Intramolecular triplet energy transfer in pyrene-metal polypyridine dyads: A strategy for extending the triplet lifetime of the metal complex

DOI: 10.1002/(SICI)1521-3765(19991105)5:11<3366::AID-CHEM3366>3.0.CO;2-I

Source and publish data:

Chemistry - A European Journal p. 3366 - 3381 (1999)

Update date:2022-08-10

Topics:

Authors:

Hissler, Muriel

Harriman, Anthony

Khatyr, Abderrahim

Ziessel, Raymond

Read Full Text PDF DownLoad Join now for total 90,000,000 free articles

Article abstract of DOI:10.1002/(SICI)1521-3765(19991105)5:11<3366::AID-CHEM3366>3.0.CO;2-I

A series of photoactive dyads bearing pyrene and metal (M = Ru(II)) or Os(II)) tris(2,2'-bipyridine) terminals bridged by an ethynylene or Pt(II) bis(σ-acetylide) moiety has been synthesized and investigated by transient spectroscopy. Selective excitation into the terminal metal complex is possible in each case and generates the lowest energy, excited triplet state localized on that molecular fragment. For both Os(II)-based dyads, the triplet state is unperturbed by the appended pyrene unit and the observed photophysical properties can be understood within the framework of the energy-gap law. The triplet state localized on the metal complex in the two Ru(II)-based dyads is involved in reversible energy transfer with the triplet associated with the pyrene unit, which is situated at slightly lower energy. When the terminal metal complex is a Ru(II) bis(2,2':6',2'-terpyridyl) fragment, however, the triplet levels are inverted such that the pyrene-like triplet state lies slightly above that of the metal complex. Kinetic spectrophotometry has allowed determination of the various rate constants and energy gaps and, on the basis of nonadiabatic electron-transfer theory, it appears that the central Pt bis(σ-acetylide) unit is a much inferior electronic conductor than is a simple ethynylene group. Reversible energy transfer of this type greatly prolongs the triplet lifetime of the Ru(II) tris(2,2'-bipyridyl) fragment. For example, equilibration between the triplet states is achieved within 10 ps for the ethynylenebridged dyad while the equilibrium mixture decays with a lifetime of about 40 μs in deoxygenated acetonitrile at room temperature.

Full text of DOI:10.1002/(SICI)1521-3765(19991105)5:11<3366::AID-CHEM3366>3.0.CO;2-I

FULL PAPER

Intramolecular Triplet Energy Transfer in Pyrene ± Metal Polypyridine Dyads:

A Strategy for Extending the Triplet Lifetime of the Metal Complex

Muriel Hissler, Anthony Harriman,* Abderrahim Khatyr and Raymond Ziessel*^[a]

Abstract: A series of photoactive dyads

bearing pyrene and metal (M Ru^II) or

Os^II) tris(2,2'-bipyridine) terminals

bridged by an ethynylene or Pt^IIbis(s-

acetylide) moiety has been synthesized

and investigated by transient spectro-

scopy. Selective excitation into the ter-

minal metal complex is possible in each

case and generates the lowest energy,

excited triplet state localized on that

molecular fragment. For both Os^II-based

dyads, the triplet state is unperturbed by

the appended pyrene unit and the ob-

served photophysical properties can be

understood within the framework of the

energy-gap law. The triplet state local-

ized on the metal complex in the two

Ru^II-based dyads is involved in reversi-

ble energy transfer with the triplet

associated with the pyrene unit, which

is situated at slightly lower energy. When

the terminal metal complex is a Ru^II

bis(2,2':6',2''-terpyridyl) fragment, how-

ever, the triplet levels are inverted such

that the pyrene-like triplet state lies

slightly above that of the metal complex.

Kinetic spectrophotometry has allowed

determination of the various rate con-

stants and energy gaps and, on the basis

of nonadiabatic electron-transfer theory,

it appears that the central Pt bis(s-

acetylide) unit is a much inferior elec-

tronic conductor than is a simple ethy-

nylene group. Reversible energy trans-

fer of this type greatly prolongs the

triplet lifetime of the Ru^IItris(2,2'-bipyr-

idyl) fragment. For example, equilibra-

tion between the triplet states is ach-

ieved within 10 ps for the ethynylene-

bridged dyad while the equilibrium

mixture decays with a lifetime of about

40 ms in deoxygenated acetonitrile at

room temperature.

Keywords: alkynes ´ energy transfer

´ luminescence ´ photochemistry ´

transition metals

triplet energy transfer between the metal complex and pyrene,

although the phosphorescence lifetime of the ruthenium(ii)

tris(polypyridine) derivative was somewhat lower, being

5.2 ms in deoxygenated methanol at 208C. It was further

reported that covalently attached naphthalene, pyrene or

anthracene units transferred singlet excitation energy to the

appended ruthenium(ii) complex while the identity of the

lowest energy triplet state was dependent on the nature of the

polycyclic chromophore.^[5]In fact, several studies have

described triplet energy transfer from a metal complex to an

aryl hydrocarbon,^[6±11]but in several cases the photosystems

are unstable with respect to sensitized oxygenation of the

polycycle.^{[6, 7, 9]}Very recently, Schmehl and co-workers^[12]

studied the photophysical properties of ruthenium(ii) tris(di-

imine) complexes linked by a single bond to naphthalene or

pyrene. It was concluded that the closely spaced terminals

remained in weak electronic communication and that triplet

states associated with these terminals were not in equilibrium

at room temperature.

Introduction

Bichromophoric molecular systems continue to provide

valuable information regarding the mechanisms of light-

induced energy- or electron-transfer processes, especially in

solution.^[1]Of the myriad dyads studied to date those based on

luminescent metal polypyridine complexes^[2]are of particular

significance since the triplet excited state can be visualized by

ultrasensitive emission spectroscopy. Within this family of

photoactive dyads, ruthenium(ii) tris(2,2'-bipyridine) com-

plexes bearing pendant aryl hydrocarbons have been studied

in some detail.^[3±12]Thus, Ford and Rodgers^[3]observed

reversible triplet energy transfer between a ruthenium(ii)

tris(polypyridine) complex and a covalently linked pyrene

moiety. Equilibration between the two triplet states localized

on the terminals of this highly flexible dyad extended the

phosphorescence lifetime of the metal complex from about

1 ms to 11.2 ms in deoxygenated solution at room temperature.

Related work by Sasse et al.^{[4, 5]}also described reversible

We now describe the results of a spectroscopic investigation

of the energy-transfer processes occurring in geometrically

constrained dyads comprising pyrene and ruthenium(ii)

tris(2,2'-bipyridine) terminals. Two types of connector are

considered: Firstly, the terminals have been joined together

by a single ethynylene group which is known to promote

strong electronic coupling along the molecular axis.^{[13, 14]}

[a] Prof. A. Harriman, Dr. R. Ziessel, Dr. M. Hissler, A. Khatyr

Laboratoire de Chimie, d'Electronique et de Photonique Moleculaires

Ecole Europeenne de Chimie, Polymeres et Materiaux

Universite Louis Pasteur, UPRES-A 7008 au CNRS, 25 rue Becquerel

F-67087 Strasbourg Cedex 2 (France)

Fax : ( 33)388136813

E-mail: ziessel@chimie.u-strasbr.fr

3366

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3366 $ 17.50+.50/0

Chem. Eur. J. 1999, 5, No. 11

3366±3381

Secondly, the conjugated bridge has been broken by incorpo-

ration of a Pt^IIfragment,^[15]which has the effect of curtailing

electronic coupling between the terminals.^[16]This strategy is

intended to provide dyads in which the terminals retain

comparable relative energy levels but for which the connector

exhibits different propensity for through-bond or through-

space energy transfer.^[11]The experimental measurements can

be conveniently categorized according to the spin multiplicity

of the initial excited state. Thus, excitation at long wavelength

(l > 500 nm) populates the triplet excited state localized on

the metal complex and this, because of the alkyne substituent,

corresponds to a metal-to-ligand charge-transfer (MLCT)

state in which an electron has been promoted from the metal

center to the functionalized 2,2'-bipyridine ligand.^[17]Alter-

natively, excitation with UV light can produce a singlet

excited state localized on the pyrene moiety, although,

because of overlapping absorption profiles, this process is

not selective. Both initial ex-

dine) complexes have been studied as model systems for the

triplet energy-transfer processes while a related ruthenium(ii)

bis(2,2':6',2''-terpyridine) complex has been investigated to

confirm the derived triplet energy spacings. Further details

concerning the photophysical properties of the metal-free

ligands will be reported elsewhere.^[18]

Results and Discussion

Synthesis: The hybrid pyr± bpy and pyr± terpy ligands were

prepared in three reactions and in excellent yield from

1-bromopyrene according to Scheme 1. The key step is a

palladium(⁰) promoted cross-coupling reaction between

1-ethynyl-pyrene and 5-bromo-2,2'-bipyridine^[19]or 4'-[{(tri-

fluoromethyl)sulfonyl}oxy]-2,2':6',2''-terpyridine.^[20]In order

cited states decay by way of

intramolecular energy transfer

and kinetic experiments have

been made in each case. The

present manuscript concen-

trates on the fate of the triplet

excited state localized on the

metal complex whilst a sepa-

rate paper^[18]deals with energy

redistribution following laser

excitation into the pyrene

chromophore. The correspond-

ing osmium tris(2,2'-bipyri-

Scheme 1. i) Trimethylsilylacetylene, n-propylamine, [Pd(PPh₃)₄] (6 mol%), 608C, 24 h; ii) KF, CH₃OH, 72 h;

iii) benzene, diisopropylamine, [Pd(PPh₃)₄] (6 mol%), 808C, 24 h.

to prepare the pyr± Pt ± bpy and pyr± Pt ± terpy metallo-

Abstract in French: Une famille de dyades photoactives

comportant une sous-unitØ pyrne et un centre mØtallique

ªRu(bpy)₃º ou ªOs(bpy)₃º pontØs par une entretoise acØtylne

ou platine(ii)-s-diacØtylne a ØtØ prØparØe, caractØrisØe et

synthons the key mono-substituted pyr± Pt ± Cl building block

has to be synthesized. Previous studies have revealed that a

square-planar trans-[Pt(PnBu₃)₂Cl₂] complex undergoes suc-

cessive substitution reactions with ethynyl-grafted oligopyr-

idines with CuI as a catalyst precursor and iPr₂NH as the

base.^[15]Unfortunately, no mono-substituted bpy or terpy

compounds could be properly isolated under these conditions.

This situation is confirmed herein by the reaction of 1-ethyn-

yl-pyrene with trans-[Pt(PnBu₃)₂Cl₂] under the experimental

conditions optimized earlier (CuI, iPr₂NH). The symmetri-

cally disubstituted pyr± Pt ± pyr was obtained in good yield

regardless of the starting molar ratio of 1-ethynyl-pyrene/Pt

(Scheme 2). However, the use of an SnMe₃-substituted

1-ethynyl-pyrene derivative, which avoids basic coupling

conditions, allows mono-functionalization of trans-

[Pt(PnBu₃)₂Cl₂] in high yield. These latter conditions afford

easy access to pyr± Pt ± Cl with only minor (6%) contami-

nation by pyr± Pt ± pyr.

ØtudiØe

par

spectroscopie

d'absorption

transitoire.

L'irradiation sØlective du centre mØtallique gØnre un Øtat

excitØ triplet localisØ sur le complexe. Dans le cas des

complexes d'osmium, cet Øtat excitØ triplet n'est pas perturbØ

par la prØsence de la sous-unitØ pyrne; les propriØtØs photo-

physiques peuvent donc etre interprØtØes dans le cadre de la loi

du seuil d'Ønergie. Par contre dans le cas des complexes de

ruthØnium, cet Øtat excitØ est en Øquilibre avec l'Øtat excitØ triplet

localisØ sur le pyrne, proche en Ønergie. Le transfert rØversible

d'Ønergie augmente de façon trs significative la durØe de vie de

l'Øtat excitØ triplet des complexes de ruthØnium. Les Øtudes

cinØtiques spectrophotomØtriques ont permis de dØterminer les

constantes de vitesse des divers processus de transfert qui ont

lieu dans les dyades. Elles ont montrØ que l'entretoise

platine(ii)-s-diacØtylne est un moins bon conducteur Ølectro-

nique qu'un coupleur acØtylnique. En effet, l'Øtat excitØ triplet

du complexe de ruthØnium(ii) se dØsactive en 40 microsecondes

pour une dyade construite avec un espacer acØtylnique tandis

que cet Øtat se dØsactive en 17 microsecondes pour une dyade

contenant un pont platine(ii)-s-diacØtylne dans une solution

d'acØtonitrile dØgazØe.

At this stage, chromatographic separation of these two

compounds is tedious as a result of their similar retention

properties. Separation of the pyr± Pt ± pyr side-product

proved to be much easier after connection of the bpy or

terpy fragment. Subsequent attachment of the chelating

center (5-ethynyl-2,2'-bipyridine^[21]or 4'-ethynyl-2,2':6',2''-ter-

pyridine^[21]) is straightforward with a copper-promoted cross-

Chem. Eur. J. 1999, 5, No. 11

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3367 $ 17.50+.50/0

3367

A. Harriman, R. Ziessel et al.

FULL PAPER

used ¹H and ¹³C NMR aug-

mented by ³¹P NMR spectro-

scopy in appropriate cases for

structural assignment and for

determination of purity. For

each compound, mass spectral

and FT-IR spectroscopic data

are provided, together with

details of the absorption spec-

trum recorded in solution.

Schemes 1 ± 4 also indicate the

abbreviations used throughout

the text to identify particular

molecular dyads and their met-

al-free ligands. Structures of the

reference compounds used for

photophysical and electro-

chemical evaluation of the

new materials are shown in

Scheme 5.

Scheme 2. i) THF, diisopropylamine, CuI (1 mol%), 72 h, RT; ii) n-butyllithium, THF, 788C; iii) THF, CuI

(3 mol%), 608C.

Spectroscopic properties of the

osmium(ii)-based dyads: Be-

coupling reaction carried out in the presence of base

(Scheme 3). The trans geometry of the square-planar plati-

num unit was confirmed by the ³¹P chemical shift (d ꢀ 4.3) and

phosphorous-platinum coupling constant (J(P,Pt) ꢀ 2300 Hz)

which are in the expected range for trans but not cis

isomers.^[22]Noteworthy, the cis isomers could not be prepared

cause the substituent has a marked effect on the absorption

spectral profile of the pyrene chromophore,^[18]there is

considerable spectral overlap between p,p* transitions asso-

ciated with the polycycle and MLCT bands localized on the

metal complex for both pyr± Os and pyr± Pt ± Os. Absorption

spectra recorded for these two dyads in acetonitrile solution

are shown in Figure 1, where the spin-allowed MLCT

transitions can be recognized as shoulders centered around

500 nm. Both compounds exhibit spin-forbidden MLCT

transitions in the far-red region of the spectrum,^{[17, 23]}however,

these facilitate selective excitation into the metal complex.

Thus, excitation at 598 nm, where the metal complex is the

sole chromophore, gives rise to very weak luminescence

centered between 700 and 800 nm (Figure 1). This emission is

attributed to the lowest energy triplet excited state localized

on the ªOs(bpy)º fragment by reference to model com-

pounds.^[17]The emission maximum (l_LUM), quantum yield

(F_LUM), and lifetime (t_T) measured in deoxygenated acetoni-

trile at 208C are somewhat dependent on the nature of the

connector Table 1), although there is no indication for

emission from the pyrene unit. For both dyads, decay of the

luminescence signal followed a single-exponential process at

all wavelengths. Corrected excitation spectra closely matched

the absorption spectrum over the entire spectral range; this

indicates that efficient energy transfer occurs from the pyrene

chromophore to the appended ªOs(bpy)º fragment in both

cases.^[18]The similarity of photophysical properties recorded

for the ªOs(bpy)º fragment and for appropriate reference

compounds^{[17, 24, 25]}suggests that there is little, if any, quench-

ing of the triplet state localized on the metal complex in these

dyads (Table 1). The triplet lifetimes recorded for these two

compounds, however, are somewhat different.

Scheme 3. i) THF, diisopropylamine, CuI (1 mol%), RT.

by the same route, starting with cis-[Pt(PnBu₃)₂Cl₂] precur-

sors. In all cases, the corresponding trans derivatives were

isolated, indicating a pronounced trans effect induced by the

electron-rich pyrene fragment. Coordination of the vacant

polypyridine ligands in the various pyrene-containing frag-

ments was readily accomplished by reaction with

[M(bpy)₂Cl₂] (M Ru or Os; bpy 2,2'-bipyridine) or [Ru-

(terpy)(dmso)Cl₂] (terpy 2,2':6',2''-terpyridine; dmso di-

methyl sulfoxide) precursors, as depicted in Scheme 4.

Although the dyads are but weakly emissive, there is

sufficient luminescence from the ªOs(bpy)º fragment in

deoxygenated acetonitrile at room temperature for the

calculation of the triplet energy. This calculation was made

The molecular structure of each new multi-component

system has been unequivocally authenticated on the basis of

spectroscopic evidence and by elemental analysis. We have

3368

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3368 $ 17.50+.50/0

Chem. Eur. J. 1999, 5, No. 11

Pyrene±Metal Polypyridine Dyads

3366±3381

Scheme 4. i) [Ru(terpy)(dmso)Cl₂], AgBF₄,

methanol, 808C, 24 h, followed by anion

metathesis; ii) cis-[Ru(bpy)₂Cl₂] ´ 2H₂O or

cis-[Os(bpy)₂Cl₂], ethanol, 808C, 24 h, fol-

lowed by anion metathesis.

Figure 1. Absorption spectra recorded for pyr± Pt ± Os (top) and pyr± Os

(bottom) in acetonitrile at 208C with the extinction coefficients expressed

in units of

¹cm ¹. The insert to each panel shows the corrected

luminescence recorded in deoxygenated acetonitrile at 208C following

excitation at 514 nm.

component is taken to represent the energy difference

between 0,0 vibronic levels in the triplet and ground states

while the energy spacing between adjacent Gaussian compo-

nents (ꢀhw) is attributed to the averaged medium frequency

vibrational mode coupled to the MLCT triplet state. The

derived parameters were used to reconstruct the reduced

luminescence spectrum according to Equation (1),^[10]where

the summation was restricted to x 5.

Scheme 5. Structures of the reference compounds.

following the procedures introduced by Meyer and co-work-

ers.^[10]Thus, the reduced emission spectrum,^{[26, 27]}as displayed

in wavenumbers, was deconvoluted into the minimum number

of Gaussian-shaped components of equal half-width (Dn_1/2).

The peak maximum (E₀) of the highest energy Gaussian

ꢀꢁ

ꢂ ꢁ ꢂꢁ

ꢃ

ꢁ

ꢂ ꢄꢂꢅ

E_oxꢀhw

S^x

E_o xꢀhw

Dn₁₌₂

L(n)

exp 4ln2

(1)

E_o

Iteration was continued until good agreement was reached

between observed and calculated spectra, allowing refine-

Chem. Eur. J. 1999, 5, No. 11

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3369 $ 17.50+.50/0

3369

A. Harriman, R. Ziessel et al.

FULL PAPER

Table 1. Photophysical properties recorded for the various dyads and

reference compounds in deoxygenated acetonitrile at room temperature.

coupling constants. While pyr± Os gives an S value similar to

that found for the parent complex [Os(bpy)₃]²(S 0.69)^[10]

the value found for pyr± Pt ± Os is larger. This parameter is

related to the nuclear displacement between ground and

triplet states^[28]so that the central Pt unit increases structural

differences between triplet and ground states. A similar

conclusion is reached by considering the larger re-organiza-

tion energy and Stokes' shift found for the Pt-linked system. It

is instructive to consider if these parameters can account for

the disparity in triplet lifetimes found for the two dyads. Thus,

within the framework of the energy-gap law as formulated by

Englman and Jortner^[28]and the application to osmium(ii)

tris(diimine) complexes by Meyer and co-workers,^[29]the

triplet lifetime is expected to increase with increasing triplet

energy, as is observed.

[c]

[f]

Compound^[a]

l_LUM[nm]^[b]F_LUM

t_T[ns]^[d]t₁[ns]^[e]A₁

t₂[ms]^[g]

pyr± Os

790

740

745

760

750

670

620

668

625

0.0012

pyr± Pt ± Os

[Os(bpy)₃]²

bpy ± Os

Os ± Pt ± Os

pyr± Ru

pyr± Pt ± Ru

[Ru(bpy)₃]²

bpy ± Ru

0.0042

0.0046

0.0040

0.0048

0.026

0.022

0.062

0.047

0.084

40000^[h]0.007^[i]0.98 42.4^[j]

17000^[h]3.2^[j]

0.97 17.4^[j]

980

1430

1300

580

Ru ± Pt ± Ru

pyr± Ru(terpy) 698

0.0050

[a] For the compound labelling see Schemes 1, 2, 4 and 5. [b] Æ 5 nm.

[c] Æ 8% for Ru^II-based and Æ25% for ªOs(bpy)º-based compounds.

[d] Æ 10%. [e] Shortest lived component detected in the transient decay

records, Æ5%. [f] Fractional contribution that the shorter-lived compo-

nent makes to the initial signal, Æ0.005. [g] Longest lived component

found in the decay records, Æ5%. [h] Measured with a 20 ns laser pulse at

532 nm. [i] By transient absorption spectroscopy. [j] By time-resolved

emission spectroscopy.

ꢁ

ꢂ

C ²2 p

gE_T

ꢀhw

k_T

exp( S)exp

t_Tꢀh ꢀhw E_T

(4)

ꢁ

ꢂ

E_T

g ln

Sꢀhw

In Equation (4), C refers to the vibrationally induced

electronic coupling matrix element but it is variations in the

final exponential term that exert the more significant effect on

the triplet lifetime. It appears, however, that the two dyads

share a common g value (Table 2) which is comparable to that

of the parent complex (g 1.77). This suggests that the triplet

lifetime is set by the amount of energy to be dissipated during

nonradiative deactivation.

ment of the parameters and calculation of the electron-

vibrational coupling constants (S). The re-organization energy

(l_T) for each compound, assumed to contain contributions

from both nuclear and solvent terms, was estimated from the

temperature ( 20 < T< 608C) dependence of the half-width

of the Gaussian components according to Equation (2), while

the triplet energy (E_T) was calculated as in Equation (3). A

complete compilation of the derived parameters is given in

Table 2.

The triplet energies for the ªOs(bpy)º fragments calculated

from this spectral curve-fitting routine remain closely com-

parable to those estimated as the crossover point between

normalized absorption and emission spectra.^[17]Comparison

with the triplet energy calculated for the parent complex

ꢁDn₁₌₂

l_T

(2)

(3)

16 k_BT lnꢁ2

E_T E_ol_T

[Os(bpy)₃]²(E_T 14700 cm )^[10]shows that the substituent

has almost no effect on the triplet energy of pyr± Pt ± Os,

Table 2. Spectroscopic properties derived from curve-fitting of the emission

spectra observed for the various dyads in deoxygenated acetonitrile at 208C.

perhaps raising it by about 100 cm , but lowering that of pyr±

Os by about 1200 cm . We might expect similar substituent

Property

pyr± Os pyr± Pt ± Os pyr± Ru pyr± Pt ± Ru pyr± Ru(terpy)

effects to operate at the opposite terminal such that the triplet

energy of the pyrene chromophore in these dyads will remain

somewhat similar to that of the parent polycycle (E_T

[a]

E₀[cm

ꢀhw [cm

]

12640

1350

1440

0.64

13710

1305

1590

0.78

14830

1490

1850

0.66

15970

1465

2105

0.82

14320

1180

1250

0.58

[b]

]

[c]

16600 cm ).^[30]On this basis, it is reasonable to suppose that,

Dn_1/2[cm

]

[d]

S [cm

l_T[cm

]

despite the presence of an alkyne substituent, the energy of

the pyrene triplet will remain well above that of the

ªOs(bpy)º fragments (Figure 2, left). This suspicion was

confirmed^[18]by room temperature phosphorescence spectra

recorded for pyr± bpy (l_LUM 648 nm) and pyr± Pt ± bpy

(l_LUM 590 nm) in micellar media where the 0,0 bands reside

at considerably higher energy than the triplet states localized

on the ªOs(bpy)º terminals.

The nature of the connector has no effect on the half-wave

potential (E_OX) measured for one-electron oxidation of the

Os^IIcation but does perturb the reductive behavior of the

coordinated bpy ligands (Table 3). By reference to cyclic

voltammograms recorded for the model compounds, appa-

rently both for pyr± Os and pyr± Pt ± Os the alkyne-substi-

tuted bpy ligand is reduced before the parent bpy ligands. The

presence of the central Pt unit raises the corresponding half-

wave potential (E_RED) by about 140 mV, although the first

[e]

]

890

13525

1.75

1110

14825

1.68

1490

16320

1.81

1920

17890

1.70

680

15000

2.09

[f]

E_T[cm

]

g^[g]

[a] Maximum of the highest energy Gaussian component obtained by fitting

the emission spectrum to a series of Gaussian profiles, Æ50 cm ¹. [b] Single-

averaged, medium-frequency vibrational mode coupled to the MLCT triplet

state, Æ50 cm ¹. [c] Half-width of the Gaussian components deconvoluted

from the emission spectrum, Æ100 cm ¹. [d] Huang ± Rhys factor calculated

from Equation (1), Æ0.04. [e] Re-organization energy associated with deacti-

vation of the lowest energy triplet state, Æ100 cm

. [f] Triplet energy,

Æ200 cm ¹. [g] Englman ± Jortner gamma factor calculated from Equation (4),

Æ0.05.

Comparison of these parameters shows that the central Pt

unit raises the triplet energy of the ªOs(bpy)º fragment by

approximately 1300 cm and slightly increases the re-organ-

ization energy accompanying decay of the triplet state. There

is a modest variation in the size of the electron ± vibrational

3370

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3370 $ 17.50+.50/0

Chem. Eur. J. 1999, 5, No. 11

Pyrene±Metal Polypyridine Dyads

3366±3381

region and weak absorbance in

the visible region. Both triplets

decay by first-order kinetics,

giving triplet lifetimes in excel-

lent agreement with those ob-

tained from time-resolved lu-

minescence spectroscopy. The

decay profiles give no indica-

tion of additional transient spe-

cies that persist on timescales

longer than about 50 ps and the

spectral properties are inde-

pendent of excitation wave-

length. This latter finding is

indicative of efficient energy

Figure 2. Simplified energy level diagrams for pyr± Pt ± Os (left) and pyr± Pt ± Ru (right). The triplet energy of

the metal complex was established by fitting the emission spectrum recorded at 208C while the triplet energy for

the pyrene unit was derived by kinetic spectrophotometry. Similar diagrams can be drawn for the corresponding

pyr± Os and pyr± Ru systems.

Table 3. Electrochemical properties measured at room temperature in deoxygen- transfer from pyrene to ªOs(bpy)º but the pyrene-like triplet

ated N,N-dimethylformamide for the various dyads and reference compounds. All

is not seen in any spectroscopic record.

processes correspond to the transfer of a single electron unless denoted otherwise

while the value given in parenthesis after the half-wave potential refers to the

potential difference between cathodic and anodic peaks in the cyclic voltammo-

grams.

Compound

E_OX[V] vs. SCE^[a]E_RED[V] vs. SCE^[b]

pyr± bpy

pyr± Pt ± bpy

pyr± Os

1.45^[c]

1.58(75), 1.83(85)

1.97(55), 2.10(60)

1.01(70), 1.29(70), 1.56(95), 1.74(80)

1.15(60), 1.35(60), 1.69(60), 1.95(80)

1.25(65), 1.54(65), 1.80(80)

1.07^[c]

0.88(85), 1.47^[c]

0.85(65), 1.11^[c]

pyr± Pt ± Os

[Os(bpy)₃]^2[d]0.83(70)

bpy ± Os

0.81(70)

1.15(70), 1.30(70), 1.59(80)

Os ± Pt ± Os

pyr± Ru

pyr± Pt ± Ru

[Ru(bpy)₃]^2[d]1.27(70)

0.79(2e,65)

1.47(2e)^[c]

1.23(2e,70), 1.53(2e,65), 1.78(2e,75)

1.05(80), 1.38(80), 1.57(85), 1.75(75)

1.23(65), 1.41(60), 1.70(70), 2.10(80)

1.35(65), 1.54(70), 1.79(75)

1.10,^[c]1.47(85)

bpy ± Ru

1.40(70)

1.22(70), 1.49(70), 1.60(70)

Ru ± Pt ± Ru

1.22(2e,75)

1.33(2e,70), 1.54(2e,70), 1.79(2e,80)

[a] One-electron half-wave potential for oxidative processes, Æ15 mV. [b] One-

electron half-wave potential for reductive processes, Æ15 mV. [c] Irreversible

electrode process. [d] Measured in acetonitrile solution.

Figure 3. Triplet ± triplet differential absorption spectra recorded for pyr±

Pt ± Os (top, delay time 50 ns) and pyr± Os (bottom, delay time 100 ps) in

deoxygenated acetonitrile following laser (FWHM 25 ps) excitation at

598 nm.

reduction step for pyr± Pt ± Os occurs at a more positive

potential than that of [Os(bpy)₃]²under similar conditions. In

fact, the Pt bis(s-acetylide) unit partially compensates for the

electron-withdrawing properties of the ethynylene group. The

second and third electrons are added to the coordinated

parent bpy ligands while the fourth electron goes to the

pyrene unit. Comparison with the parent complex shows that

the unsubstituted bpy ligands are somewhat easier to reduce

in pyr± Os and pyr± Pt ± Os; this suggests that, even in the

latter compound, there is some electron delocalization onto

the bridging unit at the first reduction stage. Incidently, the

electrochemical properties, used in conjunction with the

derived triplet energies, indicate that electron transfer to or

from the ªOs(bpy)º triplet state is thermodynamically un-

favorable for both pyr± Os and pyr± Pt ± Os.

Laser flash photolysis studies performed in deoxygenated

acetonitrile at room temperature confirmed that the lowest

energy triplet state in these dyads is that associated with the

ªOs(bpy)º fragment (Figure 3). Based on the electrochemical

results, we can conclude that these triplets are generated by

selective charge injection from the metal center to the alkyne-

substituted bpy ligand.^[17]The transient differential absorption

spectra show bleaching of the ground-state MLCT absorption

Spectroscopic properties of the ruthenium(ii) tris(2,2'-bipy-

ridyl)-based dyads: There is considerable spectral overlap

between p,p* transitions localized on the pyrene chromo-

phore and MLCT bands associated with the ªRu(bpy)º

fragment for both pyr± Ru and pyr± Pt ± Ru (Figure 4).

Although the spin-forbidden MLCT transitions are less well

developed than in the corresponding Os^II-based dyads, it is

possible to selectively excite the ªRu(bpy)º fragments at l >

500 nm. Under such conditions, emission characteristic of the

metal complex^[17]is observed for both pyr± Ru and pyr± Pt ±

Ru in deoxygenated acetonitrile at 208C (Figure 4). Relative

to appropriate reference compounds this emission is very

long-lived, although the quantum yields, spectral profiles and

emission maxima (Table 1) closely resemble those of the

reference compounds. Following excitation with a 20 ns laser

pulse, the emission decay profiles were well described in terms

of a single exponential component, at all detection wave-

lengths and regardless of excitation wavelength. Corrected

excitation spectra remain in excellent agreement with ab-

Chem. Eur. J. 1999, 5, No. 11

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3371 $ 17.50+.50/0

3371

A. Harriman, R. Ziessel et al.

FULL PAPER

calculated triplet energies it is concluded that all possible

electron-transfer events occurring via the lowest energy

MLCT triplet state of the ªRu(bpy)º fragment are thermo-

dynamically unfavorable by at least 300 meV.

Reversible triplet energy transfer in pyr± Pt ± Ru: Following

excitation of pyr± Pt ± Ru in deoxygenated acetonitrile at

208C with a 25 ps laser pulse at 532 nm, where ªRu(bpy)º is

the sole chromophore, the emission was found to decay with

dual-exponential kinetics (Figure 5). Analysis of the decon-

voluted signal showed the two lifetimes (t₁and t₂) to be (3.2 Æ

0.1) ns and (17.4 Æ 1.3) ms while the fractional amplitude of the

shorter-lived component (A₁) was (0.970 Æ 0.005). These

Figure 4. Absorption (solid curve) and corrected emission (dotted curve)

spectra recorded for pyr± Pt ± Ru (top) and pyr± Ru (bottom) in acetoni-

trile at 208C. An excitation wavelength of 514 nm was used for the emission

spectra while extinction coefficients are expressed in units of m ¹cm

sorption spectra recorded over the entire wavelength range,

indicating efficient energy transfer from pyrene to the

appended ªRu(bpy)º fragment in both cases.^[18]The pro-

longed triplet lifetime observed for these dyads is consistent^[3]

with the triplet state localized on the ªRu(bpy)º fragment

being in equilibrium with the triplet associated with the

pyrene unit.

The emission spectra were analyzed^[10]according to Equa-

tions (1) ± (3) and the derived parameters are listed in Table 2.

Again, it is apparent that the central Pt unit serves to increase

the re-organization energy accompanying deactivation of the

triplet state of the metal complex, to raise the triplet energy,

and to increase the extent of structural change between triplet

and ground states. The effects are somewhat more pro-

nounced than found for the corresponding ªOs(bpy)º-based

dyads while the triplet energies of pyr± Ru and pyr± Pt ± Ru

exceed those of pyr± Os and pyr± Pt ± Os by approximately

Figure 5. Decay profiles recorded for the emission from pyr± Pt ± Ru in

deoxygenated acetonitrile at 208C following laser (FWHM 25 ps)

excitation at 532 nm. The profiles are shown on two different timescales

with detection by a fast-response photodiode (top) and a red-sensitive

photomultiplier tube (bottom). Detection was at 650 nm in both cases. The

entire decay curve requires analysis in terms of two exponential compo-

nents with lifetimes of 3.2 ns and 17.4 ms. Top curve is shown in semi-

logarithmic form to emphasize the presence of the longer-lived component.

3000 cm . The significantly increased S and l values found for

pyr± Pt ± Ru relative to pyr± Ru indicate a more substantial

geometry change between triplet and ground states, perhaps

as a result from the differences in the extent of charge-transfer

character inherent to these states. In fact, comparison of the

E_REDvalues recorded for pyr± Ru and pyr± Pt ± Ru shows that

the presence of a central Pt unit induces a 180 mV negative

shift in the first reduction potential but the measured value is

still about 100 mV more positive than that recorded for the

parent complex. Thus, the lowest energy MLCT triplet states

of both pyr± Ru and pyr± Pt ± Ru are formed by selective

electron donation from metal center to the alkyne-substituted

bpy ligand.

Under anodic scans, the cyclic voltammograms indicate that

both the pyrene unit and the metal center can be oxidized but

there is no indication that the Pt^IIunit is redox active at

accessible potentials. For pyr± Pt ± Ru the pyrene terminal is

oxidized more easily than is the ªRu(bpy)º fragment but the

two units are oxidized simultaneously in pyr± Ru. Within

experimental limitations, there is no indication that the nature

of the connector affects E_OXfor the metal center. Upon

considering the electrochemical data (Table 3) in terms of the

lifetimes and fractional amplitudes remained independent of

both excitation and monitoring wavelength. No changes in the

spectral profile could be detected during the decay process.

Under flash photolysis conditions with excitation being made

with a 20 ns laser pulse at 532 nm the transient differential

absorption spectrum observed at all delay times corresponds

to the pyrene-like triplet (Figure 6), as already characterized

for pyr± Pt ± bpy.^[18]The pyrene-like triplet seen in the

transient spectroscopic records decayed with the same life-

time (t_T 16 Æ 2 ms) as found for the longer-lived component

in the time-resolved emission spectra. This behavior is

consistent^[3]with the three-state model shown in Figure 2,

right. Here, excitation of the dyad with visible light produces

the MLCT triplet state localized on the ªRu(bpy)º fragment.

This species decays with an inherent lifetime (t_R) to restore

the ground state but there is a competitive intramolecular

energy-transfer process that results in population of the triplet

state associated with the pyrene unit. The pyrene-like triplet

3372

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3372 $ 17.50+.50/0

Chem. Eur. J. 1999, 5, No. 11

Pyrene±Metal Polypyridine Dyads

3366±3381

intramolecular triplet energy transfer in pyr± Pt ± Ru. Thus,

immediately after excitation the transient differential absorp-

tion spectrum indicates population of the MLCT triplet state

associated with the ªRu(bpy)º fragment (Figure 7). The

spectral profiles evolve over a few nanoseconds until the

characteristic differential absorption spectrum of the pyrene-

like triplet state is established. This latter signal decays slowly

by first-order kinetics, giving an average lifetime of (15.6 Æ

1.7) ms. Analysis of the spectral records collected on short

timescales indicates that the pyrene-like triplet grows-in with

a lifetime of (2.9 Æ 0.3) ns. No other transient species are

apparent in the spectral profiles.

These studies indicate fast establishment of an equilibrium

mixture of triplet states because of the close positioning of

triplet levels associated with each terminal. Decay of this

equilibrium mixture occurs relatively slowly so that the

effective lifetime of the ªRu(bpy)º triplet is prolonged well

beyond that of the reference compounds. Under identical

conditions, the triplet lifetime (t_P) recorded for pyr± Pt ± bpy

is (30 Æ 5) ms and, assuming this lifetime also applies to the

inherent pyrene triplet state in pyr± Pt ± Ru, the inherent

triplet lifetime (t_R) of the ªRu(bpy)º fragment is calculated to

be about 1.2 ms. This value is similar to triplet lifetimes

measured for appropriate reference compounds under iden-

tical experimental conditions (Table 1).

Figure 6. Triplet ± triplet differential absorption spectrum recorded 100 ns

after excitation of pyr± Pt ± Ru in deoxygenated acetonitrile with a 20 ns

laser pulse at 532 nm. The insert shows a decay profile recorded at 415 nm.

decays via nonradiative pathways with an inherent lifetime

(t_P) and by way of reverse energy transfer to the ªRu(bpy)º

fragment. The two triplet states exist in equilibrium and decay

slowly with a common lifetime (t_T).

ꢁ

ꢂ

ꢁ

ꢂ

I_L(t) A₁exp

A₂exp

(5)

A₂A₁

t₁

t₂

k_D

(7)

t₂

t_Rt_P

Within this model, the rate constants for forward (k_F

3.0 Â 10⁸s ) and back (k_B 9.4 Â 10⁶s ) triplet energy trans-

fer can be obtained from the kinetic fit while the equilibrium

constant (K 32 Æ 2) for partitioning between the two triplets

can be calculated from the fractional amplitudes.

Reversible triplet energy transfer in pyr± Ru: Time-resolved

emission decay curves recorded for pyr± Ru in deoxygenated

acetonitrile at 208C could be analyzed satisfactorily in terms

of a single exponential processes having a lifetime (t₂) of

(42 Æ 3) ms. Transient differential absorption spectra recorded

after excitation of pyr± Ru with a 25 ps laser pulse at 532 nm,

where the ªRu(bpy)º fragment is the only absorbing species,

showed that the pyrene-like triplet was present immediately

after the pulse (Figure 8). This species decayed via first-order

A₁k_F

k_Fk_B

;

(6)

t₁

A₂k_B

This latter value allows estimation of the energy gap

between triplet states localized on ªRu(bpy)º and pyrene

terminals as being about 8.5 kJmol (DE_TT 710 cm ) such

that the energy of the pyrene-

like triplet must be about

17200 cm . As such, the pyr-

ene-like triplet lies at much

higher energy than the ªOs(b-

py)º fragment in pyr± Pt ± Os

and slightly above that of pyr-

ene itself.^[30]The derived E_Tis in

keeping with the 0,0 band found

in the phosphorescence spec-

trum of pyr± Pt ± bpy recorded

in micellar media at room tem-

perature.^[18]The overall effect

of the central Pt bis(s-acety-

lide) unit, therefore, is to raise

the triplet energies of both

Figure 7. Transient differential absorption spectra recorded at various times after excitation of pyr± Pt ± Ru in

deoxygenated acetonitrile with a 25 ps laser pulse at 532 nm. Spectra were recorded at delay times of 0, 1, 2, 3, 5,

and 10 ns. Also shown are kinetic traces obtained by overlaying spectral data collected at different delay times.

Formation of triplet pyrene can be monitored at 415 nm while decay of the triplet state localized on the

¹Ru(bpy)ª fragment can be followed at 460 nm. Solid lines drawn through the data points correspond to first-

order kinetic processes with lifetimes of 2.9 and 3.1 ns, respectively, at 415 and 460 nm.

terminals by a similar amount.

Flash photolysis studies

made with excitation at

532 nm with a 25 ps laser pulse

confirmed the occurrence of

Chem. Eur. J. 1999, 5, No. 11

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3373 $ 17.50+.50/0

3373

A. Harriman, R. Ziessel et al.

FULL PAPER

across the visible region showed the average lifetime of the

MLCT triplet of the ªRu(bpy)º fragment to be (7 Æ 2) ps

while the pyrene-like triplet did not decay over the accessible

time window (Figure 10). This fast decay process is readily

assignable to intramolecular triplet energy transfer from the

ªRu(bpy)º fragment to the nearby pyrene unit.

Figure 8. Triplet ± triplet differential absorption spectrum recorded after

excitation of pyr± Ru in deoxygenated acetonitrile with a 20 ns laser pulse

at 532 nm. The spectrum was recorded 100 ns after excitation with the

point-by-point method. The insert shows a decay profile recorded at

400 nm following excitation with a low energy laser pulse.

kinetics with a lifetime of (40 Æ 4) ms to restore the pre-pulse

baseline. The agreement between lifetimes measured by

transient absorption spectroscopy and by time-resolved

emission, taken together with the fact that only the ªRu(bpy)º

fragment emits but only the pyrene-like triplet is detected by

absorption, indicates that equilibration between the two

triplets occurs within the excitation pulse. For pyr± Ru,

therefore, the rate constant for energy transfer from the

ªRu(bpy)º triplet to the appended pyrene unit must exceed

Figure 10. Kinetic profiles obtained by overlaying spectra collected at

different delay times for the experiment reported in Figure 11. Kinetic

measurements are shown at detection wavelengths of 620 nm (top) and

410 nm (bottom). In each case, the solid line drawn through the data points

corresponds to a computer nonlinear least-squares fit to two consecutive

first-order processes having lifetimes of 0.41 and 7.1 ps. The faster step is

assigned to rapid formation of the ªRu(bpy)º triplet within the excitation

pulse while the slower step is attributed to intramolecular triplet energy

transfer.

about 2 Â 10¹⁰s .

It was not possible to detect the anticipated shorter-lived

component in time-resolved emission decay records within

the 50 ps temporal resolution available. However, excitation

of pyr± Ru with a sub-ps laser pulse at 565 nm generated the

MLCT triplet state localized on the ªRu(bpy)º fragment

(Figure 9). This species could be identified by comparison of

the observed transient differential absorption spectrum with

that recorded for the reference compound Ru ± bpy. Forma-

tion of the MLCT triplet state was complete within a few ps

but it decayed rapidly to give the transient spectrum already

seen at longer delay times and assigned to the pyrene-like

triplet. Kinetic measurements made at a series of wavelengths

In order to estimate the partitioning of triplets within the

equilibrium mixture, and thereby derive rate constants for

forward and reverse energy-transfer steps, the emission

quantum yield measured for pyr± Ru in aerated acetonitrile

was compared with that of the reference compound bpy ± Ru.

Under these conditions, the two compounds possess the same

triplet lifetime, this being set by O₂quenching, and, assuming

the radiative rate constants remain comparable, the ratio of

emission yields reflects the concentration of ªRu(bpy)º triplet

present at equilibrium. On this basis, it was concluded that

F_LUMfor the dyad is only 2% that of the mononuclear

reference compound. As such, individual rate constants for

the two energy-transfer events can be estimated from the

overall triplet lifetime, giving k_F 1.4 Â 10¹¹

¹and k_B 2.0 Â

10⁹s . These derived rates greatly exceed those found for

pyr± Pt ± Ru but the energy gap between the triplet states, as

calculated from the rates, (DE_TT 930 cm ) remains compa-

rable to that measured for the Pt bis(s-acetylide)-bridged

system. This estimate locates the energy of the pyrene-like

triplet in pyr± Ru at about 15390 cm , which is some

1200 cm ¹below that of the parent polycycle.

Deactivation of the equilibrium mixture occurs more slowly

for pyr± Ru than for pyr± Pt ± Ru and provides for an

exceptionally long-lived MLCT triplet state localized on the

Figure 9. Transient absorption spectra recorded after excitation of pyr±

Ru with a 0.35 ps laser pulse at 565 nm. Spectra were recorded at delay

times of 0, 1, 2, 5, 10, and 20 ps.

3374

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3374 $ 17.50+.50/0

Chem. Eur. J. 1999, 5, No. 11

Pyrene±Metal Polypyridine Dyads

3366±3381

ªRu(bpy)º fragment. In fact, the effective lifetime of this

latter species (t_T 42 Æ 4 ms) greatly exceeds those reported

earlier^[3±5]for pyrene ± ruthenium(ii) tris(polypyridine) dyads

with flexible connectors where triplet lifetimes of 5 and 11 ms

have been recorded. This prolongation of the triplet lifetime

reflects both the low percentage of MLCT triplet present in

the equilibrium mixture and the relatively long inherent

lifetime of the perturbing pyrene unit. This latter value (t_P

130 Æ 10 ms), being equated to the triplet lifetime of pyr± bpy,

allows estimation of t_R 1.2 ms.

between the emitting MLCT state and a higher-energy metal-

centered state,^[32]it is not easy to attribute the prolonged t_T

found for pyr± Ru(terpy) simply to a lowering of the triplet

energy. In fact, the 10-fold increase in t_Tobserved between

pyr± Ru(terpy) and terpy ± Ru provides strong support for the

idea that the appended pyrene unit prolongs the triplet

lifetime of the ªRu(terpy)º fragment.

The triplet lifetime recorded by transient absorption

spectrometry (t_T 560 ns) agrees very well with that deter-

mined by time-resolved luminescence spectroscopy (t_T

580 ns). Furthermore, the differential absorption spectrum

recorded for the lowest energy triplet state exhibits pro-

nounced bleaching of the MLCT absorption band associated

with the ªRu(terpy)º fragment (Figure 12). This observation

Triplet state properties of pyr± Ru(terpy): Additional studies

were made with the corresponding ethynylene-linked dyad in

which the terminal metal complex comprised a ruthenium(ii)

bis(2,2':6',2''-terpyridine) complex. It was considered that,

since the triplet energy of [Ru(terpy)₂]²(where terpy

2,2':6',2''-terpyridine) is lower than that of [Ru(bpy)₃]², the

triplet energies of the two terminals might be more compa-

rable than in pyr± Ru. For this new system, pyr± Ru(terpy),

the MLCT absorption band is very much better resolved than

in the other Ru^II-based compounds, appearing as a relatively

sharp and very intense (e 44800 m ¹cm ) band centered at

511 nm. The intensity of this MLCT transition is unusually

high but strong MLCT bands appear to be a characteristic of

ªRu(terpy)º-based chromophores bearing a conjugated sub-

stituent at the 4'-position.^[31]Luminescence is readily detected

in deoxygenated acetonitrile at 208C and appears as a well-

resolved maximum at 698 nm (Figure 11). The emission

Figure 12. Triplet ± triplet differential absorption spectrum recorded after

excitation of pyr± Ru(terpy) in deoxygenated acetonitrile with a 20 ns laser

pulse at 532 nm. The spectrum was recorded 100 ns after excitation using

the point-by-point method. The insert shows a decay profile recorded at

500 nm following excitation with a low energy laser pulse.

provides clear evidence that the lowest energy triplet state

present in pyr± Ru(terpy) must be localized on the metal

complex. Therefore, the triplet energy of the ªRu(terpy)º

fragment must be lower than that of the pyrene-like triplet

state in pyr± Ru(terpy) (Figure 13). Examination of the

Figure 11. Absorption (solid curve) and emission (dotted curve) spectra

recorded for pyr± Ru(terpy) in acetonitrile at room temperature. An

excitation wavelength of 514 nm was used for the emission spectrum while

extinction coefficients are expressed in units of m ¹cm

maximum (Table 1) is similar to that recorded for the

reference compound terpy ± Ru but both the emission lifetime

and quantum yield greatly exceed those found for the model.

The increased F_L, which is unusually high for a ªRu(terpy)º

fragment,^[24]can be partly explained in terms of the higher

radiative probability associated with the improved optical

absorption properties of the chromophore (Figure 11). Al-

though the photophysical properties of ªRu(terpy)º-based

chromophores are particularly sensitive to the energy gap

Figure 13. Simplified energy level diagrams for pyr± Ru(terpy). The triplet

energy of the metal complex was established by fitting the emission

spectrum recorded at 208C while the triplet energy for the pyrene unit was

derived by kinetic spectrophotometry.

Chem. Eur. J. 1999, 5, No. 11

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3375 $ 17.50+.50/0

3375

A. Harriman, R. Ziessel et al.

FULL PAPER

system with improved temporal resolution (l 565 nm;

FWHM 0.35 ps) failed to detect any absorption spectral

changes occurring on the ps timescale that could be attributed

to triplet energy transfer between the subunits.

energy-transfer step is assumed to involve through-bond

electron exchange in accordance with the Dexter mecha-

nism^[33]and, as such, it is tempting to attribute differences in

rate to the degree of electronic coupling between the

terminals. According to current theoretical models, the rate

constant for intramolecular electron exchange (k_ET) can be

expressed in terms of the Fermiꢁs golden rule [Eq. (8)]:^{[10, 34, 35]}

Analysis of the emission spectrum recorded for pyr± Ru-

(terpy) in deoxygenated acetonitrile at 208C, in strict

accordance with the protocol outlined above, shows that the

triplet energy of the ªRu(terpy)º fragment is located at

15000 cm ¹(Table 2). Notably, this luminescence spectrum is

better resolved than any bpy-based system and corresponds to

a particularly small re-organization energy. The derived

k_ET

V ²_DAFC

ꢀh

ꢁ

ꢂꢁ

ꢂ

X X

S_Dⁿ

S_A^m

exp( S_D)exp( S_A)

exp( DG*) (8)

4pl_Tk_BT

n0 m0

energy level places the ªRu(terpy)º triplet some 1320 cm

ꢁDG^o l_T mꢀhw_D nꢀhw_A

below that of the corresponding ªRu(bpy)º triplet and

approxiamtely 390 cm ¹below the pyrene-like triplet, assum-

ing the latter has the same energy as in pyr± Ru. An energy

DG*

4 l_Tk_B

Here, V_DAis the matrix element for electron exchange, l_Tis

the total re-organization energy accompanying electron

exchange, k_Bis the Boltzmann constant, and T is the absolute

temperature. The terms S_Dand S_A, respectively, refer to the

electron-vibrational coupling constants for donor and accept-

or while ꢀhw_Dand ꢀhw_A, respectively, are averaged medium

frequency vibronic modes coupled to the two triplet states.

The indices m and n are vibrational quantum numbers for

donor and acceptor species and, in carrying out the summa-

tion, we have restricted these values to m n 7 since higher

numbers had no effect on the calculated Franck ± Condon

factor (FC). The final term in the expression, DG8

( DE_TT), refers to the energy gap between triplets localized

on ªRu(bpy)º and pyrene fragments.

gap of 390 cm corresponds to an equilibrium constant K of

0.15 Æ 0.03 and implies that the equilibrium distribution

would consist of about 13% of the pyrene-like triplet. As

such, the apparent 10-fold prolongation of the MLCT triplet

lifetime implied for pyr± Ru(terpy) cannot be explained in

terms of reversible triplet energy transfer with the appended

pyrene unit. According to Equation (7), equilibration be-

tween the triplet levels would provide only a minor stabiliza-

tion of the ªRu(terpy)º triplet and it is clear that an additional

effect operates in this case. A proper understanding of this

system requires additional experimental study but, within the

context of the present investigation, it serves to show that the

triplet energy of the ethynylated pyrene unit must lie above

15000 cm ¹so that our assignment of E_T 15390 cm ¹seems

reasonable.

Each parameter needed to evaluate the Franck ± Condon

factor is available for the ªRu(bpy)º fragment (Table 2) while

the energy gap has been established accurately by kinetic

Comparison of the rates of intramolecular triplet energy

transfer: The rate constants for forward (k_F) and reverse (k_B)

energy transfer in the two ªRu(bpy)º-based dyads are

summarized in (Table 4, together with the relevant energy

gaps (DE_TT). The rates are very much faster for the

ethynylene-bridged system and it is somewhat surprising to

find that k_Bfor pyr± Ru exceeds k_Ffor pyr± Pt ± Ru consid-

ering the relative energy gaps for these processes. Each

measurements. Estimates of both ꢀhw ( 1210 cm ) and the

room-temperature re-organization energy (l_T 410 cm ) for

the pyrene-like triplet were made by fitting the phosphor-

escence spectra of pyr± bpy and pyr± Pt ± bpy recorded in

micellar media at room temperature. The average value of ꢀhw

obtained by this analysis remains comparable to those derived

for the corresponding ªRu(bpy)º fragments while the small l_T

is in keeping with the pyrene-like triplet being primarily p,p*

in character. This latter value is similar to that determined

Table 4. Comparison of properties related to the intramolecular electron-

exchange processes occurring in the ªRu(bpy)º-based dyads, as determined

in acetonitrile at 208C.

recently (l_T 260 cm ) for anthracene triplet by Meyer and

co-workers.^[10]Finally, in order to estimate S for the pyrene

triplet we have used the relationship^[28]in Equation (9), so

that an average value of S 0.34 is available for use in the

calculations. Again, this latter value is reasonable for a

polycyclic aromatic hydrocarbon^{[28, 36]}but it is much smaller

than that estimated recently for triplet anthracene (S 0.9).^[10]

Property

pyr± Ru

pyr± Pt ± Ru

[a]

[b]

k_F[10⁸s

k_B[10⁶s

]

1400

2000

930

2.00

0.35

24.4

1.6

3.0

9.6

710

1.00

0.25

7.0

[c]

DE_TT[cm

]

FC^F[10 ⁴cm

FC^R[10 ⁴cm

]

[d]

[e]

[f]

V^F_DA[cm

V^R_DA[cm

]

[g]

0.6

l_T

S ꢂ

(9)

ꢀhw

[a] Rate constant for energy transfer from triplet ªRu(bpy)º to pyrene,

Æ10%. [b] Rate constant for energy transfer from triplet pyrene to

ªRu(bpy)º, Æ10%. [c] Energy gap between triplet states localized on

ªRu(bpy)º and pyrene fragments, Æ80 cm ¹. [d] Franck ± Condon factor

calculated from Equation (8) for the forward energy-transfer step, Æ5%.

[e] Franck ± Condon factor calculated from Equation (8) for the reverse

energy-transfer step, Æ5%. [f] Electronic coupling matrix element calcu-

lated from Equation (8) for the forward energy-transfer step, Æ15%.

[g] Electronic coupling matrix element calculated from Equation (8) for

the reverse energy-transfer step, Æ20%.

Using these various parameters in conjunction with Equa-

tion (8) allows calculation^[37]of the Franck ± Condon factors

(FC) associated with each electron-exchange event (Table 4).

The derived values fall within a narrow range, with the factors

for the forward reactions exceeding those for the correspond-

ing reverse steps. In each case, energy transfer is expected to

fall well within the normal regime of a Marcus-type rate

3376

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3376 $ 17.50+.50/0

Chem. Eur. J. 1999, 5, No. 11

Pyrene±Metal Polypyridine Dyads

3366±3381

versus energy-gap profile, since DG8 < l_T, where the rate

should increase linearly with increasing FC. This is not the

case, however, and it is clear that the electronic coupling

matrix elements (V_DA) for electron exchange must vary

throughout the series (Table 4). Indeed, the bridging Pt

bis(s-acetylide) moiety decreases electronic coupling along

the molecular axis, especially for the forward energy-transfer

process, so that slow rates of energy transfer are found for

pyr± Pt ± Ru. This effect can be attributed to a combination of

the increased length of the connector and the imposition of a

high-energy barrier associated with tunnelling across the Pt

atom. A similar attenuation in the rate of electron exchange

was found for Ru ± Pt ± Os relative to Ru ± Os but a clear

interpretation was obscured by modification of the nature of

the lowest energy triplet states in these systems.^[16]Electronic

coupling is more pronounced by a factor of about three-fold

for the forward reaction than for the analogous reverse

process. This might arise from changes in the degree of orbital

overlap between the reactants since the [nominally] MLCT

and [primarily] p,p* triplets localized on ªRu(bpy)º and

pyrene fragments, respectively, are of quite different character.

comparable to other closely spaced dyads.^[41]In the present

molecules, however, the identity of the bridge is not obvious.

We have described pyr± Ru as being a tripartite donor-

connector± acceptor molecule of the type D ± C ± A where the

connector (C) is an ethynylene group. The molecule could be

described equally well as a donor± acceptor D ± A system

where the ethynylene group is too intimately associated with

one or both of the reactants to be classified as a discrete

bridge. Certainly the bridge is not a simple acetylene group

since the triplet energies of such moieties are too high to

effectively mediate through-bond electron exchange in pyr±

Ru by the superexchange mechanism.^{[42, 43]}It is interesting to

note in this context that ab initio molecular orbital calcu-

lations made at the CIS/3 ± 21G level provide some indica-

tions for how the ethynylene group might affect the triplet

energy of pyrene. These calculations indicate that the triplet

energy drops from 16500 cm for pyrene to 16000 cm for

1-ethynylpyrene while room temperature phosphorescence

spectra confirm the validity of both values. The triplet energy

of this latter compound is too high to properly reflect that of

the pyrene-like triplet in pyr± Ru but the calculated (E_T

15600 cm ) and measured (E_T 15400 cm ) triplet energies

found for pyr± bpy are in much closer agreement with the

energy of the pyrene-like triplet in the dyad, as derived from

kinetic measurements. The implication drawn from these

calculations, therefore, is that there is direct communication

between the two triplet states and that triplet energy transfer

does not involve virtual population of a triplet localized on the

ethynylene connector.

Conclusion

Four systems have now been described where an equilibrium

mixture of triplets is clearly established within the available

time frame. The flexibly linked pyrene-ªRu(bpy)º dyads

reported by Ford and Rodgers^[3]and by Sasse et al.^{[4, 5]}provide

for ªRu(bpy)º triplet lifetimes of 11.2 and 5.2 ms, respectively,

in deoxygenated solution at room temperature. Both pyr±

Pt ± Ru (t_T 17 ms) and pyr± Ru (t_T 42 ms) give much

longer-lived MLCT triplet states at room temperature with

the latter system being unique in that the equilibrium is

established very quickly after excitation. This fast formation

and slow decay of the equilibrium make pyr± Ru an interest-

ing candidate for exploitation as a photosensitizer for those

processes that might benefit from the extended triplet lifetime

of the ªRu(bpy)º fragment. The lifetime of each triplet in the

equilibrium mixture, as measured independently by transient

absorption spectrometry and time-resolved emission spectro-

scopy, is affected equally by the presence of molecular oxygen

and in aerated acetonitrile the average triplet lifetime is

220 ns. With pyr± Ru(terpy), the equilibrium favors the triplet

localized on the ªRu(terpy)º fragment so that, in terms of

Equation (7), only a slight prolongation of the triplet lifetime.

Even so, it is rare to obtain a ªRu(terpy)º derivative

possessing a triplet lifetime of 570 ns since this represents a

1000-fold increase relative to the parent complex

[Ru(terpy)₂]²in acetonitrile at 208C.^[38]

Finally, some comparison can be made between ªRu(bpy)º

and ªRu(terpy)º terminals. At first glance, the results

obtained with pyr± Ru(terpy) appear disappointing in that

the pyrene appendage provides little stabilization of the

ªRu(terpy)º triplet state^[44]because of the energy spacing.

This is a false impression, however, since the triplet lifetime

(t_T 570 ns) observed for pyr± Ru(terpy) greatly exceeds that

recorded for any other ethynylated mononuclear ªRu(terpy)º

fragment at room temperature, where t_Tvalues tend to be

about 50 ns.^[11]Although a variety of strategies^[45±50]has been

tried in an effort to prolong the triplet lifetime of mono-

nuclear ªRu(terpy)º-based compounds, the attachment of an

ethynylated-pyrene unit gives a triplet state of unsurpassed

longevity. In fact, the triplet lifetime found for pyr± Ru(terpy)

is similar to those measured for certain alkyne-bridged

binuclear ªRu(terpy)º complexes where the lifetime is

prolonged because of extensive electron delocalization over

the ditopic ligand.^{[24, 51]}Additional studies have confirmed

that the triplet lifetime increases when the energy of the

bridge approaches that of the terminals as a result of increased

electron delocalization.^[50]This is unlikely to be the case with

pyr± Ru(terpy), however, since the pyrene unit is much more

difficult to reduce than is the coordinated terpy ligand.

Instead, the extra stabilization of the MLCT triplet inferred

for pyr± Ru(terpy) relative to other ethynylated mononuclear

ªRu(terpy)º complexes is attributed to mixing between the

MLCT triplet and the pyrene-like p,p* triplet. This same

effect, by introducing more p,p* character into the MLCT

triplet, could be responsible for the decreased re-organization

energy noted for pyr± Ru(terpy).

An equally important observation made during this inves-

tigation is that pyr± Ru acts as a molecular dyad with discrete

terminals, as opposed to being a giant supermolecule with a

highly delocalized LUMO, where intramolecular electron

exchange can be treated in terms of a nonadiabatic process.

There is no indication that the rate of triplet energy transfer is

controlled by solvation dynamics^[39]or by vibronic process-

es,^[40]despite the close proximity of the reactants. Electronic

coupling at the triplet level is modest (V_DAꢃ k_BT) and

Chem. Eur. J. 1999, 5, No. 11

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3377 $ 17.50+.50/0

3377

A. Harriman, R. Ziessel et al.

FULL PAPER

from 0 to 50%. Analytically pure compound was obtained (0.057 g, 95%).

Experimental Section

¹H NMR (CDCl₃): d 8.79 ± 8.63 (m, 7H), 8.27 ± 8.00 (m, 8H), 7.90 (td, 2H,

³J 7.63 Hz, J 1.83 Hz), 7.41 ± 7.35 (m, 2H); ¹³C{¹H} NMR (CDCl₃): d

Materials: 1-Bromopyrene, tri-n-butylphosphane, 2,2'-bipyridine, and CuI

were obtained from commercial sources and used without further

purification. 5-Bromo-2,2'-bipyridine,^[19]5-ethynyl-2,2'-bipyridine,^[21]4'-

[{(trifluoromethyl)sulfonyl}oxy]-2,2':6',2''-terpyridine,^[20]

155.7, 155.6, 149.2, 136.9, 133.6, 132.3, 131.8, 131.2, 131.0, 130.1, 128.7, 128.6,

127.2, 126.3, 125.9, 125.8, 125.5, 124.6, 124.1, 122.8, 121.3, 116.7 (CC_ethynyl),

93.1(CC_ethynyl); FT-IR (KBr, cm ¹): nÄ 3039 (m), 2977 (s), 2924 (m), 2197

4'-ethynyl-

ꢄ

(C C) (m), 1653 (m), 1576 (s), 1462 (m), 1386 (s), 1262 (m), 1087 (s), 1048

2,2':6',2''-terpyridine,^[21][Pd(PPh₃)₄],^[52]trans-[Pt(PnBu₃)Cl₂],^[53]cis-[Ru-

(bpy)₂Cl₂] ´ 2H₂O,^[54]and cis-[Os(bpy)₂Cl₂]^[55]were prepared and purified

according to literature procedures. All reactions were carried out under dry

argon with the Schlenk-tube techniques. Solvents, including diisopropyl-

amine, were dried over suitable reagents and distilled under argon

immediately prior to use. The 200 (¹H) and 50 (¹³C{¹H}) MHz NMR

spectra were recorded at room temperature unless otherwise specified with

perdeuterated solvents with residual protiated solvent acting as internal

standard: for ¹H spectra d 7.26 for CDCl₃, d 5.32 for CD₂Cl₂; for ¹³C

spectra d 77.0 for CDCl₃and d 53.8 for CD₂Cl₂. All carbon signals were

detected as singlets. Fast-atom bombardement (FAB, positive mode) mass

spectra were recorded with a ZAB-HF-VB-analytical apparatus with m-

nitrobenzyl alcohol (m-NBA) as matrix. FT-IR spectra were recorded as

KBr pellets. Melting points were obtained with a capillary melting point

apparatus in open-ended capillaries and are uncorrected. Evidence for the

(s), 883 (m), 835 (s), 786 (s), 712 (s), 611 (s), 227 (vs); FAB (m-NBA): 458.2

[M H] ; anal. calcd (%) for C₃₃H₁₉N₃(M_r 457.5 gmol ¹): C 86.63, H

4.19, N 9.18; found: C 86.54, H 4.09, N 9.10.

5-(1-Ethynyl-pyrene)-2,2'-bipyridine (pyr± bpy): A similar experimental

procedure was used as that described above for preparation of 1-trime-

thylsilylethynyl-pyrene: 1-ethynyl-pyrene (0.22 mmol, 0.050 g), 5-bromo-

2,2'-bipyridine (0.22 mmol, 0.052 g), benzene (10 mL), diisopropylamine

(5 mL). After the solution was stirred for 15 hours, the solvent was removed

under vacuum. The crude product was first chromatographed on silica

eluting with a gradient of dichloromethane in hexane from 0 to 100%,

followed by a second chromatography on alumina with the same gradient.

Analytically pure compound was obtained (0.076 g, 90%). ¹H NMR

(CDCl₃): d 9.00 (d, 1H, ⁴J 1.83 Hz), 8.72 (d, 1H, J 3.67 Hz), 8.65 (d,

1H, J 9.16 Hz), 8.48 (4line multiplet, 2H), 8.16 (16line multiplet, 9H),

7.85 (td, 1H, ³J 7.63 Hz, ⁴J 1.83 Hz), 7.35 (qd, 1H, ³J 4.89 Hz, ⁴J

1.22 Hz); ¹³C{¹H} NMR (CDCl₃): d 155.5, 154.8, 151.7, 148.3, 139.3,

137.0, 132.2, 132.0, 131.9, 131.6, 131.2, 131.0, 129.7, 128.6, 128.5, 128.4, 127.2,

126.3, 125.8, 125.3, 124.6, 123.9, 121.4, 120.6, 120.4, 116.9, 92.9 (CC_ethynyl),

92.0 (CC_ethynyl); FT-IR (KBr, cm ¹): nÄ 3030 (m), 2976 (s), 2925 (m), 2194

molecular formulation is based on H and ¹³C{¹H} NMR spectra for both

structure and purity.

ꢄ

1-Trimethylsilylethynyl-pyrene (pyrC CTMS): A Schlenk flask was charg-

ed with 1-bromopyrene (0.100 g, 0.25 mmol), trimethylsilylacetylene

(0.49 mL, 0.71 mmol), [Pd(PPh₃)₄] (0.024 g, 6% mol) and Ar-purged

(nPr)NH₂(10 mL). The highly fluorescent solution was heated at 608C.

After complete consumption of the starting material (heating overnight),

the solvent was removed under vacuum. The crude product was purified by

flash chromatography on a column packed with silica gel and eluting with a

gradient of dichloromethane in hexane from 0 to 2%. The analytically pure

compound was obtained after recrystallization from hot hexane (0.097 g,

93%). R_f 0.47 (silica gel, hexane/dichloromethane 4:1 v/v); m.p.: 102 ±

1038C; ¹H NMR (CD₂Cl₂): d 8.59 (d, 1H, ³J 9.13 Hz), 8.09 (m, 8H),

0.56 (s, 9H); ¹³C{¹H} NMR (CDCl₃): d 132.5, 131.7, 131.5, 131.3, 130.2,

128.8, 128.6, 127.4, 126.6, 126.4, 125.7, 124.8, 124.6, 124.4, 117.9, 104.5

(CC_ethynyl), 100.7 (CC_ethynyl), 0.47 (Si-CH₃); FT-IR (KBr pellets, cm ¹): nÄ

ꢄ

(C C) (m), 1653 (m), 1574 (m), 1537 (w), 1452 (s), 1428 (s), 1262 (m), 1183

(m), 1088 (s), 1048 (s), 882 (m), 841 (s), 793 (s), 736 (s), 711 (s), 219 (s);

FAB (m-NBA): 381.2 [M H] ; UV/Vis (CH₂Cl₂) l_max(e) 394 (49100),

371 (51100), 308 (34700), 279 (44700), 235 (53500), 215 (27800), 213

(28100), 203 nm (27000 m ¹cm ¹); anal. calcd (%) for C₂₈H₁₆N₂(M_r

380.4 gmol ¹): C 88.40; H 4.24; N 7.36; found: C 88.37; H 4.19; N 7.30.

[Ruthenium(ii)bis(2,2'-bipyridine)(5-{1-ethynyl-pyrene}-2,2'-bipyridine)]-

(PF₆)₂[pyr± Ru]: A stirred solution of ethanol (10 mL) with cis-[Ru(bpy)_2-

Cl₂] ´ 2H₂O (0.08 mmol, 0.041 g) and pyr± bpy (0.08 mmol, 0.030 g) was

heated at 808C for 16 hours. After the solution cooled to room temper-

ature, potassium hexafluorophosphate (0.16 mmol, 0.029 g) in water

(5 mL) was added and the organic solvent was evaporated to give a red

solid. The crude precipitate was first chromatographed on alumina eluting

with a gradient of methanol in dichloromethane from 0 to 5%, followed by

a second chromatography on alumina eluting with CH₃CN to give the

required product as an analytically pure red solid (0.040 g, 47%). ¹H NMR

ꢄ

2977 (s), 2150 (C C) (s), 1652 (m), 1598 (m), 1485 (m), 1450 (m), 1406 (m),

1251 (s), 1183 (m), 1090 (m), 1048 (s), 896 (s), 846 (s), 756 (s), 716 (s), 632

(m), 323 (m); FAB (m-NBA): 298.1 [M H] , 283.1 [M CH₃] , 226.0

[M Si(CH₃)₃] ; UV/Vis (hexane) l_max(e) 377 (2100), 357 (62200), 388

(40900), 323 (17100), 277 (53500), 266 (31200), 256 (14500), 242 (39200),

230 nm (36100 m ¹cm ¹); anal. calcd (%) for C₂₁H₁₈Si (M_r

298.46 gmol ¹): C 84.51; H 6.01; found: C 84.42; H 5.98.

(CDCl₃): d 8.54 (m, 7H), 8.18 (m, 14H), 7.95 (d, 1H, J 1.52 Hz), 7.88

(4line multiplet, 2H), 7.83 (m, 3H), 7.47 (m, 5H); ¹³C{¹H} NMR (CDCl₃):

d 158.0, 157.9, 157.4, 156.7, 154.3, 152.9, 152.7, 152.5, 140.2, 138.8, 133.2,

132.8, 131.9, 131.7, 130.7, 130.0, 129.9, 128.6, 128.5, 128.0, 127.7, 127.2, 125.7,

ꢄ

1-Ethynyl-pyrene (pyrC CH): Solid KF (0.060 g, 0.9 mmol) was added to a

stirred solution of 1-trimethylsilylethynyl-pyrene (0.090 g, 0.3 mmol) in

tetrahydrofuran/methanol (40 mL 1:1 v/v). After complete consumption of

the starting material (three days), the solution was concentrated by rotary

125.6, 125.4, 125.3, 125.2, 124.8, 116.1, 96.4 (CC_ethynyl), 90.7 (CC_ethynyl); FT-IR

ꢄ

(KBr, cm ): nÄ 3036 (m), 2975 (s), 2195 (C C) (m), 1603 (m), 1465 (s),

1447 (s), 1242 (m), 1164 (m), 840 (s), 763 (s), 731 (s), 558 (s); FAB (m-

NBA): 939.0 [M PF₆] , 794.0 [M 2PF₆], 638.0 [M 2PF₆bpy]; UV/

evaporation to give

a crude product which was purified by flash

chromatography on a column packed with alumina and eluting with a

mixture of hexane/dichloromethane 99:1 to yield 1-ethynylpyrene (0.067 g,

99%). R_f 0.38 (alumina, hexane/dichloromethane 4:1 v/v); m.p.: 102 ±

1038C; ¹H NMR (CDCl₃): d 8.57 (d, 1H, ³J 9.13 Hz), 8.04 (m, 8H), 3.69

(s, 1H); ¹³C{¹H} NMR (CDCl₃): d 132.3, 131.4, 131.0, 130.8, 129.9, 128.4,

Vis (CH₃CN) l_max(e) 416 (45200), 380 (40600), 358 (33100), 353 (33100),

339 (30500), 280 (118300), 237 nm (75800 m ¹cm ¹); anal. calcd (%) for

C₄₈H₃₂N₆RuP₂F₁₂(M_r 1083.8 gmol ¹): C 53.19; H 2.98; N 7.75; found: C

52.83; H 2.63; N 7.59.

[Osmium(ii)bis(2,2'-bipyridine)(5-{1-ethynyl-pyrene}-2,2'-bipyridine)]-

(PF₆)₂[pyr± Os]: A stirred solution of ethanol (10 mL) with cis-[Os(b-

py)₂Cl₂] (0.10 mmol, 0.060 g) and pyr± bpy (0.10 mmol, 0.040 g) was heated

at 808C for 16 hours. After the solution cooled to room temperature,

potassium hexafluorophosphate (0.2 mmol, 0.038 g) in water (5 mL) was

added before the organic solvent was removed by evaporation, to give the

required product as a red solid. The crude precipitate was chromato-

graphed on alumina eluting with a gradient of dichloromethane in hexane

from 50 to 100%. This was followed by a second chromatography on silica

eluting with a gradient of methanol in dichloromethane from 0 to 3%. The

128.2, 127.0, 126.1, 125.6, 125.5, 125.1, 124.2, 123.9, 116.3, 82.8 (CC_ethynyl),

ꢄ

82.6 (CC_ethynyl); FT-IR (KBr, cm ): nÄ 3295 (C C-H) (s), 2977 (s), 2090

ꢄ

(C C) (m), 1654 (m), 1414 (m), 1262 (m), 1088 (s), 1048 (s), 880 (s), 837 (s),

756 (s), 713 (s), 643 (m), 593 (m), 218 (s); FAB (m-NBA): 226.1 [M] ,

213.1 [M CH] ; UV/Vis (CH₂Cl₂) l_max(e) 383 (2100), 374 (1700), 358

(27300), 342 (18900), 272 (15500), 262 (7400), 247 (28600), 237 (22100),

218 (12500), 210 (11400), 208 (11100), 201 nm (10700 m ¹cm ¹); anal.

calcd (%) for C₁₈H₁₀(M_r 226.28 gmol ¹): C 95.55; H 4.45; found: C 95.47;

H 4.39.

4'-(1-Ethyny-pyrene)-2,2':6',2''-terpyridine (pyr± terpy): A similar exper-

imental procedure was used as that described for preparation of 1-trime-

thylsilylethynyl-pyrene: 1-ethynyl-pyrene (0.13 mmol, 0.029 g), 4'-{[(tri-

fluoromethyl)sulfonyl]oxy}-2,2':6',2''-terpyridine (0.13 mmol, 0.050 g), ben-

zene (10 mL), diisopropylamine (1.5 mL). After the solution was stirred for

18 hours, the solvent was removed. The crude product was chromato-

graphed on alumina eluting with a gradient of dichloromethane in hexane

analytically pure compound was isolated as a dark green solid (0.029 g,

ꢄ

30%). FT-IR (KBr, cm ): nÄ 2923 (s), 2195 (C C) (m), 1601 (m), 1460 (s),

1313 (m), 1243 (m), 1109 (m), 841 (s), 764 (m), 619 (m), 556 (m); FAB

(tetraglyme): 1029.0 [M PF₆ H] , 884.0 [M 2PF₆ H] ; UV/Vis

(CH₃CN) l_max(e) 408 (23900), 380 (23800), 360 (20100), 284 (72600),

237 nm (46900 m ¹cm ¹); anal. calcd (%) for C₄₈H₃₂N₆OsP₂F₁₂(M_r

1172.9 gmol ¹): C 49.15; H 2.75; N 7.16; found: C 49.02; H 2.68; N 7.09.

3378

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3378 $ 17.50+.50/0

Chem. Eur. J. 1999, 5, No. 11

Pyrene±Metal Polypyridine Dyads

3366±3381

ꢄ

Pyr± Pt ± pyrpyr± Pt ± Cl: A Schlenk flask was charged with 1-ethynyl-

pyrene (0.05 g, 0.22 mmol) and freshly distilled tetrahydrofuran (20 mL).

The flask was cooled to 788C and n-BuLi (0.14 mL, 0.22 mmol) was

slowly added through a syringe. The solution was stirred for 30 min before

addition of SnMe₃Cl (0.066 g, 0.33 mmol). The pale yellow solution was

warmed to 208C and stirred for an additional two hours, after which was

added trans-[Pt(PnBu₃)₂Cl₂] (0.148 g, 0.22 mmol) and CuI (0.006 g,

0.0031 mmol). The mixture was maintained overnight at 608C before

removal of the solvent under vaccum. The crude precipitate was first

chromatographed on silica gel eluting with a gradient of dichloromethane

in hexane from 50 to 100%. The isolated product was dissolved in a mixture

of hexane/CH₂Cl₂1:1 and filtered over alumina. The recovered precipitate

was a mixture of pyr± Pt ± Cl (94%) and pyr± Pt ± pyr (6%). For pyr± Pt ±

pyr: ¹H NMR (CDCl₃): d 8.85 (d, 2H, ³J 9.1 Hz), 8.07 (m, 16H), 2.31

(m, 12H, P-CH₂-(CH₂)₂-CH₃), 1.47 (m, 24H, P-CH₂-(CH₂)₂-CH₃), 0.98 (t,

nÄ 2977 (m), 2926 (m), 2082 (s, C C-Pt), 1654 (m), 1458 (m), 1262 (m),

1088 (m), 1047 (m), 880 (s), 803 (m); FAB (m-NBA) m/z: 1081.5 [M

H] ; anal. calcd (%) for C₅₉H₇₃P₂N₃Pt (M_r 1081.29 gmol ¹): C 64.54; H

6.80; N 3.89; found: C 65.29; H 6.59; N 3.65.

[Ruthenium(ii) bis(2,2'-bipyridine)(trans-{pyrene-1-yl-ethynyl}-{2,2'-bipy-

ridine-5-yl-ethynyl}bis{n-butylphosphane}platinum)](PF₆)₂[pyr± Pt ± Ru]:

Following the procedure described above for preparation of pyr± Ru,

reaction of pyr± Pt ± bpy (0.030 g, 0.029 mmol) and cis-[Ru(bpy)₂Cl₂] ´

2H₂O (0.015 g, 0.029 mmol) afforded the red-orange pyr± Pt ± Ru

(0.028 g, 55%). ¹H NMR (CDCl₃): d 8.63 (d, 1H, ³J 9.1 Hz), 7.87 (m,

31H), 2.04 (m, 12H, P-CH₂-(CH₂)₂-CH₃), 1.38 (m, 24H, P-CH₂-(CH₂)₂-

CH₃), 0.83 (t, 18H, ³J 7.0 Hz, P-(CH₂)₃-CH₃); ³¹P{¹H} NMR (CDCl₃): d

6.63 (pseudot, J_P,Pt 2321 Hz); FT-IR (KBr pellet, cm ¹): nÄ 2926 (m),

ꢄ

2885 (m) 2084 (m, C C-Pt), 1593 (m), 1453 (m), 1387 (m), 1314 (m), 1110

(s), 846 (m), 769 (m), 725 (m); FAB (m-NBA) m/z: 1453.5 [M Cl], 1417.5

18H, J 7.3 Hz, P-(CH₂)₃-CH₃); ¹³C{¹H} NMR (CDCl₃): d 131.6, 131.5,

[M 2Cl]; UV/Vis (CH₃CN) l_max(e) 389 (47800), 379 (49700), 361

131.1, 129.1, 128.6, 127.4, 126.9, 126.5, 126.2, 125.8, 124.8, 124.6, 124.5, 124.4,

(39900),

280 nm

(133200 m ¹cm ¹);

anal.

calcd

(%)

for

116.3 (C C), 108.5 (C C), 26.6 ± 23.9 (P-(CH₂)₃-CH₃), 13.9 (CH₃); ³¹P{¹H}

ꢄ

C₇₄H₈₆N₆RuPtP₄F₁₂C₇H₈CH₃CN (M_r 1707.5892.1441.05 gmol ¹):

NMR (CDCl₃): d 4.89 (pseudot, J_P,Pt 2351 Hz); FT-IR (KBr pellet,

C 54.16; H 5.31; N 5.33; found: C 54.26; H 5.41; N 5.69.

ꢄ

cm ): nÄ 3033 (m), 2926 (m), 2864 (s), 2077 (s, C C-Pt), 1654 (m), 1591

[Osmium(ii)

bis(2,2'-bipyridine)(trans-{pyrene-1-yl-ethynyl}{2,2'-bipyri-

(m), 1451 (m), 1276 (m), 1215 (m), 1180 (m), 1085 (m), 1050 (s), 966 (m),

dine-5-yl-ethynyl} bis{n-butylphosphane}platinum)](PF₆)₂[pyr± Pt ± Os]:

Following the procedure described above for preparation of pyr± Os,

reaction of pyr± Pt ± bpy (0.030 g, 0.029 mmol) and cis-[Os(bpy)₂Cl₂]

(0.017 g, 0.029 mmol) afforded the red-orange pyr± Pt ± Os (0.011 g,

21%); ¹H NMR (CDCl₃): d 8.63 (d, 1H, ³J 9.1 Hz), 7.87 (m, 31H),

2.05 (m, 12H, P-CH₂-(CH₂)₂-CH₃), 1.40 (m, 24H, P-CH₂-(CH₂)₂-CH₃), 0.86

(t, 18H, ³J 7.1 Hz, P-(CH₂)₃-CH₃); ³¹P{¹H} NMR (CDCl₃): d 4.36

(pseudot, J_P,Pt 2310 Hz); FT-IR (KBr pellet, cm ¹): nÄ 2927 (s), 2863 (m),

886(m), 838 (s), 715 (s); FAB (m-NBA) m/z: 1049.3 [M H] ; UV/Vis

(CH₂Cl₂) l_max(e) 396 (91500), 383 (74600), 364 (53500), 285 (61600), 229

(81800), 213 (34200), 211 (33900), 208 (33600), 206 nm (33100 m ¹cm ¹);

anal. calcd (%) for C₆₀H₇₂P₂Pt (M_r 1050.28 gmol ¹): C 68.92, H 6.91;

found: C 68.83, H 6.80. For pyr± Pt ± Cl: ¹H NMR (CDCl₃): d 8.70 (d, 1H,

³J 9.1 Hz), 8.03 (m, 8H), 2.10 (m, 12H, P-CH₂-(CH₂)₂-CH₃), 1.47 (m,

24H, P-CH₂-(CH₂)₂-CH₃), 0.91 (t, 18H, ³J 7.3 Hz, P-(CH₂)₃-CH₃); ³¹P{¹H}

NMR (CDCl₃): d 7.95 (pseudot, J_P,Pt 2368 Hz). FT-IR (KBr pellet,

ꢄ

2083 (s, C C-Pt), 1588 (m), 1459 (m), 1104 (m), 1088 (s), 842 (m), 764 (m);

ꢄ

cm ): nÄ 2926 (s), 2098 (s, C C-Pt), 1654 (m), 1596 (m), 1458 (m), 1262

FAB (m-NBA) m/z: 1651.6 [M PF₆], 1507.6, [M H 2PF₆]; UV/Vis

(m), 1227 (m), 1088 (s), 1048 (s), 880 (m), 848 (m), 801 (m), 719 (m); FAB

(CH₃CN) lmax (e) 471 (12900), 389 (43900), 378 (46800), 283 (73 300),

226 (46400), 226 nm (48600 m ¹cm ¹); anal. calcd (%) for

C₇₄H₈₆N₆OsPtP₄F₁₂C₇H₈(M_r 1796.7 92.14 gmol ¹): C 51.51; H 5.02;

N 4.45; found: C 51.13; H 4.82; N 4.53.

(m-NBA) m/z: 860.3 [M H] ; UV/Vis (CH₂Cl₂) l_max(nm) (e) 390

(39600), 379 (52800), 360 (43800), 284 (45600), 226 (64900), 212 (35170),

203 nm (31200 m ¹cm ¹); anal. calcd (%) for C₄₂H₆₃P₂PtCl (M_r

860.46 gmol ¹): C 58.63; H 7.38; found: C 58.56; H 7.49.

[Ruthenium(ii) (2,2':6',2''-terpyridine)(4'-{1-ethynyl-pyrene}-2,2':6',2''-ter-

pyridine] (PF₆)₂[pyr± Ru(terpy)]: A stirred solution of [Ru(terpy)(dm-

so)Cl₂] (0.06 mmol, 0.032 g) and AgBF₄(0.13 mmol, 0.025 g) in Ar-purged

methanol (20 mL) was heated at 808C for eight hours. After the reaction

mixture was cooled to room temperature, the deep-red solution was

transferred through a cannula to a suspension of [4'-(1-ethylnyl-pyrene)-

terpy] (0.05 mmol, 0.025 g) in methanol (5 mL). After the reaction was

stirred for 16 hours, a solution of KPF₆(0.33 mmol, 0.060 g) in water

(10 mL) was added before removal of the solvent. The precipitate was

chromatographed on alumina, eluting with a gradient of methanol in

trans-(Pyrene-1-yl-ethynyl)-(2,2'-bipyridine-5-yl-ethynyl)-bis(tri-n-butyl-

phosphane)platinum(ii) (pyr± Pt ± bpy): The crude precipitate comprising

94% of pyr± Pt ± Cl (0.028 g, 0.033 mmol) was added to a mixture of

5-ethynyl-2,2'-bipyridine (0.006 g, 0.033 mmol), diisopropylamine (5 mL),

CuI (0.001 g, 0.005 mmol), and tetrahydrofuran (10 mL). The solution was

stirred at room temperature for 10 days before removal of the solvent

under vacuum. The crude precipitate was chromatographed on silica

eluting first with a gradient of dichloromethane in hexane from 50 to 100%

and then with a gradient methanol in dichloromethane from 0 to 5%. The

product was recrystallized from dichloromethane/hexane (76%). ¹H NMR

(CDCl₃): d 8.73 (d, 1H, ³J 9.1 Hz), 8.67 (d, 1H, ³J 3.9 Hz), 8.62 (s,

1H), 8.37 (1H, d, ³J 7.9 Hz), 8.27 (d, 1H, ³J 7.9 Hz), 7.99 (m, 9H), 7.85 (t,

1H, ³J 7.6 Hz), 7.69 (d, 1H, ³J 7.6 Hz), 2.20 (m, 12H, P-CH₂-(CH₂)₂-

CH₃), 1.49 (m, 24H, P-CH₂-(CH₂)₂-CH3), 0.94 (t, 18H, ³J 7.3 Hz,

P-(CH₂)₃-CH₃); ¹³C{¹H} NMR (CDCl₃): d 156.2, 151.5, 151.2, 149.1,

138.2, 136.8, 131.5, 131.4, 130.9, 128.9, 128.6, 127.4, 126.8, 126.5, 126.3, 125.7,

dichloromethane from 0 to 10% (0.040 g, 68%). H NMR ([D₆]acetone):

d 9.47 (brs, 3H), 8.99 (4line multiplet, 5H), 8.36 (16line multiplet, 16H),

7.93 (d, 3H, ³J 5.14 Hz), 7.42 (m, 3H); ¹³C{¹H} NMR (CD₃CN): d 158.6,

156.3, 153.8, 139.4, 139.2, 133.5, 131.9, 131.4, 130.5, 130.3, 128.8, 128.2, 127.9,

127.5, 126.4, 126.0, 125.8, 125.2, 124.8, 116.3, 97.2 (CC_ethynyl), 93.1 (CC_ethynyl);

ꢄ

FT-IR (KBr pellet, cm ): nÄ 3423 (m), 2927 (m), 2192 (m, C C), 1605 (s),

1470 (s), 1415 (m), 1241 (m), 1111 (s), 1029 (m), 843 (s);. FAB (m-NBA)

m/z: 937.2 [M PF₆] , 792.0 [M 2PF₆] , 591.0 [M 2PF₆pyr] , 567.0

ꢄ

124.8, 124.5, 124.3, 123.1, 120.8, 120.2, 115.7 (C C), 115.2 (C C), 108.5

(C C), 106.3 (C C), 23.4 ± 26.7 (P-(CH₂)₃-CH₃), 13.9 (CH₃); ³¹P{¹H} NMR

(CDCl₃): d 4.24 (pseudot, J_P,Pt 2341 Hz); FT-IR (KBr pellet, cm ¹):

ꢄ

[M 2PF₆pyrC C] ; UV/Vis (CH₃CN) l_max(e) 511 (44800), 411

(22200), 385 (31100), 365 (28200), 309 (47000), 281 (53200), 277 nm

(49100 m ¹cm ¹); anal. calcd (%) for C₄₈H₃₀N₆RuP₂F₁₂CH₃CN (M_r

1081.841.1 gmol ¹): C 53.48; H 2.96; N 8.73; found: C 53.26; H 3.02; N

8.59.

ꢄ

nÄ 2957 (m), 2928 (m), 2864 (s), 2093 (s, C C-Pt), 1713 (m), 1631 (m), 1592

(m), 1453 (m), 1262 (m), 1215 (m), 1103 (m), 906 (m), 850 (s), 803 (m), 721

(m), 618 (s); FAB (m-NBA) m/z: 1004.4 [M H]; UV/Vis (CH₂Cl₂) l_max

(e) 392 (48900), 380 (55300), 362 (53200), 344 (41800), 285 (44400), 227

(52700), 201 nm (31300 m ¹cm ¹); anal. calcd (%) for C₅₄H₇₀P₂N₂Pt (M_r

1004.2 gmol ¹): C 64.59, H 7.03, N 2.79; found: C 64.94, H 7.04, N 2.75.

Methods: Absorption spectra were recorded at ambient temperature with a

Kontron Instruments Uvikon 930 spectrophotometer. Luminescence

spectra were recorded in deoxygenated acetonitrile at 208C with

trans-(Pyrene-1-yl-ethynyl)-(2,2':2,6'-terpyridine-4'-yl-ethynyl)-bis(tri-n-

Perkin-Elmer LS50 spectrofluorimeter equipped with red-sensitive

butylphosphane)platinum(ii) (pyr± Pt ± terpy):

similar experimental

photomultiplier tube. The emission spectra were corrected for imperfec-

tions of the instrument by reference to a standard lamp. Only a small

correction factor was required for wavelengths between 550 and 800 nm

but this factor became progressively more important between 800 and

900 nm. Emission maxima were reproducible to within Æ5 nm. Quantum

yields were calculated relative to ruthenium(ii) and osmium(ii) tris(2,2'-

bipyridyl) complexes in acetonitrile^{[29, 56]}with dilute solutions after

deoxygenation by purging with argon. Luminescence quantum yields were

taken as the average of three separate determinations and were reprodu-

cible to within Æ8% but, because of the low emission probability for the

procedure was followed as that described for preparation of pyr± Pt ±

bpy: 4'-ethynyl-2,2':2,6'-terpyridine (0.044 mmol, 0.011 g), a mixture of

pyr± Pt ± Cl and pyr± Pt ± pyr containing 0.038 g of pyr± Pt ± Cl. Yield

0.029 g, 60%; ¹H NMR (CDCl₃): d 9.30 (m, 3H), 8.56 (d, 2H, ³J

10.4 Hz), 8.37 (s, 2H), 8.05 (m, 8H), 8.27 (td, 2H, ³J 7.6 Hz, ⁴J

1.8 Hz), 7.34 (dd, 1H, ³J 4.9 Hz, ⁴J 1.2 Hz), 7.30 (dd, 1H, ³J 4.9 Hz,

⁴J 1.2 Hz), 2.20 (m, 12H, P-CH₂-(CH₂)₂-CH₃), 1.51 (m, 24H, P-CH₂-

(CH₂)₂-CH₃), 0.96 (t, 18H, ³J 7.3 Hz, P-(CH₂)₃-CH₃); ³¹P{¹H} NMR

(CDCl₃): d 4.30 (pseudot, J_P,Pt 2336 Hz); FT-IR (KBr pellet, cm ¹):

Chem. Eur. J. 1999, 5, No. 11

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3379 $ 17.50+.50/0

3379

A. Harriman, R. Ziessel et al.

FULL PAPER

osmium(ii)-based chromophores, these derived quantum yields should be

regarded as being approximate (i.e., Æ25%). Luminescence spectra used

for curve-fitting analysis were recorded with a mode-locked argon-ion laser

emitting at 514 nm as the excitation source so that a narrow slit could be

used for the emission monochromator.

with a frequency-doubled, Q-switched Nd-YAG laser (FWHM 20 ns)

using the facilities available at the Paterson Institute for Cancer Research,

Manchester (England).

Curve fitting of corrected emission spectra followed the procedure

introduced by Meyer and co-workers.^[10]Briefly, luminescence spectra

recorded for the various reference compounds were corrected for spectral

distortions of the instrument, reduced so as to display L/n³versus n where L

is the luminescence intensity at wavenumber n and normalized. Each

spectrum was deconvoluted into the minimum number of Gaussian-shaped

bands needed to give a good representation of the complete spectrum with

the commercially available PEAKFIT program. From the indiviual

Gaussian components it was possible to identify i) the energy difference

between 0,0 vibronic levels in the triplet and ground states (E₀), ii) the

average half-width at half maximum (Dn_1/2) for the series of bands, and iii)

the average energy spacing between individual vibronic bands (ꢀhw).

Subsequently, the entire emission spectrum was constructed with MATH-

CAD V6 and compared to the experimental spectrum using SCIENTIST in

order to refine the parameters and to estimate the size of the electron-

vibrational coupling constant S.

Emission lifetimes were measured following excitation of the sample with a

25 ps laser pulse at 532 nm as delivered by a frequency-doubled, mode-

locked Nd-YAG laser. For some studies, the excitation pulse was Raman

shifted with perdeuterated cyclohexane to produce excitation wavelengths

of 598 and 465 nm. The laser intensity was attenuated to 5 mJ per pulse and

incident pulses were defocussed onto an adjustable pinhole positioned in

front of the sample cuvette. Luminescence was collected with a microscope

objective lens at 908 to excitation and isolated from any scattered laser light

with non-emissive glass cut-off filters. The emergent luminescence was

focussed onto the entrance slit of a Spex high-radiance monochromator and

thereby passed to a fast-response photodiode. The output signal was

transferred to a Tektronix SCD1000 transient recorder and subsequently to

a microcomputer for storage and analysis. Approximately 500 individual

laser shots, collected at 10 Hz, were averaged for kinetic measurements.

The temporal resolution of this instrument was about 200 ps but could be

improved to about 50 ps when needed by replacing the photodiode with a

cooled microchannel plate phototube. Emission lifetimes measured with

this setup were reproducible to within Æ10%. All kinetic measurements

were made with samples previously deoxygenated by purging with argon

and the absorbance of each solution was adjusted to be about 0.08 at

532 nm. Data analysis was made by a nonlinear, least squares iterative

PM3 RHF-SCF MO calculations for the S₀and T₁levels of the pyrene

derivatives were made with the MOPAC93 program package. For the

configuration interaction calculations of the T₁state, the two highest

occupied (HOMO and HOMO-1) and two lowest unoccupied (LUMO and

LUMO 1) orbitals were taken into consideration. Ab initio MO

calculations were made at the CIS/3 ± 21G level after energy-minimization

of the structure using the AMBER force field. Reduction potentials were

measured by cyclic voltammetry carried out in deoxygenated acetonitrile

containing tetra-N-butylammonium hexafluorophosphate (0.1m) as back-

ground electrolyte using a BAS CV-50W voltammetric analyzer. Pt wires

were used as both working and counter electrodes while a saturated

calomel electrode was used as reference. The system was calibrated against

a SCE separated from the solution by a presoaked glass frit and with

ferrocene as an internal reference. Reduction potentials were reproducible

to within 15 mV.

fitting routine that utilized

a modified Levenberg-Marquardt global

minimization procedure, after deconvolution of the instrument response

function.^[57]Certain studies were made following excitation with a 20 ns

laser pulse at 532 nm using the facilities available at the Paterson Institute

for Cancer Research, Manchester (England). The signal was detected with

a red-sensitive PMT and analyzed by computer iteration.

Transient differential absorption spectra were recorded after excitation of

the sample in deoxygenated acetonitrile with a 25 ps laser pulse at 532 nm.

Where appropriate the excitation pulse was Raman shifted with perdeu-

terated cyclohexane to produce excitation wavelengths of 598 and 465 nm

or frequency-doubled to 355 nm. The monitoring beam was provided by a

pulsed, high-intensity Xe arc lamp passed through the sample at 908 to the

excitation pulse. Spectra were compiled point-by-point, with five individual

records being collected at each wavelength, with a Spex high-radiance

monochromator operated with 2 nm slits. Kinetic measurements were

made at fixed wavelength, with 300 individual laser shots being averaged

for each decay profile. The time resolution of this setup was restricted to

about 3 ns by the risetime of the PMT but was improved to about 50 ps by

replacing the Xe monitoring beam with a pulse of white light generated by

focussing residual laser light into a mixture of D₂O/H₂O. The excitation

pulse was delayed with respect to that of the continuum with a computer-

controlled optical delay stage and the two pulses were directed almost

collinearly through the sample cell. The continuum pulse was split 50:50

before the sample cell so as to provide sample and reference beams. After

passing through the sample, these beams were collected by fiber optics and

analyzed with an image-intensified, Princeton dual-diode array spectro-

graph. The spectrometer was operated at 10 Hz, with 100 individual laser

shots being averaged at each delay time. Baseline corrections were applied

and emission was subtracted from the resultant spectra by recording

control signals without the excitation or continuum pulses. Differential

absorption spectra were corrected for distortions by reference to the optical

Kerr effect obtained from CS₂.

Acknowledgement

We thank the C.N.R.S., E.C.P.M., and the Royal Society of London for

their financial support of this work. We are greatly endebted to Johnson-

Matthey Ltd. for their generous loan of precious metal salts. Acknowl-

edgement is also made to Professor Hans Lami for kindly providing access

to the variable temperature emission spectrofluorimeter. The nanosecond

laser flash photolysis studies were made at the Paterson Institute for

Cancer Research, Manchester (England). We are most grateful to the staff

of the FRRF-PICR for providing access to this equipment and to the

E.E.C. for providing financial support to operate the facility and to defray

travel expenses.

[1] A. Harriman, RSC Specialist Periodical Reports: Photochemistry 1998,

29, Ch. 1 and 7.

[2] V. Balzani, A. Juris, M. Venturi, S. Campagna, S. Serroni, Chem. Rev.

1996, 96, 759.

[3] W. E. Ford, M. A. J. Rodgers, J. Phys. Chem. 1992, 96, 2917.

[4] G. J. Wilson, W. H. F. Sasse, A. W.-H. Mau, Chem. Phys. Lett. 1996,

250, 583.

Improved time resolution was achieved using a frequency-doubled, mode-

locked Antares 76S pumped dual-jet dye laser operated with pyromethene

dye. The 565 nm output beam was split into two parts with approximately

80% and 20% of the total intensity, respectively. The most intense beam

was used as the excitation source (FHWM 350 fs) while the weaker beam

was depolarized and focussed into a 1 cm cuvette filled with water to

produce a white light continuum for use as the analyzing pulse. The

continuum was split into two equal beams before reaching the delay stage

so as to provide a reference beam by which to normalize the transient

absorption spectrum. This reference beam arrived at the sample cell about

1 ns before the excitation and analyzing beams; with the latter two pulses

passing almost collinearly through the sample. Detection and data analysis

was made as mentioned above. Certain flash photolysis studies were made

[5] G. J. Wilson, A. Launikonis, W. H. F. Sasse, A. W.-H. Mau, J. Phys.

Chem. 1997, 101, 4860.

[6] C. Weinheimer, Y. Choi, T. Caldwell, P. Gresham, J. Olmsted III, J.

Photochem. Photobiol. A 1994, 78, 119.

[7] S. Boyde, G. P. Strouse, W. E. Jones Jr., T. J. Meyer, J. Am. Chem. Soc.

1989, 111, 7448.

[8] N. B. Thornton, K. S. Schanze, New J. Chem. 1996, 20, 791.

[9] P. Belser, R. Dux, M. Book, L. De Cola, V. Balzani, Angew. Chem.

1995, 107, 634; Angew. Chem. Int. Ed. Engl. 1995, 34, 595.

[10] a) Z. Murtaza, A. P. Zipp, L. A. Worl, D. K. Graff, W. E. Jones Jr.,

W. D. Bates, T. J. Meyer, J. Am. Chem. Soc. 1991, 113, 5113; b) Z.

Murtaza, D. K. Graff, A. P. Zipp, L. A. Worl, W. E. Jones Jr., W. D.

Bates, T. J. Meyer, J. Phys. Chem. 1994, 98, 10504.

3380

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3380 $ 17.50+.50/0

Chem. Eur. J. 1999, 5, No. 11

Pyrene±Metal Polypyridine Dyads

3366±3381

[11] R. Ziessel, M. Hissler, A. El-ghayoury, A. Harriman, Coord. Chem.

Rev. 1998, 178 ± 180, 1251.

[40] J. F. Endicott, M. A. Watzky, X. Song, T. Buranda, Coord. Chem. Rev.

1997, 159, 295.

[12] J. A. Simon, S. L. Curry, R. H. Schmehl, T. R. Schatz, P. Piotrowiak, X.

Jin, R. P. Thummel, J. Am. Chem. Soc. 1997, 119, 11012.

[13] V. Grosshenny, A. Harriman, R. Ziessel, Angew. Chem. 1995, 107,

2921; Angew. Chem. Int. Ed. Engl. 1995, 34, 2705.

[14] A. C. Benniston, A. Harriman, V. Grosshenny, R. Ziessel, New J.

Chem. 1997, 21, 405.

[41] According to the nature of the bridge and the theoretical treatment

used to analyze the kinetic data, values for V_DAranging between 0.4

and 60 cm have been reported for electron-exchange reactions

involving ¹Ru(bpy)ª fragments. For further details see: refs. [10], [13],

[16], and [35] and citations therein.

[42] S. Larsson, J. Am. Chem. Soc. 1983, 103, 4034.

[15] A. Harriman, M. Hissler, R. Ziessel, A. De Cian, J. Fischer, J. Chem.

Soc. Dalton Trans. 1995, 4067.

[16] V. Grosshenny, A. Harriman, M. Hissler, R. Ziessel, J. Chem. Soc.

Faraday Trans. 1996, 92, 2223.

[43] a) M. Beer, J. Chem. Phys. 1956, 25, 745; b) Y. Nagano, T. Ikoma, K.

Akiyama, S. Tero-Kubota, J. Phys. Chem. 1998, 102, 5769.

[44] A. Harriman, M. Hissler, A. Khatyr, R. Ziessel, Chem. Commun.

1999, 735.

[17] V. Grosshenny, A. Harriman, F. M. Romero, R. Ziessel, J. Phys. Chem.

1996, 100, 17472.

[18] A. Harriman, M. Hissler, R. Ziessel, Phys. Chem. Chem. Phys. 1999, 1,

4203.

[45] For example, attaching electron donating or withdrawing groups to

the terpy ligand can increase t_Tfrom 0.56 to 11 ns, see ref. [46].

Attaching one or more phenyl rings to the 4'-position of the terpy

ligand gives a t_Tof a few ns, see ref. [31]. Attaching a vacant terpy

ligand to the coordinated terpy ligand (t_T 8 ns) and subsequent

complexation of protons, see ref. [47] or zinc cations, see ref. [48],

respectively, gives t_Tvalues of 80 and 75 ns at room temperature.

Separating the coordinated and vacant terpy ligands by an ethynylene

group increases t_Tto 55 ns, see ref. [31a] whilst complexation of zinc

cations raises this to 180 ns, see ref. [13]. The ethynylene-bridged

binuclear ¹Ru(terpy)ª complex has a t_Tof 565 ns, see ref. [13] while

the corresponding phenylene-linked binuclear complex has a t_Tof

only 4 ns, see ref. [49]. When the phenyl ring is isolated from the terpy

ligands by an ethynylene group t_Tincreases to 90 ns but replacing this

central phenyl ring with a 2,2'-bipyridyl group (t_T 100 ns) and

subsequent complexation of zinc cations gives a t_Tof 210 ns, see ref.

[50]. The longest triplet lifetime reported for any ¹Ru(terpy)ª

fragment at room temperature is the 1220 ns found for an etheny-

lene-bridged binuclear complex, see ref. [14].

[46] E. Amouyal, M. Mouallen; G. Calzaferri, J. Phys. Chem. 1991, 96,

7641.

[47] F. Barigelletti, L. Flamigni, M. Guardigli, J.-P. Sauvage, J.-P. Collin, A.

Sour, Chem. Commun. 1996, 1329.

[48] F. Barigelletti, L. Flamigni, G. Calogero, L. Hammarström, J.-P.

Sauvage, J.-P. Collin, Chem. Commun. 1998, 2333.

[49] L. Hammarström, F. Barigelletti, L. Flamigni, M. T. Indelli, N.

Armaroli, G. Calogero, M. Guardigli, A. Sour, J.-P. Collin, J.-P.

Sauvage, J. Phys. Chem. A 1997, 101, 9061.

[19] F. M. Romero, R. Ziessel, Tetrahedron Lett. 1995, 36, 6471.

[20] K. T. Potts, D. Konwar, J. Org. Chem. 1991, 56, 4815.

[21] V. Grosshenny, F. M. Romero, R. Ziessel, J. Org. Chem. 1997, 62, 1491.

[22] M. Hissler, R. Ziessel, J. Chem. Soc. Dalton Trans. 1995, 893.

[23] a) F. Felix, J. Ferguson, H. U. Güdel, A. Ludi, J. Am. Chem. Soc. 1980,

102, 4096; b) J. Ferguson, E. Krausz, Inorg. Chem. 1987, 26, 1383.

[24] a) A. Harriman, R. Ziessel, Chem. Commun. 1996, 1717; b) A.

Harriman, R. Ziessel, Coord. Chem. Rev. 1998, 171, 331.

[25] A. Harriman, F. M. Romero, R. Ziessel, A. C. Benniston, J. Phys.

Chem. A 1999, 103, 5399.

[26] I. R. Gould, D. Noukakis, L. Gomez-Jahn, R. H. Young, J. L. Good-

man, S. Farid, Chem. Phys. 1993, 176, 439.

[27] J. Cortes, H. Heitele, J. Jortner, J. Phys. Chem. 1994, 98, 2527.

[28] R. Englman, J. Jortner, Mol. Phys. 1970, 18, 145.

[29] J. V. Caspar, E. M. Kober, B. P. Sullivan, T. J. Meyer, J. Am. Chem.

Soc. 1982, 104, 630.

[30] M. Striles, L. J. Cline Love, Anal. Chem. 1980, 52, 1559.

[31] a) A. C. Benniston, V. Grosshenny, A. Harriman, R. Ziessel, Angew.

Chem. 1995, 106, 1956; Angew. Chem. Int. Ed. Engl. 1994, 33, 1884;

b) F. Barigelletti, L. Flamigni, V. Balzani, J.-P. Collin, J.-P. Sauvage, A.

Sour, E. C. Constable, A. M. W. Cargill-Thompson, J. Am. Chem. Soc.

1994, 116, 7692.

[32] J. R. Winkler, T. L. Netzel, C. Creutz, N. Sutin, J. Am. Chem. Soc. 1987,

109, 2381.

[33] D. L. Dexter, J. Chem. Phys. 1953, 21, 836.

[50] A. El-ghayoury, A. Harriman, M. Hissler, R. Ziessel, Angew. Chem.

1998, 110, 1804; Angew. Chem. Int. Ed., 1998, 37, 1717.

[51] a) S. Boyde, G. F. Strouse, W. E. Jones Jr., T. J. Meyer, J. Am. Chem.

Soc. 1990, 112, 7395; b) A. J. Downard, G. E. Honey, L. F. Phillips, P. J.

Steel, Inorg. Chem. 1991, 30, 2260; c) G. F. Strouse, J. R. Schoonover,

R. Duesing, S. Boyde, W. E. Jones Jr., T. J. Meyer, Inorg. Chem. 1995,

34, 473.

[34] a) J. Ulstrup, J. Jortner, J. Chem. Phys. 1975, 63, 4358; b) G. Orlandi,

S. Monti, F. Barigelletti, V. Balzani, Chem. Phys. 1980, 52, 313;

c) E. M. Kober, B. P. Sullivan, T. J. Meyer, Inorg. Chem. 1984, 23, 2098.

[35] a) L. Y. Liang, A. I. Baba, W. Y. Kim, S. J. Atherton, R. H. Schmehl,

J. Phys. Chem. 1996, 100, 18408; b) J. R. Shaw, G. S. Sadler, W. F.

Wacholtz, C. K. Ryu, R. H. Schmehl, New J. Chem. 1996, 20, 749.

[36] J. P. Bryne, E. Y. McCoy, I. G. Ross, Aust. J. Chem. 1965, 18, 1589.

[37] K. Razi Naqvi, C. Steel, Spectrosc. Lett. 1993, 26, 1761.

[38] a) At room temperature, [Ru(terpy)₂]²is very weakly emissive and

the triplet lifetime, measured by transient absorption spectroscopy,

has been reported as 250 ps in water, see ref. [32] while values of

565 ps, see ref. [13] or ca. 1 ns, see ref. [38b] have been found in

acetonitrile. The fast rate of radiationless deactivation is attributed, in

part, to mixing between the MLCT triplet and a metal-centered state

lying at higher energy. At 77 K, the triplet lifetime increases to ca.

10 ms, see ref. [13]; b) M. Maestri, N. Armaroli, V. Balzani, E. C.

Constable, A. M. W. Cargill-Thompson, Inorg. Chem. 1995, 34, 2759.

[39] a) H. Sumi, R. A. Marcus, J. Chem. Phys. 1985, 84, 4272. b) J. Jortner,

M. Bixon, J. Chem. Phys. 1988, 88, 167; c) M. Bixon, J. Jortner, Chem.

Phys. 1993, 176, 467.

[52] D. R. Coulson, Inorg. Synth. 1972, 13, 121.

[53] G. B. Kauffman, L. A. Teter, Inorg. Syn. 1963, 7, 245.

[54] G. Sprintschnik, H. W. Sprintschnik, P. P. Kirsch, D. Whitten, J. Am.

Chem. Soc. 1977, 99, 4947.

[55] E. M. Kober, J. V. Caspar, B. P. Sullivan, T. J. Meyer, Inorg. Chem.

1988, 27, 4587

[56] E. M. Kober, J. L. Marshall, W. J. Dressick, B. P. Sullivan, J. V. Caspar,

T. J. Meyer, Inorg. Chem. 1985, 24, 2755.

[57] H. Lami, T. Piermont, Chem. Phys. 1992, 163, 149.

Received: March 11, 1999 [F1668]

Chem. Eur. J. 1999, 5, No. 11

ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999

0947-6539/99/0511-3381 $ 17.50+.50/0

3381

Products guided by the article

Product name:1-ETHYNYLPYRENE

Cas No:34993-56-1

R&D Labs maybe for 34993-56-1

Jiangsu Wanlong Chemical Co., Ltd.

website:http://www.wanlongchem.com

Contact:+86-511-86810993 0086-511-8681 0888;

Address:Quanzhou Town, Danyang City, Jiangsu
Zhengzhou Gecko Scientific Inc.

Contact:0371-88884176

Address:56 Hongzhuan Road, Zhengzhou, China
Ji'an Kexin Trade Co., Ltd.(expird)

Contact:86-0796-8187704 18507063190

Address:ji'an jiangxi
Dezhou Longteng Chemical Co., Ltd.

website:http://www.sodium-methoxide.cn/

Contact:0086-18866052283

Address:Xinhua Industrial Zone, Dezhou City, Shandong Province, China
MENGNA

Contact:18698110882

Address:1303 No2 building,LuoMa Garden,YongAn Road,Hexi District,Tianjin city

Relevant to this article

Doi:10.1021/ja00770a055
(1972)
Doi:10.1002/jsfa.2740091007
(1958)
Doi:10.1021/jo01028a516
(1964)
Three-dimensional porphyrin-based covalent organic frameworks with tetrahedral building blocks for single-site catalysis

Doi:10.1039/c9nj04017j
(2019)
Structure Evolution and Hydrogenation Performance of IrFe Bimetallic Nanomaterials

Doi:10.1021/acs.langmuir.5b04566
(2016)
Highly efficient, green, rapid, and chemoselective oxidation of sulfur-containing compounds in the presence of an MCM-41@creatinine@M (M = la and Pr) mesostructured catalyst under neat conditions

Doi:10.1039/c7nj05189a
(2018)

Article Doi

Intramolecular triplet energy transfer in pyrene-metal polypyridine dyads: A strategy for extending the triplet lifetime of the metal complex

DOI: 10.1002/(SICI)1521-3765(19991105)5:11<3366::AID-CHEM3366>3.0.CO;2-I

Source and publish data:

Authors:

Article abstract of DOI:10.1002/(SICI)1521-3765(19991105)5:11<3366::AID-CHEM3366>3.0.CO;2-I

Full text of DOI:10.1002/(SICI)1521-3765(19991105)5:11<3366::AID-CHEM3366>3.0.CO;2-I

Products guided by the article

R&D Labs maybe for 34993-56-1

Relevant to this article

Hot Product