Synthesis of exceptionally stable iron and ruthenium η1-tetrahedro-tetraphosphorus complexes: Evidence for a strong temperature dependence of M-P4 π back donation

Article abstract of DOI:10.1002/1521-3773(20011015)40:20<3910::AID-ANIE3910>3.0.CO;2-2

At variance with white phosphorus, the most reactive allotrope of the element, which is unstable and ignites spontaneously in air, the new η¹-tetrahedro-tetraphosphorus complexes [Cp*M(PR₃)₂(η¹-P₄]Y (M= Fe, Ru; Y=Cl, PF₆, BPh₄, BAr₄′; Cp=C₅Me₃ the structure of the [Cp*Fe(Ph₂PCH₂CH₂PPh₂) (η¹-P₄)]⁺ ion is depicted) exhibit a surprising and unprecedented thermal stability and an astonishing reluctance to react with oxygen and other oxidants.

Full text of DOI:10.1002/1521-3773(20011015)40:20<3910::AID-ANIE3910>3.0.CO;2-2

COMMUNICATIONS
Reaction of [Cp*M(L2)Cl] (Cp* ˆ h5-C5Me5; M ˆ Fe,[10]
Synthesis of Exceptionally Stable Iron and
Ruthenium h1-tetrahedro-Tetraphosphorus
Complexes: Evidence for a Strong Temperature
Dependence of M P4 p Back Donation**
L2 ˆ Ph2PCH2CH2PPh2;
M ˆ Ru,[11]
L ˆ PEt3,
L2 ˆ
Ph2PCH2CH2PPh2) with one equivalent of P4 in THF at room
temperature under a nitrogen atmosphere leads to removal of
chloride and coordination of the P4 molecule through one
lone pair of electrons on phosphorus (Scheme 1). Addition of
Â
Isaac de los Rios, Jean-Rene Hamon, Paul Hamon,
Claude Lapinte,* Loïc Toupet, Antonio Romerosa, and
Maurizio Peruzzini*
In recent years the coordination chemistry of white
phosphorus has been carefully investigated and a plethora
of transition metal complexes containing Px fragments derived
from either the degradation (P, P2, and P3 ligands) of the P4
tetrahedron or the reaggregation (up to P12) of fragments
thereof have been synthesized and characterized.[1] Consid-
erable effort has been invested in the synthesis of stable
coordination compounds containing the intact P4 molecule,
but in spite of the many attempts and of the existence of an old
example, [{N(CH2CH2PPh2)3}Ni(h1-P4)] (1),[2] it is generally
considered more easy to reductively open one,[3] two,[4] or
three[5] bonds of the P4 tetrahedron than to coordinate the
P4 molecule as an intact h1-end-on ligand. Only recently,
however, h1-P4 compounds have been described either as
thermally unstable species, [{P(CH2CH2PPh2)3}M(h1-P4)]OTf
(M ˆ Co (2), Rh (3); OTf ˆ OSO2CF3)[6] and [M(CO)3-
(PR3)2(h1-P4)] (M ˆ W, R ˆ Cy (4), iPr (5); M ˆ Mo, R ˆ Cy
(6)),[7] or as a slightly air-sensitive solid, [{MeC(CH2P-
Ph2)}Re(CO)2(h1-P4)]Y (Yˆ OTf, BPh4 (7)).[8] In all of these
compounds, which were characterized by either X-ray dif-
fraction (1, 4) or 31P NMR spectroscopy (2 ± 7), the P4
molecule is bonded to the organometallic fragment through
a lone pair of electrons on one of the phosphorus atoms.
Beautiful examples of thermally unstable side-on P4 com-
plexes have recently been reported.[9]
Scheme 1. Synthesis of the P4 ± transition metal complexes 8 ± 10. 8: M ˆ
Fe, L ˆ 1/2 dppe; 9: M ˆ Ru, L ˆ PEt3; 10: M ˆ Ru, L ˆ 1/2 dppe; dppe ˆ
Ph2PCh2CH2PPh2.
ethanol/n-hexane (1:2 v/v) and slow concentration of the
resulting solution under a brisk current of nitrogen affords
reddish purple (M ˆ Fe) or orange (M ˆ Ru) microcrystalline
solids of formula [Cp*M(L2)(h1-P4)]Cl (M ˆ Fe, L2 ˆ
Ph2PCH2CH2PPh2 (8-Cl); M ˆ Ru, L ˆ PEt3 (9-Cl), L2 ˆ
Ph2PCH2CH2PPh2 (10-Cl)) in excellent yields (>80%). Meta-
thesis with NH4PF6, NaBPh4, or NaBAr4 (Arˆ B(CF3)3C6H2)
yields the corresponding hexafluorophosphate or tetraaryl-
borate salts, respectively. Complexes 8 ± 10 are thermally and
air stable in the solid state, and dissolve in most common
organic solvents. The iron derivative slowly decomposes in
halogenated hydrocarbons to give back the chloride adduct,
while the ruthenium salts are readily crystallized from
dichloromethane/ethanol mixtures without decomposition
even under aerobic conditions.
The coordination of the intact P4 molecule as a tetrahedro-
h1-P4 ligand has been ascertained by X-ray diffraction analysis
on crystals of 8-BPh4 grown by slow diffusion of n-pentane
into a saturated solution of the complex in THF.[12] The X-ray
analysis reveals the expected h1-coordination of the tetra-
phosphorus ligand (Figure 1) with the iron atom Fe1 coordi-
nated in a roughly octahedral arrangement by the two dppe
P-donor atoms and the Cp* ring occupying three contiguous
sites of the coordination polyhedron.
We report here on the synthesis and the chemical-physical
characterization of a family of novel h1-tetrahedro-tetraphos-
phorus complexes of iron and ruthenium of formula
[Cp*M(L)2(h1-P4)]Y (L ˆ phosphane or diphosphane ligand;
M ˆ Fe, Ru) which exhibit an exceptional stability in compar-
ison to the few known h1-P4 derivatives as well as to the free
phosphorus molecule. Some preliminary data illustrating the
electronic structure and reactivity of the new tetraphosphorus
complexes are also provided.
The Fe PP distance is 2.1621(7) Š and is considerably
4
shorter than the Fe P distances in the related complexes
prepared by Scherer et al., namely exo,exo-butterfly complex
[*] Dr. M. Peruzzini, Dr. I. de los Rios
Istituto per lo Studio della Stereochimica ed Energetica dei Composti
di Coordinazione, CNR
[{(C5H2tBu3)(CO)Fe}2(m,h1:1-P4)] (dFe ˆ 2.3516 Š)[13] and
Pav
the diiron species [{(C5H3tBu2)Fe}(m,h4:1-P4){Fe(CO)(C5H3-
tBu2)}],[14] in which the exo-cyclic Fe h1-P4 separation
(2.211(1) Š) is still slightly longer than in 8-BPh4. The bond
lengths within the P4 molecule are slightly dissimilar; the basal
P P separation ranges between 2.2033(11) and 2.2363(11) Š
Via J. Nardi 39, 50132 Firenze (Italy)
Fax (‡39)055-2478366
E-mail: peruz@fi.cnr.it
Dr. C. Lapinte, Dr. J.-R. Hamon, Dr. P. Hamon, Dr. L. Toupet
Â
Universite de Rennes 1, UMR CNRS 6509
and hence is longer than the dP
distances which vary
Pbasal
apical
35042 Rennes (France)
from 2.1589(10) to 2.1664(10) Š. This type of deformation of
the coordinated tetraphosphorus ligand is opposite to that
Fax : (‡33)99-28-67-00
E-mail: Claude.Lapinte@univ-rennes1.fr
found in 1 (dP
ˆ 2.20(3) Š; dP
ˆ 2.09(3) Š),[2] but
Prof. Dr. A. Romerosa
Pbasal
Pbasal
apical
basal
Â
Area de Quimica Inorganica, Universidad de Almería
parallels the values reported by Scheer for the tungsten
04071, Almería (Spain)
derivative 4 (av dP
ˆ 2.190 Š; av dP
ˆ 2.172 Š).[7]
Pbasal
Pbasal
apical
basal
[**] This work was supported by INTAS (project 00-00018) and by the EC
through COSTD17 Action (WGD17/003) and the RTN program
(contract HPRM-CT-2000-0010). I. de los Ríos thanks ªMinisterio de
Ciencia y Tecnologiaº (Spain) for a postdoctoral grant.
The P4 tetrahedron in the latter compound is less compressed
than that in 8, probably reflecting the differences in the
distribution of the overall electron density in the two
3910
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3910 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 20

COMMUNICATIONS
from d ˆ 289.22 at 808C to 303.01 at 408C (D ˆ d313K
d193K ˆ 13.79). Neither the PX (D ˆ 2.74) nor the dppe (D ˆ
1.24) doublets exhibit such unusual temperature dependence
of the chemical shift.[16]
The reactivity of the robust tetrahedro-tetraphosphorus
complexes described here is under scrutiny. Preliminary data
indicate that once coordinated to the metal center the P4
molecule is extremely resistant to oxidants, such as meta-
chloroperbenzoic acid, H2O2, or O2, and nucleophiles, such as
water and alcohols, but still endowed with reactivity towards
electrophilic transition metal moieties. Thus, bimetallic clus-
ters incorporating a bridging P4 ligand may be readily
synthesized. As an example the heterobimetallic complexes
[Cp*Fe(dppe)(m,h1:1-P4){Re(CO)2[MeC(CH2PPh2)3]}](BPh4)2,
[Cp*Ru(PEt3)2(m,h1:1-P4){Re(CO)2[MeC(CH2PPh2)3]}]BPh4)2
and [Cp*Ru(dppe)(m,h1:1-P4){W(CO)5}]BPh4 were prepared
by treating 8, 9, and 10 in THF with [{MeC(CH2PPh2)3}-
Re(CO)2(OTf)] and [W(CO)5(thf)], respectively.[17] Remark-
ably, the reaction of 10 with iodine is straightforward and
produces four equivalents of PI3 at room temperature. One
molecule of PI3 is efficiently scavenged by the metal fragment
affording the very rare PI3 metal complex [Cp*Ru(dppe)-
(PI3)]BPh4.[18]
Figure 1. Structure of the complex cation in 8-BPh4 (ORTEP drawing).
Selected bond lengths [Š] and angles [8]: Fe-P1 2.2343(7), Fe-P2 2.2502(7),
Fe-P3 2.1621(7), P3-P4 2.1664(10), P3-P5 2.1615(10), P3-P6 2.1589(10), P4-
*
P6 2.2112(12), P4-P5 2.2363(11), P5-P6 2.2033(11), Fe-Cpcentroid 1.758(2);
P3-Fe-P1 88.62(3), P3-Fe-P2 93.02(3), P1-Fe-P2 85.41(3).
isoelectronic d6 complexes (W0 vs. FeII) and the cationic
charge of 8.[15]
Experimental Section
The 57Fe Mössbauer spectra recorded in the temperature
window between 77 and 350 K exhibit QS parameters typical
for [Cp*Fe(dppe)(L)]Y complexes containing an h1-coordi-
nated L ligand, but quite large isomeric shifts ranging from
General procedure for the synthesis of complexes 8 ± 10: A solution of
[Cp*M(L2)Cl] (M ˆ Fe, L2 ˆ Ph2PCH2CH2PPh2; M ˆ Ru, L ˆ PEt3, L2 ˆ
Ph2PCH2CH2PPh2) (0.50 mmol) in THF (20 mL) was treated at 258C with
a small excess of white phosphorus (0.08 g, 0.64 mmol) under nitrogen.
Addition of ethanol/n-hexane (10 mL, 1:2 v/v) and slow concentration of
the resulting solution under a current of nitrogen gave purple (8) or orange
(9, 10) microcrystals of the h1-P4 complexes [Cp*M(L2)(h1-P4)]Cl (M ˆ Fe,
L2 ˆ Ph2PCH2CH2PPh2 (8-Cl); M ˆ Ru, L ˆ PEt3 (9-Cl), L2 ˆ
Ph2PCH2CH2PPh2 (10-Cl)) in yields higher than 80%. The solid complexes
were filtered under nitrogen and washed twice with ethanol (2 Â 3 mL) and
light petroleum ether (2 Â 5 mL). Metathesis with NaBPh4 in THF/EtOH
afforded the corresponding tetraphenylborate salts.
1
0.31 to 0.40 mms . In the rich family of [Cp*Fe(dppe)]
complexes, d values from 57Fe Mössbauer spectra generally
1
fall in the range between 0.15 and 0.30 mms and increase
when the electron-releasing properties of L decrease. It is also
noteworthy that for 8-BPh4 the d parameter depends on the
temperature; this is rather unusual for iron piano-stool
complexes. The decrease of d with temperature suggests that
the electron density at the metal center increases with
temperature. Accordingly, the p back donation could be
weaker at high temperature.
The P4 molecule remains firmly coordinated to the metal
center in solution; the complexes 8 ± 10 exhibit a temperature-
invariant first-order A2MX3 pattern in the 31P NMR spectra.
The four naked phosphorus atoms form the MX3 part of the
spectrum with the three nonmetalated atoms exhibiting a
doublet at d ꢀ 484 roughly independent of the metal and the
ancillary phosphane ligand(s). In contrast, the unique P atom
of the P4 unit coordinated to the metal (PM) features a quartet
of triplets shifted downfield with respect to the signal for free
P4 (d ˆ 526.9). Remarkably, the iron complex displays the
largest coordination chemical shift (dM ˆ 299.54) suggesting
a considerable perturbation of the electronic density in the P4
tetrahedron and pointing to a strong p back donation from the
metal to the P4 ligand.
8-BPh4: Purple crystals, yield 80%; 31P{1H} NMR (81.01 MHz, [D6]acetone,
208C, H3PO4, A2MX3 pattern): dA ˆ 87.49, dM
ˆ
299.54, dX ˆ 482.12,
J(PA,PM) ˆ 39.1, J(PA,PX) ꢀ 0, J(PM,PX) ˆ 228.9 Hz; 1H NMR (200.13 MHz,
[D6]acetone, 208C, TMS): d ˆ 1.49 (br d, J(H,PM) ˆ 1.8 Hz, 15H;
‡
C5(CH3)5); MS (FAB ‡ , o-nitrophenyl n-octylether): m/z (%): [M ], 713
‡
(4), [M
P4], 589 (100); elemental analysis calcd (%) for C60H59BFeP6: C
69.79, H 5.76; found: C 69.45, H 5.79.
9-BPh4: Orange crystals, yield 86%; 31P{1H} NMR (81.01 MHz, CD2Cl2,
208C, H3PO4, A2MX3 pattern): dA ˆ 20.97, dM
ˆ
332.05, dX ˆ 480.96,
J(PA,PM) ˆ 55.3, J(PA,PX) ꢀ 0, J(PM,PX) ˆ 228.9 Hz; 1H NMR (200.13 MHz,
CD2Cl2, TMS): d ˆ 1.61 (dt, J(H,PM) ˆ 5.1, J(H,PA) ˆ 1.5 Hz, 15H;
C5(CH3)5); 13C{1H} NMR (50.32 MHz, CD2Cl2, 208C, TMS): d ˆ 9.95 (s;
P(CH2CH3)3), 11.23 (s; C5(CH3)5), 22.20 (td, J(C,PA) ˆ 13.3, J(C,PM) ˆ
5 Hz, P(CH2CH3)3), 96.90 (s; C5(CH3)5); elemental analysis calcd (%) for
C46H65BP6Ru: C 60.33, H 7.15; found: C 60.03, H 7.33.
10-BPh4: Orange crystals, yield 90%; 31P{1H} NMR (81.01 MHz, CD2Cl2,
208C, H3PO4, A2MX3 pattern): dA ˆ 70.44, dM
ˆ
308.46, dX ˆ 490.29,
J(PA,PM) ˆ 44.1, J(PA,PX) ꢀ 0, J(PM,PX) ˆ 233.5 Hz; 1H NMR (200.13 MHz,
CD2Cl2, TMS): d ˆ 1.53 (dt, J(H,PM) ˆ 5.0, J(H,PA) ˆ 1.9 Hz, 15H;
C5(CH3)5); 13C{1H} NMR (50.32 MHz, CD2Cl2, 208C, TMS): d ˆ 10.54 (s;
C5(CH3)5), 22.50 (vt, J(C,PA) ˆ 22.3 Hz, CH2), 97.13 (s; C5(CH3)5); MS
‡
‡
(FAB ‡ , o-NPOE): m/z (%): [M ] 759 (13), [M
P4] 635 (100); elemental
Consistent with this intriguing view of the electronic
structure of 8, variable-temperature 31P{1H} NMR measure-
ments show that the high-field PM multiplet experiences a
quite surprising temperature dependence moving steadily
analysis calcd (%) for C60H59BP6Ru: C 66.86, H 5.52; found: C 66.67, H
5.38.
Received: July 6, 2001 [Z17445]
Angew. Chem. Int. Ed. 2001, 40, No. 20
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3911 $ 17.50+.50/0
3911

COMMUNICATIONS
[1] a) K. H. Whitmire, Adv. Organomet. Chem. 1998, 42, 2; b) O. J.
Scherer, Acc. Chem. Res. 1999, 32, 751; c) M. Peruzzini, I. de los Rios,
A. Romerosa, F. Vizza, Eur. J. Inorg. Chem. 2001, 593; d) M. Ehses, A.
Romerosa, M. Peruzzini, Top. Curr. Chem. 2001, 220, 107.
[2] P. Dapporto, S. Midollini, L. Sacconi, Angew. Chem. 1979, 91, 510;
Angew. Chem. Int. Ed. Engl. 1979, 18, 469. The related complex
[{N(CH2CH2PPh2)3}Pd(h1-P4)] was also mentioned, see: P. Dapporto,
L. Sacconi, P. Stoppioni, F. Zanobini, Inorg. Chem. 1981, 20, 3834.
[3] A. P. Ginsberg, W. E. Lindsell, J. Am. Chem. Soc. 1971, 93, 2082.
[4] O. J. Scherer, M. Swarowsky, H. Swarowsky, G. Wolmershäuser,
Organometallics 1989, 8, 841.
[5] F. Cecconi, C. A. Ghilardi, S. Midollini, A. Orlandini, Inorg. Chem.
1986, 25, 1766.
[6] M. Di Vaira, M. Ehses, M. Peruzzini, P. Stoppioni, Eur. J. Inorg. Chem.
2000, 2193.
[7] T. Gröer, G. Baum, M. Scheer, Organometallics 1998, 17, 5916.
[8] M. Peruzzini, L. Marvelli, A. Romerosa, R. Rossi, F. Vizza, F.
Zanobini, Eur. J. Inorg. Chem. 1999, 931.
A Dual Channel Fluorescence Chemosensor for
Anions Involving Intermolecular Excited State
Proton Transfer**
Kihang Choi and Andrew D. Hamilton*
Effective fluorescence chemosensors must convert molec-
ular recognition into changes in fluorescence that are both
highly sensitive and easy to detect. A key issue in sensor
design is the connection of substrate binding in a recognition
domain to photophysical changes in a fluorophore with
optimal sensitivity.[1] In recent years there has been great
interest in anion recognition and sensing, because of their
importance in biological and environmental settings.[2] Many
fluorescence anion sensors utilizing competitive binding,[3]
photo-induced electron transfer,[4] metal-to-ligand charge
transfer,[5] and excimer/exiplex formation[6] mechanisms have
been developed. Surprisingly, the strategy of linking a
fluorophore with emission from an internal charge transfer
(CT) excited state to an ion-binding domain, while widely
used for cation sensing, has been rarely exploited for anion
sensing.[1a, 7] Anion binding close to the fluorophore could
lead to the stabilization of positive charge developed in the
fluorophore excited state and to the opening of another
fluorescence emission channel through intermolecular excited
state proton transfer (ESPT)[8] (Figure 1). Herein we report
the preparation of anion sensors using this strategy and show
that one of them can function as a dual-channel sensor system.
As the anion-binding domain we used macrocycle 1 which
can bind certain anions strongly and selectively through
hydrogen bonding to the three amide NH groups.[9] In a first
design, we attached covalently a fluorophore onto the
periphery of 1. 4-Trifluoromethyl-7-aminocoumarin, which
possesses an excited state where negative charge is transferred
from the nitrogen atom to the coumarin ring,[10] was con-
nected to the monocarboxy derivative of 1 to give 2 (Boc ˆ
tert-butoxycarbonyl).[11, 12] This fluorophore-appended macro-
cycle showed modest changes in the intensity of its fluores-
cence emission upon addition of different anions. The binding
constants were determined by fluorescence titration and
showed high selectivity for tetrahedral anions such as H2PO4
(Table 1), which mirrors the properties of 1. Although this
high selectivity (especially for phosphate over chloride) is
desirable for sensing applications,[2a,e] the changes in emission
intensity and wavelength (‡2 nm for H2PO4 ) induced by
anion binding were small. It appears that the negative charge
on the bound anion is not effectively positioned to interact
with the increasing positive charge on the nitrogen atom of
the fluorophore, and has little effect on the excited-state
energy.[13]
[9] I. Krossing, J. Am. Chem. Soc. 2001, 123, 4603.
[10] C. Roger, P. Hamon, L. Toupet, H. Rabaa, J.-Y. Saillard, J.-R. Hamon,
C. Lapinte, Organometallics 1991, 10, 1045.
Â
[11] A. Coto, M. Jimenez, M. C. Puerta, P. Valerga, Organometallics 1998,
20, 4392.
[12] Crystal structure analyses: Data collection was performed at 110 K on
a Nonius Kappa CCD diffractometer (MoKa (0.70926 Š) radiation). A
purple platelet (0.32 Â 0.26 Â 0.08 mm) of 8-BPh4 was mounted on the
diffractometer and the structure was solved (SHELX93) by the
Patterson heavy-atom method and successive interpretation of the
difference Fourier maps, followed by least-squares refinement. All
atoms were refined anisotropically. a ˆ 11.9990(2), b ˆ 12.0900(3), c ˆ
18.1490(4) Š, a ˆ 81.3670(10), b ˆ 82.6420(10), g ˆ 82.8610(10)8. Vˆ
3
2566.71(10) Š3. 1calcd (294 K) ˆ 1.336 gcm , Z ˆ 2, space group P1,
Å
12231 measured reflections, 12231 unique(Rint ˆ 0.0280), 8808 ob-
served (F > 4); m(MoKa) ˆ 0.521 mm 1; R1 ˆ 0.0483 and wR2 (all
data) ˆ 0.0801, GOF ˆ 1.017, max/min largest residual peak: 0.641
and 0.667 eŠ 3. The X-ray structure of the ruthenium complex 10-
BPh4 was also determined and the structure shares the same general
crystallographic features as the iron analogue and will be published
elsewhere. Crystallographic data (excluding structure factors) for the
structure reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary publica-
tion no. CCDC-165695. Copies of the data can be obtained free of
charge on application to CCDC, 12 Union Road, Cambridge
CB21EZ, UK (fax: (‡44)1223-336-033; e-mail: deposit@ccdc.cam.
ac.uk).
[13] O. J. Scherer, T. Hilt, G. Wolmershäuser, Organometallics 1998, 17,
4110.
[14] O. J. Scherer, G. Schwarz, G. Wolmershäuser, Z. Anorg. Allg. Chem.
1996, 622, 951.
‡
[15] Theoretical modeling of the structure of the exo-HP4 cation points to
a similar distortion of the P4 tetrahedron with apical P P distances
(2.122 Š) shorter than basal P P separations (2.269 Š). J.-L. Abboud,
Â
 Ä
M. Herreros, R. Notario, M. Essefar, O. Mo, M. Yanez, J. Am. Chem.
Soc. 1996, 118, 1126.
[16] The occurrence of a low-lying paramagnetic excited state to explain
the large temperature shifts of dPM can reasonably be excluded on the
basis of the linewidth of the 31P NMR signals. 31P NMR resonances are
usually not observed in paramagnetic [Cp*(dppe)Fe] derivatives. See:
J. C. Fettinger, S. P. Mattamana, R. Poli, R. D. Rogers, Organometal-
lics, 1996, 15, 4221; V. Guillaume, V. Mahias, A. Mari, C. Lapinte,
Organometallics, 2000, 19, 1422.
[17] I. de los Rios, M. Peruzzini, A. Romerosa, unpublished results.
[18] The deep orange complex [Cp*Ru(dppe)(PI3)]BPh4 may be prepared
from 10-BPh4 and excess I2 in CHCl3. The compound was charac-
terized by elemental analysis and NMR spectroscopy. I. de los Rios,
M. Peruzzini, A. Romerosa, unpublished observations.
[*] Prof. A. D. Hamilton, K. Choi
Department of Chemistry, Yale University
New Haven, CT 06520 (USA)
Fax : (‡1)203-432-3221
E-mail: andrew.hamilton@yale.edu
[**] We thank the National Institutes of Health for financial support of this
work.
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
3912
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4020-3912 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 20