Highly Steroselective Kinetic Enolate Formation: Steric vs Electronic Effects

Full text of DOI:10.1021/jo971260a

7516
J . Org. Chem. 1997, 62, 7516-7519
is not clear whether such enhanced Z preference is a
High ly Ster eoselective Kin etic En ola te
result of thermodynamic or kinetic deprotonation and, if
the latter, how the stereoselectivity is controlled by the
nature of active base species. This has prompted us to
investigate in more detail the relative importance of
steric vs electronic factors in the process of kinetic
enolization. We set out to study enolate selectivity from
six ketones (1a -f) in THF by four lithium amide bases:
lithium N-isopropyl(trimethylsilyl)amide (2), lithium N-
tert-butyl(trimethylsilyl)amide (3), lithium N-isopropyl-
anilide (4), and lithium N-(trimethylsilyl)anilide (5). Our
results suggest that excellent purity of E- or Z-enolate
can be selectively achieved by modifying the steric and
electronic nature of the bases.
F or m a tion : Ster ic vs Electr on ic Effects
Linfeng Xie,* Kurt M. Isenberger, Gary Held, and
Linnea M. Dahl
Department of Chemistry, University of Wisconsin,
Oshkosh, Wisconsin 54901
Received J uly 11, 1997
The utility of enolates has been well exemplified by
their applications in the synthesis of many biologically
important natural products via aldol condensation reac-
tions.1 Since specific biological activities of many natural
products are regulated by their stereochemistry, excellent
stereoselection during their synthesis is highly de-
manded.2 Previous studies have shown that kinetic aldol
diastereoselectivity depends to a large degree on the
geometry (cis or trans) of preformed enolates.3 One
important method of forming enolates is to deprotonate
a carbonyl compound with a base under either kinetic or
thermodynamic conditions. The more desirable kinetic
deprotonation usually results in regioselectively less
substituted enolates from unsymmetric ketones. Vari-
ables that affect such regioselectivity have been previ-
ously reviewed.4 However, factors governing stereoselec-
tivity of kinetic enolate formation are less well understood.
In this paper, we report the effects of the steric and
eletronic nature of lithium amide bases on the stereose-
lectivity of kinetic enolate formation.
Among many lithium amides used in kinetic enolate
formation, the most popular ones include lithium diiso-
propylamide (LDA), lithium tetramethylpiperidide
(LTMP), and lithium hexamethyldisilazide (LHMDS).
Through their elegant work with ethyl ketones, Heath-
cock and co-workers have reported the stereoselectivity
of kinetic enolization of many ketones with these bases.3a
Other researchers have studied related systems,5 em-
ploying different reaction conditions that result in either
thermodynamic or kinetic control. While LTMP was
shown to give slightly better kinetic selectivity in favor
of the E (or trans) enolate than LDA, LHMDS, a base
with comparable steric hindrance to both LDA and
LTMP, was reported to produce significantly more Z (or
cis) enolate. Use of strong cation solvating agents such
as hexamethylphosphoramide (HMPA) in enolization also
led to a substantial increase in the Z preference.5b,d,6 It
Resu lts a n d Discu ssion
Six acyclic ketones were subject to kinetic enolization
in THF with bases 2-5, which were prepared by treating
their corresponding amines7 with n-butyllithium. The
enolates formed were quenched with chlorotrimethylsi-
lane to give trimethylsilyl enol ethers, which were then
1
analyzed on GC and H NMR.8 The resulting E/Z ratios
are summarized in Table 1. The E/Z ratios previously
reported for LDA, LTMP, and LHMDS are also included
for comparisons.
We tested the temperature dependence of enolate
selectivity for 3-pentanone (1a ) by each base in order to
find optimum condition for maximum stereoselectivity.
It is clear from Table 1 that the E/Z ratios given by 2
and 4 show negligible variation with temperatures rang-
ing from 23 to -78 °C, but those by 3 and 5 are highly
temperature dependent. It is most interesting to note
that 3 gave the best E selectivity at room temperature
(entries 7-9 and 35-37), while 5 afforded outstanding
Z selectivity at -78 °C (entries 10-12). This striking
preference for the Z-enolate exhibited by 5 will be
discussed later in more detail. Comparison of the data
in Table 1 also indicates that both 2 and 4 gave similar
E/Z selectivity to that of LDA for all ketones studied. This
can be rationalized in view of the transition state model
proposed by Ireland and others (Scheme 1).6,9 The cyclic
chairlike transition state involves a concerted proton
transfer from the carbonyl compound to the base and the
lithium cation coordination to the oxygen. The amount
of E- and Z-enolates is determined by the energy differ-
ence between the two competing transition states Aq and
Bq that lead to the E- and Z-enolates, respectively. The
stability of the two transition states is, in turn, deter-
mined by both steric and electronic factors within the
structures. Due to the comparable steric hindrance10 of
an isopropyl, a phenyl, and a trimethylsilyl group, the
* Phone: (414) 424-0436. Fax: (414) 424-2042. e-mail: xie@uwosh.
edu.
(1) For recent reviews, see: (a) Evans, D. A.; Dart, M. J .; Duffy, J .
L.; Yang, M. G. J . Am. Chem. Soc. 1996, 118, 4322. (b) Braun, M.;
Sacha, H. J . Prakt. Chem./ Chem.-Ztg. 1993, 335, 653. (c) Heathcock,
C. H. In Asymmetric Synthesis; Morrison, J . D., Ed.; Academic Press:
New York, 1984; Vol. 3, p 111. (d) Masamune, S.; McCarthy, P. A. In
Macrolide Antibiotics; Omura, S., Ed.; Academic Press: New York,
1984; p 127. (e) J uaristi, E.; Beck, A. K.; Hansen, J .; Matt, T.;
Mukhopadhyay, T. Synthesis 1993, 12, 1271.
(6) Ireland, R. E.; Mueller, R. H.; Willard, A. K. J . Am. Chem. Soc.
1976, 98, 2868.
(7) N-isopropyl(trimethylsilyl)amine is not commercially available
and was prepared according to the procedure described: Courtois, G.;
Miginiac, L. J . Organomet. Chem. 1988, 340, 127.
(8) (a) House, H. O.; Czuba, L. J .; Gall, M.; Olmstead, H. D. J . Org.
Chem. 1969, 34, 2324. (b) Cazeau, P.; Duboudin, F.; Moulines, F.;
Babot, O.; Dunogues, J . Tetrahedron 1987, 43, 2075.
(9) Other modified transition state models have also been proposed.
See (a) Moreland, D. W.; Dauben, W. G. J . Am. Chem. Soc. 1985, 107,
2264. (b) Reference 5d. (c) Narula, A. S. Tetrahedron Lett. 1981, 22,
4119.
(2) Bartlett, P. A. Tetrahedron 1980, 19, 557.
(3) (a) Heathcock, C. H.; Buse, C. T.; Kleschick, W. A.; Pirrung, M.
C.; Sohn, J . E.; Lampe, J . J . Org. Chem. 1980, 45, 1066. (b) Evans, D.
A.; Vogel, E.; Nelson, J . V. J . Am. Chem. Soc. 1979, 101, 6120. (c)
Hirama, M.; Masamune, S. Tetradedron Lett. 1979, 2225. (d) Dubois,
J . E.; Fellman, P. Tetrahedron Lett. 1975, 1225.
(4) (a) d'Angelo, J . Tetrahedron 1976, 32, 2979. (b) Evans, D. A. In
Asymmetric Synthesis; Morrison, J . D., Ed.; Academic Press: New
York, 1984; Vol. 3, p 1.
(5) (a) Spears, G. W.; Caufield, C. E.; Still, W. C. J . Org. Chem. 1987,
52, 1226. (b) Xie, L.; Saunders, W. H., J r. J . Am. Chem. Soc. 1991,
113, 3123. (c) Masamune, S.; Ellingboe, J . W.; Choy, W. J . Am. Chem.
Soc. 1982, 104, 5526. (d) Fataftah, Z. A.; Kopka, I. E.; Rathke, M. W.
J . Am. Chem. Soc. 1980, 102, 3959.
(10) The A values for i-Pr, Ph, and TMS on a cyclohexane ring are
2.2, 2.8, and 2.5, respectively. For a list of A values, see: (a) Eliel, E.
L; Wilen, S. H. In Stereochemistry of Organic Compounds; J ohn Wiley
and Sons, Inc.: New York, 1993; pp 696-7. (b) Hirsch, J . A. Top.
Stereochem. 1967, 1, 199.
S0022-3263(97)01260-7 CCC: $14.00 © 1997 American Chemical Society

Notes
J . Org. Chem., Vol. 62, No. 21, 1997 7517
Ta ble 1. Ster eoselectivity of Kin etic En ola te
TMS and Ph groups in 2 and 4, respectively. Neverthe-
less, the steric effect remains predominant over the
electronic effect.
F or m a tion : Effects of Ster ic a n d Electr on ic Na tu r e of
Lith iu m Am id e Ba sesa
entry ketones
condns
LDA,
-70 °C
LTMP,
E/Z entry ketones condns
E/Z
From the transition state model in Scheme 1, one
would expect a sterically more demanding R2 T L1
interaction to destabilize the TS Bq and lead to an
increased E preference. This has indeed been generally
supported by literature data. For example, the highly
hindered amide base11 lithium tert-butyl-tert-octylamide,
has been reported to give predominantly the E-enolate.
However, the preparation of this amide base and other
bulky bases such as lithium di-tert-butylamide is rather
difficult. On the other hand, 3 is a highly hindered base
and can be readily prepared from commercially available
N-tert-butyl(trimethylsilyl)amine. It is remarkable to
note that the best E/Z selectivity can be achieved with 3
at room temperature (see entries 7-9 and 35-37). This
finding is of great synthetic significance because most
reported enolate forming reactions had to be carried out
at 0 °C or lower to achieve better selectivity. Compared
to the popular bases LDA and LTMP, 3 yields signifi-
cantly enhanced E selectivity for all ketones studied. For
example, it gave better than 90% (and as high as 94%)
of E-enolate from ketones 1a , 1b, 1e, and 1f in THF at
23 °C. Even with a larger R1 group, as in 1-phenyl-1-
propanone (1d ) and 2,2-dimethyl-3-pentanone (1c), 3 still
afforded 48% and 11% of E-enolate, respectively, (entries
31 and 25) as compared to less than 2% E-enolate given
by the other bases. The enhanced E selectivity is
conceivably due to the severe R2 T L1 interaction imposed
by 3 in the transition state Bq, which compensates for
the R1TR2 interaction in Aq. It is worthwhile noting that
3 was previously reported to give only slightly better
regioselectivity than LDA when employed in the enoliza-
tion of unsymmetric ketones.12
1
2
3
1a
1a
1a
70:30b 21
22
80:20b 23
24
1b
1c
1c
1c
5, -78 °C 5:95c
LDA
<2:98b
LTMP
<2:98b
-70 °C
LHMDS, <2:98b
-70 °C
LHMDS, 34:66b
-70 °C
25
26
27
28
1c
1c
1c
1d
3, 23 °C
4, 0 °C
5, -78 °C 0:100
LDA,
11:89
2:98
4
5
6
7
8
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
2, -78 °C 81:19
2, 0 °C
2, 23 °C
3, -78 °C 77:23
3, 0 °C
3, 23 °C
4, -78 °C 66:34
4, 0 °C
4, 23 °C
5, -78 °C 7:93
5, 0 °C
5, 23 °C
LDA,
-70 °C
LTMP,
86:14
82:18
<2:98b
-70 °C
92:8
94:6
29
1d
LHMDS, <2:98b
-78 °C
9
10
11
12
13
14
15
16
30
31
32
33
1d
1d
1d
1e
2, -78 °C 2:98
3, 23 °C
5, -78 °C 0:100
LDA,
68:32
68:32
48:52
87:13d
19:81
23:77
-60 °C
34
1e
1e
1e
1e
1e
1e
1f
1f
2, -78 °C 89:11e
3, -78 °C 77:23e
40:60b 35
36
68:32b 37
38
3, 0 °C
3, 23 °C
4, 0 °C
90:10e
93:7e
17
18
1b
1b
-70 °C
80:20e
LHMDS, 2:98b
39
40
41
5, -78 °C 10:90e
-70 °C
3, 23 °C
4, 0 °C
3, 23 °C
91:9
19
20
1b
1b
94:6c
5, -78 °C 2:98
42:58c
a
All results were obtained with a base-to-ketone ratio of 1.1:1.
Increasing the ratio to 2:1 gave identical E/Z ratios within
b
experimental error. Reference 3a. c Less than 1% of the more
d
substituted regioisomeric enolate was also obtained. Reference
5b. e The combined yield of the E- and Z-enolates from methylene
deprotonation is less than 16%. The major product derived from
enolization at the methyl carbon accounts for 84-91% of the total
enolates, as would be expected from a kinetic deprotonation.
Sch em e 1. Kin etic En ola te F or m a tion : Ir ela n d ’s
Tr a n sition Sta te Mod el
We now return to the remarkable reversed Z selectivity
given by 5. We undertook the study of this base to gain
a better understanding of why Z-enolate is favored by
lithium disilylamides (such as LHMDS) more than
lithium dialkylamides (such as LDA). For all ketones
studied, outstanding Z selectivity (greater than 90% in
all cases, entries 13, 21, 27, 32, 39, and 41) was observed
with 5. The high Z selectivity displayed by 5 is of
particular synthetic interest since previous Z selectivity
was most commonly achieved through the use of a
considerable amount of HMPA,5b,d,6 a toxic and carcino-
genic solvent additive.13 Our experimental evidence
strongly suggests that this Z preference afforded by 5 is
a result of kinetic deprotonation rather than thermody-
namic enolate equilibration. First, enolates were formed
in quantitative yields and no ketones in excess could be
detected in our reactions. Second, the less substituted
enolates accounted for 84% of the total enolates produced
from 1e and 99% of the total enolates from 1b, as
expected from a kinetically controlled enolization. Third,
we allowed the enolates from 1a to equilibrate in the
presence of excess ketone (1a ) and found the thermody-
namic Z selectivity (E/Z ) 12:88)14 to be lower than the
kinetic Z selectivity given by 5. If the deprotonation of
competing steric interactions (R1 T R2 and R2 T L1)
should be more or less the same for 2 (L1 ) i-Pr, L2 )
TMS), 4 (L1 ) i-Pr, L2 ) Ph), and LDA (L1 ) L2 ) i-Pr).
As a result, comparable E/Z ratios are observed for all
three bases. It is important to note that electronic nature
(and basicity) of the nitrogen has changed as one i-Pr
group in LDA is replaced by the electron-withdrawing
(11) Corey, E. J .; Gross, A. W. Tetrahedron Lett. 1984, 25, 495.
(12) Prieto, J . A.; Suarez, J .; Larson, G. L. Synth. Commun. 1988,
18, 253.
(13) (a) Ashby, J .; Styles, J . A.; Anderson, D. Br. J . Cancer 1977,
36, 564. (b) Ashby, J .; Styles, J . A.; Paton, D. Br. J . Cancer 1978, 38,
418. (c) Lloyd, J . W. J ., Am. Ind. Hyg. Assoc. 1975, 36, 917.
(14) The equilibrium constant for 1a enolates was measured in the
presence of 20% excess ketone 1a . The ratio is independent of the
nature of bases used.

7518 J . Org. Chem., Vol. 62, No. 21, 1997
Notes
2,2-Dimethyl-3-pentanone from Wiley Organics was dried over
3 Å molecular sieves before use. The identity and purity of all
ketones were confirmed by 1H NMR to indicate no detectable
amount of impurity or water.
the ketones by 5 is under kinetic control, then a promi-
nent question is, why did the change from LDA to base
5 (by replacing the i-Pr’s with Ph and TMS groups)
increase the Z selectivity so dramatically? From our
earlier discussion, it is clear that the change from LDA
to 2 (by replacing one i-Pr with a TMS group) or 4 (by
replacing one i-Pr with a Ph group) caused only a small
to negligible change in E/Z ratios. The intrinsic steric
hindrance is thus nearly the same for bases 2, 4, and
LDA. If a replacement of an i-Pr by a TMS or a Ph group
caused negligible change in steric hindrance, then the
intrinsic steric hindrance of base 5 is expected to be
similar to 2, 4, and LDA. However, we believe that the
replacement of both isopropyl groups in LDA with
electron-withdrawing TMS or phenyl groups significantly
reduced the electron density (and the basicity) on the
nitrogen in 5. As a result, the electrostatic interaction
between the less basic nitrogen and the lithium ion and
the proton in the transition state is much weaker, leading
to a loose (or late) transition state. Consequently, the
R2 T L1 interaction is significantly weakened and the
transition state Bq leading to the Z-enolate is favored.
The change in the electronic nature of the bases has thus
become a dominant factor in the enolization by base 5.
Similarly, the previously reported Z selectivity given by
lithium disilylamides such as LHMDS and lithium bis-
(diphenylmethyl)silylamide can be rationalized. It is
interesting that the latter, with intrinsically more hin-
dered diphenylmethyl groups, gave much better Z
selectivity5c than LHMDS. This is consistent with the
above rationale that electronic stabilization due to phenyl
groups on the silicon further loosened the transition state
in favor of Bq. Further experimental study of such
electronic effect is now under progress.
N-ter t-Bu tyl(tr im eth ylsilyl)a m in e. Aldrich 98% N-tert-
butyl(trimethylsilyl)amine was dried over CaH2 and distilled
immediately prior to use.
N-Isop r op yla n ilin e. TCI America 99+% N-isopropylaniline
was dried over 3 Å molecular sieves in a nitrogen atmosphere
and used without further purification. Its purity was checked
with 1H NMR and GC-MS.
N-(Tr im et h ylsilyl)a n ilin e. Pfaltz & Bauer 97% N-(tri-
methylsilyl)aniline from was dried over 3 Å molecular sieves in
a nitrogen atmosphere and used without further purification.
The purity was confirmed by both 1H NMR and GC-MS.
N-Isop r op yl(tr im eth ylsilyl)a m in e. This compound was
synthesized following a modified literature procedure.7 A 1 L
round-bottom flask was filled with 250 mL of dry diethyl ether
and 120 g of isopropylamine (2.0 mol). The content was stirred
by a mechanical stirrer and cooled to 0 °C. Then 82 g (0.76 mol)
of chlorotrimethylsilane was added dropwise to the flask via a
dropping funnel, and white precipitates of isopropylammonium
chloride started to form. After the addition of chlorotrimethyl-
silane, the mixture was stirred for additional 20 min before being
allowed to warm to room temperature and was subsequently
refluxed for 1 h.
The reaction mixture was allowed to cool to room temperature
after refluxing, and the ammonium salt was filtered away by
suction filtration with a frit. The flask and the salt were washed
twice with 100 mL of dry ether, and the ethereal solution was
combined with the mother liquor. The ether solvent and iso-
propylamine were removed by fractional distillation, and 55 g
of the product, N-isopropyl(trimethylsilyl)amine, was collected
(55%). The product was further purified by distillation via a
micro spinning band distillation system. bp 100-101 °C (lit.7,15
bp 98 and 101 °C). 1H NMR (CDCl3): δ (ppm from TMS) 0.05
(s, 9H), 0.13 (s, 1H, NH), 1.03 (d, 6H), 2.92-3.08 (m, 1H). 13C
NMR (CDCl3): δ (ppm from TMS) 0.4, 27.8, 43.0. The amine
was dried over calcium hydride and distilled immediately before
use.
In conclusion, we have reported highly stereoselective
kinetic enolate formation employing lithium amide bases
with different steric and electronic characteristics. Ex-
cellent E or Z stereoselectivity can be achieved with
sterically hindered 3 or electronically dominated 5. We
conclude that enhanced Z selectivity is observed only
when both alkyl groups in LDA are changed to electron-
withdrawing groups (such as TMS or Ph). Both excellent
E selectivity by lithium N-tert-butyl(trimethylsilyl)amide
(3) at room temperature and Z selectivity by lithium
N-(trimethylsilyl)anilide are of particular synthetic im-
portance.
The above nonaqueous workup is essential to preventing
N-isopropyl(trimethylsilyl)amine from hydrolyzing. Over a time
period of several weeks, it was found that N-isopropyl(trimeth-
ylsilyl)amine had hydrolyzed to give 3-5% of hexamethyldisi-
loxane, which was difficult to separate due to its boiling point
being identical with that of the amine. Control experiments,
however, have shown that the presence of 3-5% hexamethyl-
disiloxane in the amine has no effect on the enolate stereose-
lectivity.
Tr im eth ylsilyl En ol Eth er s. Two previously reported
procedures8 were used in the synthesis of authentic samples of
trimethylsilyl enol ethers from our ketones. Lithium diisopro-
pylamide in THF at 0 °C was used for preparing kinetically
controlled trimethylsilyl enol ethers, while triethylamine in
refluxing DMF was used for preparing the enol ethers under
thermodynamic control. The product enol ethers were separated
Exp er im en ta l Section
Gen er a l. Glassware and syringe needles were dried at 140
°C overnight in an oven. Gastight syringes were dried in a
desiccator under vacuum (e0.1 Torr) for at least 1 h before use.
The enolization experiments were carried out in Schlenk flasks
under a nitrogen atmosphere. 1H NMR was obtained on a
1
and collected on a preparative GC. Their H NMR spectra were
compared to those reported in the literature and their structures
assigned on the basis of the methods developed by other authors.
The structure assignment was further validated using mass
spectrometry on a GC-MS.
J EOL-GSX 270 MHz spectrometer with
a superconducting
magnet. The quantitative analysis of trimethylsilyl enol ethers
was carried out on an HP-5890 series GC-MS with a 5791 mass
detector and on the NMR.
Solven t. Aldrich 99.5% tetrahydrofuran was dried over Na/
benzophenone under N2 until a purple color persists and distilled
immediately prior to use.
n -Bu tyllith iu m . Aldrich 1.6 M n-butyllithium in hexane was
titrated with reagent grade diphenylacetic acid for its exact
concentration. In most cases the n-butyllithium concentration
was found to be within 10% of what was labeled on the bottle.
Keton es. 3-Pentanone, 2-methyl-3-pentanone, and 2-pen-
tanone (Aldrich) were distilled on a B/R Model 8T micro spinning
band distillation system and dried over 3 Å molecular sieves.
Aldrich propiophenone and 2,6-dimethyl-4-heptanone were dried
over 3 Å molecular sieves and used without further purification.
Gen er a l P r oced u r e of En oliza tion . 2-5 were prepared
by adding 1.1 mmol of n-butyllithium to a stirring solution of
1.2 mmol of the corresponding amines in 5 mL of dry THF at 0
°C under nitrogen. The solution was allowed to stir for 15 min.
The solution was then kept at 0 °C and raised to room
temperature or lowered to -78 °C, depending on the conditions
desired. Ketone (1 mmol) was then slowly added by a gastight
syringe, and the reaction was stirred for an additional 15 min
before quenching the enolates with 1.2 mmol of freshly distilled
chlorotrimethylsilane (over CaH2). After stirring for 10 min, the
reaction mixture was allowed to warm to room temperature and
poured into 10 mL of pentane in a small separatory funnel. The
(15) Berger, A.; Magnuson, J . A. Anal. Chem. 1964, 36, 1156.

Notes
J . Org. Chem., Vol. 62, No. 21, 1997 7519
mixture was washed twice with 5 mL each of saturated sodium
bicarbonate solution and once with brine solution, and the
organic layer was dried over anhydrous magnesium sulfate. The
solution was filtered and the product ratios determined by GC-
MS. At least two individual experiments (normally three or four
experiments) for each entry in Table 1 were run to ensure
reproducible results. Usually two or three GC injections were
made for each experimental run. To ensure kinetically con-
trolled conditions, we also carried out the above reactions using
a base/ketone molar ratio of 2:1. The E/Z ratios showed no
difference from those obtained with a base/ketone ratio of 1.1:1.
Con tr ol Exp er im en ts. Two individual experiments were
carried out for the enolization of 3-pentanone by LDA and
LHMDS in THF at -78 °C, and the E/Z ratios obtained were
shown to be identical with the literature results within experi-
mental error. Another control experiment was carried out to
investigate the effect of 3-5% of hexamethyldisiloxane on the
product enolate ratio when lithium N-isopropyl(trimethylsilyl)-
amide was used as the base. Thus lithium N-isopropyl(tri-
methylsilyl)amide was prepared in THF as described above. The
solvent and hexamethyldisiloxane were then removed by apply-
ing a high vacuum. A white solid presumably due to lithium
N-isopropyl(trimethylsilyl)amide was obtained. The desired
amount of dry THF (5 mL in all experiments) was then added
to dissolve the solid, and thereafter the procedure described in
Gen er a l P r oced u r e of En oliza tion was followed. The product
ratio was found to be, within experimental error, identical with
that given by the same base when 3-5% of hexamethyldisiloxane
was present. We conclude that the effect due to the presence of
3-5% of hexamethyldisiloxane is negligible.
Ack n ow led gm en t. We thank the University of
WisconsinsOshkosh Faculty Development Program for
the financial support of this project. We also acknowl-
edge the National Science Foundation and the College
of Letters and Science at UWsOshkosh for the funding
of the J EOL-GSX 270 MHz superconducting NMR
spectrometer.
J O971260A