Oxidatively induced reactions of a diimine platinum(IV) tetramethyl complex

Article abstract of DOI:10.1016/j.jorganchem.2007.02.040

The oxidation of the Pt(IV) tetramethyl complex [Ar{single bond}N{double bond, long}CH{single bond}CH{double bond, long}N{single bond}Ar]PtMe₄ (Ar = 2,6-Me₂C₆H₃) has been investigated in acetonitrile and dichloromethane. Cyclic voltammetry demonstrates that the irreversible oxidation of [Ar{single bond}N{double bond, long}CH{single bond}CH{double bond, long}N{single bond}Ar]PtMe₄ occurs at a slightly less positive oxidation potential than the irreversible oxidation of the analogous Pt(II) species [Ar{single bond}N{double bond, long}CH{single bond}CH{double bond, long}N{single bond}Ar]PtMe₂. The product distribution arising from the oxidation depends strongly on the reaction conditions and includes cationic Pt(IV) species (acetonitrile, dichloromethane solvents) and Pt(II) species (dichloromethane only). Evidence is presented that suggests that homolytic cleavage of a weakened Pt{single bond}C bond in [ArN {double bond, long} CH {single bond} CH {double bond, long} NAr] PtMe₄^{{radical dot} +} is involved in the oxidatively induced reactions.

Full text of DOI:10.1016/j.jorganchem.2007.02.040

Journal of Organometallic Chemistry 692 (2007) 3223–3230
www.elsevier.com/locate/jorganchem
Oxidatively induced reactions of a diimine platinum(IV)
tetramethyl complex
*
Bror J. Wik, Mats Tilset
Department of Chemistry, University of Oslo, P.O. Box 1033 Blindern, N-0315 Oslo, Norway
Received 22 December 2006; received in revised form 23 February 2007; accepted 23 February 2007
Available online 12 March 2007
Abstract
The oxidation of the Pt(IV) tetramethyl complex [ArAN@CHACH@NAAr]PtMe4 (Ar = 2,6-Me2C6H3) has been investigated in ace-
tonitrile and dichloromethane. Cyclic voltammetry demonstrates that the irreversible oxidation of [ArAN@CHACH@NAAr]PtMe4
occurs at
a slightly less positive oxidation potential than the irreversible oxidation of the analogous Pt(II) species
[ArAN@CHACH@NAAr]PtMe2. The product distribution arising from the oxidation depends strongly on the reaction conditions
and includes cationic Pt(IV) species (acetonitrile, dichloromethane solvents) and Pt(II) species (dichloromethane only). Evidence is pre-
sented that suggests that homolytic cleavage of a weakened PtAC bond in ½ArN@CHACH@NArꢀPtMeÅ4þ is involved in the oxidatively
induced reactions.
Ó 2007 Elsevier B.V. All rights reserved.
Keywords: Platinum methyl; Pt(IV); Oxidation; Cyclic voltammetry; Homolysis; Diimine
1. Introduction
tions [9–11]. The up-to-date mechanistic insight that
pertains to the CAH activation step in Scheme 1 was
recently reviewed [6]. Some time ago [12], we investigated
the chemical and electrochemical oxidation chemistry of
(NAN)PtMe2 complexes, obviously of relevance to the oxi-
dation step in Scheme 1. The primary oxidatively induced
reaction in solution was found to be a methyl transfer pro-
cess, formally a disproportionation of two transient Pt(III)
species, to furnish the cationic Pt(II) and Pt(IV) products
that are depicted in Scheme 2.
Cyclic voltammograms of (NAN)PtMe2 species reveal
well-defined, reversible reduction waves, whereas the
oxidation process is irreversible and rather ill-defined [12]
with peak potentials around +0.6 V vs. Cp2Fe0/+; thermo-
dynamically significant electrode potentials could not be
extracted with the desired accuracy. Bercaw and co-work-
ers [13] investigated the Pt(II)/Pt(IV) oxidation process
via outer-sphere oxidations (i.e. equilibrium measurements
of anionic ligand exchange reactions). Kaim and
co-workers reported a rare case of a reversible one-electron
oxidation of a diimine Pt(II) complex, (CyAN@CHACH@
NACy)Pt(mesityl)2 [14]. Interestingly, irreversible electro-
More than 35 years ago, Garnett and Hodges [1,2] dem-
onstrated that aromatic CAH bonds could be activated by
aqueous Pt(II) salts, and the Shilov group [3–5] discovered
shortly thereafter that even the more unreactive aliphatic
CAH bonds can be similarly activated and functionalized.
Since then, substantial efforts have been made to under-
stand the mechanistic details of such reactions, both on
what is commonly termed the ‘‘Shilov system’’ and on
organometallic model systems [6–8]. As a result of these
efforts, the general mechanism that is depicted in Scheme
1 has emerged for the CAH functionalization by the Shilov
system.
We and others have reported that protonation of a-dii-
mine Pt(II) complexes (NAN)PtMe2 (NAN = aryl-
N@CRACR@N-aryl; aryl = substituted phenyl and
R = H/Me) in trifluoroethanol generates species that acti-
vate aliphatic and aromatic CAH bonds under mild condi-
*
Corresponding author.
E-mail address: mats.tilset@kjemi.uio.no (M. Tilset).
0022-328X/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.02.040

3224
B.J. Wik, M. Tilset / Journal of Organometallic Chemistry 692 (2007) 3223–3230
C-H activation
R—H,—H+
Pt wire, and the Ag wire reference electrode assembly
was filled with acetonitrile/0.01 M AgNO3/0.1 M
Bu4NþPFꢁ6 . The reference electrode was calibrated against
Cp2Fe0/+, which was also used as reference in this work.
The positive feedback iR compensation circuitry of the
potentiostat was employed; the separation of anodic and
cathodic peaks for Cp2Fe oxidation was 60–62 mV in
acetonitrile.
PtII
ROH + H+
PtII
R
PtIV
Functionalization
Oxidation
H2O
PtII
PtIV
R
The diimine ligand ArAN@CHACH@NAAr [15],
Pt2Me8(lASMe2)2 [16], Pt2Me4(lASMe2)2 [17], and
Cp2Fe+OTfꢁ [18] were prepared by published procedures.
All other reagents were purchased from commercial suppli-
ers and used as obtained.
Scheme 1.
0.5(N–N)PtMe(NCMe)+
0.5(N–N)PtMe3(NCMe)+
+
–e–
MeCN
•+
(N–N)PtMe2
(N–N)PtMe2
Scheme 2.
2.2. Synthesis of [ArAN@CHACH@NAAr]PtMe4
[Ar = 2,6-Me2C6H3] (1)
chemical oxidation potentials for Pt(II) and Pt(IV) com-
plexes (NAN)PtMe2 and (NAN)PtMe4 were found to be
quite similar, and it was concluded [14] that ‘‘. . .the bind-
ing of two additional methyl carbanions compensates
almost exactly for the effect of the higher metal oxidation
state.’’ If Pt(IV) complexes are indeed as readily oxidized
as closely related Pt(II) species that are involved in CAH
activation reactions, it will obviously be of great relevance
to understand the oxidation chemistry of Pt(IV) species in
detail in relation to the Shilov chemistry.
In this work, we present some results of an investigation
of the oxidation chemistry of a Pt(IV) diimine complex
(ArAN@CHACH@NAAr)PtMe4 (1) where Ar = 2,6-
Me2C6H3. It will be demonstrated that a homolytic PtAMe
bond cleavage is likely to be a key step in much of the oxi-
datively induced reactivity of 1.
The diimine ligand (110 mg, 0.42 mmol) was added to
a suspension of Pt2Me8(l-SMe2)2 (120 mg, 0.19 mmol) in
dry diethyl ether (20 mL) and the mixture was stirred
for 17 h. The product was isolated by removal of the
solvent, and washed with pentane. Removal of the resid-
ual solvents by evaporation produced pure material in
1
2
76% yield. H NMR (200 MHz, CDCl3) d ꢁ0.05 (s, J-
2
(
195PtAH) = 46.0 Hz, 6H, PtAMeax), 0.60 (s, J(195PtAH) =
72.8 Hz, 6H, PtAMeeq), 2.24 (s, 12H, ArAMe), 7.1 (m,
3
6H, ArAH), 8.75 (s, J(195PtAH) = 29.7 Hz, 2H, NCH).
2.3. Synthesis of [ArAN@CHACH@NAAr]PtMe2
[Ar = 2,6-Me2C6H3] (2)
The following is an adaption of a published procedure
[12]. The diimine ligand (203 mg, 0.77 mmol) was added
to a solution of Pt2Me4(l-SMe2)2 (200 mg, 0.35 mmol) in
toluene (10 mL). The reaction mixture was stirred for 3 h
and filtered, and the solvent was removed under vacuum.
The residue was dissolved in ether and filtered before the
addition of pentane. Large crystals of the wanted
compound formed (85 mg, 54%) after storage at ꢁ30 °C
for 15 h. 1H NMR (200 MHz, CDCl3) d 1.66 (s,
2J(195PtAH) = 87.6 Hz, 6H, PtAMe), 2.25 (s, 12H,
2. Experimental part
2.1. General procedures
1H NMR spectra were recorded on Bruker Avance DPX
1
200 and 300 instruments. H NMR shifts are reported in
parts per million relative to tetramethylsilane, with the
residual solvent peak as internal standard. All handling
of organometallic compounds were done under an atmo-
sphere of N2 with the use of standard Schlenk or drybox
techniques. The Pt(IV) complex 1 is particularly photosen-
sitive [14], so all handling of this material was done with
minimum exposure to light. Elemental analyses were per-
formed by Ilse Beetz Mikroanalytisches Laboratorium,
Kronach, Germany. For electrochemical measurements,
acetonitrile and dichloromethane containing the support-
ing electrolyte were passed through a column of active neu-
tral alumina to remove water and acidic impurities. The
electrolyte was freed of air by purging with purified and
solvent saturated argon, and all measurements were carried
out under a blanket of argon. Electrochemical measure-
ments were performed with an EG&G-PAR Model 273
Potentiostat/Galvanostat. The working electrodes were Pt
disc electrodes (d = 0.6 mm), the counter electrode was a
ArAMe),
ꢂ7.2
(m,
6H,
ArAH),
9.39
(s,
3J(195PtAH) = 34.7 Hz, 2H, NCH). 13C{1H} NMR
(300 MHz, CD3CN) d ꢁ14.4 (1J(195PtAC) = 802.9 Hz,
Pt–Me), 17.5 (ArAC), 127.3, 128.5, 130.3, 149.2
(2J(195PtAC) = 16.0 Hz, Cipso), 165.8 (2J(195PtAC) =
10.2 Hz, N@CH). Anal. Calc. for C20H26N2Pt: C, 49.07;
H, 5.35; N, 5.72; Pt, 39.85. Found: C, 50.03; H, 5.47; N,
5.62; Pt, 38.9%.
2.4. Synthesis of [ArAN@CHACH@NAAr]-
PtMe3(NCMe)+OTfꢁ (Ar = 2,6-Me2C6H3) (3 Æ OTfꢁ)
A solution of 2 (20.5 mg, 0.042 mmol) in acetonitrile
(3 mL) was cooled at ꢁ12 °C under an N2 atmosphere.
To this solution, CF3SO3Me (5.5 lL, 0.05 mmol, ca.
1 equiv.) was added. The reaction mixture was stirred for

B.J. Wik, M. Tilset / Journal of Organometallic Chemistry 692 (2007) 3223–3230
3225
3.5 h during which it was allowed to reach ambient temper-
ature. The color of the mixture changed from deep purple
to clear orange. Filtration, evaporation of the solvent, and
washing twice with ether left the product as a yellow solid
less than 1 equiv.) was dissolved in CD2Cl2 (350 lL). This
solution was transferred to the NMR tube by a syringe,
while keeping the tube in the dark to avoid photo-induced
reactions. A 1H NMR spectrum was recorded and CH3CN
(50 lL) was added. The resulting NMR spectra displayed
mainly 3 and 4 and small quantities of unreacted 1.
1
(23.3 mg, 80%). H NMR (200 MHz, CD3CN) d 0.83 (s,
2J(195PtAH) = 69 Hz,
6H,
PtAMe),
0.83
(s,
2J(195PtAH) = 76 Hz, 3H, PtAMe), 2.12 (s, 3H, CH3CN),
2.23 (s, 6H, ArAMe), 2.29 (s, 6H, ArAMe), 7.25 (m, 6H,
Ar–H), 8.93 (s, 3J(195PtAH) = 27 Hz, 2H, NCHCHN).
13C{1H} NMR (300 MHz, CD3CN) d ꢁ4.13, ꢁ2.89
(PtAMe), 19.1, 19.5 (ArAMe), 129.1, 130.1, 130.2, 130.8,
130.9, 145.6, 171.5. Anal. Calc. for C24H32F3N3O3PtS: C,
41.50; H, 4.64; N, 6.05; Pt, 28.08. Found: C, 42.12; H,
4.75; N, 6.45; Pt, 27.82%.
2.8. Constant-potential electrochemical oxidation of 1 in
acetonitrile
All electrolysis experiments were performed in an H-
shaped cell, the compartments of which were separated
by a medium-frit glass junction. Platinum-gauze working
electrodes were used. In the experiments, 1 (5.2 mg,
0.01 mmol) was added to acetonitrile/0.05 M Me4NþBF4ꢁ
(25 mL) electrolyte in the electrolysis cell. The solutions
were electrolyzed at constant potentials until the current
had dropped to less than 0.2 mA. The electrolyzed solution
was immediately transferred to a round-bottomed flask,
and the solvent was removed by vacuum transfer. The res-
idue was extracted with dichloromethane (leaving the elec-
trolyte behind), and the extract was filtered. The
dichloromethane was removed from the filtrate. The resi-
2.5. Generation of [ArAN@CHACH@NAAr]-
PtMe(NCMe)+OTfꢁ (Ar = 2,6-Me2C6H3) (4 Æ OTfꢁ) in
an NMR tube
To an NMR tube equipped with a Teflon needle valve
and vent, 2 (10 mg, 0.02 mmol) was added. Acetonitrile-
d3 (700 lL) was then added by vacuum transfer into the
tube. To the resulting solution, triflic acid (2.0 lL,
0.022 mmol) was added. The purple color immediately
1
due was dissolved in dichloromethane-d2, and a H NMR
1
changed to dark yellow. The H NMR spectrum demon-
spectrum was acquired. The electrolyses were performed
at E = 0.7 V vs. Cp2Fe0/+, and three separate experiments
demonstrated the passage of 1.03, 1.01, and 1.00 faraday/
strated quantitative conversion to the same compound that
appeared as one of the products in the oxidations that were
performed in CD2Cl2 (see below). 1H NMR (200 MHz,
CD3CN) d 0.19 (s, CH4, ca. 70% obsd. yield), 0.68 (s,
2J(195PtAH) = 76.3 Hz, 3H, PtAMe), 2.24 (s, 6H,
ArAMe), 2.34 (s, 6H, ArAMe), 7.25 (m, 6H, ArAH),
1
mol of charge. The H NMR spectrum showed 3 as the
only recognizable product. 1H NMR of 3 (200 MHz,
CD2Cl2): d 0.86 (s, 2J(195PtAH) = 71.1 Hz, 9H Pt–Me),
ꢂ2.3 (s, 12H, ArAMe), 7.22 (m, 6H, ArAH), 8.99 (s,
3J(195PtAH) = 26.5 Hz, 2H, NCH).
3
8.87 (s, J(195PtAH) = 113.5 Hz, 1H, NCHC0H0N0), 8.90
(s, 3J(195PtAH) = 42.9 Hz, 1H, NCHC0H0N0). 13C{1H}
NMR (300 MHz, CD3CN) d ꢁ13.1, 17.7 (ArAMe), 17.9,
129.0, 129.1, 129.3, 129.6, 130.1, 131.3, 145.66, 145.73,
167.7, 176.2.
2.9. Variable-temperature NMR spectra of 3
1H NMR spectra of 3 Æ OTfꢁ (ꢂ5 mg) in CDCl3
(0.7 mL) were recorded at temperatures ranging from
ꢁ45 to +55 °C. The spectra were recorded on the Bruker
Avance 300 MHz instrument, with evaporation of liquid
nitrogen as the cooling source. A small amount (10 lL)
of acetonitrile was used to ‘‘spike’’ the sample after the
measurements to confirm and identify the resonance corre-
sponding to free acetonitrile. The spectra are further
described in the following section.
2.6. Chemical oxidation of 1 in acetonitrile-d3
Inside a drybox, an NMR tube with a screw cap and
septum was loaded with 1 (4.3 mg, 0.0083 mmol) and ace-
tonitrile-d3 (400 lL). In a glass vial with screw cap and sep-
tum, Cp2Fe+OTfꢁ (2.6 mg, 0.0078 mmol, slightly less than
1 equiv.) was dissolved in acetonitrile-d3 (300 lL). This
solution was then transferred to the NMR tube by a syr-
inge, while keeping the tube in the dark to avoid photo-
induced reactions. The resulting NMR spectra revealed
several PtAMe resonances, including small quantities of
unreacted 1, but the main product and the only identifiable
platinum compound was 3, formed in ca. 70% yield.
3. Results and discussion
3.1. Preparative work
The Pt(IV) complex [ArAN@CHACH@NAAr]PtMe4
(1) was readily prepared from the diimine ligand and
Pt2Me8(SMe2)2 as described in Section 2. The Pt(II) ana-
logue [ArAN@CHACH@NAAr]PtMe2 (2) was similarly
prepared from the diimine and Pt2Me4(SMe2)2. Treatment
2.7. Chemical oxidation of 1 in dichloromethane-d2
Inside a drybox, an NMR tube with a screw cap and sep-
tum was loaded with 1 (7.5 mg, 0.015 mmol), and CD2Cl2
(350 lL) was added by syringe. In a glass vial with screw
cap and septum, Cp2Fe+OTfꢁ (4.4 mg, 0.013 mmol, slightly
of
[ArAN@CHACH@NAAr]PtMe3(NCMe)+OTfꢁ (3 Æ OTf),
whereas protonation of in acetonitrile furnished
2 with methyl triflate in acetonitrile generated
2

3226
B.J. Wik, M. Tilset / Journal of Organometallic Chemistry 692 (2007) 3223–3230
[ArAN@CHACH@NAAr]PtMe(NCMe)+OTfꢁ (4 Æ OTf).
Species 3 and 4 are of relevance as products from the oxi-
dation of 1 (vide infra). The 1H NMR spectra of 1, 2, and 4
are quite straightforwardly interpreted, with diagnostic
2J(195PtAH) satellites around the PtAMe resonances, and
are consistent with the structures and coordination geome-
tries depicted in Scheme 3.
equatorial PtAMe signals are quite different and two sets
of satellites are observed. The fact that the two sets of sat-
ellites have not coalesced establishes that rapid interconver-
sion of axial and equatorial methyls does not occur on the
measurement time scale in acetonitrile-d3. Similar dynamic
behavior, temperature, and/or solvent effects haveꢁprevi-
ously been observed for other ðNANÞPtMeþ3 OTf [19–
21] complexes. In the present case, this behavior may be
understood with Scheme 4 as the basis: In dichloromethane
or chloroform solvents, acetonitrile dissociation leads to
the five-coordinate intermediate in a medium that contains
very little free acetonitrile. The five-coordinate therefore
has ample time to scramble the axial and equatorial
PtAMe groups before acetonitrile reassociation. On the
other hand, in acetonitrile solvent, ligand reassociation
occurs much faster, and this inhibits the scrambling pro-
cess. (Recent isolated five-coordinate Pt(IV) trialkyl species
have a near square pyramidal structure with distinct axial
and equatorial positions in the solid state; these sites
undergo rapid exchange in solution [22–25]). The involve-
ment of acetonitrile dissociation was demonstrated by the
addition of 10 lL of acetonitrile to a sample of 3 Æ OTf in
chloroform-d. A new resonance arising from the uncoordi-
nated acetonitrile (d 2.06 at ꢁ45 °C) appeared. When the
sample was heated to 55 °C, the two resonances from free
and coordinated (d 2.37 at ꢁ45 °C) acetonitrile coalesced
into one broad peak at d 2.21, indicating facile exchange
between coordinated and uncoordinated acetonitrile at
55 °C.
Whereas the NMR spectra of 1, 2, and 4 were straight-
forwardly assigned and reflected the molecular symmetry,
the appearance of the 1H NMR PtAMe signals from 3 were
more complicated as they turned out to be highly solvent
(Fig. 1) and temperature dependent. The signal arising
from the three PtAMe groups appeared as one broadened
signal with broadened 195Pt satellites in dichloromethane-
d2 and chloroform-d. This indicates that the axial PtAMe
group and the two equatorial ones undergo fairly rapid
exchange on the measurement time scale. This was corrob-
orated by variable-temperature NMR spectra in CDCl3: at
+55 °C, one sharp PtAMe signal is seen as a weighted aver-
age for all three methyl groups. Below 0 °C, the axial and
at the two equatorial PtAMe groups give rise to separate
signals in a 1:2 ratio. It is also noteworthy that at the
low-temperature limit, the aryl-Me groups of 3 gave rise
to two signals that correspond to orientations ‘‘above’’
and ‘‘below’’ the coordination plane of Pt (a consequence
of the aryl groups being more or less perpendicularly ori-
ented with respect to the coordination plane, with a steri-
cally imposed restricted rotation around the NAC(aryl)
bonds). At the high-temperature limit, these two signals
had also undergone coalescence. However, when the spec-
trum was acquired in acetonitrile-d3, sharp PtAMe peaks
were observed (Fig. 1, bottom). The central peaks of the
axial and equatorial PtAMe groups were however indistin-
guishable due to coincidental chemical shift equivalence.
However, the 2J(195PtAH) couplings for the axial and
3.2. Cyclic voltammetry
Acetonitrile/0.1 M
Bu4NþPFꢁ6
and
THF/0.2 M
Bu4NþPFꢁ6 solutions of 1 were examined by cyclic voltam-
metry at 20 °C. In a cyclic voltammogram of the acetoni-
Me
μ
[Pt2(Me)8( -SMe2)2]
N
N
μ
N
N
Me
Me
N
N
Me
Me
[Pt2(Me)4( -SMe2)2]
Pt
Pt
Et2O
Toluene
Me
1
2
MeOSO2CF3
MeCN
Me
+
Pt
N
N
NCMe
Me
N
N
Me
Me
+
Pt
NCMe
4
3
Scheme 3.

B.J. Wik, M. Tilset / Journal of Organometallic Chemistry 692 (2007) 3223–3230
3227
Fig. 1. PtAMe region of the ambient-temperature 1H NMR spectrum of 3 Æ OTf in three solvents. A: CD3CN, B: CDCl3, C: CD2Cl2.
Ar
N
Ar
N
Me
+
Pt
Me
+
Pt
Me
Me
Me
Me
+
NCMe
10
0
N
N
NCMe
Ar
Ar
Scheme 4.
trile solution of 1 (Fig. 2, top), two waves can be seen. At
ꢁ1.47 V vs. Cp2Fe0/+, a reversible reduction produces 1Åꢁ
via a ligand-based reduction [14] of the diimine moiety.
In addition, a chemically irreversible oxidation wave is seen
with a peak potential of ca. 0.55 V during the anodic-going
scan. These observations qualitatively resemble results pre-
viously obtained for 2, which underwent a ligand-centered
reduction at ꢁ1.36 V and an irreversible oxidation process
at 0.6–0.7 V vs. Cp2Fe0/+ [12]. In THF, the cyclic voltam-
mogram of 1 exhibited two reversible reduction waves at
ꢁ1.57 and ꢁ2.66 V vs. Cp2Fe0/+ respectively and an irre-
versible oxidation wave at ca. 0.50 V. When the cyclic vol-
tammetry was performed in the opposite direction (anodic
sweep first) in acetonitrile, a new reversible wave appeared
at ꢁ0.96 V, with a peak current ca. 50% of that for the ini-
tial oxidation of 1. This redox couple appears to arise from
a product that must be formed by reaction of 1Å+ that was
generated as a transient during the irreversible oxidation.
This reversible wave is attributed to the cationic Pt(IV) spe-
cies 3, which was voltammetrically observed and isolated
after an exhaustive electrolysis of 1 in acetonitrile (vide
infra). The assignment of this oxidation wave is further
supported by an earlier report on closely related complexes
[12]. The presence of 3 is readily seen as a persistent wave in
Fig. 1, bottom, which depicts a multi-scan cyclic voltam-
mogram in which each scan is immediately followed by a
new one. (The apparent reversibility of the oxidation wave
at +0.5 V in this voltammogram is an artifact caused by
fouling of the electrode surface.)
-10
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
10
0
-10
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
E
(V vs. Cp2Fe0/+
)
Fig. 2. Cyclic voltammograms for 1 with an initial cathodic-going
scan direction. Conditions: 1 mM solution in acetonitrile/0.1 M
Bu4NþPFꢁ6 at 20 °C (d = 0.6 mm Pt disc electrode, voltage sweep
rate m = 1 V/s). Top: Single-scan CV. Bottom: Continuous multi-scan
CV.

3228
B.J. Wik, M. Tilset / Journal of Organometallic Chemistry 692 (2007) 3223–3230
3.3. Exhaustive electrolysis of 1
of the Pt(II) cation 4 among the minor products. Methane
or ethane, which might result from oxidatively induced
PtAMe homolysis or concerted reductive eliminations,
respectively, were not detected by 1H NMR when the
chemical oxidation was performed in acetonitrile-d3. Con-
cerning the fate of the fourth methyl group in 1, which
must somehow be cleaved off when 3 is formed, it is impor-
tant to note that methylferrocene, (C5H4Me)(C5H5)Fe, is
formed in addition to ferrocene itself. The a and b protons
of the substituted Cp ring appear as an AA0BB0 pattern at
d 3.92–4.08, readily distinguished from the overlapping
signals of the Cp rings of ferrocene and methylferrocene
at d 4.13. Integration of the a and b proton signals indi-
cated a ca. 3:1 molar ratio of ferrocene to methylferrocene.
The overall stoichiometry of the reaction appeared to be
1:1 Cp2Fe+ to Pt in acetonitrile, in agreement with the
exhaustive electrolysis experiments described above. The
findings are qualitatively summarized (but not stoichiomet-
rically balanced) in Scheme 5, top.
To elucidate the nature of the product that was observed
by cyclic voltammetry after the irreversible oxidation of 1,
exhaustive electrochemical bulk oxidation was performed.
An acetonitrile/0.05 M Me4NþBF4ꢁ solution of 1 was oxi-
dized at a Pt mesh electrode at a working electrode poten-
tial of +0.7 V vs. Cp2Fe0/+. Three separate experiments
demonstrated the passage of 1.0 faraday/mol of charge,
i.e. the oxidation of 1 is a one-electron process. The solu-
tion from the anodic compartment of the electrolysis cell
was collected, the solvent was evaporated, the solid residue
extracted with dichloromethane (leaving the electrolyte
behind) and filtered, and the solid material was investigated
by 1H NMR spectroscopy. The 1H NMR spectrum
demonstrated that [ArAN@CHACH@NAAr]PtMe3-
(NCMe)+ (3) was the major product, and in fact the only
recognizable product of the electrochemical oxidation. This
discounts the possible formation of 4 under these
conditions.
When dichloromethane-d2 was used for a similar
oxidation reaction, resonances that might result from
3.4. Chemical oxidation of 1
[ArAN@CHACH@NAAr]PtMe3(L)+
and
[ArAN@
CHACH@NAAr]PtMe(L)+ (L = OTfꢁ or CD2Cl2) species
were seen in a rather complex spectrum with numerous
overlapping signals (the combined yield is impossible to
accurately measure but is estimated at ca. 50–60%). These
species differ from 3 and 4 only in the identity of the aux-
iliary ligand L. These assignments are supported by the
finding that addition of acetonitrile to the solution after
complete oxidation produced 3 and 4, readily identified
by NMR (see above). The oxidation in dichloromethane-
d2 also led to the production of substantial quantities of
Ferrocenium triflate Cp2Fe+OTfꢁ was selected as a
homogeneous oxidant despite the fact that the electron-
transfer from 1 to Cp2Fe+ is strongly unfavorable based
on the observed irreversible peak oxidation potential for
1: The unfavorable electron-transfer should eventually be
driven to completion by the rapid follow-up reaction that
consumes 1Å+. (A similar approach was used when the oxi-
dation of analogues of 2 was investigated [12]). Indeed,
when Cp2Fe+OTfꢁ was added to a solution of 1 in aceto-
nitrile-d3 in a sealed NMR, a slow change in color from
dark purple to brown was observed. The reaction resulted
1
ethane, qualitatively identified by H NMR and by mass
spectrometry analysis of the headspace above the solution
after opening of the NMR tube. The stoichiometry of the
reaction appeared to be ca. 1:1 Cp2Fe+ to Pt also in dichlo-
romethane; again, the ferrocene was accompanied by meth-
ylferrocene, but the latter was seen in smaller quantities
than was the case in acetonitrile. The findings are summa-
rized (but not stoichiometrically balanced) in Scheme 5,
bottom.
1
in a complex reaction mixture with a number of H NMR
signals that could be attributed to PtAMe protons. An
1
analysis of the H NMR spectrum from the oxidation of
1 revealed that 3-d3 (containing a deuterated acetonitrile
ligand) was the major (ca. 70% yield) PtAMe containing
product, and also the only identifiable PtAMe containing
product. Importantly, there were no detectable quantities
Ar
N
Ar
Me
Pt
Me
+
Pt
Me
Me
N
Me
Me
CD3CN
+ Cp2Fe
+ Cp2Fe+OTf–
ca. 3:1
+ (C5H5)(C5H4Me)Fe
N
N
Me
NCMe
Ar
Ar
1
3
Ar
N
Ar
Me
+
Pt
+
Me
Me
N
L
Ar
N
L = CD2Cl2 / OTf–
Pt
+
Me
Pt
N
N
Me
Me
Me
CD2Cl2
+ Cp2Fe+OTf–
L
Ar
Ar
N
Me
Ar
+ Cp2Fe
ca. 10:1
+ C2H6
1
+ (C5H5)(C5H4Me)Fe
Scheme 5.

B.J. Wik, M. Tilset / Journal of Organometallic Chemistry 692 (2007) 3223–3230
3229
Unfortunately, we have not succeeded with completely
accounting for the mass balance in these reactions, and this
will necessarily render any mechanistic discussion very
uncertain. However, some main features and inferences will
be presented in the following.
trile-d3. In dichloromethane-d2, ethane is observed, along
with the formation of 3 and 4. We surmise that PtAC
homolytic cleavage still occurs (evidenced by the appear-
ance of methylferrocene). However, there is now a possibil-
ity that direct ethane elimination occurs from 1Å+; the
resulting 2Å+ would then give rise to 3 and 4 by a mecha-
nism analogous to that described earlier [12] for the oxida-
tion of (diimine)PtMe2 complexes.
3.5. Mechanistic implications
In agreement with observations made on a related sys-
tem by Kaim and co-workers [14], we find that the Pt(IV)
complex [ArAN@CHACH@NAAr]PtMe4 is at least as
readily oxidized as the Pt(II) analogue [ArAN@
CHACH@NAAr]PtMe4, as inferred from the irreversible
cyclic voltammetry oxidation waves. As expressed by
Kaim, the strong donor methyl groups therefore offset
the effect of the higher formal oxidation state of Pt. The
previous study [14], which was supported by DFT calcula-
tions and electronic spectra, showed that the HOMO of the
Pt(IV) species is an antisymmetric PtACaxial r-orbital com-
bination mixed with some pp (Pt) contributions. A one-
electron oxidation of 1 therefore should tend to weaken
the axial PtAMe bonds and the incipient cation 1Å+ should
be activated with respect to PtAC bond cleavage, be it by
homolytic PtAC cleavage or concerted C-C bond coupling
reactions.
3.6. Concluding remarks
In this study, we have demonstrated that closely related
Pt(IV) and Pt(II) species undergo facile oxidation reactions
under quite similar reaction conditions to give similar – but
not completely identical – products and product distribu-
tions. These findings suggest that when redox processes
are involved in catalytic cycles that involve Pt(II) and
Pt(IV) species, it may be useful to keep in mind that oxida-
tion of Pt(IV) species should also be considered as a key to
understanding the overall process and the resulting product
distributions.
Acknowledgements
We gratefully acknowledge financial support from the
Norwegian Research Council (NFR) in the form of a sti-
pend to B.J.W. and generous contributions from the
NMR facilities at the Department of Chemistry, University
of Oslo.
An oxidatively induced reductive elimination of ethane
from 1Å+ would produce 2Å+ in addition to ethane, and
the Pt-containing products should then be the same as
those seen during the oxidation of 2 and analogues, i.e. a
1:1 mixture of 3 and 4 resulting from intermolecular methyl
group transfer. Since oxidation of 1 in acetonitrile does not
produce ethane, results in only 3 and not 4, and oxidation
with Cp2Fe+ produces significant amounts of methylferro-
cene (which was not observed when Pt(II) species were oxi-
dized [12]), it is apparent that the initial reaction cannot be
a simple reductive elimination. Instead, we propose that the
incipient 1Å+ with its weakened axial PtAMe bonds under-
goes homolytic cleavage of a PtAC bond. This leaves the
five-coordinate fragment ½ArAN@CH–CH@NAArꢀPtMe3þ
which, in acetonitrile solution, is efficiently trapped to pro-
duce 3 as the only observable Pt product. The fate of the
lost methyl group is, unfortunately, to a great extent
unknown. The appearance of methylferrocene does how-
ever suggest that at least in part, Cp2Fe+ is capable of act-
ing as a methyl radical acceptor; methyl transfer to one Cp
ring in Cp2Fe+ produces protonated methylferrocene as
the obvious precursor to methylferrocene. Thus far, the
resulting stoichiometry would be 2:1 Cp2Fe+ to Pt con-
trasting the observed 1:1 stoichiometry. The accompanying
proton that would be generated might be a source of meth-
ane by protonation of PtAMe containing species; however,
methane was never observed. Alternatively, the proton
might act as a one-electron oxidant towards 1 or Cp2Fe;
in this case, a 1:1 stoichiometry would result in accordance
with observations.
References
[1] J.L. Garnett, R.J. Hodges, J. Am. Chem. Soc. 89 (1967)
4546.
[2] R.J. Hodges, J.L. Garnett, J. Phys. Chem. 72 (1968) 1673.
[3] A.E. Shilov, G.B. Shul’pin, Chem. Rev. 97 (1997) 2879.
[4] N.F. Gol’dshleger, M.B. Tyabin, A.E. Shilov, A.A. Shteinman, Zh.
Fiz. Khim. 43 (1969) 2174.
[5] A.E. Shilov, G.B. Shul’pin, Activation and Catalytic Reactions of
Saturated Hydrocarbons in the Presence of Metal Complexes, Kluwer
Academic, Dordrecht, 2000.
[6] M. Lersch, M. Tilset, Chem. Rev. 105 (2005) 2471.
[7] U. Fekl, K.I. Goldberg, Adv. Inorg. Chem. 54 (2003) 259.
[8] S.S. Stahl, J.A. Labinger, J.E. Bercaw, Angew. Chem., Int. Ed. 37
(1998) 2181.
[9] H. Heiberg, L. Johansson, O. Gropen, O.B. Ryan, O. Swang, M.
Tilset, J. Am. Chem. Soc. 122 (2000) 10831.
[10] L. Johansson, M. Tilset, J.A. Labinger, J.E. Bercaw, J. Am. Chem.
Soc. 122 (2000) 10846.
[11] H.A. Zhong, J.A. Labinger, J.E. Bercaw, J. Am. Chem. Soc. 124
(2002) 1378.
[12] L. Johansson, O.B. Ryan, C. Rømming, M. Tilset, Organometallics
17 (1998) 3957.
[13] J.D. Scollard, M. Day, J.A. Labinger, J.E. Bercaw, Helv. Chim. Acta
84 (2001) 3247.
[14] W. Kaim, A. Klein, S. Hasenzahl, H. Stoll, S. Zalis, J. Fiedler,
Organometallics 17 (1998) 237.
[15] J.M. Kliegman, R.K. Barnes, J. Org. Chem. 35 (1970)
3140.
[16] M. Lashanizadehgan, M. Rashidi, J.E. Hux, R.J. Puddephatt,
S.S.M. Ling, J. Organomet. Chem. 269 (1984) 317.
[17] J.D. Scott, R.J. Puddephatt, Organometallics 2 (1983) 1643.
´
The oxidation of 1 with Cp2Fe+ in dichloromethane-d2
produced different products from those seen in acetoni-

3230
B.J. Wik, M. Tilset / Journal of Organometallic Chemistry 692 (2007) 3223–3230
[18] Z.W. Li, A. Yeh, H. Taube, Inorg. Chem. 33 (1994) 2874.
[19] G.S. Hill, G.P.A. Yap, R.J. Puddephatt, Organometallics 18 (1999)
[22] U. Fekl, W. Kaminsky, K.I. Goldberg, J. Am. Chem. Soc. 123 (2001)
6423.
1408.
[23] S. Reinartz, P.S. White, M. Brookhart, J.L. Templeton, J. Am.
Chem. Soc. 123 (2001) 6425.
[20] R.M. Gschwind, S. Schlecht, J. Chem. Soc., Dalton Trans. (1999)
1891.
[24] U. Fekl, K.I. Goldberg, J. Am. Chem. Soc. 124 (2002) 6804.
[25] U. Fekl, W. Kaminsky, K.I. Goldberg, J. Am. Chem. Soc. 125 (2003)
15286.
[21] R. van Asselt, E. Rijnberg, C.J. Elsevier, Organometallics 13 (1994)
706.