This is the accepted version of the journal article: Reyes-Mateo, Kevin; Marquet, Jordi; Hernando, Jordi; [et al.]. «Photother- mal polymerization of benzoxazines». Polymer Chemistry, Vol. 13, Issue 36 (September 2022), p. 5256-5264. DOI 10.1039/D2PY00635A This version is available at https://ddd.uab.cat/record/267122 under the terms of the license Please do not adjust margins ARTICLE Photothermal polymerization of benzoxazines Kevin Reyes-Mateo,a Jordi Marquet,a Jordi Hernando*, a and Rosa M. Sebastián*, a Received 00th January 20xx, Accepted 00th January 20xx Despite the use of catalysts and initiators, high polymerization temperatures are typically required for polybenzoxazines, a DOI: 10.1039/x0xx00000x severe constraint for their general application as high performance thermosets. In this work we pioneer a novel photothermal strategy for the synthesis of polybenzoxazines that takes places under light irradiation at ambient conditions. This method capitalizes on the intriguing optical properties of benzoxazine-catalyst systems, which show visible light absorption and locally generate large heat upon illumination that triggers polymerization. In this way a large variety of benzoxazine monomers can be converted into polymer materials that retain some of the relevant features of polybenzoxazines – e.g., high char yields. In addition, our photothermal approach allows spatial control of benzoxazine polymerization, which we exploit for the fabrication of polybenzoxazine patterns via photolithography. Introduction temperature has been explored, though with limited results. For instance, Yagci and co-workers targeted this goal using onium salts A vast number of polymers are synthesized and processed at high as photoinitiators under UV irradiation. Unfortunately, only temperatures, 1 an energy-intensive strategy that might eventually complex, low molecular weight polymer structures were obtained compromise their mass production and use. Photopolymerization 32 due to incomplete polymerization, which makes this strategy emerges as a much-desired, environmentally-friendly alternative in better suited for the cross-linking of main chain polybenzoxazine those cases, which should not only allow reducing the energy 33 precursors. Similarly, we could merely report low polymerization demand for polymer formation but also grant direct access to a degrees for the photochemical ring-opening polymerization of Bz in variety of materials under ambient conditions through advanced 2,3 4,5 water under direct UV photoexcitation, which was even inhibited fabrication techniques– e.g., photolithography and 3D printing. 34 for monomers bearing electron-withdrawing substituents. A class of polymers that would largely benefit from these features 6,7 The reason why all these strategies fail to ensure the efficient are polybenzoxazines (PBz), a distinctive type of phenolic resins production of PBz at low temperatures can be ascribed to their that are attracting much attention from industry because of their particular polymerization mechanism: while the initial ring-opening superior properties relative to traditional thermosets – e.g., near- 8- 10 step of the monomers can be facilitated by initiators, catalysts zero volume change upon polymerization, low surface energy, 11 12 13 14- 16 and/or light, the subsequent electrophilic aromatic substitution and high chemical resistance and char yields, among others. reactions needed for extensive polymer network growth still Although PBz precursors – i.e., 1,3-benzoxazines (Bz) – can be 17 require high thermal energy (Scheme 1). Herein we propose to readily prepared from low-cost and commercially available circumvent this obstacle by means of a photothermal approach, reagents, their ring-opening polymerization (ROP) is extremely where local heat is generated in situ under irradiation by relaxation energy demanding, as it takes places at particularly high 35,36 6,15 of light absorbing species through electron-phonon interaction. temperatures (Tp >180 °C). This is therefore a critical obstacle Interestingly, this would allow complying with the high temperature that precludes the widespread application of PBz. demand of Bz polymerization without the need of external heating, Several approaches have so far been explored to decrease the 17 while concomitantly benefitting from the inherent advantages of polymerization temperature of PBz. On the one hand, we and light-induced processes – i.e., precise time and space control over others have investigated the use of initiators and catalysts for this 18,19 chemical reactivity (Scheme 1). purpose – e.g., Brönsted and Lewis acids –, 20- 24 which can be To date a narrow range of photothermally-induced either added as separate components to the polymerization 37,38 18-24 polymerization processes have been described, which mainly mixture or directly incorporated into the Bz monomers in a comprise the photothermal curing of polysiloxanes, 39- 41 the latent state. 25- 28 However, only moderate Tp reduction has been 42- 52 photothermal initiation of acrylate polymerization and the accomplished in this way for conventional monofunctional 53,54 photothermal polymerization of polyurethanes. In all these benzoxazines, although the introduction of additional cases, reactions are promoted photothermally that take place at functionalities may further lower the onset reaction 20,29- 31 relatively moderate temperatures (T < 100 °C), while the aid of temperature. Alternatively, Bz photopolymerization at room external photothermal agents is required that eventually lie 37-43,53,54 entrapped in the final material – e.g., metal nanostructures, a 44,46,48,51 45,47,49,50 Departament de Química, Universitat Autònoma de Barcelona, 08193, Cerdanyola carbon (nano)materials or organic dyes, which are del Vallès, Barcelona, Spain. Email: jordi.hernando@uab.cat, rosamaria.sebastian@uab.es responsible for light absorption and heat generation. In contrast †Electronic supplementary information (ESI) available: Materials and methods and further experimental data (NMR spectra, UV-vis absorption spectra, DSC and TGA thermograms). See DOI: XXXXXXXXXXXXXXX Please do not adjust margins Please do not adjust margins ARTICLE Journal Name Catalyst, initiatior and/or light O N Ph As expected, none of the benzoxazines synthesized showed significant absorbance in the visible region when investigated by UV-vis absorption spectroscopy in organic solution (Fig. 2a). O N Ph OH N Ph R OH Instead, they presented defined absorption bands in the UV N spectrum with maxima at λabs < 325 nm. Consequently, they cannot Ph Ring opening Polymerization be used as direct photoabsorbers of visible light to promote R R High T R High T n photothermal effects. Interestingly, a dramatic change in optical Bz PBz properties was observed upon addition of LiI, a metal salt RT RT previously identified by us as one of the most efficient catalysts of 23 Bz polymerization. In particular, a clear color darkening was Photothermal effects Photothermal effects perceived when directly adding this compound to liquid Bz3 or Scheme 1 While current strategies to reduce the polymerization preparing solid Bz-catalyst mixtures for the rest of monomers by temperature of benzoxazines focus on facilitating the initial ring evaporation of the corresponding acetone solutions (Fig. 2b). opening reaction of the monomer, we propose herein a photothermal approach that should favor both this step and the subsequent electrophilic substitution reactions leading to polymer growth. with these precedents, in this work we aim to (a) explore the more demanding photothermal polymerization of benzoxazines, which typically takes place at much higher temperatures, and (b) avoid the use of additional photoabsorbers. Instead, we intend to capitalize on the intriguing optical properties of Bz-catalyst systems, thus reducing the cost and complexity of the process. Results and discussion Photothermal properties of benzoxazine-catalyst systems To test our photothermal strategy toward benzoxazine polymerization, a variety of previously reported Bz monomers were synthesized: several 1,3-benzoxazines bearing electron-donating 34,55- 59 and electron-withdrawing substituents (Bz1-Bz8) and 60 bisbenzoxazine Bz9 (Fig. 1). Most monofunctional Bz were prepared through a solvent-free Mannich-type reaction by mixing 1,3,5-triphenyl-1,3,5-triazine or methylamine, paraformaldehyde and the respective phenol (Bz1-Bz5, Bz7). In the case of Bz6 and Bz8, 1,4-dioxane also had to be added as a solvent to avoid concomitant thermal polymerization of the monomer formed. For bisbenzoxazine Bz9, the Mannich-type reaction was conducted using aniline, paraformaldehyde and bisphenol A as a bifunctional phenol. All monomers were isolated as white or slightly yellow solids, except for Bz3 that was obtained as an orangish oil. Ph O N Ph R= H, Bz1 (56%) N CH3, Bz2 (73%) OCH3, Bz3 (77%) O O CN, Bz4 (27%) COOCH3, Bz5 (60%) N R NO2, Bz6 (57%) Ph Fig. 2 a) UV-vis absorption spectra of all prepared (bis)benzoxazines in Bz1-Bz8 CHO, Bz7 (30%) Bz9 (54%) DCM (cBz = 1.0 10-5 M). b) UV-vis absorption spectra of pure benzoxazine CF3, Bz8 (57%) Bz3 and of its mixture with LiI (50 mol % relative to monomer content). Images of Bz3 before (left) and after (right) the mixture are shown as Fig. 1 Benzoxazine monomers synthesized via a Mannich-type reaction insets. c) Temperature variation in liquid Bz3 and Bz3-LiI (50 mol % between an amine, formaldehyde and the corresponding phenol. Yields relative to monomer content, 0.5 mL) under laser irradiation at λexc = are shown in parentheses (see Table S1 and S2 in the ESI† for further 532 nm (spot diameter = 1 cm). Maximum temperature values reached details). after around 5 min of irradiation at each power. 2 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Journal Name ARTICLE To further investigate this behavior, Bz samples with high LiI this goal, we focused our attention on the peaks arising from the amounts were prepared (50 mol % relative to monomer methylene oxazine protons of the monomers, whose defined concentration) and analyzed by UV-vis absorption as neat liquids signals disappear upon polymerization and give rise to the (Bz3) or concentrated acetone solutions (Bz1, Bz2, Bz4-Bz9). As formation of a variety of new resonances at δ ∼ 3.5 – 6.0 that are shown in Fig. 2b and Figs. S1-S9 in the ESI†, a clear red-shifting of generally attributed to the large number of different structures the absorption spectrum into the visible range was observed in all formed upon polymerization23,24 (Fig. 3a). By comparing the the cases and, indeed, the emergence of a new defined visible integrals of these different sets of monomer and polymer signals, absorption band could be identified for most samples (λabs ∼ 400 - we could estimate the evolution of the polymerization process in 24 550 nm). Importantly, none of these spectral changes could be time using a methodology previously reported by us (see the ESI† attributed to the intrinsic absorption properties of the catalyst, for further details). which is a white solid that only absorbs in the UV region (Fig. S10 in Interestingly, none of the benzoxazine monomers polymerized the ESI†). Instead, they should arise from its interaction with the Bz neither upon irradiation in the absence of the catalyst nor in the monomers. Based on our previous suggestion that metal ions such dark when mixed with LiI at room temperature. These results are + as Li coordinate to the heteroatoms of the oxazine ring to favor Bz consistent with (a) the negligible visible absorbance of pure Bz1- 23,24 ring-opening polymerization, herein we hypothesize that the Bz9, and (b) the limited capacity of catalysts such as LiI to decrease resulting complexes present bathochromically displaced absorption. their Tp, which still lie above 100 ºC even for the 15 mol % Actually, a similar behavior has already been reported for aromatic concentration tested (Table S3 and Figs. S12-S20 in the ESI†). By chromophores bearing O- and N-coordinating groups upon 61 complexation to lithium ions. To demonstrate such interaction in + the case of Bz monomers, we investigated the coordination of Li ions to Bz3 by means of additional spectrophotometric experiments. This allowed us to determine a 1:1 stoichiometry and -1 a low association constant (Ka = 0.48 M ) for the Bz3-LiI complex, which is consistent with the high concentrations needed to observe absorption changes upon metal salt addition (Fig. S11a in the ESI†). Because of the red-shifted absorption of Bz-LiI systems, they can be excited with visible light. No emission was subsequently detected for any of these samples, which suggests that their photoexcited species mainly decay via internal conversion – i.e., generating heat by converting electronic excitation into vibrational energy. This photothermal effect was proven by irradiating a neat solution of LiI in Bz3 (50 mol % relative to monomer content) with increasing intensities of green laser light (λexc = 532 nm) while monitoring temperature with an IR probe. After a short incubation time (∼ 5 min), constant temperatures were reached above room temperature, whose actual values rose with the photoexcitation power (Fig. 2c). Surprisingly, a sudden change in color from orange to red was observed for the Bz3-LiI solution when photothermally - heated with the highest power intensities assayed (1.0 - 1.1 W cm 2 ), which suggested us that monomer reactivity was taking place at the temperatures reached in this case. By contrast, neither color change nor temperature increase was observed when irradiating pure Bz3 samples at the same conditions (Fig. 2c). Photothermal polymerization of benzoxazines Taking advantage of the photothermal behavior of benzoxazine-LiI Fig. 3 a) Variation of the 1H NMR spectrum (CDCl3, 360 MHz) of the Bz3- systems, we attempted their photoinduced polymerization in liquid LiI system during photoirradiation at tirr = 0, 15, 30, 40 and 50 min (cLiI = (Bz3) and solid powder samples (Bz1, Bz2, Bz4-Bz9). In a first step, 15 mol % relative to monomer content; λexc = 532 nm, power density = 1.1 W cm-2). The black circles indicate the position of the signals of the we prepared Bz-LiI systems with a 15 mol % catalyst concentration methylene oxazine protons that disappear upon polymerization. The relative to monomer content, which were irradiated with a pulsed insets show photographs of Bz3-LiI before and after irradiation, which -2 demonstrates the phototransformation of the initial liquid monomer laser at λexc = 532 nm and 1.1 W cm for tirr = 100 min at ambient into a solid polymer material. b) Evolution of the reaction mixture conditions. As all the resulting samples were found to be soluble in temperature and conversion during visible light irradiation of Bz3-LiI (cLiI organic solvents except for bisbenzoxazine Bz9, we could estimate = 15 mol % relative to monomer content; λexc = 532 nm, power density = 1 the polymerization efficiency for Bz1-Bz8 by H NMR analysis. With 1.1 W cm-2). This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 3 Please do not adjust margins Please do not adjust margins ARTICLE Journal Name contrast, illumination of Bz-LiI systems led to monomer pattern of the Bz monomer on the photopolymerization process. polymerization for all monobenzoxazines Bz1-Bz8 (Fig. S22 in the With this goal, we considered the case of Bz10, a monobenzoxazine ESI†), which were converted from liquid or solid powders to bearing methyl substituents both in ortho and para positions (see homogeneous solid materials. Actually, full monomer conversion section 2.4 in the ESI†). Upon addition of 15 mol % of LiI relative to was observed in all the cases for irradiation times shorter than 100 monomer content and irradiation of the resulting powder mixture -2 min. For instance, quantitative polymerization at these conditions at λexc = 532 nm and 1.1 W cm , less efficient photopolymerization was observed for Bz3 at tirr = 50 (Bz3-LiI) and Bz6 at tirr = 70 min. In was observed for Bz10 relative to p-substituted benzoxazines Bz1- the case of Bz9, an insoluble solid was obtained in this way, a result Bz9. Thus, only 58% monomer conversion was determined after tirr that is in agreement with the capacity of this compound to lead to = 100 min for Bz10, which could not be fully polymerized even after 60 extensively cross-linked polymer networks. Actually, when tirr = 300 min (89% conversion). This result is consistent with the incubating the material obtained upon photoirradiation of Bz9 with lower reactivity reported for o,p-disubstituted benzoxazines due to 1 organic solvents and analyzing the resulting filtrates by H NMR, no the blocking of the preferred sites for polymerization - i.e., the sites residual monomer signals could be identified, thus confirming ortho and para to the phenolic position -, which typically react at 62 quantitative conversion also for this monomer. higher temperatures through the arylamine group. Actually, this To investigate the mechanism for Bz photopolymerization at was proven by DSC measurements in our case, as we determined a our experimental conditions, we simultaneously monitored the Tp, onset = 271.2 ºC for the pure o,p-dimethylsubstituted Bz10 that is temperature changes and monomer conversion during the about 10 ºC higher than for the pure p-methylsubstituted Bz2 (Table irradiation of Bz3-LiI with green laser light (Fig. 3b). A fast increase S3 and Fig. S21 in the ESI†). Interestingly, these results are in in temperature until T ∼ 80 ºC was registered for this sample after a agreement with the conclusions from our mechanistic investigation: few minutes (tirr < 10 min), while negligible reactivity was observed. the efficiency of our photopolymerization strategy depends on the Consequently, photothermal effects imparted after visible light capacity to generate sufficiently high photothermal heat upon absorption must be responsible for such initial heating. Next, Bz irradiation as to reach the temperatures needed for the onset of polymerization was triggered and most of the monomer was thermal polymerization for each given monomer-catalyst system, a consumed (∼ 80%) without a significant rise in temperature (T ∼ 90 temperature threshold that can be largely affected by the Bz ºC), which only eventually increased up to T ∼ 100 ºC during the last substitution pattern and the electronic nature of the substituents. stages of the reaction. Interestingly, these temperature values lie close to the lower limit of the polymerization temperatures Optimization of the photothermal polymerization of benzoxazines determined for Bz3 with 15 mol % content of LiI in the dark by DSC Once demonstrated the capacity to extensively photopolymerize (Tp, onset = 141.2 ºC, Table S3 in the ESI†) and, indeed, we observed Bz-LiI systems without the need of external heating, we pursued the that such a monomer-catalyst system polymerizes when externally optimization of the reaction conditions. Most of these studies were heated at T ∼ 90 ºC for 1 h. Accordingly, our measurements support conducted for Bz3 because of its high visible absorption and the photothermal nature of the light-induced reaction observed for reactivity when mixed with LiI as well as its liquid nature, which Bz3-LiI – i.e., it is not only related with the absorption red-shift facilitates the preparation and manipulation of the monomer- accomplished upon monomer-metal ion interaction, but also to the catalyst samples. photothermal effects produced under illumination. To start with, we investigated the effect of other metal salts This conclusion was further corroborated by several additional different from LiI on photoreactivity, which we selected from some experiments. First, no polymerization was observed when of the catalysts previously evaluated in our group for the thermal 24 conducting the irradiation of monomer-catalyst samples at 0 ºC in polymerization of benzoxazines: Zn(OTf)2, LiClO4 and FeCl3 (Fig. an ice bath. Second, by turning off the illumination source when S10 in the ESI†). Although all of them promoted visible light-induced reaching ca. 50% monomer conversion (tirr = 22.5 min), the polymerization when mixed with Bz3, the best results were temperature of the sample rapidly decreased and the polymerization obtained with Zn(OTf)2 (Table 1). Thus, Bz3-Zn(OTf)2 samples could process stopped (Fig. S23 in the ESI†). Finally, when restricting the be fully photopolymerized in just 6 min, which is about 10-fold excitation to a small area of a bulk mixture, frontal polymerization – faster than for Bz3-LiI under equivalent irradiation conditions. In i.e., reactivity propagation outside of the illumination region – did addition, a similar optimal behavior was preserved when not take place (Fig. S24 in the ESI†). Overall, these results attempting the photopolymerization of other monobenzoxazines unambiguously prove that light-induced polymerization at our such as Bz1, Bz2 and Bz4 and bisbenzoxazine Bz9 with Zn(OTf)2 experimental conditions proceeds at high temperatures, which we (Table 1). We ascribe this situation to two main factors. First, the achieve locally by the photothermal effects imparted upon larger absorption measured for Bz-Zn(OTf)2 in the visible region irradiation of visible-absorbing Bz-catalyst systems. As a result, our relative to Bz-LiI even at lower metal salt concentrations (Figs. S1-S4 strategy enables efficient photopolymerization of benzoxazines, in and S9 in the ESI†), probably due to the higher coordination 2+ 63 contrast to previous reports where light was only employed for capacity of Zn ions. As a result, irradiation of Bz-Zn(OTf)2 must 32,34 monomer initiation and did not induce photothermal heating. lead to much stronger photothermal effects. Second, the larger To further explore the scope of our photothermal strategy capacity of Zn(OTf)2 to reduce the Tp of Bz in the dark (Tp, onset = 56.4 towards PBz, we also investigated the effect of the substitution ºC for Bz3), as proven by DSC (Table S3 in the ESI†). Consequently, 4 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Journal Name ARTICLE Table 1 Optimization of photopolymerization conditions for Bz-catalyst leaving group character and/or their capacity to generate triflic acid systemsa that could aid to the initiation of Bz monomers. 17 Monomer Catalyst ccatalystb Irradiation Conversion Next, we optimized the catalyst concentration and irradiation (mol %) Power Time (%) conditions required for efficient photothermal polymerization, for (W cm-2) (min) which we focused our attention on Bz3-LiI and Bz3-Zn(OTf)2 Bz3 Zn(OTf)2 15 1.1 10 100 c samples (Table 1). In the case of LiI, we could decrease its concentration down to 5 mol % without drastically altering the LiClO4 15 1.1 100 68 polymerization time of Bz3-LiI samples. By contrast, slight FeCl3 15 1.1 10 100 decrements in irradiation power significantly slowed down the ZnI2 15 1.1 10 77 reactivity of these systems. A somewhat opposite situation was instead encountered for Zn(OTf)2: while catalyst concentration Bz1 Zn(OTf)2 15 1.1 10 100 c could only be decreased to 10 mol % in order to preserve the Bz2 Zn(OTf)2 15 1.1 10 100 c polymerization kinetics, very fast monomer conversions were still Bz4 Zn(OTf)2 15 1.1 10 100 observed for notably lower illumination power – e.g., 87 % Bz3 polymerization could be achieved in only tirr =10 min despite Bz9 Zn(OTf)2 15 1.1 10 59 d decreasing the excitation intensity to 0.9 W cm-2. Finally, we also Bz3 LiI 10 1.1 50 89 explored the use of cw illumination instead of pulsed irradiation for LiI 5 1.1 50 89 Bz3-Zn(OTf)2. Efficient photothermally-induced polymerization was also observed in this case, though the average photoexcitation LiI 1 1.1 50 20 power had to be increased about 9-fold to preserve the very rapid Zn(OTf)2 10 1.1 10 95 reaction kinetics (Table 1). Zn(OTf)2 5 1.1 10 63 Even after optimization, a rather large concentration of catalyst Zn(OTf)2 1 1.1 10 0 is needed for our photopolymerization process to be efficient (> 5 mol %), which would lie entrapped within the solid material formed. Bz3 LiI 15 1.0 50 63 Importantly, we determined the complexation constant between Zn(OTf)2 15 1.0 10 100 the polymer obtained and the remaining metal ions to be very low -1 Zn(OTf)2 15 0.9 10 87 (Ka = 8.0 M for LiI and polymerized Bz3; Fig. S11b in the ESI†), which suggests that their effect on the structure of Zn(OTf)2 15 0.6 10 65 polybenzoxazines must be minor. Actually, this agrees with our e Zn(OTf)2 15 9.6 10 100 2+ previous studies on the coordination of Zn to model benzoxazine a λexc = 532 nm, monomer conversion determined by H NMR. b Catalyst 1 compounds, which were found to be only important for short concentration relative to monomer content. c Quantitative monomer benzoxazine dimers containing two nearby phenol groups or one conversion achieved at tirr = 6 min. d A soluble polymer material was 64 obtained for bisbenzoxazine Bz9 in this case, probably due to phenol and one amino groups. Instead, metal complexation incomplete polymerization. e Irradiation with a cw laser at λexc = 532 nm becomes much weaker in the presence of strong hydrogen-bonds instead of a pulsed laser. between phenol groups, which is typically the case in 64 polybenzoxazines. In light of this, we attempted the removal of the photothermal effects generated in this case should be more the metal ions from our photopolymerized materials, for which we effective in triggering benzoxazine polymerization. focused on the Bz3-Zn(OTf)2 system. Upon dissolution in cold However, in contrast to Bz-LiI, Bz-Zn(OTf)2 systems were found acetone where Zn(OTf)2 is very poorly soluble, we observed a 85 % not to be stable when stored for hours under ambient conditions, reduction in triflate counterions from the photoproduced polymer 19 1 which we attributed to triflic acid formation upon moisture by F NMR. In addition, we found that the H NMR of this sample absorption. For this reason, we also explored the use of ZnI2 as a did not significantly vary upon purification, which corroborates our metal salt for Bz photothermal polymerization (Table 1). Although hypothesis that the structure of the photogenerated polymers must more stable samples were obtained with this catalyst, the change of not be critically affected by the presence of residual metal ions. the salt counteranion detrimentally affected the polymerization kinetics – e.g., only 77 % monomer conversion was achieved for Thermal vs photothermal polymerization of benzoxazines Bz3-ZnI2 at the same conditions that led to complete To assess the effect of our photopolymerization process on the -2 photopolymerization for Bz3-Zn(OTf)2 (λexc = 532 nm, 1.1 W cm , tirr properties of PBz, we compared the behavior of the solid materials = 10 min). Such effect of the salt counteranion has already been obtained from the photothermal and thermal polymerization of described for the catalyzed thermal polymerization of several benzoxazine monomers (Bz1-Bz4 and Bz9, Table 2). benzoxazines, for which better results were obtained with Zn(OTf)2 First, we observed that the photothermally-produced polymers 24 than ZnCl2. Therefore, the presence of triflate counteranions also were considerably soluble in many of the organic polar solvents seems to enhance the efficiency of our photothermal tested - e.g., acetone, DMSO, ethyl acetate and CHCl3. As already polymerization method, which could be ascribed to their better discussed, the only exception to this behavior is bisbenzoxazine Bz9, This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 5 Please do not adjust margins Please do not adjust margins ARTICLE Journal Name Table 2 Properties of PBz prepared via thermal and photothermal the two monomers exhibiting the largest variation in solubility upon polymerizationa polymerization (see the ESI† for further details). For Bz3, almost no difference in gel content was observed for the polymers obtained Monomer Reaction Catalystb Structurec (%) Tg (ºC) Char d photothermally and thermally using LiI as a catalyst: 1 % and 0 %, yield Phenoxy Phenolic respectively, which indicates that essentially only highly soluble (%) linear polymer chains are obtained irrespective of the polymerization Bz1 Thermal - Not soluble e 98.4 42 method. On the other hand, very large gel contents were measured e Thermal LiI Not soluble 93.7 38 for Bz9 both upon photothermal and thermal polymerization with Thermal Zn(OTf)2 0 e 100 f 102.2 41 LiI (97 % and 98 %, respectively), which proves the formation of highly cross-linked insoluble networks in the two cases. Photothermal LiI 69 31 50.1 36 Despite the negligible effect of chemical cross-linking in the Photothermal Zn(OTf)2 74 26 53.3 44 polymer materials formed from Bz3 and, probably, the rest of p- Bz2 Thermal - Not soluble e 160.3 40 substituted monobenzoxazines tested, it must be noted that their e f solubility can be strikingly affected by another structural feature: Thermal LiI 5 95 100.7 29 the distinct types of polymerized structures that can be obtained Thermal Zn(OTf)2 15 85 60.7 34 from benzoxazine monomers. These are the phenolic-type structure Photothermal LiI 56 44 56.0 37 generated by electrophilic aromatic substitution after oxazine ring- opening and the phenoxy-type structure generated by nucleophilic Photothermal Zn(OTf)2 47 53 56.2 53 substitution, which rearranges to the more stable phenolic Bz3 Thermal - 25 75 57.7 32 6,7 structure upon heating (Scheme 3). Because of the strong Thermal LiI 0 100 88.5 43 intramolecular hydrogen bonding network that the phenolic-type polymer chains can form, a higher phenolic-to-phenoxy ratio must Thermal Zn(OTf)2 0 100 52.2 30 6,7 then result in lower solubility. To corroborate this hypothesis, we Photothermal LiI 50 50 53.4 48 estimated the contribution of phenolic and phenoxy groups to the Photothermal Zn(OTf)2 65 35 17.2 48 structure of the polymeric materials prepared, for which we applied 1 e f a H NMR analysis methodology previously established by us (Table Bz4 Thermal - 0 100 152.9 56 24 2, see the ESI† and Fig. S25-S28 for further details). As already e g Thermal LiI Not soluble - 53 described, PBz obtained through thermal polymerization at high e g Thermal Zn(OTf)2 Not soluble - 56 temperatures – e.g., at T = 180 ºC – were mainly composed of phenolic-type polymer chains (> 75%), thus explaining their low Photothermal LiI 45 55 60.6 53 solubilities. On the other hand, photothermally-produced PBz Photothermal Zn(OTf)2 55 45 58.6 60 presented a much larger percentage of phenoxy-type structures (> Bz9 Thermal - Not soluble e 166.3 34 45%), thus justifying their higher solubilities. Most probably, the Thermal LiI Not soluble e - g 29 larger phenoxy-type content observed upon photopolymerization is e g due to the lower temperatures achieved during the process, which Thermal Zn(OTf)2 Not soluble - 16 e do not enable thermal rearrangement to the more stable phenolic Photothermal LiI Not soluble 61.1 33 forms. This would explain why similar phenoxy-to-phenolic ratios h h Photothermal Zn(OTf)2 80 20 27.7 38 were obtained for all the monomers and catalysts by means of our a Thermal polymerizations were performed by heating at 180 ºC, while photothermal approach, as comparable temperatures must be photopolymerizations were conducted under pulsed laser excitation at accomplished in all the cases upon irradiation. λexc = 532 nm, power density = 1.1 W cm-2, tirr = 100 (Bz-LiI) or 10 min (Bz-Zn(OTf)2) and ambient conditions. b ccatalyst = 15 mol % relative to monomer concentration. c Determined by 1H NMR. d Measured after thermal annealing heating up to 800 ºC. e Polymer not soluble in organic solvents. f Polymer sparingly soluble in DMSO and insoluble in the rest of organic ∆ solvents tested. g Tg values could not be determined. h Structure type Ph ratio estimated for the polymerized fraction of the monomer. O N polymerization O OH N + N Ph Ph which seems to form insoluble cross-linked materials when R R R subjected to long irradiation times – e.g., for tirr = 100 min with LiI. n n By contrast, thermal polymerization led to insoluble or sparingly Bz Phenoxy-type structure Phenolic-type structure soluble PBz not only for Bz9, but also for several of the monobenzoxazines tested depending on the catalyst used (Table 2). Thermal polymerization Low High To investigate this issue in depth, we determined the gel content of Photothermal polymerization Medium Medium the photothermal and thermal polymers obtained for Bz3 and Bz9, Scheme 3 Influence of the polymerization method on the phenolic-to- phenoxy structure ratio of PBz. 6 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Journal Name ARTICLE The phenoxy-to-phenolic ratio of PBz structure also affects their strategy, as temperatures higher than 300 ºC were needed to thermal and mechanical performances.6,7 We investigated this issue observe large mass losses in all the cases (> 30%, Figs. S34-S38 and by determining the glass transition temperature (Tg) of the Table S4 in the ESI†). polymers obtained by DSC measurements (Figs. S29-S33 in the ESI†). As shown in Table 2, photothermally prepared Photolithography of benzoxazines polybenzoxazines mainly exhibited Tg ∼ 50-60 °C, whereas thermal Irrespective of the physico-chemical properties of the resulting polymerization conducted to much higher Tg values in most of the polymers, the main difference of our photopolymerization strategy cases. These results are therefore consistent with the lower relative to thermal PBz polymerization lies on the inherent phenolic content of the photoproduced polymers and clearly advantages of light-induced reactions: they can be performed with demonstrate that different PBz materials are obtained when extreme time (down to fs) and spatial (down to nm) control. As a applying our photothermal method relative to conventional thermal result, this would open new avenues for the fabrication and polymerization. However, we found that, by thermally annealing processing of PBz materials through light-induced self-healing, the photopolymerized PBz at 150 ºC for 2 h, nearly quantitative photolithography or 3D printing, among other methods. phenoxy-to-phenolic structure transformation occurred (Figs. S25- As a proof of concept, we applied our photothermal S28 in the ESI†), thus making these materials more similar to those polymerization approach to undertake the preparation of PBz directly obtained by thermal polymerization. patterns by means of photolithography under ambient conditions To further investigate this behavior, we analyzed in detail the (Fig. 4). For this, we prepared liquid Bz3-Zn(OTf)2 samples to which effect caused by thermal annealing on the properties of the we added a small amount of bisbenzoxazine Bz9 to favor chemical photoproduced polymers obtained from Bz3 using Zn(OTf)2 as a cross-linking of the resulting polymer matrices (9:1 Bz3:Bz9 ratio). catalyst. From a structural point of view, no increase in chemical After deposition onto glass substrates, these formulations were cross-linking was observed, as the gel content (2 %) remained as illuminated with a computer-controlled photolithography low as for the purely photothermally and thermally polymerized instrument equipped with a focusable cw violet laser (λexc = 405 nm, materials. However, a clear change in phenoxy-to-phenolic ratio Fig. S39 in the ESI†), thus generating defined solid features by local was measured (0:100), which matches the value determined upon photothermal polymerization. Removal of the unreacted monomer- thermal polymerization with the same catalyst. In addition, the catalyst system by rinsing with organic solvents (e.g., ethyl acetate) thermal treatment of the photoproduced polymer led to an finally allowed isolation of the complex target PBz patterns that increment in Tg (from 17.2 ºC to 63.2 ºC), which makes it closer to could not be fabricated otherwise. the value registered for the thermally-polymerized material (Tg = 52.2 ºC). Finally, we could prove that thermal annealing also largely enhances the mechanical properties of the photopolymerized materials. Thus, while we could not conduct dynamic mechanical analysis (DMA) of the photothermally-produced polymer of Bz3 because of its high brittleness, its mechanical properties could be determined after thermal treatment. In particular, the following Fig. 4 Photopatterning of PBz by (a) casting a Bz3/Bz9-Zn(OTf)2 storage moduli were registered for the thermally-annealed material formulation onto glass (9:1 Bz3:Bz9 ratio, ccatalyst = 15 mol % relative to the total monomer content), (b) irradiating this sample locally with a by DMA: 1855 MPa and 759.4 MPa at 0 ºC and 20 ºC, respectively, scannable cw violet laser (λexc = 405 nm, 33 W cm-2, 10 mm s-1) and (c) which are very resemblant to those obtained for the analogous PBz removing the excess of unreacted monomer. directly polymerized thermally (1678 MPa and 702.6 MPa at 0 ºC and 20 ºC, respectively). Consequently, all these results imply that our photopolymerization strategy can be employed to produce low- Conclusions Tg benzoxazine materials – i.e., easier moldable polymers – with the In this work we report a novel photopolymerization strategy for inherent advantages of light-induced processes – i.e., spatial and polybenzoxazines, a class of technologically-relevant thermosets time control –, which can be later toughened by thermal annealing that require high temperatures for their synthesis and processing to obtain polymers that reproduce the thermal and mechanical even when initiators or catalysts are employed. By mixing performance of conventional PBz. benzoxazine monomers with common metal salts typically used as By contrast, char yield, one of the relevant features of PBz, did polymerization catalysts, we observed the formation of complexes not seem to be affected by the polymerization method – i.e., by that exhibit visible light absorption and generate strong their phenolic-to-phenoxy ratio. Thus, similar char yield values (at photothermal heating upon photoexcitation. As a result, irradiation 800 ºC) were registered by TGA experiments for the thermally- and of benzoxazine-catalyst samples with visible light under ambient photothermally-produced materials and, in some cases, they could conditions triggers extensive photothermal polymerization, a be even higher for the photopolymerized benzoxazines (Table 2, process that can be expanded to a large variety of mono- and Figs. S34-S38 in the ESI†). In addition, thermogravimetric bifunctional monomers and be completed in very short time scales. measurements also uncovered that the thermal stability of the This is a remarkable improvement relative to previous resulting PBz does not significantly vary with the polymerization photopolymerization methods proposed for benzoxazines, which This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 7 Please do not adjust margins Please do not adjust margins ARTICLE Journal Name can only render incomplete monomer conversion. Although the K. R.-M. conducted all the experimental work. J. M., J. H. and R. M. polymer materials photoproduced herein are more soluble and S. designed and supervised the project. The manuscript was written show lower Tg values than when thermally polymerized, they through the contributions of all authors. All authors have given preserve some of the main features of polybenzozaxines – e.g., high approval to the final version of the manuscript. char yields. In addition, they can be toughened by subsequent thermal treatment, eventually exhibiting thermo-mechanical Conflicts of interest properties that are very similar to those of conventional polybenzoxazines. More importantly, our photothermal strategy There are no conflicts to declare. allows the polymerization of benzoxazines with spatiotemporal control, a feature that we have exploited to generate lithographical Acknowledgements polybenzoxazine patterns. Therefore, our work paves the way to The authors acknowledge the Agencia Estatal de Investigación the fabrication and processing of polybenzoxazines at ambient (PID2019-106171RB-100/AEI/10.13039/501100011033) and conditions using unprecedented light-based technologies such as Generalitat de Catalunya (2017 SGR00465) for financial support. K. photolithography and 3D printing. R.-M. thanks the Universitat Autònoma de Barcelona for his PIF pre- doctoral fellowship. We also thank Dr. Silvia De la Flor for Author Contributions conducting DMA experiments on our PBz samples. 26 K. Zhang, Y. Liu, M. Han and P. Froimowicz, Green Chemistry, Notes and references 2020, 22, 1209-1219. 27 B. Hao, L. Han, Y. Liu and K. Zhang, Polym. Chem., 2020, 11, 1 T. Meyer and J. Keurentjes. Handbook of polymer reaction 5800-5809. engineering. Wiley-VCH. 2005. 28 W. Zhao, B. Hao, Y. Lu and K. Zhang, Eur. Polym. J., 2022, 166, 2 T. F. Scott, B. A. Kowalski, A. C. Sullivan, C. N. Bowman and R. R. 111041. McLeod, Science, 2009, 324, 913-917. 29 M. Soto, M. Hiller, H. Oschkinat and K. Koschek, Polymers, 2016, 3 I. Kherbouche, Y. Luo, N. Félidj and C. Mangeney, Chem. Mater., 8, 278. 2020, 32, 5442. 30 H. Schäfer, A. Arnebold, J. Stelten, J. Marquet, R.M. Sebastián, 4 M. Layani, X. Wang and S. Magdassi, Adv. Mater., 2018, 30, A. Hartwig and K. Koschek, J. Polym. Sci. Part A Polym. Chem., 1706344. 2016, 54, 1243–1251. 5 L. Y. Zhou, J. Fu and Y. He, Adv. Funct. Mat., 2020, 30, 2000187. 31 S. K. Loke, E. P. S. Devaraju, V. Srinivasadesikan and R. K. 6 H. Ishida and T. Agag. Handbook of benzoxazine resins. Elsevier. Kottalanka, RSC Adv., 2020, 10, 36275-36286. 2011. 32 F. Kasapoglu, I. Cianga, Y. Yagci and T. Takeichi, T. J. Polym. Sci. 7 H. Ishida and P. Froimowicz. Advanced and emerging A Polym. Chem., 2003, 41, 3320-3328. polybenzoxazine science and technology. Elsevier. 2017. 33 Z. Deliballi, B. Kiskan and Y. Yagci, Polym. Chem., 2021, 12, 8 H. Ishida and H. Y. Low. Macromolecules, 1997, 30, 1099−1106. 5781-5786. 9 H. Ishida and D. Allen. J. Polym. Sci. Polym., Phys., 1996, 34, 34 J. Salabert, R. M. Sebastián and J. Marquet. Macromolecules, 1019−1030. 2018, 51, 3672-3679. 10 X. Liu and Y. Gu. J. Appl. Polym. Sci., 2002, 84, 1107−1113. 35 C. Wang and D. Astruc, Chem. Soc. Rev., 2014, 43, 7188-7216. 11 C. F. Wang, Y. C. Su, S. W. Kuo, C. F. Huang, Y. C. Sheen and F. C. 36 A. Gellé, T. Jin, L. de la Garza, G. D. Price, L. V. Besteiro and A. Chang. Angew. Chem. Int. Ed., 2006, 45, 2248−2251. Moores, Chem. Rev., 2020, 120, 986-1041. 12 H. D. Kim and H. Ishida. J. Appl. Polym. Sci., 2001, 79, 37 J. M. Walker, L. Gou, S. Bhattacharyya, S. E. Lindahl and J. M. 1207−1219. Zaleski, Chem. Mater., 2011, 23, 5275-5281. 13 N. K. Sini and T. Endo, Macromolecules 2016, 49, 8466–8478. 38 K. G. Stamplecoskie, C. Fasciani and J. C. Scaiano, 14 S. B. Shen and H. Ishida. Polym. Compos., 1996, 17, 710−719. Langmuir, 2012, 28, 10957-10961. 15 N. N. Ghosh, B. Kiskan and Y. Yagci, Prog. Polym. Sci., 2007, 32, 39 M. Fedoruk, M. Meixner, S. Carretero-Palacios, T. Lohmüller and 1344–1391. J. Feldmann, ACS Nano, 2013, 7, 7648-7653. 16 J. Liu and H. Ishida. Macromolecules, 2014, 47, 5682−5690. 40 R. J. Fortenbaugh and B. J. Lear, Nanoscale, 2017, 9, 8555-8559. 17 B. Lochab, M. Monisha, N. Amarnath, P. Sharma, S. Mukherjee 41 R. J. Fortenbaugh, S. A. Carrozzi and B. J. Lear, Macromolecules, and H. Ishida, Polymers, 2021, 13, 1260. 2019, 52, 3839-3844. 18 J. Dunkers and H. Ishida, J. Polym. Sci. Part A Polym. Chem., 42 W. M. Gramlich, J. L. Holloway, R. Rai, and J. A. Burdick, 1999, 37, 1913–1921 Nanotechnology, 2013, 25, 014004. 19 A. Sudo, H. Yamashita and T. Endo, J. Polym. Sci. Part A Polym. 43 H. Lee, S. Chung, M. G. Kim, L. P. Lee and J. Y. Lee, Adv. Healthc, Chem., 2011, 49, 3631–3636. Mater, 2016, 5, 1638-1645. 20 Y. X. Wang and H. Ishida. Polymer, 1999, 40, 4563-4570. 44 R. C. Steinhardt, T. M. Steeves, B. M. Wallace, B. Moser, D. A. 21 H. D. Kim and H. Ishida, Macromolecules, 2000, 33, 2839–2847. Fishman and A. P. Esser-Kahn, ACS Appl. Mater. Interfaces, 22 A. Sudo, S. Hirayama and T. Endo, J. Polym. Sci. Part A Polym. 2017, 9, 39034-39039. Chem., 2010, 48, 479– 484 45 A. H. Bonardi, F. Bonardi, F. Morlet-Savary, C. Dietlin, G. 23 C. Liu, D. Shen, R. M. Sebas�án, J. Marquet and R. Schönfeld. Noirbent, T. M. Grant and J. Lalevée, Macromolecules, 2018, 51, Macromolecules, 2011, 44, 4616−4622. 8808-8820. 24 C. Liu, D. Shen, R. M. Sebas�án, J. Marquet and R. Schönfeld. 46 A. H. Bonardi, F. Bonardi, F. Dumur, D. Gigmes, J. P. Fouassier Polymer, 2013, 54, 2873−2878. and J. Lalevée, Macromol. Rapid Commun., 2019, 40, 1900495. 25 W. Zhang, P. Froimowicz, C. R. Arza, S. Ohashi, Z. Xin and H. 47 A. H. Bonardi, F. Dumur, D. Gigmes, Y. Y. Xu and J. Lalevée, ACS Ishida, Macromolecules, 2016, 49, 7129-7140. Omega, 2020, 5, 3043-3046. 8 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Journal Name ARTICLE 48 H. Lee, S. Kim, C. Ryu and J. Y. Lee, ACS Biomater. Sci. Eng., 2020, 6, 1931-1939. 49 A. Caron, F. Dumur and J. Lalevée, J. Polym. Sci., 2020, 58, 2134- 2139. 50 A. H. Bonardi, F. Bonardi, G. Noirbent, F. Dumur, D. Gigmes, C. Dietlin and J. Lalevée. J. Polym. Sci., 2020, 58, 300-308. 51 Y. Liang, Y. Bai, A. Q. Xie, J. Mao, L. Zhu and S. Chen, Solar RRL, 2022, 6, 2100917. 52 P. Hu, H. Xu, Y. Pan, X. Sang and R. Liu, ChemPhysChem, 2022, 23, 202100670. 53 K. M. Haas and B. J. Lear, Chem. Sci., 2015, 6, 6462-6467. 54 E. N. van Burns and B. J. Lear, J. Phys. Chem. C, 2019, 123, 14774-14780. 55 R. Andreu, J. A. Reina and J. C. Ronda, J. Polym. Sci. A: Polym. Chem., 2008, 46, 3353-3366. 56 J. Dunkers and H. Ishida. Spectrochim. Acta A, 1995, 51, 855- 867. 57 M. E. Kuehne and E. A. Konopka, J. Med. Chem., 1962, 5, 257- 280. 58 H. M. Ma, Y. Liu, Y. X. Liu, J. J. Qiu and C. Liu, RSC Adv., 2015, 5, 102441-102447. 59 A. Martos, R. M. Sebastián and J. Marquet, Eur. Polym. J., 2018, 108, 20-27. 60 M. A. Espinosa, V. Cádiz and M. Galia, J. Appl. Polym. Sci., 2003, 90, 470-481. 61 E. Villemin and O. Raccurt, Coord. Chem. Rev., 2021, 435, 213801. 62 H. Ishida and D. P. Sanders, Polymer, 2001, 42, 3115-3125. 63 S. Upadhyay, A. Singh, R. Sinha, S. Omer and K. Negi, J. Mol. Struct., 2019, 1193, 89-102. 64 D. Shen, C. Liu, R. M. Sebastián, J. Marquet and R. Schönfeld, J. Appl. Polym. Sci. A Polym. Chem., 2016, 133, 44099. This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 9 Please do not adjust margins Electronic Supplementary Information (ESI) for Chemical Science. This journal is © The Royal Society of Chemistry 2022 Electronic supplementary information for Photothermal polymerization of benzoxazines Kevin Reyes‐Mateo, Jordi Marquet, Jordi Hernando* and Rosa M. Sebastián* Table of contents 1. Materials and methods .......................................................................................................... S2 2. Synthesis of Bz1‐Bz10 ............................................................................................................. S3 2.1. Synthesis of 1,3,5‐triphenyltriazine, 3...............................................................................S3 2.2. Synthesis of benzoxazines Bz1‐Bz8 ...................................................................................S3 2.3. Synthesis of benzoxazine Bz9 ............................................................................................S6 2.4. Synthesis of benzoxazine B10………………………………………………………………….……………………S6 3. Spectrophotometric characterization of Bz‐catalyst systems…………………………………………..S8 3.1. UV‐vis absorption spectra of Bz‐catalyst systems…………………………………………………………S8 3.1. Studies of metal ion coordination with Bz3 and PBz3 ………………………………..………………S13 4. Polymerization of Bz and Bz‐catalyst systems .................................................................... S14 4.1. Preparation of Bz‐catalyst systems for thermal and photothermal polymerization .....S14 4.2. Determination of monomer conversion .........................................................................S14 4.3. Thermal polymerization .................................................................................................. S14 4.4. Photothermal polymerization .........................................................................................S21 4.5. Chemical stability of Bz3‐catalyst systems ......................................................................S22 5. Analysis of polybenzoxazines ..........................................................................................S23 5.1. Determination of the phenoxy‐phenolic structure ratio ...............................................S23 5.2. Determination of Tg .........................................................................................................S25 5.3. Determination of char yield and thermal decomposition ..............................................S28 5.4. Determination of gel content.........................................................................................S31 5.5. Purification of polymer‐catalyst samples.......................................................................S31 6. Photolithography experiments .......................................................................................S33 7. NMR spectra ...................................................................................................................S34 8. References......................................................................................................................S40 S1 1. Materials and methods Materials. Reagents and solvents for the synthesis of Bz1‐Bz10 as well as the metal salts employed as polymerization catalysts were purchased and used without further purification. Thin layer chromatography (TLC) was performed using Merck aluminium backed plates of TLC silica gel 60 F254, and flash column chromatography was performed using silica gel (230‐400 mesh). Nuclear magnetic resonance (NMR). 1H NMR spectra were recorded on Bruker DPX250 (250 MHz) and DPX360 (360 MHz) spectrometers in CDCl3, acetone‐d6 and DMSO‐d6. The δ‐scale was normalized relative to the residual solvent signal (CDCl3: 7.26 ppm, acetone‐d6: 2.05 ppm and DMSO‐d6: 2.49 ppm). UV‐vis absorption spectrometry. Liquid samples and solutions were measured with an Agilent 8453 UV‐visible spectrophotometer using Hellma Analytics glass cuvettes (1‐cm and 2‐mm light path). Laser irradiation for photothermal polymerization. Unless indicated otherwise, samples were irradiated with the second harmonic (λexc = 532 nm, rep rate = 10 Hz, pulse width = 6 ns) of a Nd:YAG pulsed laser (Brilliant, Quantel). The photopolymerization of Bz3‐Zn(OTf)2 was also attempted with a cw laser at the same wavelength. Temperature sensing. Temperature monitorization during photoirradiation experiments was conducted using a Testo 735 thermocouple. Differential scanning calorimetry (DSC). Samples were analyzed with a TA Instruments Q20 calorimeter using Tzero™ pans and lids calibrated with indium (Tm = 429.75 K, ΔHm = 3267 kJ/mol). For all the samples, we used a heating rate of 10 °C/min and a N2 flow of 50 mL/min. Thermal gravimetric analysis (TGA). Samples were recorded with a TA Instruments TGA550 Discovery Series analyzed using a platinum HT crucible. For all the samples, we used a heating rate of 10 °C/min, a heating range of 30 – 800 °C, a N2 flow of 40 mL/min and an air flow of 60 mL/min. Dynamic mechanical analysis (DMA). Samples were tested in a TA Instruments DMAQ800 thermo‐dynamo‐mechanical analysis machine by using the three‐point bending mode (F = 1 N/min). DMA experiments were performed on 32 x 12 x 3 mm rectangular polymer pieces prepared by either photothermal or thermal polymerization. S2 2. Synthesis of Bz1‐Bz10 2.1. Synthesis of 1,3,5‐triphenyltriazine, 31 In a round‐bottom flask equipped with a Dean‐Stark apparatus, previously distilled aniline (1, 27.3 mL, 0.3 mol) and paraformaldehyde (2, 9.00 g, 0.3 mol) were added in 200 mL of toluene. The mixture was stirred under reflux for 2 h. The reaction crude was cooled down to room temperature, and the white precipitate formed was filtered and washed with n‐hexane. The solid was dried under reduced pressure and triazine 3 was obtained as a crystalline white solid (19.60 g, 62.2 mmol, 62 %). Spectroscopic data 1 H NMR (360 MHz, CDCl3). δ: 7.21 (t, J = 8.0 Hz, 6H), 7.01 (d, J = 8.0 Hz, 6H), 6.86 (t, J = 7.3 Hz, 3H), 4.89 (s, 6H). 2.2. Synthesis of benzoxazines Bz1‐Bz8 Method A: Bz1,1 Bz2,2 Bz3,1 Bz4,3 Bz54 and Bz75 In a round‐bottom flask, 1,3,5‐triphenyltriazine (3, 1 eq.), paraformaldehyde (2, 1 eq.) and the respective phenol (Ph1‐Ph5 and Ph7, 1 eq.) were slowly stirred in bulk and heated up to 100 °C. Once the mixture was melted, stirring was increased during the reaction time (40 min – 2 h). Next, the reaction crude was cooled down to room temperature and dissolved in Et2O (30 – 50 mL)a. The resulted organic phase was washed with 2 M NaOH (3 x 30 – 50 mL) and distilled water (3 x 30 – 50 mL) and finally dried with anhydrous Na2SO4. After filtration, the solvent was a The addition of Et2O must be done BEFORE the solution reaches room temperature, since a hard and insoluble solid is formed at this point. S3 evaporated under vacuum and the obtained benzoxazine was purified by flash column chromatography (silica gel). The exact conditions used for the synthesis and purification of Bz1‐Bz5 and Bz7 are shown in Table S1. Table S1. Experimental conditions for the synthesis of benzoxazines Bz1‐Bz5 and Bz7. Bz Amount of treaction Flash column Yield (%) Appearance phenol chromatography Bz1 2.23 g 2h Hexane/DCM 56% Yellow solid (R=H) (23.7 mmol) (2:1) (2.78 g) Bz2 1.30 g 40 min Hexane/DCM 73% White solid (R=CH3) (12.0 mmol) (3:1) (1.97 g) Bz3 3.70 g 2h Hexane/DCM 77% Yellow (R=OCH3) (29.5 mmol) (2:1) (5.54 g) liquid Bz4 1.43 g 1h Hexane/DCM 27% White solid (R=CN) (12.0 mmol) (7:3) (0.77 g) Bz5 2.10 g 2h No purification 60% Yellow solid (R=COOEt) (12.4 mmol) required (2.10 g) Bz7 1.46 g 1.5 h Hexane/DCM 30% Yellow solid (R=CHO) (12.0 mmol) (1:2) (0.84 g) Spectroscopic data Bz1: 1H NMR (360 MHz, CDCl3). δ (ppm): 7.29 (d, J = 8.1 Hz, 2H), 7.13 (m, 3H), 7.03 (d, J = 7.6 Hz, 1H), 6.95 (t, J = 7.0 Hz, 1H), 6.90 (t, J = 7.7 Hz, 1H), 6.83 (d, J = 8.2 Hz, 1H), 5.38 (s, 2H), 4.65 (s, 2H). Bz2: 1H NMR (360 MHz, CDCl3). δ (ppm): 7.25 (m, 2H), 7.10 (d, J = 7.6 Hz, 2H), 6.92 (t, J = 7.5 Hz, 2H), 6.82 (s, 1H), 6.71 (d, J = 8.3 Hz, 1H), 5.34 (s, 2H), 4.60 (s, 2H), 2.25 (s, 3H). Bz3: 1H NMR (360 MHz, CDCl3). δ (ppm): 7.29 (m, 2H), 7.12 (d, J = 8.0 Hz, 2H), 6.95 (t, J = 7.5 Hz, 1H), 6.77 (d, J = 8.8 Hz, 1H), 6.71 (dd, J = 8.8 Hz, J = 2.8 Hz, 1H), 6.58 (d, J = 2.8 Hz, 1H), 5.33 (s, 2H), 4.62 (s, 2H), 3.75 (s, 3H). Bz4: 1H NMR (360 MHz, CDCl3). δ (ppm): 7.41 (dd, J = 8.5 Hz, J = 2.1 Hz, 1H), 7.30 (m, 3H), 7.10 (d, J = 8.1 Hz, 2H), 7.00 (t, J = 7.4 Hz, 1H), 6.87 (d, J = 8.5 Hz, 1H), 5.45 (s, 2H), 4.66 (s, 2H). Bz5: 1H NMR (250 MHz, CDCl3). δ (ppm): 7.81 (dd, J = 8.7 Hz, J = 1.7 Hz, 1H), 7.76 (d, J = 2.1 Hz, 1H), 7.27 (t, J = 7.7 Hz, 2H), 7.10 (d, J = 7.6 Hz, 2H), 6.95 (t, J = 7.6 Hz, 1H), 6.82 (d, J = 8.7 Hz, 1H), 5.42 (s, 2H), 4.66 (s, 2H), 4.34 (q, J = 7.0 Hz, 2H), 1.37 (t, J = 6.8 Hz, 3H). S4 Bz7: 1H NMR (250 MHz, CDCl3). δ (ppm): 9.83 (s, 1H), 7.65 (d, J = 7.6 Hz, 1H), 7.59 (s, J = , 2H), 7.29 (d, J = 7.6 Hz, 1H), 7.11 (d, J = 8.2 Hz, 2H), 6.97 (t, J = 7.0 Hz, 1H), 6.91 (d, J = 8.8 Hz, 1H), 5.46 (s, 2H), 4.70 (s, 2H). Method B: Bz61 and Bz84 In a round‐bottom flask, 1,3,5‐triphenyltriazine (3), paraformaldehyde (2, 1 eq.) and the respective phenol (Ph6 and Ph8, 1 eq.) were mixed in 1,4‐dioxane (15 mL). The solution was stirred for a given time at 100 °C. The reaction crude was cooled down to room temperature and the solvent was evaporated under vacuum. A yellow liquid was obtained and, after rinsing with Et2O, the desired benzoxazine was achieved that was purified by flash column chromatography (silica gel). The exact conditions used for the synthesis and purification of Bz6 and Bz8 are shown in Table S2. Table S2. Experimental conditions for the synthesis of benzoxazines Bz6 and Bz8. Bz Amount of treaction Flash column Yield (%) Appearance phenol chromatography Bz6 1.58 g 18 h Hexane/DCM 57% Yellow solid (R=NO2) (5.0 mmol) (1:1) (2.20 g) Bz8 2.3 g 2h Hexane/DCM 57% White solid (R=CF3) (8.2 mmol) (1:1) (1.31 g) Spectroscopic data Bz6: 1H NMR (250 MHz, CDCl3). δ (ppm): 8.06‐7.96 (m, 2H), 7.28 (t, J = 8.1 Hz, 2H), 7.10 (d, J = 7.5 Hz, 2H), 6.99 (t, J = 7.3 Hz, 1H), 6.87 (d, J = 8.1 Hz, 1H), 5.47 (s, 2H), 4.70 (s, 2H). Bz8: 1H NMR (300 MHz, CDCl3). δ (ppm): 7.51 (s, 1H), 7.44 (d, J = 8.2 Hz, 1H), 7.28 (d, J = 7.8 Hz, 1H), 7.24 (t, J = 7.7 Hz, 1H), 7.20 (d, J = 8.0 Hz, 2H), 7.18 (t, J = 7.6 Hz, 1H), 6.93 (m, 1H), 5.57 (s, 2H), 4.83 (s, 2H). S5 2.3. Synthesis of bisbenzoxazine Bz92 In a round‐bottom flask, aniline (1, 1.8 mL, 20.0 mmol), paraformaldehyde (2, 1.20 g, 40.0 mmol) and bisphenol A (4, 2.30 g, 10.0 mmol) were mixed without solvent. The resulting solution was stirred for 45 min at 100 °C. The reaction crude was cooled down to room temperature and dissolved in Et2O (≈ 40 mL).b The resulting solution was dried in anhydrous Na2SO4 and filtered. The solvent was evaporated under vacuum and a yellowish solid foam was obtained.c The resulting product was purified by flash column chromatography (silica gel, hexane/AcOEt (9:1)). Bisbenzoxazine Bz9 was afforded in this way as a yellowish solid foam (2.50 g, 5.4 mmol, 54 %). Spectroscopic data H NMR (360 MHz, (CD3)2CO). δ (ppm): 7.23 (t, J = 8.0 Hz, 4H), 7.14 (dd, J = 9.0 Hz, J = 1.2 Hz, 1 4H), 7.00 (d, J = 2.2 Hz, 2H), 6.92 (dd, J = 8.7 Hz, J = 2.5 Hz, 2H), 6.86 (t, J = 7.2 Hz, 2H), 6.62 (d, J = 8.7 Hz, 2H), 5.40 (s, 4H), 4.63 (s, 4H), 1.56 (s, 6H). 2.4. Synthesis of benzoxazine Bz101 OH Ph Ph O N Ph Bulk N N O H3C + + N H H Ph 100 ºC n R CH3 3 2 Ph10 Bz10 b The addition of Et2O must be done BEFORE the solution reaches to room temperature, since a hard and insoluble solid is formed at this point. c A 500 mL round‐bottom flask is recommended to be used for these reagent quantities. When the solvent is evaporated under vacuum, a foam is formed that quickly expands. This is why a high volume is required as it is drying. S6 A mixture of phenol Ph10 (1.83 g, 15 mmol), paraformaldehyde 2 (450 mg, 15 mmol), and 1,3,5‐triphenylhexahydro‐1,3,5‐triazine 3 (1.575 g, 5 mmol) was stirred in air at 100 ℃ for 2 h. After cooled to room temperature, the mixture was dissolved in Et2O and washed three times with 2 M NaOH and three times with water. The organic phase was dried with anhydrous Na2SO4, the diethyl ether was removed at low pressure and Bz10 was obtained as a yellow oil without further purification (2.40 g, 10.1 mmol, 67%). Spectroscopic data H NMR (250 MHz, CDCl3) δ (ppm): 7.25‐7.31 (m, 2H, Ar‐H, overlapping with the signal of 1 CHCl3), 7.12 (d, J = 7.5 Hz, 2H, Ar‐H), 6.93 (t, J = 7.0 Hz, 1H, Ar‐H), 6.81 (s, 1H, Ar‐H), 6.69 (s, 1H, Ar‐H), 5.37 (s, 2H, O‐CH2‐N), 4.59 (s, 2H, Ar‐CH2‐N), 2.23 (s, 3H, CH3), 2.16 (s, 3H, CH3). S7 3. Spectrophotometric characterization of Bz‐catalyst systems 3.1. UV‐vis absorption spectra of Bz‐catalyst systems 2.0 Bz1 Bz1-LiI 1.6 Bz1-Zn(OTf)2 Absorbance 1.2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S1 UV‐vis absorption spectra of Bz1, Bz1‐LiI (cLiI = 50 mol % relative to monomer content) and Bz1‐Zn(OTf)2 (cZn(OTf)2 = 15 mol % relative to monomer content) in acetone (cBz1 = 3.9 M). A 2 mm‐thick glass cuvette was used. 2.0 Bz2 Bz2-LiI 1.6 Bz2-Zn(OTf)2 Absorbance 1.2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S2 UV‐vis absorption spectra of Bz2, Bz2‐LiI (cLiI = 50 mol % relative to monomer content) and Bz2‐Zn(OTf)2 (cZn(OTf)2 = 15 mol % relative to monomer content) in acetone (cBz2 = 3.9 M). A 2 mm‐thick glass cuvette was used. S8 2.0 Bz3 Bz3-LiI 1.6 Bz3-Zn(OTf)2 Absorbance 1.2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S3 UV‐vis absorption spectra of Bz3, Bz3‐LiI (cLiI = 50 mol % relative to monomer content) and Bz3‐Zn(OTf)2 (cZn(OTf)2 = 15 mol % relative to monomer content) in bulk liquid. A 2 mm‐thick glass cuvette was used. 2.0 Bz4 Bz4-LiI 1.6 Bz4-Zn(OTf)2 Absorbance 1.2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S4 UV‐vis absorption spectra of Bz4, Bz4‐LiI (cLiI = 50 mol % relative to monomer content) and Bz4‐Zn(OTf)2 (cZn(OTf)2 = 15 mol % relative to monomer content) in acetone (cBz4 = 1.2 M). A 2 mm‐thick glass cuvette was used. S9 2.0 Bz5 Bz5-LiI 1.6 Absorbance 1.2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S5 UV‐vis absorption spectra of Bz5 and Bz5‐LiI (cLiI = 50 mol % relative to monomer content) in acetone (cBz5 = 2.4 M). A 2 mm‐thick glass cuvette was used. 2.0 Bz6 1.6 Bz6-LiI Absorbance 1.2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S6 UV‐vis absorption spectra of Bz6 and Bz6‐LiI (cLiI = 50 mol % relative to monomer content) in acetone (cBz6 = 1.0 M). A 2 mm‐thick glass cuvette was used. S10 2.0 Bz7 1.6 Bz7-LiI Absorbance 1.2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S7 UV‐vis absorption spectra of Bz7 and Bz7‐LiI (cLiI = 50 mol % relative to monomer content) in acetone (cBz7 = 2.0 M). A 2 mm‐thick glass cuvette was used. 2.0 Bz8 1.6 Bz8-LiI Absorbance 1.2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S8 UV‐vis absorption spectra of Bz8 and Bz8‐LiI (cLiI = 50 mol % relative to monomer content) in acetone (cBz8 = 1.9 M). A 2 mm‐thick glass cuvette was used. S11 2.0 Bz9 Bz9-LiI 1.6 Bz9-Zn(OTf)2 Absorbance 1.2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S9 UV‐vis absorption spectra of Bz9, Bz9‐LiI (cLiI = 50 mol % relative to monomer content) and Bz9‐Zn(OTf)2 (cZn(OTf)2 = 15 mol % relative to monomer content) in acetone (cBz9 = 2.4 M). A 2 mm‐thick glass cuvette was used. 2.0 LiI Zn(OTf)2 1.6 LiClO4 FeCl3 Absorbance 1.2 ZnI2 0.8 0.4 0.0 350 400 450 500 550 600 650 700 750 800 Wavelength (nm) Fig. S10 UV‐vis absorption spectra of LiI (cLiI = 1 M), Zn(OTf)2 (cZn(OTf)2 = 0.3 M), LiClO4 (cLiClO4 = 0.3 M), FeCl3 (cFeCl3 = 0.3 M) and ZnI2 (cZnI2 = 0.3 M) in acetone. S12 3.2. Studies of metal ion coordination with Bz3 and PBz3 To determine the stoichiometries and association constants for the complexation between metal ions and benzoxazine monomers and polymers, we followed the same procedure previously used by us for similar systems.6 In particular, we used UV‐vis spectrophotometry to monitor the spectral changes induced when: (a) adding increasing concentrations of LiI to neat liquid Bz3 (Fig. S11a); (b) adding increasing concentrations of LiI to an acetone solution of a polybenzoxazine obtained by thermal polymerization of Bz3 (Fig. S11b). The differences in absorbance measured in the visible region upon metal salt addition were then plotted against the metal salt concentration (λabs = 480 and 500 nm for Fig. S11a and S11b, respectively) and fitted to the Benesi–Hildebrand equation (1). In this equation, A is the absorbance upon metal salt addition, A0 is the absorbance without metal salt, εc is the molar absorption coefficient of the complex, ε0 is the molar absorption coefficient of the initial compound, [N0] is the concentration of the initial compound, Ka is the association constant, [M] is the concentration of the metal salt for each addition, and n is the stoichiometric ratio. 1 1 1 1 1 𝐴 𝐴 𝜀 𝜀 𝑁 𝜀 𝜀 𝑁 𝐾 𝑀 Fig. S11 (a) Variation of the UV‐vis absorption spectrum of neat Bz3 upon addition of increasing concentrations of LiI relative to monomer content. (b) Variation of the UV‐vis absorption of thermally polymerized PBz3 in acetone (cPBz3 = 20 mg mL‐1) upon addition of increasing concentrations of LiI relative to polymer content. S13 4. Polymerization of Bz and Bz‐catalyst systems 4.1. Preparation of Bz‐catalyst systems for thermal and photothermal polymerization Each given (bis)benzoxazine (0.2 – 0.3 g for Bz1, Bz2 and Bz4‐Bz9, 0.5 mL for Bz3 and Bz10) was mixed with the catalyst of interest (typically, 15 mol % relative to monomer content) in a few milliliters of acetone.d Once the mixture was completely solved, acetone was removed under reduced pressure and the corresponding benzoxazine‐catalyst system was afforded. 4.2. Determination of monomer conversion The conversion of all tested benzoxazine monomers during polymerization reactions was determined by 1H NMR using previously reported equation 2.7,8 In this equation: (a) X is the integration of the 1H NMR signals characteristic of the methylene units in the oxazine ring of the unreacted monomers (δ = 4.6 and 5.4 ppm in CDCl3); (b) Y is the sum of the integrals of the methylene signals of the unreacted monomers and the formed polymers (δ = 3.5 – 6.0 ppm in CDCl3) at a given polymerization time. 𝑌 2𝑋 𝐶𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛 % 𝑌 100 2 4.3. Thermal polymerization Neat benzoxazine thermal polymerization: Each of the (bis)benzoxazines of interest was heated in a sand bath at 180 °C for 2 h. Thermal polymerization of benzoxazine‐catalyst systems: Each of the (bis)benzoxazine‐catalyst samples prepared (see section 4.1) was heated in a sand bath at 180 °C for 2 h. Determination of thermal polymerization temperature (TP): DSC was used to determine the thermal polymerization temperature of neat Bz monomers and monomer‐catalyst systems. Two main thermal transitions were identified in the thermograms of these samples: (a) an endothermic signal arising from the melting of the Bz monomer (except for liquid Bz3); (b) an exothermic peak assigned to the polymerization reaction. The onset and peak temperatures of the DSC polymerization signal determined for all the samples under study are given in Table S3, while the corresponding thermograms are shown in Figs. S12‐S21. d Huge solvent volumes are not required for LiI, since its solubility in acetone is around 400 mg/mL. By contrast, Zn(OTf)2 is barely soluble in acetone at room temperature. For this reason, the metal salt must be solved in hot acetone under vigorous stirring. S14 Table S3. Thermal polymerization temperature of neat Bz monomers and Bz‐catalyst systems. Bz Catalyst ccatalyst (mol %)a Tp, onset (ºC) Tp, peak (ºC) Bz1 ‐ ‐ 265.1 269.7 LiI 15 184.7 194.0 Zn(OTf)2 15 61.8 77.2 Bz2 ‐ ‐ 263.9 273.6 LiI 15 189.6 201.2 Zn(OTf)2 15 77.2 87.0 Bz3 ‐ ‐ 249.4 258.9 LiI 15 141.2 159.4 Zn(OTf)2 15 56.4 82.4 Bz4 ‐ ‐ 220.9 225.0 LiI 15 167.5 180.4 Zn(OTf)2 15 93.5 99.6 Bz5 ‐ ‐ 243.9 249.0 LiI 15 171.1 188.9 Bz6 ‐ ‐ 189.4 194.5 LiI 15 153.9 167.2 Bz7 ‐ ‐ 223.9 229.2 LiI 15 138.1 159.5 Bz8 ‐ ‐ 248.5 250.6 LiI 15 181.2 190.6 Bz9 ‐ ‐ 214.6 232.3 LiI 15 160.2 182.8 Bz10 ‐ ‐ 271.2 281.9 LiI 15 177.8 199.5 a Relative to monomer concentration. S15 Fig. S12 DSC thermogram of neat Bz1 and Bz1‐catalyst systems. Fig. S13 DSC thermogram of neat Bz2 and Bz2‐catalyst systems. S16 Fig. S14 DSC thermogram of neat Bz3 and Bz3‐catalyst systems. Fig. S15 DSC thermogram of neat Bz4 and Bz4‐catalyst systems. S17 Fig. S16 DSC thermogram of neat Bz5 and Bz5‐catalyst systems. Fig. S17 DSC thermogram of neat Bz6 and Bz6‐catalyst systems. S18 Fig. S18 DSC thermogram of neat Bz7 and Bz7‐catalyst systems. Fig. S19 DSC thermogram of neat Bz8 and Bz8‐catalyst systems. S19 Fig. S20 DSC thermogram of neat Bz9 and Bz9‐catalyst systems. Fig. S21 DSC thermogram of neat Bz10 and Bz10‐catalyst systems. S20 4.4. Photothermal polymerization General procedure: Each of the neat Bz monomers or monomer‐catalyst systems (prepared as described in section 4.1 were placed either within a glass cuvette (50 x 0.8 x 0.4 mm) for solid samples (Bz1, Bz2 and Bz4‐Bz9, 0.2 – 0.3 g) (Fig. S22a) or within a glass vial (Ø = 1 cm) for liquid samples (Bz3 and Bz10, 0.5 mL) (Fig. S22b). Then, they were irradiated with a pulsed green laser under ambient conditions for the required time (Fig. S22c). a b c Fig. S22 (a) Solid and (b) liquid samples used in the photothermal polymerization experiments, which were mainly conducted employing (c) the 2nd harmonic of a ns‐pulsed Nd:YAG laser as an irradiation source. 100 100 90 90 80 80 70 70 Temperature (ºC) 60 Conversion (%) 60 50 50 40 40 30 30 20 20 10 10 Temperature Conversion 0 0 0 5 10 15 20 25 30 Time (min) Fig. S23 Evolution of the reaction mixture temperature and monomer conversion during visible light irradiation of Bz3‐LiI (cLiI = 15 mol % relative to monomer concentration; λexc = 532 nm, power density = 1.1 W cm‐2). At tirr = 22.5 min when ca. 50% of monomer conversion had been S21 reached, irradiation was stopped, which resulted in a sudden termination of the polymerization process and a more gradual decrease of temperature. Frontal polymerization experiments: To investigate the self‐propagation capacity of the photothermal polymerization process, frontal polymerization experiments were carried out. In these experiments, only a small region of the Bz‐catalyst samples was irradiated to induce photopolymerization, while the monomer conversion was investigated for the nonirradiated area (Fig. S24). Covered surface Fig. S24 Schematic representation of a frontal polymerization experiment. Pulsed laser irradiation of a specific area was conducted for 100 min (λexc = 532 nm, Øspot = 1 cm, power = 1.1 W cm‐2). 4.5. Chemical stability of Bz3‐catalyst systems Benzoxazine Bz3 was mixed with LiI (15 mol % relative to monomer concentration) and Zn(OTf)2 (15 mol % relative to monomer content), respectively, in 2‐3 mL of deuterated acetone. Both solutions were stirred for several days at room temperature. 1H NMR analysis were performed every 24 h in order to check the intensity variation of benzoxazine peaks at δ  5.4 and 4.6 ppm. S22 5. Analysis of polybenzoxazines 5.1. Determination of the phenoxy‐phenolic structure ratio The phenoxy‐phenolic structure ratio of photothermally polymerized PBz‐catalyst samplese and their annealed products was determined by a variation of the procedure developed in previous studies in our research group.7,8 The 1H NMR signals located between δ  4.0 – 6.0 ppm and 3.5 – 4.0 ppm were assigned to phenoxy‐type CH2 and phenolic‐type CH2 units, respectively. If m and n represent the signal integration of these phenoxy and phenolic CH2 units, respectively, the phenoxy‐to‐phenolic ratio can be obtained as (in %): [m/(m+n)]∙100 (Figs. S25‐S28). Other typical signals found in the 1H NMR spectra of polybenzoxazines are found at: (a) δ  6.0 – 8.0 ppm, which arise from the aromatic protons of the polymers; (b) δ  8.0 – 9.5 ppm, which are generally attributed to residual imine formation during the polymerization process.9 Thermal annealing of photothermally produced PBz1‐PBz4 was performed by heating up the corresponding polymers at 150 °C for 2 h. It must be noted that this resulted in the disappearance of most of the resonances in the δ  4.0 – 6.0 ppm region, thus indicating thermal phenoxy‐to‐phenolic conversion of the polymer structure. m n Fig. S25 1H NMR spectra (360 MHz, DMSO‐d6) of the PBz obtained by photothermal polymerization of Bz1‐LiI (blue) and subsequent thermal annealing (red). e Phenoxy‐phenolic structure ratio of the photothermally polymerized Bz9‐LiI system and its annealed product could not be measured due to their poor solubility in all solvents. S23 Fig. S26 1H NMR spectra (360 MHz, DMSO‐d6) of the PBz obtained by photothermal polymerization of Bz2‐LiI (blue) and subsequent thermal annealing (red). Fig. S27 1H NMR spectra (360 MHz, DMSO‐d6) of the PBz obtained by photothermal polymerization of Bz3‐LiI (blue) and subsequent thermal annealing (red). S24 Fig. S28 1H NMR spectra (360 MHz, DMSO‐d6) of the PBz obtained by photothermal polymerization of Bz4‐LiI (blue) and subsequent thermal annealing (red). 5.2. Determination of Tg Transition glass temperatures of thermally‐polymerized PBz1‐PBz4 and PBz9 and photothermally‐polymerized PBz1‐PBz4 and PBz9 were measured by DSC analysis (Figs. S29‐ S33). S25 0.4 0.2 Photothermal polybenzoxazine (Zn(OTf)2) Heat Flow (W/g) 0.0 Photothermal polybenzoxazine (LiI) -0.2 Thermal polybenzoxazine -0.4 0 20 40 60 80 100 Exo Down Temperature (°C) Universal V4.5A TA Instruments Fig. S29 DSC thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz1. 0.4 Photothermal polybenzoxazine (Zn(OTf)2) 0.2 Photothermal polybenzoxazine 0.0 (LiI) Heat Flow (W/g) -0.2 Thermal polybenzoxazine -0.4 0.4 0.2 -0.6 0.0 -0.8 -0.2 120 140 160 180 200 220 240 260 -1.0 0 20 40 60 80 100 120 Exo Down Temperature (°C) Universal V4.5A TA Instruments Fig. S30 DSC thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz2. S26 0.3 Photothermal polybenzoxazine (Zn(OTf)2) Heat Flow (W/g) Thermal polybenzoxazine 0.1 Photothermal polybenzoxazine (LiI) -0.1 0 20 40 60 80 100 120 Exo Down Temperature (°C) Universal V4.5A TA Instruments Fig. S31 DSC thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz3. 0.4 Photothermal polybenzoxazine (Zn(OTf)2) 0.2 Photothermal polybenzoxazine Heat Flow (W/g) (LiI) 0.0 Thermal polybenzoxazine 0.4 -0.2 0.2 0.0 140 160 180 200 -0.4 0 20 40 60 80 100 120 Exo Down Temperature (°C) Universal V4.5A TA Instruments Fig. S32 DSC thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz4. S27 0.5 Thermal polybenzoxazine Heat Flow (W/g) 0.0 Photothermal polybenzoxazine (LiI) Photothermal polybenzoxazine (Zn(OTf)2) -0.5 0 20 40 60 80 100 120 Exo Down Temperature (°C) Universal V4.5A TA Instruments Fig. S33 DSC thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz9. 5.3. Determination of char yield and thermal decomposition The char yields of thermally‐polymerized PBz1‐PBz4 and PBz9 and photothermally‐ polymerized PBz1‐PBz4 and PBz9 were registered by thermogravimetric analysis (Figs. S34‐ S38). To assess the thermal stability of the polymers, we shown in Table S3 the temperatures for 5% and 10% weight loss determined from the TGA measurements. 100 Bz1 90 Bz1-LiI Bz1-Zn(OTf)2 80 Weight (%) 70 60 50 40 30 0 100 200 300 400 500 600 700 800 Temperature (ºC) Fig. S34 TGA thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz1. S28 100 Bz2 90 Bz2-LiI Bz2-Zn(OTf)2 80 Weight (%) 70 60 50 40 30 0 100 200 300 400 500 600 700 800 Temperature (ºC) Fig. S35 TGA thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz2. 100 Bz3 90 Bz3-LiI Bz3-Zn(OTf)2 80 Weight (%) 70 60 50 40 30 0 100 200 300 400 500 600 700 800 Temperature (ºC) Fig. S36 TGA thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz3. S29 100 Bz4 Bz4-LiI 90 Bz4-Zn(OTf)2 Weight (%) 80 70 60 50 0 100 200 300 400 500 600 700 800 Temperature (ºC) Fig. S37 TGA thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz4. 100 Bz9 90 Bz9-LiI Bz9-Zn(OTf)2 80 Weight (%) 70 60 50 40 30 0 100 200 300 400 500 600 700 800 Temperature (ºC) Fig. S38 TGA thermograms of the PBz materials obtained through the thermal (no catalyst) and photothermal polymerization of Bz9. S30 Table S4. T5% and T10% of neat thermal PBz1‐PBz4 and PBz9 and their photothermal PBz‐ catalyst systems. Bz Catalyst T5% (ºC) T10% (ºC) Bz1 ‐ 246 290 LiI 159 255 Zn(OTf)2 256 286 Bz2 ‐ 257 277 LiI 216 288 Zn(OTf)2 269 300 Bz3 ‐ 232 269 LiI 97 234 Zn(OTf)2 200 241 Bz4 ‐ 342 369 LiI 145 293 Zn(OTf)2 309 353 Bz9 ‐ 312 344 LiI 244 274 Zn(OTf)2 252 273 5.4. Determination of gel content The gel content of selected photothermally‐ and thermally‐prepared polymers was determined as previoulsy reported by us.10 About 0.7 g of the polymer material of interest were subjected to a 24 h Soxhlet extraction using acetone as a solvent. This led to dissolution of linear polymer chains, leaving behind the insoluble cross‐linked 3D network. Therefore, comparison of the final insoluble material left in the cellulose cartridge used in the extraction (mgel) with the initial polymer mass (m0) allowed us the calculation of the gel content using equation (3): 𝑚 𝐺𝑒𝑙 𝑐𝑜𝑛𝑡𝑒𝑛𝑡 % 100 3 𝑚 5.5. Purification of polymer‐catalyst samples To explore the purification of polymer‐catalyst samples, we selected the polymer material obtained from the photopolymerization of the Bz3‐Zn(OTf)2 system (cZn(OTf)2 = 15 mol %) as a S31 benchmark case. About 0.5 g of this material were suspended within 5 mL of cold acetone and the resulting mixture was stirred for 4 h in an ice bath. After this time, most of the initial polymer had been dissolved, which was separated from the remaining solid by filtration. After solvent evaporation of the filtrate, the dissolved polymer was analyzed by 1H NMR and F 19 NMR in DMSO‐d6 and the obtained spectra were compared with those registered before purification. In the case of 19F NMR, a controlled quantity of trifluoroacetic acid (10 µL) was added as an internal standard to quantify the decrease in the integral of the signal corresponding to the triflate anions after purification. S32 6. Photolithography experiments Photolithography experiments were performed with a NEJE Master 2/2s laser‐engraving machine equipped with a focusable 405 nm cw laser (33 W cm‐2) and the NEJE v5.6 software (Fig. S39). Samples were prepared by depositing a Bz3/Bz9‐Zn(OTf)2 formulation onto a glass substrate (9:1 Bz3:Bz9 ratio, ccatalyst = 15 mol % relative to the total monomer content), which was then subjected to photolithography experiments. To remove the excess of unreacted monomer‐catalyst system, the samples were gently rinsed with an organic solvent (e.g., ethyl acetate). Sample Fig. S39 Image of the laser‐engraving machine used for the photolithography experiments. S33 7. NMR spectra Fig. S40 1H NMR spectrum (360 MHz, CDCl3) of 1,3,5‐triphenyltriazine (3). S34 Fig. S41 1H NMR spectrum (360 MHz, CDCl3) of Bz1. Fig. S42 1H NMR spectrum (360 MHz, CDCl3) of Bz2. S35 Fig. S43 1H NMR spectrum (360 MHz, CDCl3) of Bz3. Fig. S44 1H NMR spectrum (360 MHz, CDCl3) of Bz4. S36 Fig. S45 1H NMR spectrum (360 MHz, CDCl3) of Bz5. Fig. S46 1H NMR spectrum (360 MHz, CDCl3) of Bz6. S37 Fig. S47 1H NMR spectrum (250 MHz, CDCl3) of Bz7. Fig. S48 1H NMR spectrum (300 MHz, (CD3)2CO) of Bz8. S38 Fig. S49 1H NMR spectrum (360 MHz, CDCl3) of Bz9. S39 8. References 1 R. Andreu, J. A. Reina and J. C. Ronda, J. Polym. Sci. A: Polym. Chem., 2008, 46, 3353‐3366. 2 M. A. Espinosa, V. Cádiz and M. Galia, J. Appl. Polym. Sci., 2003, 90, 470‐481. 3 J. Salabert, R. M. Sebastián and J. Marquet. Macromolecules, 2018, 51, 3672‐3679. 4 A. Martos, R. M. Sebastián and J. Marquet. Eur. Polym. J. 2018, 108, 20‐27. 5 H. M. Ma, Y. Liu, Y. X. Liu, J. J. Qiu and C. Liu, RSC Adv., 2015, 5, 102441‐102447. 6 P. Chutayothin and H. Ishida, Macromolecules, 2010, 43, 4562‐4572. 7 C. Liu, D. Shen, R. M. Sebastián, J. Marquet and R. Schönfeld, Polymer, 2013, 54, 2873−2878. 8 C. Liu, D. Shen, R. M. Sebastián, J. Marquet and R. Schönfeld. Macromolecules, 2011, 44, 4616−4622. 9 P. Chutayothin and H. Ishida, Macromolecules, 2010, 43, 4562‐4572. 10 A. Martos, M. Soto, H. Schäfer, K. Koschek, J. Marquet and R. M. Sebastián, Polymers, 2020, 12, 254. S40