One-Pot, Pd-catalyzed synthesis of trans -dihydrobenzofurans from o -aminophenols

Article abstract of DOI:10.1021/ol100433z

Figure presented An efficient and facile synthesis of trans- dihydrobenzofurans has been accomplished from o-aminophenols and phenylpropenes via a novel (one-pot) diastereoselective Pd-catalyzed oxyarylation reaction. The development and optimization of this method is described.

Full text of DOI:10.1021/ol100433z

ORGANIC
LETTERS
2010
Vol. 12, No. 9
1976-1979
One-Pot, Pd-Catalyzed Synthesis of
trans-Dihydrobenzofurans from
o-Aminophenols
Ericsson D. Coy B.,† Ljubisa Jovanovic,‡ and Michael Sefkow*,‡
UniVersita¨t Potsdam, Institut fu¨r Chemie, Karl-Liebknecht Strasse 24-25,
14476 Golm, Germany, and UniVersidad Nacional de Colombia, Facultad de Ciencias,
Departamento de Qu´ımica, Laboratorio de InVestigacio´n en Productos Naturales
Vegetales, AA 14490, Cra 30 45-03, Bogota´, D.C., Colombia
sefkow@uni-potsdam.de
Received February 20, 2010
ABSTRACT
An efficient and facile synthesis of trans-dihydrobenzofurans has been accomplished from o-aminophenols and phenylpropenes via a novel
(one-pot) diastereoselective Pd-catalyzed oxyarylation reaction. The development and optimization of this method is described.
The trans-2,3-disubstituted dihydrobenzofurans (DHBs) be-
long to an important class of heterocycles because many
biologically active, naturally occurring 8,5′-neolignans and
several synthetic derivatives possess this structural motif. The
biological properties have attracted chemists in developing
efficient syntheses for these trans-DHBs.1 Several approaches
have been described in the literature, in particular, oxidative
coupling reactions, Lewis acid catalyzed and Schmid rearrange-
ments, acid-catalyzed cycloadditions, radical-based cyclizations,
Pummerer reactions, directed R-metalations, and transition-
metal-mediated reactions, among others.1 However, most of
these methods suffer from low yields, undesired side products,
low chemo- and/or diastereoselectivities, difficulties in product
purification, and strong and/or precious reaction conditions.
A variation of the Heck reaction is the Pd-catalyzed
oxyarylation4 that seemed to be a promising methodology
for an easy and efficient synthesis of DHBs (Figure 1). This
Figure 1. Synthetic approach to trans-DHBs via oxyarylation.
reaction has been used in the past to obtain pterocarpans
from substituted o-iodo- or o-chloromercuriophenols5 and
Transition-metal-catalyzed C-C coupling reactions have
become a very powerful tool in the arsenal of organic
chemists, and palladium is the most widely used metal in
such cross-coupling reactions. The Heck reaction represents
a simple and efficient way to obtain variously substituted
alkenes and other unsaturated compounds.2,3
(2) For a recent monograph, see: Oestreich, M. The Mizoroki-Heck
Reaction; John Wiley & Sons: New York, 2009.
(3) Some recent reviews: (a) Djakovitch, L.; Pinel, C. Curr. Org. Synth.
2009, 6, 54–65. (b) Zeni, G.; Larock, R. C. Chem. ReV. 2006, 106, 4644–
4680. (c) Dounay, A. B.; Overman, L. E. Chem. ReV. 2003, 103, 2945–
2963. (d) Beletskaya, I. P.; Cheprakov, A. V. Chem. ReV. 2000, 100, 3009–
3066.
(4) (a) Li, W.; Shi, M. Eur. J. Org. Chem. 2009, 270–274. (b) Rozkhov,
R. V.; Larock, R. C. AdV. Synth. Catal. 2004, 346, 1854–1858. (c) Larock,
R. C.; Berrios-Pen˜a, N.; Narayanan, K. J. Org. Chem. 1990, 55, 3447–
3450.
† Universidad Nacional de Colombia.
‡ Universita¨t Potsdam.
(1) Sefkow, M. Synthesis 2003, 2595–2625.
10.1021/ol100433z  2010 American Chemical Society
Published on Web 03/25/2010

2-propenyl-2,3-DHBs from vinylcycloalkanes and o-iodo-
phenols.6 To our knowledge, the synthesis of trans-2-aryl-
3-methyl-DHBs by oxyarylation has never been reported.
The oxyarylation, an intermolecular C(sp2)-C(sp2) cou-
pling reaction,7 can be regarded as a pseudo [3 + 2]
cycloaddition involving a 2-hydroxyarylpalladium species as
C3 and an alkene as C2 unit. In most studies reported thus
far, o-iodophenols were employed as the C3 precursor.8
However, we found that oxyarylations using o-halophenols
provided the trans-2-aryl-3-methyl-DHBs only in 10-30%
beside substantial amounts of several unidentified side
products. Under optimized conditions, the reaction of o-
iodovanillin with anethole (1) gave the corresponding DHB
(Figure 1, X ) I) in less than 50% yield.9 Consequently,
oxyarylations using o-halophenols are not suitable for the
synthesis of this type of DHBs.
Scheme 1. Oxyarylation with Arenediazonium Salts
Aryl diazonium salts are efficient reagents in Heck
reactions.10 The influence of the substitution pattern of the
alkenes on the outcome of the Heck reaction has been
reported by Kikukawa et al.11 Unfortunately, with phenyl-
propene’s C2 moiety, the C-C cross-coupling product was
formed only in 6% yield. Based on this report we have
developed a novel Heck-type oxyarylation (Figure 1, X )
mixture. In addition to 4, two side products were often
isolated. These compounds were identified as p-anisaldehyde
5, apparently derived from the reaction of alkene 1 with
NOBF4, and 4-nitroso-DHB 6, formed by exposure of either
the starting material 3 or the product 4 to an excess of
NOBF4. Several parameters (base, catalyst, solvent, diazo-
tization reagent, times and temperatures) were optimized for
the novel “one-pot” oxyarylation protocol. In addition, the
+
N2 ) using anethole (1) and 2-hydroxybenzenediazonium
tetrafluoroborate (2) as starting materials. When Pd2(dba)3
12
is employed as catalyst, CaCO3 as base, and MeCN as
solvent, the oxyarylation product, DHB 4, is isolated in up
to 85% yield and with up to 94:6 diastereoselectivity (Scheme
1, route A). Major drawbacks of this reaction are that the
required diazonium salt is not commercially available and
the yield for its preparation13 from 2-aminophenol (3) does
not exceed 33%.
For this reason, we considered carrying out both reactions
successively in one pot by in situ generation of the diazonium
salt from 2-aminophenol (3) using NOBF4 and subsequent
oxyarylation with anethole (1) to obtain trans-4 (Scheme 1,
route B). An initial experiment carried out in the presence
of 1.2 equiv of CaCO3 and 5 mol % of Pd2(dba)3 in MeCN
at room temperature provided trans-414 in 41% yield (Table
1, entry 1).
Table 1. Base and Counterion Effect on the Oxyarylation
ratio of productsb
trans-4/cis-4/5/6
entry
base
CaCO3
yield (%) of 4a
According to the observation with isolated diazonium salts,
this reaction proceed highly diastereoselective to afford the
trans-4. The cis isomer was formed in less than 5% yield,
1
41
53
43
>10/1/<1/1.3
>10/1/<1/1
7/1/<1/0
2
3
4
5
BaCO3
MgCO3
MnCO3
ZnCO3
1
as judged by the H NMR spectrum of the crude product
57
64 (59)
38
>10/1/<1/1.3
>10/1/<1/1.5
6/1/<1/0
(5) Kiss, L.; Kurta´n, T.; Antus, S.; Brunner, H. ARKIVOC 2003, 5, 69–
76.
6
CoCO3
(6) Larock, R. C.; Yum, E. K. Tetrahedron 1996, 52, 2743–2758.
(7) Boto, A.; Ling, M.; Meek, G.; Pattenden, G. Tetrahedron Lett. 1998,
39, 8167–8170.
7
8
Bi2(CO3)3
M2CO3
49
0
6/1/<1/1.8
c
9
Ca-Gly-PO4
CaSO4
CaHPO4
Ca3(PO4)2
CaSiO3
62 (58)
43
44
49
56
>10/1/<1/1.2
6/1/0.7/1.3
7/1/1/1.1
(8) Catellani, M.; del Rio, A. Russ. Chem. Bull. 1998, 47, 928–931.
(9) Jovanovic, L. Ph.D. Thesis, Universita¨t Potsdam, 2006.
(10) For a comprehensive review, see: Roglans, A.; Pla-Quintana, A.;
Moreno-Man˜as, M. Chem. ReV. 2006, 106, 4622–4643.
(11) Kikukawa, K.; Nagira, K.; Wada, F.; Matsuda, T. Tetrahedron 1981,
37, 31–36.
10
11
12
13
14
15
7/1/<1/0
7/1/<1/1.1
>10/1/<1/<1
>10/1/<1/1.6
CaO
Ca(OH)2
42
46
(12) (a) Brunner, H.; Le Cousturier de Courcy, N.; Geneˆt, J.-P.
Tetrahedron Lett. 1999, 40, 4815–4818. (b) Brunner, H.; Le Cousturier de
Courcy, N.; Geneˆt, J.-P. Synlett 2000, 201–204.
a Calculated by 1H NMR using tetradecane as internal standard; yields
(13) Kolar, G. F. Z. Naturforsch. B 1972, 27, 1183–1185.
(14) The trans-configuration of 4 and all analogues was assigned on the
basis of the reported NMR data (trans-4): Snider, B. B.; Han, L.; Xie, C. J.
Org. Chem. 1997, 62, 6978–6984. cis-4: Cheung, W.-H.; Zheng, S.-L.; Yu,
W.-Y.; Zhou, G.-C.; Che, C.-M. Org. Lett. 2003, 5, 2535–2538.
of isolated product in parentheses. b Ratio determined by selected, dif-
ferentiable H NMR signals. c M corresponds to single valent metals such
1
as Li+, Na+, Cs+, and Ag+.
Org. Lett., Vol. 12, No. 9, 2010
1977

applicability of our methodology is demonstrated by using
a variety of different alkenes and o-aminophenols as sub-
strates.
Table 2. NOPF6 as Diazotizating Agent and Additives Effect
From the results achieved with isolated diazonium salts
we learned that the base has a strong influence on the
outcome of the oxyarylation. Therefore, the base was
systematically screened with respect to both cation (Table
1, entries 2-8) and anion (entries 9-15), while all other
parameters remained unchanged. Carbonates from double or
triple valent cations gave trans-4 in 38-64% yield, with
MgCO3 and CoCO3 being less effective and ZnCO3 most
effective. In addition, with Mg2+, Co2+, and Bi3+ as cations,
significantly more cis-product was formed (∼15%). The best
diastereoselectivities have been achieved with Mn2+ and
Zn2+. It is notable that the carbonates from monovalent
cations, such as Li+, Na+, Cs+, and Ag+, completely failed
to give any desired product. This result is unexpected since
these bases are widely used in common Heck reactions.4a
The structure of the anion does not influence the outcome
of the oxyarylation to that extent (Table 1, entries 9-15).
Several nonbasic anions (sulfate, silicate and some phos-
phates) as well as the basic oxide and hydroxide have been
examined using Ca2+ as counterion. In all cases, the yields
of 4 were in the range between 42-62%. The stereoselec-
tivity of the oxyarylation correlated with the basicity of the
counterion. Whereas the basic oxide and hydroxide gave
good stereoselectivities albeit low yields, the nonbasic anions
provided ca. 7:1 mixtures of trans- and cis-4. As an
exception, the nonbasic calcium glycero phosphate gave not
only the best yield (62%) but also the best diastereoselectivity
(>95:5) (Table 1, entry 9).15
ratio of productsb
entry
base
CaCO3
additive yield (%) of 4a trans-4/cis-4/5/6
1
2
3
4
5
6
7
8
9
53
>10/1/<1/2
>10/1/<1/1.1
7/1/1.1/0
>10/1/<1/13
>10/1/0/0
>10/1/9.5/<1
>10/1/<1/<1
>10/1/1.2/3.6
>10/1/4/3.5
3/1/3/1.5
ZnCO3
Ca-Gly-PO4
ZnCO3
ZnCO3
ZnCO3
ZnCO3
ZnCO3
ZnCO3
85 (79)
77 (73)
24
c
d
88 (82)
33
44
43
29
Bu4NCl
MS 3 Å
10 eq H2O
e
f,g
10 ZnCO3
NOPF6
27
a Calculated by 1H NMR using tetradecane as internal standard; yields
of isolated product in parentheses. b Ratio determined by selected, dif-
ferentiable H NMR signals. c Oxyarylation immediately after addition of
1
NOPF6. d Oxyarylation 24 h after addition of NOPF6. e 2 equiv of NOPF6
used for diazotization. f 1 equiv of NOPF6 added after oxyarylation, and
reaction proceeded for a further 20 h. g Formation of 6 (ca. 30%) occurred
after treatment of purified 4 with 1 equiv of NOPF6 for 20 h.
latter one. Again, best results were achieved with ZnCO3
and Ca-glycerophosphate as base (85 and 77%), though
ZnCO3 provided 4 with better trans-diastereoselectivity. A
diazotization time of 30 min is sufficient since shorter periods
of activation decreased the yields and longer diazotization
times (24 h) gave only marginal better yields (Table 2, entries
4 and 5).
Several additives were often applied in Heck reactions to
further increase the yields.19 In our hands, additives, such
as Bu4NCl, 3 Å molecular sieves, or water, do not contribute
to improve the yield of the oxyarylation (Table 2, entries
6-8). In contrast, all additives decreased the yield of
compound 4. Notably, Bu4NCl increased the yield of
aldehyde 5, which can be also prepared in 52% yield by
treatment of anethole (1) with 1 equiv of NOPF6.
Although 1 equiv of water is present throughout the
oxyarylation reaction due to the in situ diazotization, addition
of extra water decreased the yield of 4 and promoted the
formation of 6. On the other hand, removal of water by
addition of 3 Å molecular sieve also decreased the yield of
4 and provided many unidentified side products. As expected,
a second equivalent of NOPF6, either added at the beginning
of diazotization or after oxyarylation decreased the yield of
DHB and increased the formation of 5 and 6 (Table 2, entries
9 and 10).
The influence of the catalyst was examined with respect
to catalyst load, transition metal used, and nature of the
ligand. The catalyst load was studied using Pd2(dba)3 as
reference. The yield of oxyarylation product was comparable
at either 5 or 10 mol % catalyst load but significantly
Several solvents have been employed in Heck reactions16
such as NMP, DMF, DMA, THF, or CH2Cl2, but in neither
of these solvents did oxyarylation occur. Even PhCN17 gave
less satisfactory results compared to MeCN. Nitriles were
the only solvent suitable to initiate the oxyarylation from
diazonium salts, confirming the work by Eberlin et al.18 They
showed that arenediazonium salts and Pd(dba)2 in MeCN
undergo a time-dependent process involving ligand exchange.
As a result of this phenomenon, a stable cationic intermediate
having one MeCN molecule coordinated as ligand was
formed slowly. This intermediate is likely to be the key
intermediate in the Heck reaction using diazonium salts.11
Besides NOBF4, nitrosyl hexafluorophosphate is the only
other commercially available nitrosonium salt. NOPF6 was
thus examined as diazotizating agent (Table 2). The yield
was significantly improved with NOPF6 in comparison with
NOBF4, probably because of the hygroscopic nature of the
(15) It should be noted that the yields given in Tables 1-4 refer to
calculated yields based on the integrals of selected product signals in the
1H NMR spectra of the crude products (δH at 1.3 and 5.2) with those of
tetradecane added as internal standard. However, in several experiments,
we purified the products by chromatography and found marginal differences
between calculated and isolated yields (ca. 5%).
(16) Tietze, L. F.; Ila, H.; Bell, H. P. Chem. ReV. 2004, 104, 3453–
3516.
(17) (a) Moro, A. V.; Cardoso, F. S. P.; Correia, C. R. D. Tetrahedron
Lett. 2008, 49, 5668–5671. (b) Moro, A. V.; Cardoso, F. S. P.; Correia,
C. R. D. Org. Lett. 2009, 11, 3642–3645.
(18) Sabino, A. A.; Machado, A. H. L.; Correia, C. R. D.; Eberlin, M. N.
Angew. Chem., Int. Ed. 2004, 43, 2514–2518.
(19) Biffis, A.; Zecca, M.; Basato, M. J. Mol. Catal. A: Chem. 2001,
173, 249–274.
1978
Org. Lett., Vol. 12, No. 9, 2010

decreased when only 1 mol % of Pd2(dba)3 was applied. As
a consequence, all further experiments were carried out with
5 mol % catalyst load.
Likewise, the suitablility of various phenylpropenes for
their conversion to DHBs was examined (Table 4). Unsub-
Several other Pd catalysts, Pd(OAc)2, Pd(acac)2, and
[Pd(Ph3P)]4, as well as the novel N-heterocyclic carbene
palladium complexes ([(SIPr)Pd(cinnamyl)Cl],20 [Pd(IPr)Cl2]2,21
and PEPPSI-IPr22), commonly used in Heck reactions,23 were
compared with Pd2(dba)3 in their efficiency to catalyze the
oxyarylation. Only with Pd(OAc)2 was DHB 4 obtained in a
similar yield (84%, Pd2(dba)3: 85%), although the diastereose-
lectivity was decreased significantly (88:12, Pd2(dba)3: >95:5),
while Pd(acac)2 provided 4 in 58% and [Pd(Ph3P)]4 in 49%
yield. All Pd-NHC complexes failed to give the DHB. Any
catalyst comprising a different transition metal, such as Ni(II),
Cu(I), Ag(I), and Pt(0), also failed to catalyze the oxyarylation.
Having developed a working oxyarylation protocol, vari-
ous substituted 2-aminophenols were examined as substrates
for diazotization/oxyarylation (Table 3). With electron-
Table 4. Substitution Effect on Starting Material Alkene
entry
alkene
trans-ꢀ-methylstyrene
yielda (%)
1
0
2
trans-stilbene
0
3
4
5
6
(E)-1-methyl-2-(prop-1-enyl)benzene
(E)-1,3,5-trimethoxy-2-(prop-1-enyl)benzene
(E)-1,4-dimethoxy-2-(prop-1-enyl)benzene
(E)-1,2,3-trimethoxy-5-(prop-1-enyl)benzene
isoeugenol
26
18
26
73 (70)
7
85 (81)
8
isomyristicin
77 (71)
9
isosafrol
methyl isoeugenol
cis-cinnamyl alcohol
cinnamyl methyl ether
80
81
0
Table 3. Substitution Effect of Different 2-Aminophenols
10
11
12
0
a Calculated by 1H NMR using tetradecane as internal standard by means
of selected, differentiable 1H NMR signals at around δH 5.1-5.4 from the
obtained DHBs; yields of isolated products in parentheses.
entry
2-aminophenol
yielda (%)
1
2
3
4
5
6
7
2-amino-4-chlorophenol
2-amino-5-nitrophenol
2-amino-pyridin-3-ol
4-amino-3-hydroxybenzoic acid
3-amino-4-hydroxybenzoic acid
methyl 4-amino-3-hydroxybenzoate
methyl 3-amino-4-hydroxybenzoate
85 (79)
49
0
59
44
stituted arylpropenes failed to yield any oxyarylation product
(Table 4, entries 1 and 2). Good yields of DHBs (73-85%)
have been achieved with 4-hydroxy- or alkoxy-substituted
phenylpropenes unsubstituted at the ortho position of the
aromatic ring (Table 4, entries 6-10). Phenols can be used
unprotected. Any ortho substituent at the aromatic ring (Table
4, entries 3-5) dramatically reduces the reaction rate and
decreases the yield of product (18-26%). Cinnamic alcohol
and derivatives with a heteroatom at the propenyl moiety
completely failed to afford the corresponding DHBs (Table
4, entries 11 and 12).
In conclusion, a novel, convergent, stereoselective syn-
thesis of trans-DHBs from commercially available o-
aminophenols and arylpropenes was developed via diazoti-
zation and Pd-catalyzed oxyarylation in a “one-pot” process.
An important class of natural products, namely 8,5′-neolig-
nans, which comprises a similar DHB core, is accessible with
good yields and diastereoselectivities under mild conditions.
84 (79)
64
a Calculated by 1H NMR using tetradecane as internal standard by means
of selected, differentiable 1H NMR signals at around δH 5.1-5.4 from the
obtained DHBs; yields of isolated products in parentheses.
donating groups (e.g., chloro, Table 3, entry 1) or electron-
withdrawing groups (e.g., carboxylic acids or esters, Table
3, entries 5 and 7) para to the hydroxy group, acceptable
yields of the corresponding oxyarylation products (44-85%)
have been achieved. Reasonable yields of DHBs (59-84%)
were also obtained when substrates with electron-withdraw-
ing groups para to the amino group (Table 3, entries 2, 4,
and 6) were employed. Not surprisingly, 2-amino-3-hydroxy-
pyridine did not give the desired product since pyridines act
as donators for complexation of transition metals (Table 3,
entry 3). It should be noted that free carboxyl groups are
tolerated, although the yields are diminished by ∼20-25%.
Acknowledgment. We thank the Institut fu¨r Chemie at
Universita¨t Potsdam for the analytical support. E.D.C.B.
acknowledges the DAAD for a fellowship.
Supporting Information Available: Experimental pro-
cedure, spectral data, 1H and 13C NMR spectra, and a sample
calculation of the yields. This material is available free of
charge via the Internet at http://pubs.acs.org.
(20) Navarro, O.; Marion, N.; Mei, J.; Nolan, S. P. Chem. Eur. J. 2006,
12, 5142–5148.
(21) Viciu, M. S.; Kissling, R. M.; Stevens, E. D.; Nolan, S. P. Org.
Lett. 2002, 4, 2229–2231.
(22) Organ, M. G.; Abdel-Hadi, M.; Avola, S.; Hadei, N.; Nasielski, J.;
O’Brien, C. J.; Valente, C. Chem. Eur. J. 2007, 13, 150–157.
(23) Andrus, M. B.; Song, C.; Zhang, J. Org. Lett. 2002, 4, 2079–2082.
OL100433Z
Org. Lett., Vol. 12, No. 9, 2010
1979