Hydration of propargylic alcohols by ruthenium catalysts, with dominant anti-Markovnikov regioselectivity, formation of α,β-unsaturated products and catalytic decarbonylation to 1-alkenes

Article abstract of DOI:10.1002/ejic.200300562

Ruthenium catalysts - water-soluble ruthenium sulfophthalocyanine and heterogeneous ruthenium hydroxyapatite complexes - proved to be effective for the hydration of propargylic alcohols in entirely aqueous media. 1-Phenyl-2-propyn-1-ol underwent an unprecedented catalytic hydration- decarbonylation-dehydration reaction, giving rise to styrene and carbon monoxide; 2-propyn-1-ol and 3-butyn-2-ol gave predominantly the products of anti-Markovnikov addition, together with products of hydration-dehydration (α,β-rearrangement) and, to a minor extent, the decarbonylation products, ethene or propene, respectively. Hydrations were also conducted in D₂O, giving indications of the mechanism of the reactions and apparently ruling out the allenylidene route for the α,β- rearrangement. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejic.200300562

FULL PAPER
Hydration of Propargylic Alcohols by Ruthenium Catalysts, with Dominant
Anti-Markovnikov Regioselectivity, Formation of α,β-Unsaturated Products and
Catalytic Decarbonylation to 1-Alkenes
[
a]
[a]
[a]
[a]
Nicola d’Alessandro, Milena Di Deo, Monica Bonetti, Lucia Tonucci,
[b]
[a]
Antonino Morvillo, and Mario Bressan*
Keywords: Ruthenium / Alkynes / Phthalocyanines / Hydration / Decarbonylation
Ruthenium catalysts
—
water-soluble ruthenium sulfo-
dition, together with products of hydration-dehydration (α,β-
phthalocyanine and heterogeneous ruthenium hydroxy- rearrangement) and, to a minor extent, the decarbonylation
apatite complexes — proved to be effective for the hydration
of propargylic alcohols in entirely aqueous media. 1-Phenyl-
products, ethene or propene, respectively. Hydrations were
also conducted in D O, giving indications of the mechanism
2
2-propyn-1-ol underwent an unprecedented catalytic hydra- of the reactions and apparently ruling out the allenylidene
tion-decarbonylation-dehydration reaction, giving rise to
styrene and carbon monoxide; 2-propyn-1-ol and 3-butyn-2-
ol gave predominantly the products of anti-Markovnikov ad-
route for the α,β-rearrangement.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
gioselectivity of the hydration reactions, suggesting that ad-
dition of water to the π-adduct might follow Markovnikov’s
rule, eventually yielding methylketones, whereas addition to
the vinylidene intermediate would lead to an acyl complex
Alkynes are useful intermediates for the synthesis of car-
bonyl compounds through the well-known metal-catalyzed
hydration reaction. Typically, the addition reactions of
and, finally, aldehydes. However, only a few examples have
water to 1-alkynes are catalyzed by a variety of transition
reported good selectivity in the aldehyde formation,[
11,13]
as
metal salts and complexes,[1Ϫ7] including RuCl3,[8,9] all of
several alternative reaction pathways, not strictly dealing
with conventional hydration, could become dominant.
Among them, the catalytic cleavage of the carbon-carbon
triple bond is particularly attractive, since it could represent
an interesting topic from the synthetic point of view, if high
selectivities can be attained in mild experimental conditions.
So far, however, alkyne cleavage has also been demonstrated
which follow Markovnikov’s rule to give ketones. Only very
[
10]
recently
regioselectively oriented to the formation of aldehydes
anti-Markovnikov addition) in the presence of catalytic
has it been reported that the reaction can be
(
amounts of some ruthenium() organometallic derivatives,
containing either cyclopentadienyl[11,12] or indenyl[13] moiet-
ies, although temperatures of around 100 °C were still re-
quired. The accepted mechanism involves the rearrange-
ment of the π-adduct deriving from the conventional inter-
action with the terminal alkynes at the metal center into a
[
16,17]
to work effectively in only a few cases.
The potential for the use of both ruthenium catalysts and
water as reagent prompted our interest in the subject. In
the present work we tested the water-soluble ruthenium()
complex ruthenium sulfophthalocyanine (RuPcS), pre-
viously studied by us as an effective catalyst for a variety of
[13]
metal-vinylidene intermediate or, alternatively, a hydrido-
[
12]
alkynyl complex,
the latter of which has recently been
isolated as an intermediate in the activation of propargylic
[
18Ϫ22]
alcohols by [Cp*RuCl(dppe)];[
14]
organic substrates in aqueous media,
and the hetero-
both the proposed inter-
geneous ruthenium hydroxyapatite (RuHAP) catalyst, suc-
cessfully tested for the aerobic oxidation of alcohols to alde-
mediates are prone to nucleophilic attack by water at the α-
carbon atom. Previous results[
15]
showed the critical role
[
23]
hydes, reportedly via a metal-hydrido species, which pro-
ved to work satisfactorily also in aqueous media.[ This
paper deals with the catalytic hydration of a number of pro-
pargylic (2-propyn-1-ol, 3-butyn-2-ol, 1-phenyl-2-propyn-1-
ol and 1,1-diphenyl-2-propyn-1-ol) and non-propargylic al-
cohols (3-butyn-1-ol and 4-pentyn-1-ol). A variety of reac-
tion pathways were observed, depending upon the nature
of the substrate and the catalyst; among them, cleavage of
played by the π-alkyne/vinylidene equilibrium in the re-
24]
[
[
a]
b]
Universit a` ‘‘G. d’Annunzio’’ di Chieti-Pescara, Dipartimento
di Scienze,
Viale Pindaro 42, 65127, Pescara, Italy
E-mail: bressan@sci.unich.it
Universit a` di Padova, Dipartimento di Chimica Inorganica and
Centro C.N.R.,
Via Marzolo 1, 35100 Padova, Italy
810
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200300562
Eur. J. Inorg. Chem. 2004, 810Ϫ817

Hydration of Propargylic Alcohols by Ruthenium Catalysts
FULL PAPER
Scheme 1. Product distribution for the hydration of propargylic alcohols catalyzed by RuII derivatives
the triple bond, with formation of alkenes, was of particular Table 1. Product analysis of the hydration reaction of propargylic
alcohols by ruthenium catalysts[
a]
interest (Scheme 1).
Substrate Catalyst Conversion
Yields (mol %)
Results and Discussion
(mol %)
2
3
4
5
6
By-
products
The RuHAP and RuPcS catalysts proved to be effective,
at various selectivities, in the reaction of the terminal acety-
lenes 1-phenyl-2-propyn-1-ol, 2-propyn-1-ol, 3-butyn-2-ol
and 4-pentyn-1-ol with water; no reaction took place in the
absence of added catalyst or in the presence of calcium 1b
hydroxyapatite, whereas other simple ruthenium complexes,
b]
b]
[c]
1
a
RuPcS
73
70
9
10 10 Ϫ 22 7[
8
[
[c]
RuHAP
RuDMS
RuPW
RuPcS
RuHAP
RuPcS
12 20 Ϫ 12
8
5
[
c]
7 Ϫ Ϫ Ϫ nd
6 Ϫ Ϫ Ϫ nd
1
1
5
5
1
1
[c]
[d]
[d]
[f]
8
7 Ϫ 15 15[
9 Ϫ 12 16[
b]
b]
e]
70
65
70
95
2
4
1c
2 Ϫ 3 Ϫ 16[
such as [RuCl (DMSO) ] (RuDMS) and K [Ru(H O)-
2
4
5
2
[e]
[f]
RuHAP
2 Ϫ 5 Ϫ 20
PW O ] (RuPW), behaved poorly. Very different and com-
1
1
39
4-Pentyn-1-ol
5-hydroxy-2-pentanone
plex reaction patterns were found for each substrate exam-
ined, as reported in Table 1. The reactions were very slow
below 80 °C.
In the presence of RuHAP and RuPcS, the non-propar-
gylic alcohol 4-pentyn-1-ol was selectively transformed into
RuPcS
RuHAP
70
100
69
99
[a]
Substrate: 400 m; catalysts: RuPcS, 10 m; RuHAP, 40 mg/mL
[b]
(
78 mmol of Ru/L); water; 80 °C; 24 h reaction.
Measured as
their 1,2-dibromo derivatives, after capturing the evolved gases into
[
c]
[d]
5-hydroxy-2-pentanone, the expected product of the con- a bromine solution.
Glycolic and formic acid.
2-Propanol.
[e]
ventional Markovnikov addition to the triple bond. The re-
lated substrate 3-butyn-1-ol (not shown) was poorly con-
verted (about 10 %) during the 24 h reaction time and gave
only minor amounts of products from both the Markovni-
Measured after extraction with diethyl ether; polystyrene and
polymeric materials were also detected in not well-defined yields.
-Phenylethanol.
[
f]
1
kov and anti-Markovnikov addition, i.e. 4-hydroxy-2-but- the expected product of Markovnikov addition; Scheme 2),
anone and 4-hydroxybutanoic acid, respectively, together 3-hydroxy-propanoic acid (3a), acrylic acid (5a) and ethene
with other unidentified compounds.
(6a), together with minor amounts of glycolic and formic
The propargylic alcohols behaved differently. Large acid (Table 1), the latter likely arising from the oxidative
amounts of 2-propyn-1-ol (1a) were converted, in the pres- cleavage of the substrate. No evidence was obtained for the
ence of both RuHAP and RuPcS, into hydroxyacetone (2a, presence of 3-hydroxypropanal, which should be the prod-
Scheme 2. Ruthenium-catalyzed Markovnikov hydration of propargylic alcohols (H coming from water; H coming from acetylenic proton)
Eur. J. Inorg. Chem. 2004, 810Ϫ817
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
811

N. d’Alessandro, M. Di Deo, M. Bonetti, L. Tonucci, A. Morvillo, M. Bressan
FULL PAPER
uct of anti-Markovnikov addition to the triple bond; in- lowed by dehydration of the original OH group to 11 via
stead, its oxidized product 3-hydroxy-propanoic acid, was 10, or, alternatively, by an early dehydration of 8 to an al-
detected in reasonable amounts. To avoid this undesired lenylidene intermediate 9, followed by the hydration step.
oxidation, we tried to conduct the reaction under an inert Thus, the rearrangement of 2-propyn-1-ol was expected to
atmosphere (nitrogen), but the long times required for the produce acrolein; the observed formation of acrylic acid
formation of the products did not allow for a rigorous ex- could again be due to the rapid oxidation of the aldehydic
clusion of oxygen and, indeed, only minor differences were group, as in the case discussed previously. However, acrol-
observed with respect to the aerobic experiments, i.e. with ein, as other α,β-unsaturated aldehydes, such as benzal-
no evidence for the aldehydic product. It should be noted dehyde and cinnamaldehyde, and contrary to the saturated
here that aliphatic aldehydes, such as propionaldehyde and aldehydes discussed above, is very difficult to oxidize under
valeraldehyde, are easily oxidized to the corresponding car- these reaction conditions. It is therefore reasonable to as-
boxylic acids in the same reaction conditions and in the sume that acrylic acid is formed directly from the or-
absence of added ruthenium catalysts. The third major reac- ganometallic intermediate 11a via a complex rearrange-
tion product, acrylic acid, could be conceivably attributed ment-oxidation step. Finally, we also detected carbon mon-
to a dehydration of 3-hydroxypropanoic acid. However, oxide and an appreciable amount of ethene (6a; see Exp.
experiments conducted directly with β-propiolactone (a Sect.), which allowed a reasonable mass balance to be set
commercially available compound that rapidly hydrolyzes for the overall transformation of 1a.
to 3-hydroxypropanoic acid in aqueous media) showed only
The reaction of 3-butyn-2-ol (1b) with water closely re-
minimal conversion into acrylic acid (less than 5 %) under sembles that of 2-propyn-1-ol and leads to all the expected
these standard reaction conditions. The formation of the products from the addition of water according to Markov-
unsaturated product probably involves the well-known re- nikov (hydroxybutanone, 2b) and anti-Markovnikov rules
arrangement of propargylic alcohols to α,β-unsaturated al- (hydroxybutanoic acid, 3b); we also detected the decar-
dehydes, which is effectively catalyzed also by several ru- bonylated and dehydration/oxidation products, i.e. propene
thenium complexes.[ More recently, Wakatsuki has re- (6b) and 2-butenoic acid (5b), respectively, and carbon
25]
[26]
ported convincing evidence that the mechanism of the ru- monoxide.
thenium-catalyzed rearrangement of propargylic alcohols to
Hydration of 1-phenyl-2-propyn-1-ol (1c) reached 70 %
unsaturated aldehydes involves a series of organometallic and 95 % conversion (with RuPcS or RuHAP, respectively)
intermediates, arising either by an initial anti-Markovnikov of the starting material within 24 h (Table 1), but the reac-
hydration of the vinylidene intermediate 8 (Scheme 3), fol- tion led to minimal amounts of the product of Markovni-
Scheme 3. Proposed pathway for ruthenium-catalyzed anti-Markovnikov hydration of propargylic alcohols (H coming from water;
H coming from the acetylenic proton)
812
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 810Ϫ817

Hydration of Propargylic Alcohols by Ruthenium Catalysts
FULL PAPER
kov addition [1-hydroxy-1-phenylpropanone (2c)] and no toluene and the stable mono-carbonyl adduct
[
10]
conventional products from the anti-Markovnikov addition, [RuCl (CO)(PNP)]. Wakatsuki
has also reported decar-
2
i.e. 3-hydroxy-3-phenylpropanal or the corresponding car- bonylation (likely stoichiometric) during the hydration of
boxylic acid. In addition, the α,β-unsaturated aldehyde [cin- benzylacetylene in the presence of 10 % [RuCl (C H )-
2
6
6
namaldehyde (4c)], arising from the ruthenium-catalyzed re- {PPh (C F )}]. In the present case, decarbonylation turned
2
6 5
arrangement of the propargylic alcohol, was detected as a out to be catalytic, thus indicating lability of the putative
minor product. It should be recalled that in the reaction carbonylruthenium adducts with either RuHAP or RuPcS.
[
21]
conditions cinnamaldehyde, as with other α,β-unsaturated Indeed, RuPcS was previously reported by us to be able
aldehydes, does not undergo oxidation to the corresponding to add carbon monoxide, giving a carbonyl product pos-
Ϫ1
carboxylic acid and therefore the reported yields must be sessing a strong band at 1960 cm , but only under very
considered highly significant. Quite unexpectedly, the major severe conditions (100 atm) and after 24 h reaction.
product formed during the reaction was styrene (6c), which
was detected after extraction of the reaction mixtures with Hydration Reactions in D
diethyl ether (16Ϫ20 %; Table 1), together with carbon
O
2
Deuterium distribution in the products was evaluated by
H NMR spectroscopy and GC-MS measurements on the
reaction mixtures from 2-propyn-1-ol (1a) and 1-phenyl-2-
propyn-1-ol (1c). It has to be noted that acetylenic-proton
monoxide and large (but not quantified) amounts of styr-
ene-based polymeric materials, likely accounting for the fi-
nal mass balance. It should be noted that in the cases of
ethene and propene discussed previously, no water-insoluble
polymeric materials were detected. However, ethene and
propene, unlike styrene, accumulate in the gas phase and
therefore are expected to escape from the reaction mixture
before they can be polymerised. Therefore the low mass-
balance values for both 1a and 1b (Table 1) were attributed
to some further unidentified products.
No reaction was observed for the other propargylic al-
cohol, 1,1-diphenyl-2-propyn-1-ol, when suspended in
water as it is only slightly soluble. However, when the hy-
dration was carried out in an organic solvent (toluene) with
small amounts of added water and in the presence of the
lipophilic ruthenium phthalocyanine derivative RuPc, in-
stead of the water-soluble RuPcS, small amounts of 1,1-
diphenylethene and of β-phenylcinnamaldehyde were
formed, thus indicating a reactivity closely related to that
of 1-phenyl-2-propyn-1-ol in the water/RuPcS system.
A simple explanation for the formation of C(nϪ1) alkenes
1
exchange of the pure substrates in 99.9 % D O turned out
2
to be slow even at 80 °C, with 100 % deuteration being re-
ached only after 4 h (Figure 1). The rate of deuterium ex-
change in the presence of the RuPcS catalyst was similar
for 1-phenyl-2-propyn-1-ol, at least at early reaction times,
whereas in the case of 2-propyn-1-ol deuteration was much
slower and never exceeded 40 % substitution, even after
long reaction times (Figure 1). The observed different deut-
eration rates of the acetylenic proton must be compared
with the hydration rates of the corresponding substrates
(
Figure 1), clearly indicating that 2-propyn-1-ol (1a) under-
went hydration in D O mainly as the protonated RCϵCH
2
(
styrene, propene and ethene) from the corresponding pro-
pargylic alcohols (1-phenyl-2-propyn-1-ol, 3-butyn-2-ol and
-propyn-1-ol, respectively) might involve decarbonylation
of the anti-Markovnikov acyl-metal intermediate 10 to
-substituted ethanol derivatives followed by a dehydration
2
1
step (Scheme 3). 2-Propanol and 1-phenylethanol were
detected in trace amounts in the case of 1-phenyl-2-propyn-
1
-ol and 3-butyn-2-ol, respectively. However, the fact that
direct experiments conducted in the presence of ethanol,
-propanol and 1-phenylethanol under the same reaction
2
conditions showed minimal (or no) dehydration to ethene,
propene and styrene (less than 2 % in the most favorable
case, namely 1-phenylethanol), indicates that the alkenes are
formed directly from an organometallic intermediate (likely
11), in competition with the 1-substituted ethanol deriva-
tives. The formation of alkenes also implies formation of
carbon monoxide, which was always detected among the
products of the hydration reactions. Decarbonylation of
terminal acetylenes by hydration is indeed a known process,
[
27]
first reported by Bruce
and thoroughly investigated by
[
28]
Figure 1. Incorporation of deuterium into 2-propyn-1-ol (1a; open
squares and triangles, dotted lines) and 1-phenyl-2-propyn-1-ol (1c;
filled squares and triangles, solid lines); % conversions of 1a (open
Bianchini in the case of phenylacetylene, which, in
the presence of water and of the Ru
II
complex
[
RuCl (PNP)(PPh )], was quantitatively transformed into circles, dotted line) and 1c (filled circles, solid line) are also reported
2 3
Eur. J. Inorg. Chem. 2004, 810Ϫ817
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
813

N. d’Alessandro, M. Di Deo, M. Bonetti, L. Tonucci, A. Morvillo, M. Bressan
FULL PAPER
species, while 1-phenyl-2-propyn-1-ol (1c) reacts as the de- deuterium atoms in the α-position; the formula PhϪCHϭ
uterated RCϵCD species.
The rapid and almost complete deuteration of the acety-
CD was fully confirmed by the GC-MS data.
Because of the slow deuterium exchange of the acetylenic
2
lene proton of 1-phenyl-2-propyn-1-ol unfortunately pre- proton, a selective deuterium distribution in the products
cluded any significant conclusion about the possible mecha- was expected for 2-propyn-1-ol. Hydroxyacetone (2a, the
nism of the hydration of this substrate (Scheme 2 and 3). product of Markovnikov addition) exhibits the COCHH2
1
1
The H NMR spectrum of the product of Markovnikov ad- (see Scheme 2) singlet in the H NMR spectra considerably
dition — 1-hydroxy-1-phenylacetone (2c) — shows com- broadened and reduced in intensity, when compared with
plete disappearance of the singlet arising from the methyl that of the singlet of the CH OH protons, but indicative of
2
protons, in agreement with the dominant presence of the ca. 70 % deuteration (Figure 2). The mass spectrum shows
ϩ
COϪCD3 fragment (46 amu) in the mass spectrum; the the peak of the molecular ion shifted by ϩ3 amu and that
ϩ
ϩ
molecular ion CH COCH(Ph)OH is almost undetectable. of the COCH3 fragment by ϩ2 amu (45 instead of 43
3
1
The H NMR spectrum of cinnamaldehyde (4c) lacks the amu), in agreement with the dominant presence of the
signals of the aldehydic proton (doublet) and that of the CD HϪCOϪCH OD species. The dideuteration of the
2
2
proton bonded to the α-carbon, whereas the signals of the methyl group in the formed methylketone confirms the slow
PhCHϭCHϪCHO proton are covered by those of the phe- exchange of the terminal CH, since fast exchange would
nyl protons. MS measurements clearly indicated that have yielded trideuteration, as found in 1-hydroxy-1-phenyl-
1
PhCHϭCDϪCDO was the dominant species. The
H
acetone. The anti-Markovnikov addition of D O to
2
2
NMR spectrum of styrene (6c) lacks the multiplets arising RCϵCH probably involves the interconversion of an η -
from the terminal methylene group, whereas the multiplet alkyne species to a vinylidene, therefore bringing the acetyl-
of the PhCHϭ proton collapses into a very broad signal, enic hydrogen onto the carbon substituted with the R group
strongly indicative of an effective coupling with the two and eventually giving RCHDϪCDO (or RCHDϪCOOD;
1
Figure 2. H NMR spectra of the crude mixture of the hydration reaction of 2-propyn-1-ol (24 h, 80 °C) in H
2
O (A) or in D
2
O (B); the
product numbers are the same as in Scheme 1
814
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 810Ϫ817

Hydration of Propargylic Alcohols by Ruthenium Catalysts
FULL PAPER
Scheme 3), unless a rapid proton exchange is taking place
on the vinylidene intermediate. We found that 3-hydroxy-
propanoic acid (3a) exhibits a distinctive deuteration at the
1
methylene group next to the carbonyl carbon, with the H
NMR spectra showing a collapse of the two diagnostic trip-
lets of the CH signals at δ ϭ 3.75 and 2.50 ppm (J ϭ 6 Hz)
2
into a broad singlet and a triplet, respectively, in a roughly
2.5:1 intensity ratio (Figure 2) and therefore best fitting
with an HOCH ϪCHDϪCOOH species. No MS measure-
2
ments were possible, since 3-hydroxypropanoic acid was not
detectable in the GC. Acrylic acid (5a), arising from the
hydration-dehydration reaction, shows the presence of both
the CH ϭCDϪCOOD (m/z ϭ 73) and the CH ϭ
2
2
CHϪCOOD species (m/z ϭ 72), roughly in a 1:1 molar ra-
1
tio. Accordingly, the H NMR signals of the hydrogen atom
bonded to the carbon atom next to the carboxylic function
appear clearly weakened with respect to the signals of the
methylene protons (ca. 50 % in intensity; Figure 2). Also
ethene (6a), the decarbonylation product, exhibits only par-
tial deuteration. The MS spectra of its dibromo adduct
1
Figure 3. Differences in the H NMR chemical shift for 1-phenyl-
CH BrϪCH Br shows an intense fragment arising from the 2-propyn-1-ol in the presence of MPcS’s (10 m), relative to the
2
2
pure substrate: CH ϭ acetylenic proton; CHOH ϭ aliphatic pro-
ton; Ph ϭ ortho aromatic protons; Ph ϭ meta and para aro-
matic protons.
loss of one Br atom as a series of peaks at 108 and 110 amu
2
3
7
9
81
(
Br) and at 109 and 111 amu ( Br), in an intensity ratio
of ca. 1:4, therefore suggesting the presence of both the di-
ϩ
ϩ
deuterated C H D Br and monodeuterated C H DBr
2
2
2
2
3
ions. The latter two findings have some significance, since (Ϯ0.01 ppm), including the acetylenic one (Figure 3). For
they apparently rule out the putative allenylidene route (i.e. comparison purposes, we investigated other MPcS com-
8a Ǟ 9a Ǟ 11a, Scheme 3), for which complete deuteration plexes (M ϭ Fe, Co and Ni), all of them catalytically inac-
of the carbon atom next to the metal-acyl function in 11a tive for the reaction under investigation; for the 1-phenyl-
1
was expected, thus eventually yielding exclusively the CH ϭ 2-propyn-1-ol/NiPcS system significant changes in the H
2
CDϪCOOH and CD ϭCH species; the vinylidene route NMR spectra were detected by decreasing the substrate-to-
2
2
(i.e. 8a Ǟ 10a Ǟ 11a, Scheme 3) is not contradictory with metal ratios down to 1 to 1, with pronounced upfield shifts
the observed formation of partially deuterated products.
(up to 0.25 ppm) of the aromatic (Ph and Ph ) and ali-
2 3
phatic (CHOH) protons and less pronounced, but still sig-
nificant, shifts of the acetylene proton (up to 0.15 ppm; Fig-
ure 3). This finding thus strongly suggests a substrate-metal
interaction also involving the triple bond.
Interaction of the Substrates with the Metal Complexes
The nature of the catalyst apparently influences the rate
of the hydration more than it does the regioselectivity, Conclusion
which is strongly dependent upon the nature of the sub-
strate. Since an attractive explanation of the observed be-
The first noteworthy finding in this paper is the different
II
havior might be the different degree of stabilization of the reactivity exhibited by the Ru derivatives in aqueous solu-
π-adduct 7 in Scheme 2 vs. the metal-vinylidene intermedi- tion. Simple ruthenium complexes, like RuDMS and RuPW,
ate 8 in Scheme 3, induced by the various substrates, we gave almost exclusively products according to the Markov-
carefully analyzed an aqueous solution of RuPcS (10 m) nikov rule and in very low yields (at least with 2-propyn-1-
in the presence of increasing amounts of the substrates un- ol), whereas, when the metal is strongly complexed, i.e. with
der investigation, and also, for comparison purposes, of a phthalocyanine ring or a hydroxyapatite support, the re-
other saturated and unsaturated alcohols (1-butanol, 2-pro- activity was definitely enhanced. The regioselectivity was
pen-1-ol, ethanol). By increasing the substrate-to-metal ra- found to be strongly dependent upon the nature of the alky-
tios up to 20-fold, an enhancement of the measured acidity nol; in particular, only when the OH group is relatively
of the mixtures was generally observed, in some cases very close to the triple bond (propargylic alcohols) does the at-
sharp, but always irrespective of the nature of the sub- tack of water at the triple bond follow dominantly, if not
strates, and therefore hardly attributable to any significant exclusively, an anti Markovnikov pathway (in a broad
1
metal/substrate interaction. Accordingly, H NMR meas- sense), thus leading to aldehydic (or carboxylic acid) prod-
urements of the propargylic alcohols/RuPcS systems in ucts, α,β-rearrangement of the substrate and/or decar-
buffered solutions (with pHs varying between 0.5 and 9) bonylation. The selectivities in each of the above ‘‘anti-Mar-
exhibited only tiny and erratic shifts of the proton signals kovnikov’’ products were however moderate and far lower
Eur. J. Inorg. Chem. 2004, 810Ϫ817
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
815

N. d’Alessandro, M. Di Deo, M. Bonetti, L. Tonucci, A. Morvillo, M. Bressan
FULL PAPER
than those reported with other ruthenium catalysts.[10,13] for 24 h; the reaction mixtures were analyzed as above; (iii) 1,1-
diphenyl-2-propyn-1-ol (400 m) and RuPc (10 m) in a water-
saturated toluene solution were stirred at 80 °C for 24 h and di-
rectly analyzed by NMR spectroscopy and GC.
We have no data to support a real participation of the OH
group in the π-adduct/vinylidene species equilibrium, but
the fact that definite evidence of a metalϪligand interaction
was obtained only for the catalytically inactive NiPcS spec-
ies, suggests that the driving force of the RuPcS hydrations pyn-1-ol (400 m) and RuPcS (10 m) or RuHAP (13.9 mg,
O (D Ն 99.9 %) were reacted for 24 h at 80 °C.
2
Hydration Reactions with D O: 2-Propyn-1-ol or 1-phenyl-2-pro-
is essentially kinetic. The other important findings deal with 0.010 mmol) in D
2
the observed smooth decarbonylation of the propargylic al- Hydroxyacetone, acrylic acid, 1-hydroxy-1-phenylacetone, cinnam-
aldehyde and styrene were detected by GC, GC-MS and NMR
cohols to corresponding C(nϪ1) alkenes, which represents, to
spectroscopy, either directly or, when necessary, after extraction
with CDCl from the reaction mixture; 3-hydroxypropanoic was
3
directly analyzed only by NMR spectroscopy.
the best of our knowledge, the first catalytic decar-
bonylation process of terminal alkynes.
Detection of CO, Ethene and Propene: CO was captured by bub-
Experimental Section
bling the evolved reaction gas through a CH
RuCl(dppp) ]PF ; the amount of formed mono-carbonyl adduct
[RuCl(CO)(dppp) ]PF was measured both by UV/Vis and IR
2 2
Cl solution of
[
2
6
[
21]
Materials: Ruthenium tetrasulfophthalocyanine (RuPcS), cobalt
2
6
(CoPcS),[
29]
analysis, by following the disappearance of the 460 nm band of the
original complex or the appearance of the distinctive ν absorp-
tetrasulfophthalocyanine
2 4
cis-[RuCl (DMSO) ]
[
30]
39] (RuPW)[31] and ruthenium
(RuDMS),
K
5
[Ru(H
2
O)PW11
O
CO
Ϫ1
[32]
tion at 1930 cm of the carbonyl adduct.
Although both UV
hydroxyapatite (RuHAP; Ru 17 %, corresponding to 1.6 mmol/
[
23]
and IR measurements might allow, in principle, the quantitation of
the carbonyl adduct, and hence of the carbon monoxide produced,
the quantitative results were erratic, probably due to the long reac-
tion times required for the formation of the product. Ethene and
g) were prepared by published procedures. Ruthenium phthalo-
cyanine (RuPc) was prepared by the procedure of Weber and
[
29]
Bush,
starting from ruthenium trichloride, phthalic acid and
urea. Other metal complexes, organic substrates, chromatographic
and spectroscopic standards were purchased from Aldrich. D O (D
2
Ն 99 %) was obtained from Isotec Inc.
2 2
propene were captured in a CH Cl solution containing an excess
of Br ; the resulting 1,2-bromoalkane solutions were analyzed by
2
GC, GC-MS and NMR spectroscopy, by comparing the results
with those of the two commercial compounds (1,2-dibromoethane
and 1,2-dibromopropane). Further qualitative data on the pro-
duction of propene were obtained by headspace GC-MS analysis
of the final reaction mixtures.
Apparatus: Organic analyses by GC were performed on aliquots
withdrawn with a microsyringe from the aqueous reaction mixtures
and diluted 1:20 with diethyl ether, on a HP 6890 GLC instrument
equipped with FID, using a 30-m HP-5 (0.32 mm i.d.; 0.25 µm
film thickness) or a 30 m HP-Innowax (0.25 mm i.d.; 0.25 µm film
thickness) capillary column with the injection port kept at 250 °C
(carrier gas: He). The identity of each product was confirmed by
comparison of the fragmentation pattern in the mass spectrum ob-
Acknowledgments
The authors thank Mr. Antonio Ravazzolo, C.N.R., Padova, for
technical assistance.
tained with an MD 800 Fisons mass spectrometer operating in the
electron ionization mode at 70 eV. 1H NMR measurements were
performed on a Bruker Avance 300 MHz spectrometer equipped
2
with a BBO 5 mm probe, by adding small amount of D O to the
reaction mixtures (1:3); water suppression was carried out by a pre-
saturation sequence using a composite pulse (zgcppr Bruker se-
quence). A co-axial capillary tube containing a 30 m solution of
[1] [1a]
R. C. Laroci, W. W. Leong, in Comprehensive Organic Syn-
3
-(trimethylsilyl)propionic-2,2,3,3-d
4
acid sodium salt (TSP) in
thesis (Eds.: B. M. Trost, I. Fleming, M. F. Semmelhoci), Perga-
mon Press, Oxford, UK, 1991, vol. 4, 269. [ J. March, Ad-
vanced Organic Chemistry, John Wiley & Sons, New York, US,
1b]
water (D O) was used as reference. The identity of each product
2
and their quantitation were confirmed by comparison of the posi-
tion and intensity of suitable signals, by addition of measured
amounts of pure compounds to the reaction mixtures.
1
992, 762.
[
[
2]
3]
Y. Fukuda, I. Utimoto, J. Org. Chem. 1991, 56, 3729Ϫ3731.
J. Blum, H. Huminer, H. Alper, J. Mol. Catal. 1992, 75,
1
53Ϫ160.
J. W. Hartman, W. C. Hiscox, P. W. Jennings, J. Org. Chem.
993, 58, 7613Ϫ7614.
Typical Procedure for the Hydration Reactions: The progress of the
reactions was monitored by GC-MS and/or H NMR spec-
troscopy; conversions, yields and rates were reproducible to within
1
the hydration reaction: (i) 2-propyn-1-ol, 3-butyn-2-ol, 1-phenyl-2-
propyn-1-ol, 4-pentyn-1-ol, or 3-butyn-1-ol (equivalent to 400 m
of final solution) was added to an aqueous solution of RuPcS
[
4]
1
1
[5]
[6]
I. I. Meier, J. A. Marsella, J. Mol. Catal. 1993, 78, 31Ϫ42.
J. P. Damiano, M. Postel, J. Organomet. Chem. 1996, 522,
0Ϫ15 %. We describe three different experimental procedures for
303Ϫ305.
[
[
[
7]
8]
9]
W. Baidossi, M. Lahav, J. Blum, J. Org. Chem. 1997, 62,
669Ϫ672.
J. Halpern, B. R. James, A. L. W. Kemp, J. Am. Chem. Soc.
(10 m) in a 5 mL vial in air or under a nitrogen atmosphere. The
1
966, 88, 5142Ϫ5146.
T. M. M. Khan, S. B. Halligudi, S. Shuila, J. Mol. Catal. 1990,
8, 299Ϫ305.
mixture was stirred for 24 h at 80 °C and analyzed by GC-MS and/
or NMR spectroscopy. In the case of 1-phenyl-2-propyn-1-ol the
reaction mixture was also treated with CDCl (5 mL) and the ex-
3
tract analyzed by GC-MS and NMR spectroscopy; after the evap-
oration of the solvent, a polymeric film was obtained, which on the
basis of IR analysis was shown to contain also polystyrene and/or
styrene-based polymeric materials; (ii) the above substrates
5
[10]
M. Tokunaga, Y. Wakatsuki, Angew. Chem. Int. Ed. 1998, 37,
2867Ϫ2869.
[11]
T. Suzuki, M. Tokunaga, Y. Wakatsuki, Org. Lett. 2001, 3,
735Ϫ737.
[12]
M. Tokunaga, T. Suzuki, N. Koga, T. Fukushima, A. Horiuchi,
(400 m) and RuHAP (78 mg/mL) in water were stirred at 80 °C
Y. Wakatsuki, J. Am. Chem. Soc. 2001, 123, 11917Ϫ11924.
816
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 810Ϫ817

Hydration of Propargylic Alcohols by Ruthenium Catalysts
FULL PAPER
M. Bressan, N. d’Alessandro, L. Liberatore, A. Morvillo, Co-
ord. Chem. Rev. 1999, 185Ϫ186, 385Ϫ402.
K. Yamaguchi, K. Mori, T. Mizugaki, K. Ebitani, K. Kaneda,
J. Am. Chem. Soc. 2000, 122, 7144Ϫ7145.
K. Mori, I. Yamaguchi, T. Mizugaki, K. Ebitani, K. Kaneda,
Chem. Commun. 2001, 461Ϫ462.
M. P. Picquet, A. Fernandez, C. Bruneau, P. H. Dixneuf, Eur.
J. Org. Chem. 2000, 2361Ϫ2366, and references cited therein.
T. Suzuki, M. Tokunaga, Y. Wakatsuki, Tetrahedron Lett. 2002,
43, 7531Ϫ7533.
[
[
[
13]
14]
15]
[22]
P. Alvarez, M. Bassetti, J. Gimeno, G. Mancini, Tetrahedron
Lett. 2001, 42, 8467Ϫ8470.
[
[
23]
24]
E. Bustelo, M. Jimenez-Tenorio, M. C. Puerta, P. Valerga, Eur.
J. Inorg. Chem. 2001, 2391Ϫ2398.
M. P. Gamasa, J. Gimeno, B. M. Martin-Vaca, J. Borge, S.
Garcia-Granda, E. Perez-Carreno, Organometallics 1994, 13,
[25]
[26]
4045Ϫ4057.
[
[
[
[
[
[
16]
17]
18]
19]
20]
21]
S. Datta, C.-L. Chang, K.-L. Yeh, R.-S. Liu, J. Am. Chem. Soc.
003, 125, 9294Ϫ9295.
2
[
[
27]
28]
M. I. Bruce, A. G. Swincer, Aust. J. Chem. 1980, 33,
C.-H. Jun, H. Lee, C. W. Moon, H.-S. Hong, J. Am. Chem.
Soc. 2001, 123, 8600Ϫ8601.
1
471Ϫ1483.
C. Bianchini, J. A. Casares, M. Peruzzini, A. Romerosa, F.
Zanobini, J. Am. Chem. Soc. 1996, 118, 4585Ϫ4594.
J. H. Weber, D. H. Busch, Inorg. Chem. 1965, 4, 469Ϫ471.
I. P. Evans, A. Spencer, G. Wilkinson, J. Chem. Soc., Dalton
Trans. 1973, 204Ϫ209.
N. d’Alessandro, L. Tonucci, M. Bressan, L. K. Dragani, A.
Morvillo, Eur. J. Inorg. Chem. 2003, 1807Ϫ1814.
[
[
29]
30]
N. d’Alessandro, L. Liberatore, L. Tonucci, A. Morvillo, M.
Bressan, New J. Chem. 2001, 25, 1319Ϫ1324.
[31]
M. Bressan, L. Forti, F. Ghelfi, A. Morvillo, J. Mol. Catal.
N. d’Alessandro, L. Liberatore, L. Tonucci, A. Morvillo, M.
Bressan, J. Mol. Catal. 2001, 175, 83Ϫ90.
1993, 79, 85Ϫ93.
[32]
M. Bressan, P. Rigo, Inorg. Chem. 1975, 14, 2286Ϫ2288.
Received August 12, 2003
M. Bressan, N. Celli, N. d’Alessandro, L. Liberatore, A. Morv-
illo, L. Tonucci, J. Organomet. Chem. 2000, 593Ϫ594,
Early View Article
Published Online January 2, 2004
416Ϫ420.
Eur. J. Inorg. Chem. 2004, 810Ϫ817
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
817