giving rise to v-dienyl aldehydes bearing no a-substituents,
provided aldol condensation products 2l9 and 2m9,9 respectively,
as the major products. Cyclohexanol derivative 1j was exception-
ally robust and was recovered unchanged even after long heating
at 50 uC (entry 10).9 Cyclopentanol derivative 1i showed a
somewhat low yield (entry 9). In sharp contrast to these, bicyclic
diols 1d–h and 1k containing cyclohexanol and/or cyclopentanol
structural motifs smoothly underwent fragmentation and provided
the expected v-dienyl aldehydes 2 in good to excellent yields
(entries 4–8 and 11). Cyclobutanol derivatives 1a–c reacted
similarly well. These results suggest that ring strain or torsional
strain in the cycloalkanols is a key factor to promote the
dehydration fragmentation successfully. It should be also noted
that the present method is applicable to the synthesis of v-dienyl
ketones 2h (entry 8).
array of cycloalkanols, ranging from cyclobutanol to cyclodecanol
with the exception of cyclohexanol, under essentially neutral
conditions.16
Notes and references
§ General procedure (see ESI for details{): into a flask containing
Pd(PPh3)4 (29 mg, 0.025 mmol) purged with N2 were added dry toluene
(2.5 mL), syn-1k (98.1 mg, 0.5 mmol), and 9-PhBBN (0.8 mL, 0.3 M
solution, 0.25 mmol)15 via syringe at rt. The solution was stirred at 50 uC
for 24 h under N2. After usual work-up and purification, 2k was isolated in
92% yield (80.2 mg).
1 (a) Y. Sato, N. Saito and M. Mori, Tetrahedron, 1998, 54, 1153; (b)
N. Waizumi, T. Itoh and T. Fukuyama, Tetrahedron Lett., 1998, 39,
6015; (c) Y. Sato, M. Takimoto and M. Mori, Synlett, 1997, 734.
2 (a) C.-M. Yu, J. Youn, S.-K. Yoon and Y.-T. Hong, Org. Lett., 2005, 7,
4507; (b) Y. Sato, N. Saito and M. Mori, J. Org. Chem., 2002, 67, 9310;
(c) K. Shibata, M. Kimura, M. Shimizu and Y. Tamaru, Org. Lett.,
2001, 3, 2181.
3 H. Harayama, T. Kuroki, M. Kimura, S. Tanaka and Y. Tamaru,
Angew. Chem., Int. Ed. Engl., 1997, 36, 2352.
4 M. Mori, M. Kimura, Y. Takahashi and Y. Tamaru, Chem. Commun.,
2006, 4303.
5 (a) M. Kimura, M. Fukasaka and Y. Tamaru, Heterocycles, 2006, 67,
535; (b) M. Kimura, M. Futamata, R. Mukai and Y. Tamaru, J. Am.
Chem. Soc., 2005, 127, 4592. For a review, see: (c) Y. Tamaru, Eur. J.
Org. Chem., 2005, 2647.
6 Review on the Grob fragmentation: T. Constantieux and J. Rodriguez,
Sci. Synth., 2004, 26, 413.
7 Et3B is usually stable toward hydrolysis with alcohols. Hence, the
hydrolysis of Et3B with 1n might be attributed to chelation coordination
of 1n (and 1e, see ref. 8) to boron. However, the hydrolysis by a 1,3-diol
seems to be the subject of some subtle stereoelectronic effects of diols,
since Et3B withstands hydrolysis with some 1,3-diols, see: R. Mukai,
Y. Horino, S. Tanaka, Y. Tamaru and M. Kimura, J. Am. Chem. Soc.,
2004, 126, 11138.
8 The reaction features described here seem to be general for other diols.
For example, similar results were obtained for the reactions of 1e with
Et3B, Ph3B, and (C6F5)3B.
9 Isolated yields observed for the Pd-catalyzed decarboxylation:3 2j (42%),
2m (54%), 2n (85%). Isolated yields observed for the Ni-catalyzed
decarboxylation:4 2j (84%), 2m (94%), 2n (85%).
10 Use of both Pd(PPh3)4 and 9-PhBBN is essential to promote the
dehydrative fragmentation. In the absence of either of them, neither 2n
nor 2n9 was formed at all.
11 X-Ray crystallographic data of anti-1e and syn-1k (relative stereo-
chemistry) were obtained. CCDC 639709 and 631058, respectively. For
crystallographic data in CIF or other electronic format, see DOI:
10.1039/b708526e.
The most plausible mechanism is outlined in Scheme 2 using 1a
as a representative of the diols. Oxidative addition of Pd(0) to the
allylic C–OH bond of anti-1a, activated by the coordination of
9-PhBBN, with inversion of configuration would provide a cis-
oxapalladacyclopentane intermediate cis-I, being cis with respect to
the C2- and C3-substituents, as a primary intermediate, which
would lead to (Z)-2a on fragmentation. However, the selective
formation of (E)-2a over (Z)-2a (entry 1, Table 2) suggests that cis-
I would rather isomerize to a sterically less congested, more stable
trans-I via a s–p–s isomerization mechanism than undergo
fragmentation into (Z)-2a. Fragmentation through trans-I leads
to (E)-2a. For the fragmentation of syn-1a, it is sterically
impossible for the allylpalladium intermediate to form a cyclic
structure like I owing to the severe strain imposed on a trans-fused
bicyclo[3.2.0]heptane skeleton, and hence it might undergo
fragmentation through an open-chain intermediate II, with an
anti conformation regarding C1–C2 and C3–Pd bonds, and
furnish (E)-2a selectively.
The difference in reaction features between the decarboxylation
and dehydration methods, e.g., 3j A 2j (42%)3,9 under Pd(0)
catalysis and 3j A 2j (84%)4,9 under Ni(0) catalysis, while 1j A 2j
(0%, entry 10, Table 2), may be primarily attributed to the
structural differences in the intermediates I. An anionic charge
developed on the oxygen in I (O2 instead of OH) during
decarboxylation might weaken the C1–C2 bond12 and hence
facilitate the fragmentation. On the other hand, the oxygen in I
generated through dehydration may be mostly neutral, and hence
considerable amount of ring strain may be required to weaken the
C1–C2 bond.
12 Review on oxy-Cope rearrangement: T. Constantieux and J. Rodriguez,
Sci. Synth., 2004, 26, 413.
13 For the BF3?OEt2 catalyzed synthesis of v-dienyl aldehydes from
acetonides of 1,3-anti-diols of close structural similarity to 1k, see:
´
J. Barluenga, M. Alvarez-Pe´rez, K. Wuerth, F. Rodriguez and F. J.
Fan˜ana´s, Org. Lett., 2003, 5, 905.
The Grob-type fragmentation of 1,3-diols (e.g., 1 A 2) is among
the very powerful tools available for the construction of desired
molecules.6 However, harsh reaction conditions, either strongly
basic or acidic and/or high reaction temperatures, have limited its
wide use.13 Transition metal catalysis has been so far only effective
for the ring opening reaction of some strained cyclopropanol and
cyclobutanol derivatives.14 In this context, it should be noted that
the present palladium-catalyzed reaction is the first example, to the
best of our knowledge, that demonstrates the ring-opening of an
14 (a) M. Murakami, S. Ashida and T. Matsuda, J. Am. Chem. Soc.,
2006, 128, 2166. For a review on palladium-catalyzed C–C bond
cleavage reaction, see: (b) T. Nishimura and S. Uemura, Synlett, 2004,
201.
15 G. W. Kramer and H. C. Brown, J. Organomet. Chem., 1974, 73, 1.
16 We thank financial support from the Ministry of Education, Culture,
Sports, Science and Technology, Japanese Government [Grant-in-Aid
for Scientific Research (B) 19350050 and Priority Areas 17035065 and
18037059].
4506 | Chem. Commun., 2007, 4504–4506
This journal is ß The Royal Society of Chemistry 2007