Strong and directed association of porphyrins and iron(terpyridine)s using hydrogen bonding and ion pairing

Article abstract of DOI:10.1016/S0040-4020(01)01096-1

The combination of cooperative hydrogen bonding and ion pairing between cationic iron(II)terpyridines and anionic porphyrins yielded remarkably stable neutral complexes even in the highly competitive solvent DMSO. Isothermal titration calorimetry (ITC) was used to compare association constants, enthalpies and entropies of binding between various combinations of the two molecular components that make up the complexes. Steady-state luminescence studies highlighted that, as expected, the fluorescence quenching of the porphyrin is maximized in the cases where the iron(terpyridine) is strapped the tightest across the macrocycle.

Full text of DOI:10.1016/S0040-4020(01)01096-1

TETRAHEDRON
Pergamon
Tetrahedron 58 )2002) 639±651
Strong and directed association of porphyrins and
ironꢀterpyridine)s using hydrogen bonding and ion pairing
Tyler B. Norsten, Kelly Chichak and Neil R. Brandap
Department of Chemistry, University of Alberta, Edmonton, Alta., Canada T6G 2G2
Received 8 June 2001; revised 29 August 2001; accepted 1 September 2001
AbstractÐThe combination of cooperative hydrogen bonding and ion pairing between cationic iron)II)terpyridines and anionic porphyrins
yielded remarkably stable neutral complexes even in the highly competitive solvent DMSO. Isothermal titration calorimetry )ITC) was used
to compare association constants, enthalpies and entropies of binding between various combinations of the two molecular components that
make up the complexes. Steady-state luminescence studies highlighted that, as expected, the ¯uorescence quenching of the porphyrin is
maximized in the cases where the iron)terpyridine) is strapped the tightest across the macrocycle. q 2002 Elsevier Science Ltd. All rights
reserved.
1. Introduction
tionality). Systems that fall into this category are often
plagued by their functioning only in non-competitive
solvents, or by the generation of assemblies with ill-de®ned
topology and composition. The obvious answer to these
problems is to design heterodiatopic or heteropolytopic
building blocks where several different interactions are
used in concert to guide the molecular recognition process
and afford strong and selective complexation. The hydrogen
bond is a particularly prized interaction used to unite build-
ing blocks due to its directionality and the fact that hydrogen
bond donors and acceptors can be placed onto a building
block with a high degree of precision. The utility of the
hydrogen bond is greatly reduced, however, in polar com-
petitive solvents that can better solvate the hydrogen
bonding surface. The implementation of cooperative, ionic
hydrogen bonding motifs, where the hydrogen bond
partners are of opposite charge, can often overcome these
destructive solvation effects.27±31
The ef®cient collection of solar energy is the primary task of
natural light harvesting systems such as those found in
plants, photosynthetic algae and cyanobacteria. These
complicated photo-sensitive assemblies are responsible for
initiating a cascade of energy and electron transfer
reactions, ultimately affording usable forms chemical fuel.
The key to the successful operation of these assemblies lies
in their high degree of structural organization which ensures
optimal alignment and relative proximity of the molecular
components to maximize through-space communication of
the chromophores. This re®ned architectural organization
and structural integrity is a direct outcome of a multitude
of contiguous non-covalent interactions.
With this in mind, signi®cant efforts have been devoted to
design and construct arti®cial multi-component photoactive
arrays, particularly those involving porphyrins.1±26 Because
the photochemical behavior of the supramolecular
assemblies will be dictated by their topology, design
strategies must include judicial use of molecular recognition
principles and, therefore, an emphasis on the logical
programming of intermolecular attractions such as coordi-
native bonding,2±12 ion pairing,12±14 p±p stacking15 and
hydrogen bonding16±26 is paramount. Supramolecular
systems that utilize only one of these interactions, however,
do not take full advantage of the properties that each has to
offer )expressed in their relative binding strength and direc-
The formation of highly robust supramolecular assemblies
is especially critical in order to investigate energy and
electron transfer processes by luminescence spectroscopy.
In the absence of strong complexation, excessive quantities
of the quenching component must be added to produce the
observable spectroscopic response at the low concentrations
demanded by this analytical technique.
We have recently reported the self-assembly of a novel non-
covalently united porphyrin±iron)II)terpyridine complex 1
and how the porphyrin's ¯uorescence is signi®cantly
quenched by the iron)terpyridine).32 In the present article,
we expand on the synthesis, structural characterization and
luminescence properties of assembly 1 and its 1:2 counter-
part 2. We designed the building blocks in order to harness
the bene®cial recognition attributes of both ion pairing and
Keywords: hydrogen bond)s); luminescence quenching; molecular recogni-
tion, porphyrins; self-assembly.
Corresponding author. Address: Department of Chemistry, Simon Fraser
University, 8888 University Drive, Burnaby, BC, Canada V5A1S6. Tel.:
11-604-291-3594; fax: 11-604-291-3765; e-mail: nbranda@sfu.ca
p
0040±4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020)01)01096-1

640
T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
Figure 1. Schematic representation of complexes 1 and 2. Observed intermolecular nOes )represented as arrows) and complexation-induced chemical shifts in
the 1H NMR spectra )shown in parentheses) are highlighted. Double bonds and carbon-bound hydrogen atoms have been removed for clarity.
Ê
hydrogen bonding. The ®rst interaction is provided by the
attraction between the cationic iron)terpyridine) coordina-
tion compounds and the anionic porphyrinic carboxylates.
The carboxylate groups on the porphyrins also act as
charged hydrogen-bond-acceptor sites for the neutral
bidentate urea hydrogen-bond-donors ¯anking the
iron)terpyridine) component. The result is the self-assembly
of the neutral complexes 1 and 2, which retain both their
structural integrity and topology even at low concentrations
in the highly competitive solvent DMSO )Fig. 1).
)4±5 A) in 13, as compared to 7a and, consequently,
separates the two active components in space. The meta-
substitution pattern between the terperidine and urea
moieties is maintained across the xanthene backbone so as
2. Results and discussion
2.1. Synthesis of the molecular components
Terpyridine 533 was prepared from N-)2-)pyridyl)-2-
oxoethyl)pyridinium iodide )3) and )E)-3-)3000-nitro-
phenyl)-1-)pyrid-20-yl)prop-2-enone )4)33 as shown in
Scheme 1. Reduction of this product with SnCl2, followed
by subsequent dissymmertic urea formation employing 3,5-
di-tert-butylaniline as the amine capping agent and tri-
phosgene as the carbonyl linking agent conveniently yielded
urea 6a. Reacting urea 6a with iodomethane in anhydrous
DMF and in the presence of base cleanly afforded the N,N0-
dimethylated analog 6b, which will make its appearance as
a control later in this report. The ®nal iron)II) coordination
compounds 7a and 7b were isolated as purple solids by
treating 2 molar equivalents of 6a and 6b with 1 equiv. of
Fe)H2O)6)BF4)2.
We also prepared the xanthene analog 13 in order to
evaluate the effects )both structural and photophysical) of
the relative proximity between the iron)terpyridine) and
porphyrin components. The replacement of the 40-phenyl
ring in terpyridine 6a by the extended xanthene skeleton
in 12 provides a substantially deeper binding pocket
2
Scheme 1. Counterions for 7a and 7b are BF4
.

T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
641
2
Scheme 2. Counterions for 13 are BF4
.
not to alter the hydrogen bonding vectors involved in the
association between the two chromophores. The key
intermediate is the dissymetric xanthene 9, which was
conveniently prepared by a two-step nitration-alkylkation
procedure. The tert-butyl group's role was not to enhance
solubility in organic solvents as is more typical of this
group. )All studies will be performed using DMSO
solutions, after all.) It was incorporated to block the more
reactive para-site on the xanthene skeleton and direct the
subsequent formylation reaction )9!10) to produce the
desired substitution pattern. The aldehyde was transformed
in two steps into terpyridine 11 by reacting it with an excess
of 2-acetylpyridine and then with ammonium acetate. The
conversion of 11 to 13 was accomplished in an identical
fashion as already described for terpyridines 7a and 7b
)Scheme 2).
by acidi®cation to generate their corresponding carboxylic
acids )17). The tertrabutylammonium carboxylate
porphyrins 18a and 18b were generated by treating 17a
and 17b with 2 equiv. of a methanolic solution of tetra-
butylammonium hydroxide. The corresponding derivatives
of 5,10,15,20-meso-tetra-substituted porphyrins 20a and
20b were generated from the known acids )19a37 and
19b38) employing 4 equiv. of a methanolic solution of tetra-
butylammonium hydroxide )Scheme 4).
Hydrogen-bonded complexes 1 and 2 )Fig. 1) were prepared
by simply mixing methanol solutions of porphyrin 18a or
20a with iron)terpyridine) 7a in a 1:1 molar ratio for
complex 1 and a 1:2 molar ratio for complex 2. Complexes
1 and 2 precipitated from the reaction mixture and were
easily isolated in high purity and in nearly quantitative
yields. Because the association of the terpyridine and
porphyrin components is accompanied by charge-balancing
and the construction of neutral complexes, 2 and 4 equiv. of
tetrabutylammonium tetra¯uoroborate were also produced
from the self-assembly reaction. These biproducts can be
washed away during ®ltration.
All of the 5,15-meso-di-substituted porphyrins were synthe-
sized by the acid-catalyzed cross-condensation34 of the
appropriately substituted benzaldehyde )14a35 or 14b) and
dipyrrolemethane 1536 in CH2Cl2 )Scheme 3). The
porphyrin esters )16) were saponi®ed using KOH, followed

642
T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
Figure 2. Optimized structure of complex 1 as predicted by molecular
modeling.
2.2. 1H NMR characterization of the neutral complexes 1
and 2
Complexes 1 and 2 are reasonably soluble in highly polar
solvents such as DMSO and DMF, but show poor solubility
in other common organic solvents. The integration of the
signals in the 1H NMR spectrum in DMSO-d6 corresponding
to the two components that make up complex 1 yielded a
clean 1:1 stoichiometry attesting to the molecular composi-
³
tion of the complex. Upon association with the porphyrin,
the signals for the protons on the terminal pyridine rings of
7a shift signi®cantly up®eld while those for the N±H
protons of the urea shift in the opposite direction )Fig. 3).
Scheme 3.
Scheme 4.
²
Molecular modeling highlights the ideal lock-and-key ®t
between the two components making up complex 1 as well
as the close proximity between them )Fig. 2). The
importance of this spatial orientation and relative
arrangement will become apparent throughout this report.
1
Figure 3. 300 MHz H NMR spectra of DMSO-d6 solutions of: )a) iron-
)terpyridine) 7a, )b) 7a1porphyrin 18a, )c) terpyridine 13, and
)d) 131porphyrin 18a. CIS for selected hydrogen atoms are indicated by
connecting lines.
²
³
The starting geometry of the iron)terpyridine) component was obtained
from the X-ray crystal structure of the 4-ethoxyphenyl analog of 7a )see
Supporting Information).
The stoichiometry of complexes 1 and 2 were con®rmed using electro-
spray mass spectrometry, employing continuous infusion injection )see
Section 4).

T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
643
1
The H NMR complex-induced shift )CIS) values for the
slightly up®eld. We attribute the sharpness of the signals to
the fact that there is now adequate distance between the two
chromophores to permit free rotation of the terpyridine rotor
over the porphyrin plane. The reduced up®eld shifting is
easily ascribed to the fact that the pyridyl protons are
lying further from the shielding region of the macrocycle.
urea and selected terminal terpyridyl protons are shown in
Fig. 1. The signi®cant down®eld shifts )Dd greater than
3 ppm in DMSO-d6) observed for the signals of the urea
N±H protons are clearly a result of effective hydrogen
bonding, even in such a polar and competitive solvent.
The role of these hydrogen bonds is two-fold, one being
to associate the two components. The second role is to
steer the iron)terperidine) fragment into a position where
it lies directly over the porphyrin plane. This guidance is
successful as diagnosed by the up®eld shifts observed for
the hydrogen atoms lying directly over the porphyrin plane
and, thus, within the strong shielding region of the macro-
cyclic ring. GCOSY and TROESY experiments aided in
assigning the protons within complex 1. Intermolecular
nuclear Overhauser enhancements )nOes) )highlighted by
the arrows in Fig. 1) support the claim that 7a is strapped
across the porphyrin ring and not straddling its side.
The association of iron)terpyridine) 7a with the tetra-
substituted porphyrin 20a resulted in the 1:2 complex 2.
Fig. 1 shows how, in this case, the two iron)terpyridine)
units can alternately strap across opposing faces of one
1
porphyrin macrocycle. The CIS values in the H NMR
spectrum for the urea N±H and terpyridine protons are
nearly identical to those of complex 1. The only major
difference is that the signals in the H NMR spectrum of 2
1
integrate for a 1:2 stoichiometry. The addition of 2 equiv. of
7a to a methanol solution of the para-substituted porphyri-
nic analog 20b resulted in a precipitate that was insoluble
even in DMSO. Repeating the association experiment in
DMSO-d6 directly in an NMR tube resulted in an initial
spectrum where the signals for the urea N±H protons shift
1.5 ppm down®eld without any accompanying up®eld shift
of the signals for the terpyridyl protons. Aprecipitate
formed in the NMR tube after several hours of standing.
We suggest that the result of the association of 20b and
7a is a 1:2 complex where the two iron)terpyridine) frag-
ments are strapped to the side of porphyrin 20b )Fig. 4).
This stoichiometry is supported by the calorimetric studies
described later in this report.
The signals for the hydrogen atoms on the terminal pyridine
ring of the terpyridine ligand are unique in that they also
appear as broad peaks in the spectrum )Fig. 3b). We attri-
bute this to the fact that the strapped iron)terpyridine)
fragment does not have the luxury to rotate freely above
the porphyrin ring. The axis de®ned by the N±Fe±N
Ê
bonds in 7a is judged to be approximately 8.5 A above the
plane de®ned by the macrocycle which is not large enough
to allow the terpyridine to pass over the porphyrin
unencumbered. The result is that the terpyridine protons
Ê
can range in distance from approximately 3.5 to 14.5 A
from the plane of the porphyrin at any given moment afford-
ing a variety of possible rotomers that can coexist within the
NMR time-scale. Variable temperature NMR experiments
failed to alter the shape of these broadened signals.
When the para-substituted porphyrin 18b was exposed to
iron)terpyridine) 7a, a complex was isolated in a similar
manner as already described for complex 1. In this case,
the hydrogen bonds are not suitably positioned to guide
the formation of a strapped 1:1 complex, although the two
components can certainly generate a polymeric array with
1
the same 1:1 stoichiometry. The H NMR spectrum of this
complex indeed indicated a 1:1 stoichiometry, however, the
signals for the N±H protons of the urea only shift 1 ppm
down®eld. Also, there was no observable shifting of the
terpyridine signals. These observations indicate that 7a
does not reside over the plane of porphyrin 18b and instead
suggests the existence of a polymeric aggregate 7an´18bm.
The relative positioning of the two chromophores and
inability of 18b to form cooperative hydrogen bonds with
7a has a signi®cant impact on the association between the
two chromophores and the nature of the ¯uorescence
quenching as will be discussed later.
Figure 4. Proposed structure for the 2:1 complex formed when 7a assem-
bles with 20b.
2.3. Stability of the complexes
The xanthene-modi®ed iron)terpyridine) 13 provided us a
suitable model compound to evaluate the rotation of the
coordination compound over the plane of porphyrin 18a.
The strength of the binding between iron)terpyridine) 7a
and porphyrins 18a and 20a was too large to be accurately
measured by 1H NMR spectroscopy even in DMSO-d6. The
lack of any new or shifting absorption bands in the UV±Vis
region of the spectrum during the formation of 1 or 2 also
precluded this technique for the estimation of the associa-
tion constants )Ka). Isothermal titration calorimetry )ITC),
1
The H NMR spectrum of the resulting complex )Fig. 3d)
shows a similar down®eld shifting of the signals for the urea
N±Hs of 13 upon complexation to 18a as was already
observed for 7a. However, in this case, the signals corre-
sponding to the hydrogen atoms on the terminal pyridine
rings of 13 appear as sharp well-de®ned peaks that shift only

644
T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
between 18a and 7a§. Detailed calorimetric data for all
titrations are collected in Table 1.
We argue that the reduced association between the two
components in the corresponding xanthene complex
13´18a is not due to any changes in hydrogen bonding,
but instead, solely a result of changes in ion-paring and
solvation. We feel comfortable with this claim because
the hydrogen bonding sites in 7a and 13 are virtually
identical. One reason for the lower value of Ka for 13´18a
may stem from a diminished attraction of the oppositely-
charged ion pairs, as the two components are now forced to
reside further away from each other. This should, however,
afford a less negative enthalpic term for the binding of 13 by
18a which was not observed )entries 1 and 3 in Table 1). An
examination of the values of DS of association clearly shows
that the difference in Ka is entropically governed. The
process of desolvation during a self-assembly reaction
possesses an unfavorable enthalpic term )energy is required
to break the bonds between the components and solvent
molecules) and a favorable entropic term )there is an
increase in disorder as solvent is released from the compo-
nents). These arguments may apply to complexes 1 and
13´18a, although the role of desolvation in such complicated
systems is dif®cult to predict. We propose that, in the case of
13´18a, because the two components are removed from each
other in space, there are remaining solvent-accessible
surfaces on both porphyrin faces and all around the iron-
)terpyridine) fragment. On the other hand, in 1 the two
Figure 5. Heat change as iron)terpyridine)s 7a,b and 13 are added to
porphyrins 18a,b and 20a,b in DMSO: )a) 7a into 18a )W), 7b into 18a
)X), 7a into 18b )A), 13 into 18a )S), and 7a into DMSO )V); )b) 7a into
20a )W), 7b into 20a )X), 7a into 20b )A) and 7a into DMSO )V). The solid
lines represent the calculated isotherms.
Table 1. Binding stoichiometries, association constants )Ka), association enthalpy )DH) and association entropy )DS) for the binding of porphyrins 18a, 18b,
20a, and 20b with iron)terpyridine)s 7a,c and 13 in DMSO at 258C
Entry
Components
Stoichiometrya
Ka )M21
)
DH )kcal mol21
)
DS )cal mol21 deg21
)
1
2
3
4
5
18ab17a
18bb17a
18ab113
20ac17a
20bc17a
1.1^0.1
1.0^0.1
1.0^0.1
2.0^0.1
2.1^0.1
2.5£106^4.4£105
3.0£104^6.1£103
1.1£106^3.0£105
2.5£106^3.2£105
4.3£105^2.2£104
26.3^0.1
24.8^0.0
28.4^0.2
25.9^0.1
25.4^0.1
8.0^0.8
4.6^0.4
0.3^0.2
9.5^0.4
7.6^0.3
The data represent average values from duplicate ITC experiments performed using fresh solutions. Atypical experiments consisted of 60±70, 4 mL injections
of a 1.45 mm solution of the terpyridine component into the calorimetry cell containing a 0.05 or 0.1 mM solution of the porphyrin component. Aone-site
binding model was employed for all data sets.
a
Values represent equivalents of the terpyridine component per equivalent of the porphyrin component.
[Porphyrin]ˆ0.1 mM.
b
c
[Porphyrin]ˆ0.05 mM.
which measures the heat absorbed or evolved during a host±
guest binding event, was employed to determine the thermo-
dynamic parameters of the association of these high-af®nity
species. In addition to determining association constants,
ITC conveniently provides the stoichiometry of binding as
well as the enthalpy and entropy of association in a single
experiment.
components are in direct contact and, in essence, solvate
each other )at least on one face).
When the hydrogen bonds are not suitably positioned to
form a ®nite 1:1 complex, as is the case in the para-
substituted porphyrin 18b, the binding isotherm loses its
sigmoidality and substantially ¯attens out. This clearly indi-
cates that the lack of cooperative hydrogen bonding and the
separation of the ion pairs in space results in a substantially
weaker association between the molecular components
)entry 2 in Table 1).
The integration of the heat generated as aliquots of iron-
)terpyridine) 7a were added to a solution of porphyrin 18a
are plotted against the molar ratio generating the binding
isotherm shown in Fig. 5a. This curve ®t to a one-site
binding model. The steep upward sigmoidal nature of
the resulting curve in¯ecting at a molar ratio of 1
§
It should be noted that the ITC data for the titrations of 18a with 7a do not
rule out the possibility of higher-ordered equimolar aggregates )2:2, 3:3,
etc.) which will all produce identical binding isotherms, although the 1H
NMR data clearly support the strapped 1:1 complex.
indicates that
a
strong exothermic association
)Kaˆ2.5£106^4.4£105 M21) and 1:1 stoichiometry exists

T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
645
When hydrogen bonding is absent and association is purely
ionic in nature, as is the case between porphyrin 18a and the
N,N0-dimethylated iron)terpyridine) 7b, an endothermic
isotherm was produced. This can be expected for ion pair-
ing, where association is entropically driven with DH close
to zero or positive. Attempts to ®t the ITC data to a one-site
binding model were unsuccessful and resulted in an
unde®ned binding stoichiometry and an uncharacterizable
and weak association. This behavior is likely a result of non-
speci®c aggregation, typical of unde®ned ion pairing inter-
actions. These data are consistent with previous ITC studies
on non-speci®c solution based aggregation processes.39
The binding isotherm for porphyrin 20a and 7a were ®t to a
two-site binding model to yield the expected 1:2 stoichio-
metry and two similar association constants indicative of
non-cooperativity )Fig. 5b and entry 4 in Table 1). It is
not surprising that the two hydrogen-bond chelation sites
in 20a act independently as the binding of the iron)ter-
pyridine) component on one face of the macrocycle is not
expected to have any impact on the other.
The results of ITC experiment using the tetra-para-
substituted porphyrin 20b and 7a also indicate the formation
of a complex with a 1:2 stoichiometry )entry 5 in Table 1).
Because the iron)terpyridine) fragment cannot strap across
the face of the porphyrin macrocycle, as has already been
proposed )see Fig. 4), the association between 20b and 7a is
somewhat reduced as compared to the corresponding
strapped complex 2. This is surely due to a combination
of a looser ®t between 20b and 7a )adjacent carboxylate
groups on the porphyrin do not completely span the receptor
cleft provided by the 7a), as well as, the fact that the hydro-
gen bonding vectors are slightly less than ideal )908 between
adjacent carboxylates for 20b, as opposed to the more ideal
608 angles for 20a). The 7a2´20b structure proposed in Fig. 4
is supported by the increase in the value of Ka for this
complex as compared to that for 7a´18b )entries 2 and 5
in Table 1) which proves some cooperative binding is taking
place.
Figure 6. Inverse Stern±Volmer plots for ¯uorescence quenching titrations
of )a) 18a with 7a )W), 18a with 7b )X), 18b with 7a )A) and 18a with 13
)S); )lexˆ415 nm, lemˆ630 nm), and )b) 20a with 7a )W), 20a with 7b
)X), and 20b with 7a )A); )lexˆ430 nm, lemˆ650 nm). All titrations
were preformed in DMSO. [porphyrin]ˆ1.0£1026 M, [iron)terpyridine)]ˆ
2.0£1025 M.
7a are shown in Fig. 6.¶ The immediate quenching of the
¯uorescence of 18a is clearly a consequence of both the
strong association and the intimacy of the two chromo-
phores which ensures maximum through-space communica-
tion. The large extent of ¯uorescence quenching by 7a as
compared to its N,N0-dimethylated counterpart 7b, which
only slightly quenched the ¯uorescence of 18a in a
collision-dependant process, clearly indicates that an intra-
complex quenching mechanism is in effect within 1.
Although still signi®cant, the slightly reduced ¯uorescence
quenching observed as xanthene 13 was added to porphyrin
18a is most likely due to a combination of the weaker asso-
ciation between the two components and an increase in the
distance between the two chromophores. Despite the fact
that both hydrogen bonding and ion pairing are present in
the 7an´18bm polymolecular assembly, the iron)terpyridine)
The data from the titration of 20a with N,N0-dimethylated
iron)terpyridine) 7c could not be ®t to the binding models
again implying that an uncharacterizable, weakly-
associated aggregate was produced from the self-assembly
reaction as we have already argued for the association of
18a with the same iron)terpyridine) component.
Two things are evident from our calorimetric studies: )1) ion
pairing signi®cantly contributes to the association between
these anionic porphyrins and cationic iron)terpyridines),
however, )2) complementary hydrogen bonding is
ultimately responsible for the stoichiometic association
and large exothermic response.
fragment cannot form
a cooperative hydrogen-bond
strapped arrangement, resulting in reduced association
between the molecular components and a signi®cant lower-
ing of the through-space communication between the chro-
mophores. This is re¯ected in the titration of 18b with 7a.
2.4. Fluorescence quenching behavior
The proximity and relative positioning of the two compo-
nents within complexes 1 and 2 have a signi®cant impact on
the photophysical behavior of the ®nal assemblies. Studies
using steady-state ¯uorescence spectroscopy to monitor the
changes in the emission intensities of DMSO solutions of
18a and 20a as the porphyrins were treated with aliquots of
¶
The raw ¯uorescence data for the titration of 18a with 7a ®t to 1:1 binding
model and gave an association constant )Kaˆ5.24£106 M21) that was
close to that obtained for the calorimetry studies. The titration data was
analyzed using Christopher A. Hunter's 1:1 complexation model program
)Krebs Institute for Biomolecular Science, Department of Chemistry,
University of Shef®eld, UK).

646
T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
Adramatic decrease in ¯uoresence was also observed when
7a was titrated into a DMSO solution of porphyrin 20a.
When the para-substituted porphyrin 20b was titrated
with 7a, the extent of quenching response was signi®cantly
reduced, and a much shallower quenching curve was
produced. We attribute this to the fact that the iron)ter-
pyridine) component can only bind onto the side of the
porphyrin and not across its face. Consequently, the distance
between the two chromophores in the proposed complex
)Fig. 4) is greater than that in 2.
from Aldrich with the exception of aldehyde 14b which was
purchased from Acros.
1H NMR characterizations were performed on a Varian
Inova-500 instrument, working at 499.92 MHz, on a Varian
Inova-400 instrument, working at 399.96 MHz, or on a
Varian Inova-300 instrument, working at 299.96 MHz.
Chemical shifts )d) are reported in parts per million relative
to tetramethylsilane using the residual solvent peak as a
reference standard. FT-IR measurements were performed
using a Nicolet Magna-IR 750 spectrometer. UV±Vis
absorption measurements were performed using a Varian
Cary 400 Scan spectrophotometer. Low resolution mass
spectrometry measurements were preformed using an
Agilent Technologies 1100 MSD with an electrospray
source. High resolution measurements were preformed
using a Kratos MS-50 with an electron impact source or a
PerSeptive Biosystems )Mariner) MS with an electrospray
source. Solutions of complex 1, 2, and 13´18a in 10% DMF/
MeOH were prepared by dissolving the isolated complexes
in a minimal amount of DMF and then diluting with
methanol. These solutions were infused into the electro-
spray source )¯ow rateÐ30 mL/min, nebulizer pressureÐ
12 psi). All other electrospray data were obtained by
standard injection procedures.
When a suf®ciently good electron acceptor is proximal,
porphyrins are known to undergo rapid photoinduced elec-
tron transfer.40 The result is a reduction in the observed
¯uorescence of the porphyrin because the photoexcited
electron recombines with the porphyrin in a non-emitting
pathway. Alternatively, the ¯uorescence of a porphyrin can
be quenched in an energy transfer process.41 This requires
adequate overlap between the emission band of the photo-
excited molecule and absorption band of the energy
acceptor molecule. Because there is minimal overlap
between the porphyrin emission band and the metal-to-
ligand charge transfer )MLCT) absorption band of the
iron)terpyridine) in the present systems )see Supporting
Information), we believe that an energy transfer mechanism
is not favorable, although, at this point, it cannot be
completely ruled out.
4.2. Synthesis of ironꢀterpyridine)s
4.2.1. Urea 6a. Triphosgene )93 mg, 0.31 mmol) was added
to a solution of 40-)3000-aminophenyl)-2,20:6,200-terpyridine33
)302 mg, 0.93 mmol) and triethylamine )212 mg, 2.1 mmol)
in freshly distilled CH2Cl2 )50 mL). After 2 h, 3,5-di-tert-
butylamine )229 mg, 1.12 mmol) was added and the reac-
tion stirred for an additional 2 h. The CH2Cl2 was removed
under vacuum, the residue was dissolved in chloroform
)100 mL), and shaken with 5% HCl )100 mL). The resulting
yellow precipitate was washed several times with chloro-
form and then suspended in fresh chloroform )100 mL). The
suspension was shaken with 5% NaOH )50 mL) and the
layers separated. The resulting aqueous layer was extracted
once with chloroform )50 mL) and the combined organic
extracts were dried over Na2SO4. The chloroform was
removed under vacuum and the resulting residue was tri-
turated with a minimal amount of CHCl3 )2£5 mL) and
®ltered yielding a white solid. The impure mother liquor
was puri®ed by column chromatography through alumina
)Act II-III) using CHCl3 as the eluent. Combined yield 78%;
3. Conclusions
These experiments clearly highlight the need to consider a
polytopic approach in self-assembly synthesis, where the
advantageous features of more than one type of molecular
recognition interaction can be harnessed to provide access to
robust and well-ordered supramolecular assemblies. In the
present cases, ion pairing certainly contributes to the
stability of complexes 1 and 2, however, it is the cooperative
hydrogen bonds that, not only aid in maintaining the struc-
tural integrity of the assemblies, but are the critical element
necessary to align the building blocks into the optimal
spatial arrangement so that the ion pairing attractive forces
can be maximized. The complexes described in this report
also represent attractive architectures to investigate the
through-space communication of chromophores and
the motion of the building blocks within the assemblies.
The ability to ®ne tune the photonic and dynamic behavior
of the supermolecules by modifying the chromophores
)varying the metal within the metallo)terpyridine) and
using a metalloporphyrin, for example) and/or by tailoring
the size of the components )varying their proximity, for
example) is paramount if supramolecular architectures are
to be applied as the components of molecular machinery.
1
mp 236±2378C; H NMR )300 MHz, CDCl3) d 8.70 )m,
2H), 8.68 )s, 2H), 8.64 )d, Jˆ7.2 Hz, 2H), 7.86 )dt, Jˆ8.1,
1.8 Hz, 2H), 7.67±7.60 )m, 3H), 7.45 )t, Jˆ8.1 Hz, 1H),
7.33 )dd, Jˆ7.8, 4.8 Hz, 2H), 7.24 )m, 1H), 7.18 )d,
Jˆ1.2 Hz, 2H), 6.73 )br s, 1H), 6.41 )br s, 1H) 1.33 )s,
18H); 13C NMR )75 MHz, CDCl3) d 155.7, 154.9, 152.7,
150.7, 149.5, 149.4, 140.9, 138.8, 138, 137.5, 129.9, 124.5,
120.9, 120.2, 119.2, 117.8, 116.2, 115.9, 113.0, 34.5, 31.2;
IR )microscope) n 3332, 3155, 3054, 2963, 2904, 2867,
1651, 1607, 1586, 1567, 1441, 1287 cm21; HRMS )ES)
Calcd for [M1H]1 )C36H38N5O) 556.3076. Found:
556.3076.
4. Experimental
4.1. General
All solvents )Caledon) were distilled prior to use with the
exception of DMSO which was used as received. Solvents
for NMR analysis )Cambridge Isotope Laboratories) were
used as received. All synthetic precursors were purchased
4.2.2. N,N0-Dimethyled urea 6b. Asolution of
)38.0 mg, 0.068 mmol) in anhydrous DMF )3 mL) was
treated with NaH )8 mg, 0.33 mmol). After 15 min of
6a

T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
647
stirring at room temperature, iodomethane )12 mL) was
added and the reaction stirred at 508C under argon for
1.5 h. The reaction mixture was cooled to room temperature
and H2O )10 mL) was added. The mixture was extracted
with EtOAc )3£10 mL) and the combined organic extracts
were dried over Na2SO4. The solvent was removed under
reduced pressure and the residue was puri®ed by column
chromatography )1:1 hexane/EtOAc) through Alumina )Act
II-III) affording a white solid. Yield 96%; 1H NMR
)300 MHz, CDCl3) d 8.70 )m, 2H), 8.63 )d, Jˆ8.1 Hz,
2H), 8.52 )s, 2H), 7.86 )dt, Jˆ2.1, 7.5 Hz, 2H), 7.41 )d,
Jˆ7.5 Hz, 1H), 7.33 )ddd, Jˆ7.5, 4.5, 0.9 Hz, 2H), 7.17
)s, 1H), 7.14 )t, Jˆ7.5 Hz, 2H), 6.95 )dd, Jˆ1.8, 1.8 Hz,
1H), 6.86 )d, Jˆ7.8 Hz, 2H), 6.61 )d, Jˆ1.8 Hz, 2H), 3.26
)s, 3H), 3.25 )s, 3H), 1.11 )s, 18H); 13C NMR )75 MHz,
CDCl3) d 161.6, 156.2, 156.0, 151.3, 149.2, 146.5, 144.9,
139.1, 138.8, 129.0, 126.5, 124.5, 123.9, 121.3, 120.5,
119.3, 118.8, 39.7, 39.4, 34.7, 31.3; HRMS )EI) Calcd for
M1 )C38H41N5O) 583.3311. Found: 583.3310.
1600, 1577, 1516, 1499, 1481, 1424, 1407, 1395, 1386,
1362, 1330, 1307, 1295, 1266, 1212, 1146, 1123, 1110,
1088, 1076 cm21; HRMS )EI) Calcd for M1 )C19H21NO3)
311.1521. Found: 311.1513.
4.2.5. 3-tert-Butyl-9,90-dimethyl-6-nitro-xanthene car-
boxaldehyde ꢀ10). Hexamethylenetetraamine )HMTA)
)491 mg, 3.5 mmol) was added in one portion to a solution
of xanthene 9 )934 mg, 3.0 mmol) in CF3CO2H )10 mL).
After heating at re¯ux for 24 h, the reaction was cooled
and poured into 4 M HCl )100 mL). After stirring at room
temperature for 15 min, the mixture was extracted with
CH2Cl2 )3£100 mL), the combined organic extracts were
dried over Na2SO4 and the solvent evaporated under
vacuum. Puri®cation by column chromatography through
silica )CH2Cl2) afforded a white solid. Yield 628 mg
1
)62%); mp 237±2398C; H NMR )300 MHz, CD2Cl2) d
10.66 )s, 1H), 8.40 )d, Jˆ2.7 Hz, 1H), 8.14 )dd, Jˆ9.0,
2.7 Hz, 1H), 7.81 )d, Jˆ2.7 Hz, 1H), 7.74 )d, Jˆ2.7 Hz,
1H), 7.27 )d, Jˆ9.0 Hz, 1H), 1.72 )s, 6H), 1.36 )s, 9H);
13C NMR APT )125 MHz, CD2Cl2) d 188.9 )CH), 154.6
)C), 149.6 )C), 147.6 )C), 144.4 )C), 131.4 )C), 129.9
)CH), 129.9 )C), 124.0 )CH), 123.9 )CH), 123.6 )C),
123.3 )CH), 117.8 )CH), 36.9 )C), 34.9 )C), 32.7 )CH3),
31.3 )CH3); IR )microscope) n 3335, 3075, 2982, 2967,
2875, 2761, 1675, 1648, 1628, 1606, 1582, 1525, 1462,
1421, 1391, 1363, 1340, 1329, 1301, 1277, 1268, 1254,
1229, 1205, 1159, 1129, 1115, 1076 cm21; HRMS )EI)
Calcd for M1 )C20H21NO4) 339.1471. Found: 339.1471.
0
4.2.3. 9,90-Dimethyl-3-nitro-xanthene. Asolution of 9,9 -
dimethylxanthene )8) )2.1 g, 10.0 mmol) in glacial acetic
acid )50 mL) was treated with fuming HNO3 )700 mg)
and fuming H2SO4 )1 mL). After stirring at room tempera-
ture for 2 h, the reaction mixture was poured over ice
)200 mL) and neutralized by the careful addition of solid
NaHCO3. The mixture was extracted with CH2Cl2
)3£100 mL) and the combined organic extracts were dried
over Na2SO4. The solvent was removed under vacuum and
the residue puri®ed by column chromatography through
silca )50:1 hexane/EtOAc) affording a pale yellow solid.
4.2.6. Nitroterpyridine 11. Amixture of aldehyde 10
)300 mg,
1
Yield 78%; mp 102±1038C; H NMR )300 MHz, CD2Cl2)
)250 mg,
0.737 mmol),
2-acetylpyridine
d 8.37 )d, Jˆ2.7 Hz, 1H), 8.10 )dd, Jˆ9.0, 2.7 Hz, 1H), 7.47
)dd, Jˆ7.8, 2.7 Hz, 1H), 7.26 )ddd, Jˆ7.2, 1.5, 0.6 Hz, 1H),
7.16 )td, Jˆ7.5, 1.5 Hz, 1H), 7.16 )d, Jˆ9.0 Hz, 1H), 7.10
)dd, Jˆ7.8, 1.5 Hz, 1H), 1.68 )s, 6H); 13C NMR APT
)125 MHz, CD2Cl2) d 155.6 )C), 149.7 )C), 143.9 )C),
131.5 )C), 129.2 )C), 128.3 )CH), 126.7 )CH), 124.8
)CH), 123.8 )CH), 123.4 )CH), 117.5 )CH), 116.8 )CH),
34.7 )C), 32.8 )CH3); IR )microscope) n 3099, 2986,
2963, 2930, 2865, 2463, 2073, 1917, 1803, 1633, 1600,
1575, 1517, 1481, 1467, 1447, 1421, 1369, 1338, 1299,
1259, 1206, 1157, 1134, 1110, 1076, 1040 cm21; HRMS
)EI) Calcd for M1 )C15H13NO3) 255.0895. Found:
255.0896.
2.50 mmol), 5% aqueous NaOH )1 mL) and EtOH )3 mL)
was stirred at room temperature for 15 h. The reaction was
concentrated to dryness leaving a pink amphorous solid
)385 mg). The pink solid )193 mg) was suspended in glacial
acetic acid )1 mL), treated with excess NH4OAc )1 g) and
heated at re¯ux for 15 h. The reaction was cooled and
diluted with 50% aqueous EtOH, affording a yellow pre-
cipitate. The precipitate was collected by vacuum ®ltration
and washed with 50% aqueous EtOH. Recrystallization
from EtOH/H2O afforded an off-white solid. Yield 63%;
1
mp 201±2038C; H NMR )300 MHz, CD2Cl2) d 8.72 )m,
6H), 8.40 )d, Jˆ2.4 Hz, 1H), 8.06 )dd, Jˆ9.0, 2.7 Hz, 1H),
7.92 )td, Jˆ7.5, 1.8 Hz, 2H), 7.58 )d, Jˆ2.1 Hz, 1H), 7.50
)d, Jˆ2.1 Hz, 1H), 7.38 )m, 2H), 7.11 )d, Jˆ9.0 Hz, 1H),
1.77 )s, 6H), 1.41 )s, 9H); 13C NMR APT )125.7 MHz,
CD2Cl2) d 156.6 )C), 155.9 )C), 149.6 )CH), 148.1 )C),
147.7 )C), 144.8 )C), 144.0 )C), 137.2 )CH), 131.9 )C),
129.6 )C), 127.6 )C), 124.2 )CH), 123.8 )CH), 123.8
)CH), 123.1 )CH), 122.1, )CH), 121.5 )CH), 117.7 )CH),
35.4 )C), 35.1 )C), 32.4 )CH3), 31.6 )CH3); IR )microscope)
n 3062, 2963, 2869, 1728, 1627, 1605, 1582, 1567, 1542,
1524, 1489, 1468, 1440, 1420, 1388, 1364, 1342, 1266,
1226, 1215, 1173, 1117, 1089 cm21; HRMS )EI) Calcd
for M1 )C34H30N4O3) 542.2318. Found: 542.2315.
4.2.4. 6-tert-Butyl-9,90-dimethyl-3-nitroxanthene ꢀ9). A
solution of 9,90-dimethyl-3-nitroxanthene )1.91 g, 7.5
mmol) in CH2Cl2 )50 mL) was treated with tert-butyl
chloride )694 mg, 7.5 mmol), followed by FeCl3 )catalytic).
After stirring at room temperature over night, the reaction
was quenched by the addition of water )100 mL), extracted
with CH2Cl2 and dried over Na2SO4. Recrystallization from
EtOH afforded pale yellow crystals. Yield 74%; mp 162±
1
1688C )decomp.); H NMR )300 MHz, CD2Cl2) d 8.36 )d,
Jˆ2.7 Hz, 1H), 8.08 )dd, Jˆ9.0, 2.7 Hz, 1H), 7.45 )d,
Jˆ2.1 Hz, 1H), 7.29 )dd, Jˆ8.7, 2.1 Hz, 1H), 7.14 )d,
Jˆ9.0 Hz, 1H), 7.02 )d, Jˆ9.0 Hz, 1H), 1.69 )s, 6H), 1.34
)s, 9H); 13C NMR APT )125 MHz, CD2Cl2) d 155.8 )C),
147.7 )C), 147.4 )C), 143.8 )C), 131.6 )C), 128.3 )CH),
125.4 )CH), 123.7 )CH), 123.4 )CH), 123.2 )CH), 117.5
)CH), 116.2 )CH), 34.9 )C), 34.9 )C), 32.9 )CH3), 31.6
)CH3); IR )microscope) n 3078, 2966, 2905, 2869, 1631,
0
4.2.7.
4 -ꢀ3-Amino-6-tert-butyl-9,90-dimethylxanthyl)-
2,20:60,200-terpyridine. SnCl2´2H2O )1.0 g, 4.5 mmol) was
added to a hot )70±808C) solution of terpyridine 11
)200 mg, 0.37 mmol) in absolute EtOH )30 mL). After stir-
ring at 70±808C for 2±3 h, the heat was removed and the
reaction mixture was poured into an ice-cold mixture of

648
T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
diethylenetriaminepentaacetic acid )DTPA) )2.3 g,
5.9 mmol) and H2O )100 mL). The mixture was neutralized
with saturated NaHCO3, the aqueous phase was extracted
with CHCl3 )3£50 mL), and the combined organic extracts
were dried over Na2SO4. The chloroform was removed
under vacuum yielding an off-white solid. The crude
amine was carried on without further puri®cation. Yield
118.7, 117.3, 114.3, 35.5, 31.8; UV±Vis )DMSO) lmax
)nm) )log e )M21 cm21)) 279 )5.05), 288 )5.04), 328
)4.84), 573 )4.51); IR )microscope) n 3381, 3078, 2962,
2904, 2867, 1698, 1606, 1548, 1495, 1440, 1396, 1298,
1217 cm1; HRMS )ES) Calcd for [M22BF4]21
)C72H74N10O2Fe) 583.2667. Found: 583.2672.
1
4.3.2. Compound 7b. Prepared from 6b. Yield 93%; H
1
96%; mp 1308C )decomp); H NMR )500 MHz, CD2Cl2)
NMR )300 MHz, acetone-d6) d 9.58 )s, 4H), 8.99 )d,
Jˆ8.1 Hz, 4H), 8.07 )m, 8H), 7.48 )m, 6H), 7.29 )t,
Jˆ7.2 Hz, 4H), 7.22 )d, Jˆ7.2 Hz, 2H), 7.12 )s, 2H), 6.90
)d, Jˆ1.5 Hz, 4H), 3.36 )s, 6H), 3.29 )s, 6H), 1.25 )s, 36H);
13C NMR )75 MHz, acetone-d6) d 161.5, 160.8, 159.2,
153.8, 152.0, 150.8, 148.0, 146.0, 139.9, 137.3, 130.7,
128.6, 127.9, 125.1, 124.7, 124.2, 122.0, 120.8, 119.9,
39.8, 39.1, 35.4, 31.7; UV±Vis )DMSO) lmax )nm) )log e
)M21 cm21)) 289 )4.73), 326 )4.58), 576 )4.26); IR )micro-
scope) n 3080, 2962, 2902, 1662, 1617, 1593, 1541, 1447,
1409, 1287 cm21; HRMS )ES) Calcd for [M2BF4]1
)C76H82N10O2Fe BF4) 1309.6000. Found: 1309.6004.
d 8.73 )s, 2H), 8.71 )m, 4H), 7.91 )td, Jˆ7.5, 1.5 Hz, 2H),
7.53 )d, Jˆ2.5 Hz, 1H), 7.43 )d, Jˆ2.0 Hz, 1H), 7.36 )m,
2H), 6.81 )d, Jˆ9.0 Hz, 1H), 6.78 )d, Jˆ3.0 Hz, 1H), 6.51
)dd, Jˆ9.0, 2.5 Hz, 1H), 3.57 )br s, 2H), 1.67 )s, 6H), 1.40
)s, 9H); 13C NMR APT )125.7 MHz, CD2Cl2) d 156.9 )C),
155.7 )C), 149.6 )CH), 148.8 )C), 146.7 )C), 145.7 )C),
144.0 )C), 142.9 )C), 137.1 )CH), 131.7 )C), 130.8 )C),
126.9 )C), 125.7 )CH), 124.0 )CH), 123.7 )CH), 122.2
)CH), 121.4 )CH), 117.3 )CH), 114.7 )CH), 112.2 )CH),
35.3 )C), 34.9 )C), 31.7 )CH3), 31.7 )CH3); IR )microscope)
n 3440, 3350, 3250, 3060, 2963, 2868, 1681, 1626, 1585,
1567, 1547, 1503, 1469, 1439, 1389, 1362, 1295, 1276,
1243, 1218, 1151, 1125, 1090 cm21; HRMS )EI) Calcd
for M1 )C34H32N4O) 512.2576. Found: 512.2561.
1
4.3.3. Compound 13. Prepared from 12. Yield 92%; H
NMR )500 MHz, acetone-d6) d 9.60 )s, 4H), 8.98 )d,
Jˆ4.8 Hz, 4H), 8.27 )br s, 2H), 8.16 )br s, 2H), 8.10 )td,
Jˆ8.0, 1.5 Hz, 4H), 8.01 )d, Jˆ2.5 Hz, 2H), 7.96 )d,
Jˆ2.5 Hz, 2H), 7.78 )d, Jˆ2.0 Hz, 2H), 7.60 )d,
Jˆ4.5 Hz, 4H), 7.51 )dd, Jˆ8.5, 2.5 Hz, 2H), 7.45 )d,
Jˆ2.5 Hz, 4H), 7.32 )ddd, Jˆ8.0, 1.5, 1.5 Hz, 4H), 7.14
)t, Jˆ1.5 Hz, 2H), 7.14 )t, Jˆ2.5 Hz, 2H), 7.07 )d,
Jˆ9.0 Hz, 2H), 1.84 )s, 12H), 1.53 )s, 18H), 1.31 )s,
36H); 13C NMR APT )125 MHz, acetone-d6) d 161.0 )C),
159.5 )C), 154.1 )CH), 153.9 )C), 151.9 )C), 149.7 )C),
147.3 )C), 146.9 )C), 145.9 )C), 140.5 )C), 139.9 )CH),
137.4 )C), 131.8 )C), 131.1 )C), 128.5 )CH), 127.1 )CH),
126.6 )CH), 125.6 )CH), 125.5 )C), 124.8 )CH), 124.9 )CH),
119.5 )CH), 117.4 )CH), 117.2 )CH), 117.0 )CH), 114.1
)CH), 35.8 )C), 35.5 )C), 35.4 )C), 32.6 )CH3), 31.9
)CH3), 31.8 )CH3); UV±Vis )DMSO) lmax )nm) )log e
)M21 cm21)) 278 )5.20), 327 )4.89), 570 )4.51); IR )micro-
scope) n 3388, 3076, 2963, 2906, 2868, 1694, 1608, 1552,
4.2.8. Terpyridine 12. This urea was synthesized following
the same procedure used to prepare 6a. Puri®cation by
column chromatography )3:1 hexane/EtOAc) through
Alumina )Act II-III) afforded a white solid. Yield 65%;
1
mp 237±2398C; H NMR )500 MHz, CD2Cl2) d 8.73 )s,
2H), 8.70 )m, 4H), 7.90 )td, Jˆ7.5, 1.5 Hz, 2H), 7.53 )d,
Jˆ2.5 Hz, 1H), 7.48 )d, Jˆ2.5 Hz, 1H), 7.45 )d, Jˆ2.5 Hz,
1H), 7.35 )m, 2H), 7.20 )d, Jˆ1.5 Hz, 2H), 7.18 )t,
Jˆ1.5 Hz, 1H), 7.14 )dd, Jˆ9.0, 2.5 Hz, 1H), 6.95 )d,
Jˆ9.0 Hz, 1H), 6.81 )br s, 1H), 6.74 )br s, 1H), 1.67 )s,
6H), 1.39 )s, 9H), 1.29 )s, 18H); 13C NMR )125.7 MHz,
CD2Cl2) d 156.5, 155.2, 153.8, 152.2, 149.2, 148.5, 147.9,
145.8, 145.6, 137.2, 136.9, 132.9, 131.4, 129.8, 126.7,
125.6, 123.7, 123.1, 122.1, 121.7, 121.4, 119.9, 118.8,
117.5, 116.3, 35.0, 34.9, 34.7, 31.9, 31.6, 31.4; IR )micro-
scope) n 3321, 3064, 2963, 2926, 2867, 1650, 1604, 1585,
1567, 1501, 1469, 1439, 1411, 1391, 1363, 1275, 1245,
1503, 1453, 1415, 1363, 1277, 1246, 1213, 1059 cm21
;
;
1224, 1090 cm21 HRMS )ES) Calcd for [M1H]1
HRMS )ES) Calcd for [M22BF4]21 )C98H106N10O4Fe)
771.3879. Found: 771.3861.
)C49H54N5O2) 744.4272. Found: 744.4282.
4.3. General metal-coordination procedure
4.4. Synthesis of the porphyrins, general synthesis of
porphyrinic esters 16a and 16b
H2O)6)BF4)2 )0.150 mmol) was added to a suspension of
terpyridines 7a and 7b or 12 )0.30 mmol) in MeOH
)50 mL). The mixture immediately turned purple
yielding a homogenous solution, which was further stirred
for 3 h at room temperature. The volume of MeOH was
reduced to 5±10 mL, at which point 5±10 mL of Et2O
was added affording a purple precipitate. The solid was
isolated by ®ltration and washed times with cold acetone/
Et2O )1:4).
CF3CO2H )TFA) )1.25 equiv.) was added to a solution of
ethyl 3-formylbenzoate )14a) or methyl 4-formylbenzoate
)14b) )1.0 equiv.) and 2,20-dipyrrolyl ketone 1536
)1.0 equiv.) in freshly distilled degassed CH2Cl2
),5 mM). The reaction mixture was protected from light
and stirred at room temperature for 18 h, at which time
p-chloroanil )500 mg, 2.03 mmol) was added and the
mixture was stirred for an additional 30 min. Triethylamine
)3 mL) was added to neutralize the acid and the solution
concentrated under vacuum to yield a purple residue. The
crude residue was puri®ed by column chromatography
through silica by elution with CHCl3, followed by trituration
of the resulting purple solid with acetone.
1
4.3.1. Compound 7a. Prepared from 6a. Yield 95%; H
NMR )300 MHz, acetone-d6) d 9.55 )s, 4H), 9.06 )d,
Jˆ7.8 Hz, 4H), 8.70 )s, 2H), 8.51, )br s, 2H), 8.28 )br s,
2H), 8.07 )dt, Jˆ7.8, 1.2 Hz, 4H), 7.95 )d, Jˆ7.8 Hz, 2H),
7.64 )m, 8H), 7.52 )d, Jˆ1.5 Hz, 4H), 7.27 )m, 4H), 6.90 )d,
Jˆ1.5 Hz, 4H), 1.34 )s, 36H); 13C NMR )75 MHz, acetone-
d6) d 161.5, 159.3, 154.2, 153.8, 152.0, 151.9, 142.2, 140.2,
139.7, 138.3, 130.8, 128.5, 124.9, 122.5, 122.4, 121.5,
1
4.4.1. Compound 16a. Yield 32%; H NMR )300 MHz,
CDCl3) d 10.33 )s, 2H), 9.40 )d, Jˆ4.8 Hz, 4H), 9.00 )d,

T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
649
Jˆ4.8 Hz, 4H), 8.94 )s, 2H), 8.51 )d, Jˆ7.8 Hz, 2H), 8.43
)d, Jˆ7.8 Hz, 2H), 7.88 )t, Jˆ7.8 Hz, 2H), 4.48
)q, Jˆ7.0 Hz, 4H), 1.41 )t, Jˆ7.0 Hz, 6H), 23.14 )s, 2H);
13C NMR )75 MHz, CDCl3) d 166.9, 147.1, 145.5, 141.7,
138.7, 135.1, 132.0, 130.9, 129.6, 129.1, 127.2, 117.9,
105.6, 61.4, 14.4; UV±Vis )CH2Cl2) lmax )nm) )log e
)M21 cm21)) 407 )5.67), 502 )4.32), 535 )3.78), 574
)3.80), 630 )3.30); IR )microscope) n 3272, 3102, 2982,
2937, 2906, 1722, 1579, 1412, 1365, 1299, 1272,
1239 cm21; HRMS )EI) Calcd for M1 )C38H30N4O4)
606.2269. Found: 606.2269.
porphyrin )0.1 mmol) suspended in MeOH )2 mL). After
stirring at room temperature for 16 h, the solvent was
removed under vacuum affording a purple oil. This oil
was dissolved in methanol )0.25±0.50 mL) and diethyl
ether was added until the product oiled out. The supernatant
was removed and discarded and the residue was dried under
high vacuum for 1 day yielding a purple solid. The resulting
porphyrinic salts can be stored for up to 6 months in an
anhydrous environment of CaSO4.
1
4.6.1. Compound 18a. Yield 95%; H NMR )300 MHz,
CDCl3) d 10.62 )s, 2H), 9.64 )d, Jˆ4.5 Hz, 4H), 9.03 )d,
Jˆ4.5 Hz, 4H), 8.67 )s, 2H), 8.29 )d, Jˆ7.5 Hz, 2H), 8.13
)d, Jˆ7.5 Hz, 2H), 7.73 )t, Jˆ7.8 Hz, 2H), 3.15 )m, 16H),
1.55 )m, 16H), 1.29 )m, 16H), 0.92 )t, Jˆ7.2 Hz, 24H),
23.22 )s, 2H); 13C NMR )75 MHz, CDCl3) d 171.6,
145.0, 139.9, 139.4, 135.8, 135.2, 131.5, 131.3, 129.1,
126.1, 120.0, 104.9, 58.7, 23.9, 19.6, 13.5; UV±Vis
)DMSO) lmax )nm) )log e )M21 cm21)) 409 )5.49), 503
)4.27), 538 )3.87), 576 )3.79), 630 )3.41); IR )microscope)
n 3280, 3087, 2960, 2874, 1607, 1587, 1567, 1467, 1358,
1257 cm21; HRMS )ES-neg) Calcd for [M2NBu4]2
)C50H56N5O4) 790.4327, [M1H22NBu4]2 )C34H21N4O4)
549.1557 Found: 790.4335, 549.1559.
1
4.4.2. Compound 16b. Yield 24%; H NMR )300 MHz,
CDCl3) d 10.36 )s, 2H), 9.43 )d, Jˆ4.8 Hz, 4H), 9.04 )d,
Jˆ4.8 Hz, 4H), 8.50 )d, Jˆ8.1 Hz, 4H), 8.36 )d, Jˆ8.1 Hz,
4H), 4.14 )s, 6H), 23.13 )s, 2H); 13C NMR )75 MHz,
CDCl3) d 167.3, 146.7, 146.1, 145.3, 134.8, 132.0, 130.8,
129.6, 128.2, 117.9, 105.7, 52.5; UV±Vis )DMSO) lmax
)nm) )log e )M21 cm21)) 408 )5.35), 503 )3.97), 538
)3.62), 574 )3.41), 629 )3.14); IR )microscope) n 3247,
3093, 3035, 2957, 1719, 1606, 1433, 1312, 1293,
1266 cm21; HRMS )EI) Calcd for M1 )C36H26N4O4)
578.1954. Found: 578.1967.
4.5. General synthesis of porphyrins 17a and 17b
1
4.6.2. Compound 18b. Yield 92%; H NMR )300 MHz,
DMSO-d6) d 10.61 )s, 2H), 9.64 )d, Jˆ4.5 Hz, 4H), 9.05
)d, Jˆ4.5 Hz, 4H), 8.28 )d, Jˆ7.8 Hz, 4H), 8.12 )d,
Jˆ7.8 Hz, 4H), 3.15 )m, 16H), 1.55 )m, 16H), 1.29 )m,
16H), 0.92 )t, Jˆ7.2 Hz, 24H), 23.22 )s, 2H); UV±Vis
)DMSO) lmax )nm) )log e )M21 cm21)) 410 )5.54), 504
)4.23), 540 )3.95), 579 )3.78), 633 )3.48); IR )microscope)
n 3279, 3111, 1703, 1605, 1586, 1402, 1239 cm21; HRMS
)ES-neg) Calcd for [M1H22NBu4]2 )C34H21N4O4)
549.1557. Found: 549.1560.
An aqueous solution of KOH )4 mL, 2.0 M) was added to a
solution of ester 16a or 16b )0.10 mmol) in THF )20 mL).
After heating at re¯ux for 2±3 days, the mixture was cooled
to room temperature and 5% HCl was added dropwise until
the pH of the solution was 3±4. The resulting precipitate
was ®ltered and triturated with CH2Cl2 yielding the pure
porphyrin as a purple solid.
1
4.5.1. Compound 17a. Yield 93%; H NMR )300 MHz,
DMSO-d6) d 13.27 )br s, 2H), 10.68 )s, 2H), 9.69 )d,
Jˆ4.5 Hz, 4H), 9.02 )d, Jˆ4.5 Hz, 4H), 8.76 )s, 2H), 8.55
)d, Jˆ7.8 Hz, 2H), 8.45 )d, Jˆ7.8 Hz, 2H), 8.02 )t,
Jˆ7.8 Hz, 2H), 23.28 )s, 2H); UV±Vis )DMSO) lmax
)nm) )log e )M21 cm21)) 408 )5.59), 502 )4.21), 538
)3.73), 574 )3.70), 629 )3.30); IR )microscope) n 3277,
3200±2400, 1685, 1580, 1445, 1413, 1306, 1260,
1
4.6.3. Compound 20a. Yield 91%; H NMR )300 MHz,
DMSO-d6) d 8.82 )s, 8H), 8.21 )d, Jˆ6.6 Hz, 8H), 8.04
)Jˆ6.6 Hz, 8H), 3.14 )m, 32H), 1.55 )m, 32H), 1.26 )m,
32H), 0.92 )t, 48H), 22.89 )br s, 2H); UV±Vis )DMSO)
lmax )nm) )log e )M21 cm21)) 421 )5.65), 518 )4.20), 552
)4.00), 593 )3.70), 648 )3.78); IR )microscope) n 3466,
3277, 3089, 2960, 2873, 1666, 1595, 1551, 1483, 1367,
1280, 1239, 1150 cm21; HRMS )ES-neg) Calcd for
[M12H23NBu4]2 )C64H65N5O8) 1031.4826, [M13H2
4NBu4]2 )C48H29N4O8) 789.2001, [M1H23NBu4]22
)C64H63N5O8) 514.7363, [M12H24NBu4]22 )C48H28N4O8)
394.0956. Found: 1031.4839, 789.1980, 514.7344,
394.0959.
1241 cm21
;
)C34H21N4O4) 549.1557. Found: 549.1555.
HRMS )ES-neg) Calcd for [M2H]2
1
4.5.2. Compound 17b. Yield 91%; H NMR )300 MHz,
DMSO-d6) d 10.68 )s, 2H), 9.68 )d, Jˆ4.8 Hz, 4H),
9.05 )d, Jˆ4.8 Hz, 4H), 8.43 )d, Jˆ8.1 Hz, 4H), 8.41
)d, Jˆ8.1 Hz, 4H), 23.27 )s, 2H); UV±Vis )DMSO)
lmax )nm) )log e )M21 cm21)) 410 )5.35), 505 )3.97),
539 )3.64), 577 )3.48), 631 )3.20); IR )microscope)
n 3277, 3132, 1697, 1607, 1536, 1404, 1314, 1287,
1239, 1198, 1177, 1147, 1099 cm21; HRMS )ES-neg)
Calcd for [M2H]2 )C34H21N4O4) 549.1557. Found:
549.1561.
1
4.6.4. Compound 20b. Yield 88%; H NMR )300 MHz,
DMSO-d6) d 8.79 )s, 8H), 8.62 )t, Jˆ7.5, 4H), 8.25 )d,
Jˆ7.5 Hz, 4H), 8.14±8.04, )m, 4H), 7.65 )t, Jˆ7.5 Hz,
4H), 22.86 )s, 2H); 13C NMR )300 MHz, CDCl3) d
171.5, 140.9, 138.9, 135.7, 134.9, 134.4, 129.1, 125.9,
125.8, 120.7, 118.2; UV±Vis )DMSO) lmax )nm) )log e
)M21 cm21)) 422 )5.64), 518 )4.11), 553 )3.90), 591
)3.48), 648 )3.60); IR )microscope) n 3318, 2961, 2874,
4.6. General synthesis of porphyrins 18a,b and 20a,b
1607, 1588, 1566, 1486, 1370, 1252, 1158, 1109 cm21
MS )ES-neg) Calcd for [M2NBu4]2 )C96H134N7O8)
;
A0.1 M methanolic solution of tetrabutylammonium
hydroxide was prepared fresh by diluting 652.9 mg of a
40% aqueous tetrabutylammonium hydroxide solution
with methanol )10 mL). This solution )2 mL for 17a and
17b, and 4 mL for 19a37 and 19b38) were added to the
1514.0,
[M1H22NBu4]2
)C80H99N6O8)
1271.8,
[M12H23NBu4]2 )C64H64N5O8) 1030.5. Found: 1514.0,
1271.7, 1030.4.

650
T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
4.7. Synthesis of the terpyridine´porphrin complexes,
general synthesis of complexes 1 and 2
a one- or two-site non-linear regression analysis in the
Originw software package.
Asolution of iron)terpyridine) 7a )0.024 mmol) in MeOH
)12 mL) was added porphyrin 18a )0.024 mmol) or 20a
)0.012 mmol). Adark precipitate immediately formed.
After stirring at room temperature for 2±3 h, the solution
was concentrated to 1±2 mL and ®ltered. The ®lter cake was
washed several times with fresh MeOH affording a purple
solid.
4.9. Steady-state ¯uorescence titrations
Fluorescence measurements were performed using a PTI
C60 photon counting spectrophotometer. In a typical experi-
ment, the porphyrin )2.0 mL, 1.0£1026 M) in DMSO was
added to a quartz cuvette and the initial ¯uorescence pro®le
of the porphyrin was recorded )lexˆ415 nm for 18a and
18b, lexˆ430 nm for 20a and 20b, while monitoring the
¯uorescence between lˆ550±800 nm). Aliquots )10±
20 mL) of a DMSO solution of the iron)terperidine)
)2.0£1025 M) were added to the cuvette while stirring.
After each addition, the ¯uorescence spectrum was
recorded. Upon reaching 1.5±2 equiv. of iron)terperidine),
the aliquots were increased to 50±200 mL until 1.1 mL )8±
10 equiv.) was added. The raw ¯uorescence intensities
)lemˆ633 nm for 18a and 18b, lexˆ650 nm for 20a and
20b) were used to generate the inverse Stern±Volmer
quenching plots.
4.7.1. Compound 1. Yield 97%; 1H NMR )300 MHz,
DMSO-d6) d 12.10 )br s, 2H), 11.84 )br s, 2H), 10.35 )s,
2H), 9.52 )d, Jˆ4.8 Hz, 4H), 9.20 )s, 4H), 9.00 )d,
Jˆ4.8 Hz, 4H), 8.87 )s, 2H), 8.62 )d, Jˆ7.5 Hz, 2H), 8.44
)m, 8H), 8.20 )s, 2H), 7.94 )t, Jˆ7.8 Hz, 2H), 7.80 )s, 6H),
7.60 )t, Jˆ7.8 Hz, 2H), 7.09 )s, 2H), 6.37 )br s, 8H), 5.55 )br
s, 4H), 1.42 )s, 36H), 23.40 )s, 2H); MS )ES) Calcd for
[M1Na]1 )C106H94N14O6FeNa) 1738.9, [M12Na]21
)C106H94N14O6FeNa2) 880.9. Found 1738.7, 880.7.
4.7.2. Compound 2. Yield 95%; 1H NMR )300 MHz,
DMSO-d6) d 11.95 )br s, 2H), 11.48 )br s, 2H), 9.27 )s,
8H), 8.75 )s, 8H), 8.59±8.51 )m, 16H), 8.29 )br s, 8H),
8.24 )d, Jˆ7.5 Hz, 4H), 7.84 )d, Jˆ7.5 Hz, 4H), 7.80 )t,
Jˆ7.5 Hz, 4H), 7.71 )s, 8H), 7.58 )t, Jˆ7.5 Hz, 4H), 7.05
)s, 4H), 6.62 )br s, 8H), 6.54 )br s, 8H), 5.82 )br s, 8H), 1.37
)s, 72H), 23.06 )s, 2H); MS )ES) Calcd for [M12Na]21
)C192H174N36O12Fe2Na2) 1583.7, Found: 1583.5.
Acknowledgements
This work was supported by the Natural Sciences and Engi-
neering Research Council of Canada and the University of
Alberta. We are grateful to Dr Randy Whittal for obtaining
the mass spectral data of the complexes.
4.7.3. Xanthene complex 18a´13. Asolution of the iron-
)terpyridine) 13 )0.002 mmol) in acetone )2 mL) was treated
with porphyrin 18a )0.002 mmol). The solution was stirred
for 30 min and then sonicated for 10±20 s, at which point a
dark precipitate formed. The suspension was ®ltered and the
isolated solid was washed several times with acetone until
References
1. Ward, M. D. Chem. Soc. Rev. 1997, 26, 365 and references
therein.
2. Imamura, T.; Fukushima, K. Coord. Chem. Rev. 2000, 198,
133.
1
the ®ltrate was colorless. Yield 91%; H NMR )500 MHz,
3. Haycock, R. A.; Yartsev, A.; Michelsen, U.; Sundstrom, V.;
Hunter, C. A. Angew. Chem., Int. Ed. Engl. 2000, 39, 3616.
4. Ogawa, K.; Kobuke, Y. Angew. Chem., Int. Ed. Engl. 2000,
39, 4070.
DMSO-d6) d 12.14 )br s, 2H), 11.78 )br s, 2H), 10.62 )s,
2H), 9.65 )d, Jˆ4.5 Hz, 4H), 9.32 )s, 4H), 9.04 )d,
Jˆ4.5 Hz, 4H), 8.91 )br s, 2H), 8.69 )d, Jˆ7.5 Hz, 4H),
8.56 )d, Jˆ8.0 Hz, 2H), 8.50 )br s, 2H), 8.15 )d,
Jˆ7.5 Hz, 2H), 7.85 )t, Jˆ8.0 Hz, 2H), 7.78 )m, 8H), 7.70
)t, Jˆ7.5 Hz, 4H), 7.13 )d, Jˆ9.0 Hz, 2H), 7.05 )s, 2H), 7.01
)two br s, 4H), 6.96 )t, Jˆ5.0, Hz, 4H), 6.82 )d, Jˆ9.0 Hz,
2H), 1.78 )s, 12H), 1.47 )s, 18H), 1.41 )s, 36H), 23.27 )br s,
2H); MS )ES) Calcd for [M1Na]1 )C132H126N14O8FeNa)
2115.4, [M12Na]1 )C132H126N14O8FeNa2) 1069.2. Found:
2115.8, 1069.0.
5. Chichak, K.; Branda, N. R. Chem. Commun. 2000, 1211.
6. Mak, C. C.; Bampos, N.; Sanders, J. K. M. Chem. Commun.
1999, 1085.
7. Wilson, G. S.; Anderson, H. L. Chem. Commun. 1999, 1539.
8. Otsuki, J.; Harada, K.; Toyama, K.; Hirose, Y.; Araki, K.;
Seno, M.; Takatera, K.; Watanabe, T. Chem. Commun.
1998, 1515.
9. Drain, C. M.; Ni®atis, F.; Vasenko, A.; Batteas, J. D. Angew.
Chem., Int. Ed. Engl. 1998, 37, 2344.
4.8. Isothermal titration calorimetry
10. Hunter, C. A.; Shannon, R. J. Chem. Commun. 1996, 1361.
11. Hunter, C. A.; Hyde, R. A. Angew. Chem., Int. Ed. Engl. 1996,
35, 1936.
All binding experiments were preformed on an isothermal
titration calorimeter from Microcal Inc. )Northampton,
Massachusetts). In a typical ITC experiment, a 0.10 mM
solution of the porphyrin dissolved in DMSO was added
to the calorimetry cell )volumeˆ1.448 mL). A1.45 mM
solution of the iron)terpyridine) in DMSO was introduced
in 60, 4 mL injections )4 min between injections) until a
total of 240 mL of iron)terpyridine) was added. The solution
was maintained at a constant operating temperature of 258C
and continuously stirred )400 rpm) to ensure rapid mixing.
The data from the resulting injections were evaluated using
12. Agirtas, S.; Ion, R.-M.; Bekaroglu, O. Mater. Sci. Engng C
2000, 7, 105.
13. Kano, K.; Minamizono, H.; Kitae, T.; Negi, S. J. Phys. Chem
A 1997, 101, 6118.
14. Lipskier, J. F.; Tran-Thi, T. H. Inorg. Chem. 1993, 32, 722.
15. Schneider, H.-J.; Wang, M. J. Org. Chem. 1994, 59, 7464.
16. Drain, C. M.; Shi, X.; Milic, T.; Ni®atis, F. Chem. Commun.
2001, 287.
17. Masiero, S.; Gottarelli, G.; Pieraccini, S. Chem. Commun.
2000, 1995.

T. B. Norsten et al. / Tetrahedron 58 $2002) 639±651
651
18. Ikeda, C.; Nagahara, N.; Motegi, E.; Yoshioka, N.; Inoue, H.
Chem. Commun. 1999, 1759.
31. Fan, E.; Van Arman, S. A.; Kincaid, S.; Hamilton, A. J. Am.
Chem. Soc. 1993, 115, 369.
19. Drain, C. M.; Russell, K. C.; Lehn, J.-M. Chem. Commun.
1996, 337.
32. Norsten, T. B.; Chichak, K.; Branda, N. R. Chem. Commun.
2001, 1794.
20. Myles, A. J.; Branda, N. R. J. Am. Chem. Soc. 2001, 123, 177.
21. Berg, A.; Shuali, Z.; Asano-Someda, M.; Levanon, H.; Fuhs,
M.; Mobius, K.; Wang, R.; Brown, C.; Sessler, J. L. J. Am.
Chem. Soc. 1999, 121, 7433.
33. Mukkala, V.-M.; Helenius, M.; Hemmila, I.; Kankare, J.;
Takalo, H. Helv. Chim. Acta 1993, 76, 1361.
34. Lindsey, J. S.; Prathapan, S.; Johnson, T. E.; Wagner, R. W.
Tetrahedron 1994, 50, 8941.
35. Schaefer, J. P.; Lark, J. C.; Flegal, C. A.; Hoing, L. M. J. Org.
Chem. 1967, 32, 1372.
22. Osuka, A.; Yoneshima, R.; Shiratori, H.; Okada, T.; Tanigu-
chi, S.; Mataga, N. Chem. Commun. 1998, 1567.
23. Hayashi, T.; Miyahara, T.; Norihiro, K.; Tukitoshi, K.;
Masuda, H.; Ogoshi, H. J. Am. Chem. Soc. 1997, 119, 7281.
24. Arimura, T.; Brown, C. T.; Springs, S. L.; Sessler, J. L. Chem.
Commun. 1996, 2293.
36. de Groot, J. A.; Koek, J. H.; Lugtenburg, J. Recl. Trav. Chim.
Pays-Bas 1981, 100, 405.
37. Datta-gupta, N.; Jones, E.; Thomas, L. K.; Malakar, D.
J. Indian Chem. Soc. 1981, 58, 1171.
38. Suarez, M.; Salfran, E.; Rodriguez-Curiel, R. I.; Elguero, J.
Bull. Soc. Chim. Belg. 1997, 106, 211.
25. Kirby, J. P.; van Dantzig, N. A.; Chang, C. K.; Nocera, D. G.
Tetrahedron Lett. 1995, 36, 3477.
39. Corbell, J. B.; Lundquist, J. J.; Toone, E. J. Tetrahedron:
Asymmetry 2000, 11, 95.
26. Turro, C.; Chang, K.; Leroi, G. E.; Cukier, R. I.; Nocera, D. G.
J. Am. Chem. Soc. 1992, 114, 4013.
40. Kavarnos, G. J. Fundamentals of Photoinduced Electron
Transfer; VCH: New York, 1993.
27. Schmuck, C. Chem. Eur. J. 2000, 6, 709.
28. Sebo, L.; Schweizer, B.; Diederich, F. Helv. Chim. Acta 2000,
83, 80.
41. Turro, N. J. Modern Molecular Photochemistry; University
Science Books: Sausalito, 1991; pp 297±361.
29. Linton, B.; Hamilton, A. D. Tetrahedron 1999, 55, 6027.
30. Bell, T. W.; Hext, N. H.; Khasanov, A. B. Pure Appl. Chem.
1998, 70, 2371.