[2 + 2] Photodimerization of Naphthylvinylpyridines through Cation-π Interactions in Acidic Solution

Irradiation of (E)-4-(2-(2-naphthyl)vinyl)pyridine (1a) and (E)-4-(2-(1-naphthyl)vinyl)pyridine (1b) with a 250 W high-pressure mercury lamp in acidic solution afforded synHT dimers in high stereoselectivities. Similar results were obtained by visible light irradiation. On the other hand, when the reactions were carried out under neutral conditions, the stereoselectivities were very low, and the yields were decreased by visible light irradiation. Comparison of the UV-vis spectra between the acidic and the neutral conditions elucidated that the red shift was observed in acidic solutions. These results show that HCl plays essential roles not only in the preorientation of substrates through cation-π interactions, but also in the changes in the absorption properties of substrates that enable visible light reactions.

It has been reported that the size of the π-system is significantly correlated with the strength of the cation-π interactions; for example, the interaction energy of a sodium cation with naphthalene is 2-3.5 kcal/mol larger than that with benzene [23,24]. This prompted us to investigate the [2 + 2] photodimerization of naphthylvinylpyridines having more extended π-systems than those of styrylpyridines. In this paper, we report that the photodimerization of naphthylvinylpyridines 1a and 1b (Figure 1) in acidic solution provides synHT dimers with very high regio-and stereoselectivities through cation-π interaction, even under irradiation with visible light.

Results and Discussion
(E)-4-(2-(2-Naphthyl)vinyl)pyridine (1a) [25] and (E)-4-(2-(1-naphthyl)vinyl)pyridine (1b) [25] were employed as substrates for the photochemical reactions, the π-conjugate systems of which differed from each other (Figure 1). Irradiation of 1a in a 1.0 M THF solution with a 450 W high-pressure mercury lamp for 8 h under neutral conditions afforded all four possible dimers 2a-5a. SynHT dimer 3a was the major product, but the selectivity was quite low-as shown in Table 1 (entry 1). The conversion and the product ratios were determined based on 1 H-NMR spectra. Though the reaction was carried out in a 0.067 M MeOH solution due to the much lower solubility of 1a than that in THF, a similar result was observed (entry 2). Photolysis in the presence of 1 equiv. of conc. hydrochloric acid resulted in remarkable changes in the product distribution (entry 3); the synHT adduct 2a increased to become the major product along with a small amount of 3a and 4a. As the amount of HCl loading increased to 3 equiv., the yield of the synHT dimer 2a increased to 93% (entry 4).  Figure 2a,b show the 1 H-NMR spectra of 1a after irradiation in the absence and presence of 3 equiv. of HCl, respectively. A comparison clearly demonstrates the significant effect of the acid on the product distribution. Figure 2a shows multi-peaks around δ 3.7 to 4.8 for the methine protons of the four cyclobutane dimers. On the other hand, in the presence of 3 equiv. of HCl, the four methine protons appear at δ 4.59 and 4.62, indicating exclusive formation of 2a (Figure 2b).
Irradiation of 1b in the absence of the acid gave a mixture of 2b-4b and (Z)-isomer (entry 5). In the presence of 1 equiv. of HCl, the solubility of 1b decreased; therefore, the reaction was carried out in a 0.67 M MeOH solution. The acid demonstrated a significant effect similar to that in the case of 1a; synHT dimer 2b was obtained in 75% yield with no other dimers along with 25% of (Z)-isomer (entry 6). A survey of the solvents found that a 9:1 mixture of MeOH and H 2 O had a good solubility for 1b·HCl. Using this solvent system, irradiation of a 1.0 M solution of 1b in the presence of acid

Results and Discussion
(E)-4-(2-(2-Naphthyl)vinyl)pyridine (1a) [25] and (E)-4-(2-(1-naphthyl)vinyl)pyridine (1b) [25] were employed as substrates for the photochemical reactions, the π-conjugate systems of which differed from each other (Figure 1). Irradiation of 1a in a 1.0 M THF solution with a 450 W high-pressure mercury lamp for 8 h under neutral conditions afforded all four possible dimers 2a-5a. SynHT dimer 3a was the major product, but the selectivity was quite low-as shown in Table 1 (entry 1). The conversion and the product ratios were determined based on 1 H-NMR spectra. Though the reaction was carried out in a 0.067 M MeOH solution due to the much lower solubility of 1a than that in THF, a similar result was observed (entry 2). Photolysis in the presence of 1 equiv. of conc. hydrochloric acid resulted in remarkable changes in the product distribution (entry 3); the synHT adduct 2a increased to become the major product along with a small amount of 3a and 4a. As the amount of HCl loading increased to 3 equiv., the yield of the synHT dimer 2a increased to 93% (entry 4).

Results and Discussion
(E)-4-(2-(2-Naphthyl)vinyl)pyridine (1a) [25] and (E)-4-(2-(1-naphthyl)vinyl)pyridine (1b) [25] were employed as substrates for the photochemical reactions, the π-conjugate systems of which differed from each other (Figure 1). Irradiation of 1a in a 1.0 M THF solution with a 450 W high-pressure mercury lamp for 8 h under neutral conditions afforded all four possible dimers 2a-5a. SynHT dimer 3a was the major product, but the selectivity was quite low-as shown in Table 1 (entry 1). The conversion and the product ratios were determined based on 1 H-NMR spectra. Though the reaction was carried out in a 0.067 M MeOH solution due to the much lower solubility of 1a than that in THF, a similar result was observed (entry 2). Photolysis in the presence of 1 equiv. of conc. hydrochloric acid resulted in remarkable changes in the product distribution (entry 3); the synHT adduct 2a increased to become the major product along with a small amount of 3a and 4a. As the amount of HCl loading increased to 3 equiv., the yield of the synHT dimer 2a increased to 93% (entry 4).  Irradiation of 1b in the absence of the acid gave a mixture of 2b-4b and (Z)-isomer (entry 5). In the presence of 1 equiv. of HCl, the solubility of 1b decreased; therefore, the reaction was carried out in a 0.67 M MeOH solution. The acid demonstrated a significant effect similar to that in the case of 1a; synHT dimer 2b was obtained in 75% yield with no other dimers along with 25% of (Z)-isomer (entry 6). A survey of the solvents found that a 9:1 mixture of MeOH and H 2 O had a good solubility for 1b·HCl. Using this solvent system, irradiation of a 1.0 M solution of 1b in the presence of acid   Irradiation of 1b in the absence of the acid gave a mixture of 2b-4b and (Z)-isomer (entry 5). In the presence of 1 equiv. of HCl, the solubility of 1b decreased; therefore, the reaction was carried out in a 0.67 M MeOH solution. The acid demonstrated a significant effect similar to that in the case of 1a; synHT dimer 2b was obtained in 75% yield with no other dimers along with 25% of (Z)-isomer (entry 6). A survey of the solvents found that a 9:1 mixture of MeOH and H 2 O had a good solubility for 1b·HCl. Using this solvent system, irradiation of a 1.0 M solution of 1b in the presence of acid significantly improved the selectivity to 94%, suggesting the importance of concentration in obtaining good selectivities in these reactions (entries 7 and 8). significantly improved the selectivity to 94%, suggesting the importance of concentration in obtaining good selectivities in these reactions (entries 7 and 8). Another remarkable feature observed under the acidic conditions was that visible light could also induce photodimerization. When the irradiation of 1a was carried out with a 250 W high-pressure mercury lamp through a UV-cut filter for 3 h, only (Z)-isomer was produced in 7% yield ( Table 2, entry 1). On the contrary, in the presence of HCl, the reaction proceeded to provide synHT dimer 2a in a similar selectivity to that obtained by UV-light irradiation (entry 2). Irradiation of 1b in the absence of the acid gave dimers 2b and 3b in low yields, whereas in the presence of the acid, 2b was produced in a significantly high yield (entries 3 and 4). The remarkable differences in the photoreactivities whether in the presence or absence of HCl can be explained by differences in their absorption properties. UV-vis spectra for 1a, 1b, 1a·HCl, and 1b·HCl are shown in Figure 3. The longest absorption bands of 1a and 1b in MeOH appeared at 319.0 and 331.0 nm, respectively. On the other hand, in the presence of HCl these bands were shifted to longer wavelengths and observed at 331.0 and ca. 360 nm, respectively, the shoulders of which exceeded 400 nm. These changes in the absorption properties induced by the acid are thought to be responsible for the visible light photodimerization in acidic media. Another remarkable feature observed under the acidic conditions was that visible light could also induce photodimerization. When the irradiation of 1a was carried out with a 250 W high-pressure mercury lamp through a UV-cut filter for 3 h, only (Z)-isomer was produced in 7% yield ( Table 2, entry 1). On the contrary, in the presence of HCl, the reaction proceeded to provide synHT dimer 2a in a similar selectivity to that obtained by UV-light irradiation (entry 2). Irradiation of 1b in the absence of the acid gave dimers 2b and 3b in low yields, whereas in the presence of the acid, 2b was produced in a significantly high yield (entries 3 and 4). significantly improved the selectivity to 94%, suggesting the importance of concentration in obtaining good selectivities in these reactions (entries 7 and 8). Another remarkable feature observed under the acidic conditions was that visible light could also induce photodimerization. When the irradiation of 1a was carried out with a 250 W high-pressure mercury lamp through a UV-cut filter for 3 h, only (Z)-isomer was produced in 7% yield ( Table 2, entry 1). On the contrary, in the presence of HCl, the reaction proceeded to provide synHT dimer 2a in a similar selectivity to that obtained by UV-light irradiation (entry 2). Irradiation of 1b in the absence of the acid gave dimers 2b and 3b in low yields, whereas in the presence of the acid, 2b was produced in a significantly high yield (entries 3 and 4). The remarkable differences in the photoreactivities whether in the presence or absence of HCl can be explained by differences in their absorption properties. UV-vis spectra for 1a, 1b, 1a·HCl, and 1b·HCl are shown in Figure 3. The longest absorption bands of 1a and 1b in MeOH appeared at 319.0 and 331.0 nm, respectively. On the other hand, in the presence of HCl these bands were shifted to longer wavelengths and observed at 331.0 and ca. 360 nm, respectively, the shoulders of which exceeded 400 nm. These changes in the absorption properties induced by the acid are thought to be responsible for the visible light photodimerization in acidic media. The remarkable differences in the photoreactivities whether in the presence or absence of HCl can be explained by differences in their absorption properties. UV-vis spectra for 1a, 1b, 1a·HCl, and 1b·HCl are shown in Figure 3. The longest absorption bands of 1a and 1b in MeOH appeared at 319.0 and 331.0 nm, respectively. On the other hand, in the presence of HCl these bands were shifted to longer wavelengths and observed at 331.0 and ca. 360 nm, respectively, the shoulders of which exceeded 400 nm. These changes in the absorption properties induced by the acid are thought to be responsible for the visible light photodimerization in acidic media. The structures of the product dimers were determined from X-ray crystallographic analyses and the 1 H-NMR and MS spectra. The X-ray analyses were carried out for 2a, 2b·2HCl, 3b, and 4a, as single crystals suitable for X-ray analyses were obtained. Figure 4 clearly shows that 2a and 2b are synHT dimers, and 3b and 4a are synHH and antiHT dimers, respectively. Based on MS and 1 H-NMR analyses, 3a and 5a were assigned as synHH and antiHH dimers, respectively. In their MS spectra, the fragment peaks of m/z 231 and 280 were observed. The peak of m/z 231 corresponds to a naphthylethylpyridine moiety, and the peak of m/z 280 corresponds to a binaphthylethane moiety, indicating their head-to-head structures ( Figures S8, S9, S12, and S13). The methine protons at δ 4.67 and 4.69 for 3a and δ 3.88 and 3.97 for 5a are very close to those of related synHH and antiHH dimers, respectively [18]. The assignment of 4b was also performed by MS spectra; the observations of a base peak of m/z 231 with no peak of m/z 280 are characteristic of the head-to-tail structures ( Figures S10 and S11). In addition, since 2b was assigned as a synHT dimer by X-ray analysis, 4b should be an antiHT dimer.  The structures of the product dimers were determined from X-ray crystallographic analyses and the 1 H-NMR and MS spectra. The X-ray analyses were carried out for 2a, 2b·2HCl, 3b, and 4a, as single crystals suitable for X-ray analyses were obtained. Figure 4 clearly shows that 2a and 2b are synHT dimers, and 3b and 4a are synHH and antiHT dimers, respectively. Based on MS and 1 H-NMR analyses, 3a and 5a were assigned as synHH and antiHH dimers, respectively. In their MS spectra, the fragment peaks of m/z 231 and 280 were observed. The peak of m/z 231 corresponds to a naphthylethylpyridine moiety, and the peak of m/z 280 corresponds to a binaphthylethane moiety, indicating their head-to-head structures ( Figures S8, S9, S12, and S13). The methine protons at δ 4.67 and 4.69 for 3a and δ 3.88 and 3.97 for 5a are very close to those of related synHH and antiHH dimers, respectively [18]. The assignment of 4b was also performed by MS spectra; the observations of a base peak of m/z 231 with no peak of m/z 280 are characteristic of the head-to-tail structures (Figures S10 and S11). In addition, since 2b was assigned as a synHT dimer by X-ray analysis, 4b should be an antiHT dimer. The structures of the product dimers were determined from X-ray crystallographic analyses and the 1 H-NMR and MS spectra. The X-ray analyses were carried out for 2a, 2b·2HCl, 3b, and 4a, as single crystals suitable for X-ray analyses were obtained. Figure 4 clearly shows that 2a and 2b are synHT dimers, and 3b and 4a are synHH and antiHT dimers, respectively. Based on MS and 1 H-NMR analyses, 3a and 5a were assigned as synHH and antiHH dimers, respectively. In their MS spectra, the fragment peaks of m/z 231 and 280 were observed. The peak of m/z 231 corresponds to a naphthylethylpyridine moiety, and the peak of m/z 280 corresponds to a binaphthylethane moiety, indicating their head-to-head structures ( Figures S8, S9, S12, and S13). The methine protons at δ 4.67 and 4.69 for 3a and δ 3.88 and 3.97 for 5a are very close to those of related synHH and antiHH dimers, respectively [18]. The assignment of 4b was also performed by MS spectra; the observations of a base peak of m/z 231 with no peak of m/z 280 are characteristic of the head-to-tail structures ( Figures S10 and S11). In addition, since 2b was assigned as a synHT dimer by X-ray analysis, 4b should be an antiHT dimer.  To obtain mechanistic insights into the dimerization process, the dependence of product distribution on the irradiation time was investigated. Figure 5 shows a plots of the yields of synHT dimer 2a vs. irradiation time. The yield of 2a increased with increased irradiation time, with the curve following second-order kinetics. This suggests that the dimerization process does not involve any other processes, such as a reverse reaction or isomerization reaction. The second-order kinetics was confirmed by plots of t vs. 1/[1a], in which a linear correlation was observed ( Figure S14). To obtain mechanistic insights into the dimerization process, the dependence of product distribution on the irradiation time was investigated. Figure 5 shows a plots of the yields of synHT dimer 2a vs. irradiation time. The yield of 2a increased with increased irradiation time, with the curve following second-order kinetics. This suggests that the dimerization process does not involve any other processes, such as a reverse reaction or isomerization reaction. The second-order kinetics was confirmed by plots of t vs. 1/[1a], in which a linear correlation was observed ( Figure S14). The preference for the formation of the synHT dimer in the presence of HCl agrees with that in the photodimerization of styrylpyridine [18]. This synHT preference indicates that the intermediate naphthylvinylpyridinium salts form preorganized head-to-tail molecular dimers through intermolecular cation-π interactions, as shown in Scheme 1. The fact that the selectivities observed in the photodimerization of 1a and 1b are much higher than that in the case of styrylpyridine (71% selectivity) [18] suggests a significant effect of the naphthyl group on the intermolecular cation-π interactions, regardless of the differences in the π-conjugate systems. It has been reported that the larger aromatic π-conjugate systems are more effective for cation-π interactions. For example, the threshold collision-induced dissociation technique clarified that extended π-network leads to an increase in the strength of the cation-π interaction due to increased polarizability of the ligand [23]. The B3LYP calculations for the binding energy of Na + ···naphthalene predicted that the energy is larger than that of Na + ···benzene [24]. These experimental and theoretical studies agree with the higher selectivities observed in the photodimerization of 1a and 1b than that of styrylpyridine.

General Methods
Melting points were determined with a Yanaco model MP microscope (Yanaco, Tokyo, Japan). Column chromatography was carried out using silica gel 60 N (Kanto Chemical, Tokyo, Japan). Thin The preference for the formation of the synHT dimer in the presence of HCl agrees with that in the photodimerization of styrylpyridine [18]. This synHT preference indicates that the intermediate naphthylvinylpyridinium salts form preorganized head-to-tail molecular dimers through intermolecular cation-π interactions, as shown in Scheme 1. The fact that the selectivities observed in the photodimerization of 1a and 1b are much higher than that in the case of styrylpyridine (71% selectivity) [18] suggests a significant effect of the naphthyl group on the intermolecular cation-π interactions, regardless of the differences in the π-conjugate systems. To obtain mechanistic insights into the dimerization process, the dependence of product distribution on the irradiation time was investigated. Figure 5 shows a plots of the yields of synHT dimer 2a vs. irradiation time. The yield of 2a increased with increased irradiation time, with the curve following second-order kinetics. This suggests that the dimerization process does not involve any other processes, such as a reverse reaction or isomerization reaction. The second-order kinetics was confirmed by plots of t vs. 1/[1a], in which a linear correlation was observed ( Figure S14). The preference for the formation of the synHT dimer in the presence of HCl agrees with that in the photodimerization of styrylpyridine [18]. This synHT preference indicates that the intermediate naphthylvinylpyridinium salts form preorganized head-to-tail molecular dimers through intermolecular cation-π interactions, as shown in Scheme 1. The fact that the selectivities observed in the photodimerization of 1a and 1b are much higher than that in the case of styrylpyridine (71% selectivity) [18] suggests a significant effect of the naphthyl group on the intermolecular cation-π interactions, regardless of the differences in the π-conjugate systems. It has been reported that the larger aromatic π-conjugate systems are more effective for cation-π interactions. For example, the threshold collision-induced dissociation technique clarified that extended π-network leads to an increase in the strength of the cation-π interaction due to increased polarizability of the ligand [23]. The B3LYP calculations for the binding energy of Na + ···naphthalene predicted that the energy is larger than that of Na + ···benzene [24]. These experimental and theoretical studies agree with the higher selectivities observed in the photodimerization of 1a and 1b than that of styrylpyridine.

General Methods
Melting points were determined with a Yanaco model MP microscope (Yanaco, Tokyo, Japan). Column chromatography was carried out using silica gel 60 N (Kanto Chemical, Tokyo, Japan). Thin It has been reported that the larger aromatic π-conjugate systems are more effective for cation-π interactions. For example, the threshold collision-induced dissociation technique clarified that extended π-network leads to an increase in the strength of the cation-π interaction due to increased polarizability of the ligand [23]. The B3LYP calculations for the binding energy of Na + ···naphthalene predicted that the energy is larger than that of Na + ···benzene [24]. These experimental and theoretical studies agree with the higher selectivities observed in the photodimerization of 1a and 1b than that of styrylpyridine.

General Methods
Melting points were determined with a Yanaco model MP microscope (Yanaco, Tokyo, Japan). Column chromatography was carried out using silica gel 60 N (Kanto Chemical, Tokyo, Japan). Thin layer chromatography (TLC) was carried out on a Merck silica gel 60 PF 254 (Merck, Tokyo, Japan). IR spectra were obtained on FT/IR-410 spectrometer (JASCO, Tokyo, Japan) as neat films between NaCl plates, or KBr pellets. NMR spectra were recorded on JEOL EX-400 spectrometer (JEOL, Tokyo, Japan). 1 H-NMR spectra were obtained at 400 MHz as dilute solution in CDCl 3 , and the chemical shifts were reported relative to internal TMS. Low-resolution mass spectra were recorded at ionizing voltage of 70 eV by electron impact. High-resolution mass spectra were recorded on Exactive Orbitrap LC-MS (Thermo-Fischer, Yokohama, Japan). UV/vis spectra were recorded on JASCO V-650DS spectrometer (JASCO, Tokyo, Japan) in MeOH. UV irradiation was performed using a 450 W high-pressure mercury lamp (USHIO UM453, Tokyo, Japan). Visible-light irradiation was performed using a 250 W high-pressure mercury lamp (ASAHI SPECTRA REX-250, Tokyo, Japan) through quartz fibers with a UV-cut filter.

Representative Procedure for Irradiation of Naphthylvinylpyridines under Acidic Conditions
A solution of naphthylvinylpyridine (0.1 mmol) in 100 µL of MeOH containing conc. HCl was irradiated for 8 h through a Pyrex filter with a 450 W high-pressure mercury lamp. The solution was neutralized with sat. NaHCO 3 , and the solvent was evaporated to give an oil. This oil was extracted with CH 2 Cl 2 three times, and the combined organic layer was dried over anhydrous MgSO 4 . Evaporation of the solvent gave a crude product, which was subjected to column chromatography on silica gel with a 7:2:1 mixture of ethyl acetate, hexane, and methanol as an eluent. The product ratio was determined by the 1 H-NMR spectrum of the crude product. Visible-light irradiation of the above-mentioned solution was carried out with a 250 W high-pressure mercury lamp for 3 h through a UV-cut filter. The work-up followed the same procedure as that described above. generally form a stronger cation-π complex with a cation, such systems are useful not only in [2 + 2] photocycloaddition, but also in various cation-π-controlled organic reactions.