Photoinitiating Systems for LED-Cured Interpenetrating Polymer Networks

Transcription

1 Journal of Photopolymer Science and Technology Volume 28, Number 1 (2015) SPST Photoinitiating Systems for LED-Cured Interpenetrating Polymer Networks Suqing Shi 1,2, Feyza Karasu 1, Caroline Rocco 1, Xavier Allonas 1, and Céline Croutxé-Barghorn 1 * 1 Laboratory of Macromolecular Photochemistry and Engineering, University of Haute Alsace, 3b Rue Alfred Werner, Mulhouse Cedex, France *celine.croutxe-barghorn@uha.fr 2 College of Chemistry & Materials Science, Northwest University, Xi an, PR. China LED curing is becoming attractive in various applications due to specific characteristics such as long life-time, low heat generation and more energy saving. It requires the selection of suitable photoinitiating systems that exhibit long range absorption properties. For radical process, there are various commercially available photoinitiators showing a red-shift absorption. In contrast, most cationic photoinitiators have a short light absorption below 300 nm and display a limited overlap with the newly emerging visible LED light sources. Therefore, the development of photoinitiating systems with extended absorption in UVA ( > 380nm) is of paramount importance. Free radical promoted cationic photopolymerization (FRPCP) is considered as an elegant and fairly flexible type of cationic polymerization technique. This approach is explored for the curing of interpenetrating acrylate/epoxide polymer networks. Keyword: Free radical promoted cationic polymerization, Interpenetrated Polymer Networks 1. Introduction Nowadays, the current trend in photocuring is to use visible light sources arising from LED systems as they are preferred to the harmful radiations of conventional UV lamps. LED are quite attractive: they provide narrow bandwidth with stable irradiation, long life-time, low heat generation and more energy saving. The emission wavelengths of LED devices are commonly ranging in the visible region of the spectrum ( > 380 nm). LED curing of radical resins requires the selection of suitable photoinitiating systems that exhibit long range absorption. However, one of the most common drawbacks of LED systems is their low emission intensity which can cause an enhancement of the oxygen inhibition in the free radical photopolymerization. Indeed, molecular oxygen deactivates the excited states of the photoinitiator and scavenges the propagating radicals leading to peroxy radicals which are not able to continue the polymerization. Oxygen diffuses into the film and leads to a lack of conversion, tacky surface and heterogeneity in depth conversion of the coatings. Therefore, there are several physical methods -such as using inert atmosphere, oxygen barriers, improving irradiation conditions and photoinitiator concentration or chemical methods -such as using different photoinitiating systems or additives to overcome oxygen inhibition [1-5]. Amines, thiols, phosphines, boranes, germanes or metallic complexes are effective additives to obtain efficient photopolymerization under air atmosphere. In contrast to free radical photopolymerization, Received April 10, 2015 Accepted May 11,

2 LED curing of cationic resins is much more limited. Generally, cationic photopolymerization involves the generation of initiating species upon photolysis of onium salts. However, most of the currently commercial onium salts suffer from the short light absorption with max around 220~320 nm, resulting in inefficient cationic photopolymerization under LED [6-7]. To improve the overlap between the absorption wavelength of onium salts and these emerging visible light sources, design and development of various Cationic Photoinitiating Initiators (LWCPI) with extended absorptions in Long-Wavelength region ( max >350 nm), especially with sustainability, are highly focused in recent years. Among the different approaches reported in the literature, free radical promoted cationic photopolymerization (FRPCP) is considered as an elegant and fairly flexible type of cationic polymerization technique [8-13]. There are two feasible mechanisms to generate initiating species in presence of free radical source: (a) addition-fragmentation mechanism involving the addition of a free radical to an allynic onium salt and its subsequent fragmentation and (b) electron transfer (radical oxidation) mechanism involving the oxidation of a free radical by an appropriate onium salt. Efficient FRPCP usually requires the following conditions (i) long wavelength absorbing photoinitiators; (ii) radicals generation which are able to oxidized easily (with low oxidation potentials); (iii) presence of onium salts (with high reduction potentials). The present paper focus on the comparison of the polymerization efficiency of different LWCPIs in epoxide resin. In parallel, free radical polymerization of acrylate will be investigated under LED. Finally, photoinitiating systems for Interpenetrating Polymer Networks (IPNs) will be presented. IPNs are a combination of two or more polymer networks which are synthesized in the presence of another to form permanent entanglements without being bonded to each other covalently. They exhibit specific features such as high mechanical strength, good thermal stability and good chemical resistance. The use of interpenetrating polymers typically covers the area of damping materials, impact resistant materials and adhesives [14, 15]. 2. Experimental 2.1. Materials Polyethyleneglycol diacrylate (Sartomer SR-610) and (3, 4-epoxycyclohexane) methyl 3, 4-epoxycyclohexylcarboxylate (Dow Chemical, UVR 6110) were selected for this study. Bis (2,4,6-trimethylbenzoyl)-phenylphosphine oxide (BAPO), isopropyl-9h-thioxanthen-9-one (ITX), benzyl alcohol (BA), pentaerythritol tetrakis (3-mercaptopropionate) (PTTMP), N-Vinyl carbazole (NVK) and iodonium (4-methylphenyl) [4-(2-MethylPropyl) phenyl]- hexafluorophosphate(1-) (I250) were used in the different photoinitiating systems without any further purification. BAPO NVK PTTMP ITX I250 (or Ph 2 I + PF 6 ) BA Scheme 1. Chemical structures of the different compounds used in the photoinitiating systems Sample preparation and LED curing Photocurable formulations were prepared by mixing acrylate and/or epoxide resins with the photoinitiating systems (see indication in the figure caption). Formulations were applied on polypropylene substrates. Photocuring was performed at room temperature and under air, using a belt conveyor with an equivalent speed of 10 m.min -1. The LED source is a 395 nm Firejet FJ200 LED from Phoseon (12 W). 32

3 The light dose received by the samples for each pass (light exposure of 0.5s) was 310 mj/cm 2 (UVA: 10 mj/cm 2, UVV: 300 mj/cm 2 ) with an irradiance of 550 mw/cm 2 (after 10 passes). Tack-free films of about µm thick were obtained Evaluation of the conversion The neat acrylate and epoxide conversions were followed by FTIR with a Vertex 70 Bruker Optics. Conversions were calculated by following the C=C deformation band at 1640 cm 1 and the symmetrical stretching at 790 cm 1 for acrylate and epoxide, respectively. The acrylate/epoxide combination used in this study is not suitable for FTIR characterization due to overlapping of epoxide and acrylate vibrational bands and integration issues. Confocal Raman Microscopy (CRM) was thus implemented not to determine the kinetics, but the final conversion of acrylate and epoxide for each IPN network. The measurements were performed on an in Via Reflex Raman microscope from Renishaw which couples a Raman spectrophotometer with a Leica DM2500 microscope as described elsewhere [16]. 3. Results and Discussion 3.1. Photoinitiating systems for visible light curing Acrylate photopolymerization was initiated by Type I (exhibiting an absorption shifted to the red range) or Type II photoinitiators. LED curing of epoxides was carried out by free radical promoted cationic photopolymerization (FRPCP). Radicals generated by the previous mentioned photoinitiators can be oxidized by onium salts to generate reactive cations. By taking into account easy applicability and commercially availability, a few systems were chosen for this study. Radical and cationic polymerization was first performed separately to determine the efficiency of each system for both polymerization processes. Then, IPNs were photocured. Bisacyl phosphine oxide (BAPO) / I250 The radical photoinitiation mechanism is based on free radical generation produced via homolytic cleavage of BAPO with light absorption. Radicals are further oxidized by I250 to generate reactive cations as described in Scheme 2 (pathway 1). BAPO / N-vinylcarbazole (NVK) / I250 Since the oxidation potential of the radicals produced via BAPO photolysis is relatively high to enable efficient oxidation by iodonium salts, a third component, N-vinyl carbazole (NVK), was added to the photoinitiating system as a promoting agent. Radical resulting from BAPO photolysis will add to the double bond of NVK; a new carbon centered radical is produced, whose lower oxidative potential makes it easier to be oxidized by iodonium salts (Scheme 2, pathway 2). Ph 2 I + + N NVK R R Pathway 2 BAPO Visible light R Hardly oxidized Ph 2 I + + Ph +PhI R + + Ph + PhI Pathway 1 Scheme 2. Promoter assisted FRPCP of BAPO in presence of Ph 2 I + X - Isopropyl thioxanthone (ITX) / Pentaerythritol tetrakis (3-mercaptopropionate) (PTTMP) or Benzyl alcohol (BA) / I250 The radical photoinitiation mechanism is based on an excitation of a photosensitizer (ITX in this study) to triplet state. Hydrogen abstraction from H-donor (PEG-based monomers) results in a formation of ketyl radicals and alkyl radicals. These latter are 33

4 the only one able to initiate the polymerization. For cationic photoinitiation, besides direct interaction of iodonium salt with triplet state of ITX, benzyl alcohol or pentaerythritol tetrakis (3-mercaptopropionate) lead as previously described to the generation of a radical which is oxidized by iodonium salt to generate reactive cation species Polymerization kinetics in neat acrylate or epoxide resin Photopolymerizations were carried out under belt-conveyor at 10m.min -1 and 395 nm. As reported in Figure 1, for the three photoinitiating systems, acrylate conversion is higher than epoxide that exhibits a limited conversion of maximum 65% due to the fast vitrification and brittleness of the polyether network A 20 BAPO/NVK/Irg 250 BAPO/NVK/I250 ITX/Thiol ITX/PTTMP/I ITX/Benzyl ITX/BA/I250 alcohol Light dose (J/cm 2 ) B BAPO/NVK/Irg 250 BAPO/NVK/I250 ITX/Thiol ITX/PTTMP/I ITX/Benzyl ITX/BA/I250 alcohol Light dose (J/cm 2 ) Fig 1. Conversion profiles of epoxide A) at 790 cm -1 and acrylate (B) at 1640 cm -1 vs. light dose under 395 nm LED curing. [BAPO]: 2 wt%, [NVK]: 3 wt%, [ITX]: 2 wt%, [PTTMP]: 3 wt%, [BA]: 3 wt%, [I250]: 3 wt%). Better epoxide conversions and tack-free surfaces were obtained for ITX-based photoinitiating systems (Figure 1A). For radical photopolymerization, full and fast conversion was achieved with BAPO/NVK/I250. For the other photoinitiator combinations much higher light dose must be applied to obtain tack-free films. Comparison of polymerization kinetics of acrylate and epoxide as a function of time for a selected photoinitiating system highlights the possibility to further control the rate of reaction in IPN and thus the final properties (Figure 1). Indeed, radical polymerization is much faster than epoxide for BAPO/NVK/I250. In case of ITX/PTTMP/I250, the rate of both polymerizations is close to each other. Finally, for ITX/BA/I250, epoxide polymerization is faster than acrylate one LED-cured polymerization of acrylate /epoxide interpenetrating polymer networks Confocal Raman Microscopy (CRM) has been implemented to determine the final conversions for both radical and cationic polymerization. Vibrational band at 1640 cm -1 for C=C double bond stretching and 785 cm -1 for epoxy ring stretching have been followed. The conversion results can be seen in Table 1 for each PI system and for two IPN ratios: 70A30E and 30A70E. Table 1. CRM conversion of IPN systems for three different PIs. LED 395 nm. Different light doses were applied to obtain tack-free films. PI 1: BAPO (3 wt%) / NVK (3 wt %) / I250 (3 wt%), light dose: 11 J/cm 2 ). PI 2: ITX (2 wt%) / PTTMP (4 wt%) / I250 (3 wt%). PI 3: ITX (2 wt%) / BA (4 wt%) / I250 (3 wt%) IPN 70A30E 70 wt% acrylate 30 wt% epoxide 30A70E 30 wt% acrylate 70 wt% epoxide PI system by CRM Epoxide Acrylate BAPO/NVK/I ITX/PTTMP/I ITX/BA/I BAPO/NVK/I ITX/PTTMP/I ITX/BA/I

5 The trends observed for the homopolymer polymerization are confirmed in the IPNs studied. BAPO/NVK/I250 was proved to be very efficient for initiating radical polymerization. Therefore, the fast acrylate conversion in IPN limits the epoxide polymerization which levels to 55 or 28% for 70A30E and 30A70E respectively. Although the film was tack-free for 70A30E, the one prepared with the ratio of 30A70E was sticky but became completely tack-free after a few hours waiting due to post-polymerization of remaining epoxide functions. ITX/PTTMP/I250 films had similar physical behavior and conversion values as BAPO/NVK/I250 system. In contrast, ITX/BA/I250 enables higher epoxide conversion simultaneously to an efficient acrylate polymerization. Besides the better efficiency of this photoinitiating system, the plasticizing effect of acrylates may explain this result. These results highlight the possibilities of curing IPNs under visible LED irradiation and controlling the polymerization reaction between the two networks and finally the final material properties. 4. Conclusion Free radical promoted cationic photopolymerization (FRPCP) has become a highly focused technique in recent years due to the large choice of free radical sources and their tunable absorption wavelength. Most photoinitiating systems can effectively extend the absorption to long-wavelength range. Visible light LED curing become thus a reality. However, the effectiveness of generating efficient initiating species depends on different parameters. Photochemical amplified cationic process implementing the use of various promoters such as N-vinylcarbazole (NVK), or benzyl alcohols, significantly improves the cationic polymerization profiles and displays great potential in practical applications such as in Interpenetrating Polymer Networks. References 1. R. S. Davidson, in Radiation Curing in Polymer Science and Technology; Fouassier, J. P.; Rabek, J. F., Eds.; Elsevier Applied Science: London, (1993), Vol. 3, Chapter C. Belon, X. Allonas, C. Croutxé-Barghorn, J. Lalevee, J. Polym. Sci., Part A: Polym. Chem., 48 (2010) F. Courtecuisse, J. Cerezo, C. Croutxé-Barghorn, C. Dietlin, X. Allonas, J. Polym. Sci., Part A: Polym. Chem., 51 (2013) F. Courtecuisse, A. Belbakra, C. Croutxé-Barghorn, X. Allonas, C. Dietlin, J. Polym. Sci., Part A: Polym. Chem., 49 (2011), E. Andrzejewska, Polymer, 37 (1996) J.V. Crivello. Advance in Polymer Science, 62 (1984) 1. 7 A.W. Green in Industrial Photoinitiators. A Technical Guide CRC Press: New York. (2010). 8. P. E. Sundell, S. Jönsson and A. Hult, Polym. Sci. Part A: Polym. Chem., 29 (1991) G. Maniannan and J. P. Fouassier, J. Polym. Sci. Part A: Polym. Chem., 29 (1991) R. Rodrigues and M. G. Newmann, Macromol. Chem. Phys., 202 (2001) J. V. Crivello and M. Jang, Photochem. Photobiol. A: Chem., 159 (2003) J. V. Crivello and U. Bulut, Macromol. Symp., 202 (2006) 1; 13. J. V. Crivello and U. Bulut, J. Polym. Sci. Part A: Chem., 43 (2005) P. Deepalekshmi, P. M. Visakh, A. P. Mathew, A. Chandra, and S. Thomas, in Advances in Elastomers I Blends and Interpenetrating Networks, eds. Visakh, P.M., S. Thomas, A. K. Chandra, and A. P. Mathew, Springer-Verlag, Berlin Heidelberg, (2013). 15. M. de Brito, X. Allonas, C. Croutxé-Barghorn, M. Palmieri, C. Dietlin, S. Agarwal, D. Lellinger, I. Alig, Prog. Org. Coat., 73( 2 3) (2012) F. Courtecuisse, C. Dietlin, C. Croutxé-Barghorn, L. G. J. van der Ven, Appl. Spectrosc., 65 (2011)