Sequestered alkyllithiums: Why phenyllithium alone is suitable for betaine-ylide generation

Article abstract of DOI:10.1002/chem.200390061

The key step in the trans-selective modification of the Wittig reaction is the α-lithiation of the lithium bromide coordinated ylide - aldehyde adduct (the so-called "P-betaine"). Only phenyllithium effects this deprotonation rapidly and cleanly. Alkyllithiums (in particular, butyl-, sec-butyl-, and tert-butyllithium) react only sluggishly and incompletely, being tied up in very stable mixed aggregates with the lithium alkoxide part of the betaines.

Full text of DOI:10.1002/chem.200390061

FULL PAPER
Sequestered Alkyllithiums: Why Phenyllithium Alone is Suitable for
Betaine-Ylide Generation
Qian Wang,[a] Dariusz Deredas,[a] Cyril Huynh,[a] and Manfred Schlosser*[a, b]
Abstract: The key step in the trans-selective modification of the Wittig reaction is
the a-lithiation of the lithium bromide coordinated ylide aldehyde adduct (the so-
called ™P-betaine∫). Only phenyllithium effects this deprotonation rapidly and
cleanly. Alkyllithiums (in particular, butyl-, sec-butyl-, and tert-butyllithium) react
only sluggishly and incompletely, being tied up in very stable mixed aggregates with
the lithium alkoxide part of the betaines.
Keywords: betaines
labeling ¥ mixed aggregates ¥ Wittig
reactions ¥ ylides
¥
deuterium
Introduction
™Betaine ylides∫ (triphenylphosphonio-2-lithiooxyalkanides)
are generated by the a-deprotonation of lithium bromide
complexed ™phosphorus betaines∫, the zwitterionic adducts of
phosphorus ylides and carbonyl compounds. They act as the
3]
key intermediates in trans-selective[1 and ™three-dimen-
7]
sional∫ (™SCOOPY∫)[3 Wittig reactions (see Scheme 1).
Stereocontrol is accomplished by the spontaneous pyramidal
inversion of the a-lithiated betaine. At equilibrium, the threo
configuration is strongly favored and typically predominates
to the extent of ꢀ99.5%.[1, 2] In contrast, the erythro
component is formed preferentially (respectively, ꢁ2:1 and
ꢀ10:1 in the presence and absence of soluble lithium salts) if
the ylide and aldehyde combine under irreversible condi-
tions.[1, 3, 8, 9]
The crucial epimerization at the a-carbon is just one stage
in a multistep one-pot reaction sequence (see Scheme 2). Its
success critically depends on a high concentration of a soluble
lithium salt, in general lithium bromide. Its role is to suppress
the dissociation of the a-lithiated betaine ylide to an a-metal-
free betaine ylide.[10] This rules out the unconsidered use of
any commercial organolithium reagent and requires ethereal
media that are rich in tetrahydrofuran. If these conditions are
met, the trans-selective modification of the Wittig olefination
protocol works with absolute reliability as impressive appli-
cations of it to the synthesis of natural products and
Scheme 1. Betaine-ylides as the key intermediates in trans-selective and
™three-dimensional∫ olefination reactions.
At first sight, the exact nature of the organometallic reagent
would not be expected to matter. Any alkyl- or aryllithium
should be basic enough to enable the complete deprotonation
of the phosphorus betaine, the acidity of which can be
assumed to approximate that of ordinary alkyltriphenylphos-
phonium salts (pKa ꢁ20[17 19]). The consecutive treatment of
(w-hydroxyalkyl)triphenylphosphonium bromide with lithi-
um bromide containing butyllithium, an aldehyde and butyl-
lithium again has indeed been reported to give rise, after
16]
biomimetics have been amply demonstrated.[11
[a] Prof. Dr. M. Schlosser, Dr. Q. Wang, Dr. D. Deredas, C. Huynh
¬
Institut de Chimie Organique, Universite, BCh
1015 Lausanne (Switzerland)
E-mail: manfred.schlosser@epfl.ch
[b] Prof. Dr. M. Schlosser
¬
Institut de Chimie Moleculaire et Biologique
¬
¬
Ecole Polytechnique Federale, BCh
1015 Lausanne (Switzerland)
570
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0570 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2

570 574
Table 1. trans-Selective Wittig reactions involving betaine-ylides prepared
7
from ylides (H5C6)3P+ CH R'and aldehydes R CH O: Z/E ratios and, in
À
À
À
ˆ
parentheses, yields of olefins.[a]
R
R'
LiC6H5
LiCH3
3:97
LiC4H9 LIS[b] LIT[c]
50:50 33:67 17:83
H13C6
C4H9
< 0.5:99.5
(88%)[d] (87%)
< 0.5:99.5
(88%)[d] (92%)
(92%) (86%) (84%)
1.5:98.5 12:88 6:94 23:77
H5C6
C3H7
(77%) (63%) (60%)
[e]
ˆ À
H2C C CH3
(H3C)2CH
(H3C)3C
(CH2)2C6H5 < 0.5:99.5
2:98
16:84
12:88 36:64
(58%)[d] (59%)
< 0.5:99.5
(66%)[f]
94:6
(57%) (45%) (42%)
C5H11
C5H11
2:98
(59%)
98:2
16:84
(57%)
97:3
13:87
(62%)
76:24
(59%)
(63%)
(56%)
(42%)
[a] Standard working procedure: see Experimental Section. At the decisive
stage of betaine ylide formation, the tetrahydrofuran/diethyl ether ratio
approached 1:1. [b] LIS ˆ LiCH(CH3)C2H5 (sec-butyllithium). [c] LITˆ
LiC(CH3)3 (tert-butyllithium). [d] The same limit of stereoselectivity and
virtually the same yields were attained when the reaction was performed in
an approximate 7:1:2 (v/v/v) mixture of tetrahydrofuran, diethyl ether, and
cyclohexane, respectively. [e] 2-Methyl-2-propenal (methacroleine). [f] A Z/E
ratio of 3:97 and a yield of 65% were found when the reaction was performed in
an approximate 7:1:2 (v/v/v) mixture of tetrahydrofuran, diethyl ether, and
cyclohexane, respectively.
favors cis olefination under all circumstances, however
(Table 1). Pivaldehyde is known to produce exceptionally
high erythro/threo ratios, and, consequently, in the absence of
an efficient epimerization mechanism, high cis/trans ratios
Scheme 2. The multistep, if one-pot protocol for trans-selective olefin
formation. a) Lithium bromide containing phenyllithium (or any other
22]
when allowed to react with any kind of ylide.[20
À
ˆ
organolithium) in THF and diethyl ether. b) Aldehyde R CH O. c) Lith-
ium bromide containing phenyllithium in diethyl ether. d) Either 1 min at
258C or 30 min at À758C. e) Hydrogen chloride (1.0 equiv) in diethyl ether
at À758C. f) Potassium tert-butoxide. g) Some 15 min at À258C.
At first sight one might feel tempted to rationalize the
reagent effect on the E stereoselectivity by side reactions
reflecting structural individuality of the organometallics
involved. In particular, the highly reactive family of butyl-
lithiums (butyllithium, sec-butyllithium, and tert-butyllithium)
may promote substitution of a ™stationary∫[1, 3, 23] P-substitu-
ent rather than a-deprotonation. Following well-established
precedents, this group displacement at phosphorus may occur
neutralization and decomplexation, to the highly trans-
selective (Z/E ꢂ3:97) formation of alken-1-ols.[2] However,
as we have recognized meanwhile, only phenyllithium ensures
maximum stereoselectivities throughout.
The results are clear-cut. Straight-chain and monobranched
aliphatic aldehydes (heptanal, isobutyraldehyde) as well as
a,b-unsaturated aldehydes (methacroleine, benzaldehyde)
give olefins with excellent (Z/E < 1:200), acceptable (Z/E
ꢁ50), and poor (Z/E 16:1 1:1) trans selectivity depending on
what reagent was employed to generate the betaine ylide,
phenyllithium, methyllithium, or one of the butyllithiums
(butyllithium, sec-butyllithium, or tert-butyllithium), respec-
tively (see Table 1). The doubly a-branched pivaldehyde
by addition of the organolithium to the betaine and subse-
27]
quent elimination of benzene,[24
or, alternatively, by
phenyl/alkyl exchange at the level of an ylide,[28 including
a betaine ylide. The resulting P-alkyl derivative, if a betaine,
would be reluctant to undergo the required a-deprotonation,
or, if a betaine ylide, would no longer discriminate so strictly
against the erythro configuration. To account for the fairly
good performance of methyllithium in the framework of this
hypothesis, one would merely have to assume a more
pronounced decrease in nucleophilicity than in basicity with
this reagent. Phenyllithium would clearly avoid such compli-
cations as it could not cause any structural alteration if
involved in group displacement at phosphorus.
29]
Abstract in German: Der Schl¸sselschritt bei der trans-
selektiven Variante der Wittig-Reaktion ist die a-Deprotonie-
rung des mit Lithiumbromid koordinierten Ylid/Aldehyd-
Addukts (des sogenannten ™P-Betains∫). Nur Phenyllithium
vermag diese Umsetzung rasch und sauber zu bewerkstelligen.
Aliphatische Organolithium-Verbindungen (insbesondere, Bu-
tyl-, sec-Butyl- und tert-Butyllithium) reagieren nur schleppend
und unvollst‰ndig, weil ihnen Fesseln angelegt sind durch den
Lithiumalkoholat-Teil der Betaine, womit sie sehr stark
gebundene Mischaggregate bilden.
Despite its prima facie plausibility, the possibility of
P-substituent displacement had to be ruled out for two
reasons. For one thing, we were unable to detect even trace
amounts of alkyldiphenylphosphine oxides in the crude
reaction mixtures when the trans-selective olefination proto-
col had been performed with a butyllithium variety. More-
over, we have consecutively treated butyltriphenylphospho-
nium bromide, which was perdeuterated at all aromatic
positions, with unlabeled phenyllithium/lithium bromide,
benzaldehyde, phenyllithium/lithium bromide (again), hydro-
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0571 $ 20.00+.50/0
571

M. Schlosser et al.
FULL PAPER
gen chloride, and potassium tert-butoxide. The phosphonium
salt had been prepared by quaternization of perdeuterated
triphenylphosphonium, which was made from hexadeutero-
tight tetramer.[33] The homoaggregates of primary, secondary,
and tertiary alkyllithiums being less optimally constructed,[33]
heteroaggregation evidently offers better binding interactions
32]
benzene by metalation with the LIC-KOR superbase.[30
in such a situation. Lithium alcoholates are known to bind to
38]
The triphenylphosphine oxide isolated was found to consist of
the [2H15] and [2H10] isomers in a ratio of 99:1. In other words,
virtually all of the betaine ylide had been formed by direct
deprotonation rather than through a transient phosphorane
(Scheme 3).
organolithiums under mixed aggregate formation.[34
Ac-
cordingly, all members of the butyllithium family combine
strongly with betaine/lithium bromides, which represent a
special class of oligomeric alcoholates. The resulting mixed
superclusters, tied together by multiple electron-deficient
bonds, are so stable that the sequestered alkyllithium
component loses some of its inherent reactivity, being no
longer able to bring about complete betaine ylide formation
(Scheme 4).
Scheme 4. Betaine/lithium bromide adducts sequestering alkyllithium by
heteroaggregate formation.
Scheme 3. Phenyllithium acting towards a betaine/lithium bromide adduct
almost exclusively as a base and only marginally as a nucleophile.
Salt-free phenyllithium is commercially available in a 3:7
(v/v) diethyl ether/cyclohexane solvent mixture at a moderate
price (E125 per mol). It may be employed as a reagent for
trans-selective Wittig reactions if, after the evaporation of the
original solvents, the residue is taken up in tetrahydrofuran
and sufficient amounts of anhydrous lithium bromide are
added. It is more convenient to use ™homemade∫ material,
Evidently, the next step was to monitor the reaction
between betaines and butyllithiums. This was done in two
ways. On the one hand, we probed for unconsumed organo-
metallic reagent (e.g., butyllithium) by trapping it with
benzaldehyde and identifying the adduct (i.e., 1-phenylpen-
tan-1-ol). On the other hand, we quantified the amount of
isotope incorporation after successive addition of the alkyl-
lithium (for 1 h at À758C), deuterium chloride, and potassium
tert-butoxide to the betaine/lithium bromide adduct.
which can be easily prepared from bromobenzene in tetrahy-
43]
drofuran[39, 40] or diethyl ether,[41
the precious by-product
lithium bromide remaining in solution in either solvent.
For example, consecutive treatment of butyltriphenylphos-
phonium bromide in the presence of excess lithium bromide,
with phenyllithium, benzaldehyde, phenyllithium (again),
deuterium chloride, and potassium tert-butoxide afforded
pure (E)-2-[2H]1-phenyl-1-pentene (88%). When only the
first time (ylide generation) phenyllithium, but the second
time (betaine ylide generation) tert-butyllithium was em-
ployed as the base, the olefin (60%) was obtained as a 1:3
(Z/E) mixture which could be easily separated by column
chromatography. The cis isomer was found to contain no
deuterium at all, whereas the trans isomer was labeled to the
extent of 40%.
Aconsistent picture emerges from these findings. Both
deuteration and trans selectivity depend on betaine ylide
generation as a prerequisite. The a-deprotonation of the
betaine precursor can be cleanly achieved with phenyllithium,
which shows little propensity to aggregate,[33] but also, if less
perfectly, with methyllithium, which exists as an exceptionally
Experimental Section
For working habits and abbreviations, see recent publications (e.g.,
ref. [44]) which have emanated from this laboratory. 1H, 13C, and 31P
NMR spectra were recorded of samples dissolved in deuterochloroform at
400, 101, and 162 MHz, respectively, chemical shifts being given relative to
the internal standard tetramethylsilane (1H and 13C) or the external
standard 85% aqueous phosphoric acid (31P).
trans-Selective Wittig reactions: A tÀ758C, a 0.70m solution of phenyl-
lithium (28 mmol) containing one molar equivalent of lithium bromide
(28 mmol) in ethyl ether (40 mL) was added to the alkyltriphenylphos-
phonium bromide (25 mmol) in tetrahydrofuran (80 mL). After 30 min of
vigorous stirring at 258C, the reaction mixture was cooled to À758C to be
treated with the aldehyde (25 mmol) and, a few minutes later, when
complete decolorization had occurred, again with the phenyllithium/
lithium bromide complex (28 mmol) in diethyl ether (40 mL). Aclear
solution was obtained that rapidly turned cherry-red when stored each time
30 min at À758C, 258C, and again À758C. Hydrogen chloride (28 mmol) in
diethyl ether (10 mL) decolorized it instantaneously. After addition of
potassium tert-butoxide (3.4 g, 30 mmol), the reaction mixture was stirred
1 h at 258C before being poured into water (59 mL) and extracted with
572
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0572 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2

Betaine Ylide Generation
570 574
diethyl ether (3 Â 25 mL). The combined organic layers were washed with
brine (2 Â 15 mL), dried, and evaporated. The semisolid residue was
triturated with pentanes (3 Â 25 mL). The extracts were filtered and
distilled. The olefinic product was identified and its Z/E ratio determined
by comparison of the gas chromatographic retention times with those of
authentic samples: 5-dodecene[45] (b.p. 70 718C/10 Torr; n2D0 1.42998; 50 m,
MSV, 30 m, DB-210, 408C), 1-phenyl-1-pentene[46] (b.p. 111 1138C/
10 Torr; n2D01.5287; 3 m, 5% C-20M, 1608C; 3 m, 5% SE-30, 1708C),
2-methyl-6-phenyl-1,3-hexadiene[47] (b.p. 75 788C/0.5 Torr; nD20 1.5404;
30 m, DB-1, 808C; 30 m, DB-FFAP, 808C), 2-methyl-3-nonene[48] (b.p.
55 578C/17 Torr; n2D0 1.4217; 30 m, DB-1, 308C; 50 m, MSV, 408C), and
2,2-dimethyl-3-nonene[49] (b.p. 65 678C/20 Torr; n2D0 1.4281; 30 m, DB
1701, 308C; 30 m, DB-FFAP, 308C).
When the temperature had reached 258C, the organic layer was washed
with a 1.0m aqueous solution (3 Â 0.10 L) of sodium hydroxide and brine
(2 Â 50 mL) before being dried and evaporated. Recrystallization of the
residue from hexanes gave colorless platelets; 21 g (75%); m.p. 74 768C
(ref. [52]: m.p. 768C/16 Torr); 13C NMR: d ˆ 136.9 (d, J ˆ 11 Hz), 133.2 (td,
J ˆ 24, 5 Hz), 128.1 (t, J ˆ 25 Hz), 127.9 ppm (td, J ˆ 24, 6 Hz); 31P
NMR:d ˆ 5.5 ppm; elemental analysis calcd (%) for C18D15P (277.38): C
77.94, H 6.06; found: C 78.35, H 5.72.
Butyl(tri[2H5]phenyl)phosphonium bromide: Tri([2H5]phenyl)phosphine
(6.8 g, 25 mmol), 1-bromobutane (3.0 mL, 3.8 g, 28 mmol), and toluene
(10 mL) were mixed and heated 6 h to 1258C. At 25 8C, the syrup formed
was dissolved in dichloromethane, and warm ethyl acetate was added until
the solution became turbid. Upon scratching with a glass rod, crystalliza-
tion set in. Recrystallization from dichloromethane and ethyl acetate gave
colorless granules. 9.1 g (88%); m.p. (decomp) 230 2328C; 1H NMR: d ˆ
3.7 (m, 2H), 1.6 (m, 4H), 0.92 (t, J ˆ 7.1 Hz, 3H); 13C NMR: d ˆ 134.4 (d,
J ˆ 12 Hz), 133.0 (dt, J ˆ 12, 5 Hz), 129.8 (td, J ˆ 12, 6 Hz), 117.8 (d, J ˆ
86 Hz), 24.4 (d, J ˆ 5 Hz), 23.5 (d, J ˆ 17 Hz), 22.4 (d, J ˆ 51 Hz), 13.6 ppm
(s); 31P NMR: d ˆ 24.7 ppm; elemental analysis calcd (%) for C22H9D15BrP
(414.40): C 63.76, H 6.24; found: C 64.03, H 6.11.
To prepare solutions of alkyllithiums containing one molar equivalent of
lithium bromide, the original solvents (pentanes, hexanes, or cyclohexanes)
were stripped off from the commercial reagents (0.70 mol). The residue was
then dissolved in the precooled (À758C) solution of lithium bromide (61 g,
0.70 mol) in diethyl ether (1.0 L).
To determine the water content (in general some 5%) of commercial
lithium bromide, a sample was dissolved in anhydrous tetrahydrofuran and
titrated with butyllithium in the spatula tip of triphenylmethane as an
indicator until the cherry red color (of triphenylmethyllithium) persisted.
The protocol had to be modified when the ™homemade∫ lithium bromide
containing ethereal phenyllithium (see above) was replaced by the
commercial product (1.8m in a 3:7 mixture of diethyl ether and cyclo-
hexane, twice 15 mL). The missing lithium bromide (4.8 g, 55 mmol) was
dissolved in tetrahydrofuran (15 mL) and was added from the beginning to
phosphonium salt (suspended in 60 mL tetrahydrofuran).
Acknowledgement
This work was supported by the Schweizerische Nationalfonds zur
Fˆrderung der wissenschaftlichen Forschung, Bern (grants 20-49'307-96
and 20-55'303-98). The authors also wish to express their appreciation to
Professor Yu-fen Zhao at the Chemistry Departments of Tsinghua
University, Beijing, and of Xiamen University. She offered Q.W. the
opportunity to pursue his graduate studies during two years in Lausanne.
Deactivation of butyllithium by coordination: A tÀ758C, pivaldehyde
(1.1 mL, 0.86 g, 10 mmol) was added dropwise to a solution of butyllithium
(10 mmol) in tetrahydrofuran (30 mL) and diethyl ether (50 mL). After
5 min, the mixture was treated with hydrogen chloride (11 mmol) in diethyl
ether (10 mL) before being poured into water (20 mL). The aqueous phase
was saturated with sodium chloride. The organic layer was separated and
analyzed by gas chromatography (30 m, DB-1701, 808C; 30 m, DB-FFAP,
[1] M. Schlosser, K. F. Christmann, Liebigs Ann. Chem. 1967, 708, 1 35.
[2] M. Schlosser, T. Huynh Ba, B. Schaub, Tetrahedron Lett. 1985, 26,
311 314.
[3] M. Schlosser, Top. Stereochem. 1970, 5, 1 30.
[4] M. Schlosser, K. F. Christmann, Synthesis 1969, 38 39.
[5] M. Schlosser, K. F. Christmann, A. Piskala, D. Coffinet, Synthesis
1971, 29 31.
[6] M. Schlosser, D. Coffinet, Synthesis 1971, 380 381.
[7] M. Schlosser, D. Coffinet, Synthesis 1972, 575 576.
[8] M. Schlosser, B. Schaub, J. de Oliveira-Neto, S. Jeganathan, Chimia
1986, 40, 244 245.
[9] B. Schaub, S. Jeganathan, M. Schlosser, Chimia 1986, 40, 246 247.
[10] M. Schlosser, K. F. Christmann, A. Piskala, Chem. Ber. 1970, 103,
2814 2820.
[11] W. S. Johnson, Angew. Chem. 1976, 88, 33 41; Angew. Chem. Int. Ed.
Engl. 1976, 15, 9 25.
[12] M. B. Gravenstock, W. S. Johnson, B. E. McCarry, R. J. Parry, B. E.
Ratcliffe, J. Am. Chem. Soc. 1978, 100, 4274 4282.
[13] W. S. Johnson, L. R. Hughes, J. A. Kloeck, T. Niem, A. Shenvi, J. Am.
Chem. Soc. 1979, 101, 1279 1281.
[14] W. S. Johnson, B. E. McCarry, R. L. Markezich, S. G. Boots, J. Am.
Chem. Soc. 1980, 102, 352 359.
[15] W. S. Johnson, C. Newton, S. D. Lindell, Tetrahedron Lett. 1986, 27,
6027 6030.
808C); octan-1-ol as
a calibrated ™internal standard∫) revealing the
presence of 99% of 2,2-dimethyl-3-heptanol.[50]
In parallel reaction, again conducted at À758C, 2,4-dimethylpentan-3-ol
(10 mmol) was added to butyllithium (20 mmol) in tetrahydrofuran
(30 mL) and diethyl ether (5.0 mL), followed 15 min later by pivalaldehyde
(10 mmol). 2,2-Dimethylheptan-3-ol[50] was present this time in 68% yield.
In
a third experiment, pentyltriphenylphosphonium bromide (4.1 g,
10 mmol) in tetrahydrofuran (20 mL) was consecutively treated with
lithium bromide containing phenyllithium (10 mmol) in tetrahydrofuran
(8 mL) and diethyl ether (4 mL) and with heptanol (10 mmol). After
decolorization and always at À758C, butyllithium (10 mmol) in tetrahy-
drofuran (5 mL) and, 15 min later, pivalaldehyde (10 mmol) were added.
The quantity of 2,2-dimethylheptan-3-ol formed amounted to 16%.
Perdeuterophenyl substituted phosphonium salts:
[2H5]Iodobenzene: The mixture of [2H6]benzene (22 mL, 21 g, 0.25 mol;
99% isotopically pure), butyllithium (0.25 mol) in pentanes (25 mL) and
potassium tert-butoxide (28 g, 0.25 mol) was vigorously stirred for 20 h at
08C. At À758C, precooled tetrahydrofuran (0.10 mL) was added. After
15 min of continuous stirring a homogeneous solution was obtained.
Always at À758C, a solution of iodine (64 g, 0.25 mol) in tetrahydrofuran
(0.15 L) was added dropwise over the course of 30 min. According to gas
chromatographic analysis (30 m, DB-1701, 808C; 30 m, DB-FFAP, 808C),
iodobenzene (34%) was present in the reaction mixture. The solvent and
unconsumed benzene were removed by distillation. The residue was
dissolved in diethyl ether (50 mL) and thoroughly washed with a saturated
aqueous solution of sodium thiosulfate (3 Â 50 mL) and with brine (2 Â
25 mL), was dried and evaporated. Distillation afforded a colorless liquid;
14.6 g (28%); b.p. 90 928C/45 Torr (ref. [51]: b.p. 67 688C/16 Torr); nD20
1.6150.
Tri([2H5]phenyl)phosphine: A tÀ758C, [2H5]iodobenzene (63 g, 0.30 mol,
ꢁ99% isotopically pure) was added dropwise, over the course of 15 min, to
tert-butyllithium (0.60 mol), from which the commercial solvent (pentanes)
had been removed by evaporation and which then was dissolved in
precooled diethyl ether (0.30 L). Still at À758C and 15 min later, the
mixture was treated with triphenyl phosphite (26 mL, 31 g, 0.10 mol).
[16] W. S. Johnson, Y. Q. Chen, M. S. Kellogg, J. Am. Chem. Soc. 1983, 105,
6653 6656.
[17] K. Issleib, R. Lindner, Liebigs Ann. Chem. 1967, 707, 120 129.
[18] F. G. Bordwell, Acc. Chem. Res. 1988, 21, 456 463.
[19] T. A. Mastryukova, A. G. Matveeva, A. A. Grigoreva, G. F. Litov-
chenko, E. I. Matrosov, I. M. Aladzheva, O. V. Bukhovskaya, I. V.
Leonteva, M. I. Kabachnik, Zh. Obshch. Khim. 1991, 61, 2427 2433;
Russ. J. Gen. Chem. 1991, 61, 2253 2257; Chem. Abstr. 1992, 116,
194449k.
[20] V. Patil, M. Schlosser, Synlett 1993, 125 126.
[21] X.-p. Zhang, M. Schlosser, Tetrahedron Lett. 1993, 34, 1925 1928.
¬
[22] X.-p. Zhang, Ph.D. thesis, Universite de Lausanne, 1994, pp. 17 24.
[23] M. Schlosser, B. Schaub, J. Am. Chem. Soc. 1982, 104, 5821 5823.
[24] G. Wittig, G. Gei˚ler, Liebigs Ann. Chem. 1953, 580, 44 57.
Chem. Eur. J. 2003, 9, No. 2
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0573 $ 20.00+.50/0
573

M. Schlosser et al.
FULL PAPER
[25] D. Seyferth, M. A. Eisert, J. K. Heeren, J. Organomet. Chem. 1964, 2,
101 108.
[38] B. Goldfuss, S. I. Khan, K. N. Houk, Organometallics 1999, 18, 2927
2929.
[26] D. Seyferth, J. K. Heeren, W. B. Hughes, J. Am. Chem. Soc. 1965, 87,
2847 2854.
[27] D. Seyferth, W. B. Hughes, J. K. Heeren, J. Am. Chem. Soc. 1965, 87,
3467 3474.
[28] M. Schlosser, Angew. Chem. 1962, 74, 291 291; Angew. Chem. Int.
Ed. Engl. 1962, 1, 266 266.
[29] B. Schaub, T. Jenny, M. Schlosser, Tetrahedron Lett. 1984, 25, 4097
4100.
[30] M. Schlosser, J. Organomet. Chem. 1967, 8, 9 16.
[31] M. Schlosser, Pure Appl. Chem. 1988, 60, 1627 1634.
[32] M. Schlosser, Mod. Synth. Methods 1992, 6, 227 271.
[33] M. Schlosser in Organometallics in Synthesis: A Manual, 2nd ed. (Ed.:
M. Schlosser), Wiley, Chichester, 2002, pp. 1 352, specifically pp. 21
22.
[39] H. Gilman, B. J. Gaj, J. Org. Chem. 1957, 22, 1165 1168.
[40] R. Waack, M. A. Doran, J. Org. Chem. 1967, 32, 3395 3399.
[41] G. Wittig, Angew. Chem. 1940, 53, 242 247.
[42] H. Gilman, J. W. Morton, Org. React. 1954, 8, 258 304, specifically
p. 286.
[43] M. Schlosser in Organometallics in Synthesis: A Manual, 2nd ed. (Ed.:
M. Schlosser), Wiley, Chichester, 2002, p. 100.
[44] C. Bobbio, M. Schlosser, Eur. J. Org. Chem. 2001, 4533 4536.
[45] H. C. Brown, D. Basavaiah, J. Org. Chem. 1982, 47, 754 756.
[46] G. Wittig, M. Schlosser, Chem. Ber. 1961, 94, 1373 1383.
[47] E. Vedejs, W.-f. Huang, J. Org. Chem. 1984, 49, 210 212.
¬
[48] J. Zavada, M. Pankova, J. Sicher, Collect. Czech. Chem. Commun.
1973, 38, 2102 2120.
¬
[49] J. Zavada, M. Pankova, J. Sicher, Collect. Czech. Chem. Commun.
[34] T. L. Brown, J. A. Ladd, G. N. Newman, J. Organomet. Chem. 1965, 3,
1972, 37, 2414 2435.
1
6.
[50] W. F. Foley, F. J. Welch, E. M. LaCombe, H. S. Mosher, J. Am. Chem.
Soc. 1959, 81, 2779 2784.
[51] T. Joh, K. Doyama, K. Fujiwara, K. Maeshima, S. Takahahi, Organo-
metallics 1991, 10, 508 513.
[35] H. Dietrich, D. Rewicki, J. Organomet. Chem. 1981, 205, 281 289.
[36] M. Marsch, K. Harms, L. Lochmann, G. Boche, Angew. Chem. 1990,
102, 334 335; Angew. Chem. Int. Ed. Engl. 1990, 29, 308 309.
[37] G. T. DeLong, D. K. Pannell, M. T. Clark, R. D. Thomas, J. Am. Chem.
Soc. 1993, 115, 7013 7014.
[52] M. A. Bennett, D. L. Milner, J. Am. Chem. Soc. 1969, 91, 6983 6994.
Received: July 5, 2002 [F4230]
574
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0902-0574 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 2