Binding ability of first and second generation/carbazolylphenyl dendrimers with Zn(II) tetraphenylporphyrin core towards small heterocyclic substrates

Abstract

A study of complex formation of Zn(II) tetraarylporphyrin dendrimers with carbazolylphenyl branches towards 1,4-diazabicyclo-[2.2.2]octane, pyridine, imidazole, N-methylimidazole and 1,2,3-triazole was carried out by spectrophotometric and ¹H NMR titration methods. It has been shown that the binding ability of the porphyrin receptors towards mono and bidentate N-containing substrates depends on the nature, number and generation of the branches. Bulky substituents are able either to significantly reduce the binding ability of the tetrapyrrolic cores due to the shielding of the porphyrin reaction centres, or to significantly increase it by forming intramolecular cavities for complementary binding of substrates. It has been determined that due to a good geometric match of the ligand's size with the size of the intramolecular cavities of the porphyrin receptors, and by the existence of additional hydrogen bonding and/or π–π interactions between the ligand and the triazole fragments of the porphyrin the Zn-tetraarylporphyrins with eight 4-carbazolylphenyl-1,2,3-triazole end groups of the first and the second generations could be used as effective receptors for imidazole, N-methylimidazole and 1,2,3-triazole. Taking into account the fact that binding is accompanied by a clear and easily identifiable response in the UV-Vis spectra of the reaction mixture, these metalloporphyrins could be considered as molecular optical sensing devices for small heterocyclic substrates.

---

Introduction

Dendrimers are monodisperse macromolecules with highly branched three-dimensional structures. Given the fact that the size of dendrimeric macromolecules can be predicted and controlled with a high accuracy they are often called a new generation of polymers and have a great future as polyfunctional materials. The presence of channels and pores allows them to encapsulate and/or activate small guest molecules, including physiologically active ones.

According with the literature, porphyrin-based dendrimers are of great interest.^1–8 It was found that Fe(II) porphyrins containing polyethylenglycol branches have a much higher (1500 times) constant of reversible binding of O₂ compared with human hemoglobin in which the iron porphyrin (heme) is surrounded by a globular protein (globin).^9,10 In both cases, the fixation of oxygen occurs as a result of its coordination at the iron atom. It is assumed that a causal factor responsible for the affinity of O₂ to dendrimer porphyrins is the formation of hydrogen bonds between oxygen molecules and the amide groups of the branches' first generation. The design and properties of “patched dendrimers” has been described,¹¹ in which different types of oligopeptide dendrons are asymmetrically introduced on the Zn(II) porphyrin core. The “patch” gives the porphyrin dendrimer an additional interface to bind with another molecule or macromolecule. “patched dendrimers” with porphyrin cores show molecular recognition phenomena at the nanoscale, which provides good insight into the biological molecular recognition performed by proteins and enzymes.

Porphyrin-based dendrimers are often using as photofunctional artificial receptors, in which the strong photoabsorption and intense fluorescence signals of the porphyrin can respond sensitively to substrate binding.^12–17

This paper investigates the binding ability of Zn-tetraarylporphyrins with different number [two (ZnD1-G1, ZnD4-G1, ZnD7-G2), four (ZnD2-G1, ZnD5-G1, ZnD8-G2) and eight (ZnD3-G1, ZnD6-G1, ZnD9-G2)] and generation [the first (ZnD1-G1, ZnD2-G1, ZnD3-G1, ZnD4-G1, ZnD5-G1, ZnD6-G1) and the second (ZnD7-G2, ZnD8-G2, ZnD9-G2)] of carbazolylphenyl branches towards 1,4-diazabicyclo-[2.2.2]octane (L1), pyridine (L2), imidazole (L3), N-methylimidazole (L4) and 1,2,3-triazole (L5) in toluene. The dendrimers also differ by the nature of bridging spacers [oxygen (ZnD1-G1, ZnD2-G1, ZnD3-G1) and 1,2,3-triazole (ZnD4-G1, ZnD5-G1, ZnD6-G1, ZnD7-G2, ZnD8-G2, ZnD9-G2)] connecting the tetraarylporphyrin core and carbazolylphenyl fragments. Zn(II) tetraphenylporphin (ZnTPP) was taken as the object of comparison. The compounds ZnD4-G1, ZnD5-G1, ZnD6-G1, ZnD7-G2, ZnD8-G2, ZnD9-G2 were previously synthesized¹⁸ as new fluorescent switches and photoactive devices for detection of substrates of different nature.

Result and discussion

Synthesis

The synthesis of dendrimers H₂D1-G1, H₂D2-G1 and H₂D3-G1 was based on Lindsey method starting from 5-mesityldipyrromethane¹⁹ or pyrrole and carbazole-based aldehydes.

The nucleophilic substitution reaction of 4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenol (1)²⁰ and 4-bromomethylbenzaldehyde (2)²¹ in DMF resulted in the formation of 4-[(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methyl]benzaldehyde (3) (Scheme 1). Similarly, the mixture of arylaldehydes consisting of 3,5-bis(bromomethyl)-2,4,6-trimethylbenzaldehyde (4) and 3-bromomethyl-5-chloromethyl-2,4,6-trimethylbenzaldehyde (5)¹⁸ was reacted with (1) and 3,5-bis[(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methyl]-2,4,6-trimethylbenzaldehyde (6) was obtained (Scheme 1) in pure form after column chromatography purification.

Scheme 1 Synthesis of carbazole-based aldehydes.

The condensation between arylaldehyde (3) and 5-mesityldipyrromethane¹⁹ was carried out in dry CH₂Cl₂ and the presence of a Lewis acid catalyst BF₃·OEt₂ at room temperature. Then p-chloranil was used as oxidant and the reaction mixture was refluxed for 1 hour. The starting materials' concentration was optimized at 10 mM in CH₂Cl₂, the yield of 5,15-bis(2,4,6-trimethylphenyl)-10,20-bis[4-(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methylphenyl] porphyrin (7) (Scheme 2) reached 34% with 0.3 equivalent of BF₃·OEt₂. Similarly, 5,10,15,20-tetrakis[4-(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)me-thylphenyl] porphyrin (8) was obtained in 15% when arylaldehyde (3) was reacted with pyrrole under the same conditions that were used to make dendrimer (7). The synthesis of 5,10,15,20-tetrakis[3,5-bis((4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methyl)-2,4,6-trimethylphenyl] porphyrin (9) was unsuccessful when using the procedure that applied for making dendrimer (7). In the presence of 0.75% absolute ethanol in dry CH₂Cl₂, the tetrasubstituted porphyrin (9) was obtained in 5%. The increase in the amount of Lewis acid catalyst from 0.3 to 0.8 equivalent as well as the condensation time between (6) and pyrrole did not lead to any change in the yield of dendrimer (9). The low yield of making dendrimer (9) was due to the sterically hindered methyl groups at 2 and 6 positions and bulky groups at 3 and 5 positions of compound (6). Dendrimers (7), (8) and (9) were then metallated in CHCl₃ to obtain ZnD1-G1, ZnD2-G1 and ZnD3-G1 in quantitative yield.

Scheme 2 Dendrimers H₂D1-G1 (7), H₂D2-G1 (8), H₂D3-G1 (9).

Dendrimers ZnD4-G1, ZnD5-G1, ZnD6-G1, ZnD7-G2, ZnD8-G2, ZnD9-G2 (Scheme 3) were synthesized via the copper(I)-catalyzed azide–alkyne cycloaddition (CuAAC reaction or click reaction) in THF solvent under [Cu(NCCH₃)₄][PF₆] catalysis.¹⁸

Scheme 3 Structures of dendrimers ZnD4-G1, ZnD5-G1, ZnD6-G1, ZnD7-G2, ZnD8-G2, ZnD9-G2.

Binding ability

The strength of axial binding of electron donating ligands (L) on Zn(II) porphyrins (ZnP) depends on the degree of aromaticity of the tetrapyrrolic macrocycle.^22,23 The aromaticity of the tetrapyrrolic macrocycle is higher, the more strongly a zinc cation is connected with the macrocycle nitrogen atoms. The reasons of decreasing of the tetrapyrrolic macrocycle aromaticity can be both the electronic influence of the substituents, and a spatial factor causing distortion of the planar structure of the tetrapyrrolic macrocycle, especially due to unsymmetrical substitution with bulky substituents.

Next to distortion of the planar structure of the tetrapyrrolic macrocycle, bulky substituents can also create steric hindrance to the ligands axial coordination due to shielding of the metalloporphyrin reaction center from both sides or a single side of the molecule. On the other hand, highly branched bulky substituents may form intramolecular binding cavities for effective binding of guest molecules.

Axial coordination of L1–L5 (Scheme 4) to ZnP is accompanied by a characteristic red shift of the absorption bands in the UV-Vis spectra of the system ZnP–L and a high field shift of the ligand protons signals in the ¹H NMR spectra of the corresponding complexes. It should be noted that upon complexation of ZnP with monodentate ligands L2–L5, over a wide concentration range of the ligands (C_L = 1 × 10⁻⁷ to 8 × 10⁻² M), changes in the UV-Vis spectra of the reaction mixture occur with the formation of one family of spectral curves with one set of isosbestic points. The titration curve has one step, which indicates the formation of a single type of complexes in a ratio of 1 : 1. The details of the spectrophotometric and ¹H NMR titration are described in the preliminary communication.²⁴ The changes in the UV-Vis spectra of the system ZnD3-G1–L3 and the corresponding binding isotherms are depicted on Fig. S1 as an example (ESI†).

Scheme 4 Structures of ligands L1–L5.

It was found that para-substitution of the tetrapyrrolic core phenyl groups by two (ZnD1-G1) or four (ZnD2-G1) 4-(4-(3,6-bis(t-butyl)carbazol-9-ylphenyl)-oxy) fragments and by two (ZnD7-G2) or four (ZnD8-G2) 4-(4-(3,6-bis(t-butyl)carbazol-9-ylphenyl)-1,2,3-triazole) branches of the second generation leads to an increase in the stability constants of the 1 : 1 complexes between the dendrimers (ZnD1-G1, ZnD2-G1, ZnD7-G2, ZnD8-G2) and monodentate ligands L2–L5 as compared with the similar complexes of ZnTPP (Fig. S2 and Table 1) (ESI†). This could be explained by distortion of the planar structure of the tetrapyrrolic macrocycle due to substitution with bulky groups.

The decreasing of the binding ability of the para-substituted porphyrins with two (ZnD4-G1) and four (ZnD5-G1) 4-(4-(3,6-bis(t-butyl)carbazol-9-ylphenyl)-1,2,3-triazole) branches of the first generation as compared with the corresponding complexes of ZnTPP with L2–L5²⁴ probably is the result of shielding of the metalloporphyrin central zinc cation by one of the carbazolylphenyl fragments. The optimized structures of the dendrimers ZnD1-G1, ZnD4-G1 and ZnD7-G2 are given as an example of the validation of provided assumption on Fig. S3 (ESI†). Tetra-substituted dendrimers ZnD2-G1, ZnD5-G1 and ZnD8-G2 are characterized by the same features.

It should be noted that meta-octasubstitution of the tetrapyrrolic core phenyl groups by eight 4-(4-(3,6-bis(t-butyl)carbazol-9-ylphenyl)-1,2,3-triazole branches of the first (ZnD6-G1) and the second (ZnD9-G2) generations leads to an increase in the stability constants of the 1 : 1 complexes between the dendrimers and the monodentate ligands L2–L5 as compared with the similar complexes of ZnTPP and para-substituted dendrimers ZnD1-G1, ZnD2-G1, ZnD3-G1, ZnD4-G1, ZnD5-G1, ZnD7-G2, ZnD8-G2.²⁴ As could be seen from Table 1, among the complexes of ZnD6-G1 with L2–L5²⁴ the complex between ZnD6-G1 and L4 has the highest stability constant. This could be explained by a good geometric match of the ligand size to the size of the intramolecular cavities of the porphyrinic receptor. The decrease in the value of the binding constant of the complexes between ZnD9-G2 and L4 in comparison with the similar complexes of the dendrimer with L3 testifies that beside a good geometric match between host–guest molecules the formation of additional hydrogen bonding interactions between the L3 and the triazole fragments of ZnD9-G2 may be possible.

Table 1 Stability constants of 1 : 1 complexes (K_{assoc 1}, M⁻¹) between ZnP and monodentate ligands L2–L5 in toluene, C_ZnP ≈ 1.1 × 10⁻⁵ M

	L2	L3	L4	L5
ZnTPP	5800	26460	39550	480
ZnD1-G1	24 180	120 800	86 900	8030
ZnD2-G1	26 400	118 050	79 500	11 090
ZnD3-G1	30 500	250 000	186 300	87 700
ZnD4-G1	1200	7250	5050	90
ZnD5-G1	3900	11 700	8500	240
ZnD6-G1	110 000	545 600	782 500	660 000
ZnD7-G2	30 530	70 400	43 000	9050
ZnD8-G2	25 000	80 250	61 800	15 100
ZnD9-G2	115 000	600 500	360 000	810 500

---

The dendrimers ZnD6-G1, ZnD9-G2 can be seen as a “picket-fence” porphyrins with intramolecular cavities formed by the 4-carbazolylphenyl-1,2,3-triazole end groups emanating from both sides of the porphyrin core (Fig. S4) (ESI†).²⁴

On the other hand, the meta-octasubstituted dendrimer ZnD3-G1 can not form similar intramolecular cavities for the ligand due to the lack of 1,2,3-triazole bridging fragments between tetrapyrrolic core and carbazolylphenyl branches. This is the reason why the binding ability of ZnD3-G1 towards L2–L5 is much less in comparison with ZnD6-G1,²⁴ ZnD9-G2 and it is comparable while significantly higher than the corresponding values for ZnD1-G1, ZnD2-G1 (Table 1). The dependence of the stability constants of octa-substituted dendrimers ZnD3-G1, ZnD6-G1, ZnD9-G2 with L2–L5 on the nature of small N-containing organic molecules is summarized in Fig. 1.

Fig. 1 Stability constants of ZnTPP and octa-substituted dendrimers with L2–L5 in toluene, 25 °C.

In line with our interests in the supramolecular chemistry of porphyrins,^25–28 we also investigated the binding ability of ZnD1-G1, ZnD2-G1, ZnD3-G1, ZnD4-G1, ZnD5-G1, ZnD6-G1, ZnD7-G2, ZnD8-G2, ZnD9-G2²⁴ towards the bidentate ligand L1. It is well known that upon interaction of ZnP with bifunctional nitrogen containing ligands formation of the complexes in a ratio of either 1 : 1 or 2 : 1 is possible.^29–31 Spatially distorted porphyrins or porphyrins with bulky substituents do not form complexes with L1 in a ratio of 2 : 1.

The study of complex formation of dendrimers with two (ZnD1-G1, ZnD4-G1, ZnD7-G2) and four (ZnD2-G1, ZnD5-G1, ZnD8-G2) branches and the octa-substituted dendrimer ZnD3-G1 without 1,2,3-triazole bridging groups between the tetrapyrrolic core and the carbazolylphenyl fragments with L1, using the method of spectrophotometric titration, showed that these processes, similarly to the system ZnP–L1, proceed in two stages. The changes in the UV-Vis spectra of the system ZnD3-G1–L1 in toluene are depicted in Fig. 2 as an example.

Fig. 2 The changes in the UV-Vis spectra of the system ZnD3-G1–L1 in toluene at 20 °C, C_ZnD3-G1 = 0 to 1.0 × 10⁻⁴ M.

There are two families of spectral curves with two sets of isosbestic points in the UV-Vis spectra of the system. Each of them is characterized by its own step in the corresponding titration curves (Fig. S5 and S6) (ESI†). Existence of two steps in the complexation also is confirmed by the graphical dependence of lg[(A₀ − A_i)/(A_i − A_k)] from lg C_L for the system. The splitting of the ligand non-equivalent proton signal in the ¹H NMR spectrum of the complex formed at the high concentrations of the ligand according with the literature^25–28 indicates the formation of a 1 : 1 complex. One signal of the ligand equivalent protons in the spectrum of the complex at lower concentrations of the ligand reveals the formation of the 2 : 1 complex between ZnD3-G1 and L1.

It should be noted that complex formation of dendrimers ZnD6-G1,²⁴ ZnD9-G2 with L1 in toluene proceeds in a single step with the formation of only 1 : 1 complexes. Probably, the presence of the bulky branches in the first and second generations prevents two-center coordination of L1. The stability constants of the considered complexes are presented in Table 2.

Table 2 The stability constants of 1 : 1 and 2 : 1 complexes of ZnP with bidentate ligand L1 in toluene at 25 °C, C_ZnP ≈ 1.5 × 10⁻⁵ M^a

	2 : 1 complexes, Kassoc 2, (M^−2)	1 : 1 complexes, Kassoc 1, (M^−1)
a The error in determining the stability constants was 5–7% (for 1 : 1 complexes) and 10% (for 2 : 1 complexes).
ZnTPP	5.0 × 10^9	1.9 × 10^5
ZnD1-G1	6.0 × 10^9	2.1 × 10^5
ZnD2-G1	6.0 × 10^9	2.2 × 10^5
ZnD3-G1	4.0 × 10^10	2.1 × 10^5
ZnD4-G1	1.7 × 10^8	2.3 × 10^4
ZnD5-G1	1.3 × 10^9	9.7 × 10^4
ZnD6-G1	—	1.3 × 10^6
ZnD7-G2	7.0 × 10^9	2.1 × 10^5
ZnD8-G2	8.0 × 10^9	2.9 × 10^5
ZnD9-G2	—	7.4 × 10^5

---

Conclusions

Thus, the study of complex formation of Zn(II) tetraarylporphyrins with carbazolylphenyl branches by spectrophotometric and ¹H NMR titration methods showed that their binding ability towards mono and bidentate N-containing organic molecules depends on the nature, number and generation of the branches. Bulky substituents are able either to significantly reduce the binding ability of the tetrapyrrolic cores due to the shielding of the porphyrin reaction centers, or significantly increase it by forming of intramolecular cavities for complementary binding of substrates. By varying the number of the branches and the number of their generation, it is possible to develop intramolecular cavities of different shapes for selective binding of guest molecules by a good geometric match of the ligand size to the size of the cavities, and by a existence of additional π–π and/or hydrogen bonding interactions between the ligand and the triazole fragments of the porphyrin. These metalloporphyrins could be considered as a molecular optical sensing device for small heterocyclic substrates due to a clear and easily identifiable response in the UV-Vis spectra of the reaction mixture.

Experimental

General experimental methods

NMR spectra were acquired on commercial instruments (Bruker Avance 300 MHz, Bruker AMX 400 MHz or Bruker Avance II⁺ 600 MHz) and chemical shifts (δ) are reported in parts per million (ppm) referenced to tetramethylsilane (TMS) or the internal (NMR) solvent signals. Mass spectra were run using a HP5989A apparatus (CI and EI, 70 eV ionisation energy) with Apollo 300 data system or a Thermo Finnigan LCQ advantage apparatus (ESI). Exact mass measurements were acquired on a Kratos MS50TC instrument (performed in the EI mode at a resolution of 10 000). Melting points (not corrected) were determined using a Reichert Thermovar apparatus. For column chromatography, 70–230 mesh silica 60 (E. M. Merck) was used as the stationary phase. Chemicals received from commercial sources were used without further purification. MALDI-TOF mass spectrometry was carried out on Bruker Daltonics – ultraflex II & ultraflex II TOF/TOF using the matrix 2,5-dihydroxylbenzoic acid for all samples.

Spectroscopic methods and instrumentation. 1,4-Diazabicyclo-[2.2.2]octane (L1), pyridine (L2), imidazole (L3), N-methylimidazole (L4) and 1,2,3-triazole (L5) from Sigma-Aldrich were used without further purification. ¹H NMR spectra were recorded on a Bruker VC-500 (500.17 MHz) in CDCl₃ using TMS as the internal standard. UV-Vis spectra of the porphyrins and their evolution upon addition of the ligands were measured on a Carry 100 spectrophotometer.

The UV-visible absorption spectral studies reveal red shifted Soret and visible bands upon addition of the ligands to a solution of the investigated receptor porphyrins confirming that the N-containing entity of the ligands binds to the Zn-cation of the coordination centre of the tetrapyrrolic macrocycle.

The stability constants of the metalloporphyrin complexes with the ligands in ratio of 1 : 1 (K_{assoc 1}) and 2 : 1 (K_{assoc 2}) were calculated according with the literature¹⁷ based on spectrophotometric data at two wavelengths (decreasing and increasing) using the following relationships:

where, λ₁ is the decreasing wavelength, λ₂ is the increasing wavelength, [A] is the Zn-porphyrin concentration, [B] is the ligand concentration, ΔA_o is the maximal change of the optical density at the given wavelength, ΔA_i – is the change of the optical density of the solution at a given wavelength at a given concentration.

Synthesis of 4-[(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methyl]benzaldehyde (3). 4-Bromomethylbenzaldehyde (2) (200 mg, 1.1 mmol, 1 equiv.) and 4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenol (1) were stirred in DMF (10 ml) for a few minutes. Then K₂CO₃ was added and the reaction was conducted at 80 °C overnight under N₂ atmosphere. Crude product was purified by column chromatography (silica, eluent CH₂Cl₂–heptane 2 : 1) to obtain (3) (366 mg, 73%) as a white solid. M.p. 190–192 °C. ¹H NMR (300 MHz, CDCl₃, 25 °C, TMS): δ = 10.05 (s, 1H, CHO), 8.13 (s, 2H, H-carbazole), 7.95 (d, ³J_H,H = 7.92, 2H, H-Ar), 7.66 (d, ³J_H,H = 7.89, 2H, H-Ar), 7.44 (d, ³J_H,H = 8.67, 4H, H-Ar), 7.25 (d, ³J_H,H = 8.49, 2H, H-Ar), 7.14 (d, ³J_H,H = 8.64, 2H, H-Ar), 1.45 ppm (s, 18H, tert-butyl). ¹³C NMR (75 MHz, CDCl₃, 25 °C, TMS): δ = 191.86, 130.14, 128.34, 127.58, 123.53, 116.21, 115.88, 109.03 (CH-Ar), 157.25, 143.69, 142.60, 139.63, 136.09, 131.48, 123.08 ppm (C-Ar), 69.57 (CH₂), 34.71 (C, tert-butyl), 32.02 ppm (CH₃, tert-butyl). HRMS (EI): m/z calcd for C₃₄H₃₅NO₂: 489.27 [M⁺]; found 489.26 [M⁺].

Synthesis of 3,5-bis[(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methyl]-2,4,6-trimethylbenzaldehyde (6). The mixture of arylaldehydes (180 mg), consisting of 3,5-bis(bromomethyl)-2,4,6-trimethylbenzaldehyde (4) and 3-bromomethyl-5-chloromethyl-2,4,6-trimethylbenzaldehyde (5), and carbazole-based phenol (1) were dissolved in DMF (10 ml) and the mixture was stirred at room temperature for a few minutes. Subsequently, K₂CO₃ (148 mg, 1.08 mmol) and a catalytic amount of 18-crown-6 (26.4 mg, 0.1 mmol) were added and the reaction was carried out at 80 °C overnight under N₂ atmosphere. Purification was conducted via a silica column (CH₂Cl₂–heptane1.5 : 1) to obtain (6) (390 mg) as a white solid. M.p. 268–270 °C. ¹H NMR (300 MHz, CDCl₃, 25 °C, TMS): δ = 10.72 (s, 1H, CHO), 8.14 (s, 4H, H-carbazole), 7.48 (m, 8H, H-carbazole), 7.29 (d, ³J_H,H = 8.64, 4H, H-Ar), 7.21 (d, ³J_H,H = 8.67, 4H, H-Ar), 5.22 (s, 4H, CH₂), 2.66 (s, 6H, 2 × CH₃), 2.62 (s, 3H, CH₃), 1.46 ppm (s, 36H, tert-butyl). ¹³C NMR (75 MHz, CDCl₃, 25 °C, TMS): δ = 195.33 (CHO), 128.34, 123.51, 116.21, 115.64, 109.05 (CH-Ar), 157.76, 143.63, 142.60, 140.26, 139.68, 134.01, 132.51, 131.42, 123.09 (C-Ar), 64.48 (CH₂), 34.72 (C, tert-butyl), 32.04 (CH₃, tert-butyl), 16.58 (CH₃), 15.89 ppm (CH₃). MALDI-TOF: m/z calcd for C₆₄H₇₀N₂O₃: 914.54 [M⁺]; found 914.53 [M⁺].

Synthesis of 5,15-bis(2,4,6-trimethylphenyl)-10,20-bis[4-(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methylphenyl] porphyrin (7). Arylaldehyde (3) (100 mg, 0.20 mmol, 1 equiv.) and 5-mesityldipyrromethane (54 mg, 0.20 mmol, 1 equiv.) were dissolved in dry CH₂Cl₂ (20 ml) and the solution was purged with N₂ for a few minutes. Then BF₃·OEt₂ (7.5 μl, 0.06 mmol, 0.3 equiv.), in dry CH₂Cl₂ (1 ml), was added dropwise and the resulting solution was stirred at room temperature for 1 hour under N₂ atmosphere. Subsequently, p-chloranil (100 mg, 0.41 mmol, 2 equiv.) was added in powder form and the mixture was heated at reflux for 1 hour. The solvent was evaporated and then purification was carried out with column chromatography. The first flash column (silica, eluent CH₂Cl₂) was to remove dark pigments and the second one (silicagel, CH₂Cl₂–heptane 1 : 1.5) was to separate the different porphyrin fractions. Pure product (51 mg, 34%) was obtained as a purple solid. ¹H NMR (300 MHz, CDCl₃, 25 °C, TMS): δ = 8.84 (d, ³J_H,H = 4.71, 4H, H-pyrrole), 8.71 (d, ³J_H,H = 4.5, 4H, H-pyrrole), 8.29 (d, ³J_H,H = 7.71, 4H, H-Ar), 8.17 (s, 4H, H-carbazole), 7.87 (d, ³J_H,H = 7.74, 4H, H-Ar), 7.56 (d, ³J_H,H = 8.46, 4H, H-Ar), 7.49 (d, ³J_H,H = 8.64, 4H, H-Ar), 7.36 (m, 8H, H-Ar), 7.29 (s, 4H, H-mesityl), 5.50 (s, 4H, 2 × CH₂), 2.63 (s, 6H, 2 × CH₃), 1.85 (s, 12H, 4 × CH₃), 1.48 (s, 36H, tert-butyl), −2.59 ppm (s, 2H, 2 × NH). ¹³C NMR (75 MHz, CDCl₃, 25 °C, TMS): δ = 134.77, 128.40, 127.78, 125.94, 123.54, 116.20, 116.04, 109.12 (CH-Ar), 157.88, 142.56, 141.88, 139.75, 139.39, 138.40, 137.76, 136.16, 131.31, 123.10, 118.86, 118.42 (C-Ar), 70.50 (CH₂), 34.74 (C, tert-butyl), 32.06 (CH₃, tert-butyl), 21.65 (CH₃), 21.48 ppm (CH₃). MALDI-TOF: m/z calcd for C₁₀₄H₁₀₀N₆O₂: 1465.94 [M⁺]; found 1465.84 [M⁺].

Synthesis of 5,10,15,20-tetrakis[4-(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methylphenyl] porphyrin (8). Compound (3) was reacted with pyrrole under BF₃·OEt₂ catalysis using the procedure that was applied for the synthesis of dendrimer D1 (7). Crude product was purified by column chromatography (silica, CH₂Cl₂–heptane 1 : 1) to get pure compound (15%) as a purple solid. ¹H NMR (300 MHz, CDCl₃, 25 °C, TMS): δ = 8.92 (s, 8H, H-pyrrole), 8.28 (d, ³J_H,H = 7.71 Hz, 8H, H-Ar), 8.17 (s, 8H, H-carbazole), 7.86 (d, ³J_H,H = 7.71 Hz, 8H, H-Ar), 7.55 (d, ³J_H,H = 8.46 Hz, 8H, H-Ar), 7.49 (d, ³J_H,H = 8.67 Hz, 8H, H-Ar), 7.35 (d, ³J_H,H = 8.64 Hz, 16H, H-Ar), 5.45 (s, 8H, CH₂), 1.48 (s, 72H, tert-butyl), −2.70 ppm (s, 2H, 2 × NH). ¹³C NMR (75 MHz, CDCl₃, 25 °C, TMS): δ = 134.83, 128.38, 125.98, 123.55, 116.21, 116.00, 109.12 (CH-Ar), 157.86, 142.57, 141.95, 139.73, 136.27, 131.31, 123.10, 119.82 (C-Ar), 70.42 (CH₂), 34.73 (C, tert-butyl), 32.05 (CH₃, tert-butyl). MALDI-TOF: m/z calcd for C₁₅₂H₁₄₆N₈O₄: 2148.84 [M⁺]; found 2148.30 [M⁺].

Synthesis of 5,10,15,20-tetrakis[3,5-bis((4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methyl)-2,4,6-trimethylphenyl] porphyrin (9). 3,5-Bis[(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenoxy)methyl]-2,4,6-trimethylbenzaldehyde (6) (200 mg, 0.21 mmol, 1 equiv.) and pyrrole (15 μl, 0.21 mmol, 1 equiv.) were dissolved in CH₂Cl₂ (22 ml) and absolute ethanol (164 μl). The solution was purged with N₂ for 15 minutes. The reaction was carried out following the procedure described above using BF₃·OEt₂ (0.3 equiv.). Crude product was purified by column chromatography (silica, CH₂Cl₂–heptane 1 : 1) to get pure compound (10 mg, 5%) as a purple solid. ¹H NMR (600 MHz, CDCl₃, 25 °C, TMS): δ = 8.82 (s_br, 8H, H-pyrrole), 8.10 (s, 16H, H-carbazole), 7.48 (s_br, 16H, H-Ar), 7.39 (d, ³J_H,H = 7.74 Hz, 16H, H-Ar), 7.32 (s_br, 16H, H-Ar), 7.25 (s, 16H, H-Ar), 5.46 (s_br, 16H, CH₂), 2.93 (s, 12H, CH₃), 2.05 (s_br, 24H, CH₃), 1.41 ppm (s, 144H, tert-butyl). ¹³C NMR (100 MHz, CDCl₃, 25 °C, TMS): δ = 158.15, 144.97, 142.55, 140.91, 140.16, 139.72, 131.27, 130.96, 128.35, 123.47, 123.07, 116.16, 115.77, 109.08 (C, CH-Ar), 65.86 (CH₂), 34.68 (C, tert-butyl), 32.00 (CH₃, tert-butyl), 19.32 (CH₃), 16.33 ppm (CH₃). MALDI-TOF: m/z calcd for C₂₇₂H₂₈₆N₁₂O₈: 3851.26 [M⁺]; found 3851.58 [M⁺].

General procedure for synthesis of zinc(II) porphyrin

Porphyrin (15 mg, 1 equiv.) and Zn(OAc)₂·H₂O (4 equiv.) were added to a flask of 25 ml containing CHCl₃ (10 ml) and the solution was heated at reflux for 4 hours. The resulting mixture was washed three times with distilled water. The organic layer was dried over MgSO₄ and the solvent was evaporated under vacuum to obtain Zn(II)-porphyrin in pure form in quantitative yield.

Dendrimer ZnD1-G1. ¹H NMR (300 MHz, CDCl₃, 25 °C, TMS): δ = 8.93 (d, ³J_H,H = 4.71 Hz, 4H, H-pyrrole), 8.80 (d, ³J_H,H = 4.71 Hz, 4H, H-pyrrole), 8.30 (d, ³J_H,H = 7.92 Hz, 4H, H-Ar), 8.17 (d, ⁴J_H,H = 1.5 Hz, 4H, H-carbazole), 7.86 (d, ³J_H,H = 7.92 Hz, 4H, H-Ar), 7.56 (d, ³J_H,H = 8.85 Hz, 4H, H-Ar), 7.49 (dd, ³J_H,H = 8.67 Hz, ⁴J_H,H = 1.71 Hz, 4H, H-Ar), 7.35 (dd, ³J_H,H = 8.85 Hz, ⁴J_H,H = 2.46 Hz, 8H, H-Ar), 7.28 (s, 4H, H-mesityl), 5.48 (s, 4H, CH₂), 2.63 (s, 6H, CH₃), 1.84 (s, 12H, CH₃), 1.48 ppm (s, 36H, tert-butyl). ¹³C NMR (75 MHz, CDCl₃, 25 °C, TMS): δ = 134.68, 132.33, 130.85, 128.38, 127.68, 125.80, 123.55, 116.20, 116.06, 109.14 (CH-Ar), 157.89, 150.04, 149.98, 142.68, 142.56, 139.75, 139.24, 138.99, 137.49, 135.87, 131.29, 123.10, 119.78, 119.39 (C-Ar), 70.57 (CH₂), 34.74 ppm (C, tert-butyl). MALDI-TOF: m/z calcd for C₁₀₄H₉₈N₆O₂Zn: 1527.71 [M⁺]; found 1527.78 [M⁺].

Dendrimer ZnD2-G1. ¹H NMR (300 MHz, CDCl₃, 25 °C, TMS): δ = 9.03 (s, 8H, H-pyrrole), 8.30 (d, ³J_H,H = 7.71 Hz, 8H, H-Ar), 8.17 (d, ⁴J_H,H = 1.29 Hz, 8H, H-carbazole), 7.88 (d, ³J_H,H = 7.92 Hz, 8H, H-Ar), 7.56 (d, ³J_H,H = 8.67 Hz, 8H, H-Ar), 7.49 (dd, ³J_H,H = 8.67 Hz, ⁴J_H,H = 1.68 Hz, 8H, H-Ar), 7.35 (dd, ³J_H,H = 8.64 Hz, ⁴J_H,H = 2.46 Hz, 16H, H-Ar), 5.48 (s, 8H, CH₂), 1.48 (s, 72H, tert-butyl). ¹³C NMR (75 MHz, CDCl₃, 25 °C, TMS): δ = 134.70, 132.13, 128.38, 125.87, 123.54, 116.21, 116.03, 109.12 (CH-Ar), 157.89, 150.24, 142.65, 142.57, 139.75, 136.02, 131.31, 123.11, 120.80 (C-Ar), 70.54 (CH₂), 34.73 (C, tert-butyl), 32.05 (CH₃, tert-butyl). MALDI-TOF: m/z calcd for C₁₅₂H₁₄₄N₈O₄Zn: 2210.06 [M⁺]; found 2210.28 [M⁺].

Dendrimer ZnD3-G1. ¹H NMR (600 MHz, CDCl₃, 25 °C, TMS): δ = 8.88 (s_br, 8H, H-pyrrole), 8.12 (s, 16H, H-pyrrole), 7.50 (d, ³J_H,H = 8.1 Hz, 16H, H-Ar), 7.40 (d, ³J_H,H = 8.82 Hz, 16H, H-Ar), 7.32 (s_br, 16H, H-Ar), 7.27 (d, ³J_H,H = 8.82 Hz, 16H, H-Ar), 5.48 (s_br, 16H, CH₂), 2.95 (s, 12H, CH₃), 2.05 (s_br, 24H, CH₃), 1.43 (s, 144H, tert-butyl). ¹³C NMR (100 MHz, CDCl₃, 25 °C, TMS): δ = 128.34, 123.47, 116.17, 115.79, 109.09 (CH-Ar), 158.20, 150.08, 142.55, 140.72, 139.73, 131.44, 131.25, 130.80, 123.09, (C-Ar), 65.94 (CH₂), 34.68 (C, tert-butyl), 32.01 (CH₃, tert-butyl), 19.27 (CH₃), 16.30 ppm (CH₃). MALDI-TOF: m/z calcd for C₂₇₂H₂₈₄N₁₂O₈Zn: 3912.15 [M⁺]; found 3912.26 [M⁺].

Acknowledgements

This work was supported by a FP-7 grant from the EC for Research, Technological Development and Demonstration Activities, “Dendrimers for Photonic Devices” IRSES-PEOPLE-2009-247260-DphotoD, under the “International Research Staff Exchange Scheme”. Nguyen Tran Nguyen thanks Ministry of Education and Training, Vietnam International Education Development (VIED) for financial support during his Ph.D in KU Leuven, Belgium.

Notes and references

1. H. Azho, L. Baldini, J. Hong, A. J. Wilson and A. D. Hamilton, J. Am. Chem. Soc., 2006, 128, 2421–2425.
2. P. Weyermann, J.-P. Gisselbrecht, C. Boudon, F. Diederich and M. Gross, Angew. Chem., Int. Ed., 1999, 38, 3212–3215.
3. H. S. Park, Q. Lin and A. D. Hamilton, J. Am. Chem. Soc., 1999, 121, 8–13.
4. A. Zingg, B. Felber, L. Fu, J. P. Collman and F. Diederich, Helv. Chim. Acta, 2002, 85, 333–351.
5. P. Weyermann and F. Diederich, Helv. Chim. Acta, 2002, 85, 599.
6. G. J. Hawker and J. M. J. Fréchet, J. Am. Chem. Soc., 1990, 112, 7638–7647.
7. R. H. Jin, T. Aida and S. Inoue, Chem. Commun., 1993, 1260–1262.
8. Y. Tomoyose, D. L. Jiang, R. H. Jin, T. Aida, T. Yamashita, K. Horie, E. Yashima and Y. Okamoto, Macromolecules, 1996, 29, 5236–5238.
9. P. J. Dandliker, F. Diederich and J.-P. Gisselbrecht, Angew. Chem. Int., Ed. Engl., 1995, 34, 2775.
10. J. P. Collman, L. Fu, A. Zingg and F. Diederrich, Chem. Commun., 1997, 193–194.
11. S. Shinoda, J. Inclusion Phenom. Macrocyclic Chem., 2007, 59, 1–9.
12. K. Maurer, K. Hager and A. Hirsch, Eur. J. Org. Chem., 2006, 3338–3347.
13. G. L. Hawker and J. M. J. Frechet, J. Am. Chem. Soc., 1990, 112, 7638–7647.
14. J. Kress, A. Rosner and A. Hirsch, Chem.–Eur. J., 2000, 6, 247–257.
15. B. Buschhaus, W. Bauer and A. Hirsch, Tetrahedron, 2003, 59, 3899–3915.
16. B. Buschhaus and A. Hirsch, Eur. J. Org. Chem., 2005, 1148–1160.
17. B. Buschhaus, F. Hampel, S. Grimme and A. Hirsch, Chem.–Eur. J., 2005, 11, 3530–3540.
18. N. T. Nguyen, J. Hofkens, I. Scheblykin, M. Kruk and W. Dehaen, Eur. J. Org. Chem., 2014, 1766–1777.
19. T. Rohand, E. Dolusic, T. H. Ngo, W. Maes and W. Dehaen, ARKIVOC, 2007, 307–324.
20. Z.-H. Zhao, H. Jin, Y.-X. Zhang, Z. Shen, D.-C. Zou and X.-H. Fan, Macromolecules, 2011, 44, 1405–1413.
21. J.-M. Barbe, G. Canard, S. Brandes and R. Guilard, Eur. J. Org. Chem., 2005, 21, 4601–4611.
22. G. M. Mamardashvili, I. A. Shinkar, N. Zh. Mamardashvili and O. I. Koifman, Russ. J. Coord. Chem., 2007, 33(10), 774–778.
23. N. Zh. Mamardashvili and O. I. Koifman, Russ. J. Org. Chem., 2005, 41(6), 807–826.
24. N. T. Nguyen, G. Mamardashvili, M. Gruzdev, N. Mamardashvili and W. Dehaen, Supramol. Chem., 2013, 25(3), 180–188.
25. G. M. Mamardashvili, N. Zh. Mamardashvili and O. I. Koifman, Russ. J. Inorg. Chem., 2007, 52(8), 1215.
26. G. M. Mamardashvili, O. M. Kulikova, N. Zh. Mamardashvili and O. I. Koifman, Russ. J. Coord. Chem., 2008, 34(6), 427–433.
27. G. M. Mamardashvili, O. M. Kulikova and O. I. Koifman, Russ. J. Gen. Chem., 2007, 77(11), 1965–1971.
28. G. M. Mamardashvili, O. M. Kulikova, N. Zh. Mamardashvili and O. I. Koifman, Russ. J. Gen. Chem., 2008, 78(10), 1964–1971.
29. V. V. Borovkov and Y. Inoue, Top. Curr. Chem., 2006, 265, 89–146.
30. V. V. Borovkov, N. Zh. Mamardashvili and Y. Inoue, Russ. Chem. Rev., 2006, 75, 737–748.
31. B. C. Kavaric, B. Kokona, A. D. Shwab, M. A. Twomey, J. C. Paula and R. Faiman, J. Am. Chem. Soc., 2006, 128, 4166–4167.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3ra45660a

---