Rhodium-catalyzed one-pot hydroformylation-cyclization of allylbenzene derivatives: Simple and efficient route to 5,6-dihydronaphthalenes

Article abstract of DOI:10.1016/j.catcom.2010.02.014

The one-pot hydroformylation-cyclization of allylbenzene derivatives by the catalyst system Rh(CO)₂acac/ultranox626/CO/H₂/CH₂Cl₂ led chemoselectively to 5,6-dihydronaphthalene derivatives 5a-d in good yields. The addition of a catalytic amount of H₃PO₄ enhanced in situ the cyclization process via the nucleophilic attack on the carbonyl group of the linear aldehyde, and finally the elimination of alcohol. The type of substitution on phenyl group of the allylbenzene is of great importance in enhancing the cyclization process.

Full text of DOI:10.1016/j.catcom.2010.02.014

Catalysis Communications 11 (2010) 778–782
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Rhodium-catalyzed one-pot hydroformylation–cyclization of allylbenzene
derivatives: Simple and efficient route to 5,6-dihydronaphthalenes
*
Mouheddin Alhaffar, Rami Suleiman, Bassam El Ali
Chemistry Department, King Fahd University of Petroleum and Minerals (KFUPM), Dhahran 31261, Saudi Arabia
a r t i c l e i n f o
a b s t r a c t
Article history:
The one-pot hydroformylation–cyclization of allylbenzene derivatives by the catalyst system Rh(CO)2a-
cac/ultranox626/CO/H2/CH2Cl2 led chemoselectively to 5,6-dihydronaphthalene derivatives 5a–d in good
yields. The addition of a catalytic amount of H3PO4 enhanced in situ the cyclization process via the nucle-
ophilic attack on the carbonyl group of the linear aldehyde, and finally the elimination of alcohol. The
type of substitution on phenyl group of the allylbenzene is of great importance in enhancing the cycliza-
tion process.
Received 17 December 2009
Received in revised form 7 February 2010
Accepted 8 February 2010
Available online 12 February 2010
Keywords:
Allylbenzene
Eugenol
Ó 2010 Elsevier B.V. All rights reserved.
Hydroformylation
Cyclization
Dihydronaphthalenes
Rhodium
Phosphite
Ultranox626
1. Introduction
cyclodehydration have been reported for allylpyrole derivatives
[19].
The hydroformylation of functionalized unsaturated substrates,
especially when involved in tandem or domino reaction sequences,
constitutes a very versatile tool for the synthesis of fine chemicals
[1]. Thus, some important reactions such as reduction, nucleophilic
addition or cyclization can be achieved directly under the reaction
conditions of the oxo process [2].
In this communication, we report interesting results on the
chemoselective synthesis of dihydronaphthalene derivatives in
high yield via oxo process. To the best of our knowledge, this is
the first example of a simple and efficient method for the synthesis
of functionalized dihydronaphthalenes via the one-pot hydrofor-
mylation–cyclization of allylbenzene derivatives.
During the survey of the literature on hydroformylation of allyl-
benzene derivatives, we have observed in one of the publication
that some 5,6-dihydronaphthalene cyclic compounds were ob-
tained as minor products. However, only trace amount of dihydro-
naphthalene by-products were obtained in hydroformylation of
eugenol using Co2(CO)8 [3]. It is well known that some functional-
ized dihydronaphthalenes exhibit interesting physiological and
biological activities [4–6]. Moreover, these compounds are valu-
able intermediates for the synthesis of numerous biologically nat-
ural products and pharmaceuticals. Recently, some anilide
derivatives of dihydronaphthalene showed anti-HIV-1 activity
[7]. In these respects, the design and development of efficient syn-
thesis method for dihydronaphthalene derivatives are an attractive
and interesting research topic in synthetic chemistry [8–16]. The
hydroformylation of allylbenzene derivatives have been studied
carefully in the literature [17,18]. Hydroformylation followed by
2. Experimental
2.1. General
Allylbenzene derivatives, rhodium catalysts, phosphine and
phosphite ligands, and all acids are commercially available materi-
als and were used without any purification. Dry solvents have been
used in all experiments. 1H and 13C NMR spectra were recorded on
500 MHz Joel 1500 NMR machine. Chemical shifts (d) were reported
in ppm relative to tetramethyl silane (TMS) using CDCl3. IR spectra
were recorded on Perkin–Elmer 16F PC FT-IR spectrometer and re-
ported in wave numbers (cmꢀ1). Gas chromatography (GC) analyses
were realized on Agilent GC 6890 plus Series gas chromatograph
equipped with a split–splitless injector (split ratios of 20:1). The
temperature of the injector was 250 °C, with 10 psi constant pres-
sure. The column was an HP-5 column (30 m ꢁ 0.25 mm i.d.,
0.25
lm film thickness). Helium was the carrier gas at a flow rate
* Corresponding author. Tel.: +966 3 8604491; fax: +966 3 8604277.
E-mail address: belali@kfupm.edu.sa (B. EL Ali).
of 1.5 ml/min and programmed temperature was applied to obtain
1566-7367/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2010.02.014

M. Alhaffar et al. / Catalysis Communications 11 (2010) 778–782
779
the separation of the compounds; precisely the initial temperature
2.3. Characterization of dihydronaphthalenes
was 50 °C, hold 2 min, then ramped at 10 °C/min to 140 °C (hold
time 0 min), then finally ramped at 20 °C/min to 250 °C (hold time
20 min). The detector was flame ionization detector (FID), with
hydrogen and air flow of 40.0 and 450.0 ml/min, respectively. Ma-
keup flow was on with 45.0 ml/min of helium. The products of
the reactions were also analyzed on GC–MS Varian Saturn 2000
equipped with 30 m capillary column (HP-5). Thin-layer chroma-
tography (TLC) analyses were performed on silica gel Merck 60
2.3.1. 3-Methoxy-5,6-dihydronaphthalen-2-ol (5a)
White solid, mp 108–109 °C; IR (KBr) v (cmꢀ1) 1576, 1613 (C@C),
3424 (OH); 1H NMR d (CDCl3): 2.26 (m, 2H, C6-2H), 2.70 (t, 2H,
J = 8.25 Hz, C5-2H), 3.87 (s, 3H, OMe), 5.44 (s, 1H, OH), 5.91 (dt,
1H, J = 9.50, 4.30 Hz, C7-H), 6.33 (dt, J = 9.75, 1.85 Hz, C8-H), 6.62
(s, 1H, C4-H), 6.63 (s, 1H, C2-H); 13C NMR d (CDCl3): 23.2, 27.3,
56.0, 110.6, 112.4, 126.6, 127.1, 127.2, 127.5, 143.8, 145.1; GC–
MS m/z 176 (M+). Analysis calculated for C11H12O2 (176.20): C,
74.98; H, 6.86; found: C, 74.86; H, 6.83.
F254 plates (250 lm layer thickness).
2.3.2. 5,6-Dihydronaphtho[2,3-d][1,3]dioxole (5c)
2.2. General procedure for the hydroformylation of allylbenzene
derivatives
Oil; IR (neat) v (cmꢀ1) 1575, 1614 (C@C); 1H NMR d (CDCl3):
2.27 (m, 2H, C6-2H), 2.71 (t, 2H, J = 8.25 Hz, C5-2H), 5.90 (s, 2H,
OCH2O), 5.94 (dt, 1H, J = 9.45, 4.43 Hz, C7-H), 6.35 (dt, J = 9.45,
1.75 Hz, C8-H), 6.57 (s, 1H, C4-H), 6.64 (s, 1H, C2-H); 13C NMR d
(CDCl3): 23.0, 27.7, 100.6, 106.7, 108.4, 108.5, 126.5, 127.4, 127.9,
129.2, 145.9; GC–MS m/z 174 (M+). Analysis calculated for
C11H10O2 (174.18): C, 75.85; H, 5.79; found: C, 75.72; H, 5.88.
To a Parr autoclave fitted with a glass liner and stirring bar,
Rh(CO)2(acac) (0.005 mmol), ligand (0.030 mmol), allylbenzene
derivative (5.0 mmol), H3PO4 (85%) (0.14 mmol), anisole
(3.76 mmol) and CH2Cl2 (5.0 ml) are added. The autoclave is vented
three times with CO and then pressurized with CO and H2. The
mixture is stirred and heated for the required time. At the end of
the reaction, the autoclave is cooled down, and the pressure is re-
leased under fume hood. The reaction mixture is filtered and the
products are separated on silica gel column using ethyl acetate/
petroleum ether (2:8) as eluent and identified by GC, NMR, FT-IR,
GC–MS, and elemental analysis.
2.3.3. 1,3-Dimethoxy-5,6-dihydronaphthalen-2-ol (5d)
Oil; IR (neat) v (cmꢀ1) 1575, 1612 (C@C), 3424 (OH); 1H NMR d
(CDCl3): 2.24 (m, 2H, C6-2H), 2.66 (t, 2H, J = 8.25 Hz, C5-2H), 3.82
(s, 3H, OMe), 3.85 (s, 3H, OMe), 5.40 (s, 1H, OH), 5.94 (dt, 1H,
J = 9.75, 4.67 Hz, C7-H), 6.44 (s, 1H, C4-H), 6.66 (dt, J = 9.75,
R1
R1
R1
R2
R1
R2
R2
Rh(CO)2acac, Ligand
R2
CHO
+
+
Solvent, T (oC), Time (h)
CO/H2 (1: 2) (300 psi)
R3
R3
CHO
R3
R3
CHO
1
2
4
3
a: R1= H, R2= OH, R3= OCH3
b: R1= H, R2=R3= OCH3
R1
R1
R2
R3
R2
+
c: R1 = H, R2,R3 = -OCH2O-
R3
d:
R1=R3= OCH3, R2= OH
e: R1=R3= H, R2= OCH3
f:
1'
5
R1=R2=R3= H
Scheme 1.
Table 1
Hydroformylation–cyclization of eugenol 1a by Rh(CO)2acac. Effect of the type of ligand.a
Entry
Ligand
Conversion (%)b
Yield of 10a (%)b
Yield of 5a (%)b
Yield of aldehydes (%)c
Aldehydes distribution (%)d
2a
3a
28
96
19
86
95
4a
1e
2
PPh3
PPh3
Ultranox626
Ultranox626
(PhO)3P
dppp
99
98
99
98
97
3
3
21
3
4
16
1
0
55
0
65
52
0
96
22
96
29
29
2
1
4
0
4
5
0
0
71
0
81
10
0
3e
4
5
6
100
45
0
55
7
dppb
9
1
3
5
a
Reaction conditions: Rh(CO)2acac (0.005 mmol), ligand (0.03 mmol, except for PPh3 and (PhO)3P 0.06 mmol), eugenol (1a) (5.0 mmol), H3PO4 (85%) (0.14 mmol), CH2Cl2
(5.0 ml), CO/H2 (100/200 psi), 2 h, 110 °C.
b
Determined by GC using anisole as internal standard.
c
Calculated yield = conversion ꢀ (10a + 5a).
d
Determined by GC and 1H NMR.
No acid.
e

780
M. Alhaffar et al. / Catalysis Communications 11 (2010) 778–782
tant industrial products. A careful screening of the experimental
conditions including the type of ligand and the type and the
amount of additive has been made.
O
O
O
O
3.1. Effect of the type of ligand
O
P
P
O
In preliminary experiment, the hydroformylation of 1a was car-
ried out using Rh(CO)2acac catalyst precursor and PPh3 ligand to
yield the linear aldehyde 4a in high regioselectivity, traces amount
of isomerization product 10a and no cyclic products 5a (Table 1, en-
try 1). Surprisingly, the in situ addition of small amount of H3PO4
acid additive to the above reaction mixture produced the cyclic
product 5a in moderate yield but simultaneously with the disap-
pearance of any linear aldehyde product 4a, an increase in the
isomerization product 10a and less significant change in the branch
aldehyde product 3a (Table 1, entry 2). The above results gave a
clue for the formation of 5a starting from the linear aldehyde 4a.
This assumption was proved further when hydroformylation reac-
tions using the ultranox626 (Fig. 1) ligand showed similar che-
moselectivity and regioselectivity behaviors (Table 1, entries 3
and 4). The increase of the yield of linear aldehyde 4a in the ab-
sence of any acid additives for this ligand (Table 1, entry 3) allows
the formation of 5a in higher yield when a small amount of acid
was added to the reaction mixture (Table 1, entry 4). The use of dif-
ferent phosphite and diphosphine ligands does not help in increas-
ing the yield of the targeted product 5a but helps in better
understanding of the role of ligand in the hydroformylation reac-
tion (Table 1, entries 5–7). In order for the catalyst system to be ac-
Fig. 1. Structure of ultranox626.
1.85 Hz, C8-H); 13C NVR d (CDCl3): 23.1, 27.5, 56.2, 61.2, 106.7,
120.5, 121.4, 126.5, 127.1, 136.6, 143.0, 146.0; GC–MS m/z 206
(M+). Analysis calculated for C12H14O3 (206.23): C, 69.88; H, 6.84;
found: C, 69.71; H, 6.93.
3. Results and discussion
The hydroformylation of eugenol 1a (R1 = H, R2 = OH,
R3 = OCH3), adopted as a model substrate, was carried out in the
presence of Rh(CO)2acac as catalyst, different phosphine and phos-
phite ligands, and acid additives in dichloromethane as a solvent
(Scheme 1). Although the cyclic 5,6-dihydronaphthalene product
5a is the targeted product, the aldehydes 2a, 3a, 4a are also impor-
Table 2
Hydroformylation–cyclization of eugenol 1a by Rh(CO)2acac. Effect of the type of acid.a
Entry
Acid
Conversion (%)b
Yield of 10a (%)b
Yield of 5a (%)b
Yield of aldehydes (%)c
Aldehydes distribution(%)d
2a
3a
19
4a
1
–
99
99
98
99
91
97
80
0
3
6
4
0
59
65
65
54
0
96
34
29
28
26
87
41
0
0
3
4
4
0
1
0
0
81
13
10
4
2e
3
H3PO4 (85%)
H3PO4 (85%)
H3PO4 (85%)
H3PO3 (42%)
CH3COOH (99.8%)
H2SO4 (98%)
p-TsOH
84
86
92
100
14
89
0
4f
5
6
7
8
6
11
10
28
0
0
85
11
0
11
0
a
b
c
d
e
f
Reaction conditions: Rh(CO)2acac (0.005 mmol), ultranox626 (0.03 mmol), eugenol (1a) (5.0 mmol), acid (0.14 mmol), CH2Cl2 (5.0 ml), CO/H2 (100/200 psi), 2 h, 110 °C.
Determined by GC using anisole as internal standard.
Calculated yield = conversion ꢀ (10a + 5a).
Determined by GC and 1H NMR.
H3PO4 (85%) = 0.07 mmol.
H3PO4 (85%) = 0.28 mmol.
Table 3
Hydroformylation–cyclization of different allylbenzene derivatives 1a–f.a
Entry
Substrate
Conversion (%)b
Yield of 10 (%)b
Yield of 5 (%)b (%)c
Yield of aldehydesd
Aldehydes distribution (%)e
2
3
4
1
2
1a
1b
1c
1c
1d
1e
1f
98
99
99
99
99
4
7
3
6
3
65 [60]
62 [57]
38 [35]
62 [58]
76 [70]
0
29
30
58
31
20
95
63
4
5
0
4
0
1
1
86
71
44
82
93
23
19
10
24
56
14
7
3
4f
5
6
7
100
95
5
32
76
80
0
a
Reaction conditions: substrate (5.0 mmol), Rh(CO)2acac (0.005 mmol), ultranox626 (0.03 mmol), H3PO4 (85%) (0.14 mmol), CH2Cl2 (5.0 ml), 300 psi (CO/H2 = 1/2), 2 h,
110 °C.
b
Determined by GC using anisole as internal standard.
c
d
e
f
Isolated yield.
Calculated yield = conversion ꢀ (10 + 5) (%).
Determined by GC and 1H NMR.
Reaction time = 6 h.

M. Alhaffar et al. / Catalysis Communications 11 (2010) 778–782
781
R1
R1
R1
O
R2
R2
R3
R2
R3
9
1
4
Rh(CO)2acac, Ligand
CO/H2 (1:2), CH2Cl2,
110 oC, 2h
1
5
1
5
+
6
5
R3
5
4
1
H3PO4
a: R1= H, R2= OH, R3= OCH3
b:
R1= H, R2=R3= OCH3
c: R1 = H, R2,R3 = -OCH2O-
R1
OH
d
: R1=R3= OCH3, R2= OH
R2
R3
Scheme 2.
tive in forming 5a, it should be first active and selective as hydro-
formylation catalyst system toward the formation of linear alde-
hyde 4a. The low conversion obtained with diphosphine ligands
clearly supports this assumption (Table 1, entries 6 and 7).
1c, the conversion of the remained linear aldehyde 4c–5c can be
effectively done by increasing the reaction time to 6 h (Table 2, en-
try 4). The six-membered nature of the newly formed ring, to-
gether with the observation that the sum of the linear aldehyde
4 and the cyclic product 5 gives almost a constant value with re-
spect to the branch aldehyde 3, suggests that the dihydronaphtha-
lene structure comes from the linear aldehyde 4 possibly via an
intramolecular nucleophilic attack of the allylbenzene phenyl C5
carbon on the carbonyl group of the linear aldehyde. Then a tetra-
hydronaphthalene alcohol should form, which undergoes water
elimination very easily to give a double bond conjugated with
the phenyl ring (Scheme 2) [19]. The un-reactivity of 1e and 1f to-
wards the cyclization reaction (Table 3, entries 6 and 7) suggests a
great influence of substitution on phenyl ring in allylbenzene
derivative substrate in enhancing this reaction. The presence of
electron-denoting substituents on the phenyl ring gives more
nucleophilicity to the ring and hence promoting the nucleophilic
attack of the double bond on the linear aldehyde carbonyl group.
Compounds 5a–c are known dihydronaphthalene derivatives
[18,20,21], but spectral data are only reported in literature for
compound 5b [20]. Compound 5d is a new 5,6-dihydronaphtha-
lene derivative and its biological activity may be part of future
studies.
3.2. Effect of the type of acid additive
For more understanding of the role of acid in enhancing the for-
mation of the cyclic product 5a, we have carried out the hydrofor-
mylation of 1a using different acids (Table 2). Although the
presence of acid is crucial for the cyclization reaction, the amount
being added has little effect on the yield of 5a (Table 2, entries 1–
4), which suggests minor role of the acid in the hydroformylation
reaction. More isomerization product 10a and a lower yield of 5a
were obtained with phosphorous acid (Table 2, entry 5). The use
of other mineral and organic acids greatly affected the hydrofor-
mylation and cyclization reactions (Table 2, entries 6–8). Acetic
acid does not give cyclization product. p-TsOH inhibits hydrofor-
mylation reaction. Sulphuric acid mainly gives substrate, isomeri-
zation and hence the branched aldehyde 3a is the main product.
It is worth noting that the study of the effect of different tem-
peratures for the model reaction for 2 h showed a moderate con-
version of eugenol (1a) (46%) and a low yield of 5a (19%) at
temperature 80 °C, while at temperatures above 110 °C a slightly
increase in the yield (68%) of 5a was observed. The sole Rh(CO)2a-
cac, in the absence of any ligand and acid additive, was considered
as a catalyst in the hydroformylation of eugenol (1a) at room tem-
perature. Traces of products were detected after 2 h, and a very low
yield (11%) of aldehydes were obtained after 24 h with no pro-
nounced regioselectivity toward linear aldehyde (3a/4a = 1).
In general, at low and high temperatures, the product of isom-
erization of the double bond in 5a was not detected using GC–
MS and 1H NMR.
4. Conclusion
In conclusion, the synthesis of 5,6-dihydronaphthalene deriva-
tives 5a–d proceeds smoothly and efficiently in one-pot hydrofor-
mylation/cyclization of allylbenzene derivatives using the catalyst
system Rh(CO)2acac/ultranox626/H3PO4/CO/H2/CH2Cl2. The type of
ligand and acid are highly influential on the yield and chemoselec-
tivity of the reaction. The approach described here, starting from
the easily available allylbenzene derivatives 1, may become a con-
venient and competitive tool for the preparation of these
compounds.
3.3. Hydroformylation–cyclization of different allylbenzene derivatives
The hydroformylation of different allylbenzene derivatives 1a–f
(Scheme 1) were carried out by using the catalyst system
Rh(CO)2acac/ultranox626/H3PO4/CH2Cl2 (Table 3). The allylben-
zene derivatives (except for 1e and 1f) reacted efficiently to pro-
duced the corresponding 5,6-dihydronaphthalene derivatives 5a–
d in moderate to very good yields (Table 3, entries 1–5). For safrole
Acknowledgments
We thank King Fahd University of Petroleum and Minerals
(KFUPM-Saudi Arabia) for providing all support to this project. This
project has been funded by King Fahd University of Petroleum and
Minerals under Project No. IN070379.

782
M. Alhaffar et al. / Catalysis Communications 11 (2010) 778–782
[10] D.K. Rayabarapu, C.F. Chiou, C.H. Cheng, Org. Lett. 4 (2002) 1679–1682.
References
[11] M. Lautens, S. Hiebert, J.L. Renaud, Org. Lett. 2 (2000) 1971–1973.
[12] T. Hamura, M. Miyamoto, T. Matsumoto, K. Suzuki, Org. Lett. 4 (2002) 229–
232.
[13] M. Nishizawa, H. Takao, V.K. Yadav, H. Imagawa, T. Sugihara, Org. Lett. 5 (2003)
4563–4565.
[14] N. Chatani, H. Inoue, T. Ikeda, S. Murai, J. Org. Chem. 65 (2000) 4913–4918.
[15] M. Lautens, S. Hiebert, J. Am. Chem. Soc. 126 (2004) 1437–1447.
[16] M. Lautens, K. Fagnou, T. Rovis, J. Am. Chem. Soc. 122 (2000) 5650–5651.
[17] A.C. da Silva, K.C.B. de Oliveira, E.V. Gusevskaya, E.N. dos Santos, J. Mol. Catal.
A: Chem. 179 (2002) 133–141.
[18] M. Rosa Axet, S. Castillon, C. Claver, Inorg. Chim. Acta 359 (2006) 2973–
2979.
[19] R. Lazzaroni, R. Settambolo, A. Caiazzo, L. Pontorno, J. Organomet. Chem. 601
(2000) 320–323.
[20] L.F. Silva Jr., R.M.F. Sousa, H.M.C. Ferraz, A.M. Aguilar, J. Braz. Chem. Soc. 16
(2005) 1160–1173.
[21] D.C. Harrowven, M.J. Tyte, Tetrahedron Lett. 43 (2002) 5971–5972.
[1] F. Ungváry, Coord. Chem. Rev. 188 (1999) 263–296.
[2] Roberta Settambolo, Aldo Caiazzo, Raffaello Lazzaroni, Tetrahedron Lett. 42
(2001) 4045–4048.
[3] Franco Gaslini, Lucio Zion Nahum, J. Org. Chem. (1964) 1177–1180.
[4] P.K. Datta, C. Yau, T.S. Hooper, B.L. Yvon, J.L. Charlton, J. Org. Chem. 66 (2001)
8606–8611.
[5] A.J. Reynolds, A.J. Scott, C.I. Turner, M.S. Sherburn, J. Am. Chem. Soc. 125 (2003)
12108–12109.
[6] A.A. Cordi, I. Berque-Bestel, T. Persigand, J.M. Lacoste, A. Newman-Tancredi, V.
Audinot, M.J. Millan, J. Med. Chem. 44 (2001) 787–805.
[7] M. Shiraishi, Y. Aramaki, M. Seto, H. Imoto, Y. Nishikawa, N. Kanzaki, M.
Okamoto, H. Sawada, O. Nishimura, M. Baba, M. Fujino, J. Med. Chem. 43 (2000)
2049–2063.
[8] N. Asao, T. Kasahara, Y. Yamamoto, Angew. Chem., Int. Ed. 42 (2003) 3504–
3506.
[9] T. Hamura, M. Miyamoto, K. Imura, T. Matsumoto, K. Suzuki, Org. Lett. 4 (2002)
1675–1678.