The synthesis of 4′-aryl substituted terpyridines by Suzuki cross-coupling reactions: Substituent effects on ligand fluorescence

Article abstract of DOI:10.1039/b205330f

Several 4′-aryl-substituted 2,2′:6′,2″-terpyridines (tpy-C₆H₄R) have been prepared by palladium-catalysed cross-coupling of 4′-bromoterpyridine or 4′-triflate-terpyridine (triflate = trifluoromethylsulfonyloxy) with aryl boronic acids or esters, RC₆H₄B(OR′)₂ (R = H, m-NH₂, p-CHO, -NO₂, -CN, -NMe₂, -NPh₂). The new ligand 4′-mesityl-terpyridine (mesityl = 2,4,6-trimethylphenyl) was prepared in the same way. Similarly, 4′-bromophenylterpyridine (tpy-φ-Br) has been cross-coupled with aryl halides to generate several new biaryl-substituted terpyridines (tpy-φ-C₆H₄R where R = H, p-CN, NMe₂, NPh₂), together with two related compounds with pendent 3- or 4-pyridyl groups (tpy-φ-C₆H₄-py). For selected compounds, the alternative coupling strategy of reaction of a terpyridine-4-boronate or terpyridine-4-phenylboronate with the appropriate aryl halide has also been investigated (e.g. to prepare tpy-φ-C₆H₄NO₂), but was generally found to be less satisfactory. All of the compounds are fluorescent in the UV region of the spectrum, the biaryl-substituted compounds being only slightly red-shifted compared to the monoaryl systems, but with the further red-shift that accompanies protonation being more significant for the former. Fluorescence lifetimes in solution are in the range 1-5 ns. The emission spectra of the aminobiphenyl-substituted compounds (tpy-φ-C₆H₄NR″₂, where R″ = Me or Ph) display a large red-shift with increasing solvent polarity, suggesting the involvement of an intramolecular charge transfer state, as found previously for the two analogues omitting the phenyl ring (tpy-C₆H₄NR″₂). In contrast to the latter, however, protonation or binding of a Lewis acidic metal ion to the aminobiphenyl compounds serves to quench almost completely their emission.

Full text of DOI:10.1039/b205330f

View Article Online / Journal Homepage / Table of Contents for this issue
The synthesis of 4Ј-aryl substituted terpyridines by Suzuki cross-
coupling reactions: substituent effects on ligand fluorescence
Wendy Goodall, Kerstin Wild, Kathryn J. Arm and J. A. Gareth Williams*
2
Department of Chemistry, University of Durham, South Road, Durham, UK DH1 3LE.
E-mail: j.a.g.williams@durham.ac.uk
Received (in Cambridge, UK) 31st May 2002, Accepted 1st August 2002
First published as an Advance Article on the web 3rd September 2002
Several 4Ј-aryl-substituted 2,2Ј:6Ј,2Љ-terpyridines (tpy-C6H4R) have been prepared by palladium-catalysed cross-
coupling of 4Ј-bromoterpyridine or 4Ј-triflate-terpyridine (triflate = trifluoromethylsulfonyloxy) with aryl boronic
acids or esters, RC6H4B(ORЈ)2 (R = H, m-NH2, p-CHO, -NO2, -CN, -NMe2, -NPh2). The new ligand 4Ј-mesityl-
terpyridine (mesityl = 2,4,6-trimethylphenyl) was prepared in the same way. Similarly, 4Ј-bromophenylterpyridine
(tpy-ꢀ-Br) has been cross-coupled with aryl halides to generate several new biaryl-substituted terpyridines (tpy-ꢀ-
C6H4R where R = H, p-CN, NMe2, NPh2), together with two related compounds with pendent 3- or 4-pyridyl groups
(tpy-ꢀ-C6H4-py). For selected compounds, the alternative coupling strategy of reaction of a terpyridine-4-boronate
or terpyridine-4-phenylboronate with the appropriate aryl halide has also been investigated (e.g. to prepare tpy-ꢀ-
C6H4NO2), but was generally found to be less satisfactory. All of the compounds are fluorescent in the UV region of
the spectrum, the biaryl-substituted compounds being only slightly red-shifted compared to the monoaryl systems,
but with the further red-shift that accompanies protonation being more significant for the former. Fluorescence
lifetimes in solution are in the range 1–5 ns. The emission spectra of the aminobiphenyl-substituted compounds
(tpy-ꢀ-C6H4NRЉ2, where RЉ = Me or Ph) display a large red-shift with increasing solvent polarity, suggesting the
involvement of an intramolecular charge transfer state, as found previously for the two analogues omitting the phenyl
ring (tpy-C6H4NRЉ2). In contrast to the latter, however, protonation or binding of a Lewis acidic metal ion to the
aminobiphenyl compounds serves to quench almost completely their emission.
stabilisation and red-shift of the ICT state and hence to
Introduction
intriguing potential as sensory systems for these metals.6 Com-
The complexation chemistry of polypyridyl ligands has been a
vibrant area of research for several years. The intense interest in
such ligands and their complexes with transition metal ions is
due especially to their potential utility in a range of applica-
tions, for example as luminescent chemosensors, as photo-
catalysts, as components of devices for the conversion of light
into electrical energy, in new electroluminescent materials, and
even in information processing and storage in the germinating
area of molecular electronics and photonics.1a–e Apart from
the ubiquitous bidentate ligands 2,2Ј-bipyridine and 1,10-
phenanthroline, the ligand 2,2Ј:6Ј,2Љ-terpyridine has been
employed in a large number of studies, usually binding as a
tridentate ligand. The structural simplicity of bis-terpyridyl
complexes compared to their tris-bipyridyl analogues has been
widely noted2,3 (e.g. the absence of complicating factors such as
chirality or facial and meridional isomers), and this simplicity is
retained upon introduction of substituents into the 4Ј-positions
of the terpyridines, which are necessarily disposed in a trans
configuration to one another along a C2 axis.
Aryl substituents at the 4Ј-position exert a profound influ-
ence on the luminescence of bis-terpyridyl ruthenium()
complexes, and their effect on the wavelength and lifetime of
emission has been investigated in some detail.4 More recently,
the photophysical properties of uncomplexed terpyridines
have also begun to attract interest.4c,5,6 Whilst the majority of
those studied to date emit in the near ultra-violet region of the
spectrum, terpyridines substituted with electron-donating
aminophenyl substituents have been shown to possess low-
energy intramolecular charge transfer (ICT) excited states,
leading to highly solvent-dependent emission in the visible
region.5,6 We have demonstrated recently that complexation of
metal ions (e.g. Zn2ϩ or Cd2ϩ) to such terpyridines leads to a
plexes incorporating electron-rich terpyridines of this type have
also been reported recently to offer interesting non-linear
optical properties,7 whilst the incorporation of a biphenyl
substituent into the 4Ј-position has been used to influence π-
stacking interactions in the solid state.8 Clearly then, strategies
for the incorporation of aryl substituents into the 4Ј-position of
the terpyridine core are of considerable interest with respect to
many of the applications discussed above.
The “traditional” routes to 4Ј-aryl terpyridines are based on
the Tschitschibabin pyridine synthesis, involving the condensa-
tion of two equivalents of 2-acetylpyridine with the appropriate
aryl aldehyde, with formation of the central pyridine ring being
effected by reaction of the diketone intermediate with ammonia
(or a source thereof ) at elevated temperatures.9 A number of
variations have been reported over the 50 years since 4Ј-phenyl-
terpyridine was first prepared in this way,10a some involving
isolation of the intermediate enone or diketone,10b,c others
employing mild or solvent-free conditions for parts of the syn-
thesis,10c but almost all relying on a common step in which the
central pyridine ring is formed. Such methodology relies on the
availability of the appropriate aryl aldehyde, and is not com-
patible with base-sensitive functionality. Distinctly different
synthetic approaches, in which all three pyridine rings are
present in the starting materials, include the reaction of 2,2Ј-
bipyridine with 2-pyridyl-lithiums and the Stille coupling of
2,6-dibromopyridines with 2-trialkyltin pyridines.11 Although
this latter procedure has proved popular for 6,6Љ- and 4Ј-(non-
aryl)-substituted systems,11 it has not been used to provide
access to 4Ј-arylterpyridines, probably owing to the relative
inaccessibility of 4-aryl-2,6-dibromopyridines.12
This report describes some investigations into the utility of
the Suzuki cross-coupling reaction of aryl boronic acids with
DOI: 10.1039/b205330f
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681
This journal is © The Royal Society of Chemistry 2002
1669

View Article Online
aryl bromides in derivatising terpyridines at the 4Ј position.†
Recently, we reported on the preparation of terpyridine-4Ј-
phenylboronic acid, and on cross-coupling reactions of metal
complexes of this ligand with aryl bromides; complexes incor-
porating 4Ј-bromoterpyridine similarly underwent coupling
with aryl boronic acids.14 Here, results on free terpyridines are
presented, offering access to a range of new and established
ligands, and the relative merits of the two distinct coupling
strategies considered (i.e. which of the two coupling partners
carries the boronic acid). The photophysical properties of the
series of substituted ligands so prepared are also described.
group on the boronic acid increases the nucleophilicity of the
boron and hence the rate of transmetallation, making it more
competitive with the side-reactions discussed above. There
was no evidence for aryl–aryl exchange between the palladium
centre and the triphenylphosphine ligand to give 4-(dimethyl-
amino)biphenyl, although such competing reactions have
sometimes been reported to become significant in the presence
of electron-donating substituents in the aryl boronic acid.15,19
B
Couplings with 4Ј-bromoterpyridine. Given the poor
results with 4Ј-triflate-terpyridine, attention was turned to the
use of 4Ј-bromoterpyridine in its place: bromo-aromatics are
usually more reactive and more robust than their triflate
analogues. 4Ј-Bromoterpyridine (tpy-Br) may be prepared
directly from 2,6-bis(2-pyridyl)-4(1H)-pyridone by bromination
with POBr3;20 in our hands, however, this reaction required
large quantities of the brominating agent, long reaction times
and rarely gave a yield of more than 50% (significantly lower
upon scaling-up). In contrast, we have found that treatment of
4Ј-triflate-terpyridine with a solution of HBr in acetic acid at
110 ЊC leads to complete conversion into 4Ј-bromoterpyridine
within 4 h, and this procedure was therefore used preferentially.
As is evident from Table 1, 4Ј-bromoterpyridine gave greatly
superior results in cross-couplings with aryl boronic acids
compared to the triflate. Fewer side-products were formed,
facilitating purification. In line with several reports that strong
bases favour couplings with sterically crowded boronic acids,21
barium hydroxide was employed in place of sodium carbonate
in the preparation of the novel mesityl-substituted ligand L8,
which was obtained in 61% yield after recrystallisation from
ethanol. The same strategy was also successful in preparing the
4Ј-biphenyl-substituted ligand, L9, and p-substituted biaryl
analogues, L10–L15, by using 4Ј-(4-bromophenyl)terpyridine in
place of 4Ј-bromoterpyridine (Scheme 1).
Results and discussion
Synthesis of ligands by cross-coupling
A
Couplings with 4Ј-triflate-terpyridine. 4Ј-Triflate-
terpyridine is readily prepared from 2,6-bis(2-pyridyl)-4(1H)-
pyridone in around 70% yield.13b Initial investigations focused
on the use of this compound in cross-couplings with aryl
boronic acids or esters. Although sometimes more capricious
than their bromo or iodo analogues, many aryl triflates have
been employed successfully in cross-couplings of the Suzuki
type.15 Conditions frequently used for triflate couplings were
adopted, namely tetrahydrofuran as the solvent, Pd(PPh3)4 as
catalyst, and the presence of sodium carbonate to favour form-
ation of the boronate anion (added as a saturated aqueous
solution). For example, these conditions have been reported to
give good results for a triflate-substituted pyrrole.16
Although ligands L1–L4 were successfully prepared in this
way from 4Ј-triflate-terpyridine (Scheme 1), the yields of crude
product, as estimated from 1H NMR, were, at best, modest, and
reaction times for complete consumption of starting material
were long (24–72 h). Moreover, the contaminating materials
included significant amounts of triphenylphosphine and,
apparently, mixtures of tetraarylphosphonium salts, which
proved troublesome to separate from the terpyridine products,
both by crystallisation and by chromatography. The final
isolated yields were, therefore, disappointing, as evident from
Table 1. Purification could be achieved more readily in the case
of L2, since acidification led to a water-soluble tricationic
compound: impurities were removed by washing an acidic
aqueous solution with dichloromethane, and the terpyridine
subsequently isolated by basicification and back-extraction into
dichloromethane.
Although frequently well-behaved, unsatisfactory behaviour
of triflates in cross-couplings is not uncommon. A number of
studies on a variety of substrates have reported problems, often
attributed to decomposition of the catalyst and irreversible
reaction of the liberated triarylphosphine with the triflate
substrate, leading to phosphonium salts.17 In some instances,
phosphine-free palladium catalysts, including palladium acet-
ate and derivatives, are able to catalyse cross-coupling reactions
efficiently, whilst eliminating the problems of phosphine-
derived side-products.18 However, there was no evidence for the
formation of the desired products when such conditions were
employed for reactions of 4Ј-triflate-terpyridine (reaction in
acetone in the presence of 0.2 mol% palladium acetate18).
Only for the p-dimethylamino-substituted ligand, L6, was a
satisfactory yield obtained using triflate-terpyridine (Table 1).
The reaction proceeded more rapidly in this case, as monitored
by consumption of tpy-OTf on TLC, and the product was
amenable to purification using an acidification–wash–
basicification sequence. It is likely that the electron-donating
C
Couplings with terpyridine-boronic esters. Although
access to many aryl boronic acids is straightforward by the
“classical” methodology of lithiation of the analogous bromo-
aromatic and subsequent reaction with a trialkoxyboron
{usually B(OMe)3 or B(OiPr)3}, this procedure is unsuitable for
aryl groups incorporating sensitive functionality (for example,
those which would react with BuLi). In contrast, palladium-
catalysed coupling of the bromo-aromatic with tetraalkoxy-
diboron compounds offers valuable entry to such systems,
a procedure which has rapidly gained popularity since its
inception by Miyaura in 1995 (Scheme 2).22 The method tends
to be less reliable in the presence of strongly electron-
withdrawing groups, however, as they often favour competitive
homo-coupling of the starting bromo-aromatic to generate
biphenyl side-products. This undesirable effect was encountered
in this work in the preparation of the p-nitrobenzeneboronate
ester required in the synthesis of L4 (Table 1): p-nitroiodo-
benzene consistently gave a 3 : 1 mixture of the desired com-
pound and 4,4Ј-dinitrobiphenyl. The synthesis of pyridine
boronates (for L14 and L15) by this method was similarly
affected. Moreover, these boronic esters, once prepared, did not
undergo coupling as efficiently as those with electron-donating
substituents, as is evident from the isolated yields (e.g. entries
L4 and L14 in Table 1), probably due to a lower stability of the
electron-deficient boronic acids with respect to deboronation.23
The “reverse” coupling strategy as a route to aryl-substituted
terpyridines was therefore investigated, namely use of a
terpyridine-4Ј-boronate (or terpyridine-4Ј-phenylboronate) as
a common synthon for reaction with bromo-aromatics, to avoid
reliance on the availability of boronic acids or an efficient
pathway to them. We have previously reported the synthesis of
4Ј-(4-neopentylglycolatoboron)terpyridine (tpy-Bneo) and its
analogue with an interposed phenyl group, (tpy-ꢀ-Bneo), using
the Miyaura coupling with bis(neopentylglycolato)diboron
(B2neo2).14 Subsequent work with different batches of B2neo2
† Related Sonogashira couplings have been reported by others for the
introduction of alkynyl functionality into the 4Ј position,13a as have
Stille coupling reactions of 4Ј-triflate-terpyridine; e.g. with vinyltri-
butyltin to give 4Ј-vinylterpyridine,13b and with tetramethyltin to provide
4Ј-methylterpyridine.13c
1670
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681

View Article Online
Scheme 1
and of the Pd(dppf )Cl2 catalyst has revealed that formation of
nature of the pyridine ring. In an attempt to suppress this
undesirable side-reaction, some alternative conditions were
assessed for the preparation of tpy-Bneo
(a) Use of dioxane as the solvent, in place of dimethyl sulf-
oxide. Dioxane is generally found to be a less active solvent in
tpy-ꢀ-Bneo may frequently be accompanied by a small amount
of the homo-coupled, “back-to-back” bis-terpyridine (Scheme
2), sometimes up to 10 mol%. Again, as for the bromopyridines,
this is probably due to the inductively electron-withdrawing
:
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681
1671

View Article Online
Table 1 Routes to terpyridines L1–L15: coupling partners, solvent and isolated yieldsa
Ligand
L1
Starting terpyridine
tpy-OTf
Reacting aryl partner
Solvent
THF
Isolated yield (%)
10
tpy-Br
tpy-OTf
THF
THF
80
14
L2
L3
L4
tpy-OTf
THF
13
tpy-Br
tpy-OTf
DME
THF
75
9b
tpy-Br
tpy-Bneo
DME
THF
30
64
L5
tpy-Br
THF
THF
THF
DME
80
34
65
59
tpy-Bneo
tpy-OTf
tpy-Br
L6
L7
L8
tpy-Br
DME
61
L9
tpy-ꢀ-Br
tpy-ꢀ-Br
tpy-ꢀ-Bneo
tpy-ꢀ-Br
tpy-ꢀ-Br
DME
DME
DME
DME
DME
60
57
28
43
60
L10
L11
L12
L13
L14
tpy-ꢀ-Br
DME
DME
36
4
c
tpy-ꢀ-Bneo
L15
tpy-ꢀ-Br
DME
DME
67
46
tpy-ꢀ-Bneo
a Solvent employed was tetrahydrofuran (for the couplings involving triflates) or dimethoxyethane (for most reactions employing aryl bromides or
iodides). The catalyst was Pd(PPh3)4, and sodium carbonate was used as the base in each case, except for L8 where barium hydroxide was employed.
b Traces of PPh3 (< 5%) remained in the product, despite repeated recrystallisations. c The hydrochloride salt was used; immediate conversion into the
free pyridine is expected under the basic conditions employed.
cross-couplings, but has been used successfully in the reaction
of aryl triflates with bis(pinacolato)diboron.22b In the present
instance, no reaction was observed at reflux (100 ЊC), whilst a
reaction temperature of 150 ЊC under pressure led to a mixture
of the desired product, the homo-coupled terpyridine and
unsubstituted terpyridine, the latter presumably formed by
hydro-dehalogenation of the starting material under these
forcing conditions.
1672
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681

View Article Online
Scheme 2
(b) Use of Pd(CH3CN)2Cl2–tri-o-tolylphosphine as catalyst.
results, together with the problems in synthesising the terpyr-
This, too, led to contamination of the desired product by the
homo-coupled terpyridine, and proved to be inferior to
Pd(dppf )Cl2 in that only partial conversion of the starting
material was achieved, even after a reaction time of 24 h.
(c) Reaction in methanol in the presence of lithium methox-
ide and Pd(dppf )Cl2. At room temperature, these conditions
have been reported to suppress the homo-coupling of aryl
halides containing electron-withdrawing substituents, whilst
allowing the cross-coupling with the tetraalkoxodiboron to
proceed.24 However, no reaction was observed in the present
instance after 24 h. Increasing the temperature to 55 ЊC (18 h)
led to a complex mixture of the starting 4Ј-bromoterpyridine,
the desired product, the homo-coupled terpyridine, methoxy-
terpyridine, and unsubstituted terpyridine.
Of the conditions investigated, therefore, none offered any
advantage over those originally employed and, indeed, were less
successful. In fact, we have found that separation of the con-
taminating bis-terpyridine can be effected readily by exploiting
the water solubility of the boronate anion formed upon
addition of aqueous base (Scheme 2), and consequently the
original procedure clearly remains the preferred, with high
conversions and shorter reaction times (typically 5–6 h).
Subsequent cross-couplings of the terpyridine boronates
with aryl halides were successful. In the case of the 4Ј-nitro-
phenyl compound (L4), a higher yield was obtained in this
way (Table 1). In general, however, this strategy did not offer
improvements in terms of yields or reaction times; for example,
the 4Ј-cyanophenyl and 4Ј-(p-nitrobiphenyl) compounds (L5
and L11) and the pyridyl-substituted compounds L14 and L15
(pairs of entries in Table 1). In fact, in the case of L14, coupling
of the terpyridine-4Ј-phenylboronate with 4-bromopyridine
gave a very poor isolated yield, whereas the reverse coupling
with pyridine-4-boronic acid gave a satisfactory result, despite
the electron-withdrawing nature of the pyridine. Based on these
idine boronates and their tendency to pick up iron from the
dppf catalyst during their formation, the couplings of the 4Ј-
bromo and 4Ј-bromophenylterpyridines are clearly the more
attractive, in general.
NMR characterisation of the terpyridine products
1H NMR data for the ligands synthesised are collated in Tables
2 and 4 (for aryl- and biaryl-substituted ligands respectively).
Data for ligands L1–L4, L6 and L9 (which have been prepared
previously by conventional methodology9) were consistent with
the literature values, but our data at 500 MHz are included for
comparative purposes. Partial assignment of the spectra was
possible on the basis of splitting patterns and coupling con-
1
stants, whilst H–1H NOESY spectra proved most informative
in providing an unambiguous assignment of the resonances
of the pendent ring(s), owing to NOE between H3Ј and Hb (and
not Ha) (numbering system employed is shown in footnotes to
Tables 2 and 4). In none of the compounds was NOE observed
between H6 and H3Ј, which is consistent with a transoid
arrangement of the pyridine rings about the interannular C–C
bonds, in line with earlier observations and several solid-state
structures, including those of L6 and L8 obtained during this
work. A NOESY spectrum also allowed the two equally intense
singlets in the mesityl-substituted compound L8, namely those
due to H3Ј and Ha, to be distinguished. Thus, H3Ј displayed a
single cross-peak to the ortho methyl groups only (2.03 ppm,
6H), whereas Ha showed coupling to both the ortho and para
methyl groups. The resonances of the lateral pyridine rings (H2,
H3, H4, H5) and of the central pyridine ring (H3Ј) are scarcely
affected by changes in the substitution pattern in the 4Ј-aryl
pendent group, with the exception of the 4Ј-mesityl compound
L8, for which there is a large shift in the H3Ј resonance of about
0.5 ppm to low frequency. This may be rationalised in terms of
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681
1673

View Article Online
Table 2 1H NMR data for the phenyl-substituted ligands L1–L8 a
Ligand
L1
H3
H4
H5
H6
H3Ј
Ha
7.52
Hb
Hc
Other
—
8.68
d
7.90
t
7.37
dd
8.74
d
8.76
s
7.92
d
7.46
t
t
8.0
7.5
7.5, 5.0
6.5
—
7.5
7.0
7.0
L2
8.67
d
10.0
7.87
td
9.5, 1.5
7.34
dd
9.5, 6.0
8.73
d
6.0
8.70
s
—
7.28b
1H, m
7.29b
6.77
1H, dt
9.0, 2.5
7.22
1H, m
1H, br s
b
b
HdЈ
L3
L4
L5
L6
L7
L8
8.69
7.90
td
7.5, 2.0
7.37
ddd
7.5, 5.0, 1.5
8.73
d
4.5
8.77
s
—
8.06b
d
8.0
8.02b
d
8.0
—
—
—
—
—
—
10.05
(1H)
s (CHO)
d
8.1
8.67
7.90
td
7.5, 2.0
7.32
ddd
7.5, 4.5, 1.0
8.73
d
4.5
8.74
s
—
8.36
d
8.0
8.04
d
8.0
—
—
d
7.5
8.69
7.90
td
7.5, 2.0
7.32
ddd
6.0, 5.0, 1.5
8.73
d
5.0
8.74
s
—
7.82
d
10.0
8.00
d
8.5
d
8.0
8.68
7.88
7.36
ddd
7.5, 5.0, 1.0
8.75
d
5.0
8.74
s
—
6.83
d
9.0
7.90
3.05 (6H)
s (CH3)
d
8.0
td
d
c
c
8.67
7.87
td
8.0, 2.0
7.35
ddd
7.5, 5.0, 1.0
8.72
d
4.0
8.71
s
7.17
7.79
d
8.5
7.30 (4H) 7.17 (4H) 7.07
t (m-Ph) d (o-Ph) t (p-Ph)
7.5 c 7.5
d
7.5
d
c
8.64
7.83
td
7.5, 1.5
7.28
dd
7.0, 5.5
8.62
d
4.5
8.26
s
—
6.90
s
—
—
2.29 (3H) 2.03 (6H)
s (p-CH3) s (o-CH3)
—
d
7.5
1
a All spectra acquired in CDCl3 at 500 MHz and assigned using H–1H NOESY. δ in ppm relative to residual CHCl3 at 7.26 ppm; J in Hertz. All
signals integrate to two protons except where stated otherwise. b Signals too close to allow unambiguous assignment and, for L2, determination of
coupling constant. c Overlap of resonances prohibits reliable determination of coupling constants. Key to numbering:
the deshielding effect of the π-cloud of the 4Ј-pendent ring,
normally significant, being reduced as the mean torsion angle
between the pendent aryl ring and the central pyridine ring
increases. That the ortho methyl substituents of the pendent
mesityl group induce such a twisting is evident from the solid
state structure of L8, which reveals a torsion angle of 67.52(4)Њ,
compared to typical values of about 30Њ in most biphenyl struc-
tures omitting ortho substituents.‡ The effect is apparently
also transmitted to H6 and H5 to some extent, which appear at
slightly lower frequency than in the other compounds; indeed,
only for this compound does H6 appear to low frequency of H3.
The relative shifts of the protons in the pendent aryl group
are governed by the aryl substituent. For the unsubstituted
phenyl ligand L1 and all the biaryl systems L9–L15, the proton
ortho to the terpyridine (Hb) appears to high frequency of the
meta proton (Ha), reflecting the more strongly electron-
withdrawing nature of the terpyridine unit compared to a
phenyl group, with the difference being amplified by electron-
donating substituents at the para position (L6 and L7). In
contrast, the situation is reversed for the p-nitro substituted
compound, indicating that the electron-withdrawing influence
of the terpyridine is weaker than that of a nitro group. The
substituent and terpyridyl effects are almost balanced in the
formyl-substituted compound L3, where the chemical shifts are
approximately coincident. A similar trend is seen in the distal
ring of the biaryl-substituted compounds (Table 4), with HbЈ
at higher frequency in all cases, except for the electron-
withdrawing 4-nitro and 4-pyridyl systems, L11 and L14; the
balance between electron withdrawal by the substituent and
the 4Ј-phenylterpyridine fragment is attained in this case in the
cyano system (L10).
1
The 13C spectra, assigned with the aid of H–13C correlation
spectroscopy, revealed a similar striking homology in the
resonances of the pyridyl carbons; the most significant
deviation was again found in the mesityl system L8, in which
the C3Ј resonance displayed a modest shift of 4 ppm to high
frequency compared to the other compounds.
Absorption and fluorescence properties of the terpyridines
‡ Full details of the crystal structures of L8 and L6, determined by
X-ray diffraction, will be reported shortly elsewhere.
2,2Ј:6Ј,2Љ-Terpyridine itself absorbs strongly below about
320 nm, but is only weakly emissive, with a fluorescence
1674
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681

View Article Online
Table 3 13C NMR data for ligands L1–L8 a
Ligand
L1
C3
C4
C5
C6
C3Ј
Ca
Cb
Cc b
Quaternaries
Other
—
121.4
137.0
123.9
149.0
119.0
128.9
127.4
129.0
156.1, 155.8, 150.4,
138.4
L2
121.4
136.9
123.7
149.1
118.9
117.7c
129.8c
115.7
156.3, 155.8, 150.5, 113.8 (CdЈ)
146.9, 139.5
L3
L4
L5
L6
L7
L8
121.4
121.4
121.3
121.5
121.4
121.1
137.0
137.0
137.0
137.2
136.8
136.7
124.0
124.2
124.1
123.7
123.8
123.6
149.1
149.2
148.3
150.1
148.7
149.1
119.0
118.9
118.6
117.7
118.3
121.9
128.0c
124.2
132.7
112.4
129.4
128.0
130.3c
128.3
128.0
128.1
128.0
—
—
—
—
—
—
—
156.2, 155.8, 148.9, 192.0 (CHO)
144.4, 136.4
156.3, 155.6, 147.8,
144.9, 148.1
156.3, 155.7, 143.1, 112.6 CN
118.7d
156.3, 155.3, 151.1,
148.8, 125.0
40.4 (CH3)
156.3, 155.8, 149.7, 124.8, m-Ph 123.3, o-Ph
149.2, 147.4, 131.8
123.1, p-Ph
156.2, 155.5, 151.6,
137.0, 136.6, 134.9
20.9 (p-CH3) 20.6 (o-CH3)
a Spectra recorded in CDCl3 at 125.7 MHz; assigned using 1H–13C COSY, observing 1H at 499.8 MHz and decoupling 13C at 125.7 MHz. δ in ppm
relative to CDCl3 at 77.0 ppm. Numbering system employed as in footnote to Table 2. b Cc assigned only for L1 and L2 (Table 3) and CcЈ only for L9
and L15 (Table 5), where these are secondary carbon atoms. c Ca and Cb (Table 3) and CaЈ and CbЈ (Table 5) not assignable unambiguously, due to
coincidence or near-coincidence of corresponding protons in the 1H spectrum. d ipso C–CN not detected. e Overlap of 1H resonances of ortho-Ph and
HaЈ prohibits unambiguous assignment of these resonances by 1H–13C correlation spectroscopy.
maximum of 338 nm.4c As discussed recently by Araki and
co-workers,5 the introduction of aryl substituents into the 4Ј
position of the terpyridine unit can lead to quite dramatic
effects on the fluorescence intensities and, in some cases, the
wavelength of emission. In line with their observations, the 4Ј-
aryl-substituted terpyridines prepared here display longer wave-
length and more intense emission than those of unsubstituted
terpyridine (apart from the nitro-substituted system L4, which is
scarcely emissive). The absorption and emission properties in
solution in ethanol are collated in Table 6. Fluorescence life-
times have been recorded where possible, with values in the
range 1–5 ns, quite typical of conjugated aromatic molecules.
Comparison of the behaviour of 4Ј-phenylterpyridine L1 with
the newly prepared 4Ј-mesityl analogue L8 provides very strong
evidence in support of the need for attainment of an approxi-
mately coplanar conformation in the excited state in order to
favour fluorescence, as the emission of L8 is less intense than
that of L1 by a factor of 20.
The fluorescence properties of the amino-functionalised
ligands L6, L7, L12 and L13 are quite different from those of the
other ligands. The strong solvent dependence of emission from
L6 was reported recently by Araki,5 and interpreted in terms of
an intramolecular charge transfer (ICT) excited state: polar
solvents are able to stabilise such charge-separated excited
states, leading to large red shifts in the emission, as observed.
These observations were reinforced by the results of a semi-
empirical MO calculation, which showed the influence of the
amino substituent in raising the energy level of the pendent
phenyl to such an extent that the πPh—π*tpy excited state occurs
at lower energy than the “locally excited” πtpy—π*tpy state.5 We
have reported that the diphenylamino substituted system L7
displays comparable behaviour but that, in contrast to L6,
the intensity of emission remains high in ethanol (Table 7).6
This effect was attributed to the much lower basicity of the
diphenylamino pendent, and hence its weaker capacity to act as
a hydrogen bond acceptor from ethanol solvent molecules.6 The
newly prepared ligands L12 and L13 incorporate the same pair of
units as L6 and L7 (terpyridine unit and an amino-substituted
phenyl group) but with an interposed phenyl ring. In fact, the
data in Table 7 reveal strikingly that the two new systems
behave in a manner which is fully in line with the earlier ana-
logues. The absorbance maxima are again almost independent
of solvent, consistent with initial excitation to a locally excited
state as normally observed for ICT compounds. In a given
solvent, the emission maximum of both L12 and L13 occurs at
rather longer wavelength than for the corresponding analogue
L6 or L7 respectively in the same solvent, indicative of stabilisa-
tion of the excited state (of the order of 1000 cmϪ1) as a result
of the additional phenyl ring. However, a very similar pattern to
the effect of increasing polarity emerges. The dimethylamino-
substituted compounds experience the larger overall effects
(e.g. L12 shifts from 395 nm in cyclohexane to 567 nm in
ethanol, compared to 404 to 545 nm for L13), with the change
The introduction of a p-formyl (L3) or m-amino substituent
(L2) leads to a large reduction in the intensity of the emission
relative to 4Ј-phenylterpyridine, though with no significant
effect on the excited state energy, whilst a p-cyano substituent
(L5) gives rise to a small red-shift in absorption and emission,
and an increase in the emission intensity. Interest in the photo-
physics of the biphenyl-substituted terpyridine L9 and its metal
complexes has been awakened following the recognition of dif-
ferent possible π-stacking interactions in the solid-state, leading
to distinct effects on the excited states.8 The additional phenyl
ring in L9 compared to L1 leads to only a small red-shift in the
emission, although the intensity is significantly enhanced.
Again, a p-nitro substituent has a deleterious effect on emission
(L11), although to a lesser extent than in L4. The p-cyanophenyl
and 4- and 3-pyridyl substituted analogues (L10, L14 and L15)
reveal a small blue-shift in the emission maximum and some-
what longer fluorescence lifetimes.
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681
1675

View Article Online
Table 4 1H NMR data for the biphenyl-substituted ligands L9–L15 a
Ligand
L9
H3
H4
H5
H6
H3Ј
Ha
Hb
HaЈ
HbЈ
HcЈ
Other
8.69 7.89
d
7.9
7.37
dd
8.74
d
4.0
8.80 7.75 8.01 7.49
s
—
7.68
t
7.5
7.39
t, 1H
7.5
—
—
—
td
d
8.3
d
8.4
t
7.5
7.6, 1.7 6.4, 4.5
L10
L11
L12
L13
8.68 7.89
d
8.0
7.37
dd
7.0, 5.0
8.74
d
4.5
8.77 7.72 8.01 7.74
s
—
—
—
—
—
t
7.5
d
8.5
d
8.5
s (4H)
—
8.69 7.92
d
7.9
7.38
ddd
8.75
d
8.79 7.77 8.04 8.33
s
—
7.82
d
6.8
td
d
8.4
d
8.6
d
8.8
7.8, 1.8 7.4, 4.9, 1.1 4.7
8.68 7.89
d
7.9
7.36
dd
6.8, 5.0
8.75
d
4.5
8.79 7.71 7.97 6.84
s
—
7.60
d
8.7
3.02 (CH3)
s (6H)
—
t
7.4
d
8.2
d
8.2
d
8.5
8.69 7.89
7.36
ddd
8.74
d
8.79 7.72 7.99 7.17
7.56
d
7.17, 4H, d
7.0c, o-Ph
7.29, 4H, t 8.5,
m-Ph
d
td
s
d
d
d
8.0
8.0, 2.0 7.5, 5.0, 1.0 4.0
—
8.5
8.0
7.5
8.5
7.05, t 7.0,
p-Ph
L14
8.70 7.90
7.38
dd
8.75
d
4.5
8.79 7.80 8.05 8.71
7.60
d
4.5
—
—
d
td
s
d
8.0
d
8.0
d
8.0c
7.5, 1.5 7.5, 4.5
—
5.5c
L15
8.70 7.90
7.38
ddd
8.75
ddd
8.80 7.75 8.05 7.42
7.97
8.64
8.94 (HdЈ)
d
td
s
d
d
dd (1H) dd (1H)
dd, 1H d (1H)
7.8
7.8, 1.5 7.5, 4.8, 1.2 4.8, 1.8
—
8.7
8.4
8.1, 4.5 7.8, 2.4, 1.8 4.8, 1.5 2.4
a Spectral details as in footnote (a) to Table 2. Key to numbering:
from cyclohexane to THF leading to an especially pronounced
effect for the former. Significantly, the intensity of emission of
the dimethylamino-substituted compound L12, just like its
analogue L6, is greatly reduced in ethanol compared to the
other solvents (factor of about 30 in ꢀf), whereas both of the
diphenylamino compounds retain strong emission in ethanol.
This reinforces our view that the weak emission of the former
pair may be related to the capacity of the –NMe2 nitrogen to
act as a good hydrogen bond acceptor from ethanol solvent
molecules, inhibiting the twisting of the diakylamino group
which is thought to accompany formation of the ICT state. The
trend in lifetimes of emission is similar for each compound,
increasing from cyclohexane through THF to DCM and then
decreasing in ethanol.
L9, L11, L14 and L15, where the maxima are shifted to 410–
440 nm with an increase in intensity, leading to strong, blue
emission (Fig. 1), and corresponds to the recently reported
effect of coordination of Lewis acidic metal ions to L9.8
The behaviour of the amino-functionalised systems to pro-
tonation is again very different from that of the other com-
pounds. Protonation of L6 and L7 leads to the appearance of
a new, long wavelength band in the absorption spectrum
(430 nm), as reported previously, whilst coordination of zinc
leads to a very similar effect.6 Emission is further red-shifted
(by about 100 nm), attributed to stabilisation of the ICT state
by the coordinated metal ion, with a moderate decrease in
intensity. In contrast, acidification of the newly prepared
analogues, L12 and L13, leads to a long tail in the absorption
spectrum, extending to about 500 nm, rather than a well-
defined, new long wavelength band. Binding of zinc ions,
on the other hand, does lead to such a band (Fig. 2). For
both Zn2ϩ and Hϩ, the emission intensity is greatly reduced,
though again red-shifted, with a scarcely detectable signal
centred around 700 nm. Presumably, the additional rotations
and vibrations that are possible in these biaryl systems lead
to much more efficient non-radiative deactivation of the
low-energy ICT state. In terms of potential applications as
sensors or in photoactive metal complexes, therefore, the new
amino-functionalised ligands incorporating the additional
phenyl ring appear to be less attractive than those previously
described.6
Effect of protonation on emission. It is well-established that
terpyridine itself normally undergoes two successive proton-
ation processes upon acidification in solution, and that the
mono- and diprotonated species formed are characterised by
red-shifted and more intense emission compared to the free-
base.4c,27 Addition of trifluoroacetic acid (> 2 equiv.) to the
4Ј-monoaryl-substituted compounds (L1–L5) leads to small
red-shifts in the emission, with variable effects on intensity (an
initial increase at low concentrations of acid was pronounced
only for ligands L2 and L3, with all compounds displaying a
reduction in intensity under highly acidic conditions). This red-
shifting effect is more pronounced for the 4Ј-biaryl compounds
1676
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681

Table 5 13C NMR data for the biphenyl-substituted ligands L9–L15 a
Ligand
L9
C3
C4
C5
C6
C3Ј
Ca
Cb
CaЈ
CbЈ
CcЈb
Quaternaries
Other
—
121.4 136.9 123.9 149.2 118.7 127.6 127.7 128.9
127.2
127.6
156.3, 156.0, 149.8, 141.9, 140.5, 137.3
L10
L11
L12
L13
L14
L15
121.4 136.9 123.9 149.1 118.7 127.7 128.0 132.6c 127.6c
—
156.1, 156.0, 149.2, 144.8, 139.6, 138.7d
156.1, 156.0, 149.1, 147.2, 146.8, 139.2, 139.0
156.3, 155.9, 150.2, 149.9, 141.9, 135.7, 127.7
156.3, 155.9, 149.8, 147.5, 141.3, 136.9, 136.7, 134.1
156.1, 150.3, 149.4, 147.6, 139.3, 138.6
111.2 (–CN)
121.4 137.0 123.9 149.2 118.7 127.9 128.1 124.2
121.3 136.8 123.8 149.1 118.5 126.5 127.6 112.7
127.9
128.2
—
—
—
40.5 (–CH3)
121.4 136.9 149.8 149.1 118.6 127.0 127.5 124.5e 126.7
—
129.3, m-Ph 123.8, o-Phe 123.0, p-Ph
121.6 137.0 124.0 149.2 118.7 127.5 128.1 121.6
121.4 136.9 124.0 149.4 118.6 127.5 128.0 123.6
121.4
134.3
—
—
149.1
156.1, 156.0, 148.8, 138.5, 138.2, 135.9
148.2 (CdЈ)
a Spectral details as in footnote to Table 3 and atom numbering as in footnote to Table 4.

View Article Online
Table 6 Photophysical data for the ligands L1–L15 a
Absorption
λmax/nm (log10 ε)
Emission
λmax/nm
b
Ligand
L1
I/Iphtpy
τ/nsc
282 (4.57)
339, 351
348
1.00
1.2
L2
280 (4.52) 308(sh) (4.12)
0.03
—
d
d
L3
L4
L5
L6
L7
L8
283 (4.56)
339, 352
368
0.10
< 0.01
1.59
—
d
289 (4.49)
—
277 (4.62) 320(sh) (3.98)
291 (4.56) 348 (4.48)
290 (4.59) 360 (4.41)
284 (4.74)
365
3.0
531e
513e
339
0.11e
3.12e
0.05
2.6e
2.5e
d
—
L9
289 (4.80)
374
2.92
2.16
0.20
0.24e
1.88e
2.40
1.90
1.5
L10
L11
L12
L13
L14
L15
294(4.93)
346(sh) 359
367 382(sh)
567e
1.9
288 (4.80) 318 (4.61)
287 (4.81) 344 (4.28)
284 (4.50) 347 (4.32)
287 (4.53)
0.80
d
—
545e
1.5e
2.8
346(sh) 358
346(sh) 358
287 (4.59)
2.0
a In solution in ethanol at 293 3 K. b Integrated emission intensity relative to L1, upon excitation at 285 nm for isoabsorbant solutions, except for
those marked (e), which were excited at 350 nm. Spectra were acquired using excitation and emission band-passes of 2.5 nm in each case, and were
corrected for the wavelength dependence of the photomultiplier tube. The quantum yield of L1 was measured to be 0.08
0.02, using 2-
aminopyridine in 0.05 M H2SO4(aq) as standard, for which ꢀ = 0.60 0.05 for excitation at 285 nm.25 c Estimated uncertainty 0.2 ns. d Emission too
weak to allow measurement of the fluorescence lifetime using the available instrumentation. e For excitation at 350 nm.
ligands investigated all display evidence of an emissive ICT
excited state but the response to acid or metal binding is very
different according to the presence or absence of the interposed
phenyl ring.
Summary
In conclusion, palladium-catalysed, Suzuki-type cross-coupling
of aryl boronic acids with 4Ј-bromoterpyridine (or 4Ј-bromo-
phenylterpyridine) has been shown to be a viable and versatile
method for the preparation of 4Ј-aryl-substituted terpyridines.
Experimental
Although compounds with simple aryl substituents are more
readily prepared using conventional Kröhnke methodology, the
cross-coupling procedure allows the use of a common synthon,
to provide facile, divergent access to a range of systems under
mild conditions. In particular, 4Ј-biaryl-substituted systems are
also accessible in this way, overcoming the relative inaccess-
ibility of the appropriately functionalised biphenyl carboxalde-
hydes which would be required for a conventional approach.
The biaryl compounds have similar fluorescence properties to
their mono-aryl analogues, and display a significant red-shift
in emission upon protonation. The four amino-functionalised
Synthetic details
2,6-Bis(2Ј-pyridyl)-4(1H )-pyridone was prepared by base-
catalysed condensation of acetone with ethyl picolinate
followed by ring closure with ammonium acetate,28 and sub-
sequently converted to 4Ј-triflate-terpyridine upon reaction
with trifluoromethanesulfonic anhydride in pyridine.13b
4-(Diphenylamino)benzeneboronic acid neopentyl ester was
prepared as described previously,6 as were the terpyridine
boronates.14 Tetrakis(triphenylphosphine)palladium() was
1678
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681

View Article Online
Table 7 The effect of solvent on the photophysical properties of the amino-functionalised ligands L6, L7, L12, L13
Cyclohexane
THF
DCM
EtOH
L6
λmax/nm (abs)a
λmax/nm (em)b
τf/nsc
337
346
469
5.2
348
470
5.2
348
531
2.6
369, 385(sh)
1.7
d
ꢀf
0.31e
0.22
0.27
0.01
L7
λmax/nm (abs)a
λmax/nm (em)b
τf/nsc
356
392, 414(sh)
1.2
360
453
3.4
361
472
4.0
362
513
2.5
d
ꢀf
0.36
0.51
0.58
0.25
L12
λmax/nm (abs)a
λmax/nm (em)b
τf/nsc
335
350
496
2.5
350
504
2.7
348
567
—
395, 413(sh)
1.2
f
d
ꢀf
0.60e
0.52
0.64
0.02
L13
λmax/nm (abs)a
λmax/nm (em)b
τf/nsc
354
404, 425(sh)
1.2
352
472
2.3
353
494
2.9
347
545
1.5
d
ꢀf
0.60
0.40
0.43
0.15
a Position of longest wavelength absorption band. b Wavelength of maximum emission upon excitation into the lowest energy absorption band,
obtained after correction for the wavelength dependence of the photomultiplier tube. c Estimated uncertainty 0.2 ns. d Fluorescence quantum yields
were measured using an excitation wavelength of 350 nm and excitation and emission band-passes set to 2.5 nm in each case; quinine sulfate was used
as the standard (ꢀ = 0.546 in 1 M H2SO4);26 estimated uncertainty 25%. e Measured using an excitation wavelength of 340 nm: a steeply decreasing
absorbance profile at 350 nm in cyclohexane prohibits the reliable use of this wavelength. f Emission too weak to allow reliable determination of
lifetime.
Fig. 2 Absorption spectrum of L13 in ethanol (solid line), in the
presence of trifluoroacetic acid (10Ϫ2 mol dmϪ3, dashed line) and in the
presence of zinc triflate (5 equiv., dotted line); [L13] = 1.7 × 10Ϫ5 mol
dmϪ3
.
dimethoxyethane was dried over sodium wire. Proton and 13C
NMR spectra were recorded on a Varian 500 MHz instrument
(500 and 125.7 MHz respectively). Electron ionisation (EI) and
desorbed chemical ionisation (CI) mass spectra were measured
using a Micromass Autospec instrument. High resolution
spectra for accurate mass determinations were carried out
by electrospray ionisation at the EPSRC National Mass
Spectrometry Service Centre, Swansea.
Representative procedure for coupling of 4Ј-bromoterpyridine
or 4Ј-bromophenylterpyridine with an aryl boronic acid; L5 as the
example. 4Ј-Bromoterpyridine (583 mg, 1.87 mmol), 4-cyano-
benzeneboronic acid (302 mg, 2.06 mmol, 1.1 equiv.) and
dimethoxyethane or tetrahydrofuran (12 mL) were placed in an
oven-dried Schlenk tube. The solution was degassed via three
freeze–pump–thaw cycles (final p < 0.1 mm Hg), and then
placed under nitrogen. Sodium carbonate (590 mg, 5.57 mmol,
3.0 equiv.) was dissolved in the minimum volume of water in a
separate small Schlenk and degassed similarly. The catalyst,
Pd(PPh3)4 (90 mg, 0.08 mmol, 0.043 equiv.) was added to the
first Schlenk, followed immediately by the sodium carbonate
solution, transferred using a cannula. After stirring at room
temperature for 1 h, the temperature was increased to 85 ЊC for
Fig. 1 The effect of protonation on the absorption and emission
spectra of the cyanobiphenyl-substituted terpyridine L10. (a) Upper:
absorbance spectrum of L10 (10Ϫ5 mol dmϪ3 in ethanol) in the absence
of acid (solid line) and in the presence of trifluoroacetic acid at 10Ϫ4 mol
dmϪ3 (dashed line). (b) Lower: corrected fluorescence emission spectra
of L10 (10Ϫ6 mol dmϪ3 in ethanol) in the absence of acid (solid line) and
in the presence of acid (dashed line), upon excitation at the isosbestic
point (315 nm).
obtained by hydrazine reduction of a DMSO solution of PdCl2
in the presence of triphenylphosphine.29 Other reagents were
used as supplied from commercial sources. DMSO was dried
over 4 Å molecular sieves for at least 24 h before use and
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681
1679

View Article Online
Table
8
Mass spectral data for terpyridines synthesiseda and
elemental analyses for new compounds
range 10–250 MHz. A suspension of Ludox® in water was used
as the standard, which acts as a scattering sample of τ 0.0 ns.
L1 309 (Mϩ), 231 (Mϩ Ϫ py), 78 (Phϩ). L2 324 (Mϩ), 246 (Mϩ Ϫ py);
HRMS (EI): 324.1374 (Mϩ); calc. for C21H16N4, M: 324.1375. L3 337
(Mϩ). L4 354 (Mϩ), 324 (Mϩ Ϫ NO), 308 (Mϩ Ϫ NO2). L5 334 (Mϩ), 256
(Mϩ Ϫ py). L6 352 (Mϩ), 308 (Mϩ Ϫ NMe2); HRMS (EI): 352.1686
(Mϩ); calc. for C23H20N4, M: 324.1683; found C, 76.71, H, 5.58, N,
16.32; calc. for C23H20N4ؒ0.5H2O: C, 76.43, H, 5.86, N, 15.50%. L7
Acknowledgements
Financial support from the University of Durham and the
Royal Society is gratefully acknowledged. We thank the EPSRC
National Mass Spectrometry Service Centre, Swansea for
recording several high resolution mass spectra. K.W. is a
student of Universität Würzburg, Germany, and was supported
at Durham under the Socrates Exchange Scheme of the
E.U. We should also like to thank Frontier Scientific UK for
samples of boronic acids and our former colleague, Dr
Catherine J. Aspley, for pioneering studies on the terpyridine
boronates, which underpin much of the work reported here.
(using CI) 477 (M ϩ Hϩ). L8 351 (Mϩ), 350 (Mϩ Ϫ H), 335 (Mϩ
Ϫ
CH3), 273 (Mϩ Ϫ py); HRMS: 352.1812 (M ϩ Hϩ); calc. for C24H21N3,
M ϩ Hϩ: 352.1813; found C, 78.95, H, 5.90, N, 11.74; calc. for
C24H21N3ؒH2O: C, 78.00, H, 6.27, N, 11.37%. L9 385 (Mϩ), 307 (Mϩ Ϫ
py). L10 410 (Mϩ), 332 (Mϩ Ϫ py); HRMS: 411.1609 (M ϩ Hϩ); calc. for
C28H18N4, M ϩ Hϩ: 411.1609; found C, 80.24, H, 4.35, N, 13.38; calc.
for C28H18N4ؒ0.5H2O: C, 80.17, H, 4.57, N, 13.36%. L11 430 (Mϩ), 400
(Mϩ Ϫ NO), 384 (Mϩ Ϫ NO2), 352 (Mϩ Ϫ py), 322 (Mϩ Ϫ NO Ϫ py);
HRMS: 431.1501 (M ϩ Hϩ); calc. for C27H18N4O2, M ϩ Hϩ: 431.1508;
found C, 74.46, H, 4.09, N, 12.65; calc. for C27H18N4O2: C, 75.34, H,
4.21, N, 13.02%. L12 428 (Mϩ), 308 (Mϩ Ϫ C6H4NMe2); HRMS:
429.2079 (M ϩ Hϩ); calc. for C29H24N4, M ϩ Hϩ: 429.2079; found C,
75.01, H, 5.43, N, 11.63; calc. for C29H24N4ؒ2H2O: C, 74.98, H, 6.08, N,
12.05%. L13 (using CI) 553 (M ϩ Hϩ). L14 386 (Mϩ), 308 (Mϩ Ϫ py);
HRMS: 387.1610 (M ϩ Hϩ); calc. for C26H18N4, M ϩ Hϩ: 387.1609;
found C, 71.88, H, 5.18, N, 12.03; calc. for C26H18N4ؒ3H2O: C, 70.89, H,
5.49, N, 12.72%. L15 386 (Mϩ), 308 (Mϩ Ϫ py); HRMS: 387.1609 (M ϩ
Hϩ); calc. for C26H18N4, M ϩ Hϩ: 387.1609.
References
1 (a) J. M. Demas and B. A. DeGraff, Coord. Chem. Rev., 2001, 211,
317; (b) for a recent example: M. Kimura, A. Takahashi, T. Sakata
and K. Tsukahara, Bull. Chem. Soc. Jpn., 1998, 71, 1839;
(c) P. Péchy, F. P. Rotzinger, M. K. Nazeeruddin, O. Kohle,
S. M. Zakeeruddin, R. Humphry-Baker and M. Grätzel, J. Chem.
Soc., Chem. Commun., 1995, 65; (d ) M. A. Baldo, M. E. Thompson
and S. R. Forrest, Nature, 2000, 403, 750; (e) R. Ziessel, J. Inclusion
Phenom. Macrocyclic Chem., 1999, 35, 369.
2 E. C. Constable, A. M. W. Cargill Thompson, D. A. Tocher and
M. A. M. Daniels, New J. Chem., 1992, 16, 855.
3 J.-P. Sauvage, J.-P. Collin, J.-C. Chambron, S. Guillerez, C. Coudret,
V. Balzani, F. Barigelletti, L. De Cola and L. Flamigni, Chem. Rev.,
1994, 94, 993.
4 (a) M. Maestri, N. Armaroli, V. Balzani, E. C. Constable and
A. M. W. Cargill Thompson, Inorg. Chem., 1995, 34, 2759;
(b) L. Hammarström, F. Barigelletti, L. Flamigni, M. T. Indelli,
N. Armaroli, G. Calogero, M. Guardigli, A. Sour, J.-P. Collin and
J.-P. Sauvage, J. Phys. Chem. A, 1997, 101, 9061; (c) G. Albano,
V. Balzani, E. C. Constable, M. Maestri and D. R. Smith, Inorg.
Chim. Acta, 1998, 277, 225.
5 T. Mutai, J.-D. Cheon, S. Arita and K. Araki, J. Chem. Soc., Perkin
Trans. 2, 2001, 1045.
6 W. Goodall and J. A. G. Williams, Chem. Commun., 2001, 2514.
7 D. Roberto, F. Tessore, R. Ugo, S. Bruni, A. Manfredi and S. Quici,
Chem. Commun., 2002, 846.
8 N. W. Alcock, P. R. Barker, J. M. Haider, M. J. Hannon,
C. L. Painting, Z. Pikramenou, E. A. Plummer, K. Rissanen and
P. Saarenketo, J. Chem. Soc., Dalton Trans., 2000, 1447.
9 A comprehensive review of methods up to 1995 is provided by:
A. M. W. Cargill Thompson, Coord. Chem. Rev., 1997, 160, 1.
10 (a) R. L. Frank and E. F. Riener, J. Am. Chem. Soc., 1950, 72,
4182; (b) See, for example: P. Korall, A. Börje, P.-O. Norrby and
B. Åkermark, Acta Chem. Scand., 1997, 51, 760; (c) For example:
G. W. V. Cave and C. L. Raston, Chem. Commun., 2000, 2199.
11 (a) D. J. Cardenas and J.-P. Sauvage, Synlett, 1996, 916; (b) R.-A.
Fallahpour, M. Neuberger and M. Zehnder, New J. Chem., 1999, 23,
53; (c) R.-A. Fallahpour, Eur. J. Inorg. Chem., 1998, 1205.
12 The synthesis of 2,6-dibromo-4-phenylpyridine has been reported
only recently, and involves diazotisation of 4-amino-2,6-dibromo-
pyridine (itself requiring four steps for preparation from 2,6-
dibromopyridine), followed by reaction with benzene: J. Yuan,
G. Wang, K. Majima and K. Matsumoto, Anal. Chem., 2001, 73,
1869.
13 (a) V. Grosshenny, F. M. Romero and R. Ziessel, J. Org. Chem.,
1997, 62, 1491; (b) K. T. Potts and D. Konwar, J. Org. Chem., 1991,
56, 4815; (c) F. Odobel, J.-P. Sauvage and A. Harriman, Tetrahedron
Lett., 1991, 32, 5977.
14 C. J. Aspley and J. A. G. Williams, New J. Chem., 2001, 25, 1136.
15 N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457.
16 J. Cossy and D. Belotti, Tetrahedron, 1999, 55, 5145.
17 These problems are reviewed in reference 15. See also: T. Oh-e,
N. Miyaura and A. Suzuki, Synlett, 1990, 221; T. Oh-e, N. Miyaura
and A. Suzuki, J. Org. Chem., 1993, 58, 2201.
a Low resolution mass spectral data are provided for all compounds and
were obtained by electron ionisation, except where stated otherwise.
Accurate mass data are provided for those compounds which have not
been reported previously, and were obtained by electrospray ionisation,
except where marked EI, indicating electron ionisation. Elemental
analyses are given for new compounds; small deviations from calculated
values are consistent with the presence of retained water molecules,
confirmed by the 1H NMR spectra in anhydrous CDCl3.
24 h. The progress of the reaction could be monitored readily
by thin layer chromatography on alumina, eluant 70% hexane–
30% ethyl acetate, allowing consumption of the 4Ј-bromo-
terpyridine (Rf = 0.6) and formation of product (Rf = 0.2) to be
followed. (Gas chromatography was also suitable, except for
the terpyridines carrying amino functionality). After complete
consumption of starting material, the solvent was removed
under reduced pressure, and the crude residue dissolved in a
mixture of dichloromethane and water (25 mL of each). The
organic layer was separated, washed with dilute aqueous
sodium hydroxide (0.1 M, 3 × 30 mL) and dried over anhydrous
potassium carbonate. Removal of solvent led to a green-brown
residue which, after recrystallisation from ethanol, led to the
1
required product as a colourless solid (497 mg, 80%). H and
13C NMR data are given in Tables 2–5.
The ‘reverse’ coupling procedure, involving the reaction
of the terpyridine-4-boronate ester, or terpyridine-4-phenyl-
boronate ester, with an aryl halide, was carried out in the same
way, using a 10% excess of the boronate. Those reactions which
employed 4Ј-triflate-terpyridine in place of 4Ј-bromoterpyr-
idine were also performed similarly, but using tetrahydrofuran
as the solvent in place of dimethoxyethane.
Characterisation data for the terpyridines are given in Tables
2–5 (nuclear magnetic resonance) and Table 8 (mass spectra
and elemental analyses).
Absorbance and emission spectra and lifetimes
UV–visible absorbance spectra were recorded using a Bio-Tek
Instruments Uvikon-XS spectrometer using quartz cuvettes of
1 cm pathlength. Steady-state emission spectra were recorded in
1 cm pathlength cuvettes, using an Instruments S.A. Fluromax
equipped with a Hamamatsu R928 photomultipler tube;
spectra were corrected for the wavelength dependence of the
detector. Fluorescence lifetimes were measured using an
Instruments S.A. Fluorolog τ-3 instrument, by global fitting of
the demodulation and phase shift of the emission, following
excitation with sinusoidally-modulated light over the frequency
18 T. I. Wallow and B. M. Novak, J. Org. Chem., 1994, 59, 5034.
19 (a) B. E. Segelstein, T. W. Butler and B. L. Chenard, J. Org. Chem.,
1995, 60, 12; (b) K.-C. Kong and C.-H. Cheng, J. Am. Chem. Soc.,
1991, 113, 6313.
20 P. Péchy, F. P. Rotzinger, M. K. Nazeeruddin, O. Kohle,
S. M. Zakeeruddin, R. Humphry-Baker and M. Grätzel, J. Chem.
Soc., Chem. Commun., 1995, 65.
1680
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681

View Article Online
25 Principles of Fluorescence Spectroscopy, ed. J. R. Lakowicz, Kluwer
Academic/Plenum, New York, 2nd edn., 1999.
26 S. R. Meech and D. Phillips, J. Photochem., 1983, 23, 193.
27 A. Sarkar and S. Chakravorti, J. Lumin., 1995, 63, 143.
28 E. C. Constable and M. D. Ward, J. Chem. Soc., Dalton Trans., 1990,
1405.
21 T. Watanabe, N. Miyaura and A. Suzuki, Synlett, 1992, 207.
22 (a) T. Ishiyama, M. Murata and N. Miyaura, J. Org. Chem., 1995,
60, 7508; (b) T. Ishiyama, Y. Itoh, T. Kitano and N. Miyaura,
Tetrahedron Lett., 1997, 38, 3447.
23 C. Fischer and E. Havinga, Recl. Trav. Chim. Pays-Bas, 1965, 84,
439.
24 S. M. Marcuccio and H. Weigold, Int. Pat. Appl., PCT/AU00/01246.
29 D. R. Coulson, Inorg. Synth., 1972, 13, 121.
J. Chem. Soc., Perkin Trans. 2, 2002, 1669–1681
1681