Influence of the substituent R1 on the reactivity of [(η5-C5H5)Fe{(η5-C 5H4)-CHN-(R1)-OH}] {R1 = -CH 2-CH2- or 1,2-C6H4} with platinum(ii) and on the properties of the complexes

Article abstract of DOI:10.1039/b9nj00509a

The synthesis and characterization of four different types of platinum(ii) complexes derived from [(η⁵-C₅H₅) Fe{(η⁵-C₅H₄)-CHN-(CH₂-CH ₂-OH)}] (1b) are reported. The new products are the trans- and cis-isomers of [Pt{(η⁵-C₅H₅) Fe[(η⁵-C₅H₄)-CHN-(CH₂-CH ₂-OH)]}Cl₂(dmso)] {trans-(2b) and cis-(3b)} and the platinacycles [Pt{[(η⁵-C₅H₃)-CHN-(CH ₂-CH₂-OH)]Fe(η⁵-C₅H ₅)}Cl(L)] {with L = dmso (4b) or PPh₃ (5b)} or [Pt{[(η⁵-C₅H₃)-CHN-(CH₂-CH ₂-O)]Fe(η⁵-C₅H₅)}(L)] {with L = PPh₃ (6b) or dmso (7b)}. In these products 1b acts as a (N) donor ligand (in 2b and 3b), as a bidentate [C(sp², ferrocene), N] ^- (in 4b and 5b) or as a terdentate [C(sp², ferrocene), N, O]^2- ligand (in 6b and 7b). The crystal structures of 4b and 5b confirm the mode of binding of the ferrocenylaldimine (1b) and a trans-arrangement between the neutral ligand (L) and the imine nitrogen. Comparative studies of: (a) the reactivity of ligands [(η⁵-C ₅H₅)Fe{(η⁵-C₅H ₄)-CHN-(R¹)-OH}] {R¹ = -CH₂-CH ₂- or 1,2-C₆H₄} with cis-[PtCl ₂(dmso)₂] and (b) the properties of their platinum(ii) complexes are also reported and provide conclusive evidence of the influence of the nature of the R¹ group of [(η⁵-C₅H ₅)Fe{(η⁵-C₅H₄)-CHN-(R ¹)-OH}] on: (a) the spectroscopic and electrochemical properties of these ligands and their platinum(ii) complexes, (b) the cycloplatination process, and (c) the ease with which the mode of binding of the ligand changes from bidentate [C(sp², ferrocene), N]^- to terdentate [C(sp², ferrocene), N, O]^2- or vice versa. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique.

Full text of DOI:10.1039/b9nj00509a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Influence of the substituent R1 on the reactivity of [(g5-C5H5)
Fe{(g5-C5H4)–CHQN–(R1)–OH}] {R1 = –CH2–CH2– or 1,2-C6H4}
with platinum(II) and on the properties of the complexesw
a
a
b
b
´ ´ ´
´
Concepcion Lopez,* Sonia Perez, Xavier Solans, Merce Font-Bardıa and
Teresa Calvetb
Received (in Victoria, Australia) 14th July 2009, Accepted 5th November 2009
First published as an Advance Article on the web 28th January 2010
DOI: 10.1039/b9nj00509a
The synthesis and characterization of four different types of platinum(II) complexes derived from
[(Z5-C5H5)Fe{(Z5-C5H4)–CHQN–(CH2–CH2–OH)}] (1b) are reported. The new products are the
trans- and cis-isomers of [Pt{(Z5-C5H5)Fe[(Z5-C5H4)–CHQN–(CH2–CH2–OH)]}Cl2(dmso)]
{trans-(2b) and cis-(3b)} and the platinacycles [Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–OH)]
Fe(Z5-C5H5)}Cl(L)] {with L = dmso (4b) or PPh3 (5b)} or [Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–O)]
Fe(Z5-C5H5)}(L)] {with L = PPh3 (6b) or dmso (7b)}. In these products 1b acts as a (N) donor
ligand (in 2b and 3b), as a bidentate [C(sp2, ferrocene), N]ꢀ (in 4b and 5b) or as a terdentate
[C(sp2, ferrocene), N, O]2– ligand (in 6b and 7b). The crystal structures of 4b and 5b confirm the
mode of binding of the ferrocenylaldimine (1b) and a trans-arrangement between the neutral ligand
(L) and the imine nitrogen. Comparative studies of: (a) the reactivity of ligands [(Z5-C5H5)
Fe{(Z5-C5H4)–CHQN–(R1)–OH}] {R1 = –CH2–CH2– or 1,2-C6H4} with cis-[PtCl2(dmso)2] and
(b) the properties of their platinum(II) complexes are also reported and provide conclusive evidence
of the influence of the nature of the R1 group of [(Z5-C5H5)Fe{(Z5-C5H4)–CHQN–(R1)–OH}] on:
(a) the spectroscopic and electrochemical properties of these ligands and their platinum(II) complexes,
(b) the cycloplatination process, and (c) the ease with which the mode of binding of the ligand
changes from bidentate [C(sp2, ferrocene), N]ꢀ to terdentate [C(sp2, ferrocene), N, O]2– or vice versa.
Introduction
Organometallic platinum(II) compounds containing bidentate
{[C, E]ꢀ (E = N, S, P)}1,2 or terdentate {[X, C, X]ꢀ (X = N,
S, or P) or [C, N, X0]q (X0 = N0, q = ꢀ1 or X0 = S or O and
q = ꢀ1 or ꢀ2)1,3} ligands have attracted great interest recently
due to their physical, photo-optical and chemical properties,
their antitumoral and catalytic activity and their utility
Fig. 1 Chemical formulae of the ferrocenyl Schiff bases: [(Z5-C5H5)-
in macro- and supramolecular chemistry, organic and
organometallic synthesis and homogeneous catalysis.1–3
On the other hand, despite the interest of platinum(II) com-
plexes with ferrocene derivatives as ligands and the prochiral
nature of the ferrocenyl moiety in the cyclometalation process,4
compounds with ‘‘Pt[C(sp2, ferrocene), N, O]q (q = ꢀ1 or ꢀ2)’’
cores are scarce.5 Among the examples known, those derived
from [(Z5-C5H5)Fe{(Z5-C5H4)–CHQN–(C6H4–2–OH)}] (1a)
(Fig. 1) are especially relevant due to their behaviour as pH-
based molecular switches.5b In view of this and in order to
determine the effect induced by the nature of the unit R1 of
[(Z5-C5H5)Fe{(Z5-C5H4)–CHQN–(R1)–OH}] {R1 = 1,2-C6H4
Fe{(Z5-C5H4)–CHQN–(R1)–OH)}] with R1 = 1,2-C6H4 (1a) or
–CH2–CH2– (1b).
(1a) or –CH2–CH2– (1b), Fig. 1} on: (a) their reactivity in front
of platinum(II), (b) the properties of the platinacycles containing
1a or 1b acting as [C(sp2, ferrocene), N, O]2– and [C(sp2,
ferrocene), N]ꢀ ligands and (c) the utility of the platinacycles
as pH-dependent molecular switches. Here we present the study
of the reactivity of [(Z5-C5H5)Fe{(Z5-C5H4)–CHQN–(CH2–
CH2–OH)}] (1b)6 (Fig. 1) with cis-[PtCl2(dmso)2]7 as well as
acomparative study of the effect of the R1 group on the
properties of the platinacycles derived from 1b and 1a.
a
´
`
´
Departament de Quımica Inorganica, Facultat de Quımica,
´
`
Universitat de Barcelona, Martı i Franques 1-11, 08028-Barcelona,
Spain. E-mail: conchi.lopez@qi.ub.es; Fax: +34 934907725;
Tel: +34 934039134
Results and discussion
Synthesis
b
`
Departament de Cristal lografia, Mineralogia i Diposits Minerals,
Facultat de Geologia, Universitat de Barcelona, Martı i Franques
s/n, 08028-Barcelona, Spain
w CCDC reference numbers 734801 and 734802. For crystallographic
data in CIF or other electronic format see DOI: 10.1039/b9nj00509a
´
`
In order to facilitate the reading, a summary of the react-
ions studied is presented in Scheme 1 and Table 1. For
comparison purposes the conditions used previously to isolate
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
676 | New J. Chem., 2010, 34, 676–685

View Article Online
(vide infra) confirmed that the aldimine (1b) behaved as a
[C(sp2, ferrocene), N]ꢀ ligand. These results are in sharp
contrast with those obtained for [(Z5-C5H5)Fe{(Z5-C5H4)–
CHQN–(C6H4–2–OH)}] (1a), that under identical experimental
conditions,
gave
[Pt{[(Z5-C5H3)–CHQN–(C6H4–2–O)]-
Fe(Z5-C5H5)}(dmso)] with a mer-terdentate [C(sp2, ferrocene),
N, O]2– ligand.5b
The yield of 4b improved substantially when the reaction
period (t) increased from 72 h to 120 h {Table 1, entries III and
IV, respectively and Scheme 1, step (ii)} and the relative
abundance of 2b and 3b decreased. Thus, the separation of
4b was easier.
Furthermore, the reaction of 2b with the equimolar amount
of NaAcO in refluxing methanol for 120 h also produced a
mixture of 3b and 4b {Scheme 1, step (iii)}, but the molar ratio
4b : 3b (3.2 : 1.0) was smaller than that obtained when 1b was
treated with cis-[PtCl2(dmso)2] and a two-fold excess of
NaAcO (Table 1, entry IV, molar ratio 4b : 3b = 7.1 : 1.0).
Further treatment of 4b with an equimolar amount of PPh3
in benzene at 343 K, gave [Pt{[(Z5-C5H3)–CHQN–
(CH2–CH2–OH)]Fe(Z5-C5H5)}Cl(PPh3)] (5b) {Scheme 1, step
(iv)}. Despite the poor quality of the crystals obtained the
structural studies (see below) confirmed: (a) a cis-arrangement
between the phosphine and the metallated carbon and (b) the
mode of binding of the ligand, [C(sp2, ferrocene), N]ꢀ, in 5b.
It should be noted that no evidence of the opening of the
metallacycle was detected by NMR even when the molar ratio
PPh3 : 5b was 2.0 : 1.0). This suggested that the Pt–N bond of
4b and 5b exhibits low lability.
Scheme 1 (i) Cis-[PtCl2(dmso)2] {molar ratio 1b : Pt(II) = 1} in
refluxing MeOH. The molar ratio 2b : 3b is dependent on the refluxing
period (see text). (ii) Cis-[PtCl2(dmso)2] and NaAcO {molar ratios
1b : Pt(II) : AcOꢀ = 1 : 1 : 2} in a mixture toluene : MeOH (5 : 1) under
reflux followed by a SiO2 column chromatography (see text). (iii)
Treatment with an equimolar amount of NaAcO in refluxing MeOH
for 120 h. (iv) Addition of PPh3 {in a molar ratio 4b : PPh3 = 1} in
benzene at 343 K for 1.5 h, followed by a SiO2 column chromato-
graphy. (v) In CDCl3 (or CH2Cl2) followed by the addition of NaOD
(or MeOH) (see text).
trans-[Pt{(Z5-C5H5)Fe[(Z5-C5H4)–CHQN–(CH2–CH2–OH)]}
Cl2(dmso)] (2b)6 are also included in Table 1, entry I.
Due to the increasing interest in platinacycles with pincer
ligands1,3,5,8a,10 and since 4b and 5b contain an alcohol unit on
the pendant arm we decided to determine whether the presence
of a base could induce the formation of platinum(II) complexes
with ‘‘Pt{C(sp2, ferrocene), N, O}’’ cores.
In a first attempt to achieve cycloplatinated complexes,
ligand 1b6 was treated with an equimolar amount of cis-
[PtCl2(dmso)2]7 in refluxing methanol for 20 h [Table 1, entry
II and Scheme 1, step (i)]. After the work up of a SiO2 column
chromatography three products were isolated. The minor com-
ponents were identified as ferrocenecarboxaldehydez (hereinafter
referred to as FcCHO) and 2b; while characterization data of
the major product (3b) agreed with those expected for the cis-
isomer of [Pt{(Z5-C5H5)Fe[(Z5-C5H4)–CHQN–(CH2–CH2–
OH)]}Cl2(dmso)]. However, no evidence of the formation of
In a first attempt, we studied the action of NaOD (in
MeOD) on a CDCl3 solution of 5b and monitored the changes
by 31P{1H}-NMR (Fig. 2, A–D). The addition of NaOD (in
MeOD) to a freshly prepared solution of 5b in CDCl3 at 298 K
in molar ratios NaOD : 5b varying from 0.2 : 1.0 to 0.7 : 1.0}
did not produce significant changes in the 31P{1H}-NMR
spectra. However, when larger excesses of NaOD were added
(molar ratios from 0.9 : 1.0 to 1.5 : 1.0), the signals detected in
the spectra (Fig. 2, B and C) indicated the presence of two
different species in solution one of them being unreacted 5b.
The chemical shift of the new singlet (d = 16.3 ppm) as well as
the value of the coupling constant (1JP–Pt = 3498 Hz) sug-
gested, according to the literature,5b the presence of [Pt{[(Z5-
1
any platinacycle was detected by H-NMR.
In view of this, and since it has been reported that the
formation of metallacycles with [C(sp2, ferrocene), N]ꢀ or
[C(sp2, ferrocene), N, X]ꢀ (X = N0, S) ligands is often
favoured in the presence of an excess of base,5,8–10 we also
studied the reaction of 1b, cis-[PtCl2(dmso)2] and NaAcO (in a
molar ratio of 1 : 1 : 2) in a toluene–methanol (5 : 1) mixture
under reflux for 72 h. After the work-up, four compounds
were obtained, two of them were 2b and 3b, and one of the
minor components was FcCHO (Table 1, entry III). Character-
ization data of the fourth product (4b) agreed with those
expected for the platinacycle [Pt{[(Z5-C5H3)–CHQN–(CH2–
CH2–OH)]Fe(Z5-C5H5)}Cl(dmso)]. The crystal structure of 4b
C5H3)–CHQN–(CH2–CH2–O)]Fe(Z5-C5H5)}(PPh3)]
(6b),
with a [C(sp2,ferrocene), N, O]2– ligand. The complete trans-
formation of 5b into 6b was achieved when the molar ratio
NaOD : 5b was 1.8 : 1.0 {Fig. 2, D and Scheme 1, step (v)}.
Similarly, when a CDCl3 solution of 4b was treated at 298 K
with NaOD (in MeOD and in a molar ratio NaOD: 4b =
1.8 : 1.0), the 1H-NMR spectrum of the resulting solution
(Fig. 3), suggested the formation of [Pt{[(Z5-C5H3)–CHQ
N–(CH2–CH2–O)]Fe(Z5-C5H5)}(dmso)] (7b) (Scheme 1). How-
ever, when the same experiment was carried out using 3b instead
of 4b or 5b the 1H-NMR spectrum of the resulting solution
revealed the presence of large amounts of FcCHO.
z FcCHO arises form the hydrolysis of the imine. The formation of
FcCHO or C6H5CHO as side-products has also been detected in
cycloplatination of other ferrocenyladimines5,6,8 or the diimine
(1R,2R)-1,2-[{(C6H5–CHQN)}2–C6H10] (J. Bravo, C. Cativela, R.
Navarro and E. P. Urriolabeitia, J. Organomet. Chem. 2002, 650,
157–172).
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
New J. Chem., 2010, 34, 676–685 | 677

View Article Online
Table 1 Summary of the experimental conditions [reagents, solvents, temperature (T) and reaction time (t, in h)] used in the study of the reactivity
of [(Z5-C5H5)Fe{(Z5-C5H4)–CHQN–(CH2–CH2–OH)}] (1b) with cis-[PtCl2(dmso)2] (abreviated as [Pt])
Entry
Reagents (molar ratio)
Solvent
T
t
Productsa (molar ratio)
I
II
III
IV
1b and [Pt] (1 : 1)
1b and [Pt] (1 : 1)
1b, [Pt] and NaAcO (1 : 1 : 2)
1b, [Pt] and NaAcO (1 : 1 : 2)
MeOH
MeOH
Reflux
Reflux
Reflux
Reflux
3
2bb
20
72
120
2b and 3b (0.36 : 1.00)
2b, 3b and 4b (0.24 : 1.0)d
2b, 3b and 4b (0.05 : 0.14 : 1.00)
Toluene/MeOHc
Toluene/MeOHc
a
b
c
d
In these reactions small amounts of FcCHO were also isolated as by-product. Data from ref. 6. A (5 : 1) mixture. In this case traces (12 mg)
of 4b were obtained.
Fig. 2 31P{1H}-NMR spectrum (in the range ꢀ10 ppm o d o
40 ppm) of
a
freshly prepared solution of [Pt{[(Z5-
C5H3)–CHQN–(CH2–CH2–OH)]Fe(Z5-C5H5)}Cl(PPh3)]
CDCl3 (A) and after the addition of NaOD (in MeOD) in molar
(5b)
in
Fig. 3 1H-NMR spectra (400 MHz) of a freshly prepared solution
of 4b in CDCl3 (A) and after the addition of NaOD (in MeOD) in
a molar ratio NaOD : 4b = 1.8 : 1.0 (B) (Cp represents the C5H5
ring).
ratios NaOD : 5b = 1.0 : 1.0 (B), 1.5 : 1.0 (C) and 1.8 : 1.0 (D).
It should be noted that the transformations 5b - 6b (or 4b
- 7b) are slower than 5a - 6a (or 4a - 7a) and require a
larger excess of NaOD {molar ratios NaOD : (5b or 4b) =
1.8 : 1.0 versus NaOD : (5a or 4a) = 1.0 : 1.05b}.
Characterization of the compounds
In order to complete the characterization of 6b and 7b,
products 5b and 4b, were treated separately in CH2Cl2 with
NaOH {molar ratio NaOH : (4b or 5b) = 1.8 : 1.0} dissolved
in methanol at 298 K {Scheme 1, steps (v)}. After work
up a deep-purple (for 6b) or a garnet (for 7b) solid was
isolated. Elemental analyses of 6b and 7b agreed with the
proposed formulae. Their IR spectra showed the typical
band of the >CQN– group11 {at 1569 (for 6b) and 1575 cmꢀ1
(for 7b)}, but that due to the stretching of the –OH unit was
not detected. This suggested: (a) the deprotonation of the
alcohol group and (b) the existence of the Pt–O bond in
6b and 7b.
The new products (3b–7b) were characterized by elemental
analyses, mass spectrometry, infrared and UV-vis spectro-
scopies and NMR spectroscopy. In all cases elemental ana-
lyses (see Experimental) were consistent with the proposed
formulae.
The most relevant features observed in the IR spectra are:
(a) the existence of the typical band due to the stretching of the
imine group5,6,8,10,11 and (b) the presence of the absorptions of
the S-bonded dmso ligand (in 3b, 4b and 7b),12 or the
X-sensitive bands of PPh3 (in 5b and 6b)13 in platinacyles
[Pt(C,N)Cl(PPh3) or [Pt(C,N,O)(PPh3)].5 An intense band in
the range 3460–3490 cmꢀ1, assigned to the stretching of the
s(O–H) bond, was also observed in the IR spectra of com-
pounds 3b–5b. According to the literature,14 the position of
this absorption suggested the existence of intermolecular
hydrogen interactions in the solid state.14 The crystal struc-
tures of 4b and 5b (see below) confirmed this finding. It should
be noted that the absence of the band due to the stretching of
the O–H bond in the IR spectrum of 6b and 7b agreed with the
existence of the Pt–Oalcoxo bond.
1
In the H-NMR spectra of freshly prepared solutions of 6b
and 7b in CDCl3 the signals were slightly broader than
expected, in particular when compared with those of 5b and
4b, respectively. Besides, when the 31P{1H}-NMR spectrum of
a CDCl3 solution of 6b was registered after 1 h in solution at
298 K the signals due to 5b (d = 17.1 ppm, 1JP–Pt = 4017 Hz)
were also detected. These findings suggested that in 6b and 7b
the Pt–O bond is labile.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
678 | New J. Chem., 2010, 34, 676–685

View Article Online
The assignment of the signals observed in the 1H and 13C{1H}-
NMR spectra (see Experimental) was achieved with the aid of
two-dimensional NMR experiments [{1H–1H} NOESY and
COSY and {1H–13C} HSQC and HMBC]. Proton and 13C{1H} -
NMR spectra showed the typical pattern of mono-(for 3b) or
1,2-disubstituted ferrocene derivatives (for 4b–7b).4–6,8–11
1
In the H-NMR spectra the resonance of the imine proton
appeared as a singlet (for 3b, 4b and 7b) or as a doublet (for 5b
and 6b) and exhibited the typical satellites due to 195Pt
coupling. This is consistent with the presence of the s(Pt–Nimine
)
bond. The spectra of 3b, 4b and 7b showed two singlets in the
range 2.90–3.60 ppm that are attributed to the dmso ligand. As
expected,5,8,10 the resonance due to the metallated carbon (C2)
of 4b and 5b appeared at lower fields than for the free ligand or
compounds 2b and 3b.
The chemical shifts of the signals observed in the 31P{1H}-
NMR spectra of 5b {d = 17.1, 1JP–Pt = 4017 Hz} and 6b {d =
16.3 ppm, JP–Pt = 3498 Hz} are consistent with a cis-
1
arrangement between the metallated carbon and the phos-
phorous of the PPh3 ligand in good agreement with the
so-called transphobia effect.15 The differences detected in the
position of these signals {Dd = (dfor 5b ꢀ dfor 6b) E 0.8 ppm}
is similar to those reported for 5a (d = 16.2 ppm) and 6a
(d = 15.5 ppm).5b
Fig. 4 UV-vis spectra of 5 ꢂ 10ꢀ5 M solutions of the free ligand (1b),
the two isomers of [Pt{(Z5-C5H5)Fe[(Z5-C5H4)–CHQN–(CH2–
CH2–OH)]}Cl2(dmso)] {trans-(2b) and cis-(3b)} and the platinacycles
[Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–OH)]Fe(Z5-C5H5)}Cl(L)] {with
L = dmso (4b) or PPh3 (5b)} in CH2Cl2 at 298 K.
195Pt{1H}-NMR spectra of 2b–5by provided convincing
evidence of: (a) the relative arrangement of the monodentate
ligands in 2b and 3b and (b) the mode of binding of the ligand
{(N)-donor (in 3b) or [C(sp2, ferrocene), N]ꢀ (in 4b and 5b).
The variation observed in the chemical shift of 2b and 3b (Dd =
dfor 3b ꢀ dfor 2b) falls in the range expected for other pairs of
cis- and trans-isomers of complexes of general formulae:
[PtCl2(N-donor ligand)2] or [PtCl2(N-donor ligand)(L0)]
(L0 = monodentate ligand).16–18
Table 2 Summary of the experimental UV-vis spectroscopic data for
[(Z5-C5H5)Fe{(Z5-C5H4)-CHQN–(CH2–CH2–OH)}] (1b) and its
platinum(II) derivatives (2b–5b) in CH2Cl2 at 298 K: wavelengths
(li in nm) and molar extinction coefficients {ei in molꢀ1 dm2 (in
parentheses)}
Compound
l1(e1)
l2(e2)
l3(e3)
1b
2b
3b
4b
5b
273(37 400)
287(31 600)
287(25 200)
283(22 400)
290(28 800)a
325(5040)
355(5320)
358(5180)
368(4260)
378(3140)
453(3240)
473(3800)
481(3300)
518(3860)
505(3019)
The differences detected in the 195Pt {1H}-NMR spectra of
the pairs of compounds [Pt{[(Z5-C5H3)–CHQN–(R1)–OH)]-
Fe(Z5-C5H5)}Cl(L)] with L = dmso [R1 = –CH2–CH2– (4b)
(d = ꢀ3791 ppm} or 1,2-C6H4 (4a) (d = ꢀ3831 ppm)5b] or
a
Shoulder.
L = PPh3 [R1 = –CH2–CH2– (5b) (d = ꢀ4180 ppm, 1JPt–P
=
1
4017 Hz) or 1,2-C6H4 (5a) (d = ꢀ4206 ppm, JPt–P = 4233
Hz)5b] can be ascribed to the nature of the R1 group. Since it is
well-known that an upfield shift of the signal detected in the
195Pt{1H}-NMR spectra is related to a strong ligand field
interaction,5,8,10,18 the comparison of 195Pt-NMR data for 5a
(d = ꢀ4260 ppm), [Pt{[(Z5-C5H3)–CHQN–(CH2–CH2)–NMe2]
Fe(Z5-C5H5)}Cl(PPh3)] (5c) (d = ꢀ4197 ppm8a) and 5b (d =
ꢀ4180 ppm) suggests that the strength of the interaction
between the Pt(II) and the bidentate (C,N)ꢀ ligands increases
according to the sequence: 5b o 5c o 5a.
(ILT) [3d(Fe) - p* and p - p*, respectively].10,11b For the
platinum(II) complexes these bands, that are due to metal
perturbed intraligand transitions (MPILT), appeared at lower
energies than for the free ligand (1b). For the platinacycles (4b
and 5b) the magnitude of the shift is greater than for 2b and 3b.
This trend is similar to that found for compounds 1a–6a.5b
Furthermore, comparison of UV-vis data for the two
families of compounds (2a–6a and 2b–6b5b) reveals that the
replacement of the R1 moiety in 2a–6a (R1 = 1,2-C6H4) by
the –CH2–CH2– unit produces a shift (of ca. 15–20 nm) of the
absorption maxima of the two MPILT to higher energies. This
suggests that the R1 unit modifies the energy of the 3d(Fe), p
and p* levels involved in these transitions. Molecular orbital
(MO) calculations of 1a19 and of metallacycles derived from
[(Z5-C5H5)Fe{(Z5-C5H4)–CHQN–R2}] (with R2 = other sub-
stituted phenyl groups) have shown that the HOMO and
LUMO are not solely iron based. In these MO there is a
contribution of p*orbitals of the phenyl unit on the imine
nitrogen.10,11b,19 However, this is not possible in compounds
with R1 = –CH2–CH2–.
The UV-vis spectra of 5 ꢂ 10ꢀ5 M solutions of the ligand
(1b) and compounds 2b–5b were registered in CH2Cl2 at 298 K
(Fig. 4 and Table 2). A summary of the results obtained from
these studies is presented in Table 2.
All spectra (Fig. 4) showed three bands. According to
previous studies on UV-vis spectra of [(Z5-C5H5)Fe
{(Z5-C5H4)–CHQN–(R2)}] the two bands in the range
300 nm o l o 480 nm are attributed to intraligand transitions
y Due to the low stability of compounds 6b and 7b in CDCl3 and
acetonitrile-d3, their 195Pt{1H}-NMR spectrum was not registered.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
New J. Chem., 2010, 34, 676–685 | 679

View Article Online
Description of the crystal structures of 4b and 5b
These structures consist of molecules of [Pt{[(Z5-C5H3)–
CHQN–(CH2–CH2–OH)]Fe(Z5-C5H5)}Cl(L)] {L = dmso
(for 4b) or PPh3 (for 5b)}. ORTEP plots are shown in Fig. 5
and 6. In the two cases, bond lengths and angles around the
platinum (II) atom are similar to those reported for related
platinacycles of the type [Pt{k2-C(sp2,ferrocene),N-li-
gand}Cl(L)] with L = dmso or PPh3.5b,20 The values of the
bond angles C(6)–Pt–S {95.88(5)1 (in 4b)} and C(6)–Pt–P(1)
{92.5(3)1 (in 5b)} indicate that the neutral ligands L are in a
cis-arrangement to the metallated carbon, in good agreement
with the transphobia effect.15
Each one of the molecules of [Pt{[(Z5-C5H3)–CHQN–
(CH2–CH2–OH)]Fe(Z5-C5H5)}Cl(L)] contains a [5.5] bicyclic
system formed by a five-membered platinacycle and the
substituted ring of the ferrocenyl unit. In 4b and 5b the set
of atoms that define the platinacycle [Pt, N(1), C(6), C(10),
C(11)], are nearly co-planar (the puckering analyses gives t
values of 3.0 and 4.3, respectively).
Fig.
6
ORTEP plot (50% probability thermal ellipsoids) of
[Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–OH)]Fe(Z5-C5H5)}]Cl(PPh3)] (5b).
Hydrogen atoms have been omitted for clarity. Selected bond lengths
(in A) and bond angles (in 1): Pt–C(6), 2.026(9); Pt–N(1), 2.190(7);
Pt–Cl(1), 2.376(2); Pt–P(1), 2.243(3); C(6)–C(10), 1.483(11);
C(10)–C(11), 1.509(12); C(11)–N(1), 2.56(10); N(1)–C(12), 1.544(12);
C(12)–C(13), 1.592(12), C(13)–O(1), 1.440(10); C(6)–Pt–N(1), 80.7(3);
N(1)–Pt–Cl(1), 92.0(2); Cl(1)–Pt–P(1), 95.36(9); C(6)–Pt–P(1), 92.5(3);
In each molecule of [Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–OH)]-
Fe(Z5-C5H5)}Cl(L)] the platinum atom is located in a slightly
distorted square-planar environment and it is bound to the
imine nitrogen, N(1) and the C(6) carbon atom of the ferro-
cenyl unit. Thus, confirming that the mode of binding of 1b is
[C(sp2,ferrocene), N]ꢀ. One chloride and the sulfur atom of the
dmso ligand (in 4b) or the phosphorous of the PPh3 ligand (in
5b) fulfil the coordination sphere.
In the molecules of [Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–OH)]-
Fe(Z5-C5H5)}Cl(L)] {L = dmso (4b) or PPh3 (5b)}, the oxygen
atom of the –OH unit is out of the coordination plane and it is
close to the C5H5 moiety. The separation between the O(1)
atom and the hydrogen atom: H(5) in 4b or H(1) in 5b,
being: 3.06 A and 2.74 A, respectively.
Pt–C(6)–C(10),
110.1(7);
113.4(8);
C(6)–C(10)–C(11),
119.1(8);
116.4(7) and
C(10)–C(11)–N(1),
C(11)–N(1)–Pt,
C(12)–N(1)–Pt, 126.0(6).
Bond lengths and angles of the ferrocenyl moiety agree with
the values reported for other 1,2-disubstituted ferrocene
derivatives.20 The two pentagonal rings are planar, nearly
parallel [tilt angles = 2.31 (for 4b) and 1.71 (for 5b)], and they
deviate by ca. 10.61 (in 4b) and 0.51 (in 5b) from the ideal
eclipsed conformation. The separation between the Pt(II) and
the Fe(II) atoms [3.587 A (for 4b) and 3.621 A (for 5b)] clearly
exceeds the sum of the van der Waals radii21 of these atoms,
thus precluding the existence of any direct interaction.
In the crystals of 4b and 5b, there are weak intermolecular
Clꢃ ꢃ ꢃH–O interactions connecting two molecules of [Pt{[(Z5-
C5H3)–CHQN–(CH2–CH2–OH)]Fe(Z5-C5H5)}Cl(L)] {L =
dmso (in 4b) or PPh3 (in 5b)} (Fig. 7).
Electrochemical studies
In order to elucidate the effect induced by: (a) the mode of
binding of the ligand, (b) the relative arrangement of the Clꢀ
ligands in 2b and 3b, and the nature of R1 in the platinum(II)
complexes with R1 = –CH2–CH2– (in 2b–5b) or 1,2-C6H4
(2a–5a) on the electronic environment of the Fe(II), we also
undertook electrochemical studies of compounds 2b–5b based
on cyclic voltammograms.z As shown in Fig. 8 the cyclic
voltammograms exhibit an anodic peak with a directly asso-
ciated reduction peak in the reverse scan.
A summary of the most relevant electrochemical parameters
is presented in Table 3 together with those reported for
compounds 2a–5a under identical experimental conditions.
For compounds under study the DE values depart appreciably
Fig.
5 ORTEP plot (50% probability thermal ellipsoids) of
[Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–OH)]Fe(Z5-C5H5)}Cl(dmso)] (4b).
Hydrogen atoms have been omitted for clarity. Selected bond lengths
(in A) and bond angles (in 1): Pt–C(6), 2.016(5); Pt–N(1), 2.078(4);
Pt–Cl, 2.3947(19); Pt–S, 2.1977(16); C(6)–C(10), 1,427(7);
C(10)–C(11), 1.417(7); C(11)–N(1), 1.298(7); N(1)–C(12), 1.454(8);
C(12)–C(13), 1.483(11); C(13)–O(1), 1.444(10); C(6)–Pt–N(1),
80.5(2); N(1)–Pt–Cl, 93.21(15); Cl–Pt–S, 90.45(6); C(6)–Pt–S,
95.88(15); Pt–C(6)–C(10), 112.0(4); C(6)–C(10)–C(11), 115.8(5);
z
1H-NMR spectra of 2b–5b in acetonitrile-d3 (see Experimental)
indicate that these products are stable in this solvent. In the 1H-
NMR spectrum of 6b in CD3CN the signals were broader than
expected and its 31P{1H}-NMR spectrum showed two singlets at
17.3 (1JP–Pt = 4098 Hz) and at 16.7 ppm (1JP–Pt = 3425 Hz) {of
relative intensities 1.0 : 0.2}, respectively} that correspond to 5b and
6b, respectively.
C(10)–C(11)–N(1),
117.7(4);
C(11)–N(1)–Pt,
113.8(3)
and
C(12)–N(1)–Pt, 125.8(4).
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
680 | New J. Chem., 2010, 34, 676–685

View Article Online
Fig. 9 Cyclometallated compounds: [Pt{[(Z5-C5H3)–C(Me)QN–OH]-
Fe(Z5-C5H5)}Cl(dmso)] and [M{[(Z5-C5H3)–CHQN–(C6H4–2-SMe)]-
Fe(Z5-C5H5)}Cl] with M = Pd or Pt described before.10,24
from the constant value of 59 mV (theoretically expected for
an electrochemically reversible one-electron-step oxidation-
reduction process) suggesting that a structural reorganization
takes place upon oxidation.22
The cyclic voltammograms of the trans-(2b) and cis-(3b)
isomers of [Pt{(Z5-C5H5)Fe[(Z5-C5H4)–CHQN–(CH2–CH2–
OH)]}Cl2(dmso)] were similar to that of the free ligand (1b)
except for the position of the wave that appeared strongly
shifted to more anodic potentials and this suggests that the
binding of the platinum(II) to 1b inhibits the oxidation of the
ferrocenyl unit. This trend is similar to that reported for 1a
compounds 2a, 3a, the imines [(Z5-C5H5)Fe{(Z5-C5H4)–
C(R1)QN–R2}] (R1 = H, Me or Ph and R2 = phenyl or
benzyl groups) and their platinum(II) or palladium(II)
derivatives where these ligands act as an N-donor
group.5b,11b,23
Fig. 7 Schematic view of the intermolecular Clꢃ ꢃ ꢃH–O interactions
between molecules of [Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–OH)]-
Fe(Z5-C5H5)}Cl(L)] in the crystals of 4b (L = dmso) and 5b (L =
PPh3).
For the platinacycles [Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–
OH)]Fe(Z5-C5H5)}Cl(L)] {with L = dmso (4b) or PPh3 (5b)}
the ferrocene-centred transition shifts to more cathodic
potentials. This agrees with the results reported for
[Pt{[(Z5-C5H3)–C(Me)QNOH]Fe(Z5-C5H5)}Cl(dmso)]24 or
[M{[(Z5-C5H3)–CHQN–(C6H4–2-SMe)]Fe(Z5-C5H5)}Cl] (M =
Pd or Pt)10 and their corresponding free ligands (Fig. 9).
Conclusions
The studies presented here have allowed us: (a) to compare the
reactivity of the ferrocenyl Schiff bases [(Z5-C5H5)Fe{(Z5-
C5H4)–CHQN–(R1)–OH}] {with R1 = 1,2-C6H4 (1a) or
–CH2–CH2– (1b)} with respect to cis-[PtCl2(dmso)2] and (b)
to demonstrate the relevancy of the nature of the R1 moiety on
these processes and the properties of the final products.
Ligands 1a and 1b exhibit a similar behaviour in refluxing
methanol giving [trans-(2a, 2b) or cis-(3a, 3b)] isomers
of [Pt{(Z5-C5H5)Fe[(Z5-C5H4)–CHQN–(R1)–OH]}Cl2(dmso)].
However, in the presence of a two-fold excess of NaAcO
and using a toluene–MeOH (5 : 1) mixture as solvent, the final
products (7a and 4b, respectively) are markedly different.
Compounds 1a and 1b contain an acidic –OH functionality,24
but in 7a the ligand behaves as a terdentate [C(sp2, ferrocene),
N, O]2– group; while in 4b, the aldimine 1a acts as bidentate
[C(sp2, ferrocene), N)]ꢀ ligand.
The crystal structure of 1a19 shows that the oxygen atom of
the –OH group it is nearly coplanar with the imine unit and
close to the imine nitrogen; while in 1b,6 where the R1 unit is
more flexible, the location of the –OH moiety is similar to
those of 4b and 5b. This suggests that once the Pt(II) ion binds
to the nitrogen of 1a or 1b, the oxygen atom of 1a would
be closer to the metal centre than for 1b. In the presence of
Fig. 8 Cyclic voltammograms of the isomers of [Pt{(Z5-C5H5)Fe[(Z5-
C5H4)–CHQN–(CH2–CH2–OH)]}Cl2(dmso)] {trans-(2b) and cis-(3b)}
and the platinacycles {L = dmso (4b) or PPh3 (5b)} in CH3CN at
298 K and u = 100 mV sꢀ1
.
Table 3 Summary of electrochemical data {anodic (Epa) and cathodic
(Epc) potentials and separation between peaks (DE) in mV}a at a scan
speed
u =
100 mV sꢀ1 for [(Z5-C5H5)Fe{(Z5-C5H4)–CHQN–
(CH2–CH2–OH)}] (1b) and its platinum(II) derivatives (2b–5b). For
comparison purposes data for [(Z5-C5H5)Fe{(Z5-C5H4)–CHQ
N–(C6H4–2–OH)}] (1a), the isomers of [Pt{(Z5-C5H5)Fe[(Z5-
C5H4)–CHQN–(C6H4–2–OH)]}Cl2(dmso)] {trans-(2a) and cis-(3a)}
and the platinacycles [Pt{[(Z5-C5H3)–CHQN–(C6H4–2–OH)]Fe(Z5-
C5H5)}Cl(L)] {with L = dmso (4a) or PPh3 (5a)} are also enclosedb
Epa
Epc
DE
Epa
Epc
DE
1bc
2b
3b
4b
5b
98
30
68
1a
2a
3a
4a
5a
258
415
474
116
151
178
294
323
15
80
262
299
74
106
181
ꢀ98
ꢀ74
156
118
162
125
121
151
101
155
51
–3
a
In all cases the Epa and Epc values are referenced to the ferrocene/
b
ferricinium couple. Data for 1a–5a were taken from ref. 5b. Data
c
from ref. 6.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
New J. Chem., 2010, 34, 676–685 | 681

View Article Online
AcOꢀ the two ligands (1a and 1b) undergo the formation of
the platinacycles with a s{Pt–C(sp2,ferrocene)} bond. For 1a
the acidic –OH group25 is closer to the Pt(II) centre than in 1b,
and the AcOꢀ, also allows the deprotonation of the –OH unit
and the formation of the Pt–O bond giving 7a with a [C(sp2,
ferrocene), N, O]2– ligand. A similar argument can be used to
explain why 4b and 5b are less prone to form complexes with
terdentate [C(sp2, ferrocene), N, O]2– ligands than their ana-
logues with R1 = 1,2-C6H4 (4a and 5a).
constants (J) in Hz. 195Pt{1H}-NMR spectra of 4b and 5b
were obtained with a Bruker-250 MHz instrument at 298 K.
31P{1H}-NMR spectra of 5b and 6b were recorded with a
Varian-Unity-300 instrument. In all cases the solvent for the
NMR experiments was CDCl3 (99.9%) and the references
1
were SiMe4 [for H and 13C{1H}-NMR], P(OMe)3 [d(31P) =
140.17 ppm, for 31P{1H}-NMR] and H2[PtCl6] [for
195Pt{1H}-NMR]. In order to test the stability of the plati-
num(II) compounds in the solvent used for electrochemical
studies the 1H-NMR spectra of these products in acetonitrile-
d3 (99.8%) were also registered at 298 K.
Furthermore, the comparison of the electrochemical proper-
ties of ligands (1a and 1b) and those of their platinum(II)
complexes (2a–5a and 2b–5b) has also shown that the replace-
ment of the 1,2-C6H4 unit (in 1a) by the –CH2–CH2– moiety
(in 1b) increases the proclivity of the ferrocenyl unit to under-
go oxidation. In general, 2b–5b are more prone to oxidize than
their analogues with R1 = 1,2-C6H4 (2a–5a).
Some of the preparations described below required the use
of highly hazardous reagents such as benzene that should be
handled with caution!
Preparation of the compounds
It has been reported that [Pt{[(Z5-C5H3)–CH2–NMe2]-
Fe(Z5-C5H5)}Cl(dmso)], (where the environment of the Pt(II)
is very similar to that of 4b) exhibits antitumour activity,26 and
consequently, the platinacycle 4b appears to be an excellent
candidate for future studies in this area.
Cis-[Pt{(g5-C5H5)Fe[(g5-C5H4)–CHQN–(CH2–CH2–OH)]}-
Cl2 (dmso)] (3b). Cis-[PtCl2(dmso)2] (211 mg, 5.0 ꢂ 10ꢀ4 mol)
was suspended in 40 mL of MeOH. The reaction flask was
protected from the light with aluminium foil and refluxed until
complete dissolution and then it was filtered. Afterwards, an
equimolar amount of ligand 1b (129 mg, 5.0 ꢂ 10ꢀ4 mol) was
added to the filtrate and the resulting solution was refluxed for
20 h. After this period the solution was filtered and concen-
trated to dryness on a rotary evaporator. The deep-brown
residue obtained was dried in vacuum for one night and
afterwards it (260 mg) was treated with 15 mL of a CHCl3 :
acetone (8 : 1) mixture and passed through a (2.0 cm ꢂ 2.5 cm)
SiO2 column using the same CHCl3:acetone solution as eluant.
This produced the release of two bands. The first eluted one
gave, after concentration to dryness, 3b (191 mg) as a red
microcrystalline product. The next band released was collected
and concentrated to dryness giving complex 2b (68 mg) as an
orange powder. Characterization data: Anal. calc. for
C15H21Cl2FeNO2PtS: C, 29.97; H, 3.52; N, 2.33; S, 5.33%;
found: C, 30.1; H, 3.6; N, 2.4; S, 5.5%. EM(ESI+): m/z =
566.0 {[M] ꢀ Cl}+, 623.9 {[M] + Na+}, 1225.0 {2[M] +
On the other hand, we have also shown that in 6b and 7b
the Pt–O bond is more labile than in their analogues with
R1 = 1,2-C6H4 (6a and 7a). Unfortunately, this reduces the
potential utility of the pairs of products (5b and 6b) and
(4b and 7b) as molecular switches, but it can be potentially
interesting for the design of pH-dependent devices. Mainly
due to the extracelular acidity of solid tumors at present there
is an increasing interest in the design of pH-activable com-
pounds for diagnostic and/or therapeutic aims.27 Several
examples of platinum(II) complexes exhibiting higher cyto-
toxic activity with decreasing pH have been reported.28 Thus,
the lability of the Pt–O bond in 6b or 7b makes them specially
attractive from this point of view. Further work on this field
is currently under study.
Na+} and 1229 {2[M]
+
NH4+}. IR data: 1625
Experimental
{n(>CQN–)} and 3450 cmꢀ1 {n(O–H)}. 1H-NMR data in
Materials and methods
3
CDCl3:8 8.23 (s, 1H, JH–Pt = 88 –CHQN–), 4.42 (s, 5H,
C5H5), 4.83 (s, 1H, H2), 4.70 (s, 1H, H3), 4.73 (s, 1H, H4), 6.92
(s, 1H, H5), 3.70 (br m, 1H, –OH), 4.25–4.27 (m, 2H,
Compounds 1b, 2b and cis-[PtCl2(dmso)2] were prepared as
reported previously6,7 and the remaining reagents were ob-
tained from commercial sources and used as received. Except
methanol, that was HPLC-grade, the remaining solvents used
were dried and distilled before use.29 Elemental analyses were
carried out at the Serveis de Cientifico-Tecnics (Universitat
Barcelona). Mass spectra (ESI+) were performed at
the Servei d’Espectrometria de Masses (Universitat de
Barcelona). Infrared spectra were obtained with a Nicolet
400FTIR instrument using KBr pellets. UV-visible spectra of
5 ꢂ 10ꢀ5 M solutions of compounds 1b–5b in CH2Cl2 were
registered with a Cary-100 scan Varian-UV spectrometer at
298 K. Routine 1H and 13C{1H}-NMR spectra were recorded
with a Mercury-400 MHz instrument. High resolution
1H-NMR spectra and the two-dimensional [{1H–1H}-
NOESY and COSY and {1H–13C}-HSQC and HMBC]
experiments were registered with a Varian VRX-500 or a
Bruker Avance DMX-500 MHz instruments at 298 K.
The chemical shifts (d) are given in ppm and the coupling
QN–CH2–), 3.25–3.47 (m, 2H, –CH2–), 3.47 [s, 3H, JPt–H =
18, Me(dmso)] and 3.25 [s, 3H, JPt–H = 19, Me(dmso)].
3
3
1H-NMR data in acetonitrile-d3:8 8.46 (s, 1H, JH–Pt = 95,
–CHQN–), 4.44 (s, 5H, C5H5), 5.05 (s, 1H, H2), 4.90 (s, 1H,
H3), 4.82 (s, 1H, H4), 6.59 (s, 1H, H5), 3.80 (br.m, 1H, –OH),
4.205–4.30 (m, 2H, QN–CH2–), 3.28–3.67 (m, 2H, –CH2–),
3
3.42 [s, 3H, JPt–H = 19, Me(dmso)] and 3.39 [s, 3H, JPt–H
=
18, Me(dmso)]. 13C{1H}-NMR data in CDCl3:8 171.7
(–CHQN–), 75.1 (C1), 74.5 (C2), 73.3 (C3), 74.2 (C4), 75.9
(C5), 70.9 (C5H5), 45.1 and 45.0 [Me(dmso)], 60.3 (–CH2–),
69.6 (QN–CH2–). 195Pt{1H}-NMR data: ꢀ2942.
[Pt{[(g5-C5H3)–CHQN–(CH2–CH2–OH)]Fe(g5-C5H5)}Cl
(dmso)] (4b). This compound can be obtained using two
8 Numbering of the atoms refers to that shown in Scheme 1; abbre-
viations: s = singlet, d = doublet, t = triplet, m = multiplet, br =
broad.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
682 | New J. Chem., 2010, 34, 676–685

View Article Online
alternative procedures {methods (a) and (b), respectively} that
differ in the nature of the starting material. The procedure
described in method (a) allows the isolation of 4b in a high
yield and its separation is easier.
for 1.5 h. Then, the hot solution was filtered and the undis-
solved material was discarded. The filtrate was concentrated to
dryness on a rotary evaporator and the deep-red residue was
then dissolved in the minimum amount of CH2Cl2 and passed
through a small (1.5 cm ꢂ 2.0 cm) SiO2 column to remove the
free dmso. Elution with CH2Cl2 produced the release of a
purple band that was collected and concentrated to dryness
on a rotary evaporator. The solid formed was air-dried and
then dried in vacuum for 3 days (Yield: 31 mg, 77.5%).
Characterization data: Anal. calc. for C31H29ClFeNOPPt: C,
49.72; H, 3.90; N, 1.87%; found: C, 49.8; H, 3.95; N, 1.95%.
EM(ESI+): m/z = 713.1 {[M] ꢀ Cl}+. IR data: 1592
{n(>CQN–)} and 3402 cmꢀ1{n(O–H)}. 1H-NMR data in
CDCl3:8 8.25 (s, d, 1H, JH–Pt = 88 and JH–P = 9.0,
–CHQN–), 3.93 (s, 5H, C5H5), 3.39 (s, 1H, H3), 4.29 (s, 1H,
H4), 4.49 (s, 1H, H5), 4.10 and 4.29 (m, 2H, QN–CH2–), 3.81
and 3.89 (m, 2H, –CH2–), 4.21 (br. m, 1H, –OH) and 7.10–7.80
(m, 15H, aromatic protons of the PPh3 ligand). 1H-NMR data
in acetonitrile-d3:8 8.45 (s, d, 1H, 1JH–Pt = 79 and 4JH–P = 8.5,
–CHQN–), 3.58 (s, 5H, C5H5), 3.30 (s, 1H, H3), 4.15 (s, 1H,
H4), 4.60 (s, 1H, H5), 3.80 and 3.90 (m, 2H, –N–CH2–),
3.60–3.75 (m, 2H, –CH2–), 4.05 (br m, 1H, –OH), and
7.30–7.80 (m, 15H, aromatic protons of the PPh3 ligand).
13C{1H}-NMR data in CDCl3:8 176.1 (–CHQN–),
70.7 (C5H5), 86.9 (C1), 99.3 (C2), 68.3 (C3), 71.6 (C4), 76.4
(C5), 62.5 (QN–CH2–), 59.8 (–CH2–) and four additional
doublets centred at 128.2, 130.0, 131.3 and 135.1 due to the
four types of 13C nuclei of the PPh3 ligand. 31P{1H} NMR data:
in CDCl3 d = 17.1 (1JP–Pt = 4017 Hz) and in acetonitrile-d3:
17.3 (1JP–Pt = 4098 Hz). 195Pt{1H} NMR data in CDCl3:
ꢀ4180(1JP–Pt = 4017 Hz).
Method (a): Ligand 1b (129 mg, 5.0 ꢂ 10ꢀ4 mol), cis-
[PtCl2(dmso)2] (211 mg, 5.0 ꢂ 10ꢀ4 mol) were dissolved in
50 mL of toluene. Then, a solution formed by 10.0 ꢂ 10ꢀ4 mol
(41 mg) of NaAcO and 10 mL of MeOH was added. The
reaction mixture was protected from the light with aluminium
foil and refluxed for 120 h. After this period the hot solution
was filtered, and the filtrate was concentrated to dryness on a
rotary evaporator. The residue was then treated with the
minimum amount of CH2Cl2 and passed through a SiO2
column (2.5 cm ꢂ 4.0 cm). Elution with CH2Cl2 produced
the release of three orange bands that gave, after concentra-
tion, FcCHO (10 mg), 2b (9 mg) and 3b (24 mg), respectively.
Once these bands were collected a garnet one was released, and
the evaporation of the solvent produced 4b (163 mg) that was
collected, and dried in vacuum for two days.
Method (b): Compound 2b (300 mg, 5.0 ꢂ 10ꢀ4 mol) was
dissolved in 30 mL of methanol, then a solution containing an
equimolar amount of NaAcO in MeOH (5 mL) was added.
The reaction mixture was protected from the light with
aluminium foil and refluxed for 120 h. After this period, the
solution was filtered and the filtrate was concentrated to
dryness on a rotary evaporator. The deep-brown residue was
then dissolved in the minimum amount of CH2Cl2 and passed
through a SiO2 column (2.5 cm ꢂ 4.0 cm). The elution with
CH2Cl2 produced the release of two pale orange bands. The
first eluted one gave after concentration, traces (9 mg) of
FcCHO and the second one produced, after work up, 3b
(46 mg). Concentration to dryness of the garnet band released
afterwards gave 4b that was collected and dried in vacuum for
two days (138 mg).
3
4
[Pt{[(g5-C5H3)–CHQN–(CH2–CH2–O)]Fe(g5-C5H5)} (PPh3)]
(6b). Compound 5b (83 mg, 1.11 ꢂ 10ꢀ4 mol) was dissolved in
CH2Cl2 (E18 mL), then a solution formed by 8 mg (2 ꢂ 10ꢀ4
mol) of NaOH dissolved in 3 mL of MeOH was added
dropwise. Once the addition had finished, the reaction mixture
was stirred at 298 K for 30 min. After this period the
undissolved materials were removed by filtration and the
deep-purple solution was concentrated to dryness on a rotary
evaporator. The solid formed was collected and dried in
vacuum for 1 day (yield: 65 mg, 85%).
Characterization data. Anal. calc. for C15H20ClFeNO2PtS:
C, 31.90; H, 3.57; N, 2.48; S, 5.68%; found: C, 32.05; H, 3.6;
N, 2.5; S, 5.8%. EM(ESI+): m/z = 564.0 [M]+and 528.5
{[M] ꢀ Cl}+. IR data: 1597 {n(>CQN–)} and 3408 cmꢀ1
1
3
{n(O–H)}. H-NMR data in CDCl3:8 8.10 (s, 1H, JH–Pt
=
103, –CHQN–), 4.21 (s, 5H, C5H5), 4.37 (t, 1H, 3J = 1.5, H3),
4.50 (d, 1H, J = 1.5, H4), 4.58 (t, 1H, 3J = 1.5, H5), 4.2 (br m,
1H, –OH), 4.17–4.22 (br.m, 2H, QN–CH2–), 3.88 and 3.93
Characterization data. Anal. calc. for C31H28FeNOPPt: C,
3
(m, 2H, –CH2–), 3.53 [s, 3H, JPt–H = 25, Me(dmso)] and
1
52.26; H, 3.96; N, 1.97%; found: C, 52.3; H, 4.05; N, 2.0%. IR
1
3
data: 1569 cmꢀ1 {n(>CQN–)}. H-NMR data (in CDCl3):8
3.57[s, 3H, JPt–H = 21, Me(dmso)]. H-NMR data in acet-
onitrile-d3:8 8.25 (s, 1H, 3JH–Pt = 98, –CH = N–), 4.35 (s, 5H,
8.16 (s, d, 1H, 1JH–Pt = 98 and 4JH–P = 10.2, –CHQN–), 3.82
(s, 5H, C5H5), 3.15 (s, 1H, H3), 4.22 (s, 1H, H4), 4.52 (s, 1H,
H5), 3.88–4.20 (br. m, 4H, QN–CH2– and –CH2–) and
7.29–7.90 (m, 15H, aromatic protons of the PPh3 ligand).
31P{1H} NMR data for a freshly prepared solution in CDCl3
d = 16.3 (1JP–Pt = 3498 Hz) and in acetonitrile-d3: two signals
C5H5), 4.42 (t, 1H, J = 1.5 H3), 4.63 (d, 1H, J = 1.5, H4),
3
3
4.49 (t, 1H, J = 1.5, H5), 4.10–4.29 (br.m, 2H, QN–CH2–),
=
3
3.50 and 3.70 (br m, 3H, –CH2– and –OH), 3.49 [s, 3H, 3JPt–H
3
22, Me(dmso)] and 3.46[s, 3H, JPt–H = 20, Me(dmso)].
13C{1H}-NMR data in CDCl3:8 179.2(–CHQN–), 70.5
(C5H5), 84.2 (C1), 101.4 (C2), 72.1 (C3), 73.9 (C4), 74.8 (C5),
64.3 (QN–CH2–), 59.4 (–CH2–), 47.1 and 46.9 [Me(dmso)].
195Pt{1H} NMR data in CDCl3: ꢀ3791 ppm.
of relative intensities 1.0 : 0.2 centred at 16.7 ppm (1JP–Pt
=
3425 Hz) and 17.1 ppm (1JP–Pt = 4098 Hz)} were observed of
which the later is identical to that of 5b under identical
experimental conditions.
[Pt{[(g5-C5H3)–CHQN–(CH2–CH2–OH)]Fe(g5-C5H5)}Cl
(PPh3)] (5b). Compound 4b (30 mg, 5.3 ꢂ 10ꢀ5 mol) was
dissolved in 10 mL of benzene, then 14 mg (5.3 ꢂ 10ꢀ5 mol)
of PPh3 were added. The reaction mixture was heated at 70 1C
[Pt{[(g5-C5H3)–CHQN–(CH2–CH2–O)]Fe(g5-C5H5)} (dmso)]
(7b). This product was prepared as described for 6b, but using
62 mg of 4b (1.11 ꢂ 10ꢀ4 mol) as starting material (yield: 49 mg,
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
New J. Chem., 2010, 34, 676–685 | 683

View Article Online
Table 4 Crystal data and details of the refinement of the crystal structures of compounds [Pt{[(Z5-C5H3)–CHQN–(CH2–CH2–OH)]-
Fe(Z5-C5H5)}Cl (L)] {with L = dmso (4b) or PPh3 (5b)}
4b
5b
Empirical formula
Formula weight
T/K
Wavelength/A
C15H20ClFeNO2PtS
564.77
293
C31H29ClFeNOPPt
748.91
293
0.71069
0.2 ꢂ 0.1 ꢂ 0.1
0.71073
0.09 ꢂ 0.05 ꢂ 0.05
Crystal sizes/mm ꢂ mm ꢂ mm
Crystal system
Space group
Triclinic
Triclinic
ꢀ
P1
ꢀ
P1
a/A
b/A
c/A
a/1
9.841(5)
10.090(4)
10.239(4)
82.85(39
61.73(2)
76.85(3)
871.8(7)
2
2.152
9.124
540
From 2.84 to 29.93
6864
3622[0.0648]
93.5
3622
200
1.215
R1 = 0.0298
wR2 = 0.0867
R1 = 0.0305
wR2 = 0.0872
9.841(8)
11.040(6)
13.762(8)
96.46(3)
96.67(4)
107.30(4)
1400.7(16)
2
1.776
5.683
732
From 2.65 to 32.43
16 100
8823[0.0691]
93.4
8823
334
0.843
R1 = 0.0611
wR2 = 0.1027
R1 = 0.1426
wR2 = 0.1160
b/1
g/1
V/A3
Z
Dcalc/Mg mꢀ3
Absorption coefficient/mmꢀ1
F(000)
Y range for data collection
No. of reflections collected
No. of reflections unique [Rint
]
Completeness to Y = 25.00 (%)
No. of data
No. of parameters
Goodness of fit on F2
Final R indices [I > 2s(I)]
Final R indices (all data)
84%). Characterization data: Anal. calc. for C15H19FeNO2PtS:
C, 34.10; H, 3.62; N, 2.65; S, 6.07% found: C, 33.95; H, 3.7; N,
2.5; and S, 5.95%. IR data: 1575 cmꢀ1 {n(>CQN–)}. 1H-NMR
data (in CDCl3):8 8.02 (s, 1H, 3JH–Pt = 98, –CHQN–), 4.17 (s,
5H, C5H5), 4.56 (t, 1H, 3J = 1.5 H3), 4.62 (d, 1H, J = 1.5, H4),
diffractometer with image plate detector. Unit-cell para-
meters were determined from 5541 reflections (in the range
31 o Y o 311) for 4b or from 37 reflections for 5b and refined
by least-squares method. Intensities were collected with a
graphite monochromatized Mo-Ka radiation.
For 4b the number of reflections measured in the range 2.841
r Y r 29.931 was 6864 of which 3622 were non-equivalent by
symmetry {Rint(on I) = 0.064} and 3449 reflections were
assumed as observed applying the condition I > 2s(I). For
5b, 16 100 reflections were collected in the range 2.651 r Y r
32.431 of which 8823 were non-equivalent by symmetry {Rint
(on I) = 0.069} and the number of reflections assumed as
observed was 4053. Lorentz-polarization and absorption
corrections were made.
3
4.90 (t, 1H, J = 1.5, H5), 4.4–4.2 and 4.25–3.80 (2br m, 4H,
–CH2–CH2–), 3.42 [s, 3H, Me(dmso)], and 3.38 [s, 3H, Me
(dmso)].
Electrochemical studies. Electrochemical data for the com-
pounds under study were obtained by cyclic voltammetry
under nitrogen at 298 K using CH3CN (HPLC-grade) as
solvent, (Bu4N)[PF6] (0.1 M) as supporting electrolyte and
an M263A potentiostat from EG&G instruments. The mea-
sured potentials (E) were referenced to an Ag/AgNO3 (0.1 M
in CH3CN) electrode separated from the solution by a medium-
porosity fritted disk. A platinum wire auxiliary electrode was
used in conjunction with a platinum disk working Tacussel
EDI rotatory electrode (3.14 mm2). Cyclic voltammograms of
ferrocene were recorded before and after each sample to
ensure the stability of the Ag/AgNO3 electrode. Cyclic vol-
tammograms of freshly prepared solutions (10ꢀ3 M) of the
samples were run, and average values of the measured poten-
tials were referenced to ferrocene [E(Fc)] which was used as
internal reference. In all the experiments, the cyclic voltammo-
grams were registered using scan speeds varying from u =
The structure was solved by Direct methods, using
SHELXS computer program30 and refined by full-matrix
least-squares method with the SHELX97 computer
program31 using 6864 reflections (for 4b) and 16100 (for 5b)
(very negative intensities were not assumed). The function
P
minimized was
wJFo|2 ꢀ |Fc|2|2, where w = [s2(I) +
(0.0928P)2 + 4.8976P]ꢀ1 (for 4b) and w = [s2(I) +
(0.0201P)2]ꢀ1 (for 5b) and P = (|Fo| + 2|Fc|2)/3; f, f0 and f0 0
were obtained from the literature.32 All hydrogen atoms were
computed and refined using a riding model with an isotropic
temperature factor equal to 1.2 times the equivalent tempera-
ture factor of the atom to which is linked. The final R(on F)
factor was 0.054 or 0.0611 (for 4b and 5b, respectively), wR(on
|F|2) = 0.149 (for 4b) and 0.1027 (for 5b) and goodness of fit
= 1.214 or 0.843 (for 4b and 5b, respectively) for all the
observed reflections.
10 mV sꢀ1 to 100 mV sꢀ1
.
Crystallographyw. A prismatic crystal of 4b or 5b (sizes
in Table 4), was selected and mounted on a MAR345
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
684 | New J. Chem., 2010, 34, 676–685

View Article Online
16 L. Ding, Y. J. Wu and D. P. Zhou, Polyhedron, 1998, 17,
1725–1728.
Acknowledgements
17 (a) F. D. Rochon and M. Fakhfakh, Inorg. Chim. Acta, 2009, 362,
1455–1466; (b) F. D. Rochon and M. Fakhfakh, Inorg. Chim. Acta,
2009, 362, 458–470.
We are grateful to the Ministerio de Educacion y Ciencia of
´
Spain for financial support (grant: CTQ2009-11501) and
Feder funds.
18 For reviews on 195Pt-NMR: (a) B. M. Still, P. G. A. Kumar,
J. R. Aldrich-Wright and W. S. Price, Chem. Soc. Rev., 2007, 36,
665–686; (b) J. R. L. Priqueler, I. S. Butler and F. D. Rochon,
Appl. Spectrosc. Rev., 2006, 41, 185–226.
References
19 C. Lo
´
pez, R. Bosque, S. Pe
´
rez, A. Roig, E. Molins, X. Solans and
a, J. Organomet. Chem., 2006, 691, 475–484.
1 Comprehensive Organometallic Chemistry-III, ed. R. B. Crabtree
and D. M. P. Mingos, Elsevier, Oxford, UK, 2007, vol. 8,
pp. 145–194, 254–276 and 508–538.
2 For recent advances in platinacycles with (C, X) (X = N, S or P)
ligands: (a) P. Shao, Y. Li and W. Sun, Organometallics, 2008, 27,
2743–2749; (b) B. Ma, P. I. Djurovich, M. Yousufuddin, R. Bau
and M. E. Thompson, J. Phys. Chem. C, 2008, 112, 8022–8031;
(c) S. Develay and J. A. G. Williams, Dalton Trans., 2008,
4562–4564.
3 For reviews on this area: (a) M. Albrecht and G. van Koten,
Angew. Chem., Int. Ed., 2001, 40, 3750–3781; (b) H. Nishiyama,
Chem. Soc. Rev., 2007, 36, 1133–1141.
4 (a) Ferrocenes. Homogeneous Catalysis, Organic Synthesis. Materials
Science, ed. A. Togni and T. Hayashi, VCH, Weinheim (Germany),
1995; (b) Ferrocene. Ligands, Materials and Biomolecules, ed.
P. Stepnicka, Wiley, Weinheim (Germany), 2008.
M. Font-Bardı
´
20 T. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58,
380–388.
21 A. I. Kitaigorodskii, Molecular Crystals and Molecules, Academic
Press, London, UK, 1973.
22 E. R. Brown and J. R. Sandifer, Physical Methods in Chemistry,
Electrochemical Methods, ed. B. W. Rossiter and H. Hamilton,
Wiley, New York, USA, 1986, ch. 4, vol. 4.
23 R. Bosque, C. Lo
141–145.
24 A. D. Ryabov, G. M. Kazankov, I. M. Panyashkina,
O. V. Grozovsky, O. G. Dyachenko, V. A. Polyakov and
L. G. Kuzmina, J. Chem. Soc., Dalton Trans., 1997, 4385–4392.
25 F. Stoddart, in Comprehensive Organic Chemistry. The Synthesis
and Reactions of Organic Compounds, ed. D. Barton and
W. D. Ollis, Pergamon, Oxford, UK, 1979, ch. 4, vol. 1,
pp. 577–598.
´
pez and J. Sales, Inorg. Chim. Acta, 1996, 244,
5 (a) C. Lo
Eur. J. Inorg. Chem., 2006, 3974–3984; (b) S. Pe
A. Caubet, R. Bosque, X. Solans and M. Font-Bardı
Inorg. Chem., 2008, 1599–1612.
6 C. Lopez, A. Caubet, X. Solans and M. Font-Bardı
J. Organomet. Chem., 2000, 598, 87–102.
´
pez, A. Caubet, S. Pe
´
rez, X. Solans and M. Font-Bardı
rez, C. Lopez,
a, Eur. J.
´
a,
´
´
26 P. R. R. Ranatunge-Bandarage, B. H. Robinson and J. Simpson,
Organometallics, 1994, 13, 500–510.
´
27 Recent contributions on this field: (a) Y. Urano, D. Asanuma,
Y. Hama, Y. Koyama, T. Barrett, M. Kamiya, T. Nagano,
T. Watanabe, A. Hasegawa, P. L. Choyke and H. Kobayashi,
Nat. Med., 2009, 15, 104–109; (b) S. Xu, Y. Luo, R. Graeser,
A. Warnecke, F. Kratz, P. Hauff, K. Licha and R. Haag, Bioorg.
Med. Chem. Lett., 2009, 19, 1030–1034; (c) E. S. Lee, Z. Gao and
Y. H. Bae, J. Controlled Release, 2008, 132, 164–170;
(d) M. Dalmau, S. Lim and S.-W. Wang, Nano Lett., 2009, 9,
160–166.
´
´
a,
7 J. H. Price, A. N. Williamson R.F. Schramm and B. B. Wayland,
Inorg. Chem., 1972, 11, 1280–1284.
8 (a) C. Lo
Commun., 2005, 8, 631–634; (b) C. Lo
X. Solans and M. Font-Bardıa, Chem. Commun., 2004, 540–541.
´
pez, X. Solans and M. Font-Bardı
´
a, Inorg. Chem.
´
pez, A. Caubet, S. Pe
´
rez,
´
9 L. Ding, D. P. Zhou and Y. J. Wu, Polyhedron, 1998, 17,
2511–2516.
28 (a) M. Galanski, C. Baumgartner, K. Meelich, V. A. Arion,
M. Fremuth, M. A. Jakupee, P. Schluga, C. G. Hartinger,
N. Graf v. Keyserlingk and B. K. Keppler, Inorg. Chim. Acta,
2004, 357, 3237–3244; (b) S. van Rijt, S. van Zutphen, H. des Dulk,
J. Brouwer and J. Reedijk, Inorg. Chim. Acta, 2006, 359,
4125–4129.
29 D. D. Perrin, W. L. F. Armarego and D. L. Perrin, Purification of
Laboratory Chemicals, Butterworth-Heinemann, Oxford, UK, 3rd
edn, 1988.
10 S. Pe
Bardı
224–236.
11 (a) R. Bosque, C. Lo
J. Chem. Soc., Dalton Trans., 1994, 735–745; (b) C. Lo
R. Bosque, X. Solans and M. Font-Bardıa, New J. Chem., 1998,
22, 977–982.
´
rez, C. Lo
´
pez, A. Caubet, R. Bosque, X. Solans, M. Font-
a, A. Roig and E. Molins, Organometallics, 2004, 23,
´
´
pez, J. Sales, X. Solans and M. Font-Bardı
´
a,
´
pez,
´
12 K. Nakamoto, Infrared and Raman Spectra of Inorganic and
Coordination Compounds, Wiley, New York, USA, 5th edn, 1997.
13 G. B. Deacon and J. H. S. Green, Spectrochim. Acta, 1964, 24,
845–852.
14 H.-U. Gremlich, Handbook of Analytical Techniques, 2001,
465–507.
´
15 J. Vicente, J. A. Abad, A. D. Frankland and M. C. Ramırez de
30 G. M. Sheldrich, SHELXS: A computer program for determination
of crystal structures, University of Gottingen, Germany, 1997.
¨
31 G. M. Sheldrich, SHELX97: A computer program for determination
of crystal structures, University of Gottingen, Germany, 1997.
¨
32 International Tables of X-Ray Crystallography, Kynoch Press,
Birmingham, UK, vol. IV, pp. 99–100 and 149.
Arellano, Chem.–Eur. J., 1999, 5, 3066–3075.
ꢁc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010
New J. Chem., 2010, 34, 676–685 | 685