Regio- And stereoselective intermolecular hydroalkoxylation of alkynes catalysed by cationic gold(I) complexes

Article abstract of DOI:10.1002/adsc.201000094

Vinyl ethers and ketals are obtained from the reaction of phenylacetylene derivatives and dimethyl acetylenedicarboxylate (DMAD) with alcohols in good yields and levels of stereoselectivity by using cationic gold(I)-phosphine complexes as catalysts. By choosing the appropriate phosphine, the selective formation of the Z or the E isomer of the vinyl ether can be tuned, and the undesired formation of the ketal can be controlled. The isomerisation of fumarates (Z-isomer) to maleates (E-isomer) is a gold-catalysed process that can be conducted in onepot. When using polyols, 5-membered cyclic ketals are easily isolated by extraction with hexane and the gold complex can be reused.

Full text of DOI:10.1002/adsc.201000094

FULL PAPERS
DOI: 10.1002/adsc.201000094
Regio- and Stereoselective Intermolecular Hydroalkoxylation of
Alkynes Catalysed by Cationic Gold(I) Complexes
Avelino Corma,a,* Violeta R. Ruiz,a Antonio Leyva-Pꢀrez,a and Marꢁa J. Sabatera
a
Instituto de Tecnologꢁa Quꢁmica, Universidad Politꢀcnica de Valencia-Consejo Superior de Investigaciones Cientꢁficas,
Avda. de los Naranjos s/n, 46022 Valencia, Spain
Fax : (+34)-9-638-77809; phone: (+34)-9-638-77800; e-mail: acorma@itq.upv.es
Received: February 4, 2010; Revised: May 31, 2010; Published online: July 2, 2010
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.201000094.
Abstract: Vinyl ethers and ketals are obtained from of fumarates (Z-isomer) to maleates (E-isomer) is a
the reaction of phenylacetylene derivatives and di- gold-catalysed process that can be conducted in one-
methyl acetylenedicarboxylate (DMAD) with alco- pot. When using polyols, 5-membered cyclic ketals
hols in good yields and levels of stereoselectivity by are easily isolated by extraction with hexane and the
using cationic gold(I)-phosphine complexes as cata- gold complex can be reused.
lysts. By choosing the appropriate phosphine, the se-
lective formation of the Z or the E isomer of the
vinyl ether can be tuned, and the undesired forma- Keywords: alkynes; calalyst recovery; catalysis; enol
tion of the ketal can be controlled. The isomerisation ethers; gold; hydroalkoxylation
Introduction
ketals,[1a] and the intermolecular addition to acetylene
has industrial applications.[1b] These reactions are tra-
The addition of alcohols to alkynes, the so-called hy- ditionally performed under metal-catalysed conditions
droalkoxylation (Figure 1), is a 100% atom-economic since, otherwise, the addition is extremely slow or the
reaction that gives direct access to vinyl ethers and conditions are too harsh.
Figure 1. Intramolecular (A) and intermolecular (B) hydroalkoxylation of alkynes (top); gold(I) complexes used as catalysts
in this work (bottom).
Adv. Synth. Catal. 2010, 352, 1701 – 1710
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1701

Avelino Corma et al.
FULL PAPERS
The intramolecular hydroalkoxylation is clearly fav- The monoaddition mode on terminal and internal al-
oured over the intermolecular version, and many dif- kynes is in most cases strongly favoured, so vinyl
ferent metals have been reported as efficient catalysts ethers are selectively formed. In the case of alkyl pro-
for intramolecular hydroalkoxylations under mild piolates, the stereoselectivity can be switched by
conditions.[1a,2] However, this is not the case for inter- choosing the appropriate conditions, and the corre-
molecular hydroalkoxylations. Indeed, although dif- sponding maleates can be efficiently obtained for the
ferent metals,[3] including palladium,[4,5] platinum,[6] first time by a metal-catalysed hydroalkoxylation. In
mercury,[7] ruthenium,[8] copper,[9] silver,[10] zinc[11] and addition, these gold(I) complexes are highly insoluble
gold[12] have been used as catalysts, none of them ful- in hexane, which enables quantitative recovery by
fill the levels of efficiency, mildness and selectivity precipitation after the reaction and efficient reuse.
achieved for the intramolecular version. Therefore, to
find selective, efficient and environmentally friend-
ly[13] procedures for the intermolecular hydroalkoxyla- Results and Discussion
tion of alkynes is of huge interest. In such a case, a
successful procedure should accomplish the following: Catalyst Screening and Substrate Scope
(i) selective monoaddition of the alcohol to the
alkyne and (ii) stereocontrol on the final vinyl ether To catalyse the hydroalkoxylation of alkynes, gold(I)
(for internal alkynes). Concerning the first issue, the complexes containing various electron-withdrawing
literature shows that controlling a monoaddition (EWD, complexes 1a and 1b) and electron-donating
mode is difficult as alkynes generally prefer a two- phosphines (ED, complexes 1c–1e) have been pre-
fold addition mode, affording ketals instead of vinyl pared (see Figure 1 and Experimental Section), and
ethers.[11,12,14] Alkyl propiolates make an exception to reactivity results using phenylacetylene 2 as the test
this trend, as their resulting vinyl enol ethers are alkyne and 2 mol% catalyst loadings are given in
stable enough to avoid a second alcohol addition. Table 1.[23]
Nevertheless, two isomers, i.e., E: maleates and Z: fu-
It can be seen there that, when 2 was reacted with
marates can be produced and, consequently, the addi- n-butanol 3, gold(I) complexes 1a and 1b give a high
tion of alcohols to alkyl propiolates is a convenient ratio of enol ether 4 to ketal 5 (entries 1 and 2), show-
reaction to test for catalyst stereoselectivity. Different ing that a nearly complete selectivity to the monoad-
metal catalysts have been reported to perform the hy- dition product is achieved when using complexes con-
droalkoxylation of alkyl propiolates, such as lead,[15] taining the EWD phosphines. Complexes containing
palladium,[5,16] silver,[10] iridium[17] and molybdenum.[18] the ED phosphines 1c–e were also highly active cata-
However, fumarates were always the major or exclu- lysts which, conversely to their more electrophilic
sive products, coming from a trans-addition of the al- counterparts 1a and 1b, led predominantly to the
cohol across the triple bond. The only scarce exam- ketal product 5 (entries 3–5). Although it is true that
ples where maleates were the major products corre- catalyst 1d is more active than 1a (compare entries 9
spond to a stoichiometric addition of metal alkoxide and 10), the ketalisation takes place with the former
complexes (metal: Mn, Mo, Re)[19] to DMAD and at a higher rate than the hydroxyalkoxylation does,
two amine-catalysed additions.[20]
since 4 and 5 are formed concominantly even when
Given the greater difficulty to accomplish the inter- using 1 equivalent of 3. Therefore, the high selectivity
molecular process relative to its intramolecular ver- achieved by catalysts 1a and b for the enol ether 4 is
sion, stronger Lewis acids are required to improve the remarkable and does not come from a kinetic control.
reaction rate. This hypothesis is clearly corroborated A catalyst with lower Lewis acidity, such as the
by listing the different catalysts found in previous re- gold(I) complex having ClÀ as counteranion instead
À
ports for intermolecular hydroalkoxylations, that in- of the low-coordinating NTf2 , gave no conversion at
volve highly acidic metal salts containing a weakly-co- all (entry 6). The presence of the phosphine ligand in
À [5]
ordinating counteranion such as OTfÀ,[6,9,10] BF4 , or the catalyst is necessary (entry 7). Importantly, it has
À [4,6]
PF6 ,
either as the pure compound or formed in- to be pointed out that the reaction is not Brønsted
situ from a pre-catalyst. Taking this into account, we acid-catalysed (entry 8).
thought that phosphine-gold(I) complexes bearing the
low coordinating group NTf2 (Figure 1), which are by surveying an array of alkynes (Table 2). Both ter-
The preliminary scope of the reaction was explored
À
cationic and highly acidic isolable compounds that minal (entries 1 and 2) and internal (entries 3–5) aro-
can be used as catalysts in different processes involv- matic alkynes reacted well in this Au(I)-catalysed hy-
ing alkyne activation,[21,22] could be an active and se- droalkoxylation, although variable and non-negligible
lective catalyst for performing the hydroalkoxylation amounts of ketals and ketones were produced along
of alkynes. In the present work, we will show that with the desired vinyl enol ethers. Moreover, terminal
these phosphine-gold(I) complexes are indeed regio- alkyl alkynes also reacted well, although ketals were
and stereoselective catalysts for this transformation. the main product (entries 6 and 7, see also footnote
1702
asc.wiley-vch.de
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1701 – 1710

Regio- and Stereoselective Intermolecular Hydroalkoxylation of Alkynes
Table 1. Hydroalkoxylation of phenylacetylene 2 with n-butanol 3 catalysed by AuPR3NTf2 complexes 1a–e and others.[a]
Entry
3 [equiv.]
Catalyst
Conversion [%][b]
Ratio of 4/5 [%][c]
1
2
3
4
5
6
7
8
9
2
1a
1b
1c
1d
98 [69]
96
>99
98
97
0
<1
<1
62
88/5
92/2
10/88
34/63
28/69
–
1e
AuPPh3[Cd]l
AuNTf2
HNTf2
1a
–
–
1
>99/1
50/50
10
1d[e]
74
[a]
Reaction conditions: alkyne (1 mmol), alcohol (1–2 mmol), catalyst, dry solvent (0.5 mL), stirring at room temperature
for 24 h.
[b]
[c]
Average of two runs, calculated by GC; in brackets, isolated yield.
Average of two runs; remaining to fit conversion corresponds to ketones from hydrolysis and/or direct hydration of the
alkyne.
[d]
[e]
Generated in-situ by stirring AuCl+AgNTf2 for 10 min.
1 mol%.
Table 2. Scope of the hydroalkoxylation of different alkynes and alcohols catalysed by complex 1a or 1d.
Entry R1
R2
Au Catalyst (mol%) R3OH (equiv.) Conversion [%][a] Product Distribution A (Z/E):B:C [%][a]
1
2
3
4
5
6
7
8
9
10
PhOC6H4
ClC6H4
Ph
H
H
Ph
1a (2)
1a (2)
1a (5)
n-BuOH (2)
n-BuOH (2)
n-BuOH (1)
n-BuOH (2)
n-BuOH (1)
MeOH (1)
100
88
57
50 (À):0:48
23 (À):34:38
43 (30/13):2:13
88 (60/28):3:7
30 (24/6):30:7
14 (À):86:0
31 (À):69:0
0:0:60
98
1d (1)
1a (0.5)
1a (2)
62
C6H13
Ph
H
42[b]
100[b]
MeOH (2)
n-Bu 1a (5)
n-BuOH[c] (1) 60
i-PrOH[c] (1)
i-PrOH (1)
58
61
messy mixture
messy mixture
1d (5)
[a]
[b]
[c]
Determined by GC.
Determined by NMR, see also Figure S11 and S12 in Supporting Information.
Benzylic alcohols dehydrated to form the corresponding ethers under these experimental conditions.
[b]). The vinyl ethers of alkyl derivatives were not since the corresponding vinyl ethers as E and Z iso-
stable enough under the acidic conditions and formed mers are stable. Thus, the addition of different alco-
finally the corresponding ketones or uncontrolled by- hols to dimethyl acetylenedicarboxylate (DMAD)
products (entries 8–10).
catalysed by complexes 1a–d was tested (Table 3).
Remarkably, a different stereoisomer was obtained
depending on the catalyst employed, i.e., catalysts
with EWD phosphines 1a and 1b lead to butoxyma-
leate E-7 (entries 1–3)[24] and catalysts with ED phos-
Stereoselective Synthesis of Vinyl Ethers
As commented before, alkyl propiolates are suitable phines 1c and d lead to the fumarate Z-7 (entries 4
alkynes to study the stereoselectivity of the reaction, and 5), with high levels of stereoselectivity in both
Adv. Synth. Catal. 2010, 352, 1701 – 1710
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1703

Avelino Corma et al.
Table 3. Hydroalkoxylation of DMAD 6 with different alcohols catalysed by AuPR3NTf2 complexes 1a–d.[a]
FULL PAPERS
Entry
ROH
Main Product
Catalyst (mol%)
Yield [%][b]
Z/E[c]
1
2
3
4
5
1a (5)
>99 [99]
58
29
89
1/13[d]
1/1.1
1/4
1a (0.5)
1b (0.5)
1c (0.5)
1d (0.5)
n-BuOH 3
8/1
>99 [87]
19/1[d]
6
7
1a (5)
1d (0.5)
>99
>99
1/19
1/10
MeOH 8
8
1d (0.05)
1d (5)
94
4.5/1
99/1
9
i-PrOH 10
>99 [77]
[a]
Reaction conditions: alkyne (1 mmol), alcohol (2 mmol), catalyst, dry solvent (0.5 mL), stirring at room temperature for
24 h.
[b]
[c]
[d]
Average of two runs, calculated by GC; in brackets, isolated yield.
Average of two runs.
Gram scale.
cases. Yields were excellent even at the gram scale although a decrease in activity is observed. The
(see entries 1 and 5). MeOH comparably displayed an 31P NMR spectrum of the complex after the third use
enhanced reactivity profile, with a high-yielding shows in both cases a new signal at 45 ppm, corre-
access to the corresponding fumarate Z-9 being possi- sponding to the species[25] [Au
ACHTUNGRTNEUNG(PPh3)2]NTf2 (see Fig-
ble at a remarkable low catalyst loading of 0.05 mol% ure S1 in the Supporting Information). Unfortunately,
(entry 8). Curiously, the amount of catalyst 1d ap- this species is not active as catalyst. Catalyst 1d deac-
peared to greatly influence the sense of stereoselectiv- tivates similarly during the addition of n-BuOH 3 to
ity of the hydromethoxylation (compare entry 7 and DMAD 6 (see Table S2 and Figure S2 in the Support-
entry 8), what would be explained ahead by isomeri- ing Information).
sation effects (for instance, see Figure S9 in the Sup-
porting Information). A secondary alcohol such as i-
PrOH added equally well with the achievement of ex- Insights into the Mechanism
cellent yield and virtually complete stereoselectivity
(entry 9), though a larger amount of catalyst 1d was A tentative mechanism for the addition of alcohols to
required to complete the reaction (compare entries 5, the internal alkyne 6 catalysed by gold complexes 1a–
8 and 9), and tertiary alcohols do not react under our e is depicted in Scheme 1:
reaction conditions.
The different steps in the mechanism were investi-
The reusability of catalyst 1a was studied for the re- gated by kinetic and isotopic studies, and NMR ex-
action of phenylacetylene 2 with n-BuOH 3 (entry 1 periments. The reaction starts with the formation of
in Table 1, see Table S1 in the Supporting Informa- the primary Z-vinyl ether, which should take place by
tion). Two procedures can be followed for recovering a gold(I)-catalysed trans-addition of the alcohol
1a, depending on the solvent employed: with CH2Cl2, across the triple bond. The different catalytic activi-
1a is recovered by precipitation in hexane and filtra- ties of 1a (EWD phosphine) and 1d (ED phosphine),
tion, while if CH3CN is used the reaction products are measured as the initial rate (r1), were compared, and
extracted with hexane, 1a remaining in the CH3CN 1d is one order of magnitude more active than 1a
phase. The inmiscibility of the acetonitrile phase in (Table S3, see Figure S3 in the Supporting Informa-
hexane is only possible because a rather high sub- tion). In contrast, the isomerisation rate (r2) is three
strate concentration is used (see Supporting Informa- times higher for catalyst 1a compared to 1d (Table S3,
tion). Anyway, in both cases the results obtained are see Figure S4 in the Supporting Information), and
similar and the catalyst is still active after three uses, equals to r1 for 1a (r1/r2 =1.1, Table S3 in the Support-
1704
asc.wiley-vch.de
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1701 – 1710

Regio- and Stereoselective Intermolecular Hydroalkoxylation of Alkynes
Scheme 1. Proposed mechanism for the gold(I)-catalysed formation of dimethyl fumarates and maleates from DMAD 6 and
alcohols.
ing Information). This result nicely explains why E-7 in the 1H NMR spectrum of the reaction mixture,
À
À
is formed when 1a is used as catalyst and not when 1d peaking as
a singlet at 2.95 [CO2Me CH2 C-
is used instead (Figure S5 in the Supporting Informa- (OBu)2CO2Me]. The vinyl ether E-12 was smoothly
AHCTUNGTRENNUNG
tion). The reverse isomerisation, from E-7 to Z-7, is achieved by addition of CD3OD to Z-7 (Figure 2),
negligible for both catalysts (Figure S6, r3 in Table S3 which unambiguously comes from a transalkoxylation
in the Supporting Information). It should be re- process.
marked that the isomerisation is not catalysed by a
Following the first pathway (Scheme 2, PW-1), and
Brønsted acid (HNTf2, Figure S7 in the Supporting using CD3OD instead of MeOH, one should expect
Information) and that the complex 1a without the al- that the obtained vinyl ethers contain deuterium at
cohol does not produce the isomerisation either (Fig- the double bond (see above Figure 2). However, no
ure S8 in the Supporting Information). From these deuterium was found in the product, as assessed by
results it can be concluded that the addition of the al- 1H (correct integration for all the protons), 13C NMR
cohol to the triple bond and subsequent isomerisation (no splitting of the peak corresponding to the olefinic
are catalysed by gold(I). The formation of the inter- carbon) and GC-MS (no +1 peaks). Thus, the double
mediate ketal was confirmed by NMR and kinetic ex- bond regeneration should come from the release of
periments. The ketal corresponding to 7 (intermediate one of the alcohols and gold, the olefinic proton re-
II in Scheme 1, R=n-Bu) was found in small amounts maining untouched during this process. If we consider
Figure 2. Plot-time yield for the isomerisation of Z-7 to the sum of both E-isomers (a), to E-7 (b) and to E-12 (c).
Adv. Synth. Catal. 2010, 352, 1701 – 1710
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1705

Avelino Corma et al.
FULL PAPERS
Scheme 2. Proposed ways for the elimination step.
PW-2, the presence of deuterium at the double bond after storage for six months (Table S4, entry 6 in the
should also be expected and, consequently, this path- Supporting Information).
way can be rejected. Finally, taking into account the
kinetic and isotopic results, PW-3 was considered as
the plausible elimination mechanism.
Regioselective Hydroalkoxylation with Polyols
Although the kinetic approach shown here could
be considered incomplete, our conclusions are also The particular behaviour of the gold(I) catalysts 1a–e
supported from the product distribution and the iso- in terms of selectivity prompted us to study a possible
topic results. We do not know if the elimination is a regioselective addition of diols to alkynes. Moreover,
sequential E1 mechanism-type (ROH releases first, the hydrolysis of the ketals would lead to the corre-
generates a carbocation, and then gold eliminates) or sponding ketones, which represent a formal regiose-
concerted. However in either case, gold should persist lective hydration of alkynes, an issue which is still un-
all over the isomerisation process. In fact, gold may resolved. The results obtained are listed in Table 4.
exert a directing effect, forcing the two CO2Me to be
Complex 1e was the best catalyst (compare en-
eclipsed in the transition state and forming the male- tries 1–7 with 8–12), showing complete conversion
ate adduct after deauration. This directing effect of and selectivity for the addition of glycerol to phenyla-
the gold complex could be due to steric hindrance cetylene (entry 7).[26] As far as we know, no hydroal-
and/or to coordination to the opposite OR groups.
koxylations of alkynes with glycerol have been previ-
As it has been seen before, the isomerisation rate ously reported.[27] Glycerol is more active than the
from the Z to the E isomer changes dramatically as a diols (compare entries 12, 13 and 15) and the 5-mem-
function of the catalyst employed (see Table S3 and bered ketal was exclusively formed in all cases.
Figure S5 in the Supporting Information), allowing us CH3CN was the solvent of choice at room tempera-
in this way to control the stereoselectivity of the final ture since glycerol is not soluble in CH2Cl2 or toluene
product. To this respect, a further study was carried (entry 3). Good regioselectivities were obtained in
out to probe the influence of other parameters such some cases (5–3 to 1, see entries 11–15), although at
as the amount of catalyst and alcohol on the yield and low conversions. Unfortunately, the regioselectivity
selectivity of the reaction (Figure S9 in the Support- obtained for the corresponding ketones after acidic
ing Information). It was found that the reaction rates hydrolysis was low (higher: 2.5 to 1 in entry 11).
for both processes, hydroalcoxylation and isomerisa-
tion, are accelerated by increasing the amount of cat-
alyst and the alcohol concentration (see Tables S4–S6 Catalyst Recovery and Recycling
as well in the Supporting Information). This explains
why catalyst 1d is able to form the Z or the E isomer, The recovery and reuse of homogeneous gold cata-
depending on the reaction conditions (see Table 3, lysts is of interest from an economical and environ-
compare entries 7 and 8). As MeOH is ca. 10 times mental point of view. Complex 1a could be reused for
more reactive than n-BuOH, E-9 is eventually formed the addition of glycerol 13 to phenylacetylene 2 up to
with the same amount of catalyst 1d (see Table 3, six times without depletion of the catalytic activity
compare entries 5 and 7). In general, the results ob- (Figure 3).
served in Table 3 are in good agreement with this
As observed before for the formation of vinyl
study. It has to be remarked that 1d keeps its activity ethers, the reaction mixture including catalyst 1a be-
1706
asc.wiley-vch.de
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1701 – 1710

Regio- and Stereoselective Intermolecular Hydroalkoxylation of Alkynes
Table 4. Au(I)-catalysed formation of cyclic ketals from alkynes and polyols and subsequent hydrolysis.[a]
Entry
R1
Ph
R2
H
R3
Catalyst
T [8C]
Conversion [%][b]
A [%][b]
B [%][b]
C [%][b]
D [%][b]
[c]
1
2
3
4
CH2OH
AuPPh3BF4
20
30
0
90
>99
0
30
-
90
87
-
–
–
–
–
30
-
90
99
-
–
–
–
–
AuPPh3Cl
1a[d]
1b
1c
5
–
–
6
7
8
9
10
11
12
13
14
15
16
1d
1e
1a
1b
1c
1d
1e
1e
1c
90
98
6
84
92
1
6
2
–
–
3
9
90
98
2
13
5
–
–
4
11
10
21
35
28
9
Ph
n-Bu
CH2OH
80
24
15
29
65
50
14
29
3
7
3
16
33
27
11
22
2
8
17
14
3
7
1
30
22
5
12
1.5
H
Et
1e
1c
17
1.5
[a]
Reaction conditions: alkyne (0.25 mmol), diol (0.25 mmol), catalyst (0.05 mmol), AgBF4 (if necessary, 0.05 mmol), dry sol-
vent (0.25 mL), N2.
Calculated by GC, 3:1 mixture of diastereoisomers.
Formed in-situ.
[b]
[c]
[d]
The reaction did not work in dry CH2Cl2 or dry toluene as solvents.
haves as an ionic liquid when dissolved in CH3CN, al- isomerisation of the Z counterpart, which is initially
lowing extraction of the reaction products in hexane. formed through an expected trans-hydroalkoxylation
Ionic liquids have been used as reaction medium for mechanism. Given the plethora of studies arising in
gold-catalysed reactions before (including recycles).[28] the last years on C=C activation by gold, these find-
ICP Au analyses of 14 show that the amount of gold ings can be of interest.
leached in the extraction procedure is less than
Moderate to good levels of conversion and selectiv-
0.05%. In accordance, the 31P NMR spectrum of the ity are achieved when using polyols as hydroalkoxy-
catalytic species recovered after the sixth use showed lating agents. The complex 1a is reused up to six
the persistence of the phosphine ligands (see Fig- times without an observed decrease of the catalytic
ure S10 in the Supporting Information), although, activity for phenylacetylene.
again, [Au
G
The tuneable selectivity shown by catalysts 1a–e to-
ed. Nevertheless, the catalytic system remains active gether with the recovery of 1a, the easy isolation of
for this particular reaction.
the products and the mild conditions employed
render the procedure reported here a convenient pro-
cess from synthetic and environmental points of view.
Conclusions
AuPR3NTf2 complexes 1a–e are selective catalysts for
Experimental Section
the intermolecular hydroalkoxylation of electron-poor
À ꢀ À
alkynes of type R C C EWG and DMAD. In the re-
Typical Preparation of the Gold(I) Catalysts
[AuDavePhosNTf2 (1c)]
actions of phenylacetylene, the ratio vinyl ether to
ketal can be controlled by choosing the catalyst.
When using stabilised internal alkynes such as
DMAD, both Z:E isomers of the vinyl ether can be
obtained with excellent yields and stereoselectivities,
depending on the catalyst employed and the reaction
AuDavePhosCl (350 mg, 0.56 mmol, see Supporting Infor-
mation for preparation) and AgNTf2 (217 mg, 0.56 mmol)
were placed in a round-bottomed flask. Air evacuation-ni-
trogen refilling cycles were carried out and a rubber septum
was rapidly fitted after the last nitrogen refilling, a nitrogen
conditions. The E isomer comes from a gold-catalysed balloon was additionally coupled through a needle. Then,
Adv. Synth. Catal. 2010, 352, 1701 – 1710
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1707

Avelino Corma et al.
FULL PAPERS
Figure 3. Reusability study of 1a dissolved in CH3CN as “ionic liquid-type” system.[a] 3:1 mixture of diastereoisomers.
dry CH2Cl2 (15 mL) was added and the mixture was magnet-
ically stirred at room temperature for 30 min. Then, the
white solid formed (AgCl) was filtered off over celite and
the clear filtrates were concentrated to dryness to obtain
AuDavePhosNTf2 as a yellow-bright solid; yield: 490 mg
(quantitative). IR: n=2931, 2854, 1342, 1196, 1142,
temperataure for 24 h. Then, an aliquot was taken for GC
analysis and the mixture was added to n-hexane (50 mL)
and stirred for 5 min. The resulting suspension was cooled in
the fridge for 1 h and, after this time, it was left to reach
room temperature and filtered over celite. The filtrates were
concentrated under reduced pressure to give E-2-butoxybut-
2-enedioic acid dimethyl ester E-7 as a colourless liquid;
yield: 184 mg (85%, E/Z ratio=10:1). GC/MS: m/z=216
(M+·, 2%), 201 (2%), 185 (4%), 173 (2%), 129 (31%), 101
1
1057 cmÀ1; H NMR: d=7.64–7.38 (5H, mult), 7.11 (2H, t,
J=7.7 Hz), 6.99 (1H, d, J=7.0 Hz), 2.46 (6H, s), 2.28–2.00
(4H, mult), 1.86–1.62 (8H, mult), 1.48–1.18 (10H, mult);
13C NMR: d=151.2 (C), 148.1 (C,), 133.4 (CH), 132.4 (CH,),
131.6 (CH, ꢃ2), 129.3 (CH), 127.3 (CH), 124.0 (C), 123.3
(C,), 122.2 (CH), 119.4 (C, JC,F =323 Hz), 119.1 (CH), 43.4
(CH3, ꢃ2), 37.4 (CH, JC,P =34 Hz), 36.3 (CH, JC,P =34 Hz),
31.0 (CH2), 29.6 (CH2,), 26.6 (CH2), 26.5 (CH2, ꢃ2), 26.4
(CH2), 26.3 (CH2), 25.6 (CH2), 25.5 (CH2, ꢃ2); 31P NMR:
d=39.1; anal. calcd. for C28H36AuF6N2O4PS2: C 38.63, H
4.17, N 3.22, S 7.37; found: C 39.53, H 4.32, N 3.24, S 7.43;
HR-MS (ESI): m/z=590.2258, (MÀCl)+, major peak; calcd.
for C26H36AuNP: 590.2251.
1
(100%), 69 (45%); H NMR: d=5.10 (1H, s), 3.80 (3H, s),
3.79 (2H, t, J=6.5 Hz), 3.61 (3H, s), 1.66 (2H, tt, J=7.5 Hz,
6.5 Hz), 1.35 (2H, sext, J=7.5 Hz), 0.86 (3H, t, J=7.5 Hz);
13C NMR: d=66.2 (C), 163.9 (C), 162.0 (C), 92.6 (CH), 69.8
(CH2), 52.6 (CH3), 51.2 (CH3), 30.1 (CH2), 18.7 (CH2), 13.4
(CH3).
Reusability Study using Glycerol as Alcohol (Figure
3)
Complex 1a (78 mg, 5 mol%) and glycerol 13 (184 mg,
2 mmol) were placed in an oven-dried 10 mL round-bot-
tomed flask. Air evacuation-nitrogen refilling cycles were
carried out and a rubber septum was rapidly fitted after the
last nitrogen refilling, a nitrogen balloon was additionally
coupled through a needle. Then, dry CH3CN (1 mL) and
Typical Hydroalkylation Procedure with
Monoalcohols (Table 3, entry 1)
Complex 1a (39 mg, 5 mol%) was placed in an oven-dried
vial and dry CH2Cl2 (0.5 mL), but-2-ynedioic acid dimethyl
ester 6 (DMAD, 122 mL, 1 mmol) and n-butanol 3 (182 mL,
2 mmol) were sequentially added. The vial was rapidly
sealed and the mixture was magnetically stirred at room
phenylacetylene
2 (330 mL, 3 mmol) were sequentially
added and the mixture was magnetically stirred at room
temperature for 24 h. Then, n-hexane (10 mL) was added
1708
asc.wiley-vch.de
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1701 – 1710

Regio- and Stereoselective Intermolecular Hydroalkoxylation of Alkynes
and the mixture stirred for 15 min. After stopping, the top
layer (hexane) was separated with a pipette and fresh n-
hexane (10 mL) was added and the mixture stirred again for
5 min. This process was repeated once again. In one run, the
remaining catalyst was dried under vacuum for 1 h and fresh
reactants and solvent (see above) were added to perform a
new reaction. In the other run, an aliquot was taken from
the combined hexane extracts and analysed by GC. Then,
volatiles were removed under vacuum and the resulting resi-
due was purified by a rapid column chromatography (50%
AcOEt/hexane) to give 2-methyl-2-phenyl-[1,3]dioxolan-4-
yl)-methanol 14; yield: 338 mg (87%, 3:1 mixture of dia-
120, 216; Angew. Chem. Int. Ed. 2008, 47, 209; f) V.
Belting, N. Krause, Org. Lett. 2006, 8, 4489.
[3] Non-metal-catalysed intermolecular hydroalkoxylations
had also been reported, see: J. E. Murtagh, S. H.
McCooey, S. J. Connon, Chem. Commun. 2005, 227.
[4] T. Murata, Y. Mizobe, H. Gao, Y. Ishii, T. Wakabaya-
shi, F. Nakano, T. Tanase, S. Yano, M. Hidai, I. Echi-
zen, H. Nanikawa, S. Motomura, J. Am. Chem. Soc.
1994, 116, 3398.
[5] A. Avshu, R. D. OꢅSullivan, A. W. Parkins, N. W.
Alcock, R. M. Countryman, J. Chem. Soc. Dalton
Trans. 1983, 1619.
ACHTUNGTRENNUNGstereoisomers, only major diastereoisomer shown). Rf (neat
[6] Y. Kataoka, O. Matsumoto, K. Tani, Organometallics
AcOEt): 0.47. GC/MS: m/z (M+· 194)=179 (100%), 163
(10%), 105 (79%), 77 (29%), 43 (31%); IR: n=3421 (b),
2988, 2934, 2888, 1446, 1374, 1246, 1201, 1135, 1046, 1026,
1996, 15, 5246.
[7] J. Barluenga, F. Aznar, M. Bayod, Synthesis 1988, 144.
[8] C. Gemel, G. Trimmel, C. Slugovc, S. Kremel, K. Mer-
eiter, R. Schmid, K. Kirchner, Organometallics 1996,
15, 3998.
[9] S. H. Bertz, G. Dabbagh, P. Cotte, J. Org. Chem. 1982,
47, 2216.
1
940, 880, 762, 703 cmÀ1); H NMR: d=7.40 (2H, mult), 7.24
(3H, mult), 3.99 (1H, mult), 3.77 (1H, dd, J=8.1 Hz,
5.4 Hz), 3.69 (2H, tt, J=8.4 Hz, 7.1 Hz), 3.55 (1H, dd, J=
11.6 Hz, 5.3 Hz), 2.30 (1H, s), 1.60 (3H, s); 13C NMR: d=
142.9 (C), 128.2 (CH, ꢃ2), 125.2 (CH, ꢃ2), 124.7 (CH),
109.6 (C), 76.0 (CH), 65.7 (CH2), 63.3 (CH2), 28.0 (CH3);
anal. calcd. for C11H14O3: C 68.02, H 7.27; found: C 66.88, H
7.67; HRMS (ESI): m/z 195.1014 [(M+H)+, calcd. for
C11H15O3: 195.1021], 193.0879 [(MÀH)+, calcd. for
C11H13O3: 193.0865], 179.0668 [(MÀHÀCH3)+, major peak,
calcd. for C10H11O3: 179.0708].
[10] Y. Kataoka, O. Matsumoto, K. Tani, Chem. Lett. 1996,
727.
[11] K. Breuer, J. H. Teles, D. Demuth, H. Hibst, A. Schꢆ-
fer, S. Brode, H. Domgçrgen, Angew. Chem. 1999, 111,
1497; Angew. Chem. Int. Ed. 1999, 38, 1401.
[12] a) J. H. Teles, S. Brode, M. Chabanas, Angew. Chem.
1998, 110, 1475; Angew. Chem. Int. Ed. 1998, 37, 1415;
b) Y. Fukuda, K. Utimoto, J. Org. Chem. 1991, 56,
3729.
Supporting Information
The syntheses of gold catalysts 1b and 1c, reaction proce-
dures, compound characterisation, NMR spectra and addi-
tional Tables and Figure for this article are available as Sup-
porting Information.
[13] The metal species is generally not recovered at the end
of the reaction, one single exception being the Telesꢅ
work (ref.[11]), where a zinc silicate is employed as het-
erogeneous catalyst in flow. In addition, high reaction
temperatures are needed in most of the cases reported.
[14] D. Masui, Y. Ishii, M. Hidai, Chem. Lett. 1998, 717.
[15] A. G. Davies, R. J. Puddephatt, J. Chem. Soc. C 1968,
1479.
Acknowledgements
[16] Y. Kataoka, Y. Tsuji, O. Matsumoto, M. Ohashi, T. Ya-
magata, K. Tani, J. Chem. Soc. Chem. Commun. 1995,
2099.
Financial support by MAT2006 and PROMETEO from
Generalitat Valenciana are acknowledged. A. L-P. thanks
MICINN for financial support on JAE-Doctor program. V.
R. thanks Consejo Superior de Investigaciones Cientꢀficas
(CSIC) for an I3-P fellowship.
[17] M. Konkol, H. Schmidt, D. Steinborn, J. Mol. Catal. A
2007, 261, 301.
[18] J. W. Goodyear, C. W. Hemingway, R. W. Harrington,
M. R. Wiseman, B. J. Brisdon, J. Organomet. Chem.
2002, 664, 176.
References
[19] E. Hevia, J. Pꢀrez, L. Riera, V. Riera, Organometallics
2002, 21, 1750.
[1] a) For an exhaustive review see F. Alonso, I. P. Belet-
skaya, M. Yus, Chem. Rev. 2004, 104, 3079; b) Kirk-
Othmer Encyclopedia of Chem. Tech., (Eds:. J. I.
Kroschwitz, M. Howe-Grant), John Wiley & Sons, 4th
edn., 1991, Vol. 1, p 221.
[2] a) S. Seo, X. Yu, T. J. Marks, J. Am. Chem. Soc. 2009,
131, 263; b) A. Diꢀguez-Vꢄzquez, C. C. Tzschucke, J.
Crecente-Campo, S. McGrath, S. V. Ley, Eur. J. Org.
Chem. 2009, 1698; c) B. A. Messerle, K. Q. Vuong, Or-
ganometallics 2007, 26, 3031; d) B. Liu, J. K. De
Brabander, Org. Lett. 2006, 8, 4907. In the last years,
intramolecular hydroalkoxylations have been incorpo-
rated into elaborated tandem processes as one of the
main steps, that is, see e) A. Diꢀguez-Vꢄzquez, C. C.
Tzschucke, W. Y. Lam, S. V. Ley, Angew. Chem. 2008,
[20] a) For a seminal work on this reaction see E. Winter-
feldt, H. Preuss, Angew. Chem. 1965, 77, 679; b) F.
Nasiri, B. Atashkar, Monatsh. Chem. 2008, 139, 1223.
[21] The well-reported alkynophilic nature of the gold
centre (see, i.e.: a) D. J. Gorin, F. D. Toste, Nature 2007,
446, 395; b) A. S. K. Hashmi, G. J. Hutchings, Angew.
Chem. 2006, 118, 8064; Angew. Chem. Int. Ed. 2006, 45,
7896) together with its strong Lewis acidity makes
these complexes suitable candidates for catalysts in ad-
ꢀ
ditions to C C triple bonds. In fact, we have recently
reported that the hydration of alkynes can be carried
out at room temperature by using these gold complexes
as catalysts, see c) A. Leyva, A. Corma, J. Org. Chem.
2009, 74, 2067.
Adv. Synth. Catal. 2010, 352, 1701 – 1710
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1709

Avelino Corma et al.
FULL PAPERS
[22] N. Mezailles, L. Ricard, F. Gagosz, Org. Lett. 2005, 7,
4133; 1a can be purchased from Aldrich as a dimer-tol-
uene aduct.
previously reported the complex AuPPh3BF4 (formed
in situ from AuPPh3Cl and AgBF4) as active catalyst
for the formation of different ketals and cyclic thio-
[23] CH2Cl2 and CH3CN can be used as solvents. However,
CH2Cl2 was used as solvent since the gold complexes
1a–e can be recovered by precipitation with hexane
and the reactants and products are fully soluble.
[24] It has been reported that the E-isomer is less prone to
isomerise, see: M. M. Baag, A. Kar, N. P. Argade, Tet-
rahedron 2003, 59, 6489.
ACHTUNGTRENkNUNG etals from the corresponding alkynes and diols.
[27] Glycerol is a massive by-product of the biodiesel pro-
duction whose market price has dramatically dropped
in the last years, expected to be cheaper than other
common diols such as glycol, see: A. Corma, S. Iborra,
A. Velty, Chem. Rev. 2007, 107, 2411.
[28] For intramolecular hydroalkoxylation see: a) ꢇ. Aksin,
N. Krause, Adv. Synth. Catal. 2008, 350, 1106. For
others, see: b) D.-M. Cui, Y.-N. Ke, D.-W. Zhuang, Q.
Wang, C. Zhang, Tetrahedron Lett. 2010, 51, 980; c) X.
Moreau, A. Hours, L. Fensterbank, J.-P. Goddard, M.
Malacria, S. Thorimbert, J. Organomet. Chem. 2009,
694, 561.
[25] M. Bardajꢁ, P. Uznanski, C. Amiens, B. Chaudret, A.
Laguna, Chem. Commun. 2002, 598. See preparation
and use of [Au
tion.
ACHTUGNTERN(NUNG PPh3)2]NTf2 in the Supporting Informa-
[26] L. L. Santos, V. R. Ruiz, M. J. Sabater, A. Corma, Tet-
rahedron 2008, 64, 7902 and others (see ref.[2f]) have
1710
asc.wiley-vch.de
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1701 – 1710