymerizations, producing polymers with controlled molecular weights and low molecular weight distributions. Multidentate amines were also chemically grafted onto silica gels. These grafted catalyst ligand complexes also mediate living polymerizations of monomers such as methacrylates, acrylates and styrenic monomers. Also disclosed is a continuous process for atom transfer radical polymerization using a column reactor packed with a supported catalyst. Homo-polymers and random copolymers with controlled molecular weights may be produced by the reactor in a continuous manner. Block copolymers of different monomers may also be prepared using packed column reactors arranged in series.

Description

Title: SUPPORTED CATALYSTS FOR USE IN ATOM TRANSFER RADICAL POLYMERIZATION AND CONTINUOUS PROCESSES FOR ATOM TRANSFER RADICAL POLYMERIZATION

FIELD OF THE INVENTION

This invention relates to silica gel, alumina gel and polymer particles supported catalysts. More particularly, it relates to silica gel, alumina gel and crosslinked polystyrene particles supported catalysts useful for atom transfer radical homo- and co-polymerization of monomers such as methacrylic, acrylic and styrenic monomers. It also relates to continuous processes for atom transfer radical polymerization. BACKGROUND TO THE INVENTION

Atom transfer radical polymerization ("ATRP") was first discovered in 1995 by Matyjaszewski et al. (J. Am. Chem. Soc. 1995, 117, 5614) and Sawamoto et al. (Macromolecules 1995, 28, 1721). A typical ATRP system comprises a vinyl monomer, a radical initiator (organohalide), a metal halide catalyst and a ligand. ATRP has the advantage that it is capable of producing well-defined functional polymers. It is advantageous over typical anionic, cationic and other types of living polymerization, with respect to the applicability of the monomer type, tolerance to water and other protonic species, and polymerization temperature range. ATRP is also very useful to synthesize functional polymers of low molecular weight. For example, polymers with allyl, vinyl, and hydroxyl groups have been readily prepared with molecular weights less than 10⁵ (Nakagawa, Y. et al. Polym. J. 1998, 30, 138; Matyjaszewski, K. et al. /. Pol m. Sc Part A: Polym. Chem. 1998, 36, 823; Haddleton, D. M. et al. Chem. Commun. 1997, 683; Coessens, V. et al. Macromol. Rapid Commun. 1999, 20, 127; Malz, H. et al. Macromol. Chem. Phys. 1999, 200, 642).

However, one of the major challenges facing the ATRP process is its low catalyst efficiency and thus high catalyst concentration used. In a typical ATRP recipe, the initiator to catalyst ratio is usually 1:1, which is one catalyst molecule mediating only one polymer chain. The metal halide catalyst usually is about 0.1-1% (molar) of monomer, i.e. 100-1000 of the monomer to catalyst ratio. Therefore, after polymerization, additional processes are required to remove the catalyst from the product mixture, usually by passing the mixture through silica or alumina gels. This post-treatment is not only time- demanding but also costly due to the catalyst wastes (both the metal halide and ligand are expensive). An additional problem associated with the ATRP process is its residual color from the metal complex (typically copper). These drawbacks severely limit the industrial applicability of ATRP.

Efforts have been made to solve these problems by supporting the catalyst /ligand system onto a solid which may be readily recovered and re-used in subsequent polymerizations. For example, Haddleton et al. first developed a support system of a copper halide/ alky lpyridylmethanimine on amino-functionalized silica gel/crosslinked polystyrene particles. This system was used to polymerize methyl methacrylate with ethyl-2-bromoisobutyrate (Chem. Commun. 1999, 99). Haddleton et al. also extended this work to support a ruthenium (II) catalyst grafted onto 3-aminopropyl functionalized silica gel (Macromolecules 1999, 32, 4769). While producing moderately successful results, these support systems are problematic in that the catalyst/ligand is covalently bonded with the support, which involves the tedious process of functionalizing the support particle surface. Matyjaszewski et al. attempted to immobilize a copper (I) catalyst (CuBr) on amino-functionalized silica gel and cross linked polystyrene particles for polymerization of styrene and methyl methacrylate, but did not obtain good polymerization results. The polymer molecular weights were significantly higher than predicted, and the polydispersites were also high (>1.5) (Macromolecules 1999, 32, 2941).

In another approach, Haddleton et al. also attempted to polymerize a "free" ligand physically adsorbed onto a support, rather than using a catalyst/ligand complex covalently bonded to the support. However, an undesirable colour remained in solution due to the free catalyst/ligand. No catalyst recycling work was reported (Chem. Commun. 1999, 99).

Another drawback to the industrial application of ATRP is that to date, it has been possible only to conduct the polymerization under batch conditions. This requires the recovery of the catalyst/ligand system before it can be used again in another batch process.

Accordingly, there still exists a need for a catalyst/ligand system which is economical in that the catalyst can be recovered and reused, and which does not contaminate the resulting polymer. Further, there exists a need for a continuous ATRP process, thereby eliminating the need to recover and recycle the catalyst/ligand system. SUMMARY OF THE INVENTION

The present inventors have found that the use of a catalyst/ligand complex physically adsorbed onto a silica gel support for use in ATRP is cost effective, and the complex may be easily recovered and reused. Further, the resulting polymer is contaminated with little undesirable colour.

The present invention therefore relates to a catalyst/ligand system for use in an atom transfer radical polymerization (ATRP) comprising a catalyst/ligand complex associated with a support. Preferably, the catalyst ligand complex is either adsorbed onto the support or it is chemically grafted onto the support.

In one aspect of the present invention, the catalyst/ligand complex is adsorbed onto the support. The support is preferably a solid support selected from silica and alumina gel. Further, the catalyst/ligand complex may be a copper halide-multidentate amine ligand complex (CuX- mAMINE). The system may be used in the ATRP of nonpolar or low polar monomers, such as methacrylic, acrylic and styrenic monomers, in organic solvents (preferred toluene) to form corresponding homo-polymers and random- and block-copolymers. The adsorbed complex mediates a living polymerization of these monomers, yielding polymers with controlled molecular weights and narrow molecular weight distributions. The catalysts may be recovered by settlement, filtration, centrifugation, and /or other solid/liquid separation methods. The catalyst retains a significant portion of its initial activity during subsequent recycled uses. The subsequent use of the recycled catalyst provides good control over the molecular weight of the polymers obtained. The present inventors have also prepared catalyst/ligand complexes that are chemically grafted onto a support via novel linker groups. The length of these linker groups has been varied to determine the distance between the catalyst/ligand complex and the support, that provides optimum polymerization results. The catalyst/ligand complexes that are grafted onto the support are useful for the ATRP of high polar monomers such as dimethylaminoethyl methacrylate.

Therefore, the present invention further relates to a catalyst/ligand system for use in ATRP, comprising a catalyst/ligand complex grafted onto a support by a spacer group, the structure of which has been optimized to provide improved polymerization results. Preferably, the spacer group comprises 1-10 ethylene glycol moieties, or any other chemical moiety that provides a similar spatial relationship between the catalyst/ligand complex and the support. In an aspect of the present invention, the catalyst/ligand complex is a copper halide-multidentate amine ligand complex that is chemically grafted onto silica gel, alumina or cross-linked polymer particles and the structure of which has been optimized to provide improved polymerization results. In particular, new ligand /support conjugates have been prepared in which the distance between the ligand and the support has been optimized to provide improved polymerization results.

The present invention further provides a process for preparing a multidentate amine ligand system comprising a multidentate amine ligand that is grafted onto a silica gel support with a spacer group having one or more ethylene glycol moieties comprising the steps of:

(a) reacting silica gel with a linear or branched amino compound having a suitable leaving group to provide silica gel functionalized with a first linker group having one or more primary or secondary amines; (b) reacting the one or more primary or secondary amines of the first linker group with a mono- or polyethylene glycol diacrylate to provide silica gel functionalized with a first linker group and a mono- or polyethylene glycol acrylate; (c) reacting the silica gel functionalized with a first linker group and a mono- or polyethylene glycol acrylate with a multidentate amine compound to provide the multidentate amine ligand system. The present inventors have successfully used the above- described catalyst/ligand systems in ATRP reactions. Therefore, the present invention also relates to a method of atom transfer radical polymerization (ATRP) comprising the steps of:

(a) preparing a catalyst/ligand complex associated with a support; (b) combining the supported catalyst/ligand complex with a monomer to be polymerized; and

(c) adding an initiator to the combined supported catalyst/ligand complex and monomer and allowing the initiator, monomer, and supported catalyst/ligand complex to react under conditions sufficient for polymerization to occur.

Preferably, the catalyst/ligand complex is associated with the support by adsorption or chemical grafting.

The present invention also relates to a continuous process for

ATRP to prepare a polymer. The process comprises the steps of: (a) providing a packed reactor column comprising a catalyst/ligand complex associated with a support;

(b) passing a solution of a monomer to be polymerized and an initiator through the packed column reactor, allowing the monomer and initiator to come into contact with the catalyst/ligand complex under conditions for polymerization to occur; and

(d) recovering polymer as it exits the column reactor.

Preferably, the catalyst/ligand complex is associated with the support by adsorption or chemical grafting.

The present invention also relates to a method of atom transfer radical polymerization to prepare block copolymers using continuous packed column reactors in series comprising the steps of: (a) providing at least one packed reactor column comprising a catalyst/ligand complex associated with a support, the reactor columns being connected in series where multiple columns are packed;

(c) passing a first monomer to be polymerized, and an initiator through a first packed column reactor, allowing the monomer and initiator to come into contact with the catalyst/ligand complex under conditions for polymerization of the first monomer to form a first polymer;

(d) passing a further monomer, together with the first polymer, through a packed column reactor, allowing the further monomer and first polymer to come into contact with the catalyst/ligand complex under conditions for polymerization of the further monomer and first polymer to form a block copolymer, the further monomer being introduced into a packed column downstream of the first monomer;

(e) where desired, repeating step (d); and (f) collecting the block copolymer.

The approach of the present invention provides a simple, cost effective approach to the ATRP for homo-polymers and random- and block- copolymers of, for example, methacrylic, acrylic and styrenic monomers.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other advantages of the present invention will be more fully and completely understood through a consideration of the following description taken together with the accompanying figures in which:

Figure 1 is a schematic of a polymerization apparatus which may be used in accordance with the continuous ATRP process of the present invention;

Figure 2 is a schematic of a polymerization apparatus which may be used to form block copolymers in accordance with the continuous ARTP process of the invention; Figure 3 is a graph of the extent of methyl methacrylate polymerization versus time, with and without silica gel catalyzed by CuBr- HMTETA at 90°C with toluene; Figure 4 is a graph of polymethyl methacrylate molecular weight and polydispersity dependence on the conversion in the methyl methacrylate polymerization with and without silica gel;

Figure 5 is a graph of the methyl methacrylate polymerization versus time with different catalyst /initiator ratios;

Figure 6 is a graph of PMMA molecular weight and polydispersity dependence on conversion in MMA polymerization with different catalyst /initiator ratios;

Figure 7 is a graph of MMA polymerization catalyzed by CuBr- HMTETA with different concentrations of silica gel;

Figure 8 is a graph of PMMA molecular weight and polydispersity dependence on the conversion in MMA polymerization catalyzed by CuBr-HMTETA in the presence of different concentrations of silica gel; Figure 9 is a graph of MMA polymerization catalyzed by CuBr-

HMTETA/2 silica gel system with different solvent fractions;

Figure 10 is a graph of PMMA molecular weight and polydispersity dependence on MMA conversion in the MMA polymerization catalyzed by CuBr-HMTETA/2 silica gel system with different solvent fractions;

Figure 11 is a graph of the recycle of CuBr-HMLΕTA/2 silica gel system for MMA polymerization at 90° C;

Figure 12 is a graph of the recycle of CuBr-HMTETA/5 silica gel system for MMA polymerization at 90°C, with toluene/MMA=2 (w/w); Figure 13 is a graph of the recycle of CuBr-HMTETA/ 10 silica gel system for MMA polymerization at 90°C, with toluene/MMA = 2 (w/w);

Figure 14 is a graph of PMMA molecular weight and polydispersity dependence on MMA conversion in the MMA polymerization catalyzed by CuBr-HMTETA/2 silica gel and CuBr-HMTETA/5 silica gel systems;

Figure 15 is a graph of PMMA molecular weight and polydispersity dependence on the conversion of the MMA polymerization catalyzed by CuBr-HMTETA/ 10 silica gel; Figure 16 is a graph of the extent of conversion of MMA versus time, at a flow rate of 0.02 ml/min in a continuous polymerization reaction according to a process in accordance with the present invention;

Figure 17 is a graph of molecular weight and molecular weight distribution of PMMA versus time at a flow rate of 0.02 ml/min;

Figure 18 is a graph of the extent of conversion of MMA versus flow rate;

Figure 19 is a graph of the extent of conversion of MMA and In ([M]o/[M]) versus residence time of polymerization solution in a column reactor;

Figure 20 is a graph of the molecular weight and molecular weight distribution of PMM versus the extent of conversion;

Figure 21 shows the GPC curves of the polymers formed during the block copolymerization using the continuous process of the invention and the apparatus shown in Figure 2;

Figure 22 is a schematic showing how TEDETA was grafted onto silica gel;

Figure 23 is a graph showing the effect of reusing the CuBr- TEDETA/silica gel system on the extent of MMA polymerization versus time. ([CuBr] = 0.0158mol/L; Silica gel/CuBr (w/w) = 10; MMA /CuBr /initiator = 100/1.5/1; 80 °C; First use (•, O); Second use (■, D) and third use (A, Δ));

Figure 24 is a graph of PMMA molecular weight and polydispersity as a function of conversion in the MMA polymerization catalyzed by fresh and recycled CuBr-TEDETA/ silica gel catalysts. (See Figure 23 for other experimental conditions; First use (#, O); Second use (■, D) and third use (A, Δ));

Figure 25 is a graph showing the extent of MMA polymerization using fresh, recycled and regenerated CuBr-TEDETA /silica gel catalyst.

([CuBr] = 0.0158 mol/L; MMA/CuBr/initiator = 300/1.5/1; Silica gel/CuBr (w/w) = 10; 80 °C; Fresh catalyst (A, Δ); recycled catalyst(#, O); regenerated catalyst , D));

Figure 26 is a graph showing PMMA molecular weight and polydispersity as a function of conversion in the MMA polymerizations catalyzed by fresh, recycled and regenerated CuBr-TEDETA /silica gel catalyst. (See Figure 24 for other experimental conditions; Fresh catalyst (A, Δ); recycled catalyst(#, O); regenerated catalyst(B, □));

Figure 27 is a schematic showing the attachment of diethylenetriamine onto silica gel;

Figure 28 is a schematic showing the grafting of TEDETA and DiPA onto silica gel via a polyethylene glycol spacer;

Figure 29 is a graph showing the extent of MMA polymerization in phenyl ether catalyzed by CuBr with TEDETA-functionalized silica gel support having different linker groups. ([MMA]=1.38 mol/L, [MBP]=9.2 xlO^"3 mol/L, [CuBr]=1.39xlO^"2 mol/L, [TEDETA] /[CuBr] (molar) = 1, 60 °C; Support: SG-PEG TEDETA (A, Δ); SG-PEG₃-TEDETA (#, O); SG-PEG₁₀- TEDETA (■, D));

Figure 30 is a graph showing the extent of MMA polymerization in phenyl ether catalyzed by CuBr with DiPA-functionalized silica gel support having different linker groups. ([MMA]=1.38 mol/L, [MBP]=9.2 xlO^"3 mol/L,

[CuBr]= 1.39X10^"2 mol/L, [DiPA]/[CuBr] (molar) = 1; 60 °C; Support: SG-

PEG DiPA (A, Δ); SG-PEG₃-DiPA (•, O); SG-PEG₁₀-DiPA (■, D));

Figure 31 is a graph showing PMMA molecular weight and polydispersity as a function of conversion in phenyl ether with different

TEDETA-functionalized supports. (Figure 29 for other experimental conditions SG-PEG TEDETA (A, Δ); SG-PEG₃-TEDETA (#, O); SG-PEG₁₀-

TEDETA (■, D));

Figure 32 is a graph showing PMMA molecular weight and polydispersity as a function of conversion in phenyl ether with different

DiPA-functionalized supports. (See Figure 30 for other experimental conditions; SG-PEG₁-DiPA(B, D); SG-PEG₃-DiPA (•, O); SG-PEG₁₀-DiPA (A,

Δ));

Figure 33 is a graph showing MMA polymerization in phenyl ether at 80 °C catalyzed by CuBr with silica gel support with different supports. ([MMA]=1.38 mol/L, [MBP]=9.2 xlO^"3 mol/L, [CuBr]= 1.39xl0^'2 mol/L, [TEDETA] or [DiPA]/[CuBr] (molar) = 1, 80 °C; Support: SG-PEG₃-

TEDETA (■, D); SG-PEG₃-DiPA (•, O); SG-PEG_α-DiPA (A, Δ));

Figure 34 is a graph showing PMMA molecular weight and polydispersity as a function of conversion in phenyl ether at 80. (See Figure 33 for other experimental conditions; Support: SG-PEG₃-TEDETA (■, D); SG-

PEG₃-DiPA (•, O); SG-PEG DiPA (A, Δ));

Figure 35 is a graph showing MMA polymerization with different CuBr /initiator ratios and SG-PEG₃-TEDETA as support. ([MMA]=

1.38 mol/L, [MBP]= 9.2xl0^"3 mol/L, [TEDETA] /[CuBr] (molar) = 1, 60 °C; [CuBr]= 9.2xl0^"3 mol/L (■, D); [CuBr]=1.39xl0² mol/L (A, Δ));

Figure 36 is a graph showing PMMA molecular weight and polydispersity as a function of conversion in phenyl ether with different

CuBr/MBP ratios. (See Figure 35 for other experimental conditions; [CuBr]=

9.2xl0^"3 mol/L (■, D); [CuBr]=1.39xl0² mol/L (A, Δ)); Figure 37 is a graph showing the extent of MMA polymerization in phenyl ether using fresh and recycled SG-PEG₃-TEDETA supported CuBr.

([MMA]=1.38 mol/L, [MBP]=9.2xl0³ mol/L, [CuBr]= 1.39xl0^"2 mol/L,

[TEDETA] /[CuBr] (molar) = 1, 60 °C; fresh catalyst (A, Δ); first recycled catalyst (■, D)); Figure 38 is a graph showing the extent of MMA polymerization in phenyl ether using fresh and recycled SG-PEG -TEDETA supported CuBr.

([MMA]=1.38 mol/L, [MBP]=9.2xl0³ mol/L, [CuBr]= 1.39xl0^"2 mol/L,

[TEDETA] /[CuBr] (molar) = 1, 60 °C; fresh catalyst (•, o); first recycled catalyst (♦, <>)); Figure 39 is a graph showing PMMA molecular weight and polydispersity as a function of conversion with fresh and recycled catalysts on different supports. (See Figures 37 and 38 for experimental conditions. SG-

PEG₃-TEDETA: first use (A, Δ); second use (■, D); SG-PEG TEDETA: first use

(#, O); second use ( ♦ , 0)); and Figure 40 shows GPC traces of (a) PMMA macro-initiator and

(b) Poly(MMA-b-DMAEMA) block copolymer. ([MBP]=9.2 xlO^"3 mol/L, [CuBr]=1.39xl0² mol/L, [TEDETA] /[CuBr] (molar) = 1, 60 °C; MMA/MBP (molar) = 80; DMAEMA/MBP (molar) = 80). DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to catalyst/ligand support systems useful in ATRP, particularly in the ATRP of nonpolar or low polar monomers, such as methacrylate, acrylates and styrenic monomers. Preferably, the catalyst/ligand complex is either physically adsorbed or chemically grafted onto a silica gel (or alumina or crosslinked polystyrene particle) substrate, and can be recovered readily after the polymerization reaction, and reused in subsequent polymerization reactions.

The catalyst/ligand complex is preferably a copper halide -multidentate ("mAMINE") complex, or a copper halide-polyethyleneimine, or alkylated derivatives thereof. Most preferably, the catalyst/ligand complex is a copper halide - hexamethyltrimethylene tetramine complex ("CuX-HMTETA"), a copper halide - polyethyleneimine complex ("CuX- PEI"), a copper halide - tetraethyldiethylenetriamine complex ("CuX- TEDETA") or a copper halide - di(2-ρicolyl) amine complex ("CuX-DiPa"). Most preferably the catalyst is copper (I) bromide. A person skilled in the art would realize that the present invention could be extended to other metal catalysts, in particular to those metal catalysts known to work in the ATRP reaction, such as RuCl₂(PPh₃)₃. Adsorbed Catalyts /Ligand Systems

In one aspect of the present invention, the catalyst/ligand complex is associated with a support by physical adsorption. The support is preferably silica gel or alumina, most preferably, silica gel.

The silica gel substrate may be any commercial grade of silica gel generally available. More preferably, the silica gel has a mesh between about 60 and 400 and most preferably between about 230 and 400, with an average pore diameter of about 6θA. The catalyst/ligand complex may be adsorbed onto the silica gel or alumina gel substrate by any known method. For example, silica gel or alumina gel may be added to a solution of the catalyst/ligand complex and a monomer (eg. methyl methacrylate) in toluene. Upon the addition of the silica gel or alumina gel to the solution, the catalyst/ligand adsorbs onto the silica or alumina gel. Preferably the support is silica gel.

Once the catalyst/ligand complex is adsorbed onto the silica gel, an initiator may be added to the solution. Many initiators for initiating ATRP reactions are known to those skilled in the art, and any such initiator that has weak carbon-halide bond may be used. For example, the initiator may be methyl-2-bromophenylacetate ("MBP"), ethyl 2-bromoisobutyrate ("EBIB"), carbon tetrachloride (CC1₄), carbon tetrabromide (CBr₄), or methyl 2-bromo- propionate. Once the initiated is added, the mixture is reacted under conditions for polymerization to occur. By "conditions for polymerization to occur" it is meant that, upon addition of the initiator, the mixture is heated sufficiently to commence and maintain the polymerization reaction. It will be understood by those skilled in the art that the specific temperature required to initiate the polymerization depends upon the monomer used, as well as the catalyst and initiator. For the polymerization of methyl methylacrylate using the conditions reported herein, the temperature for commencing and maintaining polymerization is suitably between about 30 °C to about 100 °C, preferably from about 40 °C to about 90 °C.

The adsorbed catalyst/ligand complex may be recovered easily. The silica gel or alumina gel particles supporting the complex may be left to settle to the bottom of the solution. The liquid is removed, and the silica or alumina gel particles are washed with fresh solvent (preferably degassed toluene). The particles are then ready to be used in another polymerization reaction, according to the process described above. The rate of reaction using the silica gel supported catalyst of the present invention is approximately the same as the rate of reaction not using the silica gel supported catalyst. Both reactions exhibit first order kinetics with respect to the monomer, and constant radical concentrations during the polymerization. The molecular weight of the polymer obtained using the CuBr-HMTETA complex increases linearly with the extent of conversion. The resulting polymer has a narrow molecular weight distribution, which increases slightly with the extent of conversion of the monomer. These facts suggest that the CuBr-HMTETA complex mediates the polymerization reaction in a living manner.

The CuBr-HMTETA complex retains a large portion of its initial catalytic activity during subsequent polymerization reactions, on the order of about 80% and 50% of the initial activity during second and third polymerizations.

The first order kinetics of the reaction is not affected by the ratio of silica gel to CuBr-HMTETA complex, although the actual rate of polymerization increases with the ratio of silica gel to CuBr-HMTETA. It is preferable to use relatively low monomer concentration in solution, while maintaining the same ratio of catalyst to monomer.

Preferably, the monomer concentration should be less than about 50 wt%.

More preferably, the monomer concentration should be less than about 33 wt%. Continuous Processes

The polymerization reaction using the catalyst system of the present invention may be completed in batch processes, as above described, or in a continuous process as described below.

Referring to Figure 1, there is shown a schematic for a continuous polymerization system, referred to generally as reference numeral 10. The system 10 has a source of nitrogen, or other inert gas, 20, a reservoir 30, metering pump 40, column reactor 50, and a receiver 60. The nitrogen or other source of inert gas 20, reservoir 30, metering pump 40, column reactor 50, and receiver 60, are all connected within a closed system, which is at all times maintained under an inert atmosphere. The connections may be by any suitable means such as connecting tubing 70. Preferably, the connecting tubing 70 is made from stainless steel.

The column reactor 50 may generally be of any length or diameter. It will be appreciated that the size of the column will be dependent upon the rate at which polymerization solution is to be passed through the column. The polymer molecular weight may be controlled by the mean residence time of the polymerizing materials in the packed column reactor, which is the ratio of the effective volume of the reactor over the volumetric flow rate of the polymerizing materials through the reactor. The aspect ratio of the reactor has influence on the polymer molecular weight distribution and polydispersity. A plug-flow pattern is preferable for narrow distribution polymers. The column 50 is packed, under nitrogen or another inert gas, with a catalyst/ligand complex adsorbed or grafted onto silica or alumina gel particles, which may be prepared as described above in reference to the batch polymerization process. During the polymerization reaction, the column 50 is maintained at the desired temperature (sufficient to initiate and maintain the polymerization reaction). For example, the temperature of the column 50 may be maintained by submersing it in a bath or oven 80 and regulating the temperature of the bath.

The reservoir 30 contains a polymerization solution of monomer (such as MMA), initiator (such as MBP) and solvent (such as toluene). Each of these components is described above in relation to the batch process. Preferably, the ratio of monomer : initiator is between 30:1 and 1000:1. The reservoir 30 is connected directly to the source of nitrogen or other inert gas 20, and a positive pressure of inert gas is maintained in the reservoir at all times.

The reservoir 30 is connected to the metering pump 40, which delivers the contents of the reservoir 30 to the inlet end 70 of the column 50 at the desired flow rate. The flow rate is determined by the desired degree of polymerization and the size of the column. The higher the molecular weight desired, the slower the flow rate (with respect to the same size column), thus allowing the polymerization solution a greater residence time in the column in contact with the catalyst.

After passing through the column 50, the polymer exits the column at outlet 70 and is collected in the receiver 60.

It will be appreciated that the set up shown in Figure 1 may be adapted to produce block copolymers from multiple monomers. In this case, a number of column reactors (packed with a supported catalyst/ligand complex) equal to the number of different monomers to be block copolymerized are connected in series. A representative example of a block polymerization set up having two reactors in series is provided in Figure 2. The first monomer, together with an initiator, are passed through the first column reactor 50 to form a first polymer. Then, the first polymer, together with a second monomer, are passed through a second column reactor 80 (also packed with a supported catalyst/ligand complex), under conditions to form a block copolymer of the first polymer and the second monomer. A solution of a second monomer may be held in a second reservoir 90 attached to a second inert gas source 100 and may be injected into the system at a point downstream of the first reactor 50, at a desired flow rate by metering pump 110. The second monomer and the polymer from the first reactor are allowed to mix and then injected into the second reactor 80. Again, the second inert gas 100, reservoir 90, pump 10 and reactor 80 are all connected using suitable connecting tubing 70 so that they are maintained within a closed system, which is at all times under an inert atmosphere. It will be appreciated that this process may be repeated with any number of monomers and packed column reactors.

Alternately, a single column reactor, packed with the supported catalyst/ligand complex, may be utilized to produce a block copolymer. In this case, the column reactor has inlets at various downstream points. A first monomer and initiator are introduced into a first inlet in the packed column reactor and allowed to come into contact with the catalyst/ligand complex, thus causing the monomer to polymerize. A second monomer is introduced into the column reactor, through a second inlet located downstream of the first inlet, and comes into contact with the catalyst/ligand complex, thus causing the first polymer and second monomer to form a block copolymer. It will be appreciated that any number of monomers may be block copolymerized using this method. Grafted Catalyst /Ligand Systems

In a further aspect of the present invention, the catalyst/ligand complex for use in ATRP polymerization processes (both batch and continuous) may be grafted on to silica gel, alumina or cross-linked polymer particles. Systems where the catalyst/ligand complex are grafted to the support are useful for the ATRP polymerization of high polar monomer such as dimethylaminoethyl methylacrylate. In such systems the catalyst/ligand complex is normally connected to the support by a suitable spacer group.

The supporting spacer, through which the catalyst is attached to the particle, affects the mobility of the catalyst. A long and flexible spacer renders the catalyst high mobility, but spacer having excess length may embrace the catalyst and thus hamper the catalyst mobility. Consequently, the spacer length may have a strong effect on the catalytic activity and its ability to control the ATRP process. The present inventors have developed and prepared a new spacer group in which tetraethyldiethylenetriamine (TEDETA) and di(2-picolyl)amine (DiPA) were grafted onto silica gel particles via polyethylene glycol (PEG) spacers of different lengths for the polymerization of methacrylates to investigate the spacer effect.

The polymerization of methylmethylacrylates was investigated to determine the effect of spacer group length on the ATRP. The spacer between the catalyst and the support affected polymerization rate and the control over molecular weight and polydispersity of resulting polymers during the ATRP process. The optimal spacer comprised about 3 units of ethylene glycol, with which the grafted catalyst had the highest reaction rate and the best control over the molecular weight of PMMA. The recycled catalyst by this support also gave good activity retentions and improved initiator efficiencies. The present invention therefore provides a new catalyst/ligand complex for ATRP polymerization reactions comprising a multidentate amine ligand and copper halide that have been grafted onto silica gel, alumina or cross-linked polymer particles via a spacer group comprising one to ten, preferably three, linear polyethylene glycol moieties (PEG₃). The polyethylene glycol moieties are preferably connected to the support by a triethylenetriamine group. A person skilled in the art would understand that other spacer groups could be devised and prepared that would provide a similar spatial relationship between the ligand /catalyst and support. The catalyst/ligand complex immobilized with PEG₃ effectively mediated MMA polymerization and did not require excessive catalyst (i.e., high CuBr /initiator ratio). Recycled and regenerated catalyst/ligand complexes grafted to a solid support also effectively mediate ATRP reactions. Catalyst/ligand complexes grafted to a solid support can also be used for block co- polymerization reactions as described above for the adsorbed catalyst ligand complexes.

The present invention further provides a process for preparing a multidentate amine ligand system comprising a multidentate amine ligand that is chemically grafted onto a silica gel support with a spacer group having one or more ethylene glycol moieties. Preferably the spacer comprising the ethylene glycol moieties is attached to the silica gel support via a first linker group. Such a first linker group may be, for example, a diethylene triamine group, or any other linear or branched grouping having functional groups capable of participating in a Michael-type reaction. The diethylenetriamine group may be attached to the silica gel using standard procedures, for example as described in Kickelbick, et al. Macromolecules, 32, 2941-2947, 1999. The amino groups (or any equivalent functional group) of the diethylenetriamine-functionalized silica gel may be treated with mono- or polyethyleneglycol diacrylates under standard Michael reaction conditions to provide a silica gel support further functionalized with a mono- or polyethyleneglycol acrylate, which may again be reacted under standard Michael reaction conditions with a multidentate amine compound (the ligand) to provide the multidentate amine ligand system. The number of ethylene glycol moieties within the ligand system is controlled by the selection of the mono- or polyethylene glycol diacrylate in the first Michael reaction. A person skilled in the art would understand that compounds with functional groupings that are equivalent to those described above may be substituted in the above process without deviating from the spirit and scope of the invention.

The invention will be further understood by reference to the following examples which are not to be construed as a limitation on the invention. Those skilled in the art will appreciate that other and further embodiments are obvious and within the spirit and scope of this invention from the teachings of the examples taken with the accompanying specifications.

EXAMPLES Example 1 - Batch Process Methyl Methacrylate ("MMA") from Aldrich was distilled under vacuum and stored at -15°C before use. 1,1,4,7,10,10- Hexamethyltriethylenetetramine ("HMTETA", ligand), CuBr (catalyst), methyl 2-bromophenylacetate ("MBP", initiator) also from Aldrich were used as received. Toluene was distilled over CaH₂. Silica gel of 230-400 mesh having an average pore diameter of 60 A was obtained from SiliCycle Inc.

A typical polymerization process was as follows: the CuBr and silica gel were added to a Schlenk flask. The flask was degassed by five vacuum-nitrogen cycles. The MMA (also degassed), toluene and HMTETA were added to the flask and the mixture was bubbled with nitrogen for 5 minutes. The mixture became blue upon the ligand (HMTETA) addition. The blue silica particles quickly settled down to the bottom of the flask once the stirring stopped. The upper solution layer became colorless. The degassed initiator (MBP) was then introduced to the flask dropwise with stirring. The blue particles turned to green. The flask was then immersed into a 90°C oil bath with stirring. At different time intervals, 0.2-0.5 ml of the solution mixture was withdrawn from the flask with a nitrogen-purged syringe. The mixture was diluted with CDC1₃. The conversion was estimated from the H- NMR intensity ratio of OCH₃ signals from the polymer (3.60 ppm) and monomer (3.75 ppm). Upon completion of the polymerization, the flask was lifted from the oil bath and left still for hours. The clear upper layer solution was carefully removed by cannula under nitrogen atmosphere. The remaining solid in the flask was washed twice with 20 ml degassed toluene under nitrogen. Then the same amount of the degassed methyl methacrylate, toluene, and initiator, as in the first polymerization run, were charged to the flask. The polymerization procedure was repeated.

Nuclear Magnetic Resonance (NMR) Spectroscopy: Proton ( H) NMR spectra were recorded on a Bruker ARX-200 spectrometer at 200 MHz. H NMR chemical shifts in CDC1₃ were reported downfield from 0.00 ppm using trace of CHC1₃ signal at 7.23 ppm as an internal reference.

Molecular Weight Measurements: Number and weight average molecular weights (M_n and M_^,, respectively) were determined by gel permeation chromatography (GPC) using THF-2% (v/v) trimethylamine as solvent at 25°C with an refractive index (RI) detector. Narrow polystyrene standards (Polysciences) were used to generate a calibration curve. Data were recorded and manipulated using the Waters Millennium software package. The adsorption of the copper-HMTETA complex onto silica gel was first tested. When silica gel was added to the copper-HMTETA complex MMA toluene solution, the silica gel became blue immediately. Without stirring, the blue particles settled down to the bottom of the flask quickly. The upper layer toluene solution became colorless, indicating that the catalyst complex was adsorbed onto the silica gel.

Upon adding the initiator MBP, the blue particles immediately turned into green. Heated to 90° C, the mixture gradually became viscous. Figure 3 shows the MMA polymerization results with and without silica gel. The polymerizations with and without silica gel ("SG") (SG/CuBr/HMTETA=2:l:l (w/w/w)( , >) and SG/CuBr/HMTETA=0:l:l (w/w/w) ((A, Δ) proceeded at almost the same polymerization rate. Both the ln([M]o/[M]) vs. time plots were linear, indicating the first order kinetics with respect to monomer and the constant radical concentrations during the polymerization. These observations demonstrated that copper bromide- HMTETA complex supported on silica gel mediated the MMA polymerization in a living manner. The molar ratio of HMTETA/CuBr had a minor effect on the polymerization. Doubling the ligand content slightly accelerated the polymerization.

The dependence of the molecular weight and molecular weight distribution on the conversion is shown in Figure 4. With or without silica gel, the molecular weight distributions of the resulting polymethyl methacrylate ("PMMA") were all very narrow with the polydispersities below 1.1 at the early stage of polymerization and increasing slightly with conversion, but most remaining lower than 1.3. These results were quite different from those previously reported for ruthenium catalysts, in which the polymer polydispersities were much higher. The PMMA molecular weights in Figure 4 increased linearly with the conversion. This clearly demonstrated that the polymerization mediated by silica gel supported CuBr- HMTETA proceeded via a living polymerization process.

The PMMA molecular weights, however, were much higher than the predicted. The calculated initiator efficiencies (Mncalc/MnGPC) were about 0.5 throughout the polymerization. The low initiator efficiencies were also observed in the unsupported polymerization indicating that the low initiator efficiency was not caused by the silica gel support. Increasing the ligand /CuBr ratio to 2 did not improve the initiator efficiency.

The linear plots of ln([M]o/[M]) vs. t in Figure 3 show that the radical concentrations remained constant throughout the polymerization. The low initiator efficiency is believed to be caused by the consumption of some radicals at the very beginning of the polymerization. Lowering the catalyst concentration can decrease the radical concentration so that the side reactions of radicals can be minimized. Figure 5 shows the MMA polymerization with CuBr/MBP = 0.5. The polymerization was slower than that with CuBr/MBP=l. The radical concentration with CuBr/MBP = 0.5, calculated from the slope (kp[R*]) of the ln([M]o/[M]) vs. t plot in Figure 5, was about 0.69 of that of CuBr/MBP = 1. Figure 6 shows the molecular weight and polydispersity vs. conversion data with different copper bromide concentrations. Reducing the CuBr level to a half yielded the molecular weight of PMMA close to the theoretical values. The initiator efficiencies approached to 1. The polydispersities of PMMA were below 1.1 throughout the polymerization.

The MMA polymerizations with silica gel CuBr ratios of 2, 5, and 10 are shown in Figure 7. Changing silica gel content did not change the first order kinetics, however the polymerization rate was doubled as the ratio was increased from 2 to 5. Further increasing the ratio to 10 made no significant difference in the polymerization rate. Figure 8 shows the dependence of molecular weight and polydispersity of PMMA on the MMA conversion with the different silica gel concentrations. The molecular weights of PMMA in all the runs increased linearly with the conversion. With a silica gel: CuBr ratio of 2, the molecular weights were in accordance with theoretical values. But at higher silica gel concentrations (silica gel: CuBr ratios of 5 and 10) the molecular weights were higher than the predicted. The deviations were minor at the early stage of the polymerization, but became pronounced at high conversions. The polydispersities were all lower than 1.15 before the conversions reached 70%. With a silica gel: CuBr ratio of 5 and 10, the polydispersities increased at higher conversions, corresponding to the molecular weight increase. This may be caused by the insufficient deactivation of the radicals by CuBr₂ because of the possible diffusion limitations experienced by the polymer radicals and silica gel supported catalyst complex in the viscous media. In order to illustrate the effect of viscosity on the polymerization, the MMA polymerization mediated by CuBr-HMTETA/2 silica gel was also carried out at a high monomer concentration by decreasing the toluene amount but keeping the monomer /catalyst ratio constant. Figure 9 shows the MMA polymerization at the different monomer concentrations mediated by the CuBr-HMTETA/2 silica gel system. At the MMA concentration of 33% (toluene/MMA = 2 (w/w)), the polymerization proceeded via typical first order kinetics. However, at the 55% MMA concentration (toluene/MMA = l(w/w)), the polymerization followed the first order kinetics only in the early stage of the reaction. The rate increased rapidly when the MMA conversion was higher than 30%, indicating a rapid increase in the radical concentration based on ln([M]o/[M]) = kp[R*]t equation. Correspondingly, the molecular weight deviated from the predicted values, and the polydispersity increased after this point. In contrast, the MMA polymerization at 33% MMA concentration with the same catalyst did not show this auto-acceleration in the polymerization rate, and the molecular weight increased linearly with the conversion with the polydispersities around 1.1 throughout the polymerization (Figure 10). The high viscosity of the polymerization media may be responsible for the autoacceleration in the polymerization rate in Figure 9. When 50% MMA concentration was used, the solution became very viscous beyond 30% conversion. At high conversions, the system became so viscous that the mixture was very difficult to be sampled with a syringe. However, the viscosity of 33% MMA concentration was not high even at a 90% conversion. The diffusion abilities of the polymer chains and silica particles were significantly reduced when the media became highly viscous. Therefore, the deactivation reaction of the polymer radicals by the CuBr₂- HMTETA catalyst was restricted, resulting in less control of the polymerization. This suggests that it is desirable to use lower monomer concentrations.

Following the first polymerization run, the reaction mixture was left overnight to let the catalyst particles settle down to the bottle of the flask. Then the upper layer solution was removed and the remaining green particles were washed with degassed toluene twice. The same amounts of fresh toluene, MMA and initiator as in the first polymerization run were charged to the flask. The system was heated again for the second cycle of polymerization. Figures 11, 12 and 13 show the MMA polymerization with the recycled catalysts. The MMA polymerizations were still in the first order kinetics with respect to monomer. All the recycled catalysts had good retention of their catalytic activities, around 80% of their initial activity values (k_app ratios). The slight reduction in the catalytic activity may be caused by the loss of the supported catalysts during the polymer removing and subsequent washing. A secondary reason may be side reactions of free radicals, such as radical coupling reaction.

After the second use for MMA polymerization, some CuBr- HMTETA/2 silica gel catalyst stuck on the flask surface. But CuBr- HMTETA/5 silica gel and CuBr-HMTETA/ 10 silica gel were still easily separated from the polymer solution. The catalysts settled down to the bottom of the flask. Therefore the catalysts could be re-used for the third time. Figure 13 shows that the catalyst recycled twice still had reasonable activity, about 65% of the second use and 50% of the fresh.

Significantly, the recycled catalysts maintained excellent control over the polymerization as shown in Figures 14 and 15. The molecular weights of PMMA obtained in the second and third uses were much closer to the theoretical values than those in the first run. The M_n values of the resulting PMMA increased linearly with conversion, very similar to the first run. The polydispersities of PMMA were still very low, less than 1.15. These results indicate that the recycled catalysts still effectively mediated the polymerization in the living manner. Example 2 - Continuous Process

Methyl Methacrylate ("MMA") from Aldrich was distilled under vacuum and stored at -15°C before use. 1,1,4,7,10,10-

Hexamethyltriethylenetetramine ("HMTETA", ligand), CuBr (catalyst), methyl 2-bromophenylacetate ("MBP", initiator) also from Aldrich were used as received. Toluene was distilled over CaH₂. Chromatographic grade silica gel of 100-200 mesh was boiled in deionized water for 5 h and dried in air.

12g of the boiled silica gel was weighed into a Schlenk flask and degassed by vacuum-nitrogen for five cycles. 50ml of toluene, 0.6g CuBr and 0.958g of HMTETA were added to the flask under a nitrogen atmosphere. The mixture was bubbled with ultra high purity nitrogen for 10 min with stirring, and then stirred for an additional 3h at room temperature. The mixture was then transferred to a stainless steel column reactor under a nitrogen atmosphere. The column was 900 mm long with a 3.73 mm inner diameter. The column was connected to a metering pump under nitrogen atmosphere.

A monomer reservoir, the metering pump, the column reactor and a receiver were all connected using 1mm inner diameter stainless steel tubing. 100 ml of degassed MMA/MBP/toluene polymerization solution (ratios of MMA /toluene = 1/3 (w/w); MMA/MBP = 100 molar) was added to the reservoir. The column was immersed in a water bath set to the required temperature at 90 °C to maintain polymerization in the column. The metering pump was set to a required flow rate, delivering the polymerization solution to the column reactor. At several different intervals, the resulting polymer from the column collected in the receiver was weighed to calibrate the flow rate. Once the polymerization reaction ran for approximately 2 to 3 times of the residence time in the column, three parallel samples of 0.2 to 0.5 ml solution exiting the column were collected and diluted with CDC1₃.

The conversion of monomer was measured with H-NMR by calculating the intensity ratio of OCH₃ signals in polymer (3.60 ppm) and in monomer (3.75 ppm). The number and weight average molecular weights (M_n and M_w respectively) were determined by gel permeation chromatography (GPC) using THF-2% (v/v) trimethylamine as solvent at 25°C with differential reflective index (DRI) detector. Narrow polystyrene bands were used to generate a calibration curve. The solution eluted from the column was colourless. The reactivity stability of the column reactor was evaluated by measuring the MMA conversion at a set flow rate. The column was run continuously at 1.2 ml/h and the eluting solution was sampled at different intervals. Figure 16 shows that the MMA conversions at different times remained almost the same for more than four days, indicating good retention of catalyst activity in the column. The molecular weights of the produced polymers at different times are shown in Figure 17, and it will be seen that the changes in molecular weight were also minor.

The MMA conversion at 90°C as a function of flow rate is shown in Figure 18. The MMA conversion decreased with increasing flow rate. For example, the MMA conversion reached 87% at a flow rate of 1.2 ml/h, while it decreased to about 23% at 9.6 ml/h. Assuming no back flow within the column, the residence time (τ) of the polymerization solution in the column is the ratio of the free column volume over the volumetric flow rate. Figure 19 shows the dependence of MMA conversion and ln([M]o/[M]) on the residence time with column (τ). The MMA conversion increased smoothly with increasing residence time. The ln([M]o/[M]) v. time plot is linear, indicating that the MMA polymerization in the column reactor was in first order kinetics with respect to the monomer, as is typical for ATRP reactions.

Figure 20 shows the molecular weight and molecular weight distribution dependence on the MMA conversion. The PMMA molecular weight increased linearly with MMA conversion, indicating living polymerization. The deviation of the experimental ML_, from theoretical values was typical when a high silica gel /CuBr ratio system was used. The molecular weight distribution of the PMMA were about 1.5, somewhat larger than that obtained from a batch polymerization system. This increase in molecular weight distribution may be due to back mixing or back flow in the column reactor, and possible trapping of polymer chains in the silica gel pores, resulting in increased column residence time for a portion of the polymerization solution. Example 3 - Continuous Process Applied to Block Copolymerization The continuous block copolymerization of MMA and n-butyl methacrylate (nBMA) was carried out in two columns in series, as shown in Figure 2. The toluene solution of MMA-initiator was pumped to the first column reactor heated to 80 °C. After the polymer solution eluted from the first reactor, the second monomer, nBMA, was injected to the polymer solution and was mixed. The mixture was injected into the second column reactor. The polymer eluting from the second reactor was analyzed by GPC. Figure 21 shows the GPC curves of the polymers. The first curve with Mn of 4000, is the PMMA prepolymer, which was produced without injection of second monomer. With the second monomer flow rate of 0.6 mL/h, the produced polymer had molecular weight of 6400. Increasing nBMA flow rate to 1.2 m/h yielded polymer of Mn=11000. Further increasing nBMA flow rate to 1.86 mL/h yielded polymer of Mn=18000. These results indicate that the two reactor in series setting can produce block copolymer continuously and the block length can be controlled by flow rate. Example 4 - Grafting Supported Catalysts for Polymerization (i) CuBr-TEDETA/Silica Gel

Tetraethyldiethylene triamine (TEDETA) was grafted onto silica gel as shown in Figure 22. TEDETA was first coupled with trimethoxysilylpropyl acrylate via a Michael reaction to synthesize N,N,N'N'- tetraethyl-N"-[3-(trimethoxysilylpropoxycarbonyl)ethyl]-diethylenetriamine. 20 g of 3-(trimethoxysilyl propyl) acrylate was charged to a flask and cooled to 0 °C. 22.6 g of TEDETA was added drop wisely to the flask with stirring for 10 h. The mixture was further stirred at room temperature until no acrylate signal was detectable by NMR. The liquid was then subject to high vacuum to remove possible volatile species. A viscous liquid was finally obtained. ¹ - NMR: 3.80 ppm (t, 2H, COOCH₂); 3.35 ppm (s, 9H, SiOCH₃); 2.60 ppm (t, 2H, CH₂COO); 2.30 ppm (m, 18H, NCH₂); 1.50 ppm (m, 2H, SiCH₂CH₂); 0.80 ppm(t, 12H, NCH₂CH₃).

The attachment of TEDETA ligand onto silica gel surface was via a reaction of silicon alkoxide with silanol group on the particle surface. Hydrophilic silica gel was dried in vacuum at 100 °C for 3 days. 10 g of the dried silica gel, 5 g of synthesized product (1) and 50 ml THF were charged to a flask. The mixture was refluxed for 48 h. Silica gel was separated from the solution by centrifugation and washed 6 times with THF. The silica gel was finally dried at 50 °C under vacuum for 24h. The amount of ligand grafted onto silica gel was determined by thermogravimetric analysis (TGA). The calculated TEDETA concentration was 14.87%(w/w). Figure 23 shows the MMA polymerization catalyzed by the fresh and recycled SG-TEDETA system. The polymerizations with the fresh and recycled catalysts were also in a first order. The activity of the first recycled catalyst was reduced to about 50% of the fresh one. However, the activity of the second recycled catalyst decreased only slightly. Figure 24 shows the molecular weight and polydispersity of PMMA as a function conversion in the MMA polymerizations with the fresh and recycled catalysts. In all the cases, the molecular weights were well controlled and increased linearly with conversion. The molecular weights of PMMA obtained in the second and third runs were slightly higher than those in the first run. These results indicate that the recycled catalysts still effectively mediated the polymerization. (ii) Catalyst regeneration - CuBr-TEDETA /Silica Gel

Cu(0) was used to reduce Cu(II) to Cu(I) to regenerate the recycled catalyst. Copper turnings were stirred with the catalyst at 35 °C for overnight. The green-colored recycled catalyst became blue again. This regenerated catalyst was used for the MMA polymerization. As shown in Figure 25, the regenerated recycled catalyst showed higher activity than the recycled catalyst without regeneration, but its activity was still lower than the fresh catalyst. This results confirmed the presence of Cu(II) in the recycled catalyst. The regenerated catalyst still mediated a first order polymerization with respect to monomer. The molecular weights of PMMA from the regenerated catalyst were very close to the predicted with low polydispersities (Figure 26), similar to those from the fresh catalyst.

It is important to completely separate copper metal from the catalyst after regeneration. In the presence of copper metal, the regenerated catalyst had the same activity as the fresh one in the early stage of polymerization, but leveled off later. This catalyst had no control over polymerization. The PMMA molecular weight decreased with conversion and had a high value of polydispersity (Mw/Mn >1.7), typical of conventional free radical polymerization. It is preferred to use copper turnings rather than copper powder for regernation of the catalyst. The latter is difficult to be separated from the catalyst.

(iii) Effect of distance between ligand /catalyst and support on polymerization reaction

TEDETA or DiPA were grafted onto silica gel via polyethylene glycol spacers by two Michael reactions of acrylates with primary or secondary amines. Diethylenetriamine was attached onto the silica gel surface first by the reaction of silicon alkoxides with silanol groups on the particle surface (Figure 27). Hydrophilic silica gel was dried in vacuum at 60 °C for 3 days. lOg of the dried silica gel, 5g of N [3- (trimethoxylsilyl)propyl] diethylenetriamine and 50 ml THF were charged to a flask. The mixture was refluxed for 48 h. The silica gel was separated from the solution by centrifugation and washed 6 times with THF. It was finally dried at 50 °C under vacuum for 24h. IR: 3433 cm^"1, 2975 cm^"1, 1733 cm^"1, 1567 cm^"1, 1032 cm^"1. The amount of diethylenetriamine grafted onto silica gel was determined by thermogravimetric analysis (TGA) and elemental analysis. The calculated diethylenetriamine concentration was 4.24 %(w/w).

TEDETA or DiPA grafted onto silica gel via polyethylene glycol spacers is shown in Figure 28. lOg of diethylenetriamine-functionalized silica gel was gradually added to 30 g of polyethylene glycol diacrylates with sufficient stirring. The mixture was further stirred for 48 h at room temperature. It was then diluted with THF to 100 ml and centrifuged. The silica gel was washed with 60 ml THF 6 times to remove unreacted PEG. About 0.2 g silica gel was taken out and dried in vacuum for analysis. The remaining silica gel was stirred with 20 ml TEDETA for 48 h at room temperature. The silica gel was then separated by centrifugation and washed with 7 times with THF. Finally the silica gel was air-dried and then vacuum- dried at 40 °C for 24 h. DiPA functionalized silica gel supports were prepared by the same procedure. FTIR for silica gel grafted with TEDETA: 2970 cm^"1, 2808 cm^"1, 1738 cm^"1 (C=0), 1202 cm^"1, 1109 cm^"1; for silica gel grafted with DiPA: 2926 cm^"1, 1735 cm^"1 (C=0), 1596 cm ¹, 1109 cm ¹ (Si-O-Si), 797 cm^"1. The ligand contents were analyzed by thermogravimetric analysis (TGA) and elemental analysis. Figures 29 and 30 show the MMA polymerization mediated with CuBr supported on TEDETA or DiPA-modified silica gel via different PEG spacer lengths. The MMA polymerization rates are summarized in Table 1. The apparent polymerization rate constant k^app (i.e., k_p[R^*]) increased with the spacer length in the following order: PEG_X< PEG₁₀ < PEG₃. This suggests that when CuBr is immediately anchored on the silica gel surface, the catalyst has very limited mobility and therefore less chance to react with the dormant centers (P-Br). However, when the spacer is too long, e.g. with 10 PEG units, the PEG coil surrounding the catalytic site impedes the reaction of the catalyst with dormant active center (P-Br). It becomes clear that the length of 3 ethylene glycol units is optimal for immobilizing CuBr.

The polymerizations in Figures 29 and 30 experienced initial increases in rate before the ln([M]₀/[M]) versus time curves became linear. This suggests a decrease in the radical concentration in the early stage of polymerization based on ln([M]₀/[M])= kp[R"]t. This decrease may be caused by radical termination reactions. At the early stage of the polymerization, the low Cu(II) concentration in the solution favored the forward reaction and thus resulted in a high radical concentration. Consequently radical side reactions such as termination became significant. These side reactions consumed radicals and generated excess Cu(II). When a certain level of Cu(II) concentration was accumulated and the backward reaction reached an equilibrium with the forward reaction, the reaction proceeded smoothly in a constant radical concentration (i.e., a first order reaction). Figures 31 and 32 shows the number average molecular weight

(Mn) and polydispersity (Mw/Mn) of the resulting PMMA. The molecular weights of PMMA increased linearly with the conversion, indicating a living process in all cases. The spacer length of the catalyst support affected the initiator efficiency (Mn,theoretical/Mn,experimental) and polydispersity of PMMA. The initiator efficiency was 57% with SG-PEG_α-TEDETA, 75% with SG-PEG₃-TEDETA, and 70% with SG-PEG₁₀-TEDETA. The polydispersity of PMMA was in the range of 1.4-1.6 by SG-PEG_α-TEDETA, 1.2-1.4 by SG-PEG₃- TEDETA and 1.4-1.5 by SG-PEG₁₀-TEDETA. A similar trend was observed when DiPA was used as a ligand (Figure 32). These results indicate CuBr immobilized with 3 units of ethylene glycol regulated the polymerization most effectively. A spacer length longer or shorter than 3 PEG units may slow the deactivation reaction of Cu(II) with an active center (P^*). This observation was in agreement with the kinetic data in Figures 29 and 30.

MMA was also polymerized at 80 °C using three different combinations of ligand and spacer for CuBr (SG-PEG₃-TEDETA, SG-PEG DiPA, SG-PEG₃-DiP A) (Figure 33). The polymerization rates were higher than those at 60 °C. However, the spacer effects on the polymerization rate and molecular weight control were not as significant. The polymerizations had almost the same rates below 60 % conversion (Figure 33). The polymerization with CuBr/DiPA-PEG_j-DiPA slowed down rapidly after the 60% conversion, whereas that with SG-PEG₃-TEDETA remained almost a constant rate. Increase in viscosity at high conversions may cause the reduction in rate. A high viscosity of the solution limits the diffusion of the catalytic sites. This is particularly true for a catalyst immobilized on particles with a short spacer length.

The molecular weight developments of PMMA polymerized at 80 °C (Figure 34) were similar to those obtained at 60 °C. However, the polydispersities of PMMA at 80 °C were generally higher than those at 60 °C. The polydispersity in each case decreased gradually with the conversion and reached a minimal value at 60%, prior to its increase at high conversions.

Figures 23-26 demonstrated that it was preferable to use high CuBr /initiator ratios for supported catalysts with a short spacer (-CH₂-CH₂- CH₂- was used as the spacer), typically CuBr /initiator = 1.5, in order to achieve a controlled polymerization. However, it was found that the CuBr /initiator ratio could be lowered with long spacers. Figure 35 shows the MMA polymerization using CuBr /initiator = 1 with SG-PEG₃-TEDETA. The kinetic plots were similar to those of CuBr /initiator = 1.5 with slightly lower polymerization rates. Figure 36 shows the molecular weight and polydispersity data. With CuBr/MBP=l, the molecular progress was improved and polydispersites were even lower than those with CuBr/initiator=1.5. These comparisons elucidated that the catalyst immobilized with PEG₃ effectively mediated the MMA polymerization and did not require excessive catalyst (i.e., high CuBr /initiator ratio), (iv) Reuse of grafted catalyst

One of the major advanatages of using supported catalysts is the potential for reusing the catalyst. A series of catalyst recycling experiments were therefore carried out with SG-PEG TEDETA and SG-PEG₃-TEDETA as the support. The results are shown in Figures 37 and 38. MMA was first polymerized using silica gel supported TEDETA with CuBr /TEDETA = 1. Upon completion, the reaction mixture was centrifuged and the solution was removed with cannula under nitrogen. The catalyst was washed three times with degassed phenyl ether. The same amounts of degassed solvent, MMA and initiator as in the first run were recharged to the recycled catalyst for a second polymerization cycle at a 60 °C oil bath. Unlike the polymerization with a fresh catalyst, the recycled catalysts gave rather smooth curves without the initial jump in rate. The ln([M]₀/[M])~t plots with the recycled catalysts were linear throughout the polymerization, parallel to those with the fresh catalysts, indicating that the recycled catalysts had the same catalytic activity as the fresh ones in the late stage of polymerization.

Figure 39 shows the molecular weight and polydispersity of PMMA as a function of conversion with the fresh and recycled catalysts. In all cases, the molecular weights increased linearly with conversion. Most significantly, the molecular weights of PMMA prepared by the recycled catalysts were much closer to their theoretical values, i.e. higher initiator efficiencies. The initiator efficiency was ca. 100% for the recycled CuBr/SG-PEG₃-TEDETA and 78% for the recycled CuBr/SG-PEG_j-TEDETA, compared to 75% and 57% for their corresponding fresh catalysts. These results indicated that the recycled catalysts had better ability to regulate the chain growth. The improvement in the initiator efficiency was resulted from the presence of Cu(II) in the recycled catalysts, which suppressed the radical termination by lowering the radical concentration.

Example 5 - Block copolymerization of MMA with 2-dimethylaminoethyl methacrylate (DMAEMA)

The ability of the silica gel supported CuBr to mediate a block copolymerization was examined using MMA and DMAEMA by a re-initiation method. The DMAEMA polymerization by CuBr/SG-PEG₃-TEDETA was investigated first. Table 2 shows the DMAEMA polymerization with the CuBr/SG-PEG₃-TEDETA support. Similar to the MMA polymerization, the DMAEMA polymerization was also a first order reaction, but was much faster than the MMA polymerization. The molecular weights of poly(DMAEMA) were very closer to the predicted with low polydispersities, 1.2-1.4.

MMA (MMA/MBP=80) was first polymerized using CuBr/SG- PEG₃-TEDETA for 6 hours, yielding macro-initiator of PMMA with Mn=6937 and Mw/Mn= 1.22 (Mn theor = 6000) with 75 % conversion. The reaction was stopped and the unreacted MMA was removed by vacuum. DMAEMA was then added and the mixture was reheated to 60 °C for 8 hours. A block copolymer of P(MMA-b-DMAEMA) was isolated with 70% DMAEMA conversion. The resulting polymer had Mn of 14664 (Mn theor = 15729) and polydispersity of 1.46 without contamination of PMMA prepolymer (Figure 0). This result demonstrated that the silica gel supported CuBr could be used r the block copolymerization.

Table 1.

The apparent rate constant k^app of MMA polymerization catalyzed by CuBr immobilized on silica gel with different spacer lengths. ' ^'

SG-PEG₃-TEDETA 3.8

SG-PEG₁₀-TEDETA 3.2

SG-PEG DiPA 1.6

SG-PEG₃-DiPA 3.0

SG-PEG₁₀-DiPA 2.7

[MMA]=1.38 mol/L, [MBP]=9.2 xlO^'3 mol/L, [CuBr]= 1.39xl0^"2 mol/L, [TEDETA] or [DiPA] /[CuBr] (molar) = 1, 60 °C; Solvent: phenyl ether.

The subscript number is the ethylene glycol unit.

Table 2. DMAEMA polymerization catalyzed by CuBr supported on SG-PEG₃

TEDETA ^a

Time (min) Conv. Ln([M]₀/[M]) Mn(GPC) Mn(Theor.) Mw/Mrt

20 0 .27 0.31 8369 5525 1.21

50 0 .44 0.58 10892 9016 1.24

80 0.56 0.82 13711 11402 1.23

110 0.69 1.16 15474 14027 1.31

170 0.75 1.37 16576 15209 1.28

^a [DMAEMA]= 1.38 mol/L, [MBP]=9.2 xlO^"3 mol/L, [CuBr]= 1.39xl0^"2 mol/L, [TEDETA] /[CuBr] (molar) = 1, 60 °C; Solvent: phenyl ether.

Claims

WHAT IS CLAIMED IS:

1. A catalyst/ligand system for use in atom transfer radical polymerization, comprising a catalyst/ligand complex associated with a support.

2. The catalyst/ligand system as claimed in claim 1, wherein the catalyst/ligand complex is associated with the support by adsorption or chemical grafting.

3. The catalyst/ligand system as claimed in claim 2, wherein the catalyst/ligand complex is associated with a support selected from the group consisting of silica gel, alumina gel, crosslinked polystyrene particles, and mixtures thereof.

4. The catalyst/ligand system as claimed in claim 3, wherein the catalyst/ligand complex comprises a metal halide /multidentate amine ligand.

5. The catalyst/ligand system as claimed in claim 4, wherein the catalyst comprises a copper (I) halide.

6. The catalyst/ligand system as claimed in claim 5, wherein the catalyst comprises copper bromide.

7. The catalyst/ligand system as claimed in claim 4, wherein the multidentate amine ligand comprises at least one compound selected from the group consisting of hexamethyltriethylenetetramine, tetraethyldiethylenetriamine, polyethyleneimine, di(2-picolyl)amine, their respective derivatives, and mixtures thereof.

8. The catalyst ligand system as claimed in any one of claims 1-7, wherein the catalyst/ligand complex is associated with the support by adsorption.

9. The catalyst ligand system according to claim 8, wherein the support is selected from the group consisting of silica gel and alumina gel.

10. The catalyst/ligand system as claimed in claim 2, wherein the catalyst/ligand complex is chemically grafted onto the support via a spacer group.

11. The catalyst/ligand system as claimed in claim 10, wherein the spacer group comprises 1-10 ethylene glycol moieties, or another chemical moiety providing a similar spatial relationship between the catalyst/ligand complex and the support.

12. The catalyst/ligand system as claimed in claim 11, wherein the spacer group comprises 3 ethylene glycol moieties, or another chemical moiety providing a similar spatial relationship between the catalyst/ligand complex and the support.

13. The catalyst/ligand system as claimed in claim 12, wherein the spacer group further comprises a diethylenetriamine moiety.

14. The catalyst/ligand system as claimed in claim 11, wherein the ligand comprises at least one compound selected from the group consisting of hexamethyltriethylenetetramine, tetraethyldiethylenetriamine, polyethyleneimine, di(2-picolyl)amine, their respective derivatives, and mixtures thereof.

15. The catalyst/ligand system as claimed in claim 14, wherein the catalyst is a copper (I) halide.

16. A method of atom transfer radical polymerization comprising the steps of:

(a) preparing a catalyst/ligand complex associated with a support; (b) combining the supported catalyst/ligand complex with a monomer to be polymerized; and

(c) adding an initiator to the combined supported catalyst/ligand complex and monomer and allowing the initiator, monomer and supported catalyst/ligand complex to react under conditions sufficient for polymerization to occur.

17. The method as claimed in claim 16, wherein the catalyst/ligand complex is associated with the support by adsorption or chemical grafting.

18. The method as claimed in claim 17 wherein the support is selected from the group consisting of silica gel, alumina gel, polymer particles, and mixtures thereof.

19. The method as claimed in claim 18, wherein the catalyst comprises a metal halide, the ligand comprises an amine compound, and the initiator comprises a compound having a weak carbon-halide bond.

20. The method as claimed in claim 19, wherein:

(a) the catalyst comprises copper bromide;

(b) the ligand comprises at least one compound selected from the group consisting of hexamethyltriethylenetetramine, tetraethyldiethylenetriamine, polyethyleneimine, di(2- picolyl)amine, their respective derivatives, and mixtures thereof.

(c) the monomer comprises at least one compound selected from the group consisting of methacrylates, acrylates, styrenics, and mixtures thereof; and

(d) the initiator comprises methyl 2-bromophenylacetate.

21. The method as claimed in claim 16, wherein following the polymerization reaction, the catalyst/ligand complex is recovered for re-use in subsequent polymerization reactions.

22. The method as claimed in claim 16, wherein the catalyst/ligand complex is associated with the support by adsorption and the support comprises silica gel or alumina gel.

23. A continuous process for atom transfer radical polymerization comprising the steps of:

(a) providing a packed reactor column comprising a catalyst/ligand complex associated with a support; (b) passing solution of a monomer to be polymerized and an initiator through the packed column reactor, allowing the monomer and initiator to come into contact with the catalyst/ligand complex under conditions for polymerization to occur; and (c) recovering polymer as it exits the column reactor.

24. The process as claimed in claim 23, wherein the catalyst ligand support is associated with the support by adsorption or chemical grafting.

25. The process as claimed in claim 24, wherein the support is selected from the group consisting of silica gel, alumina gel, cross-linked polymer particles, and mixtures thereof.

26. The process as claimed in claim 25, wherein the catalyst/ligand complex comprises a metal halide/amine complex.

27. The process as claimed in claim 26, wherein the catalyst/ligand complex comprises a copper bromide-multidentate amine complex and the initiator comprises a compound having a weak carbon-halide bond.

28. The process as claimed in claim 27, wherein the initiator comprises methyl 2-bromophenylacetate and the monomer comprises at least one monomer selected from the group consisting of methacrylate, acrylate and styrenic monomers and mixtures thereof.

29. The process as claimed in claim 23, wherein the catalyst/ligand complex is associated with the support by adsorption and the support is silica gel or alumina gel.

30. A method of atom transfer radical polymerization to prepare block copolymers comprising the steps of: (a) providing at least one packed column reactor comprising a catalyst/ligand complex associated with a support, the reactor columns being connected in series where multiple columns are packed; (b) passing a first monomer to be polymerized, and an initiator, through a first packed column reactor, allowing the monomer and initiator to come into contact with the catalyst/ligand complex under conditions for polymerization of the first monomer to form a first polymer; (c) passing a further monomer, together with the first polymer, through a packed column reactor downstream of the first monomer, allowing the further monomer and first polymer to come into contact with the catalyst/ligand complex under conditions for polymerization of the further monomer and the first polymer to form a block copolymer, and

(d) collecting the block copolymer.

31. The method as claimed in claim 30, wherein after step (c), a second further monomer to be polymerized is passed through a packed column reactor, together with the block copolymer, downstream of the further monomer, allowing the second further monomer and block copolymer to come into contact with the catalyst/ligand system under conditions for polymerization of the second further monomer and the block copolymer to form a second block copolymer.

32. The method as claimed in claim 30, wherein the support is selected from the group consisting of silica gel, alumina gel, polymer particles, and mixtures thereof.

33. The method as claimed in claim 30, wherein the catalyst comprises a metal halide, the ligand comprises an amine compound, and the initiator comprises a compound having a weak carbon-halide bond.

34. The method as claimed in claim 33, wherein:

(a) the catalyst comprises copper bromide;

(b) the ligand comprises at least one compound selected from the group consisting of hexamethyltriethylenetetramine, tetraethyldiethylenetriamine, polyethyleneimine, di(2- picolyl) amine, their respective derivatives, and mixtures thereof.

(c) the monomer comprises at least one compound selected from the group consisting of methacrylates, acrylates, styrenics, and mixtures thereof; and (d) ) the initiator comprises methyl 2-bromophenylacetate.

35. The method as claimed in claim 30, wherein the catalyst/ligand complex is associated with the support by adsorption and the support is silica gel or alumina gel.

36. A process for preparing multidentate amine ligand system comprising a multidentate amine ligand that is grafted onto a silica gel support with a spacer group having one or more ethylene glycol moieties comprising the steps of:

(a) reacting silica gel with a linear or branched amino compound having a suitable leaving group to provide silica gel functionalized with a first linker group having one or more primary or secondary amines;

(b) reacting the one or more primary or secondary amines of the first linker group with a mono- or polyethylene glycol diacrylate to provide silica gel functionalized with a first linker group and a mono- or polyethylene glycol acrylate; and

(c) reacting the silica gel functionalized with a first linker group and a mono- or polyethylene glycol acrylate with a multidentate amine compound to provide the multidentate amine ligand system.