π Donation and its effects on the excited-state lifetimes of luminescent platinum(II) terpyridine complexes in solution

Article abstract of DOI:10.1021/ic4004643

Introducing electron-donating groups extends the excited-state lifetimes of platinum(II)-terpyridine complexes in fluid solution. Such systems are of interest for a variety of applications, viz., as DNA-binding agents or as components in luminescence-based devices, especially sensors. The complexes investigated here are of the form [Pt(4′-X-T)Y]⁺, where 4′-X-T denotes a 4′-substituted 2,2′:6′,2″- terpyridine ligand and Y denotes the coligand. The π-donating abilities of -X and -Y increase systematically in the orders -NHMe < -NMe₂ < -(pyrrolidin-1-yl) and -CN < -Cl < -CCPh, respectively. The results presented include crystal structures of two new 4′-NHMe-T complexes of platinum, as well as absorption, emission, and excited-state lifetime data for nine complexes. Excited-state lifetimes obtained in deoxygenated dichloromethane vary by a factor of 100, ranging from 24 μs for [Pt(4′-pyrr-T)CN] ⁺ to 0.24 μs for [Pt(4′-ma-T)Cl]⁺, where ma-T denotes 4′-(methylamino)-2,2′:6′,2″-terpyridine and pyrr-T denotes 4′-(pyrrolidin-1-yl)-2,2′:6′,2″- terpyridine. Analysis of experimental and computational results shows that introducing a simple amine group on the terpyridine and/or a π-donating coligand engenders the emitting state with intraligand charge-transfer (ILCT) and/or ligand-ligand charge-transfer (LLCT) character. The excited-state lifetime increases when the change in orbital parentage lowers the emission energy, suppresses quenching via d-d states, and encourages delocalization of the excitation onto the ligand(s). At some point, however, the energy is low enough that direct vibronic coupling to the ground-state surface becomes important, and the lifetime begins to decrease again.

Full text of DOI:10.1021/ic4004643

Article
pubs.acs.org/IC
π Donation and Its Effects on the Excited-State Lifetimes of
Luminescent Platinum(II) Terpyridine Complexes in Solution
Lauren M. Hight, Meaghan C. McGuire, Yu Zhang, Matthew A. Bork, Phillip E. Fanwick,
Adam Wasserman, and David R. McMillin*
Department of Chemistry, 560 Oval Drive, Purdue University, West Lafayette, Indiana 47907-2084, United States
S
* Supporting Information
ABSTRACT: Introducing electron-donating groups extends the excited-state
lifetimes of platinum(II)−terpyridine complexes in fluid solution. Such
systems are of interest for a variety of applications, viz., as DNA-binding
agents or as components in luminescence-based devices, especially sensors.
The complexes investigated here are of the form [Pt(4′-X-T)Y]+, where 4′-X-
T denotes a 4′-substituted 2,2′:6′,2″-terpyridine ligand and Y denotes the
coligand. The π-donating abilities of −X and −Y increase systematically in the
orders −NHMe < −NMe2 < −(pyrrolidin-1-yl) and −CN < −Cl < −CCPh,
respectively. The results presented include crystal structures of two new 4′-
NHMe-T complexes of platinum, as well as absorption, emission, and excited-
state lifetime data for nine complexes. Excited-state lifetimes obtained in
deoxygenated dichloromethane vary by a factor of 100, ranging from 24 μs for
[Pt(4′-pyrr-T)CN]+ to 0.24 μs for [Pt(4′-ma-T)Cl]+, where ma-T denotes 4′-
(methylamino)-2,2′:6′,2″-terpyridine and pyrr-T denotes 4′-(pyrrolidin-1-yl)-2,2′:6′,2″-terpyridine. Analysis of experimental and
computational results shows that introducing a simple amine group on the terpyridine and/or a π-donating coligand engenders
the emitting state with intraligand charge-transfer (ILCT) and/or ligand−ligand charge-transfer (LLCT) character. The excited-
state lifetime increases when the change in orbital parentage lowers the emission energy, suppresses quenching via d−d states,
and encourages delocalization of the excitation onto the ligand(s). At some point, however, the energy is low enough that direct
vibronic coupling to the ground-state surface becomes important, and the lifetime begins to decrease again.
Scheme 1. Structures
INTRODUCTION
■
Platinum(II) polypyridine complexes are planar d8 systems
capable of exhibiting intriguing luminescence properties.1−4
Open access to the metal center permits Pt···Pt interactions in
the solid state that often render luminescence signals sensitive
to temperature and pressure variations as well as chemical
exposure.3−8 In the same way, encounters with a ground-state
analogue can give rise to excimer emission in solution.9 Solvent-
induced quenching is another possibility when the orbital
parentage of the emitting state exhibits significant metal-to-
ligand charge-transfer (MLCT) character.10 Rapid deactivation
via thermally accessible d−d states can reduce excited-state
lifetimes, not least in platinum(II) terpyridine complexes.11
However, it is possible to mitigate the effect by introducing
electronically active groups at either end of the 2-fold axis
passing through the complex.2−4 Results presented below reveal
the effects of modifiying both ligands (Scheme 1).
amine, and routine salts, such as sodium trifluoromethanesulfonate
(Na[OTf]), was Sigma-Aldrich. Laser dyes came from Exciton Inc.,
while the solvents used were from Mallinckrodt. ZnBr2 and NaOH
also came from Mallinckrodt as did [Ru(bpy)3]Cl2, where bpy denotes
2,2′-bipyridine. The complexes [Pt(dma-T)Cl]OTf and [Pt(dma-
T)CN]OTf, where dma-T is 4′-(dimethylamino)-2,2′:6′,2″-terpyr-
idine, and [Pt(pyrr-T)Cl]OTf and [Pt(pyrr-T)CN]OTf, where pyrr-T
is 4′-(pyrrolidin-1-yl)-2,2′:6′,2″-terpyridine, were available from
previous studies,12,13 as was hydrated sodium tetrakis[3,5-bis-
(trifluoromethyl)phenyl]borate (Na[TFPB]).13
More specifically, the π-donating strength of the ligand field
increases with a change of coligand, −CN < −Cl < −CCPh,
and/or the 4′-substituent on the terpyridine ligand, −NHMe <
−NMe2 < −(pyrrolidin-1-yl).
Methods. Details for the syntheses of the 4′-(methylamino)-
2,2′:6′,2″-terpyridine (ma-T) ligand and new platinum(II) complexes
thereof follow next.
ma-T Ligand. The method employed is a variation on a published
method for installing a methylamino group.14 The first step entails
EXPERIMENTAL SECTION
Materials. The source for silver cyanide, K2PtCl4, 4′-chloro-
2,2′:6′,2″-terpyridine (Cl-T), 1,5-cyclooctadiene (COD), diethanol-
■
Received: February 21, 2013
Published: July 24, 2013
© 2013 American Chemical Society
8476
dx.doi.org/10.1021/ic4004643 | Inorg. Chem. 2013, 52, 8476−8482

Inorganic Chemistry
Article
Table 1. Photophysical Data for [Pt(4′-X-T)Y]+ Complexes in DCM at Room Temperature, Except as Noted
terpyridine
ma-T
coligand
λAbs, nm
λem, nm
497, 527
τ, μs 298 K
Φ 298 K
kr, 104 s
τ1, μs (f1) 77 K
CN
383, 399
417, 432sh
435, 464sh
395, 415
420, 444
436, 457sh
402, 428
423, 447
440, 466
1.0
0.24
4.8
22
0.025
0.013
0.12
2.5
5.4
2.5
1.2
5.8
3.6
0.63
1.8
0.90
Cl
532, 562
576
14.8 (0.65)
12.9 (0.58)
CCPh
CN16
Cl
CCPh34
CN
dma-T
pyrr-T
524, 552
535, 570, 635sh
567
0.26
1.9
5.6
24
0.11
18.6 (0.73)
15.8 (0.76)
0.20
524, 557sh
548, 583, 636sh
567
0.15
Cl
CCPh13
4.5
0.079
17.9 (0.81)
17.5 (0.74)
a
a
8.1
0.073
a
Solvent was MeCN.
treating 0.500 g (1.9 mmol) of Cl-T with 6.0 mL of N-
methylformamide and 0.526 g (5.0 mmol) of diethanolamine for 72
h at 150 °C in a thick-walled glass pressure vessel. Extracting the
residue into DCM and treating with 0.470 g (2.1 mmol) of zinc
bromide in methanol yielded Zn(ma-T)Br2 as a solid after the addition
of ether. Treatment of the zinc complex in 250 mL of DCM with 100
mL of 1.5 M NaOH(aq) resulted in the extraction of Zn(II) into the
aqueous phase. Evaporation of the organic layer yielded the free ligand.
1H NMR (300 MHz, CDCl3): δ 8.69 (d, 2H), 8.6 (d, 2H), 7.82 (t,
2H), 7.63 (s, 2H), 7.3 (t, 2H), 4.5 (s,br, 1H), 3.0 (s, 3H). MS-ESI: (m
+1)/z 262.80.
wavelength was typically that of the CT absorption maximum or,
rarely when the dye laser was not functioning, the intrinsic wavelength
of the N2 laser (337 nm). A previously described method allowed for
extracting lifetimes from the decay profiles.20
Deuterium Exchange Studies. The addition of a small excess of
DCl in CH3OD to a solution of [Pt(4′-X-T)CN]Cl dissolved in the
same solvent eventually yielded a yellow precipitate. Exposing the
isolated solid to slightly less than one equivalent of Na[TFPB] in
DCM allowed for extraction of the complex into solution. Filtration at
room temperature and then a second filtration after volume reduction
and cooling yielded a solution containing a deuterated complex. A
comparison with the mass spectrum obtained with that of a control
solution treated similarly with HCl in CH3OH instead of DCl
established that the complex exchanged one deuterium atom upon
exposure to acid in CH3OD. Because a second control experiment
involving [Pt(pyrr-T)CN]+ showed no evidence of deuterium
incorporation, there is little doubt but that exchange occurred at the
intended N−H position.
[Pt(ma-T)Y]OTf. The method reported by Annibale et al. yielded
[Pt(ma-T)Cl]OTf from the reaction of the free ligand with
Pt(COD)Cl2.15 Replacement of chloride by cyanide as the coligand
was possible using a published double-exchange method.16 The
addition of tetrabutylammonium chloride to a solution of either triflate
salt in acetone provided for isolation of the corresponding chloride
1
salts. [Pt(ma-T)Cl]+. H NMR (300 MHz, (CD3)2SO): δ 8.48 (d,
Instrumentation. The absorption spectrophotometer was a Varian
Cary 100 instrument, and the spectrofluorimeter was a Varian Cary
Eclipse. The ESI mass spectrometer was a Waters Micromass ZQ
instrument. A previous publication describes the nitrogen-pumped dye
laser and signal-acquisition equipment used for the excited-state
lifetime determinations.21 An Oxford Instruments DN1704 liquid-
nitrogen-cooled system complete with an Oxford Instruments
temperature controller was available for variable-temperature emission
and lifetime measurements. The NMR spectrometer was a Varian
Inova300. A CHI620A voltammetric analyzer served to record cyclic
voltammograms. All electrochemistry involved a platinum disc working
electrode, a silver/silver chloride in 3 M aqueous sodium chloride
reference electrode, and a platinum wire as the auxiliary electrode.
Crystal Structures. A LINUX PC platform using SHELX-97
performed the refinements for crystallographic studies.22 The
diffractometers used were a Rigaku Rapid II equipped with confocal
optics and a Nonius KappaCCD equipped with a graphite crystal,
incident beam monochromator. The crystal of [Pt(ma-T)CN]-
OTF·MeCN selected for structure determination was a red plate
having approximate dimensions of 0.13 × 0.10 × 0.04 mm. As
determined by Rigaku software product CrystalClear, the refined
mosaicity was 1.05°, indicating poor crystal quality. The program used
to determine the space group was XPREP.22 The data collection was
to a maximum 2θ of 133.2°. The solution program PATTY in
DIRDIF99 solved the structure.23 The refinement included hydrogen
atoms, but the constraint was that each remain bound to a designated
atom. Of the total of 9094 reflections collected, 3425 were unique and
useful in the refinement; however, the calculation of R1 used only the
2H), 8.32 (t, 2H), 8.01 (broad, 2H), 7.72 (t, 2H), 7.26 (broad, 2H),
2.88 (d, 3H), N−H resonance not resolved. The preparation of
[Pt(ma-T)CCPh]PF6 followed the method of Yang et al.17 It involves
charging 0.048 g (0.075 mmol) of [Pt(ma-T)Cl]PF6 and 0.028 g (0.14
mmol) of CuI into a round-bottom flask and purging with N2(g). The
next step is to add 0.030 g (0.3 mmol) of phenylacetylene and 20 mL
of dimethylformamide (DMF) while maintaining a nitrogen
atmosphere. The addition of 0.030 g (0.3 mmol) of triethylamine
promotes the reaction. After the reaction is over, the addition of excess
KPF6(aq) deposits the desired complex. Subsequent crystallization is
possible by slow diffusion of ether into an acetonitrile solution of the
complex.
[Pt(ma-T)CN]OTf·CH3 OH. Analysis Calculated for
C19H18F3N5O4PtS: C, 34.34; H, 2.73; N, 10.54%. Found: C, 34.53:
H, 2.44; N, 10.38%.
[Pt(ma-T)CCPh]PF6. Analysis Calculated for C24H19F6N4PPt: C,
40.98; H, 2.72; N, 7.96%. Found: C, 41.00: H, 2.87; N, 8.27%. [Pt(ma-
1
T)CCPh]+. H NMR (300 MHz, (CD3)2SO): δ 8.87 (d, 2H), 8.50
(broad, 2H), 8.33 (t, 2H), 8.06 (broad, 2H), 7.73 (d, 2H), 7.42 (m,
2H), 7.34 (t, 2H), 7.28 (m,1H), 2.91 (s, br, 3H), N−H resonance not
resolved.
Spectral Measurements. In most cases, measurement of room-
temperature absorption, luminescence, and excitation spectra followed
after extraction of the complex into dichloromethane with sodium
tetrakis(3,5-trifluoromethylphenyl)borate (Na[TFPB]) as a solubility
agent. The slit settings were 20 nm for all luminescence spectra
obtained in DCM and 10 nm for excitation measurements, while an
appropriate long-pass filter protected the detector from stray excitation
light. The method of Parker and Rees18 permitted determination of
quantum yields with [Ru(bpy)3]2+ in deoxygenated acetonitrile as the
standard (φ = 0.062).19 The manufacturer provided instrumental
emission intensity correction factors for the fluorimeter. Bubbling the
samples with nitrogen for 20 min in a cuvette with a septum seal
served to remove oxygen prior to quantum yield and lifetime
determinations. Samples for cryostat measurements underwent four
freeze−pump−thaw cycles. For lifetime measurements, the excitation
2
2
2823 reflections with Fo > 2σ(Fo ). Table S1 summarizes the
crystallographic data.
The crystal of [Pt(ma-T)Cl]OTF was a red needle of approximate
dimensions of 0.63 × 0.15 × 0.10 mm. The program DENZO/
SCALEPACK refined the mosaicity to be 0.95°, indicating moderate
crystal quality, and the program XPREP established the space group.
The method for solution of the structure was the same as above. Here,
there were a total of 8698 reflections, of which 4084 were unique, and
8477
dx.doi.org/10.1021/ic4004643 | Inorg. Chem. 2013, 52, 8476−8482

Inorganic Chemistry
Article
2
2
the 3867 reflections with Fo > 2σ(Fo ) contributed to the calculation
of R1. See Table S1 for more results.
longer wavelength according to the strength of the electron-
donating group on the trpy ligand, i.e., in the order ma-T <
dma-T < pyrr-T. In the case of the CCPh complexes, however,
the influence of coligand appears to be dominant because
varying the trpy ligand has little effect on the emission
maximum.
On the other hand, when the trpy ligand is constant, the
emission maximum shifts to a longer wavelength as the
coligand ranges from CN < Cl < CCPh, in keeping with the
trend observed in the CT absorption bands. Data compiled in
Table 1 reveal that [Pt(ma-T)Cl]+ is the only complex in the
series that exhibits a submicrosecond excited-state lifetime in
deoxygenated DCM solution. The [Pt(ma-T)Cl]+ complex also
exhibits the lowest emission quantum yield of about 1.3%. At
lower temperatures, however, the lifetime of [Pt(ma-T)Cl]+
lengthens considerably. Equation 1 models the behavior with
ΔEa = 1830 cm−1 and A = 2.7 × 1010 s−1.
Computational Details. Geometry optimization and adiabatic
time-dependent density functional theory (TDDFT) excitation
calculations are at the B3LYP level.24,25 The quasi-relativistic
Stuttgart/Dresden effective core potential MWB6026 and its
corresponding basis set apply to Pt, and the 6-311G**basis set
applies to other atoms.27 The optimized geometries of the first triplet
excited states of dma-T and ma-T are obtained from spin-unrestricted
SCF calculations. TDDFT calculations in these optimized geometries
employ restricted Kohn−Sham references to facilitate comparison with
the emission measurements. The polarizable continuum model
(PCM) accounts for solvent effects.28−30 A c-squared population
analysis provided percentage contributions of fragmental AOs in the
MOs.31 The quantum chemistry package Gaussian 09 supplied the
software for all calculations.32
RESULTS
■
Absorption and Emission Measurements. In keeping
with results obtained with other platinum(II) polypyridine
complexes,2,3,33 the complexes studied herein all exhibit mainly
intraligand absorptions at wavelengths below about 375 nm and
charge-transfer (CT) absorptions at longer wavelengths, in the
range of 375−475 nm. Absorption data compiled in Table 1
reveal that the energies of the CT band maxima vary with the
4′-amine substituent present on the trpy ligand, where trpy
denotes 2,2′:6′,2″-terpyridine, as well as the coligand bound to
platinum. Early experiments revealed that the counterion also
influences the absorption spectrum of [Pt(ma-T)Cl]+ in DCM
solution. Figure S1 in the Supporting Information shows that
the CT absorption exhibits more structure and occurs at
slightly shorter wavelengths when the counterion is TFPB as
opposed to the triflate ion. As the emission also shifts to shorter
wavelengths when TFPB is present, it became the counteranion
of choice for studies carried out in DCM.
The room-temperature absorption and emission spectra in
Figure 1 pertain to [Pt(ma-T)CN]+, [Pt(dma-T)CN]+, and
[Pt(ma-T)CCPh]+, representative complexes from the series.
As has been the case in previous studies,2,34 partially resolved
vibronic structure is sometimes evident in the CT emission
spectrum, except if the coligand is CCPh. The results in Table 1
reveal that when the coligand is constant, the emission shifts to
τ−1 = τ0−1 + A e−ΔE /kT
a
(1)
Both the [Pt(dma-T)CN]+ and [Pt(pyrr-T)CN]+ systems
exhibit much longer room-temperature excited state lifetimes of
22 and 24 μs, respectively. The emission quantum yields are
correspondingly higher, too. At platinum concentrations of less
than about 10 μM, self-quenching is minimal, particularly when
the lifetime is relatively short to begin with. However, at higher
concentrations self-quenching is evident with [Pt(pyrr-T)-
CCPh]+, where kq = 1.8 × 109 M−1 s−1. N−H deuteration has
no measurable impact on the lifetime of [Pt(ma-T)CN]+. In a
rigid glass at 77 K, the photoexcited states of the complexes
under investigation tend to exhibit biphasic decay, as has been
reported previously.34 Table 1 includes the lifetime of the
dominant, longer-lived component of the decay observed in a
DME glass, along with the fractional contribution to the total
signal amplitude. See the Supporting Information (Table S2)
for more details.
Data in the text of the Supporting Information confirm that
electron-donating substituents shift ligand-based reduction in
the cathodic direction, as reported for related systems.35 The
presence of an amine substitutent shifts the first reduction
potential by about 280 mV more negative, while the shift is
about half as large for the second reduction.
Crystal Structure of [Pt(ma-T)Cl]OTf. Figure 2 is a
thermal-ellipsoid view of the complex complete with an atom-
numbering scheme. The caption provides a list of bond angles
and bond distances which are similar to those of related
platinum terpyridine complexes.2,3,36 In terms of the methyl-
amino substituent, the C17−N19−C20 plane makes a dihedral
angle of 0.58° with respect to the best plane through the carbon
and nitrogen atoms of the trpy ligand. The corresponding
dihedral angle is 5.9° in a related dma-T derivative.12 The
chloride complexes stack as dimers along the b direction with a
Pt···Pt spacing of 3.86 Å, too long for significant metal−metal
interaction.36,37 As in a [Pt(dma−T)CN]+ structure,12 the
packing is not exactly head to tail, as the Cl−Pt−Pt−Cl torsion
angle is 125.22° rather than 180°. Within a dimer, the mean
planes through the platinum complexes are, however, mutually
parallel.
Crystal Structure of [Pt(ma-T)CN]OTf·CH3CN. A ther-
mal-ellipsoid view of the complex along with an anion appears
in Figure 3, which also reveals the atom-numbering scheme.
The caption includes a list of selected bond angles and bond
lengths as before. The dihedral angle relating the best plane
through the carbon and nitrogen atoms of the trpy ligand and
Figure 1. Absorption and emission spectra of representative [Pt(4′-X-
T)Y]+ complexes in DCM. Absorption spectra of [Pt(ma-T)CN]+
(thin), [Pt(dma-T)CN]+ (dot dash), and [Pt(ma-T)CCPh]+ (thick).
Corrected emission spectra (red) of [Pt(ma-T)CN]+ (thin, dashed),
[Pt(dma-T)CN]+ (dot dash), and [Pt(ma-T)CCPh]+ (thick, dashed).
Areas of emission signals reflect relative quantum yields.
8478
dx.doi.org/10.1021/ic4004643 | Inorg. Chem. 2013, 52, 8476−8482

Inorganic Chemistry
Article
Figure 2. The molecular structure of a representative cation of
[Pt(ma-T)Cl]OTf with thermal ellipsoids set at 50% probability.
Selected bond lengths (Å) and angles (deg) associated with the metal
center: N1−Pt1 2.061(70), N8−Pt1 1.920(11), N15−Pt1 2.019(14),
Pt1−Cl1 2.310(3); N1−Pt1−N8 80.2(5), N8−Pt1−Cl1 178.4(3),
N1−Pt1−Cl1 99.5(3), N1−Pt1−N15 161.1(5), N8−Pt1−N15
81.0(5), N15−Pt1−Cl1 99.3(4).
Figure 4. Calculated properties of the frontier molecular orbitals of
[Pt(dma T)CN]+ (left) and [Pt(ma T)CN]+ (right) in S0 geometries.
Energies in parentheses are in au.
rotation about the 2-fold axis that passes through the ligand.38
Here, the symmetry is only approximate because, for example,
the dma group is not exactly in the plane of the trpy ligand;
nevertheless, the HOMO of each ligand is essentially a ψ-type
orbital. The HOMO mainly localizes on the inner pyridine of
the trpy ligand, with significant contributions also coming from
the dma nitrogen, the platinum 5d(xz), and the 2p(z) orbital of
the cyanide nitrogen. The LUMO also has ψ symmetry but
largely delocalizes on the outer pyridines of the trpy ligand. The
platinum 5d(xz) and 6p(z) orbitals also contribute to the
LUMO, consistent with EHMO calculations of Collison et al.39
Hill et al. have also rationalized the electrochemistry of
platinum(II) terpyridines in terms of platinum 6p(z)
participation.40 The LUMO+1 has χ symmetry, delocalizes
over the whole trpy moiety, and includes some platinum 5d(yz)
character. The two states selected for geometry optimization
were the ground state S0 and the lowest energy triplet state T1.
Tables 2 and 3 summarize calculated framework contributions
to key molecular orbitals of the low-lying singlet and triplet
Figure 3. The molecular structure of a representative cation−anion
pair of [Pt(ma-T)CN]OTf·MeCN with thermal ellipsoids set at 50%
probability. Selected bond lengths (Å) and angles (deg) associated
with the metal center: N1−Pt1 2.037(15), N12−Pt1 1.961(12), N17−
Pt1 2.090(15), Pt1−C1 1.87(2); C1−N2 1.26(2); N1−Pt1−N12
78.8(5), N12−Pt1−C1 177.1(6), N1−Pt1−C1 98.4(6), N1−Pt1−
N17 160.9(5), N12−Pt1−N17 82.1(5), N17−Pt1−C1 100.7(6); N2−
C1−Pt1 175.3(15).
the plane that passes through C9, N201, and C201 is 4.1°.
Here, the platinum complexes also stack as dimers, but along
the c axis with a Pt···Pt spacing of 3.67 Å, and the C1−Pt−Pt−
C1 torsion angle is 129.08°. Another small difference is that the
mean planes through the paired platinum complexes are not
quite parallel and intersect at an angle of 1.3°. The distance
between N201 of the methylamino group and O93 of the
triflate anion is 2.909 Å.
MO Calculations. DFT calculations show that the
molecular orbitals of the ground-states generally delocalize
over the entire framework, though some orbitals tend to
localize in specific domains (Figure 4). For convenience,
assume that the axis passing through platinum and the cyanide
defines the x direction and that the z axis is perpendicular to
the plane of the trpy ligand. The orbital makeup proves to be
very similar in [Pt(dma-T)CN]+ and [Pt(ma-T)CN]+. In each
case, the 5d(z2) atomic orbital of platinum effectively defines
the HOMO−1. As regards the π orbitals of polypyridine
ligands, it is convenient to label them as χ or ψ if they are,
respectively, symmetric or antisymmetric with respect to
Table 2. Percentage Contribution of Fragmental AOs in
MOs of [Pt(dma-T)CN]+ and [Pt(ma-T)CN]+ Complexes in
Their S0 States
complex
fragment HOMO−1 HOMO LUMO LUMO+1
% Pt
96.3
0.6
23.7
3.1
5.4
0.9
1.7
0.0
[Pt(dma-T)CN]+
% CN
% dma
% trpy
% Pt
0.0
41.7
31.5
29.8
4.1
5.8
1.1
3.1
87.9
5.3
97.2
1.8
96.3
0.6
[Pt(ma-T)CN]+
% CN
% ma
% trpy
0.9
0.0
0.1
33.3
32.8
5.0
0.4
3.0
88.8
97.8
8479
dx.doi.org/10.1021/ic4004643 | Inorg. Chem. 2013, 52, 8476−8482

Inorganic Chemistry
Article
excited states. See Table S3 for calculated excited-state
properties.
ation is, therefore, that as the trpy ligand shifts from ma-T to
dma-T to pyrr-T, ILCT character increases and the CT
absorption shifts to a longer wavelength.
Variations in orbital parentage similarly affect the emission
results, although the energy differences between states tend to
be smaller (Table 1). Pauli-exchange effects help shape the
electron distribution in the triplet state, and as a result, the
unpaired electrons occupy complementary orbitals delocalized
to differing degrees over both outer pyridine moieties. (See the
SOMO (singly occupied molecular orbital) contours in the
Supporting Information, Figures S2 and S3.) One of the
SOMOs utilizes the 5d(xz) orbital of platinum, while the other
incorporates 5d(yz). It is also evident that orbitals from a
−NHMe or −NMe2 substituent make relatively modest
contributions to the SOMOs (Table 3). Nevertheless,
experimental results reveal that electron-donating moieties
impact the emission. The apt comparison is [Pt(trpy)CN]+,
which exhibits a 0−0 emission band at 465 nm and
Table 3. Percentage Contribution of Fragmental AOs in
MOs of the T1 States of [Pt(dma-T)CN]+ and [Pt(ma-
T)CN]+ Complexes (MS = 1)
complex
fragment
SOMO1
SOMO2
% Pt
42.6
1.8
4.9
0.7
[Pt(dma-T)CN]+
% CN
% dma
% trpy
% Pt
0.7
5.7
54.9
33.3
1.2
88.7
4.3
[Pt(ma-T)CN]+
% CN
% ma
% trpy
0.6
0.2
5.6
65.3
89.5
3
DISCUSSION
accompanying vibronic structure that is diagnostic of π−π*
■
orbital parentage.16 In contrast, the emissions from the
modified forms found in Table 1 begin at significantly longer
wavelengths of 500 nm or beyond. The signals also
progressively shift to a longer wavelength as LLCT character
increases with a change of coligand (CN < Cl < CCPh). When
the coligand is Cl or CN, the emission likewise shifts to a
longer wavelength as the electron-donating ability of the 4′-
substituent increases (ma-T < dma-T < pyrr-T). LLCT
character may dominate when the coligand is CCPh, because
the emission spectra maximize at about the same wavelength,
irrespective of the 4′-substituent.
Both ma-T structures described above entail similar face-
centered, orthorhombic packing and exhibit unusual, non-head-
to-tail stacking of the platinum complexes. Because the X-ray
data do not permit resolution of the hydrogen atoms, the
geometry at the substituent nitrogen is not entirely certain in
either ma-T complex. Nevertheless, it seems clear that the
hybridization is effectively sp2, in keeping with reported
structures involving the analogous [Pt(dma-T)CN]+ and
[Pt(pyrr-T)CN]+ complexes.12 Also, consistent with a planar
substituent, the distance of 2.909 Å between the N201 nitrogen
of the methylamino group and oxygen O93 of the triflate anion
(shown in Figure 3) is compatible with N−H···O hydrogen-
bond formation.41,42
Emission Decay. The emission energy is also a key factor
influencing the excited state lifetime and emission quantum
yield because the state energy impacts important rate processes.
Figure 5 shows the pathways that affect the dynamics of the
CT Absorption and Emission. Due to the energy
requirement for the third ionization of platinum, “pure”
MLCT absorption is much less likely in polypyridine complexes
of platinum(II) by comparison with related Fe(II), Ru(II), or
Cu(I) complexes.2 Accordingly, charge-transfer excitations in
platinum complexes typically entail accompanying admixtures
of intraligand (ILCT16,35) and/or interligand (LLCT43,44
)
character. The CT absorption maxima of the ma-T complexes
shift toward lower energies as the coligand ranges from CN to
Cl to CCPh, and the same is true for the other trpy derivatives
(Table 1). The trend follows the π-donating ability (CN < Cl <
CCPh) as revealed by the ionization energies of HCN, HCl,
and HCCPh, which range from 13.6 to 12.7 to 8.8 eV,
respectively.45 Alternatively, data from Table 2 and Ji et al.
confirm that coligand participation in the HOMO increases
from 3.1% to 41.2% to 85% as the coligand changes from CN
to Cl to CCPh.34 Both arguments suggest that the CT
excitation shifts toward lower energy as LLCT character
becomes more important.
The impact of ILCT character is apparent from the shift of
the wavelength of the absorption with a change of the 4′-
substituent on the trpy ligand. Consider the [Pt(4′-X-T)CN]+
series, where the CT absorption maximum is 383, 395, or 402
nm if the substituent is methylamino, dimethylamino, or
pyrrolidinyl, respectively. The greater electron-donating ability
of the −NR2 group as compared with the −NHR group
registers in the HOMO in that the substituent participation
increases from 33.3% to 41.7% with the addition of a second
methyl group (Table 2). The pyrrolidinyl group also involves
two R groups, but there is experimental evidence that the ring
structure further enhances the donor strength.46 The expect-
Figure 5. Decay pathways for a platinum(II) polypyridine complex
with an emissive CT excited state. Radiationless decay occurs directly
to the ground state with rate constant kn or via a thermally accessible
d−d state with rate constant kn′. The rate constant for the radiative
pathway is kr.
emitting triplet state. Rate constant kr signifies the formally
spin-forbidden radiative process, which is actually partially
allowed by virtue of spin−orbit coupling. Of course, any
diminution of heavy metal character will weaken spin−orbit
coupling and lower the emission quantum yield. Pathway kn′ in
Figure 5 accounts for nonradiative decay via a thermally
accessible d−d excited state. The kn′ process can be very
efficient and is apparently responsible for the absence of
measurable emission from many platinum(II) terpyridine
complexes, including [Pt(trpy)Cl]+, in fluid solution.3,11,33,47
This pathway becomes less important when the ligand
environment destabilizes the dσ* metal−ligand antibonding
8480
dx.doi.org/10.1021/ic4004643 | Inorg. Chem. 2013, 52, 8476−8482

Inorganic Chemistry
Article
orbital and elevates the barrier (ΔEa) against quenching. Direct
nonradiative decay to the ground state occurs by the kn
pathway.
however, remarkable. In the absence of an electron-donating
group in the 4′-position, the latter complex exhibits a 10-fold
shorter room-temperature, excited-state lifetime.36
Factors that influence kn include the energy-gap law, which
predicts that the rate of nonradiative decay increases
exponentially as the energy gap narrows.48,49 In this limit,
high frequency vibrations, such as N−H vibrations, sometimes
enhance the rate of nonradiative decay by acting as good
acceptor modes in the ground electronic state.50 Finally, a
different effect comes into play if formation of the electronic
excited state entails a significant change in the equilibrium
geometry, and the assumption of nested potential energy
surfaces no longer applies. The coupling to the ground-state
surface then becomes much stronger, and the rate of
nonradiative decay increases accordingly.51,52
The lifetime in combination with the emission quantum yield
provides an estimate of the radiative rate constant, kr. Many
phenomena contribute to each term, and no simple
interpretation explains all the kr data. Nevertheless, some
patterns emerge. The first is that when the coligand is CN, kr
decreases as the terpyridine ranges in the order: ma-T > dma-T
> pyrr-T. The explanation may simply be that the emission
probability decreases as the excitation becomes more ligand
centered.53 Indeed, the pyrr-T complex always exhibits the
smallest kr value regardless of the coligand (Table 1). Another
trend is that the kr value is always highest when the coligand is
Cl. Here, the moderately high atomic number of the chlorine
atom and/or its proximity to platinum may account for the
elevated kr values.
All of the decay constants critically depend on the orbital
parentage of the emitting state. In particular, studies have
shown that introduction of ILCT character has the effect of
reducing the rate of thermally activated decay via higher energy
d−d states and thereby increasing the excited-state life-
time.13,34,35 Consistent with this expectation, when the coligand
is Cl or CN, the emission shifts to lower energy, and the
lifetime increases with the electron-donating ability of the 4′-
substituent (ma-T < dma-T < pyrr-T). (A discussion of systems
with CCPh as coligand follows below.) Systems based on the
ma-T ligand are potentially special cases because the N−H
vibration could play an important role in the relaxation process.
However, deuteration of the amine group has no measurable
effect on the lifetime of [Pt(ma-T)CN]+. In the case of [Pt(ma-
T)Cl]+, which has the shortest excited-state lifetime in Table 1,
the temperature dependence of the lifetime reveals that a
thermally accessible excited state dramatically accelerates
radiationless decay. This d−d state occurs some 1830 cm−1
above the emitting state. Data in the Supporting Information
(Figure S4) reveal that the activation barrier is much higher in
[Pt(ma-T)CN]+ due to its relatively strong field coligand.
Additional evidence in support of this model comes from the
results obtained in a frozen glass at 77 K where population of a
higher energy d−d state is no longer feasible (Figure S5). Thus,
the principal component observed in the decay of [Pt(ma-
T)Cl]+ exhibits a lifetime of 14.8 μs, actually surpassing the
corresponding lifetime (12.9 μs) observed from [Pt(ma-
T)CCPh]+, the ma-T complex that exhibits the longest lifetime
in fluid solution (Table 1). Indeed, independent of the 4′-X-T
ligand, the lifetime is longer at 77 K with Cl as opposed to
CCPh as a coligand, in keeping with expectations based on
emission energy considerations.
The three complexes with CCPh as the coligand are
noteworthy because they all exhibit microsecond lifetimes at
room temperature. The reason is an acetylide coligand naturally
enhances the ΔEa barrier to deactivation in two ways. As a
strong σ donor, it drives up the energy of d−d states.36 At the
same time, it introduces LLCT character, which also enhances
ΔEa by lowering the energy of the emitting state.44
Complementing the CCPh ligand with an amine-substituted
4′-X-T ligand further bolsters the lifetime by introducing ILCT
character into the emitting state.13,34 Accordingly, the room
temperature lifetime increases steadily as the donor strength of
the substituent builds from methylamino to dimethylamino to
pyrrolidinyl. The overall change in lifetime is less than a factor
of 2 within the series, however, so relaxation via the d−d state
represents a minor pathway for this subset of complexes. The
contrast between these systems and [Pt(trpy)CCPh]+ is,
CONCLUSIONS
■
Simple [Pt(4′-X-T)Y]+ complexes rarely exhibit strong
emission signals in fluid solution because of strain in the
metal−ligand σ bond framework and the availability of
thermally accessible d−d excited states that quench potentially
emissive excited states. The simple expedient of incorporating a
4′-amine-substituted-terpyridine ligand can dramatically en-
hance the emission quantum yield by delocalizing the hole away
from platinum, introducing ILCT character into the emitting
state, and effectively diminishing the role of d−d states. Tuning
the amine group is easy as the π-donating strength increases in
the order methylamino < dimethylamino < pyrrolidinyl. The
results also confirm that incorporating an easily ionized
coligand such as CCPh further lowers the energy of the
emitting state by introducing LLCT character. However, at
some point, direct vibronic coupling to the ground-state energy
surface becomes important, because of the energy gap law and/
or structural reorganization induced by formal oxidation of the
coligand.2,13,54 Incorporating strong field cyanide as the
coligand is usually a more attractive option because it supports
higher energy CT emission while still suppressing the influence
of d−d states. Thus, within the series investigated, [Pt(pyrr-
T)CN]+ exhibits the longest lifetime in fluid solution. The
other factor to consider is the radiative rate constant because
[Pt(dma-T)CN]+ exhibits almost twice the emission quantum
yield, despite having a slightly shorter excited-state lifetime.
ASSOCIATED CONTENT
■
S
* Supporting Information
Contour plots and density profiles of the SOMOs of the T1
states of [Pt(dma-T)CN]+ and [Pt(ma-T)CN]+, as well as state
energies, along with X-ray data, additional absorption and
emission (77 K) data, electrochemical potentials, and temper-
ature dependent studies of emission decay. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: mcmillin@purdue.edu.
Notes
The authors declare no competing financial interest.
8481
dx.doi.org/10.1021/ic4004643 | Inorg. Chem. 2013, 52, 8476−8482

Inorganic Chemistry
Article
(32) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.
Gaussian 09; Gaussian, Inc.: Wallingford, CT, 2009.
(33) Aldridge, T. K.; Stacy, E. M.; McMillin, D. R. Inorg. Chem. 1994,
33, 722−727.
ACKNOWLEDGMENTS
■
The NSF funded this research via grants CHE 0847229
(D.R.M.) and CHE 1149968 (A.W.). The authors are grateful
to Dr. Tong Ren for the use of electrochemical equipment and
to Julia Savchenko for technical assistance. Dr. David Stewart
assisted with the double-exponential fits.
REFERENCES
■
(1) Chan, C. W.; Cheng, L. K.; Che, C. M. Coord. Chem. Rev. 1994,
132, 87−97.
(34) Ji, Z. Q.; Azenkeng, A.; Hoffmann, M.; Sun, W. F. Dalton Trans.
2009, 7725−7733.
(2) McGuire, R.; McGuire, M. C.; McMillin, D. R. Coord. Chem. Rev.
2010, 254, 2574−2583.
(35) Crites, D. K.; Cunningham, C. T.; McMillin, D. R. Inorg. Chim.
Acta 1998, 273, 346−353.
(36) Yam, V. W. W.; Tang, R. P. L.; Wong, K. M. C.; Cheung, K. K.
Organometallics 2001, 20, 4476−4482.
(37) Martin, D. S. Adv. Chem. Ser. 1974, 254−275.
(38) Orgel, L. E. J. Chem. Soc. 1961, 3683−3686.
(39) Collison, D.; Mabbs, F. E.; McInnes, E. J. L.; Taylor, K. J.;
Welch, A. J.; Yellowlees, L. J. J. Chem. Soc., Dalton Trans. 1996, 329−
334.
(3) Eryazici, I.; Moorefield, C. N.; Newkome, G. R. Chem. Rev. 2008,
108, 1834−1895.
(4) Williams, J. A. G. Top. Curr. Chem. 2007, 281, 205−268.
(5) Houlding, V. H.; Miskowski, V. M. Coord. Chem. Rev. 1991, 111,
145−152.
(6) Drew, S. M.; Janzen, D. E.; Buss, C. E.; MacEwan, D. I.; Dublin,
K. M.; Mann, K. R. J. Am. Chem. Soc. 2001, 123, 8414−8415.
(7) Castellano, F. N.; Pomestchenko, I. E.; Shikhova, E.; Hua, F.;
Muro, M. L.; Rajapakse, N. Coord. Chem. Rev. 2006, 250, 1819−1828.
(8) Rivera, E. J.; Barbosa, C.; Torres, R.; Grove, L.; Taylor, S.;
Connick, W. B.; Clearfield, A.; Colon, J. L. J. Mater. Chem. 2011, 21,
15899−15902.
(40) Hill, M. G.; Bailey, J. A.; Miskowski, V. M.; Gray, H. B. Inorg.
Chem. 1996, 35, 4585−4590.
(41) Zucchi, G.; Scopelliti, R.; Bunzli, J. C. G. J. Chem. Soc., Dalton
Trans. 2001, 1975−1985.
(9) Connick, W. B.; Geiger, D.; Eisenberg, R. Inorg. Chem. 1999, 38,
3264−3265.
(42) Schiessl, W.; Puchta, R.; Bugarcic, Z. D.; Heinemann, F. W.; van
Eldik, R. Eur. J. Inorg. Chem. 2007, 1390−1404.
(10) Tears, D. K. C.; McMillin, D. R. Coord. Chem. Rev. 2001, 211,
195−205.
(43) Bevilacqua, J. M.; Eisenberg, R. Inorg. Chem. 1994, 33, 2913−
2923.
(11) McMillin, D. R.; Moore, J. J. Coord. Chem. Rev. 2002, 229, 113−
121.
(44) Zhou, X.; Zhang, H. X.; Pan, Q. J.; Xia, B. H.; Tang, A. C. J.
Phys. Chem. A 2005, 109, 8809−8818.
(45) Lias, S. G.; Liebman, J. F.; Linstrom, P. J., Mallard, W. G. NIST
Chemistry WebBook, NIST Standard Reference Database Number 69;
National Institute of Standards and Technology: Gathersburg, MD,
retrieved January, 2013.
(12) Clark, M. L.; Green, R. L.; Johnson, O. E.; Fanwick, P. E.;
McMillin, D. R. Inorg. Chem. 2008, 47, 9410−9418.
(13) Lazzaro, D. P.; Fanwick, P. E.; McMillin, D. R. Inorg. Chem.
2012, 51, 10474−10476.
(14) Cho, Y. H.; Park, J. C. Tetrahedron Lett. 1997, 38, 8331−8334.
(15) Annibale, G.; Pitteri, B.; Wilson, M. H.; McMillin, D. Inorg.
Synth. 2004, 34, 76−81.
(46) Spivey, A. C.; Arseniyadis, S. Angew. Chem., Int. Ed. 2004, 43,
5436−5441.
(47) Yip, H. K.; Cheng, L. K.; Cheung, K. K.; Che, C. M. J. Chem.
Soc., Dalton Trans. 1993, 2933−2938.
(16) Wilson, M. H.; Ledwaba, L. P.; Field, J. S.; McMillin, D. R.
Dalton Trans. 2005, 2754−2759.
(48) Englman, R.; Jortner, J. Mol. Phys. 1970, 18, 145−164.
(49) Caspar, J. V.; Kober, E. M.; Sullivan, B. P.; Meyer, T. J. J. Am.
Chem. Soc. 1982, 104, 630−632.
(17) Yang, Q. Z.; Wu, L. Z.; Wu, Z. X.; Zhang, L. P.; Tung, C. H.
Inorg. Chem. 2002, 41, 5653−5655.
(18) Parker, C. A.; Rees, W. T. Analyst (Cambridge, U. K.) 1960, 85,
587−600.
(50) Turro, N. J.; Ramamurthy, V.; Scaiano, J. C. Principles of
Molecular Photochemistry: An Introduction; University Science Books:
Sausalito, CA, 2009.
(19) Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583−
5590.
(51) Gelbart, W. M.; Rice, S. A.; Freed, K. F. J. Chem. Phys. 1970, 52,
5718−5732.
(20) Liu, F.; Cunningham, K. L.; Uphues, W.; Fink, G. W.; Schmolt,
J.; McMillin, D. R. Inorg. Chem. 1995, 34, 2015−2018.
(21) Cunningham, K. L.; Hecker, C. R.; McMillin, D. R. Inorg. Chim.
Acta 1996, 242, 143−147.
(22) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112−122.
(23) Beurskens, P. T.; Beurskens, G.; deGelder, R.; Garcia-Granda,
S.; Gould, R. O.; Smits, J. M. M. The DIRDIF2008 Program System;
Crystallography Laboratory, Univ. of Nijmegen: The Netherlands,
2008.
(52) DeArmond, M. K.; Hillis, J. E. J. Chem. Phys. 1971, 54, 2247−
2253.
(53) A theoretical analysis would depend on higher order calculations
involving spin−orbit coupling, which is beyond the scope of this work.
(54) Clark, M. L.; Diring, S.; Retailleau, P.; McMillin, D. R.; Ziessel,
R. Chem.Eur. J. 2008, 14, 7168−7179.
(24) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(25) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623−11627.
(26) Andrae, D.; Haussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123−141.
(27) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys.
1980, 72, 650−654.
(28) Cances, E.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107,
3032−3041.
(29) Cossi, M.; Barone, V.; Mennucci, B.; Tomasi, J. Chem. Phys. Lett.
1998, 286, 253−260.
(30) Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 106, 5151−5158.
(31) Ros, P.; Schuit, G. C. A. Theor. Chim. Acta 1966, 4, 1−12.
8482
dx.doi.org/10.1021/ic4004643 | Inorg. Chem. 2013, 52, 8476−8482