Neutral ligands for selective chloride anion complexation: (α,α, α,α)-5,10,15,20-tetrakis(2- (arylurea)phenyl)porphyrins

Article abstract of DOI:10.1021/ja982052i

A series of neutral, urea-appended, free-base porphyrins and their Zn(II) complexes have been synthesized and characterized. The (α,α, α,α)- 5,10,15,20-tetrakis(2-(arylurea)phenyl)porphyrins bind strongly (K (M^-1) > 10³-10⁵

Full text of DOI:10.1021/ja982052i

11684
J. Am. Chem. Soc. 1998, 120, 11684-11692
Neutral Ligands for Selective Chloride Anion Complexation:
(R,R,R,R)-5,10,15,20-Tetrakis(2-(arylurea)phenyl)porphyrins
Raymond C. Jagessar,† Maoyu Shang,‡ W. Robert Scheidt,*,‡ and Dennis H. Burns*,†
Contribution from the Department of Chemistry, Wichita State UniVersity, Wichita, Kansas 67260, and
Department of Chemistry and Biochemistry, UniVersity of Notre Dame, Notre Dame, Indiana 46556
ReceiVed June 12, 1998
Abstract: A series of neutral, urea-appended, free-base porphyrins and their Zn(II) complexes have been
synthesized and characterized. The (R,R,R,R)-5,10,15,20-tetrakis(2-(arylurea)phenyl)porphyrins bind strongly
(K (M-1) > 103-105) to chloride anion in DMSO-d6 and also in the more competitive solvent system DMSO-
d6/D2O (88:12, v/v) and bromide anion in DMSO-d6, as revealed by 1H NMR titration studies. The porphyrin
derivatives exhibited significant binding selectivity since they complexed with the spherical Cl- and Br- to a
much greater extent than with the tetrahedral H2PO4- and HSO4- and the trigonal NO3- anions in DMSO-d6.
Indeed, the selectivity trend Cl- > Br- . H2PO4 > HSO4 > NO3 is novel for any neutral urea-anion
binding system. On the other hand, the corresponding metalloporphyrins exhibited a decrease in binding
strength and selectivity in DMSO-d6. The stoichiometry of binding for the anions and porphyrins was determined
to be 1:1. The enthalpy of complexation was determined to be highly favorable and the entropy of complexation
determined to be unfavorable from a variable-temperature 1H NMR experiment with a 1:1 tetrabutylammonium
bromide/porphyrin complex. X-ray crystallography revealed (R,R,R,R)-5,10,15,20-tetrakis(2-(4-chlorophenyl-
urea)phenyl)porphyrin to be the first coordination complex of an anion (chloride and bromide) bound by a
neutral free-base porphyrin. The binding motif consisted of the halide, buried deep within the porphyrin pocket,
bound by two adjacent urea functional groups via four hydrogen bonds, with the two remaining urea functional
groups involved in hydrogen bonding to solvent molecules. The crystal structure of the tetrabutylammonium
halide-porphyrin complex showed additional Coulombic interaction between the electron-poor sulfur of a
pocket-bound, hydrogen-bonded DMSO and halide anion.
-
-
-
Introduction
with the addition of functional groups containing H bonds
capable of forming bi- and trifurcated bonding arrays.
The design and fabrication of abiotic supramolecular receptors
for anion recognition is an area of much current activity.1
Particularly challenging is the preparation of receptors that
exhibit a high degree of discrimination toward specific anions.
Nature, as exemplified by the phosphate and sulfate binding
proteins, has surmounted this difficulty by complexing anions
exclusively via hydrogen bonding, with selectivity ratios of
phosphate over sulfate, or sulfate over phosphate, greater than
105.2 Abiotic hosts that utilize hydrogen-bonding motifs
potentially combine the benefits of electroneutrality with
directional Coulombic interactions. Thus, functional groups
capable of H bonding, when positioned correctly in the
supramolecular matrix, would be able to differentiate between
the three-dimensional shapes (linear, trigonal, tetrahedral, or
spherical) of anions. The anion selectivity would be amplified
However, there are only a few reported preparations of neutral
abiotic anion receptors capable of achieving selective recogni-
tion, and those that do employ H-bonding motifs bind most
strongly to phosphate and carboxylate anions.3 Especially
lacking in the arsenal of anion receptors are those capable of
the selective recognition of the chloride anion. The recognition
of the chloride anion is a very significant event in biological
systems as the first step in its ion transport. Chloride ion
channels are critical to respiration, with their facilitated exchange
of chloride for bicarbonate anions in erythrocytes.4 Chloride
ion channels also are essential to the modulation of the central
(3) (a) Bu¨hlmann, P.; Nishizawa, S.; Xiao, K. P.; Umezawa, Y.
Tetrahedron 1997, 53, 1647-1654. (b) Cameron, B. R.; Loeb, S. J. J. Chem.
Soc., Chem. Commun. 1997, 573-574. (c) Beer, P. D.; Graydon, A. R.;
Johnson, A. O. M.; Smith, D. K. Inorg. Chem. 1997, 36, 2112-2118. (d)
Nishizawa, S.; Bu¨hlmann, P.; Iwao, M.; Umezawa, Y. Tetrahedron Lett.
1995, 36, 6483-6486. (e) Scheerder, J.; Engbersen, J. F. J.; Casnati, A.;
Ungaro, R.; Reinhoudt, D. N. J. Org. Chem. 1995, 60, 6448-6454. (f)
Scheerder, J.; Fochi, M.; Engbersen, J. F. J.; Reinhoudt, D. N. J. Org. Chem.
1994, 59, 7815-7820. (g) Kelly, T. R.; Kim, M. H. J. Am. Chem. Soc.
1994, 116, 7072-7080. (h) Boerrigter, H.; Grave, L.; Nissink, J. W. M.;
Chrisstoffels, L. A. J.; van der Mass, J. H.; Verboom, W.; de Jong, F.;
Reinhoudt, D. N. J. Org. Chem. 1998, 63, 4174-4180.
* To whom correspondence should be addressed (D.H.B.). E-mail:
burns@wsuhub.uc.twsu.edu.
† Wichita State University.
‡ University of Notre Dame.
(1) For recent reviews, see: (a) Schmidtchen, F. P.; Berger, M. Chem.
ReV. 1997, 1609-1646. (b) Beer, P. D. J. Chem. Soc., Chem. Commun.
1996, 689-696. (c) Scheerder, J.; Engberson, J. F. J.; Reinhoudth, D. N.
Recl. TraV. Chim. Pays-Bas 1996, 115, 307-320. (d) Lehn, J.-M.
Supramolecular Chemistry; VCH: Weinheim, 1995. (e) Izatt, R. M.; Pawlak,
K.; Bradshaw, J. S. Chem. ReV. 1995, 95, 2529-2586. (f) Dietrich, B. Pure
Appl. Chem. 1993, 65, 1457-1464.
(2) (a) Pflugrath, J. W.; Quiocho, F. A. Nature 1985, 314, 257. (b) He,
J. J.; Quiocho, F. A. Science 1991, 251, 1479. (c) Luecke, H.; Quiocho, F.
A. Nature, 1990, 347, 402. (d) Kanyo, Z. F.; Christianson, D. W. J. Biol.
Chem. 1991, 266, 4264.
(4) Stryer, L. Biochemistry; W. H. Freeman: New York, 1988; Chapter
37.
(5) (a) Koo, Y. H.; Delannoy, M.; Pedersen. P. L. Biochemistry 1997,
36, 5053- 5064. (b) Hanrahan, J. W.; Tabcharani, J. A.; Becq, F.; Mathews,
C. J.; Augustinas, O.; Jensen, T. J.; Chang, X.-B.; Riordan, J. R. Function
and Dysfunction of the CFTR Chloride Channel in Ion Channels and Genetic
Diseases; Dawson, D. C., Frizzel, R. A., Eds.; The Rockefeller University
Press: New York, 1995.
10.1021/ja982052i CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/29/1998

Ligands for SelectiVe Chloride Anion Complexation
J. Am. Chem. Soc., Vol. 120, No. 45, 1998 11685
substituents (chloro and fluoro) to the para position of the phenyl
rings of 3 and 4 allow for some modulation of the supramol-
ecule’s anion binding strength and selectivity. In this paper,
we fully detail the synthesis and characterization of porphyrins
2-8, including the X-ray crystal characterizations of 7, and 3
complexed with bromide and chloride anion, a van’t Hoff
analysis of 4 with bromide anion, and have enlarged the number
of binding studies to include more porphyrins (5, 6, and 8),
more anions (Br-, HSO4-), and a more competitive DMSO-
d6/D2O solvent system.
e
Results and Discussion
Receptor Synthesis. The synthesis of 5,10,15,20-meso-
tetrakis(o-aminophenyl)porphyrin (H2TAPP, 1) was accom-
plished using the method of Collman,9 from which the R,R,R,R-
isomer of H2TAPP was obtained in 65% yield using the
atropisomerization methodology of Lindsey.10 The urea ap-
pended porphyrins 2-5 (Figure 1) were synthesized by the
addition of 4 equiv of the requisite aromatic isocyanate to the
R,R,R,R-atropisomer of H2TAPP in chloroform at room tem-
perature and were obtained as purple microcrystalline solids in
yields ranging from 70 to 85% after purification by silica gel
column chromatography and subsequent recrystallization from
CH2Cl2/hexane. The model tetraphenylporphyrin (TPP) was
prepared using standard literature procedure.11
Figure 1. Urea-appended porphyrin illustrating labeled protons of
interest.
nervous system through the hyperpolarization of neuronal
membranes.4 Additionally, a mutated chloride channel protein
has been implicated as the causative factor of the genetic disease
cystic fibrosis.5 Model systems undoubtedly will prove useful
in elucidating the mechanism of transport of chloride across
membranes, and of utmost importance will be their ability to
discriminate for the chloride anion.
1H NMR and UV/vis Characterization of Receptors. The
1H NMR spectra of 2-5, recorded in DMSO-d6 at ambient
temperature, revealed a broad singlet in the range -2.60 to
-2.66 ppm due to the porphyrin NH protons resonance Hh
(Figure 1), whereas the typical eight â-pyrrole protons (Ha)
appeared as a singlet in the range 8.7-8.8 ppm. The singlet
resonance of the â-pyrrole protons was indicative of the R,R,R,R
conformation of the molecules. The resonances for the attached
urea NH protons (Hf,g) and the porphyrin meso-phenyl ring
protons (Hb-e) were generally found between 7.5 and 8.5 ppm.
In all cases, the urea NH protons were observed as two broad
singlets, with one resonating more than 0.5 ppm downfield from
the other. The proton resonance farthest downfield is presum-
ably the set of (mostly Hg) protons, which the X-ray crystal
analysis (vide infra) showed was H-bonded to three DMSO
molecules. The proton resonances of zinc complexes 6-8, in
comparison to the free-base ligands, were noted at higher field,
and the R,R,R,R conformation of the molecule was retained.
The UV/vis spectra of porphyrins 2-5 revealed porphyrinoid
features with intense Soret and four phyllo type (IV > II > III
> I) Q-bands in the range 421-424 nm and 518-650 nm,
respectively. In comparison to TPP, the Soret (B) and Q-bands
We have recently communicated preliminary (phosphate,
nitrate, and chloride) anion coordination studies in DMSO of
superstructured porphyrins (2-4, Figure 1) functionalized with
hydrogen-bonding urea moieties.6 These macrocycles are the
first example of a class of neutral, free-base porphyrins that
can bind anions and have been demonstrated to be selective
chloride anion receptors in a highly competitive solvent. For
comparison, the expanded porphyrins can bind anions only when
protonated and, thus, exhibit modest anion selectivity due to
nondirectional Coulombic interactions3e (for example, sapphyrin
binds phosphate and fluoride ions with similar binding constants
in methanol7). Cobaltocenium- and ferrocene-appended por-
phyrins have recently been reported that exhibit anion binding.1b,8
The former system requires the combination of the positively
charged cobalticinium center and amide hydrogen-bonding
environment for anion complexation, whereas neutral ferrocene
amide atropisomeric porphyrins do not complex anions unless
metalated. Both systems show little anion selectivity (no better
than 4:1 binding selectivity between halides, nitrate, and sulfate
ions in acetonitrile and dichloromethane, respectively). While
anion receptors incorporating the positively charged centers as
the main ingredients for anion complexation have been exten-
sively employed by several groups,1 the preparation of neutral
abiotic anion receptors that exhibit selective and exceptionally
strong anion complexation is scarce.
1
for these new ligands were red shifted. As with H NMR, the
optical absorption data for the zinc complexes strongly suggest
that there was no gross change in macrocyclic structure upon
metalation. Only the disappearance of the Q-bands (III and IV)
and the appearance of the R and â bands were observed for
these complexes, consistent with the optical absorption spec-
troscopy observed for other metalated zinc porphyrins.12 The
proposed structures 2-8 were corroborated by high-resolution
FAB-mass spectrometry, elemental analysis, and X-ray crystal-
lography (see the Experimental Section).
An important feature of supramolecules 2-4 is the cylindrical
orientation of the urea groups on the porphyrin scaffolding that
allow 2-4 to exhibit remarkable selectivity in their recognition
of the spherical chloride anion over that of the trigonal nitrate
or tetrahedral phosphate anions. In fact, porphyrins 2-4
recognize chloride anion in DMSO with a selectivity of 1000:1
compared to nitrate or 280:1 compared to phosphate anions.
Furthermore, electronic tuning via the addition of electronegative
Anion Coordination Studies. Quantification of the associa-
tion constants of neutral hosts 2-5, 6, and 8 was accomplished
(6) Jagessar, R. C.; Burns, D. H. J. Chem. Soc., Chem. Commun. 1997,
1685-1686.
(9) Collman, J. P.; Gagne, R. R.; Reed, C. A.; Halbert, T. R.; Lang, G.;
Robinson, W. T. J. Am. Chem. Soc. 1975, 97, 1429.
(10) Lindsey, J. S. J. Org. Chem. 1980, 45, 5215.
(11) Adler, A. D.; Longo, F. R.; Finarelli, J. D.; Goldmacher, J.; Assour,
J.; Korsakoff, L. J. Org. Chem. 1967, 32, 476.
(12) Maillard, P.; Guerqurn, K.; Huel, J. L.; Momenteau, M. J. Org.
Chem. 1993, 58, 2774.
(7) Kra´l, V.; Furuta, H.; Shreder, K.; Lynch, V.; Sessler, J. L. J. Am.
Chem. Soc. 1996, 118, 1595-1607.
(8) (a) Beer, P. D.; Drew, M. G. B.; Hesek, D.; Jagessar, R. J. Chem.
Soc., Chem. Commun. 1995, 1187. (b) Beer, P. D.; Drew, M. G. B.; Jagessar,
R. J. Chem. Soc., Dalton Trans. 1997, 881.

11686 J. Am. Chem. Soc., Vol. 120, No. 45, 1998
Jagessar et al.
Table 1. Association Constants for Porphyrins 2-5, 6, and 8 and
Cl-, Br-, NO3-, HSO4-, and H2PO4 Anions in DMSO-d6 and
-
DMSO-d6/D2O (88:12)a
K (M-1
)
K (M-1
)
K (M-1
)
K (M-1
)
K (M-1
)
-
-
-
porphyrin
Cl-
Br-
HSO4
NO3
H2PO4
2
>1 × 105
1.36 × 103 b
>1 × 105
1.02 × 103 b
>1 × 105
9.73 × 103 b
>1 × 105
c
1.01 × 104 115
1.004 × 104 137
9.99 × 103 147
1.005 × 104 226
90
400
3
4
5
60
300
55
1.4 × 103
9.56 × 103
163
6
8
9.5 × 103
1.51 × 103
d
d
23
d
49
489
9.82 × 103 1.1 × 103
a Association constants determined from nonlinear regression analysis
using the program EQNMR, with errors ranging from (10 to 20%.
b Association constants determined in DMSO-d6/D2O 88:12, v/v. c No
association constant was determined due to precipitation in DMSO-
d6/D2O 88:12, v/v. d Broadness in porphyrin proton resonances made
it impossible to determine association constant accurately.
-
Figure 2. Titration curves of porphyrin 4 plus anions: b ) H2PO4
,
differently para-substituted porphyrins 2-5 in this solvent.15
These association constants are far larger than those that have
been previously reported for urea- and nonurea-functionalized
anion receptors that bind chloride.1,7,9,14,16 To ascertain if the
para substituents modulated the porphyrins ability to bind
chloride anion, a more competitive solvent system was sought.
Titration of the porphyrin receptors in a 12% D2O/DMSO-d6
solvent mixture with TBACl resulted in a decrease in the
respective association constants by a factor of approximately
102. This placed the binding constants in a range that allowed
selectivity trends with the various porphyrins to be determined
and for a comparison to those already determined for the other
anions with titration in DMSO-d6.
O ) Cl-, 1 ) Br-, 0 ) HSO4-, and ∇ ) NO3- (tetrabutylammonium
counterion).
by following the titrations of the anions with 1H NMR
spectroscopy in the strongly solvating solvent DMSO-d6 and,
where necessary, in the more highly competitive mixed solvent
system DMSO-d6/D2O (88:12, v/v). The anions Cl-, Br-,
NO3-, HSO4-, and H2PO4 were added as their tetrabutylam-
-
monium salts. Examples of titration binding curves with
porphyrin 4 and representative anions are shown in Figure 2.
The association constants were determined by nonlinear regres-
sion analysis using the computer program EQNMR.13
In all cases, shifts were observed for the receptor porphyrin
meso-phenyl, â-pyrroles, porphyrin NH, urea NH, and urea
phenyl protons. Three sets of protons exhibited invariant shifts
upon complexation: the â-pyrrole protons always shifted
upfield, the urea NH protons always moved downfield, and the
porphyrin NH always moved downfield. Significant shifts were
observed for the urea NH protons, with the farthest downfield
of the two urea NH resonances the most perturbed. For
example, downfield shifts of 0.5 ppm were observed after the
addition of 1 equiv of chloride, and downfield shifts of 1 ppm
were observed after the addition of 2-3 equiv of dihydrogen
phosphate anion. The large shift of the urea NH protons was
indicative of their essential role in the anion recognition process
via hydrogen bonding.3e-f,14 The fact that the â-pyrrolic protons
remained a sharp singlet upon anion complexation suggested
that the anion was bound in the urea pocket of the porphyrin.
The consistent upfield shift of the â-pyrrole protons coupled
with the downfield shift of the porphyrin inner imino protons
upon anion complexation indicated that the anion was affecting
(decreasing) the porphyrin ring current. Thus, the porphyrin
macrocycle was both a scaffold for the urea functional groups
and concomitantly a sensor for the anion’s electrostatic field.
The association constants for 2-5, 6, and 8 with various
anions in DMSO-d6 are presented in Table 1. For all free-base
porphyrins, the binding of Cl- was calculated to be greater than
105 M-1 in DMSO-d6 (indicative of an extremely stable receptor
complex formation), so it was impossible to judge any selectivity
in binding that might be occurring for this anion between the
The modification of the urea phenyl ring with electron-
withdrawing para substituents was anticipated to increase the
acidity of the urea NH’s and in turn increase the strength of the
hydrogen bonds formed between the urea NH’s and the
anions.3a,14a Porphyrin 5, with a para-substituted nitro group,
did exhibit larger binding constants for the anions, and at the
-
same time the selectivity trends (Cl > Br- > H2PO42 > HSO4
,
NO3-) were the same as for the rest of the receptors. Variations
in binding constants (between 2- to 10-fold differences) were
noted for the halides, nitrate, or bisulfate anions between the
different porphyrin receptors. As shown in Table 1, the anion
whose binding selectivity changed most significantly with the
different porphyrin receptors was dihydrogen phosphate. The
binding constants increased by a factor of 24 (from 400 for 2
to 9600 for 5) as the electron-withdrawing power of the para
substituent was increased. We first hypothesized that the large
difference in selectivities observed for porphyrin receptors 2-5
toward phosphate was due to the anion’s ability to act as a
hydrogen-bond donor as well as acceptor. However, semiem-
pirical calculations showed a trend of decreasing negative
surface electrostatic potentials around the carbonyl and aromatic
ring when phenylurea was elaborated with electron-withdrawing
para substituents, a model inconsistent with the above specula-
tion. The binding constant of the tetrahedral bisulfate anion
was smaller than that of the tetrahedral dihydrogen phosphate
anion, and this selectivity may be due primarily to the greater
basicity of the dihydrogen phosphate anion, allowing it to form
stronger hydrogen bonds with the urea NHs. This same property
(13) Hynes, M. J. J. Chem. Soc., Dalton Trans. 1993, 311.
(14) (a) Hughes, M. P.; Smith, B. D. J. Org. Chem. 1997, 62, 4492-
4499. (b) Hamann, B. C.; Branda, N. R.; Rebek, J. Tetrahedron Lett. 1993,
34, 6837. (c) Fan, E.; van Arman, S. A.; Kincaid, S.; Hamilton, A. D. J.
Am. Chem. Soc. 1993, 115, 369.
(15) Binding constants greater than 105 M-1 cannot be determined
accurately: Wang, T.; Bradshaw, J. S.; Izatt, R. M. J. Heterocycl. Chem.
1994, 31, 1097.

Ligands for SelectiVe Chloride Anion Complexation
J. Am. Chem. Soc., Vol. 120, No. 45, 1998 11687
may also be responsible for the greatest differences seen in
binding constants with this anion and the different porphyrin
receptors.
The binding for bromide was high (Ka ) 1.0 × 104) but was
smaller than that of chloride. The overall strong binding of
bromide anion among the different porphyrins, however, cor-
roborated the selectivity these receptors exhibit for anions that
are spherical. The difference in binding selectivity between the
chloride and bromide anions may partially result from the
stronger hydrogen-bonding accepting properties of the chloride
anion, since it has a higher charge density than the larger
bromide anion. The selectivity trend Cl- > Br- > H2PO4
-
displayed by these receptors is novel for any urea anion-binding
system, except for Reinhoudt’s calixarenes, where the pocket
is too small for the phosphate anion to ever fit.3f Analysis of
the X-ray crystal data (vide infra) suggests that an ordered
solvent molecule, hydrogen bonded to the host in the porphyrin
pocket, also plays an important role in anion discrimination.
The four urea groups of receptors 2-5 bounded by the porphyrin
plane creates a pocket that, with the inclusion of complexed
solvent, delineates a cavity optimized for the binding of a
negatively charged spheroidal shape.
-
Figure 3. Job plot of porphyrin 4 + H2PO4 (tetrabutylammonium
counterion).
The anion coordination studies of the zinc(II) complexes of
6 and 8 showed that the porphyrins exhibited a loss both in
selectivity and binding strength, compared to their parent free-
base porphyrins. It was anticipated that the metal might act as
an additional anion-binding site orthogonal to the urea hydrogens
and, thereby, increase the overall binding interaction (as
observed with other multisite porphyrins, such as Beer’s
ferrocene porphyrin, that only binds anions when metalated8b).
The zinc(II) complexes, while exhibiting the same general
selectivity trend of Cl- > Br- > H2PO42-, NO3-, exhibited a
chloride-nitrate and chloride-dihydrogen phosphate selectivity
that was lowered compared to the corresponding free-base urea
porphyrins (Table 1). The binding constants for chloride,
bromide, and dihydrogen phosphate were three to 10 times less
than with their respective free-base porphyrin receptors. This
decrease in selectivity and binding strength for the zinc(II)
complexes may be attributable to an increase in rigidity of the
metalloporphyrin over that of the free-base receptor. This
rigidity, in turn, would affect the flexibility and positioning of
the urea groups required for optimal hydrogen bonding, as
exemplified by the free-base urea analogues. On the other hand,
zinc(II) porphyrins can be pentacoordinate, and thus, coordina-
tion to the metal could occur on the side of the porphyrin
opposite to the urea pocket (bottom-bound). Anion coordination
to the metal opposite the urea pocket would be expected to show
no change in the 1H NMR spectrum in the same way that ZnTTP
X-ray crystal structure of metalloporphyrin 7 shows DMSO
bound to the metal on the underside of the porphyrin pocket
(vide infra), demonstrating that in DMSO the solvent would
effectively compete with this kind of anion-bonding motif.
Certainly the fact that anion binding induces significant shifts
of all the respective host protons suggests that anion coordination
in solution is occurring mainly in the urea-binding pocket of
these metalloporphyrins.
Anion binding to the free-base porphyrins was also character-
ized qualitatively with UV/vis spectroscopy. UV/vis spectral
anion titration revealed anion binding through perturbation of
-
the Soret and visible Q-bands of 2-4, when Cl-, Br-, HSO4
,
H2PO4-, and NO3 anions were added as their tetrabutylam-
monium salts. Porphyrins 2-5 were first placed in a 0.1 M
solution of TBAPF6, and when titrated with the usual anions,
their Soret bands would exhibit a red shift with a concomitant
increase in intensity (the added TBAPF6 produced an environ-
ment of constant ionic strength, and the PF6 anion did not bind
to the porphyrin receptors due to its large size). Changes in
the Soret band upon titration with the anions suggest that the
anions were bound within the porphyrin-urea pocket. Under
analogous titration conditions, the model compound, tetraphen-
-
1
ylporphyrin, exhibited negligible perturbation in its H NMR
and UV/vis spectra, indicating that the changes observed with
ligands 2-8 were truly indicative of anion binding.
1
exhibits no real change in its H NMR spectrum upon anion
The very strong binding of Cl- ion allowed the mole ratio
method18 to be used in the determination of the binding
stoichiometry, which was found to be 1:1 for all four porphyrins.
Job’s method of continuous variation18 was used to determine
the stoichiometry of binding with bromine, dihydrogen phos-
phate, hydrogen sulfate, and nitrate anions for receptor 4. Figure
3 illustrates the Job plot of dihydrogen phosphate anion,
determined by utilizing the method of Tran-Thi and Lipskier19
(UV/vis spectrophotometry, 0.1 M TBAPF6). All Job plots
conformed to a 1:1 binding stoichiometry illustrated by the
inflection point at a mole fraction of 0.5. Fluoride anion, on
the other hand, had completely different binding characteristics
from all the above anions, exhibiting apparent allosteric binding
coordination. If this type of “nonspecific” binding were
extensive, it would reduce the apparent binding constant because
it would reduce the effective concentration of free anion.
However, the binding of bromide and chloride anions to ZnTTP
in dichloromethane has been determined to be relatively weak,
with K ) 15 and 300, respectively.17 Assuming a bottom-bound
anion species had a similar binding constant, its concentration
would be insignificant compared to the species where the anion
was bound in the porphyrin urea pocket (i.e., with a ratio of
binding constants of approximately 103:1). Additionally, the
(16) (a) Kavallieratos, K.; de Gala, S. R.; Austin, D. J.; Crabtree, R. H.
J. Am. Chem. Soc. 1997, 119, 2325. (b) Davis, A. P.; Perry, J. J.; Williams,
R. P. J. Am. Chem. Soc. 1997, 119, 1793. (c) Berger, M.; Schmidtchen, F.
P. J. Am. Chem. Soc. 1996, 118, 8947. (d) Gale, P. A.; Sessler, J. L.; Kra´l,
V.; Lynch, V. J. Am. Chem. Soc. 1996, 118, 5140.
(17) Nappa, M.; Valentine, J. S. J. Am. Chem. Soc. 1978, 100, 5075-
5080.
(18) Connors, K. A. Binding Constants; John Wiley & Sons: New York,
1987; pp 24-28.
(19) Lipskier, J. F.; Tran-Thi, T. H. Inorg. Chem. 1993, 32, 722-731.

11688 J. Am. Chem. Soc., Vol. 120, No. 45, 1998
Jagessar et al.
Figure 5. ORTEP diagram illustrating the molecular structure of
porphyrin 7‚5DMSO. The porphyrin plane is in the plane of the paper,
and the O-coordinated axial DMSO ligand is below the plane. Selected
atoms in the drawing are labeled with the atom names given in the
tables. 50% probability ellipsoids are displayed.
Figure 4. van’t Hoff plot of porphyrin 4 + TBABr: O ) data, b )
linear least-squares fit from which thermodynamic data was determined.
effects with a 2:1 complexation stoichiometry (to be presented
elsewhere20).
that is more centered in the pocket. The former two DMSO
solvent molecules are each bound to one urea via two hydrogen
bonds, while the latter DMSO solvent molecule is hydrogen
bonded to two ureas via two hydrogen bonds. This corroborates
1
A variable-temperature H NMR experiment utilizing the
method of Dougherty et al.21 was undertaken to examine the
thermodynamic parameters involved in anion binding. Upon
an increase in temperature, shifts in the proton resonances of
the porphyrin showed that the binding constant was diminishing,
i.e., that the pocket was becoming less amenable to tight
complexation. The van’t Hoff plot of the 1:1 complex of
porphyrin 4 and TBABr, shown in Figure 4, was used to
determine the enthalpy and entropy of complexation.. The ∆H
of the complexation was quite favorable and the ∆S of
complexation unfavorable, with values of ∆H ) -14.8 kcal/
mol and ∆S ) -32.7 eu. Similar thermodynamic parameters
have been observed with Beer’s neutral ferrocene anion recep-
tor.22 Here, the large enthalpy of complexation associated with
a large decrease in the entropy of complexation is presumably
due to the formation of several hydrogen bonds to the anion
via the urea functional groups and, possibly, a (highly ordered)
solvent molecule assisted interaction (vide infra) as well,
resulting in a large association constant.
1
what was seen in the H NMR in DMSO-d6, where the urea
hydrogens were found very far downfield due to hydrogen
bonding with the solvent. In 7, the coordinated DMSO and
the zinc metal ion are found on the side of the porphyrin plane
opposite the urea pickets. This is in distinct contrast to most
picket fence porphyrin derivatives where the ligand and metal
are found in the pocket. Clearly, the hydrogen-bonded DMSO
molecules provide substantial stabilization of extended urea
pickets.
The ORTEP diagrams of porphyrin 3 (viewed from the top),
complexed to both bromide and chloride anion (3 + TBABr‚
5DMSO and 3 + TBACl‚5DMSO), are detailed in Figures 6
and 7, respectively. Additional views of these two molecules
are given in Figures S3 and S4 of the Supporting Information.
Also presented in the Supporting Information are Figures S5
and S6, which display equivalent information for the related
solvate 3 + TBABr‚3DMSO. The disposition of the urea
functional groups about the porphyrin ring is the same as that
of the metalated species 7. The ORTEP diagrams show the
anion deeply buried in the pocket of the porphyrin receptor,
positioned over one pyrrole. The crystal structures corroborate
the Job plot 1:1 stoichiometric analysis; presumably, the buildup
of more than one negative charge within the neutral porphyrin
receptor is thermodynamically unfavorable. On opposite sides
of the anion are two adjacent ureas that are hydrogen bonded
via the four NH protons, with NH-Br distances between 2.4
and 2.8 Å and NH-Cl distances between 2.3 and 2.7 Å. A
TBA cation lies outside the pocket, and in the case of the 3 +
TBABr‚5DMSO and 3 + TBACl‚5DMSO complexes, a DMSO
molecule is positioned in the center of the pocket. The electron-
deficient sulfur is in near van der Waals contact with the chloride
or bromide anion, suggesting a Coulombic interaction. The 3
+ TBABr‚3DMSO complex shows a DMSO molecule slightly
off-center of the pocket that is positioned such that one of its
methyl groups is close to the anion, precluding a Coulombic
interaction beetween the sulfur and anion.
X-ray Crystal Structures. Crystals of free-base 3 com-
plexed with bromide and chloride anion, as well as uncomplexed
metalloporphyrin 7, suitable for X-ray crystallographic analysis
were grown by slow evaporation from DMSO. The ORTEP
diagram of the metalloporphyrin 7 (viewed from the top) is
detailed in Figure 5 with additional views (Figures S1 and S2)
in the Supporting Information. The crystal data and refinement
results of 7‚5DMSO, 3 + TBABr‚5DMSO, a related solvate 3
+ TBABr‚3DMSO, and 3 + TBACl‚5DMSO are presented in
Table 2. The crystal structure of 7‚5DMSO shows that three
of the urea groups have their NH pointed toward the inside of
the porphyrin macrocycle, and their urea carbonyls are pointed
toward the outside of the ring, while one urea has its carbonyl
and NH more or less orthogonal to the macrocyclic ring, and
there is no self-association of the urea groups. The ureas are
bound to three DMSO molecules, two of which are located
between the pickets on the periphery of the porphyrin and one
(20) Jagessar, R. C.; Burns, D. H.; Scheidt, W. R. Unpublished data.
(21) Stauffer, D. A.; Barrans, R. E.; Dougherty, D. A. J. Org. Chem.
1990, 55, 2762-2767.
(22) Beer, P. D.; Graydon, A. R.; Johnson, O. M.; Smith, D. K. Inorg.
Chem. 1997, 36, 2112-2118.
The X-ray structures in all three characterized species clearly
show that the complexation of the halide anions occurs via four

Ligands for SelectiVe Chloride Anion Complexation
Table 2. Crystal Data and Refinement Results
J. Am. Chem. Soc., Vol. 120, No. 45, 1998 11689
complex
7‚5DMSO
3 + TBABr‚3DMSO
91.32H97.98Br0.83
Cl4N12.83O7S3
1792.10
P1h
130
3 + TBABr‚5DMSO
C94.11H107.25Br0.75
Cl4N12.76O9S5
1923.50
P1h
130
3 + TBACl‚5DMSO
C
empirical form
fw, amu
space group
T, K
C82H78Cl4N12O9S5Zn
1743.03
P1h
130
C98H116Cl5N13O9S5
1957.59
P1h
130
a, Å
b, Å
c, Å
13.442(3)
14.688(2)
21.738(7)
88.54(1)
80.78(1)
77.72(1)
4139
15.349(2)
17.651(2)
18.949(2)
93.92(2)
96.97(1)
115.66(1)
4551.4
14.677(3)
15.853(2)
22.542(2)
98.41(1)
93.93(1)
106.60(1)
4938.6
14.662(1)
15.846(1)
22.527(3)
98.41(2)
93.77(1)
107.03(1)
4917.9
R, deg
â, deg
λ, deg
V, Å3
Z
2
2
2
2
final R indices,
[I > 2σ(I)]
R indices (all data)
R1 ) 0.0823,
wR2 ) 0.1572
R1 ) 0.1524,
wR2 ) 0.1955
1.031
R1 ) 0.0768,
wR2 ) 0.1759
R1 ) 0.1418,
wR2 ) 0.2180
1.016
R1 ) 0.0936,
wR2 ) 0.2276
R1 ) 0.1679,
wR2 ) 0.2841
1.014
R1 ) 0.0821,
wR2 ) 0.1903
R1 ) 0.1361,
wR2 ) 0.2298
1.019
goodness-of-fit on F2
Figure 7. ORTEP diagram illustrating the geometry of the bound Cl-
in the system 3 + TBACl‚5DMSO showing the same information as
given in Figure 6.
Figure 6. ORTEP diagram illustrating the geometry of the bound Br-
in the system 3 + TBABr‚5DMSO. The position of the two DMSO
molecules in the binding pocket are clear. The porphyrin plane is in
the plane of the paper. 50% probability ellipsoids are displayed. The
location of the two inner hydrogen atoms of the free base are also
shown.
hydrogen bonds that originate from two adjacent urea functional
groups. The question can be raised as to why is the selectivity
of this system reversed from other bis-urea systems, where
phosphate is found to be stronger binding than chloride.3a,d
Indeed, to our knowledge, the only urea receptors reported to
date that exhibit selectivity for halides are Reinhoudt’s calix-
arene-urea receptors,3e,f,h where phosphate anion is not com-
petitive because it is too large to bind within the urea cavity.
As an example, the phenyl and xanthene bis-ureas (Figure 8)
of Bu¨hlmann et al.3a,d bind dihydrogen phosphate anion stronger
than halide anions in DMSO, results opposite to those found
with our receptor. CPK model analysis and analysis of the
porphyrin crystal structures show that the two receptors exhibit
only small differences in the spacing between the urea functional
groups. It is to be noted that, unlike Bu¨hlmann’s bis-ureas, the
urea functional groups that sit over the porphyrin ring cannot
be accessed by the anion from below. However, the difference
in structures are not so dramatic that one would anticipate this
reversal of binding affinities. Indeed, the CPK analysis certainly
shows that phosphate should be able to hydrogen bond with
Figure 8. Bu¨hlmann urea receptors bound to dihydrogen phosphate
anion.
the two ureas via four hydrogen-bonding interactions (as is
postulated for Bu¨hlmann’s receptors, Figure 8) unless there is
solVent in the porphyrin pocket. Dihydrogen phosphate anion,
allowed to hydrogen bond in just such a fashion in the porphyrin
pocket, would necessitate the removal of an ordered (i.e.,
hydrogen bonded with a urea) solvent molecule of DMSO,
where the binding of chloride or bromide anion does not.
Consequently, to attain the same effective bonding interactions
(four hydrogen bonds) as the halides, the binding of the larger,
tetrahedral phosphate anion would require an additional expense
in energy due to the desolvation of DMSO from the porphyrin
pocket. This kind of solvent effect would be as likely a
contributor to the overall high affinity the porphyrin receptors

11690 J. Am. Chem. Soc., Vol. 120, No. 45, 1998
Jagessar et al.
exhibit for halide anions as is the shape of the pocket. While
a solid-state structure may not portray the solution-state structure
exactly, there are likely as many, if not more, solvent molecules
associated with the porphyrin in solution then in the crystal
structure; the crystal structure showed the porphyrin receptor
to be highly solvated (vide supra). Therefore, it is likely that
in solution a DMSO molecule is hydrogen bonded to a urea so
as to position it in the porphyrin cavity as seen in the crystal
structure. With the porphyrin-TBACl complex, the ordered
position of the pocket DMSO allows it to solvate the chloride
anion; i.e., the chloride anion and porphyrin receptor do not
need to be fully desolvated for binding to occur. This kind of
favorable solvation effect for the complexation of chloride anion
is similar to the favorable solvation effects discussed by Nolte,23
Collet,24 and Still25 to explain the observed increase in binding
constants of host-guest complexes when desolvation of the host
was unnecessary. It is also reminiscent of the effect, witnessed
by Bonnar-Law and Sanders,26 that an ordered, bound solvent
molecule has on the increase in sugar binding strength of a
steroid-capped porphyrin. Thus, complexation with four hy-
drogen bonds and concomitant continued solvation by DMSO
(both the porphyrin-host and anion-guest) would explain why
chloride anion is so tightly bound to the porphyrin receptors
2-5. Certainly, this type of bonding motif would explain the
rather large favorable enthalpy of complexation and unfavorable
entropy of complexation determined from the variable-temper-
binding to be very favorable and the entropy of complexation
to be unfavorable (∆H ) -14.8 kcal/mol and ∆S ) -32.7 eu).
The X-ray crystal structure of uncomplexed 7 showed the
metalloporphyrin to be highly solvated, with three DMSO
molecules hydrogen bonded to the urea functional groups. The
X-ray crystal structures of 3 complexed with TBABr and TBACl
showed that the halide was buried deep within the urea pickets
and bound to adjacent urea functional groups via four hydrogen
bonds. In all crystal structures, a DMSO molecule was found
in the porphyrin pocket, bound to a urea via two hydrogen
bonds. The DMSO precluded dihydrogen phosphate from the
formation of four hydrogen bonds between two adjacent ureas
without the added expense of desolvation of the porphyrin
pocket. On the other hand, the spherical halides form four
hydrogen bonds while the porphyrin still remains highly
solvated. Additionally, in the case of the 3 + TBABr‚5DMSO
and 3 + TBACl‚5DMSO complexes, the DMSO was positioned
in such a way as to allow an attractive Coulombic interaction
between the anion and the electron-poor sulfur. Thus, the
orientation and nature of the functional groups that make up
the pickets on the porphyrin (i.e., two ureas that bind the anion
and the other two that bind and order solvent) are most
complementary with the spherical halide anions. Further studies
which examine the role that competitive solvents play in the
selectivity of anion recognition are in progress.
1
Experimental Section
ature H NMR experiment. Unfortunately, we were unable to
determine solvent effects, if any, on binding by switching to
other solvents. The use of acetone-d6 in place of DMSO gave
rise to very large binding constants and the same anion
selectivities. Solubility constraints precluded the use of CD3-
OD and the non-hydrogen-bonding solvents CD3Cl or CD2Cl2.
Titration experiments with TBABr and porphyrins 2 and 3 in
88% DMSO-d6/12% D2O showed that the receptors, in this
solvent system, expressed only a slightly higher affinity for
chloride over bromide anion.
Materials and General Procedures. Reagent-grade benzene,
chloroform, and dichloromethane were obtained from Fisher and used
without further purification: methanol was distilled from Mg and stored
over 3 Å molecular sieves. Pyrrole and 2-nitrobenzaldehyde were
obtained from Aldrich. All melting points (Mel-Temp) are uncorrected.
1H NMR spectra were recorded at 300 MHz in DMSO-d6 with Me4Si
as an internal standard unless otherwise specified. Silica gel (Davisil
633) was used for column chromatography, and analytical thin-layer
chromatography (TLC) was performed using precoated Whatman PE
SIL G/UV. UV/vis spectroscopy were carried out on a Shimadzu model
1600 UV/vis spectrophotometer. FAB mass spectra were obtained on
a VG ZAB high-resolution mass spectrometer by Dr. Todd Williams
at the University of Kansas. Elemental analysis was done by Desert
Analytics, Tucson, AZ.
Binding Studies. Titration of a 0.5 mL (5 mM, where no self-
association is observed) solution of the porphyrin receptor in an NMR
tube with stepwise addition of 0-6 equiv of the anions as their
tetrabutylammonium salts was followed by 1H NMR spectroscopy, and
the spectrum was recorded after each addition. Changes in the urea
NH, â-pyrrole, meso-phenyl, and porphyrin NH protons were monitored.
Titration curves were generated by plotting the respective changes in
chemical shifts versus the molar equivalents of anion. The association
constants were determined by nonlinear regression analysis of these
curves using the computer program EQNMR and fitting the data to a
1:1 binding stoichiometric model (determined from molar ratio plots
and Job plot analysis, see below). Qualitative UV/vis anion-binding
studies were carried out by first recording the spectra of the free ligand,
followed by adding the required stoichiometric amounts of the anion
and recording changes in the UV/vis spectrum.
Conclusions
The first neutral porphyrin anion receptors 2-5, with picket
urea functional groups, have been prepared and found to bind
anions with the unique selectivity of Cl- > Br- . H2PO4
>
-
HSO4- > NO3- in DMSO. The K (M-1) of complexation for
Cl- was greater than 105, and the receptors all exhibited a Cl-
binding selectivity of greater than 10:1 with Br-, greater than
-
280:1 with H2PO4-, and greater than 1000:1 with NO3 and
HSO4-. Even in the more competitive solvent system of 88%
DMSO-d6/12% D2O, the porphyrins 2-4 bound the chloride
anion with a K (M-1) ≈ 103. The porphyrin receptors 2-5, 6,
and 8 showed similar binding constants for any one anion,
except for dihydrogen phosphate, where the K (M-1) increased
concomitantly with an increase in the urea-phenyl parasub-
stituent electron-withdrawing capability. Mole ratio and Job
analysis of anion binding showed the stoichiometry of the anions
complexation with porphyrin 4 to all be 1:1, which was
corroborated by crystal structures of the TBACl and TBABr
complexes of porphyrin 3. A van’t Hoff analysis of the
porphyrin 4-TBABr complex determined the enthalpy of
Structure Determinations. Structural analyses for crystals of 7, 3
+ TBABr-3DMSO, 3 + TBABr-5DMSO, and 3 + TBACl-5DMSO
were carried out on an Enraf-Nonius FAST area detector diffractometer
at 130 K by methods and procedures described previously.27 A brief
summary of crystal data and data collection parameters is given in Table
(23) Reek, J. N. H.; Priem, A. H.; Engelkamp, H.; Rowan, A. E.;
Elemans, J. A. A. W.; Nolte, R. J. M. J. Am. Chem. Soc. 1997, 119, 9956-
(27) Scheidt, W. E.; Turowska-Tyrk, I. Inorg. Chem. 1994, 33, 1314.
(28) Programs used in this study included SHELXS-86 (Sheldrick, G.
M. Acta Crystallogr., Sect. A 1990, A46, 467), SHELXL-93 (Sheldrick. G.
M. J. Appl. Crystallogr., manuscript in preparation), and local modifications
of Johnson’s ORTEP2. Scattering factors were taken from International
Tables for Crystallography; Wilson, A. J. C., Ed.; Kluwer Academic
Publishers: Dordrecht, 1992; Vol. C.
9964.
(24) Canceill, J.; Lacombe, L.; Collet, A. J. Am. Chem. Soc. 1986, 108,
4230.
(25) Chapman, K. T.; Still, W. C. J. Am. Chem. Soc. 1989, 111, 3075.
(26) Bonar-Law, R. P.; Sanders, J. K. M. J. Am. Chem. Soc. 1995, 117,
259.

Ligands for SelectiVe Chloride Anion Complexation
J. Am. Chem. Soc., Vol. 120, No. 45, 1998 11691
2. All structures were solved by direct methods with SHELXS-86.28
All structures were refined against F2 with the SHELXL-94 program.28
Nearly all hydrogen atoms were located in difference Fourier syntheses.
In the final refinements, the hydrogen atoms were treated as idealized
riding atoms (C-H ) 0.95 Å for R2CH, 0.98 Å for RCH3). In some
structures, bond length restraints were applied to the solvent molecules
and urea N-H bonds by the use of free variables. In 7, one DMSO
solvent molecule was found to be disordered in a way that the sulfur
and oxygen atoms were located on two sets of positions. For 3 +
TBABr-3DMSO, one butyl group of the tetrabutylammonium cation
was found to be disordered so that two carbon atom were located on
two sets of positions. In 3 + TBABr-5DMSO and 3 + TBACl-
5DMSO, which were found to be isomorphous, two DMSO solvent
molecules were found to be disordered such that the sulfur atom was
located on both sides of a plane defined by the oxygen and the two
methyl carbon atoms. Finally, in the crystals of the two bromide
complexes, the apparent occupancy of the bromide ion was found to
be less than full occupancy. This was treated as a variable in the least-
squares refinements, and the occupancy of the TBA counterion was
constrained to remain equal to that of bromide anion. We believe that
the reduced occupancy of Br- resulted from the exposure of the
crystallizing solution to ambient humidity. All data sets were corrected
for the effects of absorption using the programs DIFABS.29 Final
atomic coordinates for all four structures are available as Supporting
Information.
Job Plots. Job plots were determined by UV/vis spectroscopy using
a method adapted from Tran-Thi and Lipskier.19 A stock solution of
0.1 M TBAPF6 in dichloromethane was prepared along with separate
solutions of the porphyrin and anion of a known molarity in 0.1 M
TBAPF6 in dichloromethane. The appropriate molar ratio of porphyrin
and anion was added to a sample tube and the solution made up to a
constant volume with the stock solution of 0.1 M TBAPF6. UV/vis
spectra were recorded at the 10 different molar concentrations of
porphyrin and anion, while the total concentration of porphyrin and
anion remained the same throughout. The absorption maxima of the
Soret and Q-bands measured at a given wavelength for mixtures with
various ratios of porphyrin and anion were used in the expression:
of zinc(II) acetate dihydrate (0.05 g, 0.23 mmol) dissolved in methanol
(5 mL). The resulting solution was stirred under nitrogen for 24 h.
Solvents were removed in Vacuo, leaving a red residue. Water (10
mL) was added and the mixture stirred for 30 min, then filtered. The
crude red solid was purified by silica gel column chromatography (CH2-
Cl2/ethyl acetate, 8:1 or 4:1) followed by recrystallization from CH2-
Cl2/hexane, leaving a red powder. Yields varied between 90 and 95%.
(r,r,r,r)-5,10,15,20-Tetrakis(2-(phenylurea)phenyl)porphyrin (2).
Yield, 70%; mp 225-229 °C; UV/vis CH2Cl2 λ max (lnꢀ): 424 (11.97),
518 (8.55), 550 (8.8), 584 (8.22), 650 (7.86); 1H NMR δ - 1.5 (s, 2H),
6.66-6.69 (m, 12H), 6.88-6.89 (s, 8H), 7.38 (t, 4H, J ) 7.23 Hz),
7.66 (br s, 4H), 7.73-7.80 (m, 8H), 8.36 (d, 4H, J ) 8.79 Hz), 8.45
(br s, 4H), 8.76 (s, 8H); 13C NMR δ 115.69, 117.58, 121.51, 121.62,
122.02, 128.42, 129.09, 131.22, 131.47, 135.36, 138.91, 139.13,
152.490; HRFAB MS mz calcd for C72H54N12O4 1151.4469, found
1151.4446 (M + 1); Anal. Calcd for C72H54N12O4‚CHCl3: C, 69.00;
H, 4.36; N, 13.22. Found: C, 69.35; H, 4.75; N, 12.86.
Zn(II) (r,r,r,r)-5,10,15,20-tetrakis(2-(phenylurea)phenyl)por-
phyrin (6). Yield, 95%; UV/vis: CH2Cl2 λ max (lnꢀ): 426 (9.52),
562 (7.99), 513 (6.44); 1H NMR δ 6.62 (m, 12H), 6.98 (d, 8H), 7.2 (br
s, 4H), 7.41 (t, 4H, J ) 9.24 Hz), 7.65-7.82 (m, 8H), 8.3 (br s 4H),
8.45 (d, 4H) 8.71 (s, 8H); HRFAB MS mz calcd for C72H52N12O4Zn
1213.3604, found 1213.3557 (M + 1).
(r,r,r,r)-5,10,15,20-Tetrakis(2-(4-chlorophenylurea)phenyl)por-
phyrin (3). Yield, 76%; mp 220-224 °C; UV/vis: CH2Cl2 λ max
1
(lnꢀ): 424 (11.64), 517 (8.90), 592 (8.04), 650 (7.72); H NMR δ -
2.66 (br s 2H), 6.95 (s, 16H), 7.42 (t, 4H; J ) 7.1 Hz), 7.42 (t, 4H, J
) 7.95 Hz;), 7.65 (br s 4H), 7.70-7.81 (m, 8H) 8.37 (d, 4H J ) 8.4
Hz), 8.51 (br s, 4H), 8.78 (s, 8H); 13C NMR δ 109.5, 115.54, 119.02,
121.64, 122.36, 125.08, 128.23, 129.12, 131.20, 131.47, 135.28, 138.13,
138.93, 152.39; HRFAB MS mz calcd for C72H50N12O4Cl4 1287.2910,
found 1287.2876 (M + 1); Anal. Calcd for C72H50N12O4Cl4‚CHCl3:
C, 62.25; H, 3.64; N, 11.93. Found C, 63.04; H, 4.20; N, 11.90.
Zn(II) (r,r,r,r)-5,10,15,20-tetrakis(2-(4-chlorophenylurea)phen-
yl)porphyrin (7). Yield: 95%; mp 210 - 213 °C; UV/vis (CH2Cl2/
1
benzonitrile) λ max (lnꢀ): 428 (14.60), 475 (11.79), 556 (11.73); H
NMR δ 6.90 (s, 16H), 7.39 (t, 4H, J ) 7.44 Hz), 7.56 (br s, 4H),
7.74-7.81 (m, 8H), 8.41 (d, 4H, J ) 7.92 Hz), 8.66 (s, 8H), 8.92 (br
s, 4H); HRFAB MS mz calcd for C72H48N12O4Cl4Zn 1349.2045, found
1349.2098 (M+1); Anal. Calcd for C72H48N12O4Cl4.Zn‚CHCl3: C,
59.63; H, 3.49; N, 11.43; Found C, 59.77; H, 4.02; N, 10.46.
F(x) ) d(x) + xꢀp - ꢀp
where x is the mole fraction of anion, ꢀp is the molar absorptivity of
the porphyrin, and d(x) is the actual optical density of the solution
divided by the total concentration of porphyrin. Thus, F(x) represents
the deviation from additivity of the absorption mixture due to complex
formation. Job plots were obtained by plotting F(x) as x was
continuously varied.
(r,r,r,r)-5,10,15,20-Tetrakis(2-(4-fluorophenylurea)phenyl)por-
phyrin (4). Yield, 80%; mp 240 °C, dec; UV/vis (CH2Cl2) λ max
(lnꢀ): 424 (11.98), 518 (9.12), 553 (9.24), 593 (8.07); 1H NMR δ -2.63
(br s, 2H) 6.57-6.62 (m, 8H), 6.89-6.92 (m, 8H), 7.52 (br s 4H) 7.54
(t, 4H, J ) 7.65 Hz), 7.82-8.05 (m, 8H), 8.35 (d, 4H, J ) 8.4 Hz),
8.36 (br s 4H), 8.78 (s, 8H); 13C NMR δ 115.21, 115.50, 116.17, 119.78,
122.30, 122.78, 129.56, 131.76, 132.45, 135.86, 139.16, 153.04, 155.85,
159.01; HRFAB MS mz calcd for C72H50N12O4F4 1223.4092, found
1223.4054 (M + 1); Anal. Calcd for C72H50N12O4F4‚0.5CHCl3: C,
67.87; H, 3.96; N, 13.10. Found C, 68.52; H, 3.79; N, 12.58.
General Procedure for the Preparation of Free Base and
Metalated (r,r,r,r)-5,10,15,20-Tetrakis (2-(arylurea)phenyl)por-
phyrins. Compounds 2-5 were prepared and purified in an analogous
fashion. To the R,R,R,R isomer of H2TAPP (0.27 g, 0.4 mmol) in
chloroform (25 mL) at room temperature was added 4 equiv of the
requisite isocyanate (1.6 mmol) in a dropwise fashion via an addition
funnel over a period of 1 h. The mixture was stirred at room
temperature and the progress of the reaction monitored by TLC
analyses. When the reaction was complete (usually after 24 h), the
mixture was taken up in H2O (25 mL) and the organic layer separated,
washed with brine (24 mL), and dried over Na2SO4. Solvents were
removed in vacuo, and the crude product was purified by silica gel
column chromatography (CH2Cl2/ethyl acetate 8:1; with this solvent
system the mobility of the porphyrins were phenyl 2 > chloro 3 >
fluoro 4 > nitro 5). Further purification was done by recrystallization
from dichloromethane/hexane, producing purple microcrystalline solids
in yields ranging from 70 to 85%.
Zn(II) (r,r,r,r)-5,10,15,20-tetrakis(2-(4-fluorophenylurea)phen-
yl)porphyrin (8). Yield, 95%; mp 230-233 °C; UV/vis (CH2Cl2) λ
1
max (lnꢀ): 429 (12.09), 559 (9.05), 594 (67.50); H NMR δ 6.54-
6.60 (m; 8H), 6.74-6.76 (m, 8H), 7.34 (br s, 4H), 7.38 (t, 4H, J )
7.08 Hz) 7.70-7.87 (m, 8H), 8.47 (d, 4H, J ) 8.7 Hz), 8.65 (s, 8H),
8.80 (br s 4H); HRFAB MS mz calcd for C72H48N12O4F4.Zn 1285.3227,
found 1285.3269 (M + 1); Anal. Calcd for C72H48N12O4F4‚Zn‚
CHCl3: C, 62.36; H, 3.51; N, 11.95. Found C, 62.80; H, 3.32; N,
10.88.
(r,r,r,r)-5,10,15,20-Tetrakis(2-(4-nitrophenylurea)phenyl)por-
phyrin (5). Yield, 78%; mp 225-229 °C; UV/vis: CH2Cl2 λ max
(lnꢀ): 423 (11.87), 515 (9.59), 585 (8.50), 640 (7.51); 1H NMR δ -1.3
(s, 2H), 7.06-7.11 (m, 16H), 7.4 (t, 4H, J ) 6.84 Hz), 7.70-7.82 (m,
8H), 8.1 (s, br, 4H), 8.36 (d, 4H, J ) 8.25 Hz), 8.78 (s, 8H), 9.08 (br
s 4H); 13C NMR δ 115.88, 117.33, 118.53, 122.18, 122.56, 125.082,
125.60, 129.69, 132.2, 135.8, 139.25, 141.24, 146.24 152.46, 175.1;
HRFAB MS mz calcd for C72H50N16O12. 1331.3872, found 1331.3850
(M + 1); Anal. Calcd for C72H50N16O12‚CHCl3: C, 60.44; H, 3.54;
N, 15.44. Found C, 60.66; H, 3.86; N, 14.38.
The zinc(II) complexes were also prepared in an analogous fashion.
Typically, the requisite free-base porphyrin (4.1 × 10-6 mol) was taken
up in dichloromethane (10 mL) followed by the addition of a solution
(29) The process is based on an adaptation of the DIFABS30 logic to
area detector geometry by Karaulov: Karaulov, A. I. School of Chemistry
and Applied Chemistry, University of Wales, College of Cardiff, Cardiff
CF1 3TB, U.K., personal communication.
(30) Walker, N. P.; Stuart, D. Acta Crystallogr., Sect. A 1983, A39, 158.

11692 J. Am. Chem. Soc., Vol. 120, No. 45, 1998
Jagessar et al.
Acknowledgment. D.H.B. wishes to gratefully acknowledge
the support of this work by the National Science Foundation
(EPS-9550487), and W.R.S. wishes to gratefully acknowledge
the support of this work by the NIH (GM-28401).
diagrams for the 3 + TBABr‚3DMSO solvate, Tables S1-S24
with complete crystallographic details, atomic coordinates,
anisotropic temperature factors, fixed hydrogen atom positions,
bond distances, and bond angles for 7‚5DMSO, 3 + TBABr‚
5DMSO, 3 + TBACl‚5DMSO, and 3 + TBABr‚3DMSO (107
pages, print/PDF). See any current masthead page for ordering
information and Web access instructions.
Supporting Information Available: Figures S1 and S2
presenting additional views of 7‚5DMSO, Figures S3 and S4
giving side views of 3 + TBABr‚5DMSO and 3 + TBACl‚
5DMSO, respectively, and Figures S5 and S6 giving ORTEP
JA982052I