A study of nerve agent model organophosphonate binding with manganese-A2B-corrole and -A2B2-porphyrin systems

Article abstract of DOI:10.1016/j.poly.2009.04.020

Herein the synthesis and binding studies of novel trans-A₂B-corrole and trans-A₂B₂-porphyrin derivatives are presented in comparing manganese(III)-organophosphonate (OP) binding (e.g., Mⁿ⁺ ← O{double bond, long}

Full text of DOI:10.1016/j.poly.2009.04.020

Polyhedron 28 (2009) 2418–2430
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
A study of nerve agent model organophosphonate binding
with manganese-A2B-corrole and -A2B2-porphyrin systems
Kibong Kim a,b, Inkoo Kim a, Nilkamal Maiti a,b,1, Seong Jung Kwon a, Daniela Bucella c,
Olga A. Egorova a,b, Yoon Sup Lee a, Juhyoun Kwak a, David G. Churchill a,b,
*
a Department of Chemistry and School of Molecular Science (BK 21), Korea Advanced Institute of Science and Technology (KAIST), 373-1 Guseong-dong, Yuseong-gu,
Daejeon 305-701, Republic of Korea
b Molecular Logic Gate Laboratory, Korea Advanced Institute of Science and Technology (KAIST), 373-1 Guseong-dong, Yuseong-gu, Daejeon 305-701, Republic of Korea
c Department of Chemistry, Columbia University, 3000 Broadway, New York, New York 10027, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Herein the synthesis and binding studies of novel trans-A2B-corrole and trans-A2B2-porphyrin derivatives
are presented in comparing manganese(III)-organophosphonate (OP) binding (e.g., Mn+ O@PR(OR)2)
capabilities. H3(PFP-VC) [PFP-VC = 5,15-di(pentafluorophenyl)-10-(3-vinylphenyl)corrolate] was synthe-
sized by way of literature procedures and was characterized by a variety of 2-D NMR spectroscopic
techniques and single-crystal X-ray diffraction. These compounds represent the first example of 3-
vinyl-phenyl-containing meso-substituted corroles or porphyrins. Mn(PFP-VC) (3) was treated separately
with (CH3CH2O)2P@O(C3H6NMe2), (C4H9O)2P@O(Me), (C2H5O)2P@O(CH2COCH3), (CH3CH2O)2P@O(Me), to
give 1:1 adducts, as determined by UV–Vis spectroscopy (Job Plot), giving a red shift; Ph3P@O, was also
found to bind, but very weakly. The trans-A2B2-porphyrin analogue Mn(PFP-VP) (4) was also prepared by
way of a literature procedure; related binding studies gave 1:1 organophosphonate-Mn(PFP-VP) adducts
(Job Plot). A clean blue shift occurred for the Mn-porphyrins at higher organophosphonate loadings (Ka
values: 6.7 (0.9)–11.9 (0.4) Mꢀ1). DFT geometry optimizations of O@P(OMe)2Me binding and formal
Mn–O or P–O cleavage products in the unsubstituted neutral Mn-corrolato and -porphyrinato systems
with a range of metal-based spin states revealed greatest stability in formal phosphoryl oxygen binding
(energies: 11–13 kcal/mol) for the Mn-corrole (singlet); the Mn-porphyrin (sextet) was also quite stable.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 15 February 2009
Accepted 16 April 2009
Available online 24 April 2009
Keywords:
A2B-corrole
trans-A2B2-porphyrin
Manganese
Organophosphonate
Nerve agent sensing
UV–Vis titrations
Cyclic voltammetry
Model nerve agent
Binding study
1. Introduction
through their aromaticity on one hand, and a corrin-like structural
cavity that bears a trivalent cavity instead of a monovalent core
Macrocyclic polypyrroles are motifs of great ubiquity, constitut-
ing a range of both synthetic metal complexes and natural systems
(i.e., hemoglobin, corrin and chlorophylls) (Fig. 1). Of note, the por-
phyrin sub-class is crucial in oxygen-atom transfer [1], and in ana-
lyte binding (e.g., O2), with the ability to host a variety of Lewis
bases in forming important adducts in biology. Corroles, members
of a currently greatly expanding class of contracted porphyrin, are
gaining greater visibility through the sub-field’s major contribu-
tors: Gross [2], Gryko [3,4], Kadish [5,6], and Paolesse [7]. Corroles
differ from porphyrins in two important ways: they possess one di-
rect pyrrole–pyrrole bond in place of one methine (C–H) linkage,
and they are trivalent instead of divalent. This difference allows
for metal-corrolato species to possess a likeness to Nature’s hemes
(Fig. 1) on the other hand. These features enable a stabilization
of transition metal ions in valence states [8] that are systematically
higher than those known in the respective porphyrins.
Through the study of transition metal coordination chemistry,
adequate control of complex color, electronics, sterics, and molec-
ular geometry can, in theory, be imparted, especially through the
use of a rigid delocalized ligand corrole or porphyrin frame. A delo-
calized and rigid frame may also be fluorogenic. Mn-corrolato spe-
cies, however, do not appear to be promising in this respect. Thus,
to determine whether a porphyrin or corrole framework, or both, is
of utility in the optical or otherwise sensing of organophosphonate
is the goal of these solution studies. Since our primary focus here is
on nerve agent model [O@P] oxygen metal ion binding, we also
undertook theoretical calculations to fully determine M-corrole
O@P(OR)2R binding energies using model systems. Understanding
the features and substitutions that enhance, or detract from, the
binding capabilities of metalloporphyrin derivatives are essential
in the rational development of these systems as detection plat-
forms in hazardous analyte sensing. Generally, metallocorroles
* Corresponding author.
E-mail address: dchurchill@kaist.ac.kr (D.G. Churchill).
Present address: Midnapore College, Midnapore, Paschim Medinipur, West
1
Bengal 721101, India. Tel.: +82 42 350 2845; fax: +82 42 350 2810.
0277-5387/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2009.04.020

K. Kim et al. / Polyhedron 28 (2009) 2418–2430
2419
Aromatic Macrocycles
Contracted Porphyrin Core
NH
~2.85 Å
N
N
N
HN
~2.50 Å
HN
N
HN
HN
N
NH
N
Porphyrin Horseradish Peroxidase Core
Corrole-type Core
Vitamin B12 Corrin Cofactor Core
Fig. 1. A comparison of the macrocyclic porphyrin, corrole and corrin (vitamin B12 cofactor) frames. Note: the left and middle structures are the basis of geometry
optimization calculations, illustrated below in Figs. 11 and 12.
and metalloporphyrins show some promise as selective analyte
sensing platforms. Corroles have been explored as CO sensors of
sufficient selectivity when testing with competitive analytes such
as O2 and N2. The nature of the metal (e.g., Co3+) as well as corrole
substitution play essential parts in the Lewis acidity of the metal
center [9], which is the sensing core. Metallocorroles also have
good reproducibility and stability over a term of months as ‘‘elec-
tronic noses.” In a related report, there was no loss of corrole
‘‘monomer” from a thin layer array over a four-week period of con-
tinual use [10]. Precedent for related metalloporphyrin-based sen-
sors comes from reports by Di Natale and coworkers [11–13].
Porphyrin derivatives have also been the subject of computational
efforts in the context of analyte selectivity [14–17].
The types of analytes we chose to study here are organophosph-
onates, relatives of the extremely deadly organophosphonate nerve
agents (Fig. 2). While there have been significant efforts to fully
decontaminate all stockpiles, undertaken in good faith to abide
by the time-table set forth by the Chemical Weapons Convention
(CWC), there are still significant amounts of organophosphonate-
based chemical weapons in North America as well as in Eurasia
that, to date await complete destruction and remediation. This
impetus, along with potential terrorist agent development and
use, support the need for continued investigations into improved
real-time sensing devices [17].
and coworkers in which Ph3P@O forms an adduct with Mn(III)tpf-
corrolato; this adduct can also be formed from the manganyl
[Mn@O] species upon addition of PPh3 [18]. The crystal structure
in which triphenylphosphine oxide is coordinated to tris-pfp-Mn-
corrole [19a,19b] can be compared to one in which there exists a
free Mn(III)corrolato center [19c].
Mn-corrole-based organophosphonate nerve agent sensing can
be explored in a variety of ways. Our approach involved preparing
closely related Mn-corrolates that bear stability imparted from the
pentafluorophenyl groups but that also contain a 10-position
meso-aryl group that contains an attachment point. We were able
to develop the corrole and the related porphyrin system for com-
parison. We also go further here theoretically, in determining
how oxygen binding of organophosphonate nerve agent mimics,
and how the phosphoryl oxygen can formally migrate from the
Mn@O group; this can be treated theoretically (vide infra). While
organophosphonates have different electronic and steric properties
as the simpler Ph3P@O molecule [19], we can look at a variety
of organophosphonate species: (CH3CH2O)2P@O(C3H6NMe2),
(C4H9O)2P@O(Me),
(C2H5O)2P@O(CH2COCH3),
(CH3CH2O)2P@
O(Me), Ph3P@O (Fig. 2), and other analytes with compounds 3 or
4 (Scheme 1) [20].
In terms of the preparative chemistry of corroles, extremely
practical pathways to A2B-corroles have been reported by Gryko
and coworkers [3,4]. Thus, we have prepared meso-thienyl-A2B-
corroles species with the well-known C6F5-substitution, and also
the 3-vinylphenyl substitution which may facilitate chemisorption
via sulfhydryl groups of functionalized surfaces [21]. We have also
prepared the analogous porphyrin system to compare analyte
binding in two closely related systems.
Cyclic polypyrrole ligand-based sensing platforms may be able
to efficiently and selectively mediate in techniques such as piezo-
electric crystal sorption detection which are the centerpiece of por-
table and growingly reliable nerve agent detection/discernment.
The possibility of metallocorroles serving in nerve agent detection
and decontamination is suggestive from results reported by Gross
N
O
O
O
N
N
O
F
O
O
F
O
R
O
Tabun (GA)
A
GF
Sarin (GB), R = Me
Soman (GD), R = Bu
D
t
R
O
O
R
N
R'
O
S
R
R
B : R = Me, C : R = CH2CH2CH3
VX, R = Me R' = Me
R-VX R = H, R' = Pr
i
E
Fig. 2. d = [O@P] group. (left) Common organophosphate nerve agents. G- and V-type chemical warefare (nerve) agents. (right) Phosphorus-containing molecules used in this
study.

2420
K. Kim et al. / Polyhedron 28 (2009) 2418–2430
Scheme 1. The synthesis of corroles and porphyrins and their products of metallation reported herein. Percent yields are included next to each molecule. The crystallographic
report of 1 appears below.
the addition of benzaldehyde derivative followed by oxidation,
usually effected by DDQ. We have previously reported 2- and 3-
thienyl-dipyrromethanes [24a] as well as meso-substituted A3-
corroles that were prepared analogously as reported previously
in Polyhedron [24b]. While the A2B2 porphyrin (Compound 2) can
sometimes be satisfactorily isolated as a by-product in the cor-
role-generating reaction above [25a], this species was prepared
using more optimal conditions in accordance with a previously re-
ported literature method [25b] to give a microcrystalline sample of
2. Results and discussion
2.1. Synthesis and characterization
Corrole preparation began with the generation of dipyrrome-
thane prepared by the literature method of Lindsey and coworkers
[22] involving the condensation of pyrrole and aldehyde. The syn-
thesis of A2B corroles were adapted from well-known literature re-
ports [23,4], which begins with the dipyrromethane and involves

K. Kim et al. / Polyhedron 28 (2009) 2418–2430
2421
Fig. 3. (top) The aromatic portion of the 1H–1H COSY NMR spectrum showing all C–H protons, and (bottom) 1H–13C HMQC NMR spectrum showing all carbon resonances, for
H3(PFP-VC).

2422
K. Kim et al. / Polyhedron 28 (2009) 2418–2430
the species formulated as the Mn(III)porphyrinato chloride (4)
[25c]. Thus, a solid of compound 2 was obtained through recrystal-
lization in 37% yield (See Section 4). Suitability of this as an oxy-
gen-atom transfer catalyst has been previously reported [20b,20d].
Characterization of the new compounds 1–4 reported herein
was made mainly by spectroscopic techniques: 1H NMR, 13C
NMR and UV–Vis spectroscopic studies (e.g., Fig. 3 and Supporting
Information) as well as CV measurements. We also relied on 2-D
NMR spectra, which are used conveniently because of the fully
unsaturated nature of Compounds 1 and 2 which have resonances
gathered in the aromatic region (5–9 ppm). The substitution (as in
Compound 2) as well as other corrole features give rise to clearly
assignable cross-peaks in 2-D NMR spectra such as the 1H–1H COSY
and 1H–13C HMQC NMR spectra of H3(PFP-VC) provided in Fig. 3.
For mass spectrometry, the presence of the expected species comes
from the inspection of isotopic clusters centered at calculated val-
ues for complexes 1–4 (Supporting Information). Additional char-
acterization is provided by single-crystal X-ray diffraction for
compound 1, presented below.
fact that a crystal structure of a closely related adduct is known.
It is interesting to note that other, possibly better binding, phos-
phorus oxide species were not studied as of yet in this context.
A determination of ligand-to-metal complex stoichiometry in
this reaction mixture was performed by the method of continuous
variation. Titrimetric trials for A–D (Fig. 2) were performed, as
shown e.g., in Fig. 4, in which a portion of Mn(PFP-VC) in dichloro-
methane was titrated by (CH3CH2O)2P@O(C3H6NMe2) (A). Clear
growth of new bands appear at 473 and 634 nm. The
(CH3CH2O)2P@O(C3H6NMe2) species is interesting in that there is
a pendant amino group in virtually the same place as that seen
for VX (Fig. 2), relative to the central phosphorus [26,27]. The next
titration involved (C2H5O)2P@O(CH2COCH3) which afforded new
bands in the electronic absorption spectra at 483 and 631 nm.
The (CH3CH2O)2P@O(Me) species gave bands at 481 and 635 nm.
Finally, (C4H9O)2P@O(Me) gave bands at 481 and 631 nm (Support-
ing Information). For these four species (Figs. 4–6, and S14) the Job
plots involving the method of continuous variation were prepared;
in all these cases (Insets of Figs. 4–6, and S14), the maximum peak
at a mole fraction of ꢁ0.5 is indicative that the solution stoichiom-
etry is the expected 1:1 ligand-to-metal binding ratio. This is in
accordance with the notion that the organophosphonate species
will bind mainly through the phosphoryl [O@P] oxygen atom.
The possibility of its binding through other heteroatoms, albeit
more weakly, will not be considered herein.
There may be some concern that the Mn corrolato system va-
lence state may change during, or prior to, analyte binding in light
of (i) Mn4+ being an air stable metallocorrolate oxidation state and
(ii) the involvement of stabilizing axial ligation [25e]. Importantly,
to support our formulated Mn valence states, EPR spectroscopy for
the Ph3P@O adduct of the Mn(tpf)corrolato species reported by
Gross et al., reveals a Mn(III) (S = 2) metal center [19a]. This char-
acterization excludes the possibility for the Mn(II) ‘‘radical cation”
species [25d] and also for a Mn(IV) center. Compound 3 differs
from the Mn(tpf)corrolato species only by the 10-position aryl
group in which a 3-vinyl phenyl group replaces a pentafluorophe-
nyl group. This small substitution will unlikely change the chemis-
try dramatically. Further experimental support that we do not
2.2. UV–Vis spectroscopic detection of organophosphonate via
titrations in CH2Cl2
While fluorescence characteristics of corroles can be intense
and unique, incorporation of Mn3+ renders the complexes non-
fluorescent. Thus, herein we rely on UV–Vis absorption spectro-
scopic changes. Electronic absorption spectra of Mn(PFP-VC) (3)
and Mn(PFP-VP) (4) with different organophosphonate species
bearing various binding atoms will be presented in this section.
Detailed titrimetric spectra are provided in Figs. 4–6 for Mn cor-
roles and Figs. 7 and 8 for Mn-porphyrins. These trials all show a
diminution in the Soret band at 411 nm when Compound 3 is used.
A new band at 473–483 nm grows in as indicated by a label in each
figure. Triphenyl phosphine was also tested as an analyte; it also
has the capacity to bind through its (lone) oxygen atom, and, along
with the sterics and rigidity of the phenyl groups, does so less effi-
ciently, as shown by comparatively smaller changes in the spec-
trum (Supporting Information). This is surprising in light of the
Fig. 4. Electronic absorption spectra of Mn(PFP-VC) [4.69 ꢂ 10ꢀ6 M] treated in dichloromethane with (CH3CH2O)2P@O(C3H6NMe2) [5.06 ꢂ 10ꢀ3 M] (10–50 (by 10), 100–650
(by 50) lL). Inset: Job Plot for a (binary) mixture of 3 with (CH3CH2O)2P@O(C3H6NMe2) showing a 1:1 stoichiometric correspondence.

K. Kim et al. / Polyhedron 28 (2009) 2418–2430
2423
Fig. 5. Electronic absorption spectra of Mn(PFP-VC) [4.69 ꢂ 10ꢀ6 M] in the presence of various amounts (50–1100 (by 50)
dichloromethane. Inset: Job plot for a binary mixture of 3 with (C2H5O)2P@O(CH2COCH3).
lL) of (C2H5O)2P@O(CH2COCH3) [1.51 M] in
Fig. 6. Electronic absorption spectra of Mn(PFP-VC) [4.69 ꢂ 10ꢀ6 M] in the presence of various amounts (50–500 (by 50)
dichloromethane. Inset: Job Plot for a binary mixture of 3 with (CH3CH2O)2P@O(Me).
lL) of (CH3CH2O)2P@O(Me) [2.74 M] in
access reduced Mn-corrolato forms comes from the fact that elec-
tro-reduced forms of Mn(III) are unstable towards solvents, i.e.,
methylene chloride [25e], especially in the absence of a clear
reductant. Additional support for Mn(III)corrolato ones comes from
an analogous corrolazine compound reported by Goldberg and
coworkers in which a shoulder at 722 nm in the UV–Vis spectrum
is assigned to the Mn(III)O@PPh3 adduct [19b]. In terms of Mn-cor-
role coordination, the Mn(III) species may exist in solution with
one more axial MeCN. To further delve into the particulars of the
valence state as well as a consideration of substrate/anion binding,
we can turn to electrochemical studies discussed below that intro-
duces a detailed report by Kadish and coworkers [25f] who have
addressed how to best assign the valence state of the metal ion un-
der various conditions [8].
We also studied the related binding for the porphyrin compound
Mn(PFP-VP) with (C4H9O)2P@O(Me), (C2H5O)2P@O(CH2COCH3),
and (CH3CH2O)2P@O(Me). These analytes were used in UV–Vis
titrimetric trials whose results are plotted in Figs. 7 and 8 (and
Supporting Information). All four assays give rise to clear changes
compared to those of the Mn-corrole cases in that the Soret band
undergoes moderate diminution and blue shifting only to give a
new band in the range of 450–460 nm. While there is a clean spec-
tral change, a much greater amount of analyte was required to ef-
fect this optical change than was needed in the corrole case, thus,

2424
K. Kim et al. / Polyhedron 28 (2009) 2418–2430
Fig. 7. Electronic absorption spectra of Mn(PFP-VP) [5.00 ꢂ 10ꢀ6 M] in dichloromethane upon titration with various amounts of (C2H5O)2P@O(CH2COCH3) [1.50 M] (20–200
(by 20) lL). Inset: Job Plot for a binary mixture of 4 with (C2H5O)2P@O(CH2COCH3).
Fig. 8. Electronic absorption spectra of Mn(PFP-VP) [5.00 ꢂ 10ꢀ6 M] in dichloromethane in the presence of various amounts of (C2H5O)2P@O(Me) [1.50 M] (20–200 (by
20) lL). Inset: Job Plot for a binary mixture of 4 with (C2H5O)2P@O(Me).
these porphyrin-based systems are obviously less sensitive and
less impressive as a potential sensing platform in the context of
organophosphonates. This is in line with the tendency for a more
in-plane metal ion in which a larger deviation from the [N4] plane
may encourage stronger analyte binding. The binding constants
(Ka) values were measured by way of an equation reported by Val-
eur et al. [32] and are provided below as insets (Figs. 7 and 8). The
that for (C2H5O)2P@O(Me) is 9.9 (0.8) Mꢀ1
(C4H9O)2P@O(Me) is 11.9 (0.4) Mꢀ1
; and, that for
.
As in the corrole case, we believe that this porphyrin system
maintains an unchanged valence state upon binding. We should
note, however, that a similarity in blue shifting pattern occurs for
Mn@O formation. The oxo-Mn-(meso-tetrakis-2,3,5,6,-tetrafluoro-
N,N,N-trimethyl-4-aniliniumyl)porphyrinato species (427 nm)
gives a band at for the Mn(III)-porphyrinato at ꢁ450 nm, upon
binding constant for (C2H5O)2P@O(CH2COCH3) is 6.7 (0.9) Mꢀ1
;

K. Kim et al. / Polyhedron 28 (2009) 2418–2430
2425
studied in considerable detail and has a variety of reduction and
oxidation potentials depending on the solvent and axial anion
coordination which are similar when in non aqueous media [25f].
Data was also obtained for the porphyrin-based species 2 and 4
(Scheme 1). For the free-base porphyrin, the first reduction occurs
at ꢀ1.45, and ꢀ1.93 V, and the oxidization values occur at 0.10 and
0.83 V. For the Mn(III)-porphyrin chloride species, the first small
E1/2 reduction occurs at ꢀ0.50 with another occurring out at
ꢀ1.78 V (Inset, Fig. 9); a broad E1/2 of oxidation exists at 1.45 V.
2.4. Structural characterization of compound 1
We also provide herein the crystal structure of 5,15-bis(penta-
fluorophenyl)-10-(3-vinylphenyl)corrole (1) (Fig. 10). The molecu-
lar structure of 1 confirms the presence of the vinyl moiety found
also naturally in hemes; it points down and is suggestive of a clear
surface anchoring group [21]. This serves as the first structural
example of molecule with a 3-vinyl-phenyl dipyrrin/dipyrrome-
thane fragment. There exists a plausible hydrogen bonding net-
work between pairs of molecules as well (Supporting
Information). Crystallographic packing structures provided for this
structural solution separately along all three axes have been pre-
pared (Supporting Information).
2.5. DFT calculations on bare corrole and porphyrin model systems
Fig. 9. Cyclic voltammetry data for compounds 1–4, [3.0 mM], respectively, at room
temperature in benzonitrile. Refer to Section for further details. Half-wave
4
potentials and peak potentials for 1–4 in benzonitrile. The inset is from
measurement of compound 4 at 1.0 mM.
a
To try to understand the differences between neutral [28b] Mn–
corrole and Mn–porphyrin organophosphonate binding and re-
lated transformations, we undertook DFT calculations using an
unsubstituted corrole or porphyrin, oxygen atom, and the requiste
phosphonate/-ite fragments. While calculations have previously
been undertaken in trying to understand porphyrin M@O species,
the details of M–O–P interactions are particularly informative in
the present context. Electronic energies and geometries for formal
M–O and P–O bond cleavage products for various metal-based spin
states are illustrated in Figs. 11 and 12. The spin states in the cor-
role case were the singlet, triplet and quintet. Going from the
bench-mark ‘‘Compound I”-like metal oxo species [1,29] the differ-
ences in energies in the middle section and those on the left repre-
sent the energy of formal O–M cleavage, whereupon the O-atom
formally migrates from the metal to the phosphorus atom of the
phosponite (POMe2Me). Going from the middle set to the right side
set (in Fig. 11) formally signifies the attachment of the phosphine
atom to the manganyl O-atom. This atom then becomes engaged
as the donor atom in a Lewis base adduct with the metal center.
The presence of the O@P bond, as well as a favorable dative inter-
action of ꢁ10 kcal/mol gives this species (for the singlet) the low-
est energy state. How the O-atom rearranges from an X2(L) donor
to an L-type donor is a very interesting point for later consider-
ation. A similar profile exists for the neutral Mn-porphyrinato
and its derivatives (Fig. 12). Herein, the doublet, quartet, and sextet
spin states were calculated. A much higher energy profile for the
carbamazepine epoxidation (10,11-position oxidation) [25c]. Thus
further determinations for our porphyrin species will need to be
made through future work.
2.3. Cyclic voltammetry (CV)
Electrochemistry often enables for the useful assignment of me-
tal and ligand oxidations and affords a look at an analysis of the
state of the Mn+ of the metal complex, especially when considering
transition metal ions such as manganese. A variety of recent re-
ports of Mn-corrolates are available [25f]. The cyclic voltammo-
grams have been obtained for compounds 1–4 and are provided
below in Fig. 9. The cyclic voltammograms which include half-
wave potentials/peak potentials (versus Ag/Ag+ or Fc/Fc+ or SCE)
were obtained in PhCN. As revealed by the CV curve in Fig. 9, the
reduction potentials for 3 are ꢀ1.55 and ꢀ2.29 V (versus SCE). This
value is very similar to the reduction potentials of ꢀ1.51 and
ꢀ2.07 V (versus SCE) for the metal free corrole 1 and has been as-
signed as corrole-centered reductions. The oxidation potential for 3
(0.24 V versus SCE) is assigned to the metal center, and is less than
that for the Mn(tpf)corrole (0.71 V versus SCE), indicating 3 is more
easily oxidized than the Mn(tpf)corrole [28]. A comparison to Mn
(Mes2Ph)corrole can also be made (Table 1). This species has been
Table 1
A comparison of reduction and oxidation potentials of 3 and related compounds Mn(Mes2Ph)-corrole and Mn(tpf)corrole in various solvents.
Solvent
3rd ox. (V)
2nd ox. (V)
1st ox. (V)
1st red. (V)
Mn(Mes2Ph-Corrole) [25f]
Mn(Mes2Ph-Corrole) [25f]
Mn(Mes2Ph-Corrole) [25f]
Mn(tpf)Corrole [28]
CH2Cl2
PhCN
Pyridine
CH2Cl2
PhCN
–
1.00
1.07
–
–
–
0.32 (0.03)b
0.45 (0.12)b
0.42
0.71
0.24
–
1.83a
ꢀ1.36, ꢀ1.64
–
–
–
ꢀ1.46
–
Compound 3 [this article]
ꢀ1.55, ꢀ2.29
All values referenced to the SCE except for 3 which was reported versus the ferrocene/ferrocenium ion redox couple.; SCE = (Fc/Fc+) + 0.40 V; (Fc/Fc+) = (Ag/Ag+) ꢀ 0.092 V; E1/
TBAP = tetrabutylammonium perchlorate.
2
a
Scan rate 0.1 V/s.
b
Extra peaks are due to the varation of axial ligation. A dash indicates an unreported or unobtainable value.

2426
K. Kim et al. / Polyhedron 28 (2009) 2418–2430
Fig. 10. Top and side views of 1 (CCDC 684999). Hydrogens are retained for clarity. X-ray diffraction study parameters: chemical formula C39H18F10N4, fw 732.14, Space group
c l
= 105.165(17)°, V = 1766(4) Å3, Z = 4, = 0.121 (mmꢀ1), R1 = 0.2094, wR2 = 0.4016
ꢀ
P1, a = 10.077(13) Å, b = 13.359(18) Å, c = 14.907(19) Å,
a
= 113.113 (17)°, b = 90.83(3)°,
q
{R1 = (
R
||Fo| ꢀ |Fc||)/
R
|Fo|; wR2 = [(
R
(F2o ꢀ F2c )2)/
R
w(F2o)2]1/2}, calcd = 1.404 (mg/m3), Temp = 243(2).
Fig. 11. Relative energies and geometries for three neutral manganese(III) corrolato species.
terminal oxo-derivative is evident, when compared to the corrolato
counterpart. The oxo-intermediate is lowest for the doublet spin
state; but the sextet actually ‘‘edges out” the doublet as the lowest
energy species as the phosphonate adduct (right side set in Fig. 12).
Clearly, the P@O will never oxidize the manganese in either the
corrole or porphyrin case; and this obviously drives phosphine oxi-
dation catalysis; how phosphonites might oxidize or de-oxidize in
the context of nerve agent sensing and remediation is ‘‘food for
thought.” These calculations help in approaching or understanding
the opportunities, and limitations of catalytic decontamination of
organophosphonate-based chemical warefare agents. Structurally,
the Mn–O–P angle in the Mn(pfp)corrolato is 158.6° and the metal
exists 0.29 Å above the [N4] mean plane [19a], the calculated
Mn–O–P angles and manganese atom ‘‘out-of-plane” distances in
these model systems depend on the spin state and are provided
in Table 2 comparison for both the corrole and porphyrin adducts.
3. Conclusion
In this report we have described the preparation, characteriza-
tion, and dynamics of new A2B-corrole, trans-A2B2-porphyrin, and
their manganese complexes. To our knowledge, this is the first

K. Kim et al. / Polyhedron 28 (2009) 2418–2430
2427
Fig. 12. Relative energies and geometries for three neutral manganese(II) porphyrinato species.
Table 2
Geometrical features in the calculated structures of Mn-porphyrinato and -corrolato species.
Mn-corrole multiplicity
Distance of Mn and [N4] mean plane
Mn-corrole multiplicity
Mn–O–P angle
Singlet
Triplet
Quintet
0.54 Å
0.47 Å
0.36 Å
singlet
triplet
quintet
141.1°
136.3°
140.6°
Mn-porphyrin multiplicity
Mn-porphyrin multiplicity
Doublet
Quartet
Sextet
0.29 Å
0.38 Å
0.11 Å
doublet
quartet
sextet
139.8°
141.1°
139.8°
determination of OP sensing capabilities of metallocorroles and
metalloporphyrins. Also this is the first report of a meso 3-vinyl-
phenyl substituted polypyrrole species. The synthesis for the
metallocorroles and metalloporphyrin species is adapted from
the literature. We undertook detailed NMR spectral characteriza-
tion and obtained the structure of the free-base corrole (1) through
the use of single-crystal X-ray diffraction. We probed the possibil-
ity of selective sensing of nerve agent model organophosphonate
species. The bonding of organophosphonate to the Mnn+ center is
supported by the strong responses in the UV–Vis spectrum upon
analyte titration. This technique gave quantification of 1:1 binding
for the Mn(PFP-VC)corrole (3) (PFP-VC = 5,15-di(pentafluorophe-
nyl)-10-(3-vinylphenyl)corrolate) with (CH3CH2O)2P@O(C3H6-
NMe2), (C4H9O)2P@O(Me), (C2H5O)2P@O(Me), and (CH3COCH2)P@
O(OC2H5)2. The titrimetric data allows for the calculated Ka values
in the case of the [Mn(PFP-VP)][Cl] which ranged from 6.7 to
11.9 Mꢀ1. These studies confirm 1:1 binding through the use of
Job plots for compound 4 as well. Cyclic voltammetry of com-
pounds was acquired for 1-4; the values of the Mn(PFP-VC)corrole
can be compared to those for Mn(Mes2Ph)corrolato that appear in
a recent report from Kadish and coworkers [25f]. Modeling of
O@P(OMe)2Me binding and formal M–O or P–O cleavage products
in bare neutral Mn-corrole and -porphyrins with different metal-
based spin states was undertaken. These efforts revealed the effect
of (MeO)2P@O(Me) phosphoryl O-atom binding using three differ-
ent metal-based spin states. Regarding the minimum bare corrole/
porphyrin organophosphonate binding energies, greater stability
occurred for the Mn-corrole (singlet) and the Mn-porphyrin (sex-
tet) than for other spin states studied. From this data, we can sup-
port that the corrole is
a
more sensitive platform for
organophosphonate sensing and has a possibility to optically de-
tect organophosphonate on at least the micromolar level because
of clear changes that occur in the UV–Vis spectrum. This sensitivity
is probably inadequate for current nerve agent applications as a
stand-alone method, however. But it is hoped that this report will
enable various further studies, especially in realm of nanoscience,
and with respect to piezoelectric studies in which the response in-
volves acoustic signals of mass differences, not optical signals. Also,
these compounds can be studied in the presence of a variety of
decomposition products (CNꢀ, Fꢀ, phosphonic acids) that could
be present under real ‘‘battlefield” conditions. Future studies,
may include catalysis with these species [20b]; also an expansion

2428
K. Kim et al. / Polyhedron 28 (2009) 2418–2430
of this same analyte binding can be made to the related manga-
nese-corrolazines.
slowly. After 1 h, the reaction mixture was filtered through a silica
gel pad followed by purification via silica gel column chromatogra-
phy (CH2Cl2: hexane = 1: 1) which gave a purple solution. This
solution was then evaporated to give a purple solid (247 mg,
37.2%). 1H NMR (CDCl3): d 8.97 (d, JH–H = 4.8 Hz, H3,9,13,19), 8.80
(d, JH–H = 4.8 Hz, H4,8,14,18), 8.26 (s, 2Hf), 8.11 (d, JH–H = 7.5 Hz,
2Hj), 7.86 (d, JH–H = 7.9 Hz, 2Hh), 7.73 (t, JH–H = 7.6, 2Hi), 6.99 (dd,
JH–H = 17.6, 10.9 Hz, 2Hk), 5.96 (d, JH–H = 17.5, 2Hl-trans), 5.42 (d,
JH–H = 11.0, 2Hl-cis), ꢀ2.82 (s, 2HN–H). MALDI-TOF: m/z [M+H+]
847.184 (calc); 847.666 (obs). mN–H (KBr): 3321 cmꢀ1. Anal. Calc.
for C48H24N4F10: C, 68.09; H, 2.86; N, 6.62. Found: C, 66.99; H,
2.91; N, 6.46%.
4. Experimental
4.1. General considerations
All reagents and solvents herein were of analytical grade and
used as received from reliable commercial sources. The reagents
3-vinylbenzaldehyde, DMF, TFA, DDQ, F3BꢃOEt2, Mn(OAc)2ꢃ4H2O
were purchased from Aldrich, USA; pentafluorophenylcarboxalde-
hyde was acquired from TCI, Japan; toluene, dichloromethane, and
chloroform were acquired from Merck [30]. Dipyrromethane was
prepared by literature report [22]. The synthetic procedures for
A2B corroles and their metallocorroles were adapted from the liter-
ature [23]. All solvents used for NMR spectral analysis were pur-
chased commercially and were of spectroscopic grade. 1H NMR
spectra were measured in CDCl3 or CD2Cl2 with Bruker Avance
300 and 400 MHz spectrometers. Spectral signals were calibrated
by the protio impurity of the deuterated solvent. Infrared spectra
were recorded with a Bruker Equinox 55 spectrophotometer;
UV–Vis spectra were obtained with a JASCO V-503 UV–Vis spec-
trometer. A Vario EL III CHNS elemental analyzer was used in
obtaining microanalysis values. High resolution MALDI mass spec-
trometry was performed by the staff of the research supporting
team at KAIST on a VG AUTOSPEC ULTIMA with a trisector double
focusing magnetic sector analyzer at a resolution of 80 000.
4.4. Preparation of 5,15-bis(pentafluorophenyl)-10-(3-vinylphenyl)
corrolate manganese(III) [Mn(PFP-VC)] (3)
A small round-bottom flask was charged with 15 mL of dry
DMF, H3(PFP-VC) (0.05 g, 0.068 mmol), and Mn(OAc)2ꢃ4H2O
(0.05 g, 0.20 mmol) and was heated to reflux for 20 minutes. The
complete consumption of H3(PFP-VC) was confirmed through the
use of TLC assaying. The reaction mixture was then allowed to cool
to room temperature after which time the solvent was removed
under reduced pressure. The resulting residue was purified by sil-
ica gel column chromatography (1:1, hexane:ethyl acetate) to give
pure Mn(PFP-VC). The yield was 50.0 mg (94%). MALDI-TOF: m/z
(M+): 784.052 (calc); 784.215 (obs). Anal. Calc. for C39H15N4F10Mn:
C, 59.71; H, 1.93; N, 7.14. Found: C, 59.66; H, 3.04; N, 6.08%.
kmax (log e
/(Mꢀ1 cmꢀ1)): 601 (3.56); 411 (4.60) [soret band]; 356
(4.47).
4.2. Preparation of 5,15-bis(pentafluorophenyl)-10-(3-vinylphenyl)
corrole [H3(PFP-VC)], (1)
4.5. Preparation of 5,15-bis(pentafluorophenyl)-10,20-bis(3-
vinylphenyl)porphyrinato manganese(III) [Mn(PFP-VP)] chloride (4)
A
mixture of 5-(pentafluorophenyl)dipyrromethane (5.13 g,
16.4 mmol) and 3-vinylbenzaldehyde (1.04 mL, 8.2 mmol) and a
dichloromethane solution of TFA (80 L), were dissolved in
H2(PFP-VP) (85 mg, 0.1 mmol) and manganese acetate tetrahy-
drate (101 mg, 0.41 mmol) were dissolved in DMF (10 mL). The
reaction mixture was refluxed for 1 h, and was then allowed to cool
to room temperature. A TLC assay revealed no trace of free-base
porphyrin. The reaction mixture was then placed into a NaCl aque-
ous solution containing a portion of ice, which allowed a green pre-
cipitate to form. This solid was isolated by filtration and dried
affording a (green) solid formulated as the mono-chloride 4
(85.1 mg, 91.1%). MALDI-TOF: m/z (M–Cl+) 899.107 (calc),
l
800 mL dichloromethane. The reaction mixture was stirred for
12 h. DDQ (4.47 g, 19.6 mmol) in toluene was added dropwise over
several minutes accompanied by vigorous stirring, which contin-
ued for 15 additional minutes. The volume of the reaction mixture
was then reduced and the mixture was filtered through a pad of sil-
ica (silica gel, CH2Cl2:hexane = 1:1), followed by column chroma-
tography (silica gel, CH2Cl2:hexane 1:4) which ultimately gave a
dark green solid. This compound was further purified by prepara-
tive TLC to give pure corrole, H3(PFP-VC) (613 mg, 10.2%). MAL-
DI-TOF: m/z [M+H]+: 733.137 (calc); 733.791 (obs). 1H NMR(d,
CD2Cl2): 9.15 (d, JH–H = 4.17 Hz, 2H3,18), 8.76 (m, 4H8,9,12,13), 8.62
(d, JH–H = 3.98 Hz, 2H4,17), 8.26 (s, 1Hf), 8.09 (d, JH–H = 7.22 Hz,
1Hj), 7.83 (d, JH–H = 7.86 Hz, 1Hh), 7.74 (t, JH–H = 7.58 Hz, 1Hi),
7.00 (dd, JH–H = 17.60, 10.94, 1Hk), 5.97 (d, JH–H = 17.56, 1Hl-trans),
5.40 (d, JC–H = 10.98, 1Hl-cis). 13C-NMR: 147.87, 145.42, 143.49,
899.427 (obs). kmax (log e
/(Mꢀ1 cmꢀ1)): 609 (3.68); 576 (3.88);
475 (4.90) [Soret band]; 396 (4.52); 369 (4.67).
4.6. X-ray structural determination of compound 1
A single crystal of 1 was mounted on a goniometer and data
were collected on a Bruker P4 diffractometer equipped with a
SMART CCD detector using graphite-monochromated Mo Ka radi-
1
141.96, 141.52, 140.96, 140.37, 139.66, 137.19 (t, JC–H = 82.6 Hz,
Ck), 135.88, 134.54 (d, JC–H = 159.9 Hz, Cj), 132.93 (d, JC–H
1
1
ation (k = 0.71073 Å). The structure was solved using direct meth-
ods and standard difference map techniques, and was refined by
full-matrix least-squares procedures on F2 with SHELXTL (version
5.10). Cell parameters were determined and refined by the SMART
program [31a]. Data reduction was performed by using SAINT soft-
ware [31a]. The data were corrected for Lorentz and polarization
effects. An empirical absorption correction was applied using the
SADABS program [31b]. The structures were solved by direct meth-
ods using the program SHELXS-97 [31c] and refined by full matrix
least-squares calculations (F2) by using the SHELXL-97 software
[31d]. All non-H atoms were refined anisotropically against F2 for
all reflections. All hydrogen atoms were placed at their calculated
positions and refined isotropically. Crystal data, data collection
and refinement parameters are summarized in the caption of
Fig. 10.
=
1
164.3 Hz, Cf), 132.01, 128.11 (d, JC–H = 173.2 Hz, C8,13), 127.91 (d,
1JC–H = 159.5, Ci), 125.97 (d, JC–H = 147.6 Hz, Ch), 125.81 (d, JC–
H = 172.0 Hz, C9,12), 122.02 (d, 1J = 170.1 Hz, C4,17), 118.06 (dd,
1J = 173.1 Hz, 3.5 Hz, C3,18), 114.93 (dd, 1J = 159.7 Hz, 6 Hz, Cl),
1
1
114.47, 113.543. mN–H (KBr): 3421 cmꢀ1
.
4.3. Preparation of 5,15-bis(pentafluorophenyl)-10,20-bis(3-vinylphe-
nyl)porphyrin [H2(PFP-VP)] (2)
5-(Pentafluorophenyl)dipyrromethane (0.50 g, 1.6 mmol) and
3-(vinylphenyl)benzaldehyde (0.20 mL, 1.6 mmol) were added to
a portion of chloroform (160 mL); BF3ꢃOEt2 (65
lL) was then added
dropwise to this reaction mixture. The reaction mixture was stirred
for 1 h at RT; DDQ (0.27 g, 1.2 mmol) in toluene was then added

K. Kim et al. / Polyhedron 28 (2009) 2418–2430
2429
4.7. Modeling of analyte binding
www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
UK; fax: (+44) 1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.poly.2009.04.020.
For non-linear fitting, we have used Eq. (1) [32,33] to determine
1:1 stoichiometry in our Mn-corrole and Mn- porphyrin systems.
A ¼ A0
2
4
3
5
"
#
1=2
ꢀ
ꢁ
þ
c0 þ cM
þ
ꢀ
c0 þ cM
þ
2 ꢀ 4c0cM
References
Alim ꢀ A0
2c0
1
Ks
1
Ks
[1] B. Meunier, Heme-peroxidases, in: J.A. McCleverty, T.J. Meyer (Eds.),
Comprehensive Chemistry II, vol. 8, 2004, p. 261.
[2] I. Aviv, Z. Gross, Chem. Commun. (2007) 1987.
[3] D.T. Gryko, B. Koszarna, Org. Biomol. Chem. 1 (2003) 350.
[4] D.T. Gryko, Chem. Commun. (2000) 2243.
[5] E. VanCaemelbecke, S. Will, M. Autret, V.A. Adamian, J. Lex, J.P. Gisselbrecht, M.
Gross, E. Vogel, K.M. Kadish, Inorg. Chem. 35 (1996) 184.
ð1Þ
The parameters are as follows: A = absorbance, C0 = concentra-
tion of corrole or porphyrin; Ks = binding constant, CM = concentra-
tion of organophosphonate analyte, A0 = absorbance of corrole or
porphyrin; Alim = absorbance when analyte is bound completely.
[6] M. Autret, S. Will, E. Vancaemelbecke, J. Lex, J.P. Gisselbrecht, M. Gross, E.
Vogel, K.M. Kadish, J. Am. Chem. Soc. 116 (1994) 9141.
[7] (a) S. Nardis, D. Monti, R. Paolesse, Mini-Rev. Org. Chem. 2 (2005) 355;
(b) S. Nardis, F. Mandoj, R. Paolesse, F.R. Fronczek, K.M. Smith, L. Prodi, M.
Montalti, G. Battistini, Eur. J. Inorg. Chem. (2007) 2345.
4.8. Cyclic voltammetry (CV)
[8] G. Parkin, J. Chem. Educ. 83 (5) (2006) 791.
[9] J.-M. Barbe, G. Canard, S. Brandes, F. Jerome, G. Dubois, R. Guilard, Dalton Trans.
(2004) 1208.
Measurements were carried out with a BAS 100B instrument
(Bioanalytical Systems, Inc). A three-electrode system was used
consisting of a gold disk working electrode, a platinum wire coun-
ter electrode, and an Ag/Ag+ electrode (MF-2062 kit, Bioanalytical
Systems, Inc.) as the reference electrode. The Ag/Ag+ contained
0.1 M tetrabutylammonium perchlorate (TBAP) and 0.01 M AgNO3
in benzonitrile. In 0.1 M TBAP, the E1/2 of the ferrocene/ferroce-
nium ion couple was taken to be 0.067 V versus Ag/Ag+ in benzoni-
trile. The values in electrochemical experiments are reported
versus the ferrocene/ferrocenium ion [29] redox couple.
[10] R. Czolk, Sens. Actuators B B30 (1996) 61.
[11] R. Paolesse, C. Di Natale, A. Macagnano, F. Sagene, M.A. Scarselli, P. Chiaradia,
V.I. Troitsky, T.S. Berzina, A. D’Amino, Langmuir 15 (1999) 1268–1274.
[12] C. Di Natale, C. Goletti, R. Paolesse, M. Drago, A. Macagnano, A. Mantini, V.I.
Troitsky, T.S. Berzina, M. Cocco, A. D’Amico, Sens. Actuators B B57 (1999) 183.
[13] A. D’Amico, C. Di Natale, A. Macagnano, F. Davide, A. Mantini, E. Tarizzo, R.
Paolesse, T. Boschi, Biosens. Bioelectron. 13 (1998) 711.
[14] (a) T.A. Romanova, O.V. Kravchenko, I.I. Morgulis, A.A. Kuzubov, P.O. Krasnov,
P.V. Avramov, Koord. Khim. 30 (2004) 403;
(b) S. Han, K. Cho, J. Ihm, Phys. Rev. E: Stat. Phys., Plasmas, Fluids 19 (1999)
2218.
[15] R. Zwaans, J.H. van Lenthe, D.H.W. den Boer, Theochem 339 (1995) 153.
[16] J.W. Grate, S.N. Kaganove, Strongly Hydrogen-Bond Acidic Polymer and
Methods of Making and Using, US Patent 6015,869, January 2000.
[17] D.A. Atwood, D.G. Churchill, ‘‘Destruction and detection of chemical warfare
agents”, Chem. Rev., cr-2008-00223w, in preparation.
[18] Z. Gross, G. Golubkov, L. Simkhovich, Angew. Chem., Int. Ed. 39 (2000) 4045.
[19] (a) J. Bendix, H.B. Gray, G. Golubkov, Z. Gross, Chem. Commun. (2000) 1957;
(b) The report by D.P. Goldberg on corrolazines is particulary relevant to the
discussion herein:B.S. Mandimutsira, B. Ramdhanie, R.C. Todd, H. Wang, A.A.
Zareba, R.S. Czernuszewicz, D.P. Goldberg, J. Am. Chem. Soc. 124 (2002) 15170;
(c) S. Licoccia, E. Morgante, R. Paolesse, F. Polizio, M.O. Senge, E. Tondello, T.
Boschi, Inorg. Chem. 36 (1997) 1564.
4.9. Computational details
The GAUSSIAN 03 program was used for all calculational work [34].
All geometries were processed in silico in the gas phase at 0 K. The
following protocol was used for all calculations: (i) input geome-
tries were derived from the crystallographic structures provided
herein where possible, or by careful modification of closely related
geometries and invoking chemical intuition. (ii) Density functional
theory (DFT) geometry optimizations were performed using B3LYP
hybrid functional with the LACVP* (double zeta) basis set. The ba-
sis set used the Lanl2DZ effective core potential (ECP) for the man-
ganese atom, and 6-31G(d) for all other atoms.
[20] (a) Preliminary presentations of portions of this work include two poster
presenations: K. Kim, N. Maiti, D.G. Churchill, Abstracts of Papers, 235th ACS
National Meeting, New Orleans, LA, United States, April 6–10, 2008, INOR-
192.;
(b) K. Kim, N. Maiti, D.G. Churchill, Abstracts of Papers, 234th ACS National
Meeting, Boston, MA, United States, August 19–23, 2007, INOR-660.;
(c) D.G. Churchill, N. Maiti, S.H. Choi, Metallocorrole catalyst produced by
substituting 3-vinyl phenyl group from a corrole compound and a method for
producing the same, Repub. Korean Kongkae Taeho Kongbo (2007), CODEN:
KRXXA7 KR 2007049466 A 20070511 CAN 147:475473 AN 2007:1191159
(Note, the first two authors were erroneously listed as David, G.C. and
Nilkamal, M.).;
Acknowledgements
D.G.C. acknowledges support from the Korea Science and Engi-
neering Foundation (KOSEF) Grant (No. R01-2008-000-12388-0),
funded by the Korean Government (MOST). Professor Y.S.L. and
I.K. (Quantum Chemistry Laboratory) gratefully acknowledge (i)
financial support from the Korea Science and Engineering Founda-
tion (Nos. R01-2007-000-11015-0, R11-2007-012-03001-0) and
(ii) computational time from the supercomputing center of KISTI.
Professor Gerard Parkin (Department of Chemistry, Columbia Uni-
versity) facilitated the X-ray diffraction study of Compound 1.
MALDI-TOF analyses were obtained with the help of the KAIST re-
search supporting team. Hack Soo Shin is gratefully acknowledged
for his help in acquiring NMR data. Dr. Raj Chakrabarti (former NIH
Postdoctoral fellow at the Departments of Chemistry at M.I.T. &
Columbia University) is acknowledged for helpful discussions dur-
ing his visit to KAIST (July, 2005). Teresa Lindstead (Cambridge,
England) is thanked for her assistance in filing our crystallographic
data (CCDC 684999).
(d) Oxygen-atom transfer catalysis has previously been explored with
Compound 3: N. Maiti, K. Kim, B. Meka, D.G. Churchill, unpublished results,
2005–2007.
[21] D.L. Pilloud, C.C. Moser, K.S. Reddy, P.L. Dutton, Langmuir 14 (1998) 4809.
[22] B.J. Littler, M.A. Miller, C.H. Hung, R.W. Wagner, D.F. O’Shea, P.D. Boyle, J.S.
Lindsey, J. Org. Chem. 64 (1999) 1391.
[23] D.T. Gryko, B. Koszarna, Synthesis (2004) 2205.
[24] (a) N. Maiti, J. Lee, Y. Do, H.S. Shin, D.G. Churchill, J. Chem. Crystallogr. 35
(2005) 949;
(b) N. Maiti, J. Lee, S.J. Kwon, J. Kwak, Y. Do, D.G. Churchill, Polyhedron 25
(2006) 1519.
[25] [a] The isolation of the porphyrin from the corrole reaction involves
preparative TLC. This however would favor migration of Mn(II) which is not
ionic; as the Mn(II) migrates on silica in air, it may decompose and become the
oxo (manganyl) species (ligand oxidation may also occur). The Mn(II) species
however, is less stable than the [Mn(III)Cl] porphyrin species which can be
more completely synthesized as chloride salt as discussed herein and reported
previously.;
(b) R.D. Jones, D.A. Summerville, F. Basolo, J. Am. Chem. Soc. 100 (1978) 4416;
(c) While the formulation of the porphyrin is Mn(III), it is well known that
upon appropriate oxidant, the preparation can be made for Mn(IV) (e.g., using
tBuOOH) and Mn(V) (e.g., using H2O2):W. Nam, I. Kim, M.H. Lim, H.J. Choi, J.S.
Lee, H.G. Jang, Chem. Eur. J. 8 (2002) 2067;
Appendix A. Supplementary data
(d) D. Turner, M.J. Gunter, Inorg. Chem. 33 (1994) 1406;
(e) Z. Ou, C. Erben, M. Autret, S. Will, D. Rosen, J. Lex, E. Vogel, K.M. Kadish, J.
Porphyrins Phthalocyanines 9 (2005) 398;
CCDC 684999 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge via

2430
K. Kim et al. / Polyhedron 28 (2009) 2418–2430
(f) J. Shen, M.E. Ojaimi, M. Chkounda, C.P. Gros, J.-M. Barbe, J. Shao, R. Guilard,
(c) H.-Y. Chao, D.G. Churchill, J. Chem. Educ. 85 (2008) 1210;
(d) O.A. Egorova, O.G. Tsay, S. Khatua, J.O. Huh, D.G. Churchill, Inorg. Chem, in
press.
K.M. Kadish, Inorg. Chem. 47 (2008) 7717. See references herein for previous
electrochemical studies of a variety of Mn corroles, e.g., octaethyl substituted
corrole-based systems.
[30] With regards to laboratory safety, a useful description of some common
laboratory chemical explosion hazards has been prepared: D.G. Churchill, J.
Chem. Educ. 83 (2006) 1798.
[26] I. Bandyopadhyay, M.J. Kim, Y.S. Lee, D.G. Churchill, J. Phys. Chem. A 110 (2006)
3655.
[27] It is thought that a carefully positioned carboxylic acid or related group could
effect secondary N-binding of this analyte in the context of selective VX
binding.
[28] (a) G. Golubkov, J. Bendix, H.B. Gray, A. Mahammed, I. Goldberg, A.J. DiBilio, Z.
Gross, Angew. Chem., Int. Ed. 40 (2001) 2132;
[31] [a] SMART and SAINT, Area Detector Software Package and SAX Area Detector
Integration Program; Bruker Analytical X-ray, Madison, WI, 1997.;
[b] SADABS, Area Detector Absorption Correction Program; Bruker Analytical X-
ray: Madison, WI, 1997.;
[c] G.M. Sheldrick, SHELXS-97 program for crystal structure determination, Acta
Crystallogr. A46 (1990) 467–473.;
[d] G. Sheldrick, SHELXL-97, Universität Göttingen, Göttingen, Germany, 1999.
[32] B. Valeur, J. Pouget, J. Bourson, M. Kaschke, N.P. Ernsting, J. Phys. Chem. 96
(1992) 6545.
[b] An Mn(III) porphyrinato would require the use of an anion for
electroneutrality; this was not pursued in these calculations. The point of
this work was to seek an understanding of the diversity of the various spin
states, which was done through the analysis of neutral systems.
A
Mn(III)porphyrinato species would be more electron deficient at the metal
center (on average) and this would be expected to strengthen the
organophosphonate dative interaction.
[33] ORIGIN 6.1, Windows-Based Scientific Graphing and Data Analysis Software,
OriginLab Co. Northampton, MA, USA.
[34] GAUSSIAN 03: M.J. Frisch et al., Gaussian Inc., Wallingford CT, 2004 (see
Supplementary Supporting Information for complete citation).
[29] (a) M.K. Tse, Z. Zhang, K.S. Chan, Chem. Commun. (1998) 1199;
(b) A.E. Meier-Callahan, H.B. Gray, Z. Gross, Inorg. Chem. 39 (2000) 3605;