Luminescent platinum complexes with terdentate ligands forming 6-membered chelate rings: advantageous and deleterious effects in N∧N∧N and N∧C∧N-coordinated complexes

Article abstract of DOI:10.1021/ic9016323

Platinum(II) complexes of the form [PtLⁿCl]⁺ are reported, containing the N∧N∧N-coordinating ligands 2,6-di-(8-quinolyl) pyridlne (L¹), 2,6-di(8-quinolyl)-4-methoxypyridine (L²), or 2,6-di(7-aza-indolyl)-pyridine (L³). Metathesis of the chloride co-ligand in [PtL¹ Cl]⁺ can be accomplished under mild conditions, as exemplified by the formation of the complexes [PtL ¹OMe]⁺ and [PtL¹(C≡C-tfp)]⁺, in which L¹ remains bound as a terdentate ligand {HC≡C-tfp = 3,5-bis(trifluoromethyl)-phenylacetylene}. An N∧C∧N-coordinated, cyclometalated analogue of [PtL¹Cl]⁺ has also been prepared, namely, PtL⁴Cl where HL⁴ is 1,3-di(8-quinolyl) benzene. The common feature among the six new complexes described here Is that they contain 6-membered chelate rings, rather than the usual 5-membered rings that form when more common N∧N∧N ligands, such as 2,2′:6′, 2″-terpyrldine (tpy), bind to Pt(II). All the quinolyl-based complexes are phosphorescent in solution at room temperature, with quantum yields up to 4%. This contrasts with the well-established lack of emission from [Pt(tpy)Cl] ⁺ under these conditions. Density functional theory calculations suggest that the improvement may stem, at least in part, from the relief of ring strain associated with the larger chelate ring size, leading to a more optimal bite angle at the metal, close to 180° and hence to a stronger ligand field. Consideration of the luminescence parameters, including data at 77 K, together with absorption and electrochemical data and the results of TD-DFT calculations, suggests that the lowest-lying singlet states have metal-to-ligand charge-transfer (MLCT) character, but that the triplet state from which emission occurs has more predominant ligand-centered character. The azaindolyl complex [PtL³Cl]⁺ is not emissive at room temperature, apparently owing to a particularly small radiative rate constant. The cyclometalated complex PtL⁴Cl emits at lower energy than [PtL¹Cl] ⁺ but, in this case, the luminescence quantum yield is inferior to related complexes with 5-membered chelate rings.

Full text of DOI:10.1021/ic9016323

476 Inorg. Chem. 2010, 49, 476–487
DOI: 10.1021/ic9016323
Luminescent Platinum Complexes with Terdentate Ligands Forming
6-Membered Chelate Rings: Advantageous and Deleterious Effects in N∧N∧N and
N∧C∧N-Coordinated Complexes
Katherine L. Garner, Louise F. Parkes, Jason D. Piper, and J. A. Gareth Williams*
Department of Chemistry, University of Durham, Durham DH1 3LE, U.K.
Received August 14, 2009
Platinum(II) complexes of the form [PtLnCl]þ are reported, containing the N∧N∧N-coordinating ligands 2,6-di-
(8-quinolyl)pyridine (L1), 2,6-di(8-quinolyl)-4-methoxypyridine (L2), or 2,6-di(7-aza-indolyl)-pyridine (L3). Metathesis
of the chloride co-ligand in [PtL1Cl]þ can be accomplished under mild conditions, as exemplified by the formation of the
complexes [PtL1OMe]þ and [PtL1(CtC-tfp)]þ, in which L1 remains bound as a terdentate ligand {HCtC-tfp =
3,5-bis(trifluoromethyl)-phenylacetylene}. An N∧C∧N-coordinated, cyclometalated analogue of [PtL1Cl]þ has also
been prepared, namely, PtL4Cl where HL4 is 1,3-di(8-quinolyl)benzene. The common feature among the six new
complexes described here is∧tha∧t they contain 6-membered chelate rings, rather than the usual 5-membered rings that
form when more common N N N ligands, such as 2,20:60,200-terpyridine (tpy), bind to Pt(II). All the quinolyl-based
complexes are phosphorescent in solution at room temperature, with quantum yields up to 4%. This contrasts with the
well-established lack of emission from [Pt(tpy)Cl]þ under these conditions. Density functional theory calculations
suggest that the improvement may stem, at least in part, from the relief of ring strain associated with the larger chelate
ring size, leading to a more optimal bite angle at the metal, close to 180°, and hence to a stronger ligand field.
Consideration of the luminescence parameters, including data at 77 K, together with absorption and electrochemical
data and the results of TD-DFT calculations, suggests that the lowest-lying singlet states have metal-to-ligand charge-
transfer (MLCT) character, but that the triplet state from which emission occurs has more predominant ligand-centered
character. The azaindolyl complex [PtL3Cl]þ is not emissive at room temperature, apparently owing to a particularly
small radiative rate constant. The cyclometalated complex PtL4Cl emits at lower energy than [PtL1Cl]þ but, in this
case, the luminescence quantum yield is inferior to related complexes with 5-membered chelate rings.
Introduction
and proteins,4 and bioimaging agents.5 At the more funda-
mental level too, studies on emissive platinum complexes
have provided insight into key excited state processes such as
intersystem crossing6 and spin-orbit coupling pathways.7
A clear picture has emerged over the past 20 years of the
molecular features that are likely to favor triplet emission
over non-radiative decay in square-planar Pt(II) complexes.
Research into the design and preparation of photolumi-
nescent platinum(II) complexes has been driven by a number
of applications such as triplet-harvesting phosphors for light-
emitting devices,1 chemosensing,2 chromophores for promot-
ing photoinducedchargeseparation for artificial photosynth-
esis,3 probes for biological molecules such as nucleic acids
*To whom correspondence should be addressed. E-mail: j.a.g.williams@
durham.ac.uk.
(4) (a) Eryazici, I.; Moorefield, C. N.; Newkome, G. R. Chem. Rev. 2008,
108, 1834. (b) Wong, K.M.-C.; Tang, W. S.; Chu, B. W. K.; Zhu, N. Y.; Yam,
V.W.-W. Organometallics 2004, 23, 3459. (c) Ma, D.-L.; Shum, T.Y.-T.; Zhang,
F.; Che, C.-M.; Yang, M. Chem. Commun. 2005, 4675. (d) Wu, P.; Wong,
E.L.-M.; Ma, D.-L.; Tong, G.S.-M.; Ng, K.-M.; Che, C.-M. Chem.;Eur. J. 2009,
15, 3652. (e) Ma, D. L.; Che, C.-M.; Yan, S.-C. J. Am. Chem. Soc. 2009, 131, 1835.
(5) (a) Botchway, S. W.; Charnley, M.; Haycock, J. W.; Parker, A. W.;
Rochester, D. L.; Weinstein, J. A.; Williams, J. A. G. Proc. Natl. Acad. Sci.
U.S.A. 2008, 105, 16071. (b) Koo, C.-K.; Wong, K.-L.; Man, C.W.-Y.; Lam,
Y.-W.; So, K.-Y.; Tam, H.-L.; Tsao, S.-W.; Cheah, K.-W.; Lau, K.-C.; Yang, Y.-Y.;
Chen, J.-C.; Lam, M.H.-W. Inorg. Chem. 2009, 48, 872.
(6) Danilov, E. O.; Pomestchenko, I. E.; Kinayyigit, S.; Gentili, P. L.;
Hissler, M.; Ziessel, R.; Castellano, F. N. J. Phys. Chem. A 2005, 2465, 109.
(7) (a) Rausch, A. F.; Thompson, M. E.; Yersin, H. Chem. Phys. Lett.
2009, 468, 46. (b) Rausch A. F.; Homeier, H. H. H.; Yersin, H. Top. Organomet.
Chem., 2010, 29, 193.
(1) (a) Xiang, H.-F.; Lai, S.-W.; Lai, P. T.; Che, C.-M. Phosphorescent
platinum(II) materials for OLED applications. In Highly efficient OLEDs
with phosphorescent materials; Yersin, H., Ed.; Wiley-VCH: Weinheim, 2008.
(b) Williams, J. A. G.; Develay, S.; Rochester, D. L.; Murphy, L. Coord. Chem.
Rev. 2008, 252, 2596.
(2) (a) Fletcher, N.; Lagunas, M. C. Top. Organomet. Chem. 2010, 28, 143.
(b) Tang, W.-S.; Lu, X.-X.; Wong, K.M.-C.; Yam, V.W.-W. J. Mater. Chem.
2005, 15, 2714. (c) Siu, P. K. M.; Lai, S.-W.; Lu, W.; Zhu, N.; Che, C.-M. Eur. J.
Inorg. Chem. 2003, 2749. (d) Yang, Q.-Z.; Tong, Q.-X.; Wu, L.-Z.; Zhang, L.-P.;
€
Tung, C.-H. Eur. J. Inorg. Chem. 2004, 1948. (e) Lanoe, P.-H.; Fillaut, J.-L.;
Toupet, L.; Williams, J. A. G.; Le Bozec, H.; Guerchais, V. Chem. Commun.
2008, 4333.
(3) Chakraborty, S.; Wadas, T. J.; Hester, H.; Schmehl, R.; Eisenberg, R.
Inorg. Chem. 2005, 44, 6865.
r
pubs.acs.org/IC
Published on Web 12/11/2009
2009 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 2, 2010 477
The key is normally to make use of ligand-centered (LC) or
charge-transfer transitions, while ensuring that d-d excited
states, which involve the population of the strongly anti-
bonding dx2-y2 orbital, are displaced to higher energies not
accessible at room temperature.8 It is probably the thermally
activated population of such distorted states that accounts
for the lack of room temperature emission in simple com-
plexes like [Pt(tpy)Cl]þ, despite the rigidity associated with its
terdentate coordination.9 To raise the energy of d-d states,
ligands which exert a stronger ligand-field are required, while
retaining rigid architectures. Successful strategies include
the replacement of the chloride ligand in [Pt(tpy)Cl]þ by a
stronger-field acetylide (-CtC-Ar)10 or cyanide (-CtN)11
ligand, or replacement of the tpy ligand by a cyclometallating
analogue such as phbpy or dpyb, both of which approaches
result in room temperature emission12,13 {phbpyH = 6-phenyl-
2,20-bipyridine; dpybH = 1,3-di(2-pyridyl)benzene}.
The reason why tpy itself does not offer sufficient ligand
field strength can be traced to the poor bite angle when it
binds to large transition metal ions. Thus, the optimal
N-Pt-N angle for maximum orbital overlap would be
180°, but the structural constraints of tpy lead to a value of
only 163.5(7)° in [Pt(tpy)Cl]ClO4.14 A similar issue accounts
for the well-known difference between [Ru(tpy)2]2þ and
[Ru(bpy)3]2þ, the former being essentially non-emissive at
room temperature owing to a weaker ligand field associated
with sublinear N-Ru-N angles of 158.3(3)-159.1(2)°.15,16
A potential alternative route to emissive complexes
might be to use N∧N∧N-coordinating ligands which offer a
larger bite angle, perhaps involving 6-membered chelation.
McMillin and Thummel demonstrated that 2-(8-quinolyl)-
1,10-phenanthroline leads to a platinum complex that is
significantly emissive in solution at room temperature
(Φlum = 0.002, τ = 310 ns in CH2Cl2).17 This ligand sets
up one 5-membered and one 6-membered chelate ring when it
binds to Pt(II): the ring strain is hence alleviated and
€
the ligand field increased. Thummel and Hammarstrom
obtained room temperature emissive ruthenium(II) com-
plexes with the same ligand, where the improvement over
[Ru(tpy)3]2þ was interpreted similarly.18 Hammarstrom and
€
Johannson found that further enhancement of the emission
efficiency of [Ru(N∧N∧N)2]2þ complexes could be achieved
with the ligand 2,6-di(8-quinolyl)pyridine, which leads to two
6-membered chelates per ligand.19
In this contribution, we report on the contrasting influence
of 6-membered chelation in N∧N∧N and N∧C∧N-coordi-
nated platinum(II) complexes based on 2,6-di(8-quinolyl)-
pyridine and1,3-di(8-quinolyl)benzene. Wedemonstrate that
simple N∧N∧N complexes are accessible with emission
quantum yields comparable to [Ru(bpy)3]2þ, whereas a
detrimental effect is observed on the luminescence of the
N∧C∧N systems compared to analogues with cyclometa-
lating 5-membered chelate rings. Using the example of an
azaindole-based ligand, we also show that favorable geome-
try alone is not sufficient to ensure good emission properties
in N∧N∧N coordinated systems: electronic factors are ob-
viously also crucial.
Results and Discussion
1. Synthesis of Ligands. The complexes studied are
shown in Figure 1, and the strategies employed in the
synthesis of the requisite ligands are summarized in
Scheme 1. The N∧N∧N-coordinating ligand 2,6-di-
(8-quinolyl)pyridine L1 was prepared by a Suzuki cross-
coupling reaction between 2,6-dibromopyridine and qui-
noline-8-boronic acid, using the method described by
Johannson et al.20 We found that 2,6-dichloropyridine
could also be used, giving a comparable yield of the
product. The 4-methoxy derivative L2 was prepared in a
similar manner from 2,6-dibromo-4-methoxypyridine.21
The preparation of the new N∧C∧N proligand HL4 was
also carried out by means of a Suzuki cross-coupling
reaction, but in this case the functionality in the two units
was reversed: 1,3-benzene diboronic acid was coupled
with 2 equiv of 8-bromoquinoline.
(8) (a) Houlding, V. H.; Miskowski, V. M. Coord. Chem. Rev. 1991, 111,
145. (b) Barigelletti, F.; Sandrini, D.; Maestri, M.; Balzani, V.; von Zelewsky, A.;
Chassot, L.; Jolliet, P.; Maeder, U. Inorg. Chem. 1988, 27, 3644. (c) Williams,
J. A. G. Top. Curr. Chem. 2007, 281, 205.
(9) (a) Crites, D. K.; Cunningham, C. T.; McMillin, D. R. Inorg. Chim.
Acta 1998, 273, 346. (b) McMillin, D. R.; Moore, J. J. Coord. Chem. Rev. 2002,
229, 113.
(10) (a) Yam, V.W.-W.; Tang, R.P.-L.; Wong, K.M.-C.; Cheung, K.-K.
Organometallics 2001, 20, 4476. (b) Yang, Q.-Z.; Wu, L.-Z.; Wu, Z.-X.; Zhang,
L.-P.; Tung, C.-H. Inorg. Chem. 2002, 41, 5653. (c) Guo, F.; Sun, W.; Liu, Y.;
Schanze, K. Inorg. Chem. 2005, 44, 4055. (d) Chakraborty, S.; Wadas, T. J.;
Hester, H.; Flaschenreim, C.; Schmehl, R.; Eisenberg, R. Inorg. Chem. 2005, 44,
6284. (e) Castellano, F. N.; Pomestchenko, I. E.; Shikhova, E.; Hua, F.; Muro,
M. L.; Rajapakse, N. Coord. Chem. Rev. 2006, 250, 1819. (f) Ziessel, R.; Diring,
S.; Retailleau, P. Dalton Trans. 2006, 3285.
As an alternative form of N∧N∧N-chelating ligand, we
also investigated L3, which incorporates 7-azaindole units
in place of the 8-substituted quinolines. The combination
of copper(I) iodide with trans-1,2-diaminocyclohexane,
recently established by Buchwald and co-workers as a
particularly effective aryl amination catalyst,22 allowed
(11) Wilson, M. H.; Ledwaba, L. P.; Field, J. S.; McMillin, D. R. Dalton
Trans. 2005, 2754.
(12) (a) Cheung, T.-C.; Cheung, K.-K.; Peng, S.-M.; Che, C.-M. J. Chem.
Soc., Dalton Trans. 1996, 1645. (b) Neve, F.; Crispini, A.; Campagna, S. Inorg.
Chem. 1997, 36, 6150. (c) Lai, S.-W.; Chan, M.C.-W.; Cheung, T.-C.; Peng,
S.-M.; Che, C.-M. Inorg. Chem. 1999, 38, 4046. (d) Yip, J. H. K.; Suwarno;
Vittal, J. J. Inorg. Chem. 2000, 39, 3537. (e) Shao, P.; Li, Y.; Azenkeng, A.;
Hoffmann, M. R.; Sun, W. Inorg. Chem. 2009, 48, 2407.
(17) Hu, Y.-Z.; Wilson, M. H.; Zong, R.; Bonnefous, C.; McMillin, D. R.;
(13) (a) Williams, J. A. G.; Beeby, A.; Davies, E. S.; Weinstein, J. A.;
Wilson, C. Inorg. Chem. 2003, 42, 8609. (b) Farley, S. J.; Rochester, D. L.;
Thompson, A. L.; Howard, J. A. K.; Williams, J. A. G. Inorg. Chem. 2005, 44,
Thummel, R. P. Dalton Trans. 2005, 354.
€
(18) Abrahamsson, M.; Becker, H.-C.; Hammarstrom, L.; Bonnefous, C.;
Chamchoumis, C.; Thummel, R. P. Inorg. Chem. 2007, 46, 10354.
€
€
(19) (a) Abrahamsson, M.; Jager, M.; Osterman, T.; Eriksson, L.;
ꢀ ꢁ
9690. (c) Rochester, D. L.; Develay, S.; Zalis, S.; Williams, J. A. G. Dalton.
€
Persson, P.; Becker, H.-C.; Johansson, O.; Hammarstrom, L. J. Am. Chem.
Trans. 2009, 1728. (d) Cocchi, M.; Virgili, D.; Fattori, V.; Rochester, D. L.;
Williams, J. A. G. Adv. Funct. Mater. 2007, 17, 285.
€
€
Soc. 2006, 128, 12616. (b) Abrahamsson, M.; Jager, M.; Kumar, R. J.; Osterman,
(14) Bailey, J. A.; Hill, M. G.; Marsh, R. E.; Miskowski, V. M.; Schaefer,
W. P.; Gray, H. B. Inorg. Chem. 1995, 34, 4591.
(15) (a) Caspar, J. V.; Meyer, T. J. Inorg. Chem. 1983, 22, 2444. (b) Juris,
A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; von Zelewsky, A.
Coord. Chem. Rev. 1988, 84, 85. (c) Medlycott, E. A.; Hanan, G. S. Chem. Soc.
Rev. 2005, 34, 133.
€
T.; Persson, P.; Becker, H.-C.; Johansson, O.; Hammarstrom, L. J. Am. Chem.
Soc. 2008, 130, 15533.
€
(20) Jager, M.; Eriksson, L.; Bergquist, J.; Johansson, O. J. Org. Chem.
2007, 72, 10227.
(21) The method was loosely based on a previously reported procedure:
€
Neumann, U.; Vogtle, F. Chem. Ber. 1989, 122, 589.
(16) Lashgari, K.; Kritikos, M.; Norrestam, R.; Norrby, T. Acta Crystal-
logr. 1999, C55, 64.
(22) Antilla, J. C.; Baskin, J. M.; Barder, T. E.; Buchwald, S. L. J. Org.
Chem. 2004, 69, 5578.

478 Inorganic Chemistry, Vol. 49, No. 2, 2010
Garner et al.
Scheme 1. Strategies Employed in the Synthesis of Ligands L1-L3
and HL4a
a The numbering scheme shown is that used in the assignment of the
1H and 13C spectra. To aid comparison, we number the central ring of
HL4 in the same way as the pyridyl ring in the other ligands.
complexes with square planar geometry. As found for
complexation of terpyridine to Pt(II), the protons para to
the coordinating nitrogen atoms experience a particularly
significant deshielding effect (Δδ ∼ 0.5 ppm). However,
the position which experiences the largest perturbation
is H7 of the quinoline, ortho to the interannular bond
(Δδ = 0.68 ppm in d6-DMSO).
The azaindole ligand L3 undergoes broadly similar
behavior upon complexation, in that most signals also
0
shift to high frequency; however, in this case, position H3
within the central ring experiences a large shielding effect
of almost 1 ppm (Figure 2 and Table 1). One or both of
two possible effects may be involved here. First, the
azaindole moiety has σ-withdrawing character but is also
a potential π-donor through the indole nitrogen atom. In
the uncomplexed ligand, free rotation about the C-N
interannular bond may ensure that the former predomi-
nates, as conformations close to planarity, which would
favor the π-donation, will be sterically disfavored. Upon
complexation, the π-donation effect from azaindole to
pyridine should increase, as the conformation will be-
come locked. Second, the ligand in the complex is prob-
ably not p0lanar but partially twisted (see below). This will
lead to H3 being placed within the zone of influence of the
diamagnetic ring current of the azaindole. We note that
H6 might in that case be expected to be influenced
similarly, and indeed it is the only other signal to experi-
Figure 1. Structural formulas of the complexes investigated in this
study. The structures of [Pt(tpy)Cl]þ and Pt(dpyb)Cl are also shown for
reference.
coupling of 2,6-dibromopyridine with 2 equiv of 7-azain-
dole to be carried out under mild conditions to give L3 in
good yield (Scheme 1).
2. Synthesis and Characterization of Platinum(II)
Complexes. Complexation of the N∧N∧N-coordinating
ligands to platinum was readily achieved upon treatment
with Pt(DMSO)2Cl2 in refluxing methanol. This simple
platinum(II) complex23 has good solubility in alcohols
and does away with the need for water as a co-solvent,
which was required when K2PtCl4 was used as the plati-
num precursor. The chloride salts [PtL1-3Cl]Cl were
isolated in yields of around 50%, and converted to their
hexafluorophosphate salts, [PtL1-3Cl]PF6, upon anion
exchange with KPF6(aq).
The complexation of 2,6-di(quinolyl)pyridine, L1 to
platinum(II) results in shifts to higher frequency of all
the 1H NMR signals (Figure 2 and Table 1), as typically
observed for aromatic ligands when they form cationic
ence a shielding effect upon complexation, albeit a smaller
0
one than H3 .
The synthesis of the cyclometalated complex PtL4Cl
from ligand HL4 required a higher temperature to effect
the necessary C-H activation. In the 1H NMR spectrum,
most of the quinoline resonances are again shifted to
(23) Puddephatt, R. J. Inorg. Synth. 2002, 33, 189.

Article
Inorganic Chemistry, Vol. 49, No. 2, 2010 479
Figure 2. Left: 1H NMR spectra of L1 (top) and [PtL1Cl]Cl (bottom) in d6-DMSO at 500 MHz, 298 K. Right: Corresponding spectra of L3 (top) and
[PtL3Cl]Cl (bottom) under the same conditions. Protons are numbered according to Scheme 1. Thin arrows refer to diagnostic NOE couplings. The thick
0
arrow highlights the large upfield shift of the H3 resonance upon complexation.
Table 1. Changes in the 1H NMR Chemical Shifts of the Ligands upon
Complexation to Pt(II)
with ethynyl-3,5-bis(trifluoromethyl)-benzene,24 similar to
that used in the preparation of platinum(II) terpyridyl
acetylide complexes.25 The change from chloride to acety-
lide leads to a large deshielding of the H2 protons of L1 in the
1H NMR spectrum (Δδ = þ0.5 ppm). No other resonances
in L1 are significantly affected, so this effect is probably
through-space in origin, arising from the C-H2 bonds being
parallel and close to the CtC triple bond of the acetylide.
The identity of all complexes was further confirmed
by high-resolution mass spectrometry; a representative
example for [PtL2Cl]þ with isotope matching is provided
in Figure S1 of the Supporting Information.
3. Structures of the Complexes by Density Functional
Theory. Despite repeated attempts, including the use of
different counter-anions in the case of [PtL1-3Cl]þ, it has
not hitherto proved possible to obtain crystals of the
complexes of sufficient quality for an X-ray diffraction
study. Density functional theory (DFT) calculations have
recently proved highly successful in predicting the struc-
tures of metal complexes.26 When crystal structures have
been available for comparison, not only are qualitative
trends reproduced well, but there is frequently excellent
quantitative agreement in bond lengths and angles.27 In
the present case, we have employed DFT calculations,
using an approach now well-established for third row
transition metal complexes, to identify the energy-mini-
mized structures (B3LYP functional28 with double-ζ
basis sets, 6-31G for ligands and LANL2DZ for Pt(II),29
with the inner core electrons of Pt replaced by a relativistic
effective core potential).
[PtL1Cl]X
Hnn =a
X = Cl-b
X = PF6
[PtL3Cl]Clb
PtL4Cld
-c
2
3
4
5
6
7
30
40
þ 0.27
þ 0.20
þ 0.53
þ 0.52
þ 0.35
þ 0.68
þ 0.18
þ 0.50
þ 0.37
þ 0.26
þ 0.59
þ 0.52
þ 0.38
þ 0.54
þ 0.06
þ 0.57
þ 0.29
þ 0.39
þ 0.55
þ 0.66
- 0.08
þ 0.79
- 0.06
þ 0.20
0
þ 0.09
þ 0.54
- 0.20
- 0.27
- 0.93
þ 0.40
a Numbering system as in Scheme 1. b In d6-DMSO. c In d6-acetone.
d In CDCl3.
higher frequency upon complexation, particularly H2 (Δδ =
0.8 ppm in CDCl3), which also displays well-resolved
195Pt satellites (3J = 44 Hz). In this case, however, the
protons in the central benzene ring experience a shielding
effect (Table 1), in line with the increased electron density
in the cyclometallating ring that accompanies the Pt-C
bond formation. A similar trend has previously been
observed for complexes of 1,3-di(2-pyridyl)benzenes.13
The methoxide adduct [PtL1OMe]þ was obtained by
treatment of [PtL1Cl]PF6 with a solution of KOH in
methanol at room temperature, and is a distinctly more
vivid yellow than the pale starting material. The methoxide
protons resonate at 2.82 ppm (in CDCl3), strongly shielded
relative to the value for MeOH (3.49 ppm) (spectra are
shown in the Supporting Information). Further confirma-
tion of the identity of the product is provided by the
Nuclear Overhauser Effect (NOE) seen between this signal
and the H2 protons of the terdentate ligand.
(25) Hissler, M.; McGarrah, J. E.; Connick, W. B.; Geiger, D. K.;
Cummings, S. D.; Eisenberg, R. Coord. Chem. Rev. 2000, 208, 115.
(26) For a recent review of the application of DFT techniques to metal
The acetylide derivative [PtL1(CtC-tfp)]þ was pre-
pared by means of a copper-catalyzed metathesis reaction
ꢁ
ꢀ ꢁ
complexes and their excited states, see: Vlcek, A.; Zalis, S. Coord. Chem. Rev.
2007, 251, 258.
(27) Significantly, this is the case for [RuL1 2þ, where a structure was
]
2
available to verify the geometry predicted by DFT: see ref 19a.
(28) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C. T.; Yang,
W. T.; Parr, R. G. Phys. Rev. B. 1988, 37, 785.
(24) (a) Sonogashira, K.;Fujikura, Y.;Yatake, T.;Toyoshima, N.;Takahashi,
S.; Hagihara, N. J. Organomet. Chem. 1978, 145, 101. (b) Takahashi, S.; Kuroyama,
Y.; Sonogashira, K.; Hagihara, N. Synthesis 1980, 627.
(29) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.

480 Inorganic Chemistry, Vol. 49, No. 2, 2010
Garner et al.
Figure 3. Structures of complexes [PtL1Cl]þ, PtL4Cl, and [PtL3Cl]þ predicted by energy minimization using DFT (gas-phase at 0 K), with the viewing
angle selected to best reveal both the twist in the ligand and the planarity at the metal ion. The angle between the planes of the quinoline/azaindole rings and
the central pyridine/benzene ring is given, together with the N-Pt-N angle for trans-related N atoms. Corresponding angles for [PtL1OMe]þ are 31.6° and
174.9°, and for [PtL1(CtC-tfp)]þ 34.1° and 179.7°. Key bond lengths are also indicated.
Figure 3 shows the structures obtained in this way for
the three classes of compound. In each case, it is clear that
the ligand is not planar: on the contrary, there is a
significant twisting of the planes of the lateral quino-
line/azaindole rings relative to the central pyridyl/ben-
zene ring. Such a twisting is intuitively reasonable, since
the platinum ion would be too large to bind to the ligand
in a planar conformation, and the bite angle of the ligand
would be inappropriate. By rotation of the plane of the
lateral heterocyclic rings relative to that of the central
ring, the d8 metal ion achieves its preferred square planar
coordination. The torsion angles between the planes are
shown in Figure 3; the angle is smallest for the complex of
the azaindole ligand [PtL3Cl]þ.
in the region 0.6-0.8 V, which is tentatively attributed to
a metal-based oxidation process. Typically, Pt(II) com-
plexes with di- and tri-imines and related ligands either do
not display clear-cut oxidations, or, if they do so, they are
normally irreversible, owing to the reactivity of Pt(III) at
the vacant axial sites.31 DFT supports the notion that the
metal is involved in the oxidation: the calculated highest-
occupied molecular orbital (HOMO) of each complex
(Figure 4) is composed almost exclusively of metal and
co-ligand (Cl, OMe, or acetylide) orbitals.
The cationic quinolyl complexes display two partially
reversible reduction processes in the cyclic voltammetry,
most likely involving the π* orbitals of the terdentate ligand.
The reduction potentials for [PtL1-2Cl]þ are similar to those
displayed by [Pt(tpy)Cl]þ (Table 2), where terpyridine-
based reduction is well-established.9,32 The LUMO
(lowest unoccupied molecular orbital) and LUMOþ1 of
these complexes calculated by DFT are indeed based almost
exclusively on the terdentate ligand, spanning all five aro-
matic ring moieties, and with essentially no contribution
from the metal or co-ligand (Figure 4 and Figure S3 in
Supporting Information). Strikingly, however, the first
reduction process in the azaindole complex [PtL3Cl]þ is
cathodically displaced relative to its quinolyl analogue
by >1 V. This observation is consistent with a simplistic
picture of the azaindole unit being much more electron-rich
than a quinoline: for example, in its reactivity, azaindole
behaves as an indole, susceptible to electrophilic attack,
rather than as a quinoline. The picture is further supported
by the DFT calculations, which reveal that the LUMO in
this complex is localized on the central pyridine ring of the
ligand, with essentially no contribution from the lateral
rings (Figure 4), and in stark contrast to the delocalized
nature of the LUMO in the quinolyl complexes.
Importantly, it can be seen that the N-Pt-N angle (for
the trans-related, quinolyl nitrogen atoms) is close to 180°
in each case, suggesting that the strategy of employing
6-membered chelate rings is successful in relieving the
strain associated with the more acute angles subtended by
5-membered-chelate-forming ligands such as terpyridine.
A similar twisting of the ligands was observed for
[RuL1 ]2þ by Hammarstrom, Johansson, and co-work-
€
ers, leading to a geometry close to pure octahedral at the
2
metal (intraligand trans N-Ru-N angles are 177.6(7)°
determined by X-ray diffraction).19a Only for the meth-
oxide adduct [PtL1OMe]þ is there a significant (though
still quite small) deviation from linearity: the angle here is
174.9°. In contrast to terpyridyl complexes, this deviation
is caused by displacement of the metal ion toward the
central ring, that is, such that Npy-Pt-Nquin > 90°.
4. Electrochemistry of the Complexes and Frontier
Orbital Description. The cationic complexes [PtL1-3Cl]þ
and derivatives [PtL1OMe]þ and [PtL1(CtC-tfp)]þ were
investigated by cyclic voltammetry in acetonitrile solu-
tion, in the presence of Bu4NPF6 (0.1 M) as the support-
ing electrolyte; data are reported in Table 2 relative to
ferrocene under the same conditions (E1/2ox = 0.40 V vs
SCE30). Each complex displays an irreversible oxidation
(31) Only when there is some protection against such reactivity are
reversible oxidative processes observed; e.g., Klein, A.; Kaim, W. Organo-
metallics 1995, 14, 1176.
(32) Rakhimov, R. D.; Weinstein, Y. A.; Lileeva, E. V.; Zheligovskaya,
N. N.; Mel’nikov, M. Y.; Butin, K. P. Russ. Chem. Bull. 2003, 52, 1150.
(30) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877.

Article
Inorganic Chemistry, Vol. 49, No. 2, 2010 481
Table 2. Ground State UV-Visible Absorption and Electrochemical Data of the Platinum(II) Complexes in Solution at 298 K, and the Predominant Character and Energy of
the Lowest-Lying Singlet State Obtained by TD-DFT
complex
absorption λmax/nm a ε/M-1 cm-1
lowest S state E/eV [λ/nm]
E0ox/Vb
E0red/V b (Δ/mV)
[Pt(tpy)Cl]þ
[PtL1Cl]þ
305, 320, 340, 390, 403 c
-1.24 (70), -1.82 (70)c
-1.27 (160), -1.78
-1.41 (120), -1.98
-1.49 (100), -1.66
d
240 (67 200), 327sh (23 200), 347 (26 100)
274 (81 400), 350 (28 500), 405 (4 860)
239 (64 700), 271 (34 300), 349 (24 200), 364 (21 500)
238 (76 800), 274 (25 000), 337 (24 000)
265 (55 700), 336 (16 800)
HOMOfLUMO 2.99 [415]
HOMOfLUMO 2.44 [508]
HOMOfLUMO 2.45 [506]
0.67
0.77
0.78
0.83
0.62
0.62 e
0.35e,f
[PtL1(OMe)]þ
[PtL1(CtC-tfp)]þ
[PtL2Cl]þ
[PtL3Cl]þ
HOMOfLUMO 3.38 [366]
HOMOfLUMO 2.61 [475]
-2.73
PtL4Cl
Pt(dpyb)Cl
320 (67 500), 356 (49 900), 420 (31 500)
332 (6510), 380 (8690), 401 (7010), 454 (270), 485 (240) e
-1.57 (80), -2.26 (160)
-2.14e,f
a In solution in CH2Cl2. b In solution in MeCN except where indicated otherwise, using Bu4NPF6 (0.1 M) as the supporting electrolyte. Peak potentials
are given for the irreversible oxidations and second reductions where return wave was ill-defined. For the pseudo-reversible reductions, values refer to
E1/2 and the peak-to-peak separation is given in parentheses. Values refer to a scan rate of 300 mV s-1 and are reported relative to Fcþ|Fc. c Data from
refs 14 and 9a. d The reduction wave was poorly defined for this complex. e In CH2Cl2. f Data from ref 13a.
Figure 4. Frontier orbitals of complexes [PtL1Cl]þ, [PtL3Cl]þ, and PtL4Cl calculated by TD-DFT.
The electrochemistry of the cyclometalated complex
PtL4Cl was examined in dichloromethane solution, owing
to poor solubility in MeCN. As for the N∧N∧N analogue
[PtL1Cl]þ, an irreversible oxidation and two reduction
processes were observed. A quantitative comparison of
the potentials is undermined by the use of a different
solvent, but all the potentials are shifted to less positive/
more negative values, consistent with the more electron-
rich character expected for this charge-neutral, cyclome-
talated complex. In this case, the HOMO calculated by
DFT shows not only Pt(5d) and Cl(p) character but also a
significant contribution from the central phenyl ring
(Figure 4), whose electron density will be raised as a result
of the deprotonation that accompanies cyclometalation.
Conversely, this ring plays much less role in the LUMO
than the pyridyl ring in the N∧N∧N complexes, and the
LUMO becomes essentially localized on the lateral quino-
line moieties. Thus the picture is, in effect, the reverse of
that found in the complex of the azaindole ligand L3. A
broadly similar trend of significant participation of the
cyclometallating ring in the HOMO, but relatively low
contribution to the LUMO, has also been observed in TD-
DFT studies of Pt(dpyb)Cl and its derivatives.13c,33
solution in dichloromethane at room temperature are
shown in Figure 5, and data for all the complexes are
compiled in Table 2. Additional spectra are provided in
the Supporting Information. Each complex displays a set
of intense bands at high-energy (<280 nm), at least some
of which can be attributed to π-π* transitions within the
ligands, as they display bands in the same region. For
consideration of lower-energy features, we divide the
complexes into three groups:
(i). [PtL1X]þ (X = Cl, OMe, CtC-tfp), and [PtL2Cl]þ.
A band or, more likely, an envelope of several close bands
around 350 nm, has no counterpart in the uncoordinated
ligands. They show only a very small degree of negative
solvatochromism (e.g., Supporting Information, Figure S5).
These bands are anticipated to arise from charge-transfer
transitions involving the metal and/or co-ligand, by com-
parison with other established platinum di- and tri-imine
complexes.9,8c Time-dependent DFT (TD-DFT) calcula-
tions reveal that the lowest-lying spin-allowed singlet state
is composed mainly of HOMOfLUMO character in each
case. Hence, it is appropriate to consider the frontier orbital
pictures of Figure 4 in interpreting the nature of the lowest-
energy bands in the absorption spectra. They suggest that
the lowest-energy band be assigned as charge-transfer from
the metal and monodentate co-ligand X to the tridentate
ligand, d(Pt)/π(X) f π*(N∧N∧N). The lack of significant
solvatochromism can be understood in terms of the move-
ment of electron density from the central axis of the
5. Photophysical Properties. (a). Absorption. The
absorption spectra of representative complexes in dilute
(33) Sotoyama, W.; Satoh, T.; Sato, H.; Matsuura, A.; Sawatari, N.
J. Phys. Chem. A 2005, 109, 9760.

482 Inorganic Chemistry, Vol. 49, No. 2, 2010
Garner et al.
methoxide adduct, the experimental absorption spectrum
does reveal a tail on the red end of the spectrum of the
acetylide adduct compared to the parent chloro complex
(Figure S4, Supporting Information).
(ii). [PtL3Cl]þ. The complex of the azaindole ligand
similarly displays a quite broad band, but at higher
energy, λmax = 336 nm. Again the TD-DFT indicates
that the lowest-lying singlet state has primarily HOMOf
LUMO character. The higher energy of the band can then
be understood in terms of the destabilized LUMO, as
evident from the more negative reduction potential of this
complex (vide supra) and from the molecular orbital plot
of the LUMO in Figure 4, which was seen to be largely
localized on the pyridine ring rather than being deloca-
lized across the N∧N∧N ligand.
(iii). PtL4Cl. The cyclometalated complex has a pro-
nounced band at lower energy (λmax ∼ 420 nm) than its
N∧N∧N-coordinated analogue [PtL1Cl]þ. This is fully
consistent with cyclometalation, which tends to raise the
energy of the HOMO owing to the increased electron
density, as discussed in section 4: the substantial partici-
pation of the central metalated ring in the HOMO is
shown in Figure 4. The TD-DFT data (Table 2) confirms
that a significantly lower S1 energy is expected for this
complex compared to [PtL1Cl]þ (difference of 0.38 eV).
The band displays a larger degree of solvatochromism
than [PtL1Cl]þ (Supporting Information, Figure S6).
We note that in none of these complexes is there
evidence for any lower energy bands arising from the
direct SfT transition. This is in contrast to Pt(dpyb)Cl
and derivatives, for example, where relatively weak but
distinct low-energy bands in the absorption spectrum
(ε ∼ 200) arise from direct excitation to the triplet
state facilitated by spin-coupling associated with the
Pt(II) ion.13,35 It suggests that the oscillator strength for
these spin-forbidden transitions is small in the present set
of complexes. We return to this point in the next section.
(b). Emission. All of the quinolyl complexes are lumi-
nescent in solution at room temperature; data are com-
piled in Table 3 and representative spectra are shown in
Figure 6.
Figure 5. Normalized UV-visible absorption spectra of the Pt(II)
complexes in CH2Cl2 at 298 K. The spectra of [PtL1(CtC-tfp)]þ and
[PtL2Cl]þ are omitted here for clarity; they are provided in Figure S4 of
the Supporting Information, and data is summarized in Table 2. Inset:
excitation spectra of [PtL1OMe]þ and PtL4Cl (λem = 600 nm) in CH2Cl2
at 298 K.
molecule outward toward the quinolyl rings, rather than
from one end of the molecule to the other, such that the
change in dipole moment accompanying excitation will be
minimal.
The absorption spectrum of the methoxy-substituted
complex [PtL2Cl]þ is similar to that of [PtL1Cl]þ, but a
little shifted to the blue, probably reflecting destabiliza-
tion of the LUMO by the electron-donating effect of the
substituent (see Figure S4 in Supporting Information for
spectrum).
In the case of the methoxide derivative, the lowest-
energy band is a shoulder at 405 nm, having apparently
emerged from under the envelope of bands centered
around 350 nm (Figure 5). Qualitatively, this would be
consistent with the notion that the methoxide co-ligand
is a stronger σ-donor than chloride, such that the energy
of the CT state will be reduced. The difference in energy
relative to the center of the band in [PtL1Cl]þ (λmax
=
(i). [PtL1Cl]þ. [PtL1Cl]þ displays a broad, unstruc-
tured emission profile centered at 600 nm. The emission
displays no significant solvatochromism: the spectra
recorded in tetrahydrofuran (THF), CHCl3, and MeCN
are superimposable with the spectrum in CH2Cl2 under
the same conditions. The quantum yield of luminescence
in solution is 0.036 (degassed CH2Cl2, 298 K), a value
comparable to that of such archetypal transition metal
complexes as [Ru(bpy)3]2þ, and, strikingly, vastly super-
ior to that of the terpyridyl analogue [Pt(tpy)Cl]þ, from
which emission can scarcely be detected under these
conditions.9 The luminescence decay follows monoexpo-
nential kinetics, with a lifetime of 16 μs, indicative of
emission emanating from a triplet state promoted by the
spin-orbit coupling associated with the Pt(II) center. No
significant self-quenching was observed upon increasing
the concentration from 5 ꢀ 10-6 to 10-4 M. At 77 K in a
frozen glass, the emission becomes highly structured with
a vibrational progression of ∼1300 cm-1 (Figure 7).
347 nm) is around 4400 cm-1
.
Meanwhile, the energies of the S1 excited states pre-
dicted by TD-DFT are listed in Table 2. Although the
absolute values are underestimated compared to the
experimental ones in all cases, consideration of the trends
is informative. Indeed, for [PtL1OMe]þ compared to
[PtL1Cl]þ, there is a difference of 0.54 eV or 4400 cm-1
,
which agrees remarkably well with the experimental
value. The possibility that the low-energy band is due to
a dimer or aggregate, leading to a Pt2(σ*) f π*(N∧N∧N)
MMLCT transition34 for example, can be ruled out,
because (i) the profile of the absorption spectrum is
independent of concentration (the range up to 10-4
M
was investigated), and (ii) the bands also appear in the
luminescence excitation spectrum (see Figure 5 and Sec-
tion 5b). The TD-DFT calculations also suggest that a
similar low-energy transition should be observed in
[PtL1(CtC-tfp)]þ. Although less clear-cut than for the
(34) Miskowski, V. M.; Houlding, V. H.; Che, C.-M.; Wang, Y. Inorg.
Chem. 1993, 32, 2518.
(35) Rausch, A. F.; Murphy, L.; Williams, J. A. G.; Yersin, H. Inorg.
Chem. 2009, 28, 11407.

Article
Inorganic Chemistry, Vol. 49, No. 2, 2010 483
Table 3. Luminescence Data for the Platinum Complexes in CH2Cl2 Solution at 298 K except Where Stated Otherwise
emission 77 Kc
P
emission
max/nm
τ/μs degassed
(aerated)
kQO2/108
kr/103
knr/104
complex
λ
Φlum ꢀ 102
M-1 s-1a
s-1b
s-1b
λmax/nm
τ/μs
[Pt(tpy)Cl]þ
[PtL1Cl]þ
500, 535, 590d
600
612
593
558(sh), 594
f
0.04d
3.6
0.7
4.2
<0.01 d
16 (1.1)
1.3 (0.58)
33 (0.69)
7.3 (0.94)
40
104
6.0
76
2.9
14
476, 510, 543, 588d,e
538, 579, 623
544, 587, 628
537, 578, 622
535, 572, 616
451, 483, 519
575, 626, 676
492, 525, 562
13.3d,e
130
70
135
160
460
28
3.8
4.3
6.4
4.2
2.3
5.4
1.3
1.4
[PtL1(OMe)]þ
[PtL1(CtC-tfp)]þ
[PtL2Cl]þ
1.0
[PtL3Cl]þ
PtL4Cl
Pt(dpyb)Cl
611(sh), 645
491, 524, 562
1.6
60g
14 (f)
7.2 (0.5)g
1.1
83
7.0
5.6
9.1g
7.0
a kQO2 is the bimolecular rate constant for quenching by molecular oxygen at 298 K, estimated from the lifetime in degassed and aerated solutions,
P
assuming [O2] at 1 atm pressure of air = 2.2 mmol dm-3
.
b kr and knr are the radiative and non-radiative rate constants estimated from the quantum
yield and lifetime at 298 K. c In diethylether/isopentane/ethanol, except where stated otherwise. d Data from ref 9a. e In ethanol/methanol/dimethylform-
amide 5:5:1 (EMD) at 70 K. f Emission too weak to be detected using available instrumentation. g Data from ref 13a.
Figure 7. Emission spectra of [PtL1Cl]PF6, [PtL3Cl]PF6, and PtL4Cl at
77 K in a diethyl ether/isopentane/ethanol (2:2:1 v/v) glass. The spectra of
Figure 6. Emission spectra of selected Pt(II) complexes in CH2Cl2 at
the other complexes are omitted for clarity, but are very similar to that of
298 K, following excitation into the lowest-energy absorption band. The
[PtL1Cl]þ, being only slightly blue- and red-shifted respectively for
[PtL2Cl]þ and [PtL1OMe]þ, see Figure S7 in the Supporting Information.
spectrum of [PtL1(CtC-tfp)]þ, omitted here for clarity, is provided in
Figure S8 of the Supporting Information; see also Table 3.
approximation to make, especially given the good match
between the excitation and absorption spectra. The value
The respectable luminescence quantum yield and long
lifetime at room temperature indicate that non-radiative
decay pathways have been greatly reduced in the new
complex compared to its terpyridyl analogue. The start-
ing hypothesis that the increased chelate ring size should
reduce the ring strain, leading to a stronger ligand field
and displacement of potentially deactivating d-d states to
higher energy, thus appears to be vindicated.36
P
of knr obtained in this way is comparable to that for
Pt(dpyb)Cl and derivatives, one of the most emissive class
of platinum complexes known (Φ > 0.6),8c suggesting
that the 6-membered chelate strategy is comparable to
cyclometalation in terms of reducing non-radiative decay
pathways.
On the other hand, the factor limiting the quantum
yield of [PtL1Cl]þ is seen to be the relatively small
radiative rate constant kr, which is well over an order of
magnitude lower than for Pt(dpyb)Cl and some 2 orders
of magnitude less than the brightest iridium-based emit-
ters such as Ir(ppy)3.38 The lack of observable low-energy
transitions to the triplet state in the absorption spectrum
(vide supra) similarly indicates a low oscillator strength
for the SfT transition. Low radiative rate constants and
oscillator strengths may reflect a relatively low participa-
tion of the metal in the excited state, being typical
for ligand-centered (3LC, π-π*) transitions, since it is
the spin-orbit coupling associated with the metal ion
that promotes the formally spin-forbidden transitions.
In interpreting the above observations, it is informative
to estimate the radiative rate constant (kr) and the sum of
P
the non-radiative decay constants ( knr) from the quan-
tum yields and lifetimes. Although this approach assumes
that the emitting state is formed with unitary efficiency;
almost certainly incorrect given recent observations
of time-zero fluorescence in third row transition metal
complexes37;it is probably nevertheless a reasonable
(36) A larger gap between the emissive state and the d-d state could also be
explained simply on the basis that the emissive state in [PtL1Cl]þ is lower in
energy than in [Pt(tpy)Cl]þ, and might not necessarily mean that the d-d state
has been raised in energy. However, we note that the new complex is
significantly brighter than complexes with tpy derivatives that emit at similar
energies; e.g., see refs 9 and 8c for examples.
(37) (a) Cannizzo, A.; Blanco-Rodriguez, A. M.; El Nahhas, A.; Sebera,
ꢀ ꢁ
ꢁ
J.; Zalis, S.; Vlcek, A.; Chergui, M. J. Am. Chem. Soc. 2008, 130, 8967.
(b) Hedley, G. J.; Ruseckas, A.; Samuel, I. D. W. J. Phys. Chem. A 2009, 113, 2.
(38) Flamigni, L.; Barbieri, A.; Sabatini, C.; Ventura, B.; Barigelletti, F.
Top. Curr. Chem. 2007, 281, 143.

484 Inorganic Chemistry, Vol. 49, No. 2, 2010
Garner et al.
A ligand-centered assignment is supported by the ob-
served absence of solvatochromism in the emission
spectrum. The highly structured form of the emission at
77 K and its resemblance to that of the phosphorescence
of the uncoordinated ligand L1 (Table 4 and Figure S10 in
the Supporting Information) also supports the notion
that a predominantly ligand-centered assignment may be
appropriate.
Table 4. Phosphorescence Maxima and Lifetimes (in EPA at 77 K) of Ligands
L
1-3 and HL4
ligand
emission λmax/nm
τ/ms
L1
496, 532, 570
492, 528, 568
428, 442, 459, 474, 485
503, 536, 574
490
600
2400
560
L2
L3
HL4
€
Hammarstrom, Johansson, and co-workers similarly
found that non-radiative decay was greatly decreased in
the linear N-Pt-N geometry for the trans-disposed
nitrogens, the predicted Pt-N bond length for the central
[RuL1 ]2þ compared to [Ru(tpy)2]2þ, attributed to the less
˚
N becomes longerþthan optimal (2.11 A compared to
2
strained coordination geometry.19 In that case, however,
the broad, structureless emission band appeared at much
lower energy (λmax = 700 nm at 298 K; 673 nm at 77 K).
An unequivocal MLCT assignment is clear-cut in that
instance. The lower energy of the atomic d orbitals of
Pt(II) compared to Ru(II) would account for the greater
LC character of the Pt complex.
1
2.05 A in [PtL Cl] , Figure 3), no doubt reducing the
˚
rigidity and ligand field strength. Both factors will lead to
faster non-radiative decay in fluid solution. Meanwhile,
the higher-energy of the emissive state in this case neces-
sarily reduces the energy gap to the higher-lying d-d
states, potentially also promoting deactivation. Coupled
with the smaller kr value, these factors probably account
for the lack of emission at room temperature. We note
that Wang and co-workers have explored N∧C∧N-coor-
dinating ligands based on the 7-azaindole unit,39 and
similarly found that their Pt(II) complexes were non-
emissive at room temperature.40
(ii). [PtL2Cl]þ, [PtL1OMe]þ, and [PtL1-CtC-tfp]þ.
The acetylide adduct displays an emission spectrum very
similar to that of the parent chloro complex. The spec-
trum of the methoxide adduct is also similar but margin-
ally red-shifted compared to [PtL1Cl]þ, while that of
[PtL2Cl]þ is a little blue-shifted and shows a hint of
structure (Figure 6). In a frozen glass at 77 K, all four
complexes display very similar spectra (Figure S7 in the
Supporting Information). The similarity between the
spectra offers further support for a predominantly ligand-
centered assignment to the emissive state in these com-
plexes, since the co-ligand is seen to have little influence
on the triplet excited state energy. Again, there is essen-
tially no solvatochromism in fluid solution (Supporting
Information, Figure S8), supporting the LC assignment.
Changing the co-ligand does, however, affect the lumi-
nescence lifetime; in particular, the value is increased to
33 μs for the acetylide adduct, apparently reflecting both
reduced radiative and non-radiative decay rates (Table 3),
while the methoxide adduct has the shortest lifetime and
lowest quantum yield among the four complexes, owing
to substantially increased non-radiative decay.
(iv). PtL4Cl. The cyclometalated complex emits in
degassed solution at room temperature: λmax = 645 nm,
τ = 14 μs, Φ = 0.016 in CH2Cl2. The emission maximum
is red-shifted relative to the cationic quinolyl complexes
(Figure 6). There is a hint of vibrational structure in the
spectrum in the form of a higher-energy shoulder, similar
to that displayed by [PtL2Cl]þ. Again, there is essentially
no solvatochromism (Supporting Information, Figure S9).
At 77 K, the red-shift is more clear-cut. The spectrum at
this temperature has a profile essentially identical to that of
the cationic quinolyl complexes, but is shifted to lower-
energy by ∼1200 cm-1 (Figure 7). As in absorption, it is
reasonable to ascribe the lower excited state energy to a
destabilized HOMO arising from the effect of cyclometa-
lation. This interpretation, in terms of a greater influence
of the metal in the excited state of the cyclometalated
compound compared to the N∧N∧N complex, is rein-
forced by the observation that the uncomplexed ligands
L1 and HL4 have emission spectra at 77 K that are
essentially identical to one another (Supporting Informa-
tion, Figure S10).
It is pertinent to compare the emission properties
of PtL4Cl with those of Pt(dpyb)Cl, the analogous com-
plex incorporating pyridyl rings (coordinating through
5-membered chelates) rather than quinolines. The former
has a substantially lower luminescence quantum yield and
longer luminescence lifetime than the latter at room
temperature. The difference seems to be due primarily
to the lower radiative rate constant of PtL4Cl. The value
of 1100 s-1 (Table 3) is around 2 orders of magnitude less
than that for Pt(dpyb)Cl, while the non-radiative decay is
of a similar order in the two complexes. Recent cryogenic
studies (T = 1.2-4 K) have revealed the presence of
efficient spin-orbit coupling pathways in Pt(dpyb)Cl,
despite the apparent “LC-like” nature of the emission.35
(iii). [PtL3Cl]þ. No emission could be detected from
this complex in degassed solution at room temperature.
At 77 K, a weak, structured emission spectrum could be
recorded (Figure 7), blue-shifted substantially compared
to the other cationic complexes: λmax0-0 = 451 nm. The
higher-energy of the emissive excited state is fully con-
sistent with the conclusions of the absorption, electro-
chemical, and TD-DFT data discussed above, interpreted
in terms of a destabilized ligand π* LUMO in this
complex. The luminescence lifetime of 460 μs at 77 K is
much longer than for any of the other complexes, suggest-
ing that the radiative rate constant is low. Meanwhile,
the proportionally higher intensity of the 0-1 and 0-2
vibrational bands relative to the 0-0 in [PtL3Cl]þ com-
pared to the other complexes (i.e., larger Huang-Rhys
factor) suggests that the formation of the excited state is
accompanied by more distortion than in the other com-
plexes. One might tentatively propose that the binding
angle of L3 may be less appropriate than in the other
complexes, being adversely affected by the more acute
angle associated with the five-membered pyrrolic ring of
the azaindole unit, compared to the six-membered ben-
zenoid ring of the quinolines. We note that, in adopting
(39) Wu, Q.; Lavigne, J. A.; Tao, Y.; D’Iorio, M.; Wang, S. Chem. Mater.
2001, 13, 71.
(40) Song, D.; Wu, Q.; Hook, A.; Kozin, I.; Wang, S. Organometallics
2001, 20, 4683.

Article
Inorganic Chemistry, Vol. 49, No. 2, 2010 485
Apparently, spin-orbit coupling is less efficient in the
quinolyl-based complex than in the pyridyl analogue, but
more detailed studies of PtL4Cl at such temperatures will
be required to unravel the origins of this difference.
at the frequencies indicated. Chemical shifts (δ) are in parts
per million, referenced to residual protio-solvent resonances,
and coupling constants are in hertz. Electrospray ionization
mass spectra were acquired on a time-of-flight Micromass
LCT spectrometer. All solvents used in preparative work
were at least Analar grade, and water was purified using the
Purite system. Solvents used for optical spectroscopy were
HPLC grade.
Concluding Discussion and Summary
In this study, we have prepared new complexes of platinum
with terdentate ligands incorporating 8-substituted quino-
lines or 7-N-substituted azaindoles as the lateralcoordinating
units. The energy-minimized geometries predicted by DFT
reveal that, by twisting away from planarity, these ligands are
able to bind to platinum in a terdentate manner, in such a way
that the N-Pt-N angle is close to 180°. This contrasts with
the smaller angle of around 163° observed in [Pt(tpy)Cl]þ and
derivatives. The ability to bind linearly, while similar bond
lengths are retained, is expected to lead to an increase in the
ligand-field strength exerted by the terdentate ligand.
For the N∧N∧N-coordinating complexes based on the
quinolyl ligands ([PtL1Cl]þ and [PtL2Cl]þ), a dramatic dif-
ference in photophysical properties is observed compared to
[Pt(tpy)Cl]þ. Whereas the photoluminescence of the latter is
scarcely detectable in solution at room temperature,
[PtL1Cl]þ emits quite brightly, with a quantum yield of
3.6%. The difference may be due to the displacement of the
potentially deactivating d-d states to higher energy, as a result
of the stronger predicted ligand field. A number of observa-
tions (relatively long luminescence lifetimes, low radiative rate
constants, lack of solvatochromism, vibrationally structured
emission spectra at 77 K that have little dependence on the
identity of the co-ligand) suggest that the emission emanates
from a triplet state of predominant ligand-centered character.
In contrast, no emission can be detected from the
N∧N∧N-coordinated complex based on the azaindole ligand
[PtL3Cl]þ, despite a similar linear N-Pt-N geometry pre-
dicted in this case. Clearly, unstrained geometry alone is not
sufficient to confer good emissive properties. The electronic
properties appear to be such that the triplet radiative decay
rate is not promoted to the same extent as in the other
complexes, as manifest by the particularly long luminescence
lifetime at 77 K. Non-radiative decay is also likely to be
increased in this complex, owing to a less rigid structure and a
longer Pt-py bond.
2,6-Di(8-quinolyl)pyridine L1. This compound was initially
prepared from 2,6-dibromopyridine and quinoline-8-boronic
acid as reported previously by Johansson and co-workers.19a,20
We note that the reaction proceeded equally well from
2,6-dichloropyridine, using an otherwise identical procedure.
1H NMR data has previously been reported in CDCl3.19a To
assist comparison with the spectra of the Pt(II) complexes,
we report here data in other pertinent solvents. 1H NMR
(400 MHz, d6-acetone): δ = 9.00 (2H, dd, 3J = 4.0, 4J = 1.5, H20),
8.43 (4H, overlapping dd, H4 and H7), 8.34 (2H, d, 3J = 8.0, H3 ),
0
8.05 (2H, dd, 3J = 7.0, 4J = 1.5, H5), 7.97 (1H, t, 3J = 8.0, H4 ), 7.75
(2H,dd,3J=8.5,3J=7.0,H6), 7.58 (2H, dd, 3J=8.0,3J=4.0,H3).
1H NMR (400 MHz, d6-DMSO): δ = 8.98 (2H, dd, 3J = 4.0,
4J = 1.5, H2), 8.48 (2H, dd, 3J = 7.0, 4J = 1.5, H40 ), 8.24 (2H, dd,
3J = 7.5, 4J = 1.0, H7), 8.14 (2H, d, 3J = 8.0, H3 ), 8.01 (2H, dd,
0
3J = 8.0, 4J = 1.0, H5), 7.97 (1H, t, 3J = 8.0, H4 ), 7.74 (2H, dd,
3J = 8.0, 3J = 7.0, H6), 7.61 (2H, dd, 3J = 7.5, 3J = 4.0, H3).
4-Methoxy-2,6-di(8-quinolyl)pyridine L2. A mixture of 4-
methoxy-2,6-dibromopyridine (103 mg, 0.354 mmol), quino-
line-8-boronic acid (202 mg, 1.17 mmol), Pd(dba)2 (9 mg, 0.016
mmol) and K3PO4 (1.12 g, 5.27 mmol) in toluene (5 mL) was
degassed by five freeze-pump-thaw cycles. 2-Dicyclohexylpho-
sphino-20,60-dimethoxybiphenyl, “SPhos” (12 mg, 0.029 mmol)
was added under a flow of nitrogen and the mixture stirred at
100 °C for 18 h. After cooling to room temperature, the mixture
was filtered, the residue washed with dichloromethane, and the
solvent evaporated from the combined filtrate under reduced
pressure. Successive washing of the brown oily residue with small
portions of diethyl ether led to the product as an off-white solid. 1H
NMR (500 MHz, CDCl3) δ=9.01(2H, dd, 3J=4.0,4J=1.5,H2),
8.25 (2H, dd, 3J = 7.5, H7), 8.23 (2H, dd, 3J = 6.5, 4J = 1.5, H4),
0
7.87 (2H, dd, 3J = 7.5, 4J = 1.5, H5), 7.68 (2H, s, H3 ), 7.67 (2H, dd,
3J = 7.5, 7.5, H6), 7.45 (2H, dd, 3J = 6.5, 4J = 4.0, H3), 3.98 (3H, s,
OCH3). MS (ESþ) m/z = 364 [MþH]þ. HRMS (ESþ) m/z =
364.1443 [MþH]þ; calcd for C24H18N3O = 364.1444.
2,6-Di(N-7-azaindolyl)pyridine, L3. A mixture of 2,6-dibro-
mopyridine (200 mg, 0.84 mmol), 7-azaindole (300 mg, 2.53
mmol), K2CO3 (709 mg, 5.07 mmol), trans-1,2-diaminocyclo-
hexane (29 mg, 0.25 mmol), and CuI (8 mg, 0.042 mmol) in
dioxane (3 mL) was degassed by three freeze-pump-thaw
cycles. The mixture was then heated at reflux under a nitrogen
atmosphere for 72 h. The solvent was evaporated to yield a
brown residue, which was extracted into CH2Cl2 and washed
with water. The organic layer was dried over MgSO4, and the
solvent evaporated under reduced pressure. The resulting solid
was washed with acetonitrile to yield L3 as an off-white solid
Finally, the cyclometalated, N∧C∧N-coordinated complex
PtL4Cl proves to be emissive, but in this case, the efficiency is
reduced relative to the analogous pyridyl-based complex
with 5-membered chelates, apparently because of the triplet
radiative transition being less allowed.
Are terdentate ligands that form 6-membered chelate rings
better, from the point of view of promoting luminescence of
their platinum complexes, than those which conventionally
form 5-membered ones? Clearly, the study reveals that such
ligands may be preferable, but are not necessarily so. As
usual, a variety of factors, geometric and electronic, are
involved in determining luminescence quantum yields.
In the case of the N∧N∧N-coordinating quinolyl ligands,
performance is certainly enhanced, and the study may open
up a range of possibilites with this new class of emissive
complexes.
(154 mg, 59%). 1H NMR (CDCl3, 500 MHz): δ = 8.83 (2H, d,
0
3
3
3J = 8.0, H3 ), 8.69 (2H, d, J = 4.0, H6)0, 8.44 (2H, dd, J =
4.5, 4J = 1.5, H2), 8.06 (1H, t, 3J = 8.0, H4 ), 7.98 (2H, dd, 3J =
8.0, J = 1.5, H4), 7.19 (2H, dd, 3J = 8.0, J = 4.5, H3), 6.68
4
3
(2H, d, 3J = 4.0, H5). 1H NMR (d6-DMSO, 500 MHz): δ = 8.83
0
(2H, d, 3J = 8.0, H3 ), 8.68 (2H, d, 3J = 4.0, H6), 8.44 (2H, dd,
0
3J = 4.5, 4J = 1.5, H2), 8.22 (1H, t, 3J = 8.0, H4 ), 8.13 (2H, dd,
3J = 8.0, 4J = 1.5, H4), 7.30 (2H, dd, 3J = 8.0, 3J = 4.5, H30),
6.82 (2H, d, 3J = 4.0, H5). 13C NMR (CDCl3): δ = 149.4 (C2 ),
0
147.9 (C8), 143.5 (C2), 141.1 (C4 ), 129.4 (C4), 126.5 (C6), 123.6
0
(C9), 117.5 (C3), 111.8 (C3 ), 103.0 (C5). MS (ESþ) m/z: 312,
[MþHþ]. HRMS (ESþ) m/z: 312.1242; calcd for C19H14N5 =
312.1244. Anal. Calcd. for C19H13N5(1/4H2O): C, 72.25; H,
4.31; N, 22.17%. Found: C, 72.50; H, 4.18; N, 22.39%. Mp =
215-216 °C.
Experimental Section
1H and 13C NMR spectra, including NOESY and COSY,
were recorded on Varian or Bruker spectrometers operating

486 Inorganic Chemistry, Vol. 49, No. 2, 2010
Garner et al.
1,3-Di(8-quinolyl)benzene HL4. Benzene-1,3-diboronic acid
(0.100 g, 0.603 mmol) and 8-bromoquinoline (0.25 g, 1.201
mmol) were suspended in a mixture of toluene (6 mL), ethanol
(6 mL), and sodium carbonate solution (2M, 3 mL), and the
reaction mixture was degassed by four freeze-pump-thaw
cycles. Pd(PPh3)4 (0.030 g, 0.026 mmol) was added under a
positive pressure of nitrogen, and the solution heated under
nitrogen at reflux for 42 h. After cooling to room temperature,
water (5 mL) was added, and the product extracted into
dichloromethane (3 ꢀ 5 mL). The combined extracts were dried
over anhydrous MgSO4, filtered, and the solvent removed under
reduced pressure to yield a translucent oily solid. Purification
was carried out via column chromatography (silica, hexane/
ethyl acetate, elution gradient from 100:0 to 70:30) to yield a
colorless solid (0.15 g, 76%). Rf (silica) = 0.1 in hexane/diethyl
ether 50:50. 1H NMR (500 MHz, CDCl3) δ = 8.97 (2H, dd, 3J =
[PtL3Cl]Cl and [PtL3Cl]PF6. A mixture of L3 (0.050 g,
0.16 mmol) and Pt(DMSO)2Cl2 (0.087 g, 0.19 mmol) in metha-
nol (3 mL) was degassed by freeze-pump-thaw five times. The
mixture was refluxed under nitrogen for 3 d. The off-white
precipitate was separated from the solution by centrifugation
and washed with methanol, ethanol, diethyl ether, and dichloro-
methane to leave the product (53 mg, 57%). Conversion to the
PF6- salt was performed by dissolution in the minimum volume
of DMSO and dropwise addition of the solution to aqueous
KPF6. Data for the chloride salt: 1H NMR (d6-DMSO,
3
3
500 MHz): δ = 8.73 (2H, d, J = 6.0, H2), 8.61 (2H, d, J =
0
4.5, H6), 8.53 (2H, d, 3J 0= 7.5, H4), 8.46 (1H, t, 3J = 8.0, H4 ),
7.90 (2H, d, 3J = 8.5, H3 ), 7.58 (2H, dd, 3J = 7.5, 3J = 6.0, H3),
3
7.34 (2H, d, J = 4.5, H5). 13C NMR (d6-DMSO): δ = 146.7
0
0
(C2), 145.7 (C3 ), 145.2 (C8), 144.4 (C2 ), 135.4 (C4), 130.5 (C6),
0
125.3 (C9), 121.0 (C3), 115.7 (C4 ), 110.3 (C5). HRMS (ESþ) m/z:
540.04783 [Mþ]; calcd for C19H13N5Cl194Pt = 540.04759. Anal.
Calcd for C19H13N5Cl2Pt = C, 39.5; H, 2.3; N, 12.1%. Found
C, 39.0; H, 2.5; N, 11.7%. Data for hexafluorophosphate salt:
4.0, 4J = 2.0, H2), 8.21 (2H, dd, 3J = 8.5, 4J = 2.0, H4), 7.97
0
(1H, t, 4J = 1.5, H1 ), 7.86 (2H, dd, 3J = 7.0, H5), 7.83 (2H, dd,
0
3J = 7.0, H7), 7.78 (2H, dd, 3J = 7.5, 4J = 1.5, H3 ), 7.63, (3H,
0
1H NMR (CD3CN, 400 MHz): δ 8.66 (2H, d, 3J = 6.0, H2), 8.31
overlapping t and dd, H4 and H6), 7.42 (2H, dd, 3J = 8.5, 3J =
4.0, H3). 13C NMR (CDCl3) δ = 157.0, 150.5, 150.1, 146.3,
145.7, 141.1, 139.4, 136.5, 133.0, 130.8, 129.7, 127.8, 126.6,
121.7, 121.2. MS (ESþ) m/z = 332 [M]þ. Anal. Calcd for
C24H16N2 = C, 86.72; H, 4.85; N, 8.43%. Found C, 86.47; H,
4.93; N, 8.43%.
0
(2H, d 3J = 7.5, H4), 8.27 (1H, t, 3J = 8.5, H4 ), 8.01 (2H, d, 3J =
0
3.5, H6), 7.58 (2H, d, 3J = 8.5, H3 ), 7.38 (2H, dd, 3J = 8.0, 3J =
6.0, H3), 7.12 (2H, d, 3J = 4.0, H5).
PtL4Cl. Separate solutions of HL4 (50 mg, 0.15 mmol) in
acetonitrile (3 mL) and K2PtCl4 (70 mg, 0.19 mmol) in water
(1 mL) were degassed by three freeze-pump-thaw cycles. The
latter was added to the former by cannula under a nitrogen
atmosphere, and the mixture heated at reflux for 64 h. After
cooling, the precipitate formed was separated by centrifugation,
washed successively with water, ethanol, and diethyl ether (3 ꢀ
5 mL of each), and dried under vacuum leading to the complex
as a gray-green solid (12 mg, 14%). 1H NMR (500 MHz, CDCl3)
Platinum Complexes. [PtL1Cl]Cl and [PtL1Cl]PF6. A mix-
ture of 2,6-di(8-quinolyl)pyridine L1 (51 mg, 0.15 mmol) and
Pt(DMSO)2Cl2 (78 mg, 0.19 mmol) in methanol (1.5 mL) was
degassed by three freeze-pump-thaw cycles and then heated at
reflux under a nitrogen atmosphere for 72 h. The precipitate that
formed was collected by centrifugation and washed with water,
methanol, and diethyl ether (3 ꢀ 5 mL of each), and dried under
vacuum to give [PtL1Cl]Cl as a pale gray solid (40 mg, 44%).
Conversion to the hexafluorophosphate salt was achieved by
dissolution of the chloride salt in the mimium volume of DMSO
and addition to a saturated aqueous solution of KPF6. The
resulting precipitate was washed successively with water, metha-
nol, and diethyl ether (3 ꢀ 5 mL of each) and dried under
vacuum to give [PtL1Cl]PF6. Data for the chloride salt: 1H
3
3
δH = 9.76 (2H, d, J = 5.0, J(195Pt) = 48, H2), 8.41 (2H, d,
3J = 7.5, H4), 8.37 (2H, d, 3J = 7.5, H7), 7.86 (2H, d, 3J = 7.5,
H5), 7.72 (2H, dd, 3J = 7.5, 3J = 7.5, H6), 7.58 (2H, d, 3J = 8.00,
0
4J(195Pt) = 20, H3 ), 7.36 (3H, overlapping m, H3 and H4 ).
HRMS (ESþ) m/z: 560.0542 [Mþ]; calcd for C24H15N2ClPt =
560.0552. Anal. Calcd for C24H15N2ClPt = C, 51.3; H, 2.7; N,
5.0%. Found C, 51.1; H, 2.8; N, 4.9%.
3
NMR (d6-DMSO, 500 MHz) δ = 9.25 (2H, d, J = 4.5, H2),
[PtL1OMe]PF6. A mixture of [PtL1Cl]PF6 (30 mg, 0.042
mmol) and KOH (30 mg, 0.53 mmol) in methanol (3 mL) was
stirred at room temperature for 18 h. The solution was concen-
trated and then added to saturated aqueous KPF6 solution,
leading to a bright yellow precipitate, which was collected by
centrifugation, washed several times with water, and dried
under vacuum to give the product (16 mg, 67%). 1H NMR
(400 MHz, CDCl3) δ = 9.09 (2H, dd, 3J = 5.0, 4J = 1.5, H2),
8.74 (2H, dd, 3J = 7.5, H7), 8.67 (2H, dd, 3J = 8.0, 1.5, H4), 8.42
9.01 (2H, d, 3J = 7.5, H4), 8.92 (2H, d, 3J = 7.0, H7), 8.53 (2H, d,
0
3J = 7.5, H5), 8.47 (1H, t, 3J = 8.5, H4 ), 8.32 (2H, d, 3J = 7.5,
0
3
3
H3 ), 8.09 (2H, dd, J = 7.5, 7.0, H6), 7.81 (2H, dd, J = 7.5,
4J = 4.5, H3). MS (ESþ) m/z = 564 [Mþ]. HRMS (ESþ) m/z =
563.0606. Calcd for C23H15N335Cl195Pt = 563.0597. Anal.
Calcd for C23H15N3Cl2Pt = C, 46.1; H, 2.5; N, 7.0%. Found
C, 45.7; H, 2.7; N, 6.6%. Data for the hexafluorophosphate salt:
1H NMR (d6-acetone, 700 MHz) δ = 9.37 (2H, dd, J = 5.0,
3
0
0
4J = 1.5, H2), 9.02 (2H, dd, 3J = 7.5, 4J = 1.5, H4), 8.97 (2H, d,
(1H, t, 3J = 8.0, H4 ), 8.25 (2H, d, 3J = 8.0, H3 ), 8.16 (2H, d,
3J = 8.0, H5), 8.01 (2H, dd, 3J = 8.0, 3J = 7.5, H6), 7.71 (2H, dd,
3J = 7.5, H7), 8.57 (2H, d, 3J0 = 7.5, H5), 8.54 (1H, t, 3J = 8.5,
0
H4 ), 8.40 (2H, d, 3J = 8.5, H3 ), 8.13 (2H, dd, 3J = 7.5, 3J = 7.5,
3J = 8.0, J = 5.0, H3), 2.82 (3H, s, CH3). MS (ESþ) m/z =
4
H6), 7.84 (2H, td, 3J = 7.5, 4J = 5.0, H3). 13C NMR (d6-acetone)
559 [Mþ]. HRMS (ESþ) m/z = 559.1102. Calcd for C24H18N3-
O195Pt = 559.1094.
0
δ = 158.1 (C2), 151.5 (q), 142.0 (q), 141.9 (C4), 141.5 (C3 ), 133.9
0
(q),133.4 (C7), 132.6 (C5), 129.2 (q),129.2 (C6), 127.8 (C4 ), 123.3
(C3). MS (ESþ) m/z = 564 [Mþ]. HRMS m/z = (ESþ) 563.0611.
C23H15N335Cl195Pt requires 563.0597.
[PtL1(CtC-tfp)]PF6. Ethynyl-3,5-bis(trifluoromethyl)benzene
(250 μL, 1.41 mmol) and potassium hydroxide (30.0 mg, 0.53
mmol) in methanol (5 mL) were stirred at room temperature for
1 h. A solution of [PtL1Cl]PF6 (40.0 mg, 0.071 mmol) and
copper(I) iodide (4.0 mg, 0.021 mmol) in methanol (5 mL) was
added, and the solution stirred overnight. The yellow precipitate
formed was collected by centrifuge and washed with methanol
to yield a yellow solid (17 mg, 31%). 1H NMR (700 MHz,
[PtL2Cl]Cl and [PtL2Cl]PF6. The complex [PtL2Cl]Cl was
prepared in a similar way to [PtL1Cl]Cl above, from L2 (20 mg,
0.055 mmol) and Pt(DMSO)2Cl2, giving the product as a pale
gray solid (17 mg, 52%). Ion exchange with aqueous KPF6 gave
the hexafluorophosphate salt. Data for the chloride salt: 1H
NMR (d6-DMSO, 700 MHz) δ = 9.30 (2H, dd, 3J = 5.0, 4J =
1.5, H2), 8.97 (2H, dd, 3J = 7.5, 4J = 1.5, H4), 8.93 (2H, d, 3J =
3
4
d6-acetone) δ = 9.86 (2H, dd, J = 5.0, J = 1.0, H2), 9.07
(2H, dd, 3J = 8.0, 4J = 1.0, H4), 9.00 (2H, dd 3J = 7.5, 4J = 1.0,
0
0
7.0, H5), 8.49 (2H, d, 3J = 7.5, H7), 8.04 (1H, dd, 3J = 7.5, 3J =
H5), 8.55-8.57 (3H, m, H4 and H7), 8.47 (2H, 3J = 7.5, H3 ), 8.13
(2H, dd, 3J = 7.5, 3J = 7.5, H6), 7.88 (2H, dd, 3J = 8.0, 4J = 5.0,
H3), 7.80 (1H, s, H4-tfp), 7.78 (2H, s, H2-tfp). 13C NMR
(d6-acetone) δ = 160.3, 151.2, 142.6, 142.3, 141.7, 134.7, 133.4,
132.8, 131.8, 131.4, 131.1, 129.8, 129.2, 127.6, 123.5, 119.2, 110.2,
99.5, 97.2. 19F NMR (d6-acetone) δ = -63.9 (6F, s, CF3), -73.4
0
7.0, H6), 7.80 (2H, s, H3 ), 7.77 (2H, dd, 3J = 7.5, 4J = 5.0, H3),
4.00 (3H, s, OCH3). MS (ESþ) m/z = 594 [Mþ]. HRMS (ESþ)
m/z = 593.0713. Calcd for C24H17N335ClO 195Pt = 593.0704.
Anal. Calcd for C24H17N3Cl2OPt = C, 45.8; H, 2.7; N, 6.7%.
Found C, 45.2; H, 3.1; N, 6.3%.

Article
Inorganic Chemistry, Vol. 49, No. 2, 2010 487
2
(6F, d, J = 700, PF6-). MS (MALDIþ-TOF, DCTB matrix)
length modified with appropriate glassware to allow connection
to a high-vacuum line. Degassing was achieved via a minimum
of three freeze-pump-thaw cycles while connected to the
vacuum manifold; final vapor pressure at 77 K was <5 ꢀ
10-2 mbar, as monitored using a Pirani gauge. Luminescence
quantum yields were determined using [Ru(bpy)3]Cl2 in de-
gassed aqueous solution as the standard (φ = 0.04241); esti-
mated uncertainty in Φ is ( 20% or better.
m/z = 765.2 [M]þ.
Electrochemistry. Cyclic voltammetry was carried out using a
μAutolab Type III potentiostat with computer control and data
storage via GPES Manager software. Solutions of concentra-
tion 1 mM in CH2Cl2 were used, containing [Bu4N][BF4] as the
supporting inert electrolyte. A three-electrode assembly was
employed, consisting of a platinum working electrode, platinum
wire counter electrode, and platinum flag reference electrode.
Solutions were purged for 5 min with solvent-saturated nitrogen
gas with stirring, prior to measurements being taken without
stirring. The voltammograms were referenced to the ferrocence-
ferrocenium couple.
The luminescence lifetimes of the complexes were measured
by time-correlated single-photon counting, following excitation
at 374.0 nm with an EPL-375 pulsed-diode laser. The emitted
light was detected at 90° using a Peltier-cooled R928 PMT after
passage through a monochromator. The estimated uncertainty
in the quoted lifetimes is (10% or better. Lifetimes at 77 K in
excess of 10 μs were measured by multichannel scaling following
excitation with a μs-pulsed xenon lamp; an excitation wave-
length of 374 nm (band-pass 5 nm) was selected with a mono-
chromator. Bimolecular rate constants for quenching by
molecular oxygen, kQ, were determined from the lifetimes in
degassed and air-equilibrated solution, taking the concentration
DFT Calculations. B3LYP density functional calculations
were performed using the Gaussian03 software package. Double-ζ
quality basis sets were employed for the ligands (6-31G) and the
Ir ion (LANL2DZ). The inner core electrons of Ir were replaced
with a relativistic effective core potential, leaving the outer
core [(5s)2(5p)6] electrons and the (5d)6 valence electrons of
Ir(III). The geometries were fully optimized without symmetry
constraints.
of oxygen in CH2Cl2 at 0.21 atm O2 to be 2.2 mmol dm-3 42
.
Photophysical Measurements. Absorption spectra were mea-
sured on a Biotek Instruments XS spectrometer, using quartz
cuvettes of 1 cm path length. Steady-state luminescence spectra
were measured using a Jobin Yvon FluoroMax-2 spectrofluori-
meter, fitted with a red-sensitive Hamamatsu R928 photomul-
tiplier tube; the spectra shown are corrected for the wavelength
dependence of the detector, and the quoted emission maxima
refer to the values after correction. Samples for emission mea-
surements were contained within quartz cuvettes of 1 cm path
Acknowledgment. We thank EPSRC and the University of
Durham for financial support, Frontier Scientific for supplying key
reagents, and Octavia Blackburn for preparing a sample of [PtL3Cl]þ.
Supporting Information Available: Representative mass spec-
trum and additional 1H NMR spectra; frontier orbital plots of
the complexes, including HOMO-1 and LUMOþ1, obtained
by DFT calculations; absorption and emission spectra of com-
plexes not shown in the main text; emission spectra of ligands at
77 K; influence of solvent on absorption and emission spectra.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(41) Van Houten, J.; Watts, R. J. J. Am. Chem. Soc. 1976, 98, 4853.
(42) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochem-
istry, 2nd ed.; Marcel Dekker: New York, 1993.