2.2.1 Synthesis and Structural Characterizations
Figure 2-1. Schematic representation of the synthesis of imine-linked pyrene-anthracence COF (Py-An COF) with 1,3,6,8- tetrakis(p-formylphenyl) pyrene (TFPPy) at the vertices and 2,6-diaminoanthracene (DAAn) on the edges of the tetragonal framework.
Py-An COF was synthesized by using the imine formation reaction of 1,3,6,8- tetrakis(p-formylphenyl) pyrene (TFPPy) and 2,6-diaminoanthracene (DAAn) in o-dichlorobenzene/ n-butanol under solvothermal conditions (Figure 2-1). 1,3,6,8- Tetrakis(p-formylphenyl) pyrene (TFPPy) and 2,6-diaminoanthracene (DAAn) have low solubility in common organic solvents. It is difficult to achieve highly crystalline and porous COF using monomers with low solubility. So the author screened and optimized the solvothermal conditions, including the solvent, reaction temperature and time, and the catalyst concentration (Table 2-1). Single-component solvents, such as dioxane, dimethylacetamide (DMAc), m-cresol, o-dichlorobenzene (o-DCB), and mesitylene, resulted in amorphous and nonporous materials. The author thus
investigated the synthesis of crystalline and porous COFs via two-component solvent systems; a mixture of o-DCB and n-butanol produced crystalline and porous COFs.
By optimizing the catalyst concentration and reaction time, the author developed solvothermal conditions using an o-DCB/n-butanol solvent mixture (1/1 v/v) in the presence of a 1 M acetic acid catalyst at 120 ˚C for 5 days to synthesize COFs.
Table 2-1. Reaction conditions of preparing Py-An COF.(TFPPy 15 mg, DAAn 10mg)
No. Solvent
Catalyst:
Acetic Acid
Temp. (˚C)
Time (day)
Yield (%)
XRD Intensity (Counts)
1
Dioxane 0.5 mL
(6 M), 0.1 mL 120 5 50 No Peak
2
DMAc 0.5 mL
(6 M), 0.1 mL 120 5 40 No Peak
3
m-Cresol 0.5 mL
(6 M), 0.1 mL 120 5 76 No Peak
4
o-Dichlorobenzene 0.5 mL
(6 M), 0.1 mL 120 5 60 No Peak
5
Mesitylene 0.5 mL
(6 M), 0.1 mL 120 5 69 No Peak
6
o-DCB/Dioxane 0.25 mL/0.25 mL
(6 M), 0.1 mL 120 3 75 9295
7
o-DCB/n-BuOH 0.25 mL/0.25 mL
(6 M), 0.1 mL 120 5 88 16811
8
Mesitylene/ Dioxane 0.25 mL/0.5 mL
(6 M), 0.1 mL 120 5 66 No Peak
9
DMAc/Mesitylene 0.05 mL/0.45 mL
(6 M), 0.1 mL 120 5 82 No Peak
10
DMSO/Mesitylene 0.05 mL/0.45 mL
(6 M), 0.1 mL 120 5 80 No Peak
11
o-DCB/Dioxane 0.3 mL/0.2 mL
(6 M), 0.1 mL 120 5 80 8241
12
o-DCB/Dioxane 0.2 mL/0.3 mL
(6 M), 0.1 mL 120 5 76 1499
13
o-DCB/n-BuOH 0.3 mL/0.2 mL
(6 M), 0.1 mL 120 5 72 10280
14
o-DCB/n-BuOH 0.2 mL/0.3 mL
(6 M), 0.1 mL 120 5 82 12809
15
o-DCB/n-BuOH 0.25 mL/0.25 mL
(6 M), 0.1 mL 120 3 73 8241
16
o-DCB/n-BuOH 0.25 mL/0.25 mL
(6 M), 0.1 mL 120 7 82 12110
17
o-DCB/n-BuOH 0.4 mL/0.4 mL
(6 M), 0.1 mL 120 5 86 11904
18
o-DCB/n-BuOH 0.25 mL/0.25 mL
(3M), 0.1 mL 120 5 80 6373
The typical synthesis methods for the Py-An COF: A 10-mL pyrex tube was charged with TFPPy (15.0 mg, 0.024 mmol), DAAn (10 mg, 0.048 mmol), o-DCB/n-BuOH (0.5 mL, 1/1 by vol.) and 0.1 mL AcOH (6 M) and the mixture was sonicated for 2 minutes, degassed by three freeze-pump-thaw cycles. The tube was sealed under vacuum by flame and heated at 120 ˚C for 5 days. The reaction mixture was cooled to room temperature and the fluffy orange precipitate was collected by centrifugation. The precipitate was washed with DMF (2 mL × 3) and THF (2 mL × 5), and then dried under vacuum at 120 ˚C overnight. The isolated yield is 88%.
Figure 2-2. FT-IR spectra of TFPPy (black), DAAn (blue), and Py-An COF (red).
The formation of imine linkages was confirmed by the Fourier transform infrared (FT-IR) spectroscopy, which exhibited a stretching vibration band 1627.6 cm-1 assignable to the C=N bond (Figure 2-2). Elemental analysis reveals that the content of elements is close to the theoretical value (Table 2-2).
Table 2-2. Elemental analysis of Py-An COF
C (%) H (%) N (%)
Py-An COF (C72H42N4)n
Calcd. 89.69 4.40 5.82
Found 81.63 4.76 4.35
2.2.2 Morphologies
Figure 2-3. FE SEM images of the Py-An COF.
Figure 2-4. HR TEM images of the Py-An COF.
Field-emission scanning electron microscopy (FE-SEM) and High-resolution transmission electron microscopy (HR-TEM) were used to investigate the morphology of the Py-An COF. The FE- SEM revealed regular belt morphology (Figure 2-3). HR-TEM revealed tetragonal porous textures (Figure 2-4), which were close to the lattice resolved by X-ray diffraction (XRD).
2.2.3 PXRD Pattern and Theoretical Calculations
Figure 2-5. XRD pattern of Py-An COF
The XRD patterns revealed that the Py-An COF was a crystalline material (Figure 2-6A, red curve). The most intense peak at 2θ = 3.14˚ corresponds to the (110) facet of a square lattice. The other minor diffraction peaks at 4.67, 6.38, 9.66, 12.89 and 25.1˚ are assigned to the (200), (220), (330), (440) and (001) facets, respectively. To clarify the lattice packing, the author constructed and evaluated two typical modes of crystal lattices, i.e., the eclipsed AA and staggered AB modes, using Materials Studio software package. The simulation of the XRD pattern of the AA stacking mode (Figure 2-6A, green curve) matched the experimental data both in peak position and intensity, while the staggered AB-stacking mode produced a profile (Figure 2-6A, orange curve) that was inconsistent with the experimental pattern. The author reconstructed the eclipsed AA mode by using an optimal monoclinic space group (C2/m), which gave rise to a more plausible layer morphology and higher correlation with the experimental XRD pattern (Figure 2-6 B, blue curve). The Pawley refinements were subsequently performed within the Materials Studio Reflex
Chapter 2
44
of a = 38.4173 Å, b = 44.4373 Å, c = 3.9932 Å, α = 66.75˚, and β = γ = 90˚. The refined profile matched the experimental XRD pattern very well (Figure 2-6B, blue curve) as evidenced by their negligible difference (black curve). The AA stacking mode constituted 1D channels with pore size of 2.4 nm (Figure 2-6C), whereas the AB mode resulted in overlapped pores (Figure 2-6D).
Figure 2-6. (A) XRA profiles of Py-An-COF experimentally observed (red), simulated by using AA-stacking (green) and AB-stacking (orange) modes. (B) XRA profiles of Py-An-COF experimentally observed (red), Pawley refinement (blue), and their difference (black). Crystal structures of (C) the AA-stacking and (D) the AB-stacking modes.
COMMUNICATION Journal Name
2 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx
clarify the lattice packing, we constructed and evaluated
two typical modes of crystal lattices, i.e., the eclipsed AA and staggered AB modes, using Materials Studio software package. The simulation of the XRD pattern of the AA stacking mode (green curve) matched the experimental data both in peak position and intensity, while the staggered AB-stacking mode produced a profile (orange curve) that was inconsistent with the experimental pattern.
We reconstructed the eclipsed AA mode by using an optimal monoclinic space group (C2/m), which gave rise to a more plausible layer morphology and higher correlation with the experimental XRD pattern (Fig. 1B, blue curve). The Pawley refinements were subsequently performed within the Materials Studio Reflex Plus Module to determine the unit cell parameters, producing the unit- cell parameters of a = 38.4173 Å, b = 44.4373 Å, c = 3.9932 Å, α = 66.75°, and β = γ = 90°. The refined profile matched the experimental XRD pattern very well (Fig. 1B, blue curve) as evidenced by their negligible difference (black curve). The AA stacking mode constituted 1D channels with pore size of 2.4 nm (Fig. 1C), whereas the AB mode resulted in overlapped pores (Fig. 1D).
The Py-An COF exhibited typical type-IV nitrogen sorption isotherms collected at 77 K (Fig. 2A), which indicated that the Py-An COF was a mesoporous material.
The Brunauer–Emmett–Teller (BET) surface area and pore volume were estimated as high as 1479 m
2g
−1and 0.7 cm
3g
−1, respectively. The pore-size distribution profile (Fig. 2B) revealed that the Py-An COF contained only one type of mesopore with size of 2.4 nm.
Fig. 1 (A) XRA profiles of Py-An-COF experimentally observed (red), Pawley refinement (blue), and their difference (black). (B) XRA profiles of Py-An-COF experimentally observed (red), simulated by using AA- stacking (green) and AB-stacking (orange) modes. Crystal structures of (C) the AA-stacking and (D) the AB-stacking modes.
Fig. 2 (A) Nitrogen sorption isotherm curves measured at 77 K (red circle:
desorption, blue circle: adsorption). (B) Pores size (red circle) and pore size distribution (blue circle) profiles.
The stability was investigated by dispersing the COF samples in different solvents, such as hexane, THF, acetone, methanol, ethanol, and water at 25 °C for 24 h (Fig. S5). XRD patterns of the samples exhibit similar XRD patterns, which indicated that the Py-An COF retained crystallinity in these solvents.
The ordered anthracene columnar π-walls and 1D open channels together with solvent stability make the Py-An COF an interesting material for heterogeneous catalysis. We chose the Diels-Alder reaction of 9- hydroxymethylanthracene (0.05 mmol) and N-substituted maleimide derivatives (0.05 mmol) in the presence of Py- An COF (10 mg) catalyst and observed that the Diels- Alder reaction proceeded smoothly and cleanly at room temperature and 1 bar. The Diels-Alder adduct formed quantitatively after 6-h reaction (Table 1, entry 1, > 99%
yield based on
1H NMR analysis, Fig. S6). By contrast, control experiments without Py-An COF under otherwise identical conditions resulted in only 24% yield. This result indicates that the Py-An COF facilitate the Diels-Alder reaction under ambient conditions. To investigate the reactant scope, a variety of maleimide derivatives with different N-substituents, including N-benzyl, N-(p- boromo)phenyl, N-(p-nitro)phenyl, N-ethyl, N-cyclohexyl (entries 2-6) were used for the reaction with 9- hydroxymethylanthracene. In all cases, the Py-An COF catalysts significantly enhanced the yields, compared to the controls without the COF catalyst. To the best of our knowledge, the Py-An COF exhibited the highest catalytic activities among the heterogeneous catalysts reported to date that work at elevated temperatures.
5The using of neat water as an environmentally benign solvent has received considerable attention with respect to green chemistry. We observed that the Py-An COF enabled the use of water as solvent for the Diels-Alder reactions. Dramatically, the Py-An COF enhanced the reaction yield by 6 fold (Table 1, entry 7, increased from 13% to 91%) compared to the control experiment.
A long catalyst lifetime and the capability of cycle use are highly desired for applications. The Py-An COF catalyst was easily separated from the reaction mixture and recovered; centrifuge and subsequent rinsing with solvents and water refreshed the catalyst for the next-
ChemComm A cc ep te d Ma n u sc ri p t
Published on 12 May 2015. Downloaded by OKAZAKI NATIONAL RESEARCH INST on 17/05/2015 04:00:16.
View Article Online DOI: 10.1039/C5CC03457D
2.2.4 Porosity
Figure 2-7. Nitrogen sorption isotherm curves measured at 77 K (red circle:
adsorption, blue circle: desorption)
Figure 2-8. Pores size (red circle) and pore size distribution (blue circle) profiles.
The Py-An COF exhibited typical type-IV nitrogen sorption isotherms collected at 77 K (Figure 2-7), which indicated that the Py-An COF was a mesoporous material. The Brunauer-Emmett-Teller (BET) surface area and pore volume were estimated as high as 1479 m2g−1 and 0.7 cm3g−1, respectively. The pore-size
distribution profile (Figure 2-8) revealed that the Py-An COF contained only one type of mesopore with size of 2.4 nm.
To investigate the porosity further, adsorption experiments were carried out at room temperature by adding Py-An COF to anthracene solution in ethanol (Figure 2-9). The author found that the absorption intensity of the solution decreased gradually, which indicates that the Py-An COF could encapsulate anthracene into the pores. From the structure simulation, the author found that the anthracene building blocks stack in face-to-face mode to form the COF skeleton, which makes the pore wall covered with well-ordered protons. This special structure may encapsulate and stack aromatic molecules to the pore wall by the C-H···π interaction.55
Figure 2-9. Time-dependent electronic absorption spectral change of ethanol solution of anthracene (0.1 mM) upon addition of the Py-An COF (10 mg).
2.2.5 Chemical Stability
Figure 2-10. TGA curve of Py-An COF
Figure 2-11. XRD patterns of the Py-An COF upon 1-day treatment in different solvents.
Thermal gravimetric analysis was measured to investigate the thermal stability of Py-An COF. Thermal gravimetric analysis shows that Py-An COF is stable up to 400
˚C (Figure 2-10).
The chemical stability of the Py-An COF was investigated by dispersing the COF samples in different solvents such as hexane, THF, acetone, methanol, ethanol and water at 25 ˚C for 24 h (Figure 2-11). The author found that no decomposition occurs to the COF samples in these conditions, there is nearly no weight loss (< 0.1 wt%).
PXRD was measured after dried under vacuum at 120 ˚C for 12 h. All the samples remained to exhibit intense XRD patterns, which indicates that the Py-An COF kept its high crystallinity in these conditions. The Py-An COF shows excellent stability in these solvents.