pH-Switchable, Photoactive Microporous Polymer Networks from ...

Article pubs.acs.org/Macromolecules

Teaching New Tricks to an Old Indicator: pH-Switchable, Photoactive Microporous Polymer Networks from Phenolphthalein with Tunable CO2 Adsorption Power Baris Kiskan,*,† Markus Antonietti, and Jens Weber* Department of Colloid Chemistry, Max Planck Institute of Colloids and Interfaces, Science Park Golm, D-14424 Potsdam, Germany S Supporting Information *

ABSTRACT: Switchable, organic microporous networks were synthesized by Sonogashira coupling of tetrabromophenolphthalein with 1,4-diethynylenebenzene using optimized reaction conditions. The resulting networks are microporous and have specific surface areas exceeding 800 m2 g−1. The microporosity and the pore polarity are sensitive to the pH value as evidenced by nitrogen and carbon dioxide physisorption experiments. The switching between the open and closed form of the lactone ring is reversible, but some porosity is lost throughout the process. The colored, alkaline salts of these networks are photochemically active, as shown by the effective heterogeneous photosensitization of the photopolymerization of methyl methacrylate with visible light.

■

INTRODUCTION Microporous polymers recently attracted a great deal of attention. Major advances have been made regarding the synthesis of various architectures, including both soluble and cross-linked polymers.1−4 So far, stable permanent microporosity in polymers requires the use of highly rigid linkages. Consequently, most microporous polymers are based on fully aromatic, conjugated motifs (e.g., CMPs)2,5,6 or comparable fully aromatic high-performance polymer systems (e.g., polyimides).7−10 Such polymers were suggested for applications in various fields, e.g., gas storage, separation, and catalysis. On top of that, there is always an interest in materials, which can change their properties by applying an external stimulus, i.e., in switchable materials. Recently, switchable microporosity has been demonstrated for molecular organic solids or metal organic frameworks,11,12 but to the best of our knowledge there are no reports on switchable microporous polymer networks. Here we present microporous conjugated networks based on a phenolphthalein building block as the switching unit. Phenolphthalein is a well-known, easy to synthesize indicator molecule which undergoes a structure change upon changes in pH value. A rigid, contorted molecule as such, the central carbon atom switches from a sp3 hybridization in neutral state to a sp2 state at high pH, going along with the formation of a more polar quinonic structure giving rise to the well-known color change from transparent to red. At high pH, the lacton ring in the phenolphthalein is opened and forms a salt. Additionally, one of the phenol groups is transformed into a phenate. The ease of access and the contorted nature were employed before by using phenolphthalein ethers as model glass builders13 or for high gas permeation membrane polymers.14 In those both cases, modification however involves © 2012 American Chemical Society

the phenol groups, which irreversibly blocks the molecule in the nonconjugated lactone structure. This is why it is intended to incorporate the phenolphthalein tectons keeping the phenol groups intact. We then expect the network structure to be switchable with pH, while the micropore morphology and polarity should be sensitive toward the conformation change of the central building unit. Furthermore, using bases with different counterion size (e.g., Li+ vs Cs+) should result in a further fine-tuning of the pore size and the pore volume as well as in the introduction of potential gas selectivities.

■

MATERIALS AND METHODS

Materials. All the chemicals and solvents were obtained from Sigma-Aldrich and used as received. Anhydrous grade N,Ndimethylformamide, triethylamine, and diisoproylamine were used throughout (Sigma-Aldrich). All chemicals used had a purity of 97% or greater. Methods. Gas Sorption Analysis. Polymer surface areas and pore size distributions were measured by nitrogen adsorption and desorption at 77.3 K using Autosorb 1MP Instrument (Quantachrome Instruments). Carbon dioxide sorption isotherms were collected at 273 K using the same device. High purity gases were used. Data evaluation was performed using the AS1Win Software from Quantachrome Instruments. Pore size distributions and pore volumes were derived from the adsorption branches of the isotherms using the quenched solid density functional theory (QSDFT, N2, assuming carbon adsorbent with slit pores) or the grand canonical Monte Carlo (GCMC) method (CO2, assuming carbon adsorbent with slit pores). Samples were degassed at 105 °C for 24 h under high vacuum before Received: December 9, 2011 Revised: January 23, 2012 Published: January 31, 2012 1356

dx.doi.org/10.1021/ma202675v | Macromolecules 2012, 45, 1356−1361

Macromolecules

Article

analysis. CO2 heats of sorption were calculated from isotherms obtained at 273 and 283 K using the AS1Win software. Spectroscopy. FT-IR spectra were collected in attenuated total reflection (ATR) mode on a Thermo Nicolet FT-IR Nexus 470. 13C CP-MAS NMR experiments were acquired on a Bruker DMX 300WB (7 T) spectrometer using a 4 mm zirconia rotors spinning at the MAS frequency of νMAS = 11 kHz. Cross-polarization transfers were performed using adiabatic tangential ramps, the contact time was tCP = 3 ms, recycle delay was 3 s, and Spinal64 decoupling was applied during signal acquisition. The FIDs were subjected to an exponential multiplication (EM) function with a line broadening value of 40 Hz prior to Fourier transform (collected by Simone Mascotto at Materials Chemistry Group, Dipartimento di Ingegneria dei Materiali e Tecnologie Industriali, Via Mesiano 77, 38123 Trento, Italy). Scanning electron microscopy (SEM) was undertaken using a Gemini 1550 (LEO, Zeiss AG). Acceleration voltage: 3 kV; working distance: 5 mm. Samples were coated with a thin layer of Pd/C before investigation. Molecular Modeling. Geometry optimizations were undertaken using Gaussian 03W.15 Models of the central motif were optimized using the DFT calculations (B3LYP/6-31G). Frequency calculations were undertaken to ensure that the obtained structures are minimum structures. UV−vis spectra were calculated using the SWizard Program (S.I. Gorelsky, SWizard program) to analyze the output of the TD calculations performed using Gaussian 03. Typical input files are given in the Supporting Information. Synthesis of Phenolphthalein-Based Networks. All of the networks were synthesized by Sonogashira−Hagihara cross-coupling condensation reaction of aryl halides and arylethynylenes. In a typical procedure, 3′,3″,5′,5″-tetrabromophenolphthalein (600 mg, 1 mmol), 1,4-diethynylbenzene (360 mg, 2.8 mmol), tetrakis(triphenylphosphine)palladium(0) (40 mg, 0.035 mmol), and copper(I) iodide (20 mg, 0.1 mmol) were placed in a dry 25 mL Schlenk tube. The solids were dissolved in a mixture of anhydrous N,Ndimethylformamide (DMF) (5 mL) and anhydrous diisopropylamine (DIA) (5 mL). The content was degassed by the freeze−thaw method and backfilled with argon gas three times before closing the tube. Then the tube was placed in oil bath at 90 °C and stirred for 5 days. The solid product was then collected by filtration and washed well with DMF, acetone, THF, and methanol before undergoing Soxhlet extraction with methanol for 16 h. The products were then placed in 1,4-dioxane for solvent exchange, and after three extractions all the content was freeze-dried and finally vacuum-dried at 40 °C to give insoluble yellowish solids. Preparation of Network Salts. All the network salts are prepared in 1 M of the corresponding methanol−water (v/v, 1:1) solutions of LiOH, NaOH, KOH, and CsOH. The solid networks treated with those solutions for 24 h; solid products were then collected by filtration and washed well with copious amounts of water, methanol, and 1,4-dioxane. The network salts are freezed and vacuum-dried subsequently. Photopolymerization of MMA with Na-N1. Typical polymerization procedure: A stirred mixture of MMA (0.5 mL, 4.7 M), Na-N1 (10 mg), and triethylamine was irradiated in a photoreactor consisting of a 300 W xenon lamp equipped with a 420 nm cutoff filter and a water cooling system under an Ar atmosphere for 3 h. At the end of irradiation, the bulk solid was dissolved in THF, filtered, and precipitated in methanol. The collected polymers were dried under vacuum at 40 °C. Conversion: 45−82%. Mn: typically around 140 000 g mol−1 (GPC, eluent: THF, against PMMA standards)

Scheme 1. Synthesis Pathway and pH Switch of Phenolphthalein Networksa

a

(a) Synthetic pathway toward phenolphthalein networks; (b) structure of network N1 and the corresponding sodium salt form Na-N1,

used base from triethylamine (TEA) to diisoproylamine (DIA), however, resulted in an increase of the maximum specific surface areas up to 813 m2 g−1. This effect was also observed for the synthesis of plain CMP-1,17 and we consequently do not regard it as a special feature of the chosen monomer system. We suspect that this increase is due to altered solvent structures and conditions (the base makes up half of the used solvent volume),16 thus creating some type of solvent template effect, or due to a more effective cross-coupling (DIA is typically used for Sonogashira cross-couplings in organic chemistry instead of TEA). As DIA resulted in networks of higher porosity, we restricted ourselves to its use for the rest of the study. In order to study the switching upon pH with sufficient precision, we synthesized a single large batch of the parent network (N1), from which all switching experiments were undertaken. We are confident that this allows direct comparability of the samples. Treatment of the network N1 with alkali bases (LiOH, NaOH, KOH, and CsOH) in methanol−water mixture (v/v, 1:1) resulted in the formation of structurally switched networks (Scheme 1b, Li-N1, K-N1, etc.). After isolation of the switched networks by filtration, their identity was confirmed by spectroscopy (FTIR, solid-state CP-MAS 13C NMR). Furthermore, the color of the materials changed from yelloworange in the lactone form to dark yellow-greenish in the salt form. This color change is very well-known from phenolphthalein derivatives and indicates already a successful transformation. On a macroscopic scale, almost no changes in the network morphology were observed as evidenced by SEM (Figure 1). Upon freeze-drying, an open, spongelike network structure is observed, having its origin most probable in the polymerization induced phase separation. The networks feature macroporosity, which is in accordance with the nitrogen porosimetry measurements (see below). Such an open structure is beneficial for applications, which demand high mass transport, as for example catalysis. SEM analysis was also helpful in to exclude the presence of excess base. If excess base is present, it forms crystallites upon drying, which can be clearly identified (not shown). The absence of these crystallites indicates sufficient purification.

■

RESULTS AND DISCUSSION Network Synthesis and Switching. The synthesis of the networks starts from the commercial tetrabromo derivative (1). Sonogashira coupling with 1,4-diethynylbenzene (2) yields the microporous network (see Scheme 1). Initial experiments were undertaken following the conditions described by Cooper and co-workers.16 The resulting networks had apparent specific surface areas (BET model) of SBET = 275 m2 g−1. Changing the 1357


Macromolecules

Article

Additional CO2 sorption experiments confirmed the presence of microporosity. By comparison of the pore size distributions (PSD) obtained from CO2 and N2 isotherms, it is possible to estimate the predominant pore sizes. We found ultramicropores (pore diameters ∼0.6 nm, from CO2 data, see Figure S3) as well as larger pores (mean diameter ∼1.1−1.3 nm, from N2 data, see Figure 2 and Figure S3). As CO2 at 273 K can only

Figure 2. (a) N2 adsorption/desorption isotherms of N1, its salt forms and networks after multiple switches, data obtained at 77.3 K, isotherms are vertically offset by 100 m2 mmol−1 STP for better visibility, (b) PSD of selected samples as obtained from QSDFT analysis (carbon slit pores) of the N2physisorption isotherms, (c) CO2 adsorption/desorption isotherms of selected samples, obtained at 273.15 K, (d) isosteric heat of CO2 adsorption for N1, Li-N1, and CsN1, obtained from CO2 adsorption isotherms at 273 and 283 K as a function of CO2 loading.

Figure 1. 3D model of the optimized geometry of the central network motif in neutral form (a, N1) and in salt form (b, Na-N1); SEM micrograph of freeze-dried N1 (b) and freeze-dried Na-N1 (c); solidstate CP-MAS 13C NMR spectra of N1 (d) and Na-N1 (e); ATRFTIR spectra of N1 (f) and Na-N1 (g).

The switch process was also analyzed on a molecular scale using spectroscopy. FTIR analysis showed a strong decrease and a slight shift of the CO band of the lacton ring at 1774 to 1758 cm−1. The band at 1600 cm−1 shifts to 1572 cm−1; moreover, the appearance of new bands at 1553 cm−1 that could be attributed to carboxylate and/or conjugated CO groups are evidence of successful transformation (Figure 1). The switch is reversible, and after switching back by acid treatment, the bands at 1758, 1572 cm−1 shift back to 1775, 1602 cm−1 and the new band at 1553 cm−1 disappears (see Figure S1). Solid-state NMR could also be used to prove the change of the chemical structure. The formation of carboxylate is evidenced by the appearance of a weak peak at 174 ppm, while the intensities of peaks at 91, 147, and 169 ppm decrease due to the ring-opening of the lactone (Figure 1). Porosity Analysis. Porosity analysis of the parent network N1 was undertaken by gas sorption analysis (N2 at 77.3 K and CO2 at 273 K).18 The apparent specific surface area was found to be SBET = 813 m2 g−1. The isotherm was of type I shape, indicating predominantly microporosity. At high relative pressures, we found a large additional N2 uptake, which can be related to condensation in the macropores of the network.

probe ultramicropores, analysis of the N2 adsorption isotherm resulted in higher specific surface areas compared to the results of the CO2 analysis, the difference being due to the presence of larger micropores. Porosity analysis of the switched polymers was again performed by N2 and CO2 gas adsorption/desorption (Figure 2 and Figure S3). Table 1 summarizes all data, and Figure 2 shows representative N2 and CO2 physisorption isotherms obtained at 77 and 273 K, respectively. In the following, we will discuss porosity parameters normalized with regard to the molecular weight of the repeating unit, i.e., surface areas are given in m2 mmol−1 instead of m2 g−1. Doing so, we reduce the problem of the high specific weight of the Cs+ ion when comparing the different materials. All salt forms show a pronounced microporosity as indicated by their N2 adsorption behavior. Interestingly, the material LiN1 showed a higher surface area compared to its parental network (please note that 813 m2 g−1 correspond to 457 m2 mmol−1). The incorporation of the large Cs+ ions also did not reduce the molar surface area. The subtle differences between the materials become visible in the CO2 sorption. Both Cs-N1 and Li-N1 show a higher CO2 uptake, as compared to N1. This 1358


Macromolecules

Article

method). Frequency calculations indicated that the shown structures are indeed ground states. From the graphical representation it is obvious that rather large deformations of the polymer network must occur upon switching pH, and we speculate that these deformations may result in some irreversible changes of the network morphology as observed experimentally. In order to see how much porosity is lost in each cycle, we switched the reversed network again to its Cs+ form (noted as Cs-N1 (second switch)) and measured the porosity. We found that again some porosity is lost while again the pore size distribution and character was kept upon switching (higher CO2 uptake compared to revCs-N1). To conclude the section on porosity analysis, we shortly discuss the energetics of the CO2 adsorption process. The isosteric heats of sorption qst (i.e., a measure of the interaction strength between the polymer and CO2 on a molecular level) were determined for N1, Li-N1, and Cs-N1 from CO2 adsorption isotherms captured at 273 and 283 K (Figure 2). All materials show a decrease of qst with increasing loading, which is commonly observed for microporous materials, as high energy adsorption sites are covered first. Interestingly, qst at 1 mmol g−1 loading followed the order Li-N1 (23 kJ mol−1) < N1 (26.8 kJ mol−1) < Cs-N1 (30 kJ mol−1), which implies that the interaction strength can be lowered or increased by switching of N1. This result is in agreement with reports on cation-modified zeolites, where higher interaction strength was also found for Cs+-modified materials compared to Li+ and partly related to a coupling to the framework.20,21 The modification is not only described by the nature of the introduced cations, but it also induces a change in the basicity of the anions (carboxylates, hydroxylates) located at the polymer backbone. This can be interpreted as a tunable micropore polarity, a concept which might be important not only for gas sorption but also for catalytic applications. Photocatalytic Properties. As mentioned above, quantum chemical calculations were undertaken on the central motif of the networks (see Figure 1). The salt form, as all phenolphthalein derivatives, adsorbs light in the visible range

Table 1. Porosity Parameters of the Materials sample N1 Li-N1 Na-N1 K-N1 Cs-N1 revLi-N1 revCs-N1 Cs-N1 (2nd)

SQSDFTa (m2 mmol−1)

SBETb (m2 mmol−1)

SCO2b,c (m2 mmol−1)

Vmicrob,d (cm3 mmol−1)

480 551 483 401 467 388 344 393

457 545 451 416 458 371 321 378

309 416 398 n.d. 429 256 240 351

0.174 0.210 0.176 0.164 0.180 0.142 0.122 0.178

a

Cumulative surface area of pores smaller 25 nm obtained from the QSDFT (quenched solid density functional theory) method. bUsed pressure range: 0.03 < p/p0 < 0.1, normalized by molecular weight of the repeating unit. cGCMC (grand canonical Monte Carlo) model. d Determined from N2 uptake at p/p0 = 0.1; n.d. = not determined.

indicates that the switching mainly affects the very local structure and the related microporosity. This is also seen in the PSD, where the fraction of larger pores is diminished in the case of the salts (see Figure 2 and Figure S3). After analysis of the salt form, the samples Li-N1 and Cs-N1 were treated with aqueous HCl in order to induce the switch back to N1 (named reversed Li-N1 and reversed Cs-N1). Porosity analysis indicated that the switch back is not fully reversible as the achieved surface areas were approximately 20− 25% lower than that of the original network. It is however interesting to note that the pore character was successfully switched back, as judged from the pore size distribution and the difference between the surface areas obtained from CO2 and N2 sorption. The fraction of large pores which were present in the parent network did reappear (Figure 2). This indicates that the structural changes induced by opening of the lacton ring and by the change of the hybridization state of the central C atom are indeed responsible for the observed differences in pore size.19 We modeled the central motif of the polymer network (Figure 1) in the neutral and the salt form. Geometry optimization was performed using Gaussian 03 software (B3LYP/6-31G

Scheme 2. Use of Phenolphthalein Networks for Photopolymerization: Assumed Photoinitiation Mechanisma

a

Inset: photographs of the mixture before and after irradiation with λ > 420 nm. 1359


Macromolecules

Article

polarizability of the wall structure. This effect could also be of interest for the generation of size selective gas separation materials,25 the switching of chemical adsorption, and many more. In this contribution, we exemplified a photoelectrochemical function of the conjugated salt form as an efficient heterogeneous photosensitizer for photopolymerization reactions. Future studies will target a better understanding of the switch process (including the forces acting on the network) as well as a more detailed study of the heterogeneous photopolymerization process.

(see Scheme 1 and Figure S4). Being a high surface area solid at the same time, it can be potentially used as a photosensitizer. Such molecules absorb the energy of light, use it for an electron/hole separation along the structure, and transfer one of the charges to acceptor molecules, which can use it for a variety of reactions, such as initiating a polymerization. The initiation of such a photopolymerization is called type II photoinitiated polymerization.22Common type II initiators include aromatic ketones, xanthenes, coumarins, and many others. In our case, Na-N1 was exemplarily investigated for its use as photoinitiator. It can act as type II photoinitiator due to its conjugated structure with a triplet excited state which reacts with hydrogen donor triethylamine (TEA) to produce first an radical cation from TEA, which can then react via proton abstraction and electron transfer to the initiating TEA-CH radical (see Scheme 2). Deactivating processes can also occur, such as monomer and oxygen quenching, etc., which compete with the formation of radicals.23 Hence, all our experiments were performed under argon gas to prevent quenching. The reactions were performed in bulk methyl methacrylate (MMA) in the presence of TEA under irradiation with a daylight lamp equipped with a 420 nm filter. The reaction resulted in the formation of a bulk solid poly(methyl methacrylate) (PMMA) with typical conversions of ∼45−80%. The PMMA could be dissolved and analyzed by size exclusion chromatography (SEC). The obtained polymer had a number-average molecular weight (Mn) of Mn = 138 000 g mol−1 and a polydispersity of 4.8. As the photosensitizer is of heterogeneous nature, it can be separated very easily from the reaction product by centrifugation and filtration. That way, our heterogeneous photosensitizer prevented the coloring of the polymer, which is a common problem in dye-sensitized photopolymerizations. Cycling of the photosensitizer Na-N1 was also performed; for the second use of the Na-N1 the conversion of the photopolymerization dropped down to 25%. The resulting PMMA had a comparable molecular weight (Mn: 146 500 g mol−1). The lower conversion might be due to the consumption of the ketonic species in the first polymerization. At the moment we do not know the fate of the carbon radical left at the polymeric framework. For benzophenones in solution, they undergo recombination, which is however difficult for the cross-linked polymer. Given the lower conversion in the second run, it seems plausible that the radicals saturate during the catalysis or the work-up. This question will however be dealt with in a more detailed study on the photopolymerization mechanism. So far, the lowered conversion is still comparable with most reported conversions in homogeneous visible light free radical photopolymerizations.24

■

ASSOCIATED CONTENT

S Supporting Information *

Gas sorption data and analysis; details of the molecular modeling; experimental details of the photopolymerization process. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected] (J.W.), [email protected] (B.K.); Tel +49-331-5679569, Fax +49-331-5679502. Present Address †

Department of Chemistry, Istanbul Technical University, 34469, Maslak, Istanbul, Turkey.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Dr. Simone Mascotto (Materials Chemistry Group, Trento, Italy) is highly acknowledged for 13C solid-state NMR measurements. We furthermore thank Prof. Y. Yagci and Prof. A. I. Cooper for helpful discussions. The Max-Planck Society is acknowledged for financial support.

■

REFERENCES

(1) Thomas, A.; Kuhn, P.; Weber, J.; Titirici, M.-M.; Antonietti, M. Macromol. Rapid Commun. 2009, 30, 221−236. (2) Cooper, A. I. Adv. Mater. 2009, 21, 1291−1295. (3) McKeown, N. B.; Budd, P. M. Macromolecules 2010, 43, 5163− 5176. (4) Thomas, A. Angew. Chem., Int. Ed. 2010, 49, 8328−8344. (5) Weber, J.; Thomas, A. J. Am. Chem. Soc. 2008, 130, 6334−6335. (6) Chen, L.; Honsho, Y.; Seki, S.; Jiang, D. J. Am. Chem. Soc. 2010, 132, 6742−6748. (7) Ritter, N.; Senkovska, I.; Kaskel, S.; Weber, J. Macromolecules 2011, 44, 2025−2033. (8) Weber, J.; Antonietti, M.; Thomas, A. Macromolecules 2008, 41, 2880−2885. (9) Farha, O. K.; Spokoyny, A. M.; Hauser, B. G.; Bae, Y.-S.; Brown, S. E.; Snurr, R. Q.; Mirkin, C. A.; Hupp, J. T. Chem. Mater. 2009, 21, 3033−3035. (10) Sydlik, S. A.; Chen, Z.; Swager, T. M. Macromolecules 2011, 44, 976−980. (11) Jones, J. T. A.; Holden, D.; Mitra, T.; Hasell, T.; Adams, D. J.; Jelfs, K. E.; Trewin, A.; Willock, D. J.; Day, G. M.; Bacsa, J.; Steiner, A.; Cooper, A. I. Angew. Chem. 2011, 123, 775−779. (12) Rabone, J.; Yue, Y.-F.; Chong, S. Y.; Stylianou, K. C.; Bacsa, J.; Bradshaw, D.; Darling, G. R.; Berry, N. G.; Khimyak, Y. Z.; Ganin, A. Y.; Wiper, P.; Claridge, J. B.; Rosseinsky, M. J. Science 2010, 329, 1053−1057. (13) Chang, I.; Fujara, F.; Geil, B.; Heuberger, G.; Mangel, T.; Sillescu, H. J. Non-Cryst. Solids 1994, 172−174, 248−255.

■

CONCLUSIONS To sum up, the synthesis and characterization of a switchable, conjugated microporous network based on the phenolphthalein motif were presented. It was shown that the pH sensitivity of phenolphthalein allowed the introduction of functionality into microporous polymers, i.e., a chemical benefit above mere surface area and porosity. The networks change their porosity with regard to pH due to the well-known structural transformation of phenolphthalein. The switch has an impact on the surface area and pore size distribution. Smaller pore sizes are generally observed for the salt forms. The size of the introduced counterions is geometry-wise only of minor importance but contributes to the overall polarity and 1360


Macromolecules

Article

(14) Pixton, M. R.; Paul, D. R. J. Polym. Sci., Part B: Polym. Phys. 1995, 33, 1353−1364. (15) Frisch, M. J.et al. Gaussian 03, Revision E.01; Gaussian, Inc.: Wallingford, CT, 2004. (16) Dawson, R.; Laybourn, A.; Khimyak, Y. Z.; Adams, D. J.; Cooper, A. I. Macromolecules 2010, 43, 8524−8530. (17) Kiskan, B.; Weber, J. ACS Macro Lett. 2012, 1, 37−40. (18) Weber, J.; Schmidt, J.; Thomas, A.; Böhlmann, W. Langmuir 2010, 26, 15650−15656. (19) Whether the simultaneous formation of phenates also has an impact on the porosity based on intermolecular interactions cannot be discussed within this work as the two effects cannot be decoupled. (20) Pirngruber, G. D.; Raybaud, P.; Belmabkhout, Y.; Č ejka, J.; Zukal, A. Phys. Chem. Chem. Phys. 2010, 12, 13534. (21) Garrone, E.; Bonelli, B.; Lamberti, C.; Civalleri, B.; Rocchia, M.; Roy, P.; Otero Areán, C. J. Chem. Phys. 2002, 117, 10274. (22) Green, W. A. Industrial Photoiniators; CRC Press, Taylor and Francis Group: Boca Raton, FL, 2010. (23) Yagci, Y.; Jockusch, S.; Turro, N. J. Macromolecules 2010, 43, 6245−6260. (24) Tunc, D.; Yagci, Y. Polym. Chem. 2011, 2, 2557−2563. (25) Liu, Q.; Mace, A.; Bacsik, Z.; Sun, J.; Laaksonen, A.; Hedin, N. Chem. Commun. 2010, 46, 4502−4504.

1361
