Remarkable acceleration of template-directed photodimerisation of 9-phenylethynylanthracene derivatives assisted by complementary salt bridge formation

Abstract

The photoirradiation of 9-phenylethynylanthracene in degassed chloroform and benzene afforded not only a [4 + 2]-anti Diels–Alder addition dimer, but also a [4 + 4]-anti-dimer as a minor product for the first time as revealed by single-crystal X-ray analysis, while the anthracene residue was quantitatively oxidised in undegassed dilute chloroform, giving the corresponding endoperoxides. The photochemical reactions of carboxylic acid monomers bearing a 9-phenylethynylanthracene unit at one and both ends were further investigated in the presence and absence of the complementary amidine dimer as the template. It was found that a similar photooxidation reaction of the monomers was significantly suppressed in the presence of the template even in undegassed chloroform. In addition, the template-directed photodimerisation of the mono- and di-9-phenylethynylanthracene-bound monomers was remarkably accelerated 30- or 61-fold in the degassed chloroform, giving the [4 + 2]-anti- and [4 + 4]-anti-dimers as major and minor products, respectively, whereas the di-9-phenylethynylanthracene-bound monomer was preferentially photo-polymerised in the absence of the template.

---

Introduction

Photochemical reactions often promote the formation of products inaccessible by thermal reactions, leading to one of the key reactions in the state-of-the-art organic synthesis.¹ However, it remains difficult to control the selectivity and specificity during the photochemical transformations in homogeneous solutions in a predictable way, thus producing a mixture of regioisomers along with stereoisomers. Therefore, the template-directed photoreaction has been developed not only to improve the efficiency, but also to control the regioselectivity and/or enantioselectivity.²

Among a variety of photoreactions, the photodimerisation of anthracene and its derivatives is one of the most well-known photochemical reactions and has been extensively investigated.³ The photodimers are connected by two covalent bonds resulting from the [4 + 4] cycloaddition and revert to anthracenes thermally or under UV irradiation by using light below 300 nm. Taking advantage of this feature, the anthracene skeleton has been successfully applied to two-dimensional polymers,⁴ supramolecular polymers,⁵ reversible cross-linking reactions,⁶ photochemical molecular switches⁷ and photoinduced shape-changeable materials.⁸ Another important synthetic feature during the anthracene photodimerisation is that substituted anthracenes at specific positions afford regio- and/or stereoisomers.³ For instance, the irradiation of 9-substituted anthracenes gives two regioisomeric dimers, the [4 + 4]-anti and -syn photodimers (Fig. 1a), and the thermodynamically stable [4 + 4]-anti dimer is, in general, produced as a major product, whereas the energetically unfavourable [4 + 4]-syn dimer is regioselectively formed in specific environments, such as within micelles.⁹ Supramolecular approaches using cyclic host molecules with a rigid concave cavity, such as cyclodextrins^5f,10 and cucurbit[n]urils,^{5d,e,6e,10f,h} capable of encapsulating two anthracene molecules in close proximity, and DNA as a template,¹¹ have also been developed to improve the reaction yield and to control the regio- and stereo-selectivities during the photodimerisation of anthracene and its derivatives.

Fig. 1 Photoreaction of (a) 9-substituted anthracene, (b) 9-phenylethynylanthracene (1_M) and (c) 9-(4-mercaptoalkylphenylethynyl)anthracenes (1_M-SH) and their photoproducts.

On the other hand, when exposed to light in the presence of oxygen, anthracene derivatives readily react with singlet oxygen to produce endoperoxides,¹² which eventually revert to the parent anthracenes and oxygen under thermolysis (Fig. 1a). However, anthracene can be effectively protected from photooxidation and dimerised in the cavity of cyclodextrin^5f,10 and in the crystal state^12e even in the presence of oxygen.

Becker and Andersson reported that the irradiation of 9-phenylethynylanthracene (1_M) in solution readily promoted a [4 + 2] Diels–Alder addition reaction between the ethynyl moiety and the central ring of the other anthracene residue of 1_M over the [4 + 4] cycloaddition reaction, predominantly producing a [4 + 2]-anti dimer ([4 + 2]-anti-2_MM, Fig. 1b).¹³ In contrast, Weiss and co-workers demonstrated the energetically unfavourable [4 + 4]-syn dimer formation of thiolated 9-phenylethynylanthracenes (1_M-SH) (Fig. 1c) upon irradiation that proceeded in a highly regioselective manner when 1_M-SH was mixed in self-assembled alkane thiolate monolayers on flat Au surfaces on which the anthracene moieties were favourably arranged in such a way that the intermolecular [4 + 4]-syn cycloaddition ([4 + 4]-syn-2_MM) could be possible.¹⁴ The regioselective [4 + 4]-syn dimer formation was monitored by scanning tunneling microscopy (STM)^14a and surface-enhanced Raman spectroscopy (SERS),^14b thus showing a significant decrease in conductivity and disappearance of the peaks due to the anthracene moiety, respectively. Recently, Klajn and co-workers also demonstrated that 1_M-SH immobilised on metallic (Au and Pd) nanoparticles dimerised in a regioselective way to yield a [4 + 4]-syn cycloaddition product under photoirradiation (Fig. 1c), based on the time-dependent absorption spectral changes.¹⁵

We previously reported a series of m-terphenyl-based complementary double helices that could be rationally designed and synthesised based on a modular strategy we developed using amidinium–carboxylate salt bridges through which the double helices are stabilised by double hydrogen bonds with a well-defined directionality even in polar solvents.¹⁶ Therefore, various types of functional linkers, such as Pt(II)-acetylide^16b,e,j,l,n and azobenzene linkages,^16d,i,m can be introduced between the m-terphenyl units while maintaining the double-helical structures.¹⁶ The complementary double-helical framework stabilised by salt bridges has recently been applied to the template-directed synthesis of complementary double helices through imine-bond forming reactions between a carboxylic acid and amidine monomers bearing either a formyl or an amino group at one end.^16i,m The complementary dimeric templates linked by an azobenzene residue significantly accelerated the imine-bond forming reactions. However, the reactions along the complementary dimer strands as a template were limited to imine-bond forming reactions.

In this work, we synthesised new carboxylic acid monomers (1_C and 3_C) bearing an anthracene group at one and both ends (Fig. 2a), respectively, and investigated their template-directed photodimerisation reactions in solution in the absence and presence of the complementary amidine dimer (T_AA) connected by a p-diethynylbenzene unit as the template (Fig. 2d). The monomers 1_C and 3_C possess one or two 9-phenylethynylanthracene moieties identical to 1_M, hence we anticipated the regioselective photodimerisation or polymerisation of 1_C or 3_C, respectively, giving products with a specific regioselectivity among the following four possible regioisomers ([4 + 2]-anti-, [4 + 2]-syn-, [4 + 4]-anti- and [4 + 4]-syn-isomers, Fig. 2b and c) along with acceleration of the photoreactions in the presence of the rigid template T_AA through the salt bridges, which permit the 9-phenylethynylanthracene moieties to be arranged in close proximity (Fig. 2d).

Fig. 2 (a) Chemical structures of 9-phenylethynylanthracene-bound carboxylic acid monomers (1_C and 3_C), monomeric amidine (A) and dimeric amidine template (T_AA). (b and c) Possible photoreactions (photooxidation and cycloaddition reactions) of the carboxylic acid monomers ((b) 1_C and (c) 3_C) and their photoproducts. (d) Schematic illustration for the template-directed photodimerisation of the 9-phenylethynylanthracene-bound monomers (1_C and 3_C).

For comparison, the photochemical dimerisation of 1_M (Fig. 1b),¹³ a model compound of 1_C and 3_C, was also thoroughly investigated under various conditions. We found that a [4 + 4]-anti-dimer ([4 + 4]-anti-2_MM) was also produced as a minor product along with the major Diels–Alder addition product, [4 + 2]-anti-2_MM, (Fig. 1b) and then their structures were unambiguously determined by single-crystal X-ray analysis.

Results and discussion

Synthesis

9-Phenylethynylanthracene (1_M) was prepared according to the reported method.¹⁷ The achiral carboxylic acid monomers (1_C and 3_C) bearing an anthracene group at one and both ends were newly synthesised according to Schemes S3 and S5,† respectively, (see the ESI†). The p-phenylene-linked optically active amidine dimer (T_AA)^16g and its monomeric amidine (A)^16a were also prepared according to the reported methods.

Photoreactions of model monomer (1_M)

The photoreaction of 1_M (0.50 mM) was first investigated in undegassed CDCl₃ at 25 °C (run 1 in Table 1) upon light irradiation over 400 nm. Reaction progress was monitored by ¹H NMR spectroscopy. The peak intensity of the anthracene proton (H_a) at the 10-position in 1_M decreased with time, whereas a new singlet signal appeared in a relatively low magnetic field of 6.08 ppm, which can be assigned to the bridge-head proton (H_b) of endoperoxide (1_M-O₂) and its peak intensity gradually increased as the reaction progressed, reaching 89% after irradiation for 3 min (Fig. 3a and S6a†). This assignment was supported by a molecular ionic peak at m/z = 309.19 ([1_M-O₂–H]⁻) in its negative-mode electron-spray ionisation (ESI) mass spectrum of the reaction mixture and also by the absorption spectrum of 1_M after irradiation for 3 min, in which the peaks due to the anthracene moiety almost disappeared (Fig. S6†). It was noted that the obtained 1_M-O₂ was not stable in solution and was thermally converted back to 1_M without decomposition at 25 °C.^12b,d,f Thus, the rate constant for the retro-photooxidation reaction was estimated to be 4.7 × 10⁻⁶ s⁻¹ on the basis of the time-dependent ¹H NMR spectral changes (Fig. S7†).

Fig. 3 Partial ¹H NMR (500 MHz, 25 °C) spectra of 1_M (0.50 mM) in (a) (run 1 in Table 1) undegassed and (b) (run 5 in Table 1) degassed CDCl₃ before (top) and after (bottom) irradiation of light (>400 nm). Full-scale spectra are shown in Fig. S6a and S11a,† respectively.

Table 1 Results of photoreactions of 1_M under various reaction conditions

Run	1M			Product yield (%) (Reaction rate 10^−3k (s^−1))				Sup. Fig. no.
	Solvent (conc. (mM))	Irradiation time (min)	Conv. (%) (consumption rate 10^−3k (s^−1))	1M-O2	[4 + 2]-anti-2MM	[4 + 4]-anti-2MM	[4 + 2]/[4 + 4]
a The maximum yield of 1M-O2 was 18% after 8 min irradiation of light (>400 nm) (see Fig. S8a). b The maximum yield of 1M-O2 was 38% after 8 min irradiation of light (>400 nm) (see Fig. S9a). c The maximum yield of 1M-O2 was 15% after 8 min irradiation of light (>400 nm) (see Fig. S10a). d Not available in ref. 13.
1	Undegassed CDCl3 (0.50)	3	89 (16)	89 (16)	—	—	—	Fig. S6
2	Undegassed CDCl3 (8.0)	30	88 (2.0)	12^a (0.71)	60 (0.93)	14 (0.14)	4.3	Fig. S8
3	Undegassed benzene-d6 (0.50)	16	88 (3.2)	32^b (1.0)	42 (0.96)	8 (0.14)	5.3	Fig. S9
4	Undegassed benzene-d6 (8.0)	30	96 (1.9)	9^c (0.37)	70 (1.0)	12 (0.13)	5.8	Fig. S10
5	Degassed CDCl3 (0.50)	30	88 (1.7)	—	64 (1.1)	23 (0.21)	2.8	Fig. S11
6	Degassed CDCl3 (8.0)	45	95 (1.5)	—	77 (1.0)	18 (0.15)	4.3	Fig. S12
7	Degassed benzene-d6 (0.50)	30	90 (1.6)	—	74 (1.1)	15 (0.16)	4.9	Fig. S13
8	Degassed benzene-d6 (8.0)	45	97 (1.3)	—	80 (1.0)	15 (0.13)	5.3	Fig. S14
9	Degassed benzene (8.0)	90	—^d	—^d	75	8	9.4	Ref. 13

---

In contrast, after light irradiation of 1_M (0.50 mM) for 30 min in degassed CDCl₃ at 25 °C (run 5 in Table 1), the ¹H NMR spectrum showed two new sets of singlet peaks due to the bridge-head protons (H′_b and H′′_b) along with a new anthracene proton (H_a′), indicating the formation of two photodimers (Fig. 3b and S11a†), which were unambiguously identified as [4 + 2]-anti-2_MM and [4 + 4]-anti-2_MM, respectively, by X-ray crystallographic analyses (Fig. 4) together with 2D NMR spectroscopy (Fig. S41 and S44†). Becker and Andersson reported that the photoreaction of 1_M (8.0 mM) in degassed benzene gave [4 + 2]-anti-2_MM in ca. 75% yield together with a by-product (ca. 8%) which was assumed to be an intermolecular [4 + 4] cycloadduct in spite of no structural information (run 9 in Table 1).¹³ Thus, we examined the photoreaction of 1_M (8.0 mM) under almost the same conditions in degassed benzene-d₆, affording [4 + 2]-anti-2_MM (80%) and [4 + 4]-anti-2_MM (15%) after 45 min irradiation (run 8 in Table 1 and Fig. S14†). The regioisomer ratio (5.3) was different from that reported by Becker and Andersson (9.4), but the difference may not be important.

Fig. 4 The crystal structures of (a) [4 + 2]-anti-2_MM (CCDC 1503775) and (b) [4 + 4]-anti-2_MM (CCDC 1503774).

The effects of oxygen, solvent (CDCl₃ and benzene-d₆) and concentration (0.50 and 8.0 mM) on the photooxidation and regioselectivity ([4 + 2]/[4 + 4]) during the photoreaction of 1_M were further investigated at 25 °C and the results are summarised in Table 1. In dilute undegassed solution, the photooxidation of 1_M preferentially took place more than that in the concentrated solution. In particular, irradiation of a dilute solution of 1_M (0.50 mM) in undegassed CDCl₃ (run 1 in Table 1) quantitatively produced 1_M-O₂ due to the heavy-atom effect of the solvent, which inhibits the photodimerisation by promoting intersystem crossing.^12c Although the concentration and solvent effects of the photodimerisation on regioselectivity were not significant, the yield of the major product, [4 + 2]-anti-2_MM, tended to increase with the increasing concentration of 1_M from 0.50 to 8.0 mM on changing the solvent from CDCl₃ to benzene-d₆.

We then calculated the structures of four possible photodimers (Fig. 1b) using the density functional theory (DFT) and found that their stabilities decrease in the following order: [4 + 2]-anti-2_MM > [4 + 2]-syn-2_MM > [4 + 4]-anti-2_MM > [4 + 4]-syn-2_MM (Fig. S5†), thus supporting the experimental results that the major product, [4 + 2]-anti-2_MM, is much more stable than the minor one, [4 + 4]-anti-2_MM, by 170.6 kJ mol⁻¹, whereas the corresponding syn-photodimers ([4 + 2]-syn-2_MM and [4 + 4]-syn-2_MM) could not be observed at all in the ¹H NMR spectra under the present conditions even though the syn-Diels–Alder adduct, [4 + 2]-syn-2_MM, is the second most stable photodimer. The reason is not clear, but it was suggested that it was due to the more favourable centrosymmetric-oriented complex formation ((1_M)₂), which could generate [4 + 2]-syn-2_MM,¹³ or these may be thermally unstable and immediately revert to the parent 1_M in solution.

On the other hand, an energetically unfavourable [4 + 4]-syn-2_MM derivative was reported to form in a highly regioselective fashion on self-assembled flat metal surfaces upon photoirradiation (Fig. 1c) as revealed by STM,^14a Raman^14b and the time-dependent absorption spectral changes.¹⁵

Fig. S15† shows the experimental (top) and DFT simulated (bottom) Raman spectra of 1_M, 1_M-O₂, [4 + 2]-anti-2_MM and [4 + 4]-anti-2_MM, which are in good agreement, although 1_M-O₂ contains 1_M generated from 1_M-O₂via thermolysis (Fig. S15b†).

The peaks due to the anthracene residue of 1_M completely disappeared in the Raman spectrum of [4 + 4]-anti-2_MM (Fig. S15d†), while these remained in the Raman spectrum of [4 + 2]-anti-2_MM (Fig. S15c†). Similar distinct spectral changes were also observed in the absorption spectra of 1_M, [4 + 2]-anti-2_MM and [4 + 4]-anti-2_MM (Fig. S2a†). We note, however, that the Raman spectra of 1_M-O₂ and [4 + 4]-anti-2_MM, in which an anthracene moiety no longer exists, were quite similar to each other except for a weak peak at 921 cm⁻¹ observed for 1_M-O₂, which can be assigned to the –O–O– bond vibration (Fig. S15b†).¹⁸ These results combined with the absorption spectra of 1_M-O₂ and [4 + 4]-anti-2_MM (Fig. S2a†) suggested that it might be difficult to assign the photochemical reaction products of 1_M and its derivatives, 1_M-O₂ or [4 + 4]-anti-2_MM, based on the Raman^14b and absorption spectroscopies.¹⁵ However, it is apparent that the photooxidation is protected in the absence of oxygen.

Template effects on the photoreactions of mono-9-phenylethynylanthracene-bound carboxylic acid monomer

The mono-9-phenylethynylanthracene-bound carboxylic acid monomer (1_C) used in this study contains a 9-phenylethynylanthracene moiety identical to 1_M, probably showing a similar photoreactivity to 1_M in the absence of the template. Based on the photoreaction results of the model monomer 1_M (Table 1), we employed a dilute CDCl₃ (0.50 mM) solution throughout the following photoreactions. As anticipated, the irradiation of 1_C in undegassed CDCl₃ at 25 °C (run 1 in Table 2) resulted in the formation of endoperoxide (1_C-O₂)¹⁹ in 89% yield after 3 min as evidenced by the bridge-head proton (H_b) that newly appeared as a singlet at 6.08 ppm (Fig. 5a and S16a†); the chemical shift was very similar to that of 1_M-O₂ (Fig. 3a and S6a†). The structure of 1_C-O₂ was characterised and identified by comparing the ¹H NMR spectrum of 1_C-O₂ with that of 1_M-O₂ (Fig. S1†) and ESI-mass measurements (Fig. S16c†).

Fig. 5 Partial ¹H NMR (500 MHz, undegassed CDCl₃, 25 °C) spectra of 1_C (0.50 mM) in the absence (run 1 in Table 2) (a) and presence of A (0.50 mM) (run 2 in Table 2) (b) and T_AA (0.25 mM) (run 3 in Table 2) (c) before (top) and after (bottom) irradiation of light (>400 nm). Full-scale spectra are shown in Fig. S16a, S18a and S19a.†

Table 2 Results of photoreactions of 1_C

Run	1C				Product yield (%) (reaction rate 10^−3k (s^−1))				Sup. Fig. no.
	Additive (conc. (mM))	Solvent (conc. (mM))	Irradiation time (min)	Conv. (%) (consumption rate 10^−3k (s^−1))	1C-O2	[4 + 2]-anti-2CC	[4 + 4]-anti-2CC	[4 + 2]/[4 + 4]
a 1C-O2 formed during the initial stage was gradually converted into an unknown compound probably due to photolysis products of 1C-O2 upon further photoirradiation. The reaction rate was therefore estimated based on the time–conversion relationship during the initial stage. The maximum yield of 1C-O2 was 10% after 30 s irradiation of light (>400 nm) (see Fig. S19a). b The peaks for (A)2·[4 + 2]-anti-2CC were too broad to estimate its reaction rate (see Fig. S22a).
1	—	Undegassed CDCl3 (0.50)	3	89 (11)	89 (11)	—	—	—	Fig. S16
2	A (0.50)	Undegassed CDCl3 (0.50)	3	89 (10)	89 (10)	—	—	—	Fig. S18
3	TAA (0.25)	Undegassed CDCl3 (0.50)	2	92 (25)	10^a (4.4)	63 (11)	18 (2.5)	3.5	Fig. S19–20
4	—	Degassed CDCl3 (0.50)	60	85 (0.66)	—	60 (0.40)	25 (0.11)	2.4	Fig. S21
5	A (0.50)	Degassed CDCl3 (0.50)	90	68 (0.18)	—	19 (—)^b	17 (0.036)	1.1	Fig. S22–24
6	TAA (0.25)	Degassed CDCl3 (0.50)	2	89 (20)	—	68 (12)	20 (2.9)	3.4	Fig. S25–26

---

The photoreactions of 1_C (0.50 mM) in the presence of the template T_AA (0.25 mM) and its monomeric amidine A (0.50 mM)²⁰ were then investigated in undegassed CDCl₃ at 25 °C. The monomeric amidine A was used for the control experiment to evaluate the template effect of T_AA on the photoreaction. The ¹H NMR spectra of 1_C (0.50 mM) in the presence of A (0.50 mM) or T_AA (0.25 mM) showed the characteristic peaks for the NH protons in the low magnetic field at ca. 13.3 ppm, indicating the salt bridge formation (Fig. S18a and S19a†).

The irradiation of A·1_C (run 2 in Table 2) also produced only one set of ¹H NMR signals, such as H_b (6.09 ppm), resulting from the preferential formation of 1_C-O₂ complexed with A (Fig. 5b and S18a†) as observed in the photooxidation of 1_C in the absence of A (run 1 in Table 2). In sharp contrast, upon the irradiation of T_AA·(1_C)₂ (run 3 in Table 2), the peak intensity of H_a decreased and almost disappeared within 2 min, whereas two new sets of singlet peaks due to the bridge-head protons (H′_b and H′′_b) appeared at 6.21 and 5.10 ppm, respectively (Fig. 5c and S19a†). These peaks could be clearly identified as the complexes of T_AA·[4 + 2]-anti-2_CC (63%) and T_AA·[4 + 4]-anti-2_CC (18%) based on the ¹H NMR spectra of the 1 : 1 mixtures of the isolated [4 + 2]-anti-2_CC and [4 + 4]-anti-2_CC after purification with T_AA, respectively (Fig. S28†). Interestingly, the photooxidation reaction of 1_C was significantly protected in the presence of the template T_AA, affording 1_C-O₂ in less than 10% yield (Fig. S19a†) and the intermolecular photodimerisation selectively proceeded due to the close proximity of the two anthracene moieties arranged along the template, as supported by the decrease and red-shift of the absorption peaks of 1_C upon the addition of T_AA, indicative of the π-stacked arrangements of the anthracene moieties (Fig. S20†).

In the same way, the photoreactions of 1_C (0.50 mM) in the absence and presence of A (0.50 mM) or T_AA (0.25 mM) were performed in degassed CDCl₃ at 25 °C (runs 4–6 in Table 2), in which the photooxidation was completely prohibited as observed in the model reaction of 1_M under the same conditions. In the absence of A and T_AA, 1_C was gradually converted to the photodimers, producing [4 + 2]-anti-2_CC and [4 + 4]-anti-2_CC in 60 and 25% yields, respectively, after irradiation for 60 min (run 4 in Table 2 and Fig. 6a and S21a†).

Fig. 6 Partial ¹H NMR (500 MHz, degassed CDCl₃, 25 °C) spectra of 1_C (0.50 mM) in the absence (run 4 in Table 2) (a) and presence of A (0.50 mM) (run 5 in Table 2) (b) and T_AA (0.25 mM) (run 6 in Table 2) (c) before (top) and after (bottom) irradiation of light (>400 nm). * denotes the peak due to 1,1,2,2-tetrachloroethane used as an internal standard. Full-scale spectra are shown in Fig. S21a, S22a and S25a.† The peaks for (A)₂·[4 + 2]-anti-2_CC were too broad to estimate its yield. The yield of (A)₂·[4 + 2]-anti-2_CC was estimated to be 19% from the integral ratios of the peaks for isolated [4 + 4]-anti-2_CC and [4 + 2]-anti-2_CC (see Fig. S22b†).

Quite interestingly, the photodimerisation of 1_C in the presence of the template T_AA (run 6 in Table 2) took place much faster than that in the absence of T_AA, affording the T_AA·[4 + 2]-anti-2_CC (68%) and T_AA·[4 + 4]-anti-2_CC (20%) complexes after 2 min of light irradiation (Fig. 6c and S25a†).²¹ Contrary to our expectation, however, we could not observe a specific regioselectivity during the photodimerisation, producing photodimers with an almost similar regioselectivity ([4 + 2]/[4 + 4]) to that in the absence of the template T_AA.

To estimate the reaction rate constant (k) of 1_C, the conversions of 1_C were estimated from the integral ratios of the peaks for H_a (T_AA·(1_C)₂), H′′_b ([4 + 2]-anti-2_CC), H′′_b ([4 + 4]-anti-2_CC) and the internal standard (1,1,2,2-tetrachloroethane) based on the ¹H NMR spectral changes and were plotted versus the reaction time (Fig. 7, S21, S23 and S25†). The experimental data were then fitted to a first-order kinetic model using eqn (1),

ln(C/C₀) = −kt (1)

where C is the concentration of 1_C, and t is the reaction time. As shown in Fig. 7, the photodimerisation was significantly accelerated in the presence of T_AA probably because of the close proximity of the anthracene moiety of 1_C along the template (Fig. S26†). The k values in the presence and absence of T_AA were then estimated to be 20 × 10⁻³ s⁻¹ and 0.66 × 10⁻³ s⁻¹, respectively, by the least-squares curve fitting method as shown in Fig. 7. Thus, the photodimerisation of 1_C was remarkably accelerated 30-fold (k(T_AA·(1_C)₂)/k(1_C) = 30) in the presence of T_AA through the salt bridges, although the template-directed regioselectivity control during the photoreaction was not achieved.

Fig. 7 Kinetic plots of the photodimerisation of 1_C (0.50 mM) in degassed CDCl₃ at 25 °C in the absence (blue line) and presence of A (black line) and T_AA (red line). The reaction rates were estimated from the integral ratios of the peaks for H_a (1_C) and the internal standard (1,1,2,2-tetrachloroethane) based on the ¹H NMR spectral changes shown in Fig. S21a, S22a and S25a.†

In the presence of A, however, the photodimerisation (run 5 in Table 2) of 1_C took place very slowly with the k value of 0.18 × 10⁻³ s⁻¹ (Fig. 7 and S23†). The ¹H NMR spectrum of A·1_C after irradiation for 90 min was rather complicated except for the signals of (A)₂·[4 + 4]-anti-2_CC (17%), and those of (A)₂·[4 + 2]-anti-2_CC were not clearly observed because the signals were too broad (Fig. 6b and S22a†). The resulting photodimers with carboxylic acid residues were then isolated from the reaction mixture by silica gel column chromatography and the yield of (A)₂·[4 + 2]-anti-2_CC was estimated to be 19% based on the integral ratio ([4 + 2]-anti-2_CC/[4 + 4]-anti-2_CC) in the ¹H NMR spectrum (Fig. S22b†). Molecular mechanics (MM) calculations for the A·1_C salt-bridged complex suggested that one of the terminal TMS units of A is located close to the anthracene moiety of 1_C and a reactive acetylene residue of 1_C is likely sandwiched between the TMS and the phenyl group of the amidine residue of A (Fig. S24†). Thus, the photodimerisation of 1_C, in particular, the formation of [4 + 2]-anti-2_CC may be considerably retarded once complexed with A due to such steric effects, leading to the decrease in the reaction rate.

Template effects on the photoreactions of di-9-phenylethynylanthracene-bound carboxylic acid monomer

The photoreactions of the di-9-phenylethynylanthracene-bound carboxylic acid monomer (3_C) (0.50 mM) in the absence and presence of the template T_AA (0.25 mM) were also investigated in undegassed and degassed CDCl₃ at 25 °C (Table 3). The irradiation of 3_C in undegassed CDCl₃ (run 1 in Table 3 and Fig. S29†) readily promoted the photooxidation to produce a mixture of mono- (3_C-O₂) and di-endoperoxides (3_C-2O₂) in 50% yield as identified by the ESI-mass (Fig. S29c†) and NMR analysis (Fig. S1†), which eventually reverted to the parent 3_C and further converted in part to unknown compounds by thermolysis (Fig. S30†). In the presence of T_AA, the photooxidation of 3_C was significantly suppressed as anticipated (12% yield, run 2 in Table 3)²² and the [4 + 2]- and [4 + 4]-anti-4_CC photodimers were mainly produced in 60 and 17% yields, respectively, on the basis of the ¹H NMR spectra of the products by comparing them with those of the photodimers of 2_CC formed in the presence of T_AA (Fig. S37a–d†). We found that one of the terminal anthracene units of 4_CC complexed with T_AA was further oxidised (Fig. S31b†), resulting in the formation of the mono- and/or di-oxidised dimers, [4 + 2]-anti-4_CC-O₂ and [4 + 4]-anti-4_CC-O₂.

Table 3 Results of photoreactions of 3C

Run	3C				Products yield (%) (reaction rate 10^−3k (s^−1))						Sup. Fig. no.
	Additive (conc. (mM))	Solvent (conc. (mM))	Irradiation time (min)	Conv. (%) (consumption rate 10^−3k (s^−1))	poly-3C		3C-O2 and 3C-2O2	[4 + 2]-anti-4CC	[4 + 4]-anti-4CC	[4 + 2]/[4 + 4]
					[4 + 2]-anti	[4 + 4]-anti
a Total yield of mono- (3C-O2) and di-endoperoxides (3C-2O2) of 3C, which were gradually converted into unknown compounds probably due to photolysis upon further photoirradiation (see Fig. S30). b One of the terminal anthracene units of [4 + 2]- and [4 + 4]-anti-4CC was oxidised (see Fig. S31). c The conversion of 3C and its consumption rate were estimated based on the formation of poly-3C using ^1H NMR (see Fig. S35a). d The number-average molecular weight (Mn) was determined by SEC using polystyrene standards in THF containing 0.1 wt% TBAB as the eluent (see Fig. 8B(a)). e It was difficult to estimate the yields of [4 + 2]- and [4 + 4]-anti-4CC because their proton NMR signals were overlapped with those of poly-3C (see Fig. S35a).
1	—	Undegassed CDCl3 (0.50)	3	82 (9.9)	—		50^a (4.2)	—	—	—	Fig. S29
2	TAA (0.25)	Undegassed CDCl3 (0.50)	2	97 (38)	—		12 (1.4)	60^b (19)	17^b (3.6)	3.5	Fig. S31–34
3	—	Degassed CDCl3 (0.50)	60	22^c (0.61)^c	72^dMn = 3.2 × 10^3		—	—^e	—^e	3.0	Fig. S35
					54 (0.39)	18 (0.010)
4	TAA (0.25)	Degassed CDCl3 (0.50)	2	95 (37)	—		—	74 (23)	20 (4)	3.7	Fig. S36–38

---

In strong contrast, the ¹H NMR spectrum of 3_C (0.50 mM) after irradiation for 60 min in degassed CDCl₃ (run 3 in Table 3) became significantly broadened, giving signals due to the [4 + 2]- and [4 + 4]-anti dimerisations (H′_b and H′′_b), indicating the formation of a random copolymer (poly-3_C) composed of [4 + 2]- and [4 + 4]-anti units (Fig. 8A(a) and S35a†). The size-exclusion chromatography (SEC) analysis further supported the polymerisation of 3_C (Fig. 8B(a)).²³ Interestingly, the photoreaction of T_AA·(3_C)₂ in degassed CDCl₃ (run 4 in Table 3) showed two sharp bridge-head peaks (H′_b and H′′_b), suggesting the formation of the photodimers, T_AA·[4 + 2]-anti-4_CC and T_AA·[4 + 4]-anti-4_CC, while the other anthracene moieties remained unreacted as evidenced by the ¹H NMR and absorption spectra (Fig. 8A(b), S36a and S38†). The SEC analysis of the products further supported this conclusion (Fig. 8B(b–d)).

Fig. 8 (A) Partial ¹H NMR (500 MHz, degassed CDCl₃, 25 °C) spectra of 3_C (0.50 mM) in the absence (run 3 in Table 3) (a) and presence of T_AA (run 4 in Table 3) (b) before (top) and after (bottom) irradiation of light (>400 nm). * denotes the peak due to 1,1,2,2-tetrachloroethane used as an internal standard. Full-scale spectra are shown in Fig. S35a and S36a.† (B) SEC chromatograms of (a) poly-3_C produced in a degassed CDCl₃ solution of 3_C (0.50 mM) upon irradiation with light (>400 nm) for 60 min (see Fig. S35†), (b) isolated carboxylic acid dimers obtained from T_AA·(3_C)₂ (0.25 mM) after irradiation of light (>400 nm) for 120 s in degassed CDCl₃ (see Fig. S37g†), (c) dimer [4 + 2]-anti-2_CC, and (d) monomer 3_C. These carboxylic acids were converted into the corresponding methyl esters by treatment with (trimethylsilyl)diazomethane before SEC analysis.

These results indicated that the dimeric template T_AA selectively promoted the photodimerisation of the anthracene units of the bimolecular 3_C located in close proximity at the center (Fig. S38†), whereas the further polymerisation of the resultant T_AA·[4 + 2]-anti-4_CC and T_AA·[4 + 4]-anti-4_CC dimers was exclusively protected due to the steric effects between the remaining anthracene units at the ends and the TMS and phenyl groups of the amidine residues of T_AA as observed in the photodimerisation of A·1_C (run 5 in Table 2).

In the same way, the k values of 3_C in degassed CDCl₃ in the presence and absence of T_AA were estimated to be 37 × 10⁻³ s⁻¹ and 0.61 × 10⁻³ s⁻¹, respectively, (Fig. S35d and S36c†). Thus, the photodimerisation of 3_C was accelerated 61-fold (k(T_AA·(3_C)₂)/k(3_C) = 61) in the presence of T_AA. Moreover, the dimerisation of 3_C was found to take place faster than that of 1_C in the presence of T_AA by a factor of 1.9 (k(T_AA·(3_C)₂)/k(T_AA·(1_C)₂) = 1.9). This indicated that 3_C can more effectively dimerise along the template than 1_C because 3_C possesses two anthracene moieties at both ends (Fig. S37e†).

Conclusions

In conclusion, we have performed a close inspection of the photochemical reactions of 9-phenylethynylanthracene under various reaction conditions and proved the [4 + 4]-anti and [4 + 2]-anti dimers structures produced during the photodimerisation¹³ by single-crystal X-ray analysis. We also found a remarkable template effect of the amidine dimer on the photoreactions of the mono- and di-9-phenylethynylanthracene-bound carboxylic acid monomers. The detailed kinetic and structural studies revealed that the [4 + 4]- and [4 + 2]-anti photodimers (2_CC and 4_CC) were selectively produced along the template and the reaction rate was accelerated 30 or 61-fold without side reactions, oxidation and polymerisation. The observed noticeable enhancement of the dimerisation rate constants could be ascribed to the two anthracene moieties of the monomers located in close proximity at the reaction site along the rigid template assisted by the amidinium–carboxylate salt bridges. Although the template-directed, specific regioselective-photodimerisation was not achieved, the present template-directed photodimerisation system based on salt bridges will be further applied to the template-directed enantioselective photodimerisation of prochiral anthracene derivatives^10,24 because the amidine dimer used in the present study is optically active.²⁵ Studies along this line are now underway in our laboratory.

Acknowledgements

This work was supported in part by JSPS KAKENHI (Grant-in-Aid for Scientific Research (S), No. 25220804 (E. Y.) and Grant-in-Aid for Young Scientists (B), No. 26810048 (D. T.)). J. T. expresses his thanks for the JSPS Research Fellowship for Young Scientists (No. 8886). The authors thank Professor Hiroki Iida for his help with the single crystal X-ray diffraction analysis.

Notes and references

1. (a) N. Hoffmann, Chem. Rev., 2008, 108, 1052–1103; (b) T. Bach and J. P. Hehn, Angew. Chem., Int. Ed., 2011, 50, 1000–1045; (c) N. Hoffmann, J. Photochem. Photobiol., C, 2014, 19, 1–19; (d) N. Hoffmann, Synthesis, 2016, 1782–1802.
2. (a) N. D. McClenaghan and D. M. Bassani, Int. J. Photoenergy, 2004, 6, 185–192; (b) J. Svoboda and B. König, Chem. Rev., 2006, 106, 5413–5430; (c) C. Yang, Chin. Chem. Lett., 2013, 24, 437–441; (d) B. Bibal, C. Mongin and D. M. Bassani, Chem. Soc. Rev., 2014, 43, 4179–4198; (e) N. Vallavoju and J. Sivaguru, Chem. Soc. Rev., 2014, 43, 4084–4101; (f) C. Yang and Y. Inoue, Chem. Soc. Rev., 2014, 43, 4123–4143; (g) R. Brimioulle, D. Lenhart, M. M. Maturi and T. Bach, Angew. Chem., Int. Ed., 2015, 54, 3872–3890; (h) V. Ramamurthy and J. Sivaguru, Chem. Rev., 2016, 116, 9914–9993.
3. (a) H.-D. Becker, Chem. Rev., 1993, 93, 145–172; (b) H. Bouas-Laurent, A. Castellan, J.-P. Desvergne and R. Lapouyade, Chem. Soc. Rev., 2000, 29, 43–55; (c) H. Bouas-Laurent, A. Castellan, J.-P. Desvergne and R. Lapouyade, Chem. Soc. Rev., 2001, 30, 248–263.
4. (a) P. Kissel, R. Erni, W. B. Schweizer, M. D. Rossell, B. T. King, T. Bauer, S. Götzinger, A. D. Schlüter and J. Sakamoto, Nat. Chem., 2012, 4, 287–291; (b) R. Bhola, P. Payamyar, D. J. Murray, B. Kumar, A. J. Teator, M. U. Schmidt, S. M. Hammer, A. Saha, J. Sakamoto, A. D. Schlüter and B. T. King, J. Am. Chem. Soc., 2013, 135, 14134–14141; (c) P. Kissel, D. J. Murray, W. J. Wulftange, V. J. Catalano and B. T. King, Nat. Chem., 2014, 6, 774–778; (d) M. J. Kory, M. Wörle, T. Weber, P. Payamyar, S. W. van de Poll, J. Dshemuchadse, N. Trapp and A. D. Schlüter, Nat. Chem., 2014, 6, 779–784; (e) P. Payamyar, K. Kaja, C. Ruiz-Vargas, A. Stemmer, D. J. Murray, C. J. Johnson, B. T. King, F. Schiffmann, J. VandeVondele, A. Renn, S. Götzinger, P. Ceroni, A. Schütz, L.-T. Lee, Z. Zheng, J. Sakamoto and A. D. Schlüter, Adv. Mater., 2014, 26, 2052–2058; (f) D. J. Murray, D. D. Patterson, P. Payamyar, R. Bhola, W. Song, M. Lackinger, A. D. Schlüter and B. T. King, J. Am. Chem. Soc., 2015, 137, 3450–3453; (g) P. Payamyar, M. Servalli, T. Hungerland, A. P. Schütz, Z. Zheng, A. Borgschulte and A. D. Schlüter, Macromol. Rapid Commun., 2015, 36, 151–158.
5. (a) J.-F. Xu, Y.-Z. Chen, L.-Z. Wu, C.-H. Tung and Q.-Z. Yang, Org. Lett., 2013, 15, 6148–6151; (b) P. Wang, J. Hu, S. Yang, B. Song and Q. Wang, Chem. – Asian J., 2014, 9, 2880–2884; (c) P. Wei, X. Yan and F. Huang, Chem. Commun., 2014, 50, 14105–14108; (d) Z. Ji, Y. Li, Y. Ding, G. Chen and M. Jiang, Polym. Chem., 2015, 6, 6880–6884; (e) Z. Yu, J. Zhang, R. J. Coulston, R. M. Parker, F. Biedermann, X. Liu, O. A. Scherman and C. Abell, Chem. Sci., 2015, 6, 4929–4933; (f) X. Zhang, Y. Gao, Y. Lin, J. Hu and Y. Ju, Polym. Chem., 2015, 6, 4162–4166.
6. (a) Y. Zheng, M. Micic, S. V. Mello, M. Mabrouki, F. M. Andreopoulos, V. Konka, S. M. Pham and R. M. Leblanc, Macromolecules, 2002, 35, 5228–5234; (b) L. A. Connal, R. Vestberg, C. J. Hawker and G. G. Qiao, Adv. Funct. Mater., 2008, 18, 3315–3322; (c) P. Froimowicz, H. Frey and K. Landfester, Macromol. Rapid Commun., 2011, 32, 468–473; (d) L. A. Wells, M. A. Brook and H. Sheardown, Macromol. Biosci., 2011, 11, 988–998; (e) F. Biedermann, I. Ross and O. A. Scherman, Polym. Chem., 2014, 5, 5375–5382; (f) L. López-Vilanova, I. Martinez, T. Corrales and F. Catalina, Eur. Polym. J., 2014, 56, 69–76; (g) H. Xie, C.-Y. Cheng, L. Du, C.-J. Fan, X.-Y. Deng, K.-K. Yang and Y.-Z. Wang, Macromolecules, 2016, 49, 3845–3855; (h) H. Xie, M.-J. He, X.-Y. Deng, L. Du, C.-J. Fan, K.-K. Yang and Y.-Z. Wang, ACS Appl. Mater. Interfaces, 2016, 8, 9431–9439.
7. (a) F. Moriwaki, A. Ueno, T. Osa, F. Hamada and K. Murai, Chem. Lett., 1986, 1865–1868; (b) Y. Molard, D. M. Bassani, J.-P. Desvergne, P. N. Horton, M. B. Hursthouse and J. H. R. Tucker, Angew. Chem., Int. Ed., 2005, 44, 1072–1075; (c) C. Schäfer, R. Eckel, R. Ros, J. Mattay and D. Anselmetti, J. Am. Chem. Soc., 2007, 129, 1488–1489; (d) C.-K. Liang, G. V. Dubacheva, T. Buffeteau, D. Cavagnat, P. Hapiot, B. Fabre, J. H. R. Tucker and D. M. Bassani, Chem. – Eur. J., 2013, 19, 12748–12758; (e) H. Wang, F. Liu, R. C. Helgeson and K. N. Houk, Angew. Chem., Int. Ed., 2013, 52, 655–659.
8. (a) R. O. Al-Kaysi, A. M. Müller and C. J. Bardeen, J. Am. Chem. Soc., 2006, 128, 15938–15939; (b) R. O. Al-Kaysi and C. J. Bardeen, Adv. Mater., 2007, 19, 1276–1280; (c) J. T. Good, J. J. Burdett and C. J. Bardeen, Small, 2009, 5, 2902–2909; (d) L. Zhu, R. O. Al-Kaysi and C. J. Bardeen, J. Am. Chem. Soc., 2011, 133, 12569–12575; (e) T. Kim, L. Zhu, L. J. Mueller and C. J. Bardeen, J. Am. Chem. Soc., 2014, 136, 6617–6625.
9. (a) T. Wolff, J. Photochem., 1981, 16, 343–346; (b) T. Wolff and N. Müller, J. Photochem., 1983, 23, 131–140; (c) T. Wolff, N. Müller and G. von Bünau, J. Photochem., 1983, 22, 61–70; (d) A. Schütz and T. Wolff, J. Photochem. Photobiol., A, 1997, 109, 251–258.
10. (a) T. Tamaki, Chem. Lett., 1984, 13, 53–56; (b) T. Tamaki and T. Kokubu, J. Inclusion Phenom., 1984, 2, 815–822; (c) A. Ueno, F. Moriwaki, Y. Iwama, I. Suzuki, T. Osa, T. Ohta and S. Nozoe, J. Am. Chem. Soc., 1991, 113, 7034–7036; (d) A. Nakamura and Y. Inoue, J. Am. Chem. Soc., 2003, 125, 966–972; (e) C. Yang, A. Nakamura, G. Fukuhara, Y. Origane, T. Mori, T. Wada and Y. Inoue, J. Org. Chem., 2006, 71, 3126–3136; (f) C. Yang, T. Mori, Y. Origane, Y. H. Ko, N. Selvapalam, K. Kim and Y. Inoue, J. Am. Chem. Soc., 2008, 130, 8574–8575; (g) C. Ke, C. Yang, T. Mori, T. Wada, Y. Liu and Y. Inoue, Angew. Chem., Int. Ed., 2009, 48, 6675–6677; (h) C. Yang, C. Ke, W. Liang, G. Fukuhara, T. Mori, Y. Liu and Y. Inoue, J. Am. Chem. Soc., 2011, 133, 13786–13789; (i) J. Yao, Z. Yan, J. Ji, W. Wu, C. Yang, M. Nishijima, G. Fukuhara, T. Mori and Y. Inoue, J. Am. Chem. Soc., 2014, 136, 6916–6919.
11. T. Ihara, T. Fujii, M. Mukae, Y. Kitamura and A. Jyo, J. Am. Chem. Soc., 2004, 126, 8880–8881.
12. (a) N. Sugiyama, M. Iwata, M. Yoshioka, K. Yamada and H. Aoyama, Bull. Chem. Soc. Jpn., 1969, 42, 1377–1379; (b) N. J. Turro, M.-F. Chow and J. Rigaudy, J. Am. Chem. Soc., 1981, 103, 7218–7224; (c) N. Toshima, T. Sugano and H. Hirai, Can. J. Chem., 1984, 62, 2047–2053; (d) J.-M. Aubry, C. Pierlot, J. Rigaudy and R. Schmidt, Acc. Chem. Res., 2003, 36, 668–675; (e) E. Berni, C. Dolain, B. Kauffmann, J.-M. Léger, C. Zhan and I. Huc, J. Org. Chem., 2008, 73, 2687–2694; (f) W. Fudickar and T. Linker, J. Am. Chem. Soc., 2012, 134, 15071–15082.
13. H.-D. Becker and K. Andersson, J. Photochem., 1984, 26, 75–77.
14. (a) M. Kim, J. N. Hohman, Y. Cao, K. N. Houk, H. Ma, A. K.-Y. Jen and P. S. Weiss, Science, 2011, 331, 1312–1315; (b) Y. B. Zheng, J. L. Payton, T.-B. Song, B. K. Pathem, Y. Zhao, H. Ma, Y. Yang, L. Jensen, A. K.-Y. Jen and P. S. Weiss, Nano Lett., 2012, 12, 5362–5368.
15. T. Zdobinsky, P. S. Maiti and R. Klajn, J. Am. Chem. Soc., 2014, 136, 2711–2714.
16. (a) Y. Tanaka, H. Katagiri, Y. Furusho and E. Yashima, Angew. Chem., Int. Ed., 2005, 44, 3867–3870; (b) Y. Furusho, Y. Tanaka and E. Yashima, Org. Lett., 2006, 8, 2583–2586; (c) M. Ikeda, Y. Tanaka, T. Hasegawa, Y. Furusho and E. Yashima, J. Am. Chem. Soc., 2006, 128, 6806–6807; (d) Y. Furusho, Y. Tanaka, T. Maeda, M. Ikeda and E. Yashima, Chem. Commun., 2007, 3174–3176; (e) T. Hasegawa, Y. Furusho, H. Katagiri and E. Yashima, Angew. Chem., Int. Ed., 2007, 46, 5885–5888; (f) H. Ito, Y. Furusho, T. Hasegawa and E. Yashima, J. Am. Chem. Soc., 2008, 130, 14008–14015; (g) T. Maeda, Y. Furusho, S.-i. Sakurai, J. Kumaki, K. Okoshi and E. Yashima, J. Am. Chem. Soc., 2008, 130, 7938–7945; (h) H. Iida, M. Shimoyama, Y. Furusho and E. Yashima, J. Org. Chem., 2010, 75, 417–423; (i) H. Yamada, Y. Furusho, H. Ito and E. Yashima, Chem. Commun., 2010, 46, 3487–3489; (j) H. Ito, M. Ikeda, T. Hasegawa, Y. Furusho and E. Yashima, J. Am. Chem. Soc., 2011, 133, 3419–3432; (k) H. Yamada, Y. Furusho and E. Yashima, J. Am. Chem. Soc., 2012, 134, 7250–7253; (l) H. Yamada, Z.-Q. Wu, Y. Furusho and E. Yashima, J. Am. Chem. Soc., 2012, 134, 9506–9520; (m) J. Tanabe, D. Taura, H. Yamada, Y. Furusho and E. Yashima, Chem. Sci., 2013, 4, 2960–2966; (n) M. Horie, N. Ousaka, D. Taura and E. Yashima, Chem. Sci., 2015, 6, 714–723.
17. C.-W. Wan, A. Burghart, J. Chen, F. Bergstrom, L. B.-Å. Johansson, M. F. Wolford, T. G. Kim, M. R. Topp, R. M. Hochstrasser and K. Burgess, Chem. – Eur. J., 2003, 9, 4430–4441.
18. (a) Y. Shiraishi, S. Kanazawa, Y. Sugano, D. Tsukamoto, H. Sakamoto, S. Ichikawa and T. Hirai, ACS Catal., 2014, 4, 774–780; (b) H. Zhang, L.-H. Guo, L. Zhao, B. Wan and Y. Yang, J. Phys. Chem. Lett., 2015, 6, 958–963.
19. The photooxidised 1_C-O₂ gradually decomposed under shielded light, producing the monomer 1_C together with a small amount of an unknown compound (Fig. S17†).
20. The association constant (K_a) between 1_C and A could be roughly estimated to be ca. 3.5 × 10⁶ M⁻¹ in CDCl₃ by comparison of the K_a value between analogous amidine and carboxylic acids.¹⁶.
21. The template T_AA possesses four alkyne units that are not bonded to the anthracene residue. However, these alkyne units did not participate in the photoreactions because the conversion (89%) of 1_C after photoirradiation for 2 min in the presence of T_AA is almost consistent with the total yield (68 and 20%) of the resulting carboxylic acid photodimers ([4 + 2]-anti-2_CC and [4 + 4]-anti-2_CC) (run 6 in Table 2). In addition, the structures of the dimer products isolated were fully characterised, indicating that the photodimers were not composed of T_AA at all. Becker and Andersson also reported that the intermolecular photodimerisation between anthracene and diphenylacetylene did not take place, giving no [4 + 2]-cycloadduct upon light irradiation over 400 nm.¹³ This also suggests that the alkyne units of T_AA will not react with the anthracene unit(s) of 1_C and 3_C upon photoirradiation (>400 nm).
22. The peak assignments were performed by comparing the ¹H NMR spectrum of a mixture of 3_C-O₂ and 3_C-2O₂ in the presence of T_AA (Fig. S33†).
23. Poly-3_C was converted to the corresponding methyl esters by treatment with (trimethylsilyl)diazomethane. The number average molecular weight (M_n) and its distribution (M_w/M_n) were estimated to be 3.2 × 10³ and 1.5, respectively, by SEC using polystyrene standards and THF containing 0.1 wt% tetrabutylammonium bromide as the eluent. The M_n value corresponds to ca. 4 repeating units in a single polymer chain.
24. (a) Y. Ishida, Y. Kai, S.-y. Kato, A. Misawa, S. Amano, Y. Matsuoka and K. Saigo, Angew. Chem., Int. Ed., 2008, 47, 8241–8245; (b) A. Dawn, T. Shiraki, S. Haraguchi, H. Sato, K. Sada and S. Shinkai, Chem. – Eur. J., 2010, 16, 3676–3689; (c) Y. Ishida, A. S. Achalkumar, S.-y. Kato, Y. Kai, A. Misawa, Y. Hayashi, K. Yamada, Y. Matsuoka, M. Shiro and K. Saigo, J. Am. Chem. Soc., 2010, 132, 17435–17446; (d) D. Fuentealba, H. Kato, M. Nishijima, G. Fukuhara, T. Mori, Y. Inoue and C. Bohne, J. Am. Chem. Soc., 2013, 135, 203–209; (e) Y. Ishida, Y. Matsuoka, Y. Kai, K. Yamada, K. Nakagawa, T. Asahi and K. Saigo, J. Am. Chem. Soc., 2013, 135, 6407–6410; (f) Y. Kawanami, H. Umehara, J.-i. Mizoguchi, M. Nishijima, G. Fukuhara, C. Yang, T. Mori and Y. Inoue, J. Org. Chem., 2013, 78, 3073–3085; (g) G. Fukuhara, H. Umehara, S. Higashino, M. Nishijima, C. Yang, T. Mori, T. Wada and Y. Inoue, Photochem. Photobiol. Sci., 2014, 13, 162–171; (h) G. Fukuhara, K. Iida, Y. Kawanami, H. Tanaka, T. Mori and Y. Inoue, J. Am. Chem. Soc., 2015, 137, 15007–15014; (i) G. Fukuhara, K. Iida, T. Mori and Y. Inoue, J. Photochem. Photobiol., A, 2016, 331, 76–83; (j) M. M. Maturi, G. Fukuhara, K. Tanaka, Y. Kawanami, T. Mori, Y. Inoue and T. Bach, Chem. Commun., 2016, 52, 1032–1035.
25. The complexes of the achiral 1_C and 3_C with the optically active amidine template T_AA exhibited apparent Cotton effects in the absorption region of the anthracene residues (ca., 350–440 nm) accompanied by a decrease and red-shift in their absorption spectra (Fig. S20, S26, S34 and S38†), indicating that the two anthracene units of the monomers appear to be arranged in a chiral fashion along the chiral template. These results seem to be promising for realising such a template-directed enantioselective photodimerisation.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details and additional spectroscopic data. CCDC 1503774 and 1503775. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ob02087a

---