The relationship between the electrochemical and chemical oxidation of ferrocene-containing carbonyl-phosphane-β-diketonato-rhodium(I) complexes - Cytotoxicity of [Rh(FcCOCHCOPh)(CO)(PPh3)]

Article abstract of DOI:10.1002/ejic.201100007

Ferrocene-containing rhodium(I) complexes of the type [Rh(FcCOCHCOR)(CO) (PPh₃)] with R = Fc (ferrocenyl), 1, Ph (phenyl), 2, CH₃, 3, and CF₃, 4, have been studied by cyclic voltammetry (CV) and bulk electrolysis techniques in CH₃CN. Two isomers for complexes 2-4 were detected. Results are consistent with Rh^I being oxidised first in an electrochemically irreversible two-electron transfer process, followed by the electrochemical reversible oxidation of each ferrocenyl group in a one-electron transfer process at slightly larger potentials. Only the E_pa(Rh) CV peaks resolved into two separate waves, each one being associated with an isomer. Slow isomerisation kinetics allowed the detection of two isomers of 2-4. E_pa(Rh) and other compound physical parameters relates to the second-order rate constant for the oxidative addition of CH₃I to complexes 1-4 in acetone. Rhodium(III) reduction of oxidised 1, 2, 3 and 4 was observed at large negative potentials, while for [Rh(FcCOCHCOCF ₃)(CH₃)(I)(CO)(PPh₃)] (9) it was observed at -1.492 V vs. Fc/Fc⁺. Complex 2 was slightly more cytotoxic than the free FcCOCH₂COPh ligand but less cytotoxic than cisplatin.

Full text of DOI:10.1002/ejic.201100007

FULL PAPER
DOI: 10.1002/ejic.201100007
The Relationship between the Electrochemical and Chemical Oxidation of
Ferrocene-Containing Carbonyl-Phosphane-β-Diketonato-Rhodium(I)
Complexes – Cytotoxicity of [Rh(FcCOCHCOPh)(CO)(PPh3)]
Jeanet Conradie[a] and Jannie C. Swarts*[a]
Keywords: Rhodium / Carbonyl complexes / Sandwich complexes / Cytotoxicity / Cyclic voltammetry
Ferrocene-containing rhodium(I) complexes of the type
[Rh(FcCOCHCOR)(CO)(PPh3)] with R = Fc (ferrocenyl), 1, Ph
(phenyl), 2, CH3, 3, and CF3, 4, have been studied by cyclic
voltammetry (CV) and bulk electrolysis techniques in
two separate waves, each one being associated with an iso-
mer. Slow isomerisation kinetics allowed the detection of two
isomers of 2–4. Epa(Rh) and other compound physical param-
eters relates to the second-order rate constant for the oxidat-
CH3CN. Two isomers for complexes 2–4 were detected. Re- ive addition of CH3I to complexes 1–4 in acetone. Rhodi-
sults are consistent with RhI being oxidised first in an electro-
chemically irreversible two-electron transfer process, fol-
lowed by the electrochemical reversible oxidation of each
ferrocenyl group in a one-electron transfer process at slightly
larger potentials. Only the Epa(Rh) CV peaks resolved into
um(III) reduction of oxidised 1, 2, 3 and 4 was observed at
large negative potentials, while for [Rh(FcCOCHCOCF3)-
(CH3)(I)(CO)(PPh3)] (9) it was observed at –1.492 V vs. Fc/
Fc+. Complex 2 was slightly more cytotoxic than the free
FcCOCH2COPh ligand but less cytotoxic than cisplatin.
transfer processes[11] due to their high thermal stability, re-
versible redox behaviour, and their chemical modification
possibilities.
Introduction
Dicarbonylrhodium(I) complexes were made famous by
the use of [Rh(CO)2(I)2]– as catalyst in the Monsanto pro-
cess of converting methanol to acetic acid.[1] The rate-de-
termining step of this homogeneous catalytic process in-
volves oxidative addition. During oxidative addition, the re-
dox state of the catalytic metal centre changes from rhodi-
um(I) to rhodium(III). The kinetics of CH3I oxidative ad-
dition in complexes of the type [Rh(β-diketonato)(CO)-
(PPh3)], where the β-diketonato either contains a metallo-
cene group,[2,3] or not,[4] have been thoroughly studied. In
contrast, no electrochemical properties relating to ferro-
cene-containing β-diketonato carbonyl phosphane rhodium
complexes have to the best of our knowledge been pub-
lished. This is surprising because the reactivity of the rho-
dium-based catalytic species is related to the ease of rho-
dium oxidation during the course of the oxidative addition
reaction.
Since the ferrocenyl group is a strongly electron-donating
group, it accelerates oxidative addition substantially.[3]
Quantification ofthe effects that ligands have on the rhodi-
um(I) redox (oxidation) potential may assist in designing
effective rhodium-based catalysts. Several physical proper-
ties of such ligands may be used to predict the oxidation
potential of a rhodium(I) centre in a specific class of com-
pounds, here [Rh(RCOCHCORЈ)(CO)(PPh3)] complexes
with R = ferrocenyl, Fc. These especially include Gordy
scale group electronegativities, χR, of ligand side groups and
“observed” ligand pKЈa values. Published pKa data for free
β-diketones, which exist as keto and enol isomers in equilib-
rium,[12] are referred to as “apparent pKЈa values”[13] be-
cause in the experimentally obtained pKЈa values, no at-
tempt was made to separate enol and keto tautomer pKa
values. Gordy scale group electronegativities, χR, of the R
groups of β-diketones RЈCOCH2COR, are empirical num-
bers that express the combined tendency of not only one
atom, but a group of atoms, like R = CF3 or ferrocenyl
(Fc), to attract electrons (including those in a covalent
bond) as a function of the number of valence electrons, n,
and the covalent radius, r (in Å), of groups as discussed
elsewhere.[14] It is empirically adjusted to be aligned with
Pauling atomic electronegativities.
Ferrocenes in general are studied as donors in energy
transfer processes,[5] because they enhance catalytic activity
in many reactions,[3,6] as high-burning rate composite pro-
pellant catalyst,[7] as anticancer drugs[8] and as a strong
electron-donating substituent to manipulate electron den-
sity on complexes.[9,10] They are often used in electron
[a] Department of Chemistry, University of the Free State,
P. O. Box 339, Bloemfontein 9300, Republic of South Africa
Fax: +27-51-444-6384
It has previously been shown that the pKЈa of the β-di-
ketone RЈCOCH2COR is linearly dependent on the sum of
the group electronegativities of RЈ and R, (χRЈ + χR) by
pKЈa = –3.484(χRЈ + χR) + 24.6.[3] The formal reduction
E-mail: swartsjc@ufs.ac.za
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201100007,
Eur. J. Inorg. Chem. 2011, 2439–2449
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2439

J. Conradie, J. C. Swarts
FULL PAPER
potentials (in V vs. Fc/Fc+ in CH3CN) of a series of ferro-
Complex 1, [Rh(FcCOCHCOFc)(CO)(PPh3)], contains a
cene containing β-diketones FcCOCH2COR, also relates[9] symmetrical β-diketonato ligand with two ferrocenyl side
linear to the group electronegativity, χR, of R by EoЈ = groups and exhibits three oxidation peaks during the cyclo-
0.114χR – 0.025. Moreover, the sum of the group electro- voltammetric anodic cycle in the region 0–0.6 V vs. Fc/Fc+,
negativities of RЈ and R of the β-diketonato ligand see Figure 1. The electrochemically irreversible anodic oxi-
(RЈCOCHCOR)– coordinated to [Rh(RЈCOCHCOR)(CO)- dation peak Epa at 0.108 V vs. Fc/Fc+ is assigned to the
(PPh3)] was shown to determine the second-order rate con- two-electron oxidation of rhodium(I) to rhodium(III). The
stant, k2, of oxidative addition of CH3I to these rhodium two electrochemically reversible couples with formal re-
complexes by the equation lnk2 = –3.14(χRЈ + χR) + 10.0.[3] duction potentials, EoЈ, of 0.200 and 0.312 V vs. Fc/Fc+
In this communication we specifically report on the influ- correspond to the two successive one-electron redox process
ence of Gordy scale group electronegativities of side groups of each of the two ferrocenyl groups of the (FcCOCH-
of ferrocenyl-containing β-diketones on the oxidation COFc)– ligand of 12+. The two ferrocenyl groups of the free
potential of the rhodium centre in the complexes ligand, FcCOCH2COFc, are also oxidised in two successive
[Rh(FcCOCHCOR)(CO)(PPh3)] with R = Fc, Ph, CH3 and redox steps during cyclic voltammetry.[9] The two ferrocene-
CF3. We generalise our results by relating χR, β-diketone based formal oxidation potentials of the free ligand are 89
pKЈa, Epa of the rhodium(I) centre and rhodium complex and 92 mV smaller (less positive) than the corresponding
νCO IR stretching frequencies to each other, and more im- EoЈ values of 12+
portantly, to the rate of CH3I oxidative addition to these
complexes via k2, the second-order rate constant of this re-
action. The cytoxicity of [Rh(FcCOCHCOPh)(CO)(PPh3)]
is also discussed.
.
Results and Discussion
The electrochemical behaviour of [Rh(FcCOCHCOR)-
(CO)(PPh3)] complexes with R = ferrocenyl = (C5H5)-
Fe(C5H4) = Fc, 1, phenyl = C6H5, 2, CH3, 3 and CF3, 4,
was studied in acetonitrile containing 0.100 moldm–3 tetra-
n-butylammonium hexafluorophosphate {[N(nBu4)][PF6]}
utilizing a glassy carbon working electrode. Choice of this
solvent ensured a two-electron oxidation process for the
electrochemical oxidation of the rhodium(I) centre to en-
sure that it is comparable with the two-electron chemical
oxidation of rhodium(I) during oxidative addition with
CH3I. Complexes 1–4 have three (for 1) or two (in 2, 3 and
4) metal redox centres; the iron nucleus of the ferrocenyl
group(s) and the square planar rhodium(I) centre, see
Scheme 1.
Figure 1. Cyclic voltammograms of a 1 mmoldm–3 solution of
[Rh(FcCOCHCOFc)(CO)(PPh3)] (1), measured in 0.1 moldm–3
[N(nBu4)][PF6]/CH3CN at scan rates of 50 (smallest currents), 100,
150, 200 and 250 mV s–1 on a glassy carbon working electrode at
25.0(1) °C. Scans are initiated in the positive direction, as indicated
by the arrow. The rhodium and two ferrocene oxidation peaks are
identified as Rh, Fc(1) and Fc(2), respectively.
Electron transfer approached electrochemical reversibil-
ity at slow scan rates with ΔEp = Epa – Epc = 74 or 79 mV
for the redox couples of the two ferrocenyl groups of 1 at
scan rate 100 mV s–1, see Table 1 (see Table S1 in the
Supporting Information; scan rates 50–250 mV s–1). Ideal
electrochemical reversibility is characterised by ΔEp
=
59 mV.[15]
Different formal reduction potentials for side groups
on symmetrical complexes in which mixed-valent (i.e.
differently charged) intermediates are generated, here
the ferrocenyl and the ferrocenium groups in
[RhIII(Fc+COCHCOFc)(CO)(PPh3)]2+, are well known in
systems that allow electron delocalisation, either through
bridge mediated paths or from a direct metal–metal interac-
tion.[16] The inequivalence of the ferrocenyl (Fc) and ferro-
cenium (Fc+) groups of this intermediate is highlighted in
Scheme 1. Rhodium complexes containing three (in 1) or two (in
2, 3 and 4) metal redox centres. An equilibrium exists between the
two isomers of [Rh(FcCOCHCOR)(CO)(PPh3)] giving Kc = [iso-
mer 1]/[isomer 2][10] at 298 K in CDCl3 (value in brackets is for
CD3CN) when R ϶ Fc. For 4, Kc in CD3CN gave poor resolved
broad NMR peaks; from electrochemical studies, Kc ≈ 0.6. The RhI
nucleus of 1 is the most electron-rich in the series 1–4, while for 4
it is the most electron deficient, see ref.[3]
terms of their group electronegativities: χFc = 1.87, χFc+
=
2.82.[9,13,17] The electron-withdrawing power of the ferro-
cenium group is almost as high as that of the CF3 group
(χCF3 = 3.01). As there is good communication (by conjuga-
2440
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2439–2449

Fc-Containing Rhodium(I) Complexes
Table 1. Electrochemical data for the ferrocenyl groups in ferrocenyl groups of the two isomers of 2, 3 and 4. These
[Rh(FcCOCHCOCR)(CO)(PPh3)] and [Rh(FcCOCHCOCF3)-
latter peaks could not be well resolved under our experi-
(CH3)(I)(CO)(PPh3)] (9) from 0.7 mmoldm–3 solutions at a scan
mental conditions.
rate of 100 mVs–1 measured in 0.1 moldm–3 [N(nBu4)][PF6]/
CH3CN on a glassy carbon electrode at 25.0(1) °C vs. Fc/Fc+. Fer-
rocene itself showed under identical conditions reversible electro-
chemical behaviour (ΔEp, = 66 mV, ipc/ipa = 1.00 at scan rate ν =
100 mV s–1, EoЈ = 0.077 V vs. a Ag/Ag+ reference electrode).
Epa /V[a]
ΔEp /mV[b] EoЈ /V[c] ipc /μA[a] ipc/ipa
1; R = Fc
0.237[d]
74[d]
0.200[d]
11[d]
,
1.00[d]
(1.00)[e]
7.5 [f,g]
7.9 [f,g]
7.7 [f,g]
0.98
(0.352)[e]
(79)[e]
117[g]
102[g]
122[g]
79
(0.321)[e] 11[e]
2; R = C6H5 0.331
0.273
0.250
0.380
0.230
12
11
10
12.9
3; R = CH3
4; R = CF3
9
0.301
0.441
0.269
[a] Peak anodic potential or peak cathodic current. [b] ΔEp = Epa
–
Epc. [c] EoЈ = (Epa + Epc)/2. [d] Values for the first oxidized ferro-
cenyl group. [e] Values for the second oxidized ferrocenyl group. [f]
Measured values for two unresolved ferrocenium reduction peaks.
[g] ΔEp and ipc/ipa are unusually large because of two unresolved
Fc peaks from two isomers that lead to peak broadening.
Figure 2. Cyclic voltammograms of 0.7 mmoldm–3 solutions of
tion) between the Fc and Fc+ β-diketonato-pendent side [Rh(FcCOCHCOR)(CO)(PPh3)] complexes 1 (R = Fc), 2 (R =
C6H5), 3 (R = CH3) and 4 (R = CF3) at scan rate 100 mV s–1
groups on the pseudo-aromatic core of the intermediate
measured in 0.1 moldm–3 [N(nBu4)][PF6]/CH3CN on a glassy car-
[RhIII(Fc+COCHCOFc)(CO)(PPh3)]2+, oxidation of the
bon working electrode at 25.0(1) °C. Complexes 2, 3 and 4 exist
second Fc group to yield the final oxidised species
as two isomers. The two rhodium and two unresolved ferrocene
[RhIII(Fc+COCHCOFc+)(CO)(PPh3)]2+ takes place at a
oxidation peaks of 2, 3 and 4 are identified as Rh(1), Rh(2) and
Fc(1,2). The Rh(2) peak of 3 overlaps indistinguishable with the
ferrocenyl peaks. Complex 1 having a symmetrical β-diketonato li-
gand cannot have two isomers but it has two ferrocenyl groups, the
redox peaks of which are labelled Fc(1) and Fc(2).
more positive potential than observed for the oxidation of
the first Fc group.
Scheme 2 represents the electrochemical processes associ-
ated with 1.
An alternative but less likely explanation for the multiple
redox peaks observed in Figure 2 exists. Rather than as-
signing the two consecutively observed RhI ǞRhIII oxi-
dation peaks for 2, 3 and 4 (Figure 2) to two isomeric forms
as shown in Scheme 1, one may consider to assign the first
oxidation RhI wave to a RhI ǞRhII conversion and the sec-
ond to a RhII ǞRhIII conversion as described for the five-
coordinate complex [RhI(H)(CO)(PPh3)3].[18] We do not
favour including identification of a rhodium(II) species in
CH3CN for the present compound series for the following
reasons: Since four coordinate rhodium(II) complexes are
known to be labile, except in the case where RhII dimers
are coordinated to acetate ligands, we propose that, al-
though the oxidation RhI ǞRhIII in successive observable
steps in CH3CN as solvent is not impossible, it is rather
unlikely. Experimentally if RhI ǞRhII oxidation corre-
sponds to the first observed oxidation peak and
RhII ǞRhIII corresponds to the second, one would expect
that the peak currents, ipa, for RhI ǞRhII oxidation as well
as for RhII ǞRhIII oxidation should be the same for all
Scheme 2. Electrochemical scheme highlighting the electrochemical
reversible ferrocene-based and electrochemical irreversible rho-
dium-based electron transfer processes. Species 12+ is a 14 electron
complex that must be ligated by two additional ligands. The most
likely candidate for this is the solvent, CH3CN, but CO bridges and
–
PF6 coordination are two further possibilities.
Complexes 2–4, containing unsymmetrical β-diketonato complexes. This was not found to be the case. In addition,
ligands exist in solution as two isomers, see Scheme 1.[10] from the ferrocene waves, a one-electron transfer process in
This equilibrium is slow to be reached, see the kinetic dis- the present complexes under our experimental conditions is
cussion below, and enabled us to observe each isomer by found to vary between 10 and 12 μA (Table 1). The values
1H NMR separately, and to determine Kc values (defined of the peak anodic currents for the first rhodium oxidation
in Scheme 1) in different solvents, including CD3CN.[10] The shown in Figure 2, ipa(Rh1), of the first isomer of 2, 3 and
CVs of 2, 3 and 4 (Figure 2) are thus expected to exhibit 4 measured 17, 21 and 15 μA, respectively (Table 2). This is
two electrochemically irreversible anodic rhodium(I) oxi- too large for a one-electron transfer process. The sum of
dation peaks and two peaks that are associated with the the two rhodium peak anodic currents, ipa(Rh1) + ipa(Rh2)
Eur. J. Inorg. Chem. 2011, 2439–2449
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2441

J. Conradie, J. C. Swarts
FULL PAPER
Table 2. Electrochemical and oxidative addition kinetic data related to rhodium(I) oxidation in [Rh(β-diketonato)(CO)(PPh3)] complexes.
[b]
β-diketonato
Epa(Rh) /V [ipa /μA][a]
Epc(Rh) /V
pKЈa
k2 /mol–1 dm3 s–1 [c] νCO /cm–1 [d]
1
2
3
4
5
6
7
8
FcCOCHCOFc
FcCOCHCOC6H5
FcCOCHCOCH3
FcCOCHCOCF3
0.108 [26][a]
–0.69
13.1(1)
10.41(2)
0.155[h]
1977
1977
1980
1986
1979
1980
1983
1983
1: 0.123 [17], 2: 0.273 [8][a]
1: 0.154 [21], 2: 0.253 [3][a,e]
1: 0.196 [15], 2: 0.327 [9][a]
1: –0.75, 2: –[g]
0.0409(3)
0.0455(6)
0.00370(4)
0.00961
0.00930
0.00112
0.00146
1: –0.86, 2: –0.45[g] 10.01(1)
1: –0.70, 2: –[g]
6.56(3)
9.35
8.7
6.3
6.3
C6H5COCHCOC6H5 0.308[f]
CH3COCHCOC6H5 0.336[f]
CF3COCHCOC6H5 0.448[f]
CF3COCHCOCH3
0.491[f]
0.573[f]
10 CF3COCHCOCF3
4.71(1)
[a] Oxidation peak potentials were measured on 0.7 moldm–3 substrate concentration in 0.1 moldm–3 [N(nBu4)][PF6]/CH3CN on a glassy
carbon electrode at 25.0(1) °C at a constant sweep rate of 100 mVs–1 referenced against Fc/Fc+; values in square brackets are ipa values
for isomers 1 and 2, respectively. [b] pKЈa of the free β-diketone from ref.[13,20,21]. [c] Chemical oxidative addition rate constants k2 are
for the first step of the oxidative addition reaction between various [Rh{β-diketonato(CO)(PPh3)}] complexes and CH3I in acetone at
25.0(1) °C as described in ref.[2]. [d] The IR carbonyl stretching frequency of [Rh(β-diketonato)(CO)(PPh3)] complexes, see ref.[10]. [e]
Estimated values only because of closely overlapping peaks. [f] Results from ref.[22]. [g] Epc(Rh) of the other isomer could not clearly be
identified. [h] Value in chloroform.
for 2, 3 and 4 was however 25, 24 and 24 μA, respectively,
the expected values for two-electron transfer for the isomer
mixture. In addition, a fast two-electron transfer process of
a single compound having no isomers should exhibit a sin-
gle ipa peak of ca. 24 μA, i.e. twice the value of the ferrocene
measurements. Complex 1 does not have any isomeric forms
because the β-diketonato ligand is symmetrical. For this
complex, only a single peak associated with RhI oxidation
is observed, the size of which (ipa = 26 μA) is also consistent
with a two-electron transfer process, Figures 1 and 2,
Table 2.
The large ferrocenyl ΔEp values of 2–4 (Table 1) are not
indicative of electrochemical irreversibility. Rather, it is the
result of two unresolved ferrocenyl peaks from the two
[Rh(FcCOCHCOR)(CO)(PPh3)] isomers that lead to peak
broadening. The difficulties that are encountered in resolv- Figure 3. Time trace showing the conversion for isomer 2 to isomer
1
(reaction defined in Scheme 1) at 20 °C in CDCl3 for
ing closely overlapping waves in cyclic voltammetry are well
[Rh(FcCOCHCOCH3)(CO)(PPh3)], 3. Insert: A kinetic first order
treatment of data for this process utilizing the first order kinetic
iu0 – iu
discussed elsewhere.[15,19]
To further strengthen the argument that the two sets of
two-electron rhodium(I) oxidation peaks observed in Fig-
ure 2 corresponds to the two equilibrium species shown in
equation ln(
ϱ) = kobst. iut is the average integration units of
iut – iuϱ
one H of isomer 1 in 3 at time t and is proportional to the concen-
Scheme 1, we attempted to follow the kinetics of conversion tration of isomer 1; iu0 is the average integration unit at time t = 0
1
and iuϱ the average integration units at time t = ϱ (i.e. after equilib-
of isomer 2 to isomer 1 by H NMR spectroscopy. Poor
rium has set in).
solubility and concomitant low integration accuracies of
complexes 2–4 in CD3CN prevented collection of reliable
kinetic data to study the equilibrium isomerization kinetics scan from peak potential Rh(1) which is associated with
of 2–4 in this solvent. For 2 and 3, equilibrium in CD3CN oxidation of the rhodium nucleus of the first isomer, to
was reached in less than 5 min, and for 4, equilibrium in peak potential Rh(2), which is associated with the oxidation
CD3CN was reached within one hour. However, in CDCl3, of the rhodium nucleus of the second isomer, is much
it was possible to measure the conversion of isomer 2 to shorter than t1/2. This means rhodium oxidation in both
isomer 1 of complex 3 with time, see Figure 3. The half-life isomers will take place on a CV timescale before concentra-
of this transformation is t1/2 = 143 s and the observed rate tion changes can be completed to reinstate Kc after the first
constant of this transformation is kobs = 0.0048(6) s–1 at rhodium isomer has been oxidised. Figure 4 shows the
20 °C.
NMR of 3 in CDCl3 at equilibrium as well as the NMR of
This slow rate of isomerization is very useful because it 3 enriched in isomer 1.
means in principle it is possible to distinguish between two
The above described electrochemical results imply a two-
different rhodium-based oxidation peaks during the electro- electron transfer for the RhI/III couple and a one-electron
chemical oxidation of the rhodium nucleuses of two isomers transfer process for the Fc/Fc+ couple. This combined
of [Rh(FcCOCHCOR)(CO)(PPh3)]. The time required (less three-electron transfer process was confirmed by bulk elec-
than 3 s at the slowest scan rate) in a CV experiment to trolysis for 2–4. Figure 5 shows current changes and the
2442
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2439–2449

Fc-Containing Rhodium(I) Complexes
[Rh(CH3COCHCOCH3)(CO)(PPh3)],
[Rh(CH3COCH-
COC6H5)(CO)(PPh3)] (6) and [Rh(CF3COCHCOC6H5)-
(CO)(PPh3)] (7) in 0.2 moldm–3 [N(nBu4)][BF4]/CH2Cl2 at
a Pt-disc electrode. The big difference in the results of Pom-
beiro[23] and the present communication is the solvent used.
Pombeiro used CH2Cl2, while in this study CH3CN was
preferred. Geiger,[24] and others[15a,25] have shown that the
use of CH2Cl2 over CH3CN and also [N(nBu)4][B(C6F5)4]
over [N(nBu4)][PF6] as supporting electrolyte often lead to
detection of unstable species such as ruthenocenium radical
cations.
To establish whether it is possible to observe RhIII re-
duction in our compounds, several experiments were per-
formed. These are shown in Figure 6 for 3. Upon scanning
in a positive direction from –0.27 V till 0.71 V (bottom
graph) the rhodium centre of both isomers of 3 were oxid-
ized to a RhIII species, the second of which was closely over-
lapping with the oxidation waves of the ferrocene centres
(Table 2). During the reverse cycle, apart from observing
the ferrocenium reduction at 0.199 V, two further small
cathodic peaks were observed at –0.45 and –0.86 V respec-
tively. These peaks are indicative of two separate RhIII re-
ductions associated with each of the two isomers of 3. In a
separate experiment, when the reversal potential of the first
forward scan was lowered from 0.71 V to 0.185 V, the cath-
odic peak at –0.45 V was not observed (Figure 6, bottom
grey scale). Since this reversal potential was small enough
to disallow oxidation of the second RhIII isomer during the
first anodic sweep at ca. 0.253 V (Table 2) it can be con-
cluded that the –0.45 V cathodic peak must belong to RhIII
reduction of one of the two isomers of oxidised 3 (isomer
2, Rh2 in our notation).
Figure 4. The 3–6 ppm region of
a
1H NMR spectrum of
[Rh(FcCOCHCOCH3)(CO)(PPh3)], 3 enriched in isomer 1 (bot-
tom) and in equilibrium with isomer 2 (top, isomers and Kc are
defined in Scheme 1) at 293 K in CDCl3 giving Kc = [0.217/1 +
0.441/2.040 + 0.437/2.080 + 1.064/4.920]/4 = 0.22[10] Isomer 1: 4.19
(s, 5 H, C5H5), 4.37 (t, 2 H, C5H4), 4.79 (t, 2 H, C5H4), 5.73 (s, 1
H, CH); Isomer 2: 3.91 (s, 5 H, C5H5), 4.08 (t, 2 H, C5H4), 4.15 (t,
2 H, C5H4), 5.78 (s, 1 H, CH).
amount of electrons transferred per molecule during the
bulk electrolysis of 4 at an applied potential of 920 mV vs.
Fc/Fc+.
Figure 5. The electric current vs. time (i vs. t) response and total
It
amount of electrons transferred per molecule ( , n = number of
nF
mol of 4, F = Faraday’s constant) upon performing bulk electrolysis
at 0.923 V vs. Fc/Fc+ on [Rh(FcCOCHCOCF3)(CO)(PPh3)] (4).
Figure 6. Cyclic voltammogram of 2 mmoldm–3 [Rh(FcCOCH-
COCH3)(CO)(PPh3)] (3) measured in 0.1 moldm–3 [N(nBu4)][PF6]/
CH3CN at a scan rate of 100 mV s–1 on a glassy carbon working
electrode at 25.0(1) °C. The arrows indicate initial scan directions.
See Table 1, footnote g, for explanation of small ipc/ipa ratios.
Our observed two-electron transfer for rhodium(I) oxi-
dation obtained in a 0.1 moldm–3 [N(nBu4)][PF6]/CH3CN
solution is in agreement with those described by Lam-
precht,[22] whom used the same solvent/electrolyte system as
us to study 5–10 (see Table 2 for compound formulas), but
contrasts the results obtained by Pombeiro[23] for other
To show that the –0.86 V cathodic peak represents RhIII
rhodium(I) complexes. Pombeiro[23] demonstrated a one- reduction of the other oxidized isomer of 3 (Rh1 in our
electron oxidation process for the oxidation of rhodium in notation), another experiment was conducted in which the
Eur. J. Inorg. Chem. 2011, 2439–2449
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2443

J. Conradie, J. C. Swarts
FULL PAPER
scan direction was negative from the onset potential at see Figure 7. Only a electrochemically reversible Fc/Fc+
–0.27 V, Figure 6 (top grey scale). Such a procedure meant couple relating to the ferrocenyl group of [Rh(FcCOCH-
no RhIII generation has occurred before the scan direction
was reversed. In this case, the –0.86 V cathodic peak was
also absent which proved this peak was associated with
RhIII reduction of isomer 1 of oxidized 3 which is generated
at Epa(Rh1) = 0.154 V.
The electrochemical Scheme that summarizes the above
described results for 2, 3 and 4 are shown in Scheme 3.
The exact formula of the unstable in situ formed RhIII
complexes shown in Scheme 3 is unknown. Spectro-electro-
chemistry failed to give additional information on the po-
tential structure. Rhodium(III) complexes normally are oc-
tahedral, so two further ligands is required to complete the
newly generated rhodium(III) coordination sphere. The two
ligands that will coordinate to these electrochemically in
situ formed, RhIII complex, apart from CO, PPh3 and the
β-diketonato, is unknown. However, it can only be species
–
originating from the supporting electrolyte anions, PF6 , or
more likely, solvent molecules, CH3CN.
The structures of [RhI(FcCOCHCOCF3)(CO)(PPh3)][26]
and [RhIII(FcCOCHCOCF3)(CH3)(I)(CO)(PPh3)] (9) was
published[2] and the geometry of other RhIII-β-diketonato
Figure 7. Cyclic voltammograms of 0.7 mmoldm–3 [RhI(FcCOCH-
COCF3)(CO)(PPh3)] (4) (top at 100 mV s–1) and 1.5 mmoldm–3
derivatives were predicted by means of DFT geometry opti-
mization techniques.[27] The electrochemistry of this known [RhIII(FcCOCHCOCF3)(CH3)(I)(CO)(PPh3)] (9) [middle at scan
rates of 50, 100, 150, 200 and 250 (largest ipa) mVs–1] measured in
and stable RhIII derivative 9 containing an un-oxidized
0.1 moldm–3 [N(nBu4)][PF6]/CH3CN at a glassy carbon working
ferrocenyl group were investigated cyclic voltammetrically
electrode at 25.0(1) °C. The CV of 9 exhibits an electrochemically
to put the observed rhodium(III) reductions of complexes
1–4 in perspective (Figure 7). Complex 9 currently repre-
sents the only available stable rhodium(III) derivative of 1–
4, and its structure differs from that proposed in Scheme 3
by having the two solvent molecules replaced by an iodo
and a methyl group. As was expected, no initial oxidation
reversible Fc/Fc+ couple which corresponds to the formal reduction
potential of the ferrocenyl group of the (FcCOCHCOCF3)– ligand
coordinated to 9 and an electrochemically irreversible one-electron
transfer process corresponding to the reduction of the rhodium(III)
centre. The linear sweep voltammogram (LSV) of 9 (bottom) at a
scan rate of 2 mV s–1 show a one-electron transfer process for both
the oxidation of the ferrocenyl group and the reduction of rhodium
of RhI to RhIII could be seen in the 0–0.5 V potential range, of complex 9.
Scheme 3. Electrochemical oxidation of [Rh(FcCOCHCOR)(CO)(PPh3)] complexes 1–4. Isomers in Scheme may interchange. x denotes
the fraction of reduced rhodium(I) complexes existing as isomer that oxidizes first, while y denotes the fraction of oxidized compound
existing as isomer 2 that is reduced on the reverse cycle. x and y are not necessarily the same. Rhodium(III) reduction is in the above
scheme indicated to generate either rhodium(I) or rhodium(II) because it was not possible to determine the oxidation state of the
reduced species unambiguously. However, the reduction product of 9, [RhIII(FcCOCHCOCF3)(CH3)(I)(CO)(PPh3)], was shown to be a
rhodium(II) species (Figure 7).
2444
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2439–2449

Fc-Containing Rhodium(I) Complexes
Table 3. Electrochemical data for the oxidation and reduction of the ferrocenyl group and the rhodium nucleus of the complex
[RhIII(FcCOCHCOCF3)(CH3)(I)(CO)(PPh3)] (9) measured in 0.1 moldm–3 [N(nBu4)][PF6]/CH3CN on a glassy carbon electrode at 25.0(1)
°C vs. Fc/Fc+. The concentration of the complex was 1.5 mmoldm–3 in CH3CN.
Ferrocenyl group
Rhodium nucleus
ν/mVs–1
Epa/V
ΔEp/mV
EoЈ/V
ipa/μA
ipc/ipa
Epa/V
Epc /V
ipa/μA
ipc/μA
50
0.268
0.269
0.273
0.275
0.277
74
79
84
90
94
0.231
0.230
0.231
0.230
0.230
18.2
25.8
32.4
36.4
40.8
0.98
0.98
0.98
0.97
0.97
–
–
–1.536
–1.542
–1.566
–1.567
–1.568
23.3
29.1
36.8
40.0
42.5
–
–
4.0
4.4
5.1
100
150
200
250
–0.459
–0.504
–0.517
COCF3)(CH3)(I)(CO)(PPh3)] (9) with ΔEp = 74 mV at scan
With respect to group electronegativity sums, (χRЈ + χR),
rate 50 mV s–1 was observed in the potential range –0.2 till and pKЈa values, the apparent acid strength of the free β-
1 V vs. Fc/Fc+ (Table 3). The formal reduction potential of diketones RЈCOCH2COR, it was shown in ref.[3] that
the ferrocenyl group in 9, EoЈ = 0.230 V, is 150 mV less posi-
lnk2 = –3.14 (χRЈ + χR) + 10.0
and that
(1a)
tive than the Fc/Fc+ couple for [RhIII(FcCOCH-
COCF3)(CO)(PPh3)(solvent?)2]. This implies the two un-
known ligands of the in situ formed [RhIII(FcCOCH-
COCF3)(CO)(PPh3)(solvent?)2] complex must have strong
electron withdrawing properties compared to CH3 and I.
However, during the reverse cathodic cycle, reduction of
lnk2 = 0.83 pKЈa – 11.6
(1b)
where k2 is the second-order rate constant of the oxidative
addition of CH3I to 1–8 in acetone at 25.0(1) °C. From this
study, the relationship between Epa(Rh2) where Rh2 relates
to the second isomer (see Scheme 1, Figure 2 and Table 2)
and lnk2 for 1–8 is shown in Figure 8 (a), black line. The
linear tendency between these data points fits the equation
the rhodium(III) centre of 9 was observed as at Epc
=
–1.492 V vs. Fc/Fc+ at a scan rate of 100 mV s–1. This result
parallel the rhodium(III) reduction of 4 also shown in Fig-
ure 7. The Epc value of 4 is 0.75 V more positive than that
of 9, and like the potential of the ferrocenyl-based wave, is
also consistent with the oxidised product of 4 having two
stronger electron-withdrawing groups coordinated to it
than methyl and iodo. The peak currents of the ferrocene
oxidation and the rhodium(III) reduction are approxi-
mately the same (Table 3). This is consistent with a one-
electron transfer process for each. Linear sweep voltamme-
try of 9 also shows a one-electron transfer process for both
the oxidation of the ferrocenyl group and the reduction of
rhodium of complex 9. The reduced rhodium(II) species
that formed at –1.492 V is re-oxidised at ca. –0.5 V in an
electrochemically irreversible process.
lnk2 = –14.721Epa(Rh(2)) + 0.195; R2 = 0.95
(1c)
Similarly, the linear relationship between ferrocenyl EoЈ
values summarize in Table 1 and k2 (insert in Figure 8, a)
was determined as
lnk2 = –20.5 EoЈ(Fc) + 2.11; R2 = 0.99
(1d)
The plot of νCO, the carbonyl stretching frequencies of
[Rh(RЈCOCHCOR)(CO)(PPh3)] vs. Epa(Rh) was not as
clearly linear as the Epa(Rh2)/lnk2 relationship, Figure 8 (a,
broken line). Especially complexes 2 and 4 were clearly off
the predicted trend. By ignoring data for complexes 2 and
4, the equation
The rhodium-based peak anodic potential of 1–4 is a
measure of the energy required to remove electrons from
the rhodium centre, i.e.
νCO = 16.1Epa(Rh) + 1975; R2 = 0.91
(2a)
[RhI(β-diketonato)(CO)(PPh3)] + energyǞ
[RhIII(β-diketonato)(CO)(PPh3)] + 2e–
was fitted to the graph.
In ref.[10] it was shown that
The more positive the rhodium centre becomes under the
influence of the coordinating ligands, the more energy will
be required to remove electrons from it. This corresponds
to a higher Epa (electrochemical oxidation) and slower ki-
netic rate of chemical (e.g. with CH3I) oxidation. Param-
eters that are related to the electron density on the rhodium
centre in [RhI(β-diketonato)(CO)(PPh3)] is the pKЈa value
of the β-diketone RЈCOCH2COR coordinated to the rho-
dium complex (RЈ = Fc or any other group), and the group
electronegativities χRЈ or χR of the RЈ and R side groups on
the β-diketonato ligand.[10] In turn, an increase in the elec-
tron density on the Rh metal centre will result in lower CO
νCO = 5(χRЈ + χR) + 1959
and that
(2b)
(2c)
νCO = –1.3 pKЈa + 1993
By manipulation of Equations 1a and 2b, it can be
shown that the activity of oxidative addition of CH3I ad-
dition to [Rh(RЈCOCHCOR)(CO)(PPh3)] may be estab-
lished as
lnk2 = –0.628 νCO + 1241
The linear trend between Epa of Rh and the sum of the
infrared stretching frequencies, νCO, in [RhI(β-diketonato)- group electronegativities χRЈ + χR in Figure 8 (b, broken
(CO)(PPh3)].[10]
line) indicates that the RhI core becomes increasingly diffi-
Eur. J. Inorg. Chem. 2011, 2439–2449
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2445

J. Conradie, J. C. Swarts
FULL PAPER
Figure 8. (a) Linear dependence (solid black line) between the kinetic parameter k2 {the rate constant of the oxidative addition reaction
between iodomethane and [Rh(β-diketonato)(CO)(PPh3)]; k2 was experimentally determined in ref.[3]} and Epa of Rh in [Rh(β-diketonato)-
(CO)(PPh3)] measured at a scan rate of 100 mV s–1 and 25 °C. Points indicated with an asterisk * refer to the first RhI oxidation peak
(Table 2) and were not used in the linear fits. The broken line represents the relationship between νCO and Epa in the [Rh(β-diketonato)-
(CO)(PPh3)] complexes 1–8. Insert: linear relationship between ferrocenyl formal reduction potentials and k2. (b) The relationship between
Epa(Rh(2)) and the pKЈa of the free β-diketones (solid black line), or the sum of the group electronegativities of RЈ and R (χRЈ + χR) of
the β-diketonato ligand (RЈCOCHCOR)– coordinated to the metal complexes 1–9 (broken line).
Epa(Rh) = 0.191(χRЈ + χR) – 0.547; R2 = 0.96
Epa(Rh) = –0.054 pKЈa + 0.817; R2 = 0.99
Epa(Rh) = 0.056 νCO – 111; R2 = 0.91
(5a)
(5b)
(5c)
(5d)
cult to oxidize as the R groups of the coordinating β-diket-
onato ligands become more electronegative. More electron
density is withdrawn from the rhodium(I) centre by the
(CF3COCHCOCF3)– ligand of 10 than by any other β-diket-
onato ligand in the compound series 1–8 and 10. Apparent
group electronegativities in Gordy scale (in brackets) of the
substituents R increase in the order Fc(1.87) Ͻ C6H5(2.21)
≈ CH3(2.34) Ͻ CF3(3.01),[9,13] which explains the progress-
ive electron deficiency of the metal centre in moving
from [Rh(FcCOCHCOFc)(CO)(PPh3)] (1) to the
[Rh(CF3COCHCOCF3)(CO)(PPh3)] complex 10. The
Epa(Rh)/(χRЈ + χR) relationship of Figure 8 (b) fits the
equation
Epa(Rh) = –0.064 lnk2 + 0.029; R2 = 0.95
Equation array (6), obtained either by combination and
manipulation of the equations given above, and confirmed
by fitting the data of Tables 1 and 2 to a spread sheet fitting
program, allows the estimation of EoЈ(Fc) for [Rh(FcCOCH-
COR)(CO)(PPh3)]; χFc = 1.87 in this equation array.
EoЈ(Fc) = 0.154(χRЈ + χR) – 0.377; R2 = 0.96
EoЈ(Fc) = –0.028 pKЈa + 0.552; R2 = 0.95
EoЈ(Fc) = –0.049 lnk2 + 0.107; R2 = 0.99
(6a)
(6b)
(6c)
(χRЈ+χR) = 5.02Epa(Rh) + 2.95; R2 = 0.96
(3)
The linear relationship between the observed acidity of
the free β-diketonato ligand, pKЈa and Epa, Figure 8 (b),
black solid line, were exceptionally good. Only 4 deviated
slightly from the pKЈa – Epa trend and was ignored in the
data fitting that gave the equation
EoЈ(Fc) could not be estimated accurately from a fitted
equation linking it to νCO, because the data correlation pa-
rameter R2 (0.79) deviated to much from unity.
The cytotoxicity of ferrocene-containing complexes are
frequently dependent on the formal oxidation potential of
the ferrocenyl group. Related to ferrocene-containing
alcohols, it was found that smaller EoЈ values lead to more
favourable (higher) cytotoxicity.[28] In contrast, the free β-
diketones FcCOCH2COR which were the ligands in 1–4
followed exactly the opposite trend.[8b] Two mechanisms by
which the ferrocenyl group destroys antineoplastic growths
were identified. The first was shown to involve homolytic
pKЈa = –18.2Epa(Rh) + 14.9; R2 = 0.99
From ref.[3], it was found that
pKЈa = –3.484(χRЈ + χR) + 24.6
Rearrangement of equation (2c) gives
pKЈa = –0.769 νCO + 1533
(4a)
(4b)
(4c)
Combination and manipulation of the above equation action, i.e. radical induced electron transfer processes[29] be-
sets resulted in equation array 5 as methods to estimate tween a ferrocenium group and water, inter alia to generate
Epa(Rh) for complexes of the type [Rh(β-diketonato)- hydroxy radicals which cleave DNA strands. This implies a
(CO)(PPh3)].
ferrocene-containing drug must, after it is administered to
2446
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2439–2449

Fc-Containing Rhodium(I) Complexes
the human body, first be oxidized by redox-active body en- metal–CO ligand combination will probably cause 2 to be
zymes to the ferrocenium species to show antineoplastic ac- scavenged in vivo by the body’s defense mechanism, the
tivity. Indications are that the cut-off formal reduction po- macrophage from the reticuloendothelial system, thereby li-
tential of the ferrocenyl group where this cannot happen miting its capability to gain access to neoplastic growths in
any more is ca. +0.21 V vs. SCE (saturated calomel refer- vivo.[33]
ence electrode) or 0.52 V vs. Fc/Fc+.[8a,8b] EoЈ values of the
ferrocenyl group of all four complexes 1–4 are less than
0.52 V, Table 1. In the second mechanism, the ferrocenyl Conclusions
group itself acts as a reducing agent when it reduces the
We showed in this study that in acetonitrile, complexes
tyrosyl radical of the R2 subunit of the enzyme ribonucleo-
of the type [Rh(FcCOCHCOR)(CO)(PPh3)] with R = Fc
tide reductase.[30] The active site of dimeric R2 consists of
(1), Ph (2), CH3 (3), or CF3 (4) undergo irreversible two-
a tyrosil radical and two FeIII centers which are μ-oxo
electron electrochemical oxidation of the rhodium(I) centre
bridged. This enzyme catalyses the reduction of ribonucleo-
before ferrocenyl oxidation takes place. Complexes 2–4 hav-
ing unsymmetric β-diketonato ligands, i.e. R ϶ Fc, exist
as two isomers in solution; both isomers exhibits distinct
tides to deoxyribonucleotides, a key step in DNA syntheses,
and it’s inactivation is therefore a goal in chemotherapy.[31]
To investigate the possibility of compounds of the type
electrochemical behavior. Electrochemical observation of
[Rh(FcCOCHCOR)(CO)(PPh3)] having antineoplastic
the rhodium and ferrocene oxidation of both isomers was
properties, the cytotoxicity of 2 was determined against the
possible because of the slow rate of isomerisation between
HeLa cell line (human cervix epitheloid; ATCC CCL-2,
isomers. Isomerisation of 3 was shown to have a half-life of
American Type Culture Collection, Manassas, Virginia,
USA). Cell survival was measured by means of the colori-
143 seconds in CDCl3 at 20 °C utilizing 1H NMR tech-
niques.
metric
3-(4,5-dimethylthiazol-2-yl)-diphenyltetrazolium
An electrochemical study of the chemically prepared rho-
dium(III) complex [RhIII(FcCOCHCOCF3)(CH3)(I)(CO)-
(PPh3)] clearly showed a one-electron rhodium(III) re-
duction process at –1.492 V vs. Fc/Fc+, while RhIII re-
duction of electrochemically generated RhIII complexes of
1–4 was observed at higher potentials. Epa(Rh) of 1–4 is
related to the rate of chemical oxidative addition of CH3I
to 1–4 by Epa(Rh) = –0.064 lnk2 + 0.026. Both the reactivity
of 1–4 towards oxidative addition and Epa(Rh) was shown
to depend linearly on the relative electron richness of the
rhodium nucleus. This led to linear relationships between
complex reactivity expressed as lnk2 and β-diketonato pKЈa
values, complex νCO infrared stretching frequencies. The
cytotoxic determination of 2 showed that [Rh(FcCOCH-
COPh)(CO)(PPh3)] is not significantly more active against
neoplastic growths than the free FcCOCH2COPh ligand
despite the presence of both CO and the rhodium centre.
bromide (MTT) assay and a cell survival curve as a function
of concentration of 2 is shown in Figure 9.
Figure 9. Effect of [Rh(FcCOCHCOPh)(CO)(PPh3)], 2 on the sur-
vival of HeLa human cancer cells after 7 days of incubation mea-
sured as a percentage of untreated controls. Each end-point repre-
sents the mean of three experiments Ϯ standard error of the mean.
The mean drug concentration of 2 from 3 experiments
causing 50% cell growth inhibition, the IC50 value, was
42.2 μmoldm–3. Complex 2 was substantially less cytotoxic
Experimental Section
than cisplatin [Pt(NH)3Cl2], which has IC50
=
2.3 μmoldm–3, but slightly more cytotoxic than the free li-
gand, FcCOCH2COPh, (IC50 = 54.2 μmoldm–3) under
identical conditions.[8b] The lowest IC50 value correspond
to the more active compound. Compound 2 is about half as
active against neoplastic cells as the cyclooctadiene complex
Materials and Cytoxicity Determination: Complexes 1–4[10] and 9[26]
were synthesized as described earlier. Dry acetonitrile for electro-
chemical measurements was obtained by refluxing under nitrogen
over calcium hydride, distillation onto alumina for storage, and re-
distilled just prior to use. The supporting electrolyte, electrochemi-
cal grade [N(nBu4)][PF6] from Fluka, was used as received. Cyto-
toxicity tests were performed as described before.[8b]
[Rh(FcCOCHCOPh)(cod)] which exhibits IC50
=
28.3 μmoldm–3.[32] These results are consistent with the be-
tadiketonato ligand of 2 being responsible for most of the
cell damage. The slightly higher activity of 2 as compared
to the free ligand, FcCOCH2COPh, must arise in part from
the presence of the toxic CO group in 2, but this is not the
only reason. The rhodium centre itself also exhibits neo-
plastic activity as demonstrated for the cod complex.[32] Re-
sults are interpreted to imply that the (CO)(PPh3) ligand
combination of 2 is not as amenable to cell internalization
Electrochemical Measurements: Cyclic voltammetry measurements
on ca. 0.7–1.0 mmoldm–3 solutions of 1–4 and 9 in dry acetonitrile
containing 0.100 moldm–3 tetra-n-butylammonium hexafluoro-
phosphate {[N(nBu4)][PF6], Fluka, electrochemical grade} as sup-
porting electrolyte were conducted under a blanket of purified ar-
gon at 25.0(1) °C utilizing a BAS model CV-27 voltammograph
interfaced with a personal computer. A three-electrode cell, which
utilized a Pt auxiliary electrode, a glassy carbon working electrode
with surface area 7.07 mm2 pre-treated by polishing on a Buehler
as linear or cyclic aliphatics such as cod. Also, the Rh microcloth first with 1 micron and then 1/4 micron diamond paste,
Eur. J. Inorg. Chem. 2011, 2439–2449
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2447

J. Conradie, J. C. Swarts
FULL PAPER
and a Ag/Ag+ reference electrode was used. The reference electrode
was constructed from a silver wire inserted into a solution of
0.0100 moldm–3 AgNO3 and 0.1 moldm–3 [N(nBu4)][PF6] in
CH3CN in a luggin capillary with a vycor tip.[9,34] Data, uncor-
rected for junction potentials, were collected with an Adalab-PCTM
and AdaptTM data acquisition kit (Interactive Microwave, Inc.)
with locally developed software, and analyzed with Hyperplot
(JHM International, Inc.). Successive experiments under the same
experimental conditions showed that all formal reduction and oxi-
dation potentials were reproducible within 5 mV. All cited poten-
tials are reported against the Fc/Fc+ couple as suggested by IU-
PAC,[35] but were experimentally measured against Ag/Ag+. The
Fc/Fc+ couple exhibited under our experimental conditions EoЈ =
0.077 V vs. Ag/Ag+, ipc/ipa = 0.98 and ΔEp = 74 mV. Bulk electro-
lyses were carried out utilising a BAS CV-27 voltammograph at
25.0(1) °C in 2.5 cm3 acetonitrile. A three-electrode cell equipped
with a Pt wire auxiliary electrode [isolated from the sample by me-
ans of a 0.1 moldm–3 (nBu)4NPF6/CH3CN salt bridge], a glassy
carbon working electrode (electro-active areaϮ3 cm2) and the Ag/
Ag+ reference electrode described above were employed. Current
readings and the integrated current (Coulomb units) were recorded
manually at different times, t, during the course of the experiments.
Constructing of decay currents was required due to the overlap of
the different oxidation and reduction peaks. Realistic peak anodic
and cathodic currents of the cyclic voltammograms of complexes
1–4 were determined by constructing a decay current for each over-
lapping peak according to a method described in Figure S1 in the
Supplementary Material.
J. C. Swarts, T. G. Vosloo, S. J. Cronje, W. C. Du Plessis, C. E. J.
Van Rensburg, E. Kreft, J. E. Van Lier, Anticancer Res. 2008,
28, 2781; c) B. Weber, A. Serafin, J. Michie, C. E. J.
Van Rensburg, J. C. Swarts, L. Baum, Anticancer Res. 2004, 24,
763.
W. C. Du Plessis, J. J. C. Erasmus, G. J. Lamprecht, J. Conradie,
T. S. Cameron, M. A. S. Aquino, J. C. Swarts, Can. J. Chem.
1999, 77, 378.
J. Conradie, G. J. Lamprecht, S. Otto, J. C. Swarts, Inorg. Chim.
Acta 2002, 328, 191.
a) A. Auger, J. C. Swarts, Organometallics 2007, 26, 102; b)
W. L. Davis, R. F. Shago, E. H. G. Langner, J. C. Swarts, Poly-
hedron 2005, 24, 1611.
For a discussion of the kinetics of keto-enol conversion of fer-
rocene β-diketones see: W. C. Du Plessis, W. L. Davis, S. J.
Cronje, J. C. Swarts, Inorg. Chim. Acta 2001, 314, 97. A general
treatment of β-diketone tautomers can be found in: R. C.
Mehrotra, R. Bphra, D. P. Gaur, Metal β-diketonates and Allied
Derivates, Academic Press, London, 1978.
W. C. Du Plessis, T. G. Vosloo, J. C. Swarts, J. Chem. Soc., Dal-
ton Trans. 1998, 2507.
a) P. R. Wells, in Progress in Physical Organic Chemistry, John
Wiley & Sons, New York, 1968; vol. 6, p. 111–145; b) R. E.
Kagarise, J. Am. Chem. Soc. 1955, 77, 1377.
[9]
[10]
[11]
[12]
[13]
[14]
[15]
a) H. J. Gericke, N. I. Barnard, E. Erasmus, J. C. Swarts, M. J.
Cook, M. A. S. Aquino, Inorg. Chim. Acta 2010, 363, 2222; b)
D. H. Evans, K. M. O’Connell, R. A. Peterson, M. J. Kelly, J.
Chem. Educ. 1983, 60, 290; c) P. T. Kissinger, W. R. Heineman,
J. Chem. Educ. 1983, 60, 702; d) J. J. Van Benshoten, L. Y.
Lewis, W. R. Heineman, J. Chem. Educ. 1983, 60, 772; e) G. A.
Mobbott, J. Chem. Educ. 1983, 60, 697.
Supporting Information (see footnote on the first page of this arti-
cle): A diagram to explain how peak currents and potentials may
be estimated for closely overlapping peaks, and a Table giving po-
tentials at scan rates of 50–250 mV s–1.
[16]
a) C. Creutz, H. Taube, J. Am. Chem. Soc. 1969, 91, 3988; b)
M. J. Cook, I. Chambrier, G. White, E. Fourie, J. C. Swarts,
Dalton Trans. 2009, 1136; c) N. Van Order, W. E. Geiger, T. E.
Bitterwolf, A. L. Reingold, J. Am. Chem. Soc. 1987, 109, 5680;
d) D. T. Pierce, W. E. Geiger, Inorg. Chem. 1994, 33, 373; e) A.
Auger, A. J. Muller, J. C. Swarts, Dalton Trans. 2007, 3623; f)
W. E. Geiger, N. Van Order, D. T. Pierce, T. E. Bitterwolf, A. L.
Reingold, N. D. Chasteen, Organometallics 1991, 10, 2403.
J. Conradie, T. S. Cameron, M. A. S. Aquino, G. J. Lamprecht,
J. C. Swarts, Inorg. Chim. Acta 2005, 358, 2530.
a) D. Menglet, A. M. Bond, K. Coutinho, R. S. Dickson, G. G.
Lazarev, S. A. Olsen, J. R. Pilbrow, J. Am. Chem. Soc. 1998,
120, 2086; b) F. Barriere, W. E. Geiger, Organometallics 2001,
20, 2133; c) I. Kovacik, H. Gevert, M. Schmittel, R. Söllner,
Inorg. Chim. Acta 1998, 276, 435.
Acknowledgments
[17]
[18]
Financial assistance from the Central Research Fund of the Uni-
versity of the Free State is gratefully acknowledged.
[1] a) P. M. Maitlis, A. Haynes, G. J. Sunley, M. J. Howard, J.
Chem. Soc., Dalton Trans. 1996, 2187; b) L. Cavallo, M. Sola,
J. Am. Chem. Soc. 2001, 123, 12294; c) A. Haynes, P. M.
Maitlis, G. E. Morris, G. J. Sunley, H. Adams, P. W. Badger,
C. M. Bowers, D. B. Cook, P. I. P. Elliot, T. Ghaffer, H. Green,
T. R. Griffin, M. Payne, J. M. Pearson, M. J. Taylor, P. W. Vick-
ers, R. J. Watt, J. Am. Chem. Soc. 2007, 126, 2847; d) J. Jones,
Platinum Met. Rev. 2000, 44, 94.
[19]
[20]
D. E. Richardson, H. Taube, Inorg. Chem. 1981, 20, 1287.
J. Stary, The Solvent Extraction of Metal Chelates, MacMillan,
New York, 1964, p. 196.
M. Ellinger, H. Duschner, K. Starke, J. Inorg. Nucl. Chem.
1978, 40, 1063.
[21]
[22]
[23]
[24]
[2] J. Conradie, G. J. Lamprecht, A. Roodt, J. C. Swarts, Polyhe-
dron 2007, 26, 5075.
D. Lamprecht, G. J. Lamprecht, Inorg. Chim. Acta 2000, 309,
72.
[3] J. Conradie, J. C. Swarts, Organometallics 2009, 28, 1018.
[4] a) J. A. Venter, J. G. Leipoldt, R. van Eldik, Inorg. Chem. 1991,
30, 2207; b) G. J. J. Steyn, A. Roodt, J. G. Leipoldt, Inorg.
Chem. 1992, 31, 3477; c) M. M. Conradie, J. Conradie, Inorg.
Chim. Acta 2008, 361, 2285; d) N. F. Stuurman, J. Conradie, J.
Organomet. Chem. 2009, 694, 259.
[5] F. Spanig, C. Kolvacs, F. Hauke, K. Ohlubo, F. Fukuzumi,
D. M. Guldi, A. Hirsch, J. Am. Chem. Soc. 2009, 131, 8180.
[6] a) A. Klapars, K. R. Clampos, C. Chen, R. P. Volante, Org.
Lett. 2005, 7, 1185; b) Q. Shen, S. Shekhar, J. P. Stambuli, J. F.
Hartwig, Angew. Chem. 2005, 117, 1395; Angew. Chem. Int. Ed.
2005, 44, 1371; c) V. Percec, L.-Y. Bae, D. H. Hill, J. Org. Chem.
1995, 60, 1060.
M. Fatima, G. C. da Silva, A. M. Trzeciak, J. J. Ziolkowski,
A. J. L. Pombeiro, J. Organomet. Chem. 2001, 620, 174.
a) F. Barriere, R. U. Kirss, W. E. Geiger, Organometallics 2005,
24, 48, and references cited therein; b) S. Trupia, A. Nafady,
W. E. Geiger, Inorg. Chem. 2003, 42, 5480; c) F. Barriere, N.
Camire, W. E. Geiger, U. T. Mueller-Westerhoff, R. Sanders, J.
Am. Chem. Soc. 2002, 124, 7262; d) F. Barriere, W. E. Geiger,
J. Am. Chem. Soc. 2006, 128, 3980; e) A. Nafady, T. T. Chin,
W. E. Geiger, Organometallics 2006, 25, 1654; f) D. S. Chong,
J. Slote, W. E. Geiger, J. Electroanal. Chem. 2009, 630, 28.
a) J. C. Swarts, A. Nafady, J. H. Roudebush, S. Trupia, W. E.
Geiger, Inorg. Chem. 2009, 48, 2156; b) E. Fourie, J. C. Swarts,
I. Chambrier, M. J. Cook, Dalton Trans. 2009, 1145; c) K. C.
Kemp, E. Fourie, J. Conradie, J. C. Swarts, Organometallics
2008, 27, 353; d) I. Chambrier, D. L. Hughes, J. C. Swarts, B.
Isara, M. J. Cook, Chem. Commun. 2006, 3504.
[25]
[7] P. J. Swarts, M. Immelman, G. J. Lamprecht, S. E. Greyling,
J. C. Swarts, S. Afr. J. Chem. 1997, 50, 208.
[8] a) J. C. Swarts, D. M. Swarts, M. D. Maree, E. W. Neuse, C.
La Madeleine, J. E. van Lier, Anticancer Res. 2001, 21, 2033; b)
2448
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2439–2449

Fc-Containing Rhodium(I) Complexes
[26] G. J. Lamprecht, J. C. Swarts, J. Conradie, J. G. Leipoldt, Acta
Crystallogr., Sect. C 1993, 49, 82.
Biophys. Acta 1995, 1247, 215; b) J. C. Swarts, A. G. Sykes,
Anti-Cancer Drug Des. 1994, 9, 41; c) J.-Y. Han, J. C. Swarts,
A. G. Sykes, Inorg. Chem. 1996, 35, 4629; d) S. Nyholm, L.
Thelander, A. Gräslund, Biochemistry 1993, 32, 11569; e) E.
Artin, J. Wang, G. J. S. Lohman, K. Yokoyama, G. Yu, R. G.
Griffin, G. Bar, J. Stubbe, Biochemistry 2009, 48, 11622.
[32] J. Conradie, J. C. Swarts, Dalton Trans. 2011, DOI: 10.1039/
C1DT00013F.
[27] For DFT calculated structures of the products of oxidative ad-
dition of CH3I to [Rh(ThCOCHCOR)(CO)(PPh3)] and
[Rh(FcCOCHCOCF3)(CO)(PPh3)], see ref.[4c] and M. M. Con-
radie, J. J. Conradie, S. Afr. J. Chem. 2008, 61, 102.
[28] R. F. Shago, J. C. Swarts, C. E. J. Van Rensburg, Anticancer
Res. 2007, 27, 3431.
[29] D. Osella, M. Ferrali, P. Zanello, F. Laschi, M. Fontani, C.
Nervi, G. Carvigiolio, Inorg. Chim. Acta 2000, 306, 42.
[30] A. Liu, D. N. Leese, J. C. Swarts, A. G. Sykes, Inorg. Chim.
Acta 2002, 337, 83.
[31] a) J. C. Swarts, M. A. S. Aquino, K.-Y. Lam, A. G. Sykes, Bi-
ochim. Biophys. Acta Protein Struct. Mol. Enzymol. Biochim.
[33] M. D. Maree, E. W. Neuse, E. Erasmus, J. C. Swarts, Metal-
Based Drugs 2008, xdoi:10.1155/2008/217573.
[34] A. Auger, J. C. Swarts, Organometallics 2007, 26, 102.
[35] G. Gritzner, J. Kuta, Pure Appl. Chem. 1984, 56, 461.
Received: January 4, 2011
Published Online: April 14, 2011
Eur. J. Inorg. Chem. 2011, 2439–2449
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2449