Carbazole Derivatives with Thermally Activated Delayed Fluorescence

Jun 28, 2017 - Carbazole Derivatives with Thermally Activated Delayed Fluorescence Property as Photoinitiators/Photoredox Catalysts for LED 3D Printin...
0 downloads 0 Views 9MB Size
Article pubs.acs.org/Macromolecules

Carbazole Derivatives with Thermally Activated Delayed Fluorescence Property as Photoinitiators/Photoredox Catalysts for LED 3D Printing Technology Assi Al Mousawi,†,‡ Diego Magaldi Lara,§ Guillaume Noirbent,§ Frederic Dumur,*,∥ Joumana Toufaily,‡ Tayssir Hamieh,‡ Thanh-Tuan Bui,§ Fabrice Goubard,§ Bernadette Graff,† Didier Gigmes,∥ Jean Pierre Fouassier,† and Jacques Lalevée*,† †

UMR CNRS 7361 − UHA, Institut de Science des Matériaux de Mulhouse IS2M, 15, rue Jean Starcky, 68057 Mulhouse, Cedex, France Laboratoire de Matériaux, Catalyse, Environnement et Méthodes analytiques (MCEMA-CHAMSI), EDST, Université Libanaise, Campus Hariri, Hadath, Beyrouth, Liban § Laboratoire de Physicochimie des Polymères et des Interfaces LPPI, Université de Cergy-Pontoise, 5 mail Gay Lussac, Neuville-sur-Oise, 95031 Cergy-Pontoise, Cedex, France ∥ Aix Marseille Univ., CNRS, ICR UMR 7273, F-13397 Marseille, France ‡

S Supporting Information *

ABSTRACT: This paper is devoted to the effect of a thermally activated delayed fluorescence (TADF) property in new photoinitiators/photoredox catalysts. Four carbazole derivatives A1−A4 exhibiting a TADF character are synthesized and proposed for the first time as high performance visible light photoinitiators/metal-free photoredox catalysts, in the presence of an amine or/and an iodonium salt, for both the free radical polymerization (FRP) of (meth)acrylates and the cationic polymerization (CP) of epoxides upon visible light exposure using light-emitting diodes (LEDs) at 405, 455, and 477 nm. Interestingly, the impact of the substituent effect on the excited state lifetimes and therefore on the photoinitiating ability of a series of substituted carbazoles was clearly evidenced and examined. Upon bromination of the carbazole core, clear effects on the excited state lifetimes and light absorption were demonstrated, enabling to tune the initiator performance. Excellent polymerization initiating abilities are found, and high final monomer conversions are obtained. The use of these novel carbazolebased systems in photocurable cationic formulations for LED projector 3D printing is particularly outlined. TADF molecules allow a more efficient reaction from the excited singlet state as a result of their prolonged lifetimes; i.e., this effect is well highlighted through a comparison with previously published none-TADF metal-free photoredox catalysts. A full picture of the involved photochemical mechanisms is also provided. Carbazoles exhibiting a TADF character pave the way toward metal-free photoredox catalysts active in both oxidative and reductive cycles with efficiency on par with those of the traditional metal-based photoredox catalysts/photoinitiators.



INTRODUCTION Recently, carbazole derivatives have been investigated as additives for cationic polymerization (CP) (e.g., N-vinylcarbazole (NVK)1−4 or 9H-carbazole-9-ethanol (CARET)5) or as organic photoredox catalysts (OPC; they also referred to metal-free photoredox catalysts) active in both oxidative and reductive cycles to initiate the free radical polymerization (FRP) of meth(acrylates) or the CP of epoxides.6 These carbazolebased OPCs are fluorescent and characterized by a short excited singlet state (S1) lifetime (nanosecond scale) and photosensitize the decomposition of an iodonium salt to initiate a polymerization process. Recently, N-vinylcarbazole has been also reported as a versatile photoinaddimer of photopolymerization under UV LED.7 © XXXX American Chemical Society

It is generally agreed that photoinitiators (PI) operating in the S1 state should have lifetimes as long as possible to lead to high yields of initiating species through the PI/additive interaction. This was a question for most of the already used OPCs. Increasing the S1 lifetime remains a challenge in photochemistry/ photopolymerization processes to increase the yields of redox reaction from the S1 state. In 2012, Chihaya Adachi has popularized a new class of fluorescent materials based on carbazole moieties and characterized by a small S1−T1 energy splitting so that the S1 state of Received: May 28, 2017 Revised: June 13, 2017

A

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Δ(S1−T1) energy barrier can be easily overcomed by the absorption of heat and the up-conversion of the electrons of the triplet state to the singlet state provides compounds with excited state lifetimes comparable to that of phosphorescent materials, i.e., in the microsecond time scale. As interesting feature, these materials can be metal-free, addressing the issue of the transition metal toxicity. However, all the cabazoles already reported were efficient upon UV irradiation. Unfortunately, this technology is known to be potentially toxic for the operators and is associated with high-energy consumption. To date, such TADF materials were not reported as visible light photoredox catalysts for polymerization reactions.11 To get a deeper insight into this new generation of visible light photoredox catalysts exhibiting TADF character, four carbazole derivatives (A1−A4 in Scheme 1) have been examined as photoinitiators of polymerization, two of them (A1 and A4) being the starting compounds of refs 8 and 12 and exhibiting a thermally activated delayed fluorescence. Compounds A2 and A4 sport halogen substituents (it has been evidenced that the attachment of the halogen atoms to standard TADF molecules could considerably increase the delayed part of fluorescence to the desexcitation decay by increasing the rate of RISC due to the heavy atom effect11), whereas A1 and A3 are free of halogen. 9H-Carbazole-9-ethanol CARET used previously in our papers5,13 is set here as a reference carbazole derivative. These carbazoles will be incorporated into two-component (PI/iodonium salt (Iod) or PI/amine (EDB)) and sometimes three-component (PI/Iod/EDB) photoinitiating systems PISs to photoinduce the generation of reactive species (radicals or cations) for both the FRP of acrylates and methacrylates and/or the CP of a diepoxide upon near-UV or visible light (LED@375 nm, LED@405 nm, LED@455 nm, and LED@477 nm). The performance of these new TADF carbazoles will be also compared to the well-known BAPO photoinitiator (bis(2,4,6-trimethylbenzoyl)phenylphosphine oxide) or to other recently proposed

Scheme 1. Different Carbazole Derivatives (Noted A1−A4) Investigated in This Study

these materials can be thermally repopulated from the triplet state (T1) by mean of a reverse intersystem crossing (RISC), giving rise to a thermally activated delayed fluorescence (TADF).8 To reach this goal, the fundamental principle to construct these materials was the following: the highest occupied molecular orbital (HOMO) must be spatially separated from the lowest unoccupied molecular orbital (LUMO): this can be reached by mean of a rigid and highly twisted structure. By introducing a large dihedral angle and by interrupting the conjugation between the electron withdrawing and releasing units, the resulting materials could be circumvented from an excessive electron density delocalization, avoiding the overlap of the frontier molecular orbitals.9,10 By applying these design features of steric hindrance and consequent twist between the donors and the acceptors, a situation where the energy difference between the singlet and the triplet state (Δ(S1−T1)) could be lowered below 0.1 eV was achieved. As a consequence, the small Scheme 2. Other Used Chemical Compounds

B

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules carbazole derivatives but without TADF properties6 (e.g., C2 in Scheme 2). The use of these new high performance photosensitive systems in LED projector 3D printing experiments is also provided. The additional cyano group (for A1 and A4 compared to A2 and A3) as well as the halogen substituents introduced on the carbazole core (for A2 and A4) should affect the absorption as well as the photochemical/electrochemical properties for A1−A4. The expected long S1 lifetimes should be more favorable for the interactions with additives. A detailed analysis of the absorption properties, the steady state photolysis, the molecular orbitals, the excited state processes, and the production of radicals will be presented, and the structure/reactivity/ efficiency relationships are discussed.



Synthesis of 2,3,5,6-Tetrakis(3,6-dibromo-N-carbazolyl)terephthalonitrile A4. To a stirred solution of NaH (60% dispersed in mineral oil, 9 mmol) in 10 mL of dry DMF at 0 °C, 3,6-dibromocarbazole (487 mg, 1.5 mmol) is added in one portion under an argon atmosphere. The mixture is allowed to reach room temperature and stirred for 30 min. After the 2,3,5,6-tetrafluorobenzonitrile (40 mg, 0.2 mmol) is added, then the reaction mixture is heated up to 135 °C to let it react overnight. The reaction mixture is cooled to room temperature, solvent removed by decantation, and product filtered and abundantly washed with EtOH in order to remove the excess of reactants. The mixture was purified by column chromatography (hexane/ethyl acetate 1/1). A bright brown amorphous solid was obtained (20 mg, 13% yield). 1H NMR (CDCl3) δ 7.20−7.43 (m, 8H), 7.51−7.78 (m, 8H), 7.96−8.13 (m, 8H). HRMS (ESI MS) m/z: theor: 1411.5529 found: 1411.5532 ([M]+• detected). Anal. Calcd for C56H24Br8N6: C, 47.4; H, 1.7; N, 5.9. Found: C, 47.5, H, 1.6; N, 5.9%. Other Chemical Compounds. 9H-Carbazole-9-ethanol (CARET), phenyl-N-tert-butylnitrone (PBN), and ethyl 4-(dimethylamino)benzoate (EDB) were obtained from Sigma-Aldrich (Scheme 2). Bis(4-tert-butylphenyl)iodonium hexafluorophosphate (Iod or Speedcure 938) and bis(2,4,6-trimethylbenzoyl)phenylphosphine oxide (BAPO or Irgacure 819) were obtained from Lambson. Trimethylolpropane triacrylate (TMPTA) and (3,4-epoxycyclohexane)methyl 3,4- epoxycyclohexylcarboxylate (EPOX; Uvacure 1500) were obtained from Allnex and used as benchmark monomers for radical and cationic photopolymerization. Bisphenol A-glycidyl methacrylate (BisGMA), triethylene glycol dimethacrylate (TEGDMA), and urethane dimethacrylate (UDMA) were obtained from Sigma-Aldrich and used with the highest purity available. C2 was prepared by the procedure presented by us in ref 6. Irradiation Sources. The following light-emitting diodes (LEDs) were used as irradiation sources: (i) LED@375 nm = incident light intensity at the sample surface: I0 ≈ 40 mW cm−2; (ii) LED@405 nm

EXPERIMENTAL PART

Chemical Compounds. Synthesis of the Carbazole Derivatives (A1−A4). All reagents and solvents were purchased from Aldrich or Alfa Aesar and used as received without further purification. Mass spectroscopy was performed by the Spectropole of Aix-Marseille University. ESI mass spectral analyses were recorded with a 3200 QTRAP (Applied Biosystems SCIEX) mass spectrometer. The HRMS mass spectral analysis was performed with a QStar Elite (Applied Biosystems SCIEX) mass spectrometer. Elemental analyses were recorded with a Thermo Finnigan EA 1112 elemental analysis apparatus driven by the Eager 300 software. 1H and 13C NMR spectra were determined at room temperature in 5 mm o.d. tubes on a Bruker Avance 400 spectrometer of the Spectropole: 1H (400 MHz) and 13C (100 MHz). The 1H chemical shifts were referenced to the solvent peak DMSO (2.49 ppm), and the 13C chemical shifts were referenced to the solvent peak DMSO (49.5 ppm). All these carbazole photoinitiators were prepared with analytical purity up to accepted standards for new organic compounds (>98%) which was checked by high field NMR analysis. 2,3,5,6-Tetra(9H-carbazol-9-yl)terephthalonitrile A1 was synthesized as previously reported in the literature, without modifications and obtained in similar yields.12 Synthesis of 2,3,5,6-Tetrakis(3,6-dibromo-9H-carbazol-9-yl)benzonitrile A2. To a stirred solution of NaH (60% dispersed in mineral oil, 7.5 mmol) in dry DMF (8 mL) at 0 °C under an argon atmosphere, 3,6-dibromocarbazole (975 mg, 3.0 mmol) is added in one portion. The mixture is allowed to warm up to room temperature. After 30 min, 2,3,5,6-tetrafluorobenzonitrile (122 mg, 0.7 mmol) is added, and the reaction mixture is heated up to 135 °C overnight. The reaction mixture is then cooled, quenched with water, and extracted with CHCl3 (50 mL × 5). All organic fractions are gathered, washed with brine, dried with anhydrous MgSO4, and filtered. Solvent is removed under vacuum, and the product is abundantly washed with EtOH giving the desired product as bright yellow amorphous solid (400 mg, 40% yield). 1H NMR (DMSO-d6) δ 7.46 (dd, 8H, J = 8.1 Hz, J = 2.1 Hz), 7.80 (d, 8H, J = 8.5 Hz), 8.31 (d, 8H, J = 2.1 Hz), 8.93 (s, 1H). HRMS (ESI MS) m/z: theor: 1386.5577 found: 1386.5579 ([M]+• detected). Anal. Calcd for C55H25Br8N5: C, 47.3; H, 1.8; N, 5.0. Found: C, 47.6, H, 1.6; N, 5.2%. Synthesis of 2,3,5,6-Tetrakis(N-carbazolyl)benzonitrile A3. To a stirred solution of NaH (60% dispersed in mineral oil, 29 mmol) in dry DMF (25 mL) at 0 °C, carbazole (1.67 g, 10 mmol) is added in one portion under an argon atmosphere. The mixture is allowed to reach room temperature and stirred for 30 min. After the 2,3,5,6-tetrafluorobenzonitrile (280 mg, 1.6 mmol) is added, then the reaction mixture is heated up to 135 °C to let it react overnight. The reaction mixture is cooled to room temperature, solvent removed by decantation, and product filtered and abundantly washed with EtOH in order to remove the excess of reactants. One product is obtained. A bright yellow amorphous solid was obtained (600 mg, 50% yield). 1H NMR (CDCl3) δ 7.09−7.23 (m, 16H), 7.3−7.41 (m, 8H), 7.74−7.85 (m, 8H), 8.45 (s, 1H). 13C NMR (CDCl3) δ 109.2, 109.9, 113.0, 118.2, 120.2, 120.3, 123.9, 124.2, 125.5, 125.7, 136.6, 136.9, 137.8, 138.8, 139.2. HRMS (ESI MS) m/z: theor: 763.2736 found: 763.2735 ([M]+• detected). Anal. Calcd for C55H33N5: C, 86.5; H, 4.3; N, 9.2. Found: C, 86.6, H, 4.6; N, 9.2%.

Figure 1. Absorption spectra of the carbazole derivatives in DCM and CARET in methanol.

Table 1. Absorption Properties for A1−A4 vs CARET and BAPO

A1 A2 A3 A4 CARET BAPO a

C

λmax (nm)/ ε (M−1 cm−1)

ε@405 nm (M−1 cm−1)

ε@455 nm (M−1 cm−1)

ε@470 nm (M−1 cm−1)

330/∼8800 340/∼40000 333/∼33000 349/∼18000 343/∼4100 370/∼950

∼1350 ∼7800 ∼5700 ∼3300 n.a.a ∼500

∼2500 ∼450 ∼50 ∼1200 n.a. n.a.

∼2500 n.a. n.a. ∼800 n.a. n.a.

n.a: no absorption of light. DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 2. Contour plots of HOMOs and LUMOs for the A1−A4 structures optimized at the B3LYP/6-31G* level of theory. (I0 ≈ 110 mW cm−2); (iii) LED@455 nm (I0 ≈ 80 mW cm−2), (iv) LED@477 nm (I0 ≈ 80 mW cm−2). Free Radical (FRP) and Cationic (CP) Photopolymerization. The two-component photoinitiating systems (PISs) are mainly based on A/iodonium salt (0.5%/1% w/w) for both CP and FRP. The weight percent of the photoinitiating system is calculated from the monomer content. The photosensitive thin formulations (∼25 μm of thickness) were deposited on BaF2 pellets under air for the CP of EPOX, while for

the FRP of TMPTA and BisGMA/TEGDMA it was done in laminate (the formulation is sandwiched between two polypropylene films to reduce the O2 inhibition). The 1.4 mm thick samples of BisGMA/ TEGDMA were also polymerized under air. The evolution of the epoxy group content and the double bond content of (meth)acrylate functions were continuously followed by real time FTIR spectroscopy (JASCO FTIR 4100) at about 790 and 1630 cm−1, respectively. The evolution of the methacrylate characteristic D

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules peak for the thick samples (1.4 mm) was followed in the near-infrared range at ∼6160 cm−1. The procedure used to monitor the photopolymerization profile has been described in detail in refs 14 and 15. Redox Potentials. Carbazole derivatives oxidation potentials (Eox and Ered vs SCE) were measured in acetonitrile by cyclic voltammetry with tetrabutylammonium hexafluorophosphate (0.1 M) as the supporting electrolyte. The free energy change ΔGet for an electron transfer reaction was calculated from the equation16

ΔGet = Eox − Ered − E* + C

(see also Table 1). The three new compounds (A2−A4) are characterized by high molar extinction coefficients ε in the 300−430 nm range (e.g., ε(A2) ∼ 7800 M−1 cm−1 at 405 nm). The absorptions of the known A1 and A4 compounds are lower (ε(A1) ∼ 1350 M−1 cm−1 and ε(A4) ∼ 3300 M−1 cm−1 at 405 nm), but they spread up to 530 nm (ε(A1) = 2500 M−1 cm−1 and ε(A4) = 1200 M−1 cm−1 at 455 nm). Remarkably, the A1−A4 absorptions are excellent in the 300−550 nm spectral range, ensuring a good overlap with the emission spectra of the LEDs used in this work (e.g., at 375, 405, 455, and 477 nm). The optimized geometries as well as the frontier orbitals (highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO)) are shown in Figure 2. In fact, the HOMO is localized on the electron-donating carbazole moieties, whereas the LUMO is located on the electron-accepting cyano functions of the benzonitrile central core. As expected from the design of a TADF material, a complete isolation of the electron releasing and electron withdrawing parts is observed by means of the highly twisted structure. As evidenced in Figure 2, it is obvious that the clear charge transfer character corresponding to delocalization of electrons upon HOMO to LUMO transition (π → π* lowest energy transition) due to light excitation. On a related level, the presence of an additional cyano substituent for A1 (compared to A3) on the benzonitrile central core leads to a clear red-shift in its spectrum. By increasing the electron accepting ability of the central core, the energy level of the LUMO is lowered thereby red-shifting the charge transfer band, the HOMO level of the spatially isolated carbazole being unchanged. In the same context, the addition of bromide groups on the para position relative to the carbazole scaffold nitrogen group also leads to a slight red-shift in the spectrum (A2 compared to A3) inducing consequently an increase of HOMO. Photophysical and Electrochemical Properties of A1−A4. The redox potentials of the carbazole derivatives were measured in acetonitrile by cyclic voltammetry (CV) (Figure 3A and Table 2). The excited state energy is evaluated from the crossing point of the absorption and fluorescence spectra (e.g., for A2 ES1 = 2.8 eV; Figure 3B and Table 2). The A1−A4 fluorescence lifetimes (τ) were determined using laser flash photolysis LFP experiments. In fact, as expected, the A1−A4 singlet state lifetimes are very long (e.g., τA3 = 7.24 μs; Table 2). The fluorescence is quenched by oxygen (e.g., for A2; τA2 = 7.8 μs (under N2) and 755 ns (under air); the quenching rate constant kq = 5.4 × 108 M−1 s−1 in DCM was obtained by a

(1)

where Eox, Ered, E*, and C are the oxidation potential of the electron donor, the reduction potential of the electron acceptor, the excited state energy level (determined from luminescence experiments, see Fluorescence Experiments section) and the Coulombic term for the initially formed ion pair, respectively. C is neglected as usually done in polar solvents. ESR Spin Trapping (ESR-ST) Experiments. The ESR-ST experiments were carried out using an X-band spectrometer (Bruker EMXPlus). LED@405 nm was used as irradiation source for triggering the production of radicals at room temperature (RT) under N2-saturated tert-butylbenzene and trapped by phenyl-N-tert-butylnitrone (PBN) according to a procedure described elsewhere in detail.17 The ESR spectra simulations were carried out with the PEST WINSIM program. Light Absorption Properties. The UV−vis absorption properties of the compounds were studied using a JASCO V730 spectrometer. Fluorescence Experiments. The fluorescence properties of the compounds were studied using a JASCO FP-6200 spectrometer. Computational Procedure. Molecular orbital calculations were carried out with the Gaussian 03 suite of programs.18,19 The electronic absorption spectra for the different compounds were calculated with the time-dependent density functional theory at the mPW1PW91/6-31G* level of theory on the relaxed geometries calculated at the UB3LYP/ 6-31G* level of theory. 3D Printing Experiments. For 3D printing experiments, a LED projector @405 nm (Thorlabs) was used. The photosensitive cationic resin was polymerized under air, and the generated patterns were analyzed by a numerical optical microscope (DSX-HRSU from OLYMPUS Corp.).20−22 Laser Flash Photolysis (LFP). Nanosecond LFP experiments were carried out using Luzchem LFP 212 spectrometer. For the excitation, a Q-switched nanosecond Nd/YAG laser (λexc = 355 nm, ∼6−8 ns pulses; energy reduced down to 10 mJ) from Continuum (Minilite) was used.17



RESULTS AND DISCUSSION Light Absorption Properties of A1−A4. The UV−vis absorption spectra of the new proposed photoinitiators/photoredox catalysts in dichloromethane (DCM) are reported in Figure 1

Figure 3. (A) Reduction potential determination of A3 in acetonitrile. (B) Singlet state energy determination. E

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules Stern−Volmer treatment (Figure S1 in Supporting Information with [O2] = 2.2 × 10−3 M).23 Cationic Photopolymerization (CP) of Epoxides. A full study on the CP of diepoxides (e.g., EPOX) in thin films (∼25 μm) was performed under air using the new compounds A1−A4. Initially starting with LED@405 nm as a convenient soft irradiation source, the CP in the presence of two-component photoinitiating systems based on A/Iod combinations (0.5%/1% w/w) is very efficient in term of Rp (rate of polymerization) and final

epoxy function conversion (FC) (e.g., FC = 54% with both A2 and A3 compared to 51% and 47% with A4 and A1, respectively; Figure 4A, curves 4 and 3 vs curves 5 and 2, respectively; Table 3). The same holds true but with slightly lower FCs using higher wavelengths such as LED@455 nm (e.g., FC ∼54% and ∼46% for A3 and A2, respectively, compared to 34% and 27% for A4 and A1, respectively; Figure 4B, curves 3 and 2 vs curves 4 and 1, respectively; Table 3). The lower FCs can be likely partly related to the fact that the LED@455 nm has a lower intensity (see the Experimental Part). A new peak ascribed to the polyether network arises at ∼1080 cm−1 (Figure 3C) in the FTIR spectra. Iod alone does not initiate the polymerization. Remarkably, a very high rate of polymerization (Rp), when using both LED@405 nm and LED@455 nm, was clearly achieved with the A3/Iod system compared to A2/Iod and A1/Iod systems or to CARET/Iod set as a reference for which almost no polymerization occurs (i.e., in Figure 4D, the polyether peak is not observed for CARET/Iod compared to A2/Iod system as e.g. in Figure 4C). The comparison of the As with CARET is convenient (all compounds hold up at least one carbazole moiety). In an additional comparison with a well-known BAPO/Iod1

Table 2. Photophysical/Photochemical and Electrochemical Properties of A1−A4

A1 A2 A3 A4

Eox [V]a

Ered [V]a

ES1 [eV]

τb [μs] (under air; under N2) in DCM

ET1 [eV]

1.61 1.62 1.61

A4 > A1 ≫ BAPO > CARET. Obviously, this behavior seems not to be conveniently connected to the absorption properties of the carbazole derivatives as A3 being the most efficient PI and having lower extinction coefficients than the others (e.g., ε@455 nm is ∼50 M−1 cm−1 compared to that of A2 and A1 (∼450 and ∼2500 M−1 cm−1) at 455 nm; see Table 1). This behavior calls attention and thus will be discussed below in detail. Free Radical Photopolymerization (FRP). Photopolymerization of Acrylates (TMPTA). The FRP of TMPTA in thin films (∼25 μm), in laminate, and in the presence of the different A/Iod or A/EDB couples (Figure 5 and Table 3) is quite efficient using LED@405 nm (Iod alone, EDB alone, or A alone does not work). The experimental results show that carbazole derivatives are quite efficient in both photo-oxidation processes (electron transfer from *A to Iod) as well as photoreduction processes (electron transfer from EDB to *A). This efficient dual behavior will be discussed below. The efficiency trend in FRP of the A/Iod or A/EDB couples using LED@405 nm is in accordance with the trend observed in CP (i.e., A3 > A2 > A4 > A1 ≫ CARET). In comparison with a well-known radical initiator BAPO1 (0.5% w/w) for FRP, A3/Iod is found quite efficient (e.g., FC ∼ 61% for BAPO (0.5% w/w) and FC ∼ 58% for A3/Iod (0.5%/1% w/w). The A3/Iod two-component PIS exhibits a noteworthy efficiency: FC ∼ 58% > FC of A2 − 50% > FC of A4 − 47% > FC of A1 − 23% (Figure 5A, curve 4 vs curve 3 vs curve 5 vs curve 2, respectively; Table 3). Similarly, the A3/EDB couple exhibits a superiority over the others: FC ∼ 48% ≫ FC of A2 − 31% ≫ FC of A1 − 15% (Figure 5B, curve 3 vs curve 2 vs curve 1, respectively; Table 3). Thus, A3 is considered as the most efficient photoinitiator among the present series of carbazoles. Remarkably, the photo-oxidative process in A/Iod couples shows preponderance over the photoreductive process in A/EDB (e.g., FC ∼ 50% with A2/Iod vs 31% with A2/EDB; Figure 5D curve 2 vs curve 1; Table 3). The same holds true with A1 (Figure 5C, curve 2 vs curve 1; Table 3) and A3 (Figure 5E, curve 2 vs curve 1; Table 3). The structure/reactivity relationships will also be discussed below. A photoredox catalyst behavior was observed upon using the A/Iod/EDB three-component PIS under exposure to the LED@405 nm. A clear increase of the performance is noted (e.g., FC increases up to 47% with A1/Iod/EDB compared to 23% and 15% with A1/Iod and A1/EDB, respectively; Figure 5C, curve 3 vs curves 2 and 1; Table 3). Similarly, with A2 and A3 an enhancement is well observed (e.g., FC = 56% with A2/Iod/ EDB compared to 50% and 31% for both A2/Iod (0.5%/1%) and A2/EDB, respectively; Table 3). In comparison with our recently published none-TADF carbazole derivatives (C1−C4),6 the maximal acrylate function conversion (FC) of TMPTA monomer previously reached was about 50% with two-component PIS C2/Iod (0.5%/1% w/w) (Figure 5, curve 4) and 57% with three-component PIS C2/Iod/ EDB (0.5%/1%/1% w/w), whereas higher FCs are reached with our new TADF molecules (e.g., FC = 58% with two-component PIS A3/Iod (0.5%/1% w/w) (Figure 5, curve 2) and 62% with three-component PIS A3/Iod/EDB (0.5%/1%/1% w/w)

n.p. = no polymerization. From ref 6.

61% LED@405 nm b a

57% 50% n.p.a n.p.

A1 A2 A3 A4 C2b CARET BAPO

12% (at t = 400 s) 15%

47% 56% 62% 15% 31% 48%

23% LED@405 nm; 19% LED@455 nm 50% LED@405 nm; 23% LED@455 nm 58% LED@405 nm; 47% LED@405 nm 56% LED@405 nm

n.p. 65% LED@405 nm; 5% LED@477 nm

A/Iod (0.5%/1% w/w) LED@405 nm A/Iod/EDB (0.5%/1%/1% w/w) LED@405 nm A/EDB (0.5%/1% w/w) A/Iod (0.5%/1% w/w)

LED@455 nm A /Iod(0.5%/1% w/w) 27% 46% 54% 34% LED@405 nm A /Iod(0.5%/1% w/w) 47% 55% 55% 50% LED/PIS

methacrylate function conv FC for BisGMA/TEGDMA (%) (at t = 100 s): thick sample (1.4 mm) under air acrylate function conv FC for TMPTA (%) (at t = 100 s): thin sample (25 μm) in laminate epoxy function conv FC (%) for EPOX (at t = 800 s): thin sample (25 μm) under air

Table 3. Final Reactive Function Conversions (FC): Epoxy for EPOX; Acrylate for TMPTA and Methacrylate for BisGMA/TEGDMA Using Different Photoinitiating Systems and Different LEDs for Irradiation

Macromolecules

G

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 5. (A) Polymerization profiles of TMPTA (acrylate function conversion vs irradiation time) in laminate upon exposure to LED@405 nm in the presence of (1) A1/Iod (0.5%/1% w/w); (2) A2/Iod (0.5%/1% w/w); (3) A3/Iod (0.5%/1% w/w); (4) A4/Iod (0.5%/1% w/w). (B) Polymerization profiles of TMPTA (acrylate function conversion vs irradiation time) in laminate upon exposure to LED@405 nm in the presence of (1) A1/EDB (0.5%/1% w/w); (2) A2/EDB (0.5%/1% w/w); (3) A3/EDB (0.5%/1% w/w). (C) Polymerization profiles of TMPTA (acrylate function conversion vs irradiation time) in laminate upon exposure to LED@405 nm in the presence of (1) A1/EDB (0.5%/1% w/w); (2) A1/Iod (0.5%/1% w/w); and (3) A1/Iod/EDB (0.5%/1%/1% w/w). (D) Polymerization profiles of TMPTA (acrylate function conversion vs irradiation time) in laminate upon exposure to LED@405 nm in the presence of (1) A2/EDB (0.5%/1% w/w); (2) A2/Iod (0.5%/1% w/w); and (3) A2/Iod/EDB (0.5%/1%/1% w/w). (E) Polymerization profiles of TMPTA (acrylate function conversion vs irradiation time) in laminate upon exposure to LED@405 nm in the presence of (1) A3/EDB (0.5%/1% w/w); (2) A3/Iod (0.5%/1% w/w); and (3) A3/Iod/EDB (0.5%/1%/1% w/w); (4) C2/Iod (0.5%/1% w/w) from ref 6. The irradiation starts at t = 10 s. H

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 6. (A) Polymerization profiles of BisGMA/TEGDMA (methacrylate function conversion vs irradiation time) under air (1.4 mm thick sample) in the presence of two-component photoinitiating system A2/Iod (0.5%/1% w/w) upon exposure to (1) LED@405 nm and (2) LED@477 nm. The irradiation starts at t = 10 s. (B) Photos of a BisGMA/TEGDMA thick film (1.4 mm) upon irradiation with the LED@405 nm for 100 s in the presence of A2/Iod (0.5%/1% w/w) under air before and after polymerization. The irradiation starts for t = 10 s.

the A3/Iod (0.5%/1% w/w) system with urethane dimethacrylate (UDMA) resin. Thick samples up to 500 μm were obtained with high spatial resolution and short writing time ( 370 nm) is formed. Similarly, in the photolysis of A2/EDB, another photoproduct is observed after short irradiation times (e.g., 2 min Figure 9C) and then is quickly subjected to bleaching as the irradiation is going on. Accordingly, it is noteworthy to mention that the photobleaching character is in line with the high reactivity/efficiency of the systems in polymerization. A1−A4 as Electron Donors in the A/Iod Two-Component PIS. The free energy change ΔGet for the electron transfer reaction between A1−A4 as electron donors and Iod as an electron acceptor was calculated from the classical Rehm−Weller equation (eq 1) and using the oxidation potentials Eox and the excited state energies ES1 of A1−A4 (Table 3). Remarkably, 1 A/Iod is highly favorable (e.g., ΔGet ∼ −1.1 eV for 1A3/Iod) compared to the other 1A/Iod (e.g., ΔGet ∼ −1.02 eV for 1A2/Iod and ∼ −0.69 eV for 1A1/Iod; Table 4), and high electron transfer quantum yields (ϕet) are calculated according to eq 2 for 1A3/Iod compared to the other couples, e.g., ϕet (1A3/Iod) ∼ 0.83 > ϕet (1A2/Iod) ∼ 0.78 ≫ ϕet (1A1/Iod) ∼ 0.25). As known, such electron transfer reactions are followed by the formation of reactive species (Ar• and A•+) capable of initiating FRP and CP (r1−r2).

(Figure 5, curve 3, and Table 3) under the same conditions. Thus, we can claim that the new TADF molecules are better photoinitiators/photoredox catalysts in terms of efficiency (FC) but also rate of polymerization (Rp) for FRP compared to our previous none-TADF carbazole-based molecules in FRP. This will be discussed in terms of electron transfer quantum yields for photoinitiator/additives interaction in the photochemical mechanisms. Nevertheless, it is hard to compare the two behaviors for CP, since the radical cation plays a crucial role in CP and therefore the chemical structure of both TADF and none-TADF molecules is another parameter to be taken into account. Photopolymerization of Methacrylates. Interestingly, the A2/Iod (0.5%/1% w/w) couple efficiently initiates the FRP of 1.4 mm thick samples of a blend of methacrylates (Bis-GMA/ TEGDMA 70%/30% w/w) under air upon irradiation with the LED@405 nm (Figure 6A, curve 1). An inhibition time for polymerization under air is observed. Remarkably, a tack-free polymer is obtained and exhibits good bleaching properties (photo taken before and after polymerization; Figure 6B). However, a very poor performance of A2/Iod was observed when using LED@477 nm (methacrylate conversion FC for BisGMA/ TEGDMA FC ∼ 5%; Figure 6A, curve 2). This is related to the fact that A2 has no absorption in the emission region of LED@477 nm (Figure 1 and Table 1). 3D Printing Using A3/Iod Based System upon LED@ 405 nm Projector. Seeking out low shrinkage effect that usually occurs in radical photopolymerization, 3D printing experiments were carried out through a cationic process. Using a LED projector at 405 nm (Thorlabs, 110 mW/cm2), some 3D printing experiments (Figure 7) were carried under air using the A3/Iod (0.5%/1% w/w) system which was characterized by high reactivity in the CP of epoxides (see above). This LED projector technology really shows an advanced step among other laserbased 3D printing strategies as the entire layer is projected at one time. By means of a numerical optical microscope, profilometric experiments were also carried out showing clearly the dimensions of the printed specimen (Figure 7C or 7E). More profilometric experiments were carried out showing the good spatial resolution that could be achieved with the proposed system (Figure S2). Laser Write Experiments Using A3/Iod System. By means of laser diode at 405 nm of intensity 100 mW/cm−2, laser write experiments were successfully performed under air using

ϕet(S1) = K sv[Iod]/(1 + KSV[Iod])

(2)

A → 1A(hv)

(r1)

A + Ar2I+ → A•+ + Ar2I• → A•+ + Ar • + PhI

(r2)

1

The fluorescence emission is gradually quenched by Iod (Figure 10A). This quenching was also carried out in LFP experiments using eq 3 (e.g., kq = ∼2 × 109 M−1 s−1 for A3/Iod; Figure 10B) (e.g., Stern−Volmer plot of the 1A lifetime vs [Iod] I

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 7. Cationic photopolymerization experiments using a LED projector at 405 nm: (A) number (5) and patterns. (B) Letters. (C) Characterization in 3D by profilometry of “1” thin sample (30 μm). (D) number (“0”). (E) Characterization in 3D by profilometry of thick sample (400 μm) easily observed by using a numerical optical microscope.

is given in Figure S3). A new transient species was observed (Figure 10C) which may correspond to the formation of the A3•+ radical cation (r2). The presence of Ar• (r2) is confirmed by ESR-ST. Indeed, the phenyl radicals (Ar•) were easily detected as PBN/Ar• radical adducts in the irradiated

A/Iod systems (e.g., for A2/Iod; Figure 11; hyperfine coupling constants (hfcs): aN = 14.2 G and aH = 2.14 G; reference values in ref 17). 1/τ = 1/τ0 + kq[Iod] J

(3) DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 8. Free radical photopolymerization experiments for laser write at 405 nm: (A) “IS2M” logo; (B) characterization in 3D by profilometry of thick sample (500 μm).

Figure 9. (A) Photolysis of A2 alone, (B) A2/Iod photolysis, and (C) A2/EDB photolysis. Upon exposure to LED@375 nm in DCM. K

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

the formation of the EDB(−H•) aminoalkyl radical (r4) capable of initiating FRP. In fact, no clear reduction peak was obtained for A1 or A2 using CV, but upon comparison with the blank (just with electrolyte), the reduction potentials can be estimated from the variation of the curve shape (e.g., Ered for A2 < −1.5 eV; Table 4). For the free energy changes the photoreduction process is less favorable than photooxidation process (A/Iod), e.g., ΔGet 1A2/EDB > −0.2 eV vs 1A2/Iod ∼ −1.02 eV (Table 4). However, the bimolecular quenching fluorescence quenching rate constants are high (e.g., KSV = 382 M−1 for A2/EDB; using the lifetime determined in DCM: kq = ∼2.7 × 108 M−1 s−1) showing favorable (r3). Accordingly, the electron transfer quantum yield is almost quantitative (0.95; Table 4).

Table 4. Parameters Characterizing the Chemical Mechanisms Associated with the A/Iod and A/EDB Systems in Acetonitrile and DCM A1/Iod A2/Iod A3/Iod A4/Iod A1/EDB A2/EDB A3/EDB

ΔGetS1a [eV]

KSV (M−1)

Φet(S1)

ΔGetT1a [eV]

−0.69 −1.02 −1.1

18 210 274 200

0.25 0.788 0.835 0.7

−0.31 −0.76

382

0.95

>0.1 >−0.2 −0.2

ES1(A4) = 2.72 eV > ES1(A1) = 2.5 eV; Table 2). Remarkably, the reactivity (A3 > A2 > A4 > A1) follows the electron transfer quantum yield for (r2) (Table 4) showing the crucial role of the oxidation process generating A•+ and Ar• to initiate CP and FRP, respectively. For FRP, TADF molecules (A1−A4) are better PIs/PCs than none-TADF carbazoles (C1−C4)6 as mentioned above. In fact, to support this claim, a comparison is set for electron transfer quantum yield calculations originated from singlet states (e.g., ϕetS1= 0.835 for A3/Iod vs ϕetS1= 0.05 for C3/Iod in ref 6). Accordingly, these data explain the main interest from synthesizing the new TADF molecules, as S1 is thermally populated allowing the reaction through a singlet state pathway.

Figure 11. ESR-ST spectra obtained upon irradiation (LED@405 nm) of a A2/Iod solution: experimental (lower spectrum), and simulated (upper spectrum) in tert-butylbenzene as a solvent.

Scheme 3. Carbazole Photoredox Catalytic Cycles



CONCLUSION In the present paper, carbazole derivatives characterized by a TADF behavior are proposed for the development of new high performance photoinitiators or metal-free photoredox catalysts upon near-UV or visible LEDs for the photoinitiation of both the cationic polymerization of epoxides and the free radical polymerization of (meth)acrylates. High final conversions and polymerization rates are obtained. These new initiating systems incorporated in photocurable cationic resins have found an interesting application in LED projector 3D printing. This unique property of delayed fluorescence issued from the up-conversion of the electrons of the triplet state to the singlet state confers to A1−A4 a clear elongation of the excited state lifetime. By means of their structural specificity (i.e., a high internal twist), these molecules clearly outperform the traditional photoinitiators (BAPO) for cationic polymerization and even most of the previously reported carbazole-based photoinitiators for radical polymerization. With high molecular weights, a low migration of these PIs/PCs can also be expected. Future developments of other carbazole derivatives activable at other irradiation wavelengths and suitable for 3D printing resins are currently in progress.

quantum yields are ∼0.8 and ∼0.95 with Iod and EDB, respectively (Table 4). The A•+ radical cation is reduced by EDB as the reduction potential of A•+ (e.g., 1.62 V for A2•+ in Table 4) is higher than the oxidation potential of EDB (1.1 V). In the same way, the A•− radical anion, which can act as inhibitor for CP, is oxidized easily by Iod as the reduction potential of Iod (−0.2 V) is higher than the oxidation potential of A•− (Table 2). The concomitant regeneration of A ensures a photoredox catalyst behavior in line with the observed improved efficiency of the polymerization (A1/Iod/EDB better than A1/Iod and A1/EDB). Structure/Reactivity/Efficiency Relationships. In both FRP and CP, A3 is much more efficient compared to other As. It is usually expected that compounds having higher molar extinction coefficients (they absorb more light as a better overlapping with the irradiation source emission occurs) lead to higher photoinitiating abilities. However, in our case, A3 having the lowest extinction coefficient at 455 nm among the A series shows a better efficiency and reactivity in FRP and CP compared to the other As (A3 > A2 > A4 > A1) despite the fact that their



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.macromol.7b01114. Figure S1: Stern−Volmer plot of the 1A2 lifetime vs [O2]; Figure S2: characterization in 3D by profilometry of thick sample (200 μm) by using a numerical optical microscope; Figure S3: Stern−Volmer plot of the 1A2 lifetime vs [Iod] (PDF) M

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules



Resins for LED Projector 3D Printing Applications. Macromolecules 2017, 50 (3), 746−753. (14) Lalevée, J.; Blanchard, N.; Tehfe, M.-A.; Peter, M.; Morlet-Savary, F.; Gigmes, D.; Fouassier, J. P. Efficient Dual Radical/Cationic Photoinitiator under Visible Light: A New Concept. Polym. Chem. 2011, 2 (9), 1986−1991. (15) Lalevée, J.; Blanchard, N.; Tehfe, M.-A.; Peter, M.; Morlet-Savary, F.; Fouassier, J. P. A Novel Photopolymerization Initiating System Based on an Iridium Complex Photocatalyst. Macromol. Rapid Commun. 2011, 32 (12), 917−920. (16) Rehm, D.; Weller, A. Kinetics of Fluorescence Quenching by Electron and H-Atom Transfer. Isr. J. Chem. 1970, 8, 259−271. (17) Lalevée, J.; Blanchard, N.; Tehfe, M.-A.; Morlet-Savary, F.; Fouassier, J. P. Green Bulb Light Source Induced Epoxy Cationic Polymerization under Air Using Tris(2,2′-bipyridine)ruthenium(II) and Silyl Radicals. Macromolecules 2010, 43 (24), 10191−10195. (18) Foresman, J. B.; Frisch, A. Exploring Chemistry with Electronic Structure Methods, 2nd ed.; Gaussian Inc.: Pittsburgh, PA, 1996. (19) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, J. R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; M. Wong, W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 03, Revision B-2; Gaussian Inc.: Pittsburgh, PA, 2003. (20) Zhang, J.; Dumur, F.; Xiao, P.; Graff, B.; Bardelang, D.; Gigmes, D.; Fouassier, J. P.; Lalevée, J. Structure Design of Naphthalimide Derivatives: Toward Versatile Photoinitiators for Near-UV/Visible LEDs, 3D Printing, and Water-Soluble Photoinitiating Systems. Macromolecules 2015, 48 (7), 2054−2063. (21) Xiao, P.; Dumur, F.; Zhang, J.; Fouassier, J. P.; Gigmes, D.; Lalevée, J. Copper Complexes in Radical Photoinitiating Systems: Applications to Free Radical and Cationic Polymerization upon Visible LEDs. Macromolecules 2014, 47 (12), 3837−3844. (22) Al Mousawi, A.; Kermagoret, A.; Versace, D.-L.; Toufaily, J.; Hamieh, T.; Graff, B.; Dumur, F.; Gigmes, D.; Fouassier, J. P.; Lalevée, J. Copper Photoredox Catalysts for Polymerization upon near UV or Visible Light: Structure/Reactivity/Efficiency Relationships and Use in LED Projector 3D Printing Resins. Polym. Chem. 2017, 8 (3), 568−580. (23) Montalti, M.; Credi, A.; Prodi, L.; Teresa Gandolfi, M. Handbook of Photochemistry, 3rd ed.; CRC Press: 2006. (24) Fouassier, J.-P.; Lalevée, J. Photoinitiators for Polymer Synthesis, Scope, Reactivity, and Efficiency; Wiley-VCH Verlag GmbH & Co.KGaA: Weinheim, 2012.

AUTHOR INFORMATION

Corresponding Authors

*E-mail [email protected] (J.L.). *E-mail [email protected] (F.D.). ORCID

Jacques Lalevée: 0000-0001-9297-0335 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors thank the “Agence Nationale de la Recherche” (ANR) for the grant “FastPrinting”. The Lebanese group thank “The Association of Specialization and Scientific Guidance” (Beirut, Lebanon) for funding and supporting this scientific work.



REFERENCES

(1) Hua, Y.; Crivello, J. V. Synergistic Interaction of Epoxides and NVinylcarbazole during Photoinitiated Cationic Polymerization. J. Polym. Sci., Part A: Polym. Chem. 2000, 38 (19), 3697−3709. (2) Ortiz, R. A.; López, D. P.; Cisneros, M.; de, L. G.; Valverde, J. C. R.; Crivello, J. V. A Kinetic Study of the Acceleration Effect of Substituted Benzyl Alcohols on the Cationic Photopolymerization Rate of Epoxidized Natural Oils. Polymer 2005, 46 (5), 1535−1541. (3) Lalevée, J.; Tehfe, M.-A.; Zein-Fakih, A.; Ball, B.; Telitel, S.; MorletSavary, F.; Graff, B.; Fouassier, J. P. N-Vinylcarbazole: An Additive for Free Radical Promoted Cationic Polymerization upon Visible Light. ACS Macro Lett. 2012, 1 (7), 802−806. (4) Xiao, P.; Zhang, J.; Dumur, F.; Tehfe, M. A.; Morlet-Savary, F.; Graff, B.; Gigmes, D.; Fouassier, J. P.; Lalevée, J. Visible Light Sensitive Photoinitiating Systems: Recent Progress in Cationic and Radical Photopolymerization Reactions under Soft Conditions. Prog. Polym. Sci. 2015, 41, 32−66. (5) Al Mousawi, A.; Dietlin, C.; Graff, B.; Morlet-Savary, F.; Toufaily, J.; Hamieh, T.; Fouassier, J. P.; Chachaj-Brekiesz, A.; Ortyl, J.; Lalevée, J. Meta-Terphenyl Derivative/Iodonium Salt/9H-Carbazole-9-Ethanol Photoinitiating Systems for Free Radical Promoted Cationic Polymerization upon Visible Lights. Macromol. Chem. Phys. 2016, 217 (17), 1955−1965. (6) Al Mousawi, A.; Dumur, F.; Garra, P.; Toufaily, J.; Hamieh, T.; Graff, B.; Gigmes, D.; Fouassier, J. P.; Lalevée, J. Carbazole Scaffold Based Photoinitiator/Photoredox Catalysts: Toward New High Performance Photoinitiating Systems and Application in LED Projector 3D Printing Resins. Macromolecules 2017, 50, 2747. (7) Xiao, P.; Lalevée, J.; Zhao, J.; Stenzel, M. H. N-Vinylcarbazole as Versatile Photoinaddimer of Photopolymerization under Household UV LED Bulb (392 nm). Macromol. Rapid Commun. 2015, 36 (18), 1675−1680. (8) Uoyama, H.; Goushi, K.; Shizu, K.; Nomura, H.; Adachi, C. Highly Efficient Organic Light-Emitting Diodes from Delayed Fluorescence. Nature 2012, 492 (7428), 234−238. (9) Zucchero, A. J.; McGrier, P. L.; Bunz, U. H. F. Cross-Conjugated Cruciform Fluorophores. Acc. Chem. Res. 2010, 43 (3), 397−408. (10) Zucchero, A. J.; Wilson, J. N.; Bunz, U. H. F. Cruciforms as Functional Fluorophores: Response to Protons and Selected Metal Ions. J. Am. Chem. Soc. 2006, 128 (36), 11872−11881. (11) Luo, J.; Zhang, J. Donor−Acceptor Fluorophores for VisibleLight-Promoted Organic Synthesis: Photoredox/Ni Dual Catalytic C(sp3)−C(sp2) Cross-Coupling. ACS Catal. 2016, 6 (2), 873−877. (12) Kretzschmar, A.; Patze, C.; Schwaebel, S. T.; Bunz, U. H. F. Development of Thermally Activated Delayed Fluorescence Materials with Shortened Emissive Lifetimes. J. Org. Chem. 2015, 80 (18), 9126− 9131. (13) Al Mousawi, A.; Poriel, C.; Dumur, F.; Toufaily, J.; Hamieh, T.; Fouassier, J. P.; Lalevée, J. Zinc Tetraphenylporphyrin as High Performance Visible Light Photoinitiator of Cationic Photosensitive N

DOI: 10.1021/acs.macromol.7b01114 Macromolecules XXXX, XXX, XXX−XXX