Dual supermesityl stabilization: A room-temperature-stable 1,2,4-triphosphole radical, sigmatropic hydrogen rearrangements, and tetraphospholide anion

Article abstract of DOI:10.1002/ejic.200800265

Cesium 3,5-bis(2,4,6-tri-tert-butylphenyl)-1,2,4-triphospholide (12) and cesium 5-(2,4,6-tri-tert-butylphenyl)tetraphospholide (13) were synthesized and isolated with flat fivemembered rings, which are an indication of the aromaticity in these anions. Compound 13 is the first example of a stable tetraphospholide anion, which is structurally characterized. Kinetic stabilization of the 1,2,4-triphospholide system by two supermesityl groups resulted in the detection of the room-temperature-stable radical 17 and the observation of a series of successive sigmatropic hydrogen shifts in the first stable 1H-1,2,4-triphosphole 14 with a P-H bond. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejic.200800265

SHORT COMMUNICATION
DOI: 10.1002/ejic.200800265
Dual Supermesityl Stabilization: A Room-Temperature-Stable
1,2,4-Triphosphole Radical, Sigmatropic Hydrogen Rearrangements, and
Tetraphospholide Anion[‡]
Alex S. Ionkin,*[a] William J. Marshall,[a] Brian M. Fish,[a] Alexander A. Marchione,[a]
Laurie A. Howe,[a] Fredric Davidson,[a] and Charles N. McEwen[a]
Keywords: ESR spectroscopy / Phosphaalkynes / Phosphorus / Cesium / Heterocycles
Cesium 3,5-bis(2,4,6-tri-tert-butylphenyl)-1,2,4-triphosphol-
two supermesityl groups resulted in the detection of the
ide (12) and cesium 5-(2,4,6-tri-tert-butylphenyl)tetraphos- room-temperature-stable radical 17 and the observation of a
pholide (13) were synthesized and isolated with flat five-
membered rings, which are an indication of the aromaticity
in these anions. Compound 13 is the first example of a stable
tetraphospholide anion, which is structurally characterized.
Kinetic stabilization of the 1,2,4-triphospholide system by
series of successive sigmatropic hydrogen shifts in the first
stable 1H-1,2,4-triphosphole 14 with a P–H bond.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
cules. The introduction of two supermesityl groups has been
recently shown to stabilize such intriguing diradicals as 5–
7 and the diphosphorus analogue 8 of a cyclic carbene
(Scheme 1).[3–5]
Introduction
In this report we describe the use of CsF in the formation
of P=C and PϵC bonds by the hexamethyldisiloxane elimi-
nation route.[1] It has been shown that the reaction between
simple tris(trimethylsilyl)phosphane (2), as a source of
phosphorus, and 2,4,6-tri-tert-butylbenzoyl chloride (3)
does not take place at room temperature. More active metal
bis(trimethylsilyl)phosphides, e.g., lithium bis(trimethylsilyl)-
phosphide, are required for the reaction to take place.[2]
Thus, it is an attractive possibility for cesium bis(trimethyl-
silyl)phosphide (4) to facilitate the introduction of not only
one, but two supermesityl groups to the core of the mole-
Results and Discussion
Three-component reaction between tris(trimethylsilyl)
phosphane (2), cesium fluoride (1) and 2,4,6-tri-tert-butyl-
benzoyl chloride (3) resulted in variable yields of six prod-
ucts (4, 9–13), depending on the reaction time (Scheme 2).
Shorter reaction times (up to 1 d in dioxane, at 90 °C) led
predominantly to the first four products in Scheme 2.
Longer reaction times (up to 5 d) led to the formation of
cesium 3,5-bis(2,4,6-tri-tert-butylphenyl)-1,2,4-triphosphol-
ide (12) and cesium 5-(2,4,6-tri-tert-butylphenyl)tetraphos-
pholide (13). Continued heating of the reaction mixture (up
to 10 d) favors the formation of 13. Two-component reac-
tion between tris(trimethylsilyl)phosphane (2) and cesium
fluoride (1) (3 h in dioxane at 90 °C) led exclusively to the
formation of cesium bis(trimethylsilyl)phosphide (4). Pro-
gress of the reactions can be accurately monitored by 31P
NMR spectroscopy, since the compounds of interest have
distinctive chemical shifts: δ = –276.10 (4);[6] 33.2 (10);[7]
Scheme 1.
3
117.24 (11); 267.87 (d), 287.55 (t) (integral ratio 2:1; JPP
[‡] DuPont contribution #8849
= 36.7 Hz; 12) ppm. The most deshielded resonances were
observed for 13. They represent an AAЈMMЈ spin system
at δ = 341.60 ppm for the PA and PAЈ atoms and at δ =
386.50 ppm for the PM and PMЈ atoms. The simulated coup-
[a] DuPont Central Research & Development, Experimental Sta-
tion, Wilmington,
Delaware 19880-500, USA
Fax: +1-302-695-1675
E-mail: alex.s.ionkin@usa.dupont.com
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
1
3
ling constants are JAAЈ = –485.60 Hz, JMMЈ = 52.20 Hz,
1JAM = –486.80 Hz, JAMЈ = 2.00 Hz.
2
2386
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 2386–2390

Dual Supermesityl Stabilization
to form infinite chains (e.g., in 12; Figure 1) or a three-
dimensional network (e.g., in 13; Figure 2) in the solid state.
Use of large alkali cations (cesium in our case) leads to
increased coordination numbers and the ability to form co-
ordination polymers (two- or three-dimensional) in the so-
lid state. The distances between the cesium and phosphorus
atoms are in the range of 3.60–4.24 Å, which are consistent
with η1- and η2-coordinations, and are within such dis-
tances reported in the Cambridge Structural Database. The
longest Cs–P distance of 4.24 Å is an η2-coordination bond;
therefore, the actual Cs–P bond should be shorter and
should be pointed to the center of the P=P bond. The bond
lengths between the cesium and oxygen atoms are in the
range of 2.91–3.24 Å, which are the normal values in the
Cambridge Structural Database, with 3.21 Å as a median
for Cs–O bonds.
Scheme 2. Three-component reaction between tris(trimethylsilyl)-
phosphane (2), cesium fluoride (1) and 2,4,6-tri-tert-butylbenzoyl
chloride (3). Coordinated solvents are omitted for clarity. Mes* =
supermesityl (2,4,6-tri-tert-butylphenyl).
Figure 1. Packing diagram of 12. Compound 12 forms infinite
chains with four dioxane molecules bonded to each cesium atom.
Thermal ellipsoids are drawn at the 50% probability level.[18]
The formation of the three products 4, 10 and 11 is ex-
pected on the basis of mechanistic studies between lithium
bis(trimethylsilyl)phosphide and 2,4,6-tri-tert-butylbenzoyl
chloride[7] and tris(trimethylsilyl)phosphane and cesium
tert-butoxide.[6] The formation of the second three products
9, 12 and 13 is new and specific to the use of CsF.
The cesium atom is octahedral in 12 (four η1-coordina-
tion bonds to dioxane and one η2-coordination bond to
P=P) and in 13 (three η1-coordination bonds to dioxane
and three η2-coordination bonds to P=P). The cesium
atoms in 12 and in 13 have the potential ability to coordi-
nate to the five-membered rings by an η5-coordination
mode. However, combinations of η2-coordination bonds to
P=P and η1-coordination bonds to oxygen atoms prevail
over a possible η5-coordination mode. Theoretical calcula-
tions show that η2-coordination is energetically only 1.4
times lower than η5-coordination in the Cp–Li+ system.[8]
Favorable coordination of the cesium cation with the oxy-
gen atoms of dioxane also contributes to the overall prefer-
ence for η2- and η1-coordinations over possible η5-coordi-
Figure 2. Packing diagram of 13. Compound 13 forms a three-di-
mensional network that joins above and below the plane of the
page through dioxane molecules. Each cesium atom is bonded to
three dioxane molecules. Thermal ellipsoids are drawn at the 50%
nation in 12 and 13. The cesium cation has a clear tendency probability level.[18]
Eur. J. Inorg. Chem. 2008, 2386–2390
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2387

A. S. Ionkin et al.
SHORT COMMUNICATION
The five-membered rings in 12 and in 13 are perfectly flat, tem. Coupling reactions of the sterically less hindered tBu
and there is averaging of bond lengths within the five-mem- systems gave only organophosphorus cage compounds with
bered rings, which is an indication of the aromaticity in these the general formula P6C4tBu4.[12] The structure of 15 was
triphospholide and tetraphospholide anions.[9] Compound investigated by X-ray analysis (Figure 4). The two triphos-
13 is the first example of a tetraphospholide anion that is phole rings are not parallel to each other and are not π-
structurally characterized. Sodium tetraphospholide was ob- stacked. They are connected by a P–P bond (2.17 Å), which
served in the reaction mixture, but never was isolated.[10] The is the longest P–P bond in this compound. The P–P bond
ability to isolate 12 and 13 comes from a combination of length inside the triphosphole ring is 2.12 Å. The P=C bond
factors such as the ability of the large cesium cation to dis- lengths in the ring are within the normal range (between
perse its positive charge and thus to stabilize the unusual 1.72 and 1.78 Å). Characterization of 15 in solution by 31P
anions, a presence of sterically hindered supermesityl groups,
and a good crystallinity of these species. To quantify which
factor is more important is somewhat difficult.
The room-temperature 31P NMR spectrum of 3,5-
bis(2,4,6-tri-tert-butylphenyl)-1H-1,2,4-triphosphole
(14)
shows only one broad signal at δ = 314.21 ppm (Figure 3A).
A series of successive [1,5]-sigmatropic hydrogen shifts be-
tween P-1 and P-2 is occurring so quickly that only the P-
4 signal is manifested in the 31P NMR spectrum. Slowing
down the exchange by cooling to –80 °C produces the ex-
pected spectrum for 14 containing three chemical shifts: a
singlet at δ = 310.22 ppm (P-4), a doublet at δ = 299.20 ppm
(1JPP = 405.9 Hz, P-2), and a doublet at δ = 76.40 ppm (P-
1) with the same coupling constant (Figure 3B). The hydro-
gen atom is located at the phosphorus atom in 1-position of
the ring. The most shielded signal (P-1) shows an additional
splitting from the hydrogen atom in the 31P NMR spectrum
Figure 4. ORTEP drawing of 15. Thermal ellipsoids are drawn at
the 50% probability level.[18]
1
with a coupling constant of JPP = 128.2 Hz. There are no
studies on the hydrogen sigmatropic rearrangements for tri-
phosphole systems in the literature to date. Although sup-
rafacial [1,5]-sigmatropic shifts of a triphenylstannyl group
are reported in the triphosphole system.[11]
Figure 3. 31P NMR spectra of 14: (A) at 30 °C; (B) at –80 °C.
The first stable P–H triphosphole 14 was shown to have
a fruitful reactivity. It can be converted back to the cesium
salt 12 by reaction with metallic cesium at room tempera-
ture. Exposure of this P–H compound to air for 24 h led to
the formation of the P–P dimer 15 of 1H-1,2,4-triphosphole
(Figure 4 and Scheme 3) and a small amount of 2,4,6-tri-
tert-butylbenzoic acid (16). Compound 15 is the first exam-
ple of a 2,2Ј,4,4Ј-tetraphospha-1,1Ј-biphospholyl ring sys- Scheme 3.
2388
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 2386–2390

Dual Supermesityl Stabilization
NMR spectroscopy was not possible; no 31P NMR reso- to any of these previously reported species supports the
nance at all was detected in the region between δ = 1000 conclusion that the unpaired electron in 17 is fully delocal-
and –700 ppm in [D8]toluene. It was discovered that dimer ized over the cyclopentadienyl ring.[17]
15 underwent P–P dissociation, forming the persistent free
The minor species detected in the ESR spectrum has not
radical species 17 (Scheme 3). It is theorized that the pres- been identified. Attributed to it are two resonances of equal
ence of 17 effected paramagnetic broadening of the NMR amplitude, the separation of which is strongly temperature-
signals, rendering them undetectable; P–P dissociations in dependent. At 298 K, the two resonances are separated by
related tBu cage structures have been observed upon irradi- 13 G. This separation increases with decreasing tempera-
ation with UV light or daylight.[12]
ture; at the lowest temperature studied, 185 K, the separa-
The ESR spectrum of 17 in [D8]toluene at 298 K is tion is approximately 83 G (Figure 5b). The analyte was still
shown in Figure 5a. The primary feature is a broad 1:3:3:1 fully soluble at 185 K by visual inspection. The nature of
quartet with a hyperfine coupling of approximately 19 G this apparent kinetic effect is unclear.
and an isotropic ge of 2.002. The hyperfine coupling pattern
A solution of the unpurified product mixture containing
is consistent with the resonance structure of 17, in which 17 in [D8]tetrahydrofuran evidenced many persistent free
an unpaired electron in a π-orbital interacts almost equally radical species by ESR spectroscopy, perhaps attributable
with the three spin 1/2 31P nuclei. Neither the magnitude of to supermesityl-centered free radicals. Since 17 was believed
the hyperfine coupling constant nor the isotropic g factor to be the sterically most hindered of all of the radical spe-
is consistent with an unpaired electron localized on a single cies, 2,2,6,6-tetramethyl-4-oxopiperidinoxy free radical
phosphorus atom.[13] Both the g factor and the hyperfine (TEMPO) was added to the product solution at ambient
coupling constant are comparable to the somewhat similar temperature under nitrogen in an attempt to quench the less
1,3-diphosphacyclobuten-4-yl radical synthesized by Ito et hindered free radical species, and leave 17 as the remaining
al.[14] In that case, unlike 17, the magnitudes of the 31P hy- persistent radical; 18 h after addition of TEMPO, a simpli-
perfine coupling constants were different, suggesting partial fied ESR spectrum was obtained (Figure 5c), in which a
localization of the unpaired electron on the cyclobutene quartet described in the experiments above was observed as
ring. Another comparable radical, the diphosphaallyl spe- the dominant signal.
cies described by Canac et al.,[15] showed an allyl hyperfine
coupling of 9.4 G and an isotropic g of 2.0048. In contrast
to our system, the diphosphaallyl species identified as an
Conclusions
intermediate in the formation of hexaphosphapentapris-
mane by Al-Ktaifani et al.[16] showed a much higher ge
(2.013) and allyl 31P hyperfine coupling constants of 80–
84 G. The dissimilarity of the ESR signal attributed to 17
We presented here a new approach for generating P=C
and PϵC bonds in a condensation reaction between silyl-
phosphanes and aroyl chloride in the presence of CsF. A
combination of factors such as the ability of the large ce-
sium cation to disperse its positive charge and thus to stabi-
lize the unusual anions, a presence of sterically hindered
supermesityl groups, and a good crystallinity led to the iso-
lation of cesium 3,5-bis(2,4,6-tri-tert-butylphenyl)-1,2,4-tri-
phospholide (12) and cesium 5-(2,4,6-tri-tert-butylphenyl)-
tetraphospholide (13). Kinetic stabilization of the 1,2,4-tri-
phospholide system by two supermesityl groups resulted in
detection of the room-temperature-stable radical 17 and the
observation of a series of successive sigmatropic hydrogen
shifts in 14. The reactivity of 13, mechanistic studies and
use of 12 and 13 in homogeneous catalysis will be further
investigated and reported in due time.
Supporting Information (see footnote on the first page of this arti-
cle): ESR experimental details, synthesis of 4, 9, 11, 12, 13, 14 and
15.
Acknowledgments
The authors thank Professor Barry M. Trost for helpful discussions
of this study and Karin Karel for proofreading the manuscript.
Figure 5. ESR spectra of 17: (a) in [D8]toluene at 298 K; (b) in [D8]-
toluene at 185 K; (c) in [D8]tetrahydrofuran 18 h after addition of
TEMPO at room temperature; (a) and (b) are on the same horizon-
tal scale.
[1] a) K. D. Dillon, F. Mathey, J. F. Nixon, Phosphorus: The Car-
bon Copy, Wiley, Chichester, 1998; b) L. D. Quin, A Guide to
Organophosphorus Chemistry, John Wiley & Sons, Inc., New
York, 2000; c) M. Regitz, O. J. Scherer, Multiple Bonds and Low
Eur. J. Inorg. Chem. 2008, 2386–2390
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2389

A. S. Ionkin et al.
SHORT COMMUNICATION
Coordination in Phosphorus Chemistry, Georg Thieme Verlag, [11]
Stuttgart, 1990.
A. Elvers, F. W. Heinemann, B. Wrackmeyer, U. Zenneck,
Chem. Eur. J. 1999, 5, 3143.
[2] G. Märkl, H. Sejpka, Angew. Chem. 1986, 98, 286–287.
[3] H. Sugiyama, S. Ito, M. Yoshifuji, Chem. Eur. J. 2004, 10,
2700–2706.
[12]
a) M. M. Al-Ktaifani, D. P. Chapman, M. D. Francis, P. B.
Hitchcock, J. F. Nixon, L. Nyulászi, Angew. Chem. Int. Ed.
2001, 40, 3474; b) M. M. Al-Ktaifani, P. B. Hitchcock, J. F.
Nixon, Inorg. Chim. Acta 2003, 356, 103; c) R. Bartsch, J. F.
Nixon, Polyhedron 1989, 8, 2407; d) J. M. Lynam, M. C.
Copsey, M. Green, J. C. Jeffery, J. E. McGrady, C. A. Russell,
J. M. Slattery, A. C. Swain, Angew. Chem. Int. Ed. 2003, 42,
2778; e) R. K. Bansal, N. Gupta, Sci. Synth. 2004, 13, 729.
F. Gerson, W. Huber, Electron Spin Resonance of Organic Radi-
cals, Wiley-VCH, Weinheim, 2003.
S. Ito, M. Kikuchi, M. Yoshifuji, A. J. Arduengo III, T. A.
Konovalova, L. D. Kispert, Angew. Chem. Int. Ed. 2006, 45,
4341.
[4] E. Niecke, A. Fuchs, M. Baumeister, M. Nieger, W. W. Scho-
eller, Angew. Chem. Int. Ed. Engl. 1995, 34, 555.
[5] D. Martin, A. Baceiredo, H. Gornitzka, W. W. Schoeller, G.
Bertrand, Angew. Chem. Int. Ed. 2005, 44, 1700.
[6] a) F. Uhlig, S. Gremler, M. Dargatz, M. Scheer, E. Herrmann,
Z. Anorg. Allg. Chem. 1991, 606, 105; b) U. Englich, K. Hassler,
K. Ruhlandt-Senge, F. Uhlig, Inorg. Chem. 1998, 37, 3532.
[7] G. Maerkl, H. Sejpka, Tetrahedron Lett. 1986, 27, 1771.
[8] a) S. Harder, Coord. Chem. Rev. 1998, 176, 17; b) S. Harder,
Chem. Eur. J. 1999, 5, 1852; c) R. E. Dinnebier, S. van Smaalen,
F. Olbrich, S. Carlson, Inorg. Chem. 2005, 44, 964; d) M. R.
Kopp, B. Neumueller, Organometallics 1997, 16, 5623.
[9] a) A. Dransfeld, L. Nyulászi, P. von R. Schleyer, Inorg. Chem.
1998, 37, 4413; b) L. Nyulászi, J. F. Nixon, J. Organomet.
Chem. 1999, 588, 28; c) W. Egan, R. Tang, G. Zon, K. Mislow,
J. Am. Chem. Soc. 1970, 92, 1443; d) J. D. Andose, A. Rauk,
K. Mislow, J. Am. Chem. Soc. 1974, 96, 6904; e) L. Nyulászi,
Chem. Rev. 2001, 101, 1229; f) L. Nyulászi, Tetrahedron 2000,
56, 79; g) V. Caliman, P. B. Hitchcock, J. F. Nixon, J. Chem.
Soc., Chem. Commun. 1995, 1661; h) L. Nyulászi, U. Berg-
sträßer, M. Regitz, P. von R. Schleyer, New J. Chem. 1998, 22,
651; i) E. Niecke, A. Fuchs, F. Baumeister, M. Nieger, W. W.
Schoeller, Angew. Chem. Int. Ed. Engl. 1995, 34, 555.
[10] a) M. Baudler, D. Düster, D. Ouzounis, Z. Anorg. Allg. Chem.
1987, 544, 87; b) M. Baudler, J. Hahn, Z. Naturforsch. 1990,
45, 1139.
[13]
[14]
[15] Y. Canac, A. Baceiredo, W. W. Schoeller, D. Gigmes, G. Ber-
trand, J. Am. Chem. Soc. 1997, 119, 7579.
[16] M. M. Al-Ktaifani, W. Bauer, U. Bergstrasser, B. Breit, M. D.
Francis, F. W. Heinemann, P. B. Hitchcock, A. Mack, J. F.
Nixon, H. Prtizkow, M. Regitz, M. Zeller, U. Zenneck, Chem.
Eur. J. 2002, 8, 2622.
[17] J.-Y. Thépot, C. Lapinte, J. Organomet. Chem. 2002, 656, 146.
[18] CCDC-681038, -681039, -681040, -681041, -681042, -681043,
-681044 contain the supplementary crystallographic 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.
Received: March 12, 2008
Published Online: April 9, 2008
2390
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 2386–2390