Assembly of cyclometalated platinum(II) complexes via 1,1′- bis(diphenylphosphino)ferrocene ligand: Kinetics and mechanisms

Article abstract of DOI:10.1021/om101055n

The kinetics and mechanism of the reaction of the cyclometalated complexes [PtAr(C-N)(SMe₂)], 1, in which Ar is Ph, p-MeC₆H ₄, or p-MeOC₆H₄, and C-N is either ppy (deprotonated 2-phenylpyridine) or bhq (deprotonated benzo[h]quinoline), with 1,1′-bis(diphenylphosphino)ferrocene, dppf, were studied using UV-visible and ³¹P NMR spectroscopies. When 0.5 equiv of dppf was added, the binuclear Pt(II) complex [Pt₂Ar₂(C-N)₂(μ- dppf)], 2, was formed in a good yield. The complexes were fully characterized using multinuclear (¹H, ³¹P, and ¹⁹⁵Pt) NMR spectroscopy, and the structure of complex [Pt₂(p-MeOC ₆H₄)₂(bhq)₂(μ-dppf)], 2c′?CH₂Cl₂, was further identified by X-ray crystallography. On the basis of low-temperature ³¹P NMR studies involving the starting complex [Pt(p-MeC₆H₄)(ppy)(SMe ₂)], 1b, we suggest that dppf displaces the labile ligand SMe ₂ to give an uncommon complex, [Pt(p-MeC₆H ₄)(ppy)(dppf-κ¹P)], A, in which dppf- κ¹P is a monodentate dppf ligand, which rapidly forms an equilibrium with the chelating dppf isomer complex [Pt(p-MeC₆H ₄)(dppf)(ppy-κ¹C)], B, in which ppy- κ¹C is the deprotonated ppy ligand that is C-ligated with the dangling N atom. In the second step, A is reacted with the remaining second half of starting complex 1b to give the final Pt(II)-Pt(II) binuclear complex [Pt₂(p-MeC₆H₄)₂(ppy) ₂(μ-dppf)], 2b. A competitive-consecutive second-order reaction mechanism was suggested for the reaction using chemometric studies, and the rate constants at 5 °C for first and second steps were estimated as k ₂ = 10.7 ± 0.2 L mol^-1 s^-1 and k ₂′ = 0.68 ± 0.05 L mol^-1 s^-1, respectively. When the starting complex [Pt(p-MeC₆H ₄)(ppy)(SMe₂)], 1b, was reacted with 1 equiv of dppf, similarly the complex A, in equilibrium with B, was formed first, with the rate constant at 5 °C being k₂ = 10.5 ± 0.5 L mol^-1 s^-1, estimated using UV-visible spectroscopy. Subsequently, however, A and B would slowly and reversibly react with each other to form a new species, C, the structure of which, on the basis of ³¹P and ¹⁹⁵Pt NMR spectra, was proposed to be [(p-MeC₆H₄)(ppy)Pt(μ- dppf)Pt(p-MeC₆H₄)(ppy-κ¹C)(dppf- κ¹P)]; the same results were obtained when more than 1, e.g., 2, equiv of dppf was used, with a similar rate constant of k₂ = 10.6 ± 0.6 L mol^-1 s^-1. The complexes 1b and 2b were shown to have some interesting photophysical properties as investigated by absorption and electroluminescence spectroscopies.

Full text of DOI:10.1021/om101055n

ARTICLE
pubs.acs.org/Organometallics
Assembly of Cyclometalated Platinum(II) Complexes via
1,10-Bis(diphenylphosphino)ferrocene Ligand: Kinetics
and Mechanisms
S. Masoud Nabavizadeh,*,† Hashem Amini,† Hamid R. Shahsavari,† Masoud Namdar,† Mehdi Rashidi,*,†
Reza Kia,‡ Bahram Hemmateenejad,† Mohsen Nekoeinia,† Alireza Ariafard,§
Fatemeh Niroomand Hosseini,^ Alireza Gharavi,|| Ali Khalafi-Nezhad,† Mohammad Taghi Sharbati,|| and
Farhad Panahi†
†Department of Chemistry, College of Sciences, Shiraz University, Shiraz 71454, Iran
‡Department of Chemistry, Science and Research Branch, Islamic Azad University, Tehran, Iran
§Department of Chemistry, Faculty of Sciences, Central Tehran Branch, Islamic Azad University, Shahrak Gharb, Tehran, Iran
^Department of Chemistry, Islamic Azad University, Shiraz Branch, Shiraz 71993-37635, Iran
Department of Electrical and Computer Engineering, Photonics Lab, Shiraz University, Shiraz, Iran
S Supporting Information
b
ABSTRACT: The kinetics and mechanism of the reaction of the
cyclometalated complexes [PtAr(C-N)(SMe2)], 1, in which Ar is
Ph, p-MeC6H4, or p-MeOC6H4, and C-N is either ppy
(deprotonated 2-phenylpyridine) or bhq (deprotonated benzo-
[h]quinoline), with 1,10-bis(diphenylphosphino)ferrocene, dppf,
were studied using UV-visible and 31P NMR spectroscopies.
When 0.5 equiv of dppf was added, the binuclear Pt(II) complex
[Pt2Ar2(C-N)2(μ-dppf)], 2, was formed in a good yield. The
complexes were fully characterized using multinuclear (1H, 31P,
and 195Pt) NMR spectroscopy, and the structure of complex
[Pt2(p-MeOC6H4)2(bhq)2(μ-dppf)], 2c0 CH2Cl2, was further
3
identified by X-ray crystallography. On the basis of low-tempera-
ture 31P NMR studies involving the starting complex [Pt(p-
MeC6H4)(ppy)(SMe2)], 1b, we suggest that dppf displaces the labile ligand SMe2 to give an uncommon complex, [Pt(p-MeC6H4)-
(ppy)(dppf-κ1P)], A, in which dppf-κ1P is a monodentate dppf ligand, which rapidly forms an equilibrium with the chelating dppf isomer
complex [Pt(p-MeC6H4)(dppf)(ppy-κ1C)], B, in which ppy-κ1C is the deprotonated ppy ligand that is C-ligated with the dangling N
atom. In the second step, A is reacted with the remaining second half of starting complex 1b to give the final Pt(II)-Pt(II) binuclear
complex [Pt2(p-MeC6H4)2(ppy)2(μ-dppf)], 2b. A competitive-consecutive second-order reaction mechanism was suggested for the
reaction using chemometric studies, and the rate constants at 5 ꢀC forfirst and second steps were estimated as k2 = 10.7 ( 0.2 L mol-1 -
s
and k2 = 0.68 ( 0.05 L mol-1 s-1, respectively. When the starting complex [Pt(p-MeC6H4)(ppy)(SMe2)], 1b, was reacted with 1
1
0
equiv of dppf, similarly the complex A, in equilibrium with B, was formed first, with the rate constant at 5 ꢀC beingk2 = 10.5( 0.5 L mol-
1 s-1, estimated using UV-visible spectroscopy. Subsequently, however, A and B would slowly and reversibly react with each other to
form a new species, C, the structure of which, on the basis of 31P and 195Pt NMR spectra, was proposed to be [(p-MeC6H4)(ppy)Pt-
(μ-dppf)Pt(p-MeC6H4)(ppy-κ1C)(dppf-κ1P)]; the same results were obtained when more than 1, e.g., 2, equiv of dppf was used, with a
similar rate constant of k2 = 10.6 ( 0.6 L mol-1 s-1. The complexes 1b and 2b were shown to have some interesting photophysical
properties as investigated by absorption and electroluminescence spectroscopies.
’ INTRODUCTION
environment around the involved molecule.2-5 The majority of
the reported transition metal complexes with dppf are mono-
nuclear complexes bearing a single or two chelating dppf ligands
(see Scheme 1, I and II).4,5a Also, a third type of known coordi-
nation mode for dppf and related 1,10-ferrocenyl diphosphines
There has been great interest in the synthesis and investigation
of transition metal complexes containing the biphosphine ligand
1,10-bis(diphenylphosphino)ferrocene, dppf, as the complexes
have been involved in many catalysis, organic synthesis, electro-
chemical, materials science, and biological studies and appli-
cations.1 This versatile ligand adopts various modes of coordina-
tion in order to match with the steric requirement dictated by the
Received: November 10, 2010
Published: February 16, 2011
r
2011 American Chemical Society
1466
dx.doi.org/10.1021/om101055n Organometallics 2011, 30, 1466–1477
|

Organometallics
ARTICLE
Scheme 1. Some Coordination Modes of dppf
Scheme 2
(especially with Pd and Pt), where the ligand prefers a trans
coordination with a weak dative bond from iron, has been
reported (see Scheme 1, III).2b,c However, diplatinum complexes
containing bridging dppf are not common. We have recently
reported some binuclear Pt(II) and Pt(IV) complexes, each
having a dppf spacer ligand.3a In our recent report we have
shown that the great chelating appetite of dppf has caused the
rather strong cyclometalated chelate to open up from the N
donor atom in the complex [PtMe(dppf)(ppy-κ1C)], in which
ppy = deprotonated 2-phenylpyridine.3a
anticlinal staggered for Y = OH, NH2, and N(H)Me,
respectively.5d
In the present work, we have synthesized a series of cyclome-
talated diplatinum(II) complexes each containing a dppf spacer
ligand, [Pt2Ar2(C-N)2(μ-dppf)], 2, in which C-N is ppy or
deprotonated benzo[h]quinoline (bhq) and Ar is a simple
aromatic group, using a general synthetic approach involving
the ligand replacement from the corresponding precursor having
a labile ligand SMe2. In fact this type of substitution procedure is
the most common pathway used in the preparation of dppf
complexes.1 Due to the above-mentioned versatility of the
dppf in complex formation, we were prompted to perform a
systematic kinetic and mechanistic investigation on the related
replacement reactions by using UV-visible and NMR spectro-
scopies (or chemometric methods in the case where 0.5 equiv of
dppf was added) and to study the factors that control the
reactivity of dppf complexes toward chelate opening. The pre-
sence of two cycloplatinated units in the same complex in
compounds 2 may be very interesting in view of their potential
applications in a wide variety of fields. It is well known that some
platinacycles with C-N ligands have significant photophysical
properties,3b-i and we indicated that the new complexes 2
appeared to probably be suitable candidates for further studies
in these areas.
’ RESULTS AND DISCUSSIONS
Synthesis and Characterization of the Complexes. The
general synthetic route to diplatinum complexes 2 is described in
Scheme 2. The reaction of cis-[PtAr2(SMe2)2], in which Ar = Ph,
p-MeC6H4, or p-MeOC6H4, with 1 equiv of 2-phenylpyridine or
benzo[h]quinoline, in refluxing acetone, gave the cyclometalated
complexes [PtAr(C-N)(SMe2)], 1, in which C-N is either ppy
or bhq. Reaction of each of the complexes 1 with 0.5 equiv of
dppf, followed by replacement of the labile ligand SMe2, yielded
the diplatinum(II) complexes [Pt2Ar2(C-N)2(μ-dppf)], 2, in a
good yield.
The complexes 1 were characterized by their 1H NMR spectra.
In each case two well-defined peaks with a relative intensity 6:1
were observed, one for the terminal SMe2 ligand (at δ = 2.2 ppm
with 3JPtH ≈ 25 Hz) and the other for the H6 proton, related to
the CH group adjacent to ligating N atom of the C-N ligand (at
δ = 8.8 ppm for ppy and δ = 9.1 ppm for bhq). The latter signal in
each case is observed as a well-separated doublet due to coupling
with the nearest H atom (H5) with a 3JH6H5 value close to 5.5 Hz,
which is also weakly coupled to the platinum(II) center with a
3JPtH6 value around 17-19 Hz. The other C-N and Ar aromatic
hydrogen atoms appeared at the expected regions, 7-9 ppm.
In the room-temperature 31P NMR spectra of the binuclear
complexes [Pt2Ar2(C-N)2(μ-dppf)], 2, the observation of a
sharp singlet signal for each of the analogues at a chemical shift of
close to 20 ppm, which is accompanied by Pt satellites, confirms
that the two PtAr(C-N) moieties, joined together by the dppf
spacer ligand, are equivalent. The 1JPtP values for ppy complexes
are close to 2000 Hz, while the 1JPtP values for bhq complexes are
close to 2060 Hz. These 1JPtP values, which complied well with
the values obtained from the corresponding 195Pt NMR spectra,
are within the expected range,3a,5a and the difference of ca. 3% is
probably due to a greater trans influence of the metalated C atom
of ppy derivatives as compared to that of the bhq analogues. In
the 1H NMR spectrum of each of the complexes 2, the
The crystal structure of the complex [Pt2(p-MeOC6H4)2-
(bhq)2(μ-dppf)], 2c0 CH2Cl2, is also reported. To the best of
3
our knowledge this is the first X-ray structural determination of a
diplatinum complex having an “open bridge” dppf ligand,
although crystal structures of some transition metal complexes
including dppf as bridging ligand have been reported.5b-d For
example crystal structures for complexes [(C6H4CH2NMe2-C2,
N)2Pd2Cl2(μ-dppf)]5b and [{Pd[5-(COH)C6H3C(H)dNCy-
C2,N](Cl)}2(μ-dppf)]5c have been reported with the conforma-
tions of the ferrocenyl skeleton being anticlinal eclipsed,
although for a series of complexes with the general formula
[(AuSC6H4{C(dO)Y}-2)2-(μ-dppf)] the conformational
modes adopted by the ferrocene cyclopentadienyl (Cp) rings
are found as antiperiplanar staggered, synclinal eclipsed, and
1467
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
Figure 1. (a) Molecular structure of complex [Pt2(p-MeOC6H4)2-
(bhq)2(μ-dppf)], 2c0 CH2Cl2, showing 30% probability ellipsoids and
3
the atomic numbering. The H atoms are omitted for clarity [symmetry
operator for the unlabeled atoms: -x þ 1, y, -z þ 0.5]. Selected bond
distances (Å) and angles (deg): Pt(1)-C(26), 2.013(6); Pt(1)-C(7),
2.052(7); Pt(1)-N(1), 2.162(6); Pt(1)-P(1), 2.3143(19); C(26)-
Pt(1)-C(7), 90.4(3); C(26)-Pt(1)-N(1), 168.1(2); C(7)-Pt(1)-
N(1), 79.7(3); C(26)-Pt(1)-P(1), 88.9(2); C(7)-Pt(1)-P(1),
177.6(2); N(1)-Pt(1)-P(1), 101.26(17). (b) Conformation of the
cyclopentadienyl rings of the ferrocene moieties in complex 2c0. The
diagram highlights the relative dispositions of the five-membered rings as
well as the relative orientations of the platinum atoms in the structure.
Figure 2. Changes in 31P NMR spectra during the reaction of complex
[Pt(p-MeC6H4)(ppy)(SMe2)], 1b, with 0.5 equiv of dppf to form the
binuclear complex [Pt2(p-MeC6H4)2(ppy)2(μ-dppf)], 2b, via the inter-
mediate formation of complexes [Pt(p-MeC6H4)(ppy)(dppf-κ1P)], A,
and [Pt(p-MeC6H4)(dppf)(ppy-κ1C)], B: (a) spectrum immediately
after mixing the reactants at 233 K, (b) 9 min later at 233 K, (c) 23 min
later at 248 K, (d) 31 min later at 258 K, (e) 41 min later at 273 K, (f) 50
min later at 298 K.
observation of a slightly broad singlet for either of the R- or
β-protons, at δ values close to 4.4 or 3.4 ppm, respectively,
confirms the equivalency of the two PtAr(C-N) moieties in the
related complex. This is further confirmed by observation of only
one signal for the two H6 protons, related to the two CH groups
adjacent to ligating N atoms of the two C-N ligands (appearing
as a doublet signal at δ = 7.76 ppm, for the ppy complexes, and at
δ = 8.12 ppm for the bhq complexes, with 3JH6H5 and 3JPtH6 values
close to 8 and 18.5 Hz, respectively); the signals appeared in a
considerably lower field as compared to the other aromatic
protons.
centroid of the Pt1/N1/C5/C6/C7 and C4/C5/C6/C11/C12/
C13 rings.
As discussed above, a sharp singlet was observed for the two
equivalent phosphorus atoms in the room-temperature 31P
NMR spectrum of the binuclear complex [Pt2(p-MeC6H4)2-
(ppy)2(μ-dppf)], 2b. However, this singlet is considerably
broadened when the spectrum is obtained at low temperatures
(see Figure 2). It is probable that when a solid crystalline sample
of complex 2b, with its “anticlinal eclipsed” conformation, is
dissolved in solvent, a fluxionality of the type shown in Scheme 3
occurs. As such, at room temperature the fluxionality behavior
takes place very fast so that on average the “antiperiplanar
staggered” conformation, with an ideal torsion angle of 180ꢀ, is
observed, which gives rise to the observation of a sharp singlet in
the 31P NMR spectrum. At low temperatures however, the
process is slow and both conformers are observable, leading to
the observation of the broad singlet. It is therefore probable that
the energy barrier to interconverting these eclipsed and staggered
configurations is not high.
The structure of complex [Pt2(p-MeOC6H4)2(bhq)2(μ-dppf)],
2c0 CH2Cl2, was further determined by X-ray crystallography and
3
is shown in Figure 1; the selected bond lengths and bond angles are
also included in Figure 1. The geometry is best described as
distorted square-planar, and the bond angles around the Pt center
range from 79ꢀ to 101ꢀ. The chelating C-N bite angle, i.e., C(7)-
Pt(1)-N(1), amounts to 79.7(3)ꢀ, which is significantly smaller
than 90ꢀ, and implies that the chelate is probably under strain. The
Cp(centroid) Fe Cp(centroid) twist angle of dppf is 132ꢀ
3 3 3 3 3 3
(the angle has been determined as the torsion angle C33 Cp-
3 3 3
(centroid)1 Cp(centroid)2 C33A), and so dppf is arranged
Mechanism of the Diplatinum Complex Formation. The
reaction of [Pt(p-MeC6H4)(ppy)(SMe2)], 1b, with 0.5 equiv of
dppf to form the binuclear complex [Pt2(p-MeC6H4)2(ppy)2-
(μ-dppf)], 2b, was monitored using variable-temperature 31P
NMR spectroscopy (202 MHz, in the range of 233-298 K), as
shown in Figure 2.
3 3 3
3 3 3
close to the “anticlinal eclipsed” conformation;2a the ideal torsion
angle for this conformation is 144ꢀ. In the crystal structure, the
molecules form 1-D infinite chains along the (1 0 1) direction by
the intermolecular π-π interactions with the centroid (Cg1) to
centroid (Cg2) distance of 3.696(4) Å; Cg1 and Cg2 are the
1468
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
Scheme 3
Scheme 4
On the basis of the data obtained from the spectra, a mechan-
ism described in Scheme 4 is suggested to occur during the
reaction progress. Thus, 1b, containing a labile SMe2 ligand, is
first attacked by dppf to form the intermediate complex [Pt(p-
MeC6H4)(ppy)(dppf-κ1P)], A, in which dppf-κ1P is a mono-
dentate dppf ligand;6 0.5 equiv of the starting complex 1b
remains unreacted. The formation of A is confirmed by the
observation of a singlet signal at δ = -18.9 ppm in the spectrum
obtained at 233 K, due to the dangling phosphorus atom of the
dppf ligand, and a rather broad singlet signal at δ = 12.4 ppm,
which is coupled by the platinum center to give a 1JPtP value of
1842 Hz, due to the phosphorus atom of the dppf connected to
platinum. The complex A is in equilibrium with the intermediate
isomer complex [Pt(p-MeC6H4)(dppf)(ppy-κ1C)], B, in which
ppy-κ1C is the deprotonated ppy ligand that is only C-ligated
with the dangling N atom. During this dynamic process, the
dangling dppf in complex A is converted to the chelating dppf,
and as a result, the chelating C-N ligand is forced to open up
from the N site. Note that we have previously indicated that when
the Me analogue of 1b, i.e., [PtMe(ppy)(SMe2)], reacted with 1
mol of dppf, only the dppf chelate analogue complex [PtMe-
(dppf)(ppy-κ1C)] was separated.3a However in the Ar analogue
complexes 1 studied in the present work, the electron-with-
drawing ability of Ar ligands is greater than that of the Me ligand,
which favors the connection of the N ligating atom. More
importantly, the fact that coordination around the dppf chelating
B seems to be more sterically hindered than that of the dppf
dangling analogue complex A is probably responsible for the
formation of both A and B in equilibrium with each other. This
has been confirmed with further theoretical studies (see next
paragraph). The complex A is then slowly reacted with the
remaining second half of the starting complex 1b to give the final
Pt(II)-Pt(II) binuclear complex [Pt2(p-MeC6H4)2(ppy)2-
(μ-dppf)], 2b. Consistently we found that if in a similar reaction
less than 0.5 equiv of dppf is added, complex 2b is formed
accompanied by some unreacted starting complex 1b. For B two
signals appeared at δ 12.3 and 14.1 both with 1JPtP values almost
equal to 1845 Hz, which is consistent with the trans influence of
the p-MeC6H4 aryl ligand being almost equal to that of the ppy-
κ1C aryl ligand.7 In our previous report of the 31P NMR spectrum
of the analogous dppf chelating complex [PtMe(dppf)(ppy-
κ1C)],3a the two related signals, appearing at δ = 20.5 and 23.9
ppm, indicated different 1JPtP values of 1995 and 1888 Hz due to
the trans influence of Me ligand being higher than that of the ppy-
κ1C ligand. Overall, therefore, upon addition of dppf to the
solution of 1b (contributing no phosphorus signal), A is formed
in an equilibrium with B. As time passes and the temperature is
raised, A is consumed by 1b and the final diplatinum complex 2b
is formed.
A series of calculations have been performed in order to
investigate the relative stabilities of the isomers A and B
(Schemes 4 and 5) as a function of the R ligand, in which R =
Me, CtCH, CF3, H, or Ph. The lowest energy structures of
the complexes [PtR(κ2-C,N-ppy)(κ1-P-dmpf)] (isomer A) and
[PtR(κ1-C-ppy)(κ2-dmpf)] (isomer B) [dmpf = 1,10-bis(dime-
thylphosphino)ferrocene], as calculated by density functional
theory (DFT), are shown in Figure 3.8
Thus, when R = Me, CtCH, CF3, or H, the calculated
energies for isomer A are 15.05, 13.15, 17.79, and 9.39 kJ mol-
1, respectively, greater than those calculated for isomer B. The
results comply well with a greater preference for the κ1-C-isomer
A over isomer B for the case where R = Me as discussed above.
Also, consistent with related geometrical isomerization discussed
above for the A and B intermediates where R = Ph, the corre-
sponding energies calculated for both isomers are almost equal.
Although platinum(II) is a soft metal center that normally favors
1469
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
Scheme 5
complex C, during which the P dangling atom in A attacks the Pt
center of B to open up the chelating dppf. Complex C finally
dissociates and forms an equilibrium with the dimeric complex
2b by releasing free dppf. Similar results are observed when more
than 1 equiv of dppf (e.g., 2 equiv) is used in the related reaction,
except for slight variations in the ratio of the product compo-
nents, 2b:C:free-dppf. The assignments for the complexes are
indicated in Figure 4. The sharp singlet signal observed at δ =
20.6 (with 1JPtP = 2004 Hz) is due to the bridged complex 2b with
the related free dppf appearing at δ = -17.3. The sharp signal
appearing at δ = 20.9 (with 1JPtP = 2014 Hz) has almost the same
parameters as those for the rather broad signal of 2b, and we
therefore attribute it to Pa of the suggested product C with the
doublets at δ = 15.5 (2JPbPc = 15 Hz and 1JPtP = 1895 Hz) and δ =
12.2 (1JPtP = 1845 Hz and 2JPbPc = 15 Hz) being assigned to its Pc
and Pb, respectively, and the signal at δ = -17.8 to its dangling
phosphorus atom Pd. In the corresponding 195Pt NMR spectrum,
the broad doublet signal at δ = -2410 (with 1JPtP = 1990 Hz) is
assigned to the two overlapped Pt signals of 2b and C, and a
broad triplet at δ = -2820 with 1JPtP = 1890 Hz is assigned to the
second Pt atom of complex C (see Figure 4).
In order to confirm the last step of the reaction, i.e., dissocia-
tion of C to form an equilibrium with 2b and free dppf (see
Scheme 5), a pure sample of 2b was reacted with 1 (or even 2)
equiv of dppf, and on the basis of the 31P NMR spectrum we
found that complex C was formed as a mixture with 2b and free
dppf. In addition, attempts were also made to extend the length
of the chain in complex C from the terminal dangling P atom by
using other complexes bearing labile ligands, but fragmentation
of the type described in Scheme 6 may have taken place. Thus,
when a solution of the mixture {C þ 2b þ dppf}, obtained from
the reaction of 1b with 1 equiv of dppf (as described above), was
reacted with [PtR(ppy)(SMe2)], in which R = Ph or Me (equiva-
lent amount being calculated approximately as twice the amount
of the components C and free dppf in the starting mixture),
we found that in each case a mixture containing mainly unreacted
2b and an unsymmetrical complex, [(p-MeC6H4)(ppy)Pt-
(μ-dppf)PtR(ppy)] (R = Ph, 3a, or R = Me, 3b), was obtained.
Figure 3. DFT-calculated structures of (a) [PtMe(κ2-C,N-ppy)(κ1-P-
dmpf)] (isomer A), (b) [PtMe(κ1-C-ppy)(κ2-dmpf)] (isomer B), (c)
[PtPh(κ2-C,N-ppy)(κ1-P-dmpf)] (isomer A), and (d) [PtPh(κ1-C-
ppy)(κ2-dmpf)] (isomer B).
phosphorus over nitrogen coordination, the electron-withdraw-
ing ability of the Ar ligand and more importantly its steric
hindrance favor an equilibrium between A and B for the case
in which R = Ph.
Reaction of 1b with 1 equiv of dppf. The results obtained
from reaction of 1b with 1 equiv of dppf are described in
Scheme 5. On the basis of the 31P and 195Pt NMR spectra we
suggest that upon the addition of dppf at room temperature (see
Figure 4), complex 1b is completely converted to the dangling
complex A, via displacement of the SMe2 labile ligand. As was
mentioned above for the related reaction involving addition of
0.5 equiv of dppf (see Scheme 4 and the above related theoretical
discussion), A is in equilibrium with the chelate analogous B. The
two complexes A and B then are reacted with each other to form
1470
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
Figure 4. 31P (top view) and 195Pt (bottom view) NMR spectra obtained after addition of 1 equiv of dppf to a solution of the complex [Pt(p-
MeC6H4)(ppy)(SMe2)], 1b, at room temperature.
It is probable that at first complex C assembles with [PtR-
(ppy)(SMe2)] to form X, but this triplatinum(II) complex is
then fragmented to the unsymmetrical complex 3 and the dangl-
ing dppf complex A. The latter further reacts with [PtR(ppy)-
(SMe2)] to obtain 3; as expected, when R = p-MeC6H4, then 2b
is the main product (see Scheme 6).
Kinetic Studies. Addition of 0.5 equiv of dppf. In this case, we
found that the species involved in the reaction (reactants, pro-
ducts, and intermediates) have overlapped spectra, and so the
rate constant determination by conventional curve-fitting meth-
ods is not feasible. However, by using chemometric methods, as
an emerging computational tool in chemistry, we have been able
to study the complex kinetic systems involved in this process
having severe spectral overlapping. In addition, they can give
information about the number of chemical components coexist-
ing in a reaction, which is useful in mechanistic studies.
The above-mentioned mechanism of the dimer complex for-
mation was further studied by the spectrophotometric kinetics of
the initial addition of 0.5 equiv of dppf in dichloromethane. The
change in the absorbance spectra of the reaction mixture during
a typical kinetic run is shown in Figure 5. Preliminary results
show that when dppf is added to 1b in solution, the absorbance
decreases quickly in the region 330-700 nm in the first 30 min,
and after that the absorbance slightly increases along with time.
The change of absorbance at 360 nm as a function of time is
plotted as the inset in Figure 5. These results show that the
reaction mechanism of addition of dppf to 1b may be considered
as a competitive-consecutive second-order process of the form of
eq 1:
k2
f
1b þ dppf
s
A
k20
ð1Þ
A þ 1b
s 2b
f
Differentiation of the A and B intermediate species seems to be
difficult by using the data obtained from the UV-visible spectro-
scopy, as the species have very similar spectra. From the chemo-
metrics point of view, the kinetic system of eq 1 is denoted as rank
deficient since changes in the concentration of one component as
function of time can be written as a linear function of the other
species. Thus, in spite of having four chemical components, they
can be described by three noncorrelated orthogonal linear
combinations. The proposed mechanism for reaction was con-
firmed by principal component analysis (PCA).9 The main aim
of PCA is to determine the number of principal components, and
the obtained results indicate the numbers of independent sys-
tematic variations in the spectral data. It was found (see score plots
in Figure S1(a) in the Supporting Information) that the recorded
absorbance data matrix can be best explained by three sig-
nificant principal components (PCs). This implies that the
1471
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
Scheme 6
was used.10 To do so, the absorbance spectra of 1b without addition
of dppf were recorded in the same time interval as in the presence
of dppf, and a new data matrix was obtained. Thus, the augmented
data matrix is composed of the changes in absorbance spectra of
1b as a function of time in the presence and absence of dppf. The
number of rows of this matrix is 2 times each individual matrix,
whereas the number of columns is equal to each individual one. As
it is shown (see the score plots of Figure S1(b) in the Supporting
Information), the augmented data matrix is explained by four
factors, and we see that rank deficiency is broken. The augmented
data matrix is now full rank. It should be noted that augmentation of
the reaction mixture data matrix with the absorbance data matrix of
1b alone does not add a new chemical factor since 1b is presented
in both data matrices. Thus data augmentation unraveled one of
the hidden chemical factors.
Once the problem of rank deficiency was overcome, a hard-
modeling-based multivariate curve resolution (MCR) technique
was employed for determination of kinetic constants and resolu-
tion of the pure spectra of the components involved in the
reaction.11 This algorithm is explained briefly in the Supporting
Information.
Figure 5. Change in the absorbance spectra of complex 1b upon
addition of 0.5 equiv of dppf as function of time. Inset is the variation
of absorbance at 360 nm over time.
The estimated rate constants for three repeated batch pro-
cesses by applying the proposed method are k2 = 10.7 ( 0.2 L
mol-1 s-1 and k2 = 0.68 ( 0.05 L mol-1 s-1. The concentration
0
process is rank deficient, and the considered second-order con-
secutive reaction mechanism may well be correct.
Chemometrics resolution of the rank-deficient chemical sys-
tem is a difficult task, and the methods that are able to resolve
such systems have high scientific impacts. To overcome and
break the rank deficiency, a column-wise data matrix augmentation
and spectral profiles of the reactants and products calculated
based on the average rate constants are shown in Figure 6. A high
degree of similarity between the resolved pure spectra of dppf
and complex 1b and their experimental spectra was obtained.
This confirms the accuracy of the results obtained by the
employed chemometric methods. In addition to estimating the
1472
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
Figure 6. Resolved concentration profiles (a) and pure spectra (b) of the species coexisting in the studied reaction.
rate constants of such complex systems by spectrophotometric
method, another advantage of the MCR method is estimating the
pure spectra of intermediate species A, which cannot be recorded
experimentally.
Addition of 1 equiv of dppf. The reaction of 1b with 1 equiv
of dppf was monitored by the more conventional UV-visible
spectrophotometric techniques, and the data were obtained
under second-order conditions ([dppf]0 = [1b]0). The data were
fitted to eq 2 with the programs that are based on the least-
squares method.
Abst ¼ Abs¥ þ ðAbs0 - Abs¥Þ=ð1 þ ½1bꢀ0 ꢁ k2 ꢁ tÞ ð2Þ
The k2 value was evaluated by nonlinear least-squares fitting of
the absorbance-time profiles to eq 2. The same method was
used at different temperatures, and activation parameters were
obtained from the Eyring equation (eq 3 and Figure 7). The
results are collected in Table 1. It is interesting to note that the
value k2 = 10.5 ( 0.5 L mol-1 s-1 at 5 ꢀC, which is related to
the first step of the reaction, is in excellent agreement with the
value obtained from the chemometric method at a concentration
ratio of [dppf]/[1b] = 0.5:1 for the first step of the reaction.
Figure 7. Eyring plot for the reaction of 1b with dppf in CH2Cl2.
ꢀ ꢁ
ꢀ
ꢁ
k2
kB
h
ΔS‡ ΔH‡
ln
¼ ln
þ
-
ð3Þ
second-order rate constant. The rate constant was obtained as
k2 = 10.6 ( 0.6 L mol-1 s-1, in agreement with values obtained
for the 0.5:1 and 1:1 ratio cases described above. Note that the
reaction of dppf with 1b under pseudo-first-order conditions was
too fast to be measured at higher temperatures.
Electroluminescence Properties of Complexes 1b and
2b. Single-layer OLEDs were fabricated on glass substrates that
had beforehand been precleaned by ultrasonic treatment in
detergent, deionized water, acetone, and methanol, respectively,
and pretreated with oxygen plasma cleaner before use. These
layers were then precoated with a layer of indium tin oxide
(ITO). Organic layers and cathode layers (Al) were deposited by
conventional vacuum vapor deposition below 2 ꢁ 10-6 Torr.
The thickness of each layer was determined by a quartz thickness
monitor. The effective size of the light-emitting diode was 9 mm2.
The absorption and electroluminescence (EL) spectra were
measured under ambient conditions at room temperature. The
T
R
RT
Addition of Excess dppf. The addition reaction of dppf to 1b
([dppf]0 . [1b]0) in CH2Cl2 at 5 ꢀC was studied using UV-
visible spectroscopy. In each case, excess dppf at different
concentrations was used, and the change in absorbance was
monitored. The reaction followed good first-order kinetics. The
pseudo-first-order rate constants (kobs) were evaluated by non-
linear least-squares fitting of the absorbance-time profiles to a
first-order equation (eq 4):
Abst ¼ Abs¥ þ ðAbs0 - Abs¥Þ expð - kobstÞ
ð4Þ
A Graph of these pseudo-first-order rate constants against the con-
centration of dppf gave a good straight line plot passing through
the origin, showing a first-order dependence of the rate on the
concentration of dppf (Figure 8), and the slope gave an overall
1473
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
Table 1. Rate Constantsa and Activation Parameters for the Reaction of 1b with dppf in CH2Cl2
T = 5 ꢀC
T = 10 ꢀC
T = 15 ꢀC
T = 20 ꢀC
T = 25 ꢀC
T = 30 ꢀC
T = 35 ꢀC
ΔH‡/kJ mol-1
45.4 ( 1.0
ΔS‡/J K-1 mol-1
-61 ( 4
k2 (L mol-1 s-1
)
10.5
15.3
23.1
32.3
41.3
60.5
79.4
a Estimated errors in k2 values are (4%.
Figure 8. Plot of first-order rate constants (kobs/s-1) for the reaction of
1b (2 ꢁ 10-4 M) with dppf in CH2Cl2 at T = 5 ꢀC vs concentration of
dppf.
Figure 9. Absorption (dashed lines) and photoluminescence spectra
(solid lines) of 1b (blue) and 2b (red).
[Pt(C6H5)(C-N)(SMe2)], 1a or 1a0. To a solution of cis-[Pt-
(C6H5)2(SMe2)2] (200 mg, 0.42 mmol) in acetone (30 mL) was added
2-phenylpyridine (61.3 μL, 0.42 mmol for 1a) or benzo[h]quinoline
(75.8 mg, 0.42 mmol for 1a0), and the reaction mixture was refluxed for
4 h. A light green solution was formed; then the solvent was removed
under reduced pressure and the residue was triturated with cold acetone
(2 ꢁ 2 mL). The products as a light green solid were dried under
vacuum. 1a: Yield: 125 mg; 61%, mp 216 ꢀC (dec). Anal. Calcd for
C19H19NPtS: C, 46.7; H, 3.9; N, 2.9. Found: C, 46.7; H, 3.8; N, 2.8. 1H
absorption spectra were measured using a Perkin-Elmer Lambda
25 UV-vis spectrometer. The EL spectra were recorded using a
Si detector that is calibrated up to 900 nm and is thus suitable for
recording only the visible EL spectra.
Figure 9 shows a comparison between the absorption and EL
spectra of the two complexes 1b and 2b. Thus, while both
complexes show strong absorptions at λ = 320 to 345 nm, 1b has
a longer wavelength absorption peak at λ = 485 nm, which is
significantly red-shifted to λ = 637 nm for 2b. This is consistent
with the lower HOMO-LUMO energy gap in the latter com-
plex. Correspondingly, the EL of compound 2b, peaked at λ =
710 nm, is also significantly red-shifted as compared to that of 1b,
having a peak at λ = 580 nm.
3
NMR data in CDCl3: δ 2.20 (s, JPtH = 25.5 Hz, Me groups of SMe2
ligand, 6H), (aromatic protons): 7.59 (d, 3JPtHo = 57.5 Hz, 3JHoHm = 6.7
Hz, Ho of phenyl ligand, 2 Ho), 8.86 (d, 3JPtH6 = 18.8 Hz, 3JH6H5 = 5.1 Hz,
H6 of ppy ligand, 1 H6), other aromatic protons of phenyl and ppy
ligands: 6.80-8.00, 10 H. 1a0: Yield: 162 mg; 75%, mp 206 ꢀC (dec).
Anal. Calcd for C21H19NPtS: C, 49.2; H, 3.7; N, 2.7. Found: C, 49.4; H,
1
3
3.6; N, 2.7. H NMR data in CDCl3: δ 2.28 (s, JPtH = 26.3 Hz, Me
groups of SMe2 ligand, 6H), (aromatic protons): 9.13 (d, 3JPtH6 = 18.1
Hz, 3JH6H5 = 5.0 Hz, H6 of bhq ligand, 1 H6), other aromatic protons of
phenyl and bhq ligands: 7.00-8.34, 12 H.
’ EXPERIMENTAL SECTION
The 1H NMR spectra were recorded on a Bruker Avance DPX 250
MHz spectrometer, and the 31P and 195Pt NMR spectra were recorded
on a Bruker Avance DRX 500 MHz spectrometer. The operating
frequencies and references, respectively, are shown in parentheses as
follows: 1H (250 MHz, TMS), 31P (202 MHz, 85% H3PO4), and 195Pt
(107 MHz, aqueous Na2PtCl4). The chemical shifts and coupling
constants are in ppm and Hz, respectively. The microanalyses were
performed using a Thermofinnigan Flash EA-1112 CHNSO rapid
elemental analyzer. Kinetic studies were carried out by using a Perkin-
Elmer Lambda 25 spectrophotometer with temperature control using an
EYELA NCB-3100 constant-temperature bath. 2-Phenylpyridine,
benzo[h]quinoline, and 1,10-bis(diphenylphosphino)ferrocene were
purchased from Aldrich, and cis-[PtAr2(SMe2)2] (Ar = Ph, p-MeC6H4,
and p-MeOC6H4), [Pt(p-MeC6H4)(ppy)(SMe2)], 1b, [Pt(p-
MeC6H4)(bhq)(SMe2)], 1b0, [Pt(p-MeOC6H4)(ppy)(SMe2)], 1c,
and [Pt(p-MeOC6H4)(bhq)(SMe2)], 1c0, have been prepared by the
known methods.12a-d
[Pt2(C6H5)2(ppy)2(μ-dppf)], 2a. To a solution of 1a (196 mg,
0.40 mmol) in acetone (30 mL) was added 0.5 equiv of dppf (113.6 mg,
0.20 mmol), and the solution was stirred for 2 h. A yellow solution was
formed; then the solvent was removed under reduced pressure and the
residue was triturated with cold acetone (2 ꢁ 3 mL). The product as a
yellow solid was dried under vacuum. Yield: 82%, mp 266 ꢀC (dec).
Anal. Calcd for C68H54FeN2P2Pt2: C, 58.0; H, 3.9; N, 2.0. Found: C,
57.9; H, 3.8; N, 1.9. NMR data in CDCl3: 1H NMR: δ 3.35 (br s, β, β0 Cp
protons, 4 H), 4.31 (br s, R, R0 Cp protons, 4 H), (aromatic protons):
7.34 (d, 3JHoHm = 6.4 Hz, Ho of phenyl ligand, 4 Ho), 7.76 (d, 3JPtH6 = 18.5
Hz, 3JH6H5 = 8.1 Hz, H6 of ppy ligand, 2 H6), other aromatic protons of
phenyl, ppy, and dppf ligands: 6.50-8.00, 40 H. 31P NMR: δ 21.0 (s,
1JPtP = 2006 Hz, 2 P). 195Pt NMR: δ -2407 (d, 1JPtP = 2000 Hz, 2 Pt).
The following complexes were made similarly using the appropriate
starting precursors.
1474
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
[Pt2(C6H5)2(bhq)2(μ-dppf)], 2a0. This was prepared similarly
using 1a0 (200 mg, 0.39 mmol) and dppf (105.3 mg, 0.19 mmol) in
acetone (30 mL). Yield: 197 mg, 64%, mp 225 ꢀC (dec). Anal. Calcd for
C72H54FeN2P2Pt2: C, 59.4; H, 3.7; N, 1.9. Found: C, 59.6; H, 3.8; N, 2.0.
NMR data in CDCl3: 1H NMR: δ 3.40 (br s, β, β0 Cp protons, 4 H), 4.41
(br s, R, R0 Cp protons, 4 H), (aromatic protons): 8.12 (d, 3JPtH6 = 18.4
Hz, 3JH6H5 = 7.9 Hz, H6 of bhq ligand, 2 H6), other aromatic protons of
Table 2. Crystal Data and Data Collection and Structure
Refinement Details for the Complex [Pt2(p-MeOC6H4)2-
(bhq)2(μ-dppf)], 2c0 CH2Cl2
3
formula
C75H60Cl2FeN2O2P2Pt2
1685.10
fw
cryst syst
monoclinic
C2/c
aryl, bhq, and dppf ligands: 6.85-7.95, 44 H. 31P NMR: δ 20.5 (s, 1JPtP
=
space group
2069 Hz, 2 P). 195Pt NMR: δ -2433 (d, 1JPtP = 2051 Hz, 2 Pt).
a/Å
12.540(3)
14.801(3)
32.631(7)
95.88(3)ꢀ
6025(2)
Pt2(p-MeC6H4)2(ppy)2(μ-dppf)], 2b. This was prepared simi-
larly using 1b (200 mg, 0.4 mmol) and dppf (113.6 mg, 0.2 mmol) in
acetone (30 mL). Yield: 190 mg, 61%, mp 223 ꢀC (dec). Anal. Calcd for
C70H58FeN2P2Pt2: C, 58.6; H, 4.1; N, 1.9. Found: C, 57.9; H, 4.4; N, 1.8.
b/Å
c/Å
β/deg
vol/Å3
1
NMR data in CDCl3: H NMR: δ 2.17 (s, Me groups on the p-tolyl
Z
4
ligands, 6H), 3.39 (br s, β, β0 Cp protons, 4 H), 4.37 (br s, R, R0 Cp
protons, 4 H), (aromatic protons): 6.57 (d, 3JHmHo = 7.2 Hz, Hm of p-
tolyl ligand, 4 Hm), 7.08 (d, 3JPtHo = 65.5 Hz, 3JHoHm = 6.8 Hz, Ho of p-
tolyl ligand, 4 Ho), 7.76 (d, 3JPtH6 = 18.3 Hz, 3JH6H5 = 7.8 Hz, H6 of ppy
ligand, 2 H6), other aromatic protons of aryl, and dppf ligands: 6.45-
7.80, 34 H. 31P NMR: δ 20.6 (s, 1JPtP = 2004 Hz, 2 P). 195Pt NMR: δ -
2409 (d, 1JPtP = 1990 Hz, 2 Pt).
D(calc)/Mg m-3
abs coeff/mm-1
F(000)
1.764
5.064
3144
no. of collected reflns
no. of independent reflns
no. of obsd reflns [I > 2σ(I)]
data/restraints/params
Tmin/Tmax
22 925
9151 [R(int) = 0.086]
6688
[Pt2(p-MeC6H4)2(bhq)2(μ-dppf)], 2b0. This was prepared simi-
larly using 1b0 (200 mg, 0.38 mmol) and dppf (105.3 mg, 0.19 mmol) in
acetone (30 mL). Yield: 208 mg, 68%, mp 270 ꢀC (dec). Anal. Calcd for
C74H58FeN2P2Pt2: C, 59.9; H, 3.9; N, 1.9. Found: C, 59.6; H, 3.9; N, 2.0.
9151/0/387
0.4161/0.5817
1.29 and -1.61
1.056
largest diff peak, hole [e Å-3
Goof (F2)
]
1
NMR data in CDCl3: H NMR: δ 2.09 (s, Me groups on the p-tolyl
ligands, 6H), 3.44 (br s, β, β0 Cp protons, 4 H), 4.47 (br s, R, R0 Cp
R1, wR2 [I > 2σ(I)]
R1, wR2 (all data)
0.0590, 0.1369
0.0855, 0.1529
3
protons, 4 H), (aromatic protons): 6.61 (d, JHmHo = 7.7 Hz, Hm of
p-tolyl ligand, 4 Hm), 8.12 (d, 3JPtH6 = 18.7 Hz, 3JH6H5 = 8.1 Hz, H6 of bhq
ligand, 2 H6), other aromatic protons of aryl, bhq, and dppf ligands:
6.75-7.90, 38 H. 31P NMR: δ 20.0 (s, 1JPtP = 2067 Hz, 2 P). 195Pt NMR:
δ -2435 (d, 1JPtP = 2118 Hz, 2 Pt).
195Pt NMR: δ -2409 (d, 1JPtP = 1994 Hz, Pta), -2821 (t, 1JPtP = 1891
Hz, Ptb).
Monitoring the Reaction by 31P NMR Spectroscopy. A
small sample (10 mg) of 1b was dissolved in CDCl3 (0.75 mL) in an
NMR tube, and either 0.5 or 1 equiv or excess dppf was added. Then the
NMR spectra were recorded.
[Pt2(p-MeOC6H4)2(ppy)2(μ-dppf)], 2c. This was prepared si-
milarly using 1c (200 mg, 0.39 mmol) and dppf (105.3 mg, 0.19 mmol)
in acetone (30 mL). Yield: 202 mg, 66%, mp 208 ꢀC (dec). Anal. Calcd
for C70H58FeN2O2P2Pt2: C, 57.3; H, 4.0; N, 1.9. Found: C, 57.2; H, 3.8;
N, 2.1. NMR data in CDCl3: 1H NMR: δ 3.77 (s, OMe groups on the
p-anisyl ligands, 6H), 3.42 (br s, β, β0 Cp protons, 4 H), 4.32 (br s, R, R0
Cp protons, 4 H), (aromatic protons): 6.42 (d, 3JHmHo = 8.2 Hz, Hm of
p-anisyl ligand, 4 Hm), 7.76 (d, 3JPtH6 = 18.2 Hz, 3JH6H5 = 7.9 Hz, H6 of
ppy ligand, 2 H6), other aromatic protons of aryl, ppy, and dppf ligands:
6.35-7.90, 38 H. 31P NMR: δ 20.8 (s, 1JPtP = 1992 Hz, 2 P). 195Pt NMR:
δ -2405 (d, 1JPtP = 1980 Hz, 2 Pt).
Kinetic Study. A 3.0 mL portion of a 2.0 ꢁ 10-4 M solution of 1b
was transferred into a thermostated quartz cell (path length: 10 mm),
and a known volume of 0.01 M dppf was added using a microsyringe.
The absorbance spectra of the reaction mixture were rapidly monitored
with time.
X-ray Structure Determination. Single crystals of [Pt2(p-
MeOC6H4)2(bhq)2(μ-dppf)], 2c0 CH2Cl2, were grown from a con-
3
[Pt2(p-MeOC6H4)2(bhq)2(μ-dppf)], 2c0. This was prepared si-
milarly using 1c0 (200 mg, 0.36 mmol) and dppf (102.1 mg, 0.18 mmol)
in acetone (30 mL). Yield: 196 mg, 65%, mp 267 ꢀC (dec). Anal. Calcd
for C74H58FeN2O2P2Pt2: C, 58.6; H, 3.9; N, 1.8. Found: C, 58.5; H, 3.8;
N, 1.8. NMR data in CDCl3: 1H NMR: δ 3.71 (s, OMe groups on the
p-anisyl ligands, 6H), 3.42 (br s, β, β0 Cp protons, 4 H), 4.42 (br s, R, R0
Cp protons, 4 H), (aromatic protons): 6.48 (d, 3JHmHo = 8.5 Hz, Hm of
p-anisyl ligand, 2 Hm), 8.12 (d, 3JPtH6 = 19.0 Hz, 3JH6H5 = 8.1 Hz, H6 of
bhq ligand, 1 H6), other aromatic protons of aryl, bhq, and dppf ligands:
6.70-8.00, 38 H. 31P NMR: δ 20.2 (s, 1JPtP = 2055, 2 P). 195Pt NMR:
δ -2432 (d, 1JPtP = 1979, 2 Pt).
centrated CH2Cl2 solution by slow diffusion of n-hexane. A red-
orange needle of approximate 0.21 ꢁ 0.16 ꢁ 0.12 mm in size was
mounted on a glass fiber. All measurements were made on a STOE
IPDSII diffractometer with a graphite monochromator using Mo KR
radiation. The data were collected at 298 K and integrated using the
Stoe X-AREA13 software package. Crystal data and data collection and
structure refinement details are listed in Table 2. A numerical absorp-
tion correction was applied using the X-RED14 and X-SHAPE15
software programs. The data were corrected for Lorentz and polariza-
tion effects. No decay correction was applied. The structure was
solved by direct methods.16 All of the non-hydrogen atoms were
refined anisotropically. All of the hydrogen atoms were positioned
geometrically and refined with a riding model approximation with
[(p-MeC6H4)(ppy)Pt(μ-dppf)Pt(p-MeC6H4)(ppy-K1C)(dppf-
K1P)], C. In the above procedure, used to prepare 2b, when1b (200 mg,
0.4 mmol) was reacted with 1 equiv of dppf (227.2 mg, 0.4 mmol),
complex C was obtained as a mixture with 2b and dppf, and the
components of the mixture were not isolable. NMR data for complex
U
iso(H) = 1.2 or 1.5Ueq(C). Molecular graphics were performed using
SHELXTL.16 Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic Data Center,
CCDC no. 786085. Copies of this information may be obtained free of
charge from the Director, CCDC, 12 Union Road, CB2 1EZ, U.K.
(http://www.ccdc.cam.ac.uk).
1
C in CDCl3: H NMR: δ 2.18 and 2.20 (s, Me groups on the p-tolyl
ligands, 6 H), 3.49-4.80 (Cp protons, 16 H), (aromatic protons): 6.14-
8.18 (aromatic protons of p-tolyl, ppy, and dppf ligands, 80 H). 31P
NMR: δ 20.9 (s, 1JPtP = 2014, Pa, 1 P), 15.5 (d, 1JPtP = 1895, 2JPbPc = 15,
Pc, 1 P), 12.2 (d, 1JPtP = 1845, 2JPbPc = 15, Pb, 1 P), -17.8 (s, Pd, 1 P).
DFT Calculations. Gaussian 09 was used8 to fully optimize all the
structures reported in this paper at the B3LYP17 level of density
1475
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
functional theory. The effective core potential of Hay and Wadt with a
double-ξ valence basis set (LANL2DZ) was chosen to describe Pt.18
The 6-31G(d) basis set was used for other atoms.19 A polarization
function of ξf = 1.472 was also added to Pt.20 This basis set
combination will be referred to as BS1. Frequency calculations were
carried out at the same level of theory as for structural optimization.
To further refine the energies obtained from the B3LYP/BS1 calcula-
tions, we carried out single-point energy calculations for all the
structures with a larger basis set (BS2). BS2 utilizes the quadruple-ζ
valence def2-QZVP21 basis set on Pt along with the corresponding
ECP and the 6-311þG(2d,p) basis set on other atoms. We have used
the energies obtained from B3LYP/BS2//B3LYP/BS1 throughout
the paper unless otherwise stated.
’ ASSOCIATED CONTENT
S
Supporting Information. Crystallographic data in CIF
b
format and details of the chemometric method used for the
kinetic study. These materials are available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: nabavi@chem.susc.ac.ir (S.M.N.); rashidi@chem.susc.
ac.ir (M.R.).
’ ACKNOWLEDGMENT
We thank the Iran National Science Foundation (Grant No.
87041279) and the Shiraz University Research Council for
financial support. We also thank Professor James H. Espenson and
Dr. Sirous Jamali for helpful discussions and Mr. Sajedian and Mr.
Bijanzadeh for assistance with the multinuclear NMR experiments.
’ CONCLUSION
Although the complexes with monodentate dppf ligands are
rare,6a and there is no example of such a mode of dppf ligand in
platinum chemistry, we found that replacement of the labile
ligand SMe2 in the cyclometalated complex [Pt(p-MeC6H4)-
(ppy)(SMe2)], 1b, by dppf first gives the stable, uncommon
complex [Pt(p-MeC6H4)(ppy)(dppf-κ1P)], A, with a monoden-
tate dppf ligand. Complex A is obtained (under 1b:dppf = 1:0.5
ratio) in equilibrium with an isomeric form, [Pt(p-MeC6H4)-
(dppf)(ppy-κ1C)], B. The intermediate complex A is then slowly
reacted with the second half of starting complex 1b to give the
final Pt(II)-Pt(II) binuclear complex [Pt2(p-MeC6H4)2(ppy)2-
(μ-dppf)], 2b. In complex B, as compared to complex A, the
great chelating appetite of dppf has caused the cyclometalated
chelate to open up from the N donor atom. This indicates that
opening of the rather strong cyclometalated ring from the N
chelating atom has a rather comparable tendency to that of
converting a chelating dppf ligand to an uncommon case of
monodentate dppf ligand. Also, the chelating dppf ligand in
organoplatinum(II) complexes has a PPtP bite angle usually
close to 100ꢀ,4,5 which is significantly greater than 90ꢀ, and this
implies that the chelating dppf is also under strain. So as observed
in the present study (see Scheme 4) and our previous one,3 the
chelating dppf and the chelating C-N ligands are sometimes
competing with each other in “chelate opening”.
’ REFERENCES
(1) (a) Togni, A.; Hayashi, T., Ed. Ferrocenes; VCH: New York,
ꢀ
1995. (b) Stꢁepniꢀcka, P., Ed. Ferrocenes: Ligands, Materials and Biomole-
cules; John Wiley & Sons Ltd.: England, 2008. (c) Teo, P.; Koh, L. L.;
Andy Hor, T. S. Inorg. Chem. 2008, 47, 9561.
(2) (a) Bandoli, G.; Dolmella, A. Coord. Chem. Rev. 2000, 209, 161.
(b) Sato, M.; Sekino, M.; Akabori, S. J. Organomet. Chem. 1988, 344,
C31. (c) Sato, M.; Shigeta, H.; Sekino, M.; Akabori, S. J. Organomet.
Chem. 1993, 458, 199.
(3) (a) Jamali, S.; Nabavizadeh, S. M.; Rashidi, M. Inorg. Chem. 2008,
47, 5441. (b) Evans, R. C.; Douglas, P.; Winscom, C. J. Coord. Chem. Rev.
2006, 250 (15-16), 2093. (c) Lu, W.; Mi, B.-X.; Chan, M. C. W.; Hui,
Z.; Che, C.-M.; Zhu, N.; Lee, S.-T. J. Am. Chem. Soc. 2004, 126, 4958.
(d) Wong, W.-Y.; He, Z.; So, S.-K.; Tong, K.-L.; Lin, Z. Organometallic
2005, 24, 4079. (e) Sun, W.; Zhu, H.; Barron, P. M. Chem. Mater. 2006,
18, 2602. (f) Ravindranathan, D.; Vezzu, D. A. K.; Bartolotti, L.; Boyle,
P. D.; Huo, S. Inorg. Chem. 2010, 49, 8922. (g) Bai, D. R.; Baumgartner,
T. Organometallics 2010, 29, 3289. (h) Mathew, I.; Sun, W. Dalton Trans.
2010, 39, 5885. (i) Wilson, J. J.; Fedoce Lopes, J.; Lippard, S. J. Inorg.
Chem. 2010, 49, 5303.
(4) (a) Thorn, D. L.; Fultz, W. C. J. Phys. Chem. 1989, 93, 1234.
(b) Smith, D. C.; Haar, C. M., Jr.; Stevens, E. D.; Nolan, S. P.
Organometallics 2000, 19, 1427. (c) Scolaro, L. M.; Plutino, M. R.;
Romeo, A.; Romeo, R.; Ricciardi, G.; Belviso, S.; Albinati, A. Dalton
Trans. 2006, 2551.
(5) (a) Shahsavari, H. R.; Rashidi, M.; Nabavizadeh, S. M.; Habib-
zadeh, S.; Heinemann, F. W. Eur. J. Inorg. Chem. 2009, 3814. (b) Ma,
J.-F.; Yamamoto, Y. Inorg. Chim. Acta 2000, 299, 164. (c) Ares, R.;
Lꢂopez-Torres, M.; Fernꢂandez, A.; Pereira, M. T.; Alberdi, G.; Vꢂazquez-
García, D.; Fernꢂandez, J. J.; Vila, J. M. J. Organomet. Chem. 2003, 665, 76.
(d) Smyth, D. R.; Hester, J.; Young, V. G., Jr; Tiekink, E. R. T.
CrystEngComm 2002, 4, 517.
When complex 2b, with its “anticlinal eclipsed” conformation,
is dissolved in solvent, a fluxionality of the type shown in
Scheme 3 is suggested to occur so that on average the “antiperi-
planar staggered” conformation, with an ideal torsion angle of
180ꢀ, is observed.
Kinetic study of formation of 2b from the reaction of 1b with
dppf (1:0.5) is done, and a competitive-consecutive second-
order reaction mechanism is proposed. The rate constants of the
two steps were obtained using chemometric methods, as k2 =
10.7 ( 0.2 L mol-1 s-1 and k2 = 0.68 ( 0.05 L mol-1 s-1
.
0
However, when 1 equiv of dppf is used, the products were found
to be different (see Schemes 4, 5, and 6). Thus, again at first a
mixture containing complexes A and B is formed, interestingly
with a rate constant value of k2 = 10.5 ( 0.4 L mol-1 s-1 similar
to that estimated under second-order conditions by UV-visible
spectroscopy. Then by the reaction of complex A with complex
B, during which the dangling P atom in complex A would attack
the platinum center in complex B, a new aggregated species, [(p-
MeC6H4)(ppy)Pt(μ-dppf)Pt(p-MeC6H4)(ppy-κ1C)(dppf-κ1P)],
C, having a dangling dppf ligand, is formed. Attempts to extend the
length of aggregation in C were not successful and ended with chain
breaking.
(6) (a) Complexes with monodentate dppf ligands are rare, one
example being [Co2(η1-dppf)(μ2-(MeO2C)2C2)(CO)5],6b in which
the conformation of dppf is “anticlinal eclipsed”, with a Cp(centroid)
Fe Cp(centroid) twist angle of 154.9ꢀ. We also have recently
3 3 3 3 3 3
suggested the formation of an intermediate complex, [PtMe(ppy)(dppf-
κ1P)], formed during a complicated reaction:3a (b) McAdam, C. J.;
Duffy, N. W.; Robinson, B. H.; Simpson, J. J. Organomet. Chem. 1997,
527, 179.
(7) (a) Yeo, W.-C.; Vittal, J. J.; Koh, L. L.; Tan, G. K.; Leung, P.-H.
Organometallics 2006, 25, 1259. (b) Liu, F.; Pullarkat, S. A.; Tan, K.-W.;
Li, Y.; Leung, P.-H. Organometallics 2009, 28, 6254.
(8) Frisch, M. J.; et al. Gaussian 09, revision A.02; Gaussian, Inc.:
Wallingford, CT, 2009.
1476
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477

Organometallics
ARTICLE
(9) Gemperline, P. Practical Guide to Chemometrics; CRC:
New York, 2006.
(10) del Río, V.; Callao, M. P.; Larrechi, M. S.; de Espinosa, L. M.;
Ronda, J. C.; Cꢂadiz, V. Anal. Chim. Acta 2009, 649, 148.
(11) Thurston, T. J.; Brereton, R. G. Analyst 2002, 127, 659.
(12) (a) Rashidi, M.; Hashemi, M.; Khorasani-Motlagh, M.; Pudde-
phatt, R. J. Organometallics 2000, 19, 275. (b) Rashidi, M.; Fakhroeian,
Z.; Puddephatt, R. J. J. Organomet. Chem. 1991, 406, 261. (c) Pudde-
phatt, R. J.; Thomson, M. A. J. Organomet. Chem. 1982, 238, 231.
(d) Nabavizadeh, S. M.; Golbon Haghighi, M.; Esmaeilbeig, A. R.;
Raoof, F.; Mandegani, Z.; Jamali, S.; Rashidi, M.; Puddephatt, R. J.
Organometallics 2010, 29, 4893.
(13) X-AREA Program for the Acquisition and Analysis of Data,
version 1.30; Stoe & Cie GmbH: Darmstadt, Germany, 2005.
(14) X-RED Program for Data Reduction and Absorption Correction,
version 1.28b; Stoe & Cie GmbH: Darmstadt, Germany, 2005.
(15) X-SHAPE Program for Crystal Optimization for Numerical
Absorption Correction, version 2.05; Stoe & Cie GmbH: Darmstadt,
Germany, 2004.
(16) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.
(17) (a) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37,
785. (b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (c) Miehlich, B.;
Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989, 157, 200.
(18) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
(b) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284. (c) Roy, L. E.;
Hay, P. J.; Martin, R. L. J. Chem. Theory Comput. 2008, 4, 1029.
(19) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213.
(20) Ehlers, A. W.; B€ohme, M.; Dapprich, S.; Gobbi, A.; H€ollwarth,
A.; Jonas, V.; K€ohler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G.
Chem. Phys. Lett. 1993, 208, 111.
(21) Weigend, F.; Furche, F.; Ahlrichs, R. J. Chem. Phys. 2003, 119,
12753.
1477
dx.doi.org/10.1021/om101055n |Organometallics 2011, 30, 1466–1477