Wittig-olefination via an yttrium-coordinated betaine

Article abstract of DOI:10.1039/c2cc16431k

We report the synthesis of an yttrium phosphonium methylide complex and its reaction with benzophenone to form an excedingly rare hydrocarbon soluble metal-coordinated betaine. While this reaction models the C-C σ-bond formation step of the Wittig reaction under salt-conditions, addition of Ph ₃PO to the betaine complex results in formation of 1,1-diphenylethene.

Full text of DOI:10.1039/c2cc16431k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 1745–1747
www.rsc.org/chemcomm
COMMUNICATION
Wittig-olefination via an yttrium-coordinated betainew
Mark R. Crimmin*ab and Andrew J. P. Whiteb
Received 16th October 2011, Accepted 29th November 2011
DOI: 10.1039/c2cc16431k
We report the synthesis of an yttrium phosphonium methylide
complex and its reaction with benzophenone to form an excedingly
rare hydrocarbon soluble metal-coordinated betaine. While this
reaction models the C–C r-bond formation step of the Wittig
Q
reaction under salt-conditions, addition of Ph3P O to the betaine
complex results in formation of 1,1-diphenylethene.
acidic cations such as lithium are present. Schlosser, and
others, have suggested an ionic mechanism in which metal
cations catalyze the C–C s-bond forming step with subsequent
dissociation of the metal from II to form I.2,16,21–25 In spite of
this, due to the low solubility of Group 1 derivatives of II in
non-coordinating solvents, almost nothing is known of their
structures, properties and reactivity. Upon dissolution in
coordinating solvents, fast rates of olefination often complicate
the study of these compounds.
Since Vedejs’s 1973 study, oxaphosphetanes have been widely
regarded as the key intermediate in the Wittig-olefination of
carbonyls under salt-free conditions.1–4 For the reaction of
triphenylphosphonium alkylides with aldehydes and ketones,
low temperature NMR spectroscopy was used to detect and
unambigously characterize these species (Scheme 1, I). While
in related work, kinetically stabilized oxaphosphetanes have
been isolated under ambient conditions and shown to be
competent for alkene and phosphine oxide formation at
elevated temperatures.5–11
Extensive studies have demonstrated that phosphonium
ylides bind to coordinatively unsaturated lanthanide and
Group 3–5 metal centres.17,26–37 Inspired by this work, we
recently began to study the reactivity of onium ylides generated
within the coordination sphere of the rare-earth metals with the
notion that the highly electrophilic metal would perturb ylide
stability, influence reactivity and selectivity and ultimately allow
control over reaction pathways. We now report our initial
results upon the generation and reactivity of triphenyl-
phosphonium methylide at Group 3 metals.
Although the abscence of salt-free betaines as long-lived
intermediates under standard Wittig-olefination conditions12,13
is supported by extensive experimental and theoretical data,13–16
a number of reports suggest that metal-coordinated betaines
may be isolated upon precipitation from the reaction mixture
(Scheme 1, II).2,17,18 Along with the observed salt-effects on
the stereochemical outcome of the reaction of pro-chiral ylides
with pro-chiral carbonyls,19,20 this finding argues for a more
complicated mechanistic picture under conditions where Lewis
The reaction of [Y{N(SiMe3)2}3] with [Ph3PMe]I occurred
upon heating of 0.35 M d6-benzene solutions to 80 1C and could
be monitored by 31P NMR spectroscopy (see ESI,w Fig. S1).
Consumption of the phosphonium salt was observed over a 24 h
period with concomitant formation of 1 (d31P = +30.9 ppm).
A control sample of [CH2PPh3], synthesised by deprotonation of
the phosphonium iodide with [KOtBu] followed by filtration
away from potassium iodide, demonstrated different data in
d6-benzene (d31P = +22.0 ppm) suggesting that 1 contained a
metal coordinated ylide.
Performing the reaction of [Y{N(SiMe3)2}3] with [Ph3PMe]I
in alternative coordinating solvents such as diethyl ether or
THF did not lead to the isolation of 1 in a pure form.
Nevertheless reactions conducted in benzene are reproducible
and following a preparative scale reaction the product could
be crystallized from a 1 : 10 benzene to n-hexane mixture in
47% yield. X-ray quality crystals were obtained from n-hexane
(see ESIw for experimental details).
Scheme 1 Wittig-olefination of an aldehyde by [CH2PPh3].
In the solid-state the yttrium center of 1 is four coordinate
(Fig. 1). The yttrium–carbon bond length to the –CH2PPh3 ligand
is short, as may be expected for a zwitterionic ligand. At
2.460(6) A this bond length is shorter than those found in
four coordinate yttrium N-heterocyclic carbene complexes
(2.50–2.60 A)38 and longer than those found in similar complexes
containing the anionic neosilyl ligand –CH2SiMe3. For instance,
[(IPr)Y(CH2SiMe3)3] (IPr = 1,3-bis(2,6-di-iso-propylphenyl)-
imidazol-2-ylidene) provides a useful point of comparison and
a Department of Chemistry, University College London,
20 Gordon Street, London, WC1H 0AJ, United Kingdom
b Department of Chemistry, Imperial College London,
South Kensington, London, SW7 2AZ, United Kingdom.
E-mail: m.crimmin@imperial.ac.uk; Tel: 020 7594 1159
w Electronic supplementary information (ESI) available: Full Experi-
mental Details along with CIF files for 1 and 2. CCDC 843183 and
843184. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c2cc16431k
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1745–1747 1745

View Article Online
Fig. 2 POV-RAY representation of 2. H-atoms omitted for clarity.
Selected bond angles (1) and bond lengths (A), Y–O(1) 2.0717(13),
Y–N(44) 2.2399(15), Y–N(35), 2.2620(15), O(1)–C(2) 1.401(2),
C(2)–C(3) 1.529(2), C(3)–P(4) 1.8315(17), Y–O(1)–C(2) 155.24(11).
Fig. 1 POV-RAY representation of 1. H-atoms omitted for clarity.
Selected bond angles (1) and bond lengths (A), Y–C(1) 2.460(6),
P–C(1) 1.728(6), Y–N(1) 2.207(5), Y–N(2) 2.244(5), Y–I 3.0161(8),
Y–C(1)–P 138.8(3).
the retention of coordination of this motif to yttrium in solution
2
was confirmed by the small J
contains Y–C bond lengths of 2.555(2) A and 2.345(2)–2.363(2) A
from the metal to the carbene and alkyl ligands respectively.39
Consistent with the perturbation of the ylide’s structure due to
metal coordination, the P–C bond length of 1.728(6) A in 1 is
considerably longer than the average value of 1.661 A found in
[CH2PPh3],40 and lies within the range of 1.71–1.75 A estab-
lished for [{C5H4R}3Ln(CH2PPh2Me)] complexes (Ln = Nd,
Sm; R = H, t-Bu).41,42
Compound 1 reacts cleanly with benzophenone in d6-benzene
solution to form the metal-coordinated betaine 2 (d31P =
+16.0 ppm).z The reaction proceeds via insertion of the ketone
into the ylidic Y–C bond (Scheme 2). Similar addition reactions
across M–C bonds are known for rare-earth N-heterocyclic
carbene complexes, where parallels have been drawn with the
reactivity of frustrated Lewis-acid Lewis-base pairs,43,44 and
more classical rare-earth alkyl complexes.45,46 A control
experiment conducted between [CH2PPh3] and benzophenone
in the same solvent yielded 1,1-diphenylethene and triphenyl-
phosphine oxide as the only observable reaction products by
NMR spectroscopy within 15 min at 25 1C, no additional
intermediates were observed under these conditions.
coupling observed at the
89Y–13C
quartnery carbon resonance.
Single crystals of 2 could be obtained by slow diffusion of
n-hexane into a concentrated benzene solution. In the solid-state
the yttrium centre of 2 displays a similar geometry to 1, with the
triphenylphosphonium methylide replaced by the betaine ligand
(Fig. 2).y The betaine exists in a near perfect anti-periplanar
arrangement with the O–C–C–P dihedral angle being
175.88(13)1. In addition the Y–O–C of bond angle deviates
moderately from linearity taking a value of 155.24(11)1. These
data contrast with those reported for the isolated ‘betaine’,
[OC(CF3)2CH2P{(t-Bu)2Me}]ꢁMeOH.3 Synthesised via deproto-
nation of a parent phosphonium salt with NaOMe in MeOH this
compound is stabilised not only by judicious choice of substi-
tuents on phosphorous and carbon but also an intermolecular
hydrogen bond to methanol. In the solid-state, an intramolecular
electrostatic interaction between the phosphorous and oxygen
heteroatoms is suggested by the gauche conformation described
by the O–C–C–P torsion angle of 28.11 and a P–O separation of
3.121 A which resides within the sum of the van der Waals radii
(3.32 A).47 No such interaction is apparent in 2.
Berger and Neumann have reported the solid-state CP-MAS
31P NMR spectrum of the putative metal-coordinated betaine
[LiBr{OC(2-pyridyl)2CH2PPh3}] (d31P = +16.8 ppm).9 This
latter value agrees well with that determined for 2, and lies outside
the range established for oxaphosphetanes derived from triaryl-
phosphines (d31P = ꢀ55 to ꢀ75 ppm).3 In addition to the
characteristic 31P NMR data, 2 demonstrated a heavily deshielded
resonance at d5.04 ppm (d, 2J31P–1H = 10.4 Hz) in the 1H NMR
Complex 2 is an exceedingly rare example of a hydrocarbon
soluble, thermally stable metal-coordinated betaine and provides
an opportunity to study Wittig-olefination from an isolated
intermediate. In this regard, it is noteworthy that a handful of
rare-earth metal complexes, including [{(SQPPh2)2C}ScCl(Py)2],
have also been shown to be competent as ‘Tebbe’ olefination
reagents via alkylidene like reactivity of the MQCR2 unit with
ketones.48–51 While 0.4 M C6D6 solutions of 2 were stable over
24 h at 25 1C,15 addition of 1 equiv. of triphenylphosphine oxide
gave 1,1-diphenylethylene in 88% yield after 30 min at 25 1C
(Scheme 3). Although we have yet to identify the metal containing
byproducts of this reaction, the organic product, 1,1-diphenyl-
ethylene was identified by comparison of NMR spectroscopy
data to authentic samples (see ESIw for experimental details).
Similarly, heating a 0.4 M sample of 2 in d8-toluene for 30 min
at 80 1C with an excess of benzophenone (20 equiv.) gave
1,1-diphenylethylene in 83% yield based upon 2 (Scheme 4).
Samples of 2 were reacted with excess 2-chlorobenzophenone and
samples of 2-Cl (synthesized from 1 and 2-chlorobenzophenone)
were reacted with excess benzophenone, in order to investigate the
potential reversibility in the carbon–carbon s-bond forming step.
spectum and two signals at d41.2 (d, 1J
C = 31.4 Hz) and
31P–13
C = 5.6 Hz and 2J
89Y–13
83.5 ppm (dd, 2J
C = 2.8 Hz) in the
31P–13
13C NMR spectrum. These data are assigned to the methylene
and quarternary carbon positions of the betaine framework
(see Scheme 2), and while the 31P–1H and 31P–13C coupling
constants are consistent with the proposed ligand connectivity,
Scheme 2 Synthesis of an yttrium-coordinated betaine.
1746 Chem. Commun., 2012, 48, 1745–1747
c
This journal is The Royal Society of Chemistry 2012

View Article Online
9 M. Appel, S. Blaurock and S. Berger, Eur. J. Org. Chem., 2002, 1143.
10 T. Kawashima, K. Kato and R. Okazaki, J. Am. Chem. Soc., 1992,
114, 4008–4010.
11 T. Kawashima, H. Takami and R. Okazaki, J. Am. Chem. Soc.,
1994, 116, 4509–4510.
12 A. S. Ionkin, W. J. Marshall, B. M. Fish, M. F. Schiffhauer and
F. Davidson, J. Am. Chem. Soc., 2007, 129, 9210–9215.
13 F. Volatron and O. Eisenstein, J. Am. Chem. Soc., 1987, 109, 1–14.
14 V. K. Aggarwal, R. Robiette, J. Richardson and J. N. Harvey,
J. Am. Chem. Soc., 2005, 127, 13468–13469.
Scheme 3 Reaction of 2 with triphenylphosphine oxide; NMR yield
is quoted against C6Me6 as an internal standard (average of three
experiments).
15 V. K. Aggarwal, R. Robiette, J. Richardson and J. N. Harvey,
J. Am. Chem. Soc., 2006, 128, 2394–2409.
16 M. Schlosser and H. B. Tuong, Angew. Chem., Int. Ed. Engl., 1979,
18, 633.
17 R. A. Neumann and S. Berger, Eur. J. Org. Chem., 1998,
1085–1087.
18 M. Schlosser, Angew. Chem., Int. Ed. Engl., 1968, 7, 650.
19 A. B. Reitz, S. O. Nortey, A. D. Jordan, M. S. Mutter and
B. E. Maryanoff, J. Org. Chem., 1986, 51, 3302–3308.
20 W. J. Ward Jr. and W. E. McEwen, J. Org. Chem., 1990, 55,
493.
21 M. Schlosser, H. B. Tuong and C. Tarchini, Chimia, 1977, 31, 219.
22 M. Schlosser, A. Piskala, C. Tarchini and H. Batuong, Chimia,
1975, 29, 341–342.
Scheme 4 Reversibility in the Wittig-olefination of 2; NMR yields
23 U. Schroder and S. Berger, Eur. J. Org. Chem., 2000, 2601–2604.
24 M. Schlosser and K. F. Christma, Angew. Chem., Int. Ed. Engl.,
1964, 3, 636.
¨
are quoted against C6Me6 as an internal standard.
In both instances significant cross-over was observed suggesting
reversible C–C s-bond formation (Scheme 4).2,52
25 C. Geletneky, F.-H. Forsterling, W. Bock and S. Berger, Chem.
¨
Ber., 1993, 126, 2397–2401.
We have reported the synthesis of an yttrium phosphonium
ylide adduct and its reaction with benzophenone to form a
hydrocarbon soluble metal-coordinated betaine (the C–C s-bond
formation step of the Wittig reaction under salt conditions). Upon
addition of triphenylphosphine oxide to 2 C–C p-bond formation is
observed. We are continuing to study this system and the reaction
chemistry of onium ylides coordinated to electrophilic metal centres
and will report our findings in subsequent publications.
26 P. L. Watson and G. W. Parshall, Acc. Chem. Res., 1985, 18,
51–56.
27 P. L. Watson, J. Chem. Soc., Chem. Commun., 1983, 276.
28 H. Schumann, I. Albrecht, F. W. Reier and E. Hahn, Angew.
Chem., Int. Ed. Engl., 1984, 23, 522–523.
29 H. Schumann and F. W. Reier, J. Organomet. Chem., 1981,
209, C10.
30 H. Schumann and F. W. Reier, J. Organomet. Chem., 1982,
235, 287.
31 H. Schumann, F. W. Reier and M. Dettlaff, J. Organomet. Chem.,
1983, 255, 305–310.
32 H. Schumann and F. W. Reier, J. Organomet. Chem., 1984, 269,
21–27.
M.R.C is indebted to the Royal Society for the provision of
a University Research Fellowship.
33 H. Schmidbaur, Angew. Chem., Int. Ed. Engl., 1983, 22, 907–927.
34 H. Schmidbaur, Acc. Chem. Res., 1975, 8, 62–70.
35 L. E. Manzer, Inorg. Chem., 1976, 15, 2567–2569.
36 P. Berno, S. Gambarotta, S. Kotila and G. Erker, Chem. Commun.,
1996, 779–780.
37 R. Fandos, M. Gomez, P. Royo, S. Garciablanco, S. Martinezcarrera
and J. Sanzaparicio, Organometallics, 1987, 6, 1581–1583.
38 P. L. Arnold and I. J. Casely, Chem. Rev., 2009, 109, 3599–3611.
39 W. Fegler, T. P. Spaniol and J. Okuda, Dalton Trans., 2010, 39,
6774–6779.
40 J. C. J. Bart, J. Chem. Soc. B, 1969, 350.
41 W.-K. Wong, J. Guan, J. Ren and Q. Shen, Polyhedron, 1993, 12,
2749–2751.
42 W.-K. Wong, J. Guan, Q. Shen, L. Zhang and Y. Lin, Polyhedron,
1995, 14, 277–283.
43 P. L. Arnold, Z. R. Turner, R. Bellabarba and R. P. Tooze, J. Am.
Chem. Soc., 2011, 133, 11744–11756.
44 Z. R. Turner, R. Bellabarba, R. P. Tooze and P. L. Arnold, J. Am.
Chem. Soc., 2010, 132, 4050–4051.
45 Y. Yang, D. M. Cui and X. S. Chen, Dalton Trans., 2010, 39, 3959.
46 Y. Nakajima and J. Okuda, Organometallics, 2007, 26, 1270–1278.
47 A. Bondi, J. Phys. Chem., 1964, 68, 441–451.
48 D. J. Mindiola and J. Scott, Nat. Chem., 2011, 3, 15–17.
49 D. P. Mills, L. Soutar, W. Lewis, A. J. Blake and S. T. Liddle,
J. Am. Chem. Soc., 2010, 132, 14379–14381.
Notes and references
z Crystal data for 2: C44H63IN2OPSi4Y, M = 995.10, monoclinic,
P21/n (no. 14), a = 12.5750(4), b = 24.9990(7), c = 15.7491(4) A, b =
91.687(3)1, V = 4948.8(2) A3, Z = 4, Dc = 1.336 g cmꢀ3, m(Mo-Ka) =
1.964 mmꢀ1, T = 173 K, colourless needles, Oxford Diffraction
Xcalibur 3 diffractometer; 15 749 independent measured reflections
(Rint = 0.0293), F2 refinement, R1(obs) = 0.0359, wR2(all) = 0.0787,
11 490 independent observed absorption corrected reflections
[|Fo| 4 4s(|Fo|), 2y max = 641], 487 parameters. CCDC 843184.
y Crystal data for 1: C31H53IN2PSi4Y, M = 812.89, monoclinic, P21/n
(no. 14), a = 8.8866(18), b = 12.252(3), c = 36.576(7) A, b =
92.160(4)1, V = 3979.4(14) A3, Z = 4, Dc = 1.357 g cmꢀ3, m(Mo-Ka) =
2.423 mmꢀ1, T = 150 K, colourless needles, Bruker SMART APEX
CCD diffractometer; 9304 independent measured reflections (Rint =
0.0981), F2 refinement, R1(obs) = 0.0558, wR2(all) = 0.1604. CCDC
843183.
1 E. Vedejs and C. F. Marth, J. Am. Chem. Soc., 1990, 112, 3905–3909.
2 E. Vedejs, G. P. Meier and K. A. J. Snoble, J. Am. Chem. Soc.,
1981, 103, 2823–2831.
3 E. Vedejs and K. A. J. Snoble, J. Am. Chem. Soc., 1973, 95, 5778–5780.
4 A. B. Reitz, M. S. Mutter and B. E. Maryanoff, J. Am. Chem. Soc.,
1984, 106, 1873–1875.
5 F. Ramirez, C. P. Smith and J. F. Pilot, J. Am. Chem. Soc., 1968,
90, 6726–6732.
6 R. W. Saalfrank, W. Paul and H. Liebenow, Angew. Chem., Int.
Ed. Engl., 1980, 19, 713–714.
50 M. Fustier, X. F. Le Goff, P. Le Floch and N. Mezailles, J. Am.
Chem. Soc., 2010, 132, 13108–13110.
51 R. Litlabo, M. Zimmermann, K. Saliu, J. Takats, K. W. Tornroos
and R. Anwander, Angew. Chem., Int. Ed., 2008, 47, 9560–9564.
52 M. Schlosser and K. F. Christmann, Angew. Chem., Int. Ed. Engl.,
1965, 4, 689.
7 H. J. Bestmann, K. Roth, E. Wilhelm, R. Bohme and H. Burzlaff,
Angew. Chem., Int. Ed. Engl., 1979, 18, 876–877.
8 G. H. Birum and C. N. Matthews, Chem. Commun., 1967, 137.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1745–1747 1747