Method for the production of poly(2-octyl cyanoacrylate)-polyisobutylene co-network, and super initiators therefor

Abstract

A method for increasing the rate of polymerization of 2-octyl cyanoacrylate, or the rate of copolymerization of 2-octyl cyanoacrylate and a tri-telechelic star polymer comprising polyisobutylene terminated with cyanoacrylate groups ((PIB-CA).sub.3) to form a co-network, is provided. The method comprise initiating the polymerization of 2-octyl cyanoacrylate, or the copolymerization of 2-octyl cyanoacrylate and a tri-telechelic star polymer comprising polyisobutylene terminated with cyanoacrylate groups ((PIB-CA).sub.3) to form the co-network, with an initiator selected from the group consisting of cyclic tertiary aliphatic amines optionally dissolved in a non-aqueous solvent. The cyclic tertiary aliphatic amines are selected from the group consisting of azabicyclo[2.2.2]-octane (ABCO), and 1,4-diazabicyclo[2.2.2]-octane (DABCO).

Claims

1. A method for increasing the rate of polymerization of 2-octyl cyanoacrylate, or the rate of copolymerization of 2-octyl cyanoacrylate and a tri-telechelic star polymer comprising polyisobutylene terminated with cyanoacrylate groups ((PIB-CA).sub.3) to form a co-network, the method comprising: initiating the polymerization of 2-octyl cyanoacrylate, or the copolymerization of 2-octyl cyanoacrylate and (PIB-CA).sub.3 to form the co-network, with an initiator selected from the group consisting of cyclic tertiary aliphatic amines optionally dissolved in a non-aqueous solvent, wherein the cyclic tertiary aliphatic amines are selected from the group consisting of azabicyclo[2.2.2]-octane (ASCO), and 1,4-diazabicyclo[2.2.2]-octane (DABCO).

2. The method according to claim 1, wherein the non-aqueous solvent is used and is selected from the group consisting of tetrahydrofuran (THF), toluene, and combinations thereof.

3. The method according to claim 1, wherein the step of initiating further includes further initiating the polymerization or copolymerization with nucleophilic groups located on the surface to be covered by the co-network.

4. The method according to claim 3, wherein the surface to be covered is skin.

5. The method according to claim 1, wherein the polymerization of 2-octyl cyanoacrylate is provided, and wherein a set time for the resultant poly(2-octyl cyanoacrylate) is shorter than the set time for formation of poly(2-octyl cyanoacrylate) from the polymerization of 2-octyl cyanoacrylate using an aromatic tertiary amine initiator in the same solvent and initiator concentration.

6. The method according to claim 1, wherein the copolymerization of 2-octyl cyanoacrylate and (PIB-CA).sub.3 is provided, and wherein a set time for the resultant 2-octyl cyanoacrylate-(PIB-CA).sub.3 co-network is shorter than the set time for formation of a like co-network from the copolymerization of 2-octyl cyanoacrylate and (PM-CA).sub.3 using an aromatic tertiary amine initiator in the same solvent and initiator concentration.

7. The method according to claim 5, wherein the non-aqueous solvent is present and is THF.

8. The method according to claim 1, wherein the molar ratio of monomer to initiator is greater than 1000:1 and the set time is less than 300 seconds.

9. The method according to claim 8, wherein the set time is less than 120 seconds.

10. The method according to claim 5, wherein the set time for the resultant poly(2-octyl cyanoacrylate) is at least 30 seconds shorter than the set time for poly(2-octyl cyanoacrylate) using an aromatic tertiary amine initiator in the same solvent and initiator concentration.

11. The method according to claim 10, wherein the set time for resultant poly(2-octyl cyanoacrylate) is at least 60 seconds shorter than the set time for poly(2-octyl cyanoacrylate) using an aromatic tertiary amine initiator in the same solvent and initiator concentration.

12. The method according to claim 6, wherein the set time for the resultant co-network is at least 30 seconds shorter than the set time for other co-networks using an aromatic tertiary amine initiator in the same solvent and initiator concentration.

13. The method according to claim 12, wherein the set time for the resultant co-network is at least 60 seconds shorter than the set time for other co-networks using an aromatic tertiary amine initiator in the same solvent and initiator concentration.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a prior art, idealized microstructure representation of a rubbery polyisobutylene homopolymer formed of (PIB-CA).sub.3, wherein the CA-CA bonds are sufficiently short to only be crosslinkers;

(2) FIG. 2 is an idealized microstructure representation of a polyisobutylene co-network that arises from a homogeneous 50/50 wt/wt polymer blend of Oct-CA plus (PIB-CA).sub.3 upon the addition of an initiator at room temperature, wherein the poly(Oct-CA) sequences polymerized are of sufficient length (i.e., greater than 3 CA units) to form separate coalesced phases whose T.sub.g can be identified by appropriate instrumentation such as DSC or DMTA;

(3) FIG. 3A is the H NMR spectrum of an allyl three-arm star PIB intermediate;

(4) FIG. 3B is the H NMR spectrum of a hydroxyl-three arm star PIB intermediate;

(5) FIG. 3C is the H NMR spectrum of an anthracene/cyanoacrylate adduct three arm star PIB intermediate;

(6) FIG. 3D is the H NMR spectrum the cyanoacrylate-tri-telechelic PIB, (PIB-CA).sub.3 final product;

(7) FIG. 4 is the GPC trace of a representative three-arm star allyl-tri-telechelic polyisobutylene;

(8) FIG. 5 is a Proton NMR spectrum of Poly(Oct-CA);

(9) FIG. 6 is a table showing the results of an experiment showing the polymerization of Oct-CA with DMT in bulk;

(10) FIG. 7 is a table showing the results of an experiment showing the polymerization of Oct-CA with ABCO dissolved in THF;

(11) FIG. 8 is a table showing the results of an experiment showing the polymerization of Oct-CA with DABCO dissolved in THF;

(12) FIG. 9 is a table showing the results of an experiment showing the polymerization of Oct-CA with ABCO dissolved in Toluene;

(13) FIG. 10 is a table showing the results of an experiment showing the polymerization of Oct-CA with DABCO dissolved in Toluene; and

(14) FIG. 11 shows the relationship between the [Monomer]/[Initiator] Ratio and Stir/Stop Times for the experiments shown in FIGS. 6-10.

DETAILED DESCRIPTION OF THE INVENTION

(15) As noted hereinabove, the present invention seeks to provide a method for increasing the rate of polymerization of 2-octyl cyanoacrylate or the rate of copolymerization of 2-octyl-cyanoacrylate (Oct-CA) and cyanoacrylate-terminated tri-telechelic polyisobutylene ((PIB-CA).sub.3) to form a polymer composition or co-network composition, respectively, suitable for any of a number of biomedical applications, from wound closure and healing of skin tissue, to sealant for surgical cuts. Where the composition is a co-network, the co-network may comprise a copolymer of 2-octyl-cyanoacrylate (Oct-CA) and cyanoacrylate-terminated tri-telechelic polyisobutylene ((PIB-CA).sub.3). While the polymerization of Oct-CA or the copolymerization of Oct-CA with (PIB-CA).sub.3 may be initiated by the moisture (i.e., nucleophiles) within the skin, blood or other living (or dead) tissue itself when a mixture of the liquid starting materials is sprayed, coated or otherwise applied over wounds or surgical cuts, it has been found that the rate of polymerization of the Oct-CA, or the rate of copolymerization of 2-octyl-cyanoacrylate (Oct-CA) and cyanoacrylate-terminated tri-telechelic polyisobutylene ((PIB-CA).sub.3) to form the co-network, can be significantly increased by the use of particular initiators applied just before application of the liquid starting materials to the skin, wound, or surgical cut.

(16) As noted above, the starting composition is a liquid (or liquids) that can be applied by essentially any means known in the art to form a coating or film that preferably rapidly solidifies into a robust rubbery protecting barrier. In at least one embodiment, which is well known for wound closure adhesives, the liquid starting cyanoacrylate-based composition is contained or packaged within a special delivery device, wherein the composition (together with a variety of additives, modifying agents, etc.) is sealed in a thin-walled glass vial that is crushed upon deployment, and the liquid monomer(s) is then forced towards the skin through a small porous plastic sponge (typically of polypropylene or nylon) situated at the tip of the delivery port. It is generally not appreciated, but the sponge at the tip of the delivery device performs two critical functions. First, it helps evenly deliver the active ingredient over the targeted surface. Second, and perhaps more importantly, it often contains an initiator used to induce and/or accelerate the polymerization of the Oct-CA monomer(s) or the CA-terminated compounds as it is squeezed through the sponge. Absent the initiator, the set time is undesirably long, usually many minutes. For example, it has been found that the set time was less than a minute when a glass vial of a commercial Dermabond sample was crushed and the liquid was allowed to flow, as designed, through the sponge onto a glass surface with the initiator DMT. In contrast, the set time was 8-10 minutes when the vial was crushed but the liquid spilled directly over the same surface i.e., without contacting the sponge or the initiator.

(17) While the tissue, e.g., skin, or more accurately, the nucleophilic groups (OH, NH2, etc.) on the surface of the skin, can, in effect, act as a catalyst of the polymerization, i.e., be the agent that initiates the polymerization, it has been found that using the skin will only reduce the set time by a couple of minutes, i.e. from 8-10 minutes, to 6-8 minutes. Thus, the use of the skin will aid catalyzing the reaction, but will not significantly increase the rate of polymerization or copolymerization.

(18) Further, the polymer that forms must be a biocompatible biostable hydrophobic elastomeric barrier to bacterial invasion that keeps the coated skin moist, thereby promoting healing. The barrier, because of the specific catalyzed initiation mechanism, must adhere strongly by covalent bonds onto the surface of the tissue. Because of the absence of (PIB-CA).sub.3 in earlier purely polyalkyl-cyanoacrylate wound closures, earlier wound closures did not exhibit such advantageous combination of elastomeric properties.

(19) For the present invention, it has been discovered that cyclic aliphatic tertiary amines (e.g., ABCO and DABCO dissolved in dry toluene or tetrahydrofuran) very rapidly and efficiently initiate the polymerization of 2-octyl cyanoacrylate (Oct-CA) and the copolymerization of Oct-CA with three-arm star tri-telechelic polyisobutylene [(PIB-CA)3], i.e., monomers useful for the preparation of wound closure adhesives.

(20) To begin, it will be appreciated that the chemical formulas of the two starting materials are shown below as formulas (I) and (II).

(21) ##STR00001##

(22) Because both (PIB-CA).sub.3 and Oct-CA contain polymerizable cyanoacrylate (CA) groups they can readily produce polymers. Polymer (i.e., polymer or co-network) compositions can essentially be controlled by using desired amounts of the two ingredients. Overall, the co-network composition will reflect the relative composition of the starting monomers. Generally, a molar ratio of 2-octyl cyanoacrylate to (PIB-CA).sub.3 can be from about 5:1 to about 40:1, for the copolymerization reaction, wherein the copolymerization reaction is initiated by at least one initiator selected from the group consisting of cyclic aliphatic tertiary amines. In other embodiments, the molar ratio is from about 15:1 to 35:1 and in still other embodiments, the molar ratio is from about 20:1 to about 30:1. In other embodiments, and again generally, a number average molecular weight ratio of 2-octyl cyanoacrylate to (PIB-CA).sub.3 can be from about 2:1 to about 9:1, for the copolymerization reaction, wherein the copolymerization reaction is initiated by at least one initiator selected from the group consisting of cyclic aliphatic tertiary amines. In other embodiments, the weight ratio is from about 2:1 to 5:1 and in still other embodiments, the weight ratio is from about 3:1 to about 4:1. Again, the cyclic aliphatic tertiary amines may be selected from ABCO or DABCO.

(23) Because both monomers, (PIB-CA).sub.3 and Oct-CA, are liquids, their mixtures can be delivered in a number of different ways. In one embodiment, the liquids are delivered as described above as spraying or application via vial with plastic sponge tip to provide a suitable coating or film of the composition onto the desired tissue. In another embodiment, the liquids may be delivered by syringe, injecting the composition to a suitable site, again using the sponge tip. By allowing such monomer mixtures to polymerize in situ, solid rubbery plugs can form exactly where the mixture was injected, i.e., where the seal is needed. The fact that (PIB-CA).sub.3, has a high molecular weight (Mn approximately=3000 g/mol), and Oct-CA are miscible was surprising because the very similar cyanoacrylates Methyl-, Ethyl-, and Butyl-CA are completely immiscible with ((PIB-CA).sub.3) and, when such binary systems are mixed, they do not form a homogeneous phase but remain separate.

(24) For a control embodiment of the present invention, a copolymerization of (PIB-CA).sub.3 plus Oct-CA mixtures was initiated by the use of N-dimethyl-p-toluidine (DMT) initiator. Neat (PIB-CA).sub.3+Oct-CA blends, upon contact with appropriate amounts of initiator, polymerized within seconds to minutes (i.e., from about 20 to about 300 seconds) to optically transparent strong rubbers. Co-network formation was demonstrated by stir/stop studies, and their structures and properties were characterized by a battery of techniques. Select co-networks, when deposited on fresh ventral porcine skin yielded hermetically-adhering optically clear rubbery coatings appropriate for occlusive wound closures.

(25) In one control embodiment, the copolymerization of (PIB-CA).sub.3 is reacted with various amounts of Oct-CA to form co-networks by the use of DMT initiator. The characterization and testing of the resultant products have been developed. The structures of the copolymerization of these (PIB-CA).sub.3s with various proportions of Oct-CA, has been characterized and select properties of the resulting co-polymers (i.e., co-networks) studied. Extraction studies have indicated the formation of co-networks. Moreover, these copolymerizations are shown to occur fairly rapidly (i.e., approximately 20-30 seconds to 4-5 minutes). The molecular weight of the (PIB-CA).sub.3 can be adjusted to yield moderately viscous liquids for easy deposition on surfaces. Select co-networks gave optically clear strong rubbery films hermetically adhering to the surface of ventral porcine skin suitable for occlusive skin closure adhesives.

(26) The preparation and copolymerization of (PIB-CA).sub.3 with various amounts of Oct-CA to form co-networks by the use of DMT initiator is provided below. The following is an example of one control for the invention only, and therefore, should not be seen as necessarily limiting the scope of the invention, the scope and spirit of the invention being set forth in the attached claims.

(27) Synthesis of (PIB-CA).sub.3.

(28) The preparation of (PIB-CA).sub.3 is well known and has been described in at least US Patent Application Publication No. US2014/0073743 A1, the disclosure of which is hereby incorporated by reference. Briefly, the synthesis involves the living polymerization of isobutylene induced by a trifunctional initiator and termination with allyltrimethylsilane. The 3-arm star allyl-terminated intermediate so obtained is converted quantitatively to the hydroxyl or bromine terminated intermediate, which is then reacted with anthracene-protected cyanoacryloyl chloride, or, preferentially, with 2-cyanoacrylic acid.

(29) Earlier syntheses of (PIB-CA).sub.3 carried out by the use of (protected) cyanoacryloyl chloride consistently gave yellow products. Efforts to remove the color (repeated precipitations, column chromatography, treatment with activated carbon) were only partially successful. The source of the discoloration is unknown (most likely due to traces of impurities associated with the use of thionyl chloride). In contrast, esterification of (PIB-OH).sub.3 with anthracene-protected 2-cyanoacrylic acid gave colorless products (10). The following equation outlines this preferred method for the synthesis of (PIB-CA).sub.3 (The protective anthracene group, indicated by A in the semi-circle, can be readily removed by maleic anhydride):

(30) ##STR00002##

(31) Thus, in a 50 mL Schlenk flask with a magnetic stir bar were placed under a blanket of nitrogen (PIB-OH).sub.3 (1.227 g, M.sub.n=2500 g/mol), anthracene-protected 2-cyano carboxylic acid adduct (1.333 g), and 4-dimethylamino pyridine (DMAP, 71.3 mg) dissolved in dichloromethane (DCM, 25 mL). Then the solution was cooled to 0 C., N,N-dicyclohexylcarbodiimide (DCC, 1.0648 g) was added, the solution was stirred for 30 min at 0 C., and then overnight at room temperature. The precipitated urea was filtered off, the DCM was evaporated in vacuo, and the viscous residue was dissolved in THF and purified by two precipitations into methanol.

(32) Finally, to yield (PIB-CA).sub.3, the protective anthracene group was removed by treatment with maleic anhydride in refluxing xylene for 8 hrs. According to NMR analysis the yields of protection and deprotection were typically 60 and 90%, respectively. Similar yields have been obtained by others who used the same protection/deprotection technique.

(33) Polymerization of Oct-CA in Bulk

(34) A 5 ml vial containing a Teflon coated small (1 cm) magnetic stir bar was charged with Oct-CA (1.0 g), and a calculated amount of DMT initiator was injected by a micro syringe. The charge turned yellow immediately upon DMT addition due, without being bound by theory, to possibly charge complexes. The vial was capped, vigorously mixed for a few seconds, and placed on a stirring plate for stirring at 60 rpm. The gel time, i.e., the time (in seconds) stirring stopped due to the viscosity increase and gel formation, was recorded. The instant the gel formed the yellow color disappeared and the charge became colorless.

(35) Polymerization of Oct-CA with DMT Diluted with THF.

(36) Polymerizations were carried out under UHP nitrogen in 50 mL Schlenk flasks equipped with a magnetic stir bar. Flasks were sealed with a rubber septum, purged with nitrogen, flame dried, cooled to room temperature and charged with Oct-CA (1.938 g, 9.26 mmol).

(37) Polymerizations were initiated by injecting DMT (0.1224 g, 0.905 mmol) dissolved in THF (4.9 mL) under a blanket of N.sub.2 at r.t. (M.sub.0/I.sub.0=10, [Oct-CA]=1.32 mol/L). To keep the [M].sub.0/[I].sub.0 ratio constant, the volumes of Oct-CA (0.5-1.2 mL), and initiator solutions (5-100 L) were adjusted. The polymerization mixture became yellow immediately upon DMT addition. After about 10 minutes stirring, the yellow color disappeared, and the mixture became colorless. After overnight stirring at room temperature the colorless polymer was purified by three precipitations in methanol and dried in vacuum at 60 C. for 24 hours to constant weight.

(38) Copolymerization of (PIB-CA).sub.3 with Oct-CA

(39) Co-networks were prepared by copolymerizing (PIB-CA).sub.3 plus Oct-CA at room temperature. Thus, in a well-dried 10 mL screw-cap vial was placed (PIB-CA).sub.3 (0.3-1.2 g, M.sub.n=2500 or 6500 g/mol) dissolved in freshly distilled THF (3 mL). Polymerizations were induced by adding various amounts (5-150 L) of DMT initiator. The vial was capped, the solution was vigorously manually agitated for 1-2 seconds and rapidly poured into a 77 cm Teflon mold. The volatiles were evaporated in a fume hood for 1 h, and the film was dried at 60 C. to constant weight.

(40) Polymerization of Oct-CA to Thermoplastics

(41) It will be appreciated that the repeat structure of poly(Oct-CA) is shown below as Formula (III).

(42) ##STR00003##

(43) Poly(Oct-CA) is an optically transparent rather stiff thermoplastic (T.sub.g58 C., see FIGS. 8 and 9), with high permanent set and a strong tendency to creep (see FIG. 7). Thus, poly(Oct-CA) tends to slough off the skin rather rapidly after deployment. Due to its stiffness it is not used over moving or creased skin.

(44) Polymerization of (PIB-CA).sub.3 to Rubbery Networks

(45) (PIB-CA).sub.3 upon contact with nucleophiles (initiators, proteinaceous tissue, moisture) crosslinks and yields networks. It has been determined that the extent and rate of crosslinking depend on a number of factors, including the length (molecular weight) of the PIB arms (i.e., the molar concentration of CA groups), and the nature and quantity of the initiator employed. The amount of extractables obtained with a representative PIB-based homo- and co-networks prepared with (PIB-CA).sub.3 of M.sub.n=2400 g/mol using DMT initiator was determined. The small amounts of extractables (3-5%) indicate extensive network formations.

(46) Copolymerization of Oct-CA with (PIB-CA).sub.3 to Rubbery Co-Networks

(47) Prior to bulk copolymerizations, the mutual miscibility of Oct-CA and (PIB-CA).sub.3 was investigated. It was found that Oct-CA plus (PIB-CA).sub.3 having a number average molecular weight of 2500 g/mol or more (M.sub.n=at least 2500 g/mol) yield optically clear blends in all proportions, indicating that Oct-CA and (PIB-CA).sub.3 are miscible. In contrast, even small amounts (5%) of Me- and Et-CA are immiscible with (PIB-CA).sub.3 and yield hazy mixtures under the same conditions. Further, it was found that Oct-CA plus (PIB-CA).sub.3 of M.sub.n=6,500 g/mol produce hazy mixtures, indicating that the higher molecular weight (PIB-CA).sub.3 is immiscible or only partially miscible with Oct-CA Upon the addition of 10 wt % THF to the hazy Oct-CA/(PIB-CA).sub.3 (M.sub.n=6,500 g/mol) mixture, the system became optically clear, indicating miscibility in THF solution.

(48) Stress-strain profiles of a poly(Oct-CA), a homonetwork made by crosslinking (PIB-CA).sub.3, and several representative co-networks of (PIB-CA).sub.3/Oct-CA were prepared. Importantly, the stress-strain profile of poly(Oct-CA) is fundamentally different from the homo- and co-networks. While poly(Oct-CA) presents a stiff relatively weak (barely 1 MPa) thermoplastic exhibiting high permanent set (300%), the homo- and co-networks are rubbery. The homonetwork obtained (with DMT initiator) with (PIB-CA).sub.3 of M.sub.n=2400 g/mol shows low tensile strength and elongation, however, these properties increase upon Oct-CA incorporation. Specifically, a 25/75 (PIB-CA).sub.3 (M.sub.n=6500 g/mol)/Oct-CA co-network shows a yield point at 10% strain, after which it exhibits impressive rubbery properties with 6 MPa tensile strength and 180% elongation. These properties are comfortably in excess of those required of a wound closure adhesive.

(49) Moreover, tan of both products show two T.sub.gs, a low temperature transition corresponding to PIB (20 and 10 C.) and a high temperature peak due to poly(Oct-CA) sequences (60 and 70 C.). The T.sub.gs attributed to PIB are higher than values usually reported for this polymer (i.e., 50 to 70 C.) because of the relatively low molecular weight PIB in the co-networks. It is noted that the T.sub.gs due the poly(Oct-CA) are significantly higher than those (40 C.) reported by earlier investigators. The E trace of the 25% (PIB-CA).sub.3(M.sub.n=2400 g/mol)/75% Oc-CA co-network shows a well-defined low and a high temperature flow region separated by a rubbery plateau, which reflect the thermal transitions. The co-network prepared with the linear CA-PIB-CA also exhibits similar trends.

(50) Efforts to carry out DMT experiments with poly(Oct-CA) were unsuccessful because the glassy samples broke at 95 C.

(51) Polymer Chemical Considerations

(52) Since the inherent reactivity of a functional group (in this instance, the CA group) is independent of the molecular weight of the polymer it is attached to, it may be safely assumed that the reactivities of the CA groups of Oct-CA and (PIB-CA).sub.3 are essentially identical (i.e., their reactivity ratios are unity). Thus, the compositions of co-networks are controllable by controlling the relative amounts of Oct-CA and (PIB-CA).sub.3 in these miscible charges.

(53) In the presence of a stoichiometric excess of Oct-CA in a Oct-CA/(PIB-CA).sub.3 charge assembled for the synthesis of co-networks, initiation will preferentially involve Oct-CA. The first event of initiation is likely the direct addition of the initiator to Oct-CA as follows:

(54) ##STR00004##

(55) If this were true, the polymer ought to carry the initiator head group. A search of the prior art was not able to identify any reference that provided for aromatic or cyclic head groups in polymers prepared with aromatic tertiary amines (e.g., DMT) or cyclic tertiary amines. In spite of extensive research in this field by earlier workers, the exact details of initiation of anionic alkyl cyanoacrylate polymerization remain unknown.

(56) During the early stages of polymerization the viscosity of the system is relatively low and propagation, i.e., the attack of the first CA anion to Oct-CA and/or (PIB-CA).sub.3 (that yields crosslinking) is relatively unhindered as set forth in the reaction scheme below:

(57) ##STR00005##

(58) Upon further propagation steps, particularly after (PIB-CA).sub.3 incorporation, the viscosity of the system rises very rapidly and the rate of (co)polymerization necessarily drops precipitously. Ultimately, a fraction of CA groups likely become entrapped in the highly viscous matrix and propagation ceases.

(59) Due to the highly hydrophobic matrix, termination, i.e., the permanent annihilation of propagating anions, which likely involves reaction with protons (i.e., moisture), is absent or is very slow in these bulk polymerizations. The mechanical properties of the products may be controlled by controlling the relative proportions of the rubbery PIB and glassy poly(Oct-CA) segments. By increasing the length of the PIB arms, elongations increase and moduli decrease. The longer poly(Oct-CA) sequences would phase separate and may function as reinforcing sites. Ultimate properties could also be controlled by the use of mixtures of two (or more) (PIB-CA).sub.3s of different M.sub.ns, or blends of (PIB-CA).sub.3s with linear telechelic CA-PIB-CAs.

(60) Microstructure/Morphology

(61) FIG. 1 shows an idealized microstructure of a prior art network formed by the polymerization of (PIB-CA).sub.3 deposited on a proteinaceous surface (skin). These constructs are in fact PIB networks (the CA groups merely provide initiating and crosslinking sites) with two kinds of crosslinking sites: (a) the aromatic centers of the (PIB-CA).sub.3, and (b) crosslinks formed by linking two (or less likely, three) CA groups. Polymerizations may be induced by the initiator (I) (where the initiator is DMT), nucleophilic groups on the surface of the skin, or adsorbed moisture. The networks are expected to contain numerous loops and catenates, which affect ultimate load bearing properties. In FIG. 1, the wiggly lines are PIB, I is the initiator, .circle-solid. is the aromatic center of (PIB-CA).sub.3, CA--- is the cyanoacrylate available for bonding or bonded to the skin surface, the CA-CA with the I initiator are the formed crosslinkers for the network, and the CAs in circle are useless entrapped CA). The loops and entrapped/catenated crosslinks are to be noted.

(62) With more particular respect to the present invention, FIG. 2 shows an idealized microstructure of a co-network that arises from a homogeneous 50/50 wt/wt blend of Oct-CA plus (PIB-CA).sub.3 upon the addition of an initiator at room temperature. In these co-networks, the poly(Oct-CA) sequences are of sufficient length (i.e., greater than 3 CA units) to form separate coalesced phases whose T.sub.g can be identified by appropriate instrumentation. As shown in FIG. 2, the 50/50 wt/wt Poly[Oc-CA-co-(PIB-CA).sub.3] co-network is represented, with the wiggly lines being PIB, I being the initiator, CA--- being the Oct-CA bonded to or available for bonding to the skin surface, The CA of multiple units being the Poly(Oct-CA), the CAs in the circle being useless CA groups entrapped in matrix, and the .circle-solid. being the aromatic center of (PIB-CA).sub.3. Again the presence of catenated/entrapped crosslinks and PIB loops should be noted. Importantly, the polymerization of CA groups is initiated by a purposely added initiator (I), in addition to a nucleophilic group (N or O) in the epidermis (---), or by traces of moisture (not shown).

(63) In light of the foregoing, it will be appreciated that, depending upon the number of CA units, either neat (PIB-CA).sub.3, or Oct-CA plus (PIB-CA).sub.3 blends, upon contact with an initiator, produce rubbery homo-networks or co-networks, respectively, of potential use as occlusive flexible wound closure adhesives. A variety of such networks have been prepared, characterized by various techniques, and select properties, e.g., elongation, tensile strength, have been determined. Viscous blends of starting materials upon contact with appropriate amounts of initiator polymerize within a few seconds to minutes and produce rubbery strongly adherent transparent coatings on skin. The coatings follow the contours and irregularities of the skin, giving rise to smooth rubbery flexible transparent membranes, which adhere strongly to the surface and provide seamless hermetic closures.

(64) The rubbery character (stretchiness) of the co-network can be increased by increasing the concentration of the (PIB-CA).sub.3 in the co-network. Other properties can be controlled by controlling the molecular weights (Mn) of the co-network segments (i.e., the number average molecular weights of the poly(Oct-CA) and (PIB-CA).sub.3, respectively). While any molecular weight range can be used, one suitable range would be to provide the molecular weight of poly(Oct-CA) as from about 3000 g/mol to 5000 g/mol, with about 4000 g/mol being suitable for one embodiment, and that of (PIB-CA).sub.3 being from about 2000 g/mol to about 4000 g/mol, with about 3000 g/mol being suitable in one embodiment. Thus, it will be appreciated that, for the three arm tri-telechelic cyanoacrylate PIBs, each PIB arm is about 1000 g/mol. The production of these two starting molecules is well known to those of skill in the art.

(65) In order to demonstrate practice of the present invention, various polymerizations and copolymerizations were conducted which were initiated with DMT (control), ABCO, and DABCO. The following exemplified procedure and results are provided to show a detailed example of the present invention. Therefore, it should not be seen as narrowing the invention, breadth and spirit of the invention being dictated by the attached claims.

(66) Materials

(67) N,N-dimethyl-p-toluidine (DMT, 99%), purchased from Aldrich, was used as received. Azabicyclo[2.2.2]-octane 97+% (ABCO), and 1,4-diazabicyclo[2.2.2]-octane 98% (DABCO) were purchased from Alfa Aesar and were used without further purification. Ethyl-2-cyanoacrylate (Et-CA) was purchased from Loctite, and 2-octyl cyanoacrylate (Oct-CA) was purchased from Chenso, and they were used without further purification. Tetrahydrofuran, purchased from Aldrich and toluene were thoroughly dried by refluxing and distilling the solvents over sodium and benzophenone.

(68) Instruments and Procedures

(69) Proton (.sup.1H) NMR spectroscopy, (Varian Gemini 300 and 500 MHz instruments and deuterated chloroform as solvent) was used to determine chemical structures, chain-end functionalities and molecular weights (M.sub.n).

(70) Gel permeation chromatography (GPC) eluograms were obtained using a Waters GPC instrument equipped with a series of three Waters Styragel-HR columns (HR-1, HR-4E, HR-5E), a refractive index detector (Waters 2414) and a multiangle laser light scattering detector (Dawn EOS, Wyatt Technology). Samples were dissolved in THF, the flow rate was 1 mL THF/min, and column temperature was 35 C.

(71) Stress strain properties of microdumbell-shaped samples were determined with a Texture Analyzer TA.XTplus tester with a 5 kilo load cell at a crosshead speed of 5 mm/min, following ASTM D638-02a. Samples (0.2-0.25 mm thick) were punched from solution (THF) cast films.

(72) Syntheses and Chemical Manipulations-Synthesis of (PIB-CA).sub.3

(73) The synthesis started with the living polymerization of isobutylene induced by a trifunctional initiator, and the polymerization was terminated with allyltrimethyl silane. The 3-arm star allyl-terminated intermediate was converted quantitatively to a hydroxyl terminated intermediate, which was then reacted with anthracene-protected acryloyl chloride. Finally, the protective anthracene group was removed by treatment with maleic anhydride in refluxing xylene for about 10 hours to yield the target (PIB-CA).sub.3. FIGS. 3A-3C show the NMR spectra of the intermediates (allyl-, hydroxyl-, and anthracene/cyanoacrylate adduct) and FIG. 3D shows the NMR sprectra of the final product (cyanoacrylate-tri-telechelic PIBs, (PIB-CA).sub.3).

(74) FIG. 4 displays the GPC trace of the first intermediate (allyl-), indicating the presence of a homogenous well-defined material with narrow molecular weight dispersity. The GPC traces of the other intermediates were similarly narrow.

(75) Polymerization of Oct-CA in Solution

(76) The polymerization was carried out in a 50 mL Schlenk flask equipped with a magnetic stir bar under UHP nitrogen. The flask was sealed with a rubber septum and purged with nitrogen and flame dried. Then 1.938 g (9.26 mmol) Oct-CA was placed in the flask and the polymerization started by injecting a DMT initiator (0.1224 g, 0.905 mmol, dissolved in 4.9 mL THF) under N.sub.2 (t=0); M.sub.0/I.sub.0=10, [Oct-CA]=1.32 mol/L. After about 8 hours of stirring at room temperature, the system was precipitated into methanol and dried in a vacuum at 60 C. overnight. FIG. 5 shows the H NMR spectrum.

(77) Polymerization of (PIB-CA).sub.3 and Copolymerization of (PIB-CA).sub.3 with Oct-CA

(78) Networks were prepared by polymerizing (PIB-CA).sub.3 and co-networks by copolymerizing (PIB-CA).sub.3 plus Oct-CA at room temperature. Thus, in a well-dried 10 mL screw-cap vial was placed (PIB-CA).sub.3 (0.3-1.2 g, M.sub.n=2500 or 6500 g/mol) dissolved in 3 mL freshly distilled THF. Polymerizations were induced by adding various amounts (5-150 L) of the initiator (DMT in bulk, [ABCO]=0.0042 mol/L, or [DABCO]=0.0045 mol/L) dissolved in THF to the monomer solution. The vial was capped; the solution was briefly vigorously manually mixed, and then rapidly poured into a 77 cm Teflon mold. The volatiles were evaporated in a fume hood for 1 hour, and the film was dried at 60 C. to constant weight.

(79) Copolymerizations were carried out similarly, except desired amounts of Oct-CA were added to the (PIB-CA).sub.3. It is important to note that while Me- and Et-CA are immiscible, Oct-CA is miscible with (PIB-CA).sub.3 of Mn=2500 g/mol, and the latter blends are optically clear. In contrast, Oct-CA plus (PIB-CA).sub.3 of Mn=6,500 g/mol yielded hazy mixtures, indicating that the higher molecular weight (PIB-CA).sub.3 is immiscible with Oct-CA Upon the addition of 10 wt % THF to the latter mixture, the (PIB-CA).sub.3 (Mn=6,500 g/mol)/Oct-CA system becomes optically clear, which indicates miscibility.

(80) Relative Initiator Reactivities

(81) The relative reactivity of various initiators was assessed by determining the time (in seconds) that a well-defined weight of liquid Oct-CA could be stirred after adding to it a well-defined amount (in moles) of initiator at room temperature. Thus, 1 g Oct-CA and a Teflon coated small (1 cm) magnetic stir bar were placed in a 5 mL screw cap vial. At time=0 a known amount of initiator (in some instances dissolved in toluene or THF) was injected into the monomer by means of a Hamilton syringe, the vial was capped and stirring (about 60 rpm, stirring plate) was started. The gel time, i.e., the time (in seconds) stirring stopped due to the viscosity increase of the monomer, was determined, and is considered to be a measure of relative initiator activity. Both the gel time and set time are measures of the rate of crosslinking, i.e., the time it takes to convert a liquid to a solid.

EXAMPLES

(82) The experiments began using DMT, a readily available aromatic tertiary amine, frequently used as an efficient initiator for the polymerization of CAs. In the course of this work it was hypothesized that cyclic aliphatic tertiary amines, e.g., ABCO and DABCO, would be more reactive initiators than DMT (or pyridine, or other frequently used tertiary amines) due to the presence of more available, less encumbered ring N atoms in these cyclic tertiary amines. While the dangling substituents attached to N in DMT are expected to sterically hinder the approach of monomer during initiation, the substituents on N in ABCO or DABCO are forced out-of-the-way so that initiation becomes less hindered, and consequently faster. The formulas help to visualize the substituent groups on the N in these tertiary amines:

(83) ##STR00006##
Faster initiation would desirably shorten set times, and thus the rate of crosslinking could be controlled by controlling (and hopefully, reducing) the amount of the initiator employed.

(84) FIGS. 6-11 summarize the results of a series of experiments whose aim was to assess the relative initiating efficiency of DMT (the control), and ABCO and DABCO for the polymerization of Oct-CA. Other initiators, namely triethylamine and pyridine, were also tested. Neither triethylamine nor pyridine are cyclic aliphatic tertiary amines. Pyridine is aromatic, while triethylamine is not cyclic. While DMT is a liquid and can be blended as such with the liquid monomer, ABCO and DABCO are crystalline solids and thus have to first be dissolved in a solvent (e.g., toluene or THF) to be administered. The data also shows the effect on the rate of the use of small amounts of toluene and tetrahydrofuran in conjunction with ABCO, DABCO, pyridine, and DMT. This solvent requirement in conjunction with ABCO and DABCO led to a further discovery in regard to the effect of solvent polarity on the rate of Oct-CA polymerization.

(85) The desirable range of set times is between 20 and 120 seconds. As shown in the inlay of FIG. 11, the bulk polymerization of Oct-CA induced by the well-known DMT initiator (the control) requires relatively large initiator concentrations (relatively low monomer/initiator ratios on the order of a molar ratio of about 50:1 to about 100:1) for efficient polymerization within the desired set time, and lesser amounts of initiator results in much longer set times. In this system a solvent is not needed to mix the initiator and the monomer because these liquids ingredients are miscible. Heat is generated upon mixing the DMT initiator with Oct-CA and, depending the amount of DMT used, the temperature of the system can rapidly rise to 50-70 C. The Oct-CA+DMT system exhibits a significant induction period (about 60 seconds) before rapid polymerization starts. During this induction period the liquid ingredients have time to mix thoroughly. After the induction period the viscosity of the system rapidly increases and conversions reach 90-95% within seconds. Conversions were determined by FTIR spectroscopy measuring the extent of monomer (unsaturation) remaining in the system. The poly(Oct-CA) that is forming is soluble in the monomer so the polymerization is homogeneous. The Oct-CA+DMT mixture is yellow, however the color disappears the instant a semisolid colorless transparent polymer product arises.

(86) As further shown in the inlay of FIG. 11, the polymerization of Oct-CA induced by the well-known DMT initiator in vol. 5% THF (the control) also requires relatively large initiator concentrations (relatively low monomer/initiator ratios on the order of a molar ratio of about 50:1 to about 200:1) for efficient polymerization within the desired set time, and lesser amounts of initiator results in much long set times. As compared to DMT in bulk, the rate of polymerization is faster as lesser amounts of initiator is used, but still the amount of initiator is relatively high (i.e. greater than 600:1).

(87) As clearly indicated by the data in FIG. 11, the polymerizations induced by ABCO and DABCO are much faster (monomer/initiator ratios are much larger and set times [shown as Stir-Stop Time in the figure] much shorter) than those induced by DMT. The rates are also strongly affected by the nature of the solvent used to dissolve these aliphatic cyclic amines. Thus, polymerizations induced by ABCO and DABCO dissolved in THF were much faster than those in which toluene was used as initiator solvent. This effect is most likely due to faster initiation in the more polar THF, i.e., due to the formation of THF-solvated propagating ions from the neutral starting materials. The dielectric constant of THF is significantly higher than that of toluene.

(88) In reviewing FIG. 11, it will be appreciated that the polymerization of 2-octyl cyanoacrylate and/or the copolymerization of 2-octyl cyanoacrylate and (PIB-CA).sub.3 induced by ABCO and DABCO are fast (on the order of less than 10 seconds to about 300 seconds depending upon the monomer-initiator ratio employed). Furthermore, the monomer-initiator molar ratio concentration will be essentially at least ten-fold, or even closer to twenty-fold or higher. That is, the molar ratio of monomer to initiator for the present invention, i.e., where the initiator is a cyclic tertiary aliphatic amine such as ABCO or DABCO in toluene solvent, can be upwards of 2,000:1 to obtain the same set time of about 100 seconds. Even more impressive is that the molar ratio of monomer to initiator for the present invention, i.e., where the initiator is a cyclic tertiary aliphatic amine such as ABCO or DABCO in THF solvent, can be upwards of 12,000:1 to obtain the same set time of about 100 seconds. This is a significant decrease in the amount of initiator needed to initiator the polymerization or copolymerization reactions.

(89) In light of the foregoing, it will be appreciated that ABCO and DABCO are efficient super-initiators not only for the homopolymerization of Oct-CA, but also for the copolymerization of Oct-CA with CA-telechelic polyisobutylenes, e.g., three-arm star CA-terminated PIB [(PIB-CA).sub.3]. These co-networks are useful rubbery wound closure adhesives. The rate of these copolymerizations can be significantly increased and the set times of these wound closure adhesives can be significantly shortened, by the use of ABCO or DABCO.