Solid-state photolysis of α-azidoacetophenones

Article abstract of DOI:10.1021/ol061398s

(Chemical Equation Presented) Solid-state photolysis of 1 yields 2 in a crystal-to-crystal reaction. The reaction takes place by α-cleavage to form a benzoyl and an azido alkyl radical pair. The azido alkyl radicals rearrange into iminyl radicals and N₂. The iminyl and benzoyl radicals are held in close proximity within the crystal lattice, which allows them to combine and form 2. X-ray structure analysis, molecular modeling and trapping studies support this mechanism.

Full text of DOI:10.1021/ol061398s

ORGANIC
LETTERS
2006
Vol. 8, No. 19
4207-4210
Solid-State Photolysis of
-Azidoacetophenones
r
Sarah M. Mandel, Pradeep N. D. Singh, Sivaramakrishnan Muthukrishnan,
Mingxin Chang, Jeanette A. Krause, and Anna D. Gudmundsdo´ttir*
Department of Chemistry, UniVersity of Cincinnati, Cincinnati, Ohio 45221-0172
anna.gudmundsdottir@uc.edu
Received June 8, 2006
ABSTRACT
Solid-state photolysis of 1 yields 2 in a crystal-to-crystal reaction. The reaction takes place by
r
-cleavage to form a benzoyl and an azido alkyl
radical pair. The azido alkyl radicals rearrange into iminyl radicals and N2. The iminyl and benzoyl radicals are held in close proximity within
the crystal lattice, which allows them to combine and form 2. X-ray structure analysis, molecular modeling and trapping studies support this
mechanism.
Crystals can be used to control the reactivity of reactive
intermediates such as nitrenes, carbenes, biradicals, and
radical pairs.1 The rigidity of the crystals inhibits movement
and diffusion of the intermediates trapped inside, and thus,
their reactivity in the solid-state is more selective than in
solution. However, despite solid-state reactions being gener-
ally both regio- and stereoselective, they are not commonly
used in synthetic applications.2 This is mainly because
bimolecular reactions are limited by how the molecules pack
in the crystal lattice, and it is difficult to predict crystal
packing arrangements. In comparison, unimolecular reactions
in the solid-state that are controlled by intrinsic molecular
properties rather than crystal packing arrangements can be
more easily envisioned. For example, Garcia-Garibay et al.
find that ketones with radical stabilizing groups in the
R-positions undergo solid-state photodecarbonylation to form
a radical pair and CO.2,3 The constraint of the crystal lattice
forces the bonding between the radical pair to take place in
a highly chemoselective and stereospecific manner. Thus,
Garcia-Garibay et al. used solid-state photodecarbonylation
to synthesize natural products.3
Nitrogen-centered radicals are less commonly used than
carbon-based radicals in organic synthesis.4 In this paper,
we describe a new method to form iminyl radicals within
crystals and demonstrate how these radicals can be used to
selectively form carbon-nitrogen bonds. Solid-state pho-
tolysis of R-azidoacetophenone derivatives 1 (see Scheme
1) yields 2 in a crystal-to-crystal reaction. We propose that
the radical-stabilizing effect of the azido group facilitates
R-cleavage in 1 to form a benzoyl and azido alkyl radical
pair. The azido alkyl radical then rearranges to form an
iminyl radical. The iminyl and benzoyl radicals, held in close
proximity by the crystal lattice, combine to form 2.
To support this mechanism, we analyzed the X-ray
structure of 1 and trapped the benzoyl radical formed in the
solid state with oxygen. Furthermore, we used molecular
modeling to verify the radical-stabilizing effect of the azido
group. To the best of our knowledge, this is the first time
that the reactivity of azido alkyl radicals has been reported.
(1) (a) Scheffer, J. R.; Xia, W. Top. Curr. Chem. 2005, 254, 233. (b)
Garcia-Garibay, M. A. Acc. Chem. Res. 2003, 36, 491. (c) Garcia-Garibay,
M. A.; Campos, L. M. In CRC Handbook of Organic Photochemisty &
Photobiology, 2nd ed.; Horsepool, W. M., Ed.; CRC Press: Boca Raton,
2003; Chapter 48. (d) Sakamoto, M. Chem. Eur. J. 1997, 3, 684. (e)
Hollingsworth, M. D.; McBride, J. M. AdV. Photochem. 1990, 15, 279. (f)
Ramamurthy, V.; Venkatesan. K. Chem. ReV. 1987, 87, 433. (g) Desiraju,
G. R. Organic Solid-State Chemistry; Elsevier: Amsterdam, 1987.
(2) (a) Campos, L. M. Dang, H.; Ng, D.; Martinez, H. L.; Garcia-Garibay,
M. A. J. Org. Chem. 2002, 67, 3749. (b) Zhe, Y.; Ng, D.; Garcia-Garibay,
M. A. J. Org. Chem. 2001, 66, 4468. (c) Yang, Z.; Garcia-Garibay, M. A.
Org. Lett. 2000, 2, 1963.
(3) (a) Ng, D.; Yang, Z.; Garcia-Garibay, M. A. Org. Lett. 2004, 6, 645.
(b) Ellison, M.; Ng, D.; Dang, H.; Garcia-Garibay, M. A. Org. Lett. 2003,
5, 2531. (c) Resendiz, M. J. E.; Garcia-Garibay, M. A. Org. Lett. 2005, 7,
371. (d). Garcia-Garibay, M. A.; Mortko, C. J. J. Am. Chem. Soc. 2005,
127, 7994.
(4) El Kaim, L.; Meyer, C. J. Org. Chem. 1996, 61, 1556. (b) Ryu, I.;
Matsu, K.; Minakata, S.; Komatsu, M. J. Am. Chem. Soc. 1998, 120, 5838.
10.1021/ol061398s CCC: $33.50
© 2006 American Chemical Society
Published on Web 08/16/2006

Scheme 1
Scheme 4
that corresponded to trapping of triplet alkyl nitrene with a
benzoyl radical, indicating that R-cleavage in 1d is more
predominant than nitrene formation. Azide 1d must undergo
R-cleavage more efficiently than 1a-c, due to the additional
radical stabilization effect of the R-phenyl group. However,
it is possible that 3d can undergo 1,2-acyl shift and
intersystem cross to form 2d. Triplet alkyl nitrenes,5a like
aryl nitrenes,7 can be trapped with molecular oxygen to form
nitro compounds, whereas benzoyl radicals5a,8 react with
molecular oxygen to form benzoic acids. Subsequent pho-
tolysis of 1d in oxygen-saturated toluene yielded 9, 8, and
2d as the major products, thus signifying that the major
reactivity of 1d is R-cleavage.
We previously reported the photoreactivity of 1a-c in
solution and found that these molecules undergo triplet
energy transfer to form triplet alkyl nitrenes 3 (see Scheme
2).5 In competition with the triplet energy transfer, 1
Scheme 2
Solid-state irradiation of 1 cracked the crystals and turned
them into powder. Dissolution of the irradiated crystals
released gas, presumably N2 that must have been trapped in
the crystal lattice. GC-MS analysis of irradiated crystals of
1 that were dissolved in anhydrous ethyl acetate indicated
that benzamides 2 were the only product, with no remaining
starting materials. Padwa et al.9 have shown that N-methylene
benzamide is hydrolyzed to benzoic amide in the presence
of moisture, and we also found that 2a-c are not stable in
the presence of moisture.10 To confirm that 2a-c were the
primary photoproducts, the irradiated crystals were reduced
with sodium borohydrate cyanide or butyl triphenylphos-
phonium tetraborate to yield 12a-c as the only products
(Scheme 5). Product 2d, which is less sensitive to moisture,
was characterized directly.
undergoes R-cleavage to yield benzoyl radicals 4 and azido
methyl radicals. Nitrenes 3 and benzoyl radicals 4 are both
long-lived intermediates that mainly combine to form 5. We
were, however, not able to isolate any products that could
be attributed to the trapping of azido methyl radicals.
Presumably, the R-cleavage comes from the n,π* configu-
ration of the triplet ketone moiety in 1,6 whereas the π,π*
triplet state leads to energy transfer.
Since we had not previously studied the solution photo-
chemistry of 1d, we photolyzed it in argon-saturated toluene.
The major product formed was 2d, with a lesser amount of
products 6-8 (Scheme 3). All the products can be attributed
Scheme 3
Scheme 5
To further confirm that 2 is the primary photoproduct, we
followed the solid-state reaction of 1a in a KBr pellet (Figure
1) with IR spectroscopy. Irradiation of the pellet led to
depletion of the azide band and formation of new bands,
the most significant occurring at 1692 and 1643 cm-1. These
bands fit with the calculated vibrational bands (with scaling
to R-cleavage of 1d to form benzoyl and azido benzyl
radicals (see Scheme 4). The azido benzyl radical 10p can
transform into 11p and combine with a benzoyl radical to
form 2d. Abstraction of an H atom from 11p would yield 8,
whereas 6 and 7 must come from reaction of the benzoyl
radical with the solvent. We did not observe any product
(7) (a) Liu, J.; Hadad, C. M.; Platz, M. S. Org. Lett. 2005, 7, 549. (b)
Abramovitch, R. A.; Challand, S. R. Chem. Commun. 1972, 964.
(8) (a) Neville, A. G.; Brown, C. E.; Rayner, D. M.; Lusztyk, J.; Ingold,
K. U. J. Am. Chem. Soc. 1991, 113, 1869. (b) Brown, C. E.; Neville, A. G.;
Rayner, D. M.; Ingold, K. U.; Lusztyk, J. Aust. J. Chem. 1995, 48, 363.
(9) Padwa, A.; Crosby, A. J. Org. Chem. 1974, 39, 2651.
(5) (a) Singh, P. D. N.; Mandel, S. M.; Zhu, Z.; Franz, R.; Ault, B. S.;
Gudmundsdottir, A. D. J. Org. Chem. 2003, 68, 7951. (b) Mandel, S. M.;
Krause Bauer, J. A.; Gudmundsdottir, A. D. Org. Lett. 2001, 3, 523.
(6) Turro, N. Modern Molecular Photochemistry; University Press:
Menlo Park, CA, 1978; Chapter 7.
(10) Gudmundsdottir, A. D.; Liebman, J. F. Struct. Chem. 2005, 16, 155.
4208
Org. Lett., Vol. 8, No. 19, 2006

Figure 1. Photoreaction of 1a in KBr pellets.
of 0.96) for the coupled carbonyl and imine stretches at 1678
and 1646 cm-1 in 2a.11
The progress of solid-state reactions are most often limited
by the solubility of the products in the reactant crystals.12
Thus, to prevent any melting of the crystalline phase, the
solid-state photolyses of 1 were performed at 0 °C. While
photolyzing 1b, we followed the melting point of the reaction
mixture. Azide 1b melted between 52 and 54 °C, whereas
after ∼30% conversion, the reaction mixture melted between
49 and 50 °C. At 70% and 100% conversion, the irradiated
crystals melted at 62-64 and 65-66 °C, respectively. Thus,
the eutectic point for the mixed crystals of 1b and 2b must
be above 0 °C. Since, the solid-state photolysis of 1b was
done below the eutectic point of 1b and 2b, the reaction
proceeded to high conversion. Similarly, we expected the
eutectic point of 1a,c,d and their corresponding products to
be above 0 °C; however, it should be pointed out that the
crystals of 2d melt at ambient temperature.
Solid-state photolysis of 1a-c selectively yielded 2a-c;
these were not observed in solution. In comparison, solution
and solid-state irradiation of 1d gave the same major product,
2d, but it was formed exclusively in crystals. To support
the notion that formation of 2 in the solid state is a result of
R-cleavage of 1, we analyzed the crystal structures of 1a-d
(Figure 2). Compounds 1a,b adopt an approximately planar
backbone with the azido group bending out of the plane by
∼70° (torsion about C-N1). A similar geometry is observed
for 1c however the p-phenyl moiety twists out of the plane
by 37.5°. The carbonyl and the azido groups are in the syn
arrangement, thus the distance from the carbonyl carbon atom
C7 to N1 for the recombination of the radicals formed from
R-cleavage is very short: 1a, 2.48 Å; 1b-d, 2.49 Å. Azides
Figure 2. X-ray structures of 1a and 1d.
1a-c adopt a π-stacked motif with the azide groups on
adjacent molecules crossing each other. Azide 1d also packs
in a stacked manner with the azide groups pointing toward
each other.
To confirm that the solid-state reactivity of 1 comes from
R-cleavage, we trapped the intermediate formed in the solid-
state with oxygen. We ground crystals of 1d to increase their
surface area, placed them under oxygenic atmosphere and
photolyzed them. GC-MS analysis of the irradiated crystals
showed that the major product formed was benzoic acid and
benzonitrile, which support that the solid-state reaction takes
place via R-cleavage.
Photolyzing 1d in acetonitrile glass at -77 °C, also yields
2d in excellent yield and trace amounts of 7 and 8. Thus,
the immobility of the reaction medium is critical to obtain
high yields of 2 rather than the crystal packing arrangement
of 1.
Molecular modeling was also used to validate the R-cleav-
age pathway. Structural optimization of 1 and 2 (see
Supporting Information)11 shows the lowest energy confor-
mations of 1 are similar to those obtained by X-ray analysis.
The energies of the triplet-excited states in 1 were calculated
using time-dependent density functional theory (TD). The
energies of the first and second excited states of the triplet
ketone (T1 and T2, see Table 1 and Figure 3) are within 3
kcal/mol of each other in 1a, 1b, and 1d, whereas for 1c
(11) 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.; Bakken, V.; 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 C.02; Gaussian, Inc.: Wallingford, CT, 2004.
Table 1. Energies (kcal/mol) of Stationary Points on the
Triplet Surface for R-Cleavage of 1
T1
T2 TS1 4 + 10 TS2 4 + 11 + N2 2 + N2
1a 73 (π,π*) 75
1b 73 (π,π*) 74
1c 67 (π,π*) 74
1d 72 (n,π*) 75
77
77
77
73
67
67
67
53
68
68
68
56
23
23
23
15
-48
-48
-48
-60
(12) Keating A. E.; Garcia-Garibay, M. A. In Molecular and Supramo-
lecular Photochemistry; Ramamurthy, V., Schanze, K., Eds.; Marcel
Dekker: New York, 1998; Vol. 2, pp 195-248.
Org. Lett., Vol. 8, No. 19, 2006
4209

reaction i.14 The enthalpy change can be obtained as shown
in eq ii using molecular modeling to calculate the heats of
formation (see the Supporting Information). The RSE for
the azido group was calculated to be -15 kcal/mol,
demonstrating that the azido methyl radical is more stable
than a methyl radical. Thus, the azido group has a similar
radical stabilization energy as observed for phenyl and
alkenyl substituents.15 As expected, the calculations also
show that R-cleavage for 1d is more favorable than for 1a-
c, due to the additional resonance stabilization of the phenyl
substituent. This explains why 1d undergoes R-cleavage in
solution and not nitrene formation.
Figure 3. Stationary points on the triplet energy surface for 1a.
In conclusion, we have shown that the radical-stabilizing
effect of the azido groups in 1 makes R-cleavage to form
benzoyl and azido alkyl radicals feasible. The azido alkyl
radical rearranges into an iminyl radical and, since the crystal
lattice prevents any diffusion, selective bond formation
between this radical pair takes place. Thus, R-cleavage of
R-azido ketones in the solid-state can be used for selective
carbon-nitrogen bond formation.
they are slightly further apart. TD assumes a Franck-Condon
excitation of the optimized ground-state geometry; therefore,
the calculations are not accurate enough to predict whether
the n,π* or the π,π* configuration is lower in energy, but
only to confirm that T1 and T2 are of similar energies in
1a, 1b, and 1d. Our findings are also in agreement with
previous estimates of T1 and T2 energies for the valerophe-
none derivatives.13
Furthermore, we calculated the triplet transition state for
R-cleavage of 1 into radicals 4 and 10 (see Table 1). The
transition states were 1-3 kcal/mol above the n,π* config-
uration of the triplet ketone. The calculated transition state
for 10h to form 11h and N2 was less than 1 kcal/mol above
10h, and the transition state for 10p to form 11p was 2.5
kcal/mol above 10p. Thus, both the R-cleavage and the
rearrangement of 10 to 11 are easily accessible processes
arising from the triplet ketone in 1. Since we did not observe
any products in the solid-state that can be attributed to the
formation of triplet alkyl nitrenes, the crystalline phase must
be able to stabilize the n,π* configuration of the triplet ketone
better than the triplet π,π* state that is responsible for the
energy transfer to form alkyl nitrenes.
Acknowledgment. This work was supported by the
National Science Foundation, the Petroleum Research Foun-
dation, and the Ohio Supercomputer Center. X-ray data was
collected through The Ohio Crystallographic Consortium
(Ohio Board of Regents 1995 Investment Fund).
Supporting Information Available: Complete crystal-
lographic details for 1a-d (CIF). Details of experimental
procedures and calculations. This material is available free
of charge via the Internet at http://pubs.acs.org.
OL061398S
(13) (a) Wagner, P. J.; Park, B.-S. Photoinduced hydrogen atom
abstraction by carbonyl compounds. Organic Photochemistry; Padwa, A.,
Ed.; Marcel Dekker Inc. New York, 1991; Vol. 11, p 227. (b) Wagner. P.
J.; Kemppainen, A. E.; Schott, H. N. J. Am. Chem. Soc. 1973, 95, 5604.
(c) Baum, E. J.; Wan, J. K. S.; Pitts, J. N., J. Am. Chem. Soc. 1966, 88,
2652.
(14) Parkinson, C. J.; Mayer, P. M.; Radom, L. J. Chem. Soc., Perkin
Trans. 2 1999, 2305.
(15) Henry, D. J.; Parkinson, C. J.; Mayer, P. M.; Radom, L. J. Phys.
Chem. A 2001, 105, 6750.
‚CH3 + CH3X f ‚CH2X + CH4
(i)
RSE(X) ) ∆rH ) Σ∆Hf(prod) - Σ∆Hf(react)
(ii)
Generally, the radical-stabilizing energy (RSE) of a
substituent is defined as the enthalpy change in the isodemic
4210
Org. Lett., Vol. 8, No. 19, 2006