Polyphosphorylated Triphenylenes: Synthesis, Crystal Structure, and Selective Catechol Recognition

Article abstract of DOI:10.1021/jo802015k

Designed as a multivalent hydrogen bond acceptor, new receptors, Discopus 1a,b, were built from a triphenylene core surrounded by six (diaryl)phosphinate groups. An efficient synthesis was developed to prepare these elaborated structures in a high overall

Full text of DOI:10.1021/jo802015k

Polyphosphorylated Triphenylenes: Synthesis, Crystal Structure,
and Selective Catechol Recognition
Ce´cile Givelet,† Bernard Tinant,‡ Luc Van Meervelt,§ Thierry Buffeteau,†
Nathalie Marchand-Geneste,† and Brigitte Bibal*,†
UniVersite´ de Bordeaux, Institut des Sciences Mole´culaires, CNRS UMR 5255, 351 cours de la Libe´ration,
33405 Talence, France
b.bibal@ism.u-bordeaux1.fr
ReceiVed September 12, 2008
Designed as a multivalent hydrogen bond acceptor, new receptors, Discopus 1a,b, were built from a
triphenylene core surrounded by six (diaryl)phosphinate groups. An efficient synthesis was developed to
prepare these elaborated structures in a high overall yield. The X-ray structure of receptor 1b showed
strong cooperative hydrogen bonds with two water molecules and intermolecular CH-π contacts. In
chloroform, Discopus 1a,b displayed recognition properties toward dihydroxybenzenes, selectively forming
complexes with catechol derivatives 4a-c in a 1:2 (host:guest) stoichiometry. According to NMR and
microcalorimetry titrations, association constants were found in the 30-2837 M-1 range, which were
larger than those reported for curvated catechol receptors (14-120 M-1). Interestingly, Discopus present
two distinct catechol binding sites. Weak hydrogen bonding between host phosphinates and guest hydroxyl
groups was shown by infrared spectroscopy and 31P NMR. Molecular dynamics simulations and recognition
experiments suggested that a stronger hydrogen bond assisted by a π-interaction between the Discopus
core and one catechol molecule could exist within the 1:2 complex.
Introduction
ammoniums in the gas phase,3 solution,4 and solid state.5
Tetraphosphonated receptors were also efficient metallic cations
extractors,6 and di- or monophosphonated cavitands selectively
detected alcohols and amines in the gas phase.7 In these cases,
In molecular recognition, neutral polar P(V)dO groups are
well-known for accepting hydrogen bonds, mainly forming
complexes with ammoniums and metallic cations.1 To improve
weak PdO associations, multiple phosphoryl functions (phos-
phonate, phosphinate, and phosphine oxide) were largely
introduced in macrocycles, clips, and cavitands to take advan-
tages of preorganization and cooperativity.2 In particular,
polyphosphorylated resorcin[4]arenes successfully recognized
(2) (a) Caminade, A.-M.; Majoral, J.-P. Chem. ReV. 1994, 94, 1183–1213.
(b) Schrader, T. J. Inclusion Phenom. Mol. Recognit. Chem. 1999, 34, 117–
129. (c) Cherenok, S.; Dutasta, J.-P.; Kalchenko, V. Curr. Org. Chem. 2006,
10, 2307–2331. (d) Mlynarz, P.; Ruzinska, E.; Berlicki, L.; Kafarski, P. Curr.
Org. Chem. 2007, 11, 1593–1609. (e) Dutasta, J.-P. Top. Curr. Chem. 2004,
232, 55–91.
(3) Gaz phase (a) Irico, A.; Vincenti, M.; Dalcanale, E. Chem. Eur. J. 2001,
7, 2034–2042. (b) Kalenius, E.; Moiani, D.; Dalcanale, E.; Vainiotalo, P. Chem.
Commun. 2007, 3865–3867.
(4) Liquid: (a) Lippman, T.; Wilde, H.; Dalcanale, E.; Mavilla, L.; Mann,
G.; Heyer, U. J. Org. Chem. 1995, 60, 235–242. (b) Delangle, P.; Dutasta, J.-P.
Tetrahedron Lett. 1995, 36, 9325–9328. (c) Biavardi, E.; Battistini, G.; Montalti,
M.; Yebeutchou, R. M.; Prodi, L.; Dalcanale, E. Chem. Commun. 2008, 1638–
1640.
(5) Solid state: (a) De Zorzi, R.; Dubessy, B.; Mulatier, J.-C.; Geremia, S.;
Randaccio, L.; Dutasta, J.-P. J. Org. Chem. 2007, 72, 4528–4531.
(6) (a) Delangle, P.; Mulatier, J.-C.; Tinant, B.; Declercq, J.-P.; Dutasta, J.-
P. Eur. J. Org. Chem. 2001, 19, 3695–3704. (b) Bibal, B.; Tinant, B.; Declercq,
J.-P.; Dutasta, J.-P. Supramol. Chem. 2003, 15, 25–32.
† Universite´ de Bordeaux.
‡ Universite´ Catholique de Louvain, Laboratoire de cristallographie, Unite´
CSTR, 1, Place Pasteur, B-1348 Louvain-La-Neuve, Belgium.
§ Department of Chemistry, Katholieke Universiteit Leuven, Celestijnenlaan
200F, B-3001 Leuven, Belgium.
(1) (a) Corbridge, D. E. C. Phosphorus, an Outline of Its Chemistry,
Biochemistry and Technology, 4th ed.; Elsevier: Amsterdam, 1990; Chapter 14.1.
(b) Kaplan, L. J.; Weisman, G. R.; Cram, D. J. J. Org. Chem. 1979, 44, 2226–
2233. (c) Savage, P. B.; Holmgren, S. K.; Gellman, S. H. J. Am. Chem. Soc.
1993, 115, 7900–7901. (d) Savage, P. B.; Holmgren, S. K.; Gellman, S. H. J. Am.
Chem. Soc. 1994, 116, 4069–4070.
652 J. Org. Chem. 2009, 74, 652–659
10.1021/jo802015k CCC: $40.75 2009 American Chemical Society
Published on Web 12/02/2008

Polyphosphorylated Triphenylenes
SCHEME 1. Molecular Structure and Synthesis of Discopus
1a,b
FIGURE 1. Model receptor 3 and aromatic targets 4.
core participation and multivalency effects, a simplier com-
pound, 1,2-bis(diarylphosphinate) benzene 3, will be studied as
a model receptor and compared to Discopus binding abilities
(Figure 1).
receptors abilities benefit from partial or total guests encapsula-
tion into host hydrophobic cavities. Few examples of hydrogen
bonding between neutral species and mono- or polyphospho-
rylated receptors in solution have been reported.8
Results and Discussion
Synthesis. 2,3,6,7,10,11-Hexahydroxytriphenylene 2 was
synthesized using a two-step procedure12 in 77% overall yield.
Hexaphosphorylation of compound 2 was then achieved in
anhydrous conditions using commercial chloro(diphenyl)phos-
phine oxide or freshly prepared chloro-di(p-methoxybenzene)-
phosphine oxide13 in the presence of triethylamine in THF
(Scheme 1). After workup, pure compounds 1a and 1b were
isolated in 63% and 61% yield, respectively. Discopus 1a was
soluble only in chlorinated solvents, whereas Discopus 1b was
soluble in most common organic solvents. Both 1H NMR spectra
were characteristic of hexasubstituted triphenylenes, displaying
a single signal for core aromatic protons.14 31P NMR analysis
also revealed that the six diarylphosphinate groups are equivalent
(singlet at δ31P (CDCl3) 32.53 ppm for 1a and 33.28 ppm for
1b). The absorption spectra of both compounds in chloroform
showed a maximum at 274 nm (see Supporting Information),
which was comparable to the behavior of hexa(n-alkyloxy)-
triphenylenes15 in chlorinated solvents. Elaborated structures
1a,b were thus consistent with classical hexasubstituted triph-
enylenes. The same synthetic procedure was successfully used
for the preparation of model 3 (60% yield from catechol).
X-ray Analysis.16 Single crystals suitable for X-ray analysis
of Discopus 1b were grown from the slow evaporation of a
n-pentane/diethylether solution. The stoichiometry observed in
the crystal structure is 1b·Et2O·2H2O with one-half of the
complex in the asymmetric part of the unit cell (Figure 2). The
deviation from the best mean plane through the eighteen core
carbon atoms is only 0.021 Å, showing that the polyaromatic
core is not distorted by the bulky substituents (Figure 2a). Bonds
P1dO and P2dO are located under the core plane and make
an angle of, respectively, 1.64° and 1.73° with this plane,
As natural and estrogenic (multi)phenols are attractive
targets,9 we were interested in new hydrogen bond acceptor
architectures. To investigate such aromatic targets, our strategy
was to combine several phosphinate patterns with a polyaromatic
core. Indeed, polyaromatic compounds are attractive building
blocks, combining synthetic and structural advantages: easy
preparation, low-cost material, poly peripheral substituted
positions, hindered regions, and potential π-interaction as well
as photophysical properties. 2,3,6,7,10,11-Hexasubstituted triph-
enylenes were the ideal candidates to evaluate polyphosphory-
lated substituents in terms of cooperative and selective binding.
We designed Discopus 1a,b as new unpreorganized structures
displaying a nanosized discotic surface (Disco) with six
peripheral substituents (Scheme 1). This bulky phosphinate
outside rim (pus) can prevent core self-aggregation and favor
H-bonding. To the best of our knowledge, this is the first time
that neutral phosphoryl groups are incorporated to a non-
curvated or non-macrocyclic architecture for recognition
purpose.10,11
Herein this paper presents Discopus 1a,b synthesis from tri-
phenylene 2 and their binding properties toward aromatic targets
and aromatic hydrogen bond donor guests 4 (Figure 1). To evaluate
(7) Pinalli, R.; Nachtigall, F. F.; Ugozzoli, F.; Dalcanale, E. Angew. Chem.,
Int. Ed. 1999, 38, 2377–2380. (c) Pinalli, R.; Suman, M.; Dalcanale, E. Eur. J.
Org. Chem. 2004, 22, 451–462. (d) Pirondini, L.; Dalcanale, E. Chem. Soc. ReV.
2007, 36, 695–706.
(8) (a) Segura, M.; Bricoli, B.; Casnati, A.; Mun˜oz, E. M.; Sansone, F.;
Ungaro, R.; Vincent, C. J. Org. Chem. 2003, 68, 6296–6303. (b) Friedrichsen,
B. P.; Whitlock, H. W. J. Am. Chem. Soc. 1989, 111, 9132–9134. (c) Friedrichsen,
B. P.; Whitlock, H. W. J. Am. Chem. Soc. 1990, 112, 8931–8941.
(9) (a) Wiskur, S. L.; Anslyn, E. V. J. Am. Chem. Soc. 2001, 123, 10109–
10110. (b) ten Cate, M. G.; Reinhoudt, D. N.; Crego-Calama, M. J. Org. Chem.
2005, 70, 8443–8453. (c) Preziosi, P. Pure Appl. Chem. 1998, 70, 1617–1631.
(10) Substituted (metallo)porphyrins analogues as flat molecular receptors:
(a) Ogoshi, H.; Mizutani, T.; Hayashi, T.; Kuroda, Y. In The Porphyrin
Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds.; Academic Press:
New York, 2000; Vol. 6, pp 279-340. (b) Weiss, J. J. Inclusion Phenom.
Macrocyclic Chem. 2001, 40, 1–22, and references therein. (c) Sessler, J. L.;
Camiolo, S.; Gale, P. A. Coord. Chem. ReV. 2003, 240, 17–55. (d) Even, P.;
Boitrel, P. Coord. Chem. ReV. 2006, 250, 519–541.
(11) Triphenylene-based receptors: (a) O’Leary, B. M.; Grotzfeld, R. M.;
Rebek, J., Jr. J. Am. Chem. Soc. 1997, 119, 11701–11702. (b) Waldvogel, S. R.;
Wartini, A. R.; Rasmussen, P. H.; Rebek, J. Tetrahedron Lett. 1999, 40, 3515–
3518. (c) Waldvogel, S. R.; Fro¨hlich, R.; Schalley, C. A. Angew. Chem., Int.
Ed. 2000, 39, 2472–2475. (d) Schopohl, M. C.; Siering, C.; Kataeva, O.;
Waldvogel, S. R. Angew. Chem., Int. Ed. 2003, 42, 2620–2623. (e) Bomkamp,
M.; Siering, C.; Landrock, K.; Stephan, H.; Fro¨hlich, R.; Waldvogel, S. R. Chem.
Eur. J. 2007, 13, 3724–3732, and references therein. (f) Fyfe, M. C. T.; Lowe,
J. N.; Stoddart, J. F.; Williams, D. J. Org. Lett. 2000, 2, 1221–1224. (g) Badjic,
J. D.; Cantrill, S. J.; Stoddart, J. F. J. Am. Chem. Soc. 2004, 126, 2288–2289.
(12) (a) Beattie, D. R.; Hindmarsh, P.; Goodby, J. W.; Haslam, S. D.;
Richardson, R. M. J. Mater. Chem. 1992, 2, 1261–1266. (b) Zniber, R.; Achour,
R.; Cherkaoui, M. Z.; Donnio, B.; Gehringer, L.; Guillon, D. J. Mater. Chem.
2002, 12, 2208–2213.
(13) (a) Harger, M. J. P.; Westlake, S. Tetrahedron 1982, 38, 1511–1516.
(b) Clive, D. L. J.; Kang, S. J. Org. Chem. 2001, 66, 6083–6091.
(14) (a) Boden, N.; Borner, R. C.; Bushby, R. J.; Cammidge, A. N.;
Jesudason, M. V. Liq. Cryst. 1993, 15, 851–858. (b) Perez, D.; Guitia`n, E. Chem.
Soc. ReV. 2004, 33, 274–283.
(15) (a) Markovitsi, D.; Germain, A.; Millie´, P.; Le´cuyer, P.; Gallos, L. K.;
Argyrakis, P.; Bengs, H.; Ringsdorf, H. J. Phys. Chem. 1995, 99, 1005–1017,
and references therein. (b) Balzani, V.; Clemente-Leo´n, M.; Credi, A.; Lowe,
J. N.; Badjic, J. D.; Stoddart, J. F.; Williams, D. J. Chem. Eur. J. 2003, 9, 5348–
5360. (c) Li, Z.; Lucas, N. T.; Wang, Z.; Zhu, D. J. Org. Chem. 2007, 72, 3917–
3920.
(16) All measures were done with Mercury 1.4.2 program from Cambridge
Crystallographic Data Centre, U.K. Macrae, C. F.; Edgington, P. R.; McCabe,
P.; Pidcock, E.; Shields, G. P.; Taylor, R.; Towler, M.; van de Streek, J. J. Appl.
Crystallogr. 2006, 39, 453–457.
J. Org. Chem. Vol. 74, No. 2, 2009 653

Givelet et al.
In the crystal, Discopus molecules are located in parallel
planes and their cores are distant by about 13.5 Å (intercores
distances C5-C5 13.57 Å, C4-C4 13.01 Å, C7-C7 14.57 Å;
see Supporting Information). From the Cambridge Structural
Database, crystal structures of hexasubstituted triphenylenes,19
which present no core π-stacking, have intermolecular cores
distances ranging from 15.91 to 17.73 Å. Thus inter-Discopus
distances are a little bit shorter as a result of π-π and CH-π
short contacts between molecules. Indeed, two classical π-stack-
ings were observed between p-methoxyphenyl groups: an
intramolecular edge-to-face one between C43 and centroid
C31-C36 (purple center in Figure 2b, 3.668 Å) as well as an
intermolecular one between centroids C17-C22 and C31-C36
(respectively blue to yellow centers distance, 3.640 Å). More-
over, several intermolecular CH-π short contacts were high-
lighted between Discopus and other molecules (Figure 2b and
Supporting Information). A molecule of Et2O is pinched between
two diarylphosphinates substituents as the distance between C52
(Et2O) and centroid C17-C22 (blue center in Discopus) is 3.731
Å. The triphenylene core is also close to a methoxy group from
another Discopus molecule. The distance between centroid
C7-C9 (black center) and C16 is 3.776 Å, which is typical of
intermolecular CH-π contacts (3.7-3.8 Å as found in pro-
teins).20 These multiple CH-π contacts contribute to fulfill
empty π-surfaces but cannot be interpreted as CH-π interac-
tions driving the crystal packing. In the literature, CH-π contact
examples involving triphenylenes are rare. Bushby et al. reported
a crystal structure of an hepta-substituted triphenylene showing
partial π-overlapping (Car · · ·Car 3.72-3.79 Å) and also CH-π
contacts (CH · · ·Car 3.38-3.89 Å).19 In the solid state, Discopus
1b was inclined to make hydrogen bonds along with CH-π
short contacts.
Guest Affinities in Solution. Proton NMR dilution experi-
ments were conducted with compounds 1a,b in chloroform.
Critical aggregate concentrations were 6 mM for 1a and 5 mM
for 1b. Titrations were thus achieved with a 1 mM host
concentration. Phosphorus and proton NMR signals monitoring
was carried out using Discopus 1a,b and model 3 (1 mM) in
the presence of relevant guests 4 (0 to 10-25 mM). Sufficiently
soluble in chloroform, targets 4 were chosen for their various
electron-density or H-bonding abilities (Figure 1). Proton NMR
observations were achieved on guest signals because changes
in Discopus spectra were not significant.
Electron-poor (1,3,5-trinitrobenzene) and unsubstituted aro-
matics (benzene, naphthalene), as well as electron-rich benzene
derivatives (1,2-dimethoxybenzene, 1,3,5-trimethoxybenzene),
showed no π-π interaction with hosts 1a,b and 3. In contrast
to polyaromatic-based tweezers,21 Discopus 1a,b do not interact
with electronic and size complementary aromatic compounds.
FIGURE 2. X-ray crystal structure of Discopus 1b·Et2O·2H2O. (a)
One molecule of 1b·Et2O·2H2O in stick representation. (b) View of
intramolecular and intermolecular π-π and CH-π short contacts for
one molecule of 1b (capped stick). For clarity only partners moieties
(ball and stick) are shown and interactions are indicated in green dotted
lines.
TABLE 1. Hydrogen Bonds (D-H· · ·A) Geometry: Distance (d,
Å) and Angle (<DHA>, deg)
D–Ha
PdAa
d(D–H) D(H· · ·A) <DHA> d(D· · ·A)
O27–H27A P2dO20 1.07(2)
O27–H27B P1dO17 0.99(2)
1.75(2)
1.79(2)
178.6(7)
172.4(8)
2.811(3)
2.781(3)
a For atom numbering see Supporting Information.
whereas for P3dO this angle is 1.76° above the plane.
Interestingly, positions of the six diarylphosphinate substituents
are mainly governed by hydrogen bonds with water (involving
four PdO groups, Figure 2a) and CH-π short contacts (Figure
2b).
(18) Kariuki, B. M.; Harris, K. D. M.; Philp, D.; Robinson, J. M. A. J. Am.
Chem. Soc. 1997, 119, 12679–12680.
Indeed, two molecules of water are involved in cooperative
hydrogen bonds with phosphinate groups (Figure 2a). Each
water molecule is pinched between two phosphinate groups
located in the same triphenylene bay region. Distances
PdO· · ·OH2 · · ·OdP are 2.78 and 2.81 Å, respectively, and
H-bonding angles are closed to 180° (Table 1). These strong
hydrogen bonds17 are classical for phosphine oxide derivatives,18
but to the best of our knowledge, these are the first example of
a water molecule cooperative binder.
(19) (a) Boden, N.; Bushby, R. J.; Jesudason, M. V.; Sheldrick, B. Chem.
Commun. 1988, 1342–1343. (b) Bushby, R. J.; Boden, N.; Kilner, C. A.; Lozman,
O. R.; Lu, Z.; Liu, Q.; Thornton-Pett, M. A. J. Mater. Chem. 2003, 13, 470–
474.
(20) (a) Nishio, M.; Hirota, M.; Umezawa, Y. In The CH/π Interaction.
EVidence, Nature and Consequences; Wiley-VCH: New-York, 1998. (b) Taka-
hashi, H.; Tsuboyama, S.; Umezawa, Y.; Honda, K.; Nishio, M. Tetrahedron
2000, 56, 6185–6191. (c) Takemura, H.; Iwanaga, T.; Shinmyozu, T., Tetrahedron
Lett. 2005, 46, 6687–6690. (d) Brandl, M.; Weiss, M. S.; Jabs, A.; Su¨hnel, J.;
Hilgenfeld, R. J. Mol. Biol. 2001, 307, 357–377.
(21) (a) Zimmerman, S. C.; Van Zyl, C. M.; Hamilton, G. S. J. Am. Chem.
Soc. 1989, 111, 1373–1381, and references therein. (b) Kla¨rner, F.-G.; Kahlert,
B. Acc. Chem. Res. 2003, 36, 919–932. (c) Branchi, B.; Balzani, V.; Ceroni, P.;
Campan˜a Kuchenbrandt, M.; Kla¨rner, F.-G.; Bla¨ser, D.; Boese, R. J. Org. Chem.
2008, 73, 5839–5851, and references therein.
(17) Desiraju, G. R.; Steiner, T. In The Weak Hydrogen Bond in Structural
Chemistry and Biology; Oxford University Press: New York, 1999.
654 J. Org. Chem. Vol. 74, No. 2, 2009

Polyphosphorylated Triphenylenes
SCHEME 2. Structures of Catechols 4a-c That Form
Host-Guest Complexes with Discopus 1a,b and Model 3
In the presence of hydrogen bond donors, polyphosphinated
hosts had uneven performance. A selective interaction appeared
with 1,2-dihydroxybenzene (catechol) moieties (catechol 4a, 1,2-
dihydroxynaphthalene 4b, and tert-butylcatechol 4c, Scheme 2)
and 1,3-dihydroxybenzene (resorcinol), but there was no binding
with other aromatic hydrogen bond donors such as phenol
derivatives (phenol, 1-hydroxy-2-methoxybenzene), aniline
derivatives (aniline, 1,2-diaminobenzene, 1,3-diaminobenzene),
and thiol derivatives (thiobenzene, 1,2-dithiobenzene).
These preliminary results suggested that one or more hydro-
gen bond(s) may exist between polyphosphinated Discopus and
dihydroxybenzene derivatives. This weak interaction was not
observed with phenols as the strength of hydrogen bonding is
dependent upon donor-acceptor cooperativity and donor acidity,
which follows the OH > NH > SH series.22 Hydrogen bond
strength is also depending on solvent polarity and might be
absent in the presence of competiting solvents.23 Widely soluble,
Discopus 1b and guest 4b associations were evaluated in organic
solvents by the same protocol. In good hydrogen bond acceptor
solvents (DMSO, acetone, and THF), guest 4b was not
complexed by the receptor. In non-competitive solvents (CHCl3,
CH2Cl2, and acetonitrile), 1b:4b association was present through
significant NMR signal shifts. These observations are in
accordance with hydrogen bonding between Discopus and
catechol moieties.
To assess this point, we investigated the association between
1a,b and 4a by infrared spectroscopy. This technique also
proved to be sensitive to hydrogen bonding involving PdO (or
CdO) and phenol partners.24,25
Infrared Studies. As a preliminary study, infrared spectra
of catechol 4a were recorded to assign OH stretching and self-
association bands (Figure 3). Compound 4a exhibited two strong
bands at 3600 and 3558 cm-1, with similar intensities in the
OH stretching region.
The band observed at 3600 cm-1 can be ascribed to the
stretching vibration of free hydroxyl groups (νOHf), whereas
the band at 3558 cm-1 has been assigned to the stretching
vibration of hydroxyl groups that formed intramolecular hy-
drogen bonds (νOHb intra).26
As results for Discopus 1a are similar (see Support-
ing Information), only the study of 1b is reported. In CDCl3,
the infrared spectra of receptor 1b and catechol 4a, as well as
the mixture 1b:4a in the 1:2 and 1:10 ratios, were recorded in
FIGURE 3. IR spectra (CDCl3) of catechol 4a at concentrations of 5
mM (black line) and 25 mM (red line) in the 3800-3200 cm-1 region.
the 3800-2600 cm-1 (Figure 4a) and 1700-950 cm-1 (Figure
4b) regions, respectively.
Spectrum of Discopus 1b in the 3800-2600 cm-1 region was
essentially dominated by the aromatic and aliphatic C-H
stretching vibrations. The IR spectrum for the mixture 1b:4a
in the 1:2 ratio was not the simple addition of the two spectra
of 1b (at a concentration of 2.5 mM) and 4a (at a concentration
of 5 mM), indicating that interactions occurred between the two
molecules. A very broadband appears around 3200 cm-1
,
whereas the intensities of the two bands associated to the
stretching vibrations of the free (at 3600 cm-1) and associated
(at 3558 cm-1) hydroxyl groups decreased significantly. This
behavior can be easily explained considering hydrogen bonds
between catechol and receptor. The observed frequency around
3200 cm-1 proved that weak hydrogen bonds occurred, as
expected for phenols.27 The intensity of the broadband was
multiplied approximately by a factor 2 in the IR spectrum of
1b:4a in the 1:10 ratio, which was consistent with a weak
association constant between the two compounds. A broad
shoulder appeared around 3380 cm-1, which was certainly
associated to the νOHb inter mode of catechol.
The IR spectra recorded in the 1700-950 cm-1 region
(Figure 4b) for a mixture 1b:4a in the 1:2 and 1:10 ratios
were almost the addition of the two spectra of 1b and 4a,
except in the 1240-1190 cm-1 region where stretching
vibration of the diarylphosphinate group (PdO) is expected.28
Indeed, when the concentration of 4a increased, the intensity
of the band at 1228 cm-1 decreased, whereas the intensity of
the band at 1198 cm-1 increased. This last band did not appear
in the IR spectrum of 1b and can be assigned to the stretching
vibration of phosphinate groups involved in hydrogen bonds.
The splitting between the stretching vibration of free (1228
cm-1) and associated (1198 cm-1) phosphinate groups were in
agreement with those determined for hydrogen-bonded com-
plexes between phenol and phosphoryl compounds.22
(22) (a) Abraham, M. H.; Grellier, P. L.; Prior, D. V.; Duce, P. P.; Morris,
J. J.; Taylor, P. J. J. Chem. Soc., Perkin Trans. II 1989, 699–711. (b) Abraham,
M. H.; Duce, P. P.; Prior, D. V.; Barratt, D. G.; Morris, J. J.; Taylor, P. J. J. Chem.
Soc., Perkin Trans. II 1989, 1355–1375. (c) Abraham, M. H. Chem. Soc. ReV.
1993, 73–83.
(23) Reichardt, C. In SolVents and SolVent Effects in Organic Chemistry,
2nd ed.; VCH: Weinheim, 1990.
(24) Modro, A. M.; Modro, T. A. Can. J. Chem. 1999, 77, 890–894.
(25) (a) Sijbesma, R. P.; Kentgens, A. P. M.; Lutz, E. T. G.; Van der Maas,
J. H.; Nolte, R. J. M. J. Am. Chem. Soc. 1993, 115, 8999–9005. (b) 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, 9956–9964.
(26) Kjaergaard, H. G.; Howard, D. L.; Schofield, D. P.; Robinson, T. W.;
Ishiuchi, S.; Fujii, M. J. Phys. Chem. A 2002, 106, 258–266.
(27) Novak, A. In Structure and Bonding; Dunitz, J. D., Hemmerich, P.,
Holm, R. H., Ibers, J. A., Jorgensen, C. K., Neilands, J. B., Reinen, D., Williams,
R. J. P., Eds.; Springer-Verlag: Berlin, 1974; Vol. 18, pp 177-216.
(28) Socrates, G. In Infrared Characteristic Group Frequencies; Wiley
Interscience: New York, 1980.
J. Org. Chem. Vol. 74, No. 2, 2009 655

Givelet et al.
TABLE 2. Binding Constants of Phosphorylated Receptors 1a,b
and 3 toward Catechol Derivatives in CDCl3 at 298 K
a b
,
binding constants Kai (M-1
)
guest
model 3
Discopus 1a Discopus 1b
4a
92 (88)c
1407; 70 (109)c
2834; 85 (84)c
2949;d 73d
4b
4c
n.d.
n.d.
1441; 30 (51)c
1863; 97 (114)c
1887;d 189d
n.d.
144; 67 (65)c
a Monitoring guest protons by NMR. b Experiments were run twice at
least and (15% errors were estimated. c From 31P NMR titration.
d From microcalorimetric titration with (5% errors.
(Ka) ranged from 14 to 120 M-1. In the same solvent, curvated
receptors for resorcinol derivatives or dihydroxynaphthalenes
proved to be more efficient (Ka from 30 to 1 × 10+5 M-1), using
cooperative hydrogen bonds and eventually additional π-π
stacking.31 To quantify binding properties of receptors 1a,b and
3, Job plots and titrations toward catechols 4a-c were performed
32,33
1
in CDCl3 using H and 31P NMR monitoring (Table 2).
1
Both H and 31P NMR titrations demonstrated that model 3
forms a 1:1 complex with guest 4a whose association constant
is 90 ( 2 M-1. Infrared study in CDCl3 showed that catechol
function was involved in binding (decreasing intensity around
3600 cm-1 and increasing intensity around 3200 cm-1), but no
evidence of PdO implication was seen in the 1240-1190 cm-1
region due to signal overlap (see Supporting Information).
Nevertheless, 31P NMR proved to be sensitive to this association,
which probably is hydrogen bonding between host phosphinates
and guest catechol function. As expected, both Discopus
structures had recognition properties different from those of 3
(Figure 5).
Interestingly, Discopus 1a,b are associated to target 4a in
1
a 1:2 (host:guest) complex allowing H NMR monitoring
(Table 2). The corresponding association constants Ka1 and Ka2
are 1407 and 70 M-1 for 1a and slightly higher for 1b, 2834
M-1 and 85 M-1, respectively. A first catechol is thus strongly
recognized by Discopus 1a,b and a second catechol is com-
plexed as weakly as in the case of host 3. Moreover, the second
association is also observed by 31P NMR titrations (1:1 complex,
confirming Ka2 values). Only the second catechol binding site
(Ka2) involves phosphinate groups as probably hydrogen bond
acceptors. Hence Discopus 1a,b are ditopic receptors for
catechol 4a which was confirmed by microcalorimetric titrations
with 1b (sequential binding model, see Supporting Information).
Concerning the better binding ability of 1b compared to 1a,
methoxy substituents in 1b might offer potent hydrogen bonds
acceptors sites to 4a.
FIGURE 4. IR spectra (CDCl3) of Discopus 1b (black) and catechol
4a (red), as well as the mixture 1:4a in the 1:2 (blue) and 1:10 (green)
ratios, recorded in (a) the 3800-2600 cm-1 region and (b) the
1700-950 cm-1 region.
Infrared studies undoubtedly proved that catechol 4a makes
hydrogen bonds with Discopus, especially with PdO acceptors,
but other acceptors may participate in these weak associations.
Binding Studies. The quest for selective binding is one of
the key issues in developing new molecular receptors. As
previously observed, compounds 1a,b showed affinity for
catechol moieties versus phenol, aniline, and thiobenzene ones.
Few examples of catechol receptors in chloroform have been
reported.29,30 Mainly based on carbonyl (ester, amide, urea)
hydrogen bond acceptors, their associations with catechol
derivatives were weak in chloroform, and association constants
(30) Catechol receptors in aqueous medium: (a) Wiskur, S. L.; Lavigne, J. J.;
Metzger, A.; Tobey, S. L.; Lynch, V.; Anslyn, E. V. Chem. Eur. J. 2004, 10,
3792–3804. (b) Diaz, S. G.; Lynch, V.; Anslyn, E. V. J. Supramol. Chem. 2002,
2, 201–209. (c) Imada, T.; Kijima, H.; Takeuchi, M.; Shinkai, S. Tetrahedron
1996, 52, 2817–2826. (d) Bernardo, A. R.; Stoddart, J. F.; Kaifer, A. E. J. Am.
Chem. Soc. 1992, 114, 10624–10631. (e) Kimura, E.; Watanabe, A.; Kodama,
M. J. Am. Chem. Soc. 1983, 105, 2063–2066.
(31) Resorcinol and phenols recognition in organic solvents: (a) Murray,
B. A.; Whelan, G. S. Pure Appl. Chem. 1996, 68, 1561–1567. (b) Mirzoian, A.;
Kaifer, A. E. J. Org. Chem. 1995, 60, 8093–8095. (c) Reference 25b.
(32) NMR titrations were fitted using HypNMR2006, a commercial program
from Protonic Software. Frassineti, C.; Ghetti, S.; Gans, P.; Sabatini, A.; Moruzzi,
M. S.; Vacca, A. Anal. Biochem. 1995, 231, 374–382.
(33) NMR titrations were also run between Discopus 1b and resorcinol.
Proton NMR data were fitted with a (1:2) complex whose association constants
were Ka1 ) 687 M-1 and Ka2 ) 100 M-1. Phosphorus NMR data could not be
fitted in model (1:1) or (1:2) with sufficient confidence. In the case of resorcinol,
phosphinates groups involvement in recognition is more complex than in 1b:4a
complexes.
(29) Catechol and dihydroxynaphthalene receptors in chloroform: (a) Sijbes-
ma, R. P.; Nolte, R. J. M. J. Org. Chem. 1991, 56, 3122–3124. (b) Sijbesma,
R. P.; Kentgens, A. P. M.; Nolte, R. J. M. J. Org. Chem. 1991, 56, 3199–3141.
(c) Reference 25a. (d) Kawai, H.; Katoono, R.; Nishimura, K.; Matsuda, S.;
Fujiwara, K.; Tsuji, T.; Suzuki, T. J. Am. Chem. Soc. 2004, 126, 5034–5035.
(e) Cho, Y. L.; Uh, H.; Chang, S.-Y.; Chang, H.-Y.; Choi, M.-G.; Shin, I.; Jeong,
K.-S. J. Am. Chem. Soc. 2001, 123, 1258–1259.
656 J. Org. Chem. Vol. 74, No. 2, 2009

Polyphosphorylated Triphenylenes
FIGURE 5. 1H NMR titrations of hosts 1a (green diamonds), 1b (blue squares), and 3 (red triangles) toward catechol 4a (0-25 mM, CDCl3).
Inset: 31P NMR titrations of receptors 1a,b and 3 toward 4a.
Catechol 4b was also recognized by Discopus 1a,b with the
same 1:2 (host:guest) stoichiometry as 4a. Again, Discopus 1b
was a better ligand than 1a. On one hand, guest 4b had similar
association constants (Ka1 ) 1441 M-1 and Ka2 ) 30 M-1) as
4a with 1a, and on the other hand, guest 4b was less associated
(Ka1 ) 1863 M-1 and Ka2 ) 97 M-1) than 4a with 1b (confirmed
by microcalorimetric titrations). In all cases, Discopus 1b was
the best receptor.
To explain the stronger Ka1 values compared to the Ka2 ones,
we speculated that hydrogen bonding could occur with oxygens
located at the triphenylene periphery and thus it could benefit
from supplementary π-π interaction between the receptor large
π-core and catechol. Bulky guest 4c was titrated by Discopus
1b in order to evaluate the effect of steric hindrance while OH
functions are available for binding. From NMR titrations, target
4c forms 1:2 (host:guest) complex with 1b, as observed for 4a,b
compounds, but association constants were lower than those
measured between 4a,b and the same host (Table 2). Ka1
dramatically decreased to 144 M-1, while Ka2 was slightly lower
(65 M-1). Thus steric hindrance disordered the first step of
complexation while the second step (PdO· · ·H-OAr hydrogen
bond) was just affected by electronic effect. These observations
are in accordance with the hypothesis that the first adduct may
rely on an H-bond assisted by π-π interaction.
Further investigation was undertaken with molecular model-
ing to elucidate binding sites within the 1:2 association.
Molecular Dynamics Simulations. Molecular dynamics
(MD) involves the simulation of the dynamic motions of the
complex atoms using the differential equations of motion with
the molecular mechanics CVFF force field through a series of
infinitesimal time increments. A molecular modeling study was
carried out to model a 3D structure of 3:4a and 1b:4a complexes
and to get insight into major molecular interactions. A geometry
optimization of each complex was followed by MD simulations.
During MD simulation, model 3 had several types of
hydrogen bonds with catechol 4a. Hydrogen bonding acceptors
were mainly phosphinate groups PdO, triphenylene out-rim
oxygens, and to a less extend, arylmethoxy oxygens (Figure
6). These weak interactions could be multiple within the
complex (Figure 6b). This modeling is in agreement with
experimental observations using NMR and infrared spectroscopies.
Concerning the MD simulation of 1b in the presence of two
molecules 4a, it is noteworthy that one catechol remains over
FIGURE 6. Snapshots of MD simulation between model 3 and 4a.
(a) Arylmethoxy oxygen atom as hydrogen bond acceptor. (b) Phos-
phinate groups and triphenylene-linked oxygen atoms as hydrogen bond
acceptor.
the triphenylene core of Discopus 1b as a result of π-π
stacking, as the distance between aromatic moieties is about
3.57 Å (Figure 7). This catechol is also involved in a hydrogen
bond network with three different oxygen atoms (P ) O, OMe,
and triphenylene-linked oxygens) (see Supporting Information).
Additionnal interaction between catechol and methoxy groups
with complex 1b:4a may explain that, experimentally, Discopus
1b is a better ligand than 1a, which is not fitted with OMe
substituents.
J. Org. Chem. Vol. 74, No. 2, 2009 657

Givelet et al.
in vacuo to remove oxalyl chloride excess (repeated three times).
The residual yellow oil (5.340 g, 99%) was used without further
purification: 1H NMR (300 MHz, CDCl3, 295 K) δ 7.76 ppm (dd,
3JHH ) 8.7 Hz, 3JPH ) 12.3 Hz, 4H, H2), 6.98 (dd, 3JHH ) 8.7 Hz,
4JPH ) 3.4 Hz, 4H, H3), 3.83 (s, 6H, H5); 13C NMR (75.5 MHz,
4
2
CDCl3, 295 K) δ 163.3 (d, JPC ) 3.3 Hz, C4), 133.0 (d, JPC
)
13.2 Hz, C2), 124.6 (d, 1JPC ) 131.2 Hz, C1), 114.2 (d, 3JPC ) 15.9
Hz, C3), 55.5 (C5); 31P NMR (120 MHz, CDCl3, 295 K) δ 45.6
ppm (s).
2,3,6,7,10,11-Hexa(diphenylphosphinate)triphenylene 1a. To a
solution of 2,3,6,7,10,11-hexahydroxytriphenylene 2 (0.649 g, 2
mmol) and triethylamine (8 mL, 20 mmol) in dry tetrahydrofuran
(100 mL) at 0 °C was added dropwise a solution of diphenylphos-
phoric acid chloride (3.90 mL, 20 mmol) in dry tetrahydrofuran
(10 mL). The reaction mixture was stirred at reflux overnight.
Distilled water (200 mL) and dichloromethane (150 mL) were added
at room temperature. After decantation, the aqueous phase was
extracted with dichloromethane (2 × 200 mL). The combined
organic phases were washed with distilled water (2 × 400 mL)
and brine (2 × 300 mL), dried over magnesium sulfate, filtered,
and concentrated in vacuo. The beige foam was suspended in hot
dichloromethane (100 mL), filtered, and washed with dichlo-
romethane (3 × 5 mL). The filtrate was concentrated in vacuo and
submitted to the same suspension-washing process twice. After
column chromatography (dichloromethane/tetrahydrofuran 85:15),
compound 1a was isolated as a white solid (1.921 g, 63% yield):
Rf 0.24 (dichloromethane/tetrahydrofuran 85:15); mp 160 °C (dec);
IR (NaCl) ν 3004, 2856, 1593, 1503, 1465, 1419, 1210, 1126, 1002
cm-1; UV-vis (CHCl3, 5 × 10-6 M) λmax (ε) 319 (9360), 299
(24650), 274 (119794), 266 (89255), 249 nm (34800 mol-1 L cm-1);
1H NMR (300 MHz, CDCl3, 295 K) δ 8.12 (s, 6H, Hb), 7.48-7.94
FIGURE 7. Snapshot of MD simulation where two guests 4a (in
yellow) make hydrogen bonds with phosphinate groups and π-π
interactions with the triphenylene core of Discopus 1b.
The MD experiment also showed that the second 4a
compound moved outside the polyaromatic core, establishing
hydrogen bonds with PdO moieties and triphenylene-linked
oxygens. These intermolecular hydrogen bonds were conserved
during the dynamics, although the phosphinate groups involved
are different. Moreover, the dynamics simulation revealed that
intramolecular hydrogen bonds between hydroxyl groups of 4a
occurred frequently.
Hence, MD simulations for complex 1b:4a (in a 1:2 ratio)
corroborate experiments, showing that Discopus makes multiple
hydrogen bonds with 4a within two distinct binding sites. One
catechol makes hydrogen bonds at the Discopus periphery,
whereas the other one is located above triphenylene core as a
result of hydrogen bonds and π-interaction. Nevertheless, MD
predicted that both catechol molecules may interact with PdO
groups, which is in contradiction with experience: 31P NMR
proved that only one molecule of 4a was complexed by
phosphinate functions.
3
3
(dd, JHH ) 8.3 Hz, JPH ) 12.5 Hz, 24H, He), 7.33 (m, 36H, Hf
and Hg); 13C NMR (75.5 MHz, CDCl3, 295 K) δ 141.8 (d, 2JPC
)
2
6.04 Hz, Cc), 132.6 (d, Cg), 132.0 (d, JPC ) 10.6 Hz, Ce), 130.3
(d, 1JPC ) 137.0 Hz, Cd), 128.7 (d, 3JPC ) 13.6 Hz, Cf), 125.9 (Ca),
116.0 (Cb); 31P NMR (120 MHz, CDCl3, 295 K) δ 32.53 (s); LRMS
(LSIMS) m/z (%) 1525 (35, [M + H]+), 1547 (100, [M + Na]+);
HRMS (LSIMS) m/z calcd for C90H66012P6Na ([M + Na]+)
1547.282722, found 1547.287760.
Modeling and spectroscopic experiments were thus comple-
mentary to establish the structure of Discopus-catechol
associations.
2,3,6,7,10,11-Hexa(bis(4-methoxyphenyl)phosphinate)triph-
enylene 1b. To a solution of compound 2 (0.649 g, 2 mmol) and
triethylamine (8 mL, 20 mmol) in dry tetrahydrofuran (100 mL) at
0 °C was added dropwise a solution of freshly prepared bis-(4-
methoxyphenyl) phosphoric acid chloride (5.340 g, 18 mmol) in
tetrahydrofuran (10 mL). The reaction mixture was stirred at room
temperature overnight. Distilled water (100 mL) and dichlo-
romethane (150 mL) were added. After decantation, the aqueous
phase was extracted with dichloromethane (2 × 200 mL). The
combined organic phases were washed with distilled water (2 ×
400 mL) and brine (2 × 300 mL), dried over magnesium sulfate,
filtered, and concentrated in vacuo. The beige foam was dissolved
with ethyl acetate (200 mL). After precipitation, the suspension
was filtered and dried under vacuum. Compound 1b was isolated
as a white solid (2.305 g, 61%): Rf 0.23 (dichloromethane/
tetrahydrofuran 8:2); mp 156 °C (dec); IR (NaCl) ν 2972, 2836,
1589, 1513, 1454, 1418, 1255, 1137, 1013 cm-1; UV-vis (CHCl3,
5 × 10-6 M) λmax (ε) 319 (9360), 299 (24278), 274 (136653), 264
Conclusion
In summary, we have demonstrated that Discopus 1a,b can
be easily prepared in efficient overall yields. X-ray analysis
revealed that conformation of 1b, in the solid state, is governed
by PdO H-bonding to water and intermolecular CH-π short
contacts. In chloroform solution, these unpreorganized receptors
selectively interact with catechol guests in a 1:2 (host:guest)
stoichiometry. Discopus 1a,b also proved to be ditopic, allowing
distinct hydrogen bonds with, on one hand, classical phosphinate
group acceptors, as observed by infrared spectroscopy, and on
the other hand, a stronger H-bond assisted by π-interaction with
triphenylene core, relying on MD simulations. As selective
catechol receptors are needed, efforts are undertaken toward
modified phosphinated Discopus to enhance association constants.
1
(124812), 249 nm (154089 mol-1 L cm-1); H NMR (300 MHz,
CDCl3, 295 K) δ 8.06 (s, 6H, Hb), 7.87 (dd, 3JHH ) 8.7 Hz, 3JPH
)
Experimental Section
3
4
12.3 Hz, 24H, He), 6.89 (dd, JHH ) 8.7 Hz, JPH ) 2.7 Hz, 24H,
Compound 2 was prepared from commercial veratrole according
to literature procedures.12
Hf), 3.79 (s, 36H, Hh); 13C NMR (75.5 MHz, CDCl3, 295 K) δ
4
2
162.7 (d, JPC ) 3.02 Hz, Cg), 141.8 (Cc), 133.9 (d, JPC ) 12.1
Bis(4-methoxyphenyl) Phosphoric Acid Chloride.13 To a suspen-
sion of commercial bis(4-methoxyphenyl) phosphoric acid (5 g,
18 mmol) in dry dichloromethane (100 mL) at 0 °C was added
dropwise oxalyl chloride (9 mL, 45 mmol) under inert conditions.
The mixture was stirred at 0 °C for 20 min and at room temperature
overnight. The crude liquid was concentrated under reduced
pressure. Dry dichloromethane (100 mL) was added and evaporated
Hz, Ce), 125.8 (Ca), 121.9 (d, JPC ) 145.7 Hz, Cd), 116.0 (Cb),
1
3
114.2 (d, JPC ) 14.8 Hz, Cf), 55.2 (Ch); 31P NMR (120 MHz,
CDCl3, 295 K) δ 33.28 (s); LRMS (MALDI) m/z (%) 1885 (60,
[M + H]+), 1908 (100, [M + Na]+); HRMS (LSIMS) m/z calcd
for C102H90024P6 ([M + H]+) 1885.438065, found 1885.432610.
1,2-Di((bis-dimethoxyphenyl)phosphinate)benzene 3. The same
procedure as for Discopus 1a synthesis was followed using a
658 J. Org. Chem. Vol. 74, No. 2, 2009

Polyphosphorylated Triphenylenes
solution of catechol (220.2 mg, 2 mmol) and triethylamine (840
µL, 6.2 mmol) in dry THF (40 mL) and a solution of freshly
prepared bis(4-methoxyphenyl) phosphoric acid chloride (1.828 g,
6.2 mmol) in dry THF (3 mL). After reaction and extraction, the
crude product was column chromatographed on silica gel (ethyl
acetate/petroleum ether 9:1) and compound 3 was isolated as a white
solid (756 mg, 60%): Rf ) 0.25 (ethylacetate/petroleum ether 9:1);
mp 72-74 °C (dec); IR (NaCl) ν 2932, 2840, 1597, 1502, 1454,
1409, 1230, 1129, 1025 cm-1; UV-vis (CHCl3, 1.3 × 10-5 M)
λmax (ε) 280 (3932), 247 (43707), 240 nm (40098 mol-1 L cm-1);
1H NMR (250.13 MHz, CDCl3, 295 K) δ 7.78 (dd, 3JHH ) 8.7 Hz,
was added to the X-ray structure of Discopus 1b. The 3:4a (1:1
ratio) and 1b:4a (1:2 ratio) structures were subjected to energy
minimization using steepest descent followed by a conjugated
gradient until the rms gradient was less than 0.001 kcal mol-1 Å-1
.
The consistent valence force field (CVFF) in the Discover molecular
mechanics module of InsightII (Accelrys Inc.) was chosen for the
evaluation of the potential energy function and atom typing due to
its applicability in handling a wide range of organic systems. Cross-
terms were used in the energy expression and a Morse potential
was used to model the valence bond stretching term. Molecular
dynamics simulations were performed with a target temperature of
300 K reached in 20 000 fs (equilibration), while the time step was
1.5 fs. The simulations were carried out using the Verlet leapfrog
algorithm36 and the structures were stored in the computer every
0.5 ps. Molecular dynamics was monitored during 750 ps. All
simulations and visualization were performed with InsightII on a
SGI Octane computer.
NMR Titrations. NMR signal monitoring was conducted on hosts
1a,b and 3 (31P singulet) as well as guests 4a-c (protons). All
solutions were freshly prepared, and deuterated solvents were dried
over molecular sieves. A solution (250 µL) of host (∼2 mM) was
introduced in each NMR tube (12-15 experiments per titration).
Increasing aliquots of guest stock solution (∼50 mM) were added,
and the total volume (500 µL) was adjusted with CDCl3. The
titration data (∆δ ppm versus guest or host concentration) were
fitted using the nonlinear curve-fitting procedure with either a (1:
1) binding equation developed by Wilcox37 using Origin 7.0
software or a (1:2) model using HypNMR2006 program.
Microcalorimetric Titrations. Isothermal titration calorimetry
(ITC) experiments were conducted at 295 K in chloroform to allow
a direct comparison with NMR studies. Host 1b and guest solutions
4a or 4b were freshly prepared using HPLC grade solvent dried
over molecular sieves. Stock host solution (0.742 mM, CHCl3) was
placed into ITC cell (V ) 1.4192 mL). Aliquots (4 µL) of guest
solution (∼ 30 mM) were added via a computer-automated injector
at 135 s intervals. Heat changes were recorded after each addition.
Heats of dilution were measured by a blank experiment (in absence
of receptor) under the same conditions and they were subtracted
from the titration data prior to curve fitting. Additionally, an initial
2 µL injection was discarded from each data set in order to remove
the effect of titrant diffusion across the syringe tip during the
equilibration process. Titrations curves were fitting with the
sequential binding sites model (with two nonidentical binding sites)
using Origin v. 5.0 software. Experiments were duplicated at least
and data were averaged to be compared to NMR results.
3
3JPH ) 12.1 Hz, 8H, He), 7.32 (m, 2H, Hb), 6.87 (dd, JHH ) 8.9
4
Hz, JPH ) 2.8 Hz, 10H, Hf and Ha), 3.81 (s, 12H, Hh); 13C NMR
(62.9 MHz, CDCl3, 295 K) δ 162.76 (d, 4JPC ) 2.9 Hz, Cg), 141.07
(d, 2JPC ) 7.6 Hz, Cc), 133.78 (d, 2JPC ) 12.4 Hz, Ce), 124.71 (Ca),
1
3
122.37 (d, JPC ) 145.9 Hz, Cd), 121.61 (d, JPC ) 3.8 Hz, Cb),
114.12 (d, 3JPC ) 14.3 Hz, Cf), 55.33 (Ch); 31P NMR (120.1 MHz,
CDCl3, 295 K) δ 32.76 (s); LRMS (MALDI) m/z (%) 631 (90, [M
+ H]+), 653 (10, [M + Na]+); HRMS (LSIMS) m/z calcd for
C
102H90024P6 653.14784, found 653.14647.
X-ray Crystallographic Data for Discopus 1b ·Et2O·2H2O. A
crystal of 0.5 × 0.2 × 0.1 mm3 approximate dimensions was
mounted in a loop with mineral oil and transferred immediately in
a nitrogen cold gas stream for flash cooling. The data were collected
at 100 K with a Bruker SMART 6000 CCD area-detector diffrac-
tometer using Cu KR (λ ) 1.54178 Å) radiation. The unit cell
parameters were refined using all the collected spots after the
integration process. The redundancy in data allowed a semiempirical
absorption correction (SADABS version 2.03)34 to be applied on
the basis of multiple measurements of equivalent reflections. Data
reduction was accomplished with SAINT version 6.02a.34 The
structure was solved by direct methods and refined by full-matrix
least-squares on F2 using SHELXL97.35 All non-hydrogen atoms
were refined with anisotropic temperature factors. The hydrogen
atoms of the water molecule were localized by Fourier difference.
The other hydrogen atoms were calculated with AFIX and included
in the refinement with a common isotropic temperature factor. They
were allowed to ride on their parent atoms.
Crystal data for 1b·Et2O·2H2O: monoclinic, space group C2/c,
a ) 25.4815(8), b ) 21.1342(8), c ) 19.4278(7) Å, ꢀ )
111.556(3)°, V ) 9730.6 Å3, Z ) 4, Dx ) 1.36 g cm-3, 48318
total reflections collected, 9212 independent reflections of which
7667 were considered as observed (I > 2σ(I)). Data were collected
up to 2θmax ) 144°, completeness 96.5%, 640 parameters, R1 )
0049, wR2 ) 0.1273, S ) 1.027. Largest peak and hole in final
residual electron density: 0.675 and - 0.379 e Å-3. Selected
distances in Discopus 1b crystal are reported in Supporting
Information. CCDC-685480 contains the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgment. This work was supported by the French
Ministry of Research (C.G. Fellowship) and the Centre National
de la Recherche Franc¸aise. The authors thank Dr. Clarisse
Maechling from the Faculte´ de Pharmacie de Strasbourg, Illkirch
(France) for helping C.G. in the microcalorimetry experiments.
FTIR Measurements. Infrared spectra of Discopus 1a,b and
catechol 4a, as well as a mixture of 1a (or 1b) + 4a, were recorded
with a ThermoNicolet Nexus 670 FTIR spectrometer at a resolution
of 4 cm-1, by coadding 50 scans. Samples were held in a variable
path length cell with BaF2 windows. Each spectra of 1a, 1b, and
4a were measured in CDCl3 at a concentration of 2.5 mM and at
a path length of 250 µm. Additional spectra of 4a were run at
concentrations of 5 and 25 mM. Spectra of of 1a (or 1b) + 4a
were measured in a 1:2 and 1:10 molar ratios. All infrared spectra
are shown with solvent absorption subtracted out.
Supporting Information Available: Additional distances and
views of crystal 1b·Et2O·2H2O; snapshots of MD simulations
for 3:4a and 1b:4 complexes; UV-vis absorption and NMR
spectra of compounds 1a,b and 3; FTIR spectra of 3 + 4a and
1b + 4a mixtures; microcalorimetric titrations of 1b toward
4a,b; Job plots, 1H and 31P NMR titrations of hosts 1a,b and 3
toward guests 4a-c; and a cif file containing crystallographic
data for 1b·Et2O·2H2O. This material is available free of charge
via the Internet at http://pubs.acs.org.
Molecular Modeling. The molecular structure of catechol 4a was
built with AMPAC 8.16. A geometry optimization was performed
using the AM1 Hamiltonian, and the resulting catechol structure
JO802015K
(34) Bruker, SMART, SADBS and SAINT; Bruker AXS Inc.: Madison, WI,
1997.
(35) Sheldrick, G. M. SHELXS-97 and SHELXL-97, Program for Crystal
Structure Refinement; University of Go¨ttingen:Go¨ttingen, Germany, 1997.
(36) Verlet, L. Phys. ReV. 1967, 159, 98–103.
(37) Wilcox, C. S. In Frontiers in Supramolecular Organic Chemistry and
Photochemistry; Schneider, H.-J., Du¨rr, H., Eds; VCH-Weinheim: Weinheim,
Germany, 1991; pp 123-143.
J. Org. Chem. Vol. 74, No. 2, 2009 659