Polymer network material comprising a poly(glycidyl ether) structure, method of its production and use

Abstract

The present invention is related to novel preparative methods to a novel class of polymer network materials with a highly branched poly(glycidyl ether) (PGE) structure. Said polymer networks are prepared by a simple procedure involving ring-opening polymerisation and the method is applicable to a wide range of glycidyl ether containing monomers. The method comprises the step of copolymerising (A) at least one multi-topic glycidyl ether comprising at least three glycidyl ether groups with (B) at least one glycidyl ether component comprising at least one glycidyl ether group by ring opening polymerisation, wherein the multi-topic glycidyl ether (A) is glycerol glycidyl ether (GGE) having the Formula (I) and the glycidyl ether component (B) is selected from monoglycidyl ethers comprising one glycidyl ether group and diglycidyl ethers comprising two glycidyl ether groups. ##STR00001##

Claims

1. A method of preparing a polymer network material comprising a poly(glycidyl ether) structure, the method comprising the step of copolymerising (A) at least one multi-topic glycidyl ether comprising glycerol glycidyl ether (GGE) having the Formula I ##STR00009## with (B) at least one glycidyl ether component, including at least one of general Formula II and general Formula IIc: ##STR00010## wherein R.sup.2 is a branched or unbranched C1-C10 alkyl or alkylene, a branched or unbranched C2-C10 alkenyl or alkenylene, a branched or unbranched C2-C10 alkinyl or alkinylene, a branched or unbranched C1-C10 alkyl ether or alkylene ether, a branched or unbranched C2-C10 alkenyl ether or alkenylene ether, a branched or unbranched C2-C10 alkinyl ether or alkinylene ether, and m is 1 or 2, or R.sup.2 is a polyether having the structure according to general Formula III ##STR00011## wherein R.sup.3 may be a branched or unbranched C1-C10 alkylene, a branched or unbranched C2-C10 alkenylene, a branched or unbranched C2-C10 alkinylene and x is an integer in the range of from 10 to 1,000, and m is 1; ##STR00012## wherein R.sup.4 is hydrogen or methyl and x is an integer in the range of from 10 to 1,000, wherein the copolymerisation step is followed by a step of functionalizing the polymer network material by covalently binding functional groups to the polymer network material.

2. The method according to claim 1, wherein the glycidyl ether component (B) is a monoglycidyl ether having the structure according general Formula IIa ##STR00013## wherein R.sup.2 is methyl, ethyl, ethene, n-propyl, isopropyl, propenyl (allyl), n-butyl, iso-butyl, tert-isobutyl, n-pentyl, iso-pentyl, or phenyl.

3. The method according to claim 1, wherein the polyether group in Formula III has an average number molecular weight in the range of 100 to 50,000 g/mol.

4. The method according to claim 1, wherein the copolymerisation is conducted in the presence of particles being essentially insoluble in the reaction mixture, and the copolymerisation step is followed by a step of extracting the particles by dissolving them with a solvent giving rise to a porous scaffold of the polymer network material having a pore size being defined by the particle size.

5. A polymer network material comprising a poly(glycidyl ether) structure obtained by the method according to claim 1.

6. The method according to claim 1, wherein the step of copolymerising comprises ring-opening polymerisation.

7. A medical or biomedical article prepared from the polymer network material of claim 5.

8. The method according to claim 6, wherein the ring-opening polymerisation is cationically initiated.

9. The method according to claim 8, wherein the ring-opening polymerisation is cationically initiated by using a photoinitiator, a Lewis acid or a Bronsted acid.

Description

DETAILED DESCRIPTION OF THE INVENTION

(1) The present invention will hereafter be described in more detail referring to the following figures:

(2) FIG. 1: Multitopic glycidyl ethers which have been employed as comonomers testing the synthetic method to produce homogeneous homo-, co-, and terpolymer network materials.

(3) FIG. 2: Possible multitopic glycidyl ethers that could be incorporated by extending the preparative method described.

(4) FIG. 3: Reaction scheme of the synthesis of UV-crosslinked PGEs: (A) Photolysis of the photoinitiator diphenyl hexafluorophosphate (B) Structure of the crosslinker GGE showing protonated mode (C, D) Exemplary reaction for PGE formation by the monomer GGE. (E) Transparent film product.

(5) FIG. 4: General schematic overview on the process of attaching a range of synthetic functionalities to the polymer networks via use of AGE as a heterobifunctional linker.

(6) FIG. 5: Evolution of rheological storage and loss moduli (G and G respectively) during the polymerisation and hardening of GGE with varying initiator amount of 2.0, 1.0, and 0.1 wt. % under constant UV irradiation at 20 C.

(7) FIG. 6: Exemplary FT-IR spectra of a polyGGE network showing decrease of the epoxy ring absorption at 3056, 2998, 907, 839 and 756 cm.sup.1 depending on irradiation time intervals of 0, 2, 5 and 10 mins.

(8) FIG. 7: Structure-property relationship between: A) degree of swelling (Q) and network chain segment length (M.sub.n); and B) mole fraction of crosslinker GGE and resultant M.sub.n; of GGE/XGE based copolymer networks.

(9) FIG. 8: Structure-property relationship between mole fraction of GGE crosslinker and A) elastic modulus (E); B) glass transition temperature (T.sub.g), for GGE/XGE based copolymer networks.

(10) FIG. 9: Dynamic mechanical characterisation at varied temperature (DMTA) of GGE/XGE based copolymer networks: loss modulus E (A),(B),(C),(D); tan (E),(F),(G),(H), for MGE, EGE, IGE and BGE incorporated networks respectively, with 0, 20, 40, 60 wt. % of alkyl glycidyl ether incorporated for each series (nomenclature corresponds to Table 1).

(11) FIG. 10: Influence of the chemical composition on glass transition temperature, T.sub.g, of the GGE/PEGDE copolyether networks. Polymer networks are the homopolymer GGE (.square-solid.), or copolymers with: PEG.sub.(526)DE (.circle-solid.), PEG.sub.(2000)DE (.box-tangle-solidup.), or PEG.sub.(6000)DPE (.Math.). Dashed line in FIG. 1c indicates T.sub.g values predicted for PEG.sub.(526)DE copolymer networks based on the Fox equation.

(12) FIG. 11: Influence of GGE weight fraction in the comonomer feed on the Youngs elastic modulus of the GGE/PEGDE copolyether networks. Polymer networks are the homopolymer GGE (.square-solid.), or copolymers with: PEG.sub.(526)DE (.circle-solid.), PEG.sub.(6000)DE (.box-tangle-solidup.), or PEG.sub.(6000)DE (.Math.).

(13) FIG. 12: Dynamic mechanical characterisation at varied temperature of copolyether networks of GGE/PEGDE: loss modulus E (a-c); tan (d-f), for PEG.sub.(526)DE, PEG.sub.(2000)DE, and PEG.sub.(6000)DE based networks respectively, at varied wt. % incorporation. Copolymer networks contain PEGDE by wt.: 0% (-), 20% (- - - -), 40% ( . . . . ) 60% (- . - . -), 80% (- . . - . . ).

(14) FIG. 13: Viability with rat Mesenchymal Stem Cells after exposure to extracts taken from films of a homopolymer of GGE at various concentrations and extraction times.

(15) FIG. 14: LDH release in the extracellular fluid (A) and mitochondrial activity of the cells (B) after 48 h of culturing L929 Fibroblasts with pure cell culture medium (control) and with 72 h-extract of the sample; meansstandard deviation, n=6 (LDH), n=8 (MTS).

(16) The present invention relates to a method of preparing a polymer network material comprising a poly(glycidyl ether) structure and the PGEs obtainable by this method. The method comprises the step of copolymerising (A) at least one multi-topic glycidyl ether comprising at least three glycidyl ether groups with (B) at least one glycidyl ether component comprising at least one glycidyl ether group by ring opening polymerisation.

(17) FIG. 1 shows certain examples of particular multi-topic glycidyl ethers (A) and glycidyl ether components (B) which have been tested (see Examples): glycerol triglycidyl ether GGE and other derivatives wherein glycdiyl ethers are replaced by an alcohol group, poly(ethylene glycol) diglycidyl ethers PEGDE with average molecular weights of 525, 2000, and 6000 respectively, alkyl monoglycidyl ethers XGE with varying side groups X; methyl glycidyl ether MGE, ethyl glycidyl ether EGE, isopropyl glycidyl ether IGE, n-butyl glycidyl ether BGE, and allyl glycidyl ether AGE.

(18) However, the approach can be extended to any multitopic glycidyl ethers, for instance as shown in FIG. 2, particularly: monoglycidyl ethers with variable saturated and unsaturated alkyl side chains XGE, Phenyl glycidyl ether PhGE and other aryl (phenolic) glycidyl ethers, PEG digylcidyl ethers of any average molecular weight PEGDE, monomethyl PEG glycidyl ethers of any average molecular weight mPEGGE, poly(propylene glycol) diglycidyl ethers of any average molecular weight PPGDE and monomethyl poly(propylene glycol) glycidyl ethers of any average molecular weight mPPGGE, star shaped PEG constructs with multiply functionalised glycidyl ether end groups, linear polyglycerol diglycidyl ethers lin-PGDE with varying side chains and of any average molecular weight, hyperbranched polyglycerol glycidyl ethers hPGGE with hyperbranched core of any average molecular weight and with multiply functionalised gylcidyl ether end groups.

(19) In the following typical procedures are described.

(20) Polymerisation can be initiated by UV light irradiation by use of a photoinitiator (see Protocol a) below), or can be directly polymerised in bulk by addition of a suitable Lewis or strong Bronsted acid (Protocol b).

(21) All educts are liquid and fully miscible, for higher Mw PEG units it may be necessary to melt them so reaction is performed above the respective melting temperature. Purification is conducted by soxhlet extraction with ethanol (or other polar solvents); gel contents are typically above 90%. Products are formed as colourless, transparent films of varying thermal and mechanical properties. All analyses indicate homogeneous polymer network formation.

(22) Protocol a) Photoinitiated cationic ring opening polymerisation, typical procedure: Synthesis of the polymer networks is performed by a one-step procedure, which is based on photopolymerisation of the initial glycidyl ether monomer mixtures, in which the photoinitiator is dissolved. The initial mixture is poured into a mould and photopolymerisation is performed by irradiation of the mixture with a light source of a suitable wavelength. The reaction of this initial mixture, leads to a transparent film formation. The film products may then be left at dark for postcuring for several hours or days. The photocatalyst employed was diphenyliodonium hexafluorophosphate, although any derivative of such photocatalyst systems or thermolatent initiators could also be used in principle (Prog. Polym. Sci., 1998, 23, 1485-1538).

(23) The reaction scheme of the photoinitiated cationic ring opening polymerisation is shown in FIG. 3. In this example, GGE as multi-topic glycidyl ether monomer was homopolymerized. The irradiation of the monomer-initiator mixture leads to photolysis of the photoinitiator forming the superacid HPF.sub.6 (FIG. 3A). Polymerization occurs via the typical steps for cationic ring opening, however the kinetics of reaction are unusual and show strong tendencies towards frontal polymerization. During propagation, the acid-complexed, cationic epoxide intermediate is stabilised by the neighboring gylcidyl ether (FIG. 3B), which causes reactions to proceed sluggishly until exothermic reaction becomes autoaccelerated above a thermal threshold (FIGS. 3C,D). Theoretically the resulting carbocation can continue to propagate until all epoxide rings open, however steric and mobility restrictions makes this unlikely. A carbocation at any stage can be terminated by the presence of a nucleophilic impurity such as water. The resultant polymer is a densely crosslinked network structurally related to dendritic hPG polymers formed by related methods. The first step in the reaction involving the acid generation is the only step that is light-driven. The remaining steps are dark steps, which proceed under the driving force of the relief of epoxy ring strain. In such polymerizations, post curing is a crucial step, which means a long-lasting reaction after UV-irradiation, with propagation proceeding in the absence of further UV initiation. The product is transparent as shown in FIG. 3E.

(24) Protocol b) With Lewis Acids/Bronsted Acids, Typical Procedure:

(25) The cationic polymerisation can be initiated by addition of Lewis acids or strong Bronsted acids. To a prepared mixture of glycidyl ether containing monomers the Lewis or Bronsted acid is added. For instance, 1 wt. % of acid diluted in 1:1 weight ratio with diethyl ether may be added. Polymerisation occurs by vigorous reaction and the resultant films are left at room temperature, for example for 1 h. The film products are then left postcuring for several hours or days at elevated temperatures, for example for 4 days in an oven at 60 C. Preferably, BF.sub.3.O(Et).sub.2 is employed as an example of a Lewis acid; trifluoromethanesulfonic is employed as an example of a protic Bronsted acid. However, any suitable acid catalyst can be employed with particular examples being trifluoroacetic acid, para-toluenesulfonic acid, TiCl.sub.4, AlCl.sub.3, SnCl.sub.4, LiPF.sub.6 and LiClO.sub.4.

(26) Protocol c) Production of Macroporous Scaffolds by Salt Leaching, Typical Procedure:

(27) Particles of inorganic salts such as sodium chloride of predetermined size (prepared by controlled grinding and sieving) can be suspended in the viscous reaction medium using mechanical agitation with a vortex shaker. Salt content in the heterophase reaction medium can be varied in a wide range from 50 wt. % to 95 wt. %. Polymerisation following Protocol b with BF.sub.3O(Et).sub.2 is then possible. Salt can be leached out by soxhlet extraction with water to give macroporous scaffolds with defined pore sizes.

(28) Protocol d) Addition of Functional Groups by Incorporating Heterobifunctional Linkers, Typical Procedure.

(29) Polymer networks were prepared with a comonomer ratio of GGE:PEG(526)DE:AGE 19:76:5 according to method described in Protocol a. To the prepared film in water (3 ml of water per 1 g of polymer network material) was added cysteamine hydrochloride (5 eq. per mole of AGE in comonomer feed) and ammonium persulfate (1 eq. per mole of AGE in comonomer feed) and the gently stirred medium was incubated at 60 C. for 3 h. The bulk film products were then purified by soxhlet extraction with ethanol. Presence of covalently bound amino groups can be determined by dye adsorption assays (with picric acid). FIG. 4 shows a generalised scheme of the thiol-ene chemical conjugation.

(30) Reaction progress could be followed by rheological analysis to determine gelation point as shown in FIG. 5. Typically reactions proceed rapidly after an initial lag time as the exothermic reaction is autocatalytic. FTIR analysis indicates exhaustion of glycidyl ether groups in FIG. 6 Flory-Rehner theory was used to calculate crosslink density from degree of swelling in ethanol. Tensile tests were conducted to determine; Young's Modulus (E), tensile strength at yield point (.sub.y), elongation at yield point (.sub.y), tensile strength at break (.sub.B), and elongation at break (.sub.B). Dynamic mechanical thermal analyses were conducted to observe thermal transitions over a range of temperatures. These analyses indicate primarily sharp single peaks in the loss tangent (tan ), relating to the molecular relaxations of polymer chains occurring during the glass transition. This is compelling evidence for homogeneous networks with a lack of heterophase separation. As the exception, in the case of IGE and BGE containing copolymer networks, secondary beta transitions were observed which are hypothesisesd to result from side chain (isopropyl and n-butyl) relaxations above critical temperatures. Dynamic scanning calorimetry was conducted to analyse thermal transitions, only a single transition was ever observed for any composition, that relating to the glass transition.

EXAMPLES

(31) Materials

(32) Glycerol glycidyl ether (GGE) was purchased from Raschig GmbH (Ludwigshafen, Germany) and distilled under reduced pressure prior to use. Methyl glycidyl ether (MGE), ethyl glycidyl ether (EGE), isopropyl glycidyl ether (IGE), n-butyl glycidyl ether (BGE) (ABCR, Karlsruhe, Germany). Poly(ethylene glycol) diglycidyl ethers (PEGDE) with number average molecular weights of 526, 2,000, 6,000 (PDI1.1 in all cases), and photoinitiator diphenyliodonium hexafluorophosphate were purchased from Sigma Aldrich (Hannover, Germany) and used without further purification. The photoinitiator diphenyliodonium hexafluorophosphate was also purchased from Sigma Aldrich and used without further purification.

(33) Calculation of Network Properties

(34) Gel content G and degree of swelling Q is calculated according to equations 1 and 2:

(35) $\begin{array}{cc}G=\frac{m_d}{m_{\mathrm{iso}}}100\%& \left(1\right)\\ Q=1+{}_p\left[\frac{m_s}{m_d{}_s}-\frac{1}{{}_s}\right]& \left(2\right)\end{array}$
where m.sub.iso, m.sub.d and and m.sub.s are the weights of the crude, dry and swollen film, .sub.p and .sub.s are the densities of the polymer and the swelling agent respectively.

(36) Crosslink density and number average molecular weight M.sub.n between crosslinks are calculated by using the modified Flory-Rehner equations 3 to 5 and equation 6, respectively:

(37) $\begin{array}{cc}=\frac{-\left[V_r+V_r^2+\ln \left(1-V_r\right)\right]}{d_r{V}_0\left(V_r^{1/3}-\frac{V_r}{2}\right)}& \left(3\right)\\ =\frac{m_s-m_{\mathrm{iso}}}{m_s}\frac{{}_p}{{}_s}& \left(4\right)\\ V_r=\left(\frac{1}{1+}\right)& \left(5\right)\\ \stackrel{\_}{M_n}=\frac{1}{}& \left(6\right)\end{array}$
where, is the swelling coefficient, V.sub.r the volume fraction of the polymer, d.sub.r the density of polymer, V.sub.0 molar volume of the swelling agent and polymer-solvent interaction parameter, also called Flory-Huggins interaction parameter.

(38) The Flory-Huggins theory, modified by Blanks and Prausnitz (1964), allows establishing a relation between Flory-Huggins parameter and the solubility parameters of the polymer .sub.p and solvents .sub.s (eqn. 7):

(39) $\begin{array}{cc}={}_s+\frac{V_1}{\mathrm{RT}}{\left({}_p-{}_s\right)}^2& \left(7\right)\end{array}$
where V.sub.1 is the molar volume of the solvent and .sub.s the entropic contribution to . The solubility parameter of a polymer .sub.p is defined as a characteristic of a polymer used in predicting the solubility of that polymer in a given solvent. For polymers, it is usually taken to be the value of the solubility parameter of the solvent producing the solution with maximum swelling of a network of the polymer. Value of .sub.s is typically kept constant and equal to 0.34. Therefore, polymer solvent interaction parameter takes the value of 0.34 for .sub.p=.sub.s

(40) Rheology Analysis

(41) Rheological analysis of PGE films was performed on a Physica MCR 501 rheometer (Anton Paar GmbH) equipped with an external UV-light source (OmniCure UV LED spot curing system) having 365 nm UV LED head with 9,500 mW/cm.sup.2 irradiance. Starting reaction mixtures were placed on the glass plate through which UV-light irradiation passes from the source below. The gap between the glass plate and the metal plate of the measuring system was set at a distance of 0.3 mm. Shear conditions were kept at constant values for all experiments (deformation =0.5% and radial frequency =10 1/s). In Example 1 (GGE/XGE copolymer network films), the measuring temperature was kept at constant values. In Example 2 (GGE/PEGDE copolymer network films), reaction temperature for the networks containing PEG.sub.(526)DE, PEG.sub.(2000)DE and PEG.sub.(6000)DE was set to 50 C., 65 C. and 75 C. respectively.

(42) FTIR Spectroscopic Analysis

(43) FT-IR transmission spectra for PGE film samples are obtained using a Tensor 27 FT-IR Spectrometer (Bruker) with a standard DLaTGS-Detector.

(44) Thermal Analysis

(45) Thermal properties of the polymer networks were investigated by Thermal gravimetric analysis (TGA), Differential scanning calorimetry (DSC), and Dynamic mechanical analysis at varied temperature (DMTA).

(46) TGA of the samples was performed on a TG 209 apparatus (Netzsch). The film samples were heated from 25 to 400 C. at a heating rate of 10 K.Math.min.sup.1.

(47) DSC was performed on a DSC 204 apparatus (Netzsch). The film samples were heated from 25 to 250 C. (GGE/XGE copolymer network films) or to 100 C. (GGE/PEGDE copolymer network films) at a heating rate of 10 K.Math.min.sup.1, kept at this temperature for 2 minutes and cooled down to 100 C. (GGE/XGE) or to 50 C. (GGE/PEGDE) at 10 K.Math.min.sup.1 with a nitrogen purge and kept for 2 minutes at that temperature. Thermal properties were determined from second heating run at 10 K.Math.min.sup.1.

(48) DMTA was performed on an EPLEXOR QC 25 (GABO QUALIMETER Testanlagen GmbH) equipped with a 25 N load cell, at a frequency of 10 Hz and a heating rate of 2 K.Math.min.sup.1 in a temperature range between 50 C. and +100 C. (GGE/XGE) or between 30 C. to +70 C. (GGE/PEGDE).

(49) Mechanical Analysis

(50) Tensile properties of the polyether network films were determined on a Zwick tensile tester (2.5N1S, Zwick GmbH & Co, Ulm, Germany) equipped with a 50 N load cell at an elongation rate of 2 mm.Math.min.sup.1. Sample dimensions were 3 mm10 mm with a thickness of about 0.3 mm.

(51) 1. Glycerol Glycidyl Ether/Monoglycidyl Ether (GGE/XGE) Copolymer Networks

(52) 1.1. Synthesis and Network Formation

(53) GGE/XGE copolymer network films were synthesized in a one-step procedure, based on photo-polymerization of the initial glycidyl ether monomer mixture, in which the photoinitiator was dissolved. The initial monomer mixture was composed of either the GGE crosslinker itself or of a mixture comprising the GGE crosslinker and an alkyl glycidyl ether XGE, selected from methyl glycidyl ether MGE, ethyl glycidyl ether EGE, isopropyl glycidyl ether IGE, and n-butyl glycidyl ether BGE, where the alkyl glycidyl ethers XGE acted as the chain extension segments. For each mixture a weight ratio of the crosslinker to alkyl glycidyl ether GGE:XGE of 80:20, 60:40 and 40:60 was established. The photoinitiator concentration was kept at 2.0 wt-% with respect to the initial mixture amount. The unreacted mixture was poured into a mould formed by two silanized glass slides (25 mm75 mm), where a Teflon frame of 0.5 mm thickness was placed in between as the spacer to determine the thickness of the final product. Photopolymerization was performed with UVEX model SCU-110 mercury lamp, which was placed at a distance of 5 cm from glass slides. The reaction of this initial mixture, lead to a transparent film formation; the film products were left at dark for postcuring for 4 days.

(54) ##STR00008##

(55) Cross sections (10 mm20 mm) were taken from the cured transparent film samples and immersed in ethanol overnight for swelling and removal of unreacted components. The swollen films were then weighed and dried at 50 C. in high vacuum for one week until the weight reaches a constant value.

(56) 1.2 Reaction Kinetics

(57) Reaction kinetics during crosslinking were investigated by rheological analysis. Each sample was irradiated by UV until the ignition period was complete and network formation observed. Oscillatory tests were conducted to allow carrying out the measurement without any internal destruction to the sample during the curing process. Thus it was possible to examine time dependent formation of a chemical network during the measurement. Initially, G (loss modulus) >G (storage modulus), characteristic of the viscous fluid reactants. Upon irradiation a delay period is observed, followed by rapid increase in both moduli and an eventual inversion to show characteristics of a solid material where G>G. The gel time (t.sub.GT) is measured at the intersection where G=G or alternatively cited as tan =G/ G=1, and signifies the onset of the hardening process in cured thermoset networks. FIG. 5 shows typical changes in rheological parameters during the polymerization process, and is indicative of an autocatalytic process where rapid onset of gelation is observed after a delay period.

(58) The ring opening reaction could be monitored by FTIR spectroscopy. FIG. 6 shows the changes in the absorbance spectra of a reaction mixture only containing GGE after different irradiation times. The reduction of the peaks at 3056, 2998 and 907, 839 and 756 wavenumbers, which are attributed to epoxy vibrational frequencies, is indicative of the ongoing ring opening reaction. Literature states that peaks at 3056 and 2998 are particularly characteristic of oxirane compared to acyclic ethers. The broad absorbance from 3600 to 3300 cm.sup.1 is related to the hydrogen-bonded OH groups. At 3432 cm.sup.1 hydroxyl groups are formed by the acid-catalyzed epoxy ring opening reaction. High conversions are observed with no significant oxirane peaks remaining in the prepared films.

(59) 1.3. Characterisation of the Networks

(60) For each crosslinked polymer network, gel content (G), degree of swelling (Q), crosslink density (), and number average molecular weight between the crosslinks (i.e. network chain segment M.sub.n) were calculated from swelling data according to the Flory-Rehner equations as set forth above. The compositions of copolymers employed had significant influence on these parameters as shown in Table 1.

(61) TABLE-US-00001 TABLE 1 Chemical composition, degree of swelling (Q), gel content (G), density (), crosslink density (), and network chain segment length (M.sub.n) of the PGBNs studied. Q G [mol .Math. M.sub.n Series.sup.a [V .Math. V.sup.1] [%] [g .Math. cm.sup.3] cm.sup.3] .Math. 10.sup.3 [g/mol] G.sub.100 1.13 99 1.43 28.2 1.sup. 35.7 7.8 G.sub.80M.sub.20 1.27 98 1.39 15.3 1.sup. 65.0 4.1 G.sub.60M.sub.40 1.35 98 1.22 14.2 1.sup. 70.2 7.1 G.sub.40M.sub.60 1.42 97 1.17 11.3 0.1 79.1 2.7 G.sub.80E.sub.20 1.27 96 1.12 21.9 1.sup. 45.5 3.8 G.sub.60E.sub.40 1.45 95 1.07 13.3 0.1 74.9 0.4 G.sub.40E.sub.60 1.51 97 1.01 12.9 0.1 78.0 0.1 G.sub.80I.sub.20 1.28 95 1.20 19.9 5.3 50.2 1.4 G.sub.60I.sub.40 1.43 90 1.17 14.9 0.1 66.7 3.8 G.sub.40I.sub.60 1.58 94 1.03 13.0 0.4 76.7 2.1 G.sub.80B.sub.20 1.26 97 1.30 17.2 0.1 58.1 6.3 G.sub.60B.sub.40 1.35 97 1.09 16.9 0.5 58.8 1.8 G.sub.40B.sub.60 1.65 92 1.08 11.8 1.3 84.4 10 .sup.aG = GGE, M = MGE, E = EGE, I = IGE, B = BGE; numerical values indicate the weight fraction in %

(62) For each polymer network, gel content is high (92%) indicating high conversion.

(63) It can be seen in FIG. 7A that the degree of swelling (Q) correlates to the calculated chain segment length (M.sub.n, number average molecular weight of polymer chains between crosslink points). Likewise, M.sub.n inversely correlates to the amount of crosslinking GGE employed in the comonomer feed as shown in FIG. 7B. Clearly in this system crosslink density can be controlled by composition as the monoglycidyl ethers produce linear units upon polymerisation.

(64) 1.4. Thermal and Mechanical Properties

(65) Various thermal and mechanical properties of homo- and copolymer networks with varying comonomer ratios as determined by TGA, DSC and DMTA are shown in Table 2.

(66) TABLE-US-00002 TABLE 2 Mechanical properties of polyglycerol networks determined by tensile tests at room temperature; glass transition temperature (T.sub.g), Young's Modulus (E), tensile strength at yield point (.sub.y), elongation at yield point (.sub.y), tensile strength at break (.sub.B), elongation at break (.sub.B). T.sub.g, DSC E .sub.y .sub.y .sub.B .sub.B Series.sup.a [ C.] [MPa] [MPa] [%] [MPa] [%] G.sub.100 26 49.1 5 2.2 0.5 4.7 1.4 2.2 0.5 4.7 1.4 G.sub.80M.sub.20 7 33.2 2.4 4.2 0.4 13 0.7 4.0 0.5 13 0.7 G.sub.60M.sub.40 9 25.9 4.9 1.4 0.6 6.2 2.3 1.4 0.6 6.2 3.0 G.sub.40M.sub.60 28 15.9 1.9 1.3 0.2 10 1.6 1.3 0.2 10 1.7 M.sub.100 62 G.sub.80E.sub.20 17 31.6 2.4 3.5 1.0 12 3.1 3.3 1.1 12 3.1 G.sub.60E.sub.40 40 25.4 5.9 2.6 0.3 4.5 0.3 2.2 0.2 4.6 0.3 G.sub.40E.sub.60 41 16.2 1.7 0.6 0.5 3.7 0.5 0.6 0.6 3.7 0.5 E.sub.100 65 G.sub.80I.sub.20 14 45.6 0.6 3.1 0.3 7.4 0.7 2.7 0.4 7.5 0.7 G.sub.60I.sub.40 11 32.6 1.6 1.4 0.4 4.6 1.4 1.2 0.4 4.6 1.3 G.sub.40I.sub.60 20 12.2 0.5 0.5 0.1 4.9 0.5 0.4 0.1 5.0 0.5 I.sub.100 67 G.sub.80B.sub.20 4 43.8 4.2 2.4 0.5 6.0 1.3 2.2 0.6 6.1 1.2 G.sub.60B.sub.40 41 24.7 2.1 0.8 0.1 3.6 0.6 0.8 0.1 3.7 0.6 G.sub.40B.sub.60 52 16.7 4.7 0.3 0.3 2.2 1.7 0.3 0.3 2.2 1.7 B.sub.100 79 .sup.aG = GGE, M = MGE, E = EGE, I = IGE, B = BGE, numerical values indicate the weight fraction in %.

(67) Young's moduli E, determined by tensile testing, are variable in the range of 50-10 MPa and are highly tunable in a rational manner. As seen in FIG. 8A, an almost linear dependence can be drawn between the mole fraction of GGE crosslinker and the modulus value. Young's moduli were found to be highly dependent on crosslink density controlled by comonomer ratio, and less sensitive to the actual nature of the glycidyl ether side chains incorporated.

(68) As determined by DSC, glass transition temperatures T.sub.g are also proportional to crosslink density of the networks as shown in FIG. 8B. Transition temperatures T.sub.g are all below body temperature and are variable between 26 and 80 C. This variance in T.sub.g correlates to the ratio of monoglycidyl ether to crosslinker GGE and is also specific to the nature of the monoglycidyl ether sidechain. This phenomenon can be attributed to the increased mobility of longer alkyl side chains. Glass transition temperatures for copolymer networks at all ratios show a good fit to estimates derived from the Fox equation.

(69) The general trend of weight loss as analyzed by TGA is around 1 wt.-% up to 100 C., which is possibly due to loss of water.

(70) In order to gain more detailed information on the scale of molecular processes occurring during thermal transition, DMTA measurements were conducted, measuring storage and loss moduli over a temperature range. The results of DMTA measurements are shown in FIG. 9 with single peaks in loss tangent relating to glass transition (see general remarks). As the glass transition is approached loss modulus E rises to a maximum consistent with increased mobility of the polymer chains in the network. For the studied polymer networks the temperature values at the maximum of loss modulus are consistent with the glass transition temperature values obtained from DSC. As the material becomes deformable with increasing temperature, storage modulus E decreases. The rise in E concomitant with lowering of storage modulus E, leads to an increase in E/E, which value is known as the loss factor, or tan . The height and shape of tan gives information on the molecular processes occurring during the thermal transition, and the position of the highest point relates to glass transition temperature. For the studied networks, tan is generally observed as a single relatively narrow peak indicating that molecular relaxations occur in a narrow range of temperatures and a single transition occurs. This is strong support for a highly homogeneous network.

(71) Summarizing, the films prepared by bulk, highly branched GGE/XGE copolymer networks are transparent and have single thermal glass transitions below room temperature. Mechanical properties can be tuned by varying the crosslink density through incorporation of monoglycidyl ethers. Evidence points towards glycidyl ether side chains having significant influences on thermal transitions within the network. By varying comonomer ratio it is possible to control the network chain segment length which has a very clear influence on network behavior such as swelling, and on the mechanical properties. These findings are all characteristic of a homogeneous network structure.

(72) 2. Glycerol Glycidyl Ether/PEG Diglycidyl Ether (GGE/PEGDE) Copolymer Networks

(73) 2.1. Synthesis and Network Formation

(74) PEG-based polyether networks were synthesized by photopolymerization of an initial mixture of PEGDE of different molecular weight (M.sub.n=526, 2000 and 6000 g/mol) and GGE in different weight ratios, and diphenyl iodonium hexafluorophosphate as photoinitiator. The content of the photoinitiator diphenyl iodonium hexafluorophosphate was kept at 2 wt.-% with respect to the initial mixture amount. In the cases of PEG.sub.(2000)DE and PEG.sub.(6000)DE, the initial monomer mixtures were heated above the melting point of the corresponding PEGDE (T.sub.m=54 C. and 62 C., for PEG.sub.(2000)DE and PEG.sub.(6000)DE respectively). The reaction was performed using Physica MCR 501 rheometer (Anton Paar GmbH) equipped with an external UV-light source. The initial liquid mixture was placed between the glass and metal plates, which were preheated to the reaction temperature. The distance between glass and metal plates was set to 0.3 mm which resulted in a corresponding thickness of the final film product.

(75) Subsequent to the reaction of the initial mixture and network formation, the transparent film products were left in darkness for postcuring for 4 days. Cross sections (10 mm20 mm) were taken from the cured film samples and immersed in ethanol overnight for removal of unreacted components. The swollen films were then weighed and dried at 50 C. in high vacuum until the weight reached a constant value. A reswelling procedure was then performed in the same manner.

(76) 2.2. Reaction Kinetics

(77) Rheology analysis reveals the reaction kinetics and the progress of the crosslinking reaction. For thermosetting polymers rheological data is used mostly to identify gel points. Upon intense irradiation of the initial reaction mixture, the viscosity increase takes place due to gelation. Viscosity approaches infinity at the gel point, the characteristic value that indicates an infinite network. Gel time analysis of the polyether networks reveals the trend for the time of the network formation for each series composed of PEGDE with different molecular weights and in their different weight ratios. When each PEGDE series with different molecular weights is examined on their own, the increasing trend in gel time with decreasing GGE amount can be clearly distinguished (data not shown). Reaction compositions having relatively more GGE content were faster in gelation, thus in network formation. It is important to note that due to reactions requiring different temperatures in order to maintain PEGDE in the molten state, it is not possible to compare reaction kinetics between the different series.

(78) 2.3. Characterisation of the Networks and Thermal and Mechanical Properties

(79) Various network properties and thermal and mechanical properties of homo- and copolymer networks with varying comonomer ratios are shown in Table 3.

(80) TABLE-US-00003 TABLE 3 Mechanical and thermal properties of polyether networks; degree of swelling Q, glass transition temperature T.sub.g, Young's Modulus E, elongation at break .sub.B. Q T.sub.g, DSC T.sub.g (onset) T.sub.g (offset) E .sub.B Sample ID.sup.a [V .Math. V.sup.1] ( C.) ( C.) ( C.) (MPa) (%) G.sub.100 1.13 26 20 29 49.1 5.1 4.7 1.0 G.sub.80X.sub.20 1.23 11 9 13 37.1 1.6 6.9 1.0 G.sub.60X.sub.40 1.30 5 7 3 19.7 0.4 7.6 0.3 G.sub.40X.sub.60 1.37 22 24 20 14.3 0.2 11.7 1.0 G.sub.20X.sub.80 1.43 37 40 35 2.4 0.2 12.8 1.5 G.sub.80Y.sub.20 1.27 9 14 6 34.9 1.4 7.3 0.8 G.sub.60Y.sub.40 1.35 29 33 25 13.6 1.1 9.3 0.3 G.sub.40Y.sub.60 1.44 44 48 41 8.0 0.2 12.9 1.7 G.sub.80Z.sub.20 1.29 11 17 6 33.2 1.6 7.4 0.8 G.sub.60Z.sub.40 1.37 30 35 26 12.8 0.3 8.2 0.7 G.sub.40Z.sub.60 1.46 45 50 43 7.8 0.1 16.6 2.5 .sup.aG = GGE, X = PEG.sub.(526)DE, Y = PEG.sub.(2000)DE, Z = PEG.sub.(6000)DE, numerical values indicate the weight fraction in %.

(81) Gel content analysis shows for the whole series of networks a high value (min. 85%) indicating a high polymer fraction in the networks and high conversion (data not shown). Degree of swelling Q is highly dependent on the crosslinker amount in the network showing an increasing trend with the decreasing GGE amount and in the range between 1,13 (V.Math.V.sup.1) to 1,46 (V.Math.V.sup.1) for the networks that is composed of GGE and the one that is composed of PEG.sub.(6000)DE in 60 wt.% ratio.

(82) Thermal and mechanical properties of the copolyether networks were analyzed by DSC, DMTA and tensile testings. For each network, only one single transition was observed which corresponds to the glass transition. Glass transition temperatures are all below body temperature and are variable between 45 C. and +26 C. As shown in FIG. 10 glass transition temperatures T.sub.g of the copolymer networks are dependent on the GGE amount in the networks, i.e. crosslink density. T.sub.g decreases as the ratio of tritopic GGE to ditopic PEGDE is reduced due to more restricted movement of the molecular chains in the structure. The T.sub.g values observed for the PEG.sub.(526)DE networks are in close agreement with those predicted by the Fox Equation across the range of comonomer ratios employed (shown as dashed line in FIG. 10).

(83) As shown in FIG. 11 Young's modulus E is variable in the range of 2 to 50 MPa suitable for biomaterial and other applications and largely correlates with the ratio of tritopic GGE to ditopic PEGDE comonomers employed, with the mechanical properties being reliant on crosslink density. Likewise elongation at break .sub.B increased in a rational fashion from the lowest value (ca. 5%) for GGE homopolymer to the highest (ca. 17%) for the copolymer network prepared from 60:40 wt. % ratio PEG.sub.(6000)DE:GGE (Table 2).

(84) DMTA provided more detailed information about the polymer chain dynamics of the synthesized copolymer networks during thermal transition, where mechanical deformation takes place in a defined temperature range. DMTA shows sharp single peaks in the loss factor (FIG. 12a-c). The values, where the loss factor E is at maximum, are attributed to the glass temperatures for the networks and are comparable with the ones obtained from DSC. The data are indicative for a homogeneous network free of crystalline PEG rich regions. Maxima of tan curves are attributed to the term called mechanical glass transition and deviate from the values obtained from DSC by 10-15 C. (FIG. 11d-f). Intensity of tan peaks is high (>1) as an indication of good damping properties and shows an increasing trend with increasing PEGDE amount incorporated in the networks.

(85) 3. Cell Viability Test with Rat Mesenchymal Stem Cells

(86) In order to determine whether extracts were cytotoxic at low doses a preliminary series of Minimum Essential Medium (MEM) extracts were taken from homopoly(glycerol glycidyl ether) films prepared by Example 1.1 (Protocol a). These extracts were added to cell culture medium in varying concentrations in a modified procedure of the ISO 10993-5 standard cytocompatibilty testing. Importantly, the amount of film used in the extractions was half that of the ISO standard (which requires 30 cm.sup.2 of sample surface area for 10 ml of MEM eluent). Therefore the results are in this preliminary test reported as relative dilutions compared to the ISO standard.

(87) 3.1 Protocol

(88) 3.1.1. Test Medium Collection: Day0: Collect the 10 ml Soaking phosphate buffered saline (PBS) of film, refill with 10 ml cell culture medium (DMEM+10% FCS). Day1: Collect the 10 ml soaking medium and replace it with new 10 ml cell culture medium. Day2-4: Repeat step Day1 till Day4, get medium day2, day3, day4 10 ml respectively. Finally, get 5 test mediums (I, II, III, IV, V test medium at day0, day1, day2, day3, day4 respectively)

(89) 3.1.2. Cell Viability Test: (Rat Mesenchymal Stem Cells) Day0: Seed 1.5*10e.sup.5 cells in each well of 6-well plate. 20 wells should be prepared. Day1: Discard the medium in each well including the unattached cells. Wash once with PBS. Add 3 ml fresh medium (DMEM+10% FCS) containing 60 l, 30 l, 15 l, 0 l test medium I, II, III, IV, V respectively, 0 l for control. Till V medium. Day2: (1) Collect all the cells including cells in the medium of each well respectively (collect the supernatant medium and the attached cells by trypsin, get cell suspension solution). (2) Centrifuge cell suspension solution at 200 g for 10 min, discard the supernatant. Resuspend the cell pellet in 1 ml PBS. (3) Mix 10 l of 0.4% trypan blue and 10 l cell suspension (dilution of cells). Allow mixture to incubate 3 min at room temperature. (4) Count the cells number with contess (Invitrogen), count the unstained (viable) and stained (nonviable) cells separately to obtain the total number of viable cells and nonviable cells per ml of aliquot, then get the cell viability percentage. Cell viability percentage=viable cells/total cells*100%

(90) 3.2 Results

(91) Initially, low concentrations of extractant were added to the cell culture medium (varying from 0 to 60 L in 3 mL of cell culture medium). At these concentrations, shown in Table 4, the extract had no cytotoxic effect.

(92) TABLE-US-00004 TABLE 4 Viability of rat Mesenchymal Stem Cells after exposure to different concentrations of extracts taken from homopoly(glycerol glycidyl ether) films. Medium Day 0 Day 1 Day 2 Day 3 Day 4 I II III IV V Concentration (total: 3 ml) 60 l 30 l 15 l 0 l 60 l 30 l 15 l 0 l 60 l 30 l 15 l 0 l 60 l 30 l 15 l 0 l 60 l 30 l 15 l 0 l Via- 88 88 89 89 85.5 88 86 89.5 91 87.5 95.5 86.5 92.5 93 87.5 87 86.5 90 87.5 90.5 bility Per- cent- age (%)

(93) As a follow up, neat, undiluted eluent was used as cell culture medium and once again no cytotoxic effects were observed. This is shown in FIG. 13. Once again it must be stated that the amount of material used as substrate for the extraction was half that normally quoted for ISO 10993-5 standards, so this result has been quoted as a 1:2 dilution in FIG. 12, the previous results on diluted samples have also been included as 1: dilution relative to the ISO 10993-5 standard (i.e. 60 L=1:100 dilution, 30 L=1:200 dilution, 15 L=1:400 dilution). It is however promising that at this concentration no cytotoxic effects were observed.

(94) 4. Cytotoxicity Test with Fibroblast Cells

(95) Cytotoxicity test according to ISO 10993-5 standard with L-929 fibroblast cells were conducted.

(96) 4.1. Cell Culturing

(97) For extract production, 10 ml of cell culture medium without horse serum (EMEM, Biochrom, Germany) were put into a 15 ml tube (PP) with 20 cm.sup.2 of the sample (homopoly(glycerol glycidyl ether) film prepared by Example 1.1). The sample was mixed using a rotation shaker (15 rpm, 37 C., 72 h). The resulting 72 h-extract was separated from the sample by pipetting and stored at 4 C.

(98) L-929 (mouse fibroblasts, continuous cell line, originated from mice, ATCC) were cultured with the 72 h-extract for 48 h either in the undiluted extract or in extract dilated with cell culture medium at 1:10 and 1:100. As negative control (non cytotoxic) cells were cultured in the pure cell culture medium (EMEM). As positive controls (cytotoxic) cells were medium containing 1 mM CuCl.sub.2 (for MTS assay) or 0.5 Vol.-% Triton X (for LDH release), respectively.

(99) 4.2. Results

(100) After 48 h of cell growth, the cell morphology was assessed visually by phase contrast microscopy in transmission at magnifications of 20 and 40, respectively. The morphology of the L929 cells after culturing them with the undiluted 72 h-extract was different to the morphology of these cells culturing them with pure cell culture medium (negative control). The morphological changes correspond to the cytotoxicity level 2 (scale ranging from 0-4) classified to a mild toxicity.

(101) Also after 48 h of cell growth, lactate dehydrogenase LDH release in the extracellular fluid (Cytotoxicity detection KIT LDH, Roche, Germany) and the mitochondrial activity of the cells (CellTiter 96 AQ.sub.ueous Non-Radioactive Cell Proliferation Assay, Germany) were tested. Results are shown in FIG. 14 for the LDH-Relase (A) and for the mitochondrial activity (B). No significant change was observed for the release of extracellular LDH nor for the mitochondrial activity. This indicates that the sample did not influence the functional integrity of the outer cell membrane and had no impact on the cell activity.

(102) Further tests showed that the to the cytotoxicity of the films could be reduced further by additional purification steps to remove cytotoxic agents. Here the films after their preparation were continuously (soxhlet) extracted with ethanol for 10 days and subsequently washed with MEM medium (3 days with changing ever 24 h). The films purified in this way were shown to be level 1 cytotoxic, which is acceptable for use in biomaterial applications.