A polymer-supported phosphazine as a stable and practical reagent in the three-component synthesis of substituted (cyclopentadienyl)- tricarbonylrhenium complexes

Full text of DOI:10.1002/(SICI)1521-3773(19990601)38:11<1617::AID-ANIE1617>3.0.CO;2-Z

COMMUNICATIONS
L. P. F. Chibante, S. Flanagan, A. Robertson, L. J. Wilson, J. Am.
Chem. Soc. 1995, 117, 7801 ± 7804; c) P. L. Boulas, M. T. Jones, R. S.
Ruoff, D. C. Lorents, R. Malhotra, D. S. Tse, K. M. Kadish, J. Phys.
Chem. 1996, 100, 7573 ± 7579.
This indicates that the LUMO is nondegenerate for this
isomer.
The electrochemical data for the D2 and D2d isomers are in
full agreement with those previously reported[13] except for a
recent publication by Anderson et al.[14] The latter group
investigated the electrochemistry of two isomers (Table 3)
that were separated by HPLC and, on the basis of the relative
abundances and previous work, were assigned the D2- and
D2d-symmetrical structures. While the redox properties of the
D2 isomer match those reported here, the ones assigned to the
D2d isomer differ from those determined by us and others[13]
for this fullerene and rather match those of what we call here
the ªnew isomerº. On the basis of our findings we conclude
that Anderson et al. had not isolated D2d-C84 but rather the
ªnew isomerº in addition to the D2-symmetrical one.
[14] M. R. Anderson, H. C. Dorn, S. A. Stevenson, S. M. Dana, J.
Electroanal. Chem. 1998, 444, 151 ± 154.
[15] A. Herrmann, F. Diederich, C. Thilgen, H.-U. ter Meer, W. H. Müller,
Helv. Chim. Acta 1994, 77, 1689 ± 1706.
[16] A. Herrmann, M. Rüttimann, C. Thilgen, F. Diederich, Helv. Chim.
Acta 1995, 78, 1673 ± 1704.
[17] R. Taylor, J. Chem. Soc. Perkin Trans. 2 1993, 813 ± 824.
[18] E. W. Godly, R. Taylor, Pure Appl. Chem. 1997, 69, 1411 ± 1434.
[19] C. Thilgen, A. Herrmann, F. Diederich, Helv. Chim. Acta 1997, 80,
183 ± 199.
[20] The six C2-symmetrical, constitutionally isomeric bis-adducts of D2-
C84 result from cyclopropanation at the pairs of g-type bonds
C(9) C(10)/C(75) C(76), C(9) C(10)/C(17) C(18), C(9) C(10)/
C(67) C(68), C(14) C(15)/C(7) C(22), C(14) C(15)/C(63) C(78),
and C(14) C(15)/C(70) C(71); see Figure 1 for the numbering of
(fC)-D2-C84
.
Received: January 18, 1999 [Z12920IE]
German version: Angew. Chem. 1999, 111, 1716 ± 1721
[21] P. L. Boulas, Y. Zuo, L. Echegoyen, Chem. Commun. 1996, 1547 ±
1548.
[22] M. Fanti, G. Orlandi, G. Poggi, F. Zerbetto, Chem. Phys. 1997, 223,
159 ± 168.
[23] a) H. Goto, N. Harada, J. Crassous, F. Diederich, J. Chem. Soc. Perkin
Trans. 2 1998, 1719 ± 1723; b) G. Orlandi, G. Poggi, F. Zerbetto, Chem.
Phys. Lett. 1994, 224, 113 ± 117.
Keywords: Bingel reaction ´ circular dichroism ´ electro-
chemistry ´ fullerenes
[24] M. R. Anderson, H. C. Dorn, P. M. Burbank, J. R. Gibson in Recent
Advances in the Chemistry and Physics of Fullerenes and Related
Materials (Eds.: K. M. Kadish, R. S. Ruoff), The Electrochemical
Society, Pennington, NJ, 1994, pp. 448 ± 456.
[1] F. Diederich, R. Ettl, Y. Rubin, R. L. Whetten, R. Beck, M. Alvarez, S.
Anz, D. Sensharma, F. Wudl, K. C. Khemani, A. Koch, Science 1991,
252, 548 ± 551.
[2] a) D. E. Manolopoulos, P. W. Fowler, J. Chem. Phys. 1992, 96, 7603 ±
7613; b) B. L. Zhang, C. Z. Wang, K. M. Ho, J. Phys. Chem. 1992, 96,
7183 ± 7185; c) K. Raghavachari, Chem. Phys. Lett. 1992, 190, 397 ±
400; d) D. Bakowies, M. Kolb, W. Thiel, S. Richard, R. Ahlrichs, M. M.
Kappes, Chem. Phys. Lett. 1992, 200, 411 ± 417; e) S. Okada, S. Saito,
Chem. Phys. Lett. 1996, 252, 94 ± 100.
[3] a) K. Kikuchi, N. Nakahara, T. Wakabayashi, S. Suzuki, H. Shiromaru,
Y. Miyake, K. Saito, I. Ikemoto, M. Kainosho, Y. Achiba, Nature 1992,
357, 142 ± 145; b) D. E. Manolopoulos, P. W. Fowler, R. Taylor, H. W.
Kroto, D. R. M. Walton, J. Chem. Soc. Faraday Trans. 1992, 88, 3117 ±
3118; c) R. Taylor, J. Langley, A. G. Avent, T. J. S. Dennis, H. W.
Kroto, D. R. M. Walton, J. Chem. Soc. Perkin Trans. 2 1993, 1029 ±
1036; d) C. Thilgen, F. Diederich, R. L. Whetten, in Buckminsterful-
lerenes (Eds.: W. E. Billups, M. A. Ciufolini), VCH, New York, 1993,
pp. 53 ± 82.
[4] a) T. J. S. Dennis, T. Kai, T. Tomiyama, H. Shinohara, Chem. Commun.
1998, 619 ± 620; b) K. M. Allen, T. J. S. Dennis, M. J. Rosseinsky, H.
Shinohara, J. Am. Chem. Soc. 1998, 120, 6681 ± 6689.
[5] A. L. Balch, A. S. Ginwalla, J. W. Lee, B. C. Noll, M. M. Olmstead, J.
Am. Chem. Soc. 1994, 116, 2227 ± 2228.
A Polymer-Supported Phosphazine as a Stable
and Practical Reagent in the Three-Component
Synthesis of Substituted (Cyclopentadienyl)-
tricarbonylrhenium Complexes**
Filippo Minutolo and John A. Katzenellenbogen*
In recent years the use of reagents and catalysts bound to
inorganic and organic solid supports has rapidly become an
area of intense research activity,[1] since they present several
obvious advantages over their soluble counterparts: they can
be removed from the reaction mixture by simple filtration
and, often, can be recycled and used again. Herein we report
the preparation and the use of a polymer-bound stabilized
diazocyclopentadiene (C5H4N2) analogue, which can be used
as a safe and storable source of C5H4N2 in the synthesis of
substituted cyclopentadienyl ± Re(CO)3 complexes.
[6] F. Diederich, R. L. Whetten, Acc. Chem. Res. 1992, 25, 119 ± 126.
[7] a) ªScience and Technology of Fullerene Materialsº: Y. Achiba, K.
Kikuchi, Y. Aihara, T. Wakabayashi, Y. Miyake, M. Kainosho, Mater.
Res. Soc. Symp. Proc. 1995, 359, 3 ± 9; b) A. G. Avent, D. Dubois, A.
Â
Penicaud, R. Taylor, J. Chem. Soc. Perkin Trans. 2 1997, 1907 ± 1910.
Â
Â
[8] M. Saunders, H. A. Jimenez-Vazquez, R. J. Cross, W. E. Billups, C.
Gesenberg, A. Gonzalez, W. Luo, R. C. Haddon, F. Diederich, A.
Herrmann, J. Am. Chem. Soc. 1995, 117, 9305 ± 9308.
[9] a) C. Thilgen, A. Herrmann, F. Diederich, Angew. Chem. 1997, 109,
2362 ± 2374; Angew. Chem. Int. Ed. Engl. 1997, 36, 2269 ± 2280; b) C.
Thilgen, F. Diederich, Top. Curr. Chem. 1999, 199, 135 ± 171.
[10] F. Arias, Y. Yang, L. Echegoyen, Q. Lu, S. R. Wilson, in Recent
Advances in the Chemistry and Physics of Fullerenes and Related
Materials (Eds.: K. M. Kadish, R. S. Ruoff), The Electrochemical
Society, Pennington, NJ, 1995, pp. 200 ± 212.
[11] R. Kessinger, J. Crassous, A. Herrmann, M. Rüttimann, L. Echegoyen,
F. Diederich, Angew. Chem. 1998, 110, 2022 ± 2025; Angew. Chem. Int.
Ed. 1998, 37, 1919 ± 1922.
The use of ªfreeº C5H4N2 in the synthesis of halogen-
substituted cyclopentadienyl complexes of rhenium was
[*] Prof. Dr. J. A. Katzenellenbogen, Dr. F. Minutolo
Department of Chemistry, University of Illinois
600 S. Mathews Avenue, Box 37-5, Urbana, IL 61801 (USA)
Fax : (‡ 1)217-333-7325
E-mail: jkatzene@uiuc.edu
[**] This research was supported by the National Institutes of Health and
the Department of Energy. Funding for NMR instrumentation was
also provided from the W. M. Keck Foundation and the National
Science Foundation.
[12] J. M. Hawkins, M. Nambu, A. Meyer, J. Am. Chem. Soc. 1994, 116,
7642 ± 7645.
[13] a) M. S. Meier, T. F. Guarr, J. P. Selegue, V. K. Vance, J. Chem. Soc.
Chem. Commun. 1993, 63 ± 65; b) Y. Yang, F. Arias, L. Echegoyen,
Angew. Chem. Int. Ed. 1999, 38, No. 11
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1433-7851/99/3811-1617 $ 17.50+.50/0
1617

COMMUNICATIONS
reported more than twenty years ago.[2] Recently, however, we
To overcome this problem, we envisaged the use of a
polymer-bound phosphazine 2, so that the phosphane formed
at the end of the reaction could be removed by filtration,
thereby considerably simplifying the work-up and purification
procedure. The heterogeneous reagent 2 was prepared by
treating a dichloromethane suspension of triphenylphosphane
bound to a polystyrene support 3[9] with a pentane solution of
C5H4N2 [Eq. (3)]. The functionalization of the polymer was
very efficient (85%) after only 2 hours at room temperature.
found that C5H4N2 reacts simultaneously with a [fac-Re-
‡
(CO)3] precursor[4] and external nucleophiles, such as carbox-
ylates[3a] efficiently, to produce acyloxy-substituted cyclopen-
tadienyl± Re(CO)3 complexes in a one-pot reaction (three-
component reaction) and in short times. We then extended the
scope of this reaction to the synthesis of carbon-substituted
cyclopentadienyl ± Re(CO)3 complexes by utilizing boronic
acids as carbon nucleophiles;[3b] this transformation represents
a new method for C C bond formation that is based on boronic
acids, but does not require a catalyst such as palladium. These
results were very encouraging, especially in view of a possible
use of this reaction for radiolabeling biologically interesting
molecules with cyclopentadienyltricarbonyl complexes con-
taining 186Re, 188Re, and 99mTc radionuclides. However, a major
limitation for an application of the three-component synthesis
to the routine preparation of radiopharmaceuticals was the
high instability of C5H4N2.[5] In fact, this key reagent must
always be stored in solution,[6] usually in pentane, at low
temperatures (for example, 408C), in the dark; yet, in spite of
all these precautions, it cannot be stored for longer than a few
weeks, because of its rapid decomposition. All these aspects
constituted a serious limitation to the practical utilization of
this reaction in medical facilities. Therefore, we started to look
for alternative reagents that possesed the same good reactivity
of C5H4N2, together with better stability and handling proper-
ties. We focused our attention on an adduct of C5H4N2 with
triphenylphosphane, phosphazine 1,[7] which was known to be a
stable compound under ambient conditions (moist air, room
temperature, light, shocks).[8] As a matter of fact, 1 is used in
the gravimetric titration of solutions of diazocyclopentadi-
ene.[8b] We were pleased to find that 1 readily dissociates into
Ph3P and free diazo compound when dissolved in acetonitrile,
which is the solvent used in our reaction [Eq. (1)].[3] At 238C
The amount of C5H4N2 in the heterogeneous reagent 2 was
determined by several methods. Determination of the amount
of nitrogen (%N) by elemental analysis can be used as a
measurement of the extent of functionalization, but elemental
analysis by itself does not distinguish between the C5H4N2 that
is covalently bound and thereby stabilized as a phosphazine
and material that may simply be trapped in the polymeric
network, but still remaining as a free unstable diazo com-
pound. In order to be sure that all unreacted C5H4N2 had been
washed out of the polymer, we analyzed 2 by IR and solid-
state NMR spectroscopy. Free C5H4N2 has a very strong,
1
characteristic IR absorption around 2100 cm which arises
from the asymmetric stretching of the cumulated C-N-N
double bonds. The absence of such a band in the product
polymer 2 (as was also the case in the monomeric phosphazine
1), indicated that the amount of nitrogen found by elemental
analysis could be entirely attributed to polymer-bound
phosphazine. In fact, other typical bands for phosphazines
were found in the IR spectrum of 2 at nÄ ˆ 1517 (nC N), 1102
ˆ
1
(nP N), and 974 cm (nN N).
ˆ
CH3CN
(1)
P
N
Ph3P
To further validate these results, we conducted a solid-state
ªmagic angle spinningº (MAS) 31P NMR analysis of 2
(Figure 1). The 31P spectrum shows two peaks, an intense
one at d ˆ 21.10, which corresponds to functionalized P atoms,
and a weaker one at d ˆ 7.03, which arises from unreacted
phosphane groups present in the polymer. The observed
chemical shifts compare favorably to those found in the 31P
NMR spectra of their soluble counterparts in CDCl3: free
phosphazine 1 gives a reasonance at d ˆ 23.23, and Ph3P at
d ˆ 2.29. The 31P MAS NMR spectrum of the starting
polymeric phosphane 3 also showed a single peak at d ˆ
6.77, which confirmed the identity of the weaker peak
shown in Figure 1.
As we expected, the striking feature of 2 was its great
stability. This material shows no sign of decomposition
(elemental analysis, NMR and IR spectroscopy) after many
months (up to 11 as of yet) of storage on the bench, exposed to
ambient conditions without any particular precautions. To test
its reactivity, 2 was then used in the three-component reaction
with several nucleophiles. The yields obtained using hetero-
atom nucleophiles (Br and carboxylates) according to
N
N2
1
the dissociation of 1 is considerable [Eq. (2), the Kdiss value is
obtained by NMR analysis of a solution of 1 in [D3]acetoni-
trile at equilibrium], and under typical reaction conditions
‰C5H4N2Š‰Ph3PŠ
1
Kdiss (238C) ˆ
ˆ 4.0  10 3 molL
(2)
‰1Š
(808C), further dissociation generates levels of free C5H4N2
that are sufficient for an efficient reactivity, as we could verify
in preliminary experiments with carboxylates. However, the
presence of free triphenylphosphane in the reaction mixture,
produced by the dissociation of 1, made the purification of the
products very awkward, because it was difficult to separate
the phosphane from most of the product complexes. This
resulted in a decrease in the product yields when 1 was used in
place of free C5H4N2.
1618
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1433-7851/99/3811-1618 $ 17.50+.50/0
Angew. Chem. Int. Ed. 1999, 38, No. 11

COMMUNICATIONS
Et3N
OH
OH
R
∆
R
B
2
4
(5)
Re
CH3CN
CO
OC
CO
7a-e
8a-e
Table 2. Comparison of the yields of the isolated substituted cyclopenta-
dienyl ± Re(CO)3 complexes 8 obtained with boronic acids 7 using free
diazocyclopentadiene or the polymer-supported reagent 2 [Eq. (5)].
Entry Boronic
acid
R
Complex Yield [%][a] with
[b]
C5H4N2
2[c]
1
7a
8a
8b
64
53
O
2
7b
34
41
CH3
Figure 1. 31P MAS NMR (121 MHz) spectrum of 2.
Equation (4) (L ˆ CH3CN) are summarized in Table 1. Inter-
estingly, 2 showed the same good reactivity as free C5H4N2.[3a]
CH3O
3
4
5
7c
7d
7e
8c
8d
8e
74
51
72
56
45
50
L
Nu
∆
(4)
L
L
TfO–
Nu–
n-C7H15
2
Re
Re
CH3CN
OC
CO
CO
OC
CO
CO
4
6a-e
5a-e
S
Table 1. Comparison of the yields of the isolated substituted cyclopenta-
dienyl ± Re(CO)3 complexes 6 obtained with heteroatom nucleophiles 5
[a] See Table 1. [b] See ref. [3b]. [c] 0.050 mmol of 4, 0.50 mmol of boronic
acid, 1.0 mmol of Et3N, 39 mg of 2 (containing 0.075 mmol of C5H4N2),
2.5 mL of CH3CN, 808C, 45 min.
from free diazocyclopentadiene or the polymer-supported reagent
[Eq. (4)].
2
Entry
Nu
Product
Yield[%][a] with
C5H4N2
[b]
2[c]
the ones obtained with C5H4N2,[3b] but are still in
a
1
2
Br 5a
6a
6b
67
71
respectable range (around 50%), considering that this
complex transformation results in the simultaneous forma-
tion of a carbon ± carbon s bond and an h5-cyclopentadienyl ±
rhenium bond. It is important to emphasize the great
O
CH3
5b
69
67
O–
O
tolerance of this reaction towards
a wide variety of
CH3
O–
5c
5d
3
4
6c
6d
59
72
60
71
functional groups, including alcohols, phenols, amides, and
ketones. Moreover, this reaction is tolerant to sulfur-con-
taining heterocycles such as thiophenes as shown by the
synthesis of a novel thienyl-substituted complex 8e (Table 2,
entry 5).
OH
O
HO
Ph
O–
O
In conclusion, the solid-phase reagent 2 described herein
combines great safety, handling, and excellent storage proper-
ties, with good reactivity in the synthesis of substituted
cyclopentadienyl ± Re(CO)3 complexes. Its consequent suit-
ability for kit-formulations should facilitate the use of the
three-component reaction for the preparation and use of
these radiopharmaceuticals that contain organometallic com-
plexes of ReI and, by chemical analogy, TcI [11] in medical
facilities, applications that would not be possible if it were
necessary to use highly unstable reagents such as free
diazocyclopentadiene. Furthermore, the availability of this
supported/stabilized cyclopentadienyl-ring precursor should
make the synthesis of half-sandwich complexes containing
other metals practical.
5e
5
6e
60
64
Boc-NH
O–
[a] Yields of complexes purified by column chromatography are based on
the initial rhenium precursor (Et4N)2[ReBr3(CO)3][10] used to prepare
compound 4 in situ.[3a] [b] See ref. [3a]. [c] 0.050 mmol of 4, 0.10 mmol of
nucleophile, 39 mg of 2 (containing 0.075 mmol of C5H4N2), 2.5 mL of
CH3CN, 808C, 45 min; carboxylates 5b, 5d, and 5e were generated in situ
by treating the correspondent carboxylic acid (0.1 mmol) with 0.2 mmol of
Et3N; commercially available sodium salt of 5c was used.
The results obtained in the more recently described
reaction involving boronic acids as carbon nucleophiles
[Eq. (5)] are reported in Table 2. Here, the yields obtained
with 2 are generally slightly lower (apart from entry 2) than
Angew. Chem. Int. Ed. 1999, 38, No. 11
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1433-7851/99/3811-1619 $ 17.50+.50/0
1619

COMMUNICATIONS
Experimental Section
Memory of Chirality in Electron Transfer
Mediated Benzylic Umpolung Reactions of
Arene ± Cr(CO)3 Complexes**
Diazocyclopentadiene (1.78m in pentane) was prepared by a literature
method.[6b] Polymer-supported triphenylphosphane 3 and all other chem-
icals were obtained from commercial sources unless otherwise specified.
Boronic acid (7d) was prepared from 1-nonyne by a literature method.[12]
Characterization data and purification details of complexes 6a,[2] 6b ± e,[3a]
and 8a ± d[3b] have already been reported.
Hans-Günther Schmalz,* Charles B. de Koning,*
Dirk Bernicke, Stephan Siegel, and Anja Pfletschinger
Since the discovery of arene ± tricarbonylchromium com-
plexes in 1958[1] this class of compounds has been studied
extensively. It has been shown that the Cr(CO)3 group
activates the arene ligand in different ways to facilitate
transformations that cannot be achieved with the free
arenes.[2] Furthermore, it has been demonstrated that novel
and competitive strategies for the enantioselective synthesis
of complex organic molecules are possible if both the
chemical and stereochemical properties of such complexes
are exploited.[3, 4] While the vast majority of applications of
arene ± Cr(CO)3 complexes are based on polar reactions
involving coordinatively saturated (18 valence electron
(VE)) anionic or cationic intermediates,[2] it has only recently
been realized that transformations involving Cr(CO)3-com-
plexed benzylic radicals are also useful for organic synthesis.[5]
A theoretical investigation carried out in this laboratory
suggests that the parent benzyl radical complex 1a is better
described as the 17 VE resonance structure 1b. This structure
clearly indicates the presence of an exocyclic C C double
bond and shows delocalization of a significant portion of the
spin density onto the chromium atom (Scheme 1).[6]
2: A 1.78m pentane solution of diazocyclopentadiene (4 mL, 1.2 equiv) was
added to a well stirred suspension of the heterogeneous phosphane 3 (2.0 g,
ꢀ5.6 mmol of P) in CH2Cl2 (15 mL) at room temperature. The mixture was
stirred for a further 2 h, after which time Et2O (30 mL) was added and the
orange insoluble polymer was collected by filtration, washed with Et2O
(3 Â 10 mL), and dried under vacuum to yield 2 (2.4 g) as an orange
powder. Elemental analysis revealed 5.39%N, which corresponded to a
85% functionalization of the phosphane groups. IR (CH2Cl2 film): nÄ ˆ 1517
(C N), 1102 (P N), 974 cm 1 (N N); 31P MAS NMR (121 MHz, neat solid,
external reference: H3PO4, see Figure 1): d ˆ 21.10 (phosphazine), 7.03
(phosphane) .
ˆ
ˆ
8e (Table 2, entry 5): m.p. 868C; Rf (silica gel, hexane) 0.14; IR (KBr): nÄ ˆ
1
2018 (COasym), 1910 cm (COsym,); 1H NMR (500 MHz, CDCl3): d ˆ 7.32
(dd, J ˆ 5.0, 2.9 Hz, 1H, H5 thienyl), 7.28 (dd, J ˆ 2.9, 1.5 Hz, 1H, H2
thienyl), 7.07 (dd, J ˆ 5.0, 1.5 Hz, 1H, H4 thienyl), 5.67 (pseudo t, J ˆ
2.2 Hz, 2H, Ha C5H4Ar), 5.37 (pseudo t, J ˆ 2.2 Hz, 2H, Hb C5H4Ar);
13C NMR (126 MHz, CDCl3): d ˆ 194.04 (CO), 132.72 (C4 thienyl), 126.64,
126.20, 121.59 (C3 thienyl), 103.91 (C4 C5H4Ar), 84.01, 81.63 (C3 C5H4Ar);
‡
‡
MS (70 eV): m/z (%): 418 (100) [M , 187Re], 416 (62) [M , 185Re], 390 (36)
‡
‡
‡
[M
CO, 187Re], 388 (23) [M
CO, 185Re], 362 (60) [M
2CO, 187Re],
‡
360 (37) [M
2CO, 185Re].
Received: November 24, 1998 [Z12710IE]
German version: Angew. Chem. 1999, 111, 1730 ± 1732
Keywords: cyclopentadienes ´ diazo compounds ´ heteroge-
neous reactions ´ multicomponent reactions ´ rhenium.
CH3
CH2
Cr(CO)3
Cr(CO)3
[1] a) S. J. Shuttleworth, S. M. Allin, P. K. Sharma, Synthesis 1997, 1217 ±
1239, and references therein; b) G. W. Kabalka, R. M. Pagni,
Tetrahedron 1997, 53, 7999 ± 8065; c) J. H. Clark, A. P. Kybett, D. J.
Macquarrie, Supported Reagents: Preparation, Analysis, and Applica-
tions, VCH, Weinheim, 1992.
1a
1b
Scheme 1. Resonance structures for the Cr(CO)3-complexed benzyl
radical.
[2] W. A. Herrmann, Chem. Ber. 1978, 111, 2458 ± 2460.
[3] a) F. Minutolo, J. A. Katzenellenbogen, J. Am. Chem. Soc. 1998, 120,
4514 ± 4515; b) F. Minutolo, J. A. Katzenellenbogen, J. Am. Chem.
Soc. 1998, 120, 13264 ± 13265.
[4] R. Alberto, R. Schibli, P. A. Schubiger, Polyhedron 1996, 15, 1079 ±
1089, and references therein.
[5] A. G. Wedd, Chem. Ind. (London) 1970, 4, 109 ± 109.
[6] a) M. Regitz, A. Liedhegener, Tetrahedron 1967, 23, 2701 ± 2708;
b) K. J. Reimer, A. Shaver, J. Organomet. Chem. 1975, 93, 239 ± 252.
[7] For a crystallographic analysis of the electronic distribution in 1, see F.
Minutolo, S. R. Wilson, J. A. Katzenellenbogen, Acta Crystallogr.
Sect. C, in press.
[8] a) F. Ramirez, S. Levy, J. Org. Chem. 1958, 23, 2036 ± 2037; b) T. Weil,
M. Cais, J. Org. Chem. 1963, 28, 2472 ± 2472.
[9] The polymer-supported phosphane 3 was first used in the Wittig
reaction where purification problems similar to ours are often
encountered: M. Bernard, W. T. Ford, J. Org. Chem. 1983, 48, 326 ±
332.
[10] R. Alberto, A. Egli, U. Abram, K. Hegetschweiler, V. Gramlich, P. A.
Schubiger, J. Chem. Soc. Dalton Trans. 1994, 2815 ± 2820.
[11] ªTechnetium and Rheniumº: N. M. Boog, H. D. Kaesz in Compre-
hensive Organometallic Chemistry, Vol. 4 (Eds.: G. Wilkinson, G. F. A.
Stone, E. W. Abel), Pergamon, Oxford, 1982, pp. 161 ± 242; see also
ref. [3a].
As a consequence, derivatives of 1 that carry an additional
substituent in the benzylic position (for example, 2) are planar
chiral radical species and should exhibit a significant degree of
configurational stability. Indeed, the barrier for the racemi-
zation of 2 was calculated to be 13.2 kcalmol 1 (Scheme 2),[7]
which corresponds to a half-life of about one minute at
788C.[8]
[*] Prof. Dr. H.-G. Schmalz, D. Bernicke, Dr. S. Siegel,
Dipl.-Chem. A. Pfletschinger
Institut für Organische Chemie der Technischen Universität
Strasse des 17. Juni 135, D-10623 Berlin (Germany)
Fax: (‡49)30314-21105
E-mail: schmalz@wap0109.chem.tu-berlin.de
Dr. C. B. de Koning
Department of Chemistry, University of the Witwatersrand
PO WITS 2050, Johannesburg (South Africa)
Fax: (‡27)11-339-7967
E-mail: dekoning@aurum.chem.wits.ac.za
[**] This work was supported by the Volkswagenstiftung and the Fonds
der Chemischen Industrie. C.B.deK. thanks the Foundation for
Research Development (FRD) and the University of the Witwaters-
rand, South Africa for financial support. We thank Prof. Dr. W. Koch,
Technical University Berlin, for support and helpful discussions
concerning the quantum chemical calculations. We also thank
Chemetall GmbH for the generous gifts of chemicals.
[12] H. C. Brown, S. K. Gupta, J. Am. Chem. Soc. 1975, 97, 5249 ± 5255.
1620
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1433-7851/99/3811-1620 $ 17.50+.50/0
Angew. Chem. Int. Ed. 1999, 38, No. 11