Visible-Light-Curable Solvent-Free Acrylic Pressure-Sensitive Adhesives via Photoredox-Mediated Radical Polymerization

Owing to their excellent properties, such as transparency, resistance to oxidation, and high adhesivity, acrylic pressure-sensitive adhesives (PSAs) are widely used. Recently, solvent-free acrylic PSAs, which are typically prepared via photopolymerization, have attracted increasing attention because of the current strict environmental regulations. UV light is commonly used as an excitation source for photopolymerization, whereas visible light, which is safer for humans, is rarely utilized. In this study, we prepared solvent-free acrylic PSAs via visible light-driven photoredox-mediated radical polymerization. Three α-haloesters were used as additives to overcome critical shortcomings, such as the previously reported low film curing rate and poor transparency observed during additive-free photocatalytic polymerization. The film curing rate was greatly increased in the presence of α-haloesters, which lowered the photocatalyst loadings and, hence, improved the film transparency. These results confirmed that our method could be widely used to prepare general-purpose solvent-free PSAs—in particular, optically clear adhesives for electronics.


Bulk polymerization (supporting)
Table S1. The Bulk polymerization results; α = conversion of bulk polymerization determined gravimetrically; Mn and Ð represents number average molecular weight and dispersity, respectively, which were characterized by size exclusion chromatography. Detail condition was same with the Table 1 Figure S1. UV/Vis spectrum before and after bulk polymerization with 2 ppm of 4DP-IPN and 0.1 mol% of DBM (gray: before bulk polymerization, and red: after bulk polymerization).

Non-consumable PC (UV/Vis spectrum)
UV-Vis spectrum after bulk polymerization was similar with that before bulk polymerization, which suggested that 4DP-IPN was not consumed during bulk polymerization. The difference in yellow box of Figure S1 might be resulted from DBM. The residual DBM content was calculated by measuring the amount of DBM consumed during the bulk polymerization. In order to quantitatively evaluate consumed DBM, gas chromatography was employed and toluene was used as internal standard.   Pre-polymer with 5 mol% of DBM and 50 ppm of PC exhibited low molecular weight due to large amount of initiator. Before film curing, i.e. right after bulk polymerization, Mn and Mp was 84 kg/mol and 123 kg/mol, respectively. Although the crosslinker (PEGDA) was added, crosslinking network was not formed and Mp was lowered as 89 kg/mol, which might be resulted from a lot of remaining DBM (92.1%, from table of Figure S4). After film curing, the portion of "Region 1" in SEC curve, i.e. polymer initiated by remaining DBM, was larger than that of "Region 2" that represented slightly crosslinked polymer or re-initiated polymer from living chain end. When the crosslinker was not added, the portion of "Region 1" was enlarged. Figure S5. UV/Vis spectrum of PSA prepared through photoinitiator. 0.2 wt% of PEGDA was used as crosslinker. 0.3 wt% of Irgacure 184 was used as photoinitiator, and film was cured by 3 J/cm 2 .

UV/Vis spectrum of PSA prepared through UV photoinitiator
Transparency of PSA that prepared through photoinitiator was characterized by UV/Vis spectrum. After bulk polymerization (Entry 3, Table 1 of manuscript), crosslinker (PEGDA, 0.2 wt%) and photoinitiator (Irgacure 184, 0.3 wt%) were added, and then applied in a film form . The film was cured in the same way as shown in Figure S3, but UV (3 J/cm 2 ) was used instead of blue LED.   Single lap shear test was conducted to evaluate structural characteristics of prepared PSAs. The cured PSA film was cut to a certain size (length: 50 mm, and width: 25 mm), and attached to substrate (polyethylene terephthalate film, length: 50 mm, width: 25 mm, and thickness: 0.05 mm). Specimens were attached by constant force (2 kg roller for twice) and stored at room temperature for 1 day. Universal testing machine (UTM 5982, Instron) was employed and strain-stress was obtained as shown in Figure S10. Maximum force of strain-stress curve mean lap shear strength and lap shear strength of entries S2, S3, S4, 5, and 6 was shown in Figure S10. There was no big difference in lap shear strength (≈ 0.3 MPa) among aforementioned entries, and the lap shear strength of prepared PSAs were comparable to those of photo-responsive acrylic PSAs in references (0.15~0.34 MPa) [39,50,51].
Lap shear strength means resistance to fracture of internal layer ( Figure S10), whereas holding time we assessed means resistance to creep deformation (Figure 4). In addition, single lap shear test and holding test was conducted at different condition (single lap shear test: 25 o C, holding test: 50 o C) so that the results of lap shear strength could not correlate with the results of holding test. Figure S11. Characterization of Entry 6 in Table 1 (characterized after bulk polymerization). a) Nuclear magnetic resonance result ( 1 H-NMR, 300 MHz, Avance DPX-400, Bruker, solvent: CDCl3). Symbol x in graph mean peak of x that was incorporated in unreacted monomer. Symbol x' in graph mean peak of x' that was incorporated in polymer chain. b) Differential scanning calorimetry result (DSC, Q200, TA instrument, sample weight: 13 mg, sample was dried before characterization to remove unreacted monomers). Left graph presented time-temperature curve and detecting range. Right graph presented heat flow curve and Tg at detecting range.

Characterization of representative sample
After bulk polymerization, polymer chain dissolved in unreacted monomer was obtained, and its chemical structure was characterized by using 1 H-NMR ( Figure S11a). Glass transition temperature (Tg) of synthesized polymer was obtained as -28.7 o C ( Figure S11b), which was low enough to be utilized as pressure-sensitive adhesive.  Retention time / min S16  Figure S13. Density functional theory (DFT)-calculated molecular orbital (MO) diagrams and MO topologies of 4DP-IPN and α-haloesters; the time-dependent (TD)DFT calculations were performed with the B3LYP functional and 6-311G* basis set in ethyl acetate using the polarizable continuum model (PCM).