Highly efficient Rh(i) and Ir(i) single and dual metal catalysed dihydroalkoxylation reactions of alkyne diols

Article abstract of DOI:10.1039/b926773e

A highly efficient rhodium(i) and iridium(i) catalysed dihydroalkoxylation reaction of alkyne diols is employed here for the synthesis of spiroketals and a fused bicyclic ketal. The two metal catalysts show differential selectivity and efficiency for either the cyclisation of the 5-exo or 6-endo-membered rings. For the first time, a dual metal (Rh and Ir) catalyst system is effectively utilised for the formation of the 5,6-spiroketals, more efficiently than the single metal catalysts. The two different metals create a dual activation pathway to enhance the 5- and 6-membered ring closure as compared with the equivalent single catalysts.

Full text of DOI:10.1039/b926773e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Highly efficient Rh(I) and Ir(I) single and dual metal catalysed
dihydroalkoxylation reactions of alkyne diols†
a
a
b
a
Joanne Hui Hui Ho, Richard Hodgson, J o¨ rg Wagler and Barbara Ann Messerle*
Received 21st December 2009, Accepted 4th February 2010
First published as an Advance Article on the web 9th March 2010
DOI: 10.1039/b926773e
A highly efficient rhodium(I) and iridium(I) catalysed dihydroalkoxylation reaction of alkyne diols is
employed here for the synthesis of spiroketals and a fused bicyclic ketal. The two metal catalysts show
differential selectivity and efficiency for either the cyclisation of the 5-exo or 6-endo-membered rings.
For the first time, a dual metal (Rh and Ir) catalyst system is effectively utilised for the formation of the
5
,6-spiroketals, more efficiently than the single metal catalysts. The two different metals create a dual
activation pathway to enhance the 5- and 6-membered ring closure as compared with the equivalent
single catalysts.
Introduction
important. We have explored the use of rhodium(I) and irid-
ium(I) complexes, [Rh(bim)(CO) ]BPh and [Ir(PyP)(CO) ]BPh
containing the bidentate ligands, bim (bis(N-methylimidazol-2-
2
4
2
4
,
Spiroketals are an important class of heterocyclic molecules which
are found as substructures in a variety of biologically active
compounds, including HIV protease inhibitors, insect pheromones
and polyether antibiotics. Traditional approaches to the synthesis
of these bicyclic O-heterocycles often involve multi-step syntheses
and harsh reaction conditions.
transition metal-catalysed addition of oxygen nucleophiles across
an unsaturated C–C bond (hydroalkoxylation). This tandem, two-
step intramolecular dihydroalkoxylation reaction of alkyne diols is
an effective approach to bicyclic O-heterocycle synthesis because
the reactions are carried out with 100% atom efficiency and can
be highly efficient in terms of yield and reaction time (Fig. 1).
yl)methane) and PyP (1-[2-(diphenylphosphino)ethyl]pyrazole) as
8
active catalysts for the cyclisation of alkyne diols. Crabtree et al.
1
9
have also reported the use of an iridium(III) hydride for catalysing
the dihydroalkoxylation reaction of alkyne diols.
1a,1b,2
The most direct route is the
The development of catalysts which promote one-pot sequential
reactions is an important goal. In particular, the development
of multiple metal catalysts that promote transformations of two
or more sequential reaction steps either where different metals
promote successive one step reactions, or where the different
catalysts cooperatively promote the sequential one-pot process
3
10
is potentially very important. There are isolated examples of
the simultaneous use of two different metal catalysts for efficiently
10
promoting two sequential reaction steps. One example is the two-
step, one-pot production of bicyclopentenone which relies on the
Pd catalyst to promote an X–C coupling in the first step and the
Rh catalyst to promote the C–C coupling reaction in the second
10c
Fig. 1 Tandem, two-step intramolecular dihydroalkoxylation reaction of
alkyne diols.
step.
We have recently described the synthesis of a series of Rh(I)
and Ir(I) complexes containing the bis(1-pyrazolyl)methane (bpm)
Although the transition metal-catalysed intramolecular dihy-
droalkoxylation of alkyne diols is a highly efficient approach to the
synthesis of spiroketals and fused bicyclic ketals, there are only lim-
ited reports of this reaction, most likely due to the relatively large
bond enthalpies associated with typical O–H s-bonds together
11,12
ligand and their reactivity as hydroamination catalysts.
Here,
we report the catalytic activity of these complexes for the
intramolecular dihydroalkoxylation reaction of alkyne diols and
F
F
the crystal structure of [Rh(bpm)(CO)
2
]BAr
4
(3) (where BAr
4
is tetrakis[3,5-bis(trifluoromethyl)phenyl]borate). Also disclosed
herein is the first report on the use of a dual metal (Rh and Ir)
catalyst system for the catalysed dihydroalkoxylation reaction.
4
with the relatively low nucleophilicity of the oxygen atom. Previ-
ously reported metal catalysts used in the dihydroalkoxylation of
5
6
alkyne diols include palladium(II) and platinum(II) complexes,
e.g. PdCl
2
(PhCN)
2
and PdCl , as well as gold(I) and gold(III)
2
7
halides. To ultimately achieve control over product selectivity,
the design of metal catalysts with appropriate ligands will be
a
School of Chemistry, University of New South Wales, Kensington, NSW
2
052, Australia; Fax: (+61)2-93856141. E-mail: b.messerle@unsw.edu.au
b
Research School of Chemistry, Australian National University, Canberra,
ACT
†
CCDC reference number 697499. For crystallographic data in CIF or
other electronic format see DOI: 10.1039/b926773e
4
062 | Dalton Trans., 2010, 39, 4062–4069
This journal is © The Royal Society of Chemistry 2010

View Article Online
Table 1 Conversion of alkyne diol 7 into products 8 and 9 (Fig. 3), 1 mol%
Results and discussion
◦
a
catalyst, C
2 2 4
D Cl , NMR scale, 100 C
F
Crystal structure of [Rh(bpm)(CO)
2
]BAr (3)
4
Entry
Catalyst
t/h
TOF/h-
1b
A single crystal of 3 suitable for X-ray diffraction analysis was
obtained by layering hexane over a concentrated dichloromethane
solution of 3. The ORTEP representation of the structure showing
the cationic complex 3 is shown in Fig. 2. The solid-state
structure of 3 reveals a square planar geometry around the Rh
metal center, bearing two N-atoms of the bis(1-pyrazolyl)methane
ligand and the two C-atoms of the carbonyl co-ligands. The six-
membered metallocycle adopts a distorted boat conformation
and the directions of the C4-bound hydrogen atoms indicate
the envelope shape of the RhNNCNN chelate. The N–Rh–N
F
4
1
2
3
4
[Rh(bpm)(CO)
[Rh(bpm)(CO)
2
]BAr
4
3
0.22
21.00
0.58
2.50
961
90
374
58
2
]BPh
1
4
F
[Ir(bpm)(CO)
[Ir(bpm)(CO)
2
]BAr
]BPh
4
2
4
2
a
1
Determined by H NMR spectroscopy to reach 98% total conversion to
spiroketals 8 and 9. TOF (h ) is amount of product/mole of catalyst/h
at the point of 50% conversion of substrate to product.
b
-1
◦
bond angle of 88 implies only a slight distortion from the
◦
ideal 90 value and indicates a very low level of ring strain due
to the binding of the bidentate bis(1-pyrazolyl)methane ligand.
The Rh–N and Rh–C bond lengths and bond angles of 3 are
found to be comparable to the previously reported structures of
Rh/Ir(I) complexes containing the bis(1-pyrazolyl)methane and
Fig. 3 Dihydroalkoxylation of alkyne diol 7 to form two spiroketals, 8
and 9.
11
bis(N-methyl-2-imidazolyl)methane ligands.
F
-
Complexes 3 and 4, which contain the BAr
4
anion, were
much more efficient catalysts than the analogous complexes 1
-
and 2 containing the BPh
4
anion (Table 1). The reaction rates for
the cyclisation of 7 were compared using the turnover frequency
TOF) at the point of 50% conversion of substrate to product.
(
-
1
The TOF values obtained for complexes 3 and 4 are 961 h and
3
those obtained for complexes 1 and 2 (Table 1, 90 h and 58 h
-
1
74 h respectively (Table 1, entries 1 and 3), much higher than
-
1
-1
respectively, entries 2 and 4). In addition, the reaction times taken
to achieve quantitative conversion (98%) of the spiroketal products
F
-
were much shorter in the case of the BAr
4
containing complexes
(
0.22 h and 0.58 h for complexes 3 and 4 respectively) as compared
-
F -
to the analogous BPh
in place of the BPh
4
complexes. The use of the BAr
anion remarkably enhanced the catalytic
4
anion
Fig. 2 ORTEP representation and atom numbering scheme for the
cationic fragment of complex 3 in the solid-state (thermal ellipsoids at
the 50% probability level). Selected bond lengths [ A˚ ] and angles [ ]:
-
4
◦
activities of the Rh(I) and Ir(I) complexes bearing the bis(1-
Rh1–N1 2.078(2), Rh1–N4 2.074(2), Rh1–C8 1.860(3), Rh1–C9 1.867(3);
N1–Rh1–N4 87.59(8).
pyrazolyl)methane ligand for the tandem dihydroalkoxylation
F
-
reaction of alkyne diols. This is likely because the BAr
4
anion is
a more weakly coordinating anion and provides better separation
of the ions in solution, thus allowing greater availability of the
metal centre for substrate binding and in return, leading to a more
Catalysis
13
active catalyst. In addition to the weak coordinating strength of
Dihydroalkoxylation of alkyne diols
F
-
the BAr
4
anion, the low nucleophilicity, chemical inertness and
F
-
high solubility properties make the BAr
attractive anion candidate.
4
anion an even more
2
-(5-Hydroxypent-1-ynyl)benzyl alcohol (7) was chosen as a
model substrate to assess the efficiency of the transition metal
Rh(I)/Ir(I)) catalysed dihydroalkoxylation reaction of alkyne
(
Co-ligand variation. Rh(I) and Ir(I) complexes (3–6) contain-
ing dicarbonyl or 1,5-cyclooctadiene (COD) co-ligands were tested
as catalysts for the dihydroalkoxylation reaction of alkyne diol 7
diols to promote the formation of spiroketals (8 and 9) (Fig. 3). In
order to find the best catalyst to achieve the highest conversion of
the spiroketal products, we screened a series of Rh(I) and Ir(I) bis(1-
pyrazolyl)methane complexes containing different anions and co-
ligands.
(
Fig. 3). The reactions were carried out with 1 mol% catalyst,
◦
at 120 C in C
2
D
2
Cl
4
as solvent (Table 2). The high reaction
◦
temperature of 120 C is used in this set of experiments because
of the slow reaction progress when using the COD complexes (5
and 6).
Anion variation. The efficiency of cationic complexes,
[
(
Rh(bpm)(CO)
2
]BPh
4
(1), [Ir(bpm)(CO)
2
]BPh
4
F
(2), [Rh(bpm)-
(4) (where
F
CO)
2
]BAr
and BAr
4
(3) and [Ir(bpm)(CO)
2
]BAr
4
The most active catalysts were complexes 3 and 4 bearing
the dicarbonyl co-ligands, with quantitative conversion (98%)
to the spiroketals achieved in less than 0.5 h (Table 2, entries
F
BPh
4
4
are tetraphenylborate and tetrakis[3,5-
bis(trifluoromethyl)phenyl]borate respectively) as catalysts for the
intramolecular dihydroalkoxylation of alkyne diol (7) to form
-
1
-1
1 and 3, TOF = 1077 h and 1157 h respectively). These
two complexes were found to be much more active catalysts for
the dihydroalkoxylation reaction than previously reported Rh
spiroketals (8 and 9) was investigated using 1 mol% catalyst at
◦
1
00 C in C
2
D
2
4
Cl as solvent (Table 1).
This journal is © The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4062–4069 | 4063

View Article Online
Table 2 Conversion of alkyne diol 7 into products 8 and 9 (Fig. 3), 1 mol%
Table 3 Conversion of alkyne diol 7 into products 8 and 9 (Fig. 3), 1 mol%
◦
a
a
catalyst, C
2
D
2
Cl
4
, NMR scale, 120 C
catalyst, C
2
D
2
Cl
4
, NMR scale, unless otherwise stated
T/ C t/h Rel. ratio (8 : 9) TOF/h-1b
◦
Rel. ratio
Conv. [%] t/h (8 : 9)
Entry Catalyst
Entry Catalyst
TOF/h-
1b
c
1
2
25
18.1 2.4 : 1
7.5 1.1 : 1
1.8 0.8 : 1
0.2 0.7 : 1
0.3 0.4 : 1
10
40
F
c
1
2
3
4
[Rh(bpm)(CO)
2
]BAr
4
3
4
98
83
97
60
0.3 0.4 : 1
25.2 0.0 : 1
0.1 0.8 : 1
96.0 0.2 : 1
1077
3
1157
<1
60
80
F
[Rh(bpm)(COD)]BAr
5
3
4
5
255
961
1077
F
[Ir(bpm)(CO)
2
]BAr
4
4
100
120
F
[Ir(bpm)(COD)]BAr
4
6
a
1
Determined by H NMR spectroscopy to reach 98% total conversion to
b
-1
spiroketals 8 and 9. TOF (h ) is amount of product/mole of catalyst/h
at the point of 50% conversion of substrate to product.
6
7
8
80
100
120
2.1 1.2 : 1
0.6 0.9 : 1
0.1 0.8 : 1
89
374
1157
8,14
and Ir(I) complexes. Under the same reaction conditions, the
complexes containing the 1,5-cyclooctadiene co-ligand, 5 and
6, promoted the reaction with much better selectivity than the
analogous dicarbonyl complexes (entries 2 and 4), with exclusive
formation of spiroketal 9 in the case of complex 5. The very low
activity and yet high regioselectivity observed for complexes 5 and
a
1
Determined by H NMR spectroscopy to reach 98% total conversion to
6
in the dihydroalkoxylation of alkyne diol 7 was attributed to
b
-1
spiroketals 8 and 9. TOF (h ) is amount of product/mole of catalyst/h
at the point of 50% conversion of substrate to product. CDCl
solvent.
the steric bulk of the 1,5-cyclooctadiene co-ligand, which inhibits
substrate binding but leads to selectivity of product formation.
The low activity of complexes 5 and 6 could also be attributed
to the lowered p-acceptor capability of the alkene donor function
of the COD which would inhibit (or suppress) coordination of
another donor moiety
c
3
used as
spiroketal 9 is the thermodynamically more favoured product at
higher temperatures.
Regioselectivity
An interesting feature of the dihydroalkoxylation reaction of
alkyne diols is the potential for regioselectivity towards either 5-
exo or 6-endo ring cyclisation, which are both allowed by Baldwin’s
15
rules, and in particular in the two-step cyclisation process, the
potential for selectivity in the formation of mixtures of products.
Here, we investigated the variation in the relative quantities
of the two regioisomeric spiroketal products (8 and 9) formed
at different reaction temperatures from the catalysed intra-
molecular dihydroalkoxylation of alkyne diol, 7 (Fig. 3, Table 3).
F
F
[
Rh(bpm)(CO)
2
]BAr
4
(3) and [Ir(bpm)(CO)
2
]BAr
4
(4) were
chosen as the catalysts for investigation in this study as they
were shown above to be the most effective catalysts for the
dihydroalkoxylation reaction of alkyne diol 7.
Fig. 4 Relative rate of conversion for the dihydroalkoxylation of 7 to form
regioisomeric spiroketals 8 (left) and 9 (right) using 1 mol% of complex 3
at various temperatures.
The Rh complex (3) was more efficient than the Ir complex
◦
(
4) at temperatures below 120 C and the difference between
the catalytic activity of complexes 3 and 4 was greater at lower
Variation in substrate–ring size or alkyne substituent?
temperatures. More importantly, the catalysed reaction progresses
F
even at room temperature with 1 mol% of [Rh(bpm)(CO)
2
]BAr
4
The reactivity and regioselectivity of the intramolecular dihy-
droalkoxylation reaction can be influenced by the initial open
chain structure as well as the nature of the transition state that
leads to the cyclised product. One factor that can also affect the
efficiency of the catalysed cyclisation reaction of alkyne diols is the
(
3), although a longer reaction time (18 h) was required for
quantitative conversion (Table 3, entry 1). At this temperature,
product 8 was observed to be the major product.
Fig. 4 shows the relative ratio of the formation of spiroketals
◦
16
8
and 9 at a range of temperatures (60, 80 and 120 C) using
nature of the substituents on the alkyne substrate, which may lead
1
mol% of 3 in C
2
D
2
Cl
4
. At higher temperatures, improved
to differences in charge distribution between the reactant and the
transition states. The structure of the ring to be formed can also
influence the rate of cyclisation and this activation energy for ring
closure is dependent on the ring strain of the cyclised product.
To better understand the effect of ring size and alkyne substituent
on catalyst performance, we initially studied the efficiency of the
regioselectivity with preferential formation of spiroketal 9 over 8
was observed, as well as increased catalytic activity. In addition, at
higher temperatures, the amount of spiroketal 8 formed gradually
decreases over prolonged reaction times to favour the formation
of product 9. This indicates that the reaction is reversible and that
17
4
064 | Dalton Trans., 2010, 39, 4062–4069
This journal is © The Royal Society of Chemistry 2010

View Article Online
Table 4 Dihydroalkoxylation of purely aliphatic (10) or aromatic (12)
Table 5 Conversion of alkyne diol 7 into products 8 and 9 (Fig. 3), 1 mol%
F
a
alkyne diols using 1 mol% catalyst, [M] = Rh, [Rh(bpm)(CO)
2
]BAr
4
(3)
catalyst, C
2 2 4
D Cl , NMR scale, unless otherwise stated
F
◦
a
or Ir, [Ir(bpm)(CO)
2
]BAr
4
(4) in C
2 2 4
D Cl at 100 C, NMR scale
Entry
Catalyst
t/h
TOF/h-
1b
1b
Substrate
Product
[M]
t/h
TOF/h-
Rh
Ir
6.6
0.8
139
549
Rh
Ir
8.2
7.2
56
45
[
Rh] [mol%]
[Ir] [mol%]
1
2
3
4
1.0
0.0
0.5
1.0
0.0
1.0
0.5
1.0
0.22
0.58
0.13
0.10
961
374
1694
893
c
c,d
a
1
Determined by H NMR spectroscopy to reach 98% total conversion
F
(
8 and 9) using 1 mol % catalyst, [M] = Rh, [Rh(bpm)(CO)
2
]BAr
4
-1
(3)
◦
b
a
1
b
F
Determined by H NMR spectroscopy to reach 98% conversion. TOF
and/or Ir, [Ir(bpm)(CO)
the amount of product/mole of catalyst/h at the point of 50% conversion
of substrate to product. Dual catalyst system (Rh + Ir): 0.5 mol% of
catalyst 3 and 0.5 mol% of catalyst 4. Dual catalyst system (Rh + Ir):
2
]BAr
4
(4) in C
2
D
2
Cl
4
at 100 C. TOF (h ) is
-1
(
h ) is amount of product/mole of catalyst/h at the point of 50%
c
conversion of substrate to product.
d
1
8
.0 mol% of catalyst 3 and 1.0 mol% of catalyst 4, with isolated yield =
6%.
Rh and Ir catalysed dihydroalkoxylation reactions for a pair of
alkyne diols.
The reactivity of complexes [Rh(bpm)(CO)
F
2
]BAr
4
(3) and
diol of 7 (Fig. 2) using a combination of complexes 3 and 4 are
summarised in Table 5.
F
[
Ir(bpm)(CO)
2
]BAr
4
(4) were investigated for the cyclisation of
diols with different alkyne substituents, the purely aliphatic and
aromatic containing alkyne diols, 3-ethynylpentane-1,5-diol (10)
The mixture of complexes 3 and 4 catalysed the reaction at much
higher turnover rates and shorter reaction times when compared
with the reactions catalysed by the individual catalysts (3 or 4). At
the same total catalyst loading of 1 mol%, the turnover frequency
and 2,2¢-(1,2-ethynediyl)bisbenzyl alcohol (12) respectively using
◦
1
mol% of catalyst 3 or 4, in C
2
D
2
Cl
4
at 100 C. The results are
-
1
summarised in Table 4.
of the dual metal (Rh and Ir) catalyst system (TOF = 1694 h ) is
-
1
The catalysed cyclisation of aliphatic terminal alkyne diol, 3-
ethynylpentane-1,5-diol (10) affords exclusively the five-membered
bicyclic ketal 11 (Table 4) while the aromatic alkyne diol, 2,2¢-
nearly twice that of the best single catalyst (Rh; 3, TOF = 961 h )
(Table 5, entry 3). The enhanced reactivity obtained using the dual
catalyst system indicates cooperativity between the two different
complexes in this two-step reaction, despite the catalysts being
introduced as independent molecules.
(
1,2-ethynediyl)bisbenzyl alcohol (12) cyclised to form solely the
bis(benzannulated) 5,6-spiroketal 13. The use of Ir(I) complex 4
resulted in a notably shorter time for cyclising 10 than the Rh(I)
complex 3 (Rh; 6.6 h and Ir; 0.8 h) and gave a much higher turnover
A range of substrates to establish the substrate/product
dependence on the metal centre
-
1
-1
frequency (TOF = Rh; 139 h and Ir; 549 h ). For the cyclisation
of the aromatic substrate 12, the Rh(I) complex 3 achieved only
-
1
a slightly higher turnover frequency (TOF = Rh; 56 h and Ir;
Rh and Ir promote the O-cyclisation reactions with different
efficiencies, depending either on the nature of the substituents
on the alkyne substrates (aromatic versus aliphatic) or on the ring
sizes of the products (5- versus 6-membered rings). To establish
the nature of this dependence, we investigated the efficiency of
cyclisation for a series of alkyne diols. The results using 1 mol%
-
1
4
5 h ).
The cyclisation of the purely aliphatic alkyne diol (10) shows
a significant difference in reactivity between the Rh (3) and Ir
4) catalysts. The change in reactivity may be a result of the Ir(I)
(
catalyst favouring the aliphatic alkyne cyclisation, or a result of
the Ir(I) catalyst favouring the formation of the five-membered
cyclised product (11).
F
F
of [Rh(bpm)(CO)
2
]BAr
4
(3) and [Ir(bpm)(CO)
2
]BAr
4
(4) as
catalysts, both independently and as a mixed catalyst system are
summarised in Table 6.
The results presented in Table 6 show that the efficiency of
cyclisation is affected primarily by the ring sizes of the products
that are formed. In the cases of the internal aliphatic alkynes,
1-methyl-3-heptyne-1,7-diol (14) and 1-methyl-3-octyne-1,8-diol
(16), the substrate with the shorter carbon chain length, 14 gave a
much faster cyclisation rate to form the 5,5-spiroketal (15) for each
catalyst. Similarly for the cyclisation rate of the mixed alkynes,
a decreasing chain length in the order 18 > 7 > 20 leads to
an increase in substrate reactivity in the order 18 < 7 < 20 for
each catalyst. This increase in reactivity with decreasing chain
Use of a dual metal (Rh and Ir) catalyst system
The Rh (3) and Ir (4) complexes displayed differential selectivity
and efficiency either for the cyclisation of the aromatic versus
the aliphatic alkyne diols or the 5- versus 6-membered rings, so
we proposed the use of a dual metal catalyst system involving
both Rh (3) and Ir (4) complexes for the promotion of the
dihydroalkoxylation reaction with the mixed (both aliphatic and
aromatic) alkyne diol, 7, which gives the 5,6-spiroketal products.
The results of the double cyclisation reaction for the mixed alkyne
This journal is © The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4062–4069 | 4065

View Article Online
Table 6 Summary of the catalysed cyclisation of alkyne diols (7, 10, 12, 14, 16, 18 and 20) using a total of 1 mol% catalyst(s), [M] = Rh,
F
F
[
Rh(bpm)(CO)
2
]BAr
4
(3) and/or Ir, [Ir(bpm)(CO)
2
]BAr
4
(4) in C
2
D
2
Cl
4
, NMR scale, to demonstrate the effect of the substituents on the substrate and
a
ring size of product
TOF/h-
1b
Entry
Substrate
Product
Catalyst/M
t/h
1
2
3
Rh
Ir
Rh + Ir
0.15
0.18
0.11
714
1126
1014
c
4
5
6
Rh
Ir
Rh + Ir
1.03
0.55
0.32
381
372
556
7
8
9
Rh
Ir
Rh + Ir
6.60
0.80
1.20
139
549
540
c
1
1
1
0
1
2
Rh
0.28
1.73
0.59
794
177
435
Ir
c
Rh + Ir
d
1
3
Rh
Ir
0.09 (1.8 : 1)
1121
d
1
4
0.42 (3.6 : 1)
535
c
d
e
1
5
Rh + Ir
0.10 (2.2 : 1)
1259
1
6
Rh
0.22 (0.7 : 1)
961
e
e
1
1
7
8
Ir
0.58 (0.9 : 1)
0.13 (0.8 : 1)
374
1694
c
c
Rh + Ir
1
2
2
9
0
1
Rh
Ir
Rh + Ir
8.20
7.20
2.10
56
45
122
a
1
F
Determined by H NMR spectroscopy to reach 98% conversion using 1 mol% catalyst, [M] = Rh, [Rh(bpm)(CO)
2
]BAr
4
(3) and/or Ir,
F
◦
b
-1
[
Ir(bpm)(CO)
2
]BAr
4
2 2 4
(4) in C D Cl at 100 C, unless otherwise stated. TOF (h ) is the amount of product/mole of catalyst/h at the point of
c d
5
2
0% conversion of substrate to product. Dual catalyst system (Rh + Ir): 0.5 mol% of catalyst 3 and 0.5 mol% of catalyst 4. Relative ratio of product
1 : 22. Relative ratio of product 8 : 9.
e
length is possibly due to the increased steric strain in the larger
spiroketal ring systems. The relative changes in efficiency could
not be similarly attributed to the nature of the substituent on
the alkyne, in particular in the case of substrate 12 which has
two aromatic substituents and very little difference between the
reactivity of Rh and Ir.
remarkably higher in the presence of the Rh catalyst with a
1
turnover frequency – 4 times greater and reaching completion
2
about six times faster – than that required for the Ir catalyst
(entries 4 and 5). For the formation of the 5,5-spiroketals (11
and 15), both the terminal and internal alkyne diols (10 and 14)
were most efficiently cyclised using the Ir catalyst. Finally, for the
catalysed cyclisation of alkyne 20, the Ir complex generated the
5,5-spiroketal, 21 as the major product (21 : 22 = 3.6 : 1) whereas
the Rh complex generated less of product 21 (21 : 22 = 1.8 : 1).
For substrates 7, 12 and 16, the cyclisation proceeds with the
concurrent ring closure of both the 5-exo and 6-endo rings. As a
result, the use of the dual metal (Rh and Ir) catalyst system, 3 and
The observed difference in reactivity using the three different
F
catalyst systems can be attributed to the [Rh(bpm)(CO)
2
]BAr
4
(
3) favouring the 6-endo ring cyclisation while the
F
[
Ir(bpm)(CO)
This is evident in the selective formation of the 6,6-spiroketal
19), where the rate of 6-endo cyclisation of alkyne diol 18 is
2
]BAr
4
(4) favours the 5-exo cyclisation (Fig. 5).
(
4
066 | Dalton Trans., 2010, 39, 4062–4069
This journal is © The Royal Society of Chemistry 2010

View Article Online
Fig. 5 Proposed mechanism for the Rh (3) and/or Ir (4) catalysed cyclisation of alkyne diol 7.
4
, helps to promote both the 5-exo and 6-endo ring cyclisation
the spiroketal products, 8 and 9 and regenerates the catalytic
species.
in alkyne diols 7, 12 and 16 and in each case, a significant
enhancement in reactivity was observed in the presence of the
dual metal catalyst system (entries 6, 18 and 21) as compared
to the individual single catalysed reaction (entries 4, 5 and 16,
The enhanced reactivity of the dual metal catalysts is supported
by this mechanism (Fig. 5), whereby the necessary intermediates
for the two cyclisation processes giving the mixed 5,6-spiroketal
products are readily available for cyclisation by the most efficient
catalyst available.
1
7 and 19, 20). This cooperativity phenomenon observed in the
dual metal catalyst system is likely to arise from the dual activation
pathway in the dihydroalkoxylation reaction, where the Rh catalyst
preferentially promotes the 6-membered ring formation while the
Ir preferentially promotes the 5-membered cyclisation.
Conclusions
F
F
[
Rh(bpm)(CO)
2
]BAr
4
(3) and [Ir(bpm)(CO)
2
]BAr
4
(4), contain-
ing the bidentate bis(1-pyrazolyl)methane (bpm) ligand, carbonyl
Mechanism
(
CO) co-ligands and the weakly coordinating anion, tetrakis[3,5-
F
-
We have previously proposed a mechanism for the hydroalkoxyla-
tion of 4-pentyn-1-ol in the presence of methanol, catalysed by the
cationic Ir(I) complex. A similar mechanism is proposed here
bis(trifluoromethyl)phenyl]borate, BAr
4
, were found to be highly
effective catalysts for the cyclisation of alkyne diols (7, 10, 12, 14,
16, 18 and 20). In many cases, these complexes are much more
efficient than other metal catalysts reported previously for the
8
for the sequential two-step intramolecular dihydroalkoxylation
reaction of the alkyne diol, 7 (Fig. 5). The proposed mechanism
involves an initial p-coordination of the metal (Rh/Ir) complex to
the alkyne to generate intermediate A. The activated triple bond is
then subjected to nucleophilic attack by the hydroxyl group either
through the 5-exo or 6-endo cyclisation, which can be favoured
by the respective Ir (4) or Rh (3) complexes. This is subsequently
followed by protodemetalation to afford the monocyclised furan
C1 or pyran, C2 and regenerates the catalytic species (3 or 4). It
should be noted that in some cases the monocyclised products were
observed to be the minor products, such as in the cyclisation of
alkyne diol 20. Once the monocyclised products are formed, they
enter the second catalytic cycle to form the second C–O bond.
Again, the intermediate alkenes (C1, C2) can be reactivated, via
p-coordination of the complex to the alkene (D1, D2), followed
by the second intramolecular addition and rearrangement to form
intermediates (E1, E2). Finally, a reductive elimination step affords
8,9
cyclisation of these substrates.
For the formation of the 5,5-spiroketal and fused bicyclic ketals
(11 and 15), the Ir(I) complex (4) was a more active catalyst than
its Rh(I) analogue (3). In contrast, the Rh(I) complex (3) was
the more effective catalyst for the cyclisation of 2-(6-hydroxyhex-
1-ynyl)benzyl alcohol (18) to form the 6,6-spiroketal (19). The
dual metal (Rh and Ir) catalyst system is the most effective for
promoting the simultaneous double ring closure of alkyne diols
7, 12 and 16 to form 5,6- and 6,5-spiroketals (8, 9, 13, and 17)
respectively. The combination of the two metal complexes (3
and 4) worked cooperatively to promote a highly efficient dual
activation pathway for both the 5-exo and 6-endo ring cyclisation
in the dihydroalkoxylation reaction of alkyne diols. The relative
efficiencies of the Rh(I) and Ir(I) catalysts was dependent primarily
on the size of the rings formed, not the nature of the substituents
(aromatic or aliphatic) on the alkyne.
This journal is © The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4062–4069 | 4067

View Article Online
A mechanism for the dihydroalkoxylation reaction of alkyne
diols was postulated based on a previously reported mechanism.
This mechanism is based on a tandem, two step cyclisation
reaction which involves the first C–O bond formation to generate
a monocyclised intermediate followed by a second intramolecular
C–O bond addition to generate the spiroketal product.
General procedure for catalysis
Thermal catalysed dihydroalkoxylation reactions were conducted
in NMR tubes fitted with concentric Teflon (Youngs) valves under
an inert atmosphere. The catalytic reactions were performed within
the NMR spectrometer, or in an oil bath if prolonged heating was
required. The temperature in the NMR magnet was calibrated
using an Omega Microprocessor Thermometer (Model HH23)
or neat ethylene glycol. The reaction progress was monitored by
Experimental
1
H NMR spectroscopy at regular intervals. Characterisation of
8
2
1
1
-(5-Hydroxypent-1-ynyl)benzyl alcohol (7), 3-ethynylpentane-
products was confirmed by comparison to literature data. The
conversion of substrate to product was determined by integration
of the product resonances relative to the substrate resonances
8
8
,5-diol (10), 2,2¢-(1,2-ethynediyl)bisbenzyl alcohol (12),
18
-methyl-3-heptyne-1,7-diol (14),
1-methyl-3-octyne-1,8-diol
2-(6-hydroxylhex-1-ynyl)benzyl alcohol (18), 2-(4-
hydroxybut-1-ynyl)benzyl alcohol (20), [Rh(bpm)(CO)
18
(
16),
1
in the H NMR spectra. The turnover frequency (TOF) was
8
2
]BPh
4
11
calculated as the number of moles of product/moles of catalyst/h
and was calculated at the point of 50% conversion of substrate.
12
19
F
(
1), [Ir(bpm)(CO)
2
]BPh
4
11
(2), [Rh(bpm)(CO)
(4), [Rh(bpm)(COD)]BAr
(6) were prepared according to literature
2
]BAr
4
(3),
F
F
11
[
[
Ir(bpm)(CO)
Ir(bpm)(COD)]BAr
2
]BAr
4
4
(5) and
F
11
4
Characterisation data for products of the catalysed cyclisation of
alkyne diols
procedures.
Single crystal X-ray analysis was performed by Dr J o¨ rg Wa-
gler at the Research School of Chemistry, Australian National
University, Canberra. X-Ray diffraction data was collected in
U and w scans on a Nonius KappaCCD diffractometer using
Mo Ka radiation. The structure was solved with direct methods
(
SHELXS97) and refined by full-matrix least-squares refinement
2
20
1
of F using SHELXL97. Crystals of 3 suitable for X-ray crys-
tallography were grown from layering hexane over a concentrated
dichloromethane solution of complex 3. Crystallographic data for
Spiroketal (15). H NMR (500 MHz, C D Cl , 298 K): d 4.20–
2
2
4
4.14 (m, 1H, H1¢), 4.12–4.06 (m, 1H, H1), 3.94–3.75 (m, 4H, H7
and H7¢), 2.17–1.83 (m, 14H, H2, H2¢, H3, H3¢, H5, H5¢, H6
and H6¢), 1.71–1.63 (m, 1H, H2), 1.48–1.41 (m, 1H, H2¢), 1.26 (d,
3
is given in Table 7.
2
2
J(CH
3
–H1) = 6.0 Hz, 3H, CH
¢) ppm.
1
3
), 1.18 (d, J(CH
3
¢–H1¢) = 6.0 Hz,
F
Table 7 Crystallographic data for [Rh(bpm)(CO)
structure
2
]BAr
4
(3) solid-state
3H, CH
3
13
C{ H} NMR (125 MHz, C
and C4¢), 75.6 (C1), 74.0 (C1¢), 66.7, 66.6 (C7 and C7¢), 35.9, 35.3,
4.9, 34.8 (C3/5 and C3¢/5¢), 32.4, 31.9 (C2 and C2¢), 24.5, 24.3
C6 and C6¢), 22.6 (CH ) and 21.2 (CH ¢) ppm.
2
D
2
Cl
4
, 298 K): d 114.6, 114.4 (C4
F
[
Rh(bpm)(CO)
2
]BAr
4
(3)
3
(
Empirical formula
M/g mol-
Crystal system
41 4 2
C H20BF24N O
Rh
3
3
1
1170.33
Triclinic
¯
P1
Space group
a/A˚
10.3971(10)
14.9619(2)
15.1133(2)
92.0450(10)
105.2590(10)
98.5870(10)
2235.78(5)
1.738
b/A˚
c/A˚
◦
a/
◦
b/
◦
1
g /
Spiroketal (17). H NMR (500 MHz, C
4.10 (m, 2H, H1 and H1¢), 3.91–3.80 (m, 2H, H8 and H8¢), 3.55
m, 2H, H8 and H8¢), 2.11–2.04 (m, 1H, H2/2¢), 2.01–1.94 (m, 2H,
2
D
2
Cl
4
, 298 K): d 4.20–
V/A˚
3
-3
D
Z
c
/g cm
(
2
H2/2¢ and H3/3¢), 1.91–1.57 (m, 11H, H3/3¢, H6/6¢, H2/2¢, H5
and H5¢), 1.55–1.45 (m, 5H, H6/6¢ and H7 and H7¢), 1.42–1.36
T/K
100(2)
l Mo Ka/A˚
0.71073
0.524
-1
2
m Mo Ka/mm
(m, 1H, C2/2¢), 1.28 (d, J(CH –H1) = 6.2 Hz, 3H, CH ), 1.21 (d,
3
3
Crystal size/mm
0.31 ¥ 0.28 ¥ 0.14
0.815, 0.933
2.66–30.00
99.9
2
J(CH
3
1
¢–H1¢) = 6.3 Hz, 3H, CH
3
¢) ppm.
T (Gaussian)min,max
1
3
◦
C{ H} NMR (125 MHz, C D Cl , 298 K): d 105.7, 105.5 (C4
2
2
4
q range/
Completeness (%)
Index ranges
and C4¢), 76.4, 73.8 (C1 and C1¢), 61.5, 61.2 (C8 and C8¢), 38.7,
-14 ≥ h ≥ 14
3
2
7.3 (C3 and C3¢), 34.2, 33.9 (C5 and C5¢), 31.5, 31.3 (C2 and C2¢),
5.3, 25.2 (C7 and C7¢), 23.2 (CH ), 21.2 (CH ¢) and 20.3 (C6 and
-
21 ≥ k ≥ 21
21 ≥ l ≥ 21
3
3
-
C6¢) ppm.
Reflns measured
Unique reflns
R
62 552
13 056
0.0405
1.019
int
GoF(all data)
(I > 2s(I))
wR (I > 2s(I))
(all data)
wR (all data)
R
1
0.0447
0.1086
0.0549
0.1132
2
R
1
2
4
068 | Dalton Trans., 2010, 39, 4062–4069
This journal is © The Royal Society of Chemistry 2010

View Article Online
1
Spiroketal (19). H NMR (500 MHz, CDCl
3
, 298 K): d 7.16–
therein; (c) E. Genin, S. Antoniotti, V. Michelet and J.-P. Gen eˆ t, Angew.
Chem., Int. Ed., 2005, 44, 4949.
7
.14 (m, 2H, H4 & H5), 7.08–7.06 (m, 1H, H6), 7.02–7.00 (m, 1H,
4
5
X. Yu, S. Y. Seo and T. J. Marks, J. Am. Chem. Soc., 2007, 129, 7244.
Y. Wakabayashi, Y. Fukuda, H. Shiragami, K. Utimoto and H. Nozaki,
Tetrahedron, 1985, 41, 3655.
2
a
b
H3), 4.76 (d, J(H1 –H1 ) = 14.8 Hz, 2H, H1), 3.81–3.76 (m, 1H,
2
a
b
H10), 3.69–3.66 (m, 1H, H10), 2.83 (d, J(H8 –H8 ) = 16.5 Hz,
2
1
H, H8), 1.98–1.95 (m, 1H, H12), 1.84–1.80 (m, 1H, H13), 1.71–
.63 (m, 4H, H11 & H12 & H13) ppm.
6 B. Liu and J. K. De Brabander, Org. Lett., 2006, 8, 4907.
7
(a) S. Antoniotti, E. Genin, V. Michelet and J.-P. Genet, J. Am. Chem.
Soc., 2005, 127, 9976; (b) A. S. K. Hashmi, S. Schaefer, M. Woelfe,
C. Diez Gil, P. Fischer, A. Laguna, M. C. Blanco and M. C. Gimeno,
Angew. Chem., Int. Ed., 2007, 46, 6184.
1
3
1
C{ H} NMR (125 MHz, CDCl
3
, 298 K): d 133.6 (C2/7), 131.3
(
(
(
C2/7), 129.0 (C6), 126.6 (C4/5), 125.9 (C4/5), 123.9 (C3), 95.4
C9), 61.9 (C1), 61.8 (C10), 39.0 (C8), 34.9 (C13), 25.1 (C11), 18.9
C12) ppm.
8
9
B. A. Messerle; and K. Q. Vuong;, Organometallics, 2007, 26, 3031.
X. Li, A. R. Chianese, T. Vogel and R. H. Crabtree, Org. Lett., 2005,
7
, 5437.
0 (a) M. Beller, B. Zimmermann and H. Geissler, Chem.–Eur. J., 1999,
, 1301; (b) H. Siebeneicher, I. Bytschkov and S. Doye, Angew. Chem.,
1
5
Acknowledgements
Int. Ed., 2003, 42, 3042; (c) N. Jeong, S. D. Seo and J. Y. Shin, J. Am.
Chem. Soc., 2000, 122, 10220; (d) B. M. Trost and M. R. Machacek,
Angew. Chem., Int. Ed., 2002, 41, 4693; (e) S. U. Son, K. H. Park and
Y. K. Chung, J. Am. Chem. Soc., 2002, 124, 6838; (f) J. M. Lee, Y. Na,
H. Han and S. Chang, Chem. Soc. Rev., 2004, 33, 302.
1 S. L. Dabb, J. H. H. Ho, R. Hodgson, B. A. Messerle and J. Wagler,
Dalton Trans., 2009, 634.
This work was supported by the Australian Research Council and
the University of New South Wales. J. H. H. Ho is grateful to the
University of New South Wales for an University International
Postgraduate Award (UIPA). J. W. acknowledges his scholarship
by Deutscher Akademischer Austausch Dienst (DAAD). We also
thank Prof. Stuart McGregor for enlightening discussion.
1
1
2 S. Burling, L. D. Field, B. A. Messerle and P. Turner, Organometallics,
2004, 23, 1714.
1
1
3 I. Krossing and I. Raabe, Angew. Chem., Int. Ed., 2004, 43, 2066.
4 S. Selvaratnam, Joanne H. H. Ho, P. B. Huleatt, B. A. Messerle and
C. L. L. Chai, Tetrahedron Lett., 2009, 50, 1125.
Notes and references
1
5 J. E. Baldwin, J. Chem. Soc., Chem. Commun., 1976, 734.
1
(a) F. Perron and K. F. Albizati, Chem. Rev., 1989, 89, 1617 and
references therein; (b) K. T. Mead and B. N. Brewer, Curr. Org. Chem.,
16 I. G. Star a´ , A. Koll a´ rovi, F. Tepl y´ , D. Sˇ aman and P. Fiedler, Collect.
Czech. Chem. Commun., 2003, 68, 917.
2
003, 7, 227 and references therein; (c) B. D. Schwartz, P. Y. Hayes, W.
17 G. Illuminati and L. Mandolini, Acc. Chem. Res., 1981, 14, 95.
18 Y. B. Lee and J. E. Folk, Bioorg. Med. Chem., 1998, 6, 253.
19 L. D. Field, B. A. Messerle, M. Rehr, L. P. Soler and T. W. Hambley,
Organometallics, 2003, 22, 2387.
20 (a) R. H. Blessing, Acta Crystallogr., Sect. A: Found. Crystallogr., 1995,
51, 33(SORTAV); (b) G. M. Sheldrick, Crystal Structure Solution –
WinGX Version, 1986–97, Release 97–2 (SHELXS); (c) G. M. Sheldrick,
Crystral Structure Refinement – WinGX Version, 1993–7, Release 97–2
(SHELXL).
Kitching and J. J. De Voss, J. Org. Chem., 2005, 70, 3054; (d) W. Francke
and W. Kitching, Curr. Org. Chem., 2001, 5, 233 and references therein.
(a) S. Favre, P. Vogel and S. Gerber-Lemaire, Molecules, 2008, 13, 2570
and references therein; (b) P. Livant and W. Xu, J. Org. Chem., 1998,
2
3
6
3, 636; (c) K. Y. Tsang, M. A. Brimble and J. B. Bremner, Org. Lett.,
003, 5, 4425.
2
(a) K. Utimoto, Pure Appl. Chem., 1983, 55, 1845; (b) F. Alonso, I. P.
Beletskaya and M. Yus, Chem. Rev., 2004, 104, 3079 and references
This journal is © The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 4062–4069 | 4069