Palladium(II)-catalyzed ring enlargement of 2-(arylmethylene) cyclopropylcarbinols: Strong effect of substituent electronic nature on the reaction pathway

Article abstract of DOI:10.1039/b804932g

In the presence of Pd(II) catalyst and copper(II) bromide, 2-(arylmethylene)cyclopropylcarbinols undergo ring enlargement to deliver (arylcyclobutenyl)carbinols or hydrogenated furans in good yields under mild conditions; mechanisms accounting for the dis

Full text of DOI:10.1039/b804932g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Palladium(II)-catalyzed ring enlargement of
2-(arylmethylene)cyclopropylcarbinols: strong effect of
substituent electronic nature on the reaction pathwayw
Guo-Qiang Tian,a Zhi-Liang Yuan,b Zhi-Bin Zhub and Min Shi*ab
Received (in College Park, MD, USA) 25th March 2008, Accepted 25th April 2008
First published as an Advance Article on the web 8th May 2008
DOI: 10.1039/b804932g
starting material disappeared rapidly and complex product
In the presence of Pd(II) catalyst and copper(II) bromide,
2-(arylmethylene)cyclopropylcarbinols undergo ring enlarge-
ment to deliver (arylcyclobutenyl)carbinols or hydrogenated
furans in good yields under mild conditions; mechanisms ac-
counting for the distinct products have been proposed on the
basis of control and deuterium labeling experiments.
mixtures were obtained within 0.2 h (Table S1,w entry 1).
When the temperature was decreased to 0 1C, to our delight,
we found that the ring expanded products 2a and 3a were
obtained in good total yield (75%) along with moderate
selectivity, with a ratio 5 : 1, within 0.5 h (entry 2). The
structure of 2a has been further confirmed by X-ray diffraction
(ESIw).8 Compared with the method described in our previous
paper,4 this procedure needs a much lower activation energy,
indicating that the hydroxymethyl group does play an impor-
tant role in accelerating the reaction rate. Following that, we
next screened the reaction conditions in detail. When the
reaction was carried out in 1,2-dichloroethane (DCE) a similar
result was obtained, although toluene, THF, diethyl ether and
acetonitrile were not suitable solvents for this reaction (entries
3–7). Pd(OAc)2 or CuBr2 alone did not promote the transfor-
mation (entries 8 and 9). Combination of Pd(OAc)2 with other
copper salts such as Cu(OTf)2, CuCl2, Cu(OAc)2 or CuBr did
not give an effective catalytic system (entries 10, 12–14). If the
catalytic system contained K2CO3, no reaction occurred (entry
11). On the other hand, on the examination of some other
metal bromides such as NiBr2, NaBr, MgBr2, ZnBr2, and
LiBr, only NiBr2 gave a satisfactory result (entries 15–20).
These results suggest that PdBr2 might be the real active
catalyst in this transformation. Actually, the same products
were produced when either PdBr2 or PdCl2 was loaded as the
sole catalyst, although in low yields (entries 21 and 22).
Continuous experiments showed that the combination of
PdCl2, CuBr2 (0.03 or 0.06 mmol) and NaHCO3 could cata-
lyze this transformation smoothly to afford cyclobutenylcar-
binols 2a and 3a in good total yields along with high
selectivities, with the ratio 12 : 1, within 0.5 h (entries 24
and 25).
Methylenecyclopropanes (MCPs) are generally used as build-
ing blocks in organic synthesis for their ready accessibility as
well as diverse reactivity driven by the relief of ring strain.
Over the past decades, a series of review articles have been
published on the transformations of these exo-methylene
three-membered carbocycles.1 Transition metal catalysts,
now one of the most powerful tools for synthetic chemists,
have played an increasingly important role in these transfor-
mations.2,3 Recently, we showed that, catalyzed by palladium
acetate and metal bromide, MCPs could undergo ring enlarge-
ment to provide the cyclobutene motif efficiently.4 Almost at
the same time, Furstner and Aıssa reported a similar trans-
¨
¨
formation promoted by platinum.5
2-(Arylmethylene)cyclopropylcarbinols 1 are another kind
of MCP bearing an additional hydroxymethyl group6 and, as
demonstrated by our group, can undergo a variety of trans-
formations triggered by the nucleophilic hydroxyl group under
milder conditions.7 In view of this, we hypothesized that the
tethered hydroxyl group might accelerate transition metal-
assisted reactions or trap the reaction intermediates of MCPs.
Herein, we wish to report the palladium(II) chloride-catalyzed
ring enlargement of 2-(arylmethylene)cyclopropylcarbinols in
the presence of copper(II) bromide to furnish (arylcyclobute-
nyl)carbinols or hydrogenated furans in good yields depending
on the substituents on the benzene ring.
The initial examination was performed using (E)-[2-(4-
methoxybenzylidene)cyclopropyl]methanol 1a as the substrate
upon treatment with Pd(OAc)2 and CuBr2 at room tempera-
ture (20 1C) in dichloromethane (DCM). We found that the
With the optimal conditions identified, we next examined
the scope of the transformation. Substrates 1, bearing elec-
tron-donating groups on the benzene ring, underwent ring
expanding reactions smoothly to provide the corresponding
(arylcyclobutenyl)carbinols 2b–2f and 3c–3e in good total
yields and good-to-excellent selectivities within 1 h (Table 1,
a State Key Laboratory of Organometallic Chemistry, Shanghai
Institute of Organic Chemistry, Chinese Academy of Sciences, 354
Fenglin Road, Shanghai, China 200032. E-mail:
entries
1–5).
(Z)-2-(4-Methoxybenzylidene)cyclopropyl]-
Mshi@mail.sioc.ac.cn
b School of Chemistry and Molecular Engineering, East China
University of Science and Technology, 130 Meilong Road, Shanghai,
China 200237
methanol (Z)-1a delivered the same products as its (E)-isomer
in slightly lower yield and selectivity under the standard
conditions (entry 6). However, when these reaction conditions
were applied to (E)-(2-benzylidenecyclopropyl)methanol 1g
which does not have substituent on the benzene ring or
substrates 1h and 1i bearing electron-withdrawing groups on
w Electronic supplementary information (ESI) available: Experimental
procedures and spectroscopic data for all compounds. CCDC 665100
and 674227. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/b804932g
ꢀc
This journal is The Royal Society of Chemistry 2008
2668 | Chem. Commun., 2008, 2668–2670

View Article Online
Table 1 Formation of (arylcyclobutenyl)carbinols from 2-(aryl-
methylene)cyclopropylcarbinols
Scheme 2 Deuterium labeling experiment.
with an electron-withdrawing group on the benzene ring in the
presence of PdCl2 (5 mol%) and CuBr2 (20 mol%). However,
when these reactions were carried out under an air atmosphere
in the presence of PdCl2 (5 mol%) and 2.5 equiv. of CuBr2,
products 4 and anti-59 were the dominant products (Table 2,
entries 2–5). As for (E)-(2-benzylidenecyclopropyl)methanol,
1g, which does not have a substituent on the benzene ring, also
cleanly underwent this type of transformation to provide 4g
and anti-5g in good total yield (entry 1). (Z)-[2-(4-Fluoroben-
zylidene)cyclopropyl]methanol (Z)-1k gave the same products
as its (E)-isomer, but in a lower yield (Table 2, entry 6). It
should be noted that 47% of anti-5h could be converted to 4hw
if it was stored at 0 1C for 6 days, indicating that compounds 4
are more thermodynamically stable. A control experiment was
carried out to evaluate the SN20 reaction pathway of anti-5h in
the presence of NaBr (1.0 equiv.) under the standard condi-
tions (Scheme 3). It was found that 4h was formed in 16%
yield after 24 h, suggesting that this SN20 reaction was
relatively sluggish and compounds 4 and anti-5 were formed
through two different reaction pathways.
Entrya
R1/R2
Time/h
Yieldb (%) (2 : 3)
1c
2
3-BnOC6H4/H, 1b
3,4,5-(MeO)3C6H2/H, 1c
2,5-(MeO)2C6H3/H, 1d
4-MeC6H4/H, 1e
1
0.5
0.5
1
1
0.5
2
75 (2b)
89 (13 : 1) (2c : 3c)
70 (22 : 1) (2d : 3d)
70 (17 : 1) (2e : 3e)
68 (2f)
83 (5 : 1) (2a : 3a)
88 (21 : 1) (2g : 3g)
61 (2h)
3
4
5c
6
3-MeC6H4/H, 1f
H/4-(MeO)C6H4, (Z)-1a
C6H5/H, 1g
4-BrC6H4/H, 1h
7d
8c,d
9c,d
a
6
10
4-ClC6H4/H, 1i
23 (2i)
All reactions were carried out with 1 (0.3 mmol), PdCl2 (0.015
mmol), CuBr2 (0.06 mmol), NaHCO3 (0.15 mmol) in 1.0 mL of
b
dichloromethane under an argon atmosphere. Isolated yields.
c
d
A trace of 3 was observed. 2 mg of MS 4 A were added.
the benzene ring, the reactions were sluggish to afford the
corresponding products in sharply lower yields, even upon
prolonged reaction time, and most of the starting materials 1
were recovered. After a series of trials and errors, we found
that adding molecular sieves (4 A, 2 mg for 0.3 mmol of 1)
improved the situation to some extent, although a large
amount of molecular sieves would suppress the reaction. As
a result, the corresponding (arylcyclobutenyl)carbinols 2g–2i
and 3g were obtained in 23–88% total yields (entries 7–9).
To clarify the role of the hydroxymethyl group in this
transformation, a control experiment using dehydroxylated
substrate 1j as the substrate was carried out under the stan-
dard conditions, and we found that the corresponding ring
expanded products 2j and 3j were obtained in lower yield and
selectivity at elevated temperature (20 1C) even after a longer
reaction time (Scheme 1). This result clearly suggested that the
hydroxyl group plays an important role in this process. To
gain more mechanistic insight into the effects of the hydroxyl
group, we carried out deuterium labeling experiments as
shown in Scheme 2. The deuterium atom on the hydroxyl
group did not shift into the products 2 and 3 and the results
are very similar as those shown in entry 25 of Table S1.w When
the hydroxyl group was protected by a benzyl group, a similar
result was obtained after a longer reaction time (see the ESIw).
However, when the hydroxyl group was replaced by an amino
group, no reaction occurred, indicating that just the hydroxyl
group would work by accelerating the reaction rate via co-
ordination to the palladium center.
Plausible reaction mechanisms have been proposed for these
two distinct transformations (Scheme 4). In process I, the
hydroxy group in compound 1 first coordinates to PdX2 (X =
Cl or Br) to form a chelated intermediate A, which then
produces a zwitterionic intermediate B (this process might be
the rate-determining step of the transformation). This zwitter-
ionic intermediate is stabilized by the cyclopropylmethyl
group10 and reinforced by the electron-donating group sub-
stituted benzene ring. Rearrangement of this ‘‘non-classical’’
zwitterionic species produces carbene intermediates C and D
via path a and path b, respectively. The final products of
(arylcyclobutenyl)carbinols 2 and 3 are furnished by a follow-
ing 1,2-hydrogen shift.4,5 In process II, a partially charged
zwitterionic intermediate E11 rather than the totally changed
zwitterionic intermediate B might exist, which presumably
Table 2 Formation of dihydrofurans and tetrahydrofurans
Entrya
R1/R2
Time/h
Yieldb (%) (4 : 5)
As noted above, products 2 and 3 were formed in lower
yields and most of the starting materials 1 were recovered in
the absence of 4 A molecular sieves when using substrates 1
1
2
3
4
5
6
a
C6H5/H, 1g
12
12
12
12
12
32
85 (3 : 1) (4g : 5g)
81 (3 : 1) (4h : 5h)
82 (3 : 1) (4i : 5i)
97 (3 : 1) (4k : 5k)
70 (3 : 1) (4l : 5l)
27 (3 : 1) (4k : 5k)
4-BrC6H4/H, 1h
4-ClC6H4/H, 1i
4-FC6H4/H, 1k
2,3-Cl2C6H3/H, 1l
H/4-FC6H4, (Z)-1k
All reactions were carried out with 1 (0.3 mmol), PdCl2 (0.015
mmol), CuBr2 (0.75 mmol), NaHCO3 (0.40 mmol) in 1 mL of
b
dichloromethane under an air atmosphere. Isolated yields.
Scheme 1 Effects of the hydroxyl group.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 2668–2670 | 2669

View Article Online
proposed for these two distinct reaction processes. We think
that carbene species might be involved in the formation of
(arylcyclobutenyl)carbinols and a p-allylpalladium intermedi-
ate would account for the production of the dihydrofuran and
tetrahydrofuran products.
Scheme 3 Control experiment.
We thank the Shanghai Municipal Committee of Science
and Technology (04JC14083, 06XD14005) and the National
Natural Science Foundation of China for financial support
(20472096, 20672127, and 20732008).
Notes and references
1
For reviews, see: (a) P. Binger and H. M. Buch, Top. Curr. Chem.,
¨
1987, 135, 77; (b) A. Goti, F. M. Cordero and A. Brandi, Top.
Curr. Chem., 1996, 178, 1; (c) A. Brandi, S. Cicchi, F. M. Cordero
and A. Goti, Chem. Rev., 2003, 103, 1213; (d) E. Nakamura and S.
Yamago, Acc. Chem. Res., 2002, 35, 867.
2 For reviews, see: (a) I. Nakamura and Y. Yamamoto, Adv. Synth.
Catal., 2002, 344, 111; (b) M. Rubin, M. Rubina and V. Gevorg-
yan, Chem. Rev., 2007, 107, 3117.
3 For selected articles, see: (a) D. H. Camacho, I. Nakamura, S.
Saito and Y. Yamamoto, Angew. Chem., Int. Ed., 1999, 38,
3365–3367; (b) M. Lautens, C. Meyer and A. Lorenz, J. Am.
Chem. Soc., 1996, 118, 10676; (c) M. Lautens and Y. Ren, J.
Am. Chem. Soc., 1996, 118, 9597; (d) S. Saito, M. Masuda and S.
Komagawa, J. Am. Chem. Soc., 2004, 126, 10540; (e) M. Shi, B.-Y.
Wang and J.-W. Huang, J. Org. Chem., 2005, 70, 5606; (f) I.
Nakamura, B. H. Oh, S. Saito and Y. Yamamoto, Angew. Chem.,
Int. Ed., 2001, 40, 1298; (g) N. Tsukada, A. Hibuya, I. Nakamura
and Y. Yamamoto, J. Am. Chem. Soc., 1997, 119, 8123; (h) M.
Suginome, T. Matsuda and Y. Ito, J. Am. Chem. Soc., 2000, 122,
11015; (i) M. E. Scott, Y. Bethuel and M. Lautens, J. Am. Chem.
Soc., 2007, 129, 1482; (j) S. Brase and A. de Meijere, Angew.
Chem., Int. Ed. Engl., 1995, 34, 2545; (k) I. Nakamura, T. Nemoto,
Y. Yamamoto and A. de Meijere, Angew. Chem., Int. Ed., 2006, 45,
5176.
Scheme 4 Proposed mechanisms for the two distinct processes.
comes about due to the electronic nature of the electron-
withdrawing substituents on the benzene ring. Then, inter-
mediate E is attacked from two different orientations by a
bromide anion to form p-allylpalladium intermediates F and G
via path c and path d, respectively.12 Subsequent attack of the
p-allylpalladium intermediates by the tethered hydroxyl group
furnishes the corresponding two products of dihydrofuran 4
and tetrahydrofuran 5, respectively. In addition, reductive
elimination of the forming palladium(II) halide hydride
(HPdX) intermediate gives a Pd0 species, which is oxidized
by CuBr2 to regenerate the active catalyst. An alternative
mechanism of the reactions under air, similar to a Wacker-
type reaction, is provided in the ESI.w
Concerning the operation of process I, two points should be
outlined: (a) 0.2 equiv. of CuBr2 (40.05 equiv. of PdCl2) is
necessary because traces of substrate 1 follow process II and
subsequently result in the formation of catalytically inactive
Pd0 species although the bromine atom is not incorporated in
the products. (b) Adding 4 A molecular sieves could get rid of
the forming water and subsequently decrease the density of
nucleophilic bromide anions, accordingly improving the yields
of (arylcyclobutenyl)carbinols 2 and 3 especially for substrates
1 with electron-withdrawing groups on the benzene ring.
In conclusion, we have developed an efficient method for
ring enlargement of 2-(arylmethylene)cyclopropylcarbinols
catalyzed by palladium(II) chloride in the presence of
copper(II) bromide. Substrates with electron-donating groups
on the benzene ring undergo rearrangement smoothly to
provide (arylcyclobutenyl)carbinols in good yields and selec-
tivities under mild conditions. While, substrates with electron-
withdrawing groups on the benzene ring produce dihydrofur-
ans and tetrahydrofurans in good total yields in the presence
of stoichiometric amounts of CuBr2. Mechanisms have been
4 M. Shi, L.-P. Liu and J. Tang, J. Am. Chem. Soc., 2006, 128, 7430.
5 A. Furstner and C. Aıssa, J. Am. Chem. Soc., 2006, 128, 6306.
¨
¨
6 For the preparation of 2-(arylmethylene)cyclopropylcarbinols, see:
(a) A. Turcant and L. M. Corre, Tetrahedron Lett., 1976, 17, 985;
(b) L. M. Corre, A. Hercouet and B. Bessieres, J. Org. Chem., 1994,
59, 5483.
7 (a) G.-Q. Tian and M. Shi, Org. Lett., 2007, 9, 2405; (b) G.-Q. Tian
and M. Shi, Org. Lett., 2007, 9, 4917; (c) L.-X. Shao, M.-H. Qi and
M. Shi, Tetrahedron Lett., 2008, 49, 165.
8 Crystal data for 2a (CCDC 665100): C12H14O2; M = 190.23;
crystal color, habit: colorless, prismatic; crystal system: monocli-
nic; lattice type: primitive; a = 46.33(2), b = 5.461(3), c = 8.220(4)
A, b = 92.9261, V = 2077.2(18) A3; space group: P2(1)/c; Z = 8;
Dcalc = 1.217 g cmꢁ3; F000 = 816; diffractometer: Rigaku AFC7R;
residuals: R, wR: 0.0677, 0.1706. F000 = 1232; diffractometer:
Rigaku AFC7R; residuals: R; wR: 0.0928, 0.2064.
9 The anti-configuration of 5 was tentatively assigned on the basis of
NOESY spectroscopy (see the ESIw).
10 (a) F. A. Carey and R. J. Sundberg, Advanced Organic Chemistry,
Plenum Press, New York, 5th edn, 1998, pp. 221, 419; (b) C. J.
Lancelot, D. J. Cram and P. v. R. Schleyer, in Carbonium Ions, ed.
G. A. Olah and P. v. R. Schleyer, Wiley, New York, 1972, vol. III,
p. 1347.
11 (a) D. M. Kalendra and B. R. Sickles, J. Org. Chem., 2003, 68,
1594; (b) J.-M. Xu, B.-K. Liu, W.-B. Wu, C. Qian, Q. Wu and X.-
F. Lin, J. Org. Chem., 2006, 71, 3991.
12 (a) S. Ma, L. Lu and J. Zhang, J. Am. Chem. Soc., 2004, 126, 9645;
(b) I. Nakamura, H. Itagaki and Y. Yamamoto, J. Org. Chem.,
1998, 63, 6458; (c) I. Nakamura, A. I. Siriwardana, S. Saito and Y.
Yamamoto, J. Org. Chem., 2002, 67, 3445; (d) D. H. Camacho, I.
Nakamura, B. H. Oh, S. Saito and Y. Yamamoto, Tetrahedron
Lett., 2002, 43, 2903.
ꢀc
This journal is The Royal Society of Chemistry 2008
2670 | Chem. Commun., 2008, 2668–2670