Iridium(i) hydroxides in catalysis: Rearrangement of allylic alcohols to ketones

Article abstract of DOI:10.1039/c4ob01428f

The iridium(i) hydroxide complex [Ir(OH)(COD)(IⁱPr)] has been shown to be a competent catalyst for the rearrangement of allylic alcohols to ketones. Reactions proceed in short reaction times (1-1.5 h) with microwave heating, in the absence of additives. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4ob01428f

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Iridium(I) hydroxides in catalysis: rearrangement of
allylic alcohols to ketones†
Cite this: Org. Biomol. Chem., 2014,
2, 6672
1
David J. Nelson,‡ José A. Fernández-Salas, Byron J. Truscott and Steven P. Nolan*
Received 3rd June 2014,
Accepted 9th July 2014
i
The iridium(I) hydroxide complex [Ir(OH)(COD)(I Pr)] has been shown to be a competent catalyst for the
DOI: 10.1039/c4ob01428f
rearrangement of allylic alcohols to ketones. Reactions proceed in short reaction times (1–1.5 h) with
www.rsc.org/obc
microwave heating, in the absence of additives.
9
,10
Isomerisation reactions are, by their very nature, atom-efficient F sources, respectively.
This work has since been followed
transformations, and allow for the interconversion of useful by a related study on the tandem isomerisation–bromination
1
1
functional groups. Allylic alcohol rearrangements to form of allylic alcohols, to form α-bromoketones. Notably, the
ketones comprise one such type of isomerisation reaction most efficacious catalyst system for this reaction was found to
(
Scheme 1). Allylic alcohol isomerisation can be catalysed by a be 1 mol% [{Ir(Cp*)} (OH) ][OH]·11H O, which bears hydrox-
2 3 2
1
2,3
variety of metal systems, including iridium, although we ide ligands, in wet acetone at RT for 0.5–24 h. Alternative cata-
have recently reported a highly-active ruthenium system for lyst systems [Ir(Cp*)(OH ) ][SO ] and [IrCl (Cp*)] were found
2
3
4
2
2
4
this transformation. With the appropriate ligand sphere, to be less active under the same conditions. Palladium hydrox-
asymmetric variants are also possible, albeit with high load- ides (in the form of Pd(OH) on carbon) have also been used
ings of expensive catalyst complexes (ca. 5 mol%). Many to effect the isomerisation of allylic alcohols.
2
5
12
iridium-catalysed methodologies require the use of dihydrogen
Recently, we disclosed the synthesis and study of a series of
in order to generate the active iridium(III) dihydride species iridium(I) hydroxide complexes of the form [Ir(OH)(COD)-
6
–8
in situ;
the reaction solution must then be degassed to (NHC)] (COD = 1,5-cyclooctadiene, NHC = N-heterocyclic
1
3
remove excess dihydrogen and avoid competing hydrogenation carbene) (Scheme 2). These species are capable of deproto-
reactions. While this is achievable without specialist equip- nating acidic C–H, N–H and O–H bonds, activating boronic
ment, it requires a cylinder of hydrogen on hand and is less acids and silanes, or activating CO to generate new iridium(I)
2
1
4
operationally straightforward than the simple mixing of a cata- species. As the oxidation state of the new complexes remains
lyst and substrate. Recently, Ahlsten et al. reported the elegant at +1, there exists considerable scope for further reaction of
tandem allylic alcohol isomerisation/α-chlorination or α-fluori- these new species. Due to their relative rarity, little is known of
nation of a range of substrates by employing typically their ability or involvement in catalytic reactions.
0
.25–1.0 mol% of iridium(III) precursor [IrCl
2
(Cp*)]
2
(Cp* =
The reactivity of the iridium(I) hydroxide was therefore
1
,2,3,4,5-pentamethylcyclopentadienyl) as the catalyst and explored in the allylic alcohol isomerisation reaction to investi-
N-chlorosuccinimide or SelectFluor (1-chloromethyl-4-fluoro- gate the catalytic potential of iridium(I) hydroxide complexes
1
a
,4-diazoniabicyclo[2.2.2]octane bis(tetrafluoroborate)) as Cl or with substrates bearing acidic (pK ≤ ca. 30) protons. It was
reasoned that the initial reaction would be the deprotonation
of the substrate, leading to an alkoxide complex; the intra-
molecular nature of subsequent steps might then accelerate
the reaction. In particular, it was hoped that the presence of
Scheme 1 Isomerisation of allylic alcohols to form ketones.
EaStCHEM School of Chemistry, University of St Andrews, North Haugh, Fife,
KY16 9ST, UK. E-mail: snolan@st-andrews.ac.uk
†
Electronic supplementary information (ESI) available: Synthesis and characteri-
i
Me
i
Me
sation of [Ir(COD)(I Pr )(py)][PF
of isolated reaction products. See DOI: 10.1039/c4ob01428f
Current address: WestCHEM Department of Pure and Applied Chemistry,
University of Strathclyde, 295 Cathedral Street, Glasgow, G1 1XL, UK.
6
] and [Ir(COD)(I Pr
2 6
) ][PF ], and NMR spectra
‡
Scheme 2 Iridium(I) hydroxides as building blocks for organometallic
chemistry.
6672 | Org. Biomol. Chem., 2014, 12, 6672–6676
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
It was proposed that the hydroxide functionality initially
deprotonated the substrate, yielding an iridium(I) alkoxide
intermediate that could then isomerise; directly attaching the
substrate to the catalyst in this way would promote the reaction
by rendering subsequent steps where the catalyst and substrate
interact intramolecular in nature. Attempts were made to
trap this initial alkoxide species, but with little success. 1 : 1
mixtures of substrate and hydroxide rarely led to reaction at
room temperature, despite the demonstrated ability of the
Scheme 3 Trial reactions.
an internal base would preclude the need for additives such as
base or hydrogen gas.
Initially, a series of trial reactions were conducted in NMR
tubes using model substrate 1-phenylprop-2-en-1-ol and
hydroxide complex [Ir(OH)(COD)(I Pr)] (1) (I Pr = 1,3-diisopro-
pylimidazol-2-ylidene) (Scheme 3). No additives were used; the
catalyst and substrate were simply dissolved in dry, oxygen-free
1
3
hydroxide complex to react with a range of acidic protons.
The application of heat led to the allylic alcohol isomerisation
reaction taking place. The reaction of the iridium(I) hydroxide
complex with cyclohex-2-en-1-ol (1 equiv.) at room temperature
led to a 1 : 1 mixture of iridium(I) hydroxide and an unknown
i
i
1
species, as determined by the presence of H NMR signals
toluene-d and heated in an oil bath. A low catalyst loading of
8
i
corresponding to the I Pr NHC in a different environment. No
0
.25 mol% was employed, in order to be competitive with the
9
,10
signals corresponding to iridium hydride species were
state-of-the-art systems.
After mixing the components of the
1
observed on the H NMR spectrum. Unfortunately, this new
reaction mixture at room temperature, a rapid colour change
from bright yellow to orange was observed, in the same
manner as was observed in previous deprotonation experi-
species could not be isolated, although the addition of
degassed water led to isolation of only the iridium hydroxide 1.
Our hypothesis at this stage is that the deprotonation, while
feasible, is slow and disfavoured, and that heating is necessary
to overcome the subsequent barriers in the catalytic cycle.
Having established the necessary conditions, the scope and
limitations of the reaction were briefly explored (Fig. 2).
1
3
ments. However, no conversion to the product was obtained
after 15 min at room temperature, suggesting that this initial
colour change was due to deprotonation of the substrate;
unfortunately, at such a low catalyst loading, none of the
iridium species are observable by NMR spectroscopy. No con-
version was obtained overnight, but the application of heat led
to complete conversion. Hydroxide complexes of this type are
known to be capable of deprotonating ketones (such as
1
3
acetone) at the α-position, but pleasingly this side reaction
either did not occur, or otherwise did not impede productive
catalytic turnover.
i
The analogous chloride complex [IrCl(COD)(I Pr)] (2), and
i
Me
cationic complexes [Ir(COD)(I Pr )(py)][PF ] (3) and [Ir(COD)-
6
i
Me
i
Me
(
I Pr
2 6
) ][PF ] (4) (py = pyridine, I Pr = 1,3-diisopropyl-4,5-
dimethylimidazol-2-ylidene) (Fig. 1) were also tested under the
same conditions (110 °C, 16 h) but yielded poorer conversions
(24%, 38% and 15%, respectively). The hydroxide functionality
is clearly important to facilitate catalytic activity in this
reaction.
Fig. 1 Alternative complexes screened for reactivity.
Fig. 2 Scope and limitations of the thermal reaction.
Org. Biomol. Chem., 2014, 12, 6672–6676 | 6673
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Organic & Biomolecular Chemistry
Notably, most reactions proceed with 0.25 mol% iridium, and
turnover numbers (TON) of ca. 700 are achievable. Primary
alcohols can be converted to aldehydes, but with low TON
(
ca. 20); these substrates are known to be particularly challen-
2
ging in allylic alcohol isomerisation reactions. Steric bulk is
tolerated at the β-position, but α-substitution impedes the reac-
tion, presumably due to impeding the removal of the proton
that must be removed to generate the carbonyl motif. Conju-
gated substrates were successfully isomerised, although in
each case the ketone is in conjugation with an aromatic
system in the product. Alkyl and aryl-bearing substrates both
undergo reaction, although cyclohex-2-en-1-ol reacted more
slowly than oct-1-ene-3-ol.
However, the use of high temperatures for prolonged reac-
tion times is not necessarily economical, so the use of micro-
wave heating was explored instead. While there has been
much discussion regarding the efficiency of microwave heating
versus conventional heating, the consensus appears to be that
one method is not automatically superior to the other in terms
of energy efficiency, and that the most efficient method will
depend significantly on the apparatus and chemistry uti-
15
lised. However, a reduction in energy consumption is likely Fig. 3 Scope and limitations of allylic alcohol isomerisation under
microwave irradiation (isolated yields after purification by silica gel
chromatography).
to be achieved if the use of microwave heating permits a sig-
nificant reduction in reaction time; in one case study by
Kappe, a reduction in the reaction time from ca. 3–6 h to
<
5 min allowed a significant reduction in energy consumption,
reduction in reaction time compared to the thermal conditions
explored previously.
1
6
even when a relatively non-polar solvent was used (toluene).
An increase in reaction temperature, above the boiling point of
the solvent, can allow reactions to be completed in much
shorter times. A move to microwave heating was therefore
prompted primarily by a desire to render the procedure rapid
and convenient. Initial optimisation, in which various sol-
vents, temperatures and catalyst loadings were screened, high-
lighted the use of 0.1 mol% of catalyst in either THF (ε =
Equivalent results were achieved when 0.1 mol% of [IrCl-
COD)(I Pr)] (2) was used with 10 mol% KOH; while this
i
(
requires the use of base, it precludes the need to synthesise
and isolate the oxygen-sensitive iridium(I) hydroxide complex.
We hypothesise that either the hydroxide is formed in situ, or
that the substrate is deprotonated and the potassium alkoxide
complex reacts with the iridium chloride species. Iridium com-
17
17
7
.47) or toluene (ε = 2.43) at 150 °C for 0.5–1.0 h to be the
optimum conditions (Table 1). Notably, this represents a
.5-fold reduction in catalyst loading as well as a remarkable
13,18,19
plexes of this motif are generally very easy to prepare,
and tend to be air-stable, which may widen the appeal and
accessibility of this protocol.
2
With these optimised conditions in hand, the scope and
limitations of the microwave-irradiated reaction were explored
using toluene as solvent. The yields reported in Fig. 3 are iso-
lated yields after purification of the products by silica gel
chromatography. All substrates can be converted in short reac-
tion times, in very good to excellent isolated yields. This there-
fore represents a useful, general methodology for carrying out
this isomerisation reaction in the absence of base or additives.
Table 1 Screening microwave conditions for the iridium-catalysed iso-
merisation reaction of 1-phenylprop-2-en-1-ol
a
b
Entry
[Ir]/mol%
Solvent
T
Conv.
1
2
3
4
5
6
0.25
0.25
0.25
0.25
0.1
0.1
0.1
0.05
0.1
0.1
MeCN
DMF
Toluene
THF
150 °C
150 °C
150 °C
150 °C
150 °C
140 °C
140 °C
150 °C
130 °C
150 °C
35%
86%
>95%
>95%
THF
>95% Conclusions
Toluene
Toluene
THF
THF
THF
70%
>95%
65%
62%
>95%
c
7
It has been shown that iridium(I) hydroxides are capable of cat-
6
7
8
alysing the isomerisation of allylic alcohols to form ketones.
Notably, this is the first deployment of iridium(I) hydroxide in
catalysis. A range of compounds was successfully transformed
to ketones using low (typically 0.25 mol%) loadings of iridium.
d
a
1
mmol substrate in 1 mL solvent, plus indicated loading of [Ir(OH)-
i
(COD)(I Pr)], heated to
T
for 30 min by microwave i irradiation.
b
1
c
d
Determined by H NMR integration. 1 h. [IrCl(COD)(I Pr)] (2) plus Two protocols were provided: one using conventional heating,
10 mol% KOH.
and one using microwave heating to effect short and
6674 | Org. Biomol. Chem., 2014, 12, 6672–6676
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
i
6
convenient reaction times. Notably, neither protocol requires [Ir(COD)(I PrMe)(py)][PF ] (4)
the use of hydrogen or the presence of base, rendering it
In the glovebox, toluene (1 mL) was added to [(py)
2
Ir(COD)]-
] (49.8 mg, 0.083 mmol) and I PrMe (29.9 mg, 0.166 mmol,
.01 equiv.). The suspension was stirred at room temperature
highly practical and compatible with sensitive functional
groups, although the iridium chloride complex plus an alkox-
ide base can be used instead of the iridium(I) hydroxide
species.
Further work continues in our laboratories to probe
mechanistic aspects of this transformation, with the aim of
using the hydroxide species to prepare and study the postu-
lated intermediate alkoxide.
i
[PF
6
2
for 48 h, during which time a bright orange precipitate
formed. The solid was filtered off and washed with cold
pentane (3 × 0.5 mL) and dried under high vacuum. Yield:
5
2.3 mg (0.065 mmol, 79%) of a bright orange powder.
1
3
H NMR (CD
2 2 3 2
Cl ): δ 5.35 (sept, 4H, JHH = 7.0, CH(CH ) ),
3
.93–3.81 (m, 4H, COD), 2.22 (s, 12H, N(CH ) N), 2.20–2.13
3
2
3
(
7
m, 4H, COD), 1.80–1.71 (m, 4H, COD), 1.57 (d, 12H, JHH =
3
13
1
.0, CH(CH ) ), 1.24 (d, 12H, J = 7.0, CH(CH3)2). C{ H} NMR
3 2 HH
(
5
(
CD Cl ): δ 175.0 (Ir-C), 126.8 (N(CCH ) N), 73.4 (COD CH),
2 2 3 2
Experimental section
General
4.2 (CH(CH
3
)
2
), 32.4 (COD CH
2
3 2
), 22.8 (CH(CH ) ), 21.4
19
1
CH(CH ) ), 10.8 (N(CCH ) ). F{ H} NMR (CD Cl ): δ −74.0
3
2
3 2
2
2
1
31
1
1
THF and toluene were obtained from a solvent purification (d, J = 709.9). P{ H} NMR (CD Cl ): δ −144.5 (sept., J
system and stored under argon in the glovebox. Acetonitrile 709.9). The methine C signal was located using [ H, C]
=
PF
2
2
FP
13
1
3
1
and DMF were purchased as anhydrous under argon and HSQC analysis. Anal. Calcd for C29 PIr: C, 44.71;
H N F
52 4 6
stored in the glovebox. Deuterated solvents for the analysis of H, 6.50; N, 6.95. Found: C, 44.76; H, 6.36; N, 6.87.
metal complexes were dried on calcium hydride and distilled
before use and stored under argon; deuterated solvents for the
analysis of starting materials and products were used as
General procedure for the isomerisation of allylic alcohols
under thermal conditions
supplied.
In the glovebox, a vial containing a stir bar was charged with
All manipulations were performed inside an argon-filled
i
the substrate (1.0 mmol) and [Ir(OH)(COD)(I Pr)] (0.25 mol%)
glovebox or using standard Schlenk techniques under an
in toluene (1 mL). The reaction mixture was removed from the
argon atmosphere unless otherwise stated. [IrCl(COD)]
2
and
glovebox and stirred for 16 h at 110 °C. Following this, the
[
1
RhCl(COD)] were obtained from Umicore AG; complexes
and 2 were prepared using literature methods. NMR data
2
1
1
3
solvent was removed in vacuo and H NMR analysis was con-
ducted in CDCl to determine conversion relative to the start-
3
were obtained using Bruker spectrometers at 300, 400 or
1
ing material. Where the products are volatile, the reaction was
5
00 MHz ( H observe frequency); all chemical shifts are given
carried out in toluene-d
H NMR of all products were compared to literature reports.
8
and analysed without work-up.
in ppm and coupling constants in Hz, referenced to residual
1
4
1
13
1
20
proton signals ( H) or solvent signals ( C{ H}). Elemental
analyses were performed at London Metropolitan University.
General procedure for the isomerisation of allylic alcohols
under microwave irradiation
i
[
Ir(COD)(I PrMe)(py)][PF ] (3)
6
2
1
In the glovebox, a 10 mL microwave vial (fitted with septum
cap) containing a stir bar was charged with the substrate
(1.0 mmol) and [Ir(COD)(I Pr)(OH)] (0.1 mol%) in toluene
Analogous to the reported procedure,
in the glovebox,
toluene (1 mL) was added to [(py) Ir(COD)][PF ] (50.0 mg,
.083 mmol) and I PrMe (14.9 mg, 0.083 mmol, 1.00 equiv.).
The suspension was stirred at room temperature for 48 h,
during which time a yellow precipitate formed. The solid was
filtered off and washed with cold pentane (3 × 0.5 mL) and
dried in vacuo. Yield: 47.2 mg (0.067 mmol, 81%) of a yellow
2
6
i
i
0
(
1 mL). The reaction mixture was removed from the glovebox
and irradiated in the microwave at 140 °C, 150 W for 1 h. The
resultant mixture was chromatographed (flash chromato-
graphy, elution with pentane–ethyl acetate [40 : 1]) to afford
the product.
1
powder. H NMR (CD Cl ): δ 8.65–8.56 (2H, m, ArH), 7.82
tt, 1H, JHH = 7.8, 1.6, ArH), 7.54–7.46 (m, 2H, ArH), 5.90 (sept,
2
2
3
(
2
3
H, JHH = 7.1, NCHMe
2
), 3.93–3.81 (m, 2H, COD), 3.74–3.64
(
1
m, 2H, COD), 2.42–2.29 (m, 4H, COD), 2.14 (s, 6H, (NMe) ),
2
Acknowledgements
3
2
.99–1.84 (m, 4H, COD), 1.65 (d, 6 H, JHH = 7.1, NCHMe ), 1.40
3
13
1
(d, 6H, JHH = 7.1, NCHMe
2 2 2
). C{ H} NMR (CD Cl ): δ 172.5 (Ir-C), We thank Simone Manzini for synthesis of some of the sub-
1
50.6 (py), 138.7 (py), 127.2 (py), 126.5 (N(CCH ) N), 82.5 strates, Dr Catherine S. J. Cazin for use of the microwave, and
3
2
(COD CH), 64.4 (COD CH), 54.3 (CH(CH
3
)
2
), 32.8 (COD CH
2
), Melanja Smith and Dr Tomas Lebl for assistance with NMR
3
0.3 (COD CH ), 22.8 (CH(CH ), 22.6 (CH(CH )
2
3
)
2
3 2
), 10.4 spectroscopy facilities. We thank Umicore AG for a gift of
1
9
1
1
(
N(CCH ) N. F{ H} NMR (CD Cl ): δ −73.9 (d, J = 710.2). materials.
3
2
2
2
PF
3
1
1
1
P{ H} NMR (CD
2
Cl
2
): δ −144.5 (sept.,
JFP = 710.2). Anal.
We thank the EPSRC, the ERC (Advanced Investigator Award
Calcd for C24H N F PIr: C, 40.90; H, 5.29; N, 5.96. Found: ‘FUNCAT’ to SPN), and Sasol (stipend to BJT) for funding. SPN
37 3 6
C, 40.89; H, 5.25; N, 6.06.
is a Royal Society Wolfson Merit Award holder.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 6672–6676 | 6675

View Article Online
Paper
Organic & Biomolecular Chemistry
1
2 G. Sabitha, S. Nayak, M. Bhikshapathi and J. S. Yadav, Org.
Lett., 2010, 13, 382–385.
Notes and references
1
R. C. van der Drift, E. Bouwman and E. Drent, J. Organomet. 13 B. J. Truscott, D. J. Nelson, C. Lujan, A. M. Z. Slawin and
Chem., 2002, 650, 1–24. S. P. Nolan, Chem. – Eur. J., 2013, 19, 7904–7916.
R. Uma, C. Crévisy and R. Grée, Chem. Rev., 2002, 103, 27–52. 14 B. J. Truscott, D. J. Nelson, A. M. Z. Slawin and S. P. Nolan,
2
3
N. Ahlsten, A. Bartoszewicz and B. Martin-Matute, Dalton
Trans., 2012, 41, 1660–1670.
Chem. Commun., 2014, 50, 286–288.
15 J. D. Moseley and C. O. Kappe, Green Chem., 2011, 13,
4
5
S. Manzini, A. Poater, D. J. Nelson, L. Cavallo and
S. P. Nolan, Chem. Sci., 2014, 5, 180–188.
L. Mantilli, D. Gérard, S. Torche, C. Besnard and C. Mazet,
Angew. Chem., Int. Ed., 2009, 48, 5143–5147.
794–806.
16 T. Razzaq and C. O. Kappe, ChemSusChem, 2008, 1, 123–
132.
17 J. L. M. Abboud and R. Notari, Pure Appl. Chem., 1999, 71,
645–718.
6
7
L. Mantilli and C. Mazet, Tetrahedron Lett., 2009, 50, 4141–4144.
L. Mantilli, D. Gerard, S. Torche, C. Besnard and C. Mazet, 18 S. Leuthäußer, D. Schwarz and H. Plenio, Chem. – Eur. J.,
Pure Appl. Chem., 2010, 82, 1461–1469. 2007, 13, 7195–7203.
L. Mantilli, D. Gérard, C. Besnard and C. Mazet, 19 R. A. Kelly III, H. Clavier, S. Giudice, N. M. Scott,
Eur. J. Inorg. Chem., 2012, 2012, 3320–3330.
N. Ahlsten and B. Martin-Matute, Chem. Commun., 2011,
8
9
E. D. Stevens, J. Bordner, I. Samardjiev, C. D. Hoff,
L. Cavallo and S. P. Nolan, Organometallics, 2007, 27, 202–
210.
4
7, 8331–8333.
1
0 N. Ahlsten, A. Bermejo Gómez and B. Martín-Matute, 20 G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb,
Angew. Chem., Int. Ed., 2013, 52, 6273–6276.
A. Nudelman, B. M. Stoltz, J. E. Bercaw and K. I. Goldberg,
1
1 A. B. Gómez, E. Erbing, M. Batuecas, A. Vázquez-Romero
Organometallics, 2010, 29, 2176–2179.
and B. Martín-Matute, Chem. – Eur. J., 2014, DOI: 10.1002/ 21 A. C. Hillier, H. M. Lee, E. D. Stevens and S. P. Nolan,
chem.201402350. Organometallics, 2001, 20, 4246–4252.
6676 | Org. Biomol. Chem., 2014, 12, 6672–6676
This journal is © The Royal Society of Chemistry 2014