Application of UV-Vis spectroscopy to high throughput screening of hydroamination catalysts

Article abstract of DOI:10.1039/b820357c

Potential catalysts for the hydroamination of 2-(2-phenylethynyl)aniline 2 which have been identified through high throughput screening methods were investigated. Two complexes were shown to be highly active hydroamination catalysts in acetone: the in situ combinations of [Rh(CO)₂Cl] ₂-mesBIAN-NaBF₄ (mesBIAN = bis(2,4,6-trimethylphenylimino) acenapthene) and [Ir(COD)Cl]₂-NaBF₄ (COD = 1,5-cyclooctadiene). The isolated complexes [M(N-N)XCl]BF₄ (M = Rh, Ir; N-N = bidentate nitrogen donor ligand; X = CO or Cp*) were found to be inactive as catalysts for the conversion of 2 to 3. However, chloride abstraction from these complexes through the addition of AgBF₄ was found to generate extremely active catalysts. Particularly active was the complex [Rh(CO)ClmesBIAN]-AgBF₄ (5 mol%) which achieved complete conversion of 2 to 3 in 12 minutes at 50 °C. Also identified was the formation of the unusual product, N-(2-methylvinyl)-2-phenylindole 5, catalysed by [IrCp*Cl₂]₂-NaBF₄ from starting material 2via the incorporation of one molecule of acetone.

Full text of DOI:10.1039/b820357c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Application of UV-Vis spectroscopy to high throughput screening
of hydroamination catalysts
Danielle F. Kennedy, Barbara A. Messerle* and Sarah L. Rumble
Received (in Montpellier, France) 13th November 2008, Accepted 24th December 2008
First published as an Advance Article on the web 23rd January 2009
DOI: 10.1039/b820357c
Potential catalysts for the hydroamination of 2-(2-phenylethynyl)aniline 2 which have been
identified through high throughput screening methods were investigated. Two complexes were
shown to be highly active hydroamination catalysts in acetone: the in situ combinations of
[Rh(CO)2Cl]2–mesBIAN–NaBF4 (mesBIAN = bis(2,4,6-trimethylphenylimino)acenapthene) and
[Ir(COD)Cl]2–NaBF4 (COD = 1,5-cyclooctadiene). The isolated complexes [M(N–N)XCl]BF4
(M = Rh, Ir; N–N = bidentate nitrogen donor ligand; X = CO or Cp*) were found to be
inactive as catalysts for the conversion of 2 to 3. However, chloride abstraction from these
complexes through the addition of AgBF4 was found to generate extremely active catalysts.
Particularly active was the complex [Rh(CO)ClmesBIAN]–AgBF4 (5 mol%) which achieved
complete conversion of 2 to 3 in 12 minutes at 50 1C. Also identified was the formation of the
unusual product, N-(2-methylvinyl)-2-phenylindole 5, catalysed by [IrCp*Cl2]2–NaBF4 from
starting material 2 via the incorporation of one molecule of acetone.
We are particularly interested in finding highly efficient
catalysts for indole synthesis via hydroamination. Recently,
Introduction
The indole nucleus is commonly found in biologically active
we reported the application of parallel factor (PARAFAC)
molecules and is an important target for pharmaceutical
analysis to UV-Vis data obtained from high throughput
applications. The synthesis of indoles has been widely investi-
gated using homogeneous organometallic catalysts.1,2 Organo-
alkynylanilines.9 A library of rhodium and iridium complexes
catalyst screening for the intramolecular cyclisation of
metallic catalysts are effective in promoting the synthesis
of indoles under much milder conditions than would be
employed using traditional synthetic routes, and often
were generated in situ and were screened as catalysts for the
hydroamination of 2-(2-phenylethynyl)aniline 2, Scheme 1.
The progress of the hydroamination reactions was monitored
achieve better selectivity providing access to a wider range of
using UV-visible spectroscopy and PARAFAC analysis was
substituted indoles.
The intramolecular hydroamination of aromatic o-alkynyl-
used to deconvolute the spectral components due to the
solvent, starting material and product in an automated
anilines has been extensively studied previously as a route to
indoles.3 Palladium catalysts such as Pd(PPh3)4 are known to
amounts of product to starting material and product at a fixed
fashion. The catalysts were ranked using the ratio of relative
promote the intramolecular reaction at relatively high
time point. The use of PARAFAC leads to a qualitative
temperatures. Rhodium(I) and iridium(I) catalysts have
ranking of catalysts to identify any ‘hits’. By focussing only
also been studied previously, and the rhodium complex
[Rh(bis(N-methylimidazolyl)methane)(CO)2]BPh4, 1, is effec-
on the product formation, the method did not allow for the
identification of novel products.
tive under much milder conditions and lower catalyst loading
Here we report the application of an alternative method of
than is employed in palladium catalysed reactions.4 Recently,
data analysis to the screening of hydroamination catalysts
Crabtree and coworkers have identified an extremely active
using UV spectroscopy. Data from the same high throughput
screening as described previously9 were analysed indepen-
Ir(III) hydride complex which can promote intramolecular
hydroamination at room temperature (rt).5
The identification of the most effective transition metal
catalysts for any given transformation can be a time-consuming
dently and the method of data analysis was validated using a
thorough study of a portion of the library via 1H NMR
spectroscopy. ‘Hits’ identified by both methods of data
analysis (PARAFAC9 and UV) were investigated further
and expensive process when individual catalysts are synthesised
and tested. One approach to reducing the time required to
identify the most appropriate active catalysts is to utilise
and highly active catalysts were identified, synthesised and
tested as catalysts for the hydroamination of 2. As a result of
the parallel synthesis of libraries of metal complexes and
subsequent in situ high throughput screening for catalytic
activity.6–8
School of Chemistry, The University of New South Wales,
2052 Sydney, NSW, Australia. E-mail: b.messerle@unsw.edu.au;
Fax: +61 (0)2 9385 6141
Scheme 1
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
818 | New J. Chem., 2009, 33, 818–824

View Article Online
the application of this alternative method of data analysis,
a novel product, N-(2-methylvinyl)-2-phenylindole 5, was
identified.
Experimental
General procedures
The handling of all air sensitive reagents was performed in a
glove bag purged with argon. Reagents were purchased from
Aldrich and used as received unless otherwise stated. All solvents
were pre-dried and distilled under an atmosphere of argon.
Acetone and acetone-d6 were dried with CaSO4 and distilled
under argon prior to use. Methanol was dried over molecular
sieves (4 A) and distilled from CaH2. 1,1,2,2-Tetrachloro-
ethane (TCE) was dried with CaCl2 and distilled under
argon. Thf-d8 was distilled from Na–benzophenone imme-
diately prior to use. 1,2,3,4,5-Pentamethylcyclopentadiene
was purchased from Lancaster and used without further
purification. Ir(III) chloride hydrate and rhodium(III) chloride
hydrate were obtained from Precious Metals Online (PMO)
and were used without further purification. Sodium
Fig. 1 Graphical representation of well components and catalytic
conversions of 2 using direct UV analysis. Each well contains 10 mol%
[M]2, 20 mol% ligand and 20 mol% NaBF4, and the reaction was
performed at 50 1C for 18 h in acetone–MeOH (2.5 : 1) and was
analysed using a direct UV analysis.
tetrakis[(3,5-trifluoromethyl)phenyl]borate
(NaBArF24),10
diluted and analysed via UV spectroscopy. The UV absor-
bance data for each time point were normalised using an
isosbestic point in the data set and the internal standard series
included in column 12, Fig. 1, was utilised for calibration of
the UV absorbances.
[Ir(COD)Cl]2,11 [Ir(CO)2Cl]n,12 [Ir(COE)2Cl]2,11 [Rh(COD)Cl]2,13
[Rh(CO)2Cl]2,14 [Rh(COE)2Cl]2,11 [RhCpCl2]2,15 [RhCp*Cl2]2
and [IrCp*Cl2]2,15 bis(1-pyrazolyl)methane (bpm),16 bis(N-methyl-
imidazolyl)methane (bim),17 N,N-bis(p-tolyl)diazabutadiene
(ptolDAD),18
N,N-bis(p-tolyl)1,2-dimethyldiazabutadiene
(dmptolDAD),19,20 N,N-(bismesityl)1,2-dimethyldiazabuta-
diene (dmmesDAD),21 N,N-(bismesityl)diazabutadiene (mesDAD),
bis(2,4,6-trimethylphenylimino)acenapthene (mesBIAN),22,23
Synthesis of [RhClCO(MesBIAN)] (6)
[Rh(CO)2Cl]2 (27 mg, 0.07 mmol) and bis(2,4,6-trimethyl-
phenylimino)acenapthene (mesBIAN) (53 mg, 0.13 mmol)
were dissolved in dichloromethane (15 mL) to give a dark
green solution. The reaction mixture was stirred at rt, under an
atmosphere of nitrogen, for 10 min. The solvent volume was
reduced to approximately a third of the volume and hexane
(5 mL) was added. A dark green solid began to precipitate out.
The mixture was cooled in an ice-bath to give a dark green
solid and a light green filtrate. The solid was collected by
filtration, washed with hexane (2 ꢁ 1 mL) and dried in vacuo.
[RhCl(CO)(mesBIAN)] was obtained as a dark green powder
(50 mg, 70%), mp 246–252 1C (dec.); Anal (%) found: C,
63.06; H, 5.32; N, 4.28. C31H28N2OClRh ꢁ 0.5 H2O, requires
C, 62.68; H, 4.74; N, 4.79; vmax/cmꢂ1 (nujol): 1980s (Rh–CO);
1H NMR (d, 600 MHz, tetrahydrofuran-d8): 8.21 (1H, d,
2-(2-phenylethynyl)aniline 2,24 [Rh(bim)(CO)2]BPh4 1,4,25
26
[Ir(bpm)(CO)2]BPh4
(dmbpm)]BF4,
[Cp*IrCl(mesim-mim)]BF4 and [Cp*IrCl(mesBIAN)]BF4
and [Cp*IrCl(bpm)]BF4, [Cp*IrCl-
[Cp*IrCl(bim)]BF4,
[Cp*IrCl(bik)]BF4,
15
were prepared by literature methods.
1H NMR spectra were recorded on a Bruker DPX300,
DMX500 and 600 spectrometers. All spectra were recorded
1
at 323 K unless otherwise specified. H NMR chemical shifts
were referenced internally to residual solvent resonances. UV
spectra were recorded in Helma quartz 96-well plates using
Bio-Tek PowerWave XS plate reader and analysed using KC4
Microplate Data Analysis Software.
Screening of hydroamination in acetone–MeOH with NaBF4
as counterion
3
3
0
0
JH5-H4 = 8.2 Hz, 5-H), 8.17 (1H, d, JH5 -H4 = 8.2 Hz,
3
3
50-H), 7.53 (1H, dd, JH4 -H5 = 8.2 Hz, JH4 -H3 = 7.2 Hz,
40-H), 7.40 (1H, dd, 3JH4-H5 = 8.2 Hz, 3JH4-H3 = 7.2 Hz, 4-H),
0
0
0
0
Full experimental details describing the generation of the
library of rhodium and iridium complexes screened as
catalysts can be found in our previous report.9 The library
of complexes was generated in situ by reacting standard
solutions of 2 molar equivalents of ligand and 1 molar
equivalent of bimetallic metal precursor in acetone at rt,
along with 2 molar equivalents of NaBF4 which was added
as a MeOH standard solution. Fig. 1 illustrates the compo-
nents in each well of the 96-well plate. An acetone solution of
substrate 2 was then immediately added to each of the wells
and an aliquot was taken from each well for analysis. The
reaction block was sealed and heated at 50 1C for 18 hours,
after which time an aliquot was again taken from each well,
3
7.14 (2H, s, 10-H), 7.04 (2H, s, 100-H), 6.80 (1H, d, JH3 -H4
0
0
=
7.2 Hz, 30-H), 6.65 (1H, d, 3JH3-H4 = 7.2 Hz, 3-H), 2.43 (3H, s,
p-CH3), 2.404 (3H, s, p0-CH3), 2.399 (6H, s, o-CH3), 2.32 (6H,
s, o0-CH3) ppm; 13C NMR (d, 125 MHz, tetrahydrofuran-d8):
188.3 (d, 1JRh-CO = 79.8 Hz, Rh-CO), 175.0 (C10), 166.9 (C1),
148.7 (C8), 145.0 (C7), 144.4 (C80), 138.5 (C11), 137.6 (C110),
133.7 (C6), 132.7 (C50), 131.3 (C4), 130.9 (C10), 130.6 (C40),
130.5 (C90), 130.4 (C100), 130.3 (C2), 130.0 (C5), 129.6 (C9),
128.1 (C20), 125.5 (C30), 122.8 (C3), 21.9 (p0-CH3), 21.7
(p-CH3), 19.2 (o0-CH3), 18.3 (o-CH3) ppm; m/z: (ES+) 605
(M + Na, 32%), 519 (100).
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 818–824 | 819

View Article Online
Synthesis of N-(2-methylvinyl)-2-phenylindole (5)
catalysts in thf were confirmed as active catalysts using
1H NMR, with 6 achieving 495% conversion within 18 h at
50 1C, Table 1. However, in the screening where acetone was
solvent, the top PARAFAC ranked catalysts did not reach
complete conversion after 18 h using 1H NMR to monitor the
course of the reactions and several of the other catalysts
ranked in the top 10 by PARAFAC were found not to be
highly effective catalysts, Table 2. Due to the limited number
of highly active catalysts in acetone the application of
PARAFAC rankings to identify ‘hits’ in the screening results
in false positive identification of several complexes.
Under argon, 2 (100.8 mg, 0.521 mmol) and [IrCp*Cl2]2
(19.5 mg, 24.5 mmol) were dissolved in 10 mL acetone. The
mixture was refluxed at 55 1C for 18 h. The solution was
cooled and the majority of the Ir was removed via filtration.
The filtrate was run through a small silica column to remove
the remaining Ir and the acetone was removed in vacuo. The
compound 5 was isolated as a beige solid (77.8 mg, 0.33 mmol,
64%). The compound was found to be unstable in acidic
solutions losing the vinyl group generating 3. Anal (%) found:
C, 87.45; H, 6.67; N, 6.31. C17H15N requires C, 87.51; H,
In order to identify the most active catalysts from the
screening in acetone and to obtain quantitative conversions,
the UV-Vis spectra from the catalyst screening obtained in
acetone solvent were reanalysed. The spectra were normalised
using the isosbestic point of the data set. The conversion of
2-(2-phenylethynyl)aniline 2 was determined using a calibra-
tion curve generated from the internal standard series included
in the same plate. The conversion obtained in each well using
this direct UV analysis method is depicted schematically in
Fig. 1, with increasing colour intensity indicating increasing
conversion of 2-(2-phenylethynyl)aniline 2.
1
6.48; N, 6.01; H NMR (d, 300 Hz, 298 K, acetone-d6): 7.68
(2H, m, o-Ph), 7.59 (1H, m, 5-H) 7.49–7.32 (4H, m, 8-H and
m/p-Ph), 7.17 (1H, m, 7-H), 7.09 (1H, m, 6-H) 6.71 (1H, s,
3-H), 5.53 (1H, d, J = 1.51 Hz, CH2), 5.30 (1H, s, CH2), 1.79
(3H, s, Me) ppm; 13C {1H} NMR (75 MHz, acetone-d6):
d 141.4 (C10), 139.5 (C2), 138.0 (C4), 133.3 (C9), 128.4 (Ph),
127.8 (Ph), 127.6 (Ph), 122.0 (C7), 120.239 (C6 or C5), 120.220
(C6 or C5), 114.0 (CH2), 110.5 (C8), 103.2 (C3), 21.4
(CH3) ppm; GC-MS 19.43 min (ES+) m/z: 233 (M+, 95%),
217 (100).
Typical experimental procedure for the catalysed
1
hydroamination of 2 monitored by H NMR spectroscopy
1
Verification of direct UV analysis using H NMR spectroscopy
The results obtained from the direct UV analysis of the
screening results were confirmed using 1H NMR spectroscopy.
Each of the reactions from one column of the 96-well plate
(column 2 in Fig. 1 with [Ir(CO)2Cl]2 metal precursor) were
An example of a typical procedure: 2 (47.5 mg, 0.25 mmol),
[IrCp*Cl2]2 (4.8 mg, 6.0 mmol), mesBIAN (6.3 mg, 15.1 mmol)
and NaBF4 (1.5 mg, 13.7 mmol) were weighed into a NMR
tube fitted with a concentric Teflon valve. The tube was then
evacuated and acetone-d6 was vacuum-transferred into the
tube. The NMR tube was taken into an argon filled glove box
where the concentric valve was exchanged for a septum. TCE
was injected into the tube (20.9 mg, 0.124 mmol), the con-
centric valve was replaced under a cone of argon and the
sealed tube was placed into the NMR probe heated at 50 1C.
The reaction progress was monitored by 1H NMR spectro-
scopy. % Conversions given are determined directly from the
1H NMR spectra using TCE as an internal standard.
1
repeated individually and monitored using H NMR spectro-
scopy. These reactions were performed with 5 mol% [M]2,
10 mol% NaBF4 and 10 mol% ligand (where applicable) and 2
in acetone-d6. The catalyst loading was decreased in these
experiments relative to the loading used in the screening, and
the ratio of catalyst was decreased to minimize the interference
which the catalysts cause in the NMR spectra. The trend of the
1
catalytic results obtained from the H NMR study is consis-
tent with the results obtained in the UV screening, Fig. 2.
The most efficient catalyst from the screening in column
2 contained [Ir(CO)2Cl]2–dmmesDAD–NaBF4. This in situ
generated complex also proved to be the most active catalyst
Results and discussion
1
when individually repeated and monitored via H NMR.
Previously we have reported an efficient approach to the
identification of hydroamination catalysts utilising UV-Vis
spectroscopy as a screening method.9 A library of in situ
generated complexes were prepared and screened as catalysts
for the hydroamination of 2-(2-phenylethynyl)aniline 2,
Scheme 1. PARAFAC data analysis was used to identify
active catalysts. The catalysts were ranked according to the
ratio of product to starting material and product at a given
time. The PARAFAC analysis worked very well as a qualita-
tive screening technique in thf and several novel catalysts were
identified, Table 1.
To further confirm the direct UV analysis method each of
the in situ generated complexes which achieved 490% con-
versions were also repeated and monitored using 1H NMR
spectroscopy. These reactions were performed with 2.5 mol%
[M]2, 5 mol% NaBF4 and 5 mol% ligand (where applicable)
in acetone-d6 and the reactions were again monitored via
1H NMR spectroscopy. When individually repeated and
monitored, each of the four complexes achieved 495% con-
version within 18 h. These results confirm that the direct
analysis of UV-Vis spectra can be used to screen for active
catalysts using internal calibration standards to determine the
conversion of starting material.
Investigation of ‘hits’ and direct analysis of UV data
A detailed further investigation of the catalysts ranked in the
top 10 by PARAFAC analysis was performed. Each of the
reactions were repeated, generating the catalysts in situ on a
larger scale using 1H NMR spectroscopy to monitor the
conversion of 2. All but 3 of the top ten PARAFAC ranked
Synthesis of [RhCl(CO)(mesBIAN)] 6
One of the most efficient in situ generated catalysts was
[Rh(CO)2Cl]2–mesBIAN–NaBF4 achieving 495% conversion
in 5 hours. The complex assumed to be the active catalyst
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
820 | New J. Chem., 2009, 33, 818–824

View Article Online
Table 1 PARAFAC ranking for in situ generated catalysts from screening after 18 h along with efficiency of conversion of 2 by in situ generated
catalysts in isolated reactions monitored by H NMR spectroscopy in thf-d8
1
1
% Conv. via H NMR
In situ generated complex
Ranking using PARAFACa
[Ir(COD)Cl]2–NaBF4
[IrCp*Cl2]2–NaBF4
9
1
495
95
[IrCp*Cl2]2–mesBIAN–NaBF4
[IrCp*Cl2]2–bpm–NaBF4
[Rh(CO)2Cl]2–NaBF4
2
3
5
495
15
86
[Rh(CO)2Cl]2–mesBIAN–NaBF4
[Rh(CO)2Cl]2–bpm–NaBF4
[RhCpCl2]2–NaBF4
[RhCpCl2]2–bpm–NaBF4
[RhCp*Cl2]2–bpm–NaBF4
6
7
10
4
8
495
495
495
48
0
a
PARAFAC ranking gave relative efficiency of the catalysts in order 1–96.
Table 2 Comparison of PARAFAC ranking and conversion by direct UV analysis for in situ generated catalysts from screening after 18 h with
conversion of 2 by in situ generated catalysts in isolated reactions monitored by 1H NMR spectroscopy in acetone-d6
1
% Conv. via H NMR
In situ generated complex
PARAFAC ranking
% Conv. direct UV analysis
[Ir(COD)Cl]2–NaBF4
3
9
4
8
7
2
6
1
490
495
—
50
75
495
90
—
[Ir(COD)Cl]2–bpm–NaBF4
[Ir(CO)2Cl]2–dmesDAD–NaBF4
[Ir(CO)2Cl]2–bpm–NaBF4
[Rh(CO)2Cl]2–NaBF4
[IrCp*Cl2]2–mesBIAN–NaBF4
[IrCp*Cl2]2–dmesDAD–NaBF4
[Rh(CO)2Cl]2–NaBF4
50–70
70–90
70–90
70–90
490
50–70
490
80
[Rh(CO)2Cl]2–mesBIAN–NaBF4
[Rh(CO)2Cl]2–bpm–NaBF4
10
5
490
o50
495
—
[RhCl(CO)(mesBIAN)] 6 was synthesised, Scheme 2. Com-
plex 6 was synthesised using an adaptation of the method
of Gonsalvi et al., for the preparation of the complex
[RhI(CO)(dmmesDAD)], Scheme 2.27 Complex 6 was isolated
as a dark green air-stable powder, prepared in 70% yield.
Coordination of the mesBIAN ligand to the rhodium centre
was confirmed using NMR spectroscopy. In the 1H NMR
spectrum a loss of symmetry of the ligand and the downfield
shift of the ligand resonances were observed. The 13C NMR
spectrum revealed a resonance for a carbonyl bound to the
rhodium centre at 188.3 ppm, with a rhodium–carbon
coupling of 79.8 Hz. This coupling is typical of couplings seen
for rhodium–carbonyl complexes with rhodium in the +1
oxidation state.28 The IR spectrum revealed a single strong
peak for the carbonyl stretch at 1980 cmꢂ1, confirming a
mono-carbonyl complex had been formed.
Fig. 2 Comparison of in situ generated catalysts for the hydro-
amination of 2, 5 mol% [Ir(CO)2Cl]2, 10 mol% L and NaBF4 in
acetone-d6 50 1C monitored via 1H NMR. L = m none, B bim,
Investigation of [Ir(COD)Cl]2 and [Rh(CO)2Cl]2 derived
complexes as hydroamination catalysts
& bpm, K phen, ’ ptolDAD, n mesDAD,
E mesBIAN.
dmmesDAD,
One of the most active catalysts identified in the screening was
the metal precursor [Ir(COD)Cl]2. The influence of catalyst
Scheme 2
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 818–824 | 821

View Article Online
Table 3 Conversion of 2 to 3 by in situ generated complexes 5 mol% in acetone-d6, 50 1C monitored via 1H NMR spectroscopy using
1,1,2,2-tetrachloroethane as internal standard
Catalyst
TOF50%/hꢂ1
Time 495% conv.
% Conv. 18 h
a
[Ir(COD)Cl]2–NaBF4
[Ir(COD)Cl]2–NaBF4
[Ir(COD)Cl]2–mesDAD–NaBF4
5.2
1.67
0
0.4
1.7
0
2.5 h
11 h
N/A
N/A
5 h
N/A
12 min
24 min
495
495
0
80
495
0
a
[Rh(CO)2Cl]2–NaBF4
[Rh(CO)2Cl]2–mesBIAN–NaBF4
[Rh(mesBIAN)(CO)Cl]–NaBF4
[Rh(mesBIAN)(CO)Cl]–AgBF4
[Cp*IrCl(mesBIAN)]BF4–AgBF4
87
27
495
495
a
10 mol% [M].
loading, the presence of counterion, NaBF4, and the effect of
ligand on the catalytic activity were investigated, Table 3. In
the first instance, 2-(2-phenylethynyl)aniline 2 was treated with
5 mol% [Ir(COD)Cl]2 and 10 mol% NaBF4 in acetone-d6 at
in situ generated complexes of the type [Cp*Ir(N–N)Cl]BF4
(N–N = bidentate sp2-hybridised nitrogen donor ligand). The
indole product identified here by 1H NMR was, however, not
the expected 2-phenylindole 3, Scheme 1, but N-(1-methyl-
vinyl)-2-phenylindole 5, Scheme 3. This N-alkyl substituted
compound has not been reported previously. The vinylindole,
5, was isolated from the catalytic mixture and fully charac-
terised. The mechanism for the formation of this compound is
of interest; it is currently being investigated and will be the
subject of a future paper.
1
50 1C and the reaction was monitored by H NMR spectro-
scopy. Full conversion to 2-phenylindole 3 was achieved after
2.5 hours with a TOF50% of 5.2 hꢂ1.w Decreasing the catalyst
loading to 2.5 mol% [Ir(COD)Cl]2 significantly decreases the
catalytic activity per mole of catalyst decreasing the TOF50%
to 1.67 with complete conversion not achieved until 11 h.
Another series of complexes which were identified as active
hydroamination catalysts were those generated from the
[Rh(CO)2Cl]2 metal precursor. The precursor 2-(2-phenylethynyl)-
aniline 2 in the presence of 5 mol% [Rh(CO)2Cl]2 and 10 mol%
NaBF4 led to 80% conversion to 2-phenylindole 3 after 18 hours
at 50 1C in acetone. Incorporation of ligand into the catalyst
significantly increased the catalytic activity, with the in situ catalyst
generated from 2.5 mol% [Rh(CO)2Cl]2, 5 mol%, mesBIAN and
5 mol% NaBF4 achieving 495% conversion after 5 hours. The
equivalent isolated complex [Rh(CO)Cl(mesBIAN)] (either with
or without additional counterion (NaBF4)), however, did not
catalyse the conversion of 2, even after 18 hours. This indicates
that the active catalyst formed in situ is not the expected complex
[Rh(CO)Cl(mesBIAN)]. Chloride abstraction from metal com-
plexes using silver salts has been shown previously to generate an
active catalyst in situ.29–31 Chloride abstraction results in a
coordinatively unsaturated metal complex and significantly
increases the catalytic activity of an otherwise stable complex.
Abstraction of the chloride anion from the isolated complex
[Rh(CO)Cl(mesBIAN)] with AgBF4 led to a significant increase
in catalytic activity for the hydroamination of 2 generating the
most active catalyst identified in this study. A TOF50% of 87 hꢂ1
was obtained with complete conversion of 2 to 3 obtained after
12 min. This reaction rate is also amongst the best reported in the
literature for the hydroamination of 2, especially with the mild
conditions which are employed, in acetone at 50 1C.
All of the complexes generated in situ from a combination of
[IrCp*Cl2]2, bidentate N-donor ligands and NaBF4 promoted
the conversion of 2 to the unusual substituted indole 5 by
incorporating one molecule of acetone, Table 4. The most
efficient catalyst combination found was
5
mol%
[IrCp*Cl2]2–NaBF4 giving a TOF50% of 4.8 hꢂ1 and complete
conversion after 4 hours.
A series of complexes of the type [Cp*IrCl(N–N)]BF4,
presumed to be the complexes generated in situ, were syn-
thesised. The isolated complexes were tested as catalysts for
the cyclisation of 2-(2-phenylethynyl)aniline 2, Table 5. All of
the isolated complexes of this type which were tested were
inactive as catalysts for the cyclisation of 2 to form either 3 or
5. Chloride abstraction from these complexes via the addition
of one equivalent of AgBF4 to the complexes 7–9 generated
active catalysts which catalysed the hydroamination of 2 to
generate 3 with 100% conversion. Complex 10 promoted
complete conversion of the starting material 2 with 3 as the
major product and 5 as the minor product. Complexes 11 and
12 did not promote complete conversion of 2 but promoted
the generation of 3 and 5 in approximately a 1 : 1 ratio. The
abstraction of the chloride from these complexes provides a
readily available site on the metal for the binding of either 2 or
acetone, generating an active catalyst. The bidentate nitrogen
ligand was thought to be effective in stabilising the activated
catalyst as the activity of the [IrCp*Cl2]2 metal precursor after
chloride abstraction is initially high but drops off quickly as
the catalyst decomposes and deposits as an iridium film on the
reaction vessel.
Investigation of [IrCp*Cl2]2 complexes as hydroamination
catalysts—the catalysed formation of an N-alkyl substituted
indole
An interesting result identified in the high throughput catalysis
study was the complete conversion of 2-(2-phenylethynyl)-
aniline 2 by the metal precursor [IrCp*Cl2]2 as well as the
w TOF50% was calculated as the mole of product/mole of catalyst/time (h)
at 50% conversion.
Scheme 3
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
822 | New J. Chem., 2009, 33, 818–824

View Article Online
Table 4 Conversion of 2 to 5 by in situ generated complexes
5 mol% in acetone, 50 1C monitored via 1H NMR spectroscopy using
1,1,2,2-tetrachloroethane as internal standard
complexes are currently being pursued by the authors of
this paper.
Catalyst
TOF50%/hꢂ1 Time/h 495% conv.
Conclusions
[IrCp*Cl2]2
5.7
4.8
3.9
7.3
21
4
Catalysts identified through high throughput screening of a
library of Rh and Ir complexes were investigated further for
application to the hydroamination of 2-(2-phenylethynyl)-
aniline 2. Three complexes were identified to be highly active
hydroamination catalysts in acetone: the in situ combinations
of [Rh(CO)2Cl]2–mesBIAN–NaBF4 and [Ir(COD)Cl]2–NaBF4
and also Rh(CO)Cl(mesBIAN)–AgBF4 combination. The
most efficient catalyst, Rh(CO)Cl(mesBIAN) 6–AgBF4,
promoted the complete conversion of 2 to 3 within 12 minutes
at 50 1C.
[IrCp*Cl2]2–NaBF4
[IrCp*Cl2]2–NaBArF24
[IrCp*Cl2]2–AgBF4
4 (90% conv.)a
12
[IrCp*Cl2]2–mesBIAN–NaBF4 2.1
0
18 (90% conv.)a
N/A
[Cp*IrCl(mesBIAN)]BF4
a
No further conversion observed.
Table 5 Conversion (%) of 2 to 3 and/or 5 by in situ generated
complexes 5 mol% [Cp*Ir(N–N)Cl]BF4 and 5 mol% AgBF4 in
acetone, 50 1C for 3 h monitoring by H NMR
1
The isolated complexes of the type [Cp*IrCl(N–N)]BF4
were found to be inactive as hydroamination catalysts,
however, abstraction of the chloride by AgBF4 created
active catalysts in situ with the best catalyst in the series the
combination of [Cp*IrCl(mesBIAN)]BF4 9 and AgBF4.
Also identified from this work was the formation of the
unusual N-vinylindole, 5, from 2 via the incorporation of one
molecule of acetone, catalysed by [IrCp*Cl2]2–NaBF4. The
mechanism of this transformation will be the subject of a
future paper.
Catalyst
Conv.
Conv. to 3
Conv. to 5
[Cp*IrCl(bpm)]BF4 7
99
99
99
99
63
65
99
99
99
80
27
37
0
0
0
20
36
27
[Cp*IrCl(dmbpm)]BF4 8
[Cp*IrCl(mesBIAN)]BF4 9
[Cp*IrCl(bik)]BF4 10
[Cp*IrCl(bim)]BF4 11
[Cp*IrCl(mesim-mim)]BF4 12
1
Using H NMR to monitor the course of the reaction the
isolated complexes 7–9 were shown to be active catalysts for
the conversion of 2 to 3. The most active catalyst is the
combination of 5 mol% [Cp*IrCl(mesBIAN)]BF4 9 and
5 mol% AgBF4 in acetone-d6, Table 3 and Fig. 3. The catalyst
achieved complete conversion of 2 to 3 in 25 min in acetone-d6
at 50 1C. The efficiency of the silver salt AgBF4 as a catalyst
for the hydroamination of 2 was also tested, Fig. 3. It was
found to be an active catalyst for this reaction; with 5 mol%
catalyst 50% conversion was reached after 40 minutes, but
AgBF4 alone was much less efficient than the combination of
[Cp*IrCl(mesBIAN)]BF4 9 and AgBF4. However, the activity
of AgBF4 alone as a catalyst is interesting to note as it is
frequently used to generate active catalysts in situ without
consideration of its own potential role as a catalyst. Further
studies into the catalytic activity of AgBF4 and other Ag(I)
Acknowledgements
We gratefully acknowledge the financial support from the
University of New South Wales, the Australian Research
Council. We also thank the Australian government for the
provision of an Australian Postgraduate Award (D.F.K.).
References
1 I. Nakamura and Y. Yamamoto, Chem. Rev., 2004, 104,
2127–2198.
2 S. Cacchi and G. Fabrizi, Chem. Rev., 2005, 105, 2873–2920.
3 F. Pohlki and S. Doye, Chem. Soc. Rev., 2003, 32, 104–114.
4 S. Burling, L. D. Field and B. A. Messerle, Organometallics, 2000,
19, 87–90.
5 X. Li, A. R. Chianese, T. Vogel and R. H. Crabtree, Org. Lett.,
2005, 7, 5437–5440.
6 O. Lavastre and J. P. Morken, Angew. Chem., Int. Ed., 1999, 38,
3163–3165.
7 O. Lober, M. Kawatsura and J. F. Hartwig, J. Am. Chem. Soc.,
¨
2001, 123, 4366–4367.
8 J. Stambuli and J. F. Hartwig, Curr. Opin. Chem. Biol., 2003, 7,
420–426.
9 D. Ebrahimi, D. F. Kennedy, B. A. Messerle and D. B. Hibbert,
Analyst, 2008, 133, 817–822.
10 N. A. Yakelis and R. G. Bergman, Organometallics, 2005, 24,
3579–3581.
11 J. L. Herde, J. C. Lambert and C. V. Senoff, Inorg. Synth., 1974,
15, 18–19.
12 D. Roberto, E. Cariati, R. Psaro and R. Ugo, Organometallics,
1994, 13, 4227–4231.
13 C. Kocher and W. A. Herrmann, J. Organomet. Chem., 1997, 532,
¨
261–265.
14 J. A. McCleverty and G. Wilkinson, Inorg. Synth., 1990, 28, 84–86.
15 C. White, A. Yates and P. M. Maitlas, Inorg. Synth., 1992, 29,
228–234.
16 S. Julia, J. M. Mazo and L. Avila, Org. Prep. Proced. Int., 1984, 16,
´
299–307.
17 S. Elgafi, L. D. Field, B. A. Messerle, T. W. Hambley and
P. Turner, J. Chem. Soc., Dalton Trans., 1997, 2341–2345.
Fig. 3 Conversion of 2 to 3 by & 5 mol% 9,E5 mol% 9–AgBF4 and
m 5 mol% AgBF4.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 818–824 | 823

View Article Online
18 I. E. Buys, S. Elgafi, L. D. Field, T. W. Hambley and
B. A. Messerle, Inorg. Chem., 1994, 33, 1539–1542.
19 L. Stamp and H. Tom Dieck, Inorg. Chim. Acta, 1987, 129,
107–114.
20 J. M. Kleigman and R. K. Barnes, Tetrahedron, 1970, 26,
2555–2560.
25 S. Elgafi, L. D. Field, B. A. Messerle, P. Turner and
T. W. Hambley, J. Organomet. Chem., 1999, 588, 69–77.
26 L. D. Field, B. A. Messerle, L. P. Soler, M. Rehr and
T. W. Hambley, Organometallics, 2003, 22, 2387–2395.
27 L. Gonsalvi, J. A. Gaunt, H. Adams, A. Castro, G. J. Sunley and
A. Haynes, Organometallics, 2003, 22, 1047–1054.
28 B. E. Mann and B. F. Taylor, in 13C NMR Data for Organometallic
Compounds, Academic Press, London, 1981.
21 H. Tom Dieck and I. W. Renk, Chem. Ber., 1971, 104,
92–102.
22 R. van Asselt, C. J. Elsevier, W. J. Smeets, A. L. Spek and
R. Benedix, Recl. Trav. Chim. Pays-Bas, 1994, 113, 88–98.
23 M. Gasperini, F. Ragaini and S. Cenini, Organometallics, 2002, 21,
2950–2957.
29 F. E. Michael and B. M. Cochran, J. Am. Chem. Soc., 2006,
4246–4247.
30 C. Brouwer and C. He, Angew. Chem., Int. Ed., 2006, 45,
1744–1747.
24 A. Arcadi, S. Cacchi and F. Marinelli, Tetrahedron Lett., 1989, 30,
2581–2584.
31 S. R. Beeren, S. L. Dabb and B. A. Messerle, J. Organomet. Chem.,
2009, DOI: 10.1016/j.jorganchem.2008.10.038.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
824 | New J. Chem., 2009, 33, 818–824