Competition between α-cleavage and energy transfer in α-azidoacetophenones

Article abstract of DOI:10.1021/jo062160k

(Chemical Equation Presented) Molecular modeling demonstrates that the first excited state of the triplet ketone (T_1K) in azide 1b has a (π,π*) configuration with an energy that is 66 kcal/mol above its ground state and its second excited state (T_2K) is 10 kcal/mol higher in energy and has a (n,π*) configuration. In comparison, T _1K and T_2K of azide 1a are almost degenerate at 74 and 77 kcal/mol above the ground state with a (n,π*) and (π,π*) configuration, respectively. Laser flash photolysis (308 nm) of azide 1b in methanol yields a transient absorption (λ_max = 450 nm) due to formation of T_1K, which decays with a rate of 2.1 × 10 ^-5 s^-1 to form triplet alkylnitrene 2b (λ_max = 320 nm). The lifetime of nitrene 2b was measured to be 16 ms. In contrast, laser flash photolysis (308 nm) of azide 1a produced transient absorption spectra due to formation of nitrene 2a (λ _max = 320 nm) and benzoyl radical 3a (λ_max = 370 nm). The decay of 3a is 2 × 10^-5 s^-1 in methanol, whereas nitrene 2a decays with a rate of ～91 s^-1. Thus, T _1K (π,π*) in azide 1b leads to energy transfer to form nitrene 2b; however, α-cleavage is not observed since the energy of T _2K (n,π*) is 10 kcal/mol higher in energy than T _1K, and therefore, T_2K is not populated. In azide 1a both α-cleavage and energy transfer are observed from T_1K (n,π*) and T_2K (π,π*), respectively, since these triplet states are almost degenerate. Photolysis of azide 1a yields mainly product 4, which must arise from recombination of benzoyl radicals 3a with nitrenes 2a. However, products studies for azide 1b also yield 4b as the major product, even though laser flash photolysis of azide 1b does not indicate formation of benzoyl radical 3b. Thus, we hypothesize that benzoyl radicals 3 can also be formed from nitrenes 2. More specifically, nitrene 2 does undergo α-photocleavage to form benzoyl radicals and iminyl radicals. The secondary photolysis of nitrenes 2 is further supported with molecular modeling and product studies.

Full text of DOI:10.1021/jo062160k

Competition between r-Cleavage and Energy Transfer in
r-Azidoacetophenones
Sivaramakrishnan Muthukrishnan,† Sarah M. Mandel,† John C. Hackett,‡
Pradeep N. D. Singh,† Christopher M. Hadad,‡ Jeanette A. Krause,† and
Anna D. Gudmundsdo´ttir*,†
Department of Chemistry, UniVersity of Cincinnati, Cincinnati, Ohio 45221-0172 and Department of
Chemistry, The Ohio State UniVersity, Columbus, Ohio 43210-1173
Anna.Gudmundsdottir@uc.edu
ReceiVed October 17, 2006
Molecular modeling demonstrates that the first excited state of the triplet ketone (T1K) in azide 1b has a
(π,π*) configuration with an energy that is 66 kcal/mol above its ground state and its second excited
state (T2K) is 10 kcal/mol higher in energy and has a (n,π*) configuration. In comparison, T1K and T2K
of azide 1a are almost degenerate at 74 and 77 kcal/mol above the ground state with a (n,π*) and (π,π*)
configuration, respectively. Laser flash photolysis (308 nm) of azide 1b in methanol yields a transient
absorption (λmax ) 450 nm) due to formation of T1K, which decays with a rate of 2.1 × 105 s-1 to form
triplet alkylnitrene 2b (λmax ) 320 nm). The lifetime of nitrene 2b was measured to be 16 ms. In contrast,
laser flash photolysis (308 nm) of azide 1a produced transient absorption spectra due to formation of
nitrene 2a (λmax ) 320 nm) and benzoyl radical 3a (λmax ) 370 nm). The decay of 3a is 2 × 105 s-1 in
methanol, whereas nitrene 2a decays with a rate of ∼91 s-1. Thus, T1K (π,π*) in azide 1b leads to
energy transfer to form nitrene 2b; however, R-cleavage is not observed since the energy of T2K (n,π*)
is 10 kcal/mol higher in energy than T1K, and therefore, T2K is not populated. In azide 1a both R-cleavage
and energy transfer are observed from T1K (n,π*) and T2K (π,π*), respectively, since these triplet states
are almost degenerate. Photolysis of azide 1a yields mainly product 4, which must arise from recombination
of benzoyl radicals 3a with nitrenes 2a. However, products studies for azide 1b also yield 4b as the
major product, even though laser flash photolysis of azide 1b does not indicate formation of benzoyl
radical 3b. Thus, we hypothesize that benzoyl radicals 3 can also be formed from nitrenes 2. More
specifically, nitrene 2 does undergo R-photocleavage to form benzoyl radicals and iminyl radicals. The
secondary photolysis of nitrenes 2 is further supported with molecular modeling and product studies.
I. Introduction
direct irradiation or thermal activation of alkyl azides does not
lead to formation of triplet alkylnitrene intermediates at ambient
temperature.2,3 Rather, triplet sensitization must be used to form
triplet alkylnitrenes from alkyl azides.4,5 Recently, we demon-
The high-spin properties of triplet nitrenes provide a potential
application as organic magnets, and such desires have regener-
ated interest in these intermediates.1 However, triplet alkylni-
trenes have only been studied sporadically, presumably because
(1) (a) Magnetic Properties of Organic Materials; Lahti, P. M., Ed.;
Marcel Decker, Inc.: New York, 1999. (b) Tomioka, H. Triplet Carbenes.
In ReactiVe Intermediate Chemistry; Moss, R. A., Platz, M. S., Jones, M.,
Jr., Eds.; John Wiley & Sons, Inc.: Hoboken, NJ, 2004; pp 501-592.
† University of Cincinnati.
‡ The Ohio State University.
10.1021/jo062160k CCC: $37.00 © 2007 American Chemical Society
Published on Web 03/21/2007
J. Org. Chem. 2007, 72, 2757-2768
2757

Muthukrishnan et al.
SCHEME 1
SCHEME 2
SCHEME 3
strated that photolysis of azides 1 (Scheme 1) yields triplet
alkylnitrene intermediates 2 via intramolecular triplet sensitiza-
tion from the ketone chromophore to the azido moiety.4a By
photolyzing azides 1 with light above 300 nm, only the aryl
ketone chromophore absorbs the light and forms the first excited
singlet state of the ketone, which undergoes efficient intersystem
crossing to the triplet ketone.6 Energy transfer from the triplet
ketone to the azido moiety results in formation of triplet nitrene
intermediates.
first excited state of the triplet ketone (T1K) with a (π,π*)
configuration, gives the same photoproducts as R-azido-
acetophenone, which has T1K with the (n,π*) configuration.
Hence, we set out to study how the energies of the two (n,π*)
and (π,π*) configurations, T1K and T2K, in R-azidoacetophenone
derivatives affect the corresponding product formation, and this
will be investigated by comparing the photochemistry of azides
1a and 1b (Scheme 3). On the basis of the similarity of azides
1 with the analogous valerophenone derivatives, we expect azide
1a to have almost degenerate T1K and T2K whereas T1K (π,π*)
in azide 1b is expected to be almost 10 kcal/mol lower in energy
than T2K (n,π*).8
In this article, we describe how experimental laser flash
photolysis and product studies along with molecular modeling
reveal that azide 1a undergoes both R-cleavage and energy
transfer to form benzoyl radical 3a and triplet alkylnitrene 2a,
whereas azide 1b selectively forms triplet alkylnitrene 2b.
However, triplet alkylnitrenes 2 are photoreactive themselves
and undergo R-cleavage to form benzoyl and iminyl radicals.
In competition with energy transfer, azide 1 can undergo
R-cleavage to form benzoyl radical 3 and the azidomethyl (N3-
•
CH2 ) radical. The energy transfer in the R-azidoacetophenones
must come from the (π,π*) state of the triplet ketone since the
R-cleavage occurs from the (n,π*) state.7 Product studies have
shown that irradiating azide 1 yields mainly 4 and a small
amount of benzaldehyde (5), benzyl (6), and acetophenone (7)
derivatives. Presumably, triplet alkylnitrene 2 reacts with
benzoyl radicals 3 and the resulting radical abstracts a hydrogen
atom from the solvent to form 4 (Scheme 2). However,
photolysis of R-azido-(p-cyano)-acetophenone, which has the
II. Results
(2) Platz, M. S. Nitrenes. In ReactiVe Intermediate Chemistry; Moss, R.
A., Platz, M. S., Jones, M., Jr., Eds.; John Wiley & Sons, Inc.: Hoboken,
NJ, 2004; pp 375-461.
A. Phosphorescence Spectra. We measured the phospho-
rescence of 7b and azide 1b at 77 K in ethanol glasses (Figure
1) under identical experimental conditions (0.6 mmol, λex
(3) (a) Barash, L.; Wasserman, E.; Yager, W. A. J. Am. Chem. Soc. 1967,
89, 3931. (b) Radziszewski, J. G.; Downing, J. W.; Jawdosiuk, M.; Kovacic,
P.; Michl, J. J. Am. Chem. Soc. 1985, 107, 594. (c) Carrick, P. G.; Brazier,
C. R.; Bernath, P. F.; Engelking, P. C. J. Am. Chem. Soc. 1987, 109, 5100.
(g) Ferrante, R. F. J. Chem. Phys. 1987, 86, 25. (e). Gritsan, N. P.;
Likhotvorik, I.; Zhu, Z.; Platz, M. S. J. Phys. Chem. A 2001, 105, 3039.
(4) (a) Mandel, S. M.; Singh, P. N. D.; Robinson, R. M.; Zhu, Z.; Franz,
R.; Ault, B. S.; Gudmundsdo´ttir, A. D. J. Org. Chem. 2003, 68, 7951. (b)
Mandel, S. M.; Krause Bauer, J. A.; Gudmundsdottir, A. D. Org. Lett. 2001,
3, 523.
)
353 nm). We did not observe any phosphorescence for azide
1a, presumably since the triplet ketone is efficiently quenched
by intramolecular energy transfer to the azido moeity.4a Ketone
7b and azide 1b show structureless phosphorescence which is
typical9 for triplet ketones with a (π,π*) configuration. Since
7b and 1b have similar UV absorption spectra (Supporting
Information) and their phosphorescence spectra were recorded
under the same conditions, we compared their yield of emission
(5) Klima, R. F.; Gudmundsdottir, A. D. J. Photochem. Photobiol., A
2003, 162, 239.
(6) McGarry, P. F.; Doubleday, C. E., Jr.; Wu, C.-H.; Staab, H. A.; Turro,
N. J. J. Photochem. Photobiol., A 1994, 77, 109.
(7) (a) Turro, J. Modern Molecular Photochemistry; University Science
Books: Sausalito, CA, 1991; Chapter 7. (b) Garcia-Garibay, M. A.; Campos,
L. M. Photochemical decarbonylation of ketones: recent advances and
reactions in crystalline solids. In CRC Handbook of Organic Photochemistry
and Photobiology, 2nd ed.; Horspool, W., Lenci, F., Eds.; CRC Press: Boca
Raton, FL, 2004; Chapter 48.
(8) Wagner, P. J.; Kemppainen, A. E.; Schott, H. N. J. Am. Chem. Soc.
1973, 95, 5604.
(9) (a) Turro, J. Modern Molecular Photochemistry; University Science
Books: Sausalito, CA, 1991; Chapter 5. (b) Becker, R. Theory and
Interpretation of Fluorescence and Phosphorescence; Wiley: New York;
1969.
2758 J. Org. Chem., Vol. 72, No. 8, 2007

Competition in R-Azidoacetophenones
FIGURE 1. Phosphorescence spectra of azides 1b (red) and 7b (black).
SCHEME 4
SCHEME 5
and found that the yield for azide 1b is less than for 7b,
indicating that some of the phosphorescence is quenched by
energy transfer to the azido group. From the phosphorescence
spectra we can estimate that the energy of the triplet ketone in
7b and azide 1b are similar or ∼64 kcal/mol.10
In addition, smaller amounts of benzaldehyde 5, benzil 6, and
acetophenone 7 are formed as well (Scheme 4). Similar
photoproduct ratios are observed for azides 1a and 1b. As
mentioned before, product 4 must come from the initially
generated nitrene 2 being able to trap benzoyl radical 3 and the
resulting radical subsequently abstracting an H atom from the
solvent (Scheme 5). Similarly, 5 and 6 presumably come from
the benzoyl radical reacting with toluene or undergoing dimer-
ization. It is possible that nitrene 2 dimerizes to form the
azodimer that subsequently absorbs another photon to form
acetophenone radicals, which abstract a hydrogen atom from
the solvent to form 7. It is, however, possible that azide 1
undergoes â-cleavage to form azido and acetophenonyl radicals.
B. Product Studies. Product studies from the photolysis of
1a have been described previously but are included here for
comparison.4 Photolysis of azides 1a and 1b in toluene, under
the same experimental conditions, yields one major product 4.
(10) Murow, S. L. Carmichael, I.; Hug, G. L. Handbook of Photochem-
istry, Marcel Dekker: New York, 1993.
(11) Wilson, R. M; Patterson, W. S.; Austen, S. C.; Ho, D. M.; Krause-
Bauer, J. A. J. Am. Chem. Soc. 1995, 117, 7820.
J. Org. Chem, Vol. 72, No. 8, 2007 2759

Muthukrishnan et al.
Interestingly, azide 1b yields similar quantity of products 4, 5,
and 6 as azide 1a, even though they can be attributed to benzoyl
radical formation. Thus, we theorized whether nitrenes 2 can
fall apart to yield benzoyl radical 3 either thermally or
photochemically.
C. Molecular Modeling. All structures were optimized using
Gaussian0312 at the B3LYP level of theory and with the
6-31+G(d) basis set,13 unless otherwise noted. We optimized
the ground-state (S0) configuration of azides 1, triplet nitrenes
2, benzoyl radicals 3, azidomethyl radical, and methylene iminyl
radical. We calculated the IR spectra of these optimized
structures from vibrational frequency analyses at the same level
of theory. The optimized geometries of azides 1a and 1b are
very similar; both molecules are relatively flat with the exception
of the azido moiety being twisted out of the plane of the phenyl
ring, as revealed by the torsion angle between C7-C8-N1-
N2 being 62° for both 1a and 1b (Figure 2 and 3). Furthermore,
the X-ray structure of azide 1b is similar to its optimized
structure (see Supporting Information and Figure 4). In the X-ray
structure, both the carbonyl group and Me-S substituents are
rotated slightly out of the plane of the phenyl ring, presumably
to accommodate for the crystal packing arrangement of the
molecule. The twisting of 1b in the crystal lattice can be seen
by the torsion angle between C6-C1-C7-O1 and C9-S1-
C4-C5, which are 11.8° and -16.3°, respectively.
We used time-dependent density functional theory (TD-
DFT)14 to estimate the vertical excitation energies of T1K and
T2K for azides 1a and 1b (Table 1) using the geometry for the
S0 state. In azide 1a, T1K is located 75 kcal/mol above the ground
state. Visualization of the molecular orbitals for the major
components of the electronic excitation revealed that this T1K
state is dominated by a (n,π*) configuration as expected, and it
is about 1 kcal/mol more stable than T2K, which has a (π,π*)
configuration. Solvation15 does not significantly change the
relative energies of T1K and T2K in azide 1a (Table 1). The
calculated energy difference between T1K and T2K in azide 1a
fits well with what has been estimated for valerophenone and
FIGURE 2. Optimized S0, T1A, and T1K geometries of azide 1a at the
RB3LYP (S0) and UB3LYP (T1A and T1K) levels of theory. Bond
lengths are in Ångstroms.
butyrophenone derivatives.16,17 TD-DFT calculations estimated
the vertical excitation energy of T1K of azide 1b to be 66
kcal/mol above the ground state and have a (π,π*) configuration;
furthermore, T2K is 10 kcal/mol higher in energy than T1K. This
is similar to the estimated energy difference between T1K (π,π*)
and T2K (n,π*) in the analogous thiomethyl valerophenone.8
We optimized the triplet excited states of the azide moiety,
T1A, in azides 1 at the unrestricted B3LYP/6-31+G(d) level and
found that T1A was located 45 kcal/mol above the ground state.
The N1-N2-N3 bond in T1A has a bond angle of ∼120.7°, and
the N1-N2 bond length is 1.43 Å (Figure 2 and 3), whereas in
the ground state of azides 1a,b, the N1-N2 bond length is 1.23
Å. As a reflection of the increased length of the N1-N2 bond,
the calculated IR spectrum of T1A does not have an azide
stretching vibration at ∼2100 cm-1, as seen for azides 1a,b,
(12) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T. Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,
X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03,
Revision A.1; Gaussian, Inc.: Pittsburgh, PA, 2003.
but rather the N-N stretch is calculated to be ∼1720 cm-1
.
T1A can be assigned to the (n,π*) configuration of the triplet
azides based on visualization of the molecular orbitals (see
Supporting Information).
(13) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(14) (a) Density Functional Methods in Chemistry; Labanowski, J., Ed.;
Springer-Verlag: Heidelberg, 1991. (b) Parr, R. G.; Weitao, Y. Density-
Functional Theory in Atoms and Molecules; Oxford University Press: New
York, 1989. (c) Bauernschmitt, R Chem. Phys. Lett. 1996, 256, 454. (d)
Stratmann, R. J. Chem. Phys. 1998, 109, 8218. (e) Foresman, J. J. Phys.
Chem. 1992, 96, 135.
(15) (a) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. ReV. 2005, 105,
2999. (b) Mennucci, B.; Cance`s, E.;Tomasi, J. J. Phys. Chem. B 1997, 101,
10506. (c) Cance`s, E.; Mennucci, B. J. Chem. Phys. 2001, 114, 4744. (d)
Tomasi, J.; Persico, M. Chem. ReV. 1994, 94, 2027. (e) Cramer, C. J.;
Truhlar, D. J. Chem. ReV. 1999, 99, 2161.
(16) (a) He, H.-Y.; Fang, W.-H.; Phillips, D. L. J. Phys. Chem. A 2004,
108, 5386. (b) Fang, W.-H.; Phillips, D. L. Chem. Phys. Chem. 2002, 3,
889. (c) Fang, W.-H.; Phillips, D. L. J. Theor. Comput. Chem. 2003, 2, 23.
(17) (a) Wagner, P. J. Abstraction of γ-hydrogens by excited carbonyls.
In Molecular and Supramolecular Photochemistry; Griesbeck, A. G.,
Mattay, J., Eds.; Marcel Dekker: New York, 2005; Chapter 2. (b) Wagner,
P. J. In CRC Handbook of Organic Photochemistry and Photobiology;
Horspool, W. M., Song, P.-S., Eds.; CRC Press: Boca Raton, FL, 1995;
Chapter 52. (c) Wagner, P. J. Photoinduced Hydrogen Atom Abstraction
by Carbonyl Compounds. In Organic Photochemistry; Padwa, A., Ed.;
Marcel Dekker: New York, 1991; Vol. 11, Chapter 4, p 227.
2760 J. Org. Chem., Vol. 72, No. 8, 2007

Competition in R-Azidoacetophenones
TABLE 1. Comparison of the Calculated Energies for the Triplet
Excited States of Azides 1a,b
azide 1a
energy
azide 1b
energy
state
method
configuration (kcal/mol)a configuration (kcal/mol)
T1K TD-B3LYP
optimizedb
(n,π*)
75.0 (76.1)
66.0 (67.2)
65.8
(π,π*)
65.6 (65.1)
64.0 (63.0)
TURBOMOLEc
T2K TD-B3LYP
(π,π*)
(n,π*)
76.3 (77.4)
45.0
(n,π*)
(n,π*)
76.0 (78.2)
45.1
T1A optimizedb
a Numbers in parentheses are calculated energies at the PCM level with
toluene as a solvent. b Optimized using Gaussian03 at the UB3LYP level
of theory. c Optimized using TURBOMOLE with analytical derivatives at
the TD-B3LYP level of theory.
kcal/mol. In order to clarify this difference, we also optimized
T1K for azides 1a using TURBOMOLE and utilized the
analytical first derivative of TD-DFT with the B3LYP functional
as implemented in this program.18 For these calculations, the
6-31+G(d) basis set was used with the B3LYP DFT method.
However, the geometry and energy of T1K of azides 1a were
almost identical using the optimized TD-B3LYP and UB3LYP
methods. We compared the optimization of T1K in azide 1a with
the calculated energy of T1K in acetophenone. TD-B3LYP
estimates the energy of T1K to be 74 kcal/mol, which is in
excellent agreement with the measured energy of 74 kcal/mol
for the triplet ketone in acetophenone.10 In comparison, opti-
mization of T1K in acetophenone using UB3LYP/6-31+G(d)
gave the adiabatic energy of T1K (n,π*) as 69 kcal/mol, which
is 5 kcal/mol lower than its measured energy.10 Thus, UB3LYP
appears to underestimate the adiabatic energy of T1K in
acetophenone derivatives with the (n,π*) configuration, whereas
the TD-B3LYP calculations estimate these energies more
accurately.
FIGURE 3. Optimized So, T1A, and T1K geometries of azide 1b at the
RB3LYP (S0) and UB3LYP (T1A and T1K) levels of theory. Bond
lengths are in Ångstroms.
In order to evaluate whether solvation was the origin of the
difference between the calculated and experimental energies,
we did some other calculations in which a methanol molecule
was explicitly hydrogen bonded to the carbonyl group of T1K
and S0 of azide 1a, so as to estimate how hydrogen bonding of
the solvent might affect their energies. However, the energy
difference (67 kcal/mol) between S0 and T1K in azide 1a was
very similar to that in the gas phase.
UB3LYP optimization of the triplet ketone, T1K, in azide 1b
shows that the CdO bond in T1K is 1.26 Å, and its stretching
vibration occurs at 1478 cm-1, indicating that the C-O bond
has more double-bond character than for T1K in azide 1a.
Furthermore, the aromatic carbon-carbon bonds in the phenyl
ring are lengthened (Figure 3), as expected for triplet aryl
ketones with a (π,π*) configuration.16,19 The energy of the
optimized triplet ketone is 63 kcal/mol above its ground state.
Thus, the energies of the optimized structure T1K and the TD-
DFT calculation matched well with the energy of T1K obtained
from the phosphorescence spectrum of azide 1b.
We optimized the geometry of nitrenes 2 and used TD-
B3LYP to estimate the first and second excited triplet states
of the ketone moiety in the molecule and found that the T1
and T2 of the nitrenes have energies and configurations like
their corresponding azide precursors. We also used the TD-
B3LYP level to estimate the absorption spectra of nitrenes 2a,b
(Table 2).
FIGURE 4. Crystal structure of azide 1b.
We also optimized T1K for azide 1a and found that it is located
66 kcal/mol above the S0 ground state of azide 1a. In T1K, the
N1-N2-N3 bond is almost linear (173°) and the calculated
vibrational stretching frequency for the azide group is 2218
cm-1, which is similar to what was calculated for the ground
state of azide 1a (Figure 2). The C-O bond has lengthened
from 1.22 Å in the ground state to 1.31 Å, and it has a stretching
vibration at 1380 cm-1. The progression of the C-O bond and
its IR vibration fits well with what has been observed for triplet
aryl ketones with a (n,π*) configuration.16,19 The energy of T1K
obtained from the (vertical) TD-DFT calculation and the
optimized structure at the UB3LYP level differ by almost 9
(18) Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.; Horn, H.; Ko¨lmel; C. Chem. Phys.
Lett. 1989, 162, 165.
(19) Srivastava, S.; Yourd, E.; Toscano, J. P. J. Am. Chem. Soc. 1998,
120, 6173.
Similarly, we optimized the geometry of benzoyl radicals 3
and obtained their IR spectra. TD-DFT calculations were used
to estimate the absorption spectra as shown in Table 2.
J. Org. Chem, Vol. 72, No. 8, 2007 2761

Muthukrishnan et al.
TABLE 2. Major Absorption Spectral Features (nm, Oscillator Strength) for 2 and 3a
electronic transition above 300 nm
2a
2b
3a
3b
gas phase
MeOH
gas phase
MeOH
gas phase
MeOH
gas phase
MeOH
386 (0.001)
378 (0.0001)
435 (0.0002)
450 (0.0038)
362 (0.0002)
364 (0.0004)
419 (0.0009)
428 (0.0020)
344 (0.0096)
362(0.0002)
356 (0.0321)
399 (0.0399)
335 (0.001)
326 (0.0005)
347 (0.0001)
364 (0.0001)
342 (0.0003)
355 (0.0142)
342 (0.0003)
358 (0.0350)
313 (0.0001)
300 (0.0002)
332 (0.0001)
314 (0.3691)
317 (0.0015)
308 (0.0243)
318 (0.0013)
322 (0.2030)
306 (0.0113)
307 (0.1904)
320 (0.2411)
303 (0.0272)
312 (0.0952)
300 (0.3225)
309 (0.0001)
a Number in parentheses is the calculated oscillator strength (f) for the electronic transition in nanometers.
FIGURE 5. Calculated stationary points on the energy surface for R-cleavage and energy transfer in azide 1a at the UB3LYP and TD-B3LYP (in
parentheses) levels of theory.
We calculated the triplet transition states, TS1 and TS2, for
azides 1 to undergo R-cleavage to form benzoyl (PhC(dO)•)
5 and 6). IRC calculations correlated the triplet azides and triplet
nitrenes with these transition states. The transition state for the
triplet azides falling apart to yield triplet nitrene intermediates
is less than 1 kcal/mol above the transition state, indicating that
the triplet azides must be short-lived intermediates. The energy
transfer to the azido moiety must come from the (π,π*) triplet
ketones in azides 1a,b. Thus, we expect both azides 1a and 1b
to undergo energy transfer since T1K in azide 1b has a (π,π*)
configuration and because T2K (π,π*) in azide 1a is close in
energy to T1K (n,π*) and could be in equilibrium with each
other.
•
and azidomethyl (N3CH2 ) radicals (Figures 5 and 6). Intrinsic
reaction coordinate (IRC)21 calculations allowed us to correlate
the (n,π*) configuration of the triplet ketone to the benzoyl and
azidomethyl radicals to these transition states. The transition
states for R-cleavages are located only about 2 kcal/mol above
the triplet ketones with (n,π*) configuration in azides 1. Thus,
the R-cleavage is easily accessible from T1K in azides 1a,b at
room temperature; however, the initial population of T2K in azide
1b does not appear feasible. The energy gap between T1K and
T
2K in azide 1b is too large for them to be in equilibrium or to
Finally, we calculated the transition states, TS5 and TS6, for
R-cleavage of nitrenes 2 to form the corresponding benzoyl
radical 3 and methylene iminyl radical. These transitions states
were also connected to the respective reactant and products by
IRC calculation. Stationary points on the triplet energy surface
for the R-cleavage of nitrenes 2 are shown in Figures 7 and 8.
The transition states are located about 18 kcal/mol above the
ground state of nitrenes 2. Thus, at ambient temperature we
expect nitrenes 2 to undergo R-cleavage to form benzoyl radicals
photochemically rather than thermally. Both nitrenes derived
from 2a and 2b can yield benzoyl radicals 3, since this is a
rearrangement of the ground state of the triplet nitrenes. More
be strongly mixed with each other. Furthermore, intersystem
crossing from the first excited singlet state in azide 1b can
populate either T1K or T2K, but internal conversion to T1K is
expected to depopulate T2K efficiently.
We also calculated the triplet transition states, TS3 and TS4,
for formation of triplet nitrenes 2 from the azides 1a,b (Figures
(20) Srinivasan, A.; Kebede, N.; Saavedra, J. E.; Nikolaitchik, A. V.;
Brady, D. A.; Yourd, E.; Davies, K M.; Keefer, L. K.; Toscano, J. P. J.
Am. Chem. Soc. 2001, 123, 5465.
(21) (a) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154.
(b) Gonzalez, C.; Schlegel, H. B. J Phys. Chem. 1990, 94, 5523.
2762 J. Org. Chem., Vol. 72, No. 8, 2007

Competition in R-Azidoacetophenones
FIGURE 6. Calculated stationary points on the energy surface for R-cleavage and energy transfer for azide 1b at the UB3LYP and TD-B3LYP
(in parentheses) levels of theory.
FIGURE 7. Calculated stationary points on the triplet energy surface for R-cleavage of nitrene 1a at the UB3LYP and TD-B3LYP (in parentheses)
levels of theory.
specifically, either the nitrene or the ketone chromophores in
nitrenes 2 can absorb light and decay back to a vibrationally
hot ground state, which can then fall apart to yield benzoyl and
iminyl radicals.
D. Laser Flash Photolysis. Laser flash photolysis (excimer,
λ ) 308 nm, 17 ns) of azide 1b in nitrogen-saturated methanol
produces a transient absorption signal with λmax at 450 nm due
to absorption of its triplet ketone, which decays with a rate of
2.1 × 105 s-1 (Figure 9). Simultaneously with this decay, a
new transient with λmax at 320 nm forms with the same rate
constant as the triplet ketone decays, and we assign the 320
nm absorption to the triplet nitrene 2b (Figure 11). We base
this assignment on the similarity of this transient to the
calculated absorption spectra of 2b. TD-B3LYP calculations
predict that nitrene 2b has absorption bands in methanol at 450
(f ) 0.0038), 399 (f ) 0.040), 358 (f ) 0.035), 322 (f ) 0.203),
J. Org. Chem, Vol. 72, No. 8, 2007 2763

Muthukrishnan et al.
FIGURE 8. Calculated stationary points on the triplet energy surface for R-cleavage of nitrene 1b at the UB3LYP and TD-B3LYP (in parentheses)
levels of theory.
FIGURE 9. Laser flash photolysis (308 nm) of azides 1a and 1b in nitrogen-saturated methanol right after the laser pulse in a 300 ns window.
FIGURE 10. Decay of T1K in 1b at 450 nm, and formation of 2b at 350 nm.
and 320 nm (f ) 0.241). The band at 450 nm comes mainly
from an electronic transition from the lone pair on the sulfur
atom into the π* orbital, whereas the 399 nm transition is mainly
due to an electronic transition out of the π-orbitals into the half-
full orbital on the nitrogen atom. The bands around 320 nm are
mixed transitions out of the lone pair on the sulfur and oxygen
atoms as well as the half-full orbitals on the nitrogen atom and
into the π* orbitals. In oxygen-saturated ethanol solution, the
2764 J. Org. Chem., Vol. 72, No. 8, 2007

Competition in R-Azidoacetophenones
FIGURE 11. Decay at 350 nm of (a) benzoyl radical 3a and (b) nitrene 2a after initial laser flash photolysis of 1a (λ ) 308 nm) in nitrogen-
saturated methanol.
triplet ketone absorption is quenched and the yield of the nitrene
absorption is decreased, but the appearance of the spectra is
not affected (see Supporting Information), which confirms that
benzoyl radical 3b is not formed upon direct photolysis of 1b.
Benzoyl radical 3b is expected to be quenched efficiently with
molecular oxygen as 3a.4a Benzoyl radical 3b and nitrene 2b
are expected to have different absorption spectra (see Table 1);
thus, if 3b is formed from 1b, laser flash photolysis of 1b
oxygen-saturated ethanol should yield a different transient
spectrum than in argon-saturated ethanol. Thus, the laser flash
photolysis data show that benzoyl radicals 2b are not formed
from photolysis of azide 1b. Furthermore, in air-saturated
methanol solutions, the rate of decay for the triplet ketone is
1.5 × 107 s-1 (Figure 10). The decay of alkylnitrene 2b in
nitrogen-saturated methanol can be followed on the millisecond
time scale and fitted with first-order decay to be 64 s-1. In air-
and oxygen-saturated methanol solution, the decay rates for 2b
are 150 and 250 s-1, respectively. By assuming that the oxygen
concentration at saturation is 0.01 M in methanol,10 we can
estimate that nitrene 2b reacts with oxygen with a rate constant
of 3 × 104 M-1 s-1. In comparison, Liang and Schuster reported
that triplet p-nitrophenylnitrene reacts with oxygen with a rate
of less than 2 × 105 M-1 s-1 whereas Gritsan and Pritchina
estimate this same rate constant to be somewhat larger.23,24 This
rate is, however, in agreement with computational work by Liu
et al., who reported that triplet alkylnitrenes can be expected to
react similarly with oxygen as do triplet phenylnitrenes.25 This
is in contrast to Toscano and co-workers, who reported much
faster bimolecular reactions (∼109) for molecular oxygen
reacting with triplet benzyloxynitrene (3PhCH2ON),20 but Liu
et al. ascribed this unusual reactivity to an electron-transfer
component in this nitrene class.25
We previously reported that the laser flash photolysis of azide
1a results in a broad absorption band that has a maximum
around 320 nm.4a The calculated spectrum of nitrene 2a in
methanol has predicted peaks at 355 (f ) 0.014) and 308 nm (f
) 0.024), and these features agree well with the observed
transient absorption. Additionally, laser flash photolysis (λ )
308 nm) of azide 1a also shows formation of benzoyl radical
3a concurrently with that of nitrene 2a. Benzoyl radical 3a has
been reported to have a weak absorption at ∼370 nm.26 The
calculated absorption spectrum of benzoyl radical supports this
(see Table 2). We do not observe the triplet ketone in azide 1a
because the energy of the triplet ketone is higher than in 1b;
thus, it is more efficiently quenched by energy transfer and
R-cleavage. We previously estimated the lifetime of triplet
ketone in azide 1a to be between 0.9 and 0.09 ns.4a The rate
for the decay of the benzoyl radical 3a can be measured at 350
nm to be around 2 × 105 s-1, whereas the nitrene 2a decays
with a unimolecular rate constant of 91 s-1 (Figure 11).
The laser flash photolysis of azides 1 demonstrates that azide
1a forms both benzoyl radical 3a and nitrene 2a whereas 1b
only forms nitrene 2b since we did not observe any decay that
can be attributed to benzoyl radical 3b.
E. Effect of Concentration on the Product Ratio. We
irradiated five different concentrations of azides 1 in toluene
with a continuous-wave, argon-ion laser11 (λ ) 351 nm) and
analyzed the product ratios using HPLC chromatography. In
Figure 11, the ratio of ([4] + [5] + 2[6])/([4] + [7]) was plotted
versus the concentration of azides 1. Products 4, 5, and 6 are
attributed to formation of benzoyl radical 3, whereas products
4 and 7 are due to formation of triplet alkyl nitrene 2. We count
product 6 twice because it is due to dimerization of two benzoyl
radicals. Since product 4 results from both benzoyl radical 3
and nitrene 2, we count it as both the benzoyl radical and
(22) (a) Singh, P. N. D.; Mandel, S. M.; Sankaranarayanan, J.; Muth-
ukrishnan, S.; Chang, M.; Robinson, R. M.; Ault, B. S.; Gudmundsdottir,
A. D. Unpublished work. (b) Sankaranarayanan, J.; Mandel, S. M.; Krause,
J. A.; Gudmundsdottir, A. D. Acta. Crystallogr., Sect. E 2007, E63, o721.
(23) Liang, T.-Y.; Schuster, G. B. J. Am. Chem. Soc. 1987, 109, 7803.
(24) (a) Gritsan, N. P.; Pritchina, E. S. J. Inf. Rec. Mater. 1989, 17, 391.
(b) Pritchina, E. A.; Gritsan, N. P. J. Photochem. Photobiol. A 1988, 43,
165.
(26) Huggenberger, C.; Lipscher, J.; Fischer, H. J. Phys. Chem. 1980,
84, 3467.
(25) Liu, J.; Hadad, C. M.; Platz, M. S. Org. Lett. 2005, 7, 549.
J. Org. Chem, Vol. 72, No. 8, 2007 2765

Muthukrishnan et al.
FIGURE 12. Effect of concentration on product ratio for azides 1.
FIGURE 13. Effect of light intensity on product ratio from azides 1.
SCHEME 6
nitrene-based products. As more is formed of 3 and less of 2
from 1 the ratio of [4] + [5] + 2[6] versus [4] + [7] will
increase. This ratio also increases when more of 2 cleaves to
form 3. Both azides 1a and 1b yield more products attributed
to benzoyl radical formation at lower concentration, which
supports the notion that nitrenes 2a,b both cleave to form
benzoyl radicals 3a,b. Formation of nitrenes 2 in less concen-
trated solution makes it more likely that they can fall apart
photochemically to form 3 before the nitrenes scavenge other
radicals. Since the product ratios are similar for both azides 1a
and 1b, it is likely that benzoyl radicals 3 are mainly formed
from cleavage of nitrenes 2.
III. Discussion
TD-B3LYP calculations confirm that T1K (n,π*) and T2K
(π,π*) are almost degenerate in azide 1a. Laser flash photolysis
of azide 1a shows that benzoyl radical 3a and nitrene 2a are
formed, presumably from T1K and T2K, respectively. Although
UB3LYP optimizations with the 6-31+G(d) basis set underes-
timates the energy of T1K in azide 1a by ∼9 kcal/mol, this
theoretical method does demonstrate that the transition state for
the R-cleavage is only a few kcal/mol above T1K. The R-cleav-
age in azide 1a is facilitated by the radical stabilization energy
of the azido group. We have shown that azido substituents have
a radical stabilization energy of 15 kcal/mol, which is similar
to that observed for phenyl and alkenyl substituents.27 In
comparison, laser flash photolysis of azide 1b shows only
formation of nitrene 2b, which comes from energy transfer from
T1K to the azido group. As T2K (n,π*) is 10 kcal/mol higher in
energy than T1K (π,π*) in azide 1b, the (n,π*) state is not
effectively populated at ambient temperature, and therefore, we
do not observe any benzoyl radical formation upon laser flash
photolysis of azide 1b.
F. Product Ratios as a Function of Light Power. We
irradiated toluene solutions of azides 1a,b with a high-intensity,
continuous-wave, argon-ion laser at five different laser powers
and analyzed the product ratios. In Figure 12 we plot the product
ratio ([4] + [5] + 2[6])/([4] + [7]) as a function of the laser
power. As the power of the light source was increased, this ratio
increased slightly for both azides 1a and 1b, which indicates
that more of benzoyl radical-based products are formed. These
findings also support the idea that nitrenes 2 can form benzoyl
radical 3 photochemically.
(27) Mandel, S. M.; Singh, P. N. D.; Muthukrishnan, S.; Chang, M.;
Krause, J. A.; Gudmundsdottir, A. D. Org. Lett. 2006, 8, 4207.
2766 J. Org. Chem., Vol. 72, No. 8, 2007

Competition in R-Azidoacetophenones
the UB3LYP level of theory, and each transition state was
confirmed to have one imaginary vibrational frequency by analytical
determination of the second derivatives of the energy with respect
to internal coordinates. IRC calculations were used to verify that
the located transition states corresponded to the attributed reactant
and product. Vertical UV absorption spectra were calculated at the
TD-B3LYP level with the 6-31+G(d) basis set using the optimized
B3LYP/6-31+G(d) geometry for the S0 state. The T1K of azide
1a was optimized using UB3LYP and TD-B3LYP with the
6-31+G(d) basis set as implemented in the TURBOMOLE package.
Appropriate zero-point energy (ZPE) corrections were made, and
wherever it was not possible to obtain the ZPE, the uncorrected
energies were compared. The effect of solvation was calculated
using the self-consistent reaction field (SCRF) method with the
integral equation formalism polarization continuum model (IEF-
PCM)15 with methanol as the solvent.
Laser flash photolysis demonstrates that triplet alkylnitrenes
2 are long-lived intermediates. Furthermore, we have previously
shown that triplet alkylnitrenes are highly unreactive and prefer
to dimerize with another nitrene molecule rather than react with
an azide precursor (Scheme 6),22 and Schuster et al. reported
similar findings for phenylnitrene intermediates.23 Alkylnitrenes
owe their stability to the fact that they are unreactive to H-atom
abstraction from the solvent and react very slowly with O2.22
Thus, alkylnitrenes decay mainly by dimerization unless they
are formed in the presence of radicals with which they can react.
Since photolysis of azide 1a forms triplet nitrene 2a and benzoyl
radical 3a simultaneously, it is reasonable to assume that the
triplet alkylnitrene 2a intercepts benzoyl radical 3a to form the
major product 4. Since product studies for azides 1b also yield
4 as the major product, even though laser flash photolysis does
not indicate formation of benzoyl radical 3b, we theorized
whether nitrenes 2 could undergo R-cleavage to form benzoyl
and iminyl radicals. Since, nitrenes 2 are long-lived and absorb
light above 300 nm, it is reasonable that they could absorb a
photon and undergo secondary photolysis to form benzoyl and
iminyl radicals. Molecular modeling shows that the transition
state for this cleavage is located 18 kcal/mol above the ground
state of nitrene 2a, and therefore at ambient temperature, it most
likely takes place photochemically rather than thermally. In
comparison to azides 1a,b where only azide 1a can undergo
R-cleavage, both nitrenes 2a and 2b are expected to undergo
R-cleavage because it is a ground-state reactivity for the
alkylnitrenes. Thus, one hypothesis is that the ketone or the
nitrene chromophores in nitrenes 2 can absorb light and then
decay to a vibrationally hot ground state of nitrenes 2, which
falls apart to give benzoyl and iminyl radicals. Similarly, Platz
et al. have shown that phenylnitrenes are also photolabile.2,28
Additionally, the hypothesis that nitrenes 2 are photoreactive
themselves is further supported by the fact that increases in the
laser power can produce slightly more of the products derived
from benzoyl radical formation. Moreover, as the concentration
of the azide precursor decreases, more benzoyl radical products
are observed, presumably because at lower concentration
nitrenes 2 are not intercepted as efficiently in a bimolecular
reaction with other radicals and can therefore efficiently absorb
light to subsequently fall apart.
Phosphorescence. The phosphorescence spectra were obtained
on a standard phosphorimeter.
Laser Flash Photolysis. Laser flash photolysis was done with
an excimer laser (308 nm, 17 ns). These systems have been
described elsewhere.29,30 A stock solution of azides 1a and 1b in
methanol was prepared with spectroscopic-grade methanol such that
the solutions had an absorption between 0.6 and 0.8 at 308 nm.
Typically, ∼1 mL of the stock solution was placed in a 10 × 10
mm wide, 48 mm long quartz cuvette and purged with nitrogen
for 5 min. The rates were obtained by fitting an average of 3-8
kinetic traces. The absorption spectra were measured using an
excimer laser in conjunction with an optical multichannel analyzer.
Synthesis of Azide 1 and Its Photoproducts. We previously
reported the preparation of azide 1a and its photoproducts.4
2-Bromo-1-(4-methylsulfanyl-phenyl) Ethanone. 2-Bromo-1-
(4-methylsulfanyl-phenyl) ethanone was prepared according to the
literature, and its IR and 1H NMR are identical to those previously
1
reported.31 IR (CHCl3): 1694, 1590, 1477, 1428 cm-1. H NMR
(250 MHz, CDCl3): δ 2.53 (s, 3H), 4.40 (s, 2H), 7.28 (d, 8.4 Hz,
2H), 7.89 (d, 8.4 Hz, 2H) ppm.
Synthesis of 2-Azido-1-(4-methylsulfanyl-phenyl) Ethanone
(1b). Azide 1b was synthesized following a procedure similar to a
method described by Boyer et al.32 2-Bromo-1-(4-methylsulfanyl-
phenyl) ethanone (1.3 g, 5 mmol) was dissolved in ethanol (20
mL), and glacial acetic acid (1 mL) was added while the solution
was kept between 0 and 5 °C. Sodium azide (690 mg, 10 mmol)
dissolved in minimum water was added, and the reaction mixture
was shaken vigorously and kept at 4 °C for 12 h. The reaction
mixture was extracted with diethyl ether (3 × 50 mL); the organic
layer was dried over anhydrous magnesium sulfate and evaporated
under vacuum. The residue was purified on a silica gel column
eluted with 10% ethyl acetate in hexane to yield 1b as pale yellow
solid (786 mg, 3.8 mmol, 80% yield). This solid was recrystallized
from ethanol to give crystals of 1b. The X-ray structure analysis
of 1b revealed that its crystals were orthorhombic with a P212121
space group (See Supporting Information).
IV. Conclusions
Laser flash photolysis of azide 1a produces nitrene 2a and
benzoyl radical 3a, whereas azide 1b yields only nitrene 2b.
Molecular modeling demonstrates that T1K (n,π*) and T2K
(π,π*) are almost degenerate in azide 1a and that T1K undergoes
R-cleavage to form benzoyl radical 3a, whereas energy transfer
from T2K to the azido moiety forms nitrene 2a. In comparison,
1
Mp: 76-77 °C. IR (neat): 2106, 1690, 1590 cm-1. H NMR
(250 MHz, CDCl3): δ 2.53 (s, 3H), 4.52 (s, 2H), 7.28 (d, 8 Hz,
2H), 7.80 (d, 8 Hz, 2H) ppm. 13C NMR (400 MHz, CDCl3): δ
192.6, 148.1, 128.9, 126.1, 125.9, 55.3, 15.3 ppm. MS (EI): m/z
(relative intensity) 207 (M +, 15), 179 (21), 152 (60), 151 (100),
135 (13), 123 (77), 122 (11), 121 (14), 108 (68), 79 (52), 77 (31).
HRMS: m/z calcd for C9H9OSN3 [M + H]+, 207.0466; found,
207.0475. UV-vis (methanol): λmax 203, 236, 309 (ꢀ ) 8100) nm.
Synthesis of Benzil 6b. 1,2-Bis-(4-methylsulfanyl-phenyl)-
ethane-1,2-dione was synthesized via a standard benzoin condensa-
T
1K (π,π*) in azide 1b is estimated to be 10 kcal/mol lower in
energy than T2K (n,π*), and thus, we do not observe any
R-cleavage from azide 1b. However, we suggest that both
nitrenes 2a and 2b undergo photochemical R-cleavage to form
benzoyl radicals 2. This notion was further supported by
molecular modeling and product studies.
V. Experimental and Computational Methods
(29) Gritsan, N. P.; Zhai, H. B.; Yuzawa, T.; Karweik, D.; Brooke, J.;
Platz, M. S. J. Phys. Chem. A 1997, 101, 2833.
(30) Cosa, G.; Scaiano, J. C. J. Photochem. Photobiol. 2004, 80, 159.
(31) Rahim, A. M.; Rao, P. N. P.; Knaus, E . E. Bioorg. Med. Chem.
Lett. 2002, 12, 2753.
Calculations. All geometries were optimized as implemented
in the Gaussian03 programs at the B3LYP level of theory and with
the 6-31+G(d) basis set.12,13 All transition states were located at
(28) Levya, E.; Platz, M. S.; Persy, G.; Wirz, J. J. Am. Chem. Soc. 1986,
108, 3783.
(32) Boyer, J. H.; Straw, D. J. Am. Chem. Soc. 1952, 74, 4506. (b) Boyer,
J. H.; Straw, D. J. Am. Chem. Soc. 1953, 75, 1642.
J. Org. Chem, Vol. 72, No. 8, 2007 2767

Muthukrishnan et al.
tion.33 4-Methylsulfanyl-benzaldehyde (1.54 g, 10 mmol) was
dissolved in ethanol (8 mL), KCN (0.28 g, 4.3 mmol) was added,
and the mixture was refluxed on a sand bath at 95 °C for 35 min.
The mixture was cooled in an ice bath, and the crystals that formed
were filtered, washed with water, air dried, and characterized as
2-hydroxy-1,2-bis(4-methylsulfanyl-phenyl) ethanone.
samples were photolyzed with a 450 W mercury arc lamp for 2 h
and analyzed by HPLC. A similar set of samples was also irradiated
with an argon-ion laser source for 30 min. The conversion was
kept below 20%. The HPLC traces showed formation of photo-
products 4a, 5a, 6a, and 7a as well as remaining starting material.
Standards for all compounds were available either commercially
or from previous experiments.4 Similar product ratios were observed
when azide 1 was photolyzed with a 450 W mercury arc lamp and
a continuous wave argon-ion laser.
Concentration Affect on the Product Ratio from Photolysis
of Azide 1b. Five different concentrations of azide 1b (0.100, 0.07,
0.05, 0.04, 0.03 M) in freshly distilled toluene were sealed in Pyrex
tubes after purging the solution with argon for 15 min. The samples
were then photolyzed with argon-ion laser7 source for 30 min and
analyzed by HPLC. The conversion was kept below 20%. The
HPLC traces showed formation of photoproducts 4b, 5b, 6b, and
7b as well as remaining starting material. The photoproducts were
characterized by injections of authentic samples. Compounds 5b
and 7b are commercially available, whereas 6b was synthesized
as described above. Photoproduct 4b was isolated from preparative
photolysis of azide 1b.
Mp: 102-103 °C (Lit.34 102-103 °C). IR (neat): 1730 cm-1
.
Cu(OAc)2 (0.1 g, 0.5 mmol) and NH4NO3 (5 g, 62 mmol) were
dissolved in water (7 mL), and glacial acetic acid (28 mL) was
added. This mixture (7 mL) was added to 2-hydroxy-1,2-bis-(4-
methylsulfanyl-phenyl) ethanone, and the resulting solution was
refluxed at 140 °C for 1 h. The mixture was cooled in an ice bath,
and the solid that formed was filtered and purified on a short silica
column eluted with dichloromethane. The solid was recrystallized
from ethanol to yield (930 mg, 3.07 mmol, 62% yield) shiny yellow
crystals of 1,2-bis-(4-methylsulfanyl-phenyl)-ethane-1,2-dione 6b.
Mp: 161-163 °C (Lit.34 161-164 °C). IR (neat): 2970, 1654,
1537 cm-1. 1H NMR (400 MHz, CDCl3): δ 2.52 (s, 6H), 7.27 (d,
8 Hz, 4H), 7.84 (d, 8 Hz, 4H) ppm. 13C NMR (100 MHz, CDCl3):
δ 193.5, 148.8, 130.1, 129.2, 125.2, 14.6 ppm. HRMS: m/z calcd
for C16H14O2S2 [M + Na]+, 325.0333; found, 325.0331.
Effect of Light Intensity on Product Ratios Upon Irradiation
of Azides 1. Five samples of 0.1 M solutions of each of azides 1
in freshly distilled toluene were sealed in Pyrex tubes after being
flushed with argon for 15 min and irradiated with an argon laser
light source of five different light intensities (0.03, 0.06, 0.09, 0.12,
0.15 W) for 30 min. The conversion was maintained below 20%
in all instances. The product ratios (4, 5, 6, and 7) were analyzed
by comparing with the standards available by normal-phase HPLC
with ethyl acetate:hexane (from 0% ethyl acetate to 40%) as eluant
and UV detection.
Preparative Photolysis of Azide 1b. An argon-purged solution
of 1b (300 mg, 1.4 mmol) in toluene (60 mL) was irradiated with
a medium-pressure Hg arc lamp through a Pyrex filter for 14 h.
HPLC analysis of the reaction mixture showed complete decom-
position of 1b and formation of a single major product 4b and minor
amounts 5b and 7b. The solvent was removed under vacuum, and
the resulting oil was purified on a silica column eluted with 20%
ethyl acetate in hexane to obtain 4-methylsulfanyl-N-[2-(4-meth-
ylsulfanyl-phenyl)-2oxo-ethyl]-benzamide (4b) as a viscous liquid
(145 mg, 0.44 mmol, 63% yield).
IR (CHCl3): 3364, 2834, 1633, 1591 cm-1. 1H NMR (250 MHz,
CDCl3): δ 2.53 (s, 3H), 2.54 (s, 3H), 4.90 (d, 4 Hz, 2H), 7.30 (m,
4H), 7.80 (d, 8 Hz, 2H), 7.93 (d, 8 Hz, 2H) ppm. 13C NMR (250
MHz, CDCl3): δ 193.8, 188.8, 167.4, 131.2, 130.6, 129.0, 128.7,
128.2, 126.1, 125.8, 47.3, 15.7, 15.3 ppm. MS (EI): m/z (relative
intensity) 331 (M+, 17), 313 (100), 299 (5), 298 (25), 243 (6), 211
(5). HRMS: m/z calcd for C17H17O2S2N [M + H]+, 331.0701;
found, 331.0699.
Concentration Affect on the Product Ratio from Photolysis
of Azide 1a. Five different concentrations of azide 1a (0.100, 0.07,
0.05, 0.04, 0.03 M) in freshly distilled toluene were sealed in Pyrex
tubes after purging the solution with argon for 15 min. These
Acknowledgment. We thank the National Science Founda-
tion, Petroleum Research Foundation, and the Ohio Supercom-
puter Center for supporting this work. We also thank Professor
M. S. Platz at The Ohio State University and Professor J. C.
Scaiano at the University of Ottawa for use of their laser flash
photolysis instruments. We are grateful to Professor R. M.
Wilson at the University of Cincinnati for allowing us to use
his argon-ion laser.
Supporting Information Available: Cartesian coordinates,
energies, vibrational frequencies, spectra of photoproducts, and CIF
files for X-ray. This material is available free of charge via the
Internet at http://pubs.acs.org.
(33) Mayo, D. W.; Pike, R. M.; Trumper, P. K. Microscale Organic
Laboratory, 4th. Ed.; John Wiley & Sons: New York, 2000.
(34) Lombardino, J . G.; Wiseman, E. H. J. Med. Chem. 1974, 17, 1182.
JO062160K
2768 J. Org. Chem., Vol. 72, No. 8, 2007