Lithium carbenoids-ultra-reactive yet selective reagents for methylenation and halomethylenation of sulfones

Article abstract of DOI:10.1021/jo060669+

The first efficient, one-pot method for methylenation of p-toluyl sulfones (i.e., the transformation of p-MePhSO₂CHR₂ into R ₂CCH₂) is described. Methods for effecting that transformation involving alkylation of sulfones with reagents of the general formula MCH₂X, where M = SiMe₃, SnR₃, and MgCl (the Julia method) have been previously described. However, the silicon reagent is completely unreactive toward many sulfones, the tin reagent typically affords only moderate yields, and the magnesium reagent typically gives incomplete reaction (9-40% starting material). This article reports that excellent yields are obtained by alkylating the sulfone with the more ionic lithium methylene carbenoids (LiCH₂X, where X = Cl, Br, and I). For example, treatment of sulfone 1 with 3 equiv of n-BuLi (THF, -100 °C) followed by 2 equiv of CH₂Br₂ affords olefin 2 in 96% yield. Although the carbenoid is generated in the presence of a 1:2 mixture of α-lithio- sulfone and n-BuLi, it reacts selectively with the α-lithio-sulfone (K_rel = 9.6 ± 0.8 with LiCH₂Br). The chlorocarbenoid LiCH₂Cl reacts somewhat less selectively (K _rel = 3.5). The alkylsubstituted carbenoid n-BuCHBrLi reacts ≥ 40 times more slowly than LiCH₂Br, suggesting that olefination occurs by the S_N2 mechanism rather than by metal-assisted ionization (MAI), the mechanism by which cyclopropyl and vinylidene carbenoids react with nucleophiles. Finally, sulfones can be transformed into vinyl halides by treatment with 3 equiv of n-BuLi (THF, -78 °C), followed by >2 equiv of diisopropylamine, followed by 3 equiv of dihalomethane (CH₂X ₂, where X₂ = Cl₂, BrCl, Br₂, or I₂). All products are formed in high yield and purity.

Full text of DOI:10.1021/jo060669+

Lithium Carbenoids-Ultra-Reactive Yet Selective Reagents for
Methylenation and Halomethylenation of Sulfones1
Bruce A. Pearlman,*,† Sterling R. Putt, and Jeffrey A. Fleming
Chemical Research and DeVelopment, 31073-091-201, Pfizer, Incorporated, 7000 Portage Road,
Kalamazoo, Michigan 49001-0102
bruce.pearlman@spcorp.com
ReceiVed March 28, 2006
The first efficient, one-pot method for methylenation of p-toluyl sulfones (i.e., the transformation of
p-MePhSO2CHR2 into R2CCH2) is described. Methods for effecting that transformation involving alkylation
of sulfones with reagents of the general formula MCH2X, where M ) SiMe3, SnR3, and MgCl (the Julia
method) have been previously described. However, the silicon reagent is completely unreactive toward
many sulfones, the tin reagent typically affords only moderate yields, and the magnesium reagent typically
gives incomplete reaction (9-40% starting material). This article reports that excellent yields are obtained
by alkylating the sulfone with the more ionic lithium methylene carbenoids (LiCH2X, where X ) Cl, Br,
and I). For example, treatment of sulfone 1 with 3 equiv of n-BuLi (THF, -100 °C) followed by 2 equiv
of CH2Br2 affords olefin 2 in 96% yield. Although the carbenoid is generated in the presence of a 1:2
mixture of R-lithio-sulfone and n-BuLi, it reacts selectively with the R-lithio-sulfone (Krel ) 9.6 ( 0.8
with LiCH2Br). The chlorocarbenoid LiCH2Cl reacts somewhat less selectively (Krel ) 3.5). The alkyl-
substituted carbenoid n-BuCHBrLi reacts g 40 times more slowly than LiCH2Br, suggesting that olefination
occurs by the SN2 mechanism rather than by metal-assisted ionization (MAI), the mechanism by which
cyclopropyl and vinylidene carbenoids react with nucleophiles. Finally, sulfones can be transformed into
vinyl halides by treatment with 3 equiv of n-BuLi (THF, -78 °C), followed by >2 equiv of
diisopropylamine, followed by 3 equiv of dihalomethane (CH2X2, where X2 ) Cl2, BrCl, Br2, or I2). All
products are formed in high yield and purity.
SCHEME 1. Olefin Synthesis from Sulfones and “Latent
Carbenes”
Introduction
A family of methods for olefin synthesis conceptually involve
the alkylation of a metalated sulfone with a reagent with “latent
carbene” character (Scheme 1).
Vedejs et al. reported the first example of this class of
olefination reaction, where the “latent carbene” was a tosylhy-
drazone.2 Reich et al. then reported the reaction of R-lithio-
sulfones with acyl silanes to give silyl enol ethers.3 Kocienski
then reported a two-step procedure for the methylenation of
sulfones, consisting of alkylation with Me3SiCH2I, followed by
n-Bu4NF‚3H2O-induced fragmentation.4 Although effective for
many sulfones, a number of complex sulfones were reported to
undergo alkylation in low yield or not at all.5,6
† Current address: Schering-Plough Research Institute, 1011 Morris Ave.,
U-3-2 2100, Union, NJ 07083.
(1) Part 3 of series entitled “Olefin Synthesis by Reaction of Stabilized
Carbanions with Carbene Equivalents.” (a) Part 1: Pearlman, B. A.; Putt,
S. R.; Fleming, J. A. J. Org. Chem. 1985, 50, 3622. (b) Part 2: Pearlman,
B. A.; Putt, S. R.; Fleming, J. A. J. Org. Chem. 1985, 50, 3625.
(2) Vedejs, E.; Dolphin, J. M.; Stolle, W. T. J. Am. Chem. Soc. 1979,
101, 249.
(3) Reich, H. J.; Rusek, J. J.; Olson, R. E. J. Am. Chem. Soc. 1979, 101,
2225.
(4) Kocienski, P. J. Tetrahedron Lett. 1979, 2649.
(5) (a) Kocienski, P. J. J. Org. Chem. 1980, 45, 2037. (b) Kocienski, P.;
Todd, M. J. Chem. Soc., Chem. Commun. 1982, 1078. (c) Kocienski, P. J.;
Todd, M. J. Chem. Soc., Perkin Trans. 1 1983, 1777, 1783.
10.1021/jo060669+ CCC: $33.50 © 2006 American Chemical Society
Published on Web 06/24/2006
5646
J. Org. Chem. 2006, 71, 5646-5657

Lithium Carbenoids for Methylenation of Sulfones
SCHEME 2. Methylenation Step in the Meteneprost Potassium Process
SCHEME 3. Julia Process for the Methylenation of Sulfones
Methylenation of sulfone 1 is a key step in a process to
produce meteneprost potassium (Scheme 2),7 a drug candidate
in development by The Upjohn Company in the early 1980s
for cervical dilatation.
CH2ClBr in THF/Et2O/pentane at -110 °C but is “very
unstable.”14,15 Cainelli et al. found that LiCH2Br is “very
unstable and practically impossible to prepare even at -110
°C.”16 Villieras et al. found that LiCH2Br and LiCH2Cl
decomposed immediately even at -130 °C, but that mixtures
with LiBr are stable for up to 30 min in THF/Et2O/pentane at
-115 °C.17 Yet previous researchers have shown that LiCH2-
Br and LiCH2Cl, when generated in the presence of ketones,
aldehydes,16-18 alkoxyboranes,15,19 and chlorosilanes,20 form
adducts in high yield. These reactions are conducted by adding
n-BuLi or MeLi to a mixture of CH2Br2 or CH2ICl and the
electrophile (“Barbier” procedure). Thus, it was decided to
investigate the Barbier addition mode for methylenation of
R-lithio-sulfones.
Initial attempts to effect this transformation using Me3SiCH2I
resulted in no reaction.1a Aware of precedent that the reactivity
of MCH2X increases with increasing polarization of the C-M
bond,8 we explored the analogous tin reagent n-Bu3SnCH2I and
found that it alkylated sulfone 1 to afford (after fluoride-induced
fragmentation of the adduct) olefin 2 in moderate (77.7%)
yield.1a Fujita et al. independently developed a method for
sulfone methylenation using n-Bu3SnCH2I.9 Subsequently, Julia
and De Lima reported that sulfonyl-stabilized carbanions
undergo alkylation by ClMgCH2Cl (2 equiv; THF, -78 °C f
rt) to afford olefins in one step (Scheme 3).10 However, reactions
were incomplete with all sulfones studied (9-40% unreacted
sulfone). Others have encountered the same problem of incom-
plete reaction.11 Metz et al. have recently overcome this problem
by using >5 molar equivalents of IMgCH2I in the presence of
HMPA.12 Satoh et al. have extended the reaction to substituted
alkylmagnesium carbenoids.13
Speculating that the problem of incomplete reaction might
be solved by switching to a more reactive reagent, we decided
to investigate lithium carbenoids (LiCH2X, where X is Cl, Br,
or I) because lithium is much more ionic than any of the metals
previously studied.
Lithium methylene carbenoids have been extensively studied
and are known to be highly unstable. Koebrich et al. showed
that LiCH2Cl can be generated by the reaction of n-BuLi with
Results
Addition Mode with n-BuLi Added Last. A solution of
sulfone 1 and 3.0 equiv of CH2Br2 in THF was treated with 4.0
equiv of n-BuLi/hexane (THF, e-50 °C) in the hope that the
carbenoid and R-lithio sulfone would be generated at equal rates
and then react with each other efficiently (Scheme 4). In
practice, olefin 2 was formed in 64.6% yield, however, 9â-
methyl sulfone 3a was also formed in 28.4% yield. Presumably,
this byproduct was formed by the reaction of R-lithio-1 with
MeBr, the byproduct of deprotonation of sulfone 1 by LiCH2-
Br (unproven).
Addition Mode with CH2Br2 Added Last. The inverse
addition mode (CH2Br2 added last) was next investigated. With
this addition mode, the carbenoid is generated in the presence
of both R-lithio-1 and n-BuLi, thus raising the issue of whether
the carbenoid would react faster with R-lithio-1 or with n-BuLi.
(6) Metalated sulfones tend to be “somewhat unreactive” toward alkyl-
ation: Magnus, P. D. Tetrahedron 1977, 33, 2019.
(7) Pearlman, B. A. U.S. Patent 4,549,030. Oct. 22, 1985. (Examples
11A and 18).
(8) (a) Bott, R. W.; Eaborn, C.; Swaddle, T. W. J. Organomet. Chem.
1966, 5, 233. (b) Prezzavento, B. A.; Kuivila, H. G. J. Org. Chem. 1987,
52, 929.
(9) (a) Ochiai, M.; Tada, S.; Sumi, K.; Fujita, E. Tetrahedron Lett. 1982,
23, 2205. (b) Ochiai, M.; Sumi, K.; Fujita, E.; Tada, S. Chem. Pharm. Bull.
1983, 31, 3346. (c) Ochiai, M.; Ukita, T.; Fujita, E.; Tada, S. Chem. Pharm.
Bull. 1984, 32, 1829.
(14) (a) Koebrich, G.; Fischer, R. H. Tetrahedron 1968, 24, 4343. (b)
Koebrich, G. Angew. Chem., Int. Ed. Engl. 1972, 11, 473.
(15) Gurskii, M. E.; Baranin, S. V.; Mikhailov, B. M. J. Organomet.
Chem. 1984, 270, 9.
(16) Cainelli, G.; Umani Ronchi, A.; Bertini, F.; Grasselli, P.; Zubiani,
G. Tetrahedron 1971, 27, 6109.
(10) De Lima, C.; Julia, M.; Verpeaux, J.-N. Synlett 1992, 133.
(11) (a) Romo, D.; Meyer, S. D.; Johnson, D. D.; Schreiber, S. L. J.
Am. Chem. Soc. 1993, 115, 7906. (b) Smith, A. B., III; Lin, Q.; Doughty,
V. A.; Zhuang, L.; McBriar, M. D.; Kerns, J. K.; Brook, C. S.; Murase, N.;
Nakayama, K. Angew. Chem., Int. Ed. 2001, 40, 196.
(12) (a) Plietker, B.; Seng, D.; Frohlich, R.; Metz, P. Eur. J. Org. Chem.
2001, 3669. (b) Smith, A. B., III; Doughty, V. A.; Lin, Q.; Zhuang, L.;
McBriar, M. D.; Boldi, A. M.; Moser, W. H.; Murase, N.; Nakayama, K.;
Sobukawa, M. Angew. Chem., Int. Ed. 2001, 40, 191.
(17) Tarhouni, R.; Kirschleger, B.; Rambaud, M.; Villieras, J. Tetrahe-
dron Lett. 1984, 25, 835.
(18) (a) Becker, M.; Marschall, H.; Weyerstahl, P. Chem. Ber. 1975,
108, 2391. (b) Sadhu, K. M.; Matteson, D. S. Tetrahedron Lett. 1986, 27,
795. (c) Barluenga, J.; Fernandez-Simon, J. L.; Concellon, J. M.; Yus, M.
J. Chem. Soc., Chem. Commun. 1986, 1665.
(19) (a) Sadhu, K. M.; Matteson, D. S. Organometallics 1985, 4, 1687.
(b) Brown, H. C.; Singh, S. M.; Rangaishenvi, M. V. J. Org. Chem. 1986,
51, 3150.
(13) Satoh, T.; Kondo, A.; Musashi, J. Tetrahedron 2004, 60, 5453.
(20) Kobayashi, T.; Pannell, K. H. Organometallics 1991, 10, 1960.
J. Org. Chem, Vol. 71, No. 15, 2006 5647

Pearlman et al.
TABLE 1. Effect of Stoichiometry, Rate of Addition of CH2Br2, and Temperature on Yields of Methylenation and Homologationa
n-BuLi
(equiv)
CH2Br2
(equiv)
2
(%)
7b
(%)
1
(%)
4
(%)
5
(%)
6
(%)
entry
add rateb
exothermc
Kd
1
2
3
4
5
6
7
8
9
2.00
2.00
3.00
3.00
3.00
3.00
2.00
3.00
3.00
1.00
1.00
2.00
2.00
1.00
1.00
1.00
2.00
1.70
slow
fast
slow
fast
slow
fast
fast
-97 f -82 °C
-98 f -74 °C
-98 f -70 °C
-99 f -63 °C
-100 f -83 °C
-99 f -77 °C
-28 f -7 °C
-29 f +5 °C
-27 f -7 °C
81.3
90.6
94.4
96.4
62.7
81.9
69.3
86.4
69.4
0.0
0.0
1.5
0.0
1.0
0.0
9.2
4.3
0.1
17.6
7.9
2.0
1.6
2.5
9.2
8.8
2.7
2.2
1.6
5.1
7.1
1.3
0.6
0.0
5.6
0.4
0.8e
0.3e
0.9
4.8
8.8
27.2
4.9
10.4
2.1
35.5
17.2
18.4
5.5
8.0e
3.5e
8.3
fast
fast
26.2
a General procedure: a solution of sulfone 1 in THF at -100 °C or -30 °C was treated with n-BuLi/hexanes followed by CH2Br2. b Addition rate of
CH2Br2 (fast means “all at once,” slow means over 2 min). c Exotherm on addition of CH2Br2. d K ≡ kS/kB. e In these experiments, the yields of 5 and 6 do
not reflect the true extent of homologation, because not all of the alkyllithium is converted to alkyl bromide.
SCHEME 4. Addition Mode with n-BuLi Added Last
Kirmse and von Wedel,21 Huisgen and Burger,22 and Friedman
et al.23 have shown that LiCH2X reacts with n-BuLi to afford
a homologous series of alkyllithium derivatives, n-Bu(CH2)n-
Li, n ) 0, 1, 2, and so on (“homologation”). A set of
experiments was carried out in which sulfone 1 was treated in
THF at -100 °C with either 2.00 or 3.00 equiv of n-BuLi/
hexanes, followed by a solution of either 1.00 or 2.00 equiv of
CH2Br2 in THF. The stoichiometry was varied to generate
kinetic data. The experiments were done at cryogenic temper-
ature, where the carbenoid intermediate is maximally stable with
gradual addition of the CH2Br2 over 2 min to minimize the
temperature rise. The crude reaction mixtures were analyzed
by quantitative GC to determine the yields of n-C5H11Br (5)
and n-C6H13Br (6), then flash chromatographed to determine
yields of olefin 2, sulfone 1, and p-MePhSO2Me (4). Results
are compiled in Table 1.
times faster with R-lithio-1 than with n-BuLi.
(KBi - Bi - K)S(1/K) + KS + (K - 1)(D - Bi) ) 0 (1)
In the course of experimentation, it was discovered that
adding the CH2Br2 “all at once” rather than over a period of 2
min increased the yield of olefin 2 and effectively suppressed
formation of the homologated alkyl bromides 5 and 6 (see Table
1, entries 1 vs 2, 3 vs 4, and 5 vs 6). Because the period of
time during the CH2Br2 add when both n-BuLi and LiCH2Br
are present is passed without the two reacting with each other,
homologation must be slower than lithium-bromine exchange.
At the end of the add, the mixture contains an approximately
equimolar mixture of R-lithio-1 and LiCH2Br, which react
together efficiently to give olefin 2 cleanly and in high yield.
Simultaneous addition of n-BuLi and CH2Br2 was also
investigated and found to be a viable addition mode. By adding
2.5 equiv of n-BuLi and 1.1 equiv of CH2Br2 simultaneously
at -68 °C, a 91.3% yield was obtained.
The relative reactivity of R-lithio-1 versus n-BuLi (K ≡ kS/
kB) was calculated from the yields of olefin 2,24 n-C5H11Br 5,
and n-C6H13Br 6, using eq 125 (where S is the amount of sulfone,
D is the total amount of CH2Br2 added, and Bi is the initial
amount of n-BuLi) to correct for the different kinetics with
which the two species react. Because S, D, and Bi are each
known with reasonable accuracy, and mass recoveries are close
to the theoretical, the calculated values for K should be
reasonably accurate. By averaging the results of entries 1 and
3 of Table 1, it was determined that LiCH2Br reacts 9.6 ( 0.8
Two experiments (Table 1, entries 7 and 8) were run at -30
°C (initial temperature) to assess the need for cryogenic
conditions. That temperature was chosen because it can be
relatively easily reached in noncryogenic general purpose pilot/
production equipment. The experiments gave lower yields of 2
(24) The digit after the decimal in the yields of olefin 2 is not significant.
However, the digit after the decimal in the yields of the byproducts (7b, 4,
5, 6, and recovered 1) is significant. Thus, to maintain consistency of
presentation, the digit after the decimal is reported for all yields. Note that
100% minus the yield of recovered sulfone 1 is used rather than the yield
of olefin 2 in all selectivity calculations. Thus, the legitimacy of including
the digit after the decimal is a moot point.
(21) Kirmse, W.; von Wedel, B. G. Justus Liebigs Ann. Chem. 1963,
666, 1.
(22) Huisgen, R.; Burger, U. Tetrahedron Lett. 1970, 3053.
(23) Friedman, L.; Honour, R. J.; Berger, J. G. J. Am. Chem. Soc. 1970,
92, 4640.
(25) See Supporting Information (Appendix) for derivation.
5648 J. Org. Chem., Vol. 71, No. 15, 2006

Lithium Carbenoids for Methylenation of Sulfones
SCHEME 5. Reaction of Lithiated Model Sulfone with
CH2Br2
selectivity than CH2BrCl and is cheaper than CH2I2. The yield
(88%) was comparable with that obtained on a 0.9 g scale (96%;
Table 1, entry 4), and the quality was equal.
Alternative Alkyllithium Reagents. Repeating the Table 1,
entry 1 experiment but with t-BuLi instead of n-BuLi gave olefin
2 in 68% yield along with t-BuCH2Br (12) in 25.0% yield (Table
4). Analysis of the data reveals that LiCH2Br reacts with
R-lithio-1 2.8 ( 0.9 times faster than with t-BuLi, and with
t-BuLi approximately 9.6
/2.8 ) 3.4 times faster than with n-BuLi.
Previous studies have also found that t-BuLi is more nucleophilic
than n-BuLi (in THF).26
and higher yields of 5, along with high levels of vinyl bromide
7b (9.2 and 4.3%).
Adding the CH2Br2 “all at once” reduced the amount of
byproduct t-BuCH2Br 12 from 25.0 to 12.1% and increased the
yield of olefin 2 from 67.7 to 80.5% (Table 4, entries 1 vs 2).
The incomplete suppression of the byproduct is due to the fact
that t-BuLi undergoes homologation faster than n-BuLi.
Repeating the Table 1, entry 1 experiment but with MeLi
instead of n-BuLi resulted in formation of approximately equal
yields of olefin 2 and 9â-methyl sulfone 3a (Table 5), indicating
that both LiCH2Br and MeBr alkylate R-lithio-1 faster than they
are formed by halogen-metal exchange between MeLi and CH2-
Br2 (i.e., halogen-metal exchange is the rate-determining step).
To quantify the rate of halogen-metal exchange, an experiment
(Table 5, entry 2) was run at low temperature (-96 °C) and
quenched with methanol as soon as possible after the end of
the CH2Br2 add (8 s).27 From the results, it is estimated that the
half-life of sulfone 1 at -96 °C is ∼30 s.28 Halogen-metal
exchange must be slower because it is the rate-determining step.
Trisubstituted and Tetrasubstituted Olefin Synthesis.
Attempts were made to extend the new olefin synthesis to the
preparation of trisubstituted olefins (Table 6).
Olefination did occur to give trisubstituted olefin 7d, but very
slowly. The rate of alkylation of R-lithio-1 by n-BuCHBrLi
could be roughly estimated as follows. From the results of entry
2, Table 6, it is estimated that the half-life of R-lithio-1 in the
presence of n-BuCHBrLi is 170 s at -74 °C. Ample carbenoid
n-BuCHBrLi (85.6-113.3%) as well as R-lithio-1 (70.6-94.4%)
were present throughout that time frame to justify the assumption
of first-order kinetics.29 By comparison, from the results of entry
2, Table 1, it is estimated that the half-life of R-lithio-1 in the
presence of LiCH2Br at -74 °C is <4.3 s. Thus, it can be
estimated that LiCH2Br reacts g40 times more rapidly than
n-BuCHBrLi with R-lithio-1.
The mechanism of vinyl bromide formation was investigated
using a model sulfone (cis-8; Scheme 5). It was found that
R-lithio-8 reacts with CH2Br2 in the absence of n-BuLi to form
vinyl bromide 9b. Thus, the elevated level of vinyl bromide 7b
formed in the noncryogenic temperature experiments is at-
tributed to the consumption of n-BuLi prior to the end of the
CH2Br2 add by deprotonation pathways, so that there is not
enough n-BuLi left at the end of the CH2Br2 add to react with
all of the CH2Br2. Formation of vinyl bromide 7b can be
effectively suppressed by decreasing the CH2Br2 charge to 1.7
equiv (with 3.0 equiv of n-BuLi), which ensures that some
n-BuLi remains at the end of the CH2Br2 add (Table 1, entry
9). Alternatively, the vinyl bromide can be effectively suppressed
by running at cryogenic temperature (Table 1, entries 2, 4, and
6).
Compounds cis-10a and trans-10a were identified by spec-
troscopic comparison with authentic samples prepared by the
route outlined in Scheme 6.
LiCH2I. A set of experiments was done to compare LiCH2I
with LiCH2Br with both sulfone 1 and cis-8 as test substrates
(Table 2). The yields were equal within experimental error.
LiCH2Cl. With the chlorocarbenoid LiCH2Cl (generated from
either n-BuLi/CH2BrCl or t-BuLi/CH2BrCl), lower yields of
olefin 2 and higher yields of the homologated alkyl bromides
were obtained than with the bromocarbenoid (Table 3). It was
calculated that LiCH2Cl reacts with R-lithio-1 3.5 times faster
than with n-BuLi and 0.7 times as fast as with t-BuLi.
CH2Br2 was chosen for a preparative scale experiment
(starting with 50 g of sulfone 1) because it reacted with higher
As a consequence of the slow rate of olefination, carbenoid
dimerization to 5-decene (stereochemistry unassigned) was
competitive. No other significant byproducts could be detected
by TLC or GC. In particular, nC5H11SO2Ar could not be
detected, suggesting that the dimerization is not mediated by
-
ArSO2 but rather occurs by Wurtz-Fittig coupling of the
carbenoid, which is the mechanism proposed by Koebrich for
dimerization of LiCHCl2 to trans-ClCHCHCl.30
SCHEME 6. Synthesis of Bromomethyl Sulfones
J. Org. Chem, Vol. 71, No. 15, 2006 5649

Pearlman et al.
TABLE 2. LiCH2Br versus LiCH2Ia
a General procedure: a solution of sulfone in THF (0.12 M) was cooled to -100 °C, treated with 3.0 equiv of n-BuLi, recooled to -100 °C, treated with
2.5 equiv of CH2X2 in one portion, and quenched 30 s later with methanol. b By 13C NMR peak height ratio. c By integration of 1H NMR vs internal standard
(DMSO).
TABLE 3. Methylenation Experiments with LiCH2Cla
2
(%)
7a
(%)
1
(%)
4
(%)
entry
BuLi
add rateb
exotherm
RCH2Br
RCH2CH2Br
Kc
1
2
3
4
n-BuLi
n-BuLi
t-BuLi
t-BuLi
slow
fast
slow
fast
-99 f -80 °C
-101 f -85 °Cd
-100 f -78 °C
-101 f -83 °Cd
65.4
83.7
49.7
83.1
1.2
0.0
0.0
0.0
32.5
15.0
50.2
15.6
0.3
0.5
0.0
0.4
11.7
0.5
39.2
7.4
2.1
0.0
2.0
0.0
3.5
0.7
a General procedure: a solution of sulfone 1 in THF (0.12 M) was cooled to -100 °C, treated with 2.0 equiv of nBuLi/hexanes or tBuLi/pentane,
re-cooled to -100 °C, then treated with 1.0 equiv of CH2BrCl. b Add rate of CH2BrCl (fast means “all at once,” slow means over 2 min). c K ≡ kS/kB.
d Anomalously small exotherm may be attributable to low initial temperature.
Interestingly, the trans/cis isomer ratio of the product 7d
decreased continuously over the course of the reaction: 89/11
at -95 °C, 78/22 at -74 °C, and 73/27 at -62 °C. This suggests
that the two enantiomers of n-BuCHLiBr31 react at different
rates with sulfone 1 (which is enantiomerically pure because it
is synthesized from glucose) to afford different ratios of the
two geometric isomers of the product. The enantiomeric excess
of the unreacted carbenoid was not measured to test this
hypothesis.32
(26) Waack, R.; Doran, M. A. J. Org. Chem. 1967, 32, 3395.
(27) Methanol is expected to be an effective quenching agent based on
the fact that water quenches t-BuCHILi to give t-BuCH2I at below -90
°C, see: Goldstein, M. J.; Dolbier, W. R., Jr. J. Am. Chem. Soc. 1965, 87,
2293.
(28) Half-lives were calculated using the formula t1/2 ) (t ln(1/2))/(ln(1
- M), where t is the time (in seconds) and M is the yield of olefin 2
(expressed in decimal format).
(30) (a) Koebrich, G.; Merkle, H. R. Chem. Ber. 1966, 99, 1782. (b)
Koebrich, G. Angew. Chem., Int. Ed. Engl. 1967, 6, 41. (c) Koebrich, G.
Bull. Soc. Chim. Fr. 1969, 2712.
(31) Hoffmann has shown that the two enantiomers of n-BuCHBrLi do
not interconvert at -120 °C: Hoffmann, R. W.; Julius, M.; Chemla, F.;
Ruhland, T.; Frenzen, G. Terahedron 1994, 50, 6049.
(29) Halogen-metal exchange to form n-BuCHBrLi was 64% complete
in 20 s at -95 °C, by comparison of the sum of the yields of all the
carbenoid-derived products (5.9% yield of 2, 2.3% yield of 5-decene, and
113.3% yield of n-C5H11Br 5 [the product of protonation of the carbenoid])
with the yield of unreacted n-BuCHBr2 (34.3%).
5650 J. Org. Chem., Vol. 71, No. 15, 2006

Lithium Carbenoids for Methylenation of Sulfones
TABLE 4. Methylenation Experiments with t-BuLia
t-BuLi
(equiv)
CH2Br2
(equiv)
2
(%)
7b
(%)
1
(%)
4
(%)
12
(%)
entry
add rateb
exothermc
Kd
1
2
3
4
5
6
2.00
2.00
3.00
3.00
3.00
3.00
1.00
1.00
2.00
2.00
1.00
1.00
slow
fast
slow
fast
slow
fast
-99 f -84 °C
-99 f -77 °C
-96 f -68 °C
-98 f -58 °C
-100 f -80 °C
-98 f -75 °C
67.7
80.5
77.2
88.7
43.6
42.7
0.0
0.0
2.4
0.0
0.0
0.0
32.1
18.8
19.3
10.5
57.3
56.7
0.3
1.6
1.7
3.8
1.2
1.5
25.0
12.1
54.0
57.2
27.1
27.0
1.9
3.6
a General procedure: a solution of sulfone in THF (0.13 M) was cooled to -100 °C, treated with t-BuLi/pentane, recooled to -100 °C, and treated with
CH2Br2. b Add rate of CH2Br2 (fast means “all at once,” slow means over 2 min). c Exotherm on addition of CH2Br2. d K ≡ kS/kB.
TABLE 5. Experiments with MeLia
MeLi
(equiv)
CH2Br2
(equiv)
2
(%)
3a
(%)
1
(%)
4
(%)
13
(%)
entry
exotherm
1
2
3
4
5
2.00
2.00
3.00
3.00
3.00
1.00
1.00
2.00
1.99
2.00
-96 f -78 °C
51.0
17.1
53.3
64.4
44.1
48.8
11.9
46.2
36.2
55.7
0.0
71.1
0.0
2.9
0.0
0.0
0.0
0.0
1.5
0.0
4.1
4.9
2.5
-96 °Cb
-100 f -50 °C
-72 f -42 °C
-20 f +9 °C
0.0
0.0
a General procedure: a solution of sulfone 1 in THF (0.12 M) was cooled to -70 °C, treated with MeLi/Et2O, re-cooled to the indicated temperature, and
treated all at once with neat CH2Br2. b Quenched with methanol 8 s after the CH2Br2 add.
TABLE 6. Experiments with n-BuCHBr2/n-BuLia
7d
(%)
1
(%)
n-C5H11Br
n-BuCHBr2
5-decene
(%)
entry
exotherm
timeb
trans/cis
(%)
(%)
1
2
3
-102 f -95 °C
-102 f -74 °C
-97 f -62 °C
20 s
5.9
28.8
54.0
89/11
78/22
73/27
94.4
70.6
46.5
113.3
85.6
12.0
69.3
34.3
17.9
2.3
23.1
51.7
85 s
3.0 m
a General procedure: a solution of sulfone 1 in THF (19 mL/g) was cooled to -100 °C and treated with 3.0 equiv of n-BuLi, followed by 2.0 equiv of
n-BuCHBr2. b Time between the end of the n-BuCHBr2 add and the methanol quench.
The disubstituted carbenoid Me2CBrLi proved to be even less
reactive toward sulfone 1, affording tetrasubstituted olefin 15
in only 16.3% yield along with unreacted sulfone (Scheme 7).
Vinyl Halide Synthesis. Sulfones 1 and cis-8 can be
converted into vinyl halides in high yield and purity by the
application of the standard procedure, but with LDA rather than
n-BuLi (Table 7). The reaction of metalated sulfones with
dihalomethylene carbenoids to afford vinyl halides has been
previously reported by Julia et al.,33 but the procedure involved
transferring a cold (-78 °C) solution of metalated sulfone to a
cold (-78 °C) solution of carbenoid. The in situ procedure
described herein is operationally simpler. Isolated yields of
chromatographically homogeneous products are uniformly
excellent (Table 7).
(32) The stereochemistry of the sulfinate elimination step (syn-coplanar
or anti-periplanar) could also be deduced from the absolute configuration
of the unreacted, kinetically resolved carbenoid.
(33) (a) Charreau, P.; Julia, M.; Verpeaux, J. N. J. Organomet. Chem.
1989, 379, 201. (b) Charreau, P.; Julia, M.; Verpeaux, J. N. Bull. Soc. Chim.
Fr. 1990, 275.
J. Org. Chem, Vol. 71, No. 15, 2006 5651

Pearlman et al.
SCHEME 7. Attempted Synthesis of a Tetrasubstituted Olefin
TABLE 7. Synthesis of Vinyl Halidesa
LDA deprotonates CH2Br2 much faster than it deprotonates the
vinyl bromide.
The major isomer of vinyl bromide 7b was determined to be
trans by NOE (irradiation of H8 of 7b resulted in ∼10%
enhancement in the integral of H9′). Julia et al. previously
assigned the trans stereochemistry to a vinyl bromide formed
by similar chemistry.33a
The vinyl halides 7 are formed with high trans/cis stereose-
lectivity (Table 7). Interestingly, the stereoselectivity was lower
with the unsymmetrical dihalide CH2BrCl than with the sym-
metrical dihalides (CH2Cl2, CH2Br2, and CH2I2), and the ratio
decreased as the reaction progressed (Tables 7 and 8). These
observations suggest that LiCHBrCl, like n-BuCHBrLi (see
above), exists as a racemic mixture of enantiomers that do not
interconvert under the conditions of the reaction and which react
at different rates with sulfone 1 (single enantiomer).
Chlorovinylation of a Nitrile. Application of the chlorovi-
nylation process to 2-methyl-tetradecanitrile afforded vinyl
chloride 17 (as a 3:2 mixture of geometric isomers) in 55.0%
yield (Scheme 9).
Selectivity (kS/kR) of a Noncarbenoid Electrophile. A
competition experiment was carried out between R-lithio-1 and
n-BuLi for n-BuBr, a noncarbenoid electrophile (Scheme 10).
From the relative yields (17.8% R-n-butyl sulfone 3b and 35.6%
n-octane), it is estimated that n-BuBr reacts about twice as fast
with n-BuLi as with R-lithio-1.
Discussion
a General procedure: a solution of the sulfone in THF (10 mL/g) at -78
°C is treated with 3.0 equiv of n-BuLi, followed by >2.0 equiv of i-Pr2NH.
After 7 min, 3.0 equiv of CH2X2 is added all at once. b i-Pr2NH add omitted.
Reactivity. Lithium methylene carbenoids are considerably
more reactive toward lithiated sulfones than the more covalent
carbene equivalents that have been previously investigated
(MCH2I, where M ) SiMe3, SnR3, or MgCl). From the results
of Table 1, entry 2, it is estimated that the half-life of R-lithio-1
in the presence of LiCH2Br at -74 °C is <4.3 s, and from the
results of Table 5, entry 2, the half-life at -96 °C is estimated
to be ∼30 s. Thus, while the methylenation reaction is quite
fast, it is slow enough to be measurable (Table 9). Also relevant
to the discussion of rates is the observation that “all at once”
addition of CH2Br2 suppresses homologation. This indicates that
homologation is slower than bromine-lithium exchange by a
sufficient amount that very little occurs in the time period of
an “all at once” CH2Br2 addition. Thus the carbenoid, while
highly reactive, is “tame” enough to react with t-BuLi at less
than the diffusion-controlled rate.
Selectivity. To gather information about the factors that affect
partitioning between the methylenation and homologation
pathways, sets of competition experiments between R-lithio-1
and either n-BuLi or t-BuLi for LiCH2Br, LiCH2Cl, and n-BuBr
were carried out. Results are summarized in Table 10. The
following empirical trends are evident: (1) both LiCH2Br and
LiCH2Cl react faster with R-lithio-1 than with n-BuLi; (2)
LiCH2Br reacts 2.5-2.7 times more selectively than LiCH2Cl
The only caveat is that the CH2X2 must be added “all at once”
as, otherwise, the vinyl halide product will be in contact with
the LDA for a sufficient length of time to be deprotonated to a
vinylidene carbenoid species, which decompose by various
pathways depending on structure. In the case of 7b, the
corresponding vinylidene carbenoid decomposes cleanly to
cyclobutane dimer 16 by the cascade of cycloadditions depicted
in Scheme 8. Thus, an experiment in which the CH2Br2 was
added over 6 min at -78 °C (Scheme 8) afforded cyclobutane
dimer 16 in 59.2% isolated yield. Koebrich previously encoun-
tered a remarkably similar case of cyclobutane dimer formation
on treatment of a simple γ,δ-unsaturated vinyl chloride with
n-BuLi. He established that the stereochemistry was trans, head-
to-head, by X-ray crystallography.34 Accordingly, the head-to-
head stereochemistry is assigned for cyclobutane dimer 16. The
stereochemistry around the cyclopropane ring is unassigned. The
control experiment in which the CH2Br2 was added all at once
(Table 7, entry 4) gave a negligible amount of 16, showing that
(34) Koebrich, G. Angew. Chem., Int. Ed. Engl. 1973, 12, 464.
5652 J. Org. Chem., Vol. 71, No. 15, 2006

Lithium Carbenoids for Methylenation of Sulfones
SCHEME 8. Mechanism of Cyclobutane Dimer Formation
TABLE 8. Chlorovinylation of 1 with LiCHBrCla
7a
1
entry
n-BuLi
i-Pr2NH
CH2BrCl
exotherm, time
(%)
trans/cis
(%)
1
2
3
4
3.99
2.00
3.00
2.00
3.49
1.50
2.50
1.50
3.01
1.00
2.00
1.00
-99 f -55 °C over 7 m
-101 f -75 °C over 4 m
-101 f -75 °C over 5 m
-101 f -75° over 6 m
95.8
87.7
96.2
74.4
92.2/7.8
81.7/18.3
91.3/8.7
82.5/17.5
not detected
8.4%
0.9%
20.1%
a General procedure: a solution of sulfone 1 in THF (0.06 M) at -100 °C was treated with n-BuLi, treated with i-Pr2NH, re-cooled to -100 °C, and
treated all at once with a solution of CH2X2 in THF.
and t-BuLi, which is monomeric under those conditions.36 This
SCHEME 9. Chlorovinylation of a Nitrile
is compelling evidence against the involvement of aggregation
phenomena in determining the selectivity of carbenoid for
R-lithio-1 over alkyllithium.
Because the remarkable selectivity of carbenoids for the
reaction with R-lithio-sulfones is not explicable by invoking
aggregation phenomena, it presumably derives from the intrinsic
nature of the carbenoid species itself.
(with n-BuLi and t-BuLi, respectively); and (3) the selectivity
of LiCH2Br was higher in the experiments with 2 equiv of
LiCH2Br than in the experiments with 1 equiv of LiCH2Br.
Huisgen and Burger35 and Friedman et al.23 have previously
reported that LiCH2Br is more selective than LiCH2Cl for
cyclopropanation of olefins, by factors of 1.5 and 1.9, respec-
tively, in line with the findings of this study.
The possibility that the selectivity of both LiCH2Br and
LiCH2Cl for R-lithio-1 over n-BuLi and t-BuLi could be due
to formation of a heterodimer between R-lithio-1 and BuLi was
considered (Scheme 11). Because each molecule of carbenoid
LiCH2X would be generated in close proximity to a molecule
of R-lithio-1, the system would be biased in favor of methyl-
enation over homologation.
Mechanism. It is well-established that vinylidene carben-
oids37-39 and cyclopropyl carbenoids40,41 react with carbanion
nucleophiles by a metal-assisted ionization (MAI)14b,38,42 mech-
anism. This is a solvolytic SN1-type mechanism in which the
halogen associates with lithium to form a bridged zwitterionic
intermediate prior to the encounter with the nucleophile. The
bridged intermediate has been studied computationally and
verified to be the low-energy structure.43,44 However, these
special types of carbenoids are biased for displacement by
solvolytic (SN1-type) mechanisms over SN2-type mechanisms
because SN2 displacement at vinyl and cyclopropyl carbon atoms
is inherently slow.
However, a 1:1 mixture of n-BuLi and R-lithio-1 should
contain a higher percentage of heterodimer (n-BuLi)(R-lithio-
1) and less homodimer (n-BuLi)n than a 2:1 mixture, for
statistical reasons. Thus, the selectivity of carbenoid for the
reaction with R-lithio-1 versus n-BuLi should be higher with
the 1:1 mixture than with the 2:1 mixture. In fact, the selectivity
The fact that LiCH2Br reacts g40 times faster than n-
BuCHLiBr is inconsistent with the MAI mechanism because
(36) Bauer, W.; Winchester, W. R.; Schleyer, P. v. R. Organometallics
1987, 6, 2371.
(37) Koebrich, G.; Ansari, F. Chem. Ber. 1967, 100, 2011.
(38) (a) Duraisamy, M.; Walborsky, H. M. J. Am. Chem. Soc. 1984, 106,
5035. (b) Walborsky, H. M.; Duraisamy, M. Tetrahedron Lett. 1985, 26,
2743.
(39) Oku, A.; Harada, T.; Hattori, K.; Nozaki, Y.; Yamaura, Y. J. Org.
Chem. 1988, 53, 3089.
(40) (a) Taylor, K. G.; Chaney, J. J. Am. Chem. Soc. 1976, 98, 4158.
(b) Taylor, K. G.; Chaney, J.; Deck, J. C. J. Am. Chem. Soc. 1976, 98,
4163. (c) Taylor, K. G. Tetrahedron 1982, 38, 2751.
(41) Kitatani, K.; Yamamoto, H.; Hiyama, T.; Nozaki, H. Bull. Chem.
Soc. Jpn. 1977, 50, 2158.
(42) (a) Rachon, J.; Goedken, V.; Walborsky, H. M. J. Am. Chem. Soc.
1986, 108, 7435. (b) Topolski, M.; Duraisamy, M.; Rachon, J.; Gawronski,
J.; Gawronska, K.; Goedken, V.; Walborsky, H. M. J. Org. Chem. 1993,
58, 546.
(43) (a) Mareda, J.; Rondan, N. G.; Houk, K. N.; Clark, T.; Schleyer, P.
v. R. J. Am. Chem. Soc. 1983, 105, 6997. (b) Schleyer, P. v. R.; Clark, T.;
Kos, A. J.; Spitznagel, G. W.; Rohde, C.; Arad, D.; Houk, K. N.; Rondan,
N. G. J. Am. Chem. Soc. 1984, 106, 6467.
(44) For two excellent recent reviews, see: (a) Braun, M. In The
Chemistry of Alkyllithium Compounds; Rappoport, Z., Marek, I., Eds.; John
Wiley and Sons: New York, 2004; Chapter 13, pp 829-900. (b) Boche,
G.; Lohrenz, J. C. W. Chem. ReV. 2001, 101, 697.
with the 2:1 mixture was higher, with both n-BuLi (kS/kB
)
8.8 for 1:1 and 10.4 for 2:1) and t-BuLi (kS/kB ) 1.9 for 1:1
and 3.6 for 2:1). Thus, the heterodimer (BuLi)(LiCH2Br) either
is not formed or, if formed, dissociates into the solvent separated
component molecules faster than they react with each other.
Matteson has reported that addition of 0.94 equiv of n-BuLi/
hexane to a solution of triisopropylborate and 1.0 equiv of CH2-
ClI in THF at -78 °C afforded diisopropyl (chloromethyl)-
boronate in 84% yield,19a which also shows that any initially
formed heterodimer or heterotetramer (n-BuLi)n(LiCH2Cl), if
formed, must dissociate into the component molecules faster
than they react with each other.
Moreover, the carbenoid is selective for R-lithio-1 over both
n-BuLi, which exists as a mixture of dimers and tetramers under
the conditions of the methylenation reaction (THF, -100 °C),
(35) Burger, U.; Huisgen, R. Tetrahedron Lett. 1970, 3049.
J. Org. Chem, Vol. 71, No. 15, 2006 5653

Pearlman et al.
SCHEME 10. Competition Experiment between r-Lithio-1 and n-BuLi for n-BuBr
TABLE 9. Effect of Metal M on Rate and Yield of Methylenation by MCH2Ia
entry
M (in MCH2I)
Krel (1 f 18)
yield (1 f 2; %)
1b
2b
3
4
5
Me3Si
n-Bu3Sn
n-Pr
e1
0.0
77.7
6 × 102
104
Me3Sn
Li
104
85.8
86.7
95.4
>2 × 105 c
6
Lid
>3 × 105 e
a Relative rates measured in THF at -78 °C by competition experiments. Data in entries 1-4 taken from ref 1a. b Data obtained using cis-8: see footnote
11 of ref 1a for explanation of calculation procedure. c Lower limit, based on the fact that methylenation is at least 86.7% complete (>3 half-lives, assuming
bimolecular kinetics) in 30 s at -65 °C. d Experiment with LiCH2Br. e Lower limit, based on fact that reaction is at least 95.4% complete (>4 half-lives)
in 30 s at -73 °C. The half-life at -96 °C is <30 s (Table 5, entry 2).
SCHEME 11. Heterodimer Hypothesis to Rationalize Selectivity of LiCH2X for r-Lithio-1 over BuLi
TABLE 10. Selectivities of Electrophiles for r-Lithio-1 vs n-BuLi
and t-BuLi
invoking a lithium-induced lengthening and, thus, weakening
of the carbon-bromine bond, as proposed by Bent45 and
Seebach et al.46
The observation that LiCH2Br is ∼2.6 times more selective
than LiCH2Cl toward methylenation versus homologation is hard
to understand in light of computational studies showing that
the two carbenoids have similar lithium-halogen-bridged
structures and that their electrophilicities are “essentially the
same.”47 One possible explanation is that the LiBr evolved in
the methylenation stabilizes the LiCH2Br, rendering it more
selective in accord with the “reactivity/selectivity principle.”48
Villieras et al. have reported that LiCH2Br is stabilized by
LiBr.17 The hypothesis that LiBr enhances selectivity also would
explain the observation that selectivity is higher with 2 equiv
of LiCH2Br than with 1 equiv of LiCH2Br because the
concentration of LiBr increases over the course of the reaction.49
An alternative explanation is that the LiBr may act as a Lewis
kS/kB (n-BuLi)
kS/kB (t-BuLi)
electrophile
LiCH2Br
LiCH2Cl
CH3CH2CH2CH2Br
1:1a
2:1
10.4
1:1
2:1
3.6
8.8
3.5
1.9
0.7
∼0.5
a Ratio of alkyllithium to R-lithio-1 prior to addition of alkyl halide.
(45) Bent, H. A. Chem. ReV. 1961, 61, 275.
(46) (a) Seebach, D.; Siegel, H.; Mullen, K.; Hiltbrunner, K. Angew.
Chem., Int. Ed. Engl. 1979, 18, 784. (b) Siegel, H.; Hiltbrunner, K.; Seebach,
D. Angew. Chem., Int. Ed. Engl. 1979, 18, 785. (c) Seebach, D.; Siegel,
H.; Gabriel, J.; Hassig, R. HelV. Chim. Acta 1980, 63, 2046. (d) Seebach,
D.; Hassig, R.; Gabriel, J. HelV. Chim. Acta 1983, 66, 308.
(47) Hermann, H.; Lohrenz, J. C. W.; Kuhn, A.; Boche, G. Tetrahedron
2000, 56, 4109.
(48) (a) Leffler, J. E.; Grunwald, E. Rates and Equilibria of Organic
Reactions; John Wiley and Sons: New York, 1963; pp 156-168. (b)
Johnson, C. D. Chem. ReV. 1975, 75, 755. (c) Giese, B. Angew. Chem., Int.
Ed. Engl. 1977, 16, 125. (d) McLennan, D. J. Tetrahedron 1978, 34, 2331.
(e) Bordwell, F. G.; Hughes, D. L. J. Org. Chem. 1980, 45, 3320.
secondary alkyl halides are expected to solvolyze faster than
primary alkyl halides. Thus, it is proposed that lithium methylene
carbenoids alkylate R-lithio-sulfones by the SN2 mechanism.
Both Me3SiCH2I and R3SnCH2I also alkylate sulfones by the
SN2 mechanism.1a Koebrich has proposed that n-BuLi reacts
with LiCHCl2 by an SN2 mechanism.30 The fact that LiCH2Br
is more reactive than n-BuBr can be rationalized either by
invoking chelation between the lithium of the carbenoid and
one of the oxygens of the sulfone in the transition state or by
5654 J. Org. Chem., Vol. 71, No. 15, 2006

Lithium Carbenoids for Methylenation of Sulfones
oil was flash chromatographed on 60 g silica gel (3 cm × 6′′
column; fraction size: 50 mL); gradient elution, 10 f 20% ethyl
acetate/cyclohexane). The fractions containing the Rf ) 0.62
component were combined and concentrated to a pale yellow oil
identified as olefin 2 in pure form by 13C and 1H NMR comparison
with an authentic standard.1b Weight: 0.6034 g (mw 312.84; 1.9288
mmol, 96.4% yield). 13C NMR (75 MHz, CDCl3): δ 154.5 (s),
129.2 (d), 128.7 (d), 112.6 (d), 111.9 (s), 110.2 (t), 80.3 (d), 76.3
(d), 59.4 (d), 44.4 (t), 37.4 (d), 35.1 (t), 32.3 (t), 30.3 (t), 26.1 (q),
acid catalyst for the methylenation, as zinc salts catalyze the
Simmons-Smith cyclopropanation reaction.50
Conclusions
Lithium carbenoids are highly unstable, decomposing above
-110 °C, but this study has shown that lithium methylene and
dihalomethylene carbenoids, when generated at -100 °C in the
presence of R-lithio-sulfones (and one R-lithio-nitrile), react to
form olefins and vinyl halides in high yield and purity. The
carbenoids exhibit practical selectivity for methylenation over
homologation and other pathways. LiCH2Br was found to be
2.6 times more selective than LiCH2Cl, an effect attributed to
the LiBr byproduct. The demonstration that lithium methylene
carbenoids, though highly reactive, are capable of reacting with
high selectivity among potential partners encourages further
investigation of practical applications of these reagents in
synthesis.
1
24.5 (t). H NMR (300 MHz, CDCl3): δ 1.34 (3H, s), 1.52 (3H,
s), 1.5-2.7 (10H, m’s), 3.55 (2H, t, J ) 6.5 Hz), 4.06 (1H, m),
4.50 (1H, t, J ) 3.4 Hz), 5.01 (1H, s), 5.08 (1H, s), 5.49 (2H,
m’s), 6.04 (1H, d, J ) 3.3 Hz). MS (EI): m/e 312, 314 (P+, 2.2%);
297, 299 (P+ - CH3, 100%). Anal. Calcd for C17H2535ClO3: m/e
312.1492. Found: m/e 312.1495. The fractions containing the Rf
) 0.23 component were combined and concentrated to a white
crystalline solid (weight: 0.0494 g), consisting of a 39.3:60.7 (w/
w) mixture of sulfone 1 (mw 455.02; 19.4 mg, 0.043 mmol, 2.1%
yield) and methyl p-toluyl sulfone 4 (mw 170.23; 30.0 mg, 0.176
mmol, 8.8% yield) by 13C and 1H NMR comparison with an
authentic standard. Spectra for methyl p-toluyl sulfone 4: 13C NMR
(75 MHz, CD2Cl2): δ 145.1 (s), 138.4 (s), 130.3 (d), 127.7 (d),
Experimental Section
1
44.9 (q), 21.7 (q). H NMR (90 MHz, CDCl3): δ 2.47 (3H, s),
Sulfone 1. Sulfone 1 was synthesized by the process described
in ref 7. 13C NMR (20 MHz, CD2Cl2): δ 145.6 (s), 135.3 (s), 130.7
(d), 130.5 (d), 129.4 (d), 127.6 (d), 113.7 (d), 112.66 (s), 80.4 (d),
76.5 (d), 71.8 (d), 59.5 (d), 45.0 (t), 33.9 (d), 32.7 (t), 31.9 (t),
28.3 (t), 26.3 (q), 26.1 (q), 24.9 (t), 21.7 (q). 1H NMR (200 MHz,
CD2Cl2): δ 1.22 (3H, s), 1.32 (3H, s), 1.4-2.4 (10H, m’s), 2.40
(3H, s), 3.50 (2H, t, J ) 6.5 Hz), 3.54 (1H, ddd, J ) 12, 7, 2 Hz),
3.77 (1H, td, J ) 11.5, 6 Hz), 4.35 (1H, t, J ) 3 Hz), 5.40 (2H,
m), 5.90 (1H, d, J ) 3 Hz), 7.37 (2H, d, J ) 8 Hz), 7.75 (2H, d,
J ) 8 Hz). Mp 123-124.5 °C (white plates). Anal. Calcd for
C23H31ClO5S: C, 60.71; H, 6.87; Cl, 7.79; S, 7.05 (%). Found: C,
60.83; H, 6.83; Cl, 7.85%; S, 6.92 (%).
Olefin 2 (Fast Addition, 2 equiv of CH2Br2/n-BuLi; Table 1,
Entry 4). A solution of sulfone 1 (mw 455.02; 0.9102 g, 2.0004
mmol) in 17.0 mL THF under argon was cooled to -100 °C using
a liquid nitrogen/ethanol slush bath and treated over ∼1 min with
n-BuLi/hexanes (3.64 mL of 1.65 M solution,51 6.01 mmol, 3.00
equiv), resulting in an exotherm to -73 °C. The mixture was
recooled to -99 °C over 3 min and then treated all at once with
neat CH2Br2 (mw 173.85; d 2.477; 0.281 mL, 0.696 g, 4.00 mmol,
2.00 equiv), resulting in an exotherm to -63 °C. After 15 s, the
mixture was quenched with 1.00 mL methanol and analyzed by
GC.52 The mixture contained n-C5H11Br 5 (mw 151.05; 14.708 mg,
0.0974 mmol, 4.9% yield) and n-C6H13Br 6 (mw 165.08; 1.337
mg, 0.0081 mmol, 0.4% yield). The mixture was poured into 40
mL 3% aq NaOH, extracted with CH2Cl2 (2 × 25 mL), dried
(MgSO4), and concentrated to give a yellow oil (weight: 0.6655
g) consisting of one major (Rf ) 0.62) and one minor (Rf ) 0.23)
component by TLC (eluent: 25% ethyl acetate/cyclohexane). The
3.06 (3H, s), 7.37 (2H, d, J ) 8 Hz), 7.87 (2H, d, J ) 8 Hz). Mp
86.5-88.5 °C (cf. lit. 88 °C53). MS (EI): m/e 170 (P+, 60.5%), 91
(P+ - SO2CH3, 100%). Anal. Calcd for C8H10SO2: m/e 170.0401.
Found: m/e 170.0404.
Trisubstituted Olefin 7d (Table 6, Entry 3). A solution of
sulfone 1 (mw 455.02; 0.9102 g, 2.0004 mmol) in 17.0 mL THF
under argon was cooled to -95 °C using a liquid nitrogen/ethanol
slush bath and treated over 30 s with a solution of n-BuLi/hexane
(3.64 mL of 1.65 M solution, 6.01 mmol, 3.00 equiv), resulting in
an exotherm to -72 °C. The resulting orange reaction mixture was
cooled to -97 °C over 10 min, then treated all at once with neat
n-BuCHBr2 (mw 229.95; d 1.624; 0.567 mL, 0.921 g, 4.00 mmol,
2.00 equiv). Over the next 3 min, the temperature rose gradually
to -62 °C, at which time the reaction mixture was quenched with
1.0 mL of methanol, which discharged the orange color to yellow.
To the mixture was added 156.3 mg of n-Bu2O as GC internal
standard. The mixture was assayed twice by quantitative GC,54 and
the results were averaged. The following products were detected:
n-C5H11Br (mw 151.05; 36.14 mg, 0.239 mmol, 12.0% yield),
n-BuCHBr2 (mw 229.95; 82.11 mg, 0.357 mmol, 17.9% yield),
and cis- and/or trans-decene (mw 140.27; 144.96 mg, 1.033 mmol,
51.7% yield). The product mixture was poured into 40 mL of 3%
aq NaOH, extracted with CH2Cl2 (3 × 15 mL), dried (MgSO4),
and concentrated to a pale yellow oil (weight: 0.8881 g), consisting
of a mixture of two components (Rf ) 0.27 and 0.72) by TLC
(eluent: 25% ethyl acetate/cyclohexane). The oil was flash chro-
matographed on 60 g of silica gel (3 cm × 6′′ column; fraction
size: 50 mL; gradient elution: 5 f 20% ethyl acetate/cyclohexane).
The fractions containing the Rf ) 0.72 component were combined
and concentrated to a yellow oil identified as olefin 7d (trans/cis
) 73/27) by NMR spectroscopy. Weight: 0.3984 g (mw 368.95;
1.0798 mmol, 54.0% yield). 13C NMR (75 MHz, CD2Cl2): δ 145.0
(s), 129.4 (d), 129.3 (d), 125.8 (d), 113.1 (d), 112.0 (s), 81.2 (d),
77.0 (d), 59.8 (d), 45.0 (t), 37.6 (d), 32.9 (t), 32.0 (t), 31.5 (t), 31.1
(t), 29.3 (t), 26.3 (q), 26.2 (q), 25.0 (t), 22.7 (t), 14.2 (q). 1H NMR
(300 MHz, CD2Cl2): δ 0.88 (3H, distorted t, J ) 7.0 Hz), 1.29
(49) Since the half-life of LiCH2Br is much shorter than the CH2Br2
add time (2 min), the steady-state concentration of LiCH2Br is low
throughout the course of the reaction, while the concentration of LiBr
increases linearly with the amount of CH2Br2 that has been added.
(50) Nakamura, M.; Hirai, A.; Nakamura, E. J. Am. Chem. Soc. 2003,
125, 2341.
(51) Titrated by the procedure described in: Watson, S. C.; Eastham, J.
F. J. Organomet. Chem. 1967, 9, 165.
(52) GC assay: column, 8′ 20% SE30; temperature program, 60 f 140
°C at 4 °C/min; attenuation, 8 × 10-11; sample prep, inject 5 µL of neat
reaction mixture; retention times: ethylene, 0.57 min; CH2Br2, 4.75 min;
n-BuBr, 5.97 min; n-octane (ISTD), 9.80 min; n-C5H11Br, 10.90 min;
n-C6H13Br, 16.19 min. Yields were determined relative to n-octane (present
in commercial n-BuLi/hexane) as internal standard. No n-nonane (product
of the Wurtz-Fittig coupling of n-BuLi and n-C5H11Br) could be detected
in the crude reaction mixtures by GC. Therefore, no n-octane (product of
the Wurtz-Fittig coupling between n-BuLi and n-BuBr) was formed under
the conditions of the experiments, thus validating the use of n-octane as
internal standard.
(53) Truce, W. E.; Vriesen, C. W. J. Am. Chem. Soc. 1953, 75, 5032.
(54) GC assay: column, 8′ 20% SE30; temperature program, 60 f 180
°C at 4 °C/min.; attenuation setting, 8 × 10-11; injector temp, 275 °C;
detector temp, 290 °C; sample prep, inject 5 µL of neat reaction mixture;
retention times: n-BuBr, 4.59 min; n-octane, 7.78 min; n-C5H11Br, 8.95
min; n-Bu2O (ISTD), 12.11 min; cis-5-decene, 18.27 min; trans-5-decene,
18.45 min; n-BuCHBr2, 20.13 min. cis- and trans-5-Decene were not
sufficiently resolved to permit individual quantitation. Response factors were
determined for each component by injection of pure samples purchased
from commercial supply houses.
J. Org. Chem, Vol. 71, No. 15, 2006 5655

Pearlman et al.
(3H, s), 1.46 (3H, s), 1.3-2.5 (11H, m’s), 1.83 (2H, quint, J ) 6.9
Hz), 2.22 (2H, q, J ) 7.2 Hz), 2.61 (1H, dd, J ) 14.3, 6.2 Hz),
3.55 (2H, t, J ) 6.6 Hz), 3.99 (1H, td, J ) 11.2, 6.6 Hz), 4.46 (1H,
t, J ) 3.5 Hz), 5.3-5.6 (3H, m’s), 5.97 (1H, d, J ) 3.4 Hz). MS
(EI): m/e ) 368/370 (P+, 3.0%), 353/355 (27.3%), 251/253
(83.3%), 193 (100%). Anal. Calcd for C21H3335ClO3: m/e 368.2118.
Found: 368.2123. The fractions containing the Rf ) 0.27 component
were combined and concentrated to a white solid identified as
sulfone 1 by 13C and 1H NMR comparison with an authentic sample.
Weight: 0.4230 g (mw 455.02; 0.9296 mmol, 46.5% recovery).
Vinyl Chloride 7a (Table 7, Entry 3). A solution of sulfone 1
(mw 455.02; 0.3001 g, 0.6595 mmol) in 3.0 mL of THF was cooled
to -78 °C and treated with n-BuLi/hexane (1.65 mL of 1.6 M
solution, 2.64 mmol, 4.00 equiv), followed by neat diisopropylamine
(mw 101.19; d 0.722; 0.37 mL, 0.267 g, 2.64 mmol, 4.00 equiv).
After stirring at -78 °C for 7 min, the yellow solution was treated
with neat CH2BrCl (mw 129.38; d 1.991; 0.13 mL, 0.259 g, 2.00
mmol, 3.03 equiv), which caused the mixture to turn dark brown.
The mixture was allowed to warm to 0 °C, quenched with 0.4 mL
water, poured into 25 mL 3% aq NaOH, extracted with CH2Cl2 (3
× 25 mL), dried (MgSO4), and concentrated to a dark brown oil
(weight: 0.2588 g) consisting of one major (Rf ) 0.46) and one
minor (Rf ) 0.40) component, along with lesser amounts of more
polar components by TLC (eluent: 25% ethyl acetate/cyclohexane).
No starting sulfone 1 (Rf ) 0.08) could be detected. The dark brown
oil was flash chromatographed on 20 g of silica gel (gradient elution,
5 f 15% ethyl acetate/cyclohexane). The fractions containing the
Rf ) 0.46 and 0.40 components were combined and concentrated
to a pale yellow oil identified as vinyl chloride 7a as an 82:18
trans/cis isomer mixture by 1H and 13C NMR. None of the
corresponding vinyl bromide 7b was detected (estimated threshold
of detection: 1%). Weight: 0.2055 g (mw 347.29; 0.5917 mmol,
89.7% yield). 13C NMR (75 MHz, CD2Cl2): δ 149.2 (s), 130.1
(d), 128.4 (d), 113.8 (d), 113.2 (d), 112.3 (s), 80.1 (d), 76.6 (d),
59.9 (d), 44.9 (t), 38.0 (d), 32.8 (t), 32.6 (t), 30.6 (t), 26.20 (q),
(d; coupled to proton at 1.58), 44.4 (t; coupled to protons at 3.54),
38.9 (d; coupled to proton at 2.50), 34.9 (t; coupled to protons at
2.21 and 2.79), 32.1 (t; coupled to protons at 1.84), 30.1 (t; coupled
to protons at 2.21 and 2.50), 26.1 (q; coupled to protons at 1.51),
26.0 (q; coupled to protons at 1.32). 1H NMR (300 MHz, CDCl3):
δ 1.32 (3H, s), 1.51 (3H, s), 1.58 (1H, td, J ) 11.3, 3.4 Hz), 1.84
(2H, quint, J ) 6.8 Hz), 2.21 (4H, overlapping m’s; NOE on irr at
2.79), 2.50 (2H [H8 and one H7 proton], overlapping m’s; NOE
on irr at 6.11), 2.79 (1H [H10â], dd, J ) 15.5, 5.8 Hz), 3.54 (2H,
t, J ) 6.4 Hz), 4.08 (1H [H11], td, J ) 11.2, 6.7 Hz; NOE on irr
at 2.50, 2.79), 4.48 (1H, t, J ) 3.4 Hz), 5.48 (2H [H5 and H6],
sym m; NOE on irr at 2.50), 6.04 (1H, d, J ) 3.3 Hz), 6.11 (1H
[H9′], br s; NOE on irr at 2.50, no NOE on irr at 2.79). MS (EI):
m/e 375/377/379 (P+ - CH3, 100%), 332/334/336 (30.4%), 253,
255 (22.7%). Anal. Calcd for C16H2179Br35ClO3 (P+ - CH3): m/e
375.0362. Found: 375.0376.
Vinyl Iodide 7c (Table 7, Entry 5). A solution of sulfone 1
(mw 455.02; 0.4581 g, 1.0068 mmol) in 5 mL of THF was cooled
to -78 °C under argon and treated with n-BuLi/hexane (2.00 mL
of 1.6 M solution, 3.20 mmol, 3.18 equiv), producing an orange
color. The mixture was treated with neat diisopropylamine (mw
101.19; d 0.722; 0.56 mL, 0.404 g, 4.00 mmol, 3.97 equiv), stirred
at -78 °C for 9 min, then treated with neat CH2I2 (mw 267.84; d
3.325; 0.24 mL, 0.798 g, 2.98 mmol, 2.96), producing a dark purple
solution. The mixture was allowed to warm to 0 °C, stirred at 0 °C
for 10 min, quenched with 3% aq NaOH, and extracted with CH2-
Cl2 (3×). The purple solution was shaken with aq Na2S2O3, which
discharged the color, dried (MgSO4), and concentrated to a purple
oil (weight: 0.5037 g) consisting of one component (Rf ) 0.59)
by TLC (eluent: 30% ethyl acetate/cyclohexane); sulfone 1 (Rf )
0.21) was not detectable. The oil was flash chromatographed on
silica gel (2 cm × 6′′ column; gradient elution, 10 f 20% ethyl
acetate/cyclohexane). The fractions containing the Rf ) 0.59
component were combined and concentrated to a residue identified
as vinyl iodide 7c in pure form (as 93:7 trans/cis mixture, by ratio
of heights of peak pairs in 13C NMR spectrum) by spectroscopic
analysis. Weight: 0.3818 g (mw 438.74; 0.8702 mmol, 86.4%
yield). 13C NMR (75 MHz, CD2Cl2): δ 158.4 (s), 130.2 (d), 128.3
(d), 113.0 (d), 112.3 (s), 79.6 (d), 77.0 (d), 74.4 (d), 60.5 (d), 45.0
(t), 41.0 (d), 39.8 (t), 32.7 (t), 30.4 (t), 26.2 (q), 26.2 (q), 25.0 (t).
1H NMR (90 MHz, CDCl3): δ 1.38 (3H, s), 1.55 (3H, s), 1.5-2.9
(10H, m’s), 3.55 (2H, t, J ) 6 Hz), 4.10 (1H, td, J ) 11, 7 Hz),
4.50 (1H, t, J ) 4 Hz), 5.49 (2H, sym m), 6.04 (1H, d, J ) 3 Hz),
6.19 (1H, narrow m). MS (EI): m/e 438/440 (P+, 0.7%), 423/425
(P+ - CH3, 69.5%), 311/313 (57.7%), 81 (100%). Anal. Calcd for
C16H2135ClIO3 (P - CH3): m/e 423.0223. Found: 423.0243.
Cyclobutane Dimer 16 (Scheme 8). A solution of sulfone 1
(mw 455.02; 1.5155 g, 3.331 mmol) in 12 mL THF was cooled to
-78 °C and treated with n-BuLi/hexane (6.25 mL of 1.6 M solution,
10.00 mmol, 3.00 equiv), followed by a solution of diisopropyl-
amine (mw 101.19; 1.3461 g, 13.30 mmol, 3.99 equiv) in 2.0 mL
of THF. After stirring at -78 °C for 10 min, the yellow solution
was treated dropwise over 6 min with a solution of CH2Br2 (mw
173.83; 1.7597 g, 10.123 mmol, 3.04 equiv) in 8 mL of THF. After
the addition was complete, the reaction mixture was allowed to
warm to 0 °C over 15 min, stirred at 0 °C for 10 min, poured into
40 mL of 3% aq NaOH, extracted with CH2Cl2 (2 × 25 mL), dried
(MgSO4), and concentrated to a light brown glass (weight: 1.1701
g) consisting of one major component (Rf ) 0.39) contaminated
with trace amounts of vinyl bromide 7b (Rf ) 0.58 with shoulder
at 0.56) and sulfone 1 (Rf ) 0.20) by TLC (eluent: 20% ethyl
acetate/cyclohexane). This glass was flash chromatographed on 69
g of silica gel (eluent: 15% ethyl acetate/cyclohexane). The
fractions containing the Rf ) 0.39 component were pooled and
concentrated to a white glass, identified as cyclobutane dimer 16
as a single isomer (stereochemistry about the cyclopropane ring is
1
26.18 (q), 24.9 (t). H NMR (300 MHz, CDCl3): δ 1.30 (3H, s),
1.48 (3H, s), 1.56 (1H, td, J ) 11.4, 3.5 Hz), 1.84 (2H, quint, J )
6.9 Hz), 2.21 (4H, narrow m), 2.52 (2H, m’s), 2.79 (1H, dd, J )
15.2, 6.6 Hz), 3.55 (2H, t, J ) 6.5 Hz), 4.04 (1H, td, J ) 11.2, 6.7
Hz), 4.48 (1H, t, J ) 3.4 Hz), 5.48 (2H, sym m), 6.00 (1H, d, J )
3.4 Hz), 6.00 (1H, m). MS (EI): m/e 331/333/335 (P+ - CH3,
100%), 43 (83.6%). Anal. Calcd for C16H2135ClO3 (P+ - CH3):
m/e 331.0868. Found: 331.0852.
Vinyl Bromide 7b (Table 7, Entry 4). A solution of sulfone 1
(mw 455.02; 0.2389 g, 0.5250 mmol) in 2.0 mL of THF was cooled
to -78 °C and treated with n-BuLi/hexane (1.00 mL of 1.6 M
solution, 1.60 mmol, 3.05 equiv), which produced a bright yellow
color. The mixture was treated with neat diisopropylamine (mw
101.19; d 0.722; 0.32 mL, 0.231 g, 2.28 mmol, 4.35 equiv), stirred
at -78 °C for 7.5 min, and then treated with neat CH2Br2 (mw
173.85; d 2.477; 0.11 mL, 0.272 g, 1.57 mmol, 2.99 equiv), causing
the mixture to turn brown immediately. The mixture was warmed
to 0 °C, quenched with 0.5 mL methanol, added to 20 mL 3% aq
NaOH, extracted with CH2Cl2 (2 × 20 mL), dried (MgSO4), and
concentrated to a dark brown oil (weight: 0.2108 g) consisting of
a mixture of one major (Rf ) 0.52) and one minor (Rf ) 0.33)
component by TLC (eluent: 20% ethyl acetate/cyclohexane);
sulfone 1 (Rf ) 0.16) was not detectable. The oil was flash
chromatographed on 20 g of silica gel (2 cm × 6′′ column; gradient
elution, 10 f 20% ethyl acetate/cyclohexane). The fractions
containing the Rf ) 0.52 component were combined and concen-
trated to a brownish yellow oil identified as vinyl bromide 7b in
pure form (as 93:7 trans/cis mixture, by ratio of heights of peak
pairs in 13C NMR spectrum) by spectroscopic analysis. Weight:
0.1913 g (mw 391.74; 0.488 mmol, 93.0% yield). 13C NMR (75
MHz, CDCl3): δ 151.4 (s), 129.8 (d; coupled to proton at 5.48),
128.0 (d; coupled to proton at 5.48), 112.6 (d; coupled to proton at
6.04), 112.1 (s), 102.2 (d; coupled to proton at 6.11), 79.6 (d;
coupled to proton at 4.08), 76.3 (d; coupled to proton at 4.48), 59.6
1
undetermined) by 500 MHz H NMR, 13C NMR, MS, and TLC.
Weight: 0.6130 g (mw 621.65; 0.9861 mmol, 59.2% yield). 13C
NMR (spectrum taken on Varian XL-200 in CD2Cl2): δ 113.0 (1C,
5656 J. Org. Chem., Vol. 71, No. 15, 2006

Lithium Carbenoids for Methylenation of Sulfones
d), 112.0 (1C, s), 82.7 (1C, d), 76.3 (1C, d), 74.6 (1C, s), 58.6
(1C, d), 48.2 (1C, s), 45.4 (1C, t), 43.7 (1C, d), 33.5 (1C, t), 31.5
(2C, t), 31.5 (1C, d), 26.5 (1C, q), 25.9 (1C, q), 25.6 (1C, d), 22.4
(Rf ) 0.81 and 0.75) and polar material by TLC (eluent: 3% ethyl
acetate/cyclohexane) together with a small amount of starting nitrile
(Rf ) 0.27). The dark brown solid was flash chromatographed on
42 g of silica gel (eluant: cyclohexane). The fractions containing
the Rf ) 0.81 and 0.75 components were pooled and concentrated
to a colorless oil, identified as 1-chloro-2-methyl-1-tetradecene 17
(as a 3:2 mixture of geometric isomers) by 1H and 13C NMR.
Weight: 0.2408 g (mw 244.85; 0.9835 mmol, 55.0% yield). 13C
NMR (20 MHz, CD2Cl2): δ 139.4 (s), 112.0 (∼0.4C, d), 111.5
(∼0.6 C, d), 37.4 (t), 32.4 (t), 32.2 (t), 29.5-30.1 (5 overlapping
1
(1C, t). H NMR (500 MHz, CD2Cl2): δ 0.78 (1H, td, J ) 8.74,
5.89 Hz), 1.24 (3H, s), 1.40 (3H, s), 1.40 (1H, ddd, J ) 8.74, 7.84,
2.40 Hz), 1.49 (1H, m), 1.50 (1H, ddd, J ) 14.95, 4.91, 2.40 Hz),
1.67 (1H, m), 1.70 (1H, t, J ) 11.56 Hz), 1.72 (1H, td, J ) 11.1,
3.46 Hz), 1.80 (2H, m), 2.17 (1H, dd, J ) 11.56, 5.6 Hz), 2.31
(1H, ddd, J ) 14.95, 10.70, 7.84), 2.66 (1H, td, J ) 10.70, 4.91
Hz), 3.56 (2H, t, J ) 6.51 Hz), 3.63 (1H, ddd, J ) 11.56, 11.15,
5.6 Hz), 4.93 (1H, t, J ) 3.46 Hz), 5.91 (1H, d, J ) 3.46 Hz). MS
(CI): m/e 621/623/625 (P+ + H, 60.9%), 563/565/567 (100%).
Anal. Calcd for C34H4635Cl2O6: m/e 620.2671. Found: 620.2668.
Vinyl Chloride 17 from 2-Methyl-tetradecanitrile (Scheme
9). A solution of diisopropylamine (mw 101.19; d 0.722; 0.27 mL,
0.1949 g, 1.93 mmol, 1.08 equiv) in 5.0 mL of THF was cooled to
0 °C, treated with n-BuLi/hexane (1.18 mL of 1.6 M solution, 1.89
mmol, 1.06 equiv), stirred for 10 min, cooled to -78 °C, and treated
with a solution of 2-methyl-tetradecanitrile (mw 223.41; 399.7 mg,
1.789 mmol) in 3.0 mL THF. After 6 min at -78 °C, n-BuLi/
hexane (3.4 mL, 5.44 mmol, 3.04 equiv) was added, followed by
CH2Cl2 (mw 84.93; d 1.325; 0.34 mL, 0.450 g, 5.30 mmol, 2.96
equiv). The mixture was stirred at -78 °C for 2 min, then warmed
to 0°, poured into 75 mL of 3% aq NaOH, extracted with CH2Cl2
(2 × 75 mL), dried (MgSO4), and concentrated to a dark brown
solid (weight: 0.5807 mg) consisting of two major components
1
t’s), 28.0 (t), 27.2 (t), 23.1 (t), 20.9 (q), 14.2 (q). H NMR (90
MHz, CD2Cl2): δ 0.89 (3H, t, J ) 6 Hz), 1.30 (20H, br s), 1.72
(∼1.8H, d, J ) 1.5 Hz), 1.75 (∼1.2H, d, J ) 1.5 Hz), 2.1 (2H, m),
5.76 (1H, narrow m). MS (EI): m/e 244/246 (P+, 100%). Anal.
Calcd for C15H2935Cl: m/e 244.1958. Found: 244.1954.
Acknowledgment. The authors thank Dr. Stephen A. Mizsak
(Physical and Analytical Chemistry) for the NMR NOE experi-
ment and a referee for an extensive and very valuable review.
Supporting Information Available: Experimental procedures,
1
compound characterization data, 13C and H NMR spectra for all
new compounds, and derivation of eq 1. This material is available
free of charge via the Internet at http://pubs.acs.org.
JO060669+
J. Org. Chem, Vol. 71, No. 15, 2006 5657