Tunable Surface Area, Porosity, and Function in Conjugated Microporous Polymers

Abstract Simple inorganic salts are used to tune N‐containing conjugated microporous polymers (CMPs) synthesized by Buchwald–Hartwig (BH) cross‐coupling reactions. Poly(triphenylamine), PTPA, initially shows a broad distribution of micropores, mesopores, and macropores. However, the addition of inorganic salts affects all porous network properties significantly: the pore size distribution is narrowed to the microporous range only, mimicking COFs and MOFs; the BET surface area is radically improved from 58 m2 g−1 to 1152 m2 g−1; and variations of the anion and cation sizes are used to fine‐tune the surface area of PTPA, with the surface area showing a gradual decrease with an increase in the ionic radius of salts. The effect of the salt on the physical properties of the polymer is attributed to adjusting and optimizing the Hansen solubility parameters (HSPs) of solvents for the growing polymer, and named the Beijing–Xi'an Jiaotong (BXJ) method.

After 48 hours, the mixture was cooled to room temperature. The products were then washed with 200 mL CHCl3, ethanol, methanol and boiling water, to remove the catalyst, impurities and oligomers, followed by a 72 h Soxhlet extraction with methanol for 24 h, THF for 24 h and chloroform for 24 h, respectively.

S1.5 Synthesis of polytriphenylamine (PTPA) CMPs in different solvents
A Schlenk tube was charged with a tris(4-bromophenyl) amine (0.5 mmol), phenylenediamine (0.33 mmol to obtain the 1 the samples synthesized in different solvents without salts were obtained through the same way as above but without adding 0.5 mmol of Na2SO4.

S1.6 Characterization
A PerkinElmer Spectrum 100 spectrometer was used to obtain the attenuated total reflection

S2.1 Introduction of Hansen solubility parameters [2]
The solubility parameter approach proposed by Hansen for predicting polymer solubility and the compatibility of solvent for polymer synthesis has been widely used. There are three main types of interactions in common organic materials. The first is the nonpolar interactions (ED).
It is derived from atomic forces and has also been called dispersion interactions. Therefore, it is related to the molar volumes of the solvent. This parameter could not be tuned by salts. The second type of cohesion energy, namely the polar cohesive energy (EP), is caused by permanent V is the molar volume. The third major cohesive energy source is hydrogen bonding, EH, which is related to the polar and dispersion energies of vaporization from the total energy of vaporization. In this respect, hydrogen bonding resembles the polar interactions. Therefore, EP and EH could be tuned and should increase with the electronegativity of the salts. Generally speaking, these two solubility parameters could be tuned by various salts and solvents due to the change of dipole-permanent dipole interactions and hydrogen bonding.
The equation governing the Hansen parameters is that the total cohesion energy, E, which should be the sum of the individual energies.

E = ED+ EP+ EH
the total solubility parameter could be acquired by dividing the molar volume as follows, To evaluate the compatibility of Hansen solubility parameters between solvent and polymer, the difference between δT two materials of |δT| (|δT|=|δT1-δT2|) was applied.
If |δT|<1, then the solvent could be a good solvent; if 1<|δT|<3, then the solvent could be an intermediary solvent; otherwise, the solvent is a poor solvent and not suitable for synthesis. acetonitrile, water and 1,4-dioxane) by a high-pressure jet mill followed by ultra-sonication.

S3.1 Physical properties of the PTPAs before and after salt adding
PTPA networks were synthesized through BH coupling reactions, starting from tris(4bromophenyl)amine and phenylenediamine (Scheme 1). A broad range of salts of different anionic or cationic radii, namely NaF, NaCl, NaBr, NaI or LiNO3, NaNO3, KNO3, Ba(NO3)2 (at 0.5 mM concentration) were then added to the reaction mixtures and explored for their ability to tune the physical properties of the PTPA networks (see Table S1 for radii). The polymer is initially a light brown colour when under an inert N2 atmosphere during the polymerization reaction, but gradually turns dark blue with increased exposure to air, in a similar manner to pure poly(aniline). [4] The obtained polymers are insoluble in common organic solvents including toluene, THF, dioxane, dichloromethane, chloroform, ethanol and methanol, exhibiting their robustness and high degree of cross-linking. [1] The products also exhibit significantly higher thermal stabilities after addition of salt (Tdec>250º C before the addition of salt vs. Tdec>500º C after addition of salt, Please see Figure S2). Interestingly, some of the networks we resynthesized, like CMP-1, [5] show macroscopic gelation of the solution after salt addition during the polymerization reaction in the THF. The gels are retained upon washing with chloroform, but fragment into powders upon washing with methanol. The powders will swell when treated with solvents such as chloroform ( Figure S3 for full details). This phenomenon does not occur when the PTPA is polymerized without the addition of salts; instead, a powder-like precipitate is obtained, which does not swell upon washing with any solvents. In addition, product yields obtained after the BXJ process reached 99% (after a 72 h Soxhlet extraction with methanol for 24 h, THF for 24 h and chloroform for 24 h, respectively).
The maximum yields obtained for polymers prepared without salt tuning, however, is only 52%.

S3.2 Verification of higher polymerization degree of the PTPAs using XPS
The typical polyaniline XPS spectra shape were observed in the spectra of the PTPA networks, suggesting the polyaniline structure in the PTPA networks. In addition, the ratio of the imine increase with the surface area (please see the ratio of imine to amine in the N1s spectra and the ratio of peak area at around 286.4 eV which is assigned to the imine carbon), indicating higher polymerization degree of the polymers. [6]

S3.3 Discussion on HSPs of the PTPA in various conditions
In general, the polymerization process of PVB resins involves simultaneous vinyl polymerization, cross-linking, phase separation, microgel fusion, aggregation and pore infilling, which results in a broad PSD ranging from micropores to macropores (as found for our CMPs). [7] Many studies have used organic solvents as porogens to tune the phase separation of PVB resins during polymerization, achieving control over the pore diameter and corresponding internal surface area. [7][8] As a general guide, solvents with poor thermodynamic compatibility and low matching of their Hansen solubility parameters (HSPs) with the resultant polymer networks could result in the formation of microgels and early phase separation, leading to large average diameter pores and low BET surface areas. [7,9] The Hansen solubility parameters of PTPA were experimentally estimated in Figure S15 using the method provided in Section S2.2 and literature. [3] From Table S6 and Table 2, it can be seen that the difference of the Hansen solubility parameter (|T|) between the solvents applied in this study and PTPA was larger than 1, suggesting that the solvents were not good enough for PTPA synthesis. Therefore, low surface area and broad PSD of PTPA synthesized in these solvents could be observed before salt addition ( Figure S7 and Table S3). However, the surface area of the PTPA increases with the decrease of the difference of the Hansen solubility parameter (|T|) between the solvents applied in this study ( Table 2 and Table S3, SBET, (Dioxane-PTPA)> SBET, (THF-PTPA)> SBET, (Toluene-PTPA); |δT|(Dioxane-PTPA)< |δT|(THF-PTPA)< |δT|(Toluene-PTPA))) can be still observed, indicating that mechanism of Hansen solubility parameter proposed in this study is well applicable. After the salt adding, the permanent dipole interactions (δP) and the hydrogenbonding interactions (δH) of the solvent were increased by salts due to their influence on the ion strength of the solvent, leading to the decrease of the |δT| between polymer and solvent to be less than 1 and leading the solvent to be a good solvent for PTPA. As such, solvents with good thermodynamic compatibility and higher matching of their Hansen solubility parameters (HSPs) with the resultant polymer networks could result in the late phase separation, leading to uniform micropores and high BET surface areas. [7,9] 9

S3.4 Comparison of PTPA tuned by salts with CMP-1
CMP-1, previously synthesized by Jiang and Cooper [5,10] using a Sonogashira-Hagihara coupling, has a similar chemical structure to the polymers prepared in this investigation, and displayed similar properties such as high surface area (834 m 2 /g), well defined PSD (micropores only) and an amorphous nature. However, the authors ascribed these results to the novel rigid structure of poly(aryleneethynylene) (PAE), and did not consider the effect of the addition of salt (CuI in this case) to the starting materials on the porosity of the PAE polymer. It could be assumed that the PAE networks synthesized by Jiang could have low surface areas and broad PSDs, similar to PTPA, if CuI or other salts were not used.

S4.1.1 FTIR and SS 13 C CP/MAS NMR of the PTPA tuned by salts
Figure S1 FTIR spectra of the PTPA networks produced applying the BXJ approach, with different anions (a) and cations (b); SS 13 C CP/MAS NMR spectra of the PTPA networks tuned by salts with different anions (c) and cations (d) (asterisks mark spinning side bands).

Figure S2
TGA curve (a,b) and DTG curve (c,d) of the PTPA before and after salt tuning.

Figure S3
Photos of 200 mg of pure PTPA (left) and NaF-tuned PTPA (right) after adsorption and removal of chloroform using filtration. Figure S4 FTIR spectra of core (tris (4-bromophenyl)amine) and linker (phenylenediamine) (asterisks mark the bands assigned to C-Br bond).

Figure S7
N2 adsorption and desorption isotherms, nonlocal density functional theory-pore size distribution and cumulative pore volume of the PTPA networks tuned by salts with different ion dosage (a-c), solvents with 0.5 mM Na2SO4 (d-f) and solvents without salts (g-i) (the pink rectangular strips indicate the microporous region in the pore size distribution).

Figure S8
FTIR spectra of salts (the pink stripes indicate the characteristic peak ascribed to the salts which was also found in FTIR spectra of PTPA after salt tuning).

Control
NaF NaCl NaBr NaI Figure S10 SEM images of the PTPA before and after salt tuning (Scale bar 10 μm).

Figure S11
Correlation between UV-vis absorption peak and surface area of PTPA after salt tuning of different anions (a) and cations (b).

5) XPS
Figure S12 XPS spectra of PTPA before and after the tuning by BXJ-salt method.