A study of factors affecting α-(N-carbamoyl)alkylcuprate chemistry

Article abstract of DOI:10.1021/jo0056702

The effect of Cu(I) salt (i.e., CuCN, CuCN·2LiCl, CuI), cuprate reagent, sec-butyllithium quality, solvent, and temperature upon the chemical yields obtained in the reactions of α-(N-carbamoyl)alkylcuprates [i.e., N-Boc-protected α-aminoalkylcuprates] with (E)1-iodo-1-hexene, 5,5-dimethyl-2-cyclohexenone, methylvinyl ketone, crotonate esters, and an acid chloride has been examined. Cuprate conjugate addition and vinylation reactions can succeed with low-quality sec-butyllithium, presumably containing insoluble lithium hydride and lithium alkoxide impurities, although yields are significantly lower than those obtained with high-quality s-BuLi, α-(N-Carbamoyl)alkylcuprates prepared from high-quality sec-butyllithium are thermally stable for 2-3 h at room temperature and are equally effective when prepared from either insoluble CuCN or THF-soluble CuCN·2LiCl. Use of the latter reagent permits rapid cuprate formation at -78 °C, thereby avoiding the higher temperatures required for cuprate formation from THF-insoluble CuCN that are problematic with solutions containing thermally unstable α-lithiocarbamates.

Full text of DOI:10.1021/jo0056702

2302
J . Org. Chem. 2001, 66, 2302-2311
A Stu d y of F a ctor s Affectin g r-(N-Ca r ba m oyl)a lk ylcu p r a te
Ch em istr y
R. Karl Dieter,* Chris M. Topping, and Lois E. Nice
Hunter Laboratory, Department of Chemistry, Clemson University, Clemson, South Carolina 29634-1905
dieterr@clemson.edu
Received October 4, 2000
The effect of Cu(I) salt (i.e., CuCN, CuCN‚2LiCl, CuI), cuprate reagent, sec-butyllithium quality,
solvent, and temperature upon the chemical yields obtained in the reactions of R-(N-carbamoyl)-
alkylcuprates [i.e., N-Boc-protected R-aminoalkylcuprates] with (E)1-iodo-1-hexene, 5,5-dimethyl-
2-cyclohexenone, methylvinyl ketone, crotonate esters, and an acid chloride has been examined.
Cuprate conjugate addition and vinylation reactions can succeed with low-quality sec-butyllithium,
presumably containing insoluble lithium hydride and lithium alkoxide impurities, although yields
are significantly lower than those obtained with high-quality s-BuLi. R-(N-Carbamoyl)alkylcuprates
prepared from high-quality sec-butyllithium are thermally stable for 2-3 h at room temperature
and are equally effective when prepared from either insoluble CuCN or THF-soluble CuCN‚2LiCl.
Use of the latter reagent permits rapid cuprate formation at -78 °C, thereby avoiding the higher
temperatures required for cuprate formation from THF-insoluble CuCN that are problematic with
solutions containing thermally unstable R-lithiocarbamates.
In tr od u ction
salt often plays a significant role in the chemistry of the
resultant organocopper and cuprate reagents, and CuCN
and CuBr‚SMe2 have been promoted as the best Cu(I)
precursors.4 Organocopper reagents prepared from CuCN
and either 1 or 2 equiv of an alkyllithium reagent often
display superior attributes with regard to chemical yields,
suggesting an important role of the cyanide anion.1a The
structure of the latter reagent has been the subject of
considerable controversy centering around the location
of the cyanide ion.5 Although THF-soluble CuCN‚2LiCl
has been extensively used by Knochel in the copper-
promoted reactions of organozinc reagents,6 this source
of Cu(I) has not been widely used in the generation of
cuprates from alkyllithium or Grignard reagents.7 The
use of CuCN‚2LiCl appeared to be crucial for the conju-
gate addition8 and coupling reactions9 of R-(N-carbamoyl)-
alkylcuprates with R,â-unsaturated carboxylic acid de-
rivative and vinyl iodides, respectively. This plethora of
effects generally necessitates an empirical development
of optimal conditions for any particular reaction.
Although they effect a wide range of transformations
and are widely employed in synthesis, organocopper
reagents have a well-earned reputation for capriciousness
and sensitivity to reaction conditions.1 They are incom-
patible with air, moisture, and oxidants and often display
thermal instability in the same temperature region in
which the desired transformation occurs.2 Reactivity,
chemical yields, and regio- and stereoselectivity of orga-
nocopper-mediated reactions are dependent upon solvent,
additives (e.g., Lewis acids, TMSCl, TMSCl/HMPA, etc.),1,3
co-metal (e.g., Li, Mg, Zn), and the Cu(I) salt4 used in
reagent preparation, as well as the particular reagent
employed (e.g., Gilman reagents, mixed homocuprates,
heterocuprates, organocopper reagents). The copper(I)
* To whom correspondence should be addressed. Fax: 864-656-6613.
(1) For reviews on organocopper chemistry, see: (a) Lipshutz, B.
H.; Sengupta, S. Org. React. 1992, 41, 135. (b) Lipshutz, B. H. In
Organometallics in Synthesis: A Manual; Schlosser, M., Ed.; J ohn
Wiley & Sons: Chichester, 1994; Chapter 4. (c) Organocopper Re-
agents: A Practical Approach; Taylor, R. J . K., Ed.; Oxford University
Press: Oxford, 1994.
(2) For studies on the thermal stability of cuprates, see: (a) Posner,
G. H.; Whitten, C. E.; Sterling, J . J . J . Am. Chem. Soc. 1973, 95, 7788.
(b) Bertz, S. H.; Dabbagh, G. J . Chem. Soc., Chem. Commun. 1982,
1030.
(3) For the effect of additives, see: TMSCl: (a) Boaz, N. W.; Corey,
E. J . Tetrahedron Lett. 1985, 26, 6019. (b) Alexakis, A.; Berlan, J .;
Besace, Y. Tetrahedron Lett. 1986, 27, 1047. (c) J ohnson, C. R.; Marren,
T. J . Tetrahedron Lett. 1987, 28, 27. (d) Lindstedt, E.-L.; Nilsson, M.;
Olsson, T. J . Organomet. Chem. 1987, 334, 255. (e) Lipshutz, B. H.;
Ellsworth, E. L.; Siahaan, T. J .; Shirazi, A. Tetrahedron Lett. 1988,
29, 6677. (f) Bertz, H. B.; Miao, G.; Rossiter, B. E.; Snyder, J . P. J .
Am. Chem. Soc. 1995, 117, 11023. (g) Frantz, D. E.; Singleton, D. A.
J . Am. Chem. Soc. 2000, 122, 3288; and references therein. TMSCl/
HMPA or TMSCl/DMAP: (h) Matsuzawa, S.; Horiguchi, Y.; Nakamura,
E.; Kuwajima, I. Tetrahedron 1989, 45, 349. Lewis acids: (i) Yama-
moto, Y.; Yamamoto, S.; Yatagai, H.; Ishihara, Y.; Maruyama, K. J .
Org. Chem. 1982, 47, 119. (j) Lipshutz, B. H.; Parker, D. A.; Kozlowski,
J . A. Tetrahedron Lett. 1984, 25, 5959. (k) Lipshutz, B. H.; Sclafani,
J . A.; Takanami, T. J . Am. Chem. Soc. 1998, 120, 4021.
Reactivity can be a major problem in organocuprate
chemistry. Although numerous homo and mixed organo-
cuprates readily transfer alkyl ligands to the â-carbon
atom of R,â-enones, the conjugate addition reaction is
often sluggish with less reactive enoate substrates.
Cuprates prepared from CuCN (2RLi + CuCN),10 cuprous
(5) (a) Bertz, S. H.; Miao, G.; Eriksson, M. J . Chem. Soc., Chem.
Commun. 1996, 815. (b) Snyder, J . P.; Bertz, S. H. J . Org. Chem. 1995,
60, 4312. (c) Barnhart, T. M.; Huang, H.; Penner-Hahn, J . E. J . Org.
Chem. 1995, 60, 4310. (d) Lipshutz, B. H.; J ames, B. J . Org. Chem.
1994, 59, 7585. (e) Krause, N. Angew. Chem., Int. Ed. Engl. 1999, 38,
79; and references therein.
(6) Knochel, P.; Singer, R. D. Chem. Rev. 1993, 93, 2117.
(7) (a) Ba¨ckvall, J .-E.; Persson, E. S. M.; Bombrun, A. J . Org. Chem.
1994, 59, 4126. (b) Schlosser, M.; Bossert, H. Tetrahedron 1991, 47,
6287.
(8) (a) Dieter, R. K.; Velu, S. E. J . Org. Chem. 1997, 62, 3798. (b)
Dieter, R. K.; Lu, K. Tetrahedron Lett. 1999, 40, 4011. (c) Dieter, R.
K.; Lu, K.; Velu, S. E. J . Org. Chem. 2000, 65, 8715.
(4) Bertz, S. H.; Gibson, C. P.; Dabbagh, G. Tetrahedron Lett. 1987,
28, 4251.
(9) Dieter, R. K.; Sharma, R. R. Tetrahedron Lett. 1997, 38, 5937.
(10) Lipshutz, B. H. Tetrahedron Lett. 1983, 24, 127.
10.1021/jo0056702 CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/13/2001

R-(N-Carbamoyl)alkylcuprate Chemistry
J . Org. Chem., Vol. 66, No. 7, 2001 2303
thiophenoxides,11 and copper(I) trimethylsilylacetylide12
have provided partial solutions to this problem. The
reactivity of the latter reagent was enhanced by solvent
composition (4:1 ether/THF) and by the addition of
chlorotrimethylsilane. Gilman reagents effect high-yield
conjugate addition reactions with mono-â-alkyl-substi-
tuted enoates, enamides, and enecarbamates in ether in
the presence of TMSCl.13 Alkylcopper compounds also
participate with enoates in conjugate addition reactions
in synthetically useful yields in the presence of additives
such as BF3‚Et2O and chloro- or iodotrimethylsilane.14
During the course of our developmental work on R-(N-
carbamoyl)alkylcuprates, we experienced considerable
difficulty across the range of characteristic cuprate
reactions. Conjugate addition reactions of R-(N-carbam-
oyl)alkylcuprates prepared from CuCN to enones15b,d
appeared sensitive to diamine additives, solvent, and the
quality of the s-BuLi employed in the deprotonation of
N-tert-butoxycarbonyl (Boc)-protected amines.16 Addition-
ally, conjugate addition reactions with carboxylic acid
derivatives8 and vinylation reactions with vinyl iodides9
were only successful when CuCN‚2LiCl was employed.
Having developed the first examples of N-protected
R-aminoalkylcuprate (i.e., R-cuprio formamidines, car-
bamates, and azo compounds) participation in conjugate
addition reactions8,15 and of R-(N-carbamoyl)alkylcu-
prates in substitution reactions with acid chlorides,17a
vinyl triflates,17b vinyl iodides,9 allylic substrates,17c and
propargylic17d substrates, we now had the opportunity to
examine the influence of various reaction parameters
upon the reactions of these cuprates containing an
R-heteroatomalkyl ligand. We now report a study of the
factors affecting R-(N-carbamoyl)alkylcuprate chemistry
including the influence of s-BuLi quality, Cu(I) salt
employed, cuprate thermal stability, and additives (e.g.,
TMSCl, diamines, LiCl, s-BuOLi, etc.).
obtained by use of (-)-sparteine or freshly distilled
TMEDA, although the yields were significantly lower
than those obtained from cuprates prepared via the
intermediate R-(N-carbamoyl)alkylstannanes. Ultimately,
the use of CuCN‚2LiCl for cuprate formation reproducibly
gave high chemical yields comparable to those obtained
via the stannane protocol (eq 1). More recently, the use
of freshly distilled (-)-sparteine and careful manipulation
of reaction temperatures during cuprate preparation and
reaction afforded high chemical yields in the reactions
of R-(N-carbamoyl)alkylcuprates prepared from insoluble
CuCN with enones (eq 1), enoates, and vinyl iodides that
are comparable to those obtained with THF-soluble
CuCN‚2LiCl or via the R-(N-carbamoyl)alkylstannane
procedure. In these reactions, addition of solid CuCN
(-78 °C) to the R-(N-carbamoyl)alkyllithium reagent was
followed by stirring at -78 °C for 30 min, removal of the
cold bath permitting the cold flask to stand at ambient
air temperature for 10-15 min, and then cooling the flask
to -78 °C before addition of the electrophile. The chemi-
cal yields of enone conjugate adducts also appeared to
vary with the physical appearance of the s-BuLi em-
ployed in the carbamate deprotonations.16
Resu lts
During the early work on R-(N-carbamoyl)alkylcuprate
conjugate additions to R,â-enones, the reaction often
failed or gave capricious yields when insoluble CuCN was
used in conjunction with TMEDA employed to assist
carbamate deprotonation (eq 1).15b,d Subsequently, it was
discovered that reliable and reproducible yields could be
Given the capriciousness of these reactions and the
accumulated empirical observations required to obtain
reliable and reproducible results, we undertook a study
to examine the thermal stability of R-(N-carbamoyl)-
alkylcuprates and factors such as temperature, Cu(I) salt,
s-BuLi quality, and additives (e.g., alkoxide impurities
in the s-BuLi) that might affect the reliability of the
reactions.
The effect of s-BuLi quality was examined using the
conjugate addition reaction with cyclohexenone and the
vinylation reaction with trans-1-iodo-1-hexene (eq 2). A
few examples from a preliminary study using an acid
chloride are also included. Commercial samples of s-BuLi
range in appearance from clear colorless solutions to
those having varying amounts of particulates, assumed
to be LiH.18 Thermal and chemical (i.e., reaction with
oxygen, H2O, or CO2) decomposition of s-BuLi leads to
the formation of LiH, s-BuOLi, LiOH, Li2O, and s-BuCO2-
Li, all of which could have a deleterious effect upon R-(N-
carbamoyl)alkylcuprate formation and reactivity. The
rate of s-BuLi decomposition is dependent upon temper-
(11) Behforouz, M.; Curran, T. T.; Bolan, J . L. Tetrahedron Lett.
1986, 27, 3107.
(12) Sakata, H.; Kuwajima, I. Tetranhedron Lett. 1987, 28, 5719.
(13) Alexakis, A.; Berlan, J .; Besace, Y. Tetrahedron Lett. 1986, 27,
1047.
(14) (a) Bergdahl, M.; Lindstedt, E.-L.; Nilsson, M.; Olsson, T.
Tetrahedron 1988, 44, 2055. (b) Bergdahl, M.; Lindstedt, E.-L.; Nilsson,
M.; Olsson, T. Tetrahedron 1989, 45, 535. (c) Bergdahl, M.; Eriksson,
M.; Nilsson, M.; Olsson, T. J . Org. Chem. 1993, 58, 7238. (d) Bergdahl,
M.; Iliefski, T.; Nilsson, M.; Olsson, T. Tetrahedron Lett. 1995, 36, 3227.
(e) Eriksson, M.; Hjelmencrantz, A.; Nilsson, M.; Olsson, T. Tetrahe-
dron 1995, 51, 12631. (f) Eriksson, M.; Nilsson, M.; Olsson, T. Synlett
1994, 271. (g) Bergdahl, M.; Eriksson, M.; Nilsson, M.; Olsson, T. J .
Org. Chem. 1993, 58, 7238.
(15) (a) Dieter, R. K.; Alexander, C. W. Tetrahedron Lett. 1992, 33,
5693. (b) Dieter, R. K.; Alexander, C. W. Synlett 1993, 407. (c)
Alexander, C. W.; Lin, S.-Y.; Dieter, R. K. J . Organomet. Chem. 1995,
503, 213. (d) Dieter, R. K.; Alexander, C. W.; Nice, L. E. Tetrahedron
2000, 56, 2767.
(16) (a) Beak, P.; Lee, W. K. Tetrahedron Lett. 1989, 30, 1197. (b)
Beak, P.; Lee, W. K. J . Org. Chem. 1993, 58, 1109.
(17) (a) Dieter, R. K.; Sharma, R. R.; Ryan, W. Tetrahedron Lett.
1997, 38, 783. (b) Dieter, R. K.; Dieter, J . W.; Alexander, C. W.;
Bhinderwala, N. S. J . Org. Chem. 1996, 61, 2930. (c) Dieter, R. K.;
Velu, S. E.; Nice, L. E. Synlett 1997, 1114. (d) Dieter, R. K.; Nice, L.
E. Tetrahedron Lett. 1999, 40, 4293.
(18) FMC Lithium Division, “Butyllithium Guidelines for Safe
Handling”, FMC Corp., 449 North Cox Rd., Gastonia, NC 28054-0020,
J une 1991; pp 3-4.

2304 J . Org. Chem., Vol. 66, No. 7, 2001
Dieter et al.
Ta ble 1. Effect of s-Bu Li Qu a lity on th e Rea ction s of Or ga n ocop p er Rea gen ts P r ep a r ed fr om Ca r ba m a te 1 a n d n Equ iv
of Cu Cl‚2LiCl or Cu CN‚2LiCl w ith Cycloh exen on e (5 equ iv of TMSCl), 4-MeOC6H4COCl, or (E)-1-Iod o-1-h exen e to Affor d ,
Resp ectively, 3, 5, or 6a (Eq 2)
CuX‚2LiCl
% yield productg
NMR ISh isolatedi
s-BuLi
total
rxne
entry sourcea RLi Mb,c basec (M) appearanced n equiv
X
additive
electrophile cond solventf
1
2
3
4
5
6
7
8
9
FMC
filtered
1.0
1.0
Cl
CN
acid chloride
cyclohexenone
acid chloride
acid chloride
cyclohexenone
acid chloride
cyclohexenone
A
A
A
A
A
A
A
A
A
A
B
A
C
B
A
D
D
D
D
D
D
THF
Et2O/THF
THF
THF
Et2O/THF
THF
Et2O/THF
Et2O/THF
Et2O/THF
Et2O/THF
THF
Et2O/THF
Et2O/THF
THF
91-100 88
54
particulates
clear sample
50
100
72
94
62
42
60
57
80
1.0
1.0
1.0
1.0
1.0
1.0
0.5
CN
Cl
CN
Aldrich
filtered
CN s-BuOH 0.25j cyclohexenone
CN s-BuOH 1.0j cyclohexenone
10
particulates
particulates
CN
CN
cyclohexenone
1-iodohexene
11 FMC A
12
13
1.1
1.6
93
70
89
14 FMC A
15 FMC A
16 FMC B
17 CT A
18 CT B
19 CT B
20 CT B
21 CT B
1.1
1.1
1.7
0.7
1.5
1.5
1.5
1.5
1.3
1.3
1.7
0.7
1.5
1.5
1.5
1.5
filtered, clear
filtered, clear
clear
clear
clear
clear
clear
clear
0.5
0.5
0.5
0.5
0.5
0.5
0.5
1.0
CN
CN
CN
CN
CN
1-iodohexene
1-iodohexane
1-iodohexene
1-iodohexene
1-iodohexene
84
Et2O/THF
THF
THF
THF
THF
67-80
94-100
92
67-73
78 (71)k
92
CN s-BuOH 0.3j 1-iodohexene
CN s-BuOH 0.6j 1-iodohexene
CN s-BuOH 1.0j 1-iodohexene
THF
THF
98
e5-33
a
FMC ) milky white material with particulates. FMC A ) material containing solid particulates and approximately 2 years old at
b
time of usage. FMC B ) new material freshly acquired at time of usage and clear. M determined by Eastham procedure using s-BuOH
c
d
and 1,10-phenanthrolein as indicator. M determined by Gilman double titration using dibromoethane. Appearance of material used in
carbamate deprotonation. e Reaction conditions include the following: (i) deprotonation, (ii) cuprate formation, and (iii) reaction with
electrophile. A ) [(i) s-BuLi, (-)-sparteine, THF or Et2O, -78 °C, 1 h; (ii) -78 to -50 °C, 30-40 min; (iii) -50 °C to rt (5 equiv of TMSCl
with 2-cyclohexenone), 2-5 h]; B ) (i) s-BuLi, (-)-sparteine, THF or Et2O, -78 °C, 1 h; (ii) -78 to -50 °C, 80 min; (iii) -50 °C (60 min),
rt (30 min); C ) (i) s-BuLi, (-)-sparteine, THF or Et2O, -78 °C, 30 min; (ii) -78 °C, 30 min; (iii) -50 °C (30 min and then quench); D )
(i) s-BuLi, (-)-sparteine, THF or Et2O, -78 °C, 45 min; (ii) -78 to -50 °C, 45 min; (iii) -50 °C (45 min), rt (30 min). f Deprotonations
performed in Et2O resulted in a 1:2 Et2O/THF solvent mixture via addition of THF solutions of CuCN‚2LiCl and then electrophile. g Yields
h
represent
a
single experiment unless reported as
a
range. Yields determined by NMR internal standard method (IS) using
i
j
tetrachloroethane. Isolated yields of products purified by column chromatography. s-BuOH was added to the carbamate solution and
mmol s-BuLi ) mmol carbamate + mmol s-BuOH. Yield obtained in the absence of (-)-sparteine otherwise used to assist deprotonation.
k
ature and amounts of s-BuOLi present in solution.18 In
a set of preliminary studies, old s-BuLi (approximately
0.3 to 2.5 years) containing particulates was filtered
through Celite (flame dried under vacuum prior to use),
and both the filtered and unfiltered material was used
for deprotonation of the Boc-protected amine (eq 2, Table
1). Formation of the R-(N-carbamoyl)alkylcuprate derived
from Boc-protected pyrrolidine (1) using filtered or
unfiltered s-BuLi and CuX‚2LiCl (X ) Cl, CN), followed
by reaction with 4-methoxybenzoyl chloride (entries 1, 6
vs 3, 88-94% vs 50%), 2-cyclohexenone (entries 2, 7 vs
10, 54-62% vs 57%), or 1-iodo-1-hexene (entries 14-15
vs 11-13, 67-80% vs 70-93%) generally gave compa-
rable product yields with the filtered and unfiltered
s-BuLi, respectively. The acylation reaction displayed
significantly better yields (38-44% higher) when filtered
s-BuLi was employed compared to the other two trans-
formations. The conjugate addition and acylation reac-
tions employed the cyanocuprate reagent, RCuCNLi, and
the observed yields of the better set (i.e., filtered or
unfiltered) are only slightly lower (6-15%) than the
optimized yields obtained with high-quality s-BuLi (en-
tries 4-5).15d,17a A more careful study was performed with
the vinylation reaction using several sources of s-BuLi
(Table 1, entries 11-18) including old material (FMC A,
filtered and unfiltered), high-quality commercial material
(FMC B), and material prepared in our laboratory (CT
A & B). The samples of s-BuLi used in carbamate
deprotonation were titrated according to the modified
Gilman double-titration procedure to give a measure of
both total base and active s-BuLi.1a,19 Watson-Eastham
titrations gave molarity values for s-BuLi that were
identical to those obtained using the Gilman double-
titration method.1a,20 Filtration of FMC A removed half
of the nonalkyllithium base, and both filtered and
unfiltered samples gave good yields (entries 11-15, 67-
89%) of product, while s-BuLi containing no extraneous
(19) (a) Gilman, H.; Cartledge, F. K. J . Organomet. Chem. 1964, 2,
447. (b) Kamienski, C. W.; Esmay, D. L. J . Org. Chem. 1960, 25, 115.
(c) For use of benzoic acid in toluene, see: Turner, R. R.; Altenau, A.
G.; Cheng, T. C. Anal. Chem. 1970, 42, 1835.
(20) (a) Watson, S. C.; Eastham, J . F. J . Organomet. Chem. 1967,
9, 165. (b) For titrations in ethereal solvents, see: Ellison, R. A.; Griffin,
R.; Kotsonis, F. N. J . Organomet. Chem. 1972, 36, 209. (c) FMC
Lithium Division, “Titration Methods for Commercial Organolithium
Compounds”, FMC Corporation, 449 North Cox Rd., Gastonia, NC
28054-0020, Winter 1994. (d) FMC Lithium Division- “Techniques for
Applying Watson-Eastham Method to Butyllithium Analysis”.

R-(N-Carbamoyl)alkylcuprate Chemistry
J . Org. Chem., Vol. 66, No. 7, 2001 2305
Ta ble 2. Rea ctivity a n d Th er m a l Sta bility of
r-(N-ca r ba m oyl)a lk ylcu p r a tes P r ep a r ed fr om Cu CN or
Cu CN‚2LiCl a n d Lith ia ted N-Boc-p yr r olid in e or
N-Boc-p ip er id in e As Mea su r ed by Rea ction w ith Meth yl
Vin yl Keton e (MVK) To Affor d 7 (Eq 3)
bases (FMC B, CT A & B) gave the desired product in
nearly quantitative yield for two of the three samples
(entries 16-17) and somewhat lower yield for the third
(CT B, entry 18). A diminished yield (7% decrease) was
obtained when the deprotonation was performed in the
absence of the diamine (-)-sparteine (entry 18). This
third s-BuLi sample (CT B) did give nearly quantitative
yields in subsequent experiments probing the effect of
added s-BuOLi (entries 19-20).
The effect of s-BuOLi was examined by addition of
s-BuOH to the carbamate solution followed by sufficient
s-BuLi to deprotonate the alcohol and carbamate.21 While
the conjugate addition reaction appeared unaffected by
the addition of s-BuOLi (Table 1, entries 8-9 vs 7) the
yield in the vinylation reaction dropped dramatically with
one equivalent of s-BuOLi (e5-33%, four experiments),
although it remained largely unaffected when up to 0.6
equiv of s-BuOLi was added (entries 19-21). Neverthe-
less, over three experiments considerable variation in
product yield was observed when 0.3 (98%, 97%, 60%
yield of 6a ) or 0.6 (92%, 77%, 38% yields of 6a ) equiv of
s-BuLi was present in the reaction mixture. In the
presence of 1.0 equiv of s-BuOLi, yields of 6a increased
with increasing amounts of CuCN‚2LiCl [equivalents (%
yield): 1.15 (8%), 1.34 (13-15%), 1.39 (42-45%)]. Simi-
larly, the dialkylcuprate prepared from 1 in glassware
washed with alcoholic KOH solutions and rinsed three
times with cold water (prior to oven drying) gave slightly
lower yields of 6a than KOH washed glassware rinsed
with 10% aqueous HCl, water, and oven dried (90 vs
99%).
TMSCl cuprate
entry compd Cu source RLi:Cua (equiv)
aged
%
yieldb,c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1
1
1
1
1
1
1
1
1
1
1
1
4
4
4
CuCN‚2LiCl
CuCN
CuCN‚2LiCl
CuCN
CuCN‚2LiCl
CuCN
CuCN‚2LiCl
CuCN
CuCN‚2LiCl
CuCN
CuCN‚2LiCl
CuCN
CuCN‚2LiCl
CuCN‚2LiCl
CuCN‚2LiCl
1:1
1:1
2:1
2:1
1:1
1:1
1:1
1:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
0
0
0
0
5
5
5
5
5
5
5
5
5
5
5
<5
<5
68-72
58-85
88
84
20 °C, 2 h 67
20 °C, 2 h 81
90
92
20 °C, 2 h 96d
20 °C, 2 h >98d
>98e-g
20 °C, 1 h 75e,h
20 °C, 2 h 71e,h
a
Deprotonation [(-)-sparteine, s-BuLi, THF, -78 °C, 30-60
min] and cuprate formation [-78 °C, 30-45 min] employed
uniform conditions. Deprotonation employed FMC B [clear, 1.7 M
RLi. 1.7 M total base] unless noted. MVK was added at -50 °C,
b
The thermal stabilities of pyrrolidine-derived R-(N-
carbamoyl)alkylcuprates were probed by formation of the
cuprate reagents with CuCN‚2LiCl or CuCN at -78 °C
followed by aging of the reagent at a given temperature
for 30 min2b and then cooling the solutions to -50 °C and
quenching the reagents with either 1-iodo-1-hexene or
methyl vinyl ketone (Table 2). The reagent (2RLi +
CuCN‚2LiCl) prepared via deprotonation of N-Boc-pyr-
rolidine (1) with FMC A (filtered) gave comparable yields
of 6a when aged at -40 (75-79%) or 0 °C (80-81%) for
30 min but began to show reduced yields when aged at
20 °C (66%) that diminished to 53% after 2 h at room
temperature. Utilization of s-BuLi containing no extrane-
ous base (FMC B) gave nearly quantitative yields of 6a
after the cuprate reagent was aged for 60 min at 25 °C,
although the yield dropped to 36% after 3 h. The
cyanocuprate reagent, RCuCNLi, prepared via filtered
FMC A gave very low yields of vinylation product when
the reaction was quenched at -50 °C [aging temperature
°C (% yield): -40 (17%), 0 (10%), 25 (08)] and modest
yields when the solution was allowed to warm to room
temperature [aging temperature °C (% yield): -40 (40%),
0 (38%), 25 (40)]. High yields of the vinylation product
6a (84%) could be obtained when the cuprate reagent was
prepared from high quality s-BuLi (FMC B), aged at -40
°C for 30 min, and allowing the reaction mixture to warm
to room temperature after addition of the vinyl iodide.
The N-Boc piperidine derived cuprate, R2CuLi‚LiCN,
gave 82% for unaged reagent and appeared to decompose
more rapidly than the pyrrolidinyl cuprate (cuprate aged
1 h, 45% 6b). The N-Boc-piperidine was deprotonated in
and the reaction stirred for 30 min and then quenched with
saturated NH4Cl at -50 °C. c Yields determined by NMR internal
standard using 1,3,5-tribromobenzene. Yields reported as a range
d
for multiple experiments. s-BuLi was prepared [1.5 M RLi, 1.5
M total base, CT B]. e Deprotonation was effected in Et2O [(-)-
sparteine, s-BuLi, THF, -78 °C, 4 h] and MVK was added at -50
°C [-50 °C to rt overnight]. f Clear colorless solution of s-BuLi
g
(1.1M active RLi, 1.2 M total base) was used. Yield based upon
h
material isolated by column chromatography. Clear colorless
solution of s-BuLi (1.3 M active RLi, 1.35 M total base) was used.
Et2O [1.25 equiv s-BuLi, -78 °C, TMEDA (2.3 equiv), 4
h] and was facilitated by excess TMEDA.
The thermal stabilities of 2-pyrrolidinyl- and 2-pip-
eridinylcuprates were also measured using methyl vinyl
ketone in the conjugate addition reaction to permit a
cleaner examination of the cyanocuprate reagent RCuCN-
Li. Although the RCuCNLi reagent prepared from lithi-
ated 1 and either CuCN or CuCN‚2LiCl gave very low
yields of conjugate adduct with methyl vinyl ketone in
the absence of TMSCl (Table 2, entries 1-2), the reagent
prepared from 2RLi and either Cu(I) salt gave compa-
rable yields of conjugate adduct in both cases (entries
3-4, 58-85%). The RCuCNLi reagent gave very good
yields in the presence of TMSCl (entries 5-6) even after
aging for 2 h at room temperature (entries 7-8), although
the reagent prepared from CuCN‚2LiCl gave a lower
yield (14% decrease) after aging. Nearly quantitative
yields of conjugate adduct were obtained with the pyr-
rolidinyl cuprate prepared from 2RLi and either Cu(I)
salt in the presence of TMSCl (entries 9-10) even after
the cuprate was aged for 2 h at room temperature (entries
11-12). In the presence of TMSCl, the piperidinylcuprate
gave a quantitative yield of adduct (entry 13) that
appeared to decrease significantly upon aging (entries
14-15) consistent with the earlier result using the vinyl
(21) Corey, E. J .; Naef, R.; Hannon, F. J . J . Am. Chem. Soc. 1986,
108, 7114.

2306 J . Org. Chem., Vol. 66, No. 7, 2001
Dieter et al.
Ta ble 3. Effects of Cu (I) Sa lts in th e Rea ction s of
Cu p r a tes Der ived fr om N-Boc-P r otected
(entries 9-14). It is noteworthy that use of CuI purified
by treatment with KI1b resulted in lower yields (entry
11 vs 12) consistent with earlier observations.22
N,N-Dim eth yla m in e a n d Cu X‚2LiCl w ith Meth yl Vin yl
Keton e or 5,5-Dim eth yl-2-cycloh exen on e (Eq 4)a
Given the effect of multiple variables upon the success
of these R-(N-carbamoyl)alkylcuprate reactions, we re-
examined the use of solid CuCN taking careful precau-
tions to form the cuprate reagent from 1. When the solid
CuCN was added to a solution of lithiated 1 and the
solution was stirred at -78 °C for 45 min a good yield of
the vinylation product was obtained and higher yields
could be achieved if the solution was briefly warmed to
room temperature (Table 4, entries 1-2). Similarly high
yields could be achieved with cyclohexenone (entry 3) and
methyl crotonate (entries 5-7) with the dialkylcuprate
reagent and also with the RCuCNLi reagent in the latter
reaction (entry 7). Again, a lower yield was obtained
when the solution was maintained between -78 and -50
°C (entry 4) indicative of incomplete cuprate formation.
The use of commercially available undistilled (-)-
sparteine appeared to have no deleterious effect upon the
methyl crotonate conjugate addition reaction. The use of
CuI proved to be significantly less effective than CuCN
(entry 8) in the conjugate addition reaction assisted by
TMSCl. This is in marked contrast to the use of CuI‚
2LiCl (Table 3, entries 3, 7, 11-12) with enones. The use
of TMSI in THF is problematic since the principal
reaction pathway involves TMSI cleavage of THF. Inter-
estingly, the diastereomeric ratios of the conjugate adduct
are reversed with TMSI compared with those obtained
with TMSCl.
Since our initial success in the reactions of R-(N-
carbamoyl)alkylcuprates with R,â-unsaturated carboxylic
acid derivatives and vinylation reactions with vinyl
iodides involved cuprate reagents prepared from CuCN‚
2LiCl, the effect of CuX‚2LiX combinations on the
conjugate addition of lithium di-n-butylcuprate to an R,â-
enoate was examined. Lithium di-n-butylcuprate was
prepared (-78 °C, 45 min, THF) from LiX (X ) Cl, Br, I,
CN) solubilized CuX (X ) Cl, Br, I, CN) and reacted with
â-phenethyl crotonate in the presence, or absence, of
TMSCl in order to probe the contributing effects of
TMSCl, Cu(I) salt, and lithium halide (eq 6, Table 5).
Reaction of lithium di-n-butylcuprate prepared from 2
n-BuLi and insoluble CuCN or CuCl in THF, in the
absence of TMSCl, gave â-phenethyl 3-methylheptanoate
(13) in low yield (Table 5, entries 1-2), while use of CuCl
or CuI resulted in E to Z isomerization of the starting
crotonate along with formation of the tertiary alcohol,
4-butyl-2-octen-4-ol (entries 2-3). Cuprates prepared
from insoluble CuCN, CuCl, or CuI gave good to excellent
yields of conjugate adduct in the presence of TMSCl
(entries 4-6), although CuI gave the lowest yield. The
yields of the conjugate adducts were slightly obscured by
the tendency of the phenethyl esters to hydrolyze under
the workup conditions. The cuprate prepared from CuCN‚
2LiCl in the absence of TMSCl gave very low yields of
product (entry 7). All combinations of CuX (X ) CN, Cl,
Br, I) and LiX (X ) Cl, Br, I) gave excellent yields of the
conjugate addition product in the presence of TMSCl
(entries 8-13 and 15-17). Reagents prepared from CuX‚
2LiCN (X ) CN, I) gave only modest yields of product in
the presence of TMSCl (entry 14). Very similar results
were obtained for the mixed cuprate [n-BuCuCH2SiMe3]-
entry
CuX‚2LiCl
equivb
product
% yieldc
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Cl
Br
I
CN
Cl
Br
I
CN
Cl
Br
I
1.0
1.0
1.0
1.0
0.5
0.5
0.5
0.5
1.0
1.0
1.0
1.0
1.0
1.0
9
9
9
9
9
9
9
9
10
10
10
10
10
-
56-58
67-70
57-60
63-74
90d
52-60
80-84
61d
82-95
76
68-79e
91-94f
70-90
81-82g
I
CN
CN
a
b
Reaction conditions are shown in eq 4. Stoichiometry of 0.5
equiv corresponds to 2RLi + CuX‚2LiCl and 1.0 to RLi +
CuX‚2LiCl. c Yields were determined via 1H NMR spectroscopy
using tetrachloroethane as the internal standard. A range repre-
d
sents determinations from several absorption peaks. Isolated
yield. e Purified CuI was employed. f Nonpurified CuI was utilized.
g
Product was isolated in 85% yield. Reaction was performed with
the cuprate derived from N-Boc-pyrrolidine (1) and dimedone.
iodide. Lithiation of N-Boc-protected piperidine (4), how-
ever, is more difficult to achieve than with N-Boc
protected pyrrolidine (1) and shows greater variation
from experiment to experiment.
In an effort to obtain crystals of an R-(N-carbamoyl)-
alkylcuprate reagent, pentane was added to Et2O or THF
solutions of the pyrrolidinylcuprate (i.e., R2CuLi‚LiCN)
resulting in precipitation of solids at -50 °C. In one case,
the solution was allowed to warm to room temperature
and left standing overnight to afford a suspension of
brown solids. Addition of (E)-1-iodo-1-hexene to this
suspension afforded 6a in 80% yield, indicating consider-
able thermal stability of the cuprate reagent in this THF/
pentane solvent system.
Reaction of the cuprate derived from Boc-protected
N,N-dimethylamine (8) with methyl vinyl ketone or
dimedone was used to examine the effect of the Cu(I) salt
[i.e., CuX‚2LiCl (X ) Cl, Br, I, CN)] solubilized by LiCl
(eq 4, Table 3). Although the yields in this study are
uniformly lower for methyl vinyl ketone than the opti-
mized yields shown in Table 2 and show greater varia-
tion, the results indicate that the anionic component of
the Cu(I) salt (i.e., Cl, Br, I, CN) has no large effect upon
the yield of conjugate adduct for either the RCuCNLi
reagent (Table 3, entries 1-4) or the reagent prepared
from 2 RLi + CuX‚2LiCl (entries 5-8). One experiment
using optimized reaction conditions with CuCl gave a
very high yield of conjugate adduct (entry 5). Conjugate
addition reactions to dimedone with the RCuCNLi re-
agent gave similar results but in uniformly higher yields
(22) Bertz, S. H.; Chopra, A.; Eriksson, M.; Ogle, C. A.; Seagle, P.
Chem. Eur. J . 1999, 5, 2680.

R-(N-Carbamoyl)alkylcuprate Chemistry
J . Org. Chem., Vol. 66, No. 7, 2001 2307
Ta ble 4. Rea ction of Cu p r a tes Der ived fr om 1 Em p loyin g Solid Cu CN w ith (E)-1-Iod o-1-h exen e, 2-Cycloh exen on e, or
Meth yl Cr oton a te To Affor d , Resp ectively, 6a , 3, a n d 11 (Eq 5)
cuprate prepna
(T (°C), (min))
entry
X
CuX (equiv)
(E+)b
product
% yieldc
1
2
3
4
5
6
7
8
9
CN
CN
CN
CN
CN
CN
CN
I
0.5
0.5
0.5
0.5
0.5
0.5
1.0
0.5
0.5
0.5
-78 (45)
vinyl iodide
vinyl iodide
enoned
6a
6a
3
11
11
11
11
11
11
11
83
98
-78 (30), then rt (15)
-78 (30), then rt (15)
-78 (30), then -50 (15)
-78 (30), then rt (10)
-78 (30), then rt (10)
-78 (30), then rt (10)
-78 (30), then rt (10)
-78 (30), then rt (10)
-78 (30), then rt (10)
95e
enoated
59
enoated
>95
>95f
>95
<25
no product
35
enoated
enoated
enoateg
I
I
enoateg,h
enoated
10
a
Solid CuX was added to the R-(N-carbamoyl)alkyllithium reagent in THF and the mixture stirred at the indicated temperature for
the indicated time. b The electrophile was added between -65 and -50 °C and slowly warmed to room temperature. c Yields were determined
by NMR using 1,3,5-tribromobenzene as an internal standard unless otherwise noted. TMSCl (5 equiv) was employed. e Based upon
d
isolated product. f (-)-Sparteine was used directly from the commerical bottle without prior distillation. gTMSI (5 equiv) was employed.
h
The reaction was conducted in Et2O.
Ta ble 5. Con ju ga te Ad d ition of Lith iu m
n -Bu tyl(tr im eth ylsilylm eth yl)cu p r a te or Lith iu m
CuX‚2LiX or insoluble CuX react with R,â-enoates in
comparable yields in the presence of TMSCl.
Di-n -bu tylcu p r a te P r ep a r ed fr om Va r iou s Cop p er Sa lts
Solu bilized w ith Lith iu m Sa lts to â-P h en eth yl Cr oton a te
in THF a
Discu ssion
Organocopper reactions are particularly susceptible to
a variety of experimental factors. This combination of
factors include the particular reaction type,1 copper(I) salt
(e.g., CuI, CuCN, CuBr‚SMe2)4 and its purity,22,23 cuprate
reagent1,2b and its thermal stability,2 additives (e.g.,
Lewis acids, TMSCl, HMPA/TMSCl, etc.),1,3 solvent,22,24
and the instability of these organometallic reagents to
moisture and oxygen. Sometimes the effects are coun-
terintuitive. Small amounts of added water are reported
to accelerate conjugate addition reactions,25 and labora-
tory-purified CuI gives lower yields of conjugate adducts
than commercially available CuI of high purity.22 The
presence of Cu(II) impurities in the Cu(I) salts can
accelerate cuprate decomposition.23 In addition to issues
of cuprate preparation and cuprate thermal and chemical
stability, the actual cuprate transformation will be
subject to experimental variables. The rate of cuprate
conjugate addition reactions will be dependent upon the
nature of the transferable ligand, the presence and
perhaps nature of the lithium salts normally present in
cuprate solutions (e.g., LiI, LiBr, LiCl, or LiCN),22 the
solvent,22,24 and the presence of additives.3,22 The reac-
tions of R-(N-carbamoyl)alkylcuprates with a given R,â-
enone often displayed variable yields15d which were
attributed to the quality of the s-BuLi, the Cu(I) salt (i.e.,
CuCN or CuCN‚2LiCl), and/or to reagent thermal stabil-
ity. Although the current study does not clearly reveal
% yieldd
entry
metal salt(s)b
CuCN
CuCl
CuI
CuCN
CuCl
CuI
CuCN‚2LiCl
CuCN‚ LiCl
CuCN‚2LiCl
CuCN‚3LiCl
CuCN‚4LiCl
CuCN‚2LiBr
CuCN‚2LiI
CuCN‚2LiCN
CuCl‚2LiCl
CuBr‚2LiCl
CuI‚2LiCl
additivec
12
13
14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
83
37
33
10
9
7
TMSCl
TMSCl
TMSCl
93
91
77
3
6
4
10
8
90
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
90
96
91
91
95
96
30
94
88
94
83
10
5
8
7
0.3
3
21
2
5
3
49
CuI + TMSMLi
CuI + TMSMLi
CuCl + TMSMLi
CuCl + TMSMLi
14
85
20
85
TMSCl
80
5
a
Upon addition of phenethyl crotonate and TMSCl, the reaction
mixture was allowed to warm to room temperature over 2 h. The
b
(23) Hutchinson, D. K.; Fuchs, P. L. J . Am. Chem. Soc. 1987, 109,
4930.
metal salt was dissolved in THF, and the solution was added to
n-BuLi/THF at -78 °C. c The additive was mixed with phenethyl
(24) (a) House, H. O.; Wilkins, J . M. J . Org. Chem. 1978, 43, 2443.
(b) Hallnemo, G.; Ullenius, C. Tetrahedron 1991, 47, 4739. (c) Chris-
tenson, B.; Hallnemo, G.; Ullenius, C. Tetrahedron 1983, 39, 1621. (d)
Kingsbury, C. L.; Smith, R. A. J . J . Org. Chem. 1997, 62, 7637. (e)
Kingsbury, C. L.; Smith, R. A. J . J . Org. Chem. 1997, 62, 7637.
(25) Corey, E. J .; Hannon, F. J .; Boaz, N. W. Tetrahedron 1989, 45,
545.
d
crotonate and added to the cuprate solution at -78 °C. Yields
are based upoon isolated products purified by silica gel chroma-
tography unless otherwise noted.
Li (entries 18-21). Consistent with the R-(N-carbamoyl)-
alkylcuprates, dialkylcuprate reagents prepared from

2308 J . Org. Chem., Vol. 66, No. 7, 2001
Dieter et al.
the origin(s) of this variability, it does reveal a combina-
tion of effects that can cumulatively result in low yields.
While the presence of nonalkyllithium bases in the
s-BuLi does not have fatal consequences for the subse-
quent R-(N-carbamoyl)alkylcuprate vinylation and con-
jugate addition reactions (Table 1), their presence gen-
erally diminishes the overall yield of the cuprate
transformations even though equivalent amounts of
carbamate and s-BuLi are utilized as determined by
titration for the latter reagent. The Watson-Eastham
titration1b,20 gives an accurate determination of alkyl-
lithium concentration in direct comparison with deter-
minations made by the Gilman double-titration1b,19 pro-
cedure. The formation of less reactive mixed cuprates
[e.g., RCuHLi, RCu(OsBu)Li2a, R ) R-(N-carbamoyl)-
alkyl], diminished reactivity of the cuprates in the
presence of extraneous bases, and/or accelerated decom-
position of the organolithium or cuprate reagents could
account for the diminished yields. Vinylation of pyrro-
lidinylcuprates in the presence s-BuOLi ranges from
nearly quantitative (0.6 equiv of s-BuOLi) to low yielding
(e5-33% with 1 equiv), suggesting that less reactive
mixed alkoxycuprate reagents come into play when
stoichiometric quantities of extraneous bases are present.
This is also consistent with the observations that product
yields can be increased in the presence of 1 equiv of
s-BuOLi by increasing the amount of CuCN‚2LiCl. The
effect of extraneous bases upon cuprate reactivity is
expected to vary with the particular type of cuprate
reaction. Lithium ion-carbonyl oxygen coordination3g,24a-c,26
is particularly important in cuprate conjugate addition
reactions and is reflected in diminution of reaction rates
and product yields in the presence of polar solvents (THF,
DME),24a-c additives (e.g., HMPA, TMEDA, DMAP),3f,3h,27
or heteroatom functionality in the substrate26c that can
function as Lewis bases with lithium ions. Conjugate
addition reactions fail when crown ethers are added to
cuprate solutions unless additional amounts of LiX salts
are added.26a The presence of extraneous bases may
sequester lithium ions in aggregates involving the basic
species alone or as bases complexed to the lithium ions
in a cuprate aggregate in a manner analogous to polar
solvents and additives. The crude yield data in Table 1
are consistent with a greater sensitivity of the conjugate
addition reaction as compared to the vinylation reaction
to the presence of extraneous bases, although the mixed
alkoxy(alkyl)cuprate did give a far better yield in the
conjugate addition reaction (i.e., 60% vs e5-33%). It
should be noted that both the presence of TMSCl3g and
the use of CuCN‚LiCl in the conjugate addition reactions
tabulated in Table 1 compensate for the competition26b
between enone oxygen and base for lithium ions. This in
part could account for the greater variability and capri-
ciousness obtained with cuprates prepared from THF
insoluble CuCN and poorer quality s-BuLi (eq 1).15b,d
Although not dramatic, the quality of the s-BuLi does
appear to be related to the thermal stability of these
cuprate solutions. Cuprates prepared from s-BuLi con-
taining no extraneous bases show little to no decomposi-
tion after 2 h at room temperature (Table 2, entries 5-8
and 9-12) for both the RCuCNLi and R2CuLi‚LiCN
reagents. The rate of decomposition for the cuprate
solutions appears to be faster when s-BuLi containing
extraneous bases is employed. It is unclear whether this
declining reactivity of the cuprate solutions reflects
cuprate or R-(N-carbamoyl)alkyllithium decomposition.
Our studies4,28 and the observations of others29 have
pointed to the thermal instability of R-(N-carbamoyl)-
alkyllithium reagents. The known autocatalytic effect of
s-BuOLi upon s-BuLi decomposition18 suggests that
extraneous s-BuOLi could accentuate the thermal insta-
bility of R-(N-carbamoyl)alkyllithium reagents. The ex-
tent of R-(N-carbamoyl)alkyllithium decomposition would
be dependent upon the temperature and rate at which
cuprate formation occurs (vide infra). The influence of
solvent upon the thermal stability of these cuprate
solutions may reflect the solvent influence on a cuprate-
alkyllithium equilibrium.30 The long-term stability of
heterogeneous cuprate (12 h, room temperature) contain-
ing mixtures obtained by addition of pentane to THF
solutions of R-(N-carbamoyl)alkylcuprates may arise from
the particulate nature of the copper species or from the
absence of equilibrium amounts of free alkyllithium
components. In this scenario, polar solvents such as THF
which favor equilibrium of the cuprate with free alkyl-
lithium should be more problematic than Et2O, Et2O/
THF, or THF/pentane combinations. Cuprate/alkyllith-
ium equilibrium was not observed by NMR measurements
in Et2O and in THF solutions containing LiX salts where
the dynamic exchange of ligands was too fast to observe
on the NMR time scale even at low temperatures.30
Nevertheless, minor amounts of byproducts arising from
1,2-nucleophilic additions to enones16b,29 or transmetala-
tion of vinyl halides were not readily apparent in the
R-(N-carbamoyl)alkyllithium reactions indicating either
little or no free R-(N-carbamoyl)alkyllithium reagent in
these solutions or noncompetitive rates for these path-
ways due to concentration effects.
The conjugate addition reactivity of R-(N-carbamoyl)-
alkylcuprates prepared with high quality s-BuLi was
probed with a very reactive methyl vinyl ketone substrate
(Table 2). In the absence of TMSCl, the RCuCNLi reagent
failed to add to methyl vinyl ketone while the inherently
more reactive R2CuLi‚LiCN reagent gave modest to good
yields of the 1,4-adduct. Both reagents gave good to
excellent yields of the 1,4-adduct in the presence of
TMSCl. The inherent reactivity of the cuprates was
independent of the mode of preparation since similar
yields were obtained when either solid CuCN or CuCN‚
2LiCl was employed. Although enone oxygen-lithium ion
complexation plays a crucial role in cuprate conjugate
addition reactions, the presence of soluble lithium chlo-
ride from CuCN‚2LiCl does not result in higher yields.26
This is consistent with the observation that cuprate
reactivity is greater in conjugate addition reactions in
(26) For the importance of lithium ion carbonyl oxygen coordination
in cuprate conjugate addition reactions see: (a) Ouannes, C.; Dressaire,
G.; Langlois, Y. Tetrahedron Lett. 1977, 815. (b) Krauss, S. R.; Smith,
S. G. J . Am. Chem. Soc. 1981, 103, 141. (c) Hallnemo, G.; Ullenius, C.
Tetrahedron Lett. 1986, 27, 395. (d) Vellekoop, A. S.; Smith, R. A. J .
J . Am. Chem. Soc. 1994, 116, 2902. For a review see: (e) Smith, R. A.
J .; Vellekoop, A. S. Advances in Detailed Reaction Mechanisms; Coxon,
J . M., Ed.; J AI Press: Grenwich, CT, 1994; Vol. 3, p 79.
(28) (a) Dieter, R. K.; Li, S. J . Org. Chem. 1997, 62, 7726. (b)
Chlorotrimethylsilane quenching of N-Boc R-lithiopyrrolidine after
aging at a given temperature for 30 min. gave N-Boc 2-trimethylsi-
lylpyrrolidine [-78 °C (94%), -50 °C (71%), 0 °C (1-2%)]. Li, S. M.S.
Thesis, Clemson University, Clemson, SC, 1996.
(29) Beak, P.; Kerrick, S. T.; Wu, S.; Chu, J . J . Am. Chem. Soc. 1994,
116, 3231.
(27) (a) House, H. O.; Respess, W. L.; Whitesides, G. M. J . Org.
Chem. 1966, 31, 3128. (b) House, H. O.; Lee, T. V. J . Org. Chem. 1978,
43, 4369.
(30) Lipshutz, B. H.; Kozlowski, J . A.; Breneman, C. M. J . Am.
Chem. Soc. 1985, 107, 3197.

R-(N-Carbamoyl)alkylcuprate Chemistry
J . Org. Chem., Vol. 66, No. 7, 2001 2309
the absence of LiX.26b Kinetic isotope effects3g probed by
NMR indicate that the large TMSCl induced rate ac-
celeration arises from oxygen silylation of a cuprate-
enone d-π* complex and in THF the silylation contri-
bution overwhelms the more modest lithium ion con-
tribution. The TMSCl rate acceleration is not observed
in Et2O and the absence of oxygen silylation in a rate
determining step is confirmed by the kinetic isotope
studies.3f,g
carbamoyl)alkylcuprate reactions coupled with use of
CuCN‚2LiCl.
During the course of our efforts to unravel the apparent
beneficial use of CuCN‚2LiCl for R-(N-carbamoyl)alkyl-
cuprate preparation, we examined the reaction of n-Bu2-
CuLi prepared from a variety of CuX‚nLiX with cinnamyl
esters to see if the phenomenon was general (Table 5).
The reactivity of n-Bu2CuLi with enoates in THF appears
to be largely governed by the presence of TMSCl,3g and
previous observations4 about reactivity differences for
cuprates prepared from CuI, CuBr, or CuCl in the
absence of TMSCl raise interesting questions. All com-
binations of CuX and LiX, with the exception of LiCN,
give comparable yields of the conjugate addition product.
The difficulty of preparing pure LiCN and the failure of
lithium cyanide generated in situ by addition of butyl-
lithium to trimethylsilylcyanide to solubilize CuCN at low
temperatures complicates any interpretation of the re-
sults obtained with LiCN. This study suggests that
variation in chemical yield as a function of CuX reflects
the rate and extent of cuprate formation since all soluble
combinations of CuX‚LiX give comparable results. Nev-
ertheless, the nature of the halide counterion plays a
significant role in the extent of asymmetric conjugate
additions of chiral amidocuprates31 and in allylic substi-
tution reactions.32 The variation in the conjugate addition
yields obtained with CuX‚2LiCl may reflect subtle influ-
ences of the halide counterion that are clearly not well
understood.3f Although the trimethylsilylmethyl ligand
is an effective nontransferable ligand,33 it does not appear
to greatly accelerate the conjugate addition reaction.34
Although generally not essential for R-(N-carbamoyl)-
alkylcuprate reactivity (Tables 2 & 4), the use of THF
soluble CuX‚2LiCl appears to remove some of the capri-
ciousness of these reactions. Utilization of THF soluble
CuX‚2LiCl permits rapid cuprate formation at -78 °C
while use of insoluble CuCN generally requires warming
of the THF solution to -50 to -40 °C for some extended
period of time or to room temperature for shorter periods
of time in order to ensure complete cuprate formation.
Excellent yields were obtained with insoluble CuCN
when the cold bath (-50 to -40 °C) was removed and
the reaction flask was allowed to stand in ambient air
(25 °C) for 15 min before cooling back down to -78 °C
during the cuprate formation stage. Slow cuprate forma-
tion and cuprate formation at higher temperatures will
increase the likelihood of decomposition of the less
thermally stable R-lithio carbamates. The general vari-
ability in product yields with time and temperature
formats for cuprate preparation, substrate addition, and
subsequent reaction profile (e.g., stir at -40 to -50 °C
for a period of time before warming to room temperature
vs gradual warming of the reaction mixture from -40 to
room temperature) seem consistent with both the rate
of cuprate formation and an R-(N-carbamoyl)alkylcu-
prate/R-(N-carbamoyl)alkyllithium equilibrium30 that can
result in decomposition of the thermally less stable
alkyllithium reagent. Reaction rates will vary consider-
ably for each reaction type (i.e., 1,4-addition, vinyl
substitution, etc.) and for each substrate-cuprate pair for
a given transformation. Slower R-(N-carbamoyl)alkylcu-
prate transformations will benefit from long reaction
times at moderately low temperatures (e.g., -50 to -30
°C) rather than rapid warming to room temperature. The
use of CuCN‚2LiCl introduces two opposing effects. By
solubilizing the CuCN, the presence of lithium chloride
significantly increases the rate of cuprate formation at
-78 °C even as the added LiCl facilitates the dynamic
exchange of the ligands on copper presumably through
the intermediacy of alkyllithium species.30 There is,
however, a dramatic difference between solid CuI (25%
yield of 11) and CuI‚2LiCl (57-94% yields of 9 and 10)
that is not understood.
The successful use of insoluble CuCN in R-(N-carbam-
oyl)alkylcuprate vinylation and enoate conjugate addition
reactions (Table 4) suggests that the excess LiCl is not
playing a significant role in either reaction. It does not
appear that lithium chloride is functioning as an external
source of lithium ions and accelerating the conjugate
addition reaction by Lewis acid complexation with the
carbonyl oxygen. Similarly, the success of the vinylation
reaction with solid CuCN suggests that soluble chloride
ions are not facilitating the reductive elimination step
as has been suggested for the action of TMSCl3f and the
behavior of CuX (X ) Cl, Br, I) in allylic substitution32
reactions.
Finally, the difficulty of executing R-(N-carbamoyl)-
alkylcuprate transformations raised questions about
their inherent reactivity. In a previous study, the R-(N-
carbamoyl)alkyl and methyl ligands of a mixed cuprate
reagent, RCuMeLi (R derived from 8), were transferred
with nearly equal facility (i.e., 51% for R and 41% for
Me) upon reaction with 2-cyclohexenone.15d Additionally,
the higher yields of conjugate adducts obtained with
N-Boc pyrrolidinylcuprates as compared to N-Boc-N-
methyl(aminomethyl)cuprates with ester substrates sug-
gests that the R-(N-carbamoyl)alkyl ligand is comparable
in reactivity to the corresponding alkyl ligands.8 Never-
theless, R-(N-carbamoyl)alkylcuprate fail to transfer the
R-(N-carbamoyl)alkyl ligand in some cases (e.g., R,â-
We have also observed in our studies with R,â-enones
that reactions using freshly distilled TMEDA give gener-
ally higher yields of the conjugate adducts than distilled
TMEDA that has been stored for some time. These
variations were significantly less dramatic with the use
of (-)-sparteine or when TMEDA was used in conjunction
with CuCN‚2LiCl.15d Although the use of CuCN‚2LiCl
also appeared to minimize the effect of diamine purity
[i.e., TMEDA or (-)-sparteine], the capriciousness of any
potential diamine effect made it difficult to disentangle
it from other factors. While we cannot document the
deleterious effects of impurities in the diamines employed
to assist in carbamate deprotonation, we would recom-
mend purification by distillation prior to use in R-(N-
(31) Dieter, R. K.; Lagu, B.; Deo, N.; Dieter, J . W. Tetrahedron Lett.
1990, 31, 4105.
(32) Ba¨ckvall, J .-E.; Persson, E. S. M.; Bombrun, A. J . Org. Chem.
1994, 59, 4126 and ref 13 cited therein.
(33) Bertz, S. H.; Eriksson, M.; Miao, G.; Snyder, J . P. J . Am. Chem.
Soc. 1996, 118, 10906.
(34) Lutz, C.; J ones, P.; Knochel, P. Synthesis 1999, 312.

2310 J . Org. Chem., Vol. 66, No. 7, 2001
Dieter et al.
to be 1.51 M (total base content 1.54 M) by Gilman double
titration. A larger scale reaction based on 1.9 mol Li and 0.8
mol s-BuCl gave 370 mL of clear colorless product (1.7 M).
unsaturated lactones) where simple lithium dialkyl-
cuprates succeed.
The solution may be stored (frozen) at -20 °C for several
months with little degradation or loss of activity. If desired,
pentane (10% by volume) may be added to lower the freezing
point of the solution.
Su m m a r y
In summary, R-(N-carbamoyl)alkylcuprate reactions
afford good yields of products in conjugate addition,
vinylation, and acylation reactions even with poor quality
s-BuLi and with either relatively insoluble CuCN or THF-
soluble CuCN‚2LiCl. Excellent product yields can be
achieved by utilizing high quality s-BuLi and THF
soluble CuCN‚2LiCl which ensures rapid cuprate forma-
tion at -78 °C. Insoluble CuCN also affords excellent
product yields if extended time periods are used for
cuprate formation at -78 °C (g45 min) or the reaction
flask is allowed to stand at ambient air temperature for
short periods of time (10-15 min) before cooling back to
-78 °C. Lower quality s-BuLi (i.e., containing LiH and
lithium alkoxide impurities) appears to increase the rate
of cuprate and/or R-(N-cabamoyl)alkyllithium decomposi-
tion and high quality s-BuLi can be easily prepared with
FMC’s Lithium Powder-Stabi-Li-ze. The R-(N-carbam-
oyl)alkylcuprates examined display good thermal stabili-
ties (i.e., up to 71-98% yields after aging the cuprates
for 2 h at room temperature) when prepared under
optimal conditions in marked contrast to the R-(N-
cabamoyl)alkyllithium reagents. Difficulties in executing
the transformations appear to revolve around the ther-
mal instabilities of the lithium reagents which can be
minimized with the protocols described above. With these
caveats in mind, the reactions of R-(N-carbamoyl)alkyl-
cuprates are reliable and should prove to be useful
reagents for synthetic endeavors.
P r oced u r e for Gilm a n Dou ble Titr a tion . A 1 mL portion
of the organometallic reagent was added (cautiously) to water
(about 10 mL) and a few drops of phenolphthalein solution
were added as an indicator. This purple (due to base) solution
was titrated with standardized 0.1 M hydrochloric acid
[standardized using a NaOH solution, itself standardized using
dry potassium hydrogen phthalate as the primary standard]
to the persistent (after 30 s of swirling) colorless endpoint
permitting calculation of the total base content of the orga-
nometallic solution.
A second 1 mL portion of the organometallic reagent was
added to about 1 mL of dry 1,2-dibromoethane (stored over
activated sieves) under an inert, dry, atmosphere and stirred
for 15 min at room temperature to ensure complete reaction.
Addition of water (10 mL) by syringe followed by stirring for
5 min before opening to the atmosphere afforded a solution
which was titrated with the standardized 0.1 M HCl (using
phenolphthalein indicator as above) to give the non organo-
metallic base content.
N-Boc-2-[(E)-1-h exen yl]p yr r olid in e (6a ). (-)-Sparteine
(0.23 mL, 1.0 mmol) was added to a solution of N-Boc-
pyrrolidine (1) (171 mg, 1.0 mmol) in THF (3 mL) at room
temperature and mixed until homogeneous. After the mixture
was cooled to -78 °C, s-BuLi (1.1 mmol) was added in one
portion and the reaction mixture stirred at -78 °C for 30 min.
Meanwhile, a solution of CuCN‚2LiCl in THF was prepared
from CuCN (46 mg, 0.5 mmol), LiCl (50-60 mg, 1.15-1.4
mmol), and THF (3 mL) under an inert atmosphere. This
copper solution was added by syringe to the reaction mixture
and the solution stirred at -78 °C for 30 min. After warming
to -50 °C, neat 1-iodohex-1-ene (106 mg, 0.5 mmol) was added,
and the mixture stirred at -50 °C for 30 min before quenching
with saturated aqueous NH4Cl at -50 °C. The reaction
mixture was diluted with Et2O and filtered through a pad of
flash silica and washed with saturated aqueous NH4Cl (×2).
The aqueous washings were extracted with Et2O and the
combined organic extracts dried over anhydrous MgSO4,
filtered, and concentrated under reduced pressure to yield the
crude product 6a as a pale yellow oil (typical yield 80-90%)
as determined by isolation (column chromotography) or by 1H
NMR measurements using an internal standard: IR (neat)
1702 (vs), 970 (w), 885 (w) cm-1; 1H NMR δ 0.90 (t, 3 H), 1.03-
1.41 (m, 4 H), 1.43 (s, 9 H), 1.59-1.79 (m, 1 H), 1.79-1.85 (m,
2H), 1.85-2.11 (m, 3H), 3.37 (br s, 2 H), 4.20 [4.41] (br s, 1 H)
[rotamer], 5.34 (br s, 1 H), 5.43 (br s, 1 H); 13C NMR δ 13.7,
21.9, 22.8 (br), 28.2, 31.3, 31.6, 32.2 (br), 45.9 (br), 58.3 (br),
78.7, 130.0 (2 C), 154.7; mass spectrum m/z (intensity) EI 197
(25, M+ - C4H8), 196 (15, M+ - C4H9), 180 (17, M+ - C4H9O),
152 (6, M+ - C4H9CO2). Anal. Calcd for C15H27NO2: C, 71.15;
H, 10.67. Found: C, 70.88; H, 10.60.
N-Boc-2-[(E)-1-h exen yl]p ip er id in e (6b). Freshly distilled
TMEDA (0.35 mL, 2.3 mmol) was added to a solution of Boc-
protected pyrrolidine (4) (185 mg, 1.0 mmol) in Et2O (3 mL)
at room temperature and the solution mixed until homoge-
neous. After the mixture was cooled to -78 °C, s-BuLi (1.2
mmol) was added in one portion and the reaction mixture
stirred at -78 °C for 4 h. A solution of CuCN‚2LiCl in THF
was prepared from CuCN (46 mg, 0.5 mmol), LiCl (50-60 mg,
1.15-1.40 mmol), and THF (3 mL) under an inert atmosphere.
This copper solution was added by syringe to the reaction
mixture and the solution stirred at -78 to -65 °C over 1 h.
Neat 1-iodo-1-hexene (106 mg, 0.5 mmol) was added and the
reaction mixture stirred overnight at -65 °C to room temper-
ature before quenching with saturated aqueous NH4Cl. Work-
up procedure involved dilution with Et2O and filtration
through a pad of silica before washing with saturated aqueous
NH4Cl (×2). Aqueous washings were extracted with further
Exp er im en ta l Section
All glassware was flame dried under vacuum or an atmo-
sphere of argon prior to use. Sparteine, TMEDA, and vinyl
iodides were distilled under reduced (Kugelruhr) pressure prior
to use. THF and Et2O were distilled from Na/benzophenone
under an atmosphere of N2 immediately prior to use. LiCl was
flame dried under vacuum immediately prior to use.
P r ep a r a tion of s-Bu tyllith iu m . The preparation must be
carried out under an atmosphere of argon as lithium metal
reacts with nitrogen. Finely powdered lithium metal (7.8 g,
1.12 mmol, FMC’s Stabi-Li-ze) was added cautiously to de-
gassed dry cyclohexane (200 mL) in a large argon-purged
three- or four-neck round-bottomed flask fitted with a me-
chanical stirrer, a dry ice condenser, a pressure-equalizing
addition funnel, and a thermometer to directly measure the
reaction temperature. A petroleum ether/dry ice cold bath was
prepared to control reaction temperature if necessary. The
suspension was stirred vigorously for 5 min and warmed to
40 °C using a heat gun. The pressure-equalizing addition
funnel was charged with dry s-BuCl (43 mL, 0.40 mol).
Approximately 7% of the s-BuCl was added with vigorous
stirring and the reaction mixture warmed (if necessary) in
order to attain a temperature of 52 ( 3 °C. The remainder of
the s-BuCl was then added dropwise with constant supervision
over 45-60 min at such a rate that the reaction temperature
was maintained at 52 ( 3 °C. It is essential to keep the
reaction temperature below 60 °C [NOTE: Beware of unpre-
dictable initiation period.] After complete addition of s-BuCl
the reaction mixture was allowed to cool to room temperature
with stirring. The cool solution was transferred by cannular
into a Schlenk apparatus to allow filtration through a pad of
vacuum/flame dried Celite while maintaining an argon atmo-
sphere. A further extraction of the residual solids was carried
out with additional dry cyclohexane. The product was obtained
as a clear, colorless solution and the s-BuLi content was found

R-(N-Carbamoyl)alkylcuprate Chemistry
J . Org. Chem., Vol. 66, No. 7, 2001 2311
Et2O and the combined organic portions dried (anhydrous
MgSO4) and concentrated under reduced pressure to yield the
crude product as a pale yellow oil. Chemical yield (82%) of the
desired product (6b) was determined by 1H NMR spectroscopy
using an internal standard: IR (neat) 1702 (vs), 1679 (shoul-
crotonate (51 µL, 0.5 mmol) and TMSCl (0.32 mL, 2.5 mmol)
were added as a neat mixture via syringe, and the reaction
mixture was stirred overnight at -78 °C to room temperature
before quenching with saturated aqueous NH4Cl. Workup
procedure involved dilution with Et2O and filtration through
a pad of silica before washing with saturated aqueous NH4Cl
(×2). Aqueous washings were extracted with further Et2O and
the combined organic portions dried (anhydrous MgSO4) and
concentrated under reduced pressure to yield the crude product
as a colorless to pale yellow oil in nearly quantitative yields.
Spectral data for the corresponding ethyl ester can be found
in ref 8c.
der), 978 (w), 876 (w) cm-1 1H NMR δ 0.73-0.93 (m, 3 H),
;
1.13-1.80 (m, 9 H), 1.38 (s, 9 H), 1.89-2.07 (m, 2 H), 2.75 (t,
J ) 12.8 Hz, 1 H), 3.28 (br s, 1 H), 3.84 (d, J ) 12.0 Hz, 1 H),
4.66 (br s, 1H), 5.27-5.45 (m, 2H); 13C NMR δ 13.9, 19.5, 22.1,
25.6, 28.5, 29.4, 31.5, 32.1, 39.6, 51.9, 79.1, 128.1, 131.9, 155.4;
mass spectrum, m/z (relative intensity) EI 211 (20, M+ - C4H8),
194 (17, M+ - C4H9O), 166 (21, M+ - C4H9CO2). Anal. Calcd
for C16H29NO2: C, 71.91; H, 10.86; N, 5.24. Found: C, 72.07;
H, 11.03; N, 5.09.
Ack n ow led gm en t. This work was generously sup-
ported by the Petroleum Research Fund (ACS-PRF
33626-ACI). Support of the NSF Chemical Instrumenta-
tion Program for purchase of a J EOL 500 MHz NMR
instrument is gratefully acknowledged (CHE-9700278).
Gifts of s-BuLi and Lithium Powder-Stabi-Li-z were
generously provided by FMC Lithium Division. Our
appreciation goes to Troy Dover and Sharon Smith for
sharing the experimental art and helpful hints for the
preparation of high-quality s-BuLi. We thank Dr. Ram
R. Sharma for the acylation data.
Meth yl 1-[(1,1-Dim eth yleth oxy)ca r bon yl]-â-m eth yl-2-
p yr r olid in ep r op a n oa te (11). (-)-Sparteine (0.15 mL, 0.65
mmol) was added to a solution of N-Boc-pyrrolidine (1) (86 mg,
0.5 mmol) in THF (3 mL) at room temperature and mixed until
homogeneous. After the mixture was cooled to -78 °C, s-BuLi
(1.1 mmol) was added in one portion and the reaction mixture
stirred at -78 °C for 60 min. With the reaction mixture under
an argon atmosphere, solid CuCN (46 mg, 0.5 mmol) was
quickly added in one portion. The reaction mixture was stirred
at -78 °C for 30 min. The cold bath was then removed, and
the flask was allowed to stand at ambient air temperature for
15 min after which the flask was immersed in the -78 °C cold
bath. After 15 min at ambient temperature, clear nearly
colorless to pale yellow solutions were obtained. Methyl
J O0056702