Relationships between basicity, redox behaviour of ferrocenylamines and their reactivity with Pt[II] compounds

Article abstract of DOI:10.1016/S0022-328X(98)00603-2

pK_b values for the ferrocenylamines, [(η-C₅H₄(CH₂)_xNH₂)FeC p] x=1, 2, 3; [(η-C₅H₄CH₂NHR)FeCp] R=Me, 4, Ph, 5; {[η-C₅H₄CHR′NR₂]FeCp} R′/R=H/Me, 6, R′/R=H/Ph, 7, Me/Me, 8;[{η-C₅H₄CHRNMe₂)₂Fe] R=H 9, Me 10; [{1,2η-C₅H₃(CH₂NMe₂)(PPh₂)}FeCp] 11, {1,2η-C₅H₃[CH(Me)NMe₂](PR₂}}Fe[η-C₅H₄(PPh₂)_n] n=0, R=ⁱPr 12, Ph 13, n=1, R=Me 14, are correlated with inductive, neighbouring group and steric effects. Corresponding salts have been synthesised. The pK_b has a marked influence on their chemistry. Protonation competes with complexation but cis-PtCl₂L₂ L=1-3, 5, 7, and cis-Pt(N-N)Cl₂ L=8, 9, have been characterised. Two reversible couples [Fc⁺A/FcA], [Fc⁺AH⁺/FcAH⁺] (A=amine function) and an irreversible oxidation/protonation of A are linked by a EECE mechanism, but potentials for the first two are independent of the amine and similar to ferrocene. Nucleophilic attack by ferrocenylamines at the nitrile, protonation and ligand substitution are all observed with cis-[PtCl₂(NCR)₂].

Full text of DOI:10.1016/S0022-328X(98)00603-2

Journal of Organometallic Chemistry 564 (1998) 125–131
Relationships between basicity, redox behaviour of ferrocenylamines
and their reactivity with Pt[II] compounds
Noel W. Duffy, Jacqui Harper, P. Ramani, R. Ranatunge-Bandarage, Brian H. Robinson *,
Jim Simpson
Department of Chemistry, Uni6ersity of Otago, PO Box 56, Dunedin, New Zealand
Received 20 February 1998
Abstract
pKb values for the ferrocenylamines, [(p-C5H4(CH2)xNH2)FeCp] x=1, 2, 3; [(p-C5H4CH2NHR)FeCp] R=Me, 4, Ph, 5;
{[p-C5H4CHR%NR2]FeCp} R%/R=H/Me, 6, R%/R=H/Ph, 7, Me/Me, 8;[{p-C5H4CHRNMe2)2Fe] R=H 9, Me 10; [{1,2p-
C5H3(CH2NMe2)(PPh2)}FeCp] 11, {1,2p-C5H3[CH(Me)NMe2](PR2}}Fe[p-C5H4(PPh2)n] n=0, R=iPr 12, Ph 13, n=1, R=Me
14, are correlated with inductive, neighbouring group and steric effects. Corresponding salts have been synthesised. The pKb has
a marked influence on their chemistry. Protonation competes with complexation but cis-PtCl2L2 L=1–3, 5, 7, and cis-Pt(N–
N)Cl2 L=8, 9, have been characterised. Two reversible couples [Fc+A/FcA], [Fc+AH+/FcAH+] (A=amine function) and an
irreversible oxidation/protonation of A are linked by a EECE mechanism, but potentials for the first two are independent of the
amine and similar to ferrocene. Nucleophilic attack by ferrocenylamines at the nitrile, protonation and ligand substitution are all
observed with cis-[PtCl2(NCR)2]. © 1998 Elsevier Science S.A. All rights reserved.
Keywords: Ferrocenylamines; Platinum compounds; Basicity studies; Redox studies
1. Introduction
of ferrocenylamines and investigate whether this
parameter can be correlated with other features of their
chemistry. Ferrocenium salts are potential radiation
sensitisers [6] and consequently it is important to know
if the redox activity in the hypoxic cell could be predicted
from the pKb; this is explored in this paper. Recently, we
reported [7–9] a series of cycloplatinated ferroceny-
lamine complexes with promising cytotoxicity but, like
Neuse and co-workers [10], we found [11] that the
propensity to protonation posed difficulties in the syn-
thesis of cisplatin analogues with ferrocenylamine lig-
ands. In this paper, we describe the competition between
protonation and successful complex formation. Nucle-
ophilic attack by amines on a benzonitrile ligand acti-
vated by coordination to Pt[II] to give amidines and
amidates is well-documented [12]; the involvement of
ferrocenylamines in this type of reaction provides a final
illustration of the relationship between basicity and
reactivity.
An assessment of the electronic effects of the ferro-
cenyl moiety on the acidity and basicity of functional
groups has been the subject of many studies [1]. A recent
determination [2] of pKb values for [(p-C5H4NH2)FeCp],
[p-(m-C6H4NH2)C5H4]FeCp and [p-(p-C6H4NH2)C5H4]
FeCp of 8.81, 9.66 and 9.85 respectively, indicated that
these ferrocenylamines were weak Bronsted bases. How-
ever, it became clear to us from our work on biologically
active ferrocenylamine derivatives that many with N-
alkyl substituents behaved as stronger bases. Ferrocene
compounds with nitrogen functionality have shown an-
tiproliferative properties against a variety of carcinomas
and tumours [3,4] but we found [5] that the toxicology
of ferrocenylamines is strongly dependent on their basic-
ity. This encouraged us to determine the pKb for a range
* Corresponding author.
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(98)00603-2

126
N.W. Duffy et al. / Journal of Organometallic Chemistry 564 (1998) 125–131
2. Results and discussion
9.39). In general, the N-methyl ferrocenylamines (2–4,
6, 8–10) have lower basicity than their organic ana-
logues [14]. Interpolation of a saturated carbon group
between the amine function and the ferrocene ring
removes the neighbouring-group effect and the pKb
order is a function of the inductive capability of the
2.1. Basicity of ferrocenylamines
Titration of partially neutralised aqueous acidic solu-
tions provided comparative pKb data for the ferroceny-
lamines 1–14, a series which provided diverse electronic
and/or steric requirements. The pKb values decrease in
order: 5\7\1\14\12\13\9ꢀ11\6ꢀ2\3\
10\4\8 (Table). The value for 1 is in good agree-
ment with that determined by a different method [2]
(Table 1).
Table 1
pKb and E1/2 data for ferrocenylamines
a,b
1/2
Compound Substituents on Cp ring
pKb
E
Three factors may influence the trend in pKb. First,
the ferrocenyl group is inductively a net electron donor
[1]. Second, when the non-protonated amine is directly
bound to the cyclopentadienyl ring it is resonance-sta-
bilised by delocalisation of the nitrogen lone pair [13].
Third, tertiary amines should have larger pKb values
than primary and secondary amines because of in-
creased steric hindrance and reduced solvent stabiliza-
tion of the cation [14]. These three factors would then
impact to varying degrees on the relative thermody-
namic stabilization of the free and quaternized amine.
FcNH3+ is not resonance-stabilized but the free base is,
with the result that the basicity of 1 is two to three
orders of magnitude less than the other ferroceny-
lamines, except 5 and 7. The inductive effect of the
ferrocenyl substituent results in a lower pKb for 1
compared to the organic analogue, PhNH2 (pKb=
1
2
3
4
5
6
7
8
NH2
CH2NH2
–
–
–
–
–
–
–
–
8.24
5.12
4.95
4.65
9.73
5.17
9.54
4.30
5.53
4.72
5.50
6.38
5.62
6.73
–
0.47
0.49
0.48
0.48
0.48
0.49
0.47
0.47
0.46
0.57
0.58
–
CH2CH2NH2
CH2NH(Me)
CH2NH(Ph)
CH2NMe2
CH2NPh2
CH(Me)NMe2
CH2NMe2
CH(Me)NMe2 1-CH(Me)NMe2
CH2NMe2 2-PPh2
CH(Me)NMe2 2-PPh2
CH(Me)NMe2 2-PiPr2
CH(Me)NMe2 2-PPh2
1%-PPh2
9
1%-CH2NMe2
10
11
12
13
14
0.64
a Couple A: reference decamethylferrocene; potential for ferrocene
under the same conditions, 0.48 V.
b In acetone; square wave data at 20°C on Pt.

N.W. Duffy et al. / Journal of Organometallic Chemistry 564 (1998) 125–131
127
nitrogen and h-C- substituents, and steric effects. Thus,
the inductive effect of the ferrocenyl group is negated
and 5 and 7 have a pKb slightly higher than their aniline
counterparts. The introduction of one N-methyl sub-
stituent increases the basicity (2“4) but there is a
decrease with a second (6) due to decreased stabilisa-
tion of the sterically hindered cation 6H+. Although
this trend is identical to that in alkyl amines it is
accentuated in the ferrocenylamines; compare the pKb
MeNH2 (3.36), MeNHMe (3.29), MeNMe2 (4.28). It is
interesting that the compounds with an NMe2 on each
ring (9, 10), are less basic than their counterparts with
one (6, 8) and the introduction of a C-methyl leads to
the most basic ferrocenylamine (8). This is analogous to
the trends in pKb for arylamines. The addition of an
ortho PR2 substituent markedly reduces the basicity.
This is particularly obvious for the CH(Me)NMe2
derivatives where there is two orders of magnitude
reduction on the introduction of one PPh2 group (12)
and a further 10-fold decrease with a PPh2 in the other
Cp ring (14). The fact that the pKb is also dependent on
the nature of the phosphorus substituent (cf 12 and 13)
could be interpreted as suggesting that the trend with
PR2 substitution is an inductive effect. However, PR2 is
not an electron-withdrawing group and the evidence
above suggests that inductive effects are not transmitted
to the N terminus, so the decreased basicity is most
likely due to destabilisation of the cation. Steric effects
would seem to be a significant factor in determining the
basicity of ferrocenylamines.
Fig. 1. Cyclic voltammogram of 6 in acetone: Pt, 20°C, TMEAP 0.1
M, scan rate 400 mV s−1
.
noted with biferrocenylamines [15,16]. We therefore
carried out an electrochemical study of both the ferro-
cenylamines and their Pt[II] complexes (vide infra) to
see if the basicity impinged on the redox properties.
Ferrocenylamines 2–14 show a single chemically and
electrochemically reversible, one-electron, couple A in
their cyclic voltammograms or square wave plots when
the initiating sweep range is from 0.0 V out to E1/2
+
0.3 V (Figs. 1–3) due to the characteristic oxidation of
the ferrocene redox centre. E1/2 values for A (Table)
were independent of the solvent. What was unexpected
was that all non-protonated ferrocenylamines without
PR2 groups were oxidised at approximately the same
potential as ferrocene; normally, E+/0 [Fc] is very sensi-
tive to the substituent on the Cp ring [17]. Furthermore,
the potential was not influenced by the stereochemistry
as E1/2 [R-10=S-10]. The ground states for the ferroce-
nium ion (2E2g) and ferrocene both have the electron
density largely on the iron [18] and their energy may
not vary greatly once a CH2 isolates the amine function
from the Cp ring. Nevertheless, it is surprising that
difference in solvation energies between the neutral and
oxidised species do not have a greater impact on the
thermodynamic parameters. It was anticipated that 1
would be the most easily oxidised but strong adsorption
Orange salts (with either Cl− or PtCl4 as counterion)
were characterised in order to confirm the site of proto-
nation, investigate possible steric effects and to study
the effect of protonation on the redox properties. Typi-
cally, protonation of the amine function is manifested
by a broad w(N–H) band around 2600 cm−1 and a
1
shift downfield of all H and 13C resonances compared
to the free ligand-these are very useful diagnostic tools.
In our experience, many samples of purported com-
plexes of ferrocenylamines with metal ions have these
features in ‘pure’ samples, an important observation as
biological activity can be masked by protonated impu-
rities. The downfield NMR shift on protonation is not
as large as that occurring on complexation to a metal
ion. Consequently, the position of the NMe2 resonance
for 6, for example, is diagnostic of the free ligand
(l=2.1–2.2), protonation (l=2.4–2.7) and complexa-
tion (l=2.8–3.2).
2.2. Redox properties
An intriguing observation was made during our stud-
ies on Pt[II]-complexation and protonation of ferro-
cenylamines pKbB5 that many products appeared
from NMR results to be contaminated by traces of
paramagnetic ferrocenium salts; this has also been
Fig. 2. Scan 1: cyclic voltammogram of 8 in CH2Cl2; Pt, 20°C,
TBAPF6 0.1 M, scan rate 200 mV s−1. Scan 2: 8H+ under the same
conditions.

128
N.W. Duffy et al. / Journal of Organometallic Chemistry 564 (1998) 125–131
protonated species at the electrode surface. The shift in
potential A“C is simply a consequence of the higher
coulombic charge on the salt and consequently there is
a linear correlation between the electrostatic charge and
the difference in potential; that is, the number of NMe2
groups per redox centre (cf 6 with 9, 8 with 10). E+/0 for
5 has two components on the anodic scan but only Cc
and not Ac on the reverse scan at scan rates \200 mV
s−1 (Fig. 3). Thus, protonation rather than radical
coupling dominates. In contrast, multiple cathodic pro-
cesses are seen in voltammograms for 7, including a
species reduced at 0.1 V, which may arise from radical
coupling involving the N–Ph group. Additional processes
close to 1.5 V are seen in the phosphine derivatives,
12–14, associated with oxidation of the PPh2 group and
at potentials comparable to dppf [20], which obscure E
for these compounds.
Fig. 3. Cyclic voltammogram of 7 in CH2Cl2: Pt, 20°C, TBAPF6 0.1
M, scan rate 200 mV s−1
.
on the electrode gave non-reproducible results. A PPh2
substituent has a marked effect on E1/2 especially for those
derived from 8 (12–14) with the ferrocenyl redox centre
becoming more difficult to oxidise; this PPh2 substituent
effect is cumulative (cf 12 and 14). This trend could be
ascribed to the y-acceptor properties of the phosphine
(this is in contrast to the pKb trend, vide supra) but could
equally be due to decreased solvation of the cation and
steric hindrance associated with electron transfer at the
electrode surface.
When the potential range is extended beyond 1.1 V an
additional irreversible wave E is observed on the first
anodic scan with at least a one-electron transfer and
approximately 1.0 V positive of A (Figs. 1–3). Anodic
oxidations of aliphatic amines typically involve the
formation of the amine or ammonium salt which arise
from hydrogen abstraction from the solvent by the
generated cation radical whereas radical coupling is
favoured by aromatic amines [19]. A similar process can
be assigned to E for 2–4, 6, 8–10 and its potential is only
dependent on the N-substituent.
[(p−C5H4CH2NRR%)FeCp]
A
l[(p−C5H4CH2NRR%)Fe+Cp]
E
[[(p−C5H4CH2HN+RR%)Fe+Cp]
“C [(p−C5H4CH2HN+RR%)Fe+Cp]
(2)
In summary, the electrochemistry of ferrocenylamines
arises from an EECE mechanism involving both the
protonated and non-protonated forms (Eq 2). Controlled
potential electrolysis of a ferrocenylamines at E offers a
route to new ferrocene derivatives; this is currently being
explored.
2.3. Ferrocenylamine-platinum(II) complexes
Syntheses of cis-PtL2Cl2, from K2PtCl4, or the labile
precursors Pt(COD)Cl2 and cis-Pt(dmso)2Cl2, were un-
successful when L=4, 6, 7. In-situ monitoring of reaction
progress by 195Pt-NMR suggested that small amounts of
a cis-PtL2Cl2 complex were formed in the initial stages
of the reaction but protonation of the amine ligand
effectively competed with complexation, even in the
presence of K2CO3. In contrast, good yields of cis-
[PtL2Cl2] L=1–3, 5, cis-[PtLCl2] L=9, 10, and the
reported [21] cis-[Pt(11)Cl2] and cis-[Pt(14)Cl2], were
obtained by stirring a mixture of K2PtCl4 and the
appropriate ligand in CH2Cl2/H2O at r.t. Two related
compounds, Pt[Fc–CH(Me)NH2]2Cl2, Pt[FcCH2NH2–
CH2]2Cl2 have been prepared [10] from K2PtCl4 in
ethanol.
The dichotomy in behaviour, protonation versus com-
plexation, correlates with the ligand pKb values. Primary
amines and N-aryl secondary amines readily coordinate
to Pt[II]. With the, the most basic ferrocenylamine
ligands-secondary 4 and tertiary-N-methyl 8- the partic-
ular thermodynamic stability of the protonated species
leads to preferential salt formation. Despite the decreased
basicity of 6 the poor nucleophilicity of tertiary amines
[(p-C5H4CH2NRR%)
−e (E)
+
Fe+Cp l [(p-C5H4CH2N+ RR%)Fe Cp]
’
solv (H)
l [(p-C5H4CH2NH+RR%)Fe+Cp]
(1)
This assignment was confirmed by the isolation of
8H2+ from the bulk electrolysis of 8. An additional
cathodic wave Cc appears on the reverse sweep with the
anodic component Ca on subsequent scans. Couple C,
approximately 0.2 V positive of A is assigned to a
reversible one-electron couple involving oxidation of the
protonated ferrocenylamine to the ferrocenium species;
this is the only couple seen in the solution electrochem-
istry of the salts (Fig. 2). In the potential range 0.00–1.7
V the relative currents of the couples A and C are
determined by the scan rate but are essentially indepen-
dent of the solvent and basicity of the ferrocenylamine.
Cc dominates on the reverse scan at fast scan rates (ꢁ200
mV s−1) because of the higher concentration of the

N.W. Duffy et al. / Journal of Organometallic Chemistry 564 (1998) 125–131
129
1
towards complexation drives the equilibrium towards
protonation. The additional thermodynamic stabiliza-
tion resulting from chelate formation provides the nec-
essary impetus for complexation to dominate with the
other tertiary amines 9–14. Work with other N-alkyl
ferrocenylamines substantiates these conclusions [22].
With 7, the initially formed cis-[PtL2Cl2] rapidly de-
composed and neither the protonated nor complexed
product was isolated.
consistent with a cis-2N donor set. The H- and 13C-
NMR indicated a set of two C6H5, one p-C10H9, CH2
group and at least two CN functions. The relative
intensities of w(CꢀN) (2275 cm−1) and w(CꢁN) (1711
cm−1) (no w(N–H) of 5H+ were observed) and w(Pt–
Cl) modes in the infrared spectra depended on the
preparation. Despite the variable spectroscopic data the
analytical data were similar for each preparation and
consistent with either [(5)(PhCN)PtCl2] 15a in which 5
has simply substituted one nitrile ligand, or a four-
membered chelated amidine {[PhC(NPhCH2Fc)NH]
PtCl2}15b resulting from nucleophilic attack on the
coordinated nitrile; we suggest that the yellow solids are
mixtures of 15a and 15b.
All of the Pt[II] complexes were yellow-crystalline
solids, insoluble in water, but soluble in common or-
ganic solvents. The complex with 5 as ligand was
usually isolated as a purple compound reminiscent of
the so called ‘platinum blues’ due to the incorporation
of a dye which is often a decomposition product of 5
without the intervention of Pt[II]. Two w(Pt–Cl) bands
are observed for the complexes of 1, 3, 5, 11 which
confirms a cis-configuration but one w(Pt–Cl) band for
2 and the chelating ligands 13, 14. This is not unusual;
for example, cis-Pt(NH3)2Cl2 [23] and the red form of
Pt(bipy)Cl2 [24] have a single Pt–Cl band The 1H-
NMR spectra of the compounds with a N,N donor set
often displayed paramagnetic broadening of the reso-
nances, but the observed 195Pt-NMR were consistent
with the proposed donor set [25]. Cyclic voltammetry of
the Pt[II] complexes showed that the potential of their
reversible couples equivalent to A were :0.2 V less
than the free ligand, a result which could be interpreted
as a decrease in electron density on the ferrocenyl
moiety on complexation, but more likely to be stabilisa-
tion of the charged ferrocenium species. If there is
uncomplexed oxidised ferrocenylamine in a reaction
solution it will oxidise the Pt[II] complex, which could
account for the paramagnetic impurities noted in many
samples of the complexes, but it is not clear what
initiates the oxidation.
3. Conclusion
Ferrocenylamines display a wide range of basicity,
but not redox potential, which impacts significantly on
their chemistry. Potential data for any ferrocenylamine
needs to be carefully examined to ensure that protona-
tion has not taken place prior to the i/V scan as this
will occur in the double layer before it is obvious in the
bulk solution. The propensity for N-alkyl ferroceny-
lamines to protonate and oxidise to the ferrocenium
analogues needs to be considered when evaluating their
biological activity. Interestingly, it is the N-aryl ferro-
cenylamines which show the most biological activity [5]
suggesting that protonation of the more basic N-alkyl
ferrocenylamines may be inhibiting their in vivo activ-
ity. Ferrocenylamines and their Pt[II] complexes un-
dergo partial oxidation of the ferrocenyl redox centre in
solution, by a mechanism which is not understood,
which makes an NMR study of their interaction with
DNA and proteins difficult. Furthermore, recent cell
culture work has shown that oxidised water-soluble
ferrocenylamines are not active against hypoxic cell
lines [26]. Future work will concentrate on sugar ferro-
cenylamines which can be targeted to specific cell
functions.
2.4. Reaction of ferrocenylamines with (PhCN)2PtCl2
1
A H-NMR time-lapse study of the reaction of cis-
(PhCN)2PtCl2 with 6 in d6-benzene showed a singlet
NMe2 peak at 2.57 ppm at T=0 which indicated
immediate protonation. A broad resonance due to co-
ordinated NMe2 at 2.8–3.2 ppm grew in intensity and a
concurrent time-lapse 195Pt-NMR study showed that
after 7 h the only Pt[II] product was one with a cis-2N
coordination environment (l 2170) consistent with a
formulation cis-[(6)(PhCN)PtCl2]. Similar results were
obtained with cis-(MeCN)2PtCl2. Analytically pure
samples of cis-[(6)(CN)PtCl2] were not obtained be-
cause the salts were difficult to remove. Reaction of
cis-(PhCN)2PtCl2 with the least basic secondary amine
5 gave yellow crystalline solids as the major products
and an orange insoluble powder. Time-lapse 195Pt-
NMR showed the formation of two solution products
and the chemical shifts, 2164 and 2248 ppm, were
4. Experimental
Most of the ferrocenylamine compounds have acrid
odours, therefore all synthetic work was performed in a
fumehood. All reactions were carried out under an
atmosphere of dry nitrogen. Solvents were purified and
dried by standard methods and the ferrocenylamines
and Pt[II] precursors prepared by literature methods:
1-14 [27–31], cis-Pt(DMSO)2Cl2 [32], Pt(COD)Cl2 [33],
and Pt(PhCN)2Cl2 [12]. NMR, IR and UV–vis spectra
were recorded on a Varian 300 MHz VXR or 200 MHz
Gemini, Digilab FTIR and Perkin Elmer Lambda 9
spectrometers respectively. 195Pt-NMR were referenced

130
N.W. Duffy et al. / Journal of Organometallic Chemistry 564 (1998) 125–131
to Na2PtCl6. Microanalyses were carried out by the
Campbell Microanalytical Laboratory, University of
Otago Electrochemical measurements were performed
with a three-electrode cell using an EG & G PAR 273a
at scan rates 0.05–10 V s−1. A polished Pt electrode was
used; all potentials are referenced against decamethylfer-
rocene uncorrected for junction potentials, the support-
ing electrolyte (TEAP) 0.1 M and the substrate ca.
1×10−3 M.
CH6
CH3). 9H+ ·Cl− mp. 200°C. C19H30ClFeNP re-
quires: C, 64.74; H, 5.87; N, 3.02. Found: C, 64.68; H,
1
5.82; N, 2.85. H-NMR: 2.34 (3H, NCH3), 2.62 (3H,
NCH3), 3.98 (h-C5H5), 4.23, 4.65, 5.30 (h-C5H5), 7.31–
7.65 (10H, C6H5), 4.30–4.55 (CH2N). 10H+ ·Cl−
C38H37ClFeNP2 requires: C, 68.94; H, 5.64; N, 2.11.
1
Found: C, 69.05; H, 6.02; N, 2.21. H-NMR: 1.85 (3H,
CH3
NCH3), 3.49–4.58 (8H, p-C5H5), p-C5H3; CH
6.87–7.66 (20H, Ph).
6
CH, J=6.99 Hz), 2.18 (3H, NCH3), 2.21 (3H,
6
CH3),
4.1. Determination of Kb 6alue of ferrocenyl ligands
4.3. Preparations of cis-Pt(L2)Cl2, and cis-Pt(P-N)Cl2
A mixture of 0.1 M NaCl (5 cm3) in water and acetone
(8 cm3) was added to a cell equipped with a pH electrode
and stirrer and the required amount of ferrocenylamine
was then added to give a substrate concentration of 0.01
mol dm−3. An inert atmosphere was maintained in the
cell for the duration of the experiment. The mixture was
titrated with freshly prepared 0.01 M HCl solution with
steps of 0.1 cm3 between each addition and the pH was
recorded after each addition. All measurements were
obtained at 20°C and the solstat EPM-900 pH meter was
calibrated using appropriate buffer solutions.
Reactions of K2PtCl4 (1 mmol) in water (1 cm3) with
the ligands L (1 or 2 mmol as appropriate) in CH2Cl2 (10
cm3) were carried out with vigorous stirring. The red
−
colour of PtCl24 gradually fades from the aqueous layer
as reaction proceeds and when the aqueous phases
became clear (3–20 h) the organic phases were separated
and filtered. Some products precipitated while others
were soluble in CH2Cl2. The precipitates were washed
with CH2Cl2 and dried in vacuo to give product. For the
isolation of CH2Cl2 soluble products, hexane was added
to the filtrates at 0°C. L=1 Yield, 49%. M.p. 178–
180°C. Calcd. for C20Cl2Fe2H22N2Pt: C, 35.96; H, 3.32;
N, 4.19%. Found: C, 35.59; H, 3.62; N, 3.95%. IR (KBr,
4.2. Salts of ferrocenylamines
Hydrochloride salts of the ferrocenylamines were pre-
cipitated by bubbling HCl gas through an ether solution
of the appropriate ligand; to prepare PtCl24− salts
K2PtCl4 was dissolved in a small quantity of water and
added to above solutions. (3H+)2 ·PtCl42−-. M.p. 222–
224°C with dec. Calcd. for C24Cl4Fe2H32N2Pt: C, 36.16;
H, 4.04; N, 3.51; Cl, 17.79%. Found: C, 36.06; H, 4.10;
N, 3.80; Cl, 16.55%. 1H-NMR (dmso): 2.60(2H,
cm−1): 3444 w(N–H); 1628 wd(N–H). (nujol) 324, 309
1
w(Pt-Cl). H-NMR (l, d6-dmso): 3.32 (m, p5-C5H4
6
Fe);
3.37 (s, p5-C5H5
6 Fe); 4.20–4.40 (bs, 2H, NH2). UV–vis
(dmso, umax): 448 (m=979). L=2 Yield, 32%. M.p.
250–252°C. Anal. Calcd. for C22Cl2Fe2H26N2Pt: C,
37.96; H, 3.76; N, 4.02; Cl, 10.19%. Found: C, 37.55; H,
3.63; N, 3.82; Cl, 9.89%. IR (KBr, cm−1): 3267 w(N–H)
;
1
1577 w(N–H) (nujol): 324 w(Pt–Cl). H-NMR (d, CDCl3):
3.33 (bs, 2H, CH2
CH2
6
CH2N), 3.00(2H, CH2
6
CH2N), 4.11(p-C5H4), 4.16(p-
6
N); 4.10 (s, 7H, p5-C5H5
). UV–vis (dmso, umax): 428
6 ); 4.29 (m, 2H,
C5H4). IR (KBr, cm−1): 2854, 2958 w(NH3+). Lm,
p5-C5H5
6
); 4.90 (bs, 2H, NH2
6
(dmso): 104. UV–vis (umax, nm, dmso): 419 (m=229).
(295). L=3 Yield, 46%. M.p. 220–222°C. Calcd. for
C24Cl2Fe2H30N2Pt: C, 39.80; H, 4.17; N, 3.86; Cl, 9.79%.
Found: C, 38.96; H, 4.17; N, 3.80; Cl, 9.70%. UV–vis
(CH2Cl2, nm): 422 (658). L=5 Yield, 15%. M.p. 148–
150°C. Calcd. for C34Cl2Fe2H34N2Pt: C, 48.14; H, 4.04;
N, 3.30; Cl, 8.36%. Found: C, 47.97; H, 3.97; N, 3.20;
Cl, 10.83, 10.68, 11.01%. IR (KBr, cm−1): 3194 (b) w(N–H)
(nujol): 334, 328 w(Pt–Cl). UV–vis (CH2Cl2, nm): 345
(3455); 522 (1244). L=11 Yield, 62%. M.p. 220°C.
Calcd. for C16Cl2FeH24N2Pt.2H2O: C, 31.91; H, 4.60; N,
4.65; Cl, 11.77%. Found: C, 31.74; H, 4.55; N, 4.09; Cl,
11.77%. IR (cm−1) 326 w(Pt–Cl). UV–vis (CH2Cl2, nm):
413 (493). L=9 Yield, 45%. M.p. 230–232°C. Calcd. for
C25Cl2FeH26NPPt: C, 43.31; H, 3.78; N, 2.02; P, 4.47%.
Found: C, 43.46; H, 3.95; N, 1.91; P, 4.58%. IR (cm−1):
−
(4H+)2 ·PtCl24
.
M.p. 186°C(dec). Calcd. for
C24Cl4Fe2H33N2Pt: C, 36.16; H, 4.05; N, 3.51%. Found:
C, 36.38; H, 4.04; N, 3.93%. 1H-NMR (dmso): 3.32 (3H,
NCH3), 3.92 (2H, CH2N), 4.22 (p-C5H5), 4.26, 4.39
(p-C5H4), Lm (dmso): 100. UV–vis (umax, nm, dmso): 426
(258). (5H+)2 ·PtCl24− C34H36Cl4 Fe2N2Pt requires: C,
44.52; H, 3.52; N, 3.05. Found: C, 44.63; H, 3.67; N, 3.14.
1
M.p. 183°C dec. H-NMR (dmso): 3.74 (2H, CH2N),
4.08–4.60 (9H, p-C5H5, p-C5H4), 7.20–7.40 (m, C6H5).
6H+ ·Cl- m.p.: 158–160°C. Calcd. for C13FeH18NCl: C,
55.85; H, 6.49; N, 5.01%. Found: C, 55.68; H, 6.47; N,
4.95%. IR (KBr, cm−1) 2368–2646 (s, br) w(N–H).
1H-NMR (l, CDCl3): 2.64 (s, 6H, NCH3
6
); 4.09 (s, 2H,
Fe); 4.32 and 4.40 (p-
Fe). 7H+ ·Cl− m.p. 208°C. Calcd. for
CH2
6
N); 4.20 (s, 5H, p-C5H5
6
1
C5H4
6
313, 302 w(Pt–Cl). H-NMR (CDCl3): 2.79 and 3.43
C14FeH19N.HCl: C, 57.27; H, 6.87; N, 4.77%. Found: C,
(2×bs, 3H, NCH3
6 6 ,
); 3.70–4.60 (m, 10H, p5-C5H3
57.20; H, 6.84; N, 4.73%. 1H-NMR (d, CDCl3): 1.89 (bs,
CH2N); 6.85–8.40 (m, 10H, P(C6H6 5 2
6
) ). 31P-NMR
3H, CHCH3
6
); 2.48 and 2.58 (2 x s, 6H, NCH3
6
); 4.20 (s,
(CDCl3): −7.35 (JPt–P=3941 Hz). UV–vis (CH2Cl2, u
5H, p-C5H5Fe); 4.28, 4.34 (p-C5H4
6
6
Fe); 4.46 (bs, 1H,
nm): 461 (535).

N.W. Duffy et al. / Journal of Organometallic Chemistry 564 (1998) 125–131
131
L.Y. Kuo, J.H. Toney, T.J. Marks, Inorg. Chim. Acta 152
(1988) 117.
4.4. Reaction of Pt(PhCN)2Cl2 with 5 and 6
[5] R. Mason, K. McGrouther, P.P.R. Ranatunge, B.H. Robinson,
J. Simpson, J. Appl. Organomet. Chem. submitted for
publication.
[6] A.M. Joy, D.M.L. Goodgame, I.J. Stratford, Int. J. Rad. Oncol.
Biol. Phys. 16 (1989) 1053.
[7] (a) M.J. Cleare, Dev. Pharmacol. 3 (1983) 59. (b) S.K. Carter,
Dev. Oncol. 17 (1984) 359. (c) J. Reedijk, Pure Appl. Chem. 59
(1987) 181.
[8] P.R.R. Ranatunge-Bandarage, B.H. Robinson, J. Simpson,
Organometallics 13 (1994) 500.
[9] P.R.R. Ranatunge-Bandarage, N.W. Duffy, S.M. Johnson, B.H.
Robinson, J. Simpson, Organometallics 13 (1994) 511.
[10] N.W. Duffy, C.J. McAdam, B.H. Robinson, J. Simpson, Inorg.
Chem 13 (1994) 511.
[11] E.W. Neuse, M.G. Meirim, N.F. Blom, Organometallics 7 (1988)
1562.
Pt(PhCN)2Cl2 (0.5 g, 1.08 mol) was stirred with 5
(0.63 g, 2.16 mmol) in 75 cm3 of methanol for 3 days at
r.t. The solvent was evaporated (water bath at r.t.) and
a yellow solid 13 appeared on addition of CH2Cl2, to
the resulting brown syrup. Further solid precipitation
occurred on cooling to 4°C. The mixture was cen-
trifuged, the yellow brown pallet washed and cen-
trifuged in benzene three times to remove starting
material, washed once with hexane and dried in vacuo.
Found: C, 43.00; H, 3.51; Cl, 11.00; N, 4.00%.
C24H22Cl2FeN2Pt requires C, 43.65; H, 3.36; Cl, 10.74;
N, 4.24. IR (cm−1): 2271 (wCꢀN): 1732 (wCꢁN); 1651
(Ph of 5); 1594 (Ph of 1), 483, 443, 404, 343 (wn Pt–Cl).
1H-NMR (DMF): 7.91, 7.73, 7.56, 7.40, 7.21(m, Ph);
4.21(s, p-C5H5); 4.17, (p-C5H5,4), 10 (s, CH2). 13C-
NMR (DMF, l): 135.8; 134.3; 130.5; 130.4; 129.7;
126.6; 126.6; 71.8; 71.5; 69.3; 70.0; 68.9; 45.8. 195Pt-
NMR (DMF, l) −2164, 2248. L (DMF)=0. Pt
(PhCN)2Cl2 (0.1 g, 0.22 mmol) with 6 (0.11 g, 0.44
mmol) was stirred 2 h at r.t. in 25 cm3 benzene. The
solvent was evaporated off and then washed with cold
benzene, removing unreacted starting materials. The
brown solid was dried and recrystallised from CHCl3 to
give brown crystals. Yield 0.05 g. Typically: Found: C,
49.9; H, 6.9; Cl, 11.6; N, 4.7%. IR (cm−1); 2660.
1H-NMR; 4.19(s, Fc); 4.06(s, CH2); 2.62(s, CH3).
[12] P.R.R. Ranatunge-Bandarage, Ph.D. Thesis, University of
Otago, 1981.
[13] L. Maresca, G. Natile, F.P. Intini, F. Gasparrini, A. Tiripicchio,
M. Tiripicchio-Camellini, J. Am. Chem. Soc. 108 (1986) 1180.
[14] T.D. Turbitt, W.E. Watts, J. Chem. Soc. Dalton (1974) 1974.
[15] R.C. Weast (Ed.), Handbook of Chemistry and Physics. The
Chemical Rubber Co., CRC Press, Cleveland, OH, 1972.
[16] N.W. Duffy, M. Spescha, B.H. Robinson, J. Simpson,
Organometallics 3 (1994) 4895.
[17] B.H. Robinson, J. Simpson, D.J. Wilson, Acta Cryst. C52 (1996)
2196.
[18] H. Grimes, D. Logan, Inorg. Chim. Acta 45 (1980) 223.
[19] (a) R. Prins, A.R. Koorswagen, A.G.T.G. Kortbeek, J.
Organomet. Chem. 39 (1972) 335. (b) R. Prins, Mol. Phys. 19
(1970) 603.
[20] S.D. Ross, M. Finkelstein, E.J. Rudd, Anodic Oxidations, Aca-
demic Press, New York, 1975.
[21] (a) C.J. McAdam, N.W. Duffy, B.H. Robinson, J. Simpson, J.
Organomet. Chem. 527 (1997) 179. (b) B. Corain, B. Longato,
G. Favero, D. Ajo, G. Pilloni, U. Russo, F.R. Kriessel, Inorg.
Chim. Acta 157 (1989) 259.
[22] W.R. Cullen, C.V. Evans, N.F. Han, J. Trotter, J. Inorg. Chem.
26 (1987) 514.
[23] J. Landells, J. Kerr, University of Otago, unpublished results.
[24] D.M. Adams, J. Chatt, J. Gerrat, A.D. Westland, J. Am. Chem.
Soc. (1964) 734
Acknowledgements
We thank D. Wilson and J. McAdam for experimen-
tal assistance and the University of Otago for a scholar-
ship (PRRB-R).
[25] E. Bielli, P.M. Gidney, R.D. Gillard, B.T. Heaton, J Chem. Soc.
(1974) 2133
References
[26] P.S. Pregosin, Coord. Chem. Rev. 44 (1982) 247.
[27] J. Kerr, unpublished work.
[28] (a) R.W. Fish, M. Rosenblum, J. Org. Chem., 30 (1965) 1253.
(b) A.N. Nesmayanov, V.A. Sazonova, Dokl. Akad. Nauk. 150
(2) (1963) 321.
[29] (a) J.K. Lindsay, C.R. Hauser, J. Org. Chem., 22 (1957) 355. (b)
C.R. Hauser, J.K. Lindsay, J. Org. Chem. 22 (1957) 906.
[30] J.M. Osgerby, P.L. Pauson, J. Chem. Soc. 4 (1961) 4600.
[31] T. Hayashi, T. Mise, M. Fukushima, et al., J. Bull. Soc. Jpn. 53
(1980) 1138.
[1] (a) G. Wilkinson, F.G.A. Stone (Eds.), Comprehensive
Organometallic Chemistry, Vol 8, Ch 59. Pergamon, Oxford,
1982. (b) D.W. Slocum, C.R. Ernst, Adv. Organomet. Chem. 10
(1972), 79.
[2] A.N. Nesmeyanov, E.G. Peravalova, R.V. Golovinya, Doklady
Akad. Nauk. SSSR 103 (1995) 81–82; CA 50, 4926c.
[3] (a) I. Haiduc, C. Silvestru, Organometallics in Cancer
Chemotherapy, CRC Press, Boca Raton, 1989. (b) P.J. Ko¨pf-
Maier, H. Ko¨pf, Chem. Rev. 87 (1987) 1137. (c) P. Ko¨pf-Maier,
H. Ko¨pf, Structure and Bonding 70 (1988) 103. (d) P. Ko¨pf-
Maier, H. Ko¨pf, E.W. Neuse, J. Cancer Clin. Oncol. 108 (1984)
336.
[32] P.L. Pauson, M.A. Sandhu, W.E. Watts, J. Chem. Soc. C (1966)
251.
[33] J.H. Price, A.N. Williamson, R.F. Schramm, B.B. Wayland,
Inorg. Chem. 11 (1972) 1280.
[4] (a) B. Longato, B. Corain, G.M. Bonora, G. Pilloni, Inorg.
Chim. Acta. 137 (1987) 75. (b) M. Sataya Murthy, L.N. Rao,
.