Dark-field oxidative addition-based chemosensing: New bis-cyclometalated Pt(II) complexes and phosphorescent detection of cyanogen halides

Article abstract of DOI:10.1021/ja065645z

Heavy metal complexes that are phosphorescent at room temperature are becoming increasingly important in materials chemistry, principally due to their use in phosphorescent organic light-emitting devices (OLEDs). Their use in optical sensory schemes, however, has not been heavily explored. Homoleptic biscyclometalated Pt(II) complexes are known to undergo oxidative addition with appropriate electrophiles (principally alkyl halides) by either thermal or photochemical activation. We have applied this general reaction scheme to the development of a phosphorescence-based sensing system for cyanogen halides. To carry out structure-property relationship studies, a series of previously unreported Pt(II) complexes was prepared. Most of the complexes (excluding those that incorporated substituents on the ligands that forced steric crowding in the square plane) were strongly orange-red phosphorescent (Φ = 0.2-0.3) in a room-temperature oxygen-free solution. These sterically demanding ligands also accelerated the addition of cyanogen bromide to these complexes but slowed the addition of methyl iodide, indicating that the oxidative addition mechanisms for these two electrophiles is different. The lack of solvent-polarity effect on the addition of BrCN suggests a radical mechanism. Oxidative addition of BrCN to the metal complexes in solution or dispersed in poly(methyl methacrylate) gave blue-shifted emissive Pt(IV) complexes. The blue-shifted products give a dark-field sensing scheme that is in sharp contrast to energy transfer-based sensing schemes, which have limited signal-to-noise because of the presence of lower-energy vibronic bands of the energy donor that can overlap with the emission of the acceptor.

Full text of DOI:10.1021/ja065645z

Published on Web 12/07/2006
Dark-Field Oxidative Addition-Based Chemosensing: New
Bis-cyclometalated Pt(II) Complexes and Phosphorescent
Detection of Cyanogen Halides
Samuel W. Thomas III, Koushik Venkatesan, Peter M u¨ ller, and Timothy M. Swager*
Contribution from the Department of Chemistry, Massachusetts Institute of Technology, 77
Massachusetts AVenue, Cambridge, Massachusetts 02139
Received August 4, 2006; E-mail: tswager@mit.edu
Abstract: Heavy metal complexes that are phosphorescent at room temperature are becoming increasingly
important in materials chemistry, principally due to their use in phosphorescent organic light-emitting devices
(
OLEDs). Their use in optical sensory schemes, however, has not been heavily explored. Homoleptic bis-
cyclometalated Pt(II) complexes are known to undergo oxidative addition with appropriate electrophiles
principally alkyl halides) by either thermal or photochemical activation. We have applied this general reaction
(
scheme to the development of a phosphorescence-based sensing system for cyanogen halides. To carry
out structure-property relationship studies, a series of previously unreported Pt(II) complexes was prepared.
Most of the complexes (excluding those that incorporated substituents on the ligands that forced steric
crowding in the square plane) were strongly orange-red phosphorescent (Φ ) 0.2-0.3) in a room-
temperature oxygen-free solution. These sterically demanding ligands also accelerated the addition of
cyanogen bromide to these complexes but slowed the addition of methyl iodide, indicating that the oxidative
addition mechanisms for these two electrophiles is different. The lack of solvent-polarity effect on the addition
of BrCN suggests a radical mechanism. Oxidative addition of BrCN to the metal complexes in solution or
dispersed in poly(methyl methacrylate) gave blue-shifted emissive Pt(IV) complexes. The blue-shifted
products give a dark-field sensing scheme that is in sharp contrast to energy transfer-based sensing
schemes, which have limited signal-to-noise because of the presence of lower-energy vibronic bands of
the energy donor that can overlap with the emission of the acceptor.
Introduction
field induced by the metal-carbon bond, which raises the energy
of the nonemissive d-d metal centered transitions above the
triplet energy of the cyclometalated ligand.
The most commonly used metals in these complexes are
iridium and platinum. The presence of the heavy metal serves
to increase the intersystem crossing rate through spin-orbit
coupling and reduce the forbidden character of emission from
the triplet state of the ligand. The structures of the ligands are
typically optimized to enhance emission intensity and color
purity.
There has been an increasing interest in heavy metal
complexes that phosphoresce strongly at room temperature. The
principal reason behind this surge in research into novel and
efficient phosphors is the development of phosphorescence-
based organic light emitting devices (OLEDs), which can be
more efficient than fluorescence-based devices because the
majority of excitons created upon electron-hole recombination
1
are triplets. Phosphors are necessary to harness the energy of
these triplet excitons and convert it into useful light output.
Among the most popular classes of heavy metal complexes
used in phosphorescent OLEDs are those that are cyclometalated
with bidentate ligands, such as 2-phenylpyridine. These
complexes are often highly emissive in fluid solution in the
absence of oxygen. This property stems from the large ligand
In addition to their simple light-emitting properties, the
relatively underdeveloped bis-cyclometalated square-planar Pt-
(II) complexes can be highly reactive via oxidative addition to
2
give the corresponding Pt(IV) products. This work was pio-
neered by von Zelewsky and Balzani, who investigated the
oxidative addition reactions of simple bis-cyclometalated Pt-
3
(
II) complexes, such as Pt(ppy)2 or the strongly orange-emitting
(
1) Baldo, M. A.; O’Brien, D. F.; Thompson, M. E.; Forrest, S. R. Phys. ReV.
B. 1999, 60, 14422.
4
Pt(thpy)2 (Figure 1). They found that these complexes under-
went facile oxidative addition reactions with reactive electro-
philes, such as methyl iodine or benzyl bromide, under ambient
conditions, principally by an SN2-type mechanism, to give only
(2) (a) Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Nature (London) 2000,
4
03, 750. (b) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.;
Lee, H. E.; Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J.
Am. Chem. Soc. 2001, 123, 4304. (c) Brooks, J.; Babayan, Y.; Lamansky,
S.; Djurovich, P. I.; Tsyba, I.; Bau, R.; Thompson, M. E. Inorg. Chem.
2
002, 41, 3055. (d) Tsuboyama, A.; Iwawaki, H.; Furugori, M.; Mukaide,
5
one stereoisomeric Pt(IV) product out of a possible 11. They
T.; Kamatani, J.; Igawa, S.; Moriyama, T.; Miura, S.; Takiguchi, T.; Okada,
S.; Hoshino, M.; Ueno, K. J. Am. Chem. Soc. 2003, 125, 12971. (e) Cho,
J.-Y.; Suponitsky, K. Y.; Li, J.; Timofeeva, T. V.; Barlow, S.; Marder, S.
R J. Organomet. Chem. 2005, 690, 4090.
(3) Chassot, L.; M u¨ ller, E.; von Zelewsky, A. Inorg. Chem. 1984, 23, 4249.
(4) Chassot, L.; von Zelewsky, A. Inorg. Chem. 1987, 26, 2814.
10.1021/ja065645z CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 16641-16648
9
16641

A R T I C L E S
Thomas III et al.
Figure 1. Oxidative addition reactions of the type described by von
Zelewsky.
Figure 2. Schematic illustration of how a blue-shifting transduction event
gives a significant dark-field turn-on signal.
also found that less reactive molecules, such as chloroform,
dichloromethane, or ethyl bromide, underwent a similarly clean
field of oxygen sensing.11 The typically long-lived nature of
the triplet excited state of metal complexes make them highly
sensitive to even small concentrations of oxygen, which
decreases the emission intensity and phosphorescence lifetime
of most phosphors through triplet-triplet energy transfer to give
singlet oxygen and the ground state of the complex. This
reaction also forms the basis for photodynamic therapy, as the
singlet oxygen produced in this reaction can be highly cytotoxic
oxidative addition reaction under photolytic conditions via a
radical chain mechanism.6
von Zelewsky and co-workers also described that several of
the Pt(IV) products of these reactions were strongly emissive,
with a purely ligand-centered phosphorescence in the blue or
green region of the visible spectrum.6a,7 These emission signals
were very distinct from those of the Pt(II) starting materials,
where the contribution from a metal-to-ligand charge transfer
1
2
to cancer cells near the surface of the skin. Square planar Pt-
(
II) complexes have also been reported for the sensing of volatile
(
MLCT) state was much larger and gave a strongly red-shifted
13
8
organic compounds by reversible vapoluminescence.
emission. We became interested in this chemistry due to the
fact that these compounds have the potential to react specifically
with a wide variety of dangerous electrophilic chemicals,
including some chemical weapons. The ability to make use of
either thermal or photochemical reactions is very attractive and
highlights this transformation and the accompanying photo-
physical changes as a potentially powerful chemosensing
transduction scheme.
Additionally, there have been phosphorescent complexes
designed for the sensing of other vapors. One example is that
of Pilato, who used a platinum 1,2-enedithiolate complex to
detect fluoro and cyano phosphate esters by formation of a
1
4
strongly phosphorescent complex. The analytes activated a
primary alcohol that subsequently intramolecularly alkylated a
pyridine substituent to form a rigidified ligand around the
platinum center. Our group has utilized a similar reaction to
create a fluorescent turn-on sensor for these analytes, which
are simulants for chemical weapons, including SARIN and
In addition, the blue-shifted phosphorescence of the Pt(IV)
products of these reactions offers the advantage of negligible
background signal. Most multicolor emission-based chemosens-
ing schemes use red-shifting of the emission for signal trans-
1
5
SOMAN. The primary advantages to this type of detection
method for chemical weapons over others, such as surface
acoustic wave or interferometry, are that it is functional group
specific, simple, and highly portable.
9
duction, often by the F o¨ rster energy transfer mechanism.
However, the lower energy vibronic bands of the donor
chromophores often overlap with the acceptor emission, limiting
the maximum observable signal-to-noise and decreasing the
sensitivity of the measurement. Conversely, a blue-shifted signal
transduction event (illustrated schematically in Figure 2) can
allow for monitoring of a large portion of the desired signal
with no background emission.
Cyanogen halides (X-CN) are highly toxic blood agents that
have an effect on the human body similar to that of hydrogen
1
6
cyanide. Cyanogen chloride, a gas under ambient conditions,
is a military chemical weapon (abbreviation: CK). Reliable
detection of these poisons at trace concentrations has obvious
implications for homeland security. Herein we describe our
studies of a series of bis-cyclometalated platinium(II) complexes
as potential phosphorescent reporters of cyanogen halides via
an oxidative addition reaction.
There are multiple examples of phosphorescent complexes
used as chemosensors.10 The most notable is their use in the
(
5) von Zelewsky, A.; Suckling, A. P.; Stoeckli-Evans, H. Inorg. Chem. 1993,
2, 4585.
3
(
6) (a) Sandrini, d.; Maestri, M.; Balzani, V.; Chassot, L.; von Zelewsky, A.
J. Am. Chem. Soc. 1987, 109, 7720. (b) Rendina, L. M.; Puddephatt, R. J.
Chem. ReV. 1997, 97, 1735.
(11) (a) Jayarajah, C. N.; Yekta, A.; Manners, I.; Winnik, M. A. Macromolecules
2000, 33, 5693. (b) Costa-Fern a´ ndez, J. M.; Diaz-Barcia, M. E.; Sanz-
Medel, A. Anal. Chim. Acta 1998, 360, 17. (c) Choi, M. M. F.; Xiao, D.
Anal. Chim. Acta 1999, 387, 197.
(12) Dougherty, T. J.; Gomer, C. J.; Henderson, B. W.; Jori, G.; Kessel, D.;
Korbelik, M.; Moan, J.; Peng, Q. J. Natl. Cancer Inst. 1998, 90, 889.
(13) For a recent example of vapoluminescence-based sensing, see: Kui, S. C.
F.; Chui, S. S.-Y.; Che, C.-M.; Zhu, N. J. Am. Chem. Soc. 2006, 128, 8297.
(14) Van Houten, K. A.; Heath, D. C.; Pilato, R. S. J. Am. Chem. Soc. 1998,
120, 12359.
(
7) (a) Chassot, L.; von Zelewskky, A.; Sandrini, D.; Maestri, M.; Balzani, V.
J. Am. Chem. Soc. 1986, 108, 6084. (b) Barigelletti, F.; Sandrini, D.;
Maestri, M.; Balzani, V.; von Zelewsky, A.; Chassot, L.; Jolliet, P.; Maeder,
U. Inorg. Chem. 1988, 27, 3644.
(
8) Maestri, M.; Sandrini, D.; Balzani, V.; von Zelewsky, A.; Deuschel-
Cornioley, C.; Jolliet, P. HelV. Chim. Acta 1988, 71, 1053.
9) (a) Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. J. Am. Chem. Soc. 2003,
(
1
25, 896. (b) Ho, H.-A.; Dor e´ , K.; Boissinot, M.; Bergeron, M. G.; Tanguay,
R. M.; Boudreau, D.; Leclerc, M. J. Am. Chem. Soc. 2005, 127, 12673.
(15) Zhang, S.-W.; Swager, T. M. J. Am. Chem. Soc. 2003, 125, 3420.
(16) The Merck Index, 12th ed.; Budavari, S., Ed.; Merck Research Laborato-
ries: Whitehouse Station, NJ, 1996; p 453.
(
10) For a general review of chemical sensors, see: McQuade, D. T.; Pullen,
A. E.; Swager, T. M. Chem. ReV. 2000, 100, 2537.
16642 J. AM. CHEM. SOC.
9
VOL. 128, NO. 51, 2006

Dark-Field Oxidative Addition-Based Chemosensing
A R T I C L E S
Scheme 1. Synthesis of thpy Ligands
Scheme 2. Synthesis of Bis-cyclometalated Pt(II) Complexes 3-5
Results and Discussion
thene boronic acid and 2-bromopyridine by Suzuki coupling in
5
0-60% yield.
Synthesis. A total of eight platinum(II) complexes were
synthesized and studied in this work. Scheme 1 summarizes
the procedures used to synthesize the cyclometalating ligands.
The bi- or tricyclic ligands were readily synthesized by standard
palladium-catalyzed cross-coupling methods, except for 2-phe-
nylpyridine, which is commercially available. Non-benzo-fused
Scheme 2 summarizes the synthetic procedures used to
prepare the bis-cyclometalated Pt(II) complexes. Most of the
homoleptic complexes studied were synthesized by simple
lithiation of the ligands with tert-butyllithium in a THF/Et2O
mixture, followed by metalation with Cl2Pt(SEt2)2, as per the
method of von Zelewsky. Complex 4b could not be synthe-
sized, probably because of the steric congestion that would be
present in the desired product. Also, the heteroleptic complex
2
-thienylpyridine ligands, including the previously reported
4
parent thienylpyridine (thpy, 1a), were synthesized by either
Suzuki couplings (1a) with thiophene 2-boronic acid and
2
-bromopyridine or Stille couplings (1b-1f) with the ap-
(
5) was prepared by cracking the chloro-bridged ppy-ligated
propriately substituted reactants that were either commercially
available or readily prepared. Benzthiophene-based ligands 2a
and 2b were prepared from the commercially available thianaph-
dimer intermediate with diethyl sulfide, followed by reaction
with the lithiated thpy ligand. To our knowledge, this is the
first heteroleptic bis-cyclometalated platinum complex that has
been reported.17 All final complexes were readily chromato-
graphed under ambient conditions on silica gel and isolated as
(
17) For an example of a heteroletptic bis-cyclometalated palladium complex,
see: Jolliet, P.; Gianini, M.; von Zelewsky, A.; Bernardinelli, G.; Stoeckli-
Evans, H. Inorg. Chem. 1996, 35, 4883.
J. AM. CHEM. SOC.
9
VOL. 128, NO. 51, 2006 16643

A R T I C L E S
Thomas III et al.
Table 1. Photophysical Parameters of Pt(II) Complexes 3-5
gave blue-shifted MLCT bands. One example is complex 3a,
which in acetone had an MLCT maximum at 419 nm but in
toluene was red-shifted to 430 nm. Also, increasing the electron
density (3b, 3c) or the conjugation lengths (4a) of the ligands
tended to cause a red shift in the absorbance bands.
room-temperature solution
THF)
77 K glass
(2-MeTHF)
(
λ
max(UV/vis)a
(nm)
λ
max(phos)
τ
p
λ
max(phos)
τ
p
complex
(nm)
(
µs)
Φ
p
(nm)
(
µ
s)
3
3
3
3
3
3
4
5
a
b
c
d
e
f
306, 425
312, 433
317, 443
306, 419
309, 425
311, 438
277, 458
300, 415
582
599
613
578
588
582
630
581
5.5
0.24
0.29
0.30
0.24
0.22
<0.01
<0.01
0.05
575
590
605
569
578
572
626
567
10.3
13.0
14.2
10.9
9.7
15.2
8.7
13.2
In addition, most of the Pt(II) complexes displayed moderate
to strong phosphorescence intensity (with quantum yields of
emission between 0.05 and 0.30) in room temperature, deoxy-
genated fluid solution (Figure 5). All of the emissive complexes
phosphoresce in the orange or red region of the visible spectrum
with lifetimes on the time scale of 5-15 µs at ambient
temperature. The lower quantum yield and biexponential
character of 5 may be due to the presence of a competing,
nonemissive state involving the phenylpyridine ligand.
13.2
13.6
6.0
6.4
b
a
b
c
a
First maximum is ligand-centered absorption, and second is MLCT.
b
Not determined because of weakness of signal. c Multiexponential decay.
All these Pt(II) complexes show only weak solvatochromism
in their phosphorescence energy. The rigidochromic effect on
the emission of these complexes is a small value of 9 ( 3 nm
upon freezing the sample in a 2-methyltetrahydrofuran glass.
The rigid glass does not allow reorganization of solvent dipoles
upon generation of excited states and give strongly blue-shifted
spectra of complexes that emit from charge-transfer states. These
trends, coupled with the vibronic structure observed in the room-
temperature phosphorescence spectra, suggest that the emissive
state of these complexes is an admixture of an MLCT state and
an intraligand π-π* state. This behavior is similar to other Pt-
2
c
(II) cyclometalated complexes.
Not all of the complexes, however, emit strongly at room
temperature. Those complexes (3f and 4a) have phosphores-
cence quantum yields less than 1%. Complex 3f is a particularly
striking example, as the inclusion of only a methyl group on
the pyridine ring almost completely eliminated phosphorescence.
The relatively small effect of including a more powerful donor
in complex 3c suggests that this phosphorescence attenuation
is due to steric congestion of the square plane around the metal
center. The strong room-temperature emission of 3d and 3e also
supports the hypothesis that steric congestion is the reason for
the very weak emission of 3f and 4a. Complex 3f is very similar
to 3d and 3e, the only difference being that the methyl groups
meta or para to the pyridine nitrogen are not expected to have
repulsive interactions in the square plane. Complexes 3f and
Figure 3. Normalized UV/vis spectra of complexes 3a-d in THF.
4
a also gave broadened MLCT absorbance bands, suggesting
that there is more conformational variation in these nonemissive
complexes.
To directly observe the effects of interligand steric congestion
around the platinum metal center, single-crystal X-ray structures
were obtained for complexes 3c and 3e. ORTEP-style diagrams
of these crystal structures are displayed in Figure 6, while
crystallographic details and refinement statistics are included
in the Supporting Information. Complex 3b is only slightly
distorted out of a square planar geometry, while 3f, on the other
hand, is severely distorted away from ideal square plane
geometry because of the steric repulsion between the methyl
groups ortho to the nitrogen on the pyridine ring. Although we
were unable to grow suitable crystals from the benzthiophene-
substituted complexes, simple molecular mechanics modeling
indicates that the aryl hydrogen atoms in the 4-positions of the
thianaphthene ring system in complex 4a would have similar
steric interactions in a square planar geometry, forcing the
complex into a highly distorted conformation. Certain quinoline-
Figure 4. Normalized UV/vis spectra of complexes 3e, 3f, 4a, and 5 in
THF.
single stereoisomers, taken to be cis based on literature precedent
and several X-ray crystal structures (vide infra).
Photophysical Properties. Table 1 summarizes the photo-
physical properties of the platinum(II) complexes in deoxygen-
ated, room-temperature fluid solution and in 77 K 2-methyltetra-
hydrofuran glasses. Figures 3 and 4 show normalized absorbance
spectra of these complexes. Most of the complexes investigated
here showed sharp and distinct MLCT transitions with extinction
4
coefficients of approximately 10 , as is characteristic for the
parent Pt(thpy)2 complex (3a). These bands also showed
moderate negative solvatochromism, in that more polar solvents
8
substituted bis-cyclometalated heavy metal group 10 d com-
16644 J. AM. CHEM. SOC.
9
VOL. 128, NO. 51, 2006

Dark-Field Oxidative Addition-Based Chemosensing
A R T I C L E S
Figure 5. Room-temperature emission spectra of phosphorescent bis-cyclometalated Pt(II) complexes in THF.
Figure 6. 50% probability representation of X-ray crystal structures of 3b (left) and 3f (right).
plexes have been shown to have a severely twisted geometry
for similar reasons.17 These intramolecular repulsions may lead
to enhanced nonradiative relaxation of the excited state through
additional twisting and vibration.
Oxidative Addition In Solution. All of the reported com-
plexes take part in oxidative addition reactions under ambient
5
conditions as described by von Zelewsky. Figure 7 shows the
progress of a reaction with 3b with methyl iodide under pseudo-
first-order conditions in benzene as followed by absorbance
spectroscopy. Well-behaved isosbestic points, in addition to
well-behaved pseudo-first-order kinetics (Figure 8), characterize
this reaction. There is also a strong solvent dependence on the
rate of reaction, which is indicative of a highly polar transition
state and an SN2-type mechanism.
Upon adding cyanogen bromide to these complexes under
similar conditions, several important differences were apparent.
The first is that the reaction with cyanogen bromide was much
faster than with methyl iodide. Figure 9 compares the percent
conversion as a function of reaction time for cyanogen bromide
Figure 7. UV/vis spectra of 3b during its reaction with 1.0 M MeI in
benzene. Times elapsed, in seconds, are 15, 30, 45, 60, 90, 120, 180, 240,
and 300.
(
0.00013 M) and methyl iodide (1.0 M) with 3a. The cyanogen
with MeI, even though the concentration of BrCN is almost
10 times smaller. The nonpolar solvent benzene was used to
4
bromide reaction goes to completion well before the reaction
J. AM. CHEM. SOC.
9
VOL. 128, NO. 51, 2006 16645

A R T I C L E S
Thomas III et al.
Figure 8. Pseudo-first-order rate plot for 3b in 1.0 MeI/benzene. The
-
1
-1
Figure 10. UV/vis spectra of 3b during its reaction with 0.00013 M BrCN
in benzene. Times elapsed are in 20 s intervals.
calculated bimolecular rate constant was 0.0081 M
s .
Figure 9. Conversion versus time, as determined by UV/vis, for 3a, with
either MeI or BrCN (in benzene). A much smaller concentration of BrCN
is necessary to achieve rapid conversion.
Figure 11. First-order kinetic plot for the reaction of 3b with BrCN (2.4
× 10-4 M) in benzene. The reaction accelerates after an initial induction
period.
more effectively mimic a solid-state environment, which would
be used for sensing purposes, than a more polar solvent. The
UV/vis profiles of these reactions (Figure 10 for 3b) also show
well-behaved isosbestic points, disappearance of the Pt(II)
MLCT band, and growth of an absorbance at approximately
certain benzylic halides and also ascribed them to a radical
mechanism. This is further supported by the observation that
the reaction goes to completion much faster in benzene than in
toluene (Figure 12), suggesting that an intermediate radical may
abstract a benzylic hydrogen atom from toluene, thereby
inhibiting the reaction.
350 nm, characteristic of bis-cyclometalated Pt(IV) complexes.
In addition to a much faster reaction rate in solution, the
To ensure that oxidative addition, and not complex decom-
position, was occurring, the reaction between 3a and BrCN was
performed on a preparative (∼30 mg) scale. Upon mixing the
two in THF, a colorless product immediately precipitated. The
major product (of the two by TLC) was isolated by filtration
and washing with dichloromethane. Mass analysis of the isolated
product showed the desired molecular ion for Pt(thpy)2(Br)-
(CN) with the expected fragmentation pattern. NMR analysis
was not possible because of the very limited solubility of the
complex. No free ligand was observed upon addition of BrCN.
Although the product was not emissive in room temperature
fluid solution, it was strongly emissive at 77 K with photo-
physical parameters consistent with a bis-cyclometalated Pt-
kinetic profiles of these reactions with CNBr do not follow a
simple kinetic model. Figure 11 shows the pseudo-first-order
rate plot for the reaction of 3b with 0.00024 M CNBr in
benzene. The reaction accelerates very quickly following an
initial induction period. This difference in kinetic behavior
suggests a different mechanism than the SN2 type invoked for
the oxidative addition with methyl iodide.
A third significant difference between the reactions with
CNBr and MeI is the lack of strong solvent polarity dependence
on the rate of reaction with the Pt(II) complexes and CNBr.
Upon switching from toluene to acetone, the initial reaction rate
only increased by approximately 50%. The key differences in
these reactions are indicative of a radical, potentially chain,
mechanism operating in this reaction. von Zelewsky has reported
these reaction properties occurring between 3a complexes and
7
a
(IV) complex. This included a highly structured emission
spectrum with a maximum in the green region of the visible
spectrum and an excited-state lifetime of about 400 µs.
16646 J. AM. CHEM. SOC.
9
VOL. 128, NO. 51, 2006

Dark-Field Oxidative Addition-Based Chemosensing
A R T I C L E S
Figure 14. Normalized emission spectra of 3a in degassed THF solution
and doped into PMMA films (10% w/w).
Figure 12. Percentage conversion of 3a, in either toluene or benzene, in
the presence of 0.00024 M BrCN.
Figure 13. Conversion percentages of several Pt(II) complexes as a function
of time in 0.00013 M BrCN/benzene.
Figure 15. PMMA/Pt(II) spun-cast films of 4a (top) and 3a (bottom) under
UV light. Films on the left are unexposed, while films on the right have
been exposed to saturated BrCN for 15 s.
Table 2. Bimolecular Rate Constants of Platinum Complexes with
MeI in Benzene
likely be beneficial to any turn-on phosphorescence-based
sensing scheme, since excessive phosphorescence from the
starting “off” state may increase the background signal on top
of which the measurement must be made. The heteroleptic
complex 5 also goes to completion faster than the parent
homoleptic complex 3a, but the reason for this is presently
unclear.
k (M- s-1)
1
complex
-
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
5
a
b
c
d
e
f
3.4 × 10
8.1 × 10
4.8 × 10
8.8 × 10
5.1 × 10
0.82 × 10
0.93 × 10
5.8 × 10
-
-
-
-
-
-
-
a
In contrast to the structure-reactivity trend with BrCN, the
more sterically hindered complexes (3f, 4a) show bimolecular
rate constants with methyl iodide (in benzene) that are ap-
proximately 4-10 times smaller than the unstrained systems
Analysis of the structure-reactivity trends between different
Pt(II) complexes and BrCN yielded similar results to the study
of photophysical properties. The most reactive complexes were
those that incorporated interligand steric congestion into the
square plane of the complex (3f, 4a) even at stoichiometric
concentrations of BrCN. Figure 13 shows the conversion as a
function of time (in 0.00013 M BrCN/benzene) of several
members of the Pt(thpy)2 family. Therefore, the increase in the
energy of the sterically crowded complexes decreases the
activation barrier for oxidative addition, increasing the rate at
which these heavily distorted complexes go to octahedral
products. The fact that the least emissive Pt(II) complexes are
the most reactive toward oxidative addition with BrCN will
(Table 2). An electronic effect for this behavior was ruled out
because 3d and 3e, which have methyl groups in different
positions on the pyridine ring, do not show behavior similar to
that of 3f. This suggests that the transition state for SN2 type
oxidative addition to these complexes exacerbates the preexisting
steric congestion, possibly by forcing the complex into a square-
pyramidal geometry, whereas the transition state for the reaction
with BrCN relieves the unfavorable interactions. This difference
is particularly useful for sensing, as it could impart additional
selectivity for cyanogen halides over interferents that react by
the SN2-type mechanism.
J. AM. CHEM. SOC.
9
VOL. 128, NO. 51, 2006 16647

A R T I C L E S
Thomas III et al.
Upon exposing these doped PMMA films to saturated BrCN
vapor, complete conversion to the corresponding platinum(IV)
complexes occurred within seconds. The blue-shifted emission
of the product is apparent both spectroscopically and visually.
The spectral features of the products in the solid state are
consistent with the characteristic ligand-centered platinum(IV)
emission, both in spectral position and in the relative intensities
of the vibronic bands. Figure 15 shows a picture in which
PMMA films containing 3a and 4a have been exposed to
saturated BrCN vapor for 15 s. The spectra in Figure 16 illustrate
how using a strong spectral blue-shift as the sensing signal gives
a turn-on signal with virtually no background (dark field). This
feature is desirable in any sensing system for maximum
sensitivity to trace quantities of analyte. Current experiments
are focused on demonstrating trace (part per million) sensitivity
to the cyanogen halides. Preliminary experiments have shown
modest sensitivity to 10 ppm BrCN vapor. We are also focusing
on extending this type of chemosensing mechanism to other
dangerous chemicals, such as explosives.
Conclusion
A new dark-field turn-on optical chemosensing scheme has
been developed using bis-cyclometalated platinum complexes.
A series of these complexes was synthesized, most of which
have not been previously reported in the literature. The ligands
were chosen so that structures with a variety of electronic and
steric properties could be investigated. Most of these Pt(II)
complexes were strongly phosphorescent at room temperature
in fluid solution. It was found that cyanogen halides oxidatively
add readily to these Pt(II) complexes to give the corresponding
Pt(IV) products. Investigations of the structure-property rela-
tionships showed that those complexes that were twisted out of
the square planar geometry (and were also not phosphorescent
at room temperature in solution) reacted faster with CNBr than
those that were not. The relative lack of solvent polarity
dependence on the reaction kinetics as well as the strong
difference in kinetics between toluene and benzene strongly
suggests a radical mechanism for this reaction. Thin polymer
films containing these complexes, which were phosphorescent
at room temperature, also reacted with cyanogen halide vapors
to give the blue-shifted Pt(IV) products which were also
emissive under these conditions, giving a dark-field turn-on of
the product phosphorescence, an advantageous characteristic for
improving sensitivity toward any potential analyte. Current
efforts are directed at improving the sensitivity of cyanogen
halide detection, as well as applying this new chemosensing
technique to other potential analytes.
Figure 16. PMMA film emission spectra of 3a (a) and 4a (b) before
(dotted) and after (solid) exposure to BrCN vapor for 15 s.
Solid State Phosphorescence Sensing. Solid-state materials
are essential for the fabrication of effective vapor-phase sensory
devices. This can be readily achieved with conjugated polymer
sensors, neat thin films of which can be highly emissive when
the polymer structure is properly designed to prevent aggrega-
tion. These platinum complexes, however, were doped into a
polymer matrix in order to create a substance with desirable
material properties that could be readily cast into films. By spin-
casting from a dichloromethane solution that contained a mixture
of poly(methyl methacrylate) (PMMA) and the desired Pt(II)
complex (10% w/w relative to PMMA), transparent, glassy,
highly phosphorescent thin films were readily prepared. Many
of these films were highly emissive even under ambient
conditions because the glassy PMMA excluded enough oxygen
to allow radiative decay of the triplet excited states to be
kinetically competitive with oxygen-induced quenching. As
illustrated in Figure 14, these films showed very similar shapes
of their emission spectra to the solution data. This indicates
that at 10% loading the amount of intermolecular communica-
tion between the metal complexes was negligible. In addition,
those complexes that were not emissive in room-temperature
solution (3f, 4a) showed phosphorescence spectra in the PMMA
films that were in the same spectral region as at 77 K. This is
presumably because the rigid PMMA matrix inhibits the
intramolecular conformational changes that lead to nonradiative
deactivation of the excited states of these strained molecules.
Acknowledgment. This work was supported by the Army
Research Office and the Institute for Soldier Nanotechnologies.
Supporting Information Available: Experimental and syn-
thetic details. Crystallographic information files (CIF) for 3b
and 3f. Crystal structure and CIF file for 5. This material is
available free of charge via the Internet at http://pubs.acs.org.
JA065645Z
16648 J. AM. CHEM. SOC.
9
VOL. 128, NO. 51, 2006