Probing the mer- to fac-isomerization of tris-cyclometallated homo- and heteroleptic (C,N)3 iridium(III) complexes

Article abstract of DOI:10.1021/ic800169n

We have developed techniques which allow for covalent tethering, via a hetero cyclometallating ligand, of heteroleptic tris-cyclometallated iridium(III) complexes to polymeric supports (for application in light-emitting diode technologies). This involved the selective synthesis and thorough characterization of heteroleptic [Ir(C,N)₂(C′,N′)] tris-cyclometallated iridium(III) complexes. Furthermore, the synthesis and characterization of heteroleptic [Ir(C,N)₂OR] complexes is presented. Under standard thermal conditions for the synthesis of the facial (fac) isomer of tris-cyclometallated complexes, it was not possible to synthesize pure heteroleptic complexes of the form [Ir(C,N)₂(C′,N′)]. Instead, a mixture of homo- and heteroleptic complexes was acquired. It was found that a stepwise procedure involving the synthesis of a pure meridonial (mer) isomer followed by photochemical isomerization of this mer to the fac isomer was necessary to synthesize pure fac-[Ir(C,N)₂(C′, N′)] complexes. Under thermal isomerization conditions, the conversion of mer-[Ir(C,N)₂(C′,N′)] to fac-[Ir(C,N) ₂(C′,N′)] was also not a clean reaction, with again a mixture of homo- and heteroleptic complexes acquired. An investigation into the thermal mer to fac isomerization of both homo- and heteroleptic tris-cyclometallated complexes is presented. It was found that the process is an alcohol-catalyzed reaction with the formation of an iridium alkoxide [Ir(C,N)₂OR] intermediate in the isomerization process. This catalyzed reaction can be carried out between 50 and 100°C, the first such example of low-temperature mer-fac thermal isomerization. We have synthesized analogous complexes and have shown that they do indeed react so as to give fac-tris-cyclometallated products. A detailed explanation of the intermediates (and all of their stereoisomers, in particular when systems of the generic formula [M(a,b)₂(a′,b′')] are synthesized) formed in the mer to fac isomerization process is presented, including how the formed intermediates react further, and the stereoisomeric products they yield.

Full text of DOI:10.1021/ic800169n

Inorg. Chem. 2008, 47, 6681-6691
Probing the mer- to fac-Isomerization of Tris-Cyclometallated Homo-
and Heteroleptic (C,N)3 Iridium(III) Complexes
Aidan R. McDonald,† Martin Lutz,‡ Lars S. von Chrzanowski,‡ Gerard P. M. van Klink,†
Anthony L. Spek,‡ and Gerard van Koten*,†
Chemical Biology & Organic Chemistry, Debye Institute for Nanomaterials Science, Faculty of
Science, Utrecht UniVersity, Padualaan 8, 3584 CH Utrecht, The Netherlands, and Crystal and
Structural Chemistry, BijVoet Centre for Biomolecular Research, Faculty of Science, Utrecht
UniVersity, Padualaan 8, 3584 CH, Utrecht, The Netherlands
Received January 28, 2008
We have developed techniques which allow for covalent tethering, via a “hetero” cyclometallating ligand, of heteroleptic
tris-cyclometallated iridium(III) complexes to polymeric supports (for application in light-emitting diode technologies).
This involved the selective synthesis and thorough characterization of heteroleptic [Ir(C,N)2(C′,N′)] tris-cyclometallated
iridium(III) complexes. Furthermore, the synthesis and characterization of heteroleptic [Ir(C,N)2OR] complexes is
presented. Under standard thermal conditions for the synthesis of the facial (fac) isomer of tris-cyclometallated
complexes, it was not possible to synthesize pure heteroleptic complexes of the form [Ir(C,N)2(C′,N′)]. Instead, a
mixture of homo- and heteroleptic complexes was acquired. It was found that a stepwise procedure involving the
synthesis of a pure meridonial (mer) isomer followed by photochemical isomerization of this mer to the fac isomer
was necessary to synthesize pure fac-[Ir(C,N)2(C′,N′)] complexes. Under thermal isomerization conditions, the
conversion of mer-[Ir(C,N)2(C′,N′)] to fac-[Ir(C,N)2(C′,N′)] was also not a clean reaction, with again a mixture of
homo- and heteroleptic complexes acquired. An investigation into the thermal mer to fac isomerization of both
homo- and heteroleptic tris-cyclometallated complexes is presented. It was found that the process is an alcohol-
catalyzed reaction with the formation of an iridium alkoxide [Ir(C,N)2OR] intermediate in the isomerization process.
This catalyzed reaction can be carried out between 50 and 100 °C, the first such example of low-temperature
mer-fac thermal isomerization. We have synthesized analogous complexes and have shown that they do indeed
react so as to give fac-tris-cyclometallated products. A detailed explanation of the intermediates (and all of their
stereoisomers, in particular when systems of the generic formula [M(a,b)2(a′,b′)] are synthesized) formed in the
mer to fac isomerization process is presented, including how the formed intermediates react further, and the
stereoisomeric products they yield.
Introduction
application in combination with phosphorescent dopants,
which leads to even higher quantum efficiencies with
electrophosphoresence.2 Other applications include chemi-
luminescent devices,3 molecular oxygen sensors,4 and tem-
plates in oxalate-based chiral magnets.5
Present work in this field has been directed toward
augmenting ligand electronics so as to adjust energy levels
in tris-C,N-cyclometallated iridium(III) ([Ir(C,N)3]) com-
plexes and thus to improve quantum efficiencies and cover
the whole spectrum of color. This has resulted in a huge
range of aromatic cyclometallating ligands being tested and
used in homoligated complexes.6 Tris-cyclometallated iri-
dium complexes tend to show higher quantum yields and
Since the initial discovery by Watts and co-workers that
facial-tris[2-pyridinyl-κN-phenyl-κC2]iridium (fac-[Ir(p-
py)3]) complexes display emission quantum yields as high
as 0.4, tremendous effort has been invested into developing
similar compounds for use in organic light emitting diodes
(OLEDs).1 The photophysical properties of bis- and tris-
cyclometallated iridium(III) complexes are very interesting
for several applications. The most impressive of these is their
* Author to whom correspondence should be addressed. E-mail:
g.vankoten@uu.nl.
†
Debye Institute for Nanomaterials Science.
Bijvoet Centre for Biomolecular Research.
‡
10.1021/ic800169n CCC: $40.75  2008 American Chemical Society
Inorganic Chemistry, Vol. 47, No. 15, 2008 6681
Published on Web 06/28/2008

McDonald et al.
internal quantum efficiencies than their bis-cyclometallated
analogues. However, bis-cyclometallated complexes are still
very valuable photochemically, and thus a huge range of
charged heteroleptic [Ir(C,N)2L][X] complexes, where L is
a neutral chelating diamine or diphosphine ligand, have also
been developed.7 Neutral heteroleptic [Ir(C,N)2(E,E′)] com-
plexes exist with the heteroligand (E,E′) as a monoanionic
acetate (O,O′) or picolinate (N,O).8 Very few reports have
shown the selective synthesis of heteroleptic [Ir(C,N)2(C′,N′)]
complexes.9
[Ir(C,N)3]-type complexes have two stereochemical forms:
meridonial (mer) and facial (fac). The mer isomers possess
three nitrogens around the equator of the molecule with two
nitrogens trans to each other and the third nitrogen trans to
an aromatic carbon. In the fac-isomer all nitrogen atoms are
trans to a carbon atom. Previous work has shown the
differences between the photophysical10 and stereochemical11
properties of the mer and fac isomeric forms of the
complexes. Generally the fac isomer has an order of
magnitude longer emission lifetime (τ) than the mer.
Similarly, the quantum yield of emission (Φ) in the fac tends
to be an order of magnitude greater than that of the mer
isomer. Conversion of mer to fac isomers is possible using
both thermal and photochemical techniques.
We are interested in developing heteroleptic [Ir(C,N)2-
(C′,N′)] complexes, because this can provide an entry to tris-
cyclometallated Ir(III) complexes tethered to dendritic and
polymeric supports. Eventually, we would like to develop
OLED devices based on supported iridium complexes.
Binding of cyclometallated Ir(III) complexes to dendritic,12
linear polymeric,13 and biochemical supports14 has shown
some very promising results for the future development of
photochemical devices.
In this report, we present our initial results toward the
synthesis of heteroleptic [Ir(C,N)2(C′,N′)] complexes.15 The
synthesis of these complexes has thrown up some stumbling
blocks. It was found that using standard thermal techniques
it is not possible to synthesize pure fac heteroleptic com-
plexes, due to ligand scrambling. We devised a stepwise
procedure which involves the reaction of [Ir(ppy)2Cl] with
a heteroarene ligand, 4-methylphenyl-2-pyridine (Htolpy),
yielding a heteroleptic mer isomer, followed by conversion
of this mer to the corresponding fac isomer using photo-
chemical techniques. Furthermore, we also applied this
technique to synthesize mixed ligand functionalized species
for immobilization on polymeric supports.16 This has allowed
us to analyze the photophysical effects of the support on the
complex.17 We also investigated why, when using standard
thermal mer to fac isomerization techniques, a clean reaction
is not observed. The mechanism involved in the thermody-
namic mer to fac isomerization of tris-cyclometallated
iridium complexes was studied, and a number of analogues
of possible intermediates were synthesized. From these
investigations, we have developed novel methods to syn-
thesize both homo- and heteroleptic tris-cyclometallated fac-
(1) (a) King, K. A.; Spellane, P. J.; Watts, R. J. J. Am. Chem. Soc. 1985,
107, 1431–1432. (b) Dedeian, K.; Djurovich, P. I.; Garces, F. O.;
Carlson, G.; Watts, R. J. Inorg. Chem. 1991, 30, 1685–1688. (c)
Garces, F. O.; King, K. A.; Watts, R. J. Inorg. Chem. 1988, 27, 3464–
3471. (d) Sprouse, S.; King, K. A.; Spellane, P. J.; Watts, R. J. J. Am.
Chem. Soc. 1984, 106, 6647–6653. (e) Ohsawa, Y.; Sprouse, S.; King,
K. A.; DeArmond, M. K.; Hanck, K. W.; Watts, R. J. J. Phys. Chem.
1987, 91, 1047–1054.
(2) (a) Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Sibley, S.;
Thompson, M. E.; Forrest, S. R. Nature 1998, 395, 151–154. (b) Baldo,
M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R.
Appl. Phys. Lett. 1999, 75, 4–6. (c) Adachi, C.; Baldo, M. A.; Forrest,
S. R.; Thompson, M. E. Appl. Phys. Lett. 2000, 77, 904–906.
(3) Kim, J. I.; Shin, I.-S.; Kim, H.; Lee, J.-K. J. Am. Chem. Soc. 2005,
127, 1614–1615.
(4) (a) DeRosa, M. C.; Hodgson, D. J.; Enright, G. D.; Dawson, B.; Evans,
C. E. B.; Crutchley, R. J. J. Am. Chem. Soc. 2004, 126, 7619–7626.
(b) Gao, R.; Ho, D. G.; Hernandez, B.; Selke, M.; Murphy, D.;
Djurovich, P. I.; Thompson, M. E. J. Am. Chem. Soc. 2002, 124,
14828–14829.
(5) Clemente-Leo´n, M.; Coronado, E.; Gomez-Garcia, C. J.; Soriano-
Portillo, A. Inorg. Chem. 2006, 45, 5653–5660.
(6) (a) Huo, S.; Deaton, J. C.; Rajeswaran, M.; Lenhart, W. C. Inorg.
Chem. 2006, 45, 3155–3157. (b) Sajoto, T.; Djurovich, P. I.; Tamayo,
A.; Yousufuddin, M.; Bau, R.; Thompson, M. E.; Holmes, R. J.;
Forrest, S. R. Inorg. Chem. 2005, 44, 7992–8003. (c) Jung, S.; Kang,
Y.; Kim, H.-S.; Kim, Y.-H.; Lee, C.-L.; Kim, J.-J.; Lee, S.-K.; Kwon,
S.-K. Eur. J. Inorg. Chem. 2004, 3415–3423. (d) Okada, S.; Okinaka,
K.; Iwawaki, H.; Furugori, M.; Hashimoto, M.; Mukaide, T.; Kamatani,
J.; Ogawa, S.; Tsuboyama, A.; Tagiguchi, T.; Ueno, K. Dalton Trans.
2005, 1583–1590. (e) Su, Y.-J.; Huang, H.-L.; Li, C.-L.; Chien, C.-
H.; Tao, Y.-T.; Chou, P.-T.; Datta, S.; Liu, R.-S. AdV. Mater. 2003,
15, 884–888. (f) Grushin, V. V.; Herron, N.; LeCloux, D. D.; Marshall,
W. J.; Petrov, V. A.; Wang, Y. Chem. Commun. 2001, 1494–1495.
(7) (a) Li, J.; Djurovich, P. I.; Alleyne, B. D.; Yousufuddin, M.; Ho, N. N.;
Thomas, J. C.; Peters, J. C.; Bau, R.; Thompson, M. E. Inorg. Chem.
2005, 44, 1713–1727. (b) Nazeeruddin, M. K.; Humphry-Baker, R.;
Berner, D.; Rivier, S.; Zuppiroli, L.; Graetzel, M. J. Am. Chem. Soc.
2003, 125, 8790–8797. (c) Ionkin, A. S.; Marshall, W. J.; Fish, B. M.
Organometallics 2006, 25, 1461–1471. (d) Tamayo, A. B.; Garon,
S.; Sajoto, T.; Djurovich, P. I.; Tsyba, I. M.; Bau, R.; Thompson,
M. E. Inorg. Chem. 2005, 44, 8723–8732. (e) Yang, C.-H.; Li, S.-W.;
Chi, Y.; Cheng, Y.-M.; Yeh, Y.-S.; Chou, P.-T.; Lee, G.-H.; Wang,
C.-H.; Shu, C.-F. Inorg. Chem. 2005, 44, 7770–7780.
(11) Tamayo, A. B.; Alleyne, B. D.; Djurovich, P. I.; Lamansky, S.; Tsyba,
I.; Ho, N. N.; Bau, R.; Thompson, M. E. J. Am. Chem. Soc. 2003,
125, 7377–7387.
(12) Lo, S.-C.; Namdas, E. B.; Burn, P. L.; Samuels, I. D. W. Macromol-
ecules 2003, 36, 9721–9730.
(13) (a) Furuta, P. T.; Deng, L.; Garona, S.; Thompson, M. E.; Fréchet,
J. M. J. J. Am. Chem. Soc. 2004, 126, 15388–15389. (b) Sandee, A. J.;
Williams, C. K.; Evans, N. R.; Davies, J. E.; Boothby, C. E.; Kohler,
A.; Friend, R. H.; Holmes, A. B. J. Am. Chem. Soc. 2004, 126, 7041–
7048.
(14) Lo, K. K.-W.; Li, C.-K.; Lau, J. S.-Y. Organometallics 2005, 24, 4594–
4601.
(15) We have searched the literature to find reports on the stereochemistry
of octahedral [M(ab)2(a′b′)] complexes. There is very little reported
on such compounds, and we have therefore altered nomenclature
derived from [M(ab)2(a′a′)] systems. Please note: throughout the
following discussion, helicity of the various complexes is negated.
Any implied helicity in diagrams should be ignored. To the best of
our knowledge and understanding, the complexes are in racemic
mixtures. Any implied helicity is present to help with the understanding
of the geometrical isomer of a complex, unless otherwise stated.
(16) McDonald, A. R.; Mores, D.; de Mello Donega´, C.; van Walree, C. A.;
Klein Gebbink, R. J. M.; Lutz, M.; Spek, A. L.; Meijerink, A.; van
Klink, G. P. M.; van Koten, G. Submitted for publication.
(17) The properties of the polymer-supported complexes can be easily
altered, by both tethering to a support and altering the phobicity of
the copolymer/codendron.
(8) (a) You, Y.; Park, S. Y. J. Am. Chem. Soc. 2005, 127, 12438–12439.
(b) Kwon, T.-H.; Cho, H. S.; Kim, M. K.; Kim, J.-W.; Kim, J.-J.;
Lee, K. H.; Park, S. J.; Shin, I.-S.; Kim, H.; Shin, D. M.; Chung,
Y. K.; Hong, J.-I. Organometallics 2005, 24, 1578–1585.
(9) (a) Bo¨ttcher, H.-C.; Graf, M.; Kru¨ger, H.; Wagner, C. Inorg. Chem.
Commun. 2005, 8, 278–280. (b) Tsoboyama, A.; Takiguchi, T.; Okada,
S.; Osawa, M.; Hoshino, M.; Ueno, K. Dalton Trans. 2004, 1115–
1116. (c) Huo, S.; Deaton, J. C.; Rajeswaran, M.; Lenhart, W. C. Inorg.
Chem. 2006, 45, 3155–3157.
(10) Karatsu, T.; Nakamura, T.; Yagai, S.; Kitamura, A.; Yamaguchi, K.;
Matsushima, Y.; Iwata, T.; Hori, Y.; Hagiwara, T. Chem. Lett. 2003,
32, 886–887.
6682 Inorganic Chemistry, Vol. 47, No. 15, 2008

Isomerization of Tris-Cyclometallated Ir(III) Complexes
grade MeCN (500 mL). The solution was stirred in the presence
of a medium pressure 150 W mercury lamp for 4 days. Column
chromatography, using dichloromethane as the eluent, yielded the
desired product in quantitative yield as a bright yellow powder. 1H
NMR (300 MHz, d6-DMSO): δ 1.97 (s, 3H, CH3), 6.48 (s, 1H,
CH), 6.60 (d, 1H, CH), 6.67-6.70 (m, 4H, CH’s), 6.75-6.80 (m,
2H, CH’s), 7.02-7.11 (m, 3H, CH’s), 7.40-7.47 (m, 3H, CH’s),
7.61 (d, 1H, CH), 7.70-7.80 (m, 5H, CH’s), 8.04 (d, 1H, CH),
8.10 (2 × d, 2H, CH’s). 13C NMR (75 MHz, d6-DMSO): δ 21.491,
118.7, 119.0, 119.1, 119.4, 119.5, 120.6, 120.7, 122.1, 122.1, 122.7,
122.9, 124.1, 124.2, 129.0, 129.0, 136.2, 136.3, 136.7, 136.8, 136.9,
137.9, 141.3, 143.7, 143.8, 146.6, 146.7, 146.8, 160.8, 160.85,
160.9, 165.58, 165.6, 165.61. m/z 669.03 g/mol. C34H26IrN3 calcd.:
C, 61.06; H, 3.92; N, 6.28. Found: C, 60.98; H, 4.06; N, 6.20.
mer-Ir(tpy)2(ppy): mer-[bis(4-Methyl-(2-pyridinyl-KN)phenyl-
KC2)-mono((2-pyridinyl-KN)phenyl-KC2)]iridium(III), mer-8-
homo-N-trans. Bis(4-methyl-(2-pyridinyl-κN)phenyl-κC2)iridium-
(III) chloride (0.405 g, 0.718 mmol) was placed in glycerol (10
mL). To this was added K2CO3 (1.0 g, 10 equiv, 7.2 mmol) and
2-phenylpyridine (0.45 mL, 4 equiv, 2.87 mmol). The resulting
suspension was heated, under inert conditions, in an oil bath, to
150 °C for 24 h. This resulted in a dark brown solution. After the
solution was cooled to room temperature, deionized water was
added, and the mixture was vigorously mixed until a dark precipitate
was observed. The dark precipitate was filtered and washed with
water twice and subsequently with ethanol twice. The precipitate
was then purified using column chromatography with dichlo-
romethane as the eluent (yield 46%). The final product was bright
type complexes, including at lower temperatures. This report
contains a detailed discussion on the various stereochemical
forms of heteroleptic tris-cyclometallated iridium complexes.
Experimental Section
General Information. Standard Schlenk procedures under N2
were carried out throughout. Reactions were carried out in the
absence of light, unless otherwise stated. Reagents were used as
supplied from Acros BV or Sigma-Aldrich, unless otherwise stated.
The synthesis of 3 and 4 was carried out according to literature
procedures (see Scheme 1).1c,d The synthesis of 5 and 6 was carried
out according to the Thompson method.11 1H and 13C solution NMR
was carried out on a Varian Inova 300 spectrometer or a Varian
Oxford AS400. Elemental analyses were performed by Dornis and
Kolbe, Mikroanalytisches Laboratorium, Mu¨lheim a. d. Ruhr,
Germany. Mass spectrometry (MS) measurements were carried out
on an Applied Biosystems Voyager DE-STR matrix-assisted laser
desorption ionization-time of flight (MALDI-TOF) MS.
Photophysics. UV-vis absorption analysis was performed on a
Varian CARY 50 Scan UV-visible spectrophotometer in CH2Cl2.
Emission measurements were carried out on a SPEX FLUOROLOG
1680 0.22m Spectrometer in acetonitrile. fac-[Ir(ppy)3] in 2-MeTHF
was used as a reference (Φ ) 0.4).
Electrochemistry. CV measurements were carried out on an
EG&G Princeton Applied Research Potentiostat Model 263A.
Experiments were carried out at room temperature (20 °C). A
platinum disk working electrode was polished with alumina on felt
before use. A platinum wire was used as the counter electrode. A
silver wire was used as a pseudo-/quasi-reference electrode.
Tetrabutylammonium hexafluorophosphate (0.1 M) in MeCN was
used as the electrolyte. The scan rate was 0.1 V/s. The silver
reference electrode was calibrated using the ferrocene/ferrocenium
(Fc/Fc+) redox couple as an internal standard. The oxidation
potential of Fc/Fc+ was found to be 0.51 V against the silver
reference electrode.
1
orange in color and was a fine powder. H NMR (300 MHz, d6-
DMSO): δ 2.02 (s, 3H, CH3), 2.04 (s, 3H, CH3), 6.08 (s, 1H, CH),
6.23 (s, 1H, CH), 6.62 (d, 1H, CH), 6.71 (2 × d, 2H, CH’s),
6.80-6.90 (m, 4H,CH’s), 7.10 (t, 1H, CH), 7.48 (d, 1H, CH),
7.60-7.70 (m, 4H, CH’s), 7.77-7.84 (m, 3H, CH’s), 7.91-7.99
(m, 3H, CH’s), 8.14 (d, 1H, CH). 13C NMR (75 MHz, d6-DMSO):
δ 21.7, 118.2, 118.6, 119.3, 120.2, 121.0, 121.4, 121.75, 122.2,
122.5, 124.1, 124.3, 124.5, 129.9, 131.3, 133.7, 134.3, 135.7, 136.7,
137.8, 139.6, 139.9, 140.0, 142.5, 145.7, 147.9, 151.4, 153.1, 159.9,
167.9, 168.5, 170.6, 175.4, 177.8. m/z 683.19 g/mol. Elem anal.
for C35H28IrN3, calcd.: C, 61.56; H, 4.13; N, 6.15. Found: C, 61.51;
H, 4.12; N, 6.10.
mer-Ir(tpy)(ppy)2: mer-[mono(4-Methyl-(2-pyridinyl-KN)phe-
nyl-KC2)-bis((2-pyridinyl-KN)phenyl-KC2)]iridium(III), mer-7-
homo-N-trans. Bis((2-pyridinyl-κN)phenyl-κC)iridium(III) chloride
(0.2045 g, 0.381 mmol) was placed in glycerol (2 mL). To this
was added K2CO3 (0.525 g, 10 equiv, 3.8 mmol) and 4-methylphe-
nyl-2-pyridine (0.25 mL, 4 equiv, 1.5 mmol). The resulting
suspension was heated, under inert conditions, in an oil bath, to
150 °C for 40 h. This resulted in a dark brown solution. After the
solution was cooled to room temperature, deionized water was
added, and the mixture was vigorously mixed until a dark precipitate
was observed. The dark precipitate was filtered and washed with
water twice and subsequently with ethanol twice. The precipitate
was then purified using column chromatography with dichlo-
romethane as the eluent (yield 84%). The final product was bright
fac-Ir(tpy)2(ppy): fac-[bis(4-Methyl-(2-pyridinyl-KN)phenyl-
KC2)-mono((2-pyridinyl-KN)phenyl-KC2)]iridium(III), fac-8. Mer-
[Ir(tpy)2(ppy)] (0.1038 g, 0.15 mmol) was dissolved in spectrometric
grade MeCN (600 mL). The solution was stirred in the presence
of a medium-pressure 150 W mercury lamp for 4 days. Column
chromatography, using dichloromethane as the eluent, yielded the
desired product in quantitative yield as a bright yellow powder. 1H
NMR (300 MHz, d6-DMSO): δ 1.95 (s, 3H, CH3), 1.96 (s, 3H,
CH3), 6.45 (s, 1H, CH), 6.49 (s, 1H, CH), 6.59 (2x d, 2H, CH’s),
6.67 (2 × d, 2H, CH’s), 6.77 (m, 1H, CH), 7.02 (m, 2H, CH’s),
7.07 (t, 1H, CH), 7.35-7.40 (m, 3H, CH’s), 7.60 (2 × d, 2H, CH’s),
7.74 (m, 4H, CH’s), 8.02 (t, 2H, CH), 8.08 (d, 1H, CH). 13C NMR
(75 MHz, d6-DMSO): δ 22.2, 22.4, 119.2, 119.3, 119.4, 119.7,
120.2, 121.5, 121.6, 122.8, 123.4, 124.7, 124.8, 124.9, 129.8, 137.0,
137.4 137.52, 137.55, 137.58, 137.60, 137.7, 138.5, 138.55, 138.60,
141.90, 141.93, 144.4, 147.2, 147.3, 147.4, 161.7, 161.8, 166.3,
166.4. m/z 683.19 g/mol. Elem anal. for C35H28IrN3, calcd.: C,
61.56; H, 4.13; N, 6.15. Found: C, 61.47; H, 4.15; N, 6.11.
mer-Ir(ppy)2OPh: mer-[bis((2-Pyridinyl-KN)phenyl-KC2)-mono-
phenoxo]iridium(III), 9. Freshly distilled, degassed phenol (0.8532
g, 0.009 mol) was added to a thoroughly degassed THF (20 mL)
solution with lump Na (0.2 g, 9 mmol). This was stirred at room
temperature until all Na had reacted (H2 evolution!). The system
1
orange in color and was a fine powder. H NMR (300 MHz, d6-
DMSO): δ 2.00 (s, 3H, CH3), 6.44 (d, 1H, CH), 6.59 (d, 1H, CH),
6.71 (s, 1H, CH), 6.75-7.05 (m, 8H,CH’s), 7.50-7.75 (m, 7H,
CH’s), 7.87 (2 × d, 2H, CH’s), 7.91 (2 × d, 2H, CH’s), 8.15 (d,
1H, CH). 13C NMR (75 MHz, d6-DMSO): δ 22.29, 119.2, 119.4,
119.8, 119.9, 121.5, 122.7, 123.1, 123.2, 123.3, 124.8, 125.0, 125.2,
129.8, 130.2, 130.7, 132.7, 135.5, 137.2, 138.1, 138.5, 138.8, 143.0,
143.4, 145.4, 148.1, 150.8, 153.0, 160.0, 167.7, 168.3, 170.4, 175.5,
177.7. m/z 669.03 g/mol. C34H26IrN3 calcd.: C, 61.06; H, 3.92; N,
6.28. Found: C, 61.11; H, 3.96; N, 6.11.
fac-Ir(tpy)(ppy)2: fac-[mono(4-Methyl-((2-pyridinyl-KN)phe-
nyl-KC2)-bis((2-pyridinyl-KN)phenyl-KC2)]iridium(III), fac-7. Mer-
[Ir(tpy)(ppy)2] (0.2 g, 0.3 mmol) was dissolved in spectrometric-
Inorganic Chemistry, Vol. 47, No. 15, 2008 6683

McDonald et al.
was kept under N2, and bis((2-pyridinyl-κN)phenyl-κC2)iridium(III)
chloride (0.2 g, 4 mmol) was added with continuous stirring. The
resulting yellow solution was stirred at the same temperature for
24 h, after which it had become red. The solution was filtered to
remove Na salts. The solvent was removed from the filtrate, and
EtOH was added, precipitating a dark orange powder. Yield: 48%.
1H NMR (300 MHz, d6-DMSO): δ 5.22 (d, 2H, CH), 5.64, (d, 2H,
CH), 6.01 (t, 1H, OPhCH), 6.17 (t, 2H, CH), 6.28 (t, 2H, CH),
6.64 (t, 2H, CH), 6.83 (t, 2H, CH), 7.68 (d, 2H, CH), 7.84 (t, 2H,
CH), 8.23 (d, 2H, CH), 8.77 (d, 2H, CH). 13C NMR (75 MHz,
d6-DMSO) δ ) 113.0, 119.8, 120.7, 122.7, 123.1, 123.5, 124.4,
125.3, 127.6, 128.4, 129.3, 129.4, 138.0, 143.5, 145.3. Elem anal.
C28H21IrON2, calcd.: C, 56.65; H, 3.57; N, 4.72. Found: C, 56.79;
H, 3.69; N, 4.75.
Scheme 1. The Thompson Method for the Synthesis of Homoleptic
Tris-Cyclometallated Iridium(III) Complexes11
Table 1. Details of the X-Ray Crystal Structure Determinations
mer-7-homo-N-trans
mer-8-homo-N-trans
formula
C34H26IrN3 +
disordered solvent
668.78a
C35H28IrN3 +
disordered solvent
682.80a
mer-Ir(ppy)2OMe: mer-[bis((2-Pyridinyl-KN)phenyl-KC2)-
monomethoxo]iridium(III), 10. Lump Na (0.23 g, 10 mmol) was
added to dry, deoxygenated methanol (40 mL) and dry, degassed
THF (10 mL). Once all of the Na had dissolved, bis(2-pyridinyl-
κN)phenyl-κC2)iridium(III) chloride (0.38 g, 0.7 mmol) was added
to the solution. The resulting mixture was stirred at reflux for 2 h,
showing a color change from yellow to bright orange. After cooling,
the precipitated bright red powder was filtered and recrystallized
using a large volume of dichloromethane and hexanes. The final
compound was highly insoluble in all organic solvents.
X-Ray Crystal Structure Determinations. Reflections were
measured on a Nonius Kappa CCD diffractometer with rotating
anode (graphite monochromator, λ ) 0.71073 Å) up to a resolution
of (sin θ/λ)max ) 0.65 Å-1. The structures were solved with Direct
Methods (SIR-9718) and refined with SHELXL-9719 against the
F2 of all reflections. Non-hydrogen atoms were refined with
anisotropic displacement parameters. All hydrogen atoms were
introduced in calculated positions and refined with a riding model.
Geometry calculations and checking for higher symmetry was
performed with the PLATON program.20 Further details are given
in Table 1.
fw
cryst color
cryst size [mm3]
temp [K]
cryst syst
space group
a [Å]
yellow
yellow
0.12 × 0.09 × 0.06
0.45 × 0.24 × 0.21
150
110
monoclinic
C2/c (no. 15)
25.0705(1)
35.7759(2)
17.6973(1)
133.3535(2)
11541.81(11)
16
trigonal
j
R3c (no. 167)
20.02588(1)
20.02588(1)
36.14264(2)
b [Å]
c [Å]
ꢀ [deg]
V [Å3]
12552.593(11)
18
Z
Dx [g/cm3]
1.539a
1.626a
µ [mm-1
]
4.653a
4.815a
abs. corr. method
abs. corr. range
reflns (meas./unique)
param/restraints
R1/wR2 [I > 2σ(I)]
R1/wR2 [all reflns.]
S
multiscan
0.65-0.75
122103/13252
685/0
0.0296/0.0685
0.0453/0.0732
1.050
multiscan
0.07-0.37
51277/3176
178/0
0.0229/0.0496
0.0299/0.0514
1.088
Fmin/max [e/Å3]
-1.47/2.12
-0.77/1.62
a Derived parameters do not contain the contribution of the disordered
solvent.
mer-7-homo-N-trans. The crystal structure contains large voids
(1756 Å3/unit cell) filled with disordered solvent molecules. Their
contribution to the structure factors was secured by back-Fourier
transformation using the SQUEEZE routine of the program PLA-
TON,20 resulting in 590 electrons/unit cell.
mer-8-homo-N-trans. The crystal structure contains large voids
(708 Å3/unit cell) filled with disordered solvent molecules. Their
contribution to the structure factors was secured by back-Fourier
transformation using the SQUEEZE routine of the program PLA-
TON,20 resulting in 163 electrons/unit cell. Atoms N2 and C18
were constrained to the same coordinates and displacement
parameters (occupancy 0.5).
reaction techniques developed by Thompson.11 The fac
isomer of 5 (from 1 and 3) and 6 (from 2 and 4) was
synthesized when temperatures above 200 °C were used. At
temperatures of approximately 150 °C, the mer isomer was
formed (Scheme 1).
The reaction of 2 with 3 (and likewise 1 with 4) using
Thompson conditions for the synthesis of fac isomers did
not proceed cleanly (Figure S1, Supporting Information). 1H
NMR of the isolated products showed the presence of a
mixture of several similar complexes. UV-vis absorption
spectroscopy suggested that these were fac-type products.
MALDI-TOF mass analysis, however, showed that scram-
bling of the ligands had occurred, and in fact, a mixture of
four different complexes (fac-5-8) had been synthesized
(Figure S1, Supporting Information). Separation of the
individual complexes from this fac-5-8 mixture was at-
tempted but, in our hands, was not possible.
Results and Discussion
1. Synthesis and Characterization of Heteroleptic
Complexes [Ir(C,N)2(C′,N′)]; mer-Isomers of 7 and 8. All
experiments in the following section were carried out in the
absence of light.
The synthesis of [Ir(C,N)2Cl] complexes 3 and 4 was
carried out according to literature procedures.1c,d Tris-
cyclometallated complexes 5 and 6 were synthesized using
The reaction of 2 with 3 (and likewise 1 with 4) using the
Thompson reaction conditions for the synthesis of mer
isomers, gave only one product, a mer isomer of 7 or 8,
1
respectively (Scheme 2). H and 13C NMR of the acquired
(18) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giaco-
vazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna,
R. J. Appl. Crystallogr. 1999, 32, 115–119.
mer-7 and mer-8 gave conclusive proof that the desired
products were obtained. UV-vis absorption spectroscopy
results corroborated that a mer isomer had indeed been
synthesized and was isolated as a pure product (see the
(19) Sheldrick, G. M. SHELXL-97; University of Go¨ttingen: Go¨ttingen,
Germany, 1997.
(20) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7–13.
6684 Inorganic Chemistry, Vol. 47, No. 15, 2008

Isomerization of Tris-Cyclometallated Ir(III) Complexes
Scheme 2. Heteroleptic mer Complex Synthesis
molecules (see Experimental Section). The small differences
in the residues can be clearly seen in the fit diagram in Fig-
ure 1.
Single crystals suitable for X-ray analysis of mer-8 were
acquired by slow evaporation of an ethanol/dichloromethane
solution. The system was trigonal with space group R3c.25
The molecular structure confirmed that a mer isomer had
indeed been isolated (Figure 2, homoligand N’s trans to each
other).26 The complex is assigned the name mer-8-homo-
N-trans because the nitrogen atoms of the homoligands
(tolpy) are arranged trans to each other.
3. Stereochemical Aspects of Heteroleptic [Ir(C,N)2-
(C′,N′)] Complexes. The complexes mer-7- and mer-8-
homo-N-trans are geometrical isomers under the general
headings mer-7 and mer-8, respectively. There are three
possible geometrical isomers of both mer-7 and mer-8 which
are depicted in Figure 3. The homo-N-trans isomer is the
acquired product from the reaction of C2-trans starting
material 3. Each of the depicted geometrical isomers also
has two enantiomers/helimers, Λ and ∆. We have found no
reports commenting on any other isomer than homo-N-trans-
type isomers in bis-cyclometallated (C,N) iridium(III) com-
plexes.
It is timely at this point to discuss why the formation of
only the mer-homo-N-trans isomers is observed in the
synthesis of mer-7 and mer-8. Complexes of the type 3 or 4
are believed to be dimeric octahedral C2-trans-N isomers
(see the Supporting Information for detailed discussion). The
molecular geometry of 4 in the solid state was previously
reported showing a dimeric complex with C2-trans-N ge-
ometry.23 Previous reports point toward the likelihood that
the dimer is in equilibrium with a monomer in a noncoor-
dinating solution.27 All of these observations point to only
N-trans-type compounds, and absolutely no isomerization
to other isomers is observed (for a detailed discussion on 3
and the intermediates in the synthesis of 3 and similar
systems, see the Supporting Information).
The trans-N configuration of 3 holds when other ligands,
either monodentate or bidentate monoanionic ones, are
substituted for the halide, for example, cyanide7c and acac27a
complexes. We have also observed that no C2-trans-N to C1
or C2-trans-C (likewise tbp-trans-N to tbp-cis- or tbp-trans-
C) isomerization of these complexes occurs even under
relatively extreme conditions.28 No ligand scrambling or the
formation of non-homo-N-trans products was observed in
the synthesis of mer-7-homo-N-trans, suggesting that ho-
moligands remain bound, in the trans-N fashion, to the
iridium center, throughout the reaction.
Supporting Information for all UV-vis spectra: Figure S4
and Table S2). In homoligated complexes ([Ir(C,N)3] 5 and
6), the mer and fac isomers have dissimilar UV-vis
absorption spectra, with the mer isomer showing one
significant absorption at 272 nm, whereas the fac isomer
shows three strong absorptions at 245, 283, and 376 nm.
The synthesized heteroligated mer isomers, mer-7 and mer-
8, have a UV-vis absorption spectrum distinctly similar to
those observed for the mer isomers of 5 and 6.
2. Molecular Geometries of Acquired mer-7 and mer-
8 in the Solid State. Single crystals of mer-7, suitable for
X-ray diffraction, were acquired from slow evaporation of
a dichloromethane solution. The crystal system was mono-
clinic with the centrosymmetric space group C2/c and two
independent molecules in the asymmetric unit.21 The mo-
lecular structure of both independent molecules confirmed
that a mer isomer had indeed been isolated (Figure 1).22 It
is also noteworthy that the homoligands have retained their
position relative to each other. The C2-trans-(N) (the
molecule is C2 symmetrical with N’s lying trans to each
other) configured octahedral dimeric starting material 3 has
two ppy ligands with nitrogens aligned trans to each other
while the ortho carbons lie mutually cis to each other and
trans to the bridging halides.23 In the acquired mer-7, the
carbons of the homoligands bound to iridium are now trans
to the heteroligand. The N’s of the homoligands lie trans to
one another. Therefore, we call it mer-7-homo-N-trans15
(there are two other geometrical isomers of mer-7, this is
discussed in the following section).24 It should be noted that
15% of the unit cell is filled with disordered solvent
(21) See the Experimental Section.
(22) The three cyclometallating ligands are aligned in an octahedral
configuration around the iridium centre. The C-C and C-N bond
lengths and angles are within normal ranges expected for mer-tris-
cyclometallated iridium(III) complexes. (a) Garces, F. O.; Dedeian,
K.; Keder, N. L.; Watts, R. J Acta Crystallogr. 1993, C49, 1117–
1120. (b) See ref 11. The monoanionic heteroligand, 2-(4-methylphe-
nyl)pyridine (tolpy), has longer Ir-N bond distances (2.142(5) and
2.136(3) Å, respectively) compared to those of the homoligands (2-
(phenyl)pyridine, ppy) which have their nitrogens aligned trans to each
other (2.047(3) and 2.032(3) Å, respectively, Supporting Information,
Table S1). This can be explained by the different trans influence both
on carbon and on nitrogen. However, subtle effects, for example, the
role of the tolyl-CH3 group, could also be influencing the Ir-N or
Ir-C bond distances.
(25) See the Experimental Section.
(26) The three cyclometallating ligands are aligned in an octahedral
configuration around the iridium center. As with the previous crystal
structure, the C-C and C-N bond lengths and angles are within
normal ranges expected for mer-tris-cyclometallated iridium(III)
complexes.
(27) (a) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.;
Kwong, R.; Tsyba, I.; Bortz, M.; Mui, B.; Bau, R.; Thompson, M. E.
Inorg. Chem. 2001, 40, 1704–1711. (b) See ref 1c.
(23) (a) Schmid, B.; Garces, F. O.; Watts, R. J. Inorg. Chem. 1994, 33,
9–14. (b) Douglas, B. E.; Saito, Y. ACS Symp. Ser. 1980, 119, 338.
(24) Due to the centrosymmetry of the space group, the crystal is a racemic
mixture of both Λ and ∆ stereoisomers.
(28) This is related to the slow rate of isomerization of Ir(III), and the
disfavor of such systems to undergo Berry pseudorotation. Deeming,
A. J.; Proud, P. J.; Dawes, H. M.; Hursthouse, M. B.; J. Chem. Soc.,
Dalton Trans. 1986, 2545-2549.
Inorganic Chemistry, Vol. 47, No. 15, 2008 6685

McDonald et al.
Figure 1. (left) Displacement ellipsoid plot (50% probability level) of the two independent molecules of mer-7-homo-N-trans in the crystal structure. View
along the crystallographic a,c diagonal. Hydrogen atoms and disordered solvent molecules have been omitted for clarity. (right) Quaternion fit overlay plot
of residue 1 and the inverted residue 2.
Figure 2. Displacement ellipsoid plot (50% probability level) of the mer-
8-homo-N-trans isomer in the crystal. Hydrogen atoms and disordered
solvent molecules have been omitted for clarity. Symmetry operation i: 1/3
Figure 3. Geometrical isomers of mer-7 and mer-8 (the highlighted
+ x - y, 2/3 - y, 1/6 - z.
compounds are the synthesized species).
Because the homoligands remain trans-N in the synthesis
of mer-7-homo-N-trans, we assume they play no role in the
synthetic techniques with analogous systems were published
reaction, and their geometry never changes with respect to
during the time this work was carried out.30 The 1H and 13
C
each other during the cyclometalation reaction. If there was
any geometrical positional ligand exchange, different geo-
metrical isomers would definitely be expected.29
4. Synthesis and Characterization of Heteroleptic
Complexes [Ir(C,N)2(C′,N′)]; fac-Isomers of 7 and 8
Obtained by mer-to fac-Isomerization. All experiments in
the following section were carried out in the absence of
natural light.
The acquired complexes mer-7- and mer-8-homo-N-trans
could be quantitatively converted to fac-7 and fac-8 photo-
chemically, without scrambling of the ligands, by stirring
the complexes in deoxygenated acetonitrile at room temper-
ature over 4 days with a medium-pressure 150 W Hg UV
lamp submersed in the reaction solution (Scheme 3). Similar
NMR spectra of fac-7 and -8 showed a slight shift in the
resonance ppm of all aromatic and tolyl CH’s in comparison
with mer-7- and mer-8-homo-N-trans. UV absorption spectra
showed three distinct absorptions at 245/247 nm, 285/286
nm, and 376/376 nm, respectively (see the Supporting
Information, Figure S4), similar to that of fac-5/6. Fac-7 and
-8 have only one geometrical isomer. They both have two
enantiomers/helimers, Λ and ∆. The attained fac-7 and -8
are racemic mixtures. Crystals suitable for X-ray diffraction
of fac-8 were acquired. However, the CH3 groups could not
be localized on a specific ligand. Partially populated
tolyl-CH3 positions of all three ligands were observed (CIF
file available in the Supporting Information).
6686 Inorganic Chemistry, Vol. 47, No. 15, 2008

Isomerization of Tris-Cyclometallated Ir(III) Complexes
Scheme 3. Photochemical Conversion of mer-7- and mer-8-homo-N-trans to fac-7 and fac-8
Scheme 4. Synthesis of Alkoxide Iridium Complexes and Possible
Isomers of Alkoxide Complexes
Scheme 5. Attempted Thermodynamic Conversion of
mer-7-homo-N-trans to fac-7
very low solubility in all organic solvents and thus was not
characterized using NMR techniques.
Heating of complex 9 in phenolic solution at reflux showed
no change in the constitution of the complex; the complex
stayed in the C2-trans-N/tbp-trans-N configuration. To study
whether complexes 9 and 10 could be the starting point for
the synthesis of tris-cyclometallated [Ir(C,N)3] complexes,
they were separately dissolved in 1,2-dichlorobenzene
(mer-fac isomerization does not occur in this solvent, bp
180 °C) with 1 equiv of Hppy and heated to reflux (in
darkness). After 24 h, the reactions were stopped. With both
1
complexes 9 and 10, H NMR and UV-vis spectroscopy
both showed the presence of [tris((2-pyridinyl-κN)phenyl-
κC2)]iridium(III) complex 5. In the case of 9, more than 50%
was fac-5. Similarly, complexes 9 and 10 were reacted with
Htolpy. The formation of tris-cyclometallated products was
also observed, however, with scrambling of the ligands.
These are the only examples of nonphotochemical fac-
product formation in hydroxyl-free solvents. Obviously, 1
equiv of ROH is released during this reaction, which could
then promote mer-to-fac isomerization of any formed mer
species (see later in this report).
5. Synthesis and Characterization of Heteroleptic
[Ir(C,N)2(alkoxide)] Complexes. All experiments in the
following section were carried out in the absence of light.
We believe iridium alkoxide complexes to be intermediates
in the thermal mer-to-facisomerization. To this date, all
reports on thermal isomerization (also see later in this report)
have been carried out in alcoholic solvents. We have
therefore synthesized iridium alkoxide [Ir(C,N)2OR] com-
plexes and studied their reaction with cyclometallating
ligands. Complex 3 was added to a solution of NaOPh in
large excess. A color change was observed after heating, from
yellow to dark orange. A single isomer of 9 was isolated
6. Investigation of Thermal mer-to-fac Isomerization.
All experiments in the following section were carried out in
the absence of light.
Earlier studies have shown that heating of mer-[Ir(ppy)3]
in glycerol at 200 °C for 24 h yields fac-[Ir(ppy)3].11 Heating
of either mer-7- or -8-homo-N-trans in glycerol at 200 °C
for 24 h did not yield pure fac-7 or -8, respectively, but
instead a mixture of complexes 5-8 (Scheme 5). UV-vis
absorption suggested that fac-type isomers were present.
MALDI-TOF mass analysis showed a similar pattern of the
fac-5-8 mixture, as is depicted in Figure S1, Supporting
Information. This would indicate that scrambling of the
ligands has occurred during isomerization.
1
after 24 h. We observed a symmetrical H NMR spectrum
(Scheme 4). Two isomers are possible for 9: either a dimeric
structure comprising octahedral hexacoordinate iridium
centers in a C2-trans-N configuration or a monomeric
structure with a five-coordinate iridium center with both
nitrogens trans to each other in the case of an ideal tbp
structure (tbp-trans-N; see Scheme 4). In both structures, a
symmetrical 1H NMR spectrum is expected (11 aromatic H
resonances, 8 ppy, and 3 phenolate) which indeed has been
observed for 9. The alternative dimeric asymmetrical C1
complex would show a nonsymmetrical 1H NMR spectrum,
which is obviously not the case. Only the dimeric C2-cis
complex would also show a symmetrical 1H NMR spectrum.
However, we believe it unlikely that the C2-trans-N complex
3 would yield a C2-cis complex without any signs of a C1
complex. Moreover, previous reports of the synthesis of the
analogous hydroxyl and solvento- complexes stated that the
complex was a dimeric C2-trans-N isomer.23a Complex 10
was synthesized in a similar manner; however, it showed
A number of conditions of the thermodynamic mer-to-
fac isomerization were investigated:
(a) To investigate intermolecular interactions, isomerizations
under thermal conditions were carried out, while varying the
dilution of mer-7-homo-N-trans in glycerol. At every dilu-
tion, mer-to-fac isomerization was observed with ligand
scrambling. By 1H NMR/MALDI-TOF, there was no
(29) Springer, C. S. J. Am. Chem. Soc. 1973, 95, 1459–1467.
(30) Dedeian, K.; Shi, J.; Shepherd, N.; Forsythe, E.; Morton, D. C. Inorg.
Chem. 2005, 44, 4445–4447.
Inorganic Chemistry, Vol. 47, No. 15, 2008 6687

McDonald et al.
observable difference in the intensity of the various peaks
corresponding to compounds 5-8 when using different
dilutions. Therefore, the process of ligand scrambling is not
concentration-dependent.
(b) Upon gradual heating of a solution of mer-7-homo-N-
trans in glycerol, from 150 to 200 °C, no mer-to-fac
isomerization occurred below 191 °C. Furthermore, ligand
scrambling products 5, 6, and 8 were not observed below
191 °C.
(c) mer-to-fac isomerization was a faster process in the
absence of a base. When a base (K2CO3) was present, the
process was an order of magnitude slower. In both cases,
ligand scrambling was observed.
(d) From experiments involving the heating of fac-7 for 24 h
in glycerol at either 150 °C or at 200 °C (with and without
K2CO3 present), pure fac-7 was recovered quantitatively. Neither
mer products nor ligand scrambling products were observed in
this set of experiments.
(e) A range of solvents was then tested. In refluxing decane
(174 °C), mer-7-homo-N-trans did not dissolve or isomerize
to fac product(s), nor was ligand scrambling observed.
Moreover, with mer-7-homo-N-trans dissolved in refluxing
1,2-dichlorobenzene (180 °C) or benzonitrile (191 °C),
respectively, no isomerization or ligand scrambling was
observed, and the starting material was recovered (likewise,
when mer-5 is used in these experiments, no isomerization
is observed). However, when glycerol (in excess) was added
to the refluxing decane solution, isomerization with scram-
bling was observed. In neat 1-decanol at 200 °C, mer-7-
homo-N-trans, was converted to the fac-5, 6, 7, and 8 mixture
while decomposition was also observed. In refluxing phenol
(184 °C), complete conversion to fac isomers, with ligand
scrambling, was observed. Subsequently, it was found that,
in noncrystalline (50 °C) phenol, mer-7-homo-N-trans was
converted to the fac-5, 6, 7, and 8 mixture, however, over a
substantially longer time period (compared to the reaction
done in refluxing phenol). This is a very important point
because it was often believed that a temperature above 150
°C was necessary to facilitate isomerization. In fact, a
refluxing solution of 1,2-dichlorobenzene with mer-5 and 10
mol% phenol (compared to mer-5) showed 40% mer-to-fac
conversion in 24 h. Water, when added to a glycerolic or
phenolic solution of mer-7-homo-N-trans and heated to
reflux, had no effect on the rate of isomerization or on the
extent of scrambling.
was the easiest to monitor using 1H NMR.32 An NMR tube
at 70 °C was loaded with mer-6 and d6-phenol, and spectra
were taken at certain time increments. Figure 7 represents
the disappearance of the mer-6 complex versus formation
of non-mer-6 compounds.33
The conversion of mer-6 (all CH3’s nonequivalent) did
not show a simple disappearance of its three tolyl peaks and
a concomitant appearance of one tolyl peak for fac-6 (all
CH3’s equivalent). In fact, several new and overlapping peaks
1
were observed in the tolyl CH3 region of the H NMR
spectrum during the course of the reaction (see Supporting
Information, Figure S5). Two of the new peaks can be
assigned as free Htolpy (2.34 ppm) and fac-6 (2.26 ppm).
Two other nonequivalent peaks at 2.10 and 2.08 ppm (upfield
of all other peaks) are unidentifiable. The peak at 2.10 ppm
appears rapidly upon the introduction of mer-6 to the d6-
phenol solution. It then increases in intensity at a similar
rate to the consumption of mer-6 aromatic signals. After
approximately 40 min, the slope of the plot changes,
suggesting a change in the rate-determining process. When
the reaction is complete, thus when all mer-6 has been
consumed, there are three signals observable in the d6-phenol
1
solution H NMR spectrum. These belong to fac-6 (2.26
ppm) and a small quantity of free Htolpy (2.34 ppm). We
have been unable to identify the third peak at 2.08 ppm;
however, we believe it may be unreacted intermediate(s).
The presence of Htolpy suggests the loss of a ligand in the
isomerization process and the formation of an [Ir(C,N)2X]
intermediate possibly represented by the peaks at 2.08 or
2.10 ppm. The conversion of mer-6 is rapid until ap-
proximately 66% of it has been consumed (∼40 min). After
that, there is a slower disappearance of mer-6 until it is
completely consumed. This would suggest that there is an
equilibrium between mer-6 and an intermediate complex
which is not fac-6. fac-6 is then formed from the intermediate
and does not convert back. In fact, no report of a back
conversion, of any similar fac complexes, has ever been
made. We believe the change in slope of the plot is when
mer-6 and the intermediate(s) are in equilibrium. Extra heat
allows the reaction rate to increase substantially.
Because ligand scrambling is observed in heteroleptic
systems, we assume an iridium-carbon bond breakage. We
(31) Crude tests were done on complexes 5-8 to test their response to the
MALDI-TOF apparatus. All four complexes had very similar response
factors. Figure S1 (Supporting Information) shows a distribution pattern
of the thermal reaction of mer-7-homo-N-trans to a fac-5, 6, 7, and 8
mixture (all post-thermal isomerization spectra showed the same
distribution pattern). Complex 7 was the most abundant, while only
minute amounts of 5 were observed (max. 5%).
From the observations made with these solvent tests, we
can deduce that, under thermal conditions, the isomerization
of mer compounds to fac compounds is an alcohol-catalyzed
reaction.
It was noted that the amounts of individual fac isomers in
the fac-5, 6, 7, and 8 mixture did not vary considerably in
all experiments that were performed. This indicates that
conditions did not affect the extent of ligand scrambling and
scrambling was a statistical phenomenon.31
(32) We have attempted a wide range of experiments to elucidate the
reaction rate and carry out some kinetic tests on the thermodynamic
conversion of mer-7 to fac-5, 6, 7, and 8. 1H or 13C NMR spectroscopy
was not possible due to overlapping of the marker signals (tolyl CH3)
of fac-6, 7, and 8 and mer-7. UV-vis is not feasible here because all
products have almost exactly the same absorption spectra. When it
was attempted, no isosbestic point was observed, but this is due to
slight differences in the molar absorptivities of the complexes. The
products are not separable by GC or HPLC.
(33) The disappearance of mer-6 was measured by integrating an aromatic
peak belonging to mer-6 and comparing it to the integration of all
tolyl signals. This relies on the assumption that all mer-6 is eventually
converted to fac-6.
For real-time analysis of the thermal isomerization process
of homoleptic complexes, complex 6 was used because it
6688 Inorganic Chemistry, Vol. 47, No. 15, 2008

Isomerization of Tris-Cyclometallated Ir(III) Complexes
Scheme 6. Intermediate Formation from mer-6 during Thermal Isomerisation in d6-Phenol
propose that exchange of a N-donor pyridine ligand by
alkoxide can occur by either dissociative or associative
pathways (Scheme 6). We also propose that all Ir-C bonds
remain intact during this exchange. Ligand substitution can
now occur in one of two ways: (1) nucleophilic attack on
the Ir(III) center, of the hydroxyl, resulting in a loss of
Htolpy, and formation of an iridium(III) alkoxide species
(11) and (2) oxidative addition of the coordinated alcohol,
yielding an Ir(V) species. This species could then reductively
eliminate Htolpy and result in an iridium(III) alkoxide species
(11). The formation of an intermediate is substantiated by
the fact that an equilibrium between mer-6 and another
complex(es) was observed. The observance of a free ligand
further supports this theory. The subsequent reaction of the
intermediate 11 with released Htolpy can either yield tris-
cyclometallated mer-type complexes or yield thermodynami-
cally stable fac-type complexes. We therefore observe a
change in the reaction kinetic profile once the equilibrium
has been reached.
If we closely examine how mer-6 loses Htolpy, we can
clarify the kinetic measurements further. mer-6, which has
no geometrical isomers, can form different alkoxide inter-
mediates, 11a-c. Either a trans-N, trans-C, or cis isomer
can form (Figure 4, depicted as tbp for clarity, most likely
octahedral alkoxide bridged dimers). This observation ex-
plains why we see more than one newly formed peak in the
tolyl CH3 region of the 1H NMR kinetic measurements. There
are three possible isomers, which would all give contrasting
1H NMR signals. We cannot explain why excess Htolpy and
an unidentified compound (peak at 2.08 ppm) are present at
the end of the isomerization reaction.
intermediate complexes 11a-c, when reacting with the
formed Htolpy, are believed to behave the same as the
monodentate chloride complex 3 in the synthesis of mer-7-
homo-N-trans. That is, the remaining cyclometallated Ir-
bound ligands of 11a-c do not lose their configuration
relative to each other (the trans-N, trans-C, or cis configu-
ration is held) in the subsequent reaction. It must be noted
that a trigonal bipyramidal (tbp) depiction of compounds 11
is for illustrative purposes only and certainly is not the true
configuration. However, in terms of how the reaction
proceeds, the tbp depiction is the best for illustrative purposes
only. When 3 is reacted with Htolpy, the ppy ligands bound
to the iridium neVer lose their configuration relative to each
other. That is believed to be a result of a high energetic
barrier to Berry pseudorotation. Compared to rhodium(III),
hexa- and pentacoordinate iridium(III) complexes have an
increased effective nuclear charge, and thus metal to ligand
σ interactions are strengthened, and thus bond breakage and
subsequent rotation is unlikely.34 In this case, we believe,
as a result of experimental evidence (Supporting Information
and section 5), that Berry pseudorotation does not occur in
these complexes.
In the reaction of 11a-c with released Htolpy, only 11a
can give fac-6. Intermediates 11b and 11c will always revert
to mer-6. The reaction of 11a with Htolpy would yield either
mer-6 or fac-6. If the incoming Htolpy reacts with N trans
to N (of metal-bound ligand), then mer-6 will form. If Htolpy
reacts with N trans to C (of metal-bound ligand), then fac-6
will form. The reaction of 11b with Htolpy will yield either
Λ or ∆ enantiomers/helimers of mer-6, nothing else.
Likewise the reaction of 11c with Htolpy will yield either ∆
or Λ enantiomers/helimers of mer-6, nothing else. It must
be noted that phenoxide and methoxide are monodentate
ligands.35
In the following discussion, the monodentate alkoxide
From the observations we have made to this point, the
photochemical mer-to-fac isomerization is completely un-
related to the thermal isomerization. In the photochemical
isomerization, no ligand dissociation is believed to occur,
(34) Green, M.; Parker, G. J. J. Chem. Soc., Dalton Trans. 1974, 333–
343.
(35) All solvents or activating ligands previously used in the synthesis of
fac-[Ir(ppy)3] have been bidentate: glycerol, 2-ethoxyethanol, and acac.
We believe that a monodentate intermediate state must exist at some
stage during the reaction of intermediates of type 11. We believe the
bidentate solvents form intermediate complexes in a similar binding
manner to acac. This leads to another variable as to where the attacking
arylpyridine will bind in intermediate 11, trans to a carbon or trans to
a nitrogen. This could lead to preferences towards certain fac- or mer-
products over others.
Figure 4. Conversion of mer-6 over time at 70 °C in d6-phenol.
Inorganic Chemistry, Vol. 47, No. 15, 2008 6689

McDonald et al.
Figure 5. Ligand substitution on mer-6 yielding alkoxide intermediates 11a-c and their subsequent reaction with released Htolpy.
When the reaction was carried out photochemically, no
scrambling of the ligands was observed, as mentioned in the
synthesis of pure fac-7 and -8. If we now treat heteroleptic
complexes mer-7- and -8-homo-N-trans as we have mer-6
in the previous section, and focus on ligand substitution
reactions and subsequent reaction of the formed alkoxide
intermediates with a released ligand, we get a clear explana-
tion for ligand scrambling. In both homo- and heteroleptic
complexes, ligand substitution can be approached in two
ways: (1) there is a higher probability for one of the ligands
over the other two ligands to be substituted or (2) any of the
three ligands can be substituted and further react with the
formed intermediates. In the previous section, we saw that
only the formation of only one of the intermediates (the cis
isomer 11a) resulted in the formation of the final product
fac-6.
Figure 6. Substitution of ligand of mer-7-homo-N-trans and intermediates.
In mer-7-homo-N-trans, each ligand must be treated
separately, because they are all bound differently to the
iridium center (Figure 5).
and the isomerization is mediated through an excited state.36
It must, once again, be noted that all experiments in the
aboVe section were carried out in the absence of light.
7. Mechanism of Thermal mer-to-fac Isomerization
of Heteroleptic [Ir(C,N)2(C′,N′)] Complexes. When con-
verting mer-7-homo-N-trans to fac-7 under thermal condi-
tions, it was noted that scrambling of the ligands occurred.
Figure 5 depicts the isomers (12a-c) that can form as a
result of substitution of the ligands only in mer-7-homo-N-
trans. If ligand a is substituted, an intermediate, cis-
configured complex, will be formed. If only ligand a was
substituted, then the only products possible, as a result of
reaction of 12a with only a (Hppy), would be mer-7-homo-
N-trans and fac-7. If b dissociates, a trans-C intermediate
is formed. If only ligand b was substituted and 12b had to
(36) (a) See ref 10. (b) Karatsu, T.; Ito, E.; Yagai, S.; Kitamura, A. Chem.
Phys. Lett. 2006, 424, 353–357.
6690 Inorganic Chemistry, Vol. 47, No. 15, 2008

Isomerization of Tris-Cyclometallated Ir(III) Complexes
Under thermal reaction conditions, the conversion of the
mer-homo-N-trans isomers to the fac complexes led to ligand
scrambling. Phenol was found to be an ideal solvent to
facilitate thermal mer to fac isomerization. It was found that,
under conditions where isomerization does not occur, the
addition of 10 mol% of phenol facilitated mer-to-fac isomer-
ization. Bis-cyclometallated iridium(III) alkoxides are pro-
posed intermediates in the mer-to-fac isomerization of
[Ir(ppy)3]-type systems. Analogous trans-N configured bis-
cyclometallated iridium(III) alkoxide complexes were syn-
thesized and characterized. Furthermore, the reaction of these
complexes with Hppy-yielding complex 5 shows that Ir(III)
alkoxide species are viable intermediates.
A proposed mechanism of the thermodynamic isomeriza-
tion reaction was presented, which gives an explanation as
to why ligand scrambling is observed. Kinetic examination
of the mer-to-fac isomerization of homoleptic trisphenylpy-
ridine iridium(III) complexes supports the proposed mech-
anism. The kinetic analysis supported the theory that an
iridium(III) alkoxide intermediate is formed. A detailed
discussion on possible geometrical isomers of intermediates
in both the homo- and heteroleptic systems is presented.
This report is essential to the field of OLED devices for
three reasons. First, it demonstrates the synthesis of hetero-
leptic [Ir(C,N)2(C′,N′)] complexes, which are desired for the
covalent tethering of these organometallics to polymeric
supports. Second, the synthesis of heteroleptic
[Ir(C,N)2(C′,N′)] complexes opens up the field of electronic
fine-tuning of ligands for mixed ligand species. To this point,
all complexes developed were tris-cyclometallated homo-
leptic species, and thus relatively subtle changes in electron
densities around the iridium(III) center were not possible.
With heteroleptic [Ir(C,N)2(C′,N′)] systems, subtle electronic
changes can be introduced to provide, for example, emission
color changes on demand, while retaining the high quantum
yield values these complexes are renowned for. Third, and
most importantly, it is the first thorough investigation of mer
and fac heteroleptic octahedral organometallic species. It
gives essential insights into how fac-tris-cyclometallated
iridium(III) species can be synthesized, and the energetic
barriers to their synthesis. Furthermore, it gives an explana-
tion of the properties of solvents/materials required for the
synthesis of cyclometallated iridium(III) species, and also
suggestions for synthesizing these complexes at lower
temperatures.
Figure 7. Products from the reaction of Hppy and Htolpy with intermediates
12a-c.
rereact with only b (Hppy), mer-7-homo-N-cis or -trans
would be formed. If the tolpy (heteroligand) were substituted
by an alkoxide, a trans-N isomer would be formed (9).
Hypothesizing that only the substituted Htolpy ligand could
rereact with 12c would lead to only different enantiomers/
helimers of mer-7-homo-N-trans.
Because ligand scrambling was observed, this suggests that
not one single ligand is substituted, and that all ligands can
be substituted and react further. It could be that there is a
preference for one ligand to be substituted over another;
however, we cannot prove this experimentally.
We can now explain why we see fac complexes 5-8 after
the reaction of mer-7- and mer-8-homo-N-trans under
thermal isomerization conditions. Figure 6 only shows half
of the possible complexes that can form. The newly formed
mer-8-homo-N-trans, mer-7-homo-cis, mer-8-homo-cis, and
mer-5 can all react further to give a range of mer and fac
isomers of 5-8. Once a fac has formed, it will not rereact.
Conclusions
A number of heteroleptic tris-cyclometallated iridium(III)
[Ir(C,N)2(C′,N′)] complexes have been synthesized and fully
characterized using a range of techniques. The mer-homo-
N-trans isomer of heteroleptic complexes 7 and 8 was
synthesized and converted to the fac isomer using photo-
chemical techniques. The synthesized mer- and fac-7 and
-8 have been fully characterized. The synthesis of mer
heteroleptic isomers followed by conversion to their fac
isomer demonstrates a high-yielding, simple protocol for the
synthesis of tethered tris-cyclometallated iridum(III) com-
plexes. This is essential for materials science, which requires
covalent linkage of lumiphore materials (iridium organome-
tallics) to hole transport and electron transport layers in
LEDs.
Acknowledgment. Ciba (Basel, CH) is thanked for the
provision of [IrCl3 · 3H2O] and for financial assistance. Dr.
Paul van der Schaaf and Dr. Roger Pretot are acknowledged
for their helpful suggestions.
Supporting Information Available: Crystallographic informa-
tion files (CIF) for mer-7- and -8-homo-N-trans and fac-8, UV-vis
1
absorption and emission data for all synthesized compounds, H
NMR spectral data for the kinetic measurements in Figure 7. This
material is available free of charge via the Internet at http://
pubs.acs.org.
IC800169N
Inorganic Chemistry, Vol. 47, No. 15, 2008 6691