Synthesis and characterization of mono- and dinuclear phenanthroline- extended tetramesitylporphyrin complexes as well as UV-Vis and EPR studies on their one-electron reduced species

Article abstract of DOI:10.1039/c2dt32196c

The syntheses of mononuclear compounds based on the fused porphyrin phenanthroline ligand (H₂-1) and their corresponding dinuclear porphyrin bis-bipyridine ruthenium complexes are reported. The extended π-system of the ligand is able to store electron equivalents as could be proven by the single electron reduction with KC₈ (and/or Na(Hg)) followed by subsequent UV-Vis and EPR analysis. Electron reduction could also be achieved under light illumination in a dichloromethane-triethylamine mixture. The two coordination spheres of the ligand are different so that mononuclear or (hetero)dinuclear complexes can be isolated depending on the reaction conditions. We could successfully introduce Zn, Cu and Pd into the porphyrinic unit leading to a series of mononuclear (M-1) compounds. Further, we could attach the bis-(4,4′-di-tert-butyl-2,2′-bipyridine) ruthenium fragment (Ru(tbbpy)₂²⁺) to obtain dinuclear (M-1-Ru) metal complexes. In the case of Zn-1 an X-ray crystal structure could be obtained confirming the selective metallation in the porphyrinic unit. All metal complexes were isolated and characterized with standard analytical tools (elemental analysis, mass spectrometry and NMR (or EPR) spectroscopy).

Full text of DOI:10.1039/c2dt32196c

Dalton
Transactions
View Article Online
View Journal
PAPER
Synthesis and characterization of mono- and dinuclear
phenanthroline-extended tetramesitylporphyrin
complexes as well as UV-Vis and EPR studies on their
one-electron reduced species†
Cite this: DOI: 10.1039/c2dt32196c
Corinna Matlachowski and Matthias Schwalbe*
The syntheses of mononuclear compounds based on the fused porphyrin phenanthroline ligand (H2-1)
and their corresponding dinuclear porphyrin bis-bipyridine ruthenium complexes are reported. The
extended π-system of the ligand is able to store electron equivalents as could be proven by the single
electron reduction with KC8 (and/or Na(Hg)) followed by subsequent UV-Vis and EPR analysis. Electron
reduction could also be achieved under light illumination in a dichloromethane–triethylamine mixture.
The two coordination spheres of the ligand are different so that mononuclear or (hetero)dinuclear com-
plexes can be isolated depending on the reaction conditions. We could successfully introduce Zn, Cu and
Pd into the porphyrinic unit leading to a series of mononuclear (M-1) compounds. Further, we could
attach the bis-(4,4’-di-tert-butyl-2,2’-bipyridine) ruthenium fragment (Ru(tbbpy)22+) to obtain dinuclear
(M-1-Ru) metal complexes. In the case of Zn-1 an X-ray crystal structure could be obtained confirming
the selective metallation in the porphyrinic unit. All metal complexes were isolated and characterized
with standard analytical tools (elemental analysis, mass spectrometry and NMR (or EPR) spectroscopy).
Received 21st September 2012,
Accepted 11th December 2012
DOI: 10.1039/c2dt32196c
www.rsc.org/dalton
reduction reactions are of multi-electron nature (e.g. two elec-
tron processes), a linker with a partial electron storage ability
Introduction
Constant rising world energy consumption is met with greater to accumulate those electron equivalents is desired. Several
use of fossil fuels, which correlates to an increase in atmos- other examples for supramolecular assemblies appeared for
pheric CO2.1 Utilization of sunlight offers a very attractive applications like sulfoxidation,13,14 epoxidation,15 alcohol oxi-
alternative fuel source and, thus, the field of photodriven dation,16,17 water oxidation or anchoring to semiconductor
supramolecular catalysis became more and more important in surfaces (e.g. dye sensitized solar cells).18–22
recent years.2,3 Heterooligonuclear complexes were developed,
Extended π-systems as found in the ligands tetraazatetrapyri-
which consist of a light harvesting unit, a linker and a reaction dopentacene (tatpp)23–25 and tetrapyridophenazine (tpphz)10,26,27
centre, and were able to control selectivity in organometallic proved to be very good for electron storage of up to four elec-
catalysis (e.g. CO2 or proton reduction) with the help of trons (for tatpp). Unfortunately, the synthesis of heterodinuc-
light.4–7 The linker has a very crucial role as it should allow for lear metal complexes proved to be difficult because of
the connection of and communication between different metal solubility issues and the same phenanthroline based coordi-
centers. Additionally, the linker should possess electron nation sphere on both ends of the molecules (selectivity pro-
storage capacity for specific applications like proton blems). Nevertheless, several heterodinuclear complexes with
reduction.8–12 The photoelectron created on the photosensiti- the ligand tpphz and its derivatives were synthesized and suc-
zer side needs to be transferred via the linker to the catalytic cessfully tested in the photodriven proton reduction.10,28,29
side where the reduction of the substrate takes place. As many Most of the systems mentioned so far contain the chromopho-
2+
ric Ru(bpy)2 unit. In order to use a cheaper system, por-
phyrin based chromophores are very attractive.
Institute of Chemistry, Humboldt-Universität zu Berlin, Brook-Taylor-St. 2, 12489
Crossley et al. used the tatpp motive to connect two
porphyrinic units.30,31 For solubility reasons they used 3,5-di-
Berlin, Germany. E-mail: matthias.schwalbe@hu-berlin.de; Fax: +49-30-2093-6966;
Tel: +49-30-2093-7571
†Electronic supplementary information (ESI) available: Synthesis of Cu com-
tert-butylphenyl groups instead of phenyl groups in the meso-
position of the porphyrinic unit. They could show that both
porphyrin moieties are in electrochemical communication and
plexes and additional spectroscopic data. CCDC 888471. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/c2dt32196c
This journal is © The Royal Society of Chemistry 2012
Dalton Trans.

View Article Online
Paper
Dalton Transactions
that multiple reversible reduction events occur. Hence, those provides less steric bulk around the phenanthroline coordi-
systems should be suitable for multi-electron reduction cata- nation sphere and allows for complex formation of sterically
lysis. However, no (photo)catalytic application was tested for crowded metal precursors in a second reaction step. We found
these systems. In further studies, different groups showed that that the reaction sequence to M-1-Ru is not only suitable for
a porphyrin moiety could be combined with a phenanthroline copper porphyrins but also for palladium porphyrins (see the
moiety to allow for connection of two different metal Experimental section for further details). This is the first time
centers.32–34 But besides their spectroscopic characterization that a palladium porphyrin system is used for the described
no catalytic application was tested. Later, Liu and coworkers modifications and similar overall yields are obtained com-
used the supramolecular phenanthroline-extended tetraphe- pared to the copper analogue. The reaction sequence starts
nylporphyrin ligand to investigate a series of metalloporphyrin with the synthesis of H2TMP (tetramesitylporphyrin) following
phenanthroline Pd complexes M-Porphen-Pd, with M = Ni, Zn, the Lindsey method.42 H2TMP is then metalated with either
Mg, Cu and Co, in the Heck-reaction. The metal center in the copper or palladium. If other metal centers are used (e.g. Zn,
porphyrin moiety had an influence on the catalytic perform- Mg or Ni), the third step, the nitration in the beta-position,
ance but no influence of light illumination on the product dis- does not occur selectively. Thus, a suitable central metal ion is
tribution could be determined.35 Albeit multiple spectroscopic absolutely necessary for this reaction. Reduction of the nitro
investigations concern the reduced state of porphyrin com- group with sodium borohydride leads to the amino derivative,
pounds, applications in catalytic reactions are rare. Examples which is slightly unstable against photo-oxidation, but could
are restricted to systems containing a redox-active metal center be handled on air for a certain time. It is not possible to purify
like iron or cobalt in the porphyrin cavity. Those are able to the amino product with silica column chromatography, since
drive the reduction of protons, dioxygen or CO2 if the metal silica catalyzes the photo-oxidation of the amino compound.
center is two- or three-fold reduced by chemical, electrochemi- Hence, the amino compound is subsequently oxidized with air
cal or photochemical means.36–41
under light illumination (bulb light – 250 W). MTMCO2 (M =
An interesting question that was not answered for the por- Cu or Pd) can be isolated in good yield and purity as the
phyrin phenanthroline complexes so far is whether they are photoproduct after chromatographic purification. Interest-
able to store electrons, which can be used in subsequent cata- ingly, this oxidation proceeds about 24 times faster for the pal-
lytic transformation. Following, we will present the preparation ladium complex than for the corresponding copper complex
of the newly phenanthroline-extended tetramesitylporphyrin and is already completed after 2 hours. Palladium porphyrins
ligand H2-1 as well as its mononuclear metal complexes with are probably better sensitizers for singlet-oxygen, which is
Zn, Cu and Pd as the central metal atom in the porphyrin needed for this reaction.43 The acid-catalyzed condensation of
coordination sphere. The phenanthroline moiety of these com- the dioxo compound with 1,10-phenanthroline-5,6-diamine
plexes allows for the coordination of a second metal center (phendiamine) leads to the metalloporphyrin phenanthroline
2+
like Ru(tbbpy)2 which should (a) broaden the absorption complex (M-1) in very good yields. The last step, the complexa-
properties in the visible region and (b) decrease the reduction tion of the phen-moiety of M-1 with bis-(4,4′-di-tert-butyl-2,2′-
potential of the whole complex for thermodynamically more bipyridine) ruthenium(II) dichloride, is conducted under
suitable CO2-reduction. Hence, we present for the first time microwave irradiation for the palladium complex to get Pd-
2+
the combination of a Ru(tbbpy)2 chromophore with a H2-1 1-Ru in high yield. In contrast, the corresponding copper
system to yield dinuclear complexes of the type M-1-Ru. The complex is prepared applying reflux in DMF for prolonged
structural and spectroscopic properties of these metal com- time to give Cu-1-Ru in quantitative yield. Under microwave
pounds were analyzed in detail. Further, the single electron irradiation, more side reactions occur in the case of Cu-1-Ru
reduced species were isolated, characterized with EPR and which were not further investigated and just led to reduced
UV-Vis spectroscopy, and preliminary photochemical CO2- yield. Finally, M-1-Ru is synthesized with an overall yield of
reduction experiments were done.
51% (M = Cu) or 47% (for M = Pd), respectively, based on
H2TMP.
Another synthetic strategy is necessary to obtain M-1 com-
pounds with M different to Cu or Pd, e.g. M = Zn (Fig. 2). With
zinc as the metal center the nitration of the porphyrin in the
beta-position is not selective. Because of this, Karpishin deme-
Results and discussion
Synthesis and characterization
The synthesis of the compounds Cu-1-Ru and Pd-1-Ru is out- talated the corresponding copper complex after condensation
lined in Fig. 1. In the case of Cu-1, the precursor of Cu-1-Ru, a of the dioxochlorin derivative and 2,9-dimethyl-1,10-phenan-
slight variation of the reaction conditions published by Kar- throline-5,6-diamine.33 However, we found that this procedure
pishin and coworkers33 was used (see ESI† for further details) is not practicable since the demetalation of Cu-1 with sulfuric
and higher yields could be obtained for the photo-oxidation acid/trifluoro acetic acid does not proceed quantitatively and
step by using bulb light instead of sunlight. Moreover, Kar- selectively in our hands. It becomes impossible to separate the
pishin’s copper complex included a 2,9-dimethyl-1,10-phenan- metal free porphyrin from the metalated species. Thus, we
throline moiety instead of the simple phenanthroline moiety decided to demetalate the copper porphyrin in an earlier
in our system. Although this is only a marginal modification it step of the reaction sequence as Liu and coworkers did.35
Dalton Trans.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Dalton Transactions
Paper
Fig. 1 Reaction sequence for the synthesis of M-1-Ru (M = Cu or Pd).
Fig. 2 Reaction sequence for the synthesis of Zn-1.
H2TMP-NO2 can easily be separated by silica column chro- studies below). Further, all compounds were characterized by
matography from the corresponding copper complex. The follow- MALDI- or ESI-mass spectrometry (MS) as well as elemental
ing steps to the synthesis of metal free H2-1 proceed smoothly analysis (EA). All experimental results were in agreement with
and in analogy to the metalloporphyrins. H2-1 is obtained in the calculated ones and confirmed the molecular composition
1
an overall yield of 49% starting from CuTMP. The selective of the products. The H-NMR-spectra in CD2Cl2 of Zn-1, Pd-1
complexation of the porphyrin sphere with metal like zinc is (Fig. 3) and H2-1 are quite similar. The missing NH-signal at
possible while keeping the phenanthroline moiety metal free. −2.37 ppm in the spectra of Pd-1 and Zn-1 confirmed the pres-
The reaction of H2-1 with zinc acetate and the treatment of the ence of the metal in the porphyrin core. The chemical shifts of
reaction mixture with EDTA allows for the isolation of Zn-1 in the pyrrolic H-atoms (H-e, H-d, H-f) seem to have no corre-
65% yield. The complete synthesis of H2-1 and Zn-1 is illus- lation to the central metal. No trend can be determined and
trated in Fig. 2.
the values are close to each other for the three considered
The characterization of H2-1, Zn-1, Pd-1 and Pd-1-Ru was compounds (Table 1; note that the signals H-e and H-d could
carried out amongst others by NMR-spectroscopy, but not for not be assigned unambiguously to their respective positions in
Cu-1 or Cu-1-Ru due to its paramagnetism. Cu-1 and Cu-1-Ru the molecule and might be interchanged). All other 1H-NMR
were instead characterized by EPR-spectroscopy (see EPR signals show basically no shift upon changing the metal
This journal is © The Royal Society of Chemistry 2012
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 4 The molecular structure of Zn-1 (solvent molecules and hydrogen atoms
are omitted for clarity).
Table 2 Selected bond lengths and bond angles of Zn-1·CH3OH
Fig. 3 The 1H-NMR spectrum of Pd-1 in CD2Cl2.
Bond lengths [Å]
Bond angles [°]
Table 1 1H-NMR-data of pyrrolic H-atoms of H2-1, Zn-1 and Pd-1 in CD2Cl2 as
well as Pd-1-Ru in CD3CN (*)
Zn1–N1 2.062(4) N1–Zn1–N2 89.12(15) N1–Zn1–O1 101.41(18)
Zn1–N2 2.033(4) N2–Zn1–N3 88.57(15) N3–Zn1–O1 93.35(19)
Zn1–N3 2.103(4) N3–Zn1–N4 88.56(15) N1–Zn1–N3 165.20(16)
Zn1–N4 2.037(4) N4–Zn1–N1 89.59(16) N2–Zn1–N4 163.77(16)
Zn1–O1 2.080(5)
1H-NMR data [ppm]
Compound
H-e
H-d
H-f
H2-1
Zn-1
Pd-1
Pd-1-Ru*
8.91
8.85
8.75
8.65
8.82
8.79
8.68
8.69
8.57
8.68
8.64
8.60
remaining ones (2.033(4) Å and 2.037(4) Å). This has also
been observed in the crystal structures of the related species
zinc tetraphenylporphyrin-phenanthroline palladium-dichloride
(ZnTPPphenPdCl2) from Liu35 and iron(III) tetraphenylpor-
phyrin quinoxaline from Wojaczyński.44
center in the porphyrin core. The 1H-NMR-spectrum of Pd-
1-Ru (see Fig. S1†) in CD3CN is similar to that of Pd-1 consider-
ing the porphyrinic signals. Besides, the o-methyl-groups of
the mesityl-substituents are no longer equal because of the
influence of the attached ruthenium fragment. The chemical
shifts of the phen-protons H-a, -b and -c are influenced as well
by the close proximity of the ruthenium moiety. Noteworthy is
the strong shift of H-a from 9.04 ppm (for Pd-1) to 8.15 ppm
(for Pd-1-Ru). H-c is also shifted to lower values (from
9.27 ppm to 8.96 ppm) in contrast to H-b whose chemical shift
stays nearly the same.
The zinc atom is located 0.28 Å above the plane of the four
porphyrinic nitrogen atoms. Similarly, the zinc atom in Liu’s
complex35 is displaced 0.32 Å out of the plane toward the axial
oxygen atom of a coordinated DMSO molecule (Zn–O = 2.090
(4) Å). For Zn-1 the dihedral angle between the phen-moiety
and the modified pyrrole ring equals 14.4° which is 11° higher
than in Liu’s complex. This difference is due to the palladium
complexation of the phen-moiety in Liu’s complex. Neverthe-
less, the porphyrin plane of Zn-1 is almost planar. The pyrrole
rings are barely displaced alternatively above or below the
mean porphyrin plane with a mean difference of 0.1 Å. Both
mesityl substituents that are close to the phen-moiety are basi-
cally orthogonal to the porphyrin plane (∼85°), but the angle
between the other mesityl substituents and the porphyrin
plane is slightly smaller (∼77°).
Crystal structure of Zn-1
A single crystal X-ray diffraction study of Zn-1 confirmed the
formation of the desired complex with the zinc metal ion
residing in the porphyrin coordination sphere and the phen-
anthroline-moiety staying metal free. X-ray quality crystals
Electrochemistry
were obtained by covering a solution of Zn-1 in dichloro- Electrochemical investigations were done in dichloromethane
methane (DCM) with a layer of methanol. The crystal structure solutions using tetrabutylammonium hexafluorophosphate
of Zn-1 (Fig. 4) includes two molecules of methanol; one is (TBAP) as an electrolyte. Table 3 lists the redox potentials of
coordinated to the metal center and the non-coordinated one H2TMP, H2-1 and of the series of their metal complexes as well
is incorporated into the unit cell. Selected bond lengths and as of [Ru(tbbpy)2phen](PF6)2. The metalated TMP compounds
angles are listed in Table 2. The zinc atom resides in an showed two oxidation and one reduction event. All these pro-
approximate square pyramidal geometry. The four nitrogen cesses are based on the macrocycle. The M-1 compounds
atoms of the porphyrin macrocycle form the base and one showed two oxidation and two reduction events (Fig. 5). The
oxygen atom of a methanol molecule forms the top (Zn1–O1 = first oxidation event is shifted to higher values in comparison
2.080(5) Å). Interestingly, the metal–nitrogen distances to the to the corresponding TMP complexes due to the existence of
substituted pyrrole (Zn1–N3 = 2.103(4) Å) and to the opposite more electron-withdrawing nitrogen atoms in the system.
one (Zn1–N1 = 2.062(4) Å) are slightly longer than to the two Interestingly, the second oxidation event does not change at all
Dalton Trans.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Dalton Transactions
Paper
Table 3 Redox potentials for the prepared MTMP, M-1 and M-1-Ru compounds in DCM (*: in ACN) containing 0.1 M TBAP (EN = electronegativity)
Compound
Potential vs. Fc/Fc+ [V]
H2TMP46
−1.86
−2.04
−2.00
−1.98
−2.01
−2.25
−2.21
−2.25
−1.91
−1.81
−1.85
−1.80
0.45
0.33
0.45
0.57
0.59
0.44
0.62
0.67
0.60
0.61
0.76
ZnTMP (ENZn = 1.6)
CuTMP (ENCu = 1.9)
PdTMP (ENPd = 2.2)
H2-1
Zn-1
Cu-1
Pd-1
Cu-1-Ru
Cu-1-Ru*
Pd-1-Ru
0.70
0.88
1.11
−1.74
−1.80
−1.78
−1.80
−1.54
−1.40
−1.50
0.70
0.89
−2.11
−1.91
−2.09
−2.08
0.88
0.86
0.91
0.84
−2.31
−2.10
[Ru(tbbpy)2phen](PF6)2
properties that show the largest HOMO–LUMO gap for the Pd
complexes. The second oxidation of the complex Pd-1 could
not be observed in dichloromethane. Unfortunately, it was not
possible to use a different solvent for electrochemical investi-
gations because of solubility issues with the M-1 compounds.
The dependence of the oxidation potential on the central
metal was reported by Crossley and coworkers, too.45 They
investigated porphyrin quinoxaline complexes with zinc,
copper, nickel or palladium as a central metal and could
observe the same relation between the oxidation potential and
the electronegativity of the central metal ion as we did. Thus,
this fact seems to be a general property of (fused) porphyrin
systems. Similar behavior could be observed for the two oxi-
dation processes of the complexes M-1-Ru (M = Cu, Pd).
Fig. 5 Cyclovoltammogram and square-wave-voltammogram of Cu-1 in
CH2Cl2 containing 0.1 M TBAP.
UV-Vis spectroscopy
going from MTMP to M-1 for M = Cu and Zn. In the case of
Pd-1 a second oxidation event could not be observed probably
due to restrictions of the solvent window. Surprisingly, the
reduction potentials are nearly insensitive to the metal center.
Similar observations could be obtained for M-1-Ru (M = Cu,
Pd; Fig. S2†); however, there are two oxidation events and three
reduction events occurring in the CV. In comparison to the
redox potentials of [Ru(tbbpy)2phen](PF6)2 it is clear that the
RuII/RuIII couple overlays with the second porphyrin oxidation
(in the CV those two merge but they are slightly better resolved
in the square wave voltammograms presented in Fig. S3†) and
one tbbpy-centered reduction event adds to the two H2-1-based
reduction events. The second tbbpy-centered reduction lies
probably outside of the solvent window. (If acetonitrile (ACN)
is used as the solvent, five reduction events can be observed.)
With regard to the M-1 compounds, the first, very likely phen-
azine-centered reduction, is shifted to higher values due to
ruthenium complexation. The second (and even the third)
reduction of M-1-Ru occurs also at higher potentials than the
second reduction of M-1.
The UV-Vis absorption spectra of the MTMP, M-1 and M-1-Ru
compounds synthesized were measured in dichloromethane
and the respective maxima are summarized in Table 4. They
all represent typical porphyrin spectra with a strong absorption
band in the 420 nm region (Soret-band) and weaker bands in
the 500–650 nm region (Q-bands). Interestingly, the different
metal centers do not seem to have a great influence on the
position of the Soret-band. In contrast, the position of the
Q-bands depends strongly on the metal center. The Soret-band
structure of M-1 (M = Zn, Cu, Pd) differs significantly from
that of metal-free H2-1 (Fig. 6 (left)). It is split into two
relatively sharp peaks around 410 and 440 nm. Besides, the Soret-
bands of compounds H2-1 and M-1 are broadened in compari-
son with those of the parent MTMP compounds (see Fig. S4†),
since the π-system of the porphyrin is extended into the phen-
moiety. This also explains the bathochromic shift of the Soret
band by 18 nm going from H2TMP to H2-1. The Q-band region
is very sensitive to (a) the metal center and (b) the macrocycle.
In the case of CuTMP and PdTMP there is only one intense
band detectable at 540 nm and 526 nm while Cu-1 and Pd-1
show three Q-bands.
The influence of the different metal centers on the oxi-
dation potentials is clearly seen in a way that the more electro-
negative elements cause higher values of the redox potentials.
Thus, the Zn complexes are the easiest to oxidize and the Pd
complexes the hardest (with more than 0.2 V difference). This
observation is also in line with the UV-Vis absorption
Nevertheless, in these macrocyclic series a trend in the
Q-band region can be observed that shows the most hypso-
chromic shifts for the palladium complexes and the most
bathochromic shifts for the zinc complexes. In other words,
This journal is © The Royal Society of Chemistry 2012
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 4 UV-Vis absorption maxima and emission maxima of prepared MTMP, M-1 and M-1-Ru compounds in DCM (*: in ACN)
Absorption maxima [nm] (ε/1000 [L (mol cm)−1])
Emission
Compound
Soret-band
Q-bands
maxima [nm]
H2TMP33
ZnTMP33,47
CuTMP
PdTMP
H2-1
Zn-1
Cu-1
Pd-1
418(427)
423(603)
416(582)
419(331)
436(116)
412(81)
408(109)
403(90)
424(116)
416(128)
416(102)
514(18)
547(65)
540(24)
526(29)
527(7.4)
523(2.0)
525(6.5)
507(4.9)
583(18)
573(20)
562(22)
546(86)
585(26)
590(5.5)
646(2.8)
650, 718
595, 645
—
—
655, 726
615, 673
—
—
—
—
—
600(1.9)
570(14)
563(18)
545(24)
633(sh, 4.1)
612(sh, 3.4)
446(90)
444(94)
438(110)
472(68)
455(87)
460(69)
608(3.3)
601(8.2)
579(12)
Cu-1-Ru
Cu-1-Ru*
Pd-1-Ru
Fig. 6 UV-Vis absorption spectra in DCM of H2-1 and M-1 (M = Zn, Cu, Pd) (left), Cu-1 (dotted line) and Cu-1-Ru (solid line) (right).
the HOMO–LUMO gap increases in the order of Zn < Cu < Pd,
which is reflected in the electrochemical properties too (see
UV-Vis spectroscopy under reducing conditions
above).
The Soret-band of M-1-Ru (M = Cu, Pd) is shifted further
Stepwise reduction of M-1 (M = Zn, Cu, Pd) was investigated by
UV-Vis spectroscopy. To the best of our knowledge this is the
first time that chemical reduction was performed on these
kinds of complexes. M-1 was dissolved in THF (ACN could not
be used because of low solubility) and an excess of sodium
amalgam (5% sodium) was added. This mixture was stirred for
several hours and absorption spectra were measured every
15 minutes. The reduction reaction was slow enough to see
clear changes in the UV-Vis spectra with several isosbestic
points. Changes of the Soret- and Q-bands could be observed
and are basically equal for all three compounds. Spectral
changes are exemplarily illustrated for Pd-1 in Fig. 7 and are
discussed in the following. First, Pd-1 is reduced to its mono-
reduced species that is characterized by a diminishing Soret-
band forming a new broad maximum centered at 428 nm and
the appearance of three new bands at 587 nm, 622 nm and
676 nm in the Q-band region. Isosbestic points occur at
387 nm, 418 nm and 461 nm. Crossley observed similar
changes in the UV-Vis spectra for the electrochemical one-elec-
tron reduction of quinoxalino-porphyrins.48 The formation of
the mono-reduced species could also be confirmed by the
selective reduction of Pd-1 with only one equivalent of Na(Hg)
in THF and measuring the corresponding UV-Vis spectrum.
bathochromic and broadened compared to the M-1 com-
pounds and its shape is different (see Fig. 6 (right)). The Soret-
band is also split into two peaks where the peak at a lower
wavelength has a higher intensity than the one at a higher
wavelength. A shoulder around 500 nm appears that can tenta-
tively be assigned to MLCT transition(s) of the ruthenium
moiety. Furthermore, a band at 290 nm arises that represents
π–π*-transitions of the bipyridine ligands of the ruthenium
moiety.
Emission spectra can only be obtained for metal free
H2TMP and H2-1 as well as their zinc complexes. In the case of
copper and palladium as central metals the emission is com-
pletely quenched. H2-1 shows emission maxima at 655 and
726 nm while Zn-1 exhibits emission maxima at 615 and
673 nm after excitation in either the Soret- or the Q-band
region. The emission maxima of Zn-1 are blue-shifted and the
intensity is smaller than for H2-1 if a similar concentration of
the compounds is used. Karpishin and coworkers observed
the same trend for H2-1Me2 and Zn-1Me2.33
A similar
trend is observed for the emission maxima of H2TMP and
ZnTMP too.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Dalton Transactions
Paper
Fig. 7 Spectral changes upon mono-reduction of Pd-1 with Na(Hg) in THF (left) and upon further reduction of the mono-reduced species to the double-reduced
species (right) (dashed line: Pd-1, dotted line: mono-reduced species of Pd-1 ((Pd-1)−), dashed dotted line: double reduced species of Pd-1 ((Pd-1)2−), solid lines:
UV-Vis spectra were measured every 15 minutes and the selected ones are shown to illustrate the progress of reaction).
The double-reduced species was formed upon further reaction
with Na(Hg) and shows a thinner and blue-shifted Soret-band
at 416 nm with slightly higher molar absorptivity than the
mono-reduced species. Two new bands at 596 nm and 651 nm
in the Q-band region are formed while the maxima corres-
ponding to the mono-reduced species vanish. Here, isosbestic
points are observed at 398 nm and 424 nm. Qualitatively
Table 5 UV-Vis absorption maxima of M-1 (M = Zn, Cu, Pd) and their mono-
and double-reduced species in THF
Absorption maxima [nm]
Compound
Soret-band(s)
Q-bands
Zn-1
432
426
429
409
420
419
401
428
416
451
447
438
535
576
428
526
564
520
506
505
513
573
518
550
565
607
621
543
587
596
617
643
615(br)
603
638
706
578
622
651(br)
(Zn-1)−
678(br)
713
similar changes in the UV-Vis spectra during the reduction (Zn-1)2−
Cu-1
with Na(Hg) could also be observed for Zn-1 and Cu-1 and are
summarized in Table 5. Equal numbers of Q-bands are
(Cu-1)−
(Cu-1)2−
detected for the series of mono-reduced and double-reduced Pd-1
(Pd-1)−
(Pd-1)2−
676
species.
Illumination of a THF solution of Pd-1 containing 5% TEA
with LED light (400–800 nm) shows fast changes in the UV-Vis
spectrum (see Fig. S5†). Comparing the spectral appearance to
the ones obtained from chemical reduction leads to the con-
clusion that the major species formed is the double-reduced
species. However, the UV-Vis spectrum of the chemically
double-reduced species does not match exactly with the spec-
trum after illumination. Thus, the UV-Vis spectrum after illu-
mination is produced very likely from a mixture of multiple
species including the mono- and double-reduced form as well
as not identified decomposition products.
Mono-reduction of M-1-Ru (M = Cu, Pd) using Na(Hg) was
also investigated by UV-Vis spectroscopy – besides spectroelec-
trochemistry as discussed later. Acetonitrile was chosen as a
solvent, because the reduction proceeded too fast in THF to be
monitored by conventional UV-Vis-spectroscopy. M-1-Ru was
dissolved in ACN and an excess of sodium amalgam (5%
sodium) was added. Changes of the Soret- and Q-bands could
be observed and are basically equal for the two compounds.
Cu-1-Ru was reduced to its mono-reduced species that is
characterized by a diminishing Soret-band at 416 nm in favor
of a new (split) Soret-band at 436 and 405 nm (see Fig. 8 (left)
and Table 6). Clear isosbestic points are observed at 405 nm,
429 nm, 453 nm and 510 nm. Furthermore, three new bands
in the Q-band region appear at 550 nm, 586 nm and 711 nm –
the last one being characteristic of porphyrinic π-radical
anions.49 Similar observations could be obtained for the
corresponding Pd-complex. There are two new bands in the
Soret-band region at 399 nm and 436 nm. Isosbestic points are
detected at 399 nm, 420 nm, 445 nm, 497 nm, 543 nm and
557 nm. In addition, three new bands in the Q-band region
appear at 530 nm, 564 nm and 710 nm.
We believe that the first reduction event is centered on the
bridging-ligand as is the case for M-1. Further reduction of
M-1-Ru to the di- or triple-reduced species is complicated
because of the close proximity of the second and third
reduction events (see above). Thus, double or triple reduced
species could not be identified unambiguously by UV-Vis spec-
troscopy during the reduction process – partly due to miss-
ing isosbestic points after further reduction of mono-reduced
M-1-Ru.
UV-Vis spectroelectrochemistry
Spectroelectrochemical investigations were done in acetonitrile
solutions using tetrabutylammonium hexafluorophosphate
(TBAP) as an electrolyte. Measurements in dichloromethane
were not possible with our setup due to the low electric con-
ductivity of this solvent. Unfortunately, the M-1 complexes are
only marginally soluble in acetonitrile, which prevented a
profound investigation. Therefore, we could only investigate
This journal is © The Royal Society of Chemistry 2012
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 8 Spectral changes of Cu-1-Ru upon either mono-reduction with Na(Hg) in ACN (left) or first electrochemical reduction in ACN containing 0.1 M TBAP (right).
Table 6 UV-Vis absorption maxima of M-1-Ru (M = Cu, Pd) and their mono-
reduced species in ACN
Absorption maxima [nm]
Compound
Soret-band
Q-bands
Cu-1-Ru
416
405
409
399
455
436
446
436
571
548
551
530
612(sh)
588
584
(Cu-1-Ru)−
Pd-1-Ru
711
710
(Pd-1-Ru)−
564
Cu-1-Ru with spectroelectrochemistry (redox potentials are
listed in Table 3). Two more reduction events could be
observed in ACN compared to DCM solutions, because the
solvent window of ACN is larger. The first reduction at −1.40 V
of Cu-1-Ru was examined by UV-Vis spectroelectrochemical
measurements. Therefore the voltage was kept at −1.55 V and
the changes in the UV-Vis spectrum during this reduction are
monitored. They are essentially equal to those recorded for the
sodium amalgam reduction to the mono-reduced species (see
Fig. 8). The UV-Vis spectra at the end of both processes are
essentially the same (see Fig. S6†) with a prominent feature at
Fig. 9 Recorded and simulated EPR-spectrum of Cu-1/Cu-1-Ru. (* internal
711 nm being characteristic of a porphyrinic π-radical anion. standard: MgO : Cr3+, g’ = 1.9796).
Unfortunately, no further comment on the spectroscopic
changes during the second or third reduction can be made as
both overlay each other and, hence, no unambiguous assign-
ment is possible (similar to the chemical reduction process).
2+
Ax(Cu) ≠ Ay(Cu)). The additional Ru(tbbpy)2 fragment has no
influence on the EPR parameters.
Ligand 1 was designed to function as an electron storage
device. Thus, we investigated the possibility of storing an elec-
tron with the help of EPR-spectroscopy. One electron reduction
was done either chemically using KC8 or Na(Hg) or photoche-
mically using triethylamine (TEA) as the sacrificial electron
donor. First, we used zinc as a redox inactive central metal in
the porphyrin unit. Zn-1 was reduced by one equivalent of KC8
(or Na(Hg)) in THF. The solution was filtered subsequently to
abstract the formed graphite (or mercury) and the EPR-spec-
trum of this solution at 77 K showed a ligand based organic
radical with g = 2.003 which is very close to the value of a free
electron (g = 2.0023). The same signal could be obtained for
Pd-1; the g-value is also 2.003. Surprisingly, Crossley et al.
observed for their quinoxalino-porphyrins hyperfine coupling
to the nitrogen atoms of the porphyrin sphere,48 which we do
Electron paramagnetic resonance studies
EPR investigations were done in THF or DCM and measured at
77 K. The recorded and simulated EPR-spectra of Cu-1 are
shown in Fig. 9. The spectrum of Cu-1-Ru is equal to that of
Cu-1 and clearly demonstrates an axial CuII-signal, which
shows a hyperfine coupling with the four nitrogen atoms of
the porphyrin unit. We also determined g-values and coupling
constants for CuTMP, which agree well with reported values
(Table 7).50 The parameters for both compounds, Cu-1 and
CuTMP, are quite similar indicating that the phenanthroline
moiety has almost no influence on the EPR-spectra, except
that the parameters are barely more anisotropic (gx ≠ gy,
Dalton Trans.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Dalton Transactions
Paper
Table 7 EPR-parameters from simulated spectra of CuTMP and Cu-1/Cu-1-Ru
Compound
gx
gy
gz
Ax(Cu)a
Ay(Cu)a
Az(Cu)a
A(N)a
CuTMP50
Cu-1/Cu-1-Ru
2.031
2.053
2.031
2.057
2.185
2.190
26
17
26
20
208
190
16
15
a Hyperfine coupling constants in units of 10−4 cm−1
.
Fig. 10 Left: EPR-spectrum of photochemically mono-reduced Pd-1/Pd-1-Ru. Right: EPR-spectrum of chemically mono-reduced Pd-1/Pd-1-Ru (* internal standard:
MgO : Cr3+, g’ = 1.9796).
not observe in the case of our phenanthroline appended por-
Thus, the chemical and photochemical electron transfer
phyrins. In contrast, no organic radical based signal could be on the metalloporphyrin-phenanthroline system could be
detected when Cu-1 or Cu-1-Ru was reacted with one equival- confirmed and in this way also its electron storage capability.
ent of KC8. Instead, the copper-based signal nearly disap-
Fujita et al. could show that iron and cobalt porphyrins are
peared in both cases indicating an interaction of the unpaired able to reduce CO2 to CO.40,41 Currently, we are investigating
electron of copper with a possibly formed ligand based whether the reduced M-1 and M-1-Ru species are also able to
radical. Another possibility is that CuII is reduced to CuI which reduce CO2 or are able to drive other organic transformations.
is EPR-silent. But we disfavor this possibility for two reasons:
Preliminary CO2 reduction experiments with Cu-1 and
(a) the very similar electrochemical behavior of Cu-1 compared Cu-1-Ru in a DMF (or ACN)–triethylamine solution showed no
to Zn-1 and Pd-1 (no indication of an additional peak for the CO2 reduction, i.e. no CO formation (for experimental details
CuII/CuI couple is noticeable); thus, for M-1 the first reduction see ESI†). But this can be due to the fact that the metal center
should be centered on the ligand 1; and (b) we do not see any is not involved in the first reduction event. Besides, no
formation of free H2-1 in the ESI-MS which would result from changes in the UV-Vis spectrum of Cu-1 or Cu-1-Ru could be
the loss of CuI out of the porphyrinic pocket.51 To the best of observed during illumination with LED light in THF (or ACN)–
our knowledge there is no CuI porphyrin compound presently triethylamine and under CO2 atmosphere. Small changes are
known in the literature – very likely due to its instability.52
only observed after prolonged illumination, but no products
The photochemical electron transfer was investigated for could be isolated or characterized (e.g. by mass spectrometry)
M-1 (M = Zn, Pd, Cu), Pd-1-Ru and Cu-1-Ru. Because of very so far. Illumination of Pd-1 and Pd-1-Ru under the same con-
low solubility of the M-1 compounds in acetonitrile or THF, ditions caused strong changes in the UV-Vis spectrum, but
dichloromethane was used as a solvent. 5% TEA was added to also no reduction products could be identified so far and
the solution of M-1 or M-1-Ru in DCM and this solution was investigation of the reaction products is still in progress. In the
illuminated for 5 min with LED light (ca. 400–800 nm) and future we will investigate whether M-1 compounds with redox-
afterwards rapidly cooled to liquid nitrogen temperature. The active metal centers are photocatalysts for CO2 reduction.
EPR-spectrum at 77 K showed a ligand based organic radical
with g = 2.003 for Zn-1, Pd-1 and Pd-1-Ru which is exemplarily
illustrated in Fig. 10 for Pd-1. The EPR-signal was not quanti-
fied but the radical signal very likely does not represent all
Conclusions
reduced species present in solution. For Cu-1 and Cu-1-Ru the
double integral of the CuII-signal diminished, but again no
ligand based radical was detected. Note that illumination of
the neat DCM–TEA solution under the same conditions pro-
duced no EPR signal.
We have shown the successful synthesis of the tetramesitylpor-
phyrin phenanthroline ligand 1 as well as its metal complexes
with Zn, Cu and Pd. The full reaction sequence can be con-
ducted with palladium as the central metal center obtaining
higher overall yields than for the copper congener. In a
This journal is © The Royal Society of Chemistry 2012
Dalton Trans.

View Article Online
Paper
Dalton Transactions
subsequent step we could attach a bis-(4,4′-di-tert-butyl-2,2′- spectra were referenced to tetramethylsilane (TMS) or deuter-
bipyridine) ruthenium (Ru(tbbpy)2+) fragment to the ligand ated chloroform (CDCl3) as an internal standard (measured
resulting in the formation of heterodinuclear complexes of the values for δ are given in ppm and for J in Hz). Assignment of
type M-1-Ru with broad absorption characteristics over the signals was done with the help of 2D experiments. Elemental
visible light spectrum. The effect of the metal ion in the por- analysis was performed by the Microanalytical Lab of the Insti-
phyrinic unit on the electrochemical and spectroscopic proper- tute of Chemistry at HU Berlin using a HEKAtech EURO 3000.
ties was investigated in detail. Both characteristics are in ESI mass spectra were obtained using Thermo Finnigan
correlation with a HOMO–LUMO gap that is largest for Pd as a LCQ XP and LTQ FT instruments. MALDI-TOF spectra were
central metal and decreases to the least electronegative metal: recorded on a Bruker Daltonics REFLEX III instrument.
Zn. Thus, Pd-1 exhibits the highest first oxidation potential
Absorption spectroscopic measurements were made on
and shows the energetically highest absorption maxima in the DCM solutions of each compound using a Cary 100 UV-vis-NIR
Q-band region of the UV-Vis spectrum for the series of M-1 spectrometer from Varian employing the software Cary WinUV.
complexes. Interestingly, the influence of the metal center is SUPRASIL® Quartz cells from Hellma Analytics with a 10 mm
only marginal in the Soret band region of the UV-Vis spectrum. path length were used. Emission spectra were recorded on a
Further, M-1-Ru complexes exhibit anodically shifted Cary eclipse fluorescence spectrophotometer using SUPRASIL®
reduction potentials in comparison to M-1 and therefore M-1- Quartz cells from Hellma Analytics with a 10 mm path length.
Ru should be more suitable to catalyze CO2-reduction. We Emission measurements were made on DCM solutions of
could demonstrate by EPR spectroscopy that an electron can the investigated compounds. Cyclic voltammograms (CVs)
be transferred (and stored) in the delocalized ligand system by were performed on DCM or ACN solutions containing
either chemical reduction with Na(Hg)/KC8 or photochemical 0.1 M NBu4PF6 (tetrabutylammonium hexafluorophosphate
reduction with TEA/light in dichloromethane. The EPR studies (TBAP)) and the corresponding compound under argon atmos-
were extended to UV-Vis spectroscopic characterization (either phere at ambient temperature.
A potentiostat/galvanostat
by chemical, photochemical or electrochemical reduction) of PGSTAT 101 from Metrohm was used to record CVs and square
the mono-reduced M-1-(Ru) species. To the best of our know- wave voltammograms. A three compartment cell was outfitted
ledge this is the first time that chemical as well as electroche- with a glassy carbon button electrode as the working electrode,
mical reduction was performed on this type of ligand showing a platinum wire as the auxiliary electrode, and a silver wire as
results that are in very good agreement. Preliminary exper- a pseudo-reference electrode. All data were referenced versus
iments to use the complexes in photocatalytic CO2 reduction the ferrocene/ferrocenium couple at the end of each measure-
were not successful. Currently, we try to optimize the catalytic ment. CVs were collected at scan rates of 10–100 mV s−1
conditions and investigate the change of the metal center in Square wave voltammograms were collected at an amplitude of
.
the porphyrinic sphere.
20 mV, a step potential of 5 mV and a frequency of 25 Hz. EPR
spectra were recorded at the X-band spectrometer ERS 300
(ZWG/Magnettech Berlin/Adlershof, Germany) equipped with
a fused quartz Dewar for measurements at liquid nitrogen
temperature. The g-values were calculated with respect to a
Experimental section
General methods
Silica gel for column chromatography was obtained from VWR MgO : Cr3+ reference (g′ = 1.9796). The spectrum of Cu-1 and
or Acros (Silica Gel 60, 230–400 μm mesh). All solvents (tetra- Cu-1-Ru was simulated with the WINEPR simfonia package
hydrofuran (THF), toluene, dichloromethane (DCM), dimethyl- from Bruker for the calculation of powder spectra with
formamide (DMF), dimethylsulfoxide (DMSO), hexane, ethyl effective g values and anisotropic line widths (Gaussian line
acetate, acetonitrile (ACN) and methanol) and most starting shapes were used). Microwave reactions were carried out at
materials were obtained from Sigma-Aldrich. If necessary, sol- 10–20 W in a CEM monomode microwave (model: Discover®).
vents were purified employing an MBraun Solvent Purification
System. Starting materials were used without further purifi-
{(Pyrazo[5′,6′-e]-1′,10′-phenanthroline)[b]meso-tetramesitylpor-
phyrinato-copper(II)-bis(4,4′-di-tert-butyl-2,2′-bipyridine)-
ruthenium(II)} dihexafluorophosphate (Cu-1-Ru)
cation. Tetramesitylporphyrin (H2TMP) was prepared using the
Lindsey-method.42 5,6-Diamino-1,10-phenanthroline (phen-
diamine),53 bis(4,4′-di-tert-butyl-2,2′-bipyridine) ruthenium(II)
chloride (Ru(tbbpy)2Cl2),54 bis(4,4′-di-tert-butyl-2,2′-bipyridine)- Cu-1 (20 mg, 0.02 mmol) and Ru(tbbpy)2Cl2 (19 mg,
phenanthroline ruthenium(II) [Ru(tbbpy)2phen](PF6)2,54 CuTMP, 0.04 mmol, 2 equiv.) were dissolved in 30 mL DMF and 2.8 mL
CuTMP-NO2, CuTMP-NH2, CuTMCO2, Cu-133 and PdTMP55 water. This solution was heated at 120 °C for 64 hours. The
were synthesized according to literature procedures with only solvent was subsequently removed under reduced pressure
marginal modifications as described in ESI.† PdTMP-NO2, and the product was purified by column chromatography.
PdTMP-NH2, PdTMCO2, Pd-1,33 H2TMP-NO2, H2TMP-NH2,56 Impurities were first eluted with acetonitrile and the product
TMCO2,33 H2-135 and Zn-135 were prepared following literature finally with acetonitrile : water = 9 : 1 and 0.025 vol% of satu-
procedures and modified as outlined below.
rated aqueous KNO3. After removing the solvent the product
1H NMR spectra were recorded at ambient temperature was dissolved in ethanol and aqueous ammonium hexafluoro-
either on a Bruker DPX-300 or on an AV-400 spectrometer. All phosphate solution was added. The brown precipitate, Cu-1-Ru
Dalton Trans.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Dalton Transactions
Paper
(28 mg, 0.02 mmol, 100% based on Cu-1) was filtered off and 2,3-Dioxo-meso-tetramesitylchlorinatopalladium(II) (PdTMCO2)
washed with water and diethyl ether and dried in vacuum.
This procedure was based upon
a
literature protocol.33
Found: C, 62.29; H, 5.06; N, 8.57. Calc. for C104H104
-
PdTMP-NH2 was dissolved in 20 mL DCM and the solution
was illuminated (bulb light – 250 W) in an open vessel for 2 h
until the color of the solution turned from red to green. The
solvent was removed under reduced pressure and the solid
residue purified by column chromatography. The column was
packed with a 3 : 1 mix of hexane : DCM and then eluted with a
gradient (25–70% of DCM, 7.5% per 200 mL). PdTMCO2
(103 mg, 0.11 mmol, 80% based on PdTMP-NO2) was obtained
as a green powder.
Found: C, 73.25; H, 5.94; N, 5.62. Calc. for C56H50N4O2Pd:
C, 73.31; H, 6.11; N, 5.49; λmax (DCM)/nm 402 (ε × 10−3/dm3
mol−1 cm−1 115), 479 (15), 609 (2.3); 1H NMR (300 MHz,
CDCl3, 25 °C): δ = 1.81 (s, 12H), 1.85 (s, 12H), 2.54 (s, 6H), 2.56
(s, 6H), 7.17 (s, 4H), 7.20 (s, 4H), 8.04 (d, J = 4.9 Hz, 2H), 8.23
(d, J = 4.9 Hz, 2H), 8.24 (s, 2H); 13C NMR (75 MHz, CDCl3,
25 °C): δ = 21.2, 21.3, 21.4, 21.5, 115.5, 126.0, 127.5,
127.8, 128.0, 128.1, 128.3, 128.6, 129.0, 129.7, 131.9,
134.9, 136.5, 137.5, 137.6, 138.2, 138.5, 139.4, 141.0,
144.0, 186.6; MS(MALDI/TOF) (M+): m/z found 916.8. Calc. for
C56H50N4O2Pd 916.3.
N12P2F12RuCu·1.25 × H2O: C, 62.48; H, 5.37; N, 8.41; λmax
(DCM)/nm 286 (ε × 10−3/dm3 mol−1 cm−1 8.1), 344 (38), 424
(116), 472 (68), 583 (18), 633sh (4.1); ESI ([M − 2PF6−]2+): m/z
found 842.8425. Calc. for C104H104N12CuRu 842.8418; EPR-
parameters (THF, 77 K, simulated): gx = 2.053, gy = 2.057, gz =
2.190, Cu: Ax = 17, Ay = 20, Az = 190, N: Ax = Ay = Az = 15.
2-Nitro-meso-tetramesitylporphyrinatopalladium(II)
(PdTMP-NO2)
This procedure was based on a literature protocol.33 PdTMP
(100 mg, 0.11 mmol) was dissolved in 20 mL DCM in a 100 mL
round-bottom flask sealed with a rubber septum. NO2 was gen-
erated by the thermal decomposition of Pb(NO3)2 in a Schlenk-
tube and was transferred in portions of 1 mL by a syringe into
the reaction flask. The reaction was monitored with silica TLC
(hexane : DCM = 3 : 2). When all starting material was con-
sumed, the reaction was terminated by flushing with air to
remove the remaining NO2 gas. The solvent was removed
under reduced pressure. The solid residue was suspended in
hexane and loaded on a silica column packed with hexane.
The dark red product, PdTMP-NO2 (102 mg, 0.11 mmol, 99%),
was eluted with a gradient of DCM and hexane (DCM:
10–37.5%; 7.5% per 200 mL).
(Pyrazo[5′,6′-e]-1′,10′-phenanthroline)[b]meso-tetramesityl-
porphyrinato-palladium(II) (Pd-1)
Found: C, 73.22; H, 6.62; N, 6.58. Calc. for
C56H51N5O2Pd·1.25 × C6H14: C, 73.32; H, 6.64; N, 6.73; λmax
(DCM)/nm 421 (ε × 10−3/dm3 mol−1 cm−1 179), 529 (17), 567
This procedure was based on a literature protocol.33 PdTMCO2
(81 mg, 0.09 mmol) and phendiamine (60 mg, 0.29 mmol)
were dissolved in 165 mL dry DCM under argon atmosphere.
0.17 mL trifluoroacetic acid was added and the reaction
mixture was heated at reflux overnight. The color of the sol-
ution changed from green to dark red. The reaction mixture
was subsequently washed with 5% aqueous Na2CO3 and
water and dried over Na2SO4. The solvent was removed
under reduced pressure and the product purified with silica
column chromatography. The column was packed with DCM.
Unreacted PdTMCO2 was eluted with DCM and with a
mixture of 95% DCM and 5% methanol a violet band was col-
lected to yield Pd-1 as a dark red-brown solid (77 mg,
0.07 mmol, 80%).
1
(5.9); H NMR (300 MHz, CDCl3, 25 °C): δ = 1.83 (s, 12H), 1.84
(s, 6H), 1.87 (s, 6H), 2.57 (s, 3H), 2.60 (s, 3H), 2.61 (s, 6H), 7.16
(s, 2H), 7.25 (s, 6H), 8.53 (dd, J = 5.0 and 8.1 Hz, 2H), 8.59–8.56
(m, 4H), 8.82 (s, 1H); 13C NMR (75 MHz, CDCl3, 25 °C): δ =
21.39, 21.43, 21.47, 21.60, 21.65, 118.77, 119.84, 120.03,
121.87, 124.69, 127.71, 127.90, 128.00, 129.47, 130.63, 130.80,
131.02, 131.12, 131.19, 131.97, 134.35, 134.81, 136.85, 137.26,
138.03, 138.36, 138.90, 139.13, 140.20, 141.15, 142.25, 142.46,
142.51, 142.62, 142.75, 149.67; MS(MALDI/TOF) (M+): m/z
found 931.6. Calc. for C56H51N5O2Pd 932.5.
Found: C, 72.85; H, 5.18; N, 9.66. Calc. for C68H56N8Pd·0.5 ×
CH2Cl2: C, 72.54; H, 5.07; N, 9.88; λmax (DCM)/nm 274 (ε ×
10−3/dm3 mol−1 cm−1 28), 316 (21), 403 (90), 438 (109), 507
2-Amino-meso-tetramesitylporphyrinatopalladium(II)
(PdTMP-NH2)
This procedure was based upon
a
literature protocol.33 (4.9), 545 (24), 579 (12); 1H NMR (300 MHz, CD2Cl2, 25 °C): δ =
PdTMP-NO2 (133 mg, 0.14 mmol) was dissolved in 30 mL dry 1.84 (s, 12H), 1.90 (s, 12H), 2.63 (s, 6H), 2.88 (s, 6H), 7.32
DCM and 10 mL dry methanol under argon atmosphere and (s, 4H), 7.51 (s, 4H), 7.83 (dd, J = 4.2 and 8.1 Hz, 2H), 8.64
120 mg Pd (10% on carbon) was added. Sodium borohydride (s, 2H), 8.68 (d, J = 5.1 Hz, 2H), 8.75 (d, J = 5.1 Hz, 2H), 9.04
(75 mg, 1.72 mmol) was added to the solution in small por- (dd, J = 1.8 and 8.1 Hz, 2H), 9.27 (dd, J = 1.8 and 4.5 Hz, 2H);
tions. The reaction mixture was stirred for 1 h and all solvent 13C NMR (75 MHz, CDCl3, 25 °C): δ = 21.52, 21.58, 21.70,
was removed subsequently under reduced pressure. The solid 117.42, 121.69, 124.23, 127.91, 128.26, 128.81, 130.05, 131.00,
residue was passed through a plug of celite using DCM as an 131.31, 135.15, 137.53, 137.55, 138.00, 138.16, 138.84, 139.10,
eluent. Due to the instability of the product against photo- 139.20, 140.54, 141.02, 142.67, 148.04, 151.53, 151.56; MS
oxidation it was used immediately for the following reaction (MALDI/TOF) (MH+): m/z found 1091.7. Calc. for C68H57N8Pd
without further purification.
1091.7.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans.

View Article Online
Paper
Dalton Transactions
{(Pyrazo[5′,6′-e]-1′,10′-phenanthroline)[b]meso-tetramesitylpor- immediately for the following reaction without further
phyrinato-palladium(II)-bis(4,4′-di-tert-butyl-2,2′-bipyridine)- purification.
ruthenium(II)}dihexafluorophosphate (Pd-1-Ru)
2,3-Dioxo-meso-tetramesitylchlorin (H2TMCO2)
Pd-1 (20 mg, 18 μmol) and Ru(tbbpy)2Cl2 (19 mg, 22 μmol, 1.2
equiv.) were dissolved in 30 mL DMF and 2.8 mL water. This This procedure was based upon
a
literature protocol.33
solution was heated in a microwave cavity at 105 °C for H2TMP-NH2 was dissolved in 40 mL DCM. Silica (270 mg) was
5 hours. The solvent was subsequently removed under reduced added and the reaction mixture was illuminated (bulb light –
pressure and the product was purified by column chromato- 250 W) in an open vessel for 2 h. After removal of the solvent
graphy. Unreacted Pd-1 was first eluted with acetonitrile and under vacuum the solid residue was purified by silica column
the product finally with acetonitrile : water = 9 : 1 and 0.025 chromatography (hexane : DCM
= 3 : 2) to afford TMCO2
vol% of saturated aqueous KNO3. After removing the solvent (54 mg, 0.07 mmol, 80%) as a brown solid. The product was
the product was dissolved in ethanol and aqueous ammonium characterized according to the literature.56
hexafluorophosphate solution was added. The red precipitate,
Pd-1-Ru (28 mg, 14 μmol, 75%) was filtered off and washed (Pyrazo[5′,6′-e]-1′,10′-phenanthroline)[b]meso-tetramesityl-
with water and diethyl ether and dried in vacuum.
porphyrin-phenanthroline (H2-1)
λmax (DCM)/nm 287 (ε × 10−3/dm3 mol−1 cm−1 72), 416
(102), 460 (69), 562 (22); 1H NMR (400 MHz, CD3CN, 25 °C):
δ = 1.35 (s, 18H), 1.48 (s, 18H), 1.73–1.83 (m, 24H), 2.58
(s, 6H), 2.78 (s, 6H), 7.21 (dd, J = 1.6 and 6.3 Hz, 2H), 7.32
(s, 2H), 7.33 (s, 2H), 7.43 (s, 2H), 7.50 (m, 4 H), 7.56 (d, J = 6.0
Hz, 2H), 7.74 (d, J = 6.0 Hz, 2H), 7.88 (dd, J = 5.5 and 8.3 Hz,
2H), 8.14 (d, J = 5.5 Hz, 2H), 8.51 (d, J = 1.6 Hz, 2H), 8.57 (d, J =
2.0 Hz, 2H), 8.60 (s, 2H), 8.65 (d, J = 5.2 Hz, 2H) 8.69 (d, J =
4.8 Hz, 2H), 8.96 (d, J = 8.3 Hz, 2H); 13C NMR (125 MHz,
CD3CN, 25 °C): δ = 21.43, 21.45, 21.49, 21.53, 21.62, 21.65,
30.38, 30.49, 36.21, 36.35, 118.23, 118.62, 122. 42, 122.50,
122.87, 125.57, 127.75, 128.79, 129.09, 131.24, 131.98, 132.16,
132.17, 134.17, 137.98, 138.73, 138.86, 139.15, 139.43, 139.70,
140.05, 141.58, 142.11, 143.52, 149.68, 150.68, 152.10, 152.39,
153.93, 157.69, 158.07, 163.56, 163.74; ESI ([M − 2PF6−]2+): m/z
found 864.3295. Calc. for C104H104N12PdRu 864.3287.
This procedure was based on a literature protocol.35 TMCO2
(156 mg, 0.19 mmol) and phendiamine (140 mg, 0.65 mmol)
were dissolved in 35 mL dry DCM under argon atmosphere.
0.13 mL trifluoroacetic acid was added and the reaction
mixture was heated at reflux overnight. The reaction mixture
was subsequently washed with 5% aqueous Na2CO3 and water
and dried over Na2SO4. After removal of the solvent under
vacuum the solid residue was purified with silica column
chromatography. The column was packed with DCM and
unreacted TMCO2 was eluted with DCM, first. Changing to a
mixture of 95% DCM and 5% methanol eluted a brown band
that was collected to yield H2-1 as violet/brown solid (155 mg,
0.16 mmol, 84%).
Found: C, 82.23; H, 11.09; N, 5.89. Calc. for C68H58N8:
C, 82.73; H, 11.35; N, 5.92; λmax (DCM)/nm 436 (ε × 10−3/dm3
mol−1 cm−1 115), 527 (7.4), 600 (1.9); 1H NMR (300 MHz,
CD2Cl2, 25 °C): δ = −2.37 (s(br), 2H), 1.86 (s, 12H), 1.92
(s, 12H), 2.65 (s, 6H), 2.90 (s, 6H), 7.34 (s, 4H), 7.52 (s, 4H),
2-Nitro-meso-tetramesitylporphyrin (H2TMP-NO2)
This procedure was based upon a literature method.56 2 mL 7.83 (dd, J = 4.5 and 8.3 Hz, 2H), 8.57 (s, 2H), 8.82 (d, J =
concentrated sulfuric acid was added dropwise to a suspension 4.8 Hz, 2H), 8.91 (d, J = 4.9 Hz, 2H), 9.06 (dd, J = 1.8 and 8.4
of CuTMP-NO2 (120 mg, 0.13 mmol) in 9 mL trifluoroacetic Hz, 2H), 9.26 (dd, J = 1.8 and 4.5 Hz, 2H); 13C NMR (75 MHz,
acid and stirred for 2.5 h. The suspension was poured into CDCl3, 25 °C): δ = 21.52, 21.69, 21.74, 21.76, 115.66, 119.45,
150 mL water and neutralized with saturated aqueous NaHCO3 124.26, 127.43, 127.58, 127.91, 128.21, 129.04, 133.88, 135.31,
and extracted three times with chloroform. The organic layers 135.34, 137.63, 137.82, 137.96, 138.00, 138.28, 138.69, 139.36,
were combined, washed with water and dried over Na2SO4. 139.65, 143.81, 151.18, 151.65, 155.50; MS(MALDI/TOF) (M+):
H2TMP-NO2 was purified by silica column chromatography m/z found 987.1. Calc. for C68H58N8 987.2.
(hexane : DCM = 3 : 1) and obtained as a violet-brown solid
(86 mg, 0.10 mmol, 80%). The product was characterized (Pyrazo[5′,6′-e]-1′,10′-phenanthroline)[b]meso-tetramesityl-
according to the literature.56
porphyrinato-zinc(II)-phenanthroline (Zn-1)
This procedure was based on a literature protocol.35 H2-1
(50 mg, 0.05 mmol) was dissolved in 10 mL DCM. A solution
2-Amino-meso-tetramesitylporphyrin (H2TMP-NH2)
This procedure was based upon the literature method.56 of zinc acetate monohydrate (44 mg, 0.20 mmol) in 2.5 mL
H2TMP-NO2 (69 mg, 0.08 mmol) was dissolved in 16 mL dry methanol was added and the reaction mixture was heated at
DCM and 4 mL dry methanol under argon atmosphere and Pd reflux overnight. The solution was evaporated to dryness. The
(10% on carbon, 75 mg) was added to the solution. Sodium solid residue was dissolved in 10 mL DCM and washed with
borohydride (79 mg, 2.01 mmol) was added in small portions water. The organic layer was then stirred with 7 mL EDTA-
and the reaction mixture was stirred for 1 h. After removal of buffer (H4EDTA (29.2 mg), Na(CH3CO2) (22.0 mg), acetic acid
the solvent under vacuum the solid residue was passed (0.21 mL) in 100 mL water) for 24 h. The separated organic
through a plug of celite using DCM as an eluent. Due to the layer was washed with 5% NaHCO3 solution and water and
instability of the product against photo-oxidation it was used dried over Na2SO4. Silica column chromatography (DCM :
Dalton Trans.
This journal is © The Royal Society of Chemistry 2012

View Article Online
Dalton Transactions
Paper
methanol = 100 : 3) afforded Zn-1 as a green/violet solid
(35 mg, 0.03 mmol, 65%).
7 M. Schulz, M. Karnahl, M. Schwalbe and J. G. Vos, Coord.
Chem. Rev., 2012, 256(15–16), 1682–1705.
Found: C, 75.52; H, 5.55; N, 10.01. Calc. for
C68H56N8Zn·1.75 × H2O: C, 75.42; H, 5.54; N, 10.35; λmax
(DCM)/nm 412 (ε × 10−3/dm3 mol−1 cm−1 81), 446 (90), 523
(2.0), 570 (14), 608 (3.3); 1H NMR (300 MHz, CD2Cl2, 25 °C):
8 S. M. Molnar, G. Nallas, J. S. Bridgewater and K. J. Brewer,
J. Am. Chem. Soc., 1994, 116(12), 5206–5210.
9 J. Hawecker, J.-M. Lehn and R. Ziessel, New J. Chem., 1983,
7, 271–277.
δ = 1.85 (s, 12H), 1.90 (s, 12H), 2.65 (s, 6H), 2.90 (s, 6H), 7.33 10 S. Rau, B. Schäfer, D. Gleich, E. Anders, M. Rudolph,
(s, 4H), 7.51 (s, 4H), 7.83 (dd, J = 4.5 and 8.1 Hz, 2H), 8.68
(s, 2H), 8.79 (d, J = 4.8 Hz, 2H), 8.85 (d, J = 4.8 Hz, 2H), 9.11
M. Friedrich, H. Görls, W. Henry and J. G. Vos, Angew.
Chem., Int. Ed., 2006, 45(37), 6215–6218.
(dd, J = 1.8 and 8.1 Hz, 2H), 9.26 (dd, J = 1.8 and 4.2 Hz, 2H); 11 T. A. White, B. N. Whitaker and K. J. Brewer, J. Am. Chem.
13C NMR (75 MHz, CDCl3, 25 °C): δ = 21.52, 21.70, 21.81,
Soc., 2011, 133(39), 15332–15334.
116.04, 120.96, 124.01, 127.76, 128.12, 128.80, 130.79, 131.00, 12 M. Ogawa, G. Ajayakumar, S. Masaoka, H.-B. Kraatz and
131.66, 134.69, 137.23, 137.60, 138.58, 138.75, 139.14, 139.18, K. Sakai, Chem.–Eur. J., 2011, 17(4), 1148–1162.
139.32, 139.45, 149.07, 149.27, 149.80, 151.60, 151.70; MS 13 O. Hamelin, P. Guillo, F. Loiseau, M.-F. Boissonnet and
(MALDI/TOF) (M+): m/z found 1050.6. Calc. for C68H56N8Zn
1050.6.
S. Ménage, Inorg. Chem., 2011, 50(12), 7952–7954.
14 P. Guillo, O. Hamelin, P. Batat, G. Jonusauskas,
N. D. McClenaghan and S. Ménage, Inorg. Chem., 2012, 51
(4), 2222–2230.
X-ray crystallographic details
15 C. Herrero, J. L. Hughes, A. Quaranta, N. Cox,
A. W. Rutherford, W. Leibl and A. Aukauloo, Chem.
Commun., 2010, 46(40), 7605–7607.
16 W. Chen, F. N. Rein, B. L. Scott and R. C. Rocha, Chem.
–Eur. J., 2011, 17(20), 5595–5604.
17 W. Chen, F. N. Rein and R. C. Rocha, Angew. Chem., Int.
Ed., 2009, 48(51), 9672–9675.
18 B. Gholamkhass, K. Koike, N. Negishi, H. Hori, T. Sano
and K. Takeuchi, Inorg. Chem., 2003, 42(9), 2919–2932.
19 J. A. Treadway, J. A. Moss and T. J. Meyer, Inorg. Chem.,
1999, 38(20), 4386–4387.
20 J. J. Concepcion, J. W. Jurss, P. G. Hoertz and T. J. Meyer,
Angew. Chem., Int. Ed., 2009, 48(50), 9473–9476.
21 J. J. Concepcion, J. W. Jurss, M. K. Brennaman,
P. G. Hoertz, A. Otavio, T. Patrocinio, N. Y. M. Iha,
J. L. Templeton and T. J. Meyer, Acc. Chem. Res., 2009, 42
(12), 1954–1965.
22 F. Li, Y. Jiang, B. Zhang, F. Huang, Y. Gao and L. Sun,
Angew. Chem., Int. Ed., 2012, 51(10), 2417–2420.
23 N. R. de Tacconi, R. O. Lezna, R. Konduri, F. Ongeri,
K. Rajeshwar and F. M. MacDonnell, Chem.–Eur. J., 2005,
11(15), 4327–4339.
Data were collected at 100 K on a STOE IPDS 2T. The structure
was solved by direct methods (SHELXS-97)57 and refined by full
matrix least-squares procedures based on F2 with all measured
reflections (SHELXL-97).57 All non-hydrogen atoms were refined
anisotropically. All hydrogen atom positions were introduced at
their idealized positions and were refined using a riding model.
Crystallographic data (excluding structure factors) for Zn-1 have
been deposited at the Cambridge Crystallographic Data Centre
as supplementary publication no. CCDC 888471.
Crystallographic data for Zn-1: C68H56N8O2Zn·2CH3OH,
M = 1114.66, red crystals, orthorhombic, space group Pna21,
a = 21.916(3) Å, b = 16.8022(17) Å, c = 15.7466(17) Å, α = 90.00°,
β = 90.00°, γ = 90.00°, V = 5798.4(11) Å3, Z = 4, T = 100(2) K,
R1 = 0.0585, wR2 = 0.1092, GOF = 0.951.
Acknowledgements
We thank the German Science Foundation (DFGSCHW 1454/
4-1) for financial support, Ramona Metzinger for help with
X-ray crystallography and Prof. Stößer for his help with EPR
spectra. Further, we thank Philipp Kurz and his group for help
with spectroelectrochemistry.
24 R. Konduri, H. Ye, F. M. MacDonnell, S. Serroni,
S. Campagna and K. Rajeshwar, Angew. Chem., Int. Ed.,
2002, 41(17), 3185–3187.
25 S. Singh, N. R. de Tacconi, N. R. G. Diaz, R. O. Lezna,
J. Muñoz Zuñiga, K. Abayan and F. M. MacDonnell, Inorg.
Chem., 2011, 50(19), 9318–9328.
References
1 D. G. Nocera, Chem. Soc. Rev., 2009, 38, 13–15.
2 N. Armaroli and V. Balzani, Angew. Chem., Int. Ed., 2007, 46 26 C. Chiorboli, M. A. J. Rodgers and F. Scandola, J. Am. Chem.
(1–2), 52–66. Soc., 2003, 125(2), 483–491.
3 N. Lewis and D. G. Nocera, Proc. Natl. Acad. Sci. U. S. A., 27 W. Paw, W. B. Connick and R. Eisenberg, Inorg. Chem.,
2006, 103(43), 15729–15735. 1998, 37(16), 3919–3926.
4 S. Rau, D. Walther and J. G. Vos, Dalton Trans., 2007, 28 S. Tschierlei, M. Karnahl, M. Presselt, B. Dietzek,
915–917.
J. Guthmuller, L. González, M. Schmitt, S. Rau and J. Popp,
Angew. Chem., Int. Ed., 2010, 49(23), 3981–3984.
29 M. Karnahl, C. Kuhnt, F. Ma, A. Yartsev, M. Schmitt,
B. Dietzek, S. Rau and J. Popp, ChemPhysChem, 2011, 12
(11), 2101–2109.
5 A. Inagaki and M. Akita, Coord. Chem. Rev., 2010, 254
(11–12), 1220–1239.
6 H. Ozawa and K. Sakai, Chem. Commun., 2011, 47(8),
2227–2242.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans.

View Article Online
Paper
Dalton Transactions
30 M. J. Crossley and P. L. Burn, J. Chem. Soc., Chem. 45 M. J. Crossley, P. J. Sintic, R. Walton and J. R. Reimers, Org.
Commun., 1987, 39–40. Biomol. Chem., 2003, 1, 2777–2787.
31 K. Sendt, L. A. Johnston, W. A. Hough, M. J. Crossley, 46 P. Ochsenbein, K. Ayougou, D. Mandon, J. Fischer,
N. S. Hushand and J. R. Reimers, J. Am. Chem. Soc., 2002,
124(32), 9299–9309.
R. Weiss, R. N. Austin, K. Jayaraj, A. Gold, J. Terner and
J. Fajer, Angew. Chem., 1994, 106(3), 355–357.
32 M. J. Crossley, P. L. Burn, S. J. Langford and J. K. Prashar, 47 U. Eberhardt, W. Schwarz and H. Musso, Liebigs Ann.
J. Chem. Soc., Chem. Commun., 1995, 1921–1923. Chem., 1987, 809–810.
33 T. A. Vanelli and T. B. Karpishin, Inorg. Chem., 2000, 39(2), 48 K. M. Kadish, W. E, P. J. Sintic, Z. Ou, J. Shao, K. Ohkubo,
340–347.
S. Fukuzumi, L. J. Govenlock, J. A. McDonald, A. C. Try,
Z.-L. Cai, J. R. Reimers and M. J. Crossley, J. Phys. Chem. B,
2007, 111(30), 8762–8774.
34 T. A. Vanelli and T. B. Karpishin, Inorg. Chem., 1999, 38
(10), 2246–2247.
35 R.-S. Lin, M.-R. Li, Y.-H. Liu, S.-M. Peng and S.-T. Liu, Inorg. 49 E. Wenbo, K. M. Kadish, P. J. Sintic, T. Khoury,
Chim. Acta, 2010, 363(13), 3523–3529.
36 I. Bhugun, D. Lexa and J.-M. Saveant, J. Am. Chem. Soc.,
1996, 118(16), 3982–3983.
L. J. Govenlock, Z. Ou, J. Shao, K. Ohkubo, J. R. Reimers,
S. Fukuzumi and M. J. Crossley, J. Phys. Chem. A, 2008, 112,
556–570.
37 J. A. Goodwin and T. S. Kurtikyan, J. Porphyrins Phthalocya- 50 K. L. Cunningham, K. M. McNett, R. A. Pierce, K. A. Davis,
nines, 2011, 15, 99–105.
38 A. J. Olaya, D. Schaming, P.-F. Brevet, H. Nagatani,
H. H. Harris, D. M. Falck and D. R. McMillin, Inorg. Chem.,
1997, 112(11), 608–613.
T. Zimmermann, J. Vanicek, H.-J. Xu, C. P. Gros, 51 M. Kumar, P. Neta, T. P. G. Sutter and P. Hambright,
J.-M. Barbe and H. H. Girault, J. Am. Chem. Soc., 2012, 134
(1), 498–506 and references therein.
39 J. Rosenthal and D. G. Nocera, Acc. Chem. Res., 2007, 40(7),
543–553.
J. Phys. Chem., 1992, 96, 9571–9575.
52 We also tried to synthesize a CuI-porphyrin, but did not
succeed: reaction of [Cu(ACN)4](PF6) with TMP results in
no complex formation.
40 D. Behar, T. Dhanasekaran, P. Neta, C. M. Hosten, D. Ejeh, 53 S. Bodige and F. M. MacDonnell, Tetrahedron Lett., 1997, 38
P. Hambright and E. Fujita, J. Phys. Chem. A, 1998, 102(17),
2870–2877.
41 J. Grodkowski, D. Behar, P. Neta and P. Hambright, J. Phys.
Chem. A, 1997, 101(3), 248–254.
(47), 8159–8160.
54 S. Rau, B. Schäfer, A. Grüßing, S. Schebesta, K. Lamm,
J. Vieth, H. Görls, D. Walther, M. Rudolph, U. W. Grummt
and E. Birkner, Inorg. Chim. Acta, 2004, 357, 4496–4503.
42 J. S. Lindsey and R. W. Wagner, J. Org. Chem., 1989, 54(4), 55 M. J. Crossley, C. S. Sheehan, T. Khoury, J. R. Reimers and
828–836. P. J. Sintic, New J. Chem., 2008, 32, 340–352.
43 R. Beavington, P. A. Rees and P. L. Burn, J. Chem. Soc., 56 S. Eu, S. Hayashi, T. Umeyama, Y. Matano, Y. Araki and
Perkin Trans. 1, 1998, 2847–2852. H. Imahori, J. Phys. Chem. C, 2008, 112(11), 4396–4405.
44 J. Wojaczyński, L. Latos-Grażyński and T. Głowiak, Inorg. 57 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
Chem., 1997, 36(27), 6299–6306. logr., 2008, 64, 112–122.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2012