Towards alternatives to anodic water oxidation: Basket-handle thiolate FeIII porphyrins for electrocatalytic hydrocarbon oxidation

Article abstract of DOI:10.1002/cssc.201200572

Selective electrocatalytic oxidation of hydrocarbons to alcohols, epoxides or other (higher value) oxygenates should in principal present a useful complementary anodic half-cell reaction to cathodic generation of fuels from water or CO₂ viz. an alternative to oxygen evolution. A series of new basket-handle thiolate Fe^III porphyrins have been synthesised and shown to mediate anodic oxidation of hydrocarbons, specifically adamantane hydroxylation and cyclooctene epoxidation. We compare yields obtained by electrochemical and chemical oxidation of the thiolate porphyrins and benchmark their behaviour against that of Fe^III tetraphenyl porphyrin chloride and its tetrapentafluorophenyl analogue.

Full text of DOI:10.1002/cssc.201200572

DOI: 10.1002/cssc.201200572
Towards Alternatives to Anodic Water Oxidation:
Basket-Handle Thiolate FeIII Porphyrins for Electrocatalytic
Hydrocarbon Oxidation
Peiyi Li, Khalaf Alenezi, Saad K. Ibrahim, Joseph A. Wright, David L. Hughes, and
Christopher J. Pickett*[a]
Selective electrocatalytic oxidation of hydrocarbons to alco-
hols, epoxides or other (higher value) oxygenates should in
principal present a useful complementary anodic half-cell reac-
tion to cathodic generation of fuels from water or CO2 viz. an
alternative to oxygen evolution. A series of new basket-handle
thiolate FeIII porphyrins have been synthesised and shown to
mediate anodic oxidation of hydrocarbons, specifically ada-
mantane hydroxylation and cyclooctene epoxidation. We com-
pare yields obtained by electrochemical and chemical oxida-
tion of the thiolate porphyrins and benchmark their behaviour
against that of FeIII tetraphenyl porphyrin chloride and its tetra-
pentafluorophenyl analogue.
Introduction
The cathodic generation of fuels, such as dihydrogen from
water or carbon monoxide, formate, methanol or other feed-
stocks from CO2 using sunlight, wind or other sustainable
energy sources, could ameliorate fossil fuel dependency. The
complementary anode reaction is usually water oxidation
which gives benign, non-polluting dioxygen, but this has little
intrinsic value as a co-product. It is, therefore, of interest to
consider alternative anode chemistries, particularly those that
can employ earth-abundant electrocatalysts rather than unsus-
tainable platinum materials, currently the best material for
electrochemical water oxidation.
electrocatalytic activity of these novel molecules in hydrocar-
bon oxidation. As is seen in the cytochrome P450 enzymes,[4]
these synthetic iron porphyrins possess the structurally robust
axial thiolate coordination, which is provided by a basket-
handle arrangement[5] in our systems.[6] The rigid basket-handle
may also provide steric protection of the S-ligand from oxida-
tive degradation. The natural and synthetic catalyst systems
are understood to operate via the formation of an oxoferryl
porphyrin radical cation, which abstracts a hydrogen atom
from hydrocarbon CꢀH.[7] It would appear that there are two
principal indicators of the reactivity of the active species, the
oxidation potential E0 necessary to access the oxoferryl por-
phyrin radical cation species and the pKa of the FeIVꢀOH inter-
mediate formed by hydrogen-atom abstraction.[8] Electron-
withdrawing groups on the porphyrin ring increase the oxidis-
ing power of the ferryl porphyrin radical cation and this is
reflected by a more positive E0. On the other hand, a more
electron-donating group in the axial position will increase the
OꢀH bond strength by increasing the pKa. We have explored
this balance by synthesising the range of iron porphyrin pre-
cursors shown in Figure 1, which possess electron-donating
and -withdrawing substituent groups. We compare the electro-
catalytic behaviour of these molecules in hydrocarbon oxida-
tion against that in chemical catalysis, where the chemical
oxygen atom transfer reagent, iodosylbenzene (PhIO), was
employed.
Alternative reactions to dioxygen evolution should meet the
following demands: i) higher value products should be pro-
duced from an oxidisable substrate—the electron source;
ii) such complementary reactions should give products that are
required at a scale comparable to that of solar fuel or feed-
stock; iii) ideally, the anode chemistry should have fast kinetics
and iv) low overpotential. The conversion of hydrocarbons to
oxygenates such as alcohols or epoxides could in principle sat-
isfy the first two requirements, the third and fourth demand
challenging new chemistry.
Iron porphyrin systems with high-valent ferryl moieties gen-
erated by chemical oxygen atom transfer have been extensive-
ly studied as functional cytochrome P450 analogues and have
been shown to catalyse oxygenation of alkanes and alkenes.[1]
The demonstration of anodic oxidation of hydrocarbons using
iron porphyrin electrocatalysts is very limited. To the best of
our knowledge, there is only a single report on the oxidation
of an alkane by an iron porphyrin:[2] very little detail is provid-
ed; critically, current yields are not given. Similarly, studies on
electrochemical alkene oxidation are confined to a single
report[3] and the source of [O] for the oxygenation is not clear
beyond a single turnover.
[a] Dr. P. Li, K. Alenezi, Dr. S. K. Ibrahim, Dr. J. A. Wright, Dr. D. L. Hughes,
Prof. C. J. Pickett
Energy Materials Laboratory, School of Chemistry
University of East Anglia, Norwich Research Park
Norwich NR4 7TJ (United Kingdom)
Fax: (+44)1603-592003
E-mail: c.pickett@uea.ac.uk
Herein, we describe the synthesis of new iron porphyrin
electrocatalysts, which possess an axial thiolate ligand, and the
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201200572.
ChemSusChem 0000, 00, 1 – 16
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

C. J. Pickett et al.
Figure 1. Molecular structures of basket-handle thiolate FeIII porphyrins.
Results and Discussion
tected in the last step after iron insertion. The central 2,6-di-
alkoxy benzyl thioacetate unit was connected to the main por-
phyrin body through the formation of two amide bonds. The
length of the strap was designed to be just flexible enough to
allow the coordination of thiolate to the iron centre. Such an
arrangement introduced some rigidity into the strap, which
may enhance the stability of thiolate FeIII porphyrins. The strap-
ped porphyrins 1(Ac)–4(Ac) were synthesised by following
Synthesis and structure
A 2,6-dialkoxy benzylthiolate has been chosen as the central
component of the strap because the alkoxy functionality en-
hances the donicity of the benzylthiolate ligand (Scheme 1).
The air-sensitive thiolate was protected by an acetyl group as
benzyl thioacetate throughout the syntheses and was depro-
Scheme 1. General strategies for the synthesis of strapped porphyrins 1(Ac)–4(Ac). The side chain R (I, Me, C6F5, or polyether) may be introduced by one of
two routes.
&2
&
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 16
ÝÝ
These are not the final page numbers!

Porphyrins as Alternatives to Anodic Water Oxidation
a classical trifluoroacetic acid (TFA)-catalysed procedure for
trans-A2B2 porphyrins.[9] Two strategies (Scheme 1, Routes A
and B) were employed to synthesise the strapped porphyrin li-
gands. The first route (Route A) involved two major steps:
building the porphyrin body first, followed by attaching the
strap to the porphyrin. Alternatively, the strapped porphyrin
could be synthesised in a one-pot reaction by condensation of
a dipyrromethane with a dialdehyde strap (Route B). Attempts
were made to synthesise porphyrins 1(Ac)–3(Ac) via both
routes. The yield from each route depended on the peripheral
substituents of the porphyrins. Porphyrins 1(Ac) and 2(Ac)
were obtained in reasonable yields by Route A, whereas
Route B only produced a small amount of impure porphyrins
1(Ac) and 2(Ac). Route A did not generate any porphyrin 4(Ac),
which was therefore solely accessed by Route B. Route B was
chosen to prepare porphyrin 3(Ac) to avoid the complication
from the dealkoxylation of the meso-[2,6-bis(2-meth-
oxyethoxy)phenyl] group during the nitro reduction in
Route A.
Porphyrins 1(Ac)–2(Ac) were synthesised in reasonable yields
via Route A (Scheme 3). TFA-catalysed condensation of 5-(2-ni-
trophenyl)dipyrromethane with 4-iodo-benzaldehyde and sub-
sequent oxidation by 2,3-dichloro-5,6-dicyano-1,2-benzoqui-
none (DDQ) produced the cis- and trans-isomers of nitro pre-
cursor 10 in an isolated yield of 11%. The latter was reduced
to amino precursors by SnCl2 in HCl. The desired cis-isomer 11
was separated from the mixture of atropisomers using silica
column chromatography. Porphyrin 1(Ac) was prepared in an
isolated yield of 70% by coupling 11 with dicarboxylic acid
strap linker 8 using a Mukaiyama coupling reagent. A similar
approach was adopted to synthesise porphyrin 2(Ac). The nitro
precursor 13 (cis- and trans-isomers) was synthesised by TFA-
catalysed condensation of 5-(2,6-dimethylphenyl)dipyrrome-
thane 12 with 2-nitrobenzaldehyde and subsequent oxidation
by DDQ in an isolated yield of 11.6%. Reduction of the cis/
trans-mixture of 13 by SnCl2 in HCl led to a mixture of atro-
pisomers of the amino precursor, from which the desired cis-
isomer 14 was separated using silica column chromatography.
The latter was coupled to the dicarboxylic acid strap linker 8
via Mukaiyama coupling to afford porphyrin 2(Ac) in an isolat-
ed yield of 72%.
Scheme 2 outlines the synthetic route for the strap linkers 8
and 9. Alkylation of methylresorcinol with tert-butyl bromoace-
tate afforded 2,6-bis(tert-butoxycarbonylmethoxy)toluene (5) in
an isolated yield of 78%, which was then brominated with
N-bromosuccinimide (NBS) to produce 1-bromomentyl-2,6-bis-
(tert-butoxycarbonylmethoxy)benzene (6) in an isolated yield
of 67%. Treatment of 6 with potassium thioacetate (KSAc) pro-
duced thioacetate 7 in an isolated yield of 65%. Deprotection
of tert-butyl ester within thioacetate 7 with TFA generated di-
carboxylic acid 8 in an isolated yield of 82%. The dialdehyde
strap linker 9 was prepared by reacting dicarboxylic acid 8
with 2-amino-benzaldehyde with the aid of a Mukaiyama cou-
pling reagent in an isolated yield of 60%.
Scheme 4 summarises the synthesis of porphyrins 3(Ac) and
4(Ac) via Route B. Reaction of resorcinol with methoxyethoxy-
methyl chloride (MEMCl) yielded 1,3-bis(2-methoxyethoxy)ben-
zene (15) in an isolated yield of 67%. Formylation of 15 with
nBuLi and DMF furnished 2,6-bis(2-methoxyethoxy)benzalde-
hyde (16) in an isolated yield of 58%. TFA-catalysed condensa-
tion of aldehyde 16 with pyrrole gave dipyrromethane 17 in
an isolated yield of 39%. Porphyrin 3(Ac) was then synthesised
by TFA-catalysed condensation of dipyrromethane 17 with di-
aldehyde strap linker 9 and subsequent oxidation by DDQ.
Scheme 2. Synthesis of the strap linkers 8 and 9. Reagents and conditions: a) i) K2CO3, acetone, reflux; ii) tert-butyl bromoacetate, acetone, reflux; b) NBS, ben-
zoyl peroxide, CCl4, reflux; c) KSAc, acetone, RT; d) TFA, DCM, RT; e) 2-chloro-1-methylpyridinium iodide, Et3N, DCM, RT.
ChemSusChem 0000, 00, 1 – 16
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
3
&
ÞÞ
These are not the final page numbers!

C. J. Pickett et al.
Scheme 3. Synthesis of the strapped thiolate iron porphyrins Fe-1 and Fe-2. Reagents and conditions: a) i) TFA, DCM, RT; ii) DDQ, RT; iii) Et3N, RT; b) SnCl2,
HCl(aq), RT; c) 8, Et3N, DCM, RT; d) i) FeBr2, lutidine, toluene, reflux; ii) NaOCH3, RT.
Here, the concentration of the reactants is important as dimeri-
sation occurred at high concentrations and reaction rates and
yields were compromised at low concentrations. A series of ex-
periments were carried out under various reactant concentra-
tions, ranging from 0.36 to 3.66mm for dialdehyde strap linker
9. The best yield was achieved with a concentration of
2.88mm for 9 in an isolated yield of 2.3%. The same reactant
concentration was employed in the synthesis of porphyrin
4(Ac). Coupling of 5-(pentafluorophenyl)-dipyrromethane with
dialdehyde strap linker 9 in the presence of TFA and subse-
quent oxidation by DDQ provided porphyrin 4(Ac) in an isolat-
ed yield of 6.6%.
the porphyrin with the strap sitting across the porphyrin
plane. Some degree of ruffle deformation exists within all
three porphyrins to accommodate the relatively short-length
strap containing only eleven atoms. The average displacement
of meso-carbon attached to the strap from the least-squares
24-atom plane is around 0.5 ꢁ.
There are two ways to reach the final thiolate iron(III) por-
phyrins Fe-(1–4): iron insertion first, followed by thiol deprotec-
tion, or vice versa. Deprotection of the thiol was performed for
porphyrin 2(Ac) in the presence of NaOMe at RT. The resulting
mixture was found to contain a moderate amount of decom-
posed products. As the aerial oxidation of unprotected thiol is
facile, the cleavage of the thioacetate group was carried out
directly in situ after iron insertion with NaOMe at RT under N2.
The iron insertion into porphyrins 1(Ac)–4(Ac) were accom-
plished by refluxing porphyrins 1(Ac)–4(Ac) with FeBr2 and luti-
dine in toluene (Schemes 3 and 4). Both THF and toluene were
tried as solvents, but insertion only reached completion in re-
fluxing toluene. The fast work-up after the reaction allowed
iron(II) to be oxidised to iron(III) without significant oxidation
of the thiolate sulfur. After purification using silica column
chromatography, the desired complexes were crystallised in
a CH2Cl2/hexane solvent mixture. The isolated yields of iron(III)
1
Evidence for the strapped porphyrin is found in the H NMR
spectrum as the strap is constrained right under (or above) the
porphyrin plane. The CH2S resonance is not observed at the
usual d=4.3–4.9 ppm, but is extremely shielded by the ring
current of porphyrin to appear at d=ꢀ2.3 to ꢀ2.6 ppm. The
monomeric structures of four porphyrins were also supported
by their MALDI (matrix-assisted laser desorption/ionization)
mass spectra, for which the theoretical isotopic pattern
matched exactly with the experimental one (see the Support-
ing Information). The X-ray crystal structures of 1(Ac), 3(Ac)
and 4(Ac) (Figure 2) also confirm the monomeric structure of
&4
&
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 16
ÝÝ
These are not the final page numbers!

Porphyrins as Alternatives to Anodic Water Oxidation
Scheme 4. Synthesis of the strapped thiolate iron porphyrins Fe-3 and Fe-4. Reagents and conditions: a) i) NaH, DMF, 08C to RT; ii) MEMCl, 08C to RT;
b) i) nBuLi, THF, 08C; ii) DMF, 08C to RT; c) pyrrole, TFA, RT; d) i) 9, TFA, DCM, RT; ii) DDQ, RT; iii) Et3N, RT; e) i) FeBr2, lutidine, toluene, reflux; ii) NaOCH3, RT.
porphyrins Fe-(1–4) ranged from 50 to 63%. All four thiolate
iron porphyrins are stable in air for a brief period, but care
should be taken to avoid prolonged exposure, especially in
solution.
O and FeꢀS being 2.205 and 2.163 ꢁ. The octahedral iron is lo-
cated very close to the plane of the porphyrinato core with
only 0.017 ꢁ displacement from the mean 24-atom plane.
Overall, the structure is very similar to the ring-anchored
benzyl thiolate iron porphyrin reported by Nagano et al.[11]
In our case, the FeꢀS bond length of 2.163(3) ꢁ is slightly
shorter than their reported value of 2.1772(12) ꢁ and the
107.4(3)8 bond angle of FeꢀSꢀC is not significantly different
from Nagano’s reported value of 108.26(14)8. It is possible that
the rigidity of the constraining strap could be the reason for
the slightly shorter FeꢀS bond length seen in Fe-1.
All four thiolate iron porphyrins are crystalline, but only crys-
tals of Fe-1 and Fe-2 were suitable for X-ray structure analysis.
For Fe-2, only the atomic connectivity could be established
from the X-ray structure analysis due to the poor data set
(Figure 3). It clearly showed that the central iron is coordinated
to four nitrogen atoms of the porphyrin and to the sulfur of
the benzylthiolate strap. The sixth ligand is the oxygen atom
from the amide group of a neighbouring molecule, which has
been omitted in Figure 3. The X-ray structure analysis of
Fe-1 was good enough to provide a detailed portrait of its
structure (Figure 4). Again the porphyrinato core is ruffled,
which may be amplified because of the short strap linker. The
bending of the porphyrinato core plane toward the strap is
clearly shown in the side view of Fe-1 (Figure 4). The central
iron is coordinated to the four equatorial nitrogen atoms of
the porphyrin, an axial sulfur of benzylthiolate strap and an
axial oxygen of a THF solvent molecule.[10] The average FeꢀN
Electrochemistry
The electrooxidation of the four new thiolate FeIII porphyrins
and the known complexes [Fe(tppf20)Cl] (meso-tetrakis(penta-
fluorophenyl)porphyrin iron(III) chloride) and [Fe(tpp)Cl] (meso-
tetraphenylporphyrin iron(III) chloride) were studied by using
cyclic voltammetry in CH2Cl2 containing 0.2m Bu4NBF4 at a vit-
reous carbon electrode at RT.
Figure 5 shows a set of voltammograms recorded at various
bond length is 1.978 ꢁ, with the bond lengths of the axial Feꢀ scan rates for complex Fe-1. The primary oxidation process is
ChemSusChem 0000, 00, 1 – 16
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
5
&
ÞÞ
These are not the final page numbers!

C. J. Pickett et al.
Figure 3. Structure of Fe-2.
Figure 2. X-ray crystal structures of porphyrins 1(Ac), 3(Ac) and 4(Ac).
a diffusion-controlled reversible one-electron step. Thus: i) the
I
ox
plot of the peak current ip versus scan rate n1/2 is linear with
an intercept close to zero (Figure 5, inset a); ii) the magnitude
I
ox
of ip at each scan rate is, as expected, equal to those ob-
served for the primary FeIII/II one-electron reduction process
(Figure S1). This first oxidation is probably a metal-centred
FeIV/III-couple, rather than a ligand-based oxidation, because its
potential is sensitive to the addition of water/hydroxide as dis-
cussed below. Each of the four thiolate complexes shows a sim-
Figure 4. X-ray crystal structure of (THF)Fe-1.
proximately 320 mV more positive than that of Fe-1. This
maybe a consequence of the better electron-donating ability
of the thiolate group compared with Cl, or perhaps more likely
a switch from a metal-based oxidation in the case of Fe-1 to
a porphyrin-based oxidation, as has been established for
[Fe(tpp)Cl].[12]
I
ilar primary oxidation step (Figure S2). The E1/2 values range
from 0.76 V versus saturated calomel electrode (SCE) for the
most electron-rich porphyrin to 0.96 V versus SCE for the most
electron-poor porphyrin (Table 1). The potential for the primary
oxidation of the [Fe(tpp)Cl] complex, 1.12 V versus SCE, is ap-
Following the primary one-electron oxidation step, a secon-
dary diffusion-controlled process is observed at a more posi-
ox
tive potential (IIip vs. n1/2 is linear, Figure 5, inset b). The pro-
&6
&
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 16
ÝÝ
These are not the final page numbers!

Porphyrins as Alternatives to Anodic Water Oxidation
the experimental voltammogram recorded at 100 mVsꢀ1 are
shown in Figure S3 and the simulation parameters are also
listed in the Supporting Information. The other iron thiolate
porphyrins all show similar secondary oxidations as shown in
II ox
I ox
Figure S2; the ratio of ip to ip at 100 mVsꢀ1 in all cases is
between 2–3 and must depend upon the kinetics associated
with the chemical step.
In preparative electrolyses using the iron porphyrins as elec-
trocatalysts we found that the presence of the strong base
[NBu4][OH]·30H2O was essential for hydrocarbon oxidation in
all cases. We have, therefore, examined the effect of this re-
agent on the cyclic voltammograms of the systems, focusing
again on Fe-1. Figure 7a shows the effect of the base at a low
Figure 5. Cyclic voltammograms of Fe-1 at various scan rate in CH2Cl2 con-
taining 0.2m [Bu4N][BF4]. The inset figures are the plots of ipox vs. n1/2
.
Table 1. Oxidation potentials (V vs. SCE) in CH2Cl2 containing 0.2m
[Bu4N][BF4].
Catalyst
Potential [V]
[a]
ox[b]
red[b]
IE1/2
IIEp
IIEp
Fe-1
Fe-2
Fe-3
Fe-4
[Fe(tppf20)Cl]
[Fe(tpp)Cl]
0.88
0.8
0.76
0.96
1.58[c]/1.48[d]
1.12
1.44
1.37
1.43
1.45
1.77
1.39[e]
1.21
1.19
1.17
1.29
1.62
1.39[e]
[a] Reversible one-electron process. [b] Irreversible multi-electron process.
ox
red
[c] Ep for irreversible two-electron process. [d] Ep for irreversible one-
II
electron process. [e] Reversible one-electron process, E1/2
.
II ox
I ox
cess appears partially reversible and the plot of ip vs. ip is
linear with a slope of 2.5 as shown in Figure 6. The expected
ratio between the fast reversible two-electron and one-elec-
tron systems is (IIn/In)3/2 =2.83. However, the ratio of 2.5 that
we observe, the peak separation and the overall shape of the
voltammogram are best simulated by two consecutive one-
electron steps with a following chemical step and further elec-
tron transfer increasing the current. The digital simulation and
Figure 7. Cyclic voltammograms of Fe-1 in CH2Cl2 containing 0.2m
[Bu4N][BF4] at RT with added [Bu4N][OH]. Scan rate=0.1 Vsꢀ1
.
concentration (2.4 equivalents OHꢀ) on the cyclic voltammetry.
The key features are a negative shift in the potential for the
primary oxidation peak by about 120 mV and a reduction in
ox
the magnitude of IIip by about 20%. Figure 7b shows the
effect of increasing the reagent concentration approximately
13-fold to 32 equivalents. In this case, we see a new oxidation
peak at a relatively low potential with OHEp =0.43 V versus SCE,
which is accompanied by suppression of the low base/zero
base peaks at 0.7–0.9 V versus SCE and a negative shift of the
‘secondary’ oxidation process by about 160 mV. The increase in
the ‘low potential’ species with increasing base is shown by
Figure 8. We suggest that at low concentrations of the added
base speciation is predominately water occupying the axial
site and at higher base concentrations the majority species is
Figure 6. Plots of ipox versus ipox. Scan rate=0.1 Vsꢀ1
.
II
I
ChemSusChem 0000, 00, 1 – 16
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
7
&
ÞÞ
These are not the final page numbers!

C. J. Pickett et al.
Chemical oxidation
Chemical oxidation of adamantane and cyclooctene using
PhIO as an oxygen source was carried out in CH2Cl2 at 208C
under N2 for 1 h. Under our reaction conditions, all iron por-
phyrins catalysed the oxidation of adamantane and cyclooc-
tene to 1-adamantanol and cyclooctene oxide, respectively,
with minimal formation of other hydrocarbon oxidation prod-
ucts (Table 2).
The first and simplest iron porphyrin catalyst [Fe(tpp)Cl][1e]
provides a benchmark[13] against which we can compare hydro-
Figure 8. Cyclic voltammograms of Fe-1 in CH2Cl2 containing 0.2m
Table 2. Yields of FeIII porphyrin-catalysed adamantane/cyclooctene oxi-
dations by PhIO.[a]
[Bu4N][BF4] at RT with various [Bu4N][OH] concentrations. Scan
rate=0.1 Vsꢀ1
.
Catalyst
1-Adamantanol
yield [%][b]
Cyclooctene oxide
TON[c]
yield [%][b]
TON[c]
the hydroxide ligated complex, Scheme 5. This would be con-
sistent with the new peak at Ep =0.43 V versus SCE being asso-
ciated with the hydroxide species whilst that at 0.79 V versus
SCE being the aquo species. The irreversibility of the oxidation
of the aquo species may be attributed to proton removal by
the base; similarly, this may explain the irreversibility of the ox-
idation of the hydroxide species whereby iron(IV) oxo species
are formed (Scheme 5). The lower peak current associated with
secondary process in the presence of a base may be attributa-
ble to the formation of stable oxo species rather than destruc-
tive oxidation of the porphyrin.
Fe-1
Fe-2
Fe-3
Fe-4
[Fe(tppf20)Cl]
[Fe(tpp)Cl]
22.5
34.7
10.5
43.4
60.8
11.3
3
2
0.53
3.2
3.7
0.5
70.4
69.7
66.4
86.1
86.2
74.7
5.2
5.5
4
6
7
5.2
[a] Reactions were carried out in CH2Cl2 at 208C under N2 for 1 h. [FeIII
porphyrin]=1.0 mm; [PhIO]=10 mm; [adamantane or cyclooctene]=
0.1 m. [b] Average value of two experiments based on PhIO consumed.
[c] TON=moles of product/moles of catalyst. The theoretical maximum
TON is 10 based upon ratio of moles of PhIO/moles of catalyst.
Scheme 5. Proposed oxidation steps in absence and presence of a base.
&8
&
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 16
ÝÝ
These are not the final page numbers!

Porphyrins as Alternatives to Anodic Water Oxidation
carbon oxidation activity of the new thiolate porphyrins.
Table 2 shows that all of the thiolate complexes except Fe-3
have a higher activity than [Fe(tpp)Cl], both in terms of turn-
over number (TON) and yield of 1-adamantanol. The least elec-
tron-rich complex, Fe-4, is the best thiolate-based catalyst,
whereas the most electron-rich complex, Fe-3, is the least
active. This indicates that electron-withdrawing groups give
rise to the most reactive ferryl units.[14] This electronic effect
must override the protection against FeꢀOꢀFe m-oxo dimer
formation (deactivation) that we might expect to be afforded
by the oligoether chains of Fe-3. An earlier study of the penta-
fluorophenylporphyrin iron complex[15] showed a notably high
adamantane oxidation activity. In accord with this, we find that
[Fe(tppf20)Cl] is the most active catalyst precursor under our
conditions: it is correspondingly the hardest to oxidise electro-
chemically (Table 1). Other than electron density, the geometric
arrangement around the thiolate sulfur may also play a role.
The electron densities on Fe-1 and Fe-2 are very similar, yet
the 1-adamantanol yield employing Fe-2 (34.7%) is significant-
ly higher than that of Fe-1 (22.5%). The four ortho-methyl
groups in Fe-2 may shield the thiolate ligand from oxidative
destruction and/or prevent the formation of the inactive m-
oxo dimer. Where there is no significant electronic perturba-
tion of the porphyrin ring, as in the cases of Fe-1 and
[Fe(tpp)Cl], the sulfur axial ligand must favour hydrogen ab-
straction from the substrate, as demonstrated by the yield of
1-admantanol (22.5% for Fe-1 vs. 11.3% for Fe[(tpp)Cl]). In an
earlier study, Higuchi et al.[16] were also able to show that
a ring-anchored benzyl thiolate had a substantially better activ-
ity than [Fe(tpp)Cl] using a peroxy acid as the oxygen atom
source.
chemical oxidation with respect to chemical oxidation can be
ascribed to the competing generation of superoxide, peroxide
or dioxygen (Scheme 6). In the absence of an iron porphyrin
catalyst, cyclooctene could be electrochemically oxidised to cy-
clooctene epoxide in the presence of OHꢀ with a current effi-
ciency of 5.6% at an anode potential of 1.8 V (vs. SCE). Howev-
er, only a trace amount of 1-adamantanol was detected under
the same experimental conditions. The applied potential is
[17]
C
positive enough to produce OH radicals,
which evidently
convert cyclooctene to cyclooctene oxide, probably via perox-
ide. However, neither of these species are powerful enough to
oxidise adamantane to 1-adamantanol to a significant extent.
Thus, the presence of iron porphyrins lowers the activation
barriers for adamantane hydroxylation and cyclooctene epoxi-
dation, dramatically increasing the current efficiency for their
formation and substantially lowering the anode potentials re-
quired (Table 3). Notably, the cyclic voltammetry of Fe-1 in the
presence of adamantane at high concentration is only margin-
ally perturbed by the substrate, consistent with no direct oxi-
dation. In the presence of the base and the substrate the
cyclic voltammogram is largely dominated by the effect of
base; consistent with slow hydrocarbon oxidation kinetics,
some minor effects on peak currents are observed.
In parallel with the chemical oxidation of adamantane, the
electron density on the iron porphyrin appears to be the major
factor that determines the current efficiency and turnover
number: both increase as the electron density on the iron por-
phyrin decreases. Thus, among the four thiolate porphyrins,
the most electron-deficient Fe-4 provides the highest current
efficiency and the most electron rich Fe-3 gives the lowest cur-
rent efficiency. Again, [Fe(tppf20)Cl] exhibits overall the highest
current efficiency whilst [Fe(tpp)Cl] has the lowest. The effect
of geometric arrangement around the thiolate sulfur on the
current efficiency of electrocatalytic oxidation (9.4% for Fe-
1 vs. 9.8% for Fe-3) is not as significant as for that observed
for the chemical oxidation (22.5% for Fe-1 vs. 34.7% for Fe-3).
This is understandable since the oxidative destruction of the
thiolate ligand from a strong oxygen-transfer reagent, such as
PhIO, is no longer applicable in the case of electrocatalytic oxi-
dation. The current efficiencies of cyclooctene epoxidation are
also shown in Table 3. [Fe(tppf20)Cl] is the best electrocatalyst
in terms of current efficiency, but it also operates at the high-
est oxidation potential.
With respect to epoxidation of the more reactive cyclooc-
tene, the electronic and geometric effects of the iron porphyr-
ins on the catalytic activity are levelled out, the two most elec-
tron deficient iron porphyrins gave the highest yields of ap-
proximately 85%, with the other complexes giving yields of
about 70% (Table 2).
Electrochemical oxidation of adamantane and cyclooctene
Electrocatalytic oxidations of adamantane and cyclooctene
were carried out in wet CH2Cl2 at RT under N2 for 1.5 h using
0.2m [Bu4N][BF4] as the electrolyte and in the presence of tet-
rabutylammonium hydroxide. Based on the cyclic voltammo-
grams of four thiolate iron porphyrins, [Fe(tppf20)Cl] and
[Fe(tpp)Cl], the applied potentials for Fe-1, Fe-2, Fe-3, Fe-4,
[Fe(tppf20)Cl] and [Fe(tpp)Cl] were set at 1.36, 1.27, 1.31, 1.42,
1.70 and 1.49 V (vs. SCE), respectively. They are sufficiently pos-
itive to ensure accessing the two-electron oxidised state of the
FeIII complexes.
In general, our basket-handle thiolate complexes are clearly
superior to the prototypical [Fe(tpp)Cl] system, both in terms
of potential and current efficiency, and further modification
might increase the current efficiency without significantly in-
creasing the oxidation potential.
It was found that the presence of hydroxide is crucial. With-
out this base control, experiments show that no adamantanol
or cyclooctene oxide is formed. This is consistent with the pro-
posed electrocatalytic mechanism shown in Scheme 6. The
iron(III) hydroxide is oxidised by removal of two electrons and
a proton, assisted by free base, which thus generates the
highly reactive ferryl species. The lower efficiency of electro-
Conclusions
A series of new basket-handle thiolate FeIII porphyrins have
been synthesised and fully characterised. We have determined
the activity of these porphyrins as catalysts in chemical hydro-
carbon oxidations and have benchmarked them against iron
tetraphenylporphyrin and its tetrapentafluorophenyl analogue.
ChemSusChem 0000, 00, 1 – 16
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
9
&
ÞÞ
These are not the final page numbers!

C. J. Pickett et al.
Scheme 6. Proposed mechanism of electrocatalytic oxidation of hydrocarbons.
Table 3. Current efficiencies and turnover numbers of electrocatalytic oxi-
efficiency (14.5%). This work goes some way towards establish-
ing a proof of concept that alternatives to that of water oxida-
tion might usefully be explored in solar fuel-/feedstock-gener-
ating systems. Lower oxidation potentials and faster kinetics
are of course necessary targets.
dations of adamantane/cyclooctene catalysed by FeIII porphyrins.[a]
Catalyst
Oxidation poten- 1-Adamantanol
tial [V vs. SCE]
Cyclooctene oxide
yield [%][b] TON[c]
yield [%][b] TON[c]
Fe-1
Fe-2
Fe-3
Fe-4
1.36
1.27
1.31
1.42
9.4
9.8
2.5
14.5
35.8
2.6
2.7(28.1) 16.1
3.0(33.8) 20.9
0.7(25.9) 12.5
4.1(28.5) 28.2
7.5(29.1) 38.9
0.8(31.9) 17.4
4.6(28.7)
5.1(24.5)
3.4(27.6)
6.4(22.7)
12.3(31.4)
5.2(24.5)
[Fe(tppf20)Cl] 1.70
[Fe(tpp)Cl] 1.49
Experimental Section
Characterization
[a] Reactions were carried out in CH2Cl2 at RT under N2 for 1.5 h. [FeIII por-
phyrin]=0.45 mm; [Bu4NOH]=62 mm; [adamantane or cyclooctene]=
0.1 m; [Bu4NBF4]=0.2m; [H2O]=1.8 m. [b] Average value of two experi-
ments. [c] TON=moles of product/moles of catalyst; number in bracket is
the theoretical TON. The theoretical maximum TON is based upon the
total charge passed (F)/2 mol of catalyst.
All NMR spectra were recorded by using a Bruker Avance 300 and
referenced to the residual solvent signals. All MALDI mass spectra
were acquired by using an Applied Biosystems Voyager DE STR
MALDI equipped with linear and reflectron analysers. All high reso-
lution mass spectra were recorded by using a Thermofisher LTQ
Orbitrap XL.
Importantly, we show that in principle hydrocarbon oxida-
tion can be driven anodically using these iron porphyrins.
We report the first current efficiency data (yields) for this type
of electrochemical oxidation. We show that a current efficiency
of 36% can be achieved with an iron tetrapentafluorophenyl-
porphyrin electrocatalyst in a dichloromethane electrolyte at
+1.7 V versus SCE. Notably, the operating potential can be re-
duced to approximately +1.4 V versus SCE by using a thiolate
ligated iron porphyrin, although at the cost of a lower current
GC analyses were performed using a Hewlett–Packard 5890 II gas
chromatograph equipped with a flame ionization detector (FID)
and a 25 mꢂ0.32 mm 5% phenyl/95% dimethylpolysiloxane capil-
lary column (ID-BPX5 1.0). An injection volume of 1 mL was used in
all GC analyses. The products were identified by comparing their
retention times in GC with those of authentic samples. The yields
of oxidation products were determined by comparison against
standard curves (bromobenzene was used as an internal standard
for adamantine hydroxylation and n-dodecane for epoxidation of
cyclooctene).
&10
&
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 16
ÝÝ
These are not the final page numbers!

Porphyrins as Alternatives to Anodic Water Oxidation
Chemical oxidations were carried out in a sealed Schlenk tube
charged with the iron porphyrin catalyst, substrate, PhIO and dry
CH2Cl2 under N2 at 208C for 1 h. The reaction mixture was passed
through a short pad of dry Celite to remove unreacted PhIO. The
CH2Cl2 filtrate was injected directly to a gas chromatographer for
product analysis.
fied by using silica column chromatography (hexane/EtOAc, 10:1)
to afford 5 as a white solid (13.30 g, 78%). 1H NMR (300 MHz,
CDCl3): d=1.48 (s, 18H), 2.23 (s, 3H), 4.51 (s, 4H), 6.40 (d, J=
8.3 Hz, 2H), 7.04 ppm (t, J=8.3 Hz, 1H); 13C NMR (300 MHz, CDCl3):
d=8.63, 28.17, 66.44, 82.26, 105.28, 116.21, 126.14, 157.11,
168.37 ppm; NSI-HRMS: m/z calcd. for C19H28O6NH4: 370.2222
[M+NH4]+, found: 370.2224; elemental analysis calcd. (%) for
C19H28O6: C 64.75, H 8.01; found: C 64.88, H 8.13.
Cyclic voltammetric measurements were carried out using an Auto-
lab PGSTAT 30 potentiostat. A conventional three-electrode ar-
rangement was employed, consisting of a vitreous-carbon working
electrode, a platinum wire as the auxiliary electrode and SCE as
a reference electrode. All measurements were performed in dry
CH2Cl2 in the presence of 0.2m tetrabutylammonium tetrafluorobo-
rate as the supporting electrolyte at RT. All solutions were thor-
oughly degassed with nitrogen prior to being used, and during
the measurements a nitrogen atmosphere was maintained. The
electrocatalytic hydrocarbon oxidations were performed in a three-
electrode and three-compartment cell where adjacent compart-
ments were separated by a glass frit. A vitreous-carbon working
electrode was fitted into the central compartment, with the plati-
num-wire auxiliary electrode and Ag/AgCl (CH2Cl2, 0.5m [Bu4N]Cl)
reference electrode in the adjoining compartments. FeIII porphyrin
catalyst, [Bu4N][OH]·30H2O base, hydrocarbon substrate and
[Bu4N][BF4] (total volume 3 mL) electrolyte were added into the
working electrode compartment and dissolved in CH2Cl2 (3 mL).
The auxiliary and reference compartments were charged with
a CH2Cl2 solution containing [Bu4N][BF4] electrolyte (0.2m). All solu-
tions inside the cell were thoroughly degassed with N2 and sealed
under N2 prior to use. To avoid damage to the capillary column we
found it necessary to remove the supporting electrolyte. This was
achieved in the following way. After each electrolysis, CH2Cl2 was
removed from the solution mixture under a controlled vacuum
(42.7 kPa) at RT. This was optimal for removal of the solvent whilst
leaving the organic product intact. The residue was then treated
with dry diethyl (10 mL) ether to extract the product, leaving in-
soluble [Bu4N][BF4] which was removed by pipette filtration. The fil-
trate was evaporated under a controlled vacuum (42.7 kPa) to
remove diethyl ether. The residue was re-dissolved in dry CH2Cl2
(2 mL) with added internal reference. The resulting CH2Cl2 solution
was then subjected to GC analysis. Control experiments estab-
lished the efficacy of this extraction procedure.
1-Bromomentyl-2,6-bis(tert-butoxycarbonylmethoxy)benzene
(6): Compound 5 (13.00 g, 36.9 mmol) was dissolved in CCl4
(130 mL) before the addition of NBS (7.30 g, 0.041 mol) and benzo-
yl peroxide (90 mg, 0.372 mmol). The suspension was refluxed
overnight, during which time the pale yellow suspension turned
yellow-brown; TLC indicated that the reaction was complete. The
cooled suspension was filtered through a short pad of Celite, and
the filtrate was evaporated in vacuo to yield a brown solid. The
crude product was purified by using silica column chromatography
(hexane/EtOAc, 10:1) to afford 6 as a colourless oil (10.69 g, 67%).
1H NMR (300 MHz, CDCl3): d=1.47 (s, 18H), 4.60 (s, 2H), 4.81 (s,
2H), 6.42 (d, J=8.4 Hz, 2H), 7.17 ppm (t, J=8.3 Hz, 1H); 13C NMR
(300 MHz, CDCl3): d=23.13, 28.19, 66.52, 82.58, 105.54, 116.23,
129.93, 157.21, 167.86 ppm; ESI-HRMS: m/z calcd. for
C19H27BrO6NH4: 448.1329 [M+NH4]+, found: 448.1324.
tert-Butyl 2-{2-[(acetylsulfanyl)methyl]-3-[2-(tert-butoxy)-2-oxoe-
thoxy]phenoxy}acetate (7): KSAc (3.39 g, 29.7 mmol) was added
to 6 (10.69 g, 24.8 mmol) in acetone (200 mL). The suspension was
stirred at RT overnight, during which time the pale yellow suspen-
sion turned light brown. TLC indicated that the reaction was com-
plete. The suspension was filtered through a short pad of Celite,
and the filtrate was evaporated at reduced pressure to yield
a brown solid. The crude product was purified by using silica
column chromatography (hexane/ether, 5:1) as a pale yellow oil
1
(6.85 g, 65%). H NMR (300 MHz, CDCl3): d=1.46 (s, 18H), 2.31 (s,
3H), 4.39 (s, 2H), 4.53 (s, 4H), 6.40 (d, J=8.4 Hz, 2H), 7.11 ppm (t,
J=8.4 Hz, 1H); 13C NMR (300 MHz, CDCl3): d=22.47, 28.17, 30.42,
66.56, 82.44, 105.37, 114.31, 128.56, 157.24, 167.97, 196.24 ppm;
NSI-HRMS: m/z calcd. for C21H30O7SNH4: 444.2050 [M+NH4]+,
found: 444.2049.
2-[(Acetylsulfanyl)methyl]-3-(carboxymethoxy)phenoxyacetic
acid (8): Trifluoroacetic acid (12.51 mL, 0.161 mol) was added to 7
(6.85 g, 16.1 mmol) in CH2Cl2 (200 mL). The pale yellow solution
was stirred at RT overnight, during which time the pale yellow so-
lution turned yellow, and TLC indicated that the reaction was com-
plete. Water (200 mL) was added, and the white precipitate formed
was collected by filtration. The filtrate was washed with water (2ꢂ
200 mL) and dried over MgSO4. Removal of all solvents gave an
off-white solid, which was washed with CH2Cl2. The combined yield
Materials
All solvents were dried and degassed prior to use. All commercial
chemicals were used as received unless otherwise stated. All reac-
tions were carried out under a dry N2-atmosphere unless otherwise
stated. PhIO was prepared according to a published procedure[18]
and its purity was checked by iodometric titration to be above
99%. Starting materials, 5-(2-nitrophenyl)-dipyrromethane,[19] 4-
iodo-benzaldehyde[20] and 5-(pentafluorophenyl)-dipyrromethane[21]
were synthesised according to literature procedures.
1
of 8 was 4.14 g (82%). H NMR (300 MHz, CD3OD): d=2.29 (s, 3H),
4.36 (s, 2H), 4.68 (s, 4H), 6.55 (d, J=8.4 Hz, 2H), 7.17 ppm (t, J=
8.4 Hz, 1H); 13C NMR (300 MHz, CD3OD): d=23.09, 30.13, 66.60,
106.66, 115.41, 129.77, 158.52, 172.384, 198.18 ppm; NSI-HRMS: m/
z calcd. for C13H14O7SH: 315.0533 [M+H]+, found: 315.0537; ele-
mental analysis calcd. (%) for C13H14O7S: C 49.68, H 4.49; found:
C 49.77, H 4.57.
Synthesis
2,6-Bis(tert-butoxycarbonylmethoxy)toluene (5): K2CO3 (16.03 g,
0.116 mol) was added to methylresorcinol (6.00 g, 4.83 mmol) in
acetone (200 mL), and the dark red suspension was heated at 608C
for 2 h. The dark red suspension turned light upon addition of tert-
butyl bromoacetate (15.70 mL, 0.106 mol). The mixture was then
refluxed for two days, turning into a white suspension. TLC indicat-
ed the reaction was complete. The white suspension was filtered
through a short pad of Celite and the filtrate was evaporated at re-
duced pressure to yield a yellow solid. The crude product was puri-
2-{2-[(Acetylsulfanyl)methyl]-3-{[(2-formylphenyl)carbamoyl]me-
thoxy}phenoxy}-N-(2-formylphenyl)acetamide (9): To a suspension
of 8 (2.30 g, 7.32 mmol), 2-aminobenzaldehyde (2.21 g, 18.2 mol)
and 2-chloro-1-methylpyridinium iodide (4.50 g, 17.6 mmol) in dry
CH2Cl2 (150 mL) was added Et3N (5.05 mL, 35.9 mmol). The mixture
was stirred at RT overnight after which TLC indicated that the reac-
ChemSusChem 0000, 00, 1 – 16
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
11
&
ÞÞ
These are not the final page numbers!

C. J. Pickett et al.
tion was complete. Removal of the solvents afforded a yellow
solid. The crude product was purified by using silica column chro-
matography (CH2Cl2/ether, 20:1) to yield 9 as a white solid (2.30 g,
turned dark brown; TLC indicated that the reaction was complete.
Aqueous NaOH solution (0.1m, 100 mL) was added to quench the
reaction. The mixture was extracted with EtOAc (3ꢂ50 mL). The
combined organic layers were washed with brine (2ꢂ150 mL)
twice and dried over Na2SO4. Removal of the solvents afforded
a black oily residue. The crude product was purified by using silica
column chromatography (hexane/ether, 5:1). The pure product was
obtained by recrystallisation from a solvent mixture of CH2Cl2 and
1
60%). H NMR (300 MHz, CD2Cl2): d=2.19 (s, 3H), 4.70 (s, 4H), 4.86
(s, 2H), 6.67 (d, J=8.4 Hz, 2H), 7.24 (t, J=8.4 Hz, 1H), 7.32 (td, J=
7.5, 1.0 Hz, 2H), 7.63–7.69 (m, 2H), 7.76 (dd, J=7.6, 1.6 Hz, 2H),
8.77 (d, J=8.4 Hz, 2H), 9.960 (s, 1H), 9.958 (s, 1H), 11.85 ppm (br, s,
2H); 13C NMR (300 MHz, CDCl3): d=22.34, 30.40, 69.25, 107.06,
116.40, 120.64, 123.18, 124.08, 129.29, 136.14, 136.21, 140.05,
156.93, 168.04, 195.26, 195.67 ppm; NSI-HRMS: m/z calcd. for
C27H24N2O7SH: 521.1377 [M+H]+, found: 521.1372; elemental analy-
sis calcd. (%) for C35H53Br2O2Si2: C 62.30, H 4.65, N 5.38; found:
C 62.44, H 4.53, N 5.25.
1
hexane as a brownish white solid (2.30 g, 25%). H NMR (300 MHz,
CD2Cl2): d=2.10 (s, 6H), 5.91–5.95 (m, 3H), 6.13 (dd, J=6.0, 2.7 Hz,
2H), 6.66–6.68 (m, 2H), 7.03–7.13 (m, 3H), 8.04 ppm (s, 2H);
13C NMR (300 MHz, CD2Cl2): d=20.77, 39.10, 106.82, 108.81, 116.64,
127.41, 129.88, 131.50, 138.07, 138.25 ppm; APCI-HRMS: m/z calcd.
for C17H18N2H: 251.1543 [M+H]+; found: 251.1542; elemental analy-
sis calcd. (%) for C17H18N2: C 81.56, H 7.25, N 11.19; found: C 81.61,
H 7.34, N 11.07.
5,15-Bis-(2-nitrophenyl)-10,20-bis-(4-iodophenyl)porphyrin (10):
To a clear solution of 5-(2-nitrophenyl)-dipyrromethane (5.37 g,
20 mmol) and 4-iodo-benzaldehyde (4.64 g, 20 mmol) in degassed
CH2Cl2 (2 L) was added TFA (2.80 mL, 35.9 mmol), and the resulting
dark red solution was stirred at RT overnight. DDQ (4.77 g,
21 mmol) was then added, and the resulting dark green solution
was stirred at RT for 3 h. Et3N (6.0 mL, 0.0426 mol) was added to
the dark green mixture, which was stirred at RT for another 10 min.
The black CH2Cl2 solution was passed through a neutral alumina
column and eluted with CH2Cl2. The dark red eluent containing
cis-/trans-isomers was concentrated to afford a dark violet solid
(2.13 g, 11%). The solid was pure enough for the next reaction
Porphyrin 1(Ac): To a suspension of 11 (0.45 g, 0.502 mmol), 8
(0.20 g, 0.636 mmol) and 2-chloro-1-methylpyridinium iodide
(0.32 g, 1.25 mmol) in dry CH2Cl2 (450 mL) was added Et3N
(0.25 mL, 1.78 mmol). The mixture was stirred at RT for 4 h after
which TLC indicated that the reaction was complete. Removal of
the solvents gave rise to a dark violet residue. The crude product
was purified by using silica column chromatography (CH2Cl2/ether,
20:1) as a dark violet solid (0.41 g, 70%). Crystals for X-ray analysis
were grown by ether vapour diffusion into the concentrated
CH2Cl2 solution of 1(Ac). 1H NMR (300 MHz, CD2Cl2): d=ꢀ2.86 (s,
2H), ꢀ2.62 (s, 2H), 1.01 (s, 3H), 3.57 (d, J=14.4 Hz, 2H), 3.75 (d, J=
14.4 Hz, 2H), 5.67(d, J=8.4 Hz, 2H), 6.47 (t, J=8.4 Hz, 1H), 7.36 (s,
2H), 7.62–7.65 (br, m, 1H), 7.72–7.80 (m, 4H), 7.90 (td, J=
7.9,1.6 Hz, 2H), 8.00–8.12 (br, m, 5H), 8.63 (dd, J=7.5 Hz, J=1.4 Hz,
2H), 8.66 (dd, J=8.3, 0.9 Hz, 2H), 8.80 (m, 4H), 8.90 ppm (m, 4H);
MALDI: m/z calcd. for C57H40I2N6O5S: 1174.1 [M]+; found: 1174.1;
elemental analysis calcd. (%) for C57H40I2N6O5S: C 58.27, H 3.43,
N 7.15; found: C 58.42, H 3.34, N 7.16.
1
1
judged by H NMR. H NMR (300 MHz, D6-DMSO): d=ꢀ2.91 (s, 2H),
7.89–8.02 (m, 3H), 8.10–8.20 (m, 9H), 8.37–8.43 (m, 2H), 8.55–8.59
(m, 2H), 8.74 (d, J=4.7 Hz, 4H), 8.85 ppm (d, J=4.7 Hz, 4H); APCI-
HRMS: m/z calcd. for C44H26I2N6O4H: 957.0178 [M+H]+; found:
957.0158; elemental analysis calcd. (%) for C44H26I2N6O4: C 55.25,
H 2.74, N 8.79; found: C 55.37, H 2.63, N 8.69.
cis-5,15-Bis-(2-aminophenyl)-10,20-bis-(4-iodophenyl)porphyrin
(11): To a solution of 10 (1.46 g, 1.53 mmol) in concentrated HCl
(250 mL) was added SnCl2 (8.7 g, 0.046 mol). The dark green mix-
ture was stirred at RT for 3 days. TLC indicated that the reaction
was complete. The dark green mixture was treated with concen-
trated NH3 ·H2O at 0–58C until pH>8. To this basic suspension was
added EtOAc (250 mL), and the mixture was stirred at RT for 2 h.
The EtOAc layer was separated and the aqueous layer was extract-
ed with EtOAc (2ꢂ200 mL). The combined EtOAc layers were
washed with water until the washing had a pH of approximately 8,
and then dried over Na2SO4. The desired cis-isomer was separated
from the partially reduced product and trans-isomer by using silica
column chromatography (CH2Cl2!CH2Cl2/EtOAc 100:1) as a dark
violet crystalline solid. The isolated trans-isomer was heated at
808C in toluene with silica, during which time some trans-isomer
molecules converted to their cis-isomers. The cis-isomer was sepa-
rated from trans-isomer by using silica column chromatography in
the same way as aforementioned. The combined yield was 0.67 g
(49%). 1H NMR (300 MHz, CD2Cl2): d=ꢀ2.82 (s, 2H), 3.58 (s, 4H),
7.12–7.19 (m, 4H), 7.61 (td, J=7.7, 1.5 Hz, 2H), 7.84 (dd, J=7.4,
1.4 Hz, 2H), 7.94–7.97 (m, 4H), 8.12 (d, J=8.5 Hz, 4H), 8.85 (d, J=
4.8 Hz, 4H), 8.91 ppm (d, J=4.8 Hz, 4H); NSI-HRMS: m/z calcd. for
C44H30I2N6H: 897.0694 [M+H]+; found: 897.0694; elemental analysis
calcd. (%) for C44H30I2N6: C 58.94, H 3.37, N 9.37; found: C 59.05,
H 3.25, N 9.27.
5,15-Bis-(2-nitrophenyl)-10,20-bis-(2,6-dimethylphenyl)porphyrin
(13): To a solution of 12 (2.30 g, 9.19 mmol) and 2-nitrobenzalde-
hyde (1.39 g, 9.20 mmol) in degassed CH2Cl2 (1 L) was added TFA
(1.30 mL, 16.7 mmol). The mixture was stirred at RT overnight. DDQ
(3.13 g, 13.8 mmol) was then added to this dark red solution. The
resulting dark green solution was stirred at RT for 6 h. Et3N (3.0 mL,
21.3 mmol) was then added, and the mixture was stirred at RT for
another 10 min. The black solution was passed through a neutral
alumina column and eluted with CH2Cl2. The dark red eluent con-
taining cis-/trans-isomers was concentrated to afford a dark violet
solid (0.81 g, 12%). The desired cis-isomer was separated from the
trans-isomer by using neutral alumina column chromatography
(CH2Cl2) to afford 13 as a dark violet crystalline solid (0.32 g, 4.6%).
1H NMR (300 MHz, CDCl3): d=ꢀ2.50 (s, 2H), 1.85 (s, 6H), 1.91 (s,
6H), 7.43–7.48 (m, 4H), 7.57–7.62 (m, 2H), 7.90–8.00 (m, 4H), 8.21
(dd, J=8.1, 1.8 Hz, 2H), 8.48 (dd, J=7.7, 1.8 Hz, 2H), 8.59 (d, J=
4.8 Hz, 4H), 8.65 ppm (d, J=4.8 Hz, 4H); MALDI: calcd. m/z for
C48H36N6O4: 760.3 [M]+; found: 760.3; elemental analysis calcd. (%)
for C48H36N6O4: C 75.77, H 4.77, N 11.05; found: C 75.91, H 4.67,
N 10.91.
cis-5,15-Bis-(2-aminophenyl)-10,20-bis-(2,6-dimethylphenyl)por-
phyrin (14): To a solution of 13 (0.32 g, 0.42 mmol) in concentrated
HCl (50 mL) was added SnCl2 (0.8 g, 4.22 mmol). The dark green
mixture was stirred at RT for 20 h. The following TLC indicated that
the reaction was complete. The mixture was treated with aqueous
KOH solution at 0–58C until pH>8. EtOAc (250 mL) was added to
this basic suspension, and the mixture was stirred at RT for 2 h.
5-[2,6-Dimethylphenyl]dipyrromethane (12): 2,6-Dimethyl-benzal-
dehyde (5.00 g, 37.2 mol) was dissolved in freshly distilled pyrrole
(27 mL, 0.381 mol), and the clear solution was degassed for 30 min
before TFA (0.30 mL, 3.85 mmol) was added. The light yellow mix-
ture was then stirred at RT for 2 h, during which time the solution
&12
&
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 16
ÝÝ
These are not the final page numbers!

Porphyrins as Alternatives to Anodic Water Oxidation
The EtOAc layer was separated, and the aqueous layer was extract-
ed with EtOAc (2ꢂ200 mL). The combined EtOAc layers were
washed with water until the washing had a pH of approximately 8,
and then dried over Na2SO4. The desired cis-isomer was separated
from small amounts of trans-isomer by using silica column chroma-
tography (CH2Cl2/ether, 20:1) as a dark violet crystalline solid
Removal of the solvents gave rise to an oily residue which was pu-
rified by silica column chromatography (CH2Cl2/ether, 5:1) to afford
16 as a colourless oil (4.75 g, 58%). H NMR (300 MHz, CDCl3): d=
3.37 (br, s, 6H), 3.54–3.57 (m, 4H), 3.84–3.87 (m, 4H), 5.35 (s, 4H),
6.88 (d, J=8.4 Hz, 2H), 7.40 (t, J=8.4 Hz, 1H), 10.51 ppm (s, 1H);
13C NMR (300 MHz, CDCl3): d=59.16, 68.30, 71.62, 93.95, 108.81,
116.24, 135.74, 159.52, 189.39 ppm; ESI-HRMS: m/z calcd. for
C15H22O7Na: 337.1258 [M+Na]+; found: 337.1259.
1
1
(0.17 g, 58%). H NMR (300 MHz, CD2Cl2): d=ꢀ2.61 (s, 2H), 1.84 (s,
6H), 1.91 (s, 6H), 3.65 (s, 4H), 7.11–7.18 (m, 4H), 7.46–7.50 (m, 4H),
7.56–7.64 (m, 4H), 7.86 (dd, J=7.4, 1.4 Hz, 2H), 8.66 (d, J=4.8 Hz,
4H), 8.85 ppm (d, J=4.8 Hz, 4H). ESI-HRMS: m/z calcd. for
C48H40N6: 700.3 [M]+; found: 700.3; elemental analysis calcd. (%) for
C48H40N6: C 82.26, H 5.75, N 11.99; found: C 78.54, H 5.76, N 10.18.
2-({2,6-Bis[(2-methoxyethoxy)methoxy]phenyl}(1H-pyrrol-2-yl)-
methyl)-1H-pyrrole (17): Compound 16 (4.65 g, 0.0148 mol) was
dissolved in freshly distilled pyrrole (27 mL, 0.381 mol). The clear
solution was degassed for 30 min before addition of TFA (0.12 mL,
1.54 mmol). The light yellow mixture was stirred at RT for 10 min,
during which time the light yellow solution turned to dark brown.
The following TLC indicated that the reaction was complete. Aque-
ous NaOH solution (0.1m, 100 mL) was added to quench the reac-
tion. The mixture was extracted with EtOAc (3ꢂ50 mL). The com-
bined organic layers were washed with brine twice (2ꢂ150 mL)
and dried over Na2SO4. Removal of the solvents gave rise to
a black oily residue. The crude product was purified by using silica
column chromatography (CH2Cl2/ether, 10:1) as a pale yellow oil
Porphyrin 2(Ac): To a suspension of 14 (0.16 g, 0.228 mmol), 8
(90 mg, 0.286 mmol) and 2-chloro-1-methylpyridinium iodide
(0.15 g, 0.587 mmol) in dry CH2Cl2 (220 mL) was added Et3N
(0.16 mL, 1.13 mmol). The mixture was stirred at RT overnight. The
following TLC indicated that the reaction was complete. Removal
of the solvents gave rise to dark violet residue. The crude product
was purified by using silica column chromatography (CH2Cl2, then
CH2Cl2/ether, 50:1!20:1) to afford 2(Ac) as a dark violet solid
(0.16 g, 72%). Crystals for X-ray analysis were grown by ether
vapour diffusion into the concentrated CH2Cl2 solution of 2(Ac).
1H NMR (300 MHz, CD2Cl2): d=ꢀ2.75 (s, 2H), ꢀ2.41 (s, 2H), 0.942
(s, 3H), 1.01 (s, 3H), 1.58 (s, 3H), 1.73 (s, 3H), 2.25 (s, 3H), 3.57 (d,
J=14.3 Hz, 2H), 3.84 (d, J=14.3 Hz, 2H), 5.76 (d, J=8.4 Hz, 2H),
6.56 (t, J=8.4 Hz, 1H), 7.29 (d, J=7.6 Hz, 1H), 7.36–7.42 (m, 2H),
7.52 (t, J=7.6 Hz, 1H), 7.57–7.64 (m, 4H), 7.75 (td, J=7.5, 1.3 Hz,
2H), 7.88 (td, J=7.8, 1.6 Hz, 2H), 8.53 (d, J=4.8 Hz, 2H), 8.60 (dd,
J=8.2, 0.99 Hz, 2H), 8.66 (dd, J=7.5, 1.5 Hz, 2H), 8.70 (d, J=4.8 Hz,
2H), 8.73 (d, J=4.8 Hz, 2H), 8.83 ppm (d, J=4.7 Hz, 2H); NSI-
HRMS: m/z calcd. for C61H50N6O5SH: 979.3636 [M+H]+; found:
979.3636; C61H50N6O5SNa: 1001.3461 [M+Na]+; found: 1001.3450.
1
(2.47 g, 39%). H NMR (300 MHz, CDCl3): d=3.37 (br, s, 6H), 3.50–
3.52 (m, 4H), 3.65 (br, s, 4H), 5.17 (br, s, 4H), 5.937–5.939 (m, 2H),
6.63–6.65 (m, 2H), 6.84 (d, J=8.3 Hz, 2H), 7.16 ppm (t, J=8.3 Hz,
1H); 13C NMR (300 MHz, CDCl3): d=33.21, 59.15, 68.00, 71.76,
94.13, 106.27, 107.99, 109.63, 116.37, 121.30, 128.46, 132.81,
155.69 ppm; ESI-HRMS: m/z calcd. for C23H30N2O6Na: 453.1996
[M+Na]+; found: 453.1993.
Porphyrin 3(Ac): To a clear solution of dipyrromethane 17 (0.25 g,
0.58 mmol) and dialdehyde 9 (0.15 g, 0.288 mmol) in degassed
CH2Cl2 (100 mL) was added TFA (0.15 mL, 1.92 mmol). The mixture
was stirred at RT overnight. To this very dark pink solution was
then added DDQ (0.25 g, 1.10 mmol). The resulting dark greenish
solution was stirred at RT for 3 h before Et3N (0.30 mL, 2.13 mmol)
was added. The black mixture was stirred for 30 min, and then
passed through a short neutral alumina column, which was
washed initially with CH2Cl2, then CH2Cl2/THF (1:1). The concentrat-
ed black eluent was purified by using silica column chromatogra-
phy (EtOAc/CH2Cl2, 1:1 to 1.5:1) to yield a dark violet solid (9 mg,
2.3%). The dark violet crystals for X-ray analysis were obtained by
ether vapour diffusion into the concentrated CH2Cl2 solution of
3(Ac). 1H NMR (300 MHz, CDCl3): d=ꢀ2.60 (s, 2H), ꢀ2.37 (s, 2H),
0.52 (s, 3H), 2.70–2.75 (m, 4H), 2.86–3.01 (m, 12H), 3.22–3.29 (m,
8H), 3.36–3.39 (m, 2H), 3.51–3.54 (m, 2H), 3.65 (d, J=14.1 Hz, 2H),
3.84 (d, J=14.1 Hz, 2H), 4.05 (s, 2H), 4.60 (s, 2H), 4.75 (s, 2H), 5.23
(s, 2H), 5.78 (d, J=8.4 Hz, 2H), 6.55 (t, J=8.4 Hz, 1H), 7.13–7.21 (m,
3H), 7.42 (d, J=8.1 Hz, 2H), 7.63–7.77 (m, 6H), 7.12 (td, J=7.9,
1.6 Hz, 2H), 8.67–8.73 (m, 8H), 8.87 ppm (m, 4H); MALDI: m/z
calcd. for C73H74N6O17S: 1138.5 [M]+; found: 1138.4.
1,3-Bis[(2-methoxyethoxy)methoxy]benzene (15): To a solution of
resorcinol (10.00 g, 90.8 mmol) in dry DMF (160 mL) was added
NaH (60% dispersion in mineral oil, 8.72 g, 0.218 mol) portion by
portion at 0–58C. The off-white suspension was stirred at RT for
2 h. The suspension was cooled to 0–58C again, followed by slow
addition of MEMCl (22.60 mL, 0.199 mol). The mixture was warmed
to RT and stirred overnight; TLC indicated that the reaction was
complete. Aqueous NaHCO3 solution (0.5m, 200 mL) was added
slowly to quench the reaction. The mixture was extracted with
EtOAc (1ꢂ200 mL, 2ꢂ100 mL). The combined organic layers were
washed with brine twice (2ꢂ400 mL) and dried over MgSO4.
Removal of the solvents afforded an oily residue. The crude prod-
uct was purified by using silica column chromatography (EtOAc/
1
hexane, 1:3) to yield 15 as a colourless oil (17.50 g, 67%). H NMR
(300 MHz, CDCl3): d=3.40 (br, s, 6H), 3.54–3.57 (m, 4H), 3.80–3.83
(m, 4H), 5.25 (s, 4H), 6.70–6.75 (m, 3H), 7.13 ppm (t, J=8.3 Hz,
1H); 13C NMR (300 MHz, CDCl3): d=59.17, 67.81, 71.74, 93.63,
105.22, 109.78, 130.08, 158.47 ppm; ESI-HRMS: m/z calcd. for
C14H22O6Na: 309.1309 [M+Na]+; found: 309.1310.
Porphyrin 4(Ac): To a clear solution of 5-(pentafluorophenyl)dipyr-
romethane (0.56 g, 1.79 mmol) and dialdehyde
9
(0.47 g,
2,6-Bis[(2-methoxyethoxy)methoxy]benzaldehyde (16): To a solu-
tion of 15 (7.50 g, 26.2 mmol) in dry degassed THF (120 mL) was
added nBuLi (1.6m in hexane, 18.00 mL, 0.288 mol) at 0–58C. The
yellow solution was stirred for 45 min before dry DMF (4.0 mL,
0.0517 mol) was added at 0–58C. The light yellow mixture was
warmed to RT and stirred for 2 h after which TLC indicated that the
reaction was complete. Aqueous HCl solution (0.1m, 150 mL) was
added slowly to quench the reaction. The mixture was extracted
with EtOAc (1ꢂ150 mL, 2ꢂ100 mL). The combined organic layers
were washed with brine twice (2ꢂ350 mL) and dried over MgSO4.
0.90 mmol) in degassed CH2Cl2 (320 mL) was added TFA (0.47 mL,
6.03 mmol). The mixture was stirred at RT overnight. To this very
dark pink solution DDQ (0.80 g, 3.52 mmol) was added. The result-
ing dark greenish solution was stirred at RT for 7 h before Et3N
(0.50 mL, 3.55 mmol) was added. The black mixture was stirred for
30 min and then passed through a short neutral alumina column,
which was washed initially with CH2Cl2, then CH2Cl2/THF (1:1). The
concentrated black eluent was purified by using silica column
chromatography (CH2Cl2/ether, 100:1) to yield a dark violet solid
(65 mg, 6.6%). Dark violet crystals for X-ray analysis were obtained
ChemSusChem 0000, 00, 1 – 16
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&13
&
ÞÞ
These are not the final page numbers!

C. J. Pickett et al.
by vapour diffusion of ether into the concentrated CH2Cl2 solution
Fe-4: Porphyrin 4(Ac) (30 mg, 27.2 mmol), FeBr2 (60 mg,
0.278 mmol) and lutidine (0.10 mL) were added to dry degassed
toluene (25 mL). The suspension was refluxed overnight. The ensu-
ing TLC indicated that the reaction was complete. To this cooled
suspension was added NaOMe (0.2m, 10 mL). The mixture was
stirred at RT for 2 h before it was poured into a mixture of HCl
(0.1m, 100 mL) and toluene (25 mL). The toluene layer was separat-
ed, washed with water twice (2ꢂ50 mL) and dried over Na2SO4. Re-
moval of the solvents under vacuo resulted in a dark solid residue.
The crude product was purified by using silica column chromatog-
raphy (CH2Cl2, then CH2Cl2/THF, 50:1) to yield a brown-red solid
(15 mg, 50%) after all the solvents were evaporated under vacuo.
The brown-red solid was dried under vacuo and re-dissolved in dry
degassed CH2Cl2 (3 mL). The concentrated dark red CH2Cl2 solution
was then layered with hexane (5 mL) under N2 to yield a dark
brown-red solid, which was collected and dried under vacuo.
MALDI: m/z calcd. for C55H27F10FeN6O4S: 1113.1 [M]+; found: 1113.1.
1
of 4(Ac). H NMR (300 MHz, CD2Cl2): d ꢀ2.98 (s, 2H), ꢀ2.69 (s, 2H),
0.856 (s, 1H), 0.859 (s, 1H), 3.60 (d, J=14.3 Hz, 2H), 3.77 (d, J=
14.3 Hz, 2H), 5.69 (d, J=8.4 Hz, 2H), 6.50 (t, J=8.4 Hz, 1H), 7.43 (s,
2H), 7.79 (td, J=7.5, 1.2 Hz, 2H), 7.93 (td, J=7.9, 1.5 Hz, 2H), 8.62
(dd, J=8.2, 0.9 Hz, 2H), 8.65 (dd, J=7.5, 1.5 Hz, 2H), 8.75–8.77 (m,
2H), 8.87–8.91 (m, 4H), 9.03–9.04 ppm (m, 2H); NSI-HRMS: m/z
calcd. for C57H32F10N6O5SH: 1103.2068 [M+H]+; found: 1103.2043;
C57H32F10N6O5SNa: 1125.1893 [M+Na]+; found: 1125.1854.
Fe-1: Porphyrin 1(Ac) (60 mg, 511 mmol), FeBr2 (110 mg,
0.511 mmol) and lutidine (0.10 mL) were added to dry degassed
toluene (30 mL). The suspension was refluxed overnight. The ensu-
ing TLC indicated that the reaction was complete. To this cooled
suspension was added 0.2m NaOMe (10 mL). The mixture was
stirred at RT for 2 h before it was poured into a mixture of 0.1m
HCl (100 mL) and toluene (20 mL). The toluene layer was separat-
ed, washed with water twice (2ꢂ50 mL) and dried over Na2SO4.
Removal of the solvents under vacuo resulted in a dark solid resi-
due. The crude product was purified by using silica column chro-
matography (CH2Cl2, then hexane/THF, 1:1) to yield a brown-red
solid (38 mg, 63%) after all the solvents were evaporated under
vacuo. The brown-red solid was dried under vacuo and re-dis-
solved in dry degassed CH2Cl2 (2 mL). The concentrated dark red
CH2Cl2 solution was then layered with hexane (3 mL) under N2 to
produce a microcrystalline dark brown-red solid. MALDI: m/z calcd.
for C55H35FeI2N6O4S: 1185.0 [M]+; found: 1185.0.
X-ray crystallography
For each sample, crystals were suspended in oil, and one was
mounted on a glass fibre and fixed in the cold nitrogen stream of
the diffractometer. Data were collected by recording synchrotron
radiation (l=0.6889) using a Rigaku Saturn724+ diffractometer
equipped with confocal monochromator [1(Ac)] or MoKa (l=
0.71073 ꢁ) radiation using an Oxford Diffraction Xcalibur-3 CCD dif-
fractometer equipped with a graphite monochromator [3(Ac) and
4(Ac)] or a Bruker-Nonius Apex II diffractometer equipped with
confocal mirrors [(thf)Fe-1·(thf)·(CH2Cl2)]. Data were processed by
using the CrystalClear SM Expert [1(Ac)],[22a] CrysAlisPro [3(Ac) and
4(Ac)][22b] or DENZO and COLLECT [(thf)Fe-1·(thf)·(CH2Cl2)] soft-
ware.[22c] Structures were determined by the direct methods rou-
tines in SIR92 [1(Ac) and 4(Ac)],[22d] SHELXS [3(Ac)][22e] or SIR97
[(thf)Fe-1·(thf)·(CH2Cl2)][22f] programmes and refined by full-matrix
least-squares methods on F2 in SHELXL.[22e] Non-hydrogen atoms
were refined with anisotropic thermal parameters. Hydrogen
atoms H(1), H(21), H(167) and H(367) were freely refined in 3(Ac),
whereas all other hydrogen atoms were included in idealised posi-
tions and their Uiso values were set to ride on the Ueq values of the
parent carbon or nitrogen atom. In [(thf)Fe(1)·(thf)·(CH2Cl2)] and
1(Ac), disordered solvent regions were handled using the SQUEEZE
procedure.[22g] A summary of the crystallographic data for previous-
ly unreported structures is given in Table 4. CCDC 886113 to
CCDC 886116 contain the supplementary crystallographic data for
1(Ac), 3(Ac), 4(Ac) and [(thf)Fe-1·(thf)·(CH2Cl2)], respectively. These
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Fe-2: Porphyrin 2(Ac) (40 mg, 40.8 mmol), FeBr2 (90 mg,
0.417 mmol) and lutidine (0.10 mL) were added to dry degassed
toluene (30 mL). The suspension was refluxed overnight. The ensu-
ing TLC indicated the reaction was complete. To this cooled sus-
pension was added NaOMe (0.2m, 10 mL). The mixture was stirred
at RT for 2 h before it was poured into a mixture of HCl (0.1m,
100 mL) and toluene (20 mL). The toluene layer was separated,
washed with water twice (2ꢂ50 mL) and dried over Na2SO4. Re-
moval of the solvents under vacuo gave a dark solid residue. The
crude product was purified by using silica column chromatography
(CH2Cl2, then CH2Cl2/THF, 50:1) to yield a red-brown solid (25 mg,
61%) after all the solvents were evaporated under vacuo. The red-
brown solid was dried under vacuo and re-dissolved in dry and de-
gassed CH2Cl2 (3 mL). The concentrated dark red CH2Cl2 solution
was then layered with degassed hexane (4 mL) under N2 to give
microcrystalline dark red-brown solid. MALDI: m/z calcd. for
C59H45FeN6O4S: 989.3 [M]+; found: 989.3.
Fe-3: Porphyrin 3(Ac) (30 mg, 22.4 mmol), FeBr2 (50 mg,
0.232 mmol) and lutidine (0.10 mL) were added to dry degassed
toluene (25 mL). The suspension was refluxed overnight. The ensu-
ing TLC indicated that the reaction was complete. To this cooled
suspension was added NaOMe (0.2m, 10 mL). The mixture was
stirred at RT for 2 h before it was poured into a mixture of HCl
(0.1m, 100 mL) and toluene (25 mL). The toluene layer was separat-
ed, washed with water twice (2ꢂ50 mL) and dried over Na2SO4.
Removal of the solvents under vacuo produced a dark solid resi-
due. The crude product was purified by using silica column chro-
matography (CH2Cl2, then CH2Cl2/THF, 20:1) to yield a brown-red
solid after all the solvents were evaporated under vacuo. The
brown-red solid was dried under vacuo and re-dissolved in dry de-
gassed CH2Cl2 (2 mL). The concentrated dark red CH2Cl2 solution
was then layered with hexane (4 mL) under N2 to yield dark violet
needles (16 mg, 53%). MALDI: m/z calcd. for C71H69N6O16SFe:
1149.4 [M]+; found: 1149.4.
Acknowledgements
We thank the EPSRC and our colleagues in the SolarCAP consorti-
um (EPSRC F047878/1) for supporting this work. K.A. thanks the
University of Hail, Saudi Arabia, for providing a Scholarship. The
authors thank the EPSRC National Mass Spectrometry Service,
Swansea, for mass spectroscopy data, and the EPSRC National
Crystallography Service for collection of diffraction data.[23]
Keywords: electrocatalysis · hydrocarbons · iron · oxidation ·
porphyrins
&14
&
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 16
ÝÝ
These are not the final page numbers!

Porphyrins as Alternatives to Anodic Water Oxidation
Table 4. Crystallographic Data for 1(Ac), 3(Ac), 4(Ac) and Fe-1.
Property
1(Ac)
3(Ac)
4(Ac)
[(thf)Fe-1·(thf)·(CH2Cl2)]
formula
C57H40I2N6O5S
1174.81
monoclinic
P21/n
C73H74N6O17S
1339.44
triclinic
C57H32F10N6O5S
1102.95
triclinic
C59H43FeI2N6O5S, C4H8O, CH2Cl2
1414.73
triclinic
molecular weight [gcmꢀ3
crystal system
space group
]
¯
P1
¯
P1
¯
P1
crystal description, colour
crystal dimensions [mm]
block, red
0.02ꢂ0.03ꢂ0.05
13.292(5)
28.413(10)
14.958(5)
90
109.105(5)
90
5338(3)
4
sliver, red-brown
0.15ꢂ0.24ꢂ0.25
13.5315(9)
13.6265(9)
19.5980(12)
103.138(5)
100.302(5)
104.542(6)
3297.9(4)
2
prism, dark purple
0.17ꢂ0.20ꢂ0.35
10.7867(3)
14.5984(4)
17.4320(5)
70.899(2)
85.460(2)
70.258(3)
2439.88(12)
2
plate, brown
0.01ꢂ0.09ꢂ0.14
12.4593(5)
17.4383(9)
18.0032(8)
107.406(2)
107.097(2)
101.066(3)
3395.3(3)
2
a [ꢁ]
b [ꢁ]
c [ꢁ]
a [8]
b [8]
g [8]
V [ꢁ3]
Z
1calc [gcmꢀ3
2qmax [8]
T [K]
]
1.459
48.42
120(2)
1.349
50.0
140(1)
1.501
50.00
140(2)
1.384
50.00
120(2)
data collected/unique
41081/9358
7686
1.270
44388/11576
6475
0.127
31261/8572
6284
0.164
53971/11894
7325
1.291
‘observed’ reflections (i>2si)
m [mmꢀ1
]
R1 (I>2sI)
0.080
0.042
0.049
0.106
wR2 (all data)
0.219
0.089
0.124
0.274
largest diff. peak and hole
2.93, ꢀ1.42
0.27, ꢀ0.30
0.54, ꢀ0.39
1.63, ꢀ1.98
[1] a) C.-M. Che, V. K.-L. Lo, C.-Y. Zhou, J.-S. Huang, Chem. Soc. Rev. 2011, 40,
1950–1975; b) D. Mansuy, C. R. Chim. 2007, 10, 392–413; c) B. Meunier,
A. Robert, G. Pratviel, J. Bernadou in The Porphyrin Handbook, Vol. 4
(Eds.: K. M. Kadish, K. M. Smith, R. Guilard), Academic Press, London,
2000, pp. 119–187; d) B. Meunier, Chem. Rev. 1992, 92, 1411–1456;
e) J. T. Groves, T. E. Nemo, R. S. Myers, J. Am. Chem. Soc. 1979, 101,
1032–1033.
[12] M. A. Phillippi, H. M. Goff, J. Am. Chem. Soc. 1982, 104, 6026–6034.
[13] a) J. T. Groves, T. E. Nemo, J. Am. Chem. Soc. 1983, 105, 6243–6248;
b) M. J. Nappa, C. A. Tolman, Inorg. Chem. 1985, 24, 4711–4719; c) J. F.
Bartoli, O. Brigaud, P. Battioni, D. Mansuy, Chem. Commun. 1991, 440–
442.
[14] a) D. Dolphin, T. G. Traylor, L. Y. Xie, Acc. Chem. Res. 1997, 30, 251–259;
b) H. Fujii, J. Am. Chem. Soc. 1993, 115, 4641–4648.
[2] A. Nanthakumar, H. M. Goff, J. Am. Chem. Soc. 1990, 112, 4047–4049.
[3] J. T. Groves, J. A. Gilbert, Inorg. Chem. 1986, 25, 123–125.
[4] a) T. L. Poulos, B. C. Finzel, I. C. Gunsalus, G. C. Wagner, J. Kraut, J. Biol.
Chem. 1985, 260, 16122–16130; b) K. G. Ravichandran, S. S. Boddupalli,
C. A. Hasermann, J. A. Peterson, J. Deisenhofer, Science 1993, 261, 731–
736.
[5] J. E. Baldwin, T. Klose, M. Peters, Chem. Commun. 1976, 881–883.
[6] a) A. R. Battersby, W. Howson, A. D. Hamilton, Chem. Commun. 1982,
1266–1268; b) H.-A. Wagenknecht, W.-D. Woggon, Angew. Chem. 1997,
109, 404–407; Angew. Chem. Int. Ed. Engl. 1997, 36, 390–392; c) H. Volz,
M. Holzbecher, Angew. Chem. 1997, 109, 1510–1513; Angew. Chem. Int.
Ed. Engl. 1997, 36, 1442–1445; d) H.-A. Wagenknecht, C. Claude, W.-D.
Woggon, Helv. Chim. Acta 1998, 81, 1506–1520.
[15] M. das Dores Assis, J. R. L. Smith, J. Chem. Soc. Perkin Trans. 2 1998,
2221–2226.
[16] T. Higuchi, K. Shimada, N. Maruyama, M. Hirobe, J. Am. Chem. Soc. 1993,
115, 7551–7552.
[17] W. H. Koppenol, J. F. Liebman, J. Phys. Chem. 1984, 88, 99–101.
[18] H. Saltzman, J. G. Sharefkin, Org. Synth. 1963, 43, 60–61.
[19] K. Calderon-Kawasaki, S. Kularatne, Y. H. Li, B. C. Noll, W. R. Scheidt, D. H.
Burns, J. Org. Chem. 2007, 72, 9081–9087.
[20] J. M. Tour, A. M. Rawlett, M. Kozaki, Y. X. Yao, R. C. Jagessar, S. M. Dirk,
D. W. Price, M. A. Reed, C. W. Zhou, J. Chen, W. Y. Wang, I. Campbell,
Chem. Eur. J. 2001, 7, 5118–5134.
[21] A. Osuka, G. Noya, S. Taniguchi, T. Okada, Y. Nishimura, I. Yamazaki, N.
Mataga, Chem. Eur. J. 2000, 6, 33–46.
[7] a) J. H. Dawson, Science 1988, 240, 433–439; b) J. T. Groves, J. Inorg. Bio-
chem. 2006, 100, 434–447; c) J. Rittle, M. T. Green, Science 2010, 330,
933–937; d) B. Meunier, S. P. de Visser, S. Shaik, Chem. Rev. 2004, 104,
3947–3980.
[8] a) J. M. Mayer, Acc. Chem. Res. 1998, 31, 441–450; b) M. T. Green, Curr.
Opin. Chem. Biol. 2009, 13, 84–88.
[9] J. Manka, D. S. Lawrence, Tetrahedron Lett. 1989, 30, 6989–6992.
[10] The crystals for X-ray analysis were grown in a solvent mixture contain-
ing THF, dichloromethane, and hexane. Hence, the sixth coordination
site of iron was occupied by THF. The solid sample for catalysis was pre-
pared in solvent mixture of CH2Cl2 and hexane. It is very likely that the
iron centre is five-coordinated.
[22] a) CrystalClear-SM Expert, version 2.0 r7, Rigaku Europe, Sevenoaks, UK,
2010; b) CrysAlisPro, Oxford Diffraction Ltd., Abingdon, UK, 2010; c) Z.
Otwinowski, W. Minor in Methods in Enzymology, Vol. 276 (Eds.: C. W.
Carter Jr., R. M. Sweet), Academic Press, 1997, pp. 307–326; d) A. Alto-
mare, G. Cascarano, C. Giacovazzo, A. Guagliardi, J. Appl. Crystallogr.
1993, 26, 343–350; e) G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64,
112–122; f) A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Gia-
covazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna, J. Appl.
Crystallogr. 1999, 32, 115–119; g) P. van der Sluis, A. L. Spek, Acta Crys-
tallogr. Sect. A 1990, 46, 194–201.
[23] S. J. Coles, P. A. Gale, Chem. Sci. 2012, 3, 683–689.
[11] N. Suzuki, T. Higuchi, Y. Urano, K. Kikuchi, H. Uekusa, Y. Ohashi, T.
Uchida, T. Kitagawa, T. Nagano, J. Am. Chem. Soc. 1999, 121, 11571–
11572.
Received: August 7, 2012
Published online on && &&, 0000
ChemSusChem 0000, 00, 1 – 16
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&15
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
P. Li, K. Alenezi, S. K. Ibrahim, J. A. Wright,
D. L. Hughes, C. J. Pickett*
A preference for alcohol over oxygen:
We show that iron porphyrins can medi-
ate the anodic oxidation of hydrocar-
bons, thus providing a potential
&& – &&
complementary reaction to cathodic
generation of fuels or feedstocks as an
alternative to oxygen evolution.
Towards Alternatives to Anodic Water
Oxidation: Basket-Handle Thiolate FeIII
Porphyrins for Electrocatalytic
Hydrocarbon Oxidation
&16
&
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 16
ÝÝ
These are not the final page numbers!