Splitting the phosphorus bridge of 7-phosphanorbornadiene complexes by fluoride ion

Article abstract of DOI:10.1002/zaac.200500427

The reaction of fluoride ion with 7-phosphanorbornadiene P-W(CO) ₅ complexes yields fluorophosphido complexes which, in turn, can attack a second molecule of 7-phosphanorbornadiene to give a fluorobiphosphine complex. The corresponding anion displays a huge P-P coupling. The structure of the anionic chromium analogue has been investigated by DFT calculations. The P-P bond is relatively short at 2.20 A and displays a huge polarity suggesting an interesting chemistry. When the 7-phosphanorbornadiene P-substituent is 2-chloroethyl, the attack of the fluoride ion is followed by a cyclization and a reductive dimerization leading to the first known biphosphirane complex.

Full text of DOI:10.1002/zaac.200500427

DOI: 10.1002/zaac.200500427
Splitting the Phosphorus Bridge of 7-Phosphanorbornadiene Complexes by
Fluoride Ion
Carine Compain and Francois Mathey*
¸
Riverside, CA / USA, UCR-CNRS Joint Research Chemistry Laboratory, Department of Chemistry, University of California Riverside
Received October 31st, 2005
Dedicated to Professor Herbert W. Roesky on the Occasion of his 70th Birthday
Abstract. The reaction of fluoride ion with 7-phosphanorbornadi-
ene PϪW(CO)5 complexes yields fluorophosphido complexes
which, in turn, can attack a second molecule of 7-phosphanorbor-
nadiene to give a fluorobiphosphine complex. The corresponding
anion displays a huge PϪP coupling. The structure of the anionic
chromium analogue has been investigated by DFT calculations.
polarity suggesting an interesting chemistry. When the 7-phos-
phanorbornadiene P-substituent is 2-chloroethyl, the attack of the
fluoride ion is followed by a cyclization and a reductive dimeriz-
ation leading to the first known biphosphirane complex.
˚
The PϪP bond is relatively short at 2.20 A and displays a huge
Keywords: Phosphido complexes; Biphosphines; Biphosphirane
Introduction
only relatively mild nucleophiles selectively attack the phos-
phorus bridge of complexes 1. This pathway has been
illustrated with several carbon nucleophiles including stabil-
ized phosphorus ylids [3] and malonate anions [4]. But this
kind of chemistry is, by no way, limited to carbon nucleo-
philes. Hereafter, we demonstrate this assumption with the
fluoride ion.
An enormous amount of data has been accumulated over
the years on the uses of 7-phosphanorbornadiene com-
plexes (1) as convenient precursors of terminal phosphinid-
ene complexes [RP-M(CO)5] (M ϭ Cr, Mo, W) (2). These
transient species indeed display a quite versatile carbene-
like chemistry [1, 2]. This highly focused interest has led
those in the field to neglect another possibility involving the
nucleophilic cleavage of the phosphorus bridge as shown in
Scheme 1.
Results and Discussion
Our initial experiments have been carried out with the phe-
nyl derivative 3. After several experiments, we have found
that the most convenient source of fluoride ion was cesium
fluoride in the presence of 18-crown-6. At 70°C in THF,
the splitting reaction is complete in 1h (Scheme 2).
The same reaction goes to completion in 5 minutes at RT
with [Bu4N]F, but the isolated yield is much lower. This
less satisfactory experiment is, nevertheless, quite instructive
because it rules out a phosphinidene mechanism (even in
the presence of CuCl as a catalyst, the generation of 2
necessitates at least 60°C) and demonstrates that the forma-
tion of 5 is not due to a low concentration of fluoride ion
in the reaction medium. We are obliged to admit that the
initially formed fluorophosphide complex 4 is much more
reactive toward the phosphanorbornadiene complex 3 than
the fluoride ion. The anion thus formed 5 can be charac-
terized in situ by 31P NMR: δ31P(5) 197.2, 1JPϪP ϭ 413 Hz,
1JPϪF ϭ 997 Hz (PF); Ϫ4.7, 1JPϪP ϭ 413 Hz, 2JPϪF ϭ 113.6
Hz (PϪ). Upon protonation, it yields a 50:50 mixture of
the two diastereomers of 6: δ31P(6a)(CDCl3) 174.6, 1JPϪP ϭ
Scheme 1
This nucleophilic splitting can occur under much milder
conditions than the thermal splitting but is limited by the
presence, besides phosphorus, of another electrophilic
center, i.e. the electron-poor CϭC double bond. In practice,
* Prof. Dr. F. Mathey
1
1
UCR-CNRS Joint Research Chemistry Laboratory
Department of Chemistry, University of California Riverside
Riverside, CA 92521-0403 / USA
FAX: ϩϩ1-951-8277870
E-Mail: fmathey@citrus.ucr.edu
165.4 Hz, JPϪF ϭ 883.8 Hz (PF); 5.6, JPϪH ϭ 331.9 Hz,
2
1JPϪP ϭ 165.4 Hz, JPϪF ϭ 62.7 Hz (PH); δ31P(6b)(CDCl3)
1
1
170.3, JPϪP ϭ 155.6 Hz, JPϪF ϭ 878.2 Hz (PF); 10.0,
1JPϪH ϭ 334.1 Hz, JPϪP ϭ 155.6 Hz, JPϪF ϭ 77.9 Hz
1
2
Z. Anorg. Allg. Chem. 2006, 632, 421Ϫ424
 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
421

C. Compain, F. Mathey
Scheme 2
Figure 1 Computed structure of fluorophosphide complex 7.
˚
Selected bond lengths /A and angles /deg.: PϪCr 2.524, PϪF 1.678, PϪMe
Figure 2 HOMO of [MePFCr(CO)5]Ϫ
1.876; CrϪPϪF 103.8, CrϪPϪMe 107.1, FϪPϪMe 95.6, C(7)ϪCrϪC(4)
169.7, C(5)ϪCrϪC(8) 173.2.
(PH). On the proton spectrum, the PH resonance appears
at 6.36 for 6a and 6.61 for 6b. When comparing the data of
the anion 5 with those of 6a, b, the most striking difference
concerns the PϪP coupling. The enormous value recorded
for 5 is reminiscent of the so-called phospha-Wittig reagents
obtained by reaction of complexes such as 3 with tributyl-
phosphine [5] and implies some double bond character in
the PϪP bond. An interesting chemistry might be devel-
oped around these species.
The optimized geometry of the fluorophosphinephos-
phide complex 8 is shown in figure 3. The PϪP bond length
˚
at 2.200 A lies between that of a classical PϪP single bond
˚
In order to understand the electronic structure of 5, we
decided to perform DFT calculations [6] on the methyl ana-
logues of 4 and 5 in the pentacarbonyl-chromium series.
Geometry optimizations were carried out using the B3LYP
functional [7, 8] in combination with 6-31G(d) basis sets for
all atoms except for Cr (lanl2dz). The optimized geometry
of the fluorophosphide complex 7 is shown in figure 1. The
molecule adopts a staggered conformation, the PϪCr bond
(2.24 A) and that of a zwitterionic phospha-Wittig complex
˚
(2.156 A) [5]. The PϪP bond is highly polar: Mulliken
charges: Pᮎ 0.07, PF 0.71. The distortion of the Cr(CO)5
complexing group linked to Pᮎ is weaker than in the
fluorophosphide complex: eq. CO-Cr-eq. CO 173.3 and
179.3 vs 169.7 and 173.2 deg. This observation suggests
some delocalization of the negative charge from the tricoor-
dinate to the tetracoordinate P unit. The PϪCr bond
ᮎ
˚
˚
˚
is long at 2.524 A (vs 2.417 A for the corresponding
7-phosphanorbornadiene complex at the same computational
level) and the equatorial CO ligands bend toward phos-
phorus atom to a significant extent. This bending does not
exist in the phosphanorbornadiene precursor. The HOMO
is highly localized at P atom as shown in figure 2.
lengths are very different at 2.419 (PF) and 2.579 A (P ).
The CrϪPϪPϪCr dihedral angle is 86.2°, resulting in a mi-
nimization of the steric congestion. As shown in figures 4
and 5, the HOMO is centered on the [PCr(CO)5]Ϫ, and the
LUMO on the [PFCr(CO)5] unit, thus confirming the polar
nature of the PϪP bond.
422
zaac.wiley-vch.de
 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2006, 421Ϫ424

Splitting the Phosphorus Bridge of 7-Phosphanorbornadiene Complexes by Fluoride Ion
Figure 5 LUMO of [(OC)5CrPF(Me)PMeCr(CO)5]Ϫ (8)
Figure 3 Computed structure of fluorophosphinephosphide com-
plex 8.
˚
Selected bond lengths /A and angles /deg.: P(17)ϪCr(2) 2.419, P(17)ϪMe
1.851, P(17)ϪF 1.652, P(17)ϪP(18) 2.200, P(18)ϪMe 1.894, P(18)ϪCr(23)
2.579; FϪP(17)ϪMe 96.5, Cr(2)ϪP(17)ϪP(18) 121.5, P(17)ϪP(18)ϪCr(23)
111.5, P(17)ϪP(18)ϪMe 98.1, Cr(2)ϪP(17)ϪP(18)ϪCr(23) 86.2.
The biphosphirane complex 12 is the main product of the
reaction obtained in 30% isolated yield. The presence of the
3-membered ring is obvious from the 31P NMR data:
δ31P(12) Ϫ192.7, 1JPϪW ϭ 154 Hz, 2JPϪW ϭ 99 Hz. The 1H
spectrum shows two types of protons at 1.59 and 1.76 ppm.
The 13C spectrum shows a pseudo-triplet at 13.67 ppm. The
dimeric nature of 12 is confirmed by the mass spectrum
which displays a molecular peak at m/z 766. As far as we
know, this biphosphirane structure has not been reported
in the literature until now. We can admit that the initial step
of the reaction is the formation of the fluorophosphide
anion as in the previous case. The intramolecular cycliza-
tion leading to 11 appears to be a logical next step. But we
have no clue for the reduction of 11 into 12. The dismu-
tation of R2PF into R2P-PR2 and R2PF3 has been reported
in the literature [10, 11]. The mechanism implies the forma-
tion of an intermediate biphosphorus species R2P-PR2F2.
In our case, the formation of such an intermediate is im-
possible without breaking the strong PϪW bond and we
have no obvious solution to propose. Anyhow, it is clear
that the reaction of fluoride ion with 7-phosphanorbornad-
iene complexes is a source of interesting new species whose
chemistry remains to explore.
Figure 4 HOMO of [(OC)5CrPF(Me)PMeCr(CO)5]Ϫ (8)
We then decided to extend our initial findings to other
types of substituents. The reaction of CsF with the 7-(2-
chloroethyl)-7-phosphanorbornadiene complex 9 [9] led to
an unexpected result shown on scheme 3.
Scheme 3
Z. Anorg. Allg. Chem. 2006, 421Ϫ424
 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
zaac.wiley-vch.de
423

C. Compain, F. Mathey
Experimental Section
Synthesis of biphosphirane complex 12
In a Schlenk tube under nitrogen, a solution of precursor 9
(450 mg, 0.7 ϫ 10Ϫ3 mol) and cesium fluoride (150 mg, 10Ϫ3 mol)
in 2 ml of dry THF was heated at 70°C for 30 min. After evapor-
ation of the solvent, the product was recovered by chromatography
on silica gel (hexane/CH2Cl2 9/1 as the eluent) as brown crystals
(80 mg, 0.104 ϫ 10Ϫ3 mol). Yield ϭ 29.7 %. Anal. Calcl for
C14H8O10P2W2: C, 21.96; H, 1.05. Found: C, 21.76; H, 1.09%.
Reactions were performed under nitrogen using oven-dried glass-
ware. Dry tetrahydrofuran was obtained by distillation from
Na/benzophenone. Silica gel (70Ϫ230mesh) was used for chroma-
tographic separation. Nuclear magnetic resonance spectra were ob-
tained on a Bruker Avance 3000 and Varian Inova spectrometer
1
operating at 300.13 MHz for H, 75.45 MHz for13C, and 121.496
MHz for31P. Chemical shifts are expressed in parts per million
downfield from external TMS (1H and 13C) and external 85%
H3PO4 (31P) and data are reported as follows: chemical shift, multi-
plicity (sϭsinglet, dϭdoublet, tϭtriplet, mϭmultiplet, bϭbroad),
integration, and coupling constants in hertz. Mass spectra were
obtained on VG 7070 and Hewlett ϪPackard 5989A GC/Ms
spectrometers. Starting materials were obtained from commercial
suppliers or prepared according to literature methods.
31P NMR (CDCl3): δ ϭ Ϫ192.7 (1JPϪW ϭ 154.0 Hz, JPϪW ϭ 98.9 Hz);
1
1H NMR (CDCl3): δ ϭ 1.58 (m, 4H, PϪCH2), 1.59 (m, 4H, PϪCH2);
13C NMR (CDCl3): δ ϭ 13.7 (pseudo t, PϪCH2); EIMS (m/z): 767 (Mϩ,
58.5 %), 712 (M2ϩ Ϫ2CO, 4.5 %), 684 (Mϩ Ϫ3CO, 60.3 %), 654 (Mϩ Ϫ4CO,
52.5 %), 628 (Mϩ Ϫ5CO, 79.9 %), 597 (M2Ϫ Ϫ6CO, 87.3 %), 570 (MϪ
Ϫ7CO, 100 %), 542 (MϪ Ϫ8CO, 39.9 %), 514 (MϪ Ϫ9CO, 39.3 %), 486 (MϪ
Ϫ10CO, 29.5 %).
References
[1] F. Mathey, N. H. Tran Huy, A. Marinetti, Helv. Chim. Acta
2001, 84, 2938.
[2] K. Lammertsma, M. J. M. Vlaar, Eur. J. Org. Chem. 2002,
1127.
Synthesis of fluorobiphosphine complex 6a, b
[Bu4N]F method:
In a Schlenk tube under nitrogen, a solution of 1.2 g of the precur-
sor 3 (1.8 ϫ 10Ϫ3 mol) and 1.8 ml of [Bu4N]F (2 ϫ 10Ϫ3 mol) in
2 ml of dry THF was stirred for 5 min. at room temperature. After
evaporation of the solvent, the mixture of the two diastereoisomers
was recovered by chromatography on silica gel (hexane/CH2Cl2
8/2 as the eluent) as black crystals (200 mg, 0.226 ϫ 10Ϫ3 mol).
Overall yield: 25 %.
[3] N. Hoffmann, C. Wismach, P. G. Jones, R. Streubel, N. H.
Tran Huy, F. Mathey, Chem. Commun. 2002, 454; N. H. Tran
Huy, C. Compain, L. Ricard, F. Mathey, J. Organomet. Chem.
2002, 650, 57.
[4] C. Compain, N. H. Tran Huy, F. Mathey, Heteroatom Chem.
2004, 15, 258.
[5] P. Le Floch, A. Marinetti, L. Ricard, F. Mathey, J. Am. Chem.
Soc. 1990, 112, 2407.
[6] Gaussian 03, Revision B.05, Frisch, M. J.; Trucks, G. W.;
Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J.
R; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant,
J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Men-
nucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.;
Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.;
Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Strat-
mann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli,
C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.;
Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Mar-
tin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C.
Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; John-
son, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
Gaussian, Inc., Pittsburgh PA, 2003.
CsF method:
In a Schlenk tube under nitrogen, a solution of 0.6 g ofthe precur-
sor 3 (0.9 ϫ 10Ϫ3 mol), 70 mg of cesium fluoride (0.45 ϫ 10Ϫ3
mol) and 60 mg of 18-crown-6 in 2 ml of dry THF was heated for
4 h. at 70°C. After evaporation of the solvent, the mixture of the
two diastereoisomers was recovered by chromatography on silica
gel (hexane/CH2Cl2 8/2 as the eluent) as black crystals (185 mg,
0.21
ϫ
10Ϫ3 mol). Overall yield: 47 %. Anal. Calcd for
C22H11O10FP2W2: C, 29.85; H, 1.25. Found: C, 29.76; H, 1.14%.
Complex 6a:
31P NMR (CDCl3): δ ϭ 5.6 (1JPϪH ϭ 331.9 Hz, JPϪP ϭ 165.4 Hz, 2JPϪF
ϭ
1
62.7 Hz), 174.6 (1JPϪ1 P ϭ 165.4 Hz, JPϪ2 F ϭ 883.8 Hz); 1H NMR (CDCl3):
1
3
δ ϭ 6.36 (ddd, 1H, JPϪH ϭ 331.3 Hz, JPϪH ϭ 12.0 Hz, JFϪH ϭ 12.0 Hz,
PϪH), 7.34 (m, 10H, C6H5).
Complex 6b:
31P NMR (CDCl3): δ ϭ 10.0 ( 1JPϪH ϭ 334.1 Hz, 1JPϪP ϭ 155.6 Hz, 2JPϪF ϭ
77.9 Hz), 170.3 (1JPϪ1P ϭ 155.6 Hz, JPϪF ϭ 878.2 Hz); 1H NMR (CDCl3):
1
2
3
δ ϭ 6.61 (ddd, 1H, JPϪH ϭ 334.1 Hz, JPϪH ϭ 6.6 Hz, JFϪH ϭ 15.5 Hz,
PϪH), 7.34 (m, 10H, C6H5).
[7] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
[8] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
[9] B. Deschamps, F. Mathey, Tetrahedron Lett. 1985, 26, 4595.
[10] L. Riesel, J. Haenel, J. Fluorine Chem. 1988, 38, 335.
[11] J. Haenel, G. Ohms, L. Riesel, Z. Anorg. Allg. Chem. 1992,
607, 161.
Mixture 6a ϩ 6b:
13C NMR (CDCl3): δ ϭ 129.2 (m, C6H5), 193.5 (m, CO cis), 196.2 (m, CO
trans); EIMS (m/z): 888 (M4ϩ, 6.1 %), 746 (M2ϩ Ϫ5CO, 3.5 %), 718 (M2ϩ
Ϫ6CO, 3.5 %), 690 (M2ϩ Ϫ7CO, 3.8 %), 662 (M2ϩ Ϫ8CO, 3.7 %), 635 (M3ϩ
Ϫ9CO, 6.2 %), 604 (M Ϫ10CO, 21.0 %), 452 (PhPFW(CO)5ϩ, 100 %).
424
zaac.wiley-vch.de
 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2006, 421Ϫ424