Dictating Photoreactivity through Restricted Bond Rotations: Cross

Sep 8, 2014 - Elango Kumarasamy , Anoklase Jean-Luc Ayitou , Nandini Vallavoju , Ramya Raghunathan , Akila Iyer , Anthony Clay , Sunil Kumar Kandappa ...
1 downloads 0 Views 1MB Size
Article pubs.acs.org/JPCA

Dictating Photoreactivity through Restricted Bond Rotations: CrossPhotoaddition of Atropisomeric Acrylimide Derivatives under UV/ Visible-Light Irradiation Akila Iyer,† Steffen Jockusch,*,‡ and J. Sivaguru*,† †

Department of Chemistry and Biochemistry, North Dakota State University, Fargo, North Dakota 58108, United States Department of Chemistry, Columbia University, New York, New York 10027, United States



S Supporting Information *

ABSTRACT: Nonbiaryl atropisomeric acrylimides underwent facile [2 + 2] photocycloaddition leading to cross-cyclobutane adducts with very high stereospecificity (enantiomeric excess (ee): 99% and diastereomeric excess (de): 99%). The photoreactions proceeded smoothly in isotropic media for both direct and triplet sensitized irradiations. The reactions were also found to be very efficient in the solid state where the same cross-cyclobutane adduct was observed. Photophysical studies enabled us to understand the excited-state photochemistry of acrylimides. The triplet energy was found to be ∼63 kcal/mol. The reactions proceeded predominantly via a singlet excited state upon direct irradiation with very poor intersystem crossing that was ascertained by quantification of the generated singlet oxygen. The reactions progressed smoothly with triplet sensitization with UV or visible-light irradiations. Laser flash photolysis experiments established the triplet transient of atropisomeric acrylimides with a triplet lifetime at room temperature of ∼40 ns.

1. INTRODUCTION Controlling light-induced transformations requires an intricate understanding of excited-state properties of photoreactive chromophores.1−3 The cross-[2 + 2]-photocycloaddition involving olefins is among the most intriguing and often less investigated photoreactions.4−6 This is because the straight [2 + 2]-photocycloaddition product is generally preferred in solution and has extensive literature precedence with various [2 + 2]photocycloaddition reactions.7,8 We were interested in accessing the cross-photocyclized product because it not only gives access to structurally unique motifs but also involves unique features in the reacting chromophores that possess excited-state properties, providing insights into their molecular reactivity. To achieve this goal, we present our investigations on atropisomeric acrylimides in solution where, their photochemical and photophysical reactivities/properties are governed by the restricted bond rotation of the N−C(aryl) bond (Scheme 1). We have also extended the investigation to solidstate reactivity with complete control over reactivity due to crystalline confinement. The photochemical and photophysical properties of imides9,10 have been well investigated for their unique properties and have been employed for both industrial and medical applications.11,12 The intramolecular [2 + 2]-photocycloaddition of achiral acyclic acrylimides was first reported by LaLonde and coworkers in 1965.4 Maruyama and Ishitoku5 reported the formation of a straight [2 + 2] photocycloaddition © 2014 American Chemical Society

Scheme 1. Cross-Photocycloaddition of Atropisomeric Acrylimides 1

of achiral acyclic acrylimides that was debated by Bellus and coworkers,6 who observed the formation of head-to-tail cycloadduct (i.e., cross-[2 + 2]-photocycloaddition) irrespective of the irradiation conditions. Thus, contrasting reports exist in literature about the formation of straight versus cross-[2 + 2]photoadducts during intramolecular [2 + 2]-photocycloaddition of achiral acrylimides. We were intrigued by these reports because one could access atropisomeric acrylimides and use them for stereospecific photochemical reaction. This is because Special Issue: Current Topics in Photochemistry Received: June 7, 2014 Revised: August 13, 2014 Published: September 8, 2014 10596

dx.doi.org/10.1021/jp505678b | J. Phys. Chem. A 2014, 118, 10596−10602

The Journal of Physical Chemistry A

Article

enantiomers of 1a revealed that the N-aryl substituent was perpendicular to the plane of the CO−N−CO unit.25 A distance of 4.2 Å (Schmidt distance) between the alkene carbons is most often required for the [2 + 2]-photoreaction to occur in the solid state.27 For straight [2 + 2]-photocycloaddition of acrylimides, the bond formation upon photoexcitation has to occur between the two internal alkene carbon atoms and the two terminal carbon atoms (Figure 1, inset; Scheme 1). Inspection of the X-ray structure 1a (Figure 1) shows that the distance between the two internal alkene carbons (i.e., the distance between the C1 and C4 carbon) is 2.931 Å and the distance between the terminal alkene carbons (i.e., the distance between the C2 and C3 carbon) is 3.856 Å. The bond distances for straight [2 + 2]-photocycloaddition are within the Schmidt distance for photoreaction within the solid state. Similarly, for cross-[2 + 2]-photocycloaddition of acrylimides, the bond formation upon photo excitation has to occur between the internal alkene carbon and the terminal carbon atom (Figure 1, inset; Scheme 1). Examination of the single-crystal X-ray structure of 1a (Figure 1) shows that the distance between the internal alkene carbon bearing the phenyl substituent and the terminal alkene carbon of the methacryloyl functionality (i.e., the distance between the C1 and C3 carbon) is 3.158 Å and the distance between the internal alkene carbon bearing the methyl substituent and the terminal alkene carbon of the phenylmethacryloyl functionality (i.e., the distance between the C2 and C4 carbon) is 3.189 Å. The bond distances for cross-[2 + 2]-photocycloaddition are within the Schmidt distance for photoreaction to occur in the solid state. Additionally, the torsional angle between the double bonds (C2−C1−C3−C4 torsion angle) was 21.7°. Because the reactive double bonds were within the Schmidt distance, it opened up the possibility to investigate the influence of crystalline environment on the photochemical reactivity of atropisomeric acrylimides. Irradiation of optically pure crystals of 1a was carried out at room temperature on a Petri dish using a Rayonet reactor equipped with 254 nm tubes (16 tubes × 14 W). The reaction was clean with an exclusive formation of the cross-photocyclized product that was characterized by NMR spectroscopy and HRMS. The enantiomeric relation in the individual photoproduct was established by HPLC analysis on a chiral stationary phase that gave enantiomeric excess (ee) of 98% for the individual photoproduct. The formation of the cross-photoadduct over the straight photoadduct was quite surprising in the solid state because the shortest distance was between C2 and C3 (2.931 Å; Figure 1). For the formation of the cross-photoadduct, the shortest distance was between C2 and C3 of 3.158 Å. We believe that the exclusive formation of cross-[2 + 2] photocycloadduct 2a in the solid state was due to the minimum atomic movements required for the formation of the photoproduct, while the formation of straight [2 + 2] photoadduct would require drastic atomic movements that would not be preferred within the confined environment of a crystalline matrix. This has its origin in the orientation of the double bonds (torsional angle) and the orbital density in the double bond.25 To develop a quantitative model for understanding the photochemical reactivity of acrylimides leading to cross-cyclobutane photoproduct, we need to have a closer look at the excited state of the acrylimide chromophore. On the basis of the single-crystal X-ray structure of 1a, the orbital density was computed using Gaussian 09 at B3LYP/6-31G* level,28 which showed that the excitation is localized at the styrene chromophore (Figure 1). While one has to be

achieving high stereoselectivity in the photoproducts during photochemical transformations in solution has presented formidable challenges due to the short lifetimes of the reactive excited state(s).13−17 Over the past few years, we have utilized the approach of incorporating atropisomeric chromophores for controlling photochemical reactions leading to excellent selectivity in the photoproducts during various photochemical transformations. Our studies18−24 have demonstrated the role of axial chirality to be crucial during various asymmetric photochemical reactions viz., 6π-photocyclization,18,19 4πphotocyclization,20 Norrish−Yang reactions,21,22 [2 + 2]photocycloaddition,23 and Paternò−Büchi reactions.24 With this precedent, we were interested in evaluating atropisomeric acrylimides 1 (Scheme 1) for asymmetric photochemical reactions and in understanding their excited-state chemistry in both solution and solid state using detailed photophysical investigations.

2. RESULTS AND DISCUSSION The atropisomeric acrylimides 1 were synthesized by sequential acylation reactions from the corresponding aniline derivative,25 and were characterized by both NMR spectroscopy and mass spectrometry.25 HPLC analysis on a chiral stationary phase revealed that the atropisomeric acrylimides 1a−b were separable and likely had a high barrier for rotation around the N−C(aryl) bond. Establishing the energy barrier for rotation that is responsible for the axial chirality is fundamental to our investigation as fast rotation around the N−C(aryl) bond leads to racemization thereby eroding the absolute configuration that will result in poor selectivity in the desired stereospecific photoreaction. As atropisomers of 1a−b were stable at room temperature, measurements of racemization kinetics were performed at 45 °C in acetonitrile and methanol as solvents to ascertain the half-life of racemization (t1/2), rate constant for racemization (krac) and the activation energy for racemization (ΔG‡rac).26 Inspection of Table 1 reveals a Table 1. Racemization Rate Constant (krac), Half Life (τ1/2), and Activation Free Energy (ΔG‡rac) for Optically Pure 1a−b at 45 °Ca entry substrate 1 2 3 4

1a 1b

solvent MeCN MeOH MeCN MeOH

krac (s−1) 4.1 4.3 7.2 8.6

× × × ×

−6

10 10−6 10−6 10−6

τ1/2 (days)

ΔG‡rac (kcal·mol−1)

2.0 1.9 1.1 0.9

26.5 26.5 26.1 26.0

a

Optically pure atropisomeric acrylimides were used for the racemization studies. Values carry an error of ±5%. Racemization kinetics was followed by HPLC analysis.25

racemization barrier (ΔG‡rac) of ∼26 kcal/mol for the atropisomeric acrylimides 1 with half-life of 1−2 days at 45 °C. This suggested that atropisomeric acrylimides could be employed for stereospecific photoreactions at room temperature without any loss of axial chirality. Because we were successful in obtaining a single crystal X-ray structure of the individual atropisomers of 1a (4:1 hexanes/ chloroform as the crystallizing solvent mixture), it provided us with insights into the reactivity in the solid state as well as in solution (vide infra). The absolute configuration of individual atropisomers of 1a (Figure 1) was established by Flack parameters.25 Single-crystal X-ray structure analysis of the 10597

dx.doi.org/10.1021/jp505678b | J. Phys. Chem. A 2014, 118, 10596−10602

The Journal of Physical Chemistry A

Article

Figure 1. Top: Single-crystal X-ray structure of 1a. Bottom: Understanding photochemical reactivity based on orbital density computed with Gaussian 09 at B3LYP/6-31G* level using the single crystal X-ray structure of 1a.

Table 2. Reaction Condition Optimization for Intramolecular [2 + 2] Photoreactions of Acrylimidesa entry

cmpd

irradiation conditionsb

solvent

% convn.d

2:3

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1a

direct irradiation, bb - Pyrex cutoff/3 h/N2 direct irradiation, RR, 254 nm/5 min/N2 direct irradiation, RR, 254 nm/5 min/O2 direct irradiation, bb - Pyrex cutoff/3 h/N2 benzophenone (10 mol %)/RR ∼350 nm/3 h/N2 xanthone (10 mol %)/RR ∼350 nm/1.5 h/N2 xanthone (10 mol %)/RR ∼350 nm/30 min/N2 thioxanthone (10 mol %)/∼420 nm/15 min/N2 xanthone (10 mol %)/RR ∼350 nm/1.5 h/N2 xanthone (10 mol %)/RR ∼350 nm/30 min/N2 thioxanthone (10 mol %)/RR ∼420 nm/15 min/N2 xanthone (10 mol %)/RR ∼350 nm/3 h/N2 xanthone (10 mol %)/RR ∼350 nm/1.5 h/N2 xanthone (10 mol %)/RR ∼350 nm/30 min/N2 thioxanthone (10 mol %)/RR ∼420 nm/15 min/N2

MeCN MeCN MeCN acetonec MeCN MeCN MeOH MeCN MeCN MeOH MeCN MeCN MeCN MeOH MeCN

97 71 (70)e 68 100 100 100 (71)e 100 100 100 (86)e 100 58 100 100 100 (78)e 60

>99:1 >99:1 >99:1 >99:1 >99:1 >99:1 >99:1 >99:1 >99:1 >99:1 >99:1 82:18 80:20 82:18 82:18

1b

1cf

a

Irradiation at room temperature. Values based on 1H NMR spectroscopy (±5% error). [1a] = 2.88 mM, [1b] = 2.48 mM, and [1c] = 3.43 mM. bbb = broadband irradiation performed using 450 W mercury lamp with a Pyrex cutoff filter (99% in 2, highlighting the role of axial chirality in determining the ee in the photoproduct (Table 3). A point to note is that the cross-photoproduct 2 was favored (2:3→ 99:1) with both direct and sensitized irradiations with excellent conversions and high mass balance for axially chiral acrylimides. Detailed photophysical investigations were carried out to gain more insights into the cross-[2 + 2] photocycloaddition of atropisomeric acrylimides with 1a as the model system. Fluorescence and phosphorescence measurements on 1a were futile because they showed only negligible luminescence both at room temperature in solution and at 77 K in solid matrix, indicating a very fast decay of excited states. To overcome this limitation, we recorded the transient absorbance spectra of 1a (Figure 2) in argon-saturated acetonitrile solution using laser

in isotropic media with either methanol (MeOH) or acetonitrile (MeCN) as solvents. [2 + 2]-Photocycloaddition of atropisomeric acrylimides 1a−c was carried out under different irradiation conditions that proceeded smoothly with excellent isolated yields and mass balance. Three different sets of irradiation conditions were examined: (a) direct irradiation; (b) sensitized irradiation under UV light (e.g., xanthone as sensitizer); and (c) sensitization under visible-light irradiation (e.g., thioxanthone as sensitizer). After the photoreaction, the solvent was removed under reduced pressure, and the product(s) was purified by column chromatography. The purified photoproduct was characterized by NMR spectroscopy and HRMS. The ee in the photoproduct was determined by HPLC analysis on a chiral stationary phase. We employed 1a as a model system to optimize the irradiation conditions (Tables 2 and 3) as we had established its reactivity in the solid state. Table 3. Enantiospecific Intramolecular [2 + 2] Photoreactions of Acrylimides 1a−ba entry

substrate

atropisomerb

solvent

t (h)c

% ee (2)d

1 2 3 4 5 6

1a

M - isomer P - isomer M - isomer P - isomer Pk-A Pk-B

MeCN MeCN solid state solid state MeCN MeCN

1.5 1.5 11 11 1.5 1.5

98 98 99 99 99 99

1b

a

Irradiation at room temperature using Rayonet reactor equipped with ∼350 nm light tubes (16 tubes × 14 W) using 10 mol % of xanthone as the triplet sensitizer under constant purging of N2. [1a] = 2.88 mM and [1b] = 2.48 mM. bPk-A and Pk-B refers to the elution order of the peaks for a given pair of enantiomers/atropisomers. cPhotolysis time. d ee values (error of ±3%) from HPLC analysis on a chiral stationary phase (Chiralpak IC for 1a and RR-WHELK-01 for 1b) with hexanes/ 2-propanol as mobile phase.

Direct irradiation was carried out using a 450 W medium pressure Hg lamp placed inside a water-cooled jacket with a Pyrex cut off filter or a Rayonet reactor equipped with 254 nm tubes (16 tubes × 14 W) under a constant flow of nitrogen. In the case of atropisomeric acrylimides, the cross-[2 + 2] photocycloadduct was exclusively observed as the product in both MeOH and MeCN with excellent mass balance and conversion. The reaction was found to be very efficient upon changing the irradiation source from a 450 W medium pressure Hg lamp to a Rayonet reactor equipped with 254 nm tubes. The efficiency of the reaction was unaffected under oxygen (Table 2, entry 3; 68% conversion for 5 min irradiation) when compared with conversion under nitrogen atmospheres (Table 2, entry 2; 71% conversion for 5 min irradiation). To channel the reactivity via the triplet excited state, we then resorted to sensitized irradiations using triplet sensitizers of varying triplet energies (ET) between 78 and 63 kcal/mol.29 Efficient reaction was observed under continuous flow of nitrogen in acetone (ET between 78 kcal/mol, acting as both solvent and sensitizer) and with benzophenone (10 mol %, ET between 73 kcal/mol)29 in MeCN. The reaction was equally efficient in MeCN upon changing triplet sensitizer to xanthone (∼350 nm irradiation) and thioxanthone (∼420 nm irradiation). In the case of atropisomeric acrylimides 1a−b, exclusive formation of cross-[2 + 2]-photocyclized product was observed with excellent isolated yields (Table 2). For the corresponding achiral imide 1c, both cross-and straight [2 + 2]-photocyclized products were

Figure 2. Transient absorbance spectra monitored at 0−25 ns (red) and 120−170 ns (blue) after pulsed laser excitation (λex = 266 nm, 5 ns pulse width) of argon-saturated acetonitrile solution of 1a. Inset: Absorbance kinetic trace monitored at 320 nm.

flash photolysis (λex = 266 nm, 5 ns pulse width). We observed a transient spectrum centered around 320 nm that decayed monoexponentially with a lifetime of 40 ns (Figure 2, inset). This transient was quenched by molecular oxygen with a rate constant of 6 × 109 M−1 s−1 (Figure 3). The concurrent generation of singlet oxygen (Δ1O2) was monitored by its characteristic phosphorescence at 1270 nm (Figure 4). The main chromophore in 1a contains a styrene skeleton that is known to show triplet−triplet absorption at ∼320 nm.30 On the basis of oxygen quenching, singlet oxygen generation and similarities with the triplet absorption of styrene, we assigned the transient absorption shown in Figure 2 to the triplet absorption of 1a. Similar transient absorption centered at ∼320 nm was also observed for laser excitation (266 nm) of 1b and 1c.25 The analogous atropisomeric acrylimide 1b showed a triplet lifetime (τT) of 41 ns that was comparable to the τT of 1a, whereas the triplet lifetime of acrylimide 1c was significantly shorter (∼20 ns). The shorter triplet lifetime of 1c may be attributed to the free rotation of the N−C(aryl) bond. 10599

dx.doi.org/10.1021/jp505678b | J. Phys. Chem. A 2014, 118, 10596−10602

The Journal of Physical Chemistry A

Article

Figure 3. (A) Absorbance decay traces of 1a triplet states in argon(red) or oxygen- (blue) saturated acetonitrile solutions using laser flash photolysis (λex = 266 nm, 5 ns pulse width) monitored at 320 nm. (B) Determination of the bimolecular quenching rate constant kq of quenching of 1a triplet states by O2 from the slope of the plot of the inverse triplet lifetime versus the O2 concentrations. Figure 5. Determination of the bimolecular quenching rate constants kq of quenching of sensitizer triplet states by 1a using laser flash photolysis (λex = 355 nm, 7 ns pulse width). Inverse triplet lifetime determined from triplet absorption decay traces monitored at 620 nm for xanthone (blue) and thioxanthone (red) and 425 nm for 2acetonaphthone (green) with varying concentration of 1a in argonsaturated acetonitrile solutions. Refer to the Supporting Information for details.

reactivity upon direct irradiation of acrylimides to generate cross-cycloadducts. Laser flash photolysis experiments showed that triplet states of 1a are generated by direct excitation (Figure 2) that was quenched by molecular oxygen to produce singlet oxygen. The quantum yield of singlet oxygen formation (ΦΔ) upon direct excitation of 1a was estimated in oxygensaturated carbontetrachloride using phenalone as standard32 and was found to be as low as ∼0.01 (Figure 4A). Considering that not all triplet states of 1a are quenched in oxygen-saturated solutions due to the short triplet lifetime (Figure 3A), the triplet quantum yield of 1a is estimated to be on the order of ΦT ≈ 0.02. The low triplet quantum yield under direct irradiation is consistent with photoreaction product studies in the absence and presence of molecular oxygen. Upon changing the reaction atmosphere from N2 to O2, only a 3% decrease in conversion was observed (which is within our experimental error). If one overlooks the experimental error and assumes that the intersystem crossing to compete with a concerted reaction from the singlet manifold to some degree, the 3% decrease could be attributed to the quenching of the triplet excited state upon direct irradiation. Hence because of the low triplet quantum yield and relatively low impact of oxygen quenching on photoproduct yields, we believe that upon direct excitation of 1a, [2 + 2] photocloaddition probably proceeds mostly from the singlet-excited state in a concerted fashion. The low triplet quantum yield is quite reasonable due to a large singlet−triplet gap in ππ* excited state of acrylimide that is similar to that of the styrene chromophore.2

Figure 4. (A) Singlet oxygen phosphorescence decay traces monitored at 1270 nm generated by pulsed laser excitation (λex = 266 nm, 5 ns pulse length) of O2 saturated CCl4 solutions of 1a or phenalone with matching absorbances of 0.3 at 266 nm. (B) Singlet oxygen phosphorescence spectrum generated by steady-state irradiation (355 nm) of 1a in O2-saturated CCl4 solution.

Because the photocycloaddition of acrylimides was more efficient in the presence of triplet sensitizers, such as xanthone (∼350 nm irradiation) and thioxanthone (∼420 nm irradiation), we investigated the energy-transfer kinetics by laser flash photolysis. The bimolecular rate constants of quenching of sensitizer triplet states by 1a were determined by monitoring the sensitizer triplet absorption decays after pulsed laser excitation (355 nm) in the presence of varying concentrations of 1a. The slope of the plot of the inverse sensitizer triplet lifetime versus the quencher concentration yielded the bimolecular quenching rate constants (Figure 5). For xanthone as sensitizer, which has a triplet energy of ET ≈ 73 kcal/mol,29 a rate constant close to the diffusion limit of kq = 7.3 ± 0.2 × 109 M−1 s−1 was observed (Figure 5, blue). Upon changing the triplet sensitizer to thioxanthone (ET ≈ 63 kcal/mol),29 kq decreased by half to ∼4.0 ± 0.1 × 109 M−1 s−1 (Figure 5, red). If the triplet energy of the sensitizer was further lowered as with 2-acetonaphthone (ET ≈ 59.5 kcal/mol),29 a drop of two-order of magnitude was observed (kq to 8.6 ± 0.2 × 107 M−1 s−1) (Figure 5, green). This decrease shows that the triplet energy of 1a is close to 63 kcal/mol. This triplet energy is in agreement with the triplet energy of styrene (62 kcal/mol),31 the main chromophore in acrylimides. We have shown that photocycloaddition of acrylimides proceeds efficiently from the triplet state when a triplet sensitizer is used with sufficiently high triplet energy. The question that remains unanswered is the nature of the reactive spin state (singlet or triplet excited state) that is responsible for

3. CONCLUSIONS The present study has illustrated the use of a nonbiaryl axially chiral unit for a facile [2 + 2] photocycloaddition in acrylimides. Using laser flash photolysis, we have shown that atropisomeric imides react efficiently from both singlet (direct irradiation) and triplet (sensitized irradiation) excited states. The N− C(aryl) bond rotations not only control the regioselectivity, leading to the formation of the less studied cross-[2 + 2]10600

dx.doi.org/10.1021/jp505678b | J. Phys. Chem. A 2014, 118, 10596−10602

The Journal of Physical Chemistry A

Article

orbitals for 1a. This material is available free of charge via the Internet at http://pubs.acs.org.

photocyclized product, but also influence the enantioselectivity in the product. The acrylimides possess triplet energy around 63 kcal/mol and thus are amenable to visible-light irradiations, highlighting a clean and efficient strategy for performing photochemical transformations.33,34 Because the bicycle photoproducts have structural similarities to compounds that have been shown to be effective as aromatase inhibitors,35 our strategy can be extended to synthesize pharmaceutically relevant and useful skeletons.



*E-mail: [email protected] (S.J.). *E-mail: [email protected] (J.S.). Funding

J.S. thanks NSF for generous financial support (CHE1213880). S.J. thanks NSF for financial support (CHE1111392). The authors at NDSU thank NSF-CRIF (CHE0946990) for the purchase of departmental X-ray diffractometer.

4. EXPERIMENTAL SECTION 4.1. Procedure for Direct Irradiation of Acrylimides 1a−c. Solution of acrylimides 1a−c in methanol or acetonitrile ([1a] = 2.88 mM, [1b] = 2.48 mM, and [1c] = 3.43 mM) was irradiated at 25 °C using quartz tubes in a Rayonet reactor equipped with 254 nm tubes (16 tubes ×14 W). For broadband irradiations, the solution of substrates 1a−c in requisite solvent was irradiated in Pyrex tubes with a 450 W medium pressure mercury lamp placed inside a water-cooled quartz well. After irradiation, triphenylmethane was added as an internal standard and the solvent was evaporated under reduced pressure. Conversion and mass balance was determined by 1H NMR spectroscopy.25 4.2. Sensitized Photoreactions of Acrylimide 1a−c. Solution of acrylimide 1a−c in methanol or acetonitrile ([1a] = 2.88 mM, [1b] = 2.48 mM, and [1c] = 3.43 mM) along with 10-mol % triplet sensitizer (xanthone or thioxanthone) was irradiated at 25 °C using Pyrex tubes in a Rayonet reactor. For acetone sensitization, acetone was used as both solvent and sensitizer and was irradiated with a 450 W medium pressure mercury lamp with a Pyrex cut off filter. The Rayonet reactor was equipped with ∼350 nm tubes (16 tubes × 14 W) for triplet sensitization by xanthone and ∼420 nm tubes (16 tubes ×14 W) for triplet sensitization with thioxanthone. After irradiation, triphenylmethane was added as internal standard and the solvent was evaporated under reduced pressure. Conversion and mass balance were determined by 1H NMR spectroscopy. For irradiation of optically pure isomers, the ee values were determined by HPLC analysis on a chiral stationary phase. 4.3. Laser Flash Photolysis and Singlet Oxygen Phosphorescence. Laser flash photolysis experiments employed the pulses from a Spectra Physics GCR-150−30 Nd:YAG laser (355 nm, ca. 5 mJ/pulse, 7 ns pulse length or 266 nm, ca. 5 mJ/pulse, 5 ns pulse length) and a computercontrolled system that has been described elsewhere.36 Singlet oxygen phosphorescence measurements were performed on a modified Fluorolog-3 spectrometer (HORIBA Jobin Yvon) in conjunction with a NIR sensitive photomultiplier tube (H10330A-45, Hamamatsu). A 450 W xenon lamp was used for steady-state excitation to record singlet oxygen phosphorescence spectra, and a Spectra Physics GCR-150−30 Nd:YAG laser (266 nm, ca. 2 mJ/pulse, 5 ns) was used for pulsed excitation to collect 1O2 phosphorescence decay traces at 1270 nm, which were stored on a digital oscilloscope (TDS 360 from Tektronics).



AUTHOR INFORMATION

Corresponding Authors

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank Dr. Angel Ugrinov for solving the single-crystal XRD structures and Elango Kumarasamy, Anthony Clay, and Nandini Vallavoju for their scholarly inputs during the preparation of manuscript.

■ ■

ABBREVIATIONS I(Z), zwitterionic intermediate; CI, conical intersection REFERENCES

(1) Inoue, Y. Asymmetric photochemical reactions in solution. Chem. Rev. 1992, 92, 741−770. (2) Turro, N. J.; Ramamurthy, V.; Scaiano, J. C. Modern Molecular Photochemistry Of Organic Molecules; University Science Books: Sausalito, CA, 2010; pp 705−799. (3) Inoue, Y. Enantiodifferentiating photosensitized reactions. In Chiral Photochemistry, Inoue, Y., Ramamurthy, V., Eds.; Marcel Dekker: New York, 2004; Vol. 11, pp 129−177. (4) LaLonde, R. T.; Aksentijevich, R. I. The photolysis of N,Ndimethylacrylylmethacrylamide. Tetrahedron Lett. 1965, 6, 23−27. (5) Maruyama, K.; Ishitoku, T. Photochemical cyclization of acrylimide derivatives. Chem. Lett. 1980, 9, 359−360. (6) Alder, A.; Bühler, N.; Bellus, D. A Note on intramolecular photochemical cycloaddition of N-substituted dimethacrylimides. Helv. Chim. Acta 1982, 65, 2405−2412. (7) Srinivasan, R.; Carlough, K. H. Mercury(3P1) photosensitized internal cycloaddition reactions in 1,4-, 1,5-, and 1,6-dienes. J. Am. Chem. Soc. 1967, 89, 4932−4936. (8) Liu, R. S. H.; Hammond, G. S. Photosensitized internal addition of dienes to olefins. J. Am. Chem. Soc. 1967, 89, 4936−4944. (9) Kanaoka, Y. Photoreactions of cyclic imides. Examples of synthetic organic photochemistry. Acc. Chem. Res. 1978, 11, 407−413. (10) Sakamoto, M.; Takahashi, M.; Fujita, T.; Watanabe, S.; Iida, I.; Nishio, T.; Aoyama, H. Solid-state photochemistry: absolute asymmetric oxetane synthesis from an achiral acyclic imide using the chiral crystal environment. J. Org. Chem. 1993, 58, 3476−3477. (11) Gao, B.; Li, H.; Liu, H.; Zhang, L.; Bai, Q.; Ba, X. Water-soluble and fluorescent dendritic perylene bisimides for live-cell imaging. Chem. Commun. 2011, 47, 3894−3896. (12) Li, C.; Wonneberger, H. Perylene imides for organic photovoltaics: yesterday, today, and tomorrow. Adv. Mater. 2012, 24, 613−636. (13) Rau, H. Asymmetric photochemistry in solution. Chem. Rev. 1983, 83, 535−547. (14) Bach, T.; Bergmann, H.; Grosch, B.; Harms, K. Highly enantioselective intra- and intermolecular [2 + 2] photocycloaddition reactions of 2-quinolones mediated by a chiral lactam host: host−guest interactions, product configuration, and the origin of the stereoselectivity in solution. J. Am. Chem. Soc. 2002, 124, 7982−7990.

ASSOCIATED CONTENT

S Supporting Information *

Synthetic procedures and characterization of reactants and photoproducts, racemization kinetics, UV−vis spectra, photophysical data, crystal structures, and computed localization of 10601

dx.doi.org/10.1021/jp505678b | J. Phys. Chem. A 2014, 118, 10596−10602

The Journal of Physical Chemistry A

Article

(15) Mori, T.; Weiss, R. G.; Inoue, Y. Mediation of conformationally controlled photodecarboxylations of chiral and cyclic aryl esters by substrate structure, temperature, pressure, and medium constraints. J. Am. Chem. Soc. 2004, 126, 8961−8975. (16) Sivaguru, J.; Natarajan, A.; Kaanumalle, L. S.; Shailaja, J.; Uppili, S.; Joy, A.; Ramamurthy, V. Asymmetric photoreactions within zeolites: role of confinement and alkali metal ions. Acc. Chem. Res. 2003, 36, 509−521. (17) Veerman, M.; Resendiz, M. J. E.; Garcia-Garibay, M. A. Largescale photochemical reactions of nanocrystalline suspensions: a promising green chemistry method. Org. Lett. 2006, 8, 2615−2617. (18) Ayitou, A. J.-L.; Sivaguru, J. Light-induced transfer of molecular chirality in solution: enantiospecific photocyclization of molecularly chiral acrylanilides. J. Am. Chem. Soc. 2009, 131, 5036−5037. (19) Ayitou, A. J.-L.; Sivaguru, J. Reactive spin state dependent enantiospecific photocyclization of axially chiral α-substituted acrylanilides. Chem. Commun. 2011, 47, 2568−2570. (20) Kumarasamy, E.; Jesuraj, J. L.; Omlid, J. N.; Ugrinov, A.; Sivaguru, J. Light-induced enantiospecific 4π ring closure of axially chiral 2-pyridones: enthalpic and entropic effects promoted by Hbonding. J. Am. Chem. Soc. 2011, 133, 17106−17109. (21) Ayitou, A. J.-L.; Jesuraj, J. L.; Barooah, N.; Ugrinov, A.; Sivaguru, J. Enantiospecific photochemical Norrish/Yang type II reaction of nonbiaryl atropchiral α-oxoamides in solution - axial to point chirality transfer. J. Am. Chem. Soc. 2009, 131, 11314−11315. (22) Jesuraj, J. L.; Sivaguru, J. Photochemical type II reaction of atropchiral benzoylformamides to point chiral oxazolidin-4-ones. Axial chiral memory leading to enantiomeric resolution of photoproducts. Chem. Commun. 2010, 46, 4791−4793. (23) Kumarasamy, E.; Sivaguru, J. Light-induced stereospecific intramolecular [2 + 2]-cycloaddition of atropisomeric 3,4-dihydro-2pyridones. Chem. Commun. 2013, 49, 4346−4348. (24) Raghunathan, R.; Kumarasamy, E.; Iyer, A.; Ugrinov, A.; Sivaguru, J. Intramolecular Paterno-Buchi reaction of atropisomeric αoxoamides in solution and in the solid-state. Chem. Commun. 2013, 49, 8713−8715. (25) Refer to Supporting Information. (26) Wolf, C. Dynamic Stereochemistry of Chiral Compounds. Principles and Applications; RSC Publishing: Cambridge, U.K., 2008. (27) Schmidt, G. M. J. Photodimerization in the solid state. Pure Appl. Chem. 1971, 27, 647−678. (28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09; Gaussian, Inc.: Wallingford, CT, 2009. (29) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochemistry, 2nd ed.; Marcel Dekker: New York, 1993; pp 54−98. (30) Brede, O.; David, F.; Steenken, S. Styrene triplets: Absorption spectra, lifetimes and quantum yields. J. Photochem. Photobiol., A 1996, 97, 127−131. (31) Swiderek, P.; Fraser, M. J.; Michaud, M.; Sanche, L. Electronenergy-loss spectroscopy of the low-lying triplet states of styrene. J. Chem. Phys. 1994, 100, 70−77. (32) Schmidt, R.; Tanielian, C.; Dunsbach, R.; Wolff, C. Phenalenone, a universal reference compound for the determination of quantum yields of singlet oxygen O2(1Δg) sensitization. J. Photochem. Photobiol., A 1994, 79, 11−17. (33) Vallavoju, N.; Sivaguru, J. Supramolecular photocatalysis: combining confinement and non-covalent interactions to control light initiated reactions. Chem. Soc. Rev. 2014, 43, 4084−4101. (34) Vallavoju, N.; Selvakumar, S.; Jockusch, S.; Sibi, M. P.; Sivaguru, J. Enantioselective organo-photocatalysis mediated by atropisomeric thiourea derivatives. Angew. Chem., Int. Ed. 2014, 53, 5604−5608. (35) Stanek, J.; Alder, A.; Bellus, D.; Bhatnagar, A. S.; Haeusler, A.; Schieweck, K. Synthesis and aromatase inhibitory activity of novel 1(4-aminophenyl)-3-azabicyclo[3.1.0]hexane- and -[3.1.1]heptane-2,4diones. J. Med. Chem. 1991, 34, 1329−1334. (36) Yagci, Y.; Jockusch, S.; Turro, N. J. Mechanism of photoinduced step polymerization of thiophene by onium salts: reactions of

phenyliodinium and diphenylsulfinium radical cations with thiophene. Macromolecules 2007, 40, 4481−4485.

10602

dx.doi.org/10.1021/jp505678b | J. Phys. Chem. A 2014, 118, 10596−10602