- Email: [email protected]
Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine
Journal of Molecular Structure xxx (2018) 1e8
Contents lists available at ScienceDirect
Journal of Molecular Structure journal homepage: http://www.elsevier.com/locate/molstruc
Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine Onyinye Osisioma, Mrinal Chakraborty, Bruce S. Ault, Anna D. Gudmundsdottir* Department of Chemistry, University of Cincinnati, PO Box 210172, Cincinnati OH 45221, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 December 2017 Received in revised form 30 March 2018 Accepted 10 April 2018 Available online xxx
Irradiation of 2-azidovinylbenzene (1) in a cryogenic argon matrix with a 254 nm bandpass filter results in 2-phenyl-2H-azirine (2) as the major product, some ketenimine 5, and a small amount of cis-1, whereas irradiation of vinyl azide 1 with light above 300 nm forms only azirine 2 and ketenimine 5. Prolonged irradiation of azirine 2 through a 254 nm bandpass filter produces ylide 3 selectively. Laser flash photolysis with excitation at 308 nm demonstrated that azirine 2 forms triplet vinylnitrene 4 (l ~ 440 nm), whereas vinyl azide 1 does not exhibit transient absorption on a nanosecond timescale. DFT calculations (B3LYP/6-31 þ G(d)) were used to aid the characterization of the photoproducts and to support the proposed formation mechanisms. © 2018 Published by Elsevier B.V.
Keywords: Azirine Vinyl azide Argon matrix Wavelength-dependent photochemistry Laser flash photolysis Density functional theory calculation
1. Introduction Wavelength dependent photochemistry is fascinating because by simply using a different wavelength of light, the same reagent can be transformed into distinct products. The origin of wavelength dependent photochemistry is the excitation of separate chromophores in a reagent, resulting in the population of different excited states that react distinctively. It is also possible that excitation at different wavelengths can lead to the formation of different excited states of the same chromophore. Such examples are rare, as higher excited states generally undergo efficient internal conversion to the lowest excited state, as according to Kasha's rule, but a number of exceptions have been reported, such as azulene [1], naphthyl azides [2], and diazirines [3]. The photochemistry of nonconjugated aromatic azirine derivatives is extraordinary, as short wavelength irradiation leads to cleavage of the CeC bond of the azirine ring to form ylides, whereas longer wavelength irradiation results in the CeN bond breaking to form triplet vinylnitrenes [4,5]. For example, short wavelength irradiation of 3-methyl-2-phenyl-2H-azirine in an argon matrix yields the corresponding ylide, whereas longer wavelength
* Corresponding author. E-mail address: [email protected] (A.D. Gudmundsdottir).
irradiation forms the corresponding ketenimine (Scheme 1) [6,7]. We theorized that breakage of the CeC bond occurs following direct absorption by the azirine moiety. In contrast, longer wavelength irradiation selectively excites the phenyl moiety, which intersystem crosses to its triplet excited state, and subsequent energy transfer to the triplet excited state of the azirine chromophore results in cleavage of the CeN bond to form triplet vinylnitrene. However, as triplet vinylnitrene was not observed directly, it was theorized that the vinylnitrene intersystem crosses efficiently to form ketenimine. The hypothesis that the phenyl group serves as a triplet sensitizer was further supported by the fact that the photochemistry of 2-methyl-3-phenyl-2H-azirine, in which C]N is conjugated with the phenyl ring, does not display a wavelength dependence. In this case, the ylide is formed, regardless of the irradiation wavelength, as it is not possible to selectively irradiate the phenyl group of this molecule, and thus only singlet reactivity is observed (Scheme 2). In this article, we report the photochemistry of 2azidovinylbenzene (1) and 2-phenyl-2H-azirine (2) in cryogenic argon matrices as a function of irradiation wavelength. Because the phenyl chromophore is in conjugation with the vinyl azide moiety in 1, but is not conjugated with the C]N moiety in azirine 2, it is possible to compare their photochemistry to determine whether vinyl azide 1 reacts on its singlet or triplet surface upon direct irradiation. The photochemistry of azirine 2 is strongly wavelength
https://doi.org/10.1016/j.molstruc.2018.04.042 0022-2860/© 2018 Published by Elsevier B.V.
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
2
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
2.2. Matrix isolation Matrix isolation was performed using conventional equipment [10]. Vinyl azide 1 was deposited onto a CsI cold window cooled to 20 K by a CTI closed-cycle refrigerator. A stream of argon was passed over the volatile oil under vacuum to entrap vinyl azide 1 into the argon matrix. In two different sets of experiments, the argon matrix was irradiated over short time intervals with a medium-pressure short-arc 200 W mercury arc lamp through a Pyrex filter (>300 nm) or a 254 nm bandpass filter, which transmitted light between wavelengths of 233e271 nm. Infrared spectra were recorded at a resolution of 1 cm1 using a PerkinElmer Spectrum One FTIR spectrometer during and after matrix deposition and intermittently during irradiation. 2.3. Laser flash photolysis Laser flash photolysis was carried out using an excimer laser (308 nm, 17 ns) [11]. In a typical experiment, a stock solution of vinyl azide 1 or azirine 2 was prepared in spectroscopic-grade acetonitrile, such that the solution had an absorbance of 0.3e0.6 at 308 nm.
2.4. Photolysis of 1
Scheme 1. Photolysis of 3-methyl-2-phenyl-2H-azirine in argon matrices.
Scheme 2. Photolysis of 2-methyl-3-phenyl-2H-azirine in argon matrices.
dependent, whereas the reactivity of vinyl azide 1 is relatively less affected by the irradiation wavelength.
Vinyl azide 1 was dissolved in chloroform-d1 in a Pyrex NMR tube sealed with a rubber septum, and argon was bubbled through the solution for 17 min. The resulting solution was irradiated with 10 light-emitting diodes (LEDs; 450 ± 5 nm) in parallel alignment, each with a constant current of 2 mA and a voltage drop of 3.2e3.8 V across each diode. After irradiation for 17 min, 1H NMR spectroscopy of the reaction mixture showed the formation of azirine 2 (76%) and remaining starting material (24%). Preparative preparation of azirine 2. Vinyl azide 1 (17 mg, 0.12 mmol) was dissolved in chloroform-d1 (0.6 mL, 7.49 mmol) and purged with argon for 5 min in a Pyrex NMR tube, which was sealed with a rubber septa cap and irradiated with a mercury-argon lamp for 40 min 1H NMR spectroscopy verified the formation of azirine 2 (51% yields) [12].1H NMR (CDCl3, 400 MHz): 9.99 (d, J ¼ 2 Hz, 1H), 7.2 (m, 5H), 2.72 (d, J ¼ 2 Hz, 1H) ppm. 2.5. Calculations
2. Experimental 2.1. Synthesis of 1 Vinyl azide 1 was synthesized according to a published procedure [8]. A catalytic amount of CuSO4 (80 mg, 0.50 mmol) was added to NaN3 (390 mg, 6.0 mmol) and dissolved in methanol (15 mL). Trans-2-phenylvinylboronic acid (0.739 g, 5.0 mmol) was added to the solution, and the resulting mixture was stirred until thin layer chromatography (TLC) showed complete depletion of the starting material and formation of a new product. The solvent was removed under vacuum and the resulting oil was dissolved in petroleum ether. After washing with water, the organic layer was dried over Na2SO4. The solvent was removed under vacuum, providing vinyl azide 1 as an oil (0.4048 g, 2.79 mmol, 56% yield). Characterization of vinyl azide 1 by IR and 1H NMR spectroscopy agreed well with the data reported in the literature [9]. 1H NMR (CDCl3, 400 MHz): 6.18 (d, J ¼ 16 Hz, 1H), 6.83 (d, J ¼ 16 Hz, 1H), 7.2e7.3 (m, 5H) ppm. IR (CDCl3): 2108, 1636, 1260, 1019, 930, 798, 747, 692 cm1.
The geometries were optimized at the B3LYP level of theory with the 6-31þG(d) basis set, as implemented in the Gaussian09 program [13e15]. The transition states were confirmed to have one imaginary vibrational frequency, and intrinsic reaction coordinate calculations were used to verify that the transition states corresponded to the attributed reactant and product [16,17]. To obtain the calculated absorption spectra for the excited states and intermediates, time-dependent density functional theory (TD-DFT) calculations were conducted at the same level of theory with the same basis set.
3. Results 3.1. Photolysis of vinyl azide 1 in solution A solution of vinyl azide 1 in chloroform,-d1 was exposed to 450 ± 5 nm LED light for 17 min 1H NMR spectroscopy revealed the formation of azirine 2 (76%) and remaining starting material (Scheme 3).
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
Scheme 3. Photolysis of vinyl azide 1 in acetonitrile using 450 ± 5 nm LED light.
3.2. Matrix isolation Vinyl azide 1 was deposited in an argon matrix and irradiated with a medium-pressure short-arc 200 W lamp through a Pyrex filter. Irradiation of the matrix resulted in depletion of the bands of azide 1 at 2102, 1644, 1603, 1329, 1289, 1265, 970, 752, and 693 cm1 (Fig. 1a). After 46 min of irradiation, the bands corresponding to vinyl azide 1 were almost fully depleted. Concurrent with the depletion of the bands of vinyl azide 1, new bands were formed, the most significant of which were located at 2024, 1662, 1498, 1457, 1189, 982, 841, 772, 763, 702, and 538 cm1. The bands observed at 1662, 1498, 1457, 1189, 982, 841, 772, 763, 702, and 538 cm1 were assigned to azirine 2 as they match well with its calculated bands (Fig. 1b), and with its spectrum reported earlier by Wentrup and Freiermuth [18]. Most of the IR bands are due to
Fig. 1. a) Difference spectrum of vinyl azide 1 in an argon matrix before and after 46 min of irradiation through a Pyrex filter. Calculated and scaled spectra of b) azirine 2 and c) ketenimine 5.
3
vibrations of the phenyl moiety of azirine 1, but of those that belong to the azirine moiety, the most significant are the C]N stretch and the bending of the CH bonds. Calculations and scaling with 0.9613 [19] place the azirine C]N band at 1684 cm1, which fits well with the observed band at 1662 cm1. Similarly, we assign the bands at 841 and 763 cm1 to CH bending of the azirine ring, which are calculated and scaled to be at 815 and 754 cm1. In addition, a small band was observed at 2024 (Figs. 1c and 2a). We assign this band to ketenimine 5 based on its reported [18] and calculated spectra. The calculated spectrum has the most intense band at 2026 cm1 (calculated intensity ¼ 631). The broad band at 982 cm1 which is assigned to azirine 2, can also be assigned to ketenimine 5 as it has significant band at 976 cm1 (calculated intensity ¼ 316) after scaling (Fig. 2b). In a separate experiment, vinyl azide 1 was deposited in an argon matrix and irradiated with a mercury arc lamp using a 254 nm bandpass filter (Fig. 3a). The bands associated with vinyl azide 1 were almost fully depleted after irradiation for 18 h (Fig. 4). Concurrent with the depletion of the bands corresponding to vinyl azide 1, new bands were formed. After 40 min of irradiation, new bands were observed at 2024, 1662, 1498, 1457, 1189, 982, 841, 772, 762, 702, and 538 cm1, and as before, we assign the bands at 2024 and 982 cm1 to ketenimine 5 and the bands at 1662, 1498, 1457,
Fig. 2. a) Ketenimine band observed at 2024 cm1 in the difference spectrum of vinyl azide 1 in an argon matrix after 46 min of irradiation through Pyrex filter. b) TD-DFT calculated and scaled spectrum of ketenimine at 2024 cm1.
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
4
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
Scheme 4. Photoproducts obtained by irradiation of vinyl azide 1 and azirine 2 in argon matrices.
Fig. 3. a) Difference spectrum of vinyl azide 1 in an argon matrix after irradiation for 18 h through a 254 nm bandpass filter. Calculated and scaled IR spectra of b) ketenimine 5, c) ylide 3, and d) the cis-isomer of vinyl azide 1.
with the calculated IR spectrum (Fig. 3c). Additionally, bands at 2153 and 2024 cm1 were also formed, and as before the 2024 cm1 band is assigned to ketenimine 5, whereas we theorize that the band at 2153 cm1 is due to the formation of the cis-isomer of vinyl azide 1 (Fig. 3d). The calculated and scaled azido band for cis-1 is at 2164 cm1, which matches well with the observed band at 2153 cm1. Nevertheless, no additional bands were assigned to the cis-isomer because it is not formed in significant quantities. The irradiation of vinyl azide 1 through both 254 nm and Pyrex filters in argon matrices results in azirine 2 and ketenimine 5, and these products can be formed on either the triplet or singlet surface of vinyl azide 1. In contrast, ylide 3, which must be formed on the singlet surface of azirine 2, is only obtained by irradiation of azirine 2 through the 254 nm bandpass filter. It was not possible to unambiguously determine whether irradiation of azirine 2 through the Pyrex filter also formed ketenimine 5, as observed for 3-methyl2-phenyl-2-azirine (Scheme 1). Most likely, the trans-cis isomerization of vinyl azide 1 is only observed upon irradiation through the 254 nm bandpass filter because the absorption coefficient of the cis-isomer of 1 is less than that of the trans-isomer at 254 nm. Thus, the concentration of cis-1 is not depleted as efficiently as that of the trans-isomer (Scheme 4). In contrast, upon irradiation above 300 nm, where both isomers likely have similar absorption coefficients, complete depletion of both isomers is achieved. However, the trans-cis isomerization can take place on either the triplet or the singlet surface of vinyl azide 1. 3.3. Calculations
Fig. 4. Difference spectra obtained of vinyl azide 1 in an argon matrix after irradiation for a) 10 min, b) 40 min, c) 60 min, and d) 18 h through a 254 nm bandpass filter.
1189, 982, 841, 772, 762, 702, and 538 cm1 to azirine 2. Further, irradiation did not result in a build-up of the concentration of azirine 2. Rather, a new set of bands were formed, the major ones located at 1905, 879, 863, 759, 729, and 693 cm1, which we assign to ylide 3 based on the similarity of these bands
To support the proposed photoreaction mechanisms for vinyl azide 1 and azirine 2, we performed DFT calculations using the Gaussian09 program [13] at the B3LYP level of theory with the 631 þ G(d) basis set [14,15]. We calculated the stationary points on the triplet and singlet surfaces of vinyl azide 1 and azirine 2 (Fig. 5). The ground state (S0) structures of vinyl azide 1 and azirine 2 were optimized, and TD-DFT calculations were used to estimate the vertical excitation energies of the first excited singlet state (S1) and the first and second triplet excited states (T1 and T2) of each molecule. Optimization of vinyl azide 1 on the triplet surface yielded triplet biradical 6. Spin density calculations revealed that the unpaired spin is localized on the a-carbon (spin density of 0.66), with approximately equivalent spin densities distributed across the phenyl ring at both the ortho- (0.20 and 0.21) and para-positions (0.23). In addition, the terminal N atom of the vinyl azide moiety has a spin density of 0.31. Therefore, biradical 6 is best described as a 1,2-biradical because the a- and b-carbon atoms have the highest
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
5
Fig. 5. Optimized structures of trans-1, cis-1, 2, triplet biradical 6, T1 of 2, vinylnitrene 4, ylide 2, ketenimine 5, and T1 of 5. Distances are in Å and dihedral angles are in degrees.
spin densities and the torsional angle of HeCeCeH is 85 , which minimizes interactions between the two radical centers with the same spin. In addition, the optimized structure of T1 of 2 is located 59 kcal/ mol above S0 of 2. The CHeCH bond of the azirine moiety in T1 of 2 is elongated compared with that in its S0 (Fig. 5). Spin density calculations revealed that the unpaired spin is localized on the N
and b-carbon atoms (0.69 and 0.68, respectively), with a smaller spin density on the a-carbon (0.36), and thus we assign T1 of 2 to a (n,p*) configuration. As B3LYP underestimates the energy of triplet ketones with (n,p*) configurations, it is likely that the energy of the T1 state of 2 is underestimated as well [20]. Optimization of triplet vinylnitrene 34 resulted in two minimal energy conformers A and B, which have the same energy. As
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
6
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
expected, the calculated spin densities demonstrated that the unpaired spin is mainly localized on the Ne and b-carbon atoms (1.3 and 0.6, respectively), which verifies the significant 1,3-biradical character of triplet vinylnitrene 34. The calculated rotational barrier for vinylnitrene conformer 34A to rotate around the vinyl bond into conformer 34B is 11 kcal/mol. The open-shell singlet vinylnitrene 14 was optimized using the broken symmetry method, as described in the literature. The broken symmetry calculations were achieved using guess ¼ mix as a keyword in Gaussian09, and they resulted in vinylnitrene 14 with a total spin (〈S2〉) value of 0.97, which have significant spin contamination from the triplet state [21,22]. Both S0 and T1 of ketenimine 5 were optimized, revealing that T1 of 5 is located 37 kcal/mol above its S0. Finally, optimization of ylide 3 showed that it is best described as an allenic ylide [23,24] because both the C]N double bonds are of similar length (Fig. 5). The calculated stationary points for the formation of triplet vinylnitrene 34 on the triplet surfaces of vinyl azide 1 and azirine 2 are displayed in Figs. 6 and 7, respectively. The calculated transition state barrier for biradical 6 to form triplet vinylnitrene 34 is only 1 kcal/mol, whereas the calculated transition state barrier for T1 of 2 to form triplet vinylnitrene 34 is 15 kcal/mol. However, as mentioned earlier, it is expected that B3LYP underestimates the energy of T1 of 2, and thus it is feasible for T1 of 2 to form vinylnitrene 34. Thus, the calculations support that it is feasible to form vinylnitrene 34 from either vinyl azide 1 or azirine 2; however, simple DFT calculations cannot be used to estimate whether intersystem crossing from the singlet excited states of vinyl azide 1 and azirine 2 to their triplet surfaces is feasible. In contrast, the transition state barrier for triplet vinylnitrene 34 to form T1 of 5 is 51 kcal/mol, and thus ketenimine 5 cannot be formed on the triplet surface. Rather, it can be theorized that vinylnitrene 34 intersystem crosses to ketenimine 5. We calculated stationary points on the singlet surface of vinyl azide 1 for the formation of azirine 2 and ketenimine 5 (Fig. 8). The calculated transition state barrier for azirine 2 to form ylide 3 is 45 kcal/mol. The calculated transition state barrier for vinyl azide 1 to form azirine 2 is only 30 kcal/mol. We were not successful in optimizing the transition state for forming ketenimine 5 from vinyl azide 1, although the calculated transition state barrier for the simplest vinyl azide, H2C]CHeN, to form ketenimine, H2C]C] NH, has been calculated using DFT [25]. We calculated the transition state barrier for forming singlet vinylnitrene 14 from vinyl azide 1 as 37 kcal/mol, whereas the calculated transition state barrier for singlet vinylnitrene 14 to form ketenimine 5 is 21 kcal/
Fig. 7. Calculated stationary energy points for the formation of ketenimine 5 on the triplet surface of azirine 2. The energy of S1 of 2 was obtained from TD-DFT calculations, whereas the other energies were acquired from optimization calculations. Energies are in kcal/mol.
Fig. 8. Calculated stationary energy points for the formation of azirine 2 and singlet vinylnitrene 14 on the singlet surface of vinyl azide 1, and the secondary reactions of azirine 2 to form ylide 3 and singlet vinylnitrene 14 to form ketenimine 5. The energy of S1 of 1 was obtained from TD-DFT calculations, whereas the other energies were acquired from optimization calculations. Energies are in kcal/mol.
mol. However, it should be noted that DFT calculations are not well suited for obtaining accurate energies for open-shell biradical singlets such as vinylnitrene 34, as Wentrup et al. and Nguyen et al. have highlighted [26,27]. 3.4. Laser flash photolysis
Fig. 6. Calculated stationary energy points for the formation of triplet vinylnitrene 34 on the triplet surface of vinyl azide 1. The energy of S1 of 1 was obtained from TD-DFT calculations, whereas the other energies were acquired from optimization calculations. Energies are in kcal/mol.
To verify whether ketenimine 5 is formed on the singlet or triplet surface of vinyl azide 1, we preformed laser flash photolysis of vinyl azide 1 and azirine 2 with an excimer laser (308 nm, 17 ns) [11]. Laser flash photolysis of azirine 2 in acetonitrile resulted in a weak transient spectrum with lmax at 440 nm (Fig. 9a). The decay of the transient absorption can be fitted as a pseudo-first-order decay with a rate constant of 1.0 105 s1 (t ¼ 10 ms; Fig. 9c). Because this absorption was quenched in oxygen-saturated acetonitrile, we assign it to vinylnitrene 34. Furthermore, the major adsorption bands in the TD-DFT calculated absorption spectrum of conformers A and B of triplet vinylnitrene 34 are located above 330 nm at 428 nm (f ¼ 0.05) and 401 nm (f ¼ 0.02) (Fig. 9b), corresponding well with the observed transient absorption. In contrast, laser flash photolysis of vinyl azide 1 in argonsaturated solution did not yield any significant transient absorption on the nanosecond timescale. Thus, laser flash photolysis verified that vinyl azide 1 does not intersystem cross to its triplet
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
7
Scheme 5. Photolysis of vinyl azide 7 in a cryogenic matrix [31].
Scheme 6. Calculated rotational barriers for triplet vinylnitrenes 34, 38, 39, and 310.
Scheme 7. Cyclic triplet vinylnitrenes.
Fig. 9. a) Transient absorption spectrum obtained by laser flash photolysis of azirine 2 in argon-saturated acetonitrile. b) TD-DFT calculated electronic transitions for triplet vinylnitrene 34 (conformers A and B). c) Kinetic trace at 440 nm obtained by laser flash photolysis of azirine 2 in argon-saturated acetonitrile.
surface to yield vinylnitrene 34, nor does vinyl azide 1 yield singlet vinylnitrene 14, which can intersystem cross to its triplet configuration.
4. Discussion Both short- and long-wavelength irradiation of vinyl azide 1 at cryogenic temperature yields azirine 2 and ketenimine 5 concurrently. Both products are formed via singlet reactivity, as laser flash photolysis of vinyl azide 1 at ambient temperature does not yield triplet vinylnitrene 34. Upon excitation, vinyl azide 1 does not intersystem cross to its triplet surface or form singlet vinylnitrene 4, which then intersystem crosses to its triplet configuration. Thus, direct photolysis of vinyl azide 1 yields only singlet reactivity. These results strengthen the notion that photolysis of vinyl azides generally does not yield triplet reactivity without the involvement of a triplet sensitizer [28e30]. Our findings for vinyl azide 1 parallel those reported by Lopes et al., who showed that irradiation of vinyl azide 7 in a cryogenic matrix results in concurrent formation of the corresponding azirine and ketenimine derivatives (Scheme 5) [31].
In contrast to that of vinyl azide 1, the photochemistry of azirine 2 is strongly wavelength dependent. Irradiation of azirine 2 with 254 nm light in an argon matrix results in the formation of ylide 3, whereas longer wavelength irradiation does not produce any ylide 3. In addition, laser flash photolysis of azirine 2 with excitation at 308 nm yields triplet vinylnitrene 34. Thus, the photochemistry of azirine 2 is wavelength dependent, with short wavelength irradiation resulting in singlet reactivity and longer wavelength irradiation resulting in triplet reactivity and the formation of triplet vinylnitrene 34. Triplet vinylnitrenes 38 and 39 (Scheme 6) have been detected directly by laser flash photolysis in solution, which reveals that they are short-lived intermediates [6,32,33], with lifetimes of a few microseconds, similar to triplet vinylnitrene 34. Further, they are not stable at cryogenic temperature, forming the corresponding ketenimines. We theorized that because vinylnitrenes 38 and 39 have significant 1,3-biradical character, the rotational barrier for rotation around the vinyl bond is small, and this flexibility aids them in obtaining the geometry necessary to intersystem cross to the singlet ground state. The calculated rotational barriers for triplet vinylnitrenes 38 and 39 are 10 and 11 kcal/mol, respectively, which are similar to that calculated for vinylnitrene 34 and explains why these vinylnitrenes have similar reactivity. In comparison, vinylnitrene 310 does not form the corresponding ketenimine in matrices, but rather forms an azirine. Thus, we theorized that the phenyl substituent on the vinyl bond in vinylnitrene 310 facilitates intersystem crossing to the azirine product rather than the ketenimine [34]. In contrast, triplet vinylnitrenes 311 and 312 (Scheme 7), which
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
8
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
are part of cyclic structures, are stable at cryogenic temperature and do not intersystem cross to form azirines or ketenimines [29,30]. These cyclic triplet vinylnitrenes are stable enough to be detected and characterized directly with ESR and IR spectroscopy. 5. Conclusion Thus, we have demonstrated that the photoreactivity of vinyl azide 1 yields both azirine 2 and ketenimine 5 through singlet reactivity, either concerted via the singlet excited state of vinyl azide 1 or through a singlet vinylnitrene intermediate. In contrast, the photochemistry of azirine 2 is wavelength dependent because the phenyl and the azirine chromophores are not conjugated. Thus, it is possible to excite these two chromophores separately. Short wavelength irradiation of azirine 2 in argon matrices results in singlet reactivity to form ylide 3, whereas long wavelength irradiation yields triplet vinylnitrene 34 in solution.
[14] [15]
[16] [17] [18]
Acknowledgements This work was supported by the National Science Foundation (CHE-1464694) and the Ohio Supercomputer Center (OSC). OO is grateful for a Doctoral Enhancement Fellowship from the Department of Chemistry at The University of Cincinnati.
[19] [20]
[21]
References [22] [1] M. Beer, H.C. Longuet-Higgins, Anomalous light emission of azulene, J. Chem. Phys. 23 (1955) 1390e1391. [2] G. Burdzinski, J.C. Hackett, J. Wang, T.L. Gustafson, C.M. Hadad, M.S. Platz, Early events in the photochemistry of aryl azides from femtosecond uv/vis spectroscopy and quantum chemical calculations, J. Am. Chem. Soc. 128 (2006) 13402e13411. [3] A. Bogdanova, V.V. Popik, Experimental and rheoretical investigation of reversible interconversion, thermal reactions, and wavelength-dependent photochemistry of diazo meldrum's acid and its diazirine isomer, 6,6dimethyl-5,7-dioxa-1,2-diaza-spiro[2,5]oct-1-ene-4,8-dione1, J. Am. Chem. Soc. 125 (2003) 14153e14162. [4] J.S. Poole, Recent advances in the photochemistry of heterocyclic n-oxides and their derivatives, in: O.V. Larionov (Ed.), Heterocyclic N-oxides, Springer International Publishing, 2017, pp. 111e151. Cham. [5] H. Inui, S. Murata, Photochemistry of 2-(1-naphthyl)-2h-azirines in matrixes and in solutions: Wavelength-dependent CC and CN bond cleavage of the azirine ring, J. Am. Chem. Soc. 127 (2005) 2628e2636. [6] X. Zhang, S.K. Sarkar, G.K. Weragoda, S. Rajam, B.S. Ault, A.D. Gudmundsdottir, Comparison of the photochemistry of 3-methyl-2-phenyl-2H-azirine and 2methyl-3-phenyl-2H-azirine, J. Org. Chem. 79 (2014) 653e663. [7] G.K. Weragoda, A. Das, S.K. Sarkar, H.D.M. Sriyarathne, X. Zhang, B.S. Ault, A.D. Gudmundsdottir, Singlet photoreactivity of 3-methyl-2-phenyl-2H-azirine, Aust. J. Chem. 70 (2017) 413e420. [8] C.-Z. Tao, X. Cui, J. Li, A.-X. Liu, L. Liu, Q.-X. Guo, Copper-catalyzed synthesis of aryl azides and 1-aryl-1,2,3-triazoles from boronic acids, Tetrahedron Lett. 48 (2007) 3525e3529. [9] A.G. O'Brien, F. Levesque, P.H. Seeberger, Continuous flow thermolysis of azidoacrylates for the synthesis of heterocycles and pharmaceutical intermediates, Chem. Commun 47 (2011) 2688e2690. [10] B.S. Ault, Infrared spectra of argon matrix-isolated alkali halide salt/water complexes, J. Am. Chem. Soc. 100 (1978) 2426e2433. [11] S. Muthukrishnan, J. Sankaranarayanan, R.F. Klima, T.C.S. Pace, C. Bohne, A.D. Gudmundsdottir, Intramolecular H-atom abstraction in g-azido-butyrophenones: formation of 1,5 ketyl iminyl radicals, Org. Lett. 11 (2009) 2345e2348. [12] H. Bader, H.-J. Hansen, Zur imidazolbildung aus 2H-azirinene bzw. Vinylaziden and nitrilen unter saurekatalyse, Helv. Chim. Acta 62 (1978) 286e304. [13] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
[23] [24]
[25] [26] [27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford CT, 2016. A.D. Becke, Density-functional thermochemistry. Iii. The role of exact exchange, J. Chem. Phys. 98 (1993) 5648e5652. C. Lee, W. Yang, R.G. Parr, Development of the colle-salvetti correlation-energy formula into a functional of the electron density, Phys. Rev. B Condens. Matter 37 (1988) 785e789. C. Gonzalez, H.B. Schlegel, Reaction path following in mass-weighted internal coordinates, J. Phys. Chem. 94 (1990) 5523e5527. C. Gonzalez, H.B. Schlegel, An improved algorithm for reaction path following, J. Chem. Phys. 90 (1989) 2154e2161. C. Wentrup, B. Freiermuth, Pyrolysis of benzotriazoles. Relationships between 1- and 2-vinylbenzotriazoles, a- and b-azidostyrenes, n-phenylketenimine and indole. Pitfalls in the use of pyrolysis-mass spectrometry in mechanistic studies, J. Anal. Appl. Pyrolysis 121 (2016) 67e74. J.B. Foresman, Æ. Frisch, Exploring Chemistry with Electronic Structure Methods, Gaussian Inc., Pittsburgh, PA, 1996. S. Muthukrishnan, S.M. Mandel, J.C. Hackett, P.N.D. Singh, C.M. Hadad, ttir, Competition between a-cleavage and enJ.A. Krause, A.D. Gudmundsdo ergy transfer in a-azidoacetophenones, J. Org. Chem. 72 (2007) 2757e2768. L. Noodleman, E.J. Baerends, Electronic structure, magnetic properties, esr, and optical spectra for 2-iron ferredoxin models by lcao-x.Alpha. Valence bond theory, J. Am. Chem. Soc. 106 (1984) 2316e2327. K. Yamaguchi, F. Jensen, A. Dorigo, K.N. Houk, A spin correction procedure for unrestricted Hartree-Fock and Møller-Plesset wavefunctions for singlet diradicals and polyradicals, Chem. Phys. Lett. 149 (1988) 537e542. gue , C. Wentrup, Carbenic nitrile imines: properties and reactivity, J. Org. D. Be Chem. 79 (2014) 1418e1426. gue , C. Addicott, R. Burgard, P. Bednarek, E. Guille, I. Baraille, C. Wentrup, D. Be Nitrile ylides: allenic and propargylic structures from pyrazinylnitrenes. Experimental and theoretical characterization, J. Org. Chem. 79 (2014) 2148e2155. D.J.R. Duarte, M.S. Miranda, J.C.G. Esteves da Silva, Computational study on the vinyl azide decomposition, J. Phys. Chem. 118 (2014) 5038e5045. C. Wentrup, C.M. Nunes, I. Reva, Comment on “computational study on the vinyl azide decomposition”, J. Phys. Chem. 118 (2014) 5122e5123. M.T. Nguyen, T.D. Hang, N.P. Gritsan, V.G. Kiselev, Comment on “computational study on the vinyl azide decomposition”, J. Phys. Chem. 119 (2015) 12906e12907. S. Rajam, A.V. Jadhav, Q. Li, S.K. Sarkar, P.N.D. Singh, A. Rohr, T.C.S. Pace, R. Li, J.A. Krause, C. Bohne, B.S. Ault, A.D. Gudmundsdottir, Triplet sensitized photolysis of a vinyl azide: direct detection of a triplet vinyl azide and nitrene, J. Org. Chem. 70 (19) (2014) 9325e9334 (Ahead of Print). S.K. Sarkar, O. Osisioma, W.L. Karney, M. Abe, A.D. Gudmundsdottir, Using molecular architecture to control the reactivity of a triplet vinylnitrene, J. Am. Chem. Soc. 138 (2016) 14905e14914. S.K. Sarkar, A. Sawai, K. Kanahara, C. Wentrup, M. Abe, A.D. Gudmundsdottir, Direct detection of a triplet vinylnitrene, 1,4-naphthoquinone-2-ylnitrene, in solution and cryogenic matrices, J. Am. Chem. Soc. 137 (2015) 4207e4214. mez-Zavaglia, T.M.V.D. Pinho e Melo, R. Fausto, S. Lopes, C.M. Nunes, A. Go Structure and photochemical behaviour of 3-azido-acrylophenones: a matrix isolation infrared spectroscopy study, Tetrahedron 67 (2011) 7794e7804. S. Rajam, R.S. Murthy, A.V. Jadhav, Q. Li, C. Keller, C. Carra, T.C.S. Pace, C. Bohne, B.S. Ault, A.D. Gudmundsdottir, Photolysis of (3-methyl-2h-azirin-2-yl)-phenylmethanone: direct detection of a triplet vinylnitrene intermediate, J. Org. Chem. 76 (2011) 9934e9945. D.W. Gamage, Q. Li, R.A.A.U. Ranaweera, S.K. Sarkar, G.K. Weragoda, P.L. Carr, A.D. Gudmundsdottir, Vinylnitrene formation from 3,5-diphenyl-isoxazole and 3-benzoyl-2-phenylazirine, J. Org. Chem. 78 (22) (2013) 11349e11356. H.D.M. Sriyarathne, S.K. Sarkar, S. Hatano, M. Abe, A.D. Gudmundsdottir, Photolysis of 3,5-diphenylisoxazole in argon matrices, J. Phys. Org. Chem. 30 (2017) e3638-n/a.
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
Contents lists available at ScienceDirect
Journal of Molecular Structure journal homepage: http://www.elsevier.com/locate/molstruc
Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine Onyinye Osisioma, Mrinal Chakraborty, Bruce S. Ault, Anna D. Gudmundsdottir* Department of Chemistry, University of Cincinnati, PO Box 210172, Cincinnati OH 45221, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 December 2017 Received in revised form 30 March 2018 Accepted 10 April 2018 Available online xxx
Irradiation of 2-azidovinylbenzene (1) in a cryogenic argon matrix with a 254 nm bandpass filter results in 2-phenyl-2H-azirine (2) as the major product, some ketenimine 5, and a small amount of cis-1, whereas irradiation of vinyl azide 1 with light above 300 nm forms only azirine 2 and ketenimine 5. Prolonged irradiation of azirine 2 through a 254 nm bandpass filter produces ylide 3 selectively. Laser flash photolysis with excitation at 308 nm demonstrated that azirine 2 forms triplet vinylnitrene 4 (l ~ 440 nm), whereas vinyl azide 1 does not exhibit transient absorption on a nanosecond timescale. DFT calculations (B3LYP/6-31 þ G(d)) were used to aid the characterization of the photoproducts and to support the proposed formation mechanisms. © 2018 Published by Elsevier B.V.
Keywords: Azirine Vinyl azide Argon matrix Wavelength-dependent photochemistry Laser flash photolysis Density functional theory calculation
1. Introduction Wavelength dependent photochemistry is fascinating because by simply using a different wavelength of light, the same reagent can be transformed into distinct products. The origin of wavelength dependent photochemistry is the excitation of separate chromophores in a reagent, resulting in the population of different excited states that react distinctively. It is also possible that excitation at different wavelengths can lead to the formation of different excited states of the same chromophore. Such examples are rare, as higher excited states generally undergo efficient internal conversion to the lowest excited state, as according to Kasha's rule, but a number of exceptions have been reported, such as azulene [1], naphthyl azides [2], and diazirines [3]. The photochemistry of nonconjugated aromatic azirine derivatives is extraordinary, as short wavelength irradiation leads to cleavage of the CeC bond of the azirine ring to form ylides, whereas longer wavelength irradiation results in the CeN bond breaking to form triplet vinylnitrenes [4,5]. For example, short wavelength irradiation of 3-methyl-2-phenyl-2H-azirine in an argon matrix yields the corresponding ylide, whereas longer wavelength
* Corresponding author. E-mail address: [email protected] (A.D. Gudmundsdottir).
irradiation forms the corresponding ketenimine (Scheme 1) [6,7]. We theorized that breakage of the CeC bond occurs following direct absorption by the azirine moiety. In contrast, longer wavelength irradiation selectively excites the phenyl moiety, which intersystem crosses to its triplet excited state, and subsequent energy transfer to the triplet excited state of the azirine chromophore results in cleavage of the CeN bond to form triplet vinylnitrene. However, as triplet vinylnitrene was not observed directly, it was theorized that the vinylnitrene intersystem crosses efficiently to form ketenimine. The hypothesis that the phenyl group serves as a triplet sensitizer was further supported by the fact that the photochemistry of 2-methyl-3-phenyl-2H-azirine, in which C]N is conjugated with the phenyl ring, does not display a wavelength dependence. In this case, the ylide is formed, regardless of the irradiation wavelength, as it is not possible to selectively irradiate the phenyl group of this molecule, and thus only singlet reactivity is observed (Scheme 2). In this article, we report the photochemistry of 2azidovinylbenzene (1) and 2-phenyl-2H-azirine (2) in cryogenic argon matrices as a function of irradiation wavelength. Because the phenyl chromophore is in conjugation with the vinyl azide moiety in 1, but is not conjugated with the C]N moiety in azirine 2, it is possible to compare their photochemistry to determine whether vinyl azide 1 reacts on its singlet or triplet surface upon direct irradiation. The photochemistry of azirine 2 is strongly wavelength
https://doi.org/10.1016/j.molstruc.2018.04.042 0022-2860/© 2018 Published by Elsevier B.V.
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
2
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
2.2. Matrix isolation Matrix isolation was performed using conventional equipment [10]. Vinyl azide 1 was deposited onto a CsI cold window cooled to 20 K by a CTI closed-cycle refrigerator. A stream of argon was passed over the volatile oil under vacuum to entrap vinyl azide 1 into the argon matrix. In two different sets of experiments, the argon matrix was irradiated over short time intervals with a medium-pressure short-arc 200 W mercury arc lamp through a Pyrex filter (>300 nm) or a 254 nm bandpass filter, which transmitted light between wavelengths of 233e271 nm. Infrared spectra were recorded at a resolution of 1 cm1 using a PerkinElmer Spectrum One FTIR spectrometer during and after matrix deposition and intermittently during irradiation. 2.3. Laser flash photolysis Laser flash photolysis was carried out using an excimer laser (308 nm, 17 ns) [11]. In a typical experiment, a stock solution of vinyl azide 1 or azirine 2 was prepared in spectroscopic-grade acetonitrile, such that the solution had an absorbance of 0.3e0.6 at 308 nm.
2.4. Photolysis of 1
Scheme 1. Photolysis of 3-methyl-2-phenyl-2H-azirine in argon matrices.
Scheme 2. Photolysis of 2-methyl-3-phenyl-2H-azirine in argon matrices.
dependent, whereas the reactivity of vinyl azide 1 is relatively less affected by the irradiation wavelength.
Vinyl azide 1 was dissolved in chloroform-d1 in a Pyrex NMR tube sealed with a rubber septum, and argon was bubbled through the solution for 17 min. The resulting solution was irradiated with 10 light-emitting diodes (LEDs; 450 ± 5 nm) in parallel alignment, each with a constant current of 2 mA and a voltage drop of 3.2e3.8 V across each diode. After irradiation for 17 min, 1H NMR spectroscopy of the reaction mixture showed the formation of azirine 2 (76%) and remaining starting material (24%). Preparative preparation of azirine 2. Vinyl azide 1 (17 mg, 0.12 mmol) was dissolved in chloroform-d1 (0.6 mL, 7.49 mmol) and purged with argon for 5 min in a Pyrex NMR tube, which was sealed with a rubber septa cap and irradiated with a mercury-argon lamp for 40 min 1H NMR spectroscopy verified the formation of azirine 2 (51% yields) [12].1H NMR (CDCl3, 400 MHz): 9.99 (d, J ¼ 2 Hz, 1H), 7.2 (m, 5H), 2.72 (d, J ¼ 2 Hz, 1H) ppm. 2.5. Calculations
2. Experimental 2.1. Synthesis of 1 Vinyl azide 1 was synthesized according to a published procedure [8]. A catalytic amount of CuSO4 (80 mg, 0.50 mmol) was added to NaN3 (390 mg, 6.0 mmol) and dissolved in methanol (15 mL). Trans-2-phenylvinylboronic acid (0.739 g, 5.0 mmol) was added to the solution, and the resulting mixture was stirred until thin layer chromatography (TLC) showed complete depletion of the starting material and formation of a new product. The solvent was removed under vacuum and the resulting oil was dissolved in petroleum ether. After washing with water, the organic layer was dried over Na2SO4. The solvent was removed under vacuum, providing vinyl azide 1 as an oil (0.4048 g, 2.79 mmol, 56% yield). Characterization of vinyl azide 1 by IR and 1H NMR spectroscopy agreed well with the data reported in the literature [9]. 1H NMR (CDCl3, 400 MHz): 6.18 (d, J ¼ 16 Hz, 1H), 6.83 (d, J ¼ 16 Hz, 1H), 7.2e7.3 (m, 5H) ppm. IR (CDCl3): 2108, 1636, 1260, 1019, 930, 798, 747, 692 cm1.
The geometries were optimized at the B3LYP level of theory with the 6-31þG(d) basis set, as implemented in the Gaussian09 program [13e15]. The transition states were confirmed to have one imaginary vibrational frequency, and intrinsic reaction coordinate calculations were used to verify that the transition states corresponded to the attributed reactant and product [16,17]. To obtain the calculated absorption spectra for the excited states and intermediates, time-dependent density functional theory (TD-DFT) calculations were conducted at the same level of theory with the same basis set.
3. Results 3.1. Photolysis of vinyl azide 1 in solution A solution of vinyl azide 1 in chloroform,-d1 was exposed to 450 ± 5 nm LED light for 17 min 1H NMR spectroscopy revealed the formation of azirine 2 (76%) and remaining starting material (Scheme 3).
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
Scheme 3. Photolysis of vinyl azide 1 in acetonitrile using 450 ± 5 nm LED light.
3.2. Matrix isolation Vinyl azide 1 was deposited in an argon matrix and irradiated with a medium-pressure short-arc 200 W lamp through a Pyrex filter. Irradiation of the matrix resulted in depletion of the bands of azide 1 at 2102, 1644, 1603, 1329, 1289, 1265, 970, 752, and 693 cm1 (Fig. 1a). After 46 min of irradiation, the bands corresponding to vinyl azide 1 were almost fully depleted. Concurrent with the depletion of the bands of vinyl azide 1, new bands were formed, the most significant of which were located at 2024, 1662, 1498, 1457, 1189, 982, 841, 772, 763, 702, and 538 cm1. The bands observed at 1662, 1498, 1457, 1189, 982, 841, 772, 763, 702, and 538 cm1 were assigned to azirine 2 as they match well with its calculated bands (Fig. 1b), and with its spectrum reported earlier by Wentrup and Freiermuth [18]. Most of the IR bands are due to
Fig. 1. a) Difference spectrum of vinyl azide 1 in an argon matrix before and after 46 min of irradiation through a Pyrex filter. Calculated and scaled spectra of b) azirine 2 and c) ketenimine 5.
3
vibrations of the phenyl moiety of azirine 1, but of those that belong to the azirine moiety, the most significant are the C]N stretch and the bending of the CH bonds. Calculations and scaling with 0.9613 [19] place the azirine C]N band at 1684 cm1, which fits well with the observed band at 1662 cm1. Similarly, we assign the bands at 841 and 763 cm1 to CH bending of the azirine ring, which are calculated and scaled to be at 815 and 754 cm1. In addition, a small band was observed at 2024 (Figs. 1c and 2a). We assign this band to ketenimine 5 based on its reported [18] and calculated spectra. The calculated spectrum has the most intense band at 2026 cm1 (calculated intensity ¼ 631). The broad band at 982 cm1 which is assigned to azirine 2, can also be assigned to ketenimine 5 as it has significant band at 976 cm1 (calculated intensity ¼ 316) after scaling (Fig. 2b). In a separate experiment, vinyl azide 1 was deposited in an argon matrix and irradiated with a mercury arc lamp using a 254 nm bandpass filter (Fig. 3a). The bands associated with vinyl azide 1 were almost fully depleted after irradiation for 18 h (Fig. 4). Concurrent with the depletion of the bands corresponding to vinyl azide 1, new bands were formed. After 40 min of irradiation, new bands were observed at 2024, 1662, 1498, 1457, 1189, 982, 841, 772, 762, 702, and 538 cm1, and as before, we assign the bands at 2024 and 982 cm1 to ketenimine 5 and the bands at 1662, 1498, 1457,
Fig. 2. a) Ketenimine band observed at 2024 cm1 in the difference spectrum of vinyl azide 1 in an argon matrix after 46 min of irradiation through Pyrex filter. b) TD-DFT calculated and scaled spectrum of ketenimine at 2024 cm1.
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
4
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
Scheme 4. Photoproducts obtained by irradiation of vinyl azide 1 and azirine 2 in argon matrices.
Fig. 3. a) Difference spectrum of vinyl azide 1 in an argon matrix after irradiation for 18 h through a 254 nm bandpass filter. Calculated and scaled IR spectra of b) ketenimine 5, c) ylide 3, and d) the cis-isomer of vinyl azide 1.
with the calculated IR spectrum (Fig. 3c). Additionally, bands at 2153 and 2024 cm1 were also formed, and as before the 2024 cm1 band is assigned to ketenimine 5, whereas we theorize that the band at 2153 cm1 is due to the formation of the cis-isomer of vinyl azide 1 (Fig. 3d). The calculated and scaled azido band for cis-1 is at 2164 cm1, which matches well with the observed band at 2153 cm1. Nevertheless, no additional bands were assigned to the cis-isomer because it is not formed in significant quantities. The irradiation of vinyl azide 1 through both 254 nm and Pyrex filters in argon matrices results in azirine 2 and ketenimine 5, and these products can be formed on either the triplet or singlet surface of vinyl azide 1. In contrast, ylide 3, which must be formed on the singlet surface of azirine 2, is only obtained by irradiation of azirine 2 through the 254 nm bandpass filter. It was not possible to unambiguously determine whether irradiation of azirine 2 through the Pyrex filter also formed ketenimine 5, as observed for 3-methyl2-phenyl-2-azirine (Scheme 1). Most likely, the trans-cis isomerization of vinyl azide 1 is only observed upon irradiation through the 254 nm bandpass filter because the absorption coefficient of the cis-isomer of 1 is less than that of the trans-isomer at 254 nm. Thus, the concentration of cis-1 is not depleted as efficiently as that of the trans-isomer (Scheme 4). In contrast, upon irradiation above 300 nm, where both isomers likely have similar absorption coefficients, complete depletion of both isomers is achieved. However, the trans-cis isomerization can take place on either the triplet or the singlet surface of vinyl azide 1. 3.3. Calculations
Fig. 4. Difference spectra obtained of vinyl azide 1 in an argon matrix after irradiation for a) 10 min, b) 40 min, c) 60 min, and d) 18 h through a 254 nm bandpass filter.
1189, 982, 841, 772, 762, 702, and 538 cm1 to azirine 2. Further, irradiation did not result in a build-up of the concentration of azirine 2. Rather, a new set of bands were formed, the major ones located at 1905, 879, 863, 759, 729, and 693 cm1, which we assign to ylide 3 based on the similarity of these bands
To support the proposed photoreaction mechanisms for vinyl azide 1 and azirine 2, we performed DFT calculations using the Gaussian09 program [13] at the B3LYP level of theory with the 631 þ G(d) basis set [14,15]. We calculated the stationary points on the triplet and singlet surfaces of vinyl azide 1 and azirine 2 (Fig. 5). The ground state (S0) structures of vinyl azide 1 and azirine 2 were optimized, and TD-DFT calculations were used to estimate the vertical excitation energies of the first excited singlet state (S1) and the first and second triplet excited states (T1 and T2) of each molecule. Optimization of vinyl azide 1 on the triplet surface yielded triplet biradical 6. Spin density calculations revealed that the unpaired spin is localized on the a-carbon (spin density of 0.66), with approximately equivalent spin densities distributed across the phenyl ring at both the ortho- (0.20 and 0.21) and para-positions (0.23). In addition, the terminal N atom of the vinyl azide moiety has a spin density of 0.31. Therefore, biradical 6 is best described as a 1,2-biradical because the a- and b-carbon atoms have the highest
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
5
Fig. 5. Optimized structures of trans-1, cis-1, 2, triplet biradical 6, T1 of 2, vinylnitrene 4, ylide 2, ketenimine 5, and T1 of 5. Distances are in Å and dihedral angles are in degrees.
spin densities and the torsional angle of HeCeCeH is 85 , which minimizes interactions between the two radical centers with the same spin. In addition, the optimized structure of T1 of 2 is located 59 kcal/ mol above S0 of 2. The CHeCH bond of the azirine moiety in T1 of 2 is elongated compared with that in its S0 (Fig. 5). Spin density calculations revealed that the unpaired spin is localized on the N
and b-carbon atoms (0.69 and 0.68, respectively), with a smaller spin density on the a-carbon (0.36), and thus we assign T1 of 2 to a (n,p*) configuration. As B3LYP underestimates the energy of triplet ketones with (n,p*) configurations, it is likely that the energy of the T1 state of 2 is underestimated as well [20]. Optimization of triplet vinylnitrene 34 resulted in two minimal energy conformers A and B, which have the same energy. As
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
6
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
expected, the calculated spin densities demonstrated that the unpaired spin is mainly localized on the Ne and b-carbon atoms (1.3 and 0.6, respectively), which verifies the significant 1,3-biradical character of triplet vinylnitrene 34. The calculated rotational barrier for vinylnitrene conformer 34A to rotate around the vinyl bond into conformer 34B is 11 kcal/mol. The open-shell singlet vinylnitrene 14 was optimized using the broken symmetry method, as described in the literature. The broken symmetry calculations were achieved using guess ¼ mix as a keyword in Gaussian09, and they resulted in vinylnitrene 14 with a total spin (〈S2〉) value of 0.97, which have significant spin contamination from the triplet state [21,22]. Both S0 and T1 of ketenimine 5 were optimized, revealing that T1 of 5 is located 37 kcal/mol above its S0. Finally, optimization of ylide 3 showed that it is best described as an allenic ylide [23,24] because both the C]N double bonds are of similar length (Fig. 5). The calculated stationary points for the formation of triplet vinylnitrene 34 on the triplet surfaces of vinyl azide 1 and azirine 2 are displayed in Figs. 6 and 7, respectively. The calculated transition state barrier for biradical 6 to form triplet vinylnitrene 34 is only 1 kcal/mol, whereas the calculated transition state barrier for T1 of 2 to form triplet vinylnitrene 34 is 15 kcal/mol. However, as mentioned earlier, it is expected that B3LYP underestimates the energy of T1 of 2, and thus it is feasible for T1 of 2 to form vinylnitrene 34. Thus, the calculations support that it is feasible to form vinylnitrene 34 from either vinyl azide 1 or azirine 2; however, simple DFT calculations cannot be used to estimate whether intersystem crossing from the singlet excited states of vinyl azide 1 and azirine 2 to their triplet surfaces is feasible. In contrast, the transition state barrier for triplet vinylnitrene 34 to form T1 of 5 is 51 kcal/mol, and thus ketenimine 5 cannot be formed on the triplet surface. Rather, it can be theorized that vinylnitrene 34 intersystem crosses to ketenimine 5. We calculated stationary points on the singlet surface of vinyl azide 1 for the formation of azirine 2 and ketenimine 5 (Fig. 8). The calculated transition state barrier for azirine 2 to form ylide 3 is 45 kcal/mol. The calculated transition state barrier for vinyl azide 1 to form azirine 2 is only 30 kcal/mol. We were not successful in optimizing the transition state for forming ketenimine 5 from vinyl azide 1, although the calculated transition state barrier for the simplest vinyl azide, H2C]CHeN, to form ketenimine, H2C]C] NH, has been calculated using DFT [25]. We calculated the transition state barrier for forming singlet vinylnitrene 14 from vinyl azide 1 as 37 kcal/mol, whereas the calculated transition state barrier for singlet vinylnitrene 14 to form ketenimine 5 is 21 kcal/
Fig. 7. Calculated stationary energy points for the formation of ketenimine 5 on the triplet surface of azirine 2. The energy of S1 of 2 was obtained from TD-DFT calculations, whereas the other energies were acquired from optimization calculations. Energies are in kcal/mol.
Fig. 8. Calculated stationary energy points for the formation of azirine 2 and singlet vinylnitrene 14 on the singlet surface of vinyl azide 1, and the secondary reactions of azirine 2 to form ylide 3 and singlet vinylnitrene 14 to form ketenimine 5. The energy of S1 of 1 was obtained from TD-DFT calculations, whereas the other energies were acquired from optimization calculations. Energies are in kcal/mol.
mol. However, it should be noted that DFT calculations are not well suited for obtaining accurate energies for open-shell biradical singlets such as vinylnitrene 34, as Wentrup et al. and Nguyen et al. have highlighted [26,27]. 3.4. Laser flash photolysis
Fig. 6. Calculated stationary energy points for the formation of triplet vinylnitrene 34 on the triplet surface of vinyl azide 1. The energy of S1 of 1 was obtained from TD-DFT calculations, whereas the other energies were acquired from optimization calculations. Energies are in kcal/mol.
To verify whether ketenimine 5 is formed on the singlet or triplet surface of vinyl azide 1, we preformed laser flash photolysis of vinyl azide 1 and azirine 2 with an excimer laser (308 nm, 17 ns) [11]. Laser flash photolysis of azirine 2 in acetonitrile resulted in a weak transient spectrum with lmax at 440 nm (Fig. 9a). The decay of the transient absorption can be fitted as a pseudo-first-order decay with a rate constant of 1.0 105 s1 (t ¼ 10 ms; Fig. 9c). Because this absorption was quenched in oxygen-saturated acetonitrile, we assign it to vinylnitrene 34. Furthermore, the major adsorption bands in the TD-DFT calculated absorption spectrum of conformers A and B of triplet vinylnitrene 34 are located above 330 nm at 428 nm (f ¼ 0.05) and 401 nm (f ¼ 0.02) (Fig. 9b), corresponding well with the observed transient absorption. In contrast, laser flash photolysis of vinyl azide 1 in argonsaturated solution did not yield any significant transient absorption on the nanosecond timescale. Thus, laser flash photolysis verified that vinyl azide 1 does not intersystem cross to its triplet
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
7
Scheme 5. Photolysis of vinyl azide 7 in a cryogenic matrix [31].
Scheme 6. Calculated rotational barriers for triplet vinylnitrenes 34, 38, 39, and 310.
Scheme 7. Cyclic triplet vinylnitrenes.
Fig. 9. a) Transient absorption spectrum obtained by laser flash photolysis of azirine 2 in argon-saturated acetonitrile. b) TD-DFT calculated electronic transitions for triplet vinylnitrene 34 (conformers A and B). c) Kinetic trace at 440 nm obtained by laser flash photolysis of azirine 2 in argon-saturated acetonitrile.
surface to yield vinylnitrene 34, nor does vinyl azide 1 yield singlet vinylnitrene 14, which can intersystem cross to its triplet configuration.
4. Discussion Both short- and long-wavelength irradiation of vinyl azide 1 at cryogenic temperature yields azirine 2 and ketenimine 5 concurrently. Both products are formed via singlet reactivity, as laser flash photolysis of vinyl azide 1 at ambient temperature does not yield triplet vinylnitrene 34. Upon excitation, vinyl azide 1 does not intersystem cross to its triplet surface or form singlet vinylnitrene 4, which then intersystem crosses to its triplet configuration. Thus, direct photolysis of vinyl azide 1 yields only singlet reactivity. These results strengthen the notion that photolysis of vinyl azides generally does not yield triplet reactivity without the involvement of a triplet sensitizer [28e30]. Our findings for vinyl azide 1 parallel those reported by Lopes et al., who showed that irradiation of vinyl azide 7 in a cryogenic matrix results in concurrent formation of the corresponding azirine and ketenimine derivatives (Scheme 5) [31].
In contrast to that of vinyl azide 1, the photochemistry of azirine 2 is strongly wavelength dependent. Irradiation of azirine 2 with 254 nm light in an argon matrix results in the formation of ylide 3, whereas longer wavelength irradiation does not produce any ylide 3. In addition, laser flash photolysis of azirine 2 with excitation at 308 nm yields triplet vinylnitrene 34. Thus, the photochemistry of azirine 2 is wavelength dependent, with short wavelength irradiation resulting in singlet reactivity and longer wavelength irradiation resulting in triplet reactivity and the formation of triplet vinylnitrene 34. Triplet vinylnitrenes 38 and 39 (Scheme 6) have been detected directly by laser flash photolysis in solution, which reveals that they are short-lived intermediates [6,32,33], with lifetimes of a few microseconds, similar to triplet vinylnitrene 34. Further, they are not stable at cryogenic temperature, forming the corresponding ketenimines. We theorized that because vinylnitrenes 38 and 39 have significant 1,3-biradical character, the rotational barrier for rotation around the vinyl bond is small, and this flexibility aids them in obtaining the geometry necessary to intersystem cross to the singlet ground state. The calculated rotational barriers for triplet vinylnitrenes 38 and 39 are 10 and 11 kcal/mol, respectively, which are similar to that calculated for vinylnitrene 34 and explains why these vinylnitrenes have similar reactivity. In comparison, vinylnitrene 310 does not form the corresponding ketenimine in matrices, but rather forms an azirine. Thus, we theorized that the phenyl substituent on the vinyl bond in vinylnitrene 310 facilitates intersystem crossing to the azirine product rather than the ketenimine [34]. In contrast, triplet vinylnitrenes 311 and 312 (Scheme 7), which
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042
8
O. Osisioma et al. / Journal of Molecular Structure xxx (2018) 1e8
are part of cyclic structures, are stable at cryogenic temperature and do not intersystem cross to form azirines or ketenimines [29,30]. These cyclic triplet vinylnitrenes are stable enough to be detected and characterized directly with ESR and IR spectroscopy. 5. Conclusion Thus, we have demonstrated that the photoreactivity of vinyl azide 1 yields both azirine 2 and ketenimine 5 through singlet reactivity, either concerted via the singlet excited state of vinyl azide 1 or through a singlet vinylnitrene intermediate. In contrast, the photochemistry of azirine 2 is wavelength dependent because the phenyl and the azirine chromophores are not conjugated. Thus, it is possible to excite these two chromophores separately. Short wavelength irradiation of azirine 2 in argon matrices results in singlet reactivity to form ylide 3, whereas long wavelength irradiation yields triplet vinylnitrene 34 in solution.
[14] [15]
[16] [17] [18]
Acknowledgements This work was supported by the National Science Foundation (CHE-1464694) and the Ohio Supercomputer Center (OSC). OO is grateful for a Doctoral Enhancement Fellowship from the Department of Chemistry at The University of Cincinnati.
[19] [20]
[21]
References [22] [1] M. Beer, H.C. Longuet-Higgins, Anomalous light emission of azulene, J. Chem. Phys. 23 (1955) 1390e1391. [2] G. Burdzinski, J.C. Hackett, J. Wang, T.L. Gustafson, C.M. Hadad, M.S. Platz, Early events in the photochemistry of aryl azides from femtosecond uv/vis spectroscopy and quantum chemical calculations, J. Am. Chem. Soc. 128 (2006) 13402e13411. [3] A. Bogdanova, V.V. Popik, Experimental and rheoretical investigation of reversible interconversion, thermal reactions, and wavelength-dependent photochemistry of diazo meldrum's acid and its diazirine isomer, 6,6dimethyl-5,7-dioxa-1,2-diaza-spiro[2,5]oct-1-ene-4,8-dione1, J. Am. Chem. Soc. 125 (2003) 14153e14162. [4] J.S. Poole, Recent advances in the photochemistry of heterocyclic n-oxides and their derivatives, in: O.V. Larionov (Ed.), Heterocyclic N-oxides, Springer International Publishing, 2017, pp. 111e151. Cham. [5] H. Inui, S. Murata, Photochemistry of 2-(1-naphthyl)-2h-azirines in matrixes and in solutions: Wavelength-dependent CC and CN bond cleavage of the azirine ring, J. Am. Chem. Soc. 127 (2005) 2628e2636. [6] X. Zhang, S.K. Sarkar, G.K. Weragoda, S. Rajam, B.S. Ault, A.D. Gudmundsdottir, Comparison of the photochemistry of 3-methyl-2-phenyl-2H-azirine and 2methyl-3-phenyl-2H-azirine, J. Org. Chem. 79 (2014) 653e663. [7] G.K. Weragoda, A. Das, S.K. Sarkar, H.D.M. Sriyarathne, X. Zhang, B.S. Ault, A.D. Gudmundsdottir, Singlet photoreactivity of 3-methyl-2-phenyl-2H-azirine, Aust. J. Chem. 70 (2017) 413e420. [8] C.-Z. Tao, X. Cui, J. Li, A.-X. Liu, L. Liu, Q.-X. Guo, Copper-catalyzed synthesis of aryl azides and 1-aryl-1,2,3-triazoles from boronic acids, Tetrahedron Lett. 48 (2007) 3525e3529. [9] A.G. O'Brien, F. Levesque, P.H. Seeberger, Continuous flow thermolysis of azidoacrylates for the synthesis of heterocycles and pharmaceutical intermediates, Chem. Commun 47 (2011) 2688e2690. [10] B.S. Ault, Infrared spectra of argon matrix-isolated alkali halide salt/water complexes, J. Am. Chem. Soc. 100 (1978) 2426e2433. [11] S. Muthukrishnan, J. Sankaranarayanan, R.F. Klima, T.C.S. Pace, C. Bohne, A.D. Gudmundsdottir, Intramolecular H-atom abstraction in g-azido-butyrophenones: formation of 1,5 ketyl iminyl radicals, Org. Lett. 11 (2009) 2345e2348. [12] H. Bader, H.-J. Hansen, Zur imidazolbildung aus 2H-azirinene bzw. Vinylaziden and nitrilen unter saurekatalyse, Helv. Chim. Acta 62 (1978) 286e304. [13] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
[23] [24]
[25] [26] [27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford CT, 2016. A.D. Becke, Density-functional thermochemistry. Iii. The role of exact exchange, J. Chem. Phys. 98 (1993) 5648e5652. C. Lee, W. Yang, R.G. Parr, Development of the colle-salvetti correlation-energy formula into a functional of the electron density, Phys. Rev. B Condens. Matter 37 (1988) 785e789. C. Gonzalez, H.B. Schlegel, Reaction path following in mass-weighted internal coordinates, J. Phys. Chem. 94 (1990) 5523e5527. C. Gonzalez, H.B. Schlegel, An improved algorithm for reaction path following, J. Chem. Phys. 90 (1989) 2154e2161. C. Wentrup, B. Freiermuth, Pyrolysis of benzotriazoles. Relationships between 1- and 2-vinylbenzotriazoles, a- and b-azidostyrenes, n-phenylketenimine and indole. Pitfalls in the use of pyrolysis-mass spectrometry in mechanistic studies, J. Anal. Appl. Pyrolysis 121 (2016) 67e74. J.B. Foresman, Æ. Frisch, Exploring Chemistry with Electronic Structure Methods, Gaussian Inc., Pittsburgh, PA, 1996. S. Muthukrishnan, S.M. Mandel, J.C. Hackett, P.N.D. Singh, C.M. Hadad, ttir, Competition between a-cleavage and enJ.A. Krause, A.D. Gudmundsdo ergy transfer in a-azidoacetophenones, J. Org. Chem. 72 (2007) 2757e2768. L. Noodleman, E.J. Baerends, Electronic structure, magnetic properties, esr, and optical spectra for 2-iron ferredoxin models by lcao-x.Alpha. Valence bond theory, J. Am. Chem. Soc. 106 (1984) 2316e2327. K. Yamaguchi, F. Jensen, A. Dorigo, K.N. Houk, A spin correction procedure for unrestricted Hartree-Fock and Møller-Plesset wavefunctions for singlet diradicals and polyradicals, Chem. Phys. Lett. 149 (1988) 537e542. gue , C. Wentrup, Carbenic nitrile imines: properties and reactivity, J. Org. D. Be Chem. 79 (2014) 1418e1426. gue , C. Addicott, R. Burgard, P. Bednarek, E. Guille, I. Baraille, C. Wentrup, D. Be Nitrile ylides: allenic and propargylic structures from pyrazinylnitrenes. Experimental and theoretical characterization, J. Org. Chem. 79 (2014) 2148e2155. D.J.R. Duarte, M.S. Miranda, J.C.G. Esteves da Silva, Computational study on the vinyl azide decomposition, J. Phys. Chem. 118 (2014) 5038e5045. C. Wentrup, C.M. Nunes, I. Reva, Comment on “computational study on the vinyl azide decomposition”, J. Phys. Chem. 118 (2014) 5122e5123. M.T. Nguyen, T.D. Hang, N.P. Gritsan, V.G. Kiselev, Comment on “computational study on the vinyl azide decomposition”, J. Phys. Chem. 119 (2015) 12906e12907. S. Rajam, A.V. Jadhav, Q. Li, S.K. Sarkar, P.N.D. Singh, A. Rohr, T.C.S. Pace, R. Li, J.A. Krause, C. Bohne, B.S. Ault, A.D. Gudmundsdottir, Triplet sensitized photolysis of a vinyl azide: direct detection of a triplet vinyl azide and nitrene, J. Org. Chem. 70 (19) (2014) 9325e9334 (Ahead of Print). S.K. Sarkar, O. Osisioma, W.L. Karney, M. Abe, A.D. Gudmundsdottir, Using molecular architecture to control the reactivity of a triplet vinylnitrene, J. Am. Chem. Soc. 138 (2016) 14905e14914. S.K. Sarkar, A. Sawai, K. Kanahara, C. Wentrup, M. Abe, A.D. Gudmundsdottir, Direct detection of a triplet vinylnitrene, 1,4-naphthoquinone-2-ylnitrene, in solution and cryogenic matrices, J. Am. Chem. Soc. 137 (2015) 4207e4214. mez-Zavaglia, T.M.V.D. Pinho e Melo, R. Fausto, S. Lopes, C.M. Nunes, A. Go Structure and photochemical behaviour of 3-azido-acrylophenones: a matrix isolation infrared spectroscopy study, Tetrahedron 67 (2011) 7794e7804. S. Rajam, R.S. Murthy, A.V. Jadhav, Q. Li, C. Keller, C. Carra, T.C.S. Pace, C. Bohne, B.S. Ault, A.D. Gudmundsdottir, Photolysis of (3-methyl-2h-azirin-2-yl)-phenylmethanone: direct detection of a triplet vinylnitrene intermediate, J. Org. Chem. 76 (2011) 9934e9945. D.W. Gamage, Q. Li, R.A.A.U. Ranaweera, S.K. Sarkar, G.K. Weragoda, P.L. Carr, A.D. Gudmundsdottir, Vinylnitrene formation from 3,5-diphenyl-isoxazole and 3-benzoyl-2-phenylazirine, J. Org. Chem. 78 (22) (2013) 11349e11356. H.D.M. Sriyarathne, S.K. Sarkar, S. Hatano, M. Abe, A.D. Gudmundsdottir, Photolysis of 3,5-diphenylisoxazole in argon matrices, J. Phys. Org. Chem. 30 (2017) e3638-n/a.
Please cite this article in press as: O. Osisioma, et al., Wavelength-dependent photochemistry of 2-azidovinylbenzene and 2-phenyl-2H-azirine, Journal of Molecular Structure (2018), https://doi.org/10.1016/j.molstruc.2018.04.042