Regioselective hydration of alkynes by iron(III) Lewis/Bronsted catalysis

Article abstract of DOI:10.1002/chem.201200580

The triflimide iron(III) salt [Fe(NTf₂)₃] promotes the direct hydration of terminal and internal alkynes with very good Markovnikov regioselectivities and high yields. The enhanced carbophilic Lewis acidity of the Fe^III cation mediated by the weakly-coordinating triflimide anion is crucial for the catalytic activity. The iron(III) metal salt can be recycled in the form of the OPPh₃/[Fe(NTf₂)₃] system with similar activity and selectivity. However, spectroscopic and kinetic studies show that [Fe(NTf₂)₃] hydrolyzes under the reaction conditions and that catalytically less active BrAonsted species are formed, which points to a Lewis/Bronsted co-catalysis. This triflimide-based catalytic system is regioselective for the hydration of internal aryl-alkynes and opens the door to a new synthetic route to alkyl ketophenones. As a proof of concept, the synthesis of two antipsychotics Haloperidol and Melperone, with general butyrophenone-like structure, is shown. Just add water! The triflimide iron(III) salt [Fe(NTf₂)₃] promotes the direct hydration of terminal and internal alkynes with very good Markovnikov regioselectivities and high yields (see scheme). Copyright

Full text of DOI:10.1002/chem.201200580

FULL PAPER
DOI: 10.1002/chem.201200580
Regioselective Hydration of Alkynes by IronACTHUNRGTNEUNG(III) Lewis/Brønsted Catalysis
Jose R. Cabrero-Antonino, Antonio Leyva-Pꢀrez,* and Avelino Corma*[a]
Abstract: The triflimide iron
G
similar activity and selectivity. Howev-
er, spectroscopic and kinetic studies
show that [FeACHTUNGTERNU(NG NTf2)3] hydrolyzes under
the reaction conditions and that cata-
lytically less active Brønsted species
are formed, which points to a Lewis/
Brønsted co-catalysis. This triflimide-
based catalytic system is regioselective
for the hydration of internal aryl-al-
kynes and opens the door to a new syn-
thetic route to alkyl ketophenones. As
a proof of concept, the synthesis of two
antipsychotics Haloperidol and Melper-
one, with general butyrophenone-like
structure, is shown.
[Fe(NTf2)3] promotes the direct hydra-
ACHTUNGTRENNUNG
tion of terminal and internal alkynes
with very good Markovnikov regiose-
lectivities and high yields. The en-
hanced carbophilic Lewis acidity of the
FeIII cation mediated by the weakly-co-
ordinating triflimide anion is crucial for
Keywords: alkynes
·
anions
·
the catalytic activity. The iron
metal salt can be recycled in the form
of the OPPh3/[Fe(NTf2)3] system with
ketones · regioselectivity · iron ·
hydration
ACHTUNGTRENNUNG
Introduction
mations had to be restricted due to the high toxicity of the
Hg compounds. As possible alternative catalysts, many
other Brønsted acids and metal salts and complexes have
been investigated, and two of them, Au and Ru, deserve fur-
ther consideration. On one hand, AuI, which is isoelectronic
to HgII, shows excellent alcohol-mediated hydration rates[5]
and, on the other hand, Ru complexes catalyze anti-Markov-
nikov hydration of terminal alkynes (R’=H in Scheme 1) to
produce aldehydes.[6] As a result, the more active and selec-
tive catalysts for the hydration of the alkynes are still based
on toxic and/or expensive late-transition metals.
The hydration of alkynes is a 100% atom-economical pro-
cess that allows easy access to carbonyl compounds
(Scheme 1).[1,2] As agreed by many authors,[1] two main
issues remain to be addressed for this transformation: 1) the
use of nontoxic and inexpensive first-row transition metals
as catalysts, and 2) the regioselective hydration of asymmet-
rical internal alkynes without anchimeric assistance. Here,
advances regarding these two chemical challenges will be
presented.
Looking at the first-row transition metals, only the more
electron-rich, that is, copper and zinc, have shown some suc-
cess as catalysts for the hydration of alkynes, whereas the
catalytic behavior of the rest, to the best of our knowledge,
have not been reported.[1] An explanation for this may be
due to the fact that a “soft” Lewis acidity is needed to prop-
erly interact with the soft p-nucleophile alkyne, and only
those late-first-row transition-metal cations, having a full nd
level, present suitable Lewis acidity to activate the alkyne
towards hydration. It is well-known that metals become
more oxophilic and less carbophilic when going from right
to left in the periodic table or, in other words, harder Lewis
acids hamper alkyne activation. However, an exception can
be found in high-spin d5 FeIII compounds having a semi-full
3d level. This electronic configuration imparts to the cation,
to some extent, similar bond properties to those in full 3d
levels and, indeed, we have recently found that FeIII trifli-
Scheme 1. Hydration of alkynes catalyzed by Brønsted or Lewis acid cat-
alysts.
A plethora of Brønsted and Lewis acids have been used
as catalysts for the otherwise kinetically inhibited hydra-
tion.[3–6] Among them, HgII salts[3] present the best catalytic
activity, which has allowed in the past century the manufac-
ture of acetaldehyde from acetylene on an industrial scale.[7]
However, this and other acetylene-based hydration transfor-
[a] J. R. Cabrero-Antonino, Dr. A. Leyva-Pꢀrez, Prof. Dr. A. Corma
Instituto de Tecnologꢁa Quꢁmica
mide, [FeACTHNUGTRNEUNG(NTf2)3], catalyzes the regioselective Markovnikov
Consejo Superior de Investigaciones Cientꢁficas
Universidad Politꢀcnica de Valencia
Avda. de los Naranjos s/n, 46022 Valencia (Spain)
Fax : (+34)963879444
hydrothiolation of styrenes without poisoning of the metal
catalyst by the sulfur nucleophile.[8] The low-coordinating
anion ꢀNTf2 is essential to provide a strong but soft Lewis
acidity to the FeIII cation, increasing the carbophilicity of
E-mail: anleyva@itq.upv.es
acorma@itq.upv.es
FeIII.[9] Thus, it was envisaged that [Fe
ACHTUNGTRENNUNG
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201200580.
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Table 1. Comparison between FeCl3 and [FeACHTNUTRGNE(UNG NTf2)3] in the hydration of
phenylacetylene (1) either under oxidative chlorine-assisted conditions or
not, and screening of solvents.
a general hydration of alkynes. In addition, the ability of
FeIII to activate benzylic positions is also well-known,[10]
which might also control the regioselectivity of the process
in asymmetrical aryl alkynes. We can confirm that a general
and regioselective hydration indeed occurs, although the for-
mation of catalytically active protons in situ also takes
place, so a Lewis/Brønsted co-catalysis must be considered.
The addition of catalytic amounts of triphenylphosphane
oxide to the triflimide-based acids allows recycling of the
catalyst without loss of catalytic activity.
Entry[a]
Catalyst
FeCl3
t [h]
Solvent
DCE
2 [%][b]
1
67
19
46
39
51
96
92
72
12
17
40
20
5
2[c]
3[c,d]
4[d]
FeCl3+3AgNTf2
20
5[d]
1,4-dioxane
EtOAc
EtNO2
DME
iPrOH
MeCN
Toluene
H2O
6[d]
Results and Discussion
7[d]
8[d]
Pearsonꢃs (hardness-softness) rules dictate that hard cations
bound to soft anions become softer. This effect is predicted
by molecular orbital theory, which shows that orbital over-
lap between the occupied nd levels of the metal and unoccu-
pied orbitals of the ligands generates a more stabilized
LUMO in the final compound. To maintain a strong Lewis
acidity, the ligands must be weakly coordinating. Thus, if
Fe3+ is combined with soft, low-coordinating ligands, an in-
crease in the p-carbophilicity of the final compound should
be expected. Looking into possible candidates, on one hand,
inorganic low-coordinated ligands such as ꢀPF6, ꢀBF4 or
ꢀSbF6 are unsuitable because they are in equilibrium with
the corresponding Lewis acids: XFnꢀ! XF + Fꢀ, thus gener-
ating iron fluoride salts in situ. However, organic low-coor-
dinating anions such as triflate (ꢀOTf) or triflimide (ꢀNTf2)
might successfully induce the particular p-Lewis acidity re-
quired to activate alkynes.[11] To study the catalytic activity
of different FeIII salts and other metals in the hydration of
alkynes, the addition of water to phenylacetylene (1) was
chosen as a test reaction (Table 1). We first compared FeCl3
9[d]
10[d]
11[d,e]
12[d]
13[d]
DMF
0
[a] Standard reaction conditions: Phenylacetylene (1; 0.25 mmol, 28 mL),
water (0.75 mmol, 14 mL), catalyst (0.025 mmol), solvent (0.5 mL), 808C,
20 h. [b] GC yield using dodecane as internal standard. [c] In the pres-
ence of 5 bar O2. [d] Catalyst was preformed as [FeCl3 (10 mol%)+
3 AgNTf2 (30 mol%)] in the corresponding solvent (0.5 mL) for 30 min
at room temperature. [e] The major product observed was generated
from the hydroarylation reaction between toluene and phenylacetylene.
The catalyst screening included a variety of metal iron-
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
AHCTUNGTRENNUNG
(III) salts, using as anions Cl, Br, ꢀacac, PF6, SbF6, OTf
and ꢀNTf2, fifteen different metal triflimides [M
ACHTNUTRGNEGNU(NTf2)n, n=
1-3], and finally the Brønsted acids HOTf and HNTf2.[14]
The results for the Markovnikov hydration of phenylacety-
lene (1) with water (3 equiv.) using 1,4-dioxane as solvent,
at 808C for 20 h, are shown in Table 2. No catalytic activity
was found for ꢀCl, ꢀBr, ꢀacac, and ꢀPF6, and a moderate in-
crease occurred when ꢀSbF6 or ꢀOTf were used as anions
for FeIII (entries 1–8). However, an excellent conversion was
and [FeACHTUNGTRENNUNG(NTf2)3] because a chlorine-assisted FeCl3-catalyzed
hydration of alkynes has recently been reported (Table 1,
entries 1–4).[12] In that procedure, hydrochlorination of the
alkyne occurs in the presence of FeCl3 but only when 1,2-di-
chloroethane (DCE) is used as solvent and when the reac-
tion is run under a low pressure of oxygen, which apparently
maintains the activity of the catalyst through the hydro-
chlorination-hydrolysis process.[13] Our results indicate that
the triflimide salt performs better than FeCl3 (compare
Table 1, entries 1–3) and, moreover, oxygen was not re-
quired with the former catalyst (Table 1, entry 4). It appears
found for [FeACHTNUTRGNEUNG(NTf2)3] (96% yield to 2; entry 9). The catalyst
can also be prepared from the hexahydrate salt FeCl3·6H2O
(entry 10) and the possible catalytic activity from metal im-
purities in the iron salt was considered negligible because
the use of ultrapure (>99.99%) ironACTHNUTRGNEUNG(III) gave similar re-
sults (entry 11).
To investigate whether the high activity was correlated
with an increase in the softness of the Lewis acidity of FeIII,
Electronic Paramagnetic Resonance (EPR) and UV/Vis
spectroscopy measurements were carried out for FeCl3, [Fe-
then that [FeACHTUNGTRENNUNG(NTf2)3] catalyzes the direct addition of water
to alkyne 1 without the need to form an intermediate vinyl
chloride species and, thus, the use of toxic chlorinated sol-
vents can be avoided. To check this, different solvents were
tested (Table 1, entries 5–13) and, interestingly, those sol-
vents containing low-coordinating oxo-groups gave better
results (Table 1, entries 5–7). The rest of solvents tested, in-
cluding stronger oxo-coordinating (Table 1, entries 8 and 9),
cyano (entry 10), arene (entry 11), aqueous (entry 12), and
amide (entry 13) compounds gave yields of 2 below 50%.
The relatively nontoxic 1,4-dioxane was therefore chosen as
reaction solvent for the study.
N
ACHTUNGTERN(NUNG NTf2)3]. Results in Figure 1 show a gradual
ACHTUNGTRENNUNG
&
2
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Regioselective Hydration of Alkynes
FULL PAPER
Table 2. Catalyst screening for the Markovnikov hydration of phenylace-
tylene (1) with water (3 equiv) to give acetophenone (2).
Entry[a]
Catalyst
FeCl3
2 [%][b]
1
<5
<5
<5
0
<5
53
2[c]
3
FeBr3
FeACTHNGUTRENU(NG acac)3
FeCl3+3AgPF6
FeCl3+3AgSbF6
FeCl3+3AgOTf
4
5[d]
Figure 2. UV/Vis spectra for three ironACTHNUTRGENU(GN III) salts.
6[d]
7[d]
17
of phenylacetylene (1), obtaining yields of acetophenone (2)
of below 5% in all cases (Table 2, entries 16–23, respective-
ly). These results reflect the importance of having a strong
8
Fe
N
16
96
86
68
19
20
21
23
<5
<5
<5
<5
<5
<5
<5
0
56
90
100
99
29
9[d]
FeCl3+3AgNTf2
FeCl3·6H2O+3AgNTf2
FeCl3+3AgNTf2
FeCl2+2AgNTf2
Fe
HOTf
HNTf2
LiNTf2
AgNTf2
CuCl+AgNTf2
CuCl2+2AgNTf2
ZnCl2+2AgNTf2
MnCl2·4H2O+2AgNTf2
CoCl2+2AgNTf2
NiCl2·6H2O+2AgNTf2
RhCl3+3AgNTf2
RuCl3+3AgNTf2
Hg2Cl2+2AgNTf2
AuCl3+3AgNTf2
AuCl+AgNTf2
10[d]
11[c,d]
12[d]
13
14
15
but soft Lewis acid to activate the alkyne. Other metalACHTUNGTRENNUNG(III)
species such as Rh3+ and Ru3+ (isoelectronic to Fe3+) trifli-
mides showed good catalytic activities (Table 2, entries 24
and 25) and, as expected, Hg+ and Au3+ showed complete
conversion of 1 (Table 2, entries 26 and 27). Not surprisingly,
Au+ performed poorly (Table 2, entry 28) because no alco-
hol-assisted mechanism can operate under these reaction
conditions.
16
17
18[d]
19[d]
20[d]
21[d]
22[d]
23[d]
24[d]
25[d]
26[d,e]
27[d]
28[d]
It is remarkable that iron was found to have a comparable
catalytic activity to that of gold—the recognized most active
metal catalyst for the hydration of alkynes after mercury
salts—under our standard reaction conditions. Kinetic ex-
periments indicate a similar initial catalytic activity for both
metals (Figure 3A), even after adding substoichiometric
[a] Standard reaction conditions: Phenylacetylene (1; 0.25 mmol, 28 mL),
water (0.75 mmol, 14 mL), catalyst (0.025 mmol), 1,4-dioxane (0.5 mL),
808C, 20 h. [b] GC yield using dodecane as internal standard. [c] FeCl3
(99.99%) was used. [d] Catalyst was preformed as [MCln
(10 mol%)+nAg salt] in 1,4-dioxane (0.5 mL) for 30 min at room tem-
perature. [e] Reaction was carried with 5 mol% Hg2Cl2.
Figure 1. EPR experiments for three ironACTHNUTRGENU(GN III) salts.
FeCl3 disappears gradually when going to [FeCAHTUNGTRENNNUG
nally to [Fe
species.
ACHTUNGTRENNUNG
FeII triflimide, generated either in situ or not, is mainly in-
active (Table 2, entries 12 and 13) and so are the Brønsted
acids HOTf and HNTf2 at 10 mol% (Table 2, entries 14 and
15). The triflimide metal salts of Li+, Ag+, Cu+, Cu2+, Zn2+
, Mn2+, Co2+, and Ni2+ were also inactive for the hydration
Figure 3. Hydration of 1 in the presence of catalytic amounts of pre-
formed ferric and gold triflimide salts A) without or B) with ethylene
glycol as sub-stoichiometric additive.
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

A. Corma, A. Leyva-Pꢀrez, J. R. Cabrero-Antonino
Table 3. Fe
ACHTUNGTREN(NNUG NTf2)3-catalyzed hydration of different alkynes with water (3 equiv) to
amounts of a chelating hydroxyl-containing mole-
cule such as ethylene glycol to the reaction medium
(Figure 3B).
give the corresponding ketones (C=conversion).
In view of the good catalytic results obtained
with [Fe
ACHTUNGTRENNUNG(NTf2)3], the reactivity of various terminal
(Table 3, entries 1–11) and internal alkynes (en-
tries 12 and 13) was studied under the optimized Entry[a]
conditions, giving good to excellent yields of ketone
after isolation. Aromatic electron-rich 3, 7, 8, and 9
(entries 2, 6, 7, and 8, respectively) and aromatic
electron-poor alkynes 4, 5, and 6 (entries 3, 4, and
Alkyne
Product
t
C
Yield
[%][c]
[h]
[%][b]
1
2
20
40
96
93
94
5, respectively) performed well, and alkyl alkynes
10, 11, and 12 (entries 9, 10, and 11, respectively)
also formed the corresponding aliphatic ketones in
very good yields. Furthermore, the hydration of
symmetrical internal alkynes 13 and 14 (entries 12
and 13) was also accomplished to give the corre-
sponding ketones in good yields. The Markovnikov
product was obtained exclusively in all cases.
2
3
4
5
6
7
3b
100
4b
5b
6b
7b
8b
40
40
60
20
20
98
100
100
100
100
94
80
84
93
99
Given the good reactivity found for internal al-
kynes, and keeping in mind the regioselectivity of
FeIII with respect to the benzylic positions, we ex-
plored the [FeACHTUNGTRENNUNG(NTf2)3]-catalyzed hydration of asym-
metrically substituted aryl-alkyl alkynes. Results in
Table 4 (entries 1–3) show total regioselectivity of
the reaction to give the corresponding phenylke-
tones 16, 18, and 20. In the case of phenylenynones
(Table 4, entry 4) the hydration also occurs regiose-
lectively in the benzylic position, but the 1,3-dicar-
bonyl compound thus obtained undergoes an iron-
8
9b
20
60
60
100
100
92
79
90
79
9[d,e]
10[d,e]
10b
11b
ACHTUNGTRENNUNG
(III)-catalyzed retro-Claisen[15] reaction to form
acetophenone (2) as the only product. For aryl-aryl
asymmetrical internal alkynes (Table 4, entry 5) the
corresponding product was also obtained with full
regioselectivity albeit with lower yield.
Because the hydration of aryl-alkyl asymmetrical
internal alkynes progresses with very good regiose-
lectivity and high yields, resulting in the formation
of butyrophenones, we applied the FeIII-catalyzed
procedure to the synthesis of two commonly-used
antipsychotics with butyrophenone structure, Halo-
peridol (24) and Melperone (25; Scheme 2). Thus,
precursor 26 was prepared by Sonogashira reaction
(see the Supporting Information, Scheme S2) and
11[d,e]
12[f,e]
13[f,e]
12b
13b
14b
60
60
60
100
94
68
86
71
96
[a] Standard reaction conditions: Alkyne (0.5 mmol), water (1.5 mmol, 27 mL), FeCl3
(0.05 mmol, 8 mg), AgNTf2 (0.15 mmol, 60 mg), 1,4-dioxane (1 mL), 808C, 20 h.
[b] GC conversion refers to the disappearance of the alkyne using dodecane as inter-
nal standard. [c] Isolated product yield. [d] Catalyst 15 mol%. [e] Reaction was carried
at 958C. [f] Catalyst 20 mol%.
Table 4. Fe
ACHTUNGTREN(NNUG NTf2)3-catalyzed hydration of different asymmetrical internal alkynes with
submitted to [FeACHTUNGTRENNUNG(NTf2)3]-catalyzed hydration condi-
water (3 equiv) to give the corresponding ketones (C=conversion).
tions to smoothly achieve the common intermediate
27 in high yield and selectivity. Nucleophilic addi-
tion[16] of the corresponding amines gave Haloperi-
dol (24) and Melperone (25) in high yields.
Looking into the possibility of recycling the cata-
lyst, successive reaction cycles were performed
Entry[a] Alkyne
Main product
C
Rs[c]
Yield
[%][d]
[%][b]
with: 1) free [Fe
plex [(PPh3)2Fe(NTf2)3], and 3) the OPPh3/[Fe-
(NTf2)3] system, and also the Brønsted acid HNTf2
ACHTUNGTRENNUNG(NTf2)3], 2) the diphosphane com-
ACHTUNGTRENNUNG
1
2
100
100
>99:<1 80
ACHTUNGTRENNUNG
for comparative purposes (Figure 4).
100:0
92
A study of reusability for the three FeIII catalytic
species showed that triphenylphosphane oxide with
&
4
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Regioselective Hydration of Alkynes
Table 4. (Continued)
FULL PAPER
either [FeACHTUNGTRNEG(UN NTf2)3] or HNTf2 turns out to be the
Entry[a] Alkyne
Main product
C
[%][b]
Rs[c]
Yield
[%][d]
most stable system, being highly active after six
cycles of reaction (additional lines represent system
B). In contrast, when working with [(PPh3)2Fe-
3
4
99
100:0
100:0
99
84
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
100
AHCTUNGTRENNUNG
infer that the presence of OPPh3 in the reaction
medium has a stabilizing effect because the weak
phosphane oxide interaction (M···O=PPh3, M=
Fe3+ or H+) avoids catalyst decomposition. In the
case of the Brønsted acid HNTf2 used alone, al-
though it was certainly active under conditions of
increased temperature and time, it was not recycla-
ble. With these results in hand, we considered again
the possibility that H+ species generated in situ
were responsible for either part or all of the catalyt-
ic activity—a question of particular importance in
5
40
>99:<1 32[e]
[a] Standard reaction conditions: Alkyne (0.5 mmol), water (1.5 mmol, 27 mL), FeCl3
(0.1 mmol, 16 mg), AgNTf2 (0.3 mmol, 120 mg), 1,4-dioxane (1 mL), 958C, 60 h.
[b] GC conversion refers to the disappearance of the alkyne using dodecane as inter-
nal standard. [c] Regioselectivity determined by GC-MS and 1H NMR analysis.
[d] Isolated product yield. [e] GC yield.
metal triflate-triflimide catalyzed-reactions—although the
results in Table 2 (entries 14–15) seemed to indicate the op-
posite.[17] To address this question, the apparent activation
energy (Ea) values (Figure S3 and S6) were calculated for
[FeACTHUNTGRNEGNU(NTf2)3] and HNTf2 from kinetics parameters (Figure S1,
S2, S4, and S5) and the Ea for Fe3+ was found to be about
20 kJmolꢀ1 lower than for H+ (Table 5).
Table 5. Values of apparent activation energies for the Markovnikov hy-
dration of phenylacetylene (1) with water (3 equiv) catalyzed by
(10 mol%) of catalyst.
Catalyst
Fe(NTf2)3
HNTf2
Ea [kJmolꢀ1
]
G
110ꢁ0.1
130ꢁ0.2
Scheme 2. Synthesis of Haloperidol and Melperone with the ironACTHUNTGRNEUNG(III) re-
gioselective hydration of the internal alkyne 26 as a key step.
To further assess this question, we calculated and plotted
(Figure S7, S8, S9, and S10) the initial Turnover Frequencies
(TOF0ꢃs, calculated from initial rates) versus different con-
centrations of Fe3+ and H+ (up to 60 mol%), and significant
differences were found (compare Figure 5 and Figure 6).
Whereas an expected linear trend was determined for
Figure 4. Study of reusability for various catalytic systems for the hydra-
tion of phenylacetylene (1) to give acetophenone (2) with water
(3 equiv.) under different conditions: A) [Fe
(40 mol%), 1008C, 20 h; B) HNTf2 (60 mol%), OPPh3 (40 mol%),
1008C, 20 h; C) [(PPh3)2Fe(NTf2)3] (20 mol%), 1008C, 20 h; D) [Fe-
(NTf2)3] (10 mol%), 808C, 20 h; E) HNTf2 (10 mol%), 1008C, 48 h.
ACHTUGNTRENUN(GN NTf2)3] (20 mol%), OPPh3
Figure 5. TOF0 values versus various mol% of HNTf2 for the hydration
of phenylacetylene (1) to give acetophenone (2) with water (3 equiv) at
808C.
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

A. Corma, A. Leyva-Pꢀrez, J. R. Cabrero-Antonino
mide network and water molecules. Further studies in this
regard are in progress.
Spectroscopic evidence for the possible formation of
HNTf2 from [FeACHTNUTRGNEUNG(NTf2)3] species under the reaction condi-
tions were obtained from in situ 19F NMR measurements
(Figure 8). As can be seen, the peak corresponding to [Fe-
Figure 6. TOF0 values versus different mol% of FeCl3+3ACTHNUGTRENNUG[AgNTf2] for the
hydration of phenylacetylene (1) to give acetophenone (2) with water
(3equiv) at 808C.
HNTf2 (Figure 5), a rather unusual variation of the TOF0
was found for [FeACHTUNGTRENNUNG(NTf2)3] (Figure 6), showing no saturation
Figure 8. 19F NMR spectra of different triflimides recorded during the hy-
dration of phenylacetylene (1) with water (3 equiv) to give acetophenone
(2).
values at high amounts of catalyst but, instead, an exponen-
tial increase. This could be explained by the increasing
Brønsted acidity of the partially hydrated [Fe
species.
ACHTUGTNRENNUG(H2O)nACHTUGNTERN(NUGN NTf2)3]
AHCTUNTGRENGUN(N NTf2)3] disappears in the presence of water and a new peak
To check whether the presence of the triflimide anion is
responsible for an increase in the catalytic activity of [Fe-
ACHTUNGTRENNUNG(NTf2)3], additional experiments (Figure S11 and S12) were
(dꢂꢀ78.5 ppm), very proximal to that corresponding to
HNTf2 (dꢂꢀ78.0 ppm) appears during the reaction, disap-
pearing at the end of the reaction. Concomitantly, new sig-
nals (dꢂꢀ70.0 ppm and ꢀ72.0 ppm), perhaps corresponding
to iron triflimide-type compounds, appear at the end of the
reaction. These observations would be in accordance with
the possible reaction mechanism depicted in Scheme 3.
Addition of 2,6-di-tert-butylpyridine completely inhibits
the reaction, which confirms the need for H+ for the reac-
tion to proceed. Therefore, we must conclude that a Brønst-
carried out in which the catalytic activity for the hydration
of 1 was measured after adding nonactive triflimide salts
(Figure 7). It was found that the reaction rate was 2–3 times
higher when an equimolar amount of another triflimide salt
was added, regardless the nature of the cation (Brønsted H+
, Lewis Zn2+, organic imidazolium). Thus, it seems that the
increase in the catalytic activity of the Lewis catalyst could
be related to an acidity enhancement caused by the trifli-
Figure 7. TOF0 values of different catalytic species for the hydration of
phenylacetylene (1) to give acetophenone (2) with water (3 equiv) at
808C. In all cases, the anion was ꢀNTf2 and the reaction was catalyzed by
10 mol% of each species.
Scheme 3. Tentative mechanism for the Markovnikov hydration of phe-
nylacetylene (1) with water (3 equiv.) to give acetophenone (2), catalyzed
by [FeACHTNUGRTNEUNG
(NTf2)3]. M=Fe3+ and/or H+.
&
6
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Regioselective Hydration of Alkynes
FULL PAPER
Compound 27: FeCl3 (97%, Aldrich, 85.0 mg, 10 mol%) and AgNTf2
(97%, Aldrich, 624.0 mg, 30 mol%) were placed in a 25 mL round-bot-
tomed flask, 1,4-dioxane (5 mL) was added and the mixture was magneti-
cally stirred at RT for 30 min. Compound 26 (430 mL, 2.6 mmol) and
water (140 mL, 7.8 mmol) were added and the mixture was placed in
a pre-heated oil bath at 808C and magnetically stirred for 18 h. After
cooling, n-hexane (60 mL) was added (precipitation of the catalyst was
observed) and the liquid was filtered under gravity. The filtrate was con-
centrated under reduced pressure and the resulting yellow oil residue
was purified by flash column chromatography on silica (hexane/EtOAc,
80:20). Fractions were concentrated under reduced pressure and, after
drying under vacuum, the resulting residue compound 27 (406 mg, 85%)
was obtained as a yellow oil. Rf =0.66 (20% EtOAc/hexane). GC-MS: m/
z (%): 200 (1) [M]+·, 164 (1), 138 (43), 123 (100), 107 (10); 1H NMR
(300 MHz, CDCl3): d=8.03–7.97 (m, 2H), 7.17–7.09 (m, 2H), 3.68 (t, J=
6.2 Hz, 2H), 3.15 (t, J=7.0 Hz, 2H), 2.22 ppm (tt, J=7.0, 6.2 Hz, 2H);
ed/Lewis co-catalysis seems to operate under these reaction
conditions. Nevertheless, a neat Brønsted catalysis after hy-
drolysis of the ironACHTUNGTRENNUNG(III) triflimide salt cannot be discarded.
Conclusion
The regioselective Markovnikov hydration of terminal and
internal aryl-alkyl alkynes is achieved by using iron(III) tri-
AHCTUNGTRENNUNG
flimide compounds in catalytic amounts. Hydrolysis of the
salt readily occurs to give catalytically active protons. Kinet-
ic experiments point to a possible Lewis/Brønsted co-cataly-
sis or even just a neat Brønsted catalysis. The use of triphe-
nylphosphane oxide allows the catalyst to be recycled. The
regioselective hydration in the benzylic position gives access
to a formal acylation of aromatic rings, and the synthesis of
Haloperidol and Melperone illustrates the potential use of
this synthetic protocol in organic synthesis.
13C NMR (300 MHz, CDCl3): d=197.3 (C=O), 166.1 (d, 1J
ACHTUNGTRENNUNG
254.7 Hz; C-F), 133.2 (d, 4J(C,F)=4.3 Hz; C), 130.6 (d, 3J
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
2 ꢄCH), 115.9 (d, 2J
ACHTUNGTRENNUNG
26.7 ppm (CH2).
Haloperidol (24): Compound 27 (150 mg, 0.75 mmol), 4-(4-chlorophen-
yl)-4-hydroxypiperidine (321 mg, 1.5 mmol), and KI (2 mg, 1 mol%)
were placed in
a 10 mL cylindrical sealed vial. Anhydrous toluene
(2.5 mL) was added and the mixture was placed in a pre-heated oil bath
at 1258C and magnetically stirred for 25 h. After cooling, the reaction
mixture was washed with EtOAc (3ꢄ10 mL) and the liquid was filtered
under gravity. The filtrate was concentrated under reduced pressure and
the resulting brown oil residue was purified by flash column chromatog-
raphy on silica (EtOAc/MeOH/NH4OH, 95:5:1). Fractions were concen-
trated under reduced pressure and, after drying under vacuum, Haloperi-
dol 24 (168 mg, 60%) was obtained as a brown solid. Rf =0.6 (EtOAc/
MeOH/NH4OH, 95:5:1). GC-MS: m/z (%): 375 (1) [M]+·, 237 (98), 224
(100), 206 (26), 165 (8), 123 (16), 95 (7); 1H NMR (300 MHz, CDCl3):
d=8.03–7.95 (m, 2H), 7.41–7.35 (m, 2H), 7.32–7.26 (m, 2H), 7.17–7.08
(m, 2H), 3.53 (br s, OH), 3.01 (t, J=6.8 Hz, 2H), 2.97–2.88 (m, 2H),
2.67–2.55 (m, 4H), 2.20–2.07 (m, 2H), 2.06–1.98 (m, 2H), 1.77–1.67 ppm
(m, 2H); 13C NMR (300 MHz, CDCl3): d=198.0 (C=O), 164.0 (d, 1J-
Experimental Section
General methods: Reagents and solvents were obtained from commercial
sources (Aldrich) and were used without further purification otherwise
indicated. All the products obtained were characterised by GC-MS, 1H-
and 13C-NMR, and DEPT. When available, the characterisation given in
the literature was used for comparison. Gas chromatographic analyses
were performed in an instrument equipped with a 25 m capillary column
of 5% phenylmethylsilicone. GC/MS analyses were performed on a spec-
trometer equipped with the same column as the GC and operated under
1
the same conditions. H, 13C, DEPT and 19F were recorded in a 300 MHz
instrument using CDCl3 as solvent containing TMS as internal standard.
IR spectra of the compounds were recorded on a spectrophotometer by
impregnating the windows with a dichloromethane solution of the com-
pound and leaving to evaporate before analysis. Absorption spectra were
recorded on an UV/Vis spectrophotometer (UV0811M209, Varian) and
EPR experiments were carried out with a Bruker EMX/X equipment.
A
ACHTUNGERTN(NUNG C,F)=2.7 Hz; C), 132.9
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
(2 ꢄCH2-N), 37.6 (2 ꢄCH2), 36.0 (CH2), 21.0 ppm (CH2).
Melperone (25): Compound 27 (150 mg, 0.75 mmol), 4-methylpiperidine
(280 mL, 1.5 mmol) and KI (2 mg, 1 mol%) were placed in a 10 mL cylin-
drical sealed vial. Anhydrous toluene (2.5 mL) was added and the mix-
ture was placed in a pre-heated oil bath at 1258C and magnetically
stirred for 25 h. After cooling, the reaction mixture was washed with
EtOAc (3ꢄ10 mL) and the liquid was filtered under gravity. The filtrate
was concentrated under reduced pressure and the resulting brown oil res-
idue was purified by flash column chromatography on silica (EtOAc/
MeOH/NH4OH, 95:5:1). The fractions were concentrated under reduced
pressure and, after drying under vacuum, Melperone 25 (171 mg, 87%)
was obtained as a brown solid. Rf =0.5 (EtOAc/MeOH/NH4OH, 95:5:1).
GC-MS: m/z (%): 263 (1) [M]+·, 165 (4), 125 (54), 112 (100), 95 (9), 70
(6); 1H NMR (300 MHz, CDCl3): d=8.00–7.94 (m, 2H), 7.13–7.05 (m,
2H), 2.99–2.85 (m, 4H), 2.42 (t, J=7.3 Hz, 2H; CH2-CO), 2.02–1.88 (m,
4H), 1.64–1.53 (m, 2H), 1.42–1.27 (m, 1H), 1.26–1.13 (m, 2H), 0.88 ppm
(d, J=6.2 Hz, 3H); 13C NMR (300 MHz, CDCl3): d=198.3 (C=O), 165.5
Typical reaction procedure for acetophenone (2) (Table 3, entry 1): FeCl3
(97%, Aldrich, 8.0 mg, 10 mol%) and AgNTf2 (97%, Aldrich, 60.0 mg,
30 mol%) were placed in a 2 mL vial, and 1,4-dioxane (1 mL) was added.
The mixture was magnetically stirred at RT for 30 min, then phenylacety-
lene (56 mL, 0.5 mmol) and water (28 mL, 1.5 mmol) were added and the
mixture was placed in a pre-heated oil bath at 808C and magnetically
stirred for 20 h. After cooling, n-hexane (20 mL) was added (precipita-
tion of the catalyst was observed). The liquid was filtered under gravity
and the filtrate was concentrated under reduced pressure to give 2
(56 mg, 93%) as a yellow oil. GC-MS: m/z (%): 120 (40) [M]+·, 105
(100), 77 (40), 51 (40); 1H NMR (300 MHz, CDCl3): d=7.94–7.88 (m,
2H), 7.55–7.47 (m, 1H), 7.45–7.37 (m, 2H), 2.56 ppm (s, 3H).
Typical reaction procedure for reusing the OPPh3/[FeACTHNURTGNE(NUG NTf2)3] system
(Figure 4, A): FeCl3 (97%, Aldrich, 16.0 mg, 20 mol%), AgNTf2 (97%,
Aldrich, 120.0 mg, 60 mol%) and OPPh3 (55.6 mg, 40 mol%) were
placed in a 10 mL round-bottomed flask, then 1,4-dioxane (1.5 mL) was
added. The mixture was magnetically stirred at RT for 30 min, then phe-
nylacetylene (56 mL, 0.5 mmol) and water (28 mL, 1.5 mmol) were added
and the mixture was placed in a pre-heated oil bath at 1008C and mag-
netically stirred for 20 h. After cooling, n-hexane (20 mL) was added
(precipitation of the catalyst was observed). The liquid was decanted, the
remaining catalyst was washed with hexane, and the liquid was decanted
again. The combined organic phases were analyzed by GC. The catalyst
was dried for 4 h and used in a subsequent run.
(d, 1J
(C,F)=9.3 Hz; 2 ꢄCH), 115.5 (d, 2J
ACHTUNGTRENNUNG
N
(C,F)=3.3 Hz; C), 130.6 (d, 3J-
ACHTUNGTRENNUNG
U
N), 53.6 (2 ꢄCH2-N), 36.2 (CH2), 33.8 (2 ꢄCH2), 30.6 (CH), 21.7 (CH3),
21.4 ppm (CH2).
Acknowledgements
Synthesis of haloperidol (24) and melperone (25) from compound 26
(Scheme 2)
The work has been supported by Consolider-Ingenio 2010 (proyecto
MULTICAT), and PROMETEO from Generalitat Valenciana. J.R.C.A.
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

A. Corma, A. Leyva-Pꢀrez, J. R. Cabrero-Antonino
thanks MCIINN for a pre-doctoral FPU fellowship. A.L.P. thanks ITQ
for financial support. We thank Dr. J. A. Vidal-Moya for the EPR meas-
urements.
ki, J. Am. Chem. Soc. 2001, 123, 11917; f) T. Suzuki, M. Tokunaga,
Y. Wakatsuki, Org. Lett. 2001, 3, 735; g) D. B. Grotjahn, C. D. Incar-
vito, A. L. Rheingold, Angew. Chem. 2001, 113, 4002; Angew. Chem.
Int. Ed. 2001, 40, 3884; h) P. Alvarez, M. Bassetti, J. Gimeno, G.
Mancini, Tetrahedron Lett. 2001, 42, 8467; i) M. Tokunaga, Y. Wakat-
suki, Angew. Chem. 1998, 110, 3024; Angew. Chem. Int. Ed. 1998,
37, 2867; j) C. Bruneau, P. H. Dixneuf, Chem. Commun. 1997, 507.
[7] For mercury-based hydration of acetylene, see: Ullmannꢀs Encyclo-
pedia of Industrial Chemistry, 7th ed., Wiley, Weinheim, 2006.
[8] J. R. Cabrero-Antonino, A. Leyva-Pꢀrez, A. Corma, Adv. Synth.
Catal. 2012, 354, 678.
[9] L. Coulombel, F. Grau, M. Weꢆwer, I. Favier, X. Chaminade, A.
Heumann, J. C. Bayꢅn, P. A. Aguirre, E. DuÇach, Chem. Biodivers-
ity 2008, 5, 1070.
[10] a) J. R. Cabrero-Antonino, A. Leyva-Pꢀrez, A. Corma, Adv. Synth.
Catal. 2010, 352, 1571; b) K. Kohno, K. Nakagawa, T. Yahagi, J.-C.
Choi, H. Yasuda, T. Sakakura, J. Am. Chem. Soc. 2009, 131, 2784;
c) A. Correa, O. Garcia Mancheno, C. Bolm, Chem. Soc. Rev. 2008,
37, 1108; d) S. Enthaler, K. Junge, M. Beller, Angew. Chem. 2008,
120, 3363; Angew. Chem. Int. Ed. 2008, 47, 3317; e) C. Bolm, J.
Legros, J. Le Paih, L. Zani, Chem. Rev. 2004, 104, 6217.
[11] a) S. Antoniotti, V. Dalla, E. DuÇach, Angew. Chem. Int. Ed. 2010,
49, 7860; b) Y. Nakamura, T. Maki, X. Wang, K. Ishihara, H. Yama-
moto, Adv. Synth. Catal. 2006, 348, 1505; c) B. Mathieu, L. Ghosez,
Tetrahedron 2002, 58, 8219; d) J. Nie, H. Kobayashi, T. Sonoda,
Catal. Today 1997, 36, 81.
[12] a) X.-F. Wu, D. Bezier, C. Darcel, Adv. Synth. Catal. 2009, 351, 367;
b) P. O. Miranda, D. D. Dꢁaz, J. I. Padrꢅn, M. A. Ramꢁrez, V. S.
Martꢁn, J. Org. Chem. 2005, 70, 57. For a stoichiometric FeCl3 hydra-
tion of alkynes, see: c) J. P. Damiano, M. L. Postel, J. Organomet.
Chem. 1996, 522, 303.
[13] a) T. Nagano, T. Hayashi, Org. Lett. 2005, 7, 491; b) G. R. Cahiez, C.
Chaboche, F. Mahuteau-Betzer, M. Ahr, Org. Lett. 2005, 7, 1943.
[14] T. Tsuchimoto, T. Joya, E. Shirakawa, Y. Kawakami, Synlett 2000,
1777.
[15] a) C. B. Rao, D. C. Rao, D. C. Babu, Y. Venkateswarlu, Eur. J. Org.
Chem. 2010, 2855; b) S. Biswas, S. Maiti, U. Jana, Eur. J. Org. Chem.
2010, 2861.
[1] For a recent comprehensive review on hydration of alkynes, see: L.
Hintermann, A. Labonne, Synthesis 2007, 1121.
[2] For reviews on general hydroadditions to alkynes, see: a) A. Corma,
A. Leyva-Pꢀrez, M. J. Sabater, Chem. Rev. 2011, 111, 1657; b) F.
Alonso, I. P. Beletskaya, M. Yus, Chem. Rev. 2004, 104, 3079.
[3] The Hg-based Kutscheroff and Hennion-Nieuwland systems are
classical catalysts of choice for the hydration of alkynes, see:
a) D. B. Killian, G. F. Hennion, J. A. Nieuwland, J. Am. Chem. Soc.
1936, 58, 80; b) D. B. Killian, G. F. Hennion, J. A. Nieuwland, J. Am.
Chem. Soc. 1934, 56, 1786; c) G. F. Hennion, D. B. Killian, T. H.
Vaughn, J. A. Nieuwland, J. Am. Chem. Soc. 1934, 56, 1130; d) J. A.
Nieuwland, R. R. Vogt, W. L. Foohey, J. Am. Chem. Soc. 1930, 52,
1018; e) M. G. Kutscheroff, Ber. Dtsch. Chem. Ges. 1909, 42, 2759;
f) M. Kutscheroff, Ber. Dtsch. Chem. Ges. 1884, 17, 13.
[4] For a seminal example of palladium-catalyzed hydration of alkynes,
see: a) K. Utimoto, Pure Appl. Chem. 1983, 55, 1845; b) W. Hiscox,
P. W. Jennings, Organometallics 1990, 9, 1997; For a very recent
silver-catalyzed example, see: c) M. B. T. Thuong, A. Mann, A.
Wagner, Chem. Commun. 2012, 48, 434.
[5] For alcohol-assisted gold-catalyzed hydration of alkynes, see: a) A.
Corma, V. R. Ruiz, A. Leyva-Pꢀrez, M. J. Sabater, Adv. Synth.
Catal. 2010, 352, 1701; b) N. Marion, R. S. Ramꢅn, S. P. Nolan, J.
Am. Chem. Soc. 2009, 131, 448; c) A. Leyva, A. Corma, J. Org.
Chem. 2009, 74, 2067; d) L. L. Santos, V. R. Ruiz, M. J. Sabater, A.
Corma, Tetrahedron 2008, 64, 7902–7909; e) P. Roembke, H.
Schmidbaur, S. Cronje, H. Raubenheimer, J. Mol. Catal. A: Chem
2004, 212, 35; f) R. Casado, M. Contel, M. Laguna, P. Romero, S.
Sanz, J. Am. Chem. Soc. 2003, 125, 11925; g) E. Mizushima, K. Sato,
T. Hayashi, M. Tanaka, Angew. Chem. 2002, 114, 4745; Angew.
Chem. Int. Ed. 2002, 41, 4563; h) J. H. Teles, S. Brode, M. Chabanas,
Angew. Chem. 1998, 110, 1475; Angew. Chem. Int. Ed. 1998, 37,
1415; i) Y. Fukuda, K. Utimoto, J. Org. Chem. 1991, 56, 3729.
[6] For ruthenium-catalyzed anti-Markovnikov hydration of alkynes,
see: a) F. Boeck, T. Kribber, L. Xiao, L. Hintermann, J. Am. Chem.
Soc. 2011, 133, 8138; b) A. L. Labonne, T. Kribber, L. Hintermann,
Org. Lett. 2006, 8, 5853; c) F. Chevallier, B. Breit, Angew. Chem.
2006, 118, 1629; Angew. Chem. Int. Ed. 2006, 45, 1599; d) D. B.
Grotjahn, D. A. Lev, J. Am. Chem. Soc. 2004, 126, 12232; e) M. To-
kunaga, T. Suzuki, N. Koga, T. Fukushima, A. Horiuchi, Y. Wakatsu-
[16] C. S. Kook, M. F. Reed, G. A. Digenis, J. Med. Chem. 1975, 18, 533.
[17] a) D. C. Rosenfeld, S. Shekhar, A. Takemiya, M. Utsunomiya, J. F.
Hartwig, Org. Lett. 2006, 8, 4179; b) T. T. Dang, F. Boeck, L. Hinter-
mann, J. Org. Chem. 2011, 76, 9353.
Received: February 22, 2012
Published online: && &&, 0000
&
8
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Regioselective Hydration of Alkynes
FULL PAPER
Just add water! The triflimide iron
salt [Fe(NTf2)3] promotes the direct
hydration of terminal and internal
alkynes with very good Markovnikov
regioselectivities and high yields (see
scheme).
ACHTUNGTNER(NUNG III)
Regioselective Hydration
ACHTUNGTRENNUNG
J. R. Cabrero-Antonino,
A. Leyva-Pꢁrez,*
A. Corma* .................... &&&& — &&&&
Regioselective Hydration of Alkynes
by IronACTHNUGRTNEUNG(III) Lewis/Brønsted Catalysis
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!