Reactions of Et3ZnLi with ketones: Electronic and steric effects

Article abstract of DOI:10.1021/jo000630j

Toluene solutions of composition Et₃ZnLi react rapidly with aldehydes and ketones to form addition products. Et₃ZnNa and Et₃ZnK solutions react readily with the same substrates although metalation, as well as addition, is significant with substrates having α-hydrogens. The Et₃ZnM solutions react with 2-cyclohexenone to give mainly the 1,4-addition product. Relative rates of addition of Et₃ZnLi to substituted acetophenones give a Hammett ρ of 2.78. Addition of Et₃ZnLi to acetophenone is slowed significantly by α and ortho methyl substituents; relative rates of addition to acetophenone, o-methylacetophenone, and tert-butyl phenyl ketone are 1.00, 0.012, and 0.003.

Full text of DOI:10.1021/jo000630j

7750
J . Org. Chem. 2000, 65, 7750-7756
Rea ction s of Et3Zn Li w ith Keton es: Electr on ic a n d Ster ic Effects1
Curtis A. Musser and Herman G. Richey, J r.*
Department of Chemistry, The Pennsylvania State University, University Park, Pennsylvania l6802
hgr@chem.psu.edu
Received April 25, 2000
Toluene solutions of composition Et3ZnLi react rapidly with aldehydes and ketones to form addition
products. Et3ZnNa and Et3ZnK solutions react readily with the same substrates although metalation,
as well as addition, is significant with substrates having R-hydrogens. The Et3ZnM solutions react
with 2-cyclohexenone to give mainly the 1,4-addition product. Relative rates of addition of Et3ZnLi
to substituted acetophenones give a Hammett F of 2.78. Addition of Et3ZnLi to acetophenone is
slowed significantly by R and ortho methyl substituents; relative rates of addition to acetophenone,
o-methylacetophenone, and tert-butyl phenyl ketone are 1.00, 0.012, and 0.003.
Nearly 50 years ago Wittig and Lange reported that a
solution of composition Ph3ZnLi metalated fluorene,
added to benzophenone, and added slowly to PhCHd
CHC(dO)Ph to give 1,2- and 1,4-addition products; Ph2Zn
did not significantly undergo these reactions.2 Solutions
of composition R3ZnLi have since been found to give other
reactions not important for R2Zn or RZnX, including
metal-halogen exchange with 1,1-dibromoalkenes,3 aryl
iodides,4 and 1,1-dibromocyclopropanes.5 Most R2Zn and
RZnX compounds react sluggishly with aldehydes and
not at all with ketones.6 Addition to R2Zn compounds of
Bu4NX salts somewhat increases the rate of reaction with
benzaldehyde (and decreases the portion of reduction
product),7 and addition of t-BuOK permits addition to
CO;8 we now believe that these solutions contain orga-
nozincate species.9 Organozincates (particularly of com-
position R3ZnLi or R3ZnMgX) react with R,â-unsaturated
ketones, predominantly by 1,4-addition.10 We set out to
study reactions of organozincates with aldehydes and
ketones.
Zn and EtLi,11 Et3ZnNa and Et3ZnK were prepared from
Et2Zn and Na or K using procedures described in l858,12
and Et2Zn(t-BuOK), Et2Zn(t-BuOLi), and Et2Zn(Bu4NBr)
were prepared from Et2Zn and a salt. These formulas are
compositions of the solutions and imply neither molecular
weights nor structures of the dominant species.
Rea ction s of Or ga n ozin ca te Solu tion s w ith Ald e-
h yd es a n d Keton es. We found that several organozin-
cate solutions react with aldehydes and ketones (eqs 1
and 2). Table 1 summarizes the results of adding carbonyl
compounds to toluene solutions of organozincates at
ambient temperature. Et3ZnLi was prepared from Et2-
Et2Zn, as expected, did not react significantly with any
of the substrates in the indicated reaction times.13 Benzo-
phenone and Et2Zn in refluxing toluene were reported
to react but only to form reduction product 6d ; no reaction
was observed with Me2Zn.14 By contrast, we found rapid
reactions with several of the organozincate solutions, and
a reduction product (6) never was significant. Et3ZnLi
gave good yields of addition product with all substrates,
the 1,4-addition product 2 predominating from 1. Reac-
tions of Et3ZnNa and Et3ZnK with 1 also gave good yields
(1) Much of this work is taken from Musser, C. A. Ph.D. Disserta-
tion, The Pennsylvania State University, l994.
(2) Wittig, G.; Meyer, F. J .; Lange, G. Ann. 1951, 571, 167.
(3) Harada, T.; Katsuhira, T.; Hara, D.; Kotani, Y.; Maejima, K.;
Kaji, R.; Oku, A. J . Org. Chem. 1993, 58, 4897.
(4) Kondo, Y.; Takazawa, N.; Yamazaki, C.; Sakamaoto, T. J . Org.
Chem. 1994, 59, 4717. Also see Gruter, G.-J . M.; Akkerman, O. S.;
Bickelhaupt, F.; Smeets, W. J . J .; Spek, A. L. Recl. Trav. Chim. Pays-
Bas 1993, 112, 425.
(5) Harada, T.; Katsuhira, T.; Hattori, K.; Oku, A. J . Org. Chem.
1993, 58, 2958.
(6) Lee, V. J . In Comprehensive Organic Synthesis; Trost, B. M., Ed.;
Pergamon: Oxford, 1991; Vol. 4 (Semmelhack, M. F., Ed.), Section
1.2.2.1.4. Perlmutter, P. Conjugate Addition Reactions in Organic
Synthesis; Pergamon: Oxford, l992; Chapters 1-3.
(7) Chastrette, M.; Amouroux, R. Tetrahedron Lett. 1970, 5165.
(8) Rathke, M. W.; Yu, H. J . Org. Chem. 1972, 37, 1732.
(9) Fabicon, R. M.; Parvez, M.; Richey, H. G., J r. J . Am. Chem. Soc.
1991, 113, 1412.
(10) Knochel, P. In Comprehensive Organic Synthesis; Trost, B. M.,
Ed.; Pergamon: Oxford, l991; Vol. 1 (Schreiber, S. L., Ed.), Section
1.7.3.2. Erdik, E. Organozinc Reagents in Organic Synthesis; CRC
Press: Boca Raton, l996; Chapter 4.
(11) Schlenk, R.; Holtz, J . Ber. Dtsch. Chem. Ges. 1917, 50, 262. A
Et2Zn/EtLi ratio of 1.05 was used in our work reported in this paper.
It is unlikely therefore that tetraethylzincate was present [Uchiyama,
M.; Kameda, M.; Mishima, O.; Yokoyama, N.; Koike, M.; Kondo, Y.;
Sakamoto, T. J . Am. Chem. Soc. 1998, 120, 4934].
(12) Wanklyn, J . A. Proc. R. Soc. London 1858, 341. Wanklyn, J . A.
J ustus Liebigs Ann. Chem. 1858, 108, 67.
(13) Benzaldehyde and Pr2Zn in diethyl ether (not the toluene of
this work) at 25 °C were reported to react slowly to give approximately
equal amounts of an addition product and 6a .7
(14) Coates, G. E.; Ridley, D. J . Chem. Soc. (A) 1966, 1064.
10.1021/jo000630j CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/25/2000

Reactions of Et3ZnLi with Ketones
J . Org. Chem., Vol. 65, No. 23, 2000 7751
Ta ble 1. P r od u cts fr om Rea ction s of Or ga n ozin c Rea gen ts w ith Ald eh yd es a n d Keton esa
Et2Znb
%
Et3ZnLic
%
Et3ZnNac
%
Et3ZnKc
%
Et2Zn(t-BuOLi)d
Et2Zn(t-BuOK)d
Et2Zn(Bu4NBr)e
%
reactant
product
%
%
1
1
2
3
97
-
-
98
-
-
97
2
-
95
2
-
-
98
-
99
3
-
87
322
-
-
97
-
45
2
1
59
34
6
36
-
9
20
40
-
90
2
-
61
38
-
96
-
1
-
4a
4b
4a
5a
6a
4b
5b
6b
7b
4c
5c
4d
5d
4e
5e
7e
98
-
-
3
97
2
87
12
3
85
2
94
-
-
98
-
4
90
2
60
-
-
-
-
3
9
33
97
-
20
100
-
-
4c
4d
4e
99
-
27
71
-
97
38
37
23
62
36
-
99
38
31
28
99
-
101
-
99
-
-
98
-
101
49
-
98
-
-
55
-
98
-
94
-
40
50
-
a
d
All reactions were at ambient temperature. b Reaction time was 24 h. c Reaction time was 10 min. Reaction time was 15 min. e Reaction
time was 1 h.
Ta ble 2. Rela tive Ra tes of Ad d ition to
Su bstitu ted Acetop h en on es a n d σ Va lu es of Su bstitu en ts
of 2.15 Starting material was recovered from reactions of
Et3ZnNa and Et3ZnK with 4b, 4c, and 4e, however, and
the reactions with 4b and 4e each gave an additional
product that was indicated by GC and MS data to result
from aldol dimerization of the substrate. Some reactions
of Et3ZnNa and Et3ZnK with these R-hydrogen-contain-
ing substrates were quenched with D2O; MS analysis
showed the recovered reactants to be mainly monodeu-
terated. Deuterium incorporation and dimer formation
indicate significant metalation by Et3ZnNa and Et3ZnK.
Reactions of solutions prepared from Et2Zn and t-BuOK
with 4b, 4c, and 4e gave only recovered reactant and
dimer; D2O quenching indicated that the recovered
reactants were mainly monodeuterated. The solutions
prepared from Et2Zn and t-BuOLi gave significant ad-
dition product only with benzaldehyde, but dimeric
products and monodeuteration of recovered reactant
indicated that metalation of 4c and 4e took place. Much
lower yields of products were observed with solutions
prepared from Et2Zn and Bu4NBr, and a D2O quenching
experiment showed no significant excess deuterium in
recovered 4c.16
substituent
σ19
relative rate
3,5-(CF3)2
4-CN
4-CF3
3-CF3
3-C(dO)CH3
3-Cl
0.86a
0.66
0.54
0.43
0.38
0.37
0.34
0.23
0.12
0.06
0
34.8
62.5
13.8
13.8
9.65b
6.17
5.01
2.61
1.03
0.82
(1.00)
0.38
0.093
3-F
4-Cl
3-OCH3
4-F
none
4-Et
-0.15
-0.27
4-OCH3
Twice the 3-CF3 value. b One-half of the observed relative rate.
a
formation of addition products 9 except with 8c and 8i,
each of which formed a small amount (<5%) of a second
product.17 Nitro-substituted acetophenones gave complex
Electr on ic a n d Ster ic Effects on Ad d ition s of
Et3Zn Li to Acetop h en on es. On the basis of the results
in Table 1, Et3ZnLi was chosen for a study of structural
effects on addition rate. Reactions with benzaldehydes
and acetophenones at ambient temperature were com-
plete within 30 s, about the shortest time in which
reagents could be mixed and a sample taken, and
therefore were too fast to study by quenching and
analyzing aliquots. The half-time of a reaction of Et3ZnLi
(0.5 M) even with a dialkyl ketone, 4e (0.02 M), was much
less than 1 min.
We decided therefore to determine relative rates using
a competition procedure: the product composition when
Et3ZnLi is added to an excess of a pair of ketones could
give the relative reactivities of those ketones. Addition
of a toluene solution of Et3ZnLi to toluene solutions of
the acetophenones 8 in Table 2 (eq 3) led to >98%
product mixtures and hence were not used. Reactions
usually were quenched after 1 min, although reaction
(15) A minor product (ca. 5-10%) in the reaction of Et3ZnNa and 1
probably was a 1,3-diethylcyclohexanol isomer [Richey, H. G., J r.;
Farkas, J ., J r. Organometallics 1990, 9, 1778].
(16) Addition of a diethyl ether solution of Pr2Zn(Bu4NBr) to
benzaldehyde has been reported to be four times faster than addition
of Pr2Zn.7
(17) The minor product, more volatile than the starting material
and apparently different for the two chloroacetophenones, was not
acetophenone or m- or p-ethylacetophenone.

7752 J . Org. Chem., Vol. 65, No. 23, 2000
Musser and Richey
times of 10 min did not change yields or product
compositions (for the 8a -e pair, reaction times even of
several hours gave the same product compositions). A
study of yields (particularly with 8a , 8e, and 1) indicated
that only one-third of the ethyl groups of a Et3ZnLi
solution were incorporated into addition products.
Initial experiments with pairs of acetophenones sug-
gested that substituent effects are small, but the results
were not reproducible. We suspected that the ketones
seemed deceptively similar in reactivity because addition
was more rapid than the mixing that ideally would
ensure that Et3ZnLi always encounters a constant ratio
of the two ketones. Reproducible results finally were
obtained by using a lower temperature and very slow
addition of Et3ZnLi to the solution of ketones. The extent
to which the temperature could be lowered was limited
by decreasing solubility of some acetophenones; -41 °C
finally was selected, although m-cyano-, p-(N,N-dimethyl-
amino)-, and p-acetylacetophenone had to be eliminated
from the study because their solubilities were too low at
that temperature. A syringe pump was used to add the
Et3ZnLi solution; the needle through which addition was
made was placed below the surface of the stirred solution
of ketones so that addition was continual rather than in
drops. To ensure that the mixing rate no longer affected
the results, a series of competition reactions were run
using 8a and 8e. Since 8e was one of the most reactive
substrates, effects of incomplete mixing should be notice-
able with this reactant pair. The addition rate was
decreased until the product ratio did not change; an even
slower rate was used for the competition experiments.
F igu r e 1. Hammett plot of relative rates of addition at -41
°C of a toluene Et3ZnLi solution to acetophenones. The F
(omitting the 3,5-(CF3)2 compound) is 2.78.
Steric effects were probed by studying reactions of Et3-
ZnLi with a group of acetophenones (10a -f) substituted
with R and ortho methyl groups (eq 4). Addition product
Most ketones were compared directly with acetophe-
none (8a ). Since the reactions used 10 mol of each
acetophenone for each mole of Et3ZnLi, the ratio of
ketones did not change greatly over the course of a
reaction. When one of the pair of ketones was signifi-
cantly more reactive than the other, however, an adjust-
ment was made for depletion.18 The rates of 8f and 8m ,
very much greater than that of 8a , were measured
against 8e. To compare 8k and the much less reactive
8a , a 6:14:1 (8k :8a :Et3ZnLi) reactant composition was
used. The relative reactivity values could be checked by
comparisons other than those reported in Table 2. For
example, the data in the Table indicate that 8e is 2.24
times more reactive than 8i; an experiment using these
ketones gave a ratio of 2.40. Similar comparisons (from
Table 2, from direct comparison) were made for 8e and
8c (5.29, 6.66), 8e and 8k (1.43, 1.26), and 8d and 8g
(4.09, 3.93).
A plot of log(relative rate) versus Hammett σ values19
is shown in Figure 1. The relative rate of the m-acetyl
compound (8k ) was halved because of its two equivalent
reactive sites. We could not find a value for 3,5-(CF3)2
substitution and so doubled the value for a 3-CF3 sub-
stituent. The largest deviation in the plot was for the 3,5-
(CF3)2 compound (8m ); since a reliable σ value was not
available, this point was not used in determining the
slope, F ) 2.78 (r2 ) 0.973), of the line that best fits the
data. A plot using σ- values19 was much poorer (F ) 2.13,
r2 ) 0.932).
11 and 5-10% of recovered reactant were found in
reactions of 10b, 10c, and 10e. Only reactant was
isolated from reactions of 10f, but it had been metalated
since MS analysis, after quenching with D2O, showed
most of it to be monodeuterated. Reaction of 10d gave
both reduction (27%) and addition (70%) products.20 The
relative rates of additions to 10a -e, benzophenone, and
2-cyclohexenone were determined by competition experi-
ments. Values for 10b, 10c, 10e, and benzophenone were
determined by competition with acetophenone (10a ). The
same value for the relative rate of 10e was obtained in a
competition experiment with 8g (whose rate relative to
acetophenone was known). The relative rates of 10d and
2-cyclohexenone, the least and most reactive of this group
of substrates, were determined by competition with 10e
and 8e, respectively.
The relative rates for methyl, ethyl, isopropyl, tert-
butyl compounds 10a -d were 1.0, 0.28, 0.099, and 0.003;
the relative rate of o-methyl compound 10e was 0.012.
(18) For ketones A and B, A considerably more reactive than B, the
relative rate ratio (kA/kB) was determined as ([B]0/[P]t)ln([A]0/[A]t)
where P is the product formed from B. The adjustments were small,
the effect on rate ratio never exceeding 7%.
(20) Et3ZnNa and 10d also gave reduction (19%) and addition (79%)
(19) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
products.

Reactions of Et3ZnLi with Ketones
Ta ble 3. Effect of Ad d ed TMEDA on NMR Sp ectr a of Et3Zn Lia
1H NMR, δ
CH3CH2M
J . Org. Chem., Vol. 65, No. 23, 2000 7753
13C NMR, δ
CH3CH2M
solution
CH3CH2M
CH3N
CH2N
CH3CH2M
CH2N
58.7
CH3N
46.0
EtLi
-0.90
0.11
-0.23
1.30
1.10
1.33
Et2Zn
6.7
2.0
10.3
12.1
Et3ZnLi
TMEDA
2.29
1.46
1.49
1.76
1.85
1.99
2.09
2.10
1.60
1.62
1.83
1.88
2.05
2.02
Et3ZnLi/0.22 TMEDA
Et3ZnLi/1.0 TMEDA
Et3ZnLi/1.5 TMEDA
Et3ZnLi/2.0 TMEDA
Et2Zn/1.0 TMEDA
EtLi/1.1 TMEDA
Et3ZnLib
-0.11
0.23
1.46
1.82
1.90
1.87
1.73
1.90
1.20
4.4
3.8
13.8
14.5
56.5
57.0
45.5
45.9
0.17
0.28
0.21
-0.73
-0.27
TMEDAb
2.36
2.36
2.20
2.20
1.24c
ET3ZnLi/1.0 TMEDAb
Et2Zn/1.0 t-BuOLi
-0.27
-0.08
1.20
1.10
5.0
6.9
15.5
10.5
58.5
46.7
66.9d
35.7e
The solvent is benzene-d6 except where noted. The solvent is THF-d8. 1H NMR absorptions are reported relative to internal
a
b
-OCHDCD2- (δ 3.60) and 13C NMR absorptions relative to internal -CD2CD2O- (δ 26.50). c CH3CO absorption. CH3CO absorption.
d
e CH3CO absorption.
The relative rates for benzophenone and 2-cyclohexenone
(also relative to acetophenone) were 1.2 and 14, respec-
tively.
having R-hydrogens, but the latter does not undergo even
this reaction.
What organozinc species are responsible for the reac-
tions? NMR studies22 of benzene solutions of EtLi plus
Et2Zn indicate formation of a species having a 1:1 ratio
of the reactants, and molecular weight determinations23,24
indicate the composition to be (EtLi)2(Et2Zn)2. NMR
observations indicate that mixed species form even in
coordinating solvents: 2:1 and 3:1 (MeLi:Me2Zn) in
diethyl ether25 and 1:1 and 2:1 in THF.26 X-ray crystal
structural studies have found R3Zn- Li+ structures in
which Li+ is coordinated with TMEDA27,28 or 1,3,5-tri-
methylhexahydro-1,3,5-triazine,28,29 and there is some
NMR evidence for R3Zn- ions in solution.30 The NMR
results (Table 3) with Et3ZnLi-TMEDA solutions indi-
cate that TMEDA is coordinated to a metal (probably to
Li, as in structures27,28 mentioned above). Et3ZnNa and
Et3ZnK in benzene also are dimeric,23 but X-ray studies31
show Np3ZnNa, Np3ZnK, and (Me3SiCH2)3ZnK solids to
have essentially independent R3Zn- and Na+ or K+ ions.32
A species having a 1:1 ratio of the reactants is the
principal component in solutions prepared from equimo-
lar amounts of t-BuOK and an R2Zn compound;33,34 X-ray
analysis33 shows a Et2ZnO-t-Bu- K+ dimer whose Zn
atoms share two bridging t-BuO groups. We know of no
observation concerning t-BuOLi-R2Zn compositions ex-
Mod ifica tion of th e Or ga n ozin ca te Rea gen t. Ab-
sorptions from some of the NMR spectra of solutions
prepared by adding TMEDA to benzene-d6 solutions of
Et3ZnLi are given in Table 3. Until the TMEDA:Et3ZnLi
ratio reached 1, the ethyl absorptions shifted to higher
frequency, and the TMEDA absorptions were character-
istic of TMEDA coordinated to a metal. After the ratio
exceeded 1, the positions of the TMEDA absorptions
approximated weighted averages of those of free and
coordinated TMEDA, and the ethyl absorptions under-
went only small changes. When TMEDA was added to
1
Et3ZnLi in THF-d8, however, the H NMR absorptions
of Et3ZnLi and TMEDA were unchanged. These observa-
tions indicate that TMEDA is coordinated to a metal in
benzene-d6 solutions but not significantly in THF solu-
tions. Addition of TMEDA conceivably might form EtLi
+ Et2Zn(TMEDA) or EtLi(TMEDA) + Et2Zn, but com-
parisons with NMR spectra of EtLi-TMEDA and Et2-
Zn-TMEDA solutions show that the ethyl absorptions
of the Et3ZnLi-TMEDA solutions differ significantly
from the weighted averages expected if EtLi(TMEDA) or
Et2Zn(TMEDA) formation was appreciable. TMEDA there-
fore interacts with the organozincate without converting
large amounts of it to Et2Zn and EtLi; moreover, the
solutions containing TMEDA still gave principally 1,4-
addition with 1 (2:3 > 10), although alkyllithium com-
pounds give principally 1,2-addition.21
(22) Toppet, S.; Slinckx, G.; Smets, G. J . Organomet. Chem. 1967,
9, 205. Also see ref 21.
(23) Hein, F.; Schramm, H. Z. Phys. Chem. 1930, 151, 234.
(24) Aggregation is more extensive when [EtLi] > [Et2Zn].23
(25) Seitz, L. M.; Brown, T. L. J . Am. Chem. Soc. 1966, 88, 4140.
(26) Seitz, L. M.; Little, B. F. J . Organomet. Chem. 1969, 18, 227.
(27) Westerhausen, M.; Wieneke, M.; Ponikwar, W.; No¨th, H.;
Schwarz, W. Organometallics 1998, 17, 1438. Thiele, K.; Go¨rls, H.;
Seidel, W. Z. Anorg. Allg. Chem. 1998, 624, 555.
Discu ssion
In contrast to Et2Zn, Et3ZnLi reacts rapidly with
aldehydes and ketones to give good yields of addition
products. Et3ZnNa and Et3ZnK also react rapidly, but
metalation of substrates having R-hydrogens can be
significant. Et3ZnNa and Et3ZnK give mainly 1,4-addition
to R,â-unsaturated ketone 1, as already well-known for
Et3ZnLi. Et2Zn(t-BuOK) adds rapidly to the substrates
without R-hydrogens, but metalation is the principal
reaction with substrates having R-hydrogens. Et2Zn(t-
BuOLi) and Et2Zn(Bu4NBr) form addition product only
with benzaldehyde; the former metalates substrates
(28) Westerhausen, M.; Rademacher, B.; Schwarz, W. Z. Anorg. Allg.
Chem. 1993, 619, 675.
(29) Westerhausen, M.; Rademacher, B.; Schwarz, W.; Henkel, S.
Z. Naturforsch. B 1994, 49, 199.
(30) X-ray structures of the type (Li+)2(R4Zn)2- also are known:
Weiss, E.; Wolfrum, R. Chem. Ber. 1968, 101, 35 (l968). Fro¨hlich, H.-
O.; Kosan, B.; Mu¨ller, B.; Hiller, W. J . Organomet. Chem. 1992, 441,
177. Fro¨hlich, H.-O.; Kosan, B.; Undeutsch, B.; Go¨rls, H. J . Organomet.
Chem. 1994, 472, 1.
(31) Purdy, A. P.; George, C. F. Organometallics 1992, 11, 1955. Also
see Putzer, M. A.; Neumu¨ller, B.; Dehnicke, K. Z. Anorg. Allg. Chem.
1997, 623, 539.
(32) An X-ray structure of the type (K+)2(R4Zn)2- also is known:
Weiss, E.; Plass H. J . Organomet. Chem. 1968, 14, 21.
(33) Fabicon, R. M.; Parvez, M.; Richey, H. G., J r. J . Am. Chem.
Soc. 1991, 113, 1412.
(21) A recent example of 1,2-addition of an alkyllithium compound:
Rijnberg, E.; J astrzebski, T. B. H.; Boersma, J .; Kooijman, H.; Veldman,
N.; Spek, A. L.; van Koten, G. Organometallics 1997, 16, 2239.
(34) Fabicon, R. M. Ph.D. Dissertation, The Pennsylvania State
University, l991.

7754 J . Org. Chem., Vol. 65, No. 23, 2000
Musser and Richey
cept that in Table 3; the minimal changes in position of
NMR absorptions of the Et groups when t-BuOLi is added
to a Et2Zn solution suggest that formation of a mixed
species is insignificant. NMR studies of R2Zn-Bu4NBr
solutions hint that mixed species form, but are inconclu-
sive because changes of absorption positions are small.34
These observations suggest that organozincate species
are major components in the Et3ZnLi, Et3ZnNa, Et3ZnK,
and Et2Zn(t-BuOK) solutions. Organozincates probably
are responsible for the additions, a conclusion supported
by the observations that only one-third of the ethyl
groups of an Et3ZnLi solution add to acetophenones or
to 2-cyclohexenone. The additions cannot be due to Et2-
Zn, itself unreactive with the substrates; at least in
reactions with 1, neither are they due to EtLi, since 3
would be the principal product.21,35 The Et2Zn(t-BuOLi)
and Et2Zn(Bu4NBr) solutions are far less reactive, adding
significantly only to benzaldehyde; any organozincates
in these solutions must be less reactive or (as seems
probable with the former) present in lesser amounts.
Mech a n ism s of Rea ction s of Et3Zn Li Solu tion s
w ith Acetop h en on es. The F of 2.78 found for reactions
of Et3ZnLi solutions and acetophenones is much larger
than F values reported for most similar reactions of other
polar organometallic reagents: 0.4-1.4 for seven addi-
tions36-38 of methyl, butyl, and phenyl Grignard reagents
to acetophenones or benzophenones and 0.2-0.3 for four
additions38-41 of methyllithium, butyllithium, and phen-
yllithium to benzaldehydes or benzophenones. The only
comparable value, F ) 3.0, is for additions of t-BuMgCl
to benzophenones,37 reactions in which electron transfer
from the organometallic reactant to the ketone is gener-
ally thought to be rate-determining.42,43
sional angle48 between the aryl ring and the carbonyl
group, they diminish aryl-carbonyl conjugation in the
acetophenone and in some possible intermediates. Could
the effects of the methyl groups be due principally to
making electron transfer more difficult by increasing the
torsional angle? E1/2 in an aprotic solvent is reported to
be 0.16 V more negative for 10d than for acetophenone,49
a difference that must reflect the combined effects by the
added methyl groups of increased torsional angle and
weak electron release. The differences between 10d and
acetophenone in E1/2 and in addition rate are of equiva-
lent magnitude, consistent with an effect on electron-
transfer rate being significant.50 In the absence of more
extensive studies of E1/2 and an indication of error limits,
however, this is only suggestive. Note that E1/2 for
benzophenone is reported49 to be 0.16 V more positive
than for acetophenone, but we found additions to these
ketones to have almost identical rates.
Because of the large F, electron transfer seems probable
in the reactions of Et3ZnLi and acetophenones. Electron
transfer can be susceptible to steric hindrance. In fact,
Holm42 proposed that electron transfer is rate-determin-
ing in some additions of organomagnesium compounds
to ketones in which this step would be too endothermic
to be significant unless the R-carbon of the Grignard
reagent was partially bonded to the carbonyl carbon; the
rate of electron transfer is significant only at those
moments when the R-carbon and the carbonyl carbon are
in close proximity, a proximity that is less frequent with
increasing steric hindrance at the carbonyl carbon.
Exp er im en ta l Section
The rate-decreasing effects of R and ortho methyl
groups evident in additions to 10a -f must be principally
steric. A direct electronic effect of the only weakly
electron-releasing methyl groups cannot account for the
large rate differences (note the relatively small effect of
the para ethyl group of 8d ). In fact, for the 10a -d series,
a plot of log(relative rate) against the Taft Es parameter44
gives a reasonable correlation (δ ) 1.50, r2 ) 0.970).46,47
However, R and ortho methyl groups could have an
indirect electronic consequencesby increasing the tor-
1H NMR absorptions are reported relative to internal C5D5H
(δ 7.15) in benzene-d6 solutions, internal CHCl3 (δ 7.26) in
CDCl3 solutions, and internal -CHDCD2OCD2CD2- (δ 2.60)
in THF-d8 solutions; notations used are s, singlet; d, doublet;
t, triplet; q, quartet; m, more complex multiplet; c, complex
overlapping absorptions; br, broad. 13C NMR absorptions are
reported relative to internal C6D6 (δ 128.00) in benzene-d6
solutions, internal CDCl3 (δ 77.50) in CDCl3 solutions, and
internal -CD2CD2O- (δ 26.50) in THF-d8 solutions. AAS
analyses used an air-acetylene flame; concentrations were
obtained from a calibration curve prepared from standard
solutions.51 Procedures involving organometallic compounds
were performed under a nitrogen atmosphere using a vacuum
line or a glovebox. Elemental analyses were performed by
Galbraith Laboratories, Inc., Knoxville, TN. Benzene, toluene,
and TMEDA were distilled from CaH2; diethyl ether was
distilled from sodium benzophenone ketyl; t-BuOLi and t-
BuOK were sublimed.
(35) It is conceivable that the organozincate species dissociates to
form EtLi, which at that moment differs from and reacts in a different
manner than (EtLi)6, the dominant23 species at equilibrium.
(36) Lewis, R. N.; Wright, J . R. J . Am. Chem. Soc. 1952, 74, 1257.
Anteunis, M.; Van Schoote, J . Bull. Soc. Chim. Belg. 1963, 72, 776.
Yamataka, H.; Matsuyama, T.; Hanafusa, T. J . Am. Chem. Soc. 1989,
111, 4912.
(37) Holm, T.; Crossland, I. Acta Chem. Scand. 1971, 25, 59.
(38) Yamataka, H.; Miyano, N.; Hanfusa, T. J . Org. Chem. 1991,
56, 2573.
P r od u cts. Some samples needed as GC standards for
reaction products were obtained commercially. Several were
synthesized by literature procedures: 2,15 3,15 5e,52 and 1,3-
(39) Yamataka, H.; Fujimura, N.; Kawafuji, Y.; Hanafusa, T. J . Am.
Chem. Soc. 1987, 109, 4305.
(40) Yamataka, H.; Kawafuji, Y.; Nagareda, K.; Miyano, N.; Ha-
nafusa, T. J . Org. Chem. 1989, 54, 4706.
(41) For reactions of benzophenones with methyllithium and eth-
yllithium, we have found somewhat larger F values (∼1): Maclin, K.
M.; Musser, C. A. Unpublished work.
(42) For a recent review of mechanisms of additions of Grignard
reagents, see Holm, T. In Grignard Reagents: New Developments;
Richey, H. G., J r., Ed.; Wiley: Chichester, 2000; Chapter 1.
(43) A relatively large F, 1.96, also is reported for addition of “Me2-
CuLi” to benzophenones.39
(48) Barker, S. D.; Mirarchi, D.; Norris, R. K.; Phillips, L.; Ritchie,
G. L. D. J . Mol. Struct. 1981, 75, 265, and references therein. Also see
Andrieu, C. G.; Lumbroso, H. Z. Naturforsch A 1985, 40, 1329. Boykin,
D. W.; Balakrishnan, P.; Baumstark, A. L. Magn. Reson. Chem. 1987,
25, 248.
(49) Saturated calomel electrode in DMF with Bu4NI supporting
electrolyte: Seguin, J .-P.; Beaupe`re, D.; Bauer, P.; Uzan, R. Bull. Soc.
Chim. Fr. 1974, 167.
(50) For polarographic data relating to these ketones in aqueous
solutions and an introduction to problems in comparing measurements
for different ketones, see Elving, P. J .; Leone, J . T. J . Am. Chem. Soc.
1958, 80, 1021.
(51) Skoog, D. A.; West, D. M. Analytical Chemistry, 3rd ed;
Saunders: Philadelphia, l980; Chapter 17.
(44) The values are taken from ref 45.
(45) Meyer, A. Y. J . Chem. Soc., Perkin Trans. 2 1986, 1567.
(46) The plot against Vn values45 is less satisfactory (δ ) -55.6, r2
) 0.911).
(47) The relative rates closely parallel those observed for reactions
of the same substrates with PhMgBr: Van Schoote, J .; Anteunis, M.
Bull. Soc. Chim. Belg. 1963, 72, 784.
(52) Abbot, E. M.; Bellamy, A. J . J . Chem. Soc., Perkin Trans. 2
1978, 254.

Reactions of Et3ZnLi with Ketones
J . Org. Chem., Vol. 65, No. 23, 2000 7755
diethylcyclohexanol.15 All others were obtained by routine
additions of EtMgBr to an aldehyde or ketone. The following
compounds apparently are new (NMR spectra in CDCl3, 1H
at 300 MHz, 13C NMR at 75 MHz).
Small pieces of Na were added piece by piece to stirred Et2Zn
(4 equiv). There was an immediate exothermic reaction. The
mixture was stirred for 12 h, resulting in a fine black powder
suspended in a clear, colorless solution. Excess Et2Zn was
removed at reduced pressure leaving an oil and a black
powder. Toluene was added to dissolve the oil, and the mixture
was filtered through a fine fritted funnel. An aliquot of the
filtered solution was hydrolyzed and titrated for base to
determine the Et3ZnNa concentration. Et3ZnK was prepared
and analyzed in a similar fashion except that much longer
stirring times were required. AAS analysis of a typical solution
found the Zn:K ratio to be 1.09. The Et3ZnK could be crystal-
lized from concentrated solutions and recrystallized from
benzene or toluene. On tw o occa sion s th e cr ysta llin e
m a ter ia l d ecom p osed violen tly. Toluene solutions of Et3-
ZnNa and Et3ZnK were initially clear and colorless, but
precipitates formed within several days, presumably due to
metalation of the solvent.
P r ep a r a tion s of Et2Zn (t-Bu OLi), Et2Zn (t-Bu OK), a n d
Et2Zn (Bu 4NBr ) Solu tion s. The salt (1 equiv) was added to
a toluene solution of Et2Zn. Et2Zn(t-BuOLi) solutions were 0.10
M, and Et2Zn(t-BuOK) and Et2Zn(Bu4NBr) solutions were 0.20
M.
NMR Sa m p les. NMR samples were prepared by removing
solvent at reduced pressure from 1.0 mL of one of the above
solutions (Et3ZnLi was a viscous oil, Et3ZnNa and Et3ZnK were
waxy solids). Then the deuterated solvent (1.0 mL) and any
additives were added. The amount of added TMEDA was
determined by NMR integrations.
Rea ction s of Or ga n ozin ca tes w ith Ald eh yd es a n d
Keton es (1 a n d 4a -e). A toluene solution of the carbonyl
compound (1.0 mL, 0.15 M) and dodecane (ca. 0.03 M, known
accurately in each case) as an internal standard was added
dropwise to a stirred toluene solution of Et2Zn (1.0 mL, 0.20
M) in a small conical vial that was maintained at 0 °C. The
solutions were allowed to warm to ambient temperature and
were stirred for 24 h. A piece of ice was added followed by a
saturated aqueous NH4Cl solution (1.5 mL) and diethyl ether
(1.5 mL). The mixture was stirred briefly, the aqueous layer
was removed from the bottom of the vial with a pipet, and
MgSO4 was added to dry the remaining liquid. The mixture
was filtered through a piece of tissue (Kimwipe) that was
1
2-(3-Acetylp h en yl)-2-bu ta n ol: bp 108 °C (0.04 Torr); H
NMR 8.03 (dd, 1H, J ) 1.95, 1.82 Hz), 7.81 (ddd, 1H, J ) 7.69,
1.82, 1.65 Hz), 7.63 (ddd, 1H, J ) 7.84, 1.95, 1.82 Hz), 7.42
(dd, 1H, J ) 7.84, 7.69 Hz), 2.60 (s, 3H), 2.04 (br s, 1H), 1.85
(q, 1H, J ) 7.52 Hz), 1.84 (q, 1H, J ) 7.36 Hz), 1.56 (s, 3H),
0.78 (t, 3H, J ) 7.42 Hz); 13C NMR 198.48, 148.41, 136.92,
129.84, 128.31, 126.62, 124.66, 74.74, 36.64, 29.70, 26.69, 8.16.
Anal. Calcd for C12H16O2: C, 74.97; H, 8.39. Found: C, 74.85;
H, 8.50.
2-(4-Acetylp h en yl)-2-bu ta n ol: bp 110 °C (0.035 Torr); 1H
NMR 7.92 (d, J ) 8.62 Hz, 2H), 7.52 (dt, J ) 8.62, 1.94 Hz,
2H), 2.59 (s, 3H), 1.91 (br s, 1H), 1.86 (q, J ) 7.49 Hz, 1H),
1.84 (q, J ) 7.28 Hz, 1H), 1.56 (s, 3H), 0.79 (t, J ) 7.44 Hz,
3H); 13C NMR 197.91, 153.19, 135.45, 128.25, 125.16, 74.94,
36.58, 29.75, 26.59, 8.11. Anal. Calcd for C12H16O2: C, 74.97;
H, 8.39. Found: C, 74.87; H, 8.48.
2-(4-Cya n op h en yl)-2-bu ta n ol: mp 55-56 °C (recrystal-
lized from toluene after flash chromatrography53); 1H NMR
7.63 (ddd, J ) 8.23, 2.12, 1.56 Hz, 2H), 7.54 (ddd, J ) 8.23,
2.12, 1.56 Hz, 2H), 1.84 (q, J ) 7.53 Hz, 1H), 1.83 (q, J ) 7.36
Hz, 1H), 1.72 (br s, 1H), 1.55 (s, 3H), 0.78 (t, J ) 7.43 Hz,
3H); 13C NMR 153.09, 131.98, 125.82, 118.98, 110.51, 74.82,
36.58, 29.76, 8.03. Anal. Calcd for C11H13NO: C, 75.40; H, 7.48;
N, 7.99. Found: C, 75.29; H, 7.46; N, 7.99.
2-(4-Eth ylp h en yl)-2-bu ta n ol: bp 70 °C (0.10 Torr); 1H
NMR 7.37 (d, J ) 8.1 Hz, 2H), 7.19 (d, J ) 8.1 Hz, 2H), 2.67
(q, J ) 7.6 Hz, 2H), 1.85 (q, J ) 7.4 Hz, 1H), 1.84 (q, J ) 7.3
Hz, 1H), 1.82 (br s, 1H), 1.55 (s, 3H), 1.26 (t, J ) 7.6 Hz, 3H),
0.83 (t, J ) 7.4 Hz, 3H); 13C NMR 145.03, 142.30, 127.50,
125.39, 74.74, 36.58, 29.45, 28.30, 15.45, 8.32. Anal. Calcd for
C
12H18O: C, 80.85; H, 10.18. Found: C, 80.96; H, 10.10.
2-(3-Meth oxyp h en yl)-2-bu ta n ol:54 bp 86 °C (0.15 Torr);
1H NMR 7.25 (dd, J ) 8.2, 7.7 Hz, 1H), 7.03 (dd, J ) 2.1, 1.0
Hz, 1H), 7.00 (ddd, J ) 7.7, 1.0, 1.0 Hz, 1H), 6.78 (ddd, J )
8.2, 2.1, 1.0 Hz, 1H), 1.99 (br s, 1H), 3.81 (s, 3H), 1.83 (q, J )
7.5 Hz, 1H), 1.82 (q, J ) 7.2 Hz, 1H), 1.53 (s, 3H), 0.81 (t, J )
7.4 Hz, 3H); 13C NMR 159.39, 149.60, 128.95, 117.28, 111.41,
111.02, 74.74, 55.05, 36.51, 29.51, 8.20. Anal. Calcd for
lodged in
a pipet and then analyzed by GC. The same
C
11H16O2: C, 73.30; H, 8.95. Found: C, 73.36; H, 8.86.
2-(2-Meth ylp h en yl)-2-bu ta n ol: bp 45 °C (0.05 Torr); H
procedure was used for reactions with the Et3ZnM solutions
except that (1) an appropriate volume of the Et3ZnLi (0.1 M),
Et3ZnNa (0.32 M), or Et3ZnK (0.1 M) solution was used to
provide 0.2 mmol and (2) after addition was complete, the
reaction solutions were removed from the ice bath and stirred
at ambient temperature for only 10 min before hydrolysis. The
procedure for the reactions with Et2Zn(Bu4NBr), Et2Zn(t-
BuOLi), and Et2Zn(t-BuOK) was the same as for the reactions
with Et3ZnM except that, after removal from the ice bath, the
solution from the Et2Zn(Bu4NBr) reaction was stirred at
ambient temperature for 1 h before hydrolysis. White solids
formed upon hydrolysis of the Et2Zn(Bu4NBr) reactions, but
GC analysis showed no significant (>2%) loss of products.
Com p etition Exp er im en ts. A solution of the two ketones
(1.5 mL, 0.2 M in each) and dodecane (ca. 0.01 M, known
accurately in each case) was prepared in a small conical vial
and capped with a septum. The vial was removed from the
glovebox, and the needle of a gastight syringe (Hamilton, 1.0
mL) containing a solution of Et3ZnLi (0.30 mL, 0.10 M, 0.030
mmol) was inserted into the septum. The syringe was hori-
zontal so that it could be placed in a syringe pump (Sage
Instruments 341B), but the needle had been bent so that the
portion inserted through the septum was vertical and reached
below the surface of the solution of ketones. The vial was cooled
(-41 °C) in an acetonitrile-liquid N2 bath. The solution in the
syringe was added over 8 min, and then the solution was
allowed to stir briefly. Water (1.0 mL) was added, and the
solution became cloudy. Dilute aqueous HCl was added drop-
wise with stirring until all solids disappeared. Control experi-
ments showed that compositions did not change if solutions
were stirred for much longer times after acid had been added.
1
NMR 7.46 (dd, J ) 7.8, 2.0 Hz, 1H), 7.21-7.16 (c, 3H), 2.59 (s,
3H), 2.06 (br s, 1H), 2.01 (q, J ) 7.4 Hz, 1H), 1.97 (q, J ) 7.3
Hz, 1H), 1.65 (s, 3H), 0.86 (t, J ) 7.4 Hz, 3H); 13C NMR 144.55,
135.37, 132.42, 126.71, 126.65, 125.37, 76.01, 34.40, 28.69,
22.24, 8.51. Anal. Calcd for C11H16O: C, 80.44; H, 9.82.
Found: C, 80.30; H, 9.69.
P r ep a r a tion of Et3Zn Li Solu tion s. The preparation of
EtLi solutions was adapted from a literature procedure.11 Li
powder (2 equiv) was added to a rapidly stirred toluene
solution (1.0 M) of Et2Hg. Unreacted Li powder floated on the
surface until amalgamating with the Hg metal that was
forming. Additional Li (0.2-0.3 equiv) was added after 24 h if
unreacted Li was no longer evident. Additions of Li were
repeated every 24 h until unreacted Li persisted. The mixture
was filtered, and an aliquot of the resulting solution was
hydrolyzed and titrated for base to provide the EtLi concentra-
tion. A portion of the EtLi solution was added to Et2Zn (1.05
equiv), and the resulting Et3ZnLi solution was diluted to 0.10
M by addition of toluene. The solutions were stable for several
months, but eventually changed from colorless to yellow.
P r ep a r a tion s of Et3Zn Na a n d Et3Zn K Solu tion s. The
preparations were adapted from literature procedures.12,55
(53) Still, W. C.; Kahn, M.; Mitra, A. J . Org. Chem. 1978, 43, 2923.
(54) Since this work was completed, alcohol 9l has been noted in
the literature, but without characterization or properties: Choi, J . H.;
Youm, J . S.; Cho, C.-G.; Czae, M.-Z.; Hwang, B. K.; Kim, J . S.
Tetrahedron Lett. 1998, 39, 4835.
(55) Hein, F.; Petzchner, E.; Wagler, K.; Segitz, F. A. Z. Anorg. Allg.
Chem. 1924, 141, 161.

7756 J . Org. Chem., Vol. 65, No. 23, 2000
Musser and Richey
The aqueous layer was removed from the bottom of the vial
with a pipet, and MgSO4 was added to dry the remaining
liquid. The mixture was filtered through a piece of tissue
(Kimwipe) that was lodged in a pipet and then analyzed by
GC. The reported results are the average of three experiments,
and the range of values from individual experiments generally
was <(2%. Additional experiments were done when the
variation was larger.
Keith Maclin for performing the competition experiment
involving 10d .
Su p p or tin g In for m a tion Ava ila ble: GC conditions, re-
tention times of products, response factors used for the
products, and MS examination of products of reactions that
were quenched with D2O. This material is available free of
charge via the Internet at http://www.pubs.acs.org.
Ack n ow led gm en t. We are indebted to the National
Science Foundation for support of this work. We thank
J O000630J