Highly selective isomerization of allyloxyalcohols to cyclic acetals or 1-propenyloxyalcohols

Article abstract of DOI:10.1055/s-2004-825597

Highly selective isomerization of allyloxyalcohols either to 1-propenyl derivatives or to cyclic acetals of propanal depending on the transition metal (Ru, Rh) complex used is described together with a proposed explanation of an alternative reaction, which permits broad application of the described method.

Full text of DOI:10.1055/s-2004-825597

LETTER
1203
Highly Selective Isomerization of Allyloxyalcohols to Cyclic Acetals or
1
-Propenyloxyalcohols
a
b
b
c
H
M
ighlySelective Isom
e
a
rization of
g
Allyloxyalco
d
hols to Cyc
a
lic
A
cetals/
l
1-Prop
e
e
nyloxyalc
o
n
hols a Urbala, Nikodem Kuźnik,* Stanisław Krompiec, Józef Rzepa
a
Szczecin University of Technology, Institute of Organic Chemical Technology, Pułaskiego 10, 70-322 Szczecin, Poland
Silesian University of Technology, Faculty of Chemistry, M. Strzody 7, 44-100 Gliwice, Poland
Fax +48(32)2371205; E-mail: nikodem@zeus.polsl.gliwice.pl
b
c
Silesian University, Institute of Chemistry, Szkolna 9, 40-006 Katowice, Poland
Received 14 January 2004
The catalytic reaction of a terminal allyloxyalcohol leads
either to 1-propenyloxy alcohol or to cyclic acetal, de-
pending on the complex used as the catalyst (Scheme 1).
Abstract: Highly selective isomerization of allyloxyalcohols either
to 1-propenyl derivatives or to cyclic acetals of propanal depending
on the transition metal (Ru, Rh) complex used is described together
with a proposed explanation of an alternative reaction, which per- Several hydride ruthenium and rhodium complexes (see
mits broad application of the described method.
Table 1) catalyze the reaction towards 1-propenyloxy al-
cohols at low temperatures 60–80 °C. Most of the reac-
tions might be carried out without any solvent and the
Key words: isomerizations, cyclization, ruthenium complexes,
acetals, vinyl ethers
1
0
concentration of the catalyst is far below 1% (mol). The
dominant product is the Z isomer. Some attempts have
been made to control the E/Z ratio using
We have investigated the double-bond migration in vari-
1
1
1
2
{[RuCl (COD)] } with an addition of bulky phosphine.
ous allyl systems, including allyl ethers, sulfides, amines
2 x
3
Although only E isomer was formed, the selectivity (dou-
ble bond migration vs. cyclization) was unsatisfactory
and amides. In the case of aliphatic allyloxyalcohols
(
sometimes instead of the expected 1-propenyl products)
(
35% and 65% of the double bond migration and cycliza-
some highly symmetric species are formed, which turn
out to be cyclic 2-ethylacetals (Scheme 1).
tion products, respectively). In comparison to double-
bond migration, selective cyclization has been achieved
using a non-hydride ruthenium complex – [RuCl (PPh ) ]
2
3 3
–
at a low concentration, higher temperatures (usually 120
°
C) and without any solvents. We have also investigated
other hydroxyallyl derivatives with an aromatic core be-
tween the allyl and hydroxyl groups, like 2-allyloxyphe-
nol 7a and 2-allylophenol 8a. Double-bond migration was
successful in these cases (7b, 8b), while any attempts to
achieve cyclization failed (Scheme 2).
Scheme 1 Alternative routes of isomerization of allyloxyalcohols
1
-Propenyl ethers are of great interest among other mono-
4
mers for cationic photopolymerization. Also, simple cy-
clic acetals are widely investigated and applied in ionic
polymerization. Although, the double-bond migration
and the transformation of cyclic acetals catalyzed by
transition metal complexes have been widely investigat-
ed, only few methods of synthesizing cyclic acetals of
5
6
7
8
propanal have been published until now.
The application of transition metal complexes in the syn-
theses of cyclic acetals seems to be the most selective and
convenient method, thanks to the elimination of the sol-
vent, relatively short reaction time, and easy separation of
the products. Furthermore, double-bond migration cata-
Scheme 2 Isomerization of hydroxyallyl compounds with aromatic
core leads exclusively to double bond migration products
lyzed by ruthenium, rhodium and other transition metal In order to shed some light on this puzzling difference in
complexes is one of the most frequently used reaction to selectivity, several experiments have been designed. The
9
obtain 1-propenyl (or generally: vinyl) ethers.
first problem was to find out whether double-bond migra-
tion and cyclization are consecutive or competitive (paral-
lel) reactions (Scheme 3). 2-Allyloxyethanol (1a) was
completely isomerized to 1-propenyl derivative 1b, then
separated and subjected to the reaction with
SYNLETT 2004, No. 7, pp 1203–1206
0
3.
0
6.
2
0
0
4
Advanced online publication: 10.05.2004
DOI: 10.1055/s-2004-825597; Art ID: G01704ST
©
Georg Thieme Verlag Stuttgart · New York

1
204
M. Urbala et al.
Table 1 Results of Selective Isomerization of Allyloxyalcohols to 1-Propenyloxyalcohols or Cyclic Acetals
Cyclization Double-bond Migration
LETTER
Temp Catalyst [%mol] Product
°C];
t [h]
Substrate
Product
Catalyst [%mol]
Solv., Temp E/Z
cm / [°C];
mmol t [h]
3
[
8
3
1
2
0;
RuCl ·3H O
(0,5)
[RuClH(CO)(PPh3)3]
(0.5)
[RhH(CO)(PPh3)3]
(0.2)
–
–
80; 0.84
3
120; 0.54
1
3
2
20; [RuCl (p-Cym)]
1a
1b
2
2
1
c
(0.5)
1
2
20; [RuCl (PPh ) ]
[RuClH(CO)(PPh3)3]
(0.05)
–
–
80; 0.75
2
2
3 3
(0.1)
2
3
a
a
2b
3b
2
3
c
c
1
4
20; [RuCl (PPh ) ]
[RuClH(CO)(PPh3)3]
(0.1)
80; 0.71
2
2
3 3
(0.1)
1
3
00; [RuCl (PPh ) ]
[RuClH(CO)(PPh3)3]
(0.05)
–
80; 0.71
2
2
3 3
.5
(0.5)
[
RuClH(CO)(PPh ) ] THF
80; 0.75
3
3
3
4
5
a
a
4b
5b
(0.69)
1.38
4
5
ca
1
3
00; [RuCl (PPh ) ]
RuClH(CO)(PPh3)3]
(0.5)
–
100; 0.71
3.5
80; 0.82
2
2
3 3
.5
(0.5)
[RuClH(CO)(PPh ) ] THF
3
3
(0.83)
1.67
ca
1
1
1
1
20; [RuCl (PPh ) ]
[RuClH(CO)(PPh3)3] THF
60; 0.61
3
80; 0.43
3
2
3 3
(0.2)
(0.33)
0,5
THF
0,5
20; [RuCl (NBD)]2
[RhH(CO)(PPh3)3]
(0.37)
2
6
b
(0.94)
6a
6
cb
a
Polymeric products were also observed.
Product is a mixture of 5- and 6-membered products with their cis- and trans-isomers. Isolated yield is always higher than 85%, conversion of
b
allyloxyalcohol is always complete; selectivity higher than 98%, except for 4c (94%) and 5c (90%).
[
RuCl (PPh ) ] [which usually serves as a catalyst of the We conclude that the reactions (cyclization and double-
2 3 3
cyclization of 2-allyloxyethanol (1a) to 2-ethyldioxolane bond migration) are mainly competitive under conditions
1c)]. In accord with our expectations, under these condi- which lead to the formation of one of the isomers selec-
(
tions only a few percent of the 1-propenyl derivative 1b tively (either to cyclic acetal or to 1-propenyloxyalcohol).
underwent cyclization. However, when the reaction time Another reaction was designed to check whether any ex-
was prolonged to 10 hours it cyclized completely.
ternal hydroxyl group might be involved in forming an ac-
etal or if any other O-H addition to the C=C bond occurs
under the reaction conditions. Reaction of allyl butyl ether
9
a (or allyl benzyl ether 10a) with an equimolar amount
of benzyl alcohol 11 catalyzed by [RuCl (PPh ) ] led to
2
3 3
12
the acetal (9b and 10b) as the only product (neither allyl
nor 1-propenyl systems could be observed), see
Scheme 4. Thus, the selectivity towards the unsymmetri-
cal acetal is 100% and the yield is higher then 80% in this
method. Obtaining unsymmetrical acetals with such a
high selectivity would be difficult in any other way.
Scheme 3 Cyclization and double bond-migration as two competi-
tive reactions
Synlett 2004, No. 7, 1203–1206 © Thieme Stuttgart · New York

LETTER
Highly Selective Isomerization of Allyloxyalcohols to Cyclic Acetals/1-Propenyloxyalcohols
1205
(
4) (a) Crivello, J. V.; Jo, K. D. J. Polym. Sci., Part A: Polym.
Chem. 1993, 31, 1473. (b) Terada, Y.; Kanaoka, S.;
Higashimura, T. J. Polym. Sci., Part A: Polym. Chem. 2000,
3
8, 229.
5) (a) Kubisa, P.; Penczek, S. Prog. Polym. Sci. 1999, 24,
409. (b) Xu, Y. J.; Pan, C. Y. J. Polym. Sci., Part A: Polym.
(
1
Scheme 4 Intermolecular reaction of allyl ether and alcohol
Chem. 2000, 38, 1232. (c) Pantiru, M.; Vuluga, D. M.;
Vasilescu, D. S.; Abadie, M. J. M. Polym. Bull. 2002, 47,
4
85.
These two test reactions as well as the typical behaviour
(
(
6) McGrath, D. V.; Grubbs, R. H. Organometallics 1994, 224.
7) (a) Frauenrath, H. Synthesis 1989, 721. (b) Frauenrath, H.;
Philipps, T. Angew. Chem., Int. Ed. Engl. 1986, 25, 274.
(8) (a) Petrie, P. S. U.S. Patent 2,861,081, 1958. (b) Jones, D.
M.; Wood, N. F. J. Chem. Soc. 1965, 1560. (c) Bogatskii,
A. V.; Kamolov, G. L.; Luk’yanenko, N. G. J. Org. Chem.,
USSR 1977, 13, 467. (d) Marton, D.; Slaviero, P.;
1
3
of non-hydride complexes in double-bond migration, to-
gether with our previous findings with deuterated allyl
1
ethers, have led us to the suspicion, that depending on the
complex used as a catalyst, the mechanism of the isomer-
ization varies. While double-bond migration is most likely
achieved by adding [M]-H to C=C bond followed by b-
elimination (typical hydride mechanism), which is consis-
tent with the fact of the presence of the H ligand in the pre-
cursor complex. Cyclization products are most likely
formed via p-allyl–hydrido transient complex, which un-
dergoes a nucleophilic attack of the -OH group at the co-
ordinated p-allyl. It is supported by the fact, that ethylene
glycol monovinyl ether does not form cyclic 2-methyl-
Tagliavani, G. Gazz. Chim. Ital. 1989, 119, 359. (e) Mosca,
R.; Fagnoni, M.; Mella, M.; Albini, A. Tetrahedron 2001,
57, 10319.
(9) Krompiec, S. Zesz. Nauk. Politech. Slask., Chemia 1997, Z
136; and works cited therein.
10) Standard Procedure of Isomerization: Allyloxyalcohol
(
(scale 2–5 g), solvent (if applicable) and the catalyst (amount
given in Table 1) were put into a glass screw-capped
ampoule, purged with argon (for ca. 1 min), tightly capped
and heated in an oil bath for given period of time (Table 1).
Afterwards, solvent (if applicable) was evaporated using a
rotary evaporator. The products were separated by
distillation (760–0.1 mmHg; 1b, 2b, 1c–4c, 9b, 10b) or
column chromatography (silica gel, 200-400 mesh,
1
,3-dioxolane, because the formation of the above men-
tioned p-allyl–hydrido transient complex is not possible
for the system. The fact that ethylene glycol monovinyl
ether does not cyclize implies that hydride-alkoxy-ruthe-
nium species (which may appear in these conditions) does
not participate in forming of acetals. It is interesting to
note that long-chained allyloxyalcohols, like 6-allyloxy-
hexanol 5a and 5-allyloxypentanol 4a, also undergo selec-
tive cyclization. Even the relatively high entropy of these
alcohols does not prevent their cyclization. On the other
hand, we have shown, that allyloxyglycols with free -OH
groups (like 3-allyloxy-1,2-propandiol 6a) might be
isomerized successively to their 1-propenyloxy deriva-
tives i.e. 6b. Such reaction may be applied to protecting
and deprotecting polyhydroxy sugar systems. The superi-
ority of the described reactions over other methods is due
to the conditions and catalysts which lead to double-bond
migration or to cyclization, exclusively.
benzene–hexane 1:5; 3b–6b, 5c, 6c). (E)- and (Z)-4-(1-
1
Propenyloxy)-1-pentanols (4b): H NMR (CDCl ): d =
3
6.20 (d, 1 H, J = 12.5 Hz, CH CH=CH–), 5.93 (dt, 1 H,
3
J = 6.2 Hz, CH CH=CH–), 4.75 (dt, 1 H, J = 12.6 Hz,
CH CH=CH–), 4.36 (dt, 1 H, J = 6.8, CH CH=CH–), 3.59–
3.73 [m, 5 H, –OCH (CH ) CH OH], 1.40–1.67 [m, 9 H,
3
3
3
2
2 3
2
1
3
CH CH=, –(CH ) –]. C NMR (CDCl ): d = 145.6, 146.5
3
2 3
3
(
Z,E-CH CH=CH–), 98.5, 100.9 (Z,E-CH CH=CH), 69.0,
3
3
71.9 (=CHOCH –), 62.2 (–CH OH), 32.2 (–CH CH OH),
2 2 2 2
29.1, 29.6 (–OCH CH –), 22.2 (–CH –), 9.2, 12.6 (Z,E-
2
2
2
+
CH CH=). MS: m/e (rel. int.) = 144 (10) [M ], 87 (42), 69
3
(
100), 58 (49), 57 (29), 41 (50), 43 (7). 2-Ethyl-1,3-
1
dioxocane (4c): H NMR (CDCl ): d = 4.38 (t, 1 H, –CH–),
3.38–3.64 (m, 4 H, –OCH
3
–), 1.53–1.67 [m, 6 H, –(CH ) –],
2
2 3
1
3
1.40–1.47 (m, 2 H, CH CH –), 0.91 (t, 3 H, CH –). C NMR
3
2
3
(
(
CDCl ): d = 104.3 (–CH–), 65.4 (–OCH –), 29.7
3
2
–CH CH CH –), 26.5 (–CH CH CH –), 23.0 (CH CH –),
2
2
2
2
2
2
3
2
+
Acknowledgment
9.1 (CH
70), 69 (100), 67 (13), 57 (18), 56 (9), 41 (28), 43 (6).
11) {[RuCl (COD)] } (10 mg, 0.04 mmol) and tri(2,4,6-
3
–). MS: m/e (rel. int.) = 144 (1) [M ], 115 (67), 85
(
This work was supported by the State Committee for Scientific
Research as a research grant (No 4 T09A 132 25) for the years
(
2
x
trimethoxyphenylphosphine) (19 mg, 0.04 mmol) were
heated at 100 °C in THF (2 mL) for 3 h, then 2-allyloxy-
ethanol 1a (0.77 g, 7.6 mmol) was added and heated at 80 °C
for 3 h. Then, solvent was removed and the crude turned out
to get the mixture of cyclic product 1c and (E)-2-(1-pro-
penyloxy)ether 1b – NMR analysis.
2
003-2005.
References
(
(
(
1) Krompiec, S.; Kuźnik, N.; Bieg, T.; Adamus, B.; Majnusz,
J.; Grymel, M. Pol. J. Chem. 2000, 74, 1197.
(12) Benzyl-butyl Acetal of Propanal (9b): Allyl butyl ether 9a
(3.92 g, 34.3 mmol), benzyl alcohol 11 (3.7 g, 34.3 mmol)
and dichlorotris(triphenylphosphine)ruthenium(II) (65.8
mg, 0.069 mmol) were placed into a screw-capped ampoule.
The mixture was purged with argon flow for 1 min. Then,
after capping, the ampoule was placed into a thermostated
oil bath at 120 °C for 3 h. After the reaction the product was
2) Kuźnik, N.; Krompiec, S.; Bieg, T.; Baj, S.; Skutil, K.;
Chrobok, A. J. Organomet. Chem. 2003, 665, 167.
3) (a) Krompiec, S.; Pigulla, M.; Szczepankiewicz, W.; Bieg,
T.; Kuźnik, N.; Leszczyńska-Sejda, K.; Kubicki, M.;
Borowiak, Y. Tetrahedron Lett. 2001, 42, 7095.
(b) Krompiec, S.; Pigulla, M.; Bieg, T.; Szczepankiewicz,
1
W.; Kuźnik, N.; Krompiec, M.; Kubicki, M. J. Mol. Catal.
vacuum distilled, yield 82%, bp 98 °C (5 mmHg). H NMR
(CDCl
A: Chem. 2002, 189, 169.
3
): d = 7.38–7.22 (m, 5 H, Ph), 4.58 (dd, 2 H, J = 36.0,
1
–
1.8 Hz, PhCH O–), 4.53 [t, 1 H, J = 5.8 Hz,
OCH(CH CH )O–], 3.53 (dt, 2 H, J = 9.4, 6.5 Hz,
2
2
3
Synlett 2004, No. 7, 1203–1206 © Thieme Stuttgart · New York

1
206
M. Urbala et al.
LETTER
–
–
1
–
OCH CH CH CH ), 1.70 (qd, 2 H, J = 7.4, 5.8 Hz,
(–OCH CH CH CH ), 9.1 (–CH CH ). MS: m/e (rel. int.) =
221 (5) [M ], 193 (24), 148 (18), 108 (20), 92 (20), 91 (100),
79 (20), 57 (23).
2
2
2
3
2
2
2
3
2
3
+
CH CH ), 1.63–1.52 (m, 2 H, –OCH CH CH CH ), 1.47–
2
3
2
2
2
3
.34 (m, 2 H, –OCH CH CH CH ), 0.94 (t, 3 H, J = 7.4 Hz,
2
2
2
3
CH CH ), 0.93 (t, 3 H, J = 7.3 Hz, –OCH CH CH CH ).
(13) (a) Castiglioni, M.; Milone, L.; Osella, D.; Vaglio, G. A.;
Valle, M. Inorg. Chem. 1976, 15, 394. (b) Sherman, E. O.
Jr.; Olson, M. J. Organomet. Chem. 1979, 172, C13.
(c) Wuu, Y. M.; Zou, C.; Wrighton, M. S. Inorg. Chem.
1988, 27, 3039.
2
3
2
2
2
3
1
3
C NMR (CDCl ): d = 138.6 (Ph-C1), 128.4 (Ph-C2 and Ph-
3
C6), 127.7 (Ph-C3,Ph-C5), 127.5 (Ph-C4), 103.8
[
(
(
–OCH(CH CH )O–], 67.1 (PhCH O–), 65.2
–OCH CH CH CH ), 32.0 (–OCH CH CH CH ), 26.4
–CH CH ), 19.5 (–OCH CH CH CH ), 13.9
2 3 2
2
2
2
3
2
2
2
3
2
3
2
2
2
3
Synlett 2004, No. 7, 1203–1206 © Thieme Stuttgart · New York