The abnormal hydrolysis of 7-phosphanorbornenium salts: A case of phosphonium-phosphenium equivalence

Article abstract of DOI:10.1021/ja907789y

(Chemical Equation Presented) The hydrolysis of a phosphanorbornenium triflate gives the expected tertiary phosphine oxide by cleavage of one P-C bond of the bridge in the presence of triethylamine but affords the secondary phosphine oxide by cleavage of

Full text of DOI:10.1021/ja907789y

Published on Web 10/19/2009
The Abnormal Hydrolysis of 7-Phosphanorbornenium Salts: A Case of
Phosphonium-Phosphenium Equivalence
Rongqiang Tian,†,‡ Hui Liu,† Zheng Duan,*,† and Franc¸ois Mathey*,†,‡
Chemistry Department, International Phosphorus Laboratory, Zhengzhou UniVersity, Zhengzhou 450052, P. R.
China, DiVision of Chemistry & Biological Chemistry, School of Physical & Mathematical Sciences, Nanyang
Technological UniVersity, 21 Nanyang Link, Singapore 637371
Received September 14, 2009; E-mail: duanzheng@zzu.edu.cn; fmathey@ntu.edu.sg
Phosphonium salt hydrolysis is certainly one of the most
fundamental and well-studied reactions in organophosphorus chem-
istry.1 The accepted mechanism is schematized in eq 1.
When Wittig discovered pentaarylphosphoranes, he also showed
that when phosphorus is incorporated in a five-membered ring, one
of the favored decomposition pathways involves a reductive
elimination, as shown in eq 2.2 On this basis, it seemed quite
product of the reaction is the secondary phosphine oxide (4)9
resulting from cleavage of the phosphorus bridge (eq 5).
In a preceding paper, we demonstrated that 7-R-7-phosphanor-
possible to alter the normal course of a phosphonium salt hydrolysis
bornenes can be considered as synthetic equivalents of nucleophilic
to get a phosphinous acid3 instead of the classical phosphine oxide
phosphinidenes [RP] on the basis of their transformation into
(eq 3). In such a case, the phosphonium salt would become
phosphinites by quaternization and alcoholysis in the presence of
triethylamine.6c The intermediate phosphoniums were not character-
ized. The result of eq 4 seemed to contradict our preceding results.
Thus, we decided to monitor by NMR the reaction of (1) with
benzyl bromide in C6D6. At 120 °C, the phosphonium collapses to
give the bromophosphine (eq 6).10
synthetically equivalent to a phosphenium ion, whose chemistry is
only modestly developed at the present time.4 In order to realize
this possibility, it would suffice to find a system in which the
reductive elimination occurs faster than the normally slow release
of the R- carbanion. A structure of choice in this regard is the
7-phosphanorbornene bicyclic system. Indeed, following extensive
studies by Quin5 and us,6 it has been shown to readily lose its
phosphorus bridge under a variety of conditions through retro-
McCormack reactions. Here we report that it is possible to direct
the hydrolysis of phosphanorbornenium salts toward either the
normal or abnormal pathway by just slightly manipulating the
experimental parameters.
The quaternization of (1)6a by methyl triflate is complete at room
temperature in 1.5 h. The salt (2)7 is insoluble in most organic
In order to rationalize these experimental findings, we decided
solvents, as expected for an ionic compound. It does not react with
to perform density functional theory calculations at the B3LYP/6-
neutral water but is rapidly hydrolyzed in the presence of triethy-
lamine to give the expected product as a mixture of two isomers
311+G(d,p) level11 on the model compounds (8)-(10) (eq 7). The
(3a,b)8 (eq 4). Quite strikingly, when triethylamine is replaced by
computed structures of (8) and (9) are shown in Figure (1). The
structure of (8) shows that it is an ion pair with a P-Br separation
R-picoline, the reaction pathway changes completely, and the sole
of 2.85 Å (P-Br 2.22 Å in PBr3).12 Instead of being in the plane
of symmetry of the phosphanorbornene cation, the two Me groups
on P are displaced away from the Br atom as a result of this
† Zhengzhou University.
‡ Nanyang Technological University.
9
16008 J. AM. CHEM. SOC. 2009, 131, 16008–16009
10.1021/ja907789y CCC: $40.75  2009 American Chemical Society

C O M M U N I C A T I O N S
Soc. 1964, 86, 2378. Corfield, J. R.; Trippett, S. Chem. Commun. 1970,
1267. Allen, D. A.; Grayson, S. J.; Harness, I.; Hutley, B. G.; Mowat,
I. W. J. Chem. Soc., Perkin Trans. 2 1973, 1912. Schnell, A.; Tebby, J. C.
J. Chem. Soc., Perkin Trans. 1 1977, 9, 1883. McClelland, R. A.; McGall,
G. H.; Patel, G. J. Am. Chem. Soc. 1985, 107, 5204. Gilheany, D. G.;
Thompson, N. T.; Walker, B. J. Tetrahedron Lett. 1987, 28, 3843. Cristau,
H.-J.; Mouchet, P. Phosphorus, Sulfur Silicon Relat. Elem. 1995, 107, 135.
Lee, K. Y.; Na, J. E.; Lee, M. J.; Kim, J. N. Tetrahedron Lett. 2004, 45,
5977.
(2) Wittig, G.; Maercker, A. Chem. Ber. 1964, 97, 747.
(3) Pietrusiewicz, K. M.; Stankevicˇ, M. Sci. Synth. 2009, 42, 221.
(4) Yoshifuji, M. Sci. Synth. 2009, 42, 63.
(5) Quin, L. D.; Mesch, K. A. J. Chem. Soc., Chem. Commun. 1980, 959.
Mesch, K. A.; Quin, L. D. Tetrahedron Lett. 1980, 21, 4791. Quin, L. D.;
Szewczyk, J. J. Chem. Soc., Chem. Commun. 1984, 1551. Quin, L. D.;
Szewczyk, J. J. Chem. Soc., Chem. Commun. 1986, 844. Quin, L. D.;
Jankowski, S.; Rudsinski, J.; Sommese, A. G.; Wu, X.-P. J. Org. Chem.
1993, 58, 6212.
(6) (a) Mathey, F.; Mercier, F. Tetrahedron Lett. 1981, 22, 319. (b) Holand,
S.; Mathey, F. J. Org. Chem. 1981, 46, 4386. (c) Tian, R.; Duan, Z.; Zhou,
X.; Geng, D.; Wu, Y.; Mathey, F. Chem. Commun. 2009, 2589.
(7) (2): 31P NMR (DMSO-d6): δ 83.7. 1H NMR (DMSO-d6): δ 1.88 (s, 6H,
Me-C), 2.31 (d, JHP ) 14.4 Hz, Me-P), 3.69 (d, 2H, JHH ) 1.2 Hz, CH-
CO), 4.37 (dd, 2H, JHP ) 5.4, CH-P), 7.03 (m, 2H, NPh), 7.45-7.53 (m,
3H, NPh), 7.73-7.86 (m, 3H, PPh), 8.08-8.15 (m, 2H, PPh). 13C NMR
(DMSO-d6): δ 4.44 (d, JCP ) 51.7 Hz, Me-P), 15.40 (Me-C), 44.34 (d,
JCP ) 16.3 Hz, CH(CO)), 45.40 (d, JCP ) 55.1 Hz, CH-P), 119.30 (d, JCP
) 41.5 Hz, C ipso Ph-P), 174.03 (d, JCP ) 15.5 Hz, CO).
Figure 1. Computed structures of (8) and (9). Most significant bond lengths
(Å) and angles (deg) for (8): P-Br, 2.848; P-C2, 1.911; P-C3, 1.888;
P-C15, 1.828; P-C20, 1.832; C2-P-C3, 79.6; Br-P-C15, 84.6;
Br-P-C20, 83.3; C15-P-C20, 112.6. For (9): P-O, 1.783; P-C2, 2.02;
P-C3, 1.924; P-C15, 1.844; P-C20, 1.855; C2-P-C3, 76.3; O-P-C15,
94.0; O-P-C20, 92.3; C15-P-C20, 110.0.
(8) (3a,b) was separated by chromatography on silica gel with 20:1 CH2Cl2/
EtOH as the eluent. (3a): first eluted, yield 39.2% from (2). 31P NMR
(CDCl3): δ 41. 1H NMR (CDCl3): δ 1.75 (d, JHP ) 12.9 Hz, Me-P), 1.84
(d, JHP ) 5.1 Hz, Me-C), 1.89 (s, Me-C), 2.48 (d, 1H, JHH ) 15.3 Hz,
CH2), 3.21-3.44 (m, 4H, CH + CH2), 7.13-7.16 (m, 2H, NPh), 7.35-
7.44 (m, 3H, PPh), 7.56 (br, 3H, NPh), 7.80-7.83 (m, 2H, PPh). 13C NMR
(CDCl3): δ 15.92 (d, JCP ) 69.3 Hz, Me-P), 20.55 (d, JCP ) 3.1 Hz, Me-
(C)), 21.23 (d, JCP ) 2.3 Hz, Me-(C)), 31.61 (d, JCP ) 2.8 Hz, CH2),
39.73 (d, JCP ) 2.9 Hz, CH), 40.65 (s, CH), 46.28 (d, JCP ) 69.4 Hz,
CH-P), 121.15 (d, JCP ) 7.4 Hz, C ipso NPh), 126.18 (s, CH NPh), 128.57
(s, CH para NPh), 129.05 (d, JCP ) 11.1 Hz, CH PPh), 129.06 (s, CH
NPh), 130.46 (d, JCP ) 8.8 Hz, CH PPh), 131.95 (s, C(Me)), 132.13 (d,
JCP ) 2.7 Hz, C(Me)), 132.59 (d, JCP ) 91.3 Hz, C ipso PPh), 133.49 (d,
JCP ) 8.7 Hz, CH para PPh). (3b): yield 36.4%. 31P NMR (CDCl3): δ 38.
1H NMR (CDCl3): δ 1.22 (s, Me-C), 1.61 (d, JHP ) 5.1 Hz, Me-C), 1.91
(d, JHP ) 12 Hz, Me-P), 2.26 (d, 1H, JHH ) 15.9 Hz, CH2), 2.72 (br, 1H,
CH2), 3.18 (d, JHP ) 10.5 Hz, CH-P), 3.40 (m, CH), 3.80 (m, CH), 7.18
(m, 2H, NPh), 7.34-7.60 (m, 6H, NPh + PPh), 7.68-7.74 (m, 2H, PPh).
interaction. The strain of the bridge and the long P-C bridge bonds
explain why (8) can collapse easily to give the dimethylbromophos-
phine. The calculations show that (9) is a genuine intermediate (a
local minimum with no imaginary frequencies). Its structure displays
very long P-C bridge bonds, even longer than the corresponding
bonds in (8). The bridge is very strained, with a C-P-C angle of
76.3°, which is smaller than the corresponding angle in (8). These
data suggest that (9) loses its bridge very easily. Upon removal of
the hydroxyl proton from (9), the bicyclic structure directly collapses
to give (10) with no detectable intermediate or transition state. On
this basis, we explain our results as follows. When the hydrolysis
of (2) is conducted in weakly basic media,13 the half-life of the
intermediate hydroxyphosphorane is sufficiently long that this
species can lose its bridge to give the phosphinous acid. When the
hydrolysis is conducted in strongly basic media, the fast deproto-
nation induces the collapse toward the tertiary phosphine oxide.
13C NMR (CDCl3): δ 15.47 (d, JCP ) 66.9 Hz, Me-P), 20.16 (d, JCP
3.0 Hz, Me-(C)), 20.32 (d, JCP ) 1.7 Hz, Me-(C)), 31.12 (d, JCP ) 2.6
Hz, CH2), 39.55 (d, JCP ) 2.9 Hz, CH), 40.66 (s, CH), 47.39 (d, JCP
)
)
68.9 Hz, CH-P), 121.61 (d, JCP ) 7.4 Hz, C ipso NPh), 126.25 (s, CH
NPh), 128.57 (d, JCP ) 10.8 Hz, CH PPh), 128.64 (s, CH para NPh), 129.11
(s, CH NPh), 130.61 (d, JCP ) 8.9 Hz, CH PPh), 131.97 (s, C(Me)), 132.19
(d, JCP ) 2.6 Hz, C(Me)), 132.07 (d, JCP ) 95.6 Hz, C ipso PPh), 132.29
(d, JCP ) 9.3 Hz, CH para PPh).
(9) (4): purified by chromatography with EtOAc/CH2Cl2 as the eluent, yield
48% from (1). 31P NMR (CDCl3): δ 20.6 (d, JPH ) 474 Hz). 1H NMR
3
(CDCl3): δ 1.71 (dd, 3H, JHH ) 3.9 Hz, JHP ) 13.8 Hz, Me), 7.43-7.50
(m, 3H, Ph), 7.54 (dq, JHP ) 475 Hz, 3JHH ) 3.9 Hz, PH), 7.60-7.67 (m,
2H, Ph). 13C NMR (CDCl3): δ 16.25 (d, JCP ) 68.8 Hz, Me), 128.89 (d,
JCP ) 12.6 Hz, CH Ph), 129.49 (d, JCP ) 11.2 Hz, CH Ph), 131.97 (d, JCP
) 99.7 Hz, C ipso Ph), 132.42 (s, C para Ph).
(10) (5): 1H NMR (CDCl3): δ 1.90 (s, 3H, Me), 1.94 (s, 3H, Me), 2.36-2.47
(m, 1H, CH2), 2.61-2.70 (m, 1H, CH2), 3.49-3.59 (m, 1H, CH), 6.90 (d,
1H, dCH), 7.34-7.50 (m, 5H, Ph). 13C NMR (CDCl3): 17.15 (s, Me),
20.06 (s, Me), 30.66 (s, CH2), 38.59 (s, CH) 122.75, 126.41, 126.50, 128.34,
129.04, 132.07, 133.75, 136.89 (s, sp2-C), 166.95 (s, CO), 175.30 (s, CO).
Acknowledgment. The authors thank the National Natural
Science Foundation of China (20702050), Zhengzhou University
(P. R. China), and the Nanyang Technological University in
Singapore for the financial support of this work.
1
MS: m/z 254.0 (100, [M + H]+). (6): 31P NMR (C6D6): δ 75.6. H NMR
(C6D6): δ 3.08 (d, JHP ) 7.3 Hz, CH2). 13C NMR (C6D6): δ 42.92 (d, JCP
) 33.6 Hz, CH2). (7): yield 48% from (1). 31P NMR (CDCl3): δ 30 (d, JPH
2
) 475 Hz). 1H NMR (CDCl3): δ 3.34 (dt, 1H, JHH ) JHP ) 14.6 Hz,
2
CH2), 3.46 (dt, 1H, JHH ) JHP ) 14.6 Hz, CH2), 7.05 (m, 2H, benzyl),
7.20-7.29 (m, 3H, benzyl), 7.40-7.55 (m, 5H, Ph), 7.46 (dt, JHP ) 475
3
Hz, JHH ) 3.2 Hz, PH). 13C NMR (CDCl3): δ 38.76 (d, JCP ) 62.4 Hz,
Supporting Information Available: Complete ref 11. This material
is available free of charge via the Internet at http://pubs.acs.org.
CH2).
(11) Frisch, M. J.; et al. Gaussian 03, revision B.05; Gaussian, Inc.: Pittsburgh,
PA, 2003.
(12) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements; Pergamon:
Oxford, U.K., 1984; p 568.
(13) For Et3N, pKa 10.5 in water and 18.8 in MeCN; for R-picoline, pKa 6.0 in
water and 13.3 in MeCN. See: Clark, J.; Perrin, D. D. Q. ReV. Chem. Soc.
1964, 18, 295. Glasovac, Z.; Eckert-Maksic´, M.; Maksic´, Z. B. New
J. Chem. 2009, 33, 588.
References
(1) Fenton, G. W.; Ingold, C. K. J. Chem. Soc. 1929, 2342. Horner, L.;
Hoffmann, H.; Wippel, H. G.; Hassel, G. Chem. Ber. 1958, 91, 52. Zanger,
M.; WanderWerf, C. A.; McEwen, W. E. J. Am. Chem. Soc. 1959, 81,
3806. Aksnes, G.; Songstad, J. Acta Chem. Scand. 1962, 16, 1426. Hudson,
R. F.; Green, M. Angew. Chem., Int. Ed. Engl. 1963, 2, 11. McEwen, W. E.;
Kumli, K. F.; Blade-Font, A.; Zanger, M.; WanderWerf, C. A. J. Am. Chem.
JA907789Y
9
J. AM. CHEM. SOC. VOL. 131, NO. 44, 2009 16009