Synthesis and characterization of 4-chlorobutyl ester of 5-(8-carboxyl-1-naphthyl)-10,15,20-triphenyl-porphyrin and its zinc complex

ABSTRACT In the presence of Brønsted–Lowry acids (phenol or dry HCl), the acyl chloride, which was obtained by the reaction between 5-(8-carboxyl-1-naphthyl)-10,15,20-triphenyl-porphyrin (CNTPP) and oxalyl chloride, reacted with tetrahydrofuran and led to the 4-chlorobutyl ester, P1, as the result of the acylative cleavage. P1 and its zinc complex [ZnP1] have been characterized by 1H NMR. The structure of [ZnP1] was obtained by X-ray crystallography. Zinc is coordinated by four pyrrole nitrogens. The 8-position substituent, a 4-chlorobutyl ester group, lies above the porphyrin plane.


Introduction
Porphyrins and their derivatives have been found in many biological systems, such as hemoglobin, cytochrome P450, horseradish peroxidase, etc. Besides their importance in biological sciences, porphyrins also play important roles in material sciences, such as catalytic asymmetric synthesis, nonlinear optics, molecular devices, etc. [1][2][3][4][5] One attracting feature of the porphyrin species is that it can be functionalized at many positions, such as meso or b positions, which can lead to many kind of novel porphyrins with various structures and functions.
We have recently been working on a series of mononaphthyl substituted porphyrins. [6][7][8][9][10][11] These porphyrins can be easily functionalized at the 8-position of the naphthyl group. For example, we have designed and synthesized its phenolate derivative with a covalently attached phenol tail, the ligand P3 (shown in Scheme 1), as a structural mimic of catalases, which has a hanged phenol group at the appropriate position to further interact with the center metal. [8] In that preparation, dichloromethane was used as the solvent. When we extended it to other phenols, we tried to use tetrahydrofuran (THF) as the solvent. Interestingly, the 4-chlorobutyl ester, P1, has been obtained as the major product as shown in Scheme 1. Its zinc complex was also prepared and crystallographically characterized. Herein, we report the synthesis and characterization of P1 and [ZnP1].

General procedures
Materials and general methods All reagents were obtained from commercial sources without further purification unless otherwise noted. Anhydrous THF was dried and redistilled over sodium benzophenone ketyl. The dichloromethane (CH 2 Cl 2 ) was freshly distilled from CaH 2 under nitrogen used. CNTPP was synthesized according to the reported method. [

Synthesis of [ZnP1]
A solution of Zn(CH 3 COO) 2 ¢2H 2 O (0.125 g, 0.57 mmol) in methanol (5 mL) was added to the solution of P1 (0.22 g, 0.28 mmol) in chloroform (50 mL). The mixture was refluxed for 3 h. After the completion of the reaction, the solution was washed with water (3 £ 100 mL). The organic layer was collected, then dried over magnesium sulfate, and concentrated under reduced pressure. The residue was purified by chromatography (silica, CH 2 Cl 2 /hexane 1:1). The light purple solid was obtained. Yield 0.20 g (82%). 1

X-ray crystallography
Single crystals of [ZnP1] were obtained by slow evaporation of its solution in the mixture of dichloromethane and hexane (1:1). All measurements were made on Agilent Xcalibur diffractometer with an Atlas (Gemini Ultra Cu) detector using graphite monochromated Cu Ka (λ D 0.154178 nm). A purple crystal with the dimensions 0.35 £ 0.30 £ 0.20 mm 3 was used for the structure determination. It was glued to a glass fiber by epoxy cement and measured at 223 K. Cell refinement and data reduction were carried out with the use of the program CrysA-lisPro (Agilent Technologies, Version 1.171.36.32, 2013), and absorption corrections (multi-scan) were applied. The structure was solved by direct methods using SHELXS-97 and refined against F 2 using SHELXL-97; [12] subsequent difference Fourier syntheses led to the location of the remaining non-hydrogen atoms. For the structure refinement, all data were used including negative intensities. The asymmetric unit contains one porphyrin molecule and one methylene chloride solvate. All non-hydrogen atoms were refined anisotropically. Hydrogen atoms were added with the standard SHELXL-97 idealization methods. Brief crystal data for the structure are listed in Table 1.

Results and discussion
The compound P1 has been synthesized in anaerobic condition, and the overall procedure is summarized in Scheme 1. Our original design is to synthesize P2-P5. Unexpectedly, the reaction led directly to the production of P1, a 4-chlorobutyl ester, as the major product; P2 (or P3, P4, P5) as the minor product. Their formulae have been confirmed by electrospray-ionization (ESI) mass spectrometry (shown in the supporting information). For example, ESI revealed the ion peaks at m/z D 799.28 for the compound P1, which corresponds to [MCH] C .
In the reaction shown in Scheme 1, the butyl group was obviously from THF, and the product P1 was formed by the cleavage of THF. The cleavage of THF with acyl chlorides is a useful reaction, particularly in the synthesis of molecules where a four-carbon chain is required to be added. [13][14][15] Besides the acyl chloride, phenol is the other possible reagent involved in the reaction. So, phenol, a Brønsted-Lowry acid, could be the key reagent to promote the cleavage. We also tried to add NEt 3 during the reaction; no P1 was obtained under such condition. It confirms that the acidity of phenol is essential to the acylative cleavage of THF. Since phenols are weak Brønsted-Lowry acids, how about strong Brønsted-Lowry acid? We also tried to use dry HCl as shown in Scheme 2. The reaction led to a higher yield of P1 (91%). Furthermore,  ) 1.035 and ¡0.873 comparison experiments have been performed to benzoic chloride instead of CNTPP. As shown in Scheme 2, in the presence of excess catechol, the reaction ii led to both B1 (4-chlorobutyl ester) and B2. When dry HCl was used, the reaction iii gave B1 as the only product. The aforementioned study suggests that such acylative cleavage reaction is Brønsted-Lowry acid-promoted. As we also learned from the literature, many catalysts have been reported for similar transformation. There are metallic complexes, such as ZnCl 2, [16]  BiCl 3 , [20] and La(NO 3 ) 3, [21] or metals, such as Al, [22] Zn, [23] and Mg, [24] or non-metals, such as iodine [25] and graphite. [26] In most cases, they are Lewis acids. So, it may not be too surprising in light of the results from analogous reactions using Brønsted-Lowry acids. The detailed reaction mechanism is not clear so far. It is likely that a THF molecule is broken down and subsequently forms a chlorobutyl ester with the 8-carboxylnaphthyl group since it is known that THF is unstable under acidic condition and can react with HCl to produce 4-chlorobutanol under certain conditions. [27] 1 H NMR spectra Due to the ring current effect, NMR spectroscopy is a good method to study intramolecular interactions in porphyrins. [28] Scheme 2. i) Synthetic route to P1 using dry HCl. ii) Synthetic route to B1 by using catechol. iii) Synthetic route to B1 by using dry HCl. All these compounds have been characterized by 1 H NMR. For comparison, 4-chlorobutyl benzoate (compound B1) has been prepared. 1 H NMR spectra of P1, B1 and [ZnP1] are shown in Figure 1. The corresponding chemical shifts for 1 H NMR are listed in Table S1 in the supporting information. For P1, the signal at ¡2.67 ppm has been assigned to two inner NH protons, which is typical in free base porphyrin. [28] The signals of four protons at ¡0.58, ¡0.08, 0.33, and 2.16 ppm are assigned to the butyl group. The corresponding resonances for the compound B1 are located at 4.35 (H2), 1.95 (H3 and H4), and 3.59 (H5) according to literature. [29] If we consider that the butyl group is located above the porphyrin plane in the crystal structure (vide infra), the porphyrin ring current effect will cause their resonances shift upfield. So, these signals are assigned as shown in Figure 1. Their corresponding upfield shifts are 4.01, 2.51, 2.02, and 1.42 ppm compared with compound B1. Such upfield shifts are similar to the case of 5-(8-ethoxycarbonyl-1naphthyl)-10,15,20-triphenyl porphyrin (ENTPP). [6] For [ZnP1], the resonances of butyl group have slight downfield shifts compared with the compound P1. We notice that there is one resonance at 1.20 ppm, which is caused by the residue H 2 O in CDCl 3 . It is generally at 1.50 ppm. The upfield shift for H 2 O indicates that the water molecule is coordinated to the zinc porphyrinate. As in our previous report on [Zn(ENTPP)(H 2 O), [7] the crystal structure suggests that H 2 O is coordinated to zinc through the same side of the ester group. For [ZnP1], the coordination of H 2 O could cause the butyl group away from the porphyrin plane, therefore lead to slightly downfield shifts of their 1 H NMR resonances.

Molecular structure
One of the single crystals of [ZnP1] was measured by X-ray crystallography. Its structure was resolved in the space group P2 1 /c. One asymmetric unit contains one complete porphyrin and one methylene chloride solvate. The structure of porphyrin is displayed in Figure 2. It has one naphthyl and three phenyl groups on the meso-position. Zinc was inserted into the porphyrin core and was coordinated by four pyrrole nitrogens. It is coplanar with four nitrogen (the displacement out of four nitrogen plane is 0.03 A ), which is typical for four-coordinate species. The porphyrin core is quite planar with the maximum deviation of 24 atoms from the mean plane as 0.12 A (see Figure 3). Selected bond distances are listed in Table 2. The average Zn-N p distance is 2.035(6) A , which is slightly shorter than that in five-coordinate zinc complexes. [7] Because of the constraint of naphthyl unit, the 8-position substituents, a chlorobutyl group, lies above the porphyrin plane (vide supra). The distances between C(2), C(3), C(4), C(5), and the porphyrin plane are 3.42, 3.56, 4.56, and 4.58 A , respectively.

Conclusion
In conclusion, the 4-chlorobutyl ester of CNTPP was synthesized. Its structure was confirmed by mass spectroscopy, 1 H NMR, and X-ray crystallography. Such a product was formed by the acylative cleavage of THF. Our study suggests that the cleavage is promoted by Brønsted-Lowry acids. It provides a convenient method to prepare P1, a 4-chlorobutyl derivative of   porphyrin with a four-carbon chain. Such a compound could also be used as a precursor for other derivatives of CNTPP.

Funding
This work was supported by the Natural Science Foundation of China (No. 20971093 and 21271133) and the Priority Academic Program Development of Jiangsu Higher Education Institutions.