C-C coupling of aryl groups and allyl derivatives on Pt(II)-phenanthroline fragments: Crystal and molecular structure of the tbp [(η1, η2-2-allyl,5-methyl-phenyl)iodo(1,10-phenanthroline)platinum(II)] complex containing the N-N ligand in axial-equatorial coordination mode

Article abstract of DOI:10.1016/j.jorganchem.2004.12.010

The reactions between allyl compounds CH₂CHCH₂Fn bearing electron-withdrawing functional (Fn) groups and cationic {Pt(aryl)(1,10-phenanthroline)}⁺ fragments generated in situ are described. The aryl and platinum addition to the terminal and, respectively, internal unsaturated carbon is generally observed. The subsequent H-C _aryl bond activation, followed by HFn elimination, affords the ortho organic fragment (Pt)-aryl-CH₂CHCH₂ η¹, η²-chelate to the platinum. This process does not occur when the regiochemistry is of Markownikov type and the Pt-CH₂CH(aryl)CH ₂Fn fragment is formed. The described results are compared with those concerning the behaviour of unsubstituted α-olefins. The X-ray crystal structure of the title five-coordinate complex shows a distorted tbp arrangement of the ligands with the iodide in the equatorial plane and the unusual axial-equatorial coordination mode of the 1,10-phenanthroline.

Full text of DOI:10.1016/j.jorganchem.2004.12.010

Journal of Organometallic Chemistry 690 (2005) 2035–2043
www.elsevier.com/locate/jorganchem
C–C coupling of aryl groups and allyl derivatives on
Pt(II)-phenanthroline fragments: crystal and
molecular structure of the tbp
[(g1,g2-2-allyl,5-methyl-phenyl)iodo(1,10-phenanthroline)
platinum(II)] complex containing the N–N ligand
in axial-equatorial coordination mode
a
b,
a
Vincenzo De Felice , Augusto De Renzi *, Natascia Fraldi ,
b
Giuseppina Roviello , Angela Tuzi
b
a
`
Dipartimento STAT, Universita degli Studi del Molise, Via Mazzini, Isernia I 86170, Italy
b
`
Dipartimento di Chimica, Universita degli studi di Napoli ‘‘Federico II’’,
Complesso Universitario di Monte S. Angelo, via Cintia, Napoli I 80126, Italy
Received 30 November 2004; accepted 1 December 2004
Available online 22 January 2005
Abstract
The reactions between allyl compounds CH2@CHCH2Fn bearing electron-withdrawing functional (Fn) groups and cationic
{Pt(aryl)(1,10-phenanthroline)}+ fragments generated in situ are described. The aryl and platinum addition to the terminal and,
respectively, internal unsaturated carbon is generally observed. The subsequent H–Caryl bond activation, followed by HFn elimina-
tion, affords the ortho organic fragment (Pt)–aryl-CH2CH@CH2 g1,g2-chelate to the platinum. This process does not occur when
the regiochemistry is of Markownikov type and the Pt–CH2CH(aryl)CH2Fn fragment is formed. The described results are compared
with those concerning the behaviour of unsubstituted a-olefins. The X-ray crystal structure of the title five-coordinate complex
shows a distorted tbp arrangement of the ligands with the iodide in the equatorial plane and the unusual axial-equatorial coordi-
nation mode of the 1,10-phenanthroline.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: Platinum; Migratory insertion; C–H activation; Five-coordination
1. Introduction
replicates extensively to afford the polymeric chain.
When the insertion is followed by hydrogen shift and
reductive elimination, an olefin moiety higher than the
starting one is the final product. The last reaction path-
way is typically observed, mainly in late transition ions
chemistry, when aryl–metal bonds are involved. The
Heck processes are the most outstanding examples of
this behaviour [2].
The insertion of a C@C bond in a nearby metal-car-
bon r-bond on a metal centre is a pivotal step of most
important catalytic cycles prompted by transition
elements [1]. In the olefin polymerisation catalysis
the r-bound group is of alkylic type and the insertion
*
In this paper attention is focused on the insertion
involving aryl groups and allyl derivatives. As for
Corresponding author. Tel.: +39081674454; fax: +39081674090.
E-mail address: derenzi@chemistry.unina.it (A.D. Renzi).
0022-328X/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.12.010

2036
V.D. Felice et al. / Journal of Organometallic Chemistry 690 (2005) 2035–2043
previous studies in this field from this laboratory [3] we
recall that investigation of the links between the elec-
tronic and steric features of the two interacting moieties
and the regiochemistry and stereochemistry of the reac-
tion has been grounded on stoichiometric processes,
more suited to the characterisation of possible interme-
diates. Thus, the choice of the metal and of the ancillary
ligands was tailored to obtain moderate reactivity of the
two interacting organic moieties while avoiding inert-
ness. In this work we have attempted to examine the
insertion behaviour in {Pt(aryl)(N–N)}+ cations (N–
N = chelating nitrogen ligand) analogous to species pre-
viously used to assess the reactivity of a-olefins. For this
type of complex substrate clean insertion of R–
CH@CH2 (R = H, hydrocarbyl) in Pt–aryl bond to give
the sequence Pt–CH2–CHR-aryl was previously [3c] ob-
served, while if an alkyl group was present in place of
the aryl no insertion ensued [4] (Scheme 1).
The choice of 1,10-phenanthroline (phen), a rigid che-
late, as ancillary ligand was dictated by the need of
favouring the contemporary presence in adjacent sites
of the two organic moieties. Phenanthroline, with low
trans ability donors, is in particular expected to favour
the bonding of poorly coordinating allyl derivatives.
As for the choice of dealing with allylic substrates it is
to note: (i) the presence of the oxygenated substituents
should reasonably affect the regiochemistry and the ste-
reochemistry of the insertion and the observed influence
should add information on the reaction features; (ii) the
attainment of functionalized products is in principle
more worth of attention if the results on our system
can add ground for catalytic processes.
Solvents and reagents were of AnalaR grade and used
without further purification.
2.1. General reaction procedures
To a stirred suspension of 0.2 mmol of the neutral
[Pt(4-R0-C6H4)I(phen)] (1a–d) complex in 1.5 mL of
CD3NO2, an equimolar amount of AgBF4 dissolved in
0.5 mL of the same solvent was added at room temper-
ature. Dissolution of the platinum complex immediately
ensued, while slowly precipitating silver iodide. An equi-
molar amount of the allyl derivative dissolved in 0.5 mL
of CD3NO2 was then added and, after 5 min stirring, the
mixture was quickly filtered through a thin layer of Cel-
ite to remove silver iodide. Alternatively, AgBF4 was
added as CD3NO2 (0.5 mL)/THF (0.2 mL) solution to
the platinum complex suspension. Silver iodide immedi-
ately precipitated and, after filtration on Celite, the allyl
substrate was added as above. The reaction mixture was
then monitored by 1H NMR until signals of the reacting
allyl substrate disappeared. After filtration on Celite, the
final solution was divided in two portions (i) and (ii).
(i) The solution was evaporated until an oily residue
was obtained, that was dissolved in 0.5 mL of CDCl3.
Gaseous dry HCl was added and the mixture was stirred
for 2 h at room temperature. An orange microcrystalline
solid separated, that was identified as the [PtCl2(phen)]
complex. After filtration on Florisil, the resulting clear
1
solution was examined by H NMR. Identification of
the organic compounds was achieved by comparison
of the recorded spectra with those of authentic samples
and/or literature data.
(ii) An excess of solid NaI was added to the solution,
which was stirred for 1 h at room temperature. A yellow
solid formed, that was separated and re-crystallized
from CH2Cl2/n-hexane to give the type 300 complex.
The mother liquor was concentrated to a small volume
and diethyl ether carefully added to crystallize the type
30 complex as orange crystals. It is to note that, accord-
ing the used allyl substrate (see Section 3) both com-
plexes can be obtained or only one of them (30 or 300).
Experimental details and analytical data are reported
as supplementary material. Relevant 1H[13C] NMR data
for the 30 and 300 complexes are listed in Tables 1 and 2.
2. Experimental
1H[13C] NMR spectra were recorded on Varian XL-
200 or Varian Gemini-300 spectrometers. 1H[13C] chem-
ical shifts are reported in d (ppm) relative to the solvent
(CHCl3, d = 7.26 [77.0, 13CDCl3]; CHD2NO2, d = 4.33).
The neutral platinum(II) complexes [Pt(4-R0-C6H4)I-
(phen)] (1a–d) used as precursors were obtained
according to described procedures [3c] and their charac-
terization data are reported as supplementary material.
+
N
NCMe
(HC) = aryl
(a)
(b)
Pt
Pt
+
N
CH2CHR-(HC)
N
(HC)
CH2=CHR
Pt
+
N
NCMe
N
(HC)
(HC) = alkyl
CH2
+ MeCN
N
CHR
Scheme 1. Olefin addition to cationic Pt(II)-hydrocarbyl complexes.

V.D. Felice et al. / Journal of Organometallic Chemistry 690 (2005) 2035–2043
2037
Table 1
Selected 1H[13C] NMR data for 30complexesa [Pt{CH2CH(CH2OH)C6H4-4-R0}I(phen)]
Complex
4-R0-C6H4
b,b0
Pt–C3H2–
–C2H(C1H2OH)–
CH
H2,H9(phen)b
c,c0
R0
CH2
30a
7.0–7.4 (m)c
2.75 (t,d);
2.15 (dd,85)
3.20 (t,72);
2.18 (dd,96)
[3.4]
2.95 (m)
3.80 (m)
9.88 (d,20);
9.15 (d,54)
10.18 (d,20);
9.18 (d,54)
30b
7.30 (d)
7.38 (d)
7.10 (d)
6.85 (d)
2.35 (s)
[21.1]
3.00 (m)
[51.3]
3.90 (m)
[68.6]
[153.4, 148.8]
10.25 (d, 22);
9.20 (d, 61)
[153.5,148.7]
3.25 (t,61);
2.25 (dd,80)
[5.2]
30c
3.80 (s)
[55.1]
3.00 (m)
[51.5]
3.90 (m)
[68.8]
a
Spectra recorded in CDCl3 (reference d 7.26, CHCl3 [77.0, 13CDCl3]). The coupling constants with 195Pt (Hz) are reported in parentheses.
Abbreviations: s (singlet), d (doublet), dd (double doublet), t (triplet), m (multiplet).
b
The resonance with the higher value of the 195Pt coupling constant refers to the proton on the carbon adjacent to the nitrogen trans to the iodide.
Phenyl protons.
Coupling constant with 195Pt not evaluable.
c
d
Table 2
Selected 1H[13C] NMR data for 300 complexesa [Pt(g1,g2-R0–C6H3CHYCH@CH2)I(phen)] (Y = H, Me, CH2OH)
Complex
R0–C6H3–
Ar–C3HH– or ArC3H(Me, CH2OH)
–C2H@C1H2
–C2H@
H2,H9(phen)b
b0,c0
c
R0
@C1H2
300a
7.00 (m)c
8.20 (d,d)
3.30 (dd), 2.55 (d)
3.62 (m,93)
3.18 (d,d)
3.00 (d,53)
3.21 (d,d)
3.07 (d,69)
[60.5]
3.18 (d,d)
3.00 (d,70)
[62.3]
3.20 (d,d)
2.98 (d,67)
3.18 (d,d)
2.90 (d,d)
3.50 (d,d)
3.20 (d,d)
9.10 (d,22)
8.95 (d,d)
9.18 (d,26)
8.98 (d,d)
300b
6.95 (d)
6.85 (d)
8.25 (s,26)
2.43 (s)
[21.2]
3.36 (dd), 2.56 (d)
[40.1]
3.65 (m, 92)
[74]
[149.4.147.7]
9.15 (d,29)
8.92 (d,15)
[150.0,148.5]
8.95 (br)
e
–
300c
300d
6.90 (d)
6.58 (d)
7.22 (d)
7.10 (d)
3.85 (s)
[55.1]
3.30 (dd), 2.50 (d)
[42.3]
3.30 (dd), 2.58 (d)
3.62 (m,d)
[74]
3.61 (m, 91)
8.65 (s,17)
9.10 (d,28)
8.90 (d,d)
9.18 (d,32)
8.85 (d,d)
3.65 (m,d)
f
300b(Me)
300b(CH2OH)
6.85(m)
6.98 (d)
6.85 (d)
8.20 (s,32)
e
–
2.38 (s)
2.38 (s)
3.30 (m); 1.00 (Me,d)
2.80(m);
4.20,3.70 (CH2OH,m)
e
–
a
Spectra recorded in CDCl3 (reference d 7.26, CHCl3 [77.0, 13CDCl3). The coupling constants with 195Pt (Hz) are reported in parentheses.
Abbreviations: s (singlet), d (doublet), dd (double doublet), m (multiplet), br (broad).
b
The resonance with the higher value of the 195Pt coupling constant refers to the proton on the carbon adjacent to the equatorial nitrogen.
c
b0, c0 and d phenyl protons.
d
Coupling constant with 195Pt not evaluable.
Obscured by other signals.
Figures refer to the more abundant diastereoisomer.
e
f
2.2. Structure determination
The yellow crystal selected was prismatic, 0.03 ·
0.15 · 0.18 mm3. 5498 reflections were collected, 5067
independent (Rint = 0.0907, hmax = 28.02ꢁ). Empirical
absorption correction was applied using DIFABS program
[5]. The structure was solved by direct methods (SHELXS
program of SHELX 97 package [6]), completed by differ-
ence Fourier methods and refined by the full matrix least
squares method (SHELXL program of the same package).
Refinement was on F2 against all independent measured
reflections. All non-H atoms were anisotropic. H atoms
were placed in calculated positions except for olefinic H
atoms that were found from a difference Fourier map.
All H atoms were refined by the riding model. 235
Parameters were refined. Final results: R1 = 0.056,
wR2 = 0.1177 on F (based on 2384 reflections with
I > 2r(I)); R1 = 0.1612, wR2 = 0.1524 on F2 (all data).
Single crystals of [Pt(g1,g2-2-allyl,5-methyl-phenyl)-
I(phen)] (3b00), suitable for X-ray analysis were obtained
by slow crystallization from CHCl3 solution at low tem-
perature (277 K).
Accurate cell parameters were obtained through a
least-squares fit to the h angles of 24 strong reflections
in the range 9.88ꢁ 6 h 6 10.79ꢁ on an Enraf–Nonius
MACH 3 diffractometer, using graphite monochroma-
˚
tised Mo Ka radiation (k = 0.71069 A). Data collection
was performed on the same apparatus in the x scan
mode at 293 K. Crystals of complex 3b00, C22H19N2PtI,
˚
are orthorhombic, Pbca space group, a = 8.924(4) A,
3
b = 32.92(5) A, c = 14.303(5) A, V = 4202(7) A ,
˚
˚
˚
M = 633.40, Z = 8, Dc = 2.002 g/cm3, l = 8.155 mmꢀ1
.

2038
V.D. Felice et al. / Journal of Organometallic Chemistry 690 (2005) 2035–2043
Largest peak and hole in the last Fourier difference were
1.150 and ꢀ0.967 (e Aꢀ3), near heavy atoms.
homogeneity of the system and influencing its composi-
tion. For this reason, in case the reaction mixture had to
be NMR monitored as the deuterated solvent was used
nitromethane containing a 10% by volume of tetra-
hydrofurane, which speeds up AgI precipitation without
hindering appreciably the coordination of the allyl
substrates.
˚
3. Results and discussion
3.1. Synthesis
The allyl derivatives and the Pt(II) complexes used as
precursors of the reactive species are reported in
Scheme 2.
3.2. Reactions involving CH2@CH–CH2OH
The four complexes 1a–d were reacted with allyl alco-
hol (equimolar amount) in nitromethane in presence of
AgBF4. In all cases the comprehensive reaction path
could be depicted by Scheme 3, but for the lack of the
type 30d final product. The reactions of 1a–c were com-
plete in ca. 2 h at room temperature, while completion in
case of 1d required ca. 12 h.
In the reaction with allyl alcohol all the four 1a–d
complexes were employed. The attainment of the cat-
ionic species actually involved in the insertion reaction
can be effected in situ by treatment in nitromethane
solution with silver ions in presence of the olefin. How-
ever, no isolation and thorough characterization of the
reactive cation could be effected in this case. On the
other hand, the uptake of the poorly coordinating
CH2@CHCH2OH by analogous methyl-Pt(II) com-
pounds under these conditions is well established [4]
and affords the corresponding [PtMe(CH2@CH-
CH2OH)(N–N)]+ cationic complexes, which do not un-
dergo insertion.
Alternatively the iodide could be previously ex-
changed with a ‘‘labile’’ ligand such as acetonitrile to
form a cation. This cationic complex can be isolated,
characterized, and used in exact stoichiometric amount
for reaction with the olefin. However, olefins bearing
electron-withdrawing groups are not able to substitute
even the ‘‘labile’’ acetonitrile ligand, owing to the drain-
ing of electron density from the C@C bond.
[Notations for compounds recall species 1 as for the
letter. The apex denotes the type of proposed cationic
intermediate and the derived final product. In case of
type 200 (and 300) complexes a subscript (in parentheses)
identifies the groups on C3, if different from hydrogen].
Monitoring the reaction course by 1H NMR, the for-
mation of two platinum complexes containing a different
organic moiety bound to the metal was observed (20 and
200). Cations 20 and 200 were not isolated but their pres-
ence could be reasonably inferred on the ground of the
1H NMR spectra of the reaction mixture. For instance,
20b: 4.0 (m, CH2OH), 3.0 (m, CHAr), 2.70 (app t, Pt–
2
CHH, JPtH not evaluable) and 2.50 d (dd, Pt–CH H,
2JPtH not evaluable); 200b: 6.0 d (m, @CH, JPtH = 78
2
2
Hz), 4.62 d (d, @CHH, JPtH = 60 Hz), 4.45 (d, @CH
H, JPtH not evaluable), 4.08 (dd, Ar–CHH) and 3.20
2
The in situ procedure involves a problem. In fact, re-
moval of a iodide with Ag+ from a Pt(II) iodide complex
is not an immediate and simple reaction in absence of
donors apt to avoid the further coordinative unsatura-
tion of the d8 ion [7]. Precipitation of silver iodide is
fairly slow, thus conflicting with the requirement of
d (d, Ar–CHH). Furthermore, the HCl treatment of
the reaction mixture allowed the recovery of the alcohols
4-R0–C6H4–CHMeCH2OH (from 20a–c) and the olefins
4-R0–C6H4–CH2CH@CH2 (from 200a–d).
Type 20 compounds reasonably derive from an inser-
tion step involving the terminal carbon C3 of the allylic
system. We guess that they are moderately stabilized by
Ptꢁ ꢁ ꢁOH interaction and by the attainment of a five-
member ring. Formation of a type 200 product should
be attributed to an initial insertion step involving bond-
ing of the central carbon of the allylic system to the me-
tal, followed by a rearrangement based on the H–Caryl
activation and its oxidative addition to Pt. A similar
process has been previously observed [3] in similar sys-
tems for propene insertion, which however is completed
by the reductive elimination involving the Pt–Caryl bond,
while the final step to give the organic moiety in 200 is
water elimination to afford a coordinate unsaturated
bond. This reaction sequence is shown in Scheme 4.
After treatment of the reaction mixture with sodium
iodide, the corresponding neutral complexes 30 and 300
were obtained, separated as crystalline solids and fully
characterized by usual procedures.
5
6
C3H2=C2HC1H2Fn
7
4
2
Fn
8
3
OH
N
I
N
9
CH2OH
β
Pt
OEt, OPh
α
γ
OCOMe, OCOPh, OCOOMe
+
δ
Cl, NMe3
β'
R'
CN, SMe
γ'
CH(CH3)=CHCH2OH
R'
H
1a
CH3 1b
OCH3 1c
CF3
1d
O
Scheme 2. Platinum(II) complexes and allyl substrates.

V.D. Felice et al. / Journal of Organometallic Chemistry 690 (2005) 2035–2043
2039
N
I
Pt
N
(1)
R'
AgBF4
C3H2=C2HC1H2OH
R'
+
+
HOC1H2
C2H
N
N
Pt
R'
Pt
C3H2
C2H
C3H2
N
N
H2C1
(2')
(2'')
NaI
R'
R'
C3H2
C2H
I
N
I
Pt
Pt
N
C3H2 C2H
N
C1H2
C1H2OH
N
(3')
(3'')
Scheme 3. Reactions between 1a–d complexes and allyl alcohol.
Noteworthy features of 300 are the coordination
geometry and saturation. While the reaction mixture
spectra (example above) point to a square planar 200 spe-
cies, both the spectral evidence and one crystallographic
analysis prove that the corresponding neutral com-
pounds are five-coordinate. As for 1H NMR spectra
we can compare with the above reported signals the cor-
responding ones observed in the 300b spectrum: 3.65 d
nate structure [8], which was confirmed by X-ray
analysis.
The spectral patterns of the type 300 compounds with
different substituents R0 also disclose structural features
in agreement with the pathways in Scheme 4. Making
reference to 300b, it is clearly seen a signal pertaining to
0
the Hc (or Hc ) of the aryl bound to Pt as singlet at
3
0
0
8.25 d, JPtH = 26 Hz. Signals due to Hb and Hc (or
Hb and Hc) are two coupled (3JHH = 6 Hz) doublets cen-
tred at 6.85 and 6.95 d. Thus, the R0 group on the phe-
nyl, which is initially para to the metal, is para to the C3
2
2
(m, @CH, JPtH = 92 Hz), 3.20 d (d, @CHH, JPtH not
2
evaluable) and 3.03 d (d, @CHH, JPtH = 69Hz.). The
dramatic high-field shift clearly points to a five-coordi-
R'
R'
R'
H
+ H20
C3H2
Pt
C3H2
Pt
Pt
C3H2
C2H
H2C1
C2H
H
C2H
C1H2
C1H2
O
O
H
H
Scheme 4. Pathway to 200 reaction products.

2040
V.D. Felice et al. / Journal of Organometallic Chemistry 690 (2005) 2035–2043
organic fragment in the product. This is true in all cases.
Furthermore, according to the proposed oxidative addi-
tion step the signal of the b (or b0) aryl hydrogen is dis-
tion of crotyl alcohol with 1b required 12 h at room
temperature for completion and the reaction mixture
(only the type 200 ionic compound being observed)
afforded by iodide addition the type 300 derivative as
the final product. Treatment of the reaction mixture
with HCl gave 4-Me–C6H4–CHMeCH@CH2, as ex-
pected. Thus, the position of the CHMe group gives
further evidence that the regiochemistry of the inser-
tion step is such that the terminal C3 of the allylic sys-
tem binds to the aryl, while the internal C2 is linked to
the metal. On the other hand, the terminal final posi-
tion of the C@C bond is in keeping with a rearrange-
ment of the type depicted in Scheme 4. It is to note
that two diastereoisomeric compounds, of which one
is quite prevailing (over 90%), are observed in the
crude reaction product. They must differ for the abso-
lute configuration at the stereogenic centre C3 (com-
pare with the previous discussion). On the other
hand, as the C2 and C3 carbon atoms of the crotyl
alcohol interact with the platinum and the aryl group
via a concerted cis-addition mechanism, their absolute
configurations cannot independently vary. Therefore,
the observed diastereoisomeric ratio in the crude
200b(Me) (and 300b(Me)) product is related to the trans–
cis composition of the commercial crotyl alcohol.
Reaction of 3-buten-1-ol with 1b yielded a mixture of
two type 20 species which on treatment with HCl affor-
ded both the alcohols 4-Me–C6H4–CHMeCH2CH2OH
and 4-Me–C6H4–(CH2)3OH in ca. 2:1 ratio.
Results on these substrates converge to sustain the
attribution of the structure and of the role of cations
in Scheme 3. The CH2OH group has to be considered
an electron withdrawing substituent in this context, in
contrast with the methyl in propene. However, the ob-
served loss of regiospecifity in the insertion of allyl
alcohol (aryl addition on both the unsaturated carbon
is represented) respect to that observed with propene
(sequence Pt–CH2CHMe-aryl only) fits well with the
moderate inductive effect of CH2OH. Moreover, it is
easily understood that by balancing the electronic ef-
fect of the hydroxymethyl by introducing a methyl
group (crotyl alcohol) or by reducing that effect by
distancing the group (3-buten-1-ol) the regiospecifity
is recovered. Also, the effect of the ability of the
OH group to interact with the metal finds a rationale
on the ground of the proposed H–Caryl activation. In
fact, the Pt–OH interaction can afford ‘‘pushing’’ of
the phenyl ring far from the metal, thus contrasting
the activation of the H–Carylbond. Therefore, if the
length of the Pt-bound hydrocarbon chain allows that
the OH group can be involved in a fairly stable five-
or six-member ring, the reaction stops with attainment
of a type 20 product. No pushing away of the phenyl
is expected to be involved in shorter Pt-bound hydro-
carbon chains, which can evolve after rearrangement
to the 200 type product.
0
appeared and the Hc (or Hc ) is coupled to Pt.
The structure of 300a–d complexes deserves a few com-
ments. As these five-coordinate tbp compounds present
two stereogenic centres (the coordinated prochiral dou-
ble bond and the metal atom), four stereoisomers (actu-
ally two enantiomeric pairs) can be in principle expected.
In all cases only one enantiomeric couple was observed in
solution, thus indicating that the absolute configurations
of the two chiral centres must be strictly correlated. A
simple molecular model, built on the basis of the actual
configuration correlation of the two centres, disclosed
by the X-ray structure of 300b (see Section 3.6), has in fact
shown that too short contacts arise between the plati-
num-bound aryl and the phenanthroline heteroaromatic
ring occupying the equatorial site, on changing the abso-
lute configuration of only one centre. As for the 1,10-
phenanthroline coordination mode, we note that a large
number of five-coordinate [Pt(N–N)(olefin)ZZ0] com-
plexes (Z,Z0 = halogen and/or hydrocarbyl ligands) are
known, which are all characterized by the presence of
an N–N ligand with an in-plane sterical hindrance
(e.g., 2,9-dimethyl-1,10-phenanthroline) prompting to
disfavour the usual square-planar four-coordination.
They generally adopt a typical tbp geometry with the
nitrogen chelate lying in the equatorial plane [8]. A few
cases of axial-equatorial coordination mode have been
reported, although tendency to isomerise to the typical
cited structure via a Berry pseudo-rotation mechanism
was exhibited by those complexes in solution [9]. On
the other hand, the obtained species 300 do not display
the typical ligand arrangement above described and
isomerization was not observed in solution. By inspec-
tion of the 300b molecular model, we note that the ob-
served stereochemistry corresponds to that of the
possible intermediate in the associative ligand exchange
in which the entering iodide is substituting a leaving
group belonging to a chelate ligand (one phenanthroline
nitrogen or the double bond). Conceivably, this process
is energetically disfavoured. In addition, the cited isom-
erization via a Berry pseudo-rotation should involve a
transient different tbp environment with a phenanthro-
line nitrogen and the double bond in axial position. As
the absolute configuration of the coordinate double bond
is fixed, there are sterical restraints to the isomerization.
Therefore, once the iodide has coordinated, the attained
ligand arrangement is reasonably the most favoured and
thus it stands stable.
3.3. Reactions involving MeCH@CH–CH2OH and
CH2@CH(CH2)2OH
Results concerning the two title alcohols added fur-
ther information on the insertion regiochemistry. Reac-

V.D. Felice et al. / Journal of Organometallic Chemistry 690 (2005) 2035–2043
2041
3.4. Reactions involving allyl-ethers and esters
Reaction of 1b with two allyl-ethers and three allyl-
esters (see Scheme 2) was relatively fast (1.5–2 h) at
room temperature and afforded in quantitative yield
the same type 200b cation and thereafter type 300b neutral
five-coordinate complex as obtained in the reaction of
1b with allyl alcohol (Scheme 3). In analogy to that case,
water elimination is substituted by alcohol (from allyl-
ethers) or acid (from allyl-esters) release. Using the
reacting substrate allyl-methyl carbonate, methanol
(deriving from HOCOOMe decomposition) is detected
in solution.
The reaction of 1b and the cyclic allyl ether 2,5-
dihydrofurane presents a similar pathway and involves
rupture of the ether ring (Scheme 5).
On comparing the product with that formed from
crotyl alcohol it can be observed that the only difference
is the presence of methyl instead of CH2OH in that com-
pound. It is to note that in contrast with the crotyl deriv-
ative, only one diastereoisomer is observed for the
dihydrofurane product. This is consistent with the un-
ique cis configuration of the double bond in the starting
unsaturated cyclic ether and the cis addition mechanism
(see discussion in Section 3.3).
Fig. 1. Molecular structure of the complex [(g1,g2-2-allyl,5-methyl-
phenyl)iodo(1,10-phenanthroline)platinum(II)] (300b).
3.5. Reactions involving other allyl substrates
Pt adopts a distorted bipyramidal trigonal coordination,
with phen ligand occupying one axial and one equatorial
position, in an unusual and only one time found posi-
tion [7]. Moreover two of the other coordination posi-
tions (C9 and the double bond C1@C2) belong to the
same ligand and, as a consequence, some peculiar distor-
tions are expected. The axial positions are occupied by
N2 of phenanthroline and C9 of phenyl ligand. The an-
gle N2–Pt–C9 = 172.5(4)ꢁ shows a deformation that
could be ascribed to the steric hindrance of phenyl moi-
ety with phenanthroline ligand and to the axial-equato-
rial coordination of the phenyl moiety, this way of
coordination being dictated by the need of forming a
C9–Pt–(C1@C2) angle near to 90ꢁ. The equatorial plane
is defined by I, N1 atom of phenanthroline and the dou-
ble bond C1@C2. The major deformation in this plane
Complex 1b has been reacted with two allyl com-
pounds CH2@CH–CH2Fn where Fn is again an elec-
tron-withdrawing but not an oxygenated group, i.e.,
Fn = Cl or NMeþ3 . No variation is observed in the reac-
tion, 3b00 compound being finally isolated.
It is on the other hand that if the group Fn contains a
good soft donor, the allylic substrates coordinates to the
metal through Fn and no activation of the double bond
is obtained. This is in fact the behaviour of
CH2@CHCH2CN and of CH2@CHCH2SMe.
3.6. X-Ray structure of complex 300b
Molecular structure is reported in Fig. 1, relevant
bond lengths and bond angles are reported in Table 3.
Me
Me
H
Pt
Pt
CH CH2OH
CH
CH
CH2
CH
H2C
O
CH2
Scheme 5. Pathway to 200b(CH2OH) (and 300b(CH2OH)) reaction product for 2,5-dihydrofurane.

2042
V.D. Felice et al. / Journal of Organometallic Chemistry 690 (2005) 2035–2043
Table 3
ert Pt(alkyl)(olefin) complexes in which
energetically demanding electron transfer is instead rea-
sonably required.
a more
˚
Selected bond lengths (A) and bond angles (ꢁ) with e.s.d.’s in
parentheses
Pt–I
2.840(1)
2.14(1)
2.14(1)
2.01(1)
2.07(1)
2.10(1)
1.40(2)
1.50(2)
N(1)–Pt–N(2)
C(1)–Pt–C(9)
C(2)–Pt–C(9)
N(2)–Pt–C(1)
N(2)–Pt–C(2)
I–Pt–N(1)
77.9(4)
84.3(5)
82.5(5)
96.3(5)
93.5(4)
86.3(3)
172.5(4)
As far as the regiochemistry of the insertion is con-
cerned, we note that, in contrast with a-olefins insertion,
for the alkenes in this work the electronic and steric fea-
tures play opposite roles. It appears that the observed
regiochemistry is mainly dictated by the electronic prop-
erties of the substituent on the double bond favouring
bonding of Pt to the internal carbon. This is in good
agreement with an intramolecular nucleophilic addition
of an aryl carbanion on the coordinated double bond.
The only case of regiospecificity loss is related to the
presence of a substituent group of moderate inductive
effect as CH2OH. Actually, both the unsaturated car-
bons of the allyl alcohol were involved in the aryl migra-
tion at nearly equal amounts. The spacing out of this
group, i.e., the use of 3-butenol, restores, however not
completely, the regiochemistry corresponding to an
unsubstituted alkene.
A significant aspect of our studies concerns the eluci-
dation of the fate of the metal r-bound primary
insertion product. No organic compound, free or me-
tal-coordinated, deriving from a b-elimination process,
which is usual in the Pd(II) chemistry, was ever ob-
served. On the other hand, we recall that, when unsub-
stituted alkenes were used in the reaction with
[Pt(aryl)(N–N)(MeCN)]+ compounds [3c], the expected
r aryl-alkyl-Pt(II) cationic product was obtained and
its fully characterisation was achieved after conversion
to a neutral halo-complex (see Scheme 1(a)). A compa-
rable compound was observed in this work only in pecu-
liar conditions (see below). Allowing that in the present
study the reactions must be performed by generating in
situ the active {Pt-aryl(phen)}+ species, the comparison
has to be made with the previously obtained results in
the same experimental conditions. In all cases, the lack
of a fast pushing away of the aromatic ring far from
the platinum favours H–Carylbond activation. Its oxida-
tive addition to the metal centre is then followed by a
reductive elimination according two different paths
depending on the electronic properties of the starting al-
kene and depicted for clarity in Scheme 6.
Pt–N(1)
Pt–N(2)
Pt–C(9)
Pt–C(1)
Pt–C(2)
C(1)–C(2)
C(2)–C(3)
C(9)–Pt–N(2)
involves N1 atom as a consequence of the axial-equato-
rial coordination of phen (N1–Pt–N2 = 77.9(4)ꢁ).
The phenanthroline ligand is planar within 0.07 A
and its mean plane is coincident with the plane defined
˚
˚
by N1, Pt, N2 within 0.08 A, while a tilting is observed
with the mean plane of phenyl moiety (C8–C9–Pt–
N1 = ꢀ69(1)ꢁ), thus avoiding steric interactions between
them. The same steric interactions, mainly due to the
intramolecular contact between C8 and C11, prevent
the formation of the possible diastereoisomer resulting
from the coordination of the other enantioface of the
olefin.
Pt–N axial and equatorial distances (Pt–N1 = 2.14(1)
˚
A, Pt–N2 = 2.14(1) A) show no effect due to the different
˚
nature of their position. Pt–C9, Pt–I and C1@C2 bond
˚
lengths (2.01(1), 2.840(1) and 1.40(2) A, respectively)
are in the normal range for analogous coordinatively
saturated Pt(II) compounds [8].
4. Conclusions
The above described results, when compared with
those reported in case of unsubstituted alkenes, allow
enlightening some unequivocal features concerning the
M-hydrocarbyl/olefin insertion migratory process, at
least in Pt(II) chemistry. The required C@C bond activa-
tion via coordination to the metal can be easily attained
through a ligand substitution reaction only in the case of
simple alkenes. The presence of a vicinal electron-with-
drawing substituent reduces the C@C bond r-donor
properties at such an extent to disfavour the ligand ex-
change. Nevertheless, coordination can be achieved if
a fragment {Pt-hydrocarbyl(N–N)}+ is generated in ab-
sence of valuable coordinating species different from the
alkene itself. Moreover, the presence of an ancillary
bidentate ligand ensures the proper cis stereochemistry
to the subsequent insertion step. The facile reactivity ob-
served in case of hydrocarbyl = aryl can find a rationale
taking into account a possible overlap between a filled p
orbital of the aromatic ring with the empty p* orbital of
the double bond, that can occur when the alkene as-
sumes the ‘‘in plane’’ stereochemistry in the transition
state. This interaction should lower the activation en-
ergy for the insertion process with the respect to the in-
It is to note that while in the transient intermediate
{A} the fission of the Pt–CH2 bond is the only possible,
the cleavage of the CH2–heteroatom bond is quite pre-
ferred in {B}. The formation of an organic moiety
g1,g2-chelate to the metal atom could be the driving
force for this latter process. It is to strongly emphasize
that the above depicted pathways can only occur in ab-
sence of whatsoever species able to fill the coordinative
unsaturation at the metal atom resulting from the inser-
tion reaction. In fact, the primary fragment {Pt–CH2–
CHAr–CH2OH(phen)}+, deriving from the insertion of
the allyl alcohol with a regiochemistry different from
that expected by inductive effect, finds its stabilisation

V.D. Felice et al. / Journal of Organometallic Chemistry 690 (2005) 2035–2043
2043
R'
A
R'
+
H
N
N
(i) R = H, hydrocarbyl
Pt
Pt
CHR
R'
MeCN
CHRCH2H
NCMe
C
N
N
H2
N
Ar
X
Ag+, CH2=CHR
Pt
N
B
R'
+
H
N
N
(ii), R = CH2Fn
Pt
CH2
CH2
CH
H2C
CHR
N
N
+ HFn
Scheme 6. Olefin depending reaction paths.
through the formation of a five-member ring with a
Ptꢁ ꢁ ꢁOH interaction. The same is true in the case of
3-butenol, for which the two variations of the regio-
chemistry of the insertion afford respectively a five or
a six-member ring.
CB2 1EZ [Fax +44(1223)336-033] or e-mail: depos-
it@ccdc.cam.ac.uk or http://www.ccdc.cam.ac.uk. Sup-
plementary data associated with this article can be
found, in the online version at doi:10.1016/j.jorganchem.
2004.12.010.
As final remark we wish observe that the above reac-
tion sequence could be of synthetic usefulness allowing a
selective and easy removal of the C3 fragment from allyl-
ethers or -esters. As matter of fact, in an exploratory test
a-phenylethylalcohol was quantitatively released by
reaction of the corresponding allylether with the active
{Pt–C6H5(phen)}+ species [10].
References
[1] E.W. Abel, F.G.A. Stone, G. Wilkinson (Eds.), Comprehensive
Organometallic Chemistry II’’, vol. 12, 1995.
[2] E.W. Abel, F.G.A. Stone, G. Wilkinson (Eds.), Comprehensive
Organometallic Chemistry II’’, vol. 12, 1995 pp. 259–287 (Chapter
3.5).
[3] (a) V. De Felice, A. De Renzi, D. Tesauro, A. Vitagliano,
Organometallics 11 (1992) 3669;
Acknowledgements
(b) V. De Felice, M.E. Cucciolito, A. De Renzi, F. Ruffo, D.
Tesauro, J. Organomet. Chem. 493 (1995) 1;
The authors thank the MIUR (Programmi di Ricerca
Scientifica di Rilevante Interesse Nazionale, Cofinanzia-
mento 2002) and the Centro Interdipartimentale di Met-
`
odologie Chimico-Fisiche, Universita di Napoli
‘‘Federico II’’ for NMR and X-ray facilities.
(c) M.E. Cucciolito, A. De Renzi, I. Orabona, F. Ruffo, D.
Tesauro, J. Chem. Soc., Dalton Trans. (1998) 1675.
[4] M. Fusto, F. Giordano, I. Orabona, A. Panunzi, F. Ruffo,
Organometallics 16 (1997) 5981.
[5] N. Walker, D. Stuart, Acta Crystallogr., Sect. A 39 (1983) 158–
166.
[6] G.M. Sheldrick, SHELX-97. University of Go¨ttingen, Germany,
1997.
[7] V.G. Albano, M. Di Serio, M. Monari, I. Orabona, A. Panunzi,
F. Ruffo, Inorg. Chem. 41 (2002).
Appendix A. Supplementary data
[8] V.G. Albano, G. Natile, A. Panunzi, Coord. Chem. Rev. 133
(1994) 67–114.2672.
Experimental details and characterization data. Crys-
tallographic data for the structural analysis have been
deposited with the Cambridge Crystallographic Data
Centre, CCDC No. 249352 for the compound 3b00. Cop-
ies of this information may be obtained free of charge
from the Director, CCDC, 12, Union Road, Cambridge
[9] V.G. Albano, M. Monari, I. Orabona, F. Ruffo, A. Vitagliano,
Inorg. Chim. Acta 265 (1997) 35–46.
[10] A CpRu(II) complex was recently reported as efficient catalyst in
deprotection of hydroxyl groups protected as allyl ethers, see S.
Tanaka, H. Saburi, Y. Ishibashi, M. Kitamura, Org. Lett. 6
(2004) 1873.