Long-range through-bond photoactivated σ bond cleavage in steroids. Intramolecular sensitized debromination

Article abstract of DOI:10.1021/ol990319s

(formula presented) The photolysis of 17α-bromo-3α-(triphenylsilyloxy)-5α-androstane (2; 3αTPSO/17αBr) and 17α-bromo-3α-(triphenylsilyloxy)-5α-androstan-6-one (3; 3αTPSO/6ketone/17αBr) is described. Irradiation of 2 with 266 nm light leads to debrominatio

Full text of DOI:10.1021/ol990319s

ORGANIC
LETTERS
2000
Vol. 2, No. 1
15-18
Long-Range Through-Bond
Photoactivated σ Bond Cleavage in
Steroids. Intramolecular Sensitized
Debromination1
Wen-Shan Li and Harry Morrison*
Department of Chemistry, Purdue UniVersity, West Lafayette, Indiana 47907
morrison@science.purdue.edu
Received October 11, 1999
ABSTRACT
The photolysis of 17r-bromo-3r-(triphenylsilyloxy)-5r-androstane (2; 3rTPSO/17rBr) and 17r-bromo-3r-(triphenylsilyloxy)-5r-androstan-6-
one (3; 3rTPSO/6ketone/17rBr) is described. Irradiation of 2 with 266 nm light leads to debromination via intramolecular transfer of triplet
excitation energy with a quantum efficiency of 0.0011. Photolysis of 3 with both 266 and 308 nm light leads to debromination with quantum
efficiencies of ca. 0.0066. The debromination of 3 is attributed to activation via the ketone excited singlet state, with singlet energy transfer
from C6 to C17 ca. 35% efficient and occurring with a rate constant of 1.4 × 108 s-1.
Recently, there have been a number of reports that confirm
the capability of the steroid skeleton to function as a
“photonic wire”, thereby facilitating relatively long-range
through-bond intramolecular energy transfer.2-6 In our own
studies, we have employed photochemically active moieties
as the terminal excited-state energy acceptors, with π systems
(e.g., ketones and alkenes) filling this role. However, there
has also been extensive interest in the photochemical
activation of carbon-halogen bonds through interaction with
distal chromophores in rigid systems. These include chlo-
ronorbornenes (and analogues thereof)7 and chlorobenzo-
bicyclics.8-12 We now report that the steroid framework will
also facilitate distal cleavage of a σ, i.e., C-Br, bond. It
should be noted that the intermolecular sensitization of
carbon-halogen cleavage is well known. Though it is
typically initiated through electron-transfer chemistry, sen-
(1) Organic Photochemistry. 118. For part 117, see: Timberlake, L. D.;
Morrison, H. J. Am. Chem. Soc. 1999, 121, 3618.
(7) Pearl, D. M.; Burrow, P. D.; Nash, J. J.; Morrison, H.; Jordan, K. D.
J. Am. Chem. Soc. 1993, 115, 9876. Nash, J. J.; Carlson, D. V.; Kaspar, A.
M.; Love, D. E.; Jordan, K. D.; Morrison, H. J. Am. Chem. Soc. 1993, 115,
8969. Maxwell, B. D.; Nash, J. J.; Morrison, H. A.; Falcetta, M. L.; Jordan,
K. D. J. Am. Chem. Soc. 1989, 111, 7914.
(8) (a) Cristol, S. J.; Bindel, T. H. Org. Photochem. 1983, 6, 327. (b)
Cristol, S. J.; Aeling, E. O.; Strickler, S. J.; Ito, R. D. J. Am. Chem. Soc.
1987, 109, 7101 and other papers in this series.
(2) (a) Wu, Z-Z.; Morrison, H. J. Am. Chem. Soc. 1989, 111, 9267. (b)
Wu, Z.-Z.; Morrison, H. J. Am. Chem. Soc. 1992, 114, 4119. (c) Wu, Z.-
Z.; Nash, J. Morrison, H. J. Am. Chem. Soc. 1992, 114, 6640. (d) Agyin,
J. K.; Morrison, H.; Siemiarczuk, A. J. Am. Chem. Soc. 1995, 117, 3875.
(e) Jiang, S. A.; Xiao, C.; Morrison, H. J. Org. Chem. 1996, 61, 7045. (f)
Agyin, J. K.; Timberlake, L. D.; Morrison, H. J. Am. Chem. Soc. 1997,
119, 7945.
(3) Liang, N.; Miller, J. R.; Closs, G. L. J. Am. Chem. Soc. 1990, 112,
5353.
(9) Post, A. J.; Nash, J. J.; Love, D. E.; Jordan, K. D.; Morrison, H. J.
Am. Chem. Soc. 1995, 117, 4930 and other papers in this series.
(10) Seapy, D. G.; Gonzales, J.; Cameron, K. O. J. Photochem.
Photobiol., A 1992, 64, 35.
(11) For an interesting recent study of intramolecular electron transfer
induced electrochemically in 1-methyl-4-benzyloxycyclohexyl bromide,
see: Antonello, S.; Maran, F. J. Am. Chem. Soc. 1998, 120, 5713.
(12) Lodder, G.; Cornelisse, J. Recent AdVances in the Photochemistry
of the Carbon-Halogen Bond; Wiley and Sons, 1995.
(4) Tong, Z.-H.; Yang, G.-Q.; Wu, S.-K. Chinese J. Chem. 1990, 61.
(5) Cao, H.; Akinoto, Y.; Fujiwara, Y.; Tanimoto, Y.; Zhang, L.-P.; Tung,
C.-H. Bull. Chem. Soc. Jpn. 1995, 68, 3411. Tung, C.-H.; Zhang, L.-P.;
Yi, L. Chinese. J. Chem. 1996, 14, 377. Tung, C.-H.; Zhang, L.-P.; Yi, L.;
Cao, H.; Tanimoto, Y. J. Phys. Chem. 1996, 100, 4480. Tung, C.-H.; Zhang,
L.-P.; Yi, L.; Cao, H.; Tanimoto, Y. J. Am. Chem. Soc. 1997, 119, 5348.
(6) Zhu, Y.; Schuster, G. B. J. Am. Chem. Soc. 1993, 115, 2190.
10.1021/ol990319s CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/16/1999

sitization by energy transfer from both aromatic and ketone
donors has been reported.13-16
THF-H2O (2:1) gave a product mixture consisting of a small
amount of the debromination product 1 (3R-(triphenylsilyl-
oxy)-5R-androstane)23 admixed with uncharacterized isomers
of 2 involving transformations of the TPSO group. The
quantum efficiency for formation of 1 was determined to be
0.0011 while that for disappearance of the starting material
was 0.064. That debromination was induced by intramo-
lecular triplet sensitization was confirmed by the photolysis
of 2 (6 mM) in the presence of 6 mM 3R-hydroxy-5R-
androstan-17R-bromide (3ROH/17RBr) and, in a second
experiment, in the presence and absence of 1.2 mM cis-
piperylene. No debromination of the 3ROH/17RBr was
observed,24 and the debromination of compound 2 was
quenched by 80% by the diene.25
Our initial efforts focused on 17R-bromo-3R-(triphenyl-
silyloxy)-5R-androstane (2, 3RTPSO/17RBr). We reported
earlier that a donor (dimethylphenylsiloxy; DPSO) group at
C3 can transfer both singlet2a-c and triplet1,2d,f energy to
acceptors at C17. The latter, in particular, requires the
exchange mechanism which, with the rates observed at the
ca. 13 Å separation of the C3 and C17, necessitates that
through-bond interaction (TBI) facilitated transfer be opera-
tive.1 The DPSO group, chosen as the antenna in the earlier
work because of its favorable properties,17 was, in this study,
replaced by the TPSO group because of its diminished
sensitivity to acid.18,20,21
Irradiation of 2 (6 mM) at 266 nm22 in the presence of
NH4OH (12 mM; to serve as an HBr scavanger) in degassed
Though this chemistry demonstrated that steroid-mediated
triplet sensitization of C-Br cleavage had been accom-
plished, its low efficiency (a minimum26 of 1% assuming
that the æisc for the TPSO group is ca. 8%)1,18 prompted us
to determine whether the presence of a ketone at C6 might
facilitate energy transfer. In earlier work we observed a
significant enhancement of both singlet2a-c,e and triplet1,2e,f
(13) Cristol, S. J.; Micheli, R. P. J. Am. Chem. Soc. 1978, 100, 850 and
references therein.
(14) Charlton, J. L.; Williams, G. J. Tetrahedron Lett. 1977, 1473.
(15) Kropp, P. J.; Poindexter, G. S.; Pienta, N.; C, H. D. J. Am. Chem.
Soc. 1976, 98, 8153.
(16) Dannenberg, J. J.; Dill, K. Tetrahedron Lett. 1972, 1571.
(17) Morrison, H.; Agyin, K.; Jiang, A.; Xiao, C. Pure Appl. Chem. 1995,
67, 111.
(18) The TPSO fluorescence efficiency (0.010) and lifetime (2.1 ns) in
a model system, 3R-(triphenylsilyloxy)-5R-androstane (1), are similar to
those observed for the DPSO group (0.005 and 1.2 ns, respectively).19
(19) Wu, Z.-Z.; Morrison, H. Tetrahedron Lett. 1990, 31, 5865.
(20) Compound 2 was prepared by displacement of the triflate of 17â-
hydroxy-5R-androstan-3-one with lithium bromide, followed by reduction
with LiHAl[OC(CH3)3]3 and silylation of the C3 R-alcohol by TPSCl. The
configurations of the TPSO and bromine groups were assigned by NMR
and confirmed by X-ray analysis of 3.
(22) Using a Continuum NY-61 Nd:YAG laser equipped with a frequency
quadrupler (30 mW, 10 Hz, ca. 3.0-3.3 mJ/pulse) and with a 2× beam
enlarger in front of a 1 cm square Vycor cell.
(23) Prepared from 3R-hydroxy-5R-androstan-17-one by Wolff-Kishner
reduction of the ketone and silylation of the alcohol.
(24) Nor was there debromination of the 3ROH/17RBr by 266 nm light
when this compound was irradiated by itself, thus ensuring that the
photochemistry of 2 is initiated by light absorption by the TPSO antenna.
(25) Interestingly, the TPSO chemistry was unquenched by piperylene.
(26) This is a minimum value since the extent of recombination of the
initially created radical or ion pairs cannot be determined from the data in
hand.
(21) Satisfactory analytical and spectroscopic data have been obtained
for all new compounds reported herein.
16
Org. Lett., Vol. 2, No. 1, 2000

Figure 1. X-ray structure of 3.
energy transfer to C17 after initial intraSSET from a C3 aryl
antenna to a C6 carbonyl group (triplet energy transfer
occurring after intersystem crossing by the ketone). Toward
these ends, we prepared 17R-bromo-3R-(triphenylsilyloxy)-
5R-androstan-6-one (3, 3RTPSO/6ketone/17RBr)27 and con-
firmed the structure by X-ray analysis (Figure 1).
It is noteworthy that the TPSO moiety points away from
the C6 carbonyl group, thus lessening the potential for a
through-space interaction between the two groups. This
arrangement is contrary to that observed for an analogous
compound containing a DPSO antenna and C6 ketone (with
an alkylidene group at C17).2f Nevertheless, a comparison
of the aryl fluorescence quantum efficiencies for compounds
1-3 confirms that efficient intraSSET from the TPSO group
to the C6 ketone is occurring. Thus, æf for 1 and 2 are 1.0
× 10-2 and 9.3 × 10-3, respectively, both some 10-fold
higher than that for 3 (9.5 × 10-4). Using the standard
expression2 that æintraSSET ) [æf(1) - æf(3)]/æf(1), one
calculates that intraSSET from the TPSO group to the C6
carbonyl group is 90% efficient. (Note that the fluorescence
quantum efficiencies for 1 and 2 are comparable, evidence
that the C17 bromine has a minimal effect on the TPSO
singlet state.28)
17R-bromo-6â-hydroxy-3R-TPSO-5R-androstane (5), and a
pair of solvent photoalkylation products, 6 and 7. The latter
are presumed to be the â hydroxy, R and S tetrahydrofuryl
isomers resulting from R attack, but a definitive assignment
requires more extensive spectral characterization. The fifth
product was the anticipated debromination product, 3R-
TPSO-5R-androstan-6-one (8).21,30,31 The same products were
formed when 3 was photolyzed with 308 nm light.32
Quantum efficiencies were determined for the loss of 3 and
the formation of the products at both excitation wavelengths
and are also shown in eq 1. It is clear that activation at C17
Irradiation of 3, as described for 2 above, resulted in the
formation of five primary photoproducts (eq 1). Four of these
are derived from ketone photochemistry, i.e., the alcohols,
17R-bromo-6R-hydroxy-3R-TPSO-5R-androstane (4) and
(29) Prepared from dehydroisoandrosterone in four steps.
(30) The structures are assigned on the basis of spectral analysis. Those
for compounds 4, 5, and 8 were confirmed by comparison with indepen-
dently prepared authentic samples.
(31) This is clearly a product resulting from homolysis of the C-Br bond.
There was no evidence for the C17 alcohol one might have anticipated had
heterolysis been observed. For a recent, relevant discussion, see: Lipson,
A.; Deniz, A. A.; Peters, K. S. Chem. Phys. Lett. 1998, 288, 781 and
references therein.
(27) Prepared from testosterone via initial conversion to the 17â hydroxy,
3,6 dione, conversion of the alcohol to the 17R bromide, and selective
reduction and silylation at C3.
(28) The aryl emission of 1 was found to be unaffected by the presence
of an equimolar concentration of 3R-hydroxy-5R-androstan-6-one (3ROH/
6ketone),29 thus confirming that there is no intermolecular interaction
between the C6 carbonyl group and the TPSO singlet state at the
concentrations (ca. 10-4 M) used in the emission experiments.
(32) Using a Luminics EX-700 Pulse Master Excimer Laser at 308 nm,
30 mW power (10 Hz, 3.5 mJ/pulse).
Org. Lett., Vol. 2, No. 1, 2000
17

has been enhanced by the presence of the ketone; there is a
6-fold increase in the efficiency of the debromination reaction
of 3 relative to 2.
Alternatively, but less directly, one can determine the
reduction in æisc in the two steroids caused by the C6 ketone/
C-Br singlet interaction by measuring the relative amount
of triplet reduction at C6 by the solvent for the two ketones.
A 37% diminution in C6 reduction products is observed when
there is a bromine at C17. This compares favorably with
the 32% reduction observed for the singlet lifetime (see
above). Assuming that æisc ∼ 1.0 in the absence of the
halogen, and that all the reduction in the presence of the
bromine is due to energy transfer (i.e., æC6fC17Br ) 0.37),
one calculates kC6fC17Br ) 1.5 × 108 s-1. The average of the
two values, 1.4 × 108 s-1, may be compared to the value of
8.3 × 108 s-1 measured for triplet energy transfer from a
C6 ketone to a C17 ethylidene group.2f,35 Extending the
assumption that the 37% reduction in æisc reflects a corre-
sponding level of activation of the C17 halogen, and with
the assumption that C-Br activation quantitatively leads to
cleavage, one calculates that 0.0066/0.37 or ∼2% of C-Br
cleavage results in product.
Apparently the ketone triplet state either lacks sufficient
energy to initiate cleavage or cannot do so within its lifetime
in the THF medium. Computations are in progress, but it
seems likely that the mode of activation of the C-Br bond
is through TBI coupling of the carbonyl n,π* singlet state
with the C-Br n,σ* state, by analogy with the proposed
mode of activation of the carbon-halogen bond in the
chloronorbornenes7 and chlorobenzobicyclics.9
To confirm that excitation energy transfer in 3 is intramo-
lecular (as in 2) 6 mM 3 was photolyzed in the presence of
an equimolar concentration of 3ROH/17RBr with both 266
and 300 nm light. No debromination of the 3ROH/17RBr
was observed in either case. Nor, in fact, was debromination
observed upon the 300 nm irradiation of 3ROH/17RBr with
an equimolar concentration of 3ROH/6ketone under the usual
conditions, or by itself in an acetone solution.
Because intraSSET from the TPSO group to the C6 ketone
is so efficient, it is reasonable to assume that the excitation
energy transmitted to C17 in 3 derives solely from the
carbonyl group. Both the ketone singlet (84 kcal/mol) and
triplet (ca. 78 kcal/mol) excited states33 should be capable
of debrominating 3 (C-Br BDE ) ca. 68 kcal/mol for
isopropyl bromide).34 To determine whether it is the ketone
singlet or triplet state that is responsible for the debromina-
tion, 6 mM 3 was irradiated with 300 nm light in the presence
of 1.2 mM cis-piperylene. No quenching of debromination
to 8 was observed even though there was efficient (50-70%)
quenching of the ketone photoproducts 4-7. We thus
conclude that actiVation of the C17 C-Br bond in 3 occurs
through interaction with the C6 ketone singlet state.
The rate constant for this interaction can be estimated by
measuring the singlet lifetimes of 6-keto steroids with and
without a bromine at C17 (the TPSO group was removed
from C3 to avoid any possible interactions of the ketone
singlet with the aryl chromophore). Values of 2.5 ( 0.2 and
3.7 ( 0.2 ns were determined for the two compounds,
respectively. If one assumes that the shortened lifetime in
the presence of the C17 bromine is entirely due to the
interchromophoric interaction, one calculates a rate constant
for kC6fC17Br ) 1.3 × 108 s-1.
Acknowledgment. This work was supported by the
National Science Foundation through Grant CHE 9530650.
Supporting Information Available: Experimental pro-
cedures and characterization, including 1H and 13C NMR and
HRMS data for compounds 1-8 and 3ROH/6ketone; X-ray
crystal data for 3. This material is available free of charge
via the Internet at http://pubs.acs.org..
OL990319S
(33) Using the singlet and triplet energies for a cyclohexanone as taken
from Turro, N. J. Modern Molecular Photochemistry; Benjamin/Cum-
mings: Menlo Park, CA, 1978; p 290.
(35) Studies to determine the rate constant and overall efficiency for
triplet energy transfer from the C3 TPSO antenna to C17Br are in progress.
(34) Benson, S. W. Thermochemical Kinetics; Wiley: New York, 1968.
18
Org. Lett., Vol. 2, No. 1, 2000