Preferred formation of anti-housane (retention) in the sensitized photodenitrogenation of cyclopentane-annellated DBH-type azoalkanes through long-range steric effects in the cyclization of the triplet cyclopentane-1,3-diyl diradicals

Article abstract of DOI:10.1002/ejoc.200390098

The photolysis of the cyclopentene- and cyclopentane-annellated DBH-type azoalkanes 1a and 1b affords under singlet conditions (high-temperature direct photolysis) predominantly the inverted housanes syn-2 in similar amounts for both derivatives 2a and 2b

Full text of DOI:10.1002/ejoc.200390098

FULL PAPER
Preferred Formation of anti-Housane (Retention) in the Sensitized
Photodenitrogenation of Cyclopentane-Annellated DBH-Type Azoalkanes
through Long-Range Steric Effects in the Cyclization of the Triplet
Cyclopentane-1,3-diyl Diradicals
[a]
[a]
[a]
´
Waldemar Adam,* Manfred Diedering, and Vicente Martı
Keywords: DBH / Azoalkanes / Photolysis / Triplet diradical / Long-range steric effects / Diastereoselectivity
The photolysis of the cyclopentene- and cyclopentane-annel-
lated DBH-type azoalkanes 1a and 1b affords under singlet
pathway is mechanistically rationalized in terms of long-
range steric interactions between the annellated ring and the
conditions (high-temperature direct photolysis) predomin- gem-dimethyl-substituted methylene bridge during the cyc-
antly the inverted housanes syn-2 in similar amounts for both
lization of the cyclopentane-1,3-diyl triplet diradical 3DR
derivatives 2a and 2b. Under triplet conditions (low-temper- after ISC. In contrast, the denitrogenation of the intermediary
ature direct or benzophenone-sensitized photolysis), the
diazenyl diradical 1DZ along the SH2 pathway (inversion) for
photolysis leads to the retained housane anti-2 as the major
the singlet process is quite insensitive to these remote steric
diastereomer, but with a substantial difference in the syn/ effects.
anti-housane ratio for 2a (38:62) and for 2b (6:94). This signi-
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
ficant difference in the anti stereoselectivity of the triplet Germany, 2003)
Introduction
ored over the CϪN bond cleavage and the resulting triplet-
3
excited 1(n,π*) azoalkane loses N2 to produce the planar
Since its discovery in 1967,[1] the preferred formation of cyclopentane-1,3-diyl triplet diradical DR. The latter leads
inverted housane in the denitrogenation of exo-deuterated on ISC to a mixture of syn/anti-2a housanes; at high tem-
diazabicyclo[2.2.1]heptene ([D2]DBH) has been an issue of perature, the SH2 process prevails to afford the syn product
intensive mechanistic discussion.[2Ϫ6] Our recent work on through inversion.
the cyclopentene-annellated DBH derivative 1a has helped
3
to clarify the complexity of the double inversion process
(Scheme 1).[7] We found that in the direct photolysis at elev-
ated temperature (singlet pathway), the inverted housane
syn-2a is produced as the major diastereomer from the in-
termediary diazenyl diradical 1DZ through the SH2 pro-
1
cess.[7] In a competitive pathway, the DR intermediate is
1
formed by elimination of N2 from the DZ species and af-
fords both the inverted (syn-2a) and retained (anti-2a) hous-
anes. The product distribution in the direct photolysis is
strongly temperature-dependent: While at elevated temper-
ature the syn-2a housane is slightly favored (syn/anti ratio
62:38 at ϩ40 °C), the anti-2a housane dominates slightly at
subambient temperature (syn/anti ratio 39:61 at Ϫ75 °C).
This observation has been explained by intersystem crossing
Scheme 1
(ISC) of the n,π*-excited azoalkane, namely 11(n,π*) to
31(n,π*), which competes with the CϪN bond cleavage in
The slight preference of the anti-2a housane in the triplet
pathway has been attributed to a steric interaction between
the singlet pathway. Confirmation of the triplet pathway
was achieved by benzophenone-sensitized photolysis which
the annellated cyclopentene ring and the gem-dimethyl-sub-
afforded the same syn/anti-2a ratio as in the low-temper-
stituted methylene bridge during the ring closure of the
ature direct photolysis. Thus, at low temperature, ISC is fav-
planar nitrogen-free triplet diradical 3DR. Indeed, AM1
calculations had confirmed that the anti-2a diastereomer is
by about 6 kcal/mol of lower energy than the more strained
[a]
Institute of Organic Chemistry, University of Würzburg,
Am Hubland, 97074 Würzburg, Germany
592
 2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ193X/03/0203Ϫ0592 $ 20.00ϩ.50/0 Eur. J. Org. Chem. 2003, 592Ϫ596

Preferred Formation of anti-Housane
FULL PAPER
syn-housane.[8] Thus, only a small fraction (Ͻ 0.5 kcal/mol) (AM1 calculation indicates ca. 6 kcal/mol) had taken place
of this energy bias in favor of the anti-2a product is utilized even at ϩ40 °C.
in the transition state for cyclization, which is a con-
Expectedly, for both azoalkanes, similar trends with re-
sequence of the low (Ͻ 3 kcal/mol) energy barrier for the spect to the photolysis mode are displayed: Within the ex-
3DR Ǟ 2a process. It follows that an increase of the steric perimental error, the syn/anti-housane ratios are the same
interactions of the remote annellated ring and the gem-di- for the elevated-temperature (40 °C) direct (Entry 1) and
methyl-substituted methylene bridge should raise the energy the trans-piperylene-quenched photolysis (Entry 3), while
3
barrier of the DR Ǟ syn-2a cyclization and, consequently, the low-temperature (Ϫ75 °C) direct photolysis (Entry 2)
the formation of anti-housane should be enhanced. That gave the same ratios as the benzophenone-sensitized one
this is actually the case, we demonstrate herein for the pho- (Entry 4). For both azoalkanes, the syn-housane is slightly
todenitrogenation of the saturated azoalkane derivative 1b. preferred (2a: 62%; 2b: 53%; Entry 1) in the direct
Whereas in the direct photolysis (singlet-excited process), photolyses as well as in the quenched photolyses at 40 °C
the syn/anti-2b housane ratio is nominally affected by this (Entry 4). However, a substantial difference is observed be-
structural change, for the benzophenone-sensitized tween the product ratios for both azoalkanes in the low-
photolysis (triplet-excited process) essentially exclusively the temperature direct and sensitized photolysis, for which the
anti-2b housane is observed. This is the first example of syn/anti ratios are 39:61 for 2a and 6:94 for 2b (Entry 4).
such a high extent of retention for the cyclization of a nitro-
gen-free DR triplet diradical of the cyclopentane-1,3-diyl ation of the housane 2b were determined by following the
The activation parameters for the syn-to-anti isomeriz-
3
type.
decrease of syn-housane at five temperatures in the range
from ϩ70 to ϩ110 °C by 1H NMR spectrosopy to be Ea ϭ
26 Ϯ 2 kcal/mol and log A ϭ 12.1 Ϯ 0.9 sϪ1 (see Table 3).
The values for the housanes 2a were previously measured
Results
to be Ea ϭ 29 Ϯ 2 kcal/mol and log A ϭ 12.6 Ϯ 0.9 sϪ1 [7]
.
The known azoalkanes 1a,b were prepared according to
the Hünig route.[9] Upon irradiation with the 351-nm line
of an argon-ion laser, their denitrogenation led exclusively Discussion
to the syn and anti diastereomers of the housanes 2a,b. The
housanes anti-2a and syn-2a were obtained as reported,[7]
The azoalkanes 1a and 1b are very similar in structure,
the unknown housanes anti-2b and syn-2b were fully char- the only difference lies in the annellated ring, which is cyclo-
acterized by spectroscopic methods in analogy to 2a. The pentene in 1a but cyclopentane in 1b. Nevertheless, this ap-
quantum yield of denitrogenation of the azoalkane 1a was parently small structural variation is responsible for a sub-
determined to be Φrel(1a) ϭ 0.84 Ϯ 0.03 relative to 1b stantial difference in the syn/anti-housane product ratio of
[Φrel(1b) ϭ 1.0], see Exp. Sect. for details.
both azoalkanes, provided the photolysis conditions are ap-
Although the syn/anti ratios of the housane 2a in the de- propriate. In particular, under singlet conditions (Table 1),
nitrogenation of the azoalkane 1a had been determined be- i.e., elevated-temperature direct (Entry 1) or quenched
fore,[7] the photolyses were carried out in parallel with (Entry 3) photolysis, both azoalkanes 1a and 1b afford in
azoalkane 1b under identical conditions, to enable direct slight preference the thermodynamically less favored (sup-
comparison. The syn/anti product ratios for the direct, trip- ported by AM1 calculations and complete thermal syn-to-
let-sensitized (benzophenone), and quenched (trans- anti isomerization) inverted syn diastereomer (Entries 1 and
piperylene) photolyses are given in Table 1. The temper- 3). This is the expected behavior for the photochemical de-
ature was varied from Ϫ75 to ϩ40 °C for the three irradi- nitrogenation,[1a,1b,5d] which is taken as experimental evid-
ation modes (for additional data, see Table 2). Control ex- ence for the intervention of a singlet diazenyl diradical 1DZ
periments established that no thermal isomerization of the in the SH2 process that leads to the inverted syn-2 housane.
housane syn-2 to the thermodynamically favored anti-2 As for the effect of the annellated ring, namely cyclopentene
Table 1. Product studies of the photochemical denitrogenation of azoalkanes 1a and 1b (the corresponding irradiations of both azoalkanes
were run parallel under identical conditions)
Entry
Photolysis conditions[a]
Temperature [°C]
Product ratio[b]
syn/anti (2a)
syn/anti (2b)
1
2
3
4
direct
direct
quenched
sensitized
40
Ϫ75
40
62:38
38:62
65:35
39:61
53:47
6:94
54:46
6:94
40
[a]
In [D8]toluene; for the direct and trans-piperylene-quenched (1 ) photolyses, the 351-nm (0.8 W) line of the argon-ion laser was used,
[b]
1
for the benzophenone-sensitized (1 ) one, the 333-nm (0.6 W) line. Determined by H NMR analysis, normalized to 100%, error Ϯ5
of the stated values; hexamethyldisiloxane as internal standard; mass balance Ն 95%.
Eur. J. Org. Chem. 2003, 592Ϫ596
593

´
W. Adam, M. Diedering, V. Martı
FULL PAPER
in 1a or cyclopentane in 1b, backside displacement of N2 in
the respective singlet diazenyl diradicals 1DZ(1a) and
1DZ(1b) is essentially insensitive to such structural variation
during the inversion process.
In contrast, the photolytic deazetation of the azoalkanes
1 under triplet conditions, i.e., low-temperature direct or
benzophenone-sensitized photodenitrogenation, leads to a
small preference of the anti-2a housane from the unsatur-
ated azoalkane 1a (Entries 2 and 4), while the retention
product anti-2b is formed almost exclusively for the satur-
ated derivative 1b under the same conditions. The fact that
the anti-housane (retention) is preferred by far is unusual,
since normally the triplet denitrogenation of simple DBH
derivatives has furnished about equal amounts of retained
(ret) and inverted (inv) housanes, as shown for a few cases
below (Figure 1).[5d] The data show that the ret/inv ratio of
housanes is ca. 50:50, but certainly no preference for reten-
tion is expressed. This lack of stereoselectivity of the triplet
denitrogenation has been rationalized in terms of a planar
Scheme 2
between the gem-dimethyl-substituted methylene bridge and
the annellated ring manifests itself not only in the syn/anti
diastereoselectivity of the triplet photolysis, but also in the
activation barrier for the thermal syn-to-anti isomerization
of the diastereomeric housanes; it is by ca. 3 kcal/mol
smaller for the saturated housane 2b (Figure 2).
3
nitrogen-free triplet diradical DR (Scheme 1). The low en-
ergy barrier (Ͻ 3 kcal/mol) for cyclization by intersystem
crossing accounts for the small if not negligible substituent
effects.[2,10] Thus, the fact that the triplet denitrogenation of
the azoalkane 1b affords almost exclusively (inv/ret ratio of
6:94) the retained housane anti-2b is an exception.
Figure 1. Diastereoselectivity (ret/inv ratio) for the triplet denitro-
Figure 2. Energy profiles [kcal/mol] for the syn/anti-housane forma-
tion in the thermal syn-to-anti isomerisation of the housanes 2a,b
genation of simple DBH derivatives
1
through the singlet diradical DR (solid curve) and in the triplet-
sensitized and direct (Ϫ75 °C) photochemical denitrogenation of
the azoalkanes 1a,b after isc of the triplet diradical 3DR (dashed
curve)
What is the reason for this anti stereoselectivity in the
triplet pathway? As mentioned before, the significant differ-
ence between the azoalkane 1a and 1b lies in the annellated
ring. This structural variation becomes important during
the cyclization of the nitrogen-free planar triplet diradical
3DR (Scheme 2).[10,11] After intersystem crossing (isc) to the
Conclusion
1
singlet diradical DR, the direction of the ring closure to
the diastereomeric housanes 2 is controlled by steric effects
In summary, in this study we have shown the influence
between the gem-dimethyl-substituted methylene bridge of long-range steric interactions on the syn/anti-housane
with the annellated ring. This steric interaction is stronger distribution in the photolyses of the DBH-type azoalkanes
for the bulkier annellated cyclopentane ring during the 1a and 1b. This steric effect derives from the structural dif-
puckering motion of the ring closure in the resulting ference between cyclopentene and cyclopentane annell-
1DR(1b) diradical such that the anti-2b (path a) rather than ation. While the ring closure of the diazenyl diradical DZ
1
syn-2b (path b) is generated in high preference (Scheme 2). in the SH2 process (inversion) is relatively insensitive to such
1
For the annellated cyclopentene ring in the DR(1a) dirad- remote steric interaction for the direct photolysis (singlet
ical, the puckering motion is less obstructed during the ring pathway), it promotes cyclization of the planar nitrogen-
closure to the housane syn-2a by such steric effects and a free 3DR diradical preferably to the anti-housane (retention)
relatively large amount of inverted product is obtained (syn/ in the triplet process. Despite the high degree of retention
anti ratio 39:61).
in this triplet photolysis, this does not constitute a case of
From the syn/anti-housane ratios of the triplet pathway ‘‘stereochemical memory’’, as demonstrated for the direct
we may calculate by means of the Arrhenius equation the photolysis (singlet) of related bridgehead-substituted cyclo-
difference in the energy barriers for the ring closure of the pentene-annellated DBH-type azoalkanes,[12] because in the
1DZ diradical to the two housane diastereomers. The ∆Ea triplet process the planar nitrogen-free triplet diradical 3DR
1
1
value is 0.24 Ϯ 0.04 kcal/mol for DR(1a) and for DR(1b) intervenes, which has lost all of the original stereochemical
it is 1.4 Ϯ 0.3 (Figure 2). The difference in the steric strain information. The high anti diastereoselectivity of the pre-
594
Eur. J. Org. Chem. 2003, 592Ϫ596

Preferred Formation of anti-Housane
FULL PAPER
exo-3,3-Dimethyltricyclo[3.3.0.0.2,4]octane (syn-2b): 1H NMR
(200 MHz, [D8]toluene): δ ϭ 0.73 (s, 3 H, CH3), 1.48 (s, 3 H, CH3),
2.66Ϫ2.80 (m, 2 H) ppm; the remaining peaks overlap with signals
of the anti diastereomer. 13C NMR (50 MHz, [D8]toluene): δ ϭ
19.3 (q), 24.1 (s), 27.9 (t), 28.6 (d), 32.0 (t), 37.3 (d) ppm; the re-
maining peaks overlap with signals of the anti diastereomer.
sent case is the consequence of remote steric interactions in
the transition state for the triplet cyclization 3DR Ǟ syn-2b.
Experimental Section
General Aspects: NMR spectra were measured with a Bruker
AC200 or AC250 in [D8]toluene with hexamethyldisiloaxane as in-
ternal standard. UV absorption spectra were recorded with a Hita-
chi U 3200 spectrophotometer. Elemental analyses were performed
by the Microanalytical Division of the Institute of Inorganic Chem-
istry (University of Würzburg). Photolyses were carried out at the
333-nm and 351-nm laser lines of a continuous-wave argon-ion
laser (INNOVA 100, Coherent Company).
Diastereomeric mixture of syn/anti-2b: IR (neat): ν˜ ϭ 3012, 2945,
2869, 1454, 1376, 1255, 1117, 1020, 884, 811 cmϪ1. C10H16 (136.2):
calcd. C 88.16, H 11.84; found C 87.80, H 12.09.
Product Studies of the Azoalkanes 1: For the direct photolyses, a
sample of the azoalkane 1 (1a: ca. 0.074 mmol; 1b: ca. 0.117 mmol;
about the same optical density) in [D8]toluene (0.6 mL) and of
hexamethyldisiloxane (2 µL) was transferred to an NMR tube, de-
areated by purging with a slow stream of argon for 10 min, and
irradiated at the 351-nm laser line (0.8 W; widened to ca. 5 cm by
a lens) of the argon-ion laser at the temperature specified in
Table 2. For the quenching experiments, trans-piperylene (1 ) was
employed as triplet quencher. In the triplet-sensitized photolyses,
benzophenone (1 ) was used as sensitizer and irradiated at the
333-nm laser line (0.6 W). The results are listed in Table 2.
Synthesis of Housane 2b:
A
sample of (1α,4α,4aα,7aα)-
4,4a,5,6,7,7a-hexahydro-8,8-dimethyl-1,4-methano-1H-cyclopenta-
[d]pyridazin (1b) (50.0 mg, 0.304 mmol) was dissolved in pentane
(ca. 0.7 mL), transferred to an NMR tube, deareated by purging
with a slow stream of argon for 10 min, and irradiated at the 333-
nm, 351-nm and 364-nm lines of the argon-ion laser at 20 °C for
ca. 10 min. The pale yellow photolysate was passed through a short
column of basic alumina (ca. 2.0 g), the solvent evaporated (20 °C,
20 mbar) and the diastereomeric mixture of housanes 2b was ob-
tained in quantitative yield as colorless oils, which could not be
separated by column chromatography. Thermolysis at elevated tem-
perature (Ͼ 90 °C) led to the pure anti-2b diastereomer. The hous-
ane syn-2b was spectroscopically characterized directly in the mix-
ture of diastereomers.
Synthesis and Photolysis of the Azoalkanes 1c and 1d: The saturated
azoalkanes 1c and 1d, which are Me- and Ph-substituted on the
bridgehead positions were prepared as described in the literat-
ure.[9,13] Their photolysis afforded under all conditions exclusively
the known housanes anti-2c,d.[14]
endo-3,3-Dimethyltricyclo[3.3.0.0.2,4]octane (anti-2b): 1H NMR
(250 MHz, [D8]toluene): δ ϭ 0.75 (s, 2 H), 0.87 (s, 3 H, CH3), 1.19
(s, 3 H, CH3), 1.30Ϫ1.80 (m, 5H), 1.94Ϫ2.22 (m, 3 H) ppm. 13C
NMR (63 MHz, [D8]toluene): δ ϭ 14.3 (q), 21.2 (s), 24.6 (q), 25.7
(t), 29.7 (d), 31.7 (t), 38.8 (d) ppm.
Determination of Relative Photolysis Quantum Yields of the Azoalk-
anes 1a,b: The relative quantum yields have been determined by a
method which has been used for similar azoalkane derivatives.[15]
Aliquots (ca. 2.8 mL) of degassed toluene solutions of azoalkanes
1a and 1b (absorbance ca. 0.60 at λ ϭ 351 nm) were placed into
UV cuvettes (1 ϫ 1 cm) and irradiated at equal time intervals of
ca. 1 s with the 351-nm widened beam of the argon-ion laser. The
laser light intensity was adjusted by using the intensity regulation
(Ϯ 0.05%), supplied by the instrument. The irradiation was con-
ducted up to 7Ϫ11 data points per experiment. The decreasing ab-
sorbance (A) versus irradiation time was monitored by UV spectro-
Table 2. Product studies of the photochemical denitrogenation of
azoalkanes 1a and 1b (the corresponding irradiations of both
azoalkanes were run in parallel under identical conditions)
photometry. The plots of log [(10A Ϫ 1)/(10A Ϫ 1)] versus irradi-
0
ation time were linear with correlation coefficients (R2) greater than
0.98. The relative photoreaction quantum yields (Φrel) were calcu-
lated from the ratio of the slopes S(1a)/S(1b) ϭ ε1aΦ1a/(ε1bΦ1b),
where ε1a and ε1b are the extinction coefficients of the azoalkanes
1a and 1b at the irradiation wavelength of 351 nm (ε1a ϭ 174 Ϫ1
cmϪ1; ε1b ϭ 110 Ϫ1 cmϪ1). The relative quantum yields have been
determined to be Φrel(1a) ϭ 0.84 Ϯ 0.03·Φrel(1b).
Determination of the Activation Parameters for the Thermal syn-to-
anti Isomerization of the Housane 2b: The activation energies and
the log A values for the syn-to-anti isomerization of the housanes
2 were obtained from the first-order kinetics of the thermolysis of
the syn diastereomer (a syn/anti diastereomeric mixture of the hou-
sane was used) to the persistent anti-housanes (Table 3). The time
profile of the amount of syn- and anti-housanes was determined by
1H NMR spectroscopy as a function of temperature. From this
data, the rate constants of isomerzation were evaluated according
to first-order kinetics and the activation parameters calculated by
means of the Arrhenius equation.
[a]
In [D8]toluene; for the direct and trans-piperylene-quenched (1
) photolyses, the 351-nm (0.8 W) line of the argon-ion laser was
used, for the benzophenone-sensitized (1 ), the 333-nm (0.6 W)
line.
mass balance Ն 95%; hexamethyldisiloxane as internal standard;
error Ϯ5% of the stated values.
Determined by 1H NMR analysis; normalized to 100%;
[b]
Eur. J. Org. Chem. 2003, 592Ϫ596
595

´
W. Adam, M. Diedering, V. Martı
FULL PAPER
[3]
[4]
D. C. Sorescu, D. L. Thompson, L. M. Raff, J. Chem. Phys.
1995, 102, 7910Ϫ7924.
Table 3. Activation parameters for the syn-to-anti isomerization of
the housane 2a (the amount of syn- and anti-housanes was deter-
mined by 1H NMR spectroscopy with naphthalene as internal
standard; error Ϯ5% of the stated values)
N. Yamamoto, M. Olivucci, P. Celani, F. Bernardi, M. A.
Robb, J. Am. Chem. Soc. 1998, 120, 2391Ϫ2407.
[5] [5a]
W. Adam, T. Oppenländer, G. Zang, J. Org. Chem. 1985,
50, 3303Ϫ3312. [5b] J. S. Adams, R. B. Weisman, P. S. Engel, J.
Am. Chem. Soc. 1990, 112, 9115Ϫ9121. [5c] C. J. S. M. Simpson,
G. J. Wilson, W. Adam, J. Am. Chem. Soc. 1991, 113,
4728Ϫ4732. [5d] W. Adam, U. Denninger, R. Finzel, F. Kita, H.
Platsch, H. Walter, G. Zang, J. Am. Chem. Soc. 1992, 114,
5027Ϫ5035.
Entry
Temperature
[°C]
k·105 [sϪ1
]
Ea
log A
[sϪ1
[kcal molϪ1
]
]
1
2
3
4
5
70
80
90
100
110
2.8
5.3
21.4
55.3
129.4
[6]
[7]
[8]
[9]
M. B. Reyes, B. K. Carpenter, J. Am. Chem. Soc. 2000, 122,
10163Ϫ10176.
W. Adam, H. Garcıa, V. Martı, J. N. Moorthy, J. Am. Chem.
Soc. 1999, 121, 9475Ϫ9476.
W. Adam, H. M. Harrer, W. M. Nau, K. Peters, J. Org. Chem.
1994, 59, 3786Ϫ3797.
26 Ϯ 2
12.1 Ϯ 0.9
´
´
K. Beck, A. Höhn, S. Hünig, F. Prokschy, Chem. Ber. 1984,
117, 517Ϫ533.
[10] [10a]
S. L. Buchwalter, G. L. Closs, J. Am. Chem. Soc. 1975, 97,
[10b]
3857Ϫ3858.
S. L. Buchwalter, G. L. Closs, J. Am. Chem.
Acknowledgments
Soc. 1979, 101, 4688Ϫ4694.
[11]
[12]
The generous financial support from the Deutsche Forschungs-
gemeinschaft, the Volkswagen Stiftung and the Fonds der
Chemischen Industrie gratefully appreciated. V. M. is grateful for
a Marie-Curie fellowship from the European Commission.
W. Adam, H. M. Harrer, F. Kita, H.-G. Korth, W. M. Nau, J.
Org. Chem. 1997, 62, 1419Ϫ1426.
´
´
W. Adam, H. Garcıa, M. Diedering, V. Martı, M. Olivucci, E.
Palomares, J. Am. Chem. Soc. 2002, 124, 12192Ϫ12199.
[13] [13a]
[13b]
K. Beck, S. Hünig, Chem. Ber. 1987, 120, 477Ϫ483.
K. Beck, S. Hünig, F.-G. Klärner, P. Kraft, U. Artschwager-
Perl, Chem. Ber. 1987, 120, 2041Ϫ2151.
W. Adam, T. Heidenfelder, C. Sahin, Synthesis 1995,
1163Ϫ1170.
W. Adam, G. Fragale, D. Klapstein, W. N. Nau, J. Wirz, J. Am.
Chem. Soc. 1995, 117, 12578Ϫ12592.
Received August 29, 2002
[1] [1a]
[14]
[15]
W. R. Roth, M. Martin, Justus Liebigs Ann. Chem. 1967,
[1b]
702, 1Ϫ7.
47, 4695Ϫ4698.
W. R. Roth, M. Martin, Tetrahedron Lett. 1967,
[1c]
E. L. Allred, R. L. Smith, J. Am. Chem.
Soc. 1969, 91, 6766Ϫ6775.
C. D. Sherill, E. T. Seidl, H. F. Schaefer III, J. Phys. Chem.
1992, 96, 3712Ϫ3716.
[2]
[I02484]
596
Eur. J. Org. Chem. 2003, 592Ϫ596