Phenylnitrene Radical Cation Rearrangements - The Journal of

Oct 9, 2018 - Flow system does the dirty work. Organic chemists optimizing a reaction, like chefs perfecting a dish, execute a single transformation...
1 downloads 0 Views 1MB Size
Article pubs.acs.org/JPCA

Cite This: J. Phys. Chem. A XXXX, XXX, XXX−XXX

Phenylnitrene Radical Cation Rearrangements Didier Bégué,*,† Alain Dargelos,† and Curt Wentrup*,‡ †

CNRS/Université de Pau et des Pays de l’Adour/E2S UPPA, Institut des Sciences Analytiques et de Physicochimie pour l’Environnement et les Matériaux, UMR5254, 64000 Pau, France ‡ School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Queensland 4072, Australia

Downloaded via KAOHSIUNG MEDICAL UNIV on October 24, 2018 at 19:18:00 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: The electronic structure and the rearrangements of the phenylnitrene radical cation C6H5N.+ 2.+ have been investigated at DFT and CASPT2(7,9) levels of theory. The 2B2 state has the lowest energy of five identified electronic states, and it can undergo ring expansion to the 1-azacycloheptetetraene radical cation 4.+ with an activation energy of ca. 28 kcal/mol. Ring opening and recyclization provide a route to 5cyanocyclopentadiene radical cation 8.+, which may undergo facile 1,5-hydrogen shifts. The 2-, 3-, and 4-pyridylcarbene radical cations 31.+, 35.+ , and 39.+ interconvert with the phenylnitrene radical cation via azacycloheptatetraenes with activation barriers 72 kcal/mol) (see the Supporting Information for details). There are no other states of B2 or A1 symmetry between the 2B2 and 2B1 states. The energies of all ions calculated below will be referenced to the 2B2 state of 2.+. The calculated ionization energy for the formation of the 2B2 state of 2.+ from phenylnitrene is 8.2 eV in good agreement with the experimental values (8.04−8.21 eV).6,7

The ring opening to a cyanodienylcarbene radical cation 28.+ has a significantly higher activation barrier of about 48 kcal/ mol, but it would still be highly feasible under the 70 eV reaction conditions. The ion 28.+ can also be formed by cleaving a C−N bond in 4.+. Recyclization of 28.+ constitutes a potential route to 5-cyanocyclopentadienes 8.+. A facile 1,5-H shift in 8.+ yields the 1-cyanocyclopentadiene ion 9.+ with an activation energy of ca. 27 kcal/mol, and further 1,5-shifts of H and CN will cause complete hydrogen scrambling in the ions, thereby accounting for the hydrogen randomization observed in the mass spectra of deuterated phenyl azide.18 The activation energy of ∼27 kcal/mol is a little lower than the barriers for H-shifts in neutral cyclopentadienes.19,20 In addition, 28.+ may undergo a 1,2-H shift to yield the species 29.+, which is analogous to the known photochemical rearrangements of 3-pyridylnitrene and 3-pyridylcarbene.21 Ion 28.+ can also eliminate acetylene, HCCH, thereby providing one of the several possible routes to m/z 65 in the mass spectra of C6H5N.+ precursors (eq 2). In this reaction the energy rises monotonously without the occurrence of a discrete transition state until the products are formed about 63 kcal/mol above 28.+ (i.e., the reverse reaction is essentially barrierless).

High-resolution measurements of the masses in the range 50−91 Da demonstrate very similar compositions of isobaric ions from different precursors and show that m/z 65 is largely formed by loss of C2H2, and m/z 64 by loss of HCN (Table 2). The only significant exception is phenylsulfinylamine, PhNSO, but this can be ascribed to the fact that PhNSO undergoes an additional rearrangement and fragmentation to CO and C5H5NS,22−25 which by loss of NS generates an increased abundance of the (probably open-chain) C5H5+ ion at m/z 65. Formation of m/z 64 may occur by loss of HCN from the 5cyanocyclopentadiene ion 8.+ (eq 3). Loss of CN can contribute to the minor C5H5+ component of the m/z 65 ion (eq 3), but the antiaromaticity of cyclopentadienyl cation makes this process unfavorable.

Similarly, loss of H from 8.+ or 9.+, which would generate the antiaromatic cyanocyclopentadienyl cation (m/z 90), is a totally unimportant reaction. This is in sharp contrast to the cyanocyclopentadienyl radical, which is formed in the gasphase photolysis of 9.26 Further potential routes to C5H5+ and C5H4.+ will be considered below. As mentioned in the Introduction, the pyridylcarbenes and phenylnitrene undergo extensive interconversion under both photochemical and FVP conditions. Scheme 6 describes the corresponding rearrangements of the radical cations. The phenylnitrene and pyridylcarbene radical cations 31.+, 35.+, and 39.+ interconvert via the azacycloheptatetraenes 4.+, 33.+, and 37.+ with activation barriers no higher than 34 kcal/mol. As in the neutral series,27,28 the pyridylcarbenes are higher in energy than the phenylnitrene radical cation 2.+ (2B2). The pyridylcarbene radical cations may undergo additional rearrangements (Scheme 7), but these have mostly signifi-

Figure 2. Energy profile for phenylnitrene radical cation rearrangements. Normal font: CASPT2 calculations. In parentheses: DFT calculations. Energies relative to the doublet radical cation 2.+ (2B2) in kcal/mol.

The principal rearrangement of 2.+ (Figure 2) is the ring expansion to the azacycloheptatetraene radical cation 4.+, which has a modest activation barrier of 28.5 kcal/mol at the CASPT2 level, and thus it should be very facile under the reaction conditions. D

DOI: 10.1021/acs.jpca.8b08480 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A Table 2. High-Resolution Composition (%) of Fragment Peaks of C6H5N Precursors m/z

ion

Ph−N3 1

1,2,3-triazolo-[1,5-a]pyridine 7

Ph−NCOa

50 51

C4H2 C3HN C4H3 C3H2N C4H4 C4HN C5H3 C4H2N C5H4 C4H3N C413CH4 C5H5 C6H5N

100 10 90 100 0 8 92 14 86 80 17 3 100

100 22 78 86 10 5 95 16 84 60 12 25 100

100 13 87 94 6 12 88 16 84 72 22 6 100

52 63 64 65

91

Ph−NSOa

15 85 30 70 35 65 100

benzotriazole 14 100 20 80 97 3 7 93 13 84 88 8 4 100

a

For the low-resolution mass spectrum, see the Supporting Information, Figures S1 and S2.

could be an additional source of the abundant [M − HCN].+ ions (m/z 64) in the mass spectra of all the precursors of m/z 91 ions (Figure 1). The ring opening of the 3-pyridylcarbene ion 35.+ to 43.+ is more favorable (transition state at ∼35 kcal/ mol above the phenylnitrene ion). Alternate ring opening to 44.+ could also provide a source of HCN, but all the HCNforming reactions have high barriers (Scheme 7).

Scheme 6. Interconversion of Pyridylcarbene Radical Cations via Ring Expansion to Azacycloheptatetraenesa

Scheme 8. Transannular Cyclization, Ring Opening/Ring Contraction to Cyanocyclopentadiene and HCN Eliminationa

a

Energies in kcal/mol relative to phenylnitrene radical cation 2.+ (2B2) at the DFT level. Note: 30.+ is an intermediate, but 34.+, 36.+, and 38.+ are transition states only (see the Supporting Information for details).

Scheme 7. Ring Opening and HCN Elimination in 2- and 3Pyridylcarbene Radical Cationsa

a

a

Energies in kcal/mol relative to 2.+ (2B2) at the DFT level.

Energies in kcal/mol relative to 2.+ (2B2) at the DFT level.

Transannular cyclization of cycloheptatetraene radical cations can generate the bicyclic isomers 46.+ and 47.+ (Scheme 8). Ring opening of these leads to ions 49.+ and 50.+, which may serve as additional sources of HCN and C5H4.+ (m/z 64) as well as cyanocyclopentadienes 8.+ and 9+.. For the structures of the transition states, see the Supporting Information. Similar activation energies for transannular

cantly higher activation energies than the reactions considered above. Thus, ring-opening and a hydrogen shift can generate species 40.+ with an activation barrier of 57 kcal/mol relative to phenylnitrene. Fragmentation of 40.+ to HCN and C5H4.+ E

DOI: 10.1021/acs.jpca.8b08480 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

radical cation 4.+ with an activation energy of ca. 28 kcal/mol. Ring opening of the phenylnitrene radical cation to 28.+ and recyclization affords 5-cyanocyclopentadiene ion 8.+, in which facile 1,5-H shifts account for the hydrogen scrambling observed in the mass spectra of deuterated phenyl azides. 2Pyridylcarbene radical cation 31.+ is readily accessible from 2.+, and the 2-, 3-, and 4-pyridylcarbene radical cations interconvert via azacycloheptatetraenes with activation barriers no higher than 34 kcal/mol (Scheme 6). Transannular cyclization in azacycloheptatetraene radical cations 4.+ and 33.+ followed by ring opening or fragmentation provides further routes to 5-cyanocyclopentadiene 8.+. Fragmentation of the ensemble of isomerized ions in the mass spectrometer taking place primarily by loss of C2H2 and HCN from the C6H5N.+ ions accounts for the apparent carbon scrambling reported by Woodgate and Djerassi.33 The carbene−carbene and carbene−nitrene rearrangements, ring expansions, ring contractions, and ring openings known to take place under thermal and photochemical conditions1−4 also take place in the radical cations with activation energies readily accessible under electron-impact conditions. Consequently, the m/z 91 ”phenylnitrene” ions C6H5N.+ observed in the mass spectra can be expected to have undergone extensive isomerization and interconversion, thereby accounting for the similarity of many such mass spectra (Figure 1, Figure S1, and Figure S2), even though ions with partial retention of original structures may be sampled under collisional activation conditions.

cyclizations were computed for the corresponding neutral compounds4 and for the azepinium cation,29 which may form in the reaction of N+ ions with benzene.5 However, the reactions considered in Schemes 7 and 8 have generally high activation barriers, which makes them uncompetitive relative to the reactions described in Figure 2 and Scheme 6. Scheme 9. Potential Formation and Fragmentation of Cyclohexadienylidene Radical Cationa

a

Energies in kcal/mol relative to 2.+ (2B2) at the DFT level.

A further potential isomerization needs to be considered: a 1,3-H shift in the phenylnitrene radical cation would generate the iminocyclohexadienylidene ion 16.+ (Scheme 9). There is no evidence that the 1,3-H shift 2 → 16 takes place under thermal or photochemical conditions.4,11,12 A slow photochemical isomerization of phenyl azide 1 to benzotriazole 14 (presumably in solution) has been claimed30 but not substantiated, and numerous matrix photolyses yielding phenylnitrene observable by ESR spectroscopy have failed to yield any signals assignable to iminocyclohexadienylidene 16,4,10,31 which is otherwise observable in matrix photolyses of benzotriazole.32 Similarly, neither FVP nor matrix photolysis of benzotriazole produce an ESR signal of phenylnitrene. The high barrier of ∼57 kcal/mol for the potential isomerization 2.+ → 16.+ means that this reaction is not impossible in the mass spectrometer, but it is less favorable than the reactions described in Figure 2 and Scheme 6. This is in accord with the conclusions from CID mass spectrometry that the ions 2.+ and 16.+ retain their structures to some extent, although the general similarity of the spectra indicates extensive interconversion and/or rearrangement to a common set of structures.14 If 16.+ is formed, it can undergo ring contraction to the ketenimine 18.+ (Scheme 9). The sequence 16 → 18 → 8 → 9 is clearly the most favorable in this scheme and analogous to the reaction observed under thermal and photochemical conditions (Scheme 3). Further high-energy processes may also contribute; the nitrilium ion 53.+ represents an energy minimum and may undergo elimination of HNC to yield the cyclopentadienyl radical cation C5H4.+ (m/z 64), which can also form by elimination of HCN from the 5-cyanocyclopentadiene radical cation 8.+ as described in eq 3. Finally, ring opening of 16.+ may generate the vinylidene species 54.+ at a cost of ∼64 kcal/mol relative to 2.+.



ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpca.8b08480.



Mass spectra of phenyl isocyanate, phenylsulfinylamine, and 4-diazomethylpyridine; molecular orbitals of the five electronic states of 2.+ at the CASPT2/6-311G(d,p)) level; computational details; Cartesian coordinates; absolute energies; vibrational analysis; imaginary frequencies (PDF)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Didier Bégué: 0000-0002-4553-0166 Curt Wentrup: 0000-0003-0874-7144 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the Queensland Cyber Infrastructure Foundation at The University of Queensland and the Mésocentre de Calcul Intensif Aquitain of the Université de Bordeaux and the Université de Pau et des Pays de l’Adour.





CONCLUSION Calculations at the CASPT2(7,9) level demonstrate that the phenylnitrene radical cation 2.+ can exists in five low-lying electronic states, of which the 2B2 state is the lowest. It undergoes facile ring expansion to 1-azacycloheptetetraene

REFERENCES

(1) Nitrenes and Nitrenium Ions; Falvey, D. E., Gudmundsdottir, A. D., Eds.; John Wiley and Sons: Hoboken, NJ, 2013. (2) Wentrup, C. Flash Vacuum Pyrolysis of Azides, Triazoles, and Tetrazoles. Chem. Rev. 2017, 117, 4562−4623.

F

DOI: 10.1021/acs.jpca.8b08480 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A (3) Sheridan, R. S. Heteroarylcarbenes. Chem. Rev. 2013, 113, 7179−7208. (4) Kvaskoff, D.; Lü erssen, H.; Bednarek, P.; Wentrup, C. Phenylnitrene, Phenylcarbene and Pyridylcarbenes. Ring Contraction to Cyanocyclopentadiene and Fulvenallene. J. Am. Chem. Soc. 2014, 136, 15203−15214. (5) Di Stefano, M.; Rosi, M.; Sgamellotti, A.; Ascenzi, D.; Bassi, D.; Franceschi, P.; Tosi, P. Experimental and Theoretical Investigation of the Production of Cations Containing C-N Bonds in the Reaction of Benzene with Atomic Nitrogen Ions. J. Chem. Phys. 2003, 119, 1978− 1985. (6) Huijuan, C.; Huimin, B.; Rui, D.; Dong, W.; Lingpeng, M.; Shijun, Z.; Dianxun, W.; Mok, D. K.-W.; Chau, F.-T. First Determination of Ionization Energies of Phenylnitrene. Chem. Phys. Lett. 2003, 382, 291−296. (7) Tian, Z.-y.; Yuan, T.; Wang, J.; Li, Y.-y.; Zhang, T.-c.; Zhu, A.-g.; Qi, F. Identification and Chemistry of Phenylnitrene in Premixed Pyridine/Oxygen/Argon Flame with Tunable Synchrotron Photoionization. Chin. J. Chem. Phys. 2007, 20, 425−430. (8) Wentrup, C.; Braybrook, C.; Bégué, D.; Liu, S.; Tzschucke, C. C.; Dargelos, A. Nitrene-Nitrene Rearrangement under Thermal, Photochemical, and Electron-Impact Conditions. The 2-Azidopyridines/Tetrazolo[1,5-a]pyridines. Eur. J. Org. Chem. 2016, 2016, 4200−4206. (9) Crow, W. D.; Wentrup, C. Reactions of Excited Molecules. II. Thermal and Electron Impact Generation of Phenylnitrenes − A Facile Ring Contraction. Tetrahedron Lett. 1967, 8, 4379−4384. (10) Abe, M.; Bégué, B.; Silva, H. S.; Dargelos, A.; Wentrup, C. Triplet States of Tetrazoles, Nitrenes and Carbenes from Matrix Photolysis of Tetrazoles, and Phenylcyanamide as a Source of Phenylnitrene. J. Phys. Chem. A 2018, 122, 7276−7283. (11) Bégué, D.; Santos-Silva, H.; Dargelos, A.; Wentrup, C. Iminocyclohexadienylidenes − Carbenes or Diradicals? The HeteroWolff Rearrangement of Benzotriazoles to Cyanocyclopentadienes and 1H-Benzo[b]azirines. J. Phys. Chem. A 2017, 121, 5998−6003. (12) Wentrup, C.; Freiermuth, B.; Aylward, N. Pyrolysis of Benzotriazoles. 1-Acyl- and 1-Alkoxycarbonylbenzotriazoles: HeteroWolf Rearrangement to N-Acyl- and N-Alkoxycarbonyl-fulvenimines and Free Radical Routes to Cyanocyclopentadienes. J. Anal. Appl. Pyrolysis 2017, 128, 187−195. (13) Maquestiau, A.; Van Haverbeke, Y.; Flammang, R.; Menu, A.; Wentrup, C.; Winter, H.-W. Dissociations induites par Collisions de Cations-Radicaux [C6H5N]+. Tetrahedron Lett. 1978, 19, 1489− 1492. (14) Maquestiau, A.; Van Haverbeke, Y.; Flammang, R.; Menu, A.; Wentrup, C. É tude Structurale de Cations-Radicaux [C6H5N]+. Org. Mass Spectrom. 1978, 13, 518−526. (15) 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, revision A.2; Gaussian, Inc.: Wallingford, CT, 2009. (16) Werner, H.-J.; Knowles, P. J.; Knizia, G.; Manby, F. R.; Schütz, M.; Celani, P.; Korona, T.; Lindh, R.; Mitrushenkov, A.; Rauhut, G., et al. MOLPRO, version 2012.1, A Package of Ab Initio Programs, 2012; http://www.molpro.net, (retrieved August 1, 2018). (17) Winkler, M. Singlet-Triplet Energy Splitting and Excited States of Phenylnitrene. J. Phys. Chem. A 2008, 112, 8649−8653. (18) Kingston, D. G. I.; Henion, J. D. Hydrogen Randomization in Phenyl Azide. Org. Mass Spectrom. 1970, 3, 413−414. (19) Yamabe, S.; Tsuchida, N.; Yamazaki, S. Revisiting Hydrogen [1,5] Shifts in Cyclopentadiene and Cycloheptatriene as Bimolecular Reactions. J. Chem. Theory Comput. 2005, 1, 944−952. (20) Hess, B. A., Jr.; Baldwin, J. E. [1,5] Sigmatropic Hydrogen Shifts in Cyclic 1,3-Dienes. J. Org. Chem. 2002, 67, 6025−6033. (21) Bednarek, P.; Wentrup, C. 3-Pyridylcarbene and 3-Pyridylnitrene: Ring Opening to Nitrile Ylides. J. Am. Chem. Soc. 2003, 125, 9083−9089. (22) Job, B. E. Rearrangement Ions in the Mass Spectrum of Thionylaniline. Chem. Commun. 1967, 44−45.

(23) Bowie, J. H.; Larsson, F. C. V.; Schroll, G.; Lawesson, S. O.; Cooks, R. G. Elecron Impact Studies−IX Mass Spectra of Arylsulphinylamines Skeletal Rearrangement on Electon Impact. Tetrahedron 1967, 23, 3743−3752. (24) Siegel, A. S. Rearrangement IonsI: Mass Spectrum of Thionylaniline-1-13C. Org. Mass Spectrom. 1970, 3, 875−877. (25) Wentrup, C. Phenylnitrene from Phenylsulphinylamine. Correlation of Thermolysis with Electron Impact. Tetrahedron 1971, 27, 1027−1032. (26) Cullin, D. W.; Soundararajan, N.; Platz, M. S.; Miller, T. A. Laser-Induced Fluorescence Spectrum of the Cyanocyclopentadienyl Radical. A Band System Long Attributed to Triplet Phenylnitrene. J. Phys. Chem. 1990, 94, 8890−8896. (27) Wentrup, C. Rearrangements and Interconversions of Carbenes and Nitrenes. Top. Curr. Chem. 1976, 62, 173−251. (28) Kemnitz, C. R.; Karney, W. L.; Borden, W. T. Why Are Nitrenes More Stable Than Carbenes? An Ab Initio Study. W. T. J. Am. Chem. Soc. 1998, 120, 3499−3503. (29) Di Stefano, M.; Rosi, M.; Sgamellotti, A.; Negri, F. Reactions of N+ Ions with Benzene: A Theoretical Study on the C6NH6+ Potential Energy Surface. Chem. Phys. 2004, 302, 295−308. (30) Walker, P.; Waters, W. A. Pyrolysis of Organic Azides: A Mechanistic Study. J. Chem. Soc. 1962, 1632−1638. (31) Kuzaj, M.; Lüerssen, H.; Wentrup, C. ESR Observation of Thermally Produced Triplet Nitrenes and Photochemically Produced Triplet Cycloheptatrienylidenes. Angew. Chem., Int. Ed. Engl. 1986, 25, 480−482. (32) Murai, M.; Torres, M.; Strausz, O. P. Electron Spin Resonance of Iminocyclohexadienylidenes: Photoinduced Triplet Geometrical Isomerization. J. Am. Chem. Soc. 1980, 102, 1421−1422. (33) Woodgate, P. D.; Djerassi, C. Mass Spectrometry in Structural and Stereochemical Problems. CXCII. Skeletal Rearrangement in the Fragmentation of Phenyl Azide. Tetrahedron Lett. 1970, 11, 1875− 1878.

G

DOI: 10.1021/acs.jpca.8b08480 J. Phys. Chem. A XXXX, XXX, XXX−XXX