NMR kinetic studies on the decomposition of β-amidozinc reagents: Optimization of palladium-catalyzed cross-coupling with acid chlorides

Article abstract of DOI:10.1021/jo000558p

The decomposition of β-amidozinc reagent 4 by β-elimination has been shown to be a unimolecular process in both THF and DMF as solvent, with relative rates of 4:1 at room temperature, and activation parameters have been determined. These results indicate

Full text of DOI:10.1021/jo000558p

J . Org. Chem. 2000, 65, 7417-7421
7417
NMR Kin etic Stu d ies on th e Decom p osition of â-Am id ozin c
Rea gen ts: Op tim iza tion of P a lla d iu m -Ca ta lyzed Cr oss-Cou p lin g
w ith Acid Ch lor id es
Charles S. Dexter, Christopher Hunter, and Richard F. W. J ackson*
Department of Chemistry, Bedson Building, The University of Newcastle upon Tyne,
Newcastle upon Tyne, NE1 7RU, UK
J ason Elliott
Merck Sharp and Dohme Research Laboratories, Neuroscience Research Park, Terlings Park,
Eastwick Road, Harlow, Essex, CM20 2QR, UK
r.f.w.jackson@ncl.ac.uk
Received April 13, 2000
The decomposition of â-amidozinc reagent 4 by â-elimination has been shown to be a unimolecular
process in both THF and DMF as solvent, with relative rates of 4:1 at room temperature, and
activation parameters have been determined. These results indicate the â-elimination is a syn-
process. NMR experiments reveal that as little as 2 equiv of DMF can have a significant stabilizing
influence on reagent 4. Use of a mixture of DMA and toluene as the bulk solvent, in place of DMF,
has allowed successful palladium-catalyzed cross-coupling reactions of both 4 and the homologous
reagent 5 with acid chlorides to yield unsymmetrical ketones (nine examples).
In tr od u ction
Sch em e 1
The development of methodology for the synthesis of
amino acids continues to stimulate interest within or-
ganic chemistry. In particular, â- and γ-amino acids have
attracted much recent attention owing to their presence
1
in biologically active compounds and their use as build-
than the closely related â-amidozinc reagent 1.13,14 13C
NMR studies of reagents 4 and 5, in which downfield
shifts of the carbamate carbon are observed upon forma-
tion of zinc reagent from the iodide precursor, suggested
that intramolecular coordination of the carbamate car-
bonyl group to the zinc was occurring in THF. The effect
of this coordination was promotion of â-elimination,
presumably due to the zinc acting as an internal Lewis
acid (Scheme 1). However, when the reagents 4 and 5
were prepared in a dipolar aprotic solvent such as DMF,
suppression of the internal coordination was observed by
ing blocks in modified peptides.2
-9
We have recently
extended our existing methodology for preparing R-amino
acids using zinc reagents 1, 2, and 31
0,11
to the synthesis
of â- and γ-amino acid derivatives using the zinc reagents
and 5, respectively.12
4
1
3
C NMR, and â-elimination was significantly reduced.
Subsequent palladium(0)-catalyzed reactions allowed
cross-coupling with aryl iodides to yield phenylalanine
homologues, and copper-mediated reactions with allylic
halides and Michael acceptors all proceeded in good
We observed that â-amidozinc reagents 4 and 5, when
prepared in THF, underwent â-elimination more rapidly
12
yield.
It was evident that the requirement for a dipolar
aprotic solvent to stabilize the zinc reagents 4 and 5
would preclude reaction with some electrophiles, specif-
ically acid chlorides, which react with DMF. We have
therefore examined the decomposition of zinc reagent 4
in some detail to determine the nature of the â-elimina-
(
1) J uaristi, E. Enantioselective synthesis of â-amino acids; Wiley-
VCH: New York, 1997.
2) Daura, X.; Gademann, K.; J uan, B.; Seebach, D.; van Gunsteren,
W. F.; Mark, A. E. Angew. Chem., Int. Ed. 1999, 38, 236-240.
3) Gademann, K.; J aun, B.; Seebach, D. Helv. Chim. Acta 1999,
(
(
8
2, 1-11.
(
(
(
4) Gung, B. W.; Zou, D. J . Org. Chem. 1999, 64, 2176-2177.
5) Koert, U. Angew. Chem., Int. Ed. Engl. 1997, 36, 1836-1837.
6) Matthews, J . L.; Gademann, K.; J aun, B.; Seebach, D. J . Chem.
Soc., Perkin Trans. 1 1998, 3331-3340.
(11) J ackson, R. F. W.; Moore, R. J .; Dexter, C. S.; Elkiott, J .;
Mowbray, C. E. J . Org. Chem. 1998, 63, 7875-7884.
(12) Dexter, C. S.; J ackson, R. F. W.; Elliott, J . J . Org. Chem. 1999,
64, 7579-7585.
(
7) Seebach, D.; Matthews, J . L. J . Chem. Soc., Chem. Commun.
1
997, 2015-2022.
(
8) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173-180.
(9) Apella, D. H.; Barchi, J . J ., J r.; Durell, S. R.; Gellman, S. H. J .
(13) Dunn, M. J .; J ackson, R. F. W.; Pietruszka, J .; Turner, D. J .
Org. Chem. 1995, 60, 2210-2215.
Am. Chem. Soc. 1999, 121, 2309-2310.
10) Gair, S.; J ackson, R. F. W. Curr. Org. Chem. 1998, 2, 527-
50.
(
(14) Duddu, R.; Eckhardt, M.; Furlong, M.; Knoess, H. P.; Berger,
S.; Knochel, P. Tetrahedron 1994, 50, 2415-2432.
5
1
0.1021/jo000558p CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/12/2000

7
418 J . Org. Chem., Vol. 65, No. 22, 2000
Dexter et al.
Ta ble 1. Activa tion P a r a m eter s for th e Decom p osition
a
of 4 in THF a n d DMF
rate constant
at 25 °C, k/s
b
-1
‡
-1
∆S /J K mol-1
‡
-1
∆H /kJ mol
-
-
4
THF-d8
DMF-d7
1.02 × 10
+70
+90
-85
-32
4
0.26 × 10
F igu r e 1. Syn elimination of zinc reagent 4.
a
The estimated uncertainty in ∆H is ( 5 kJ mol-1, and in ∆S‡
‡
-
1
-1 b
(
10 J K mol
.
The rate constants were measured over the
range 25 °C to 55 °C in THF-d8, and 25 °C to 70 °C in DMF-d7.
tion as a function of solvent. We have also examined
whether the addition of small amounts of dipolar aprotic
solvents (e.g., DMA), which are less reactive than DMF
toward acid chlorides, can stabilize reagent 4 generated
in other solvents, and thus allow coupling with acid
chlorides.
F igu r e 2. Anti elimination of zinc reagent 4.
species in the transition state rather than two (Figure
). The smaller enthalpy of activation observed in THF
2
is most easily explained by intramolecular coordination
of the carbamate group to the zinc, apparently required
for elimination, already being present in the ground state.
In DMF such coordination in the ground state is appar-
ently less significant (as already observed by ∆δ in the
Resu lts a n d Discu ssion s
When the aspartic acid derived zinc reagent 4 was
allowed to decompose fully in either THF-d or DMF-d ,
8 7
1
3
12
it appeared that the sole degradation product was that
occurring via â-elimination (Scheme 1). The reaction
proceeded cleanly, and appeared amenable to kinetic
analysis.
Kin etic Stu d ies of Decom p osition of Asp a r tic
Acid Der ived Rea gen t Zin c 4. The zinc reagent was
C spectra), presumably due to intermolecular coordi-
nation of the zinc by solvent DMF. The need to lose this
solvation (at least partially) before elimination can occur
results in a higher enthalpy of activation in DMF. The
‡
most reasonable explanation for the observation that ∆S
is more negative in THF is that while in the ground state
there is strong intramolecular coordination between the
carbamate and zinc, increased solvation of the zinc is
required as the transition state is reached. In DMF, the
zinc is already solvated in the ground state, and the
decrease in entropy resulting from coordination of the
carbamate to zinc can be partially offset by the increase
of entropy due to concurrent loss of coordinated DMF as
the transition state forms.
In flu en ce of Solven t on Str u ctu r e of Zin c Re-
a gen t 1. We appreciated that the DMF which is required
to stabilize the â-amidozinc reagents 4 and 5 would be
incompatible with the use of reactive electrophilic cou-
pling partners, such as acyl chlorides. Indeed, Knochel
has reported that when cross-coupling of a mixed orga-
nozinc/copper reagent was attempted with benzoyl chlo-
ride in DMF, addition of the organometallic reagent to
the iminium species (formed from DMF and benzoyl
chloride) was observed.16
prepared in THF-d
studies.12 A plot of the percentage of starting zinc reagent,
A, against time showed exponential decay of the zinc
8
by the procedure used in our previous
%
reagent at 25 °C, while a plot of ln(%A) against time
2
generated a straight line (R ) 0.9985). This indicated
that the reaction was first order with a unimolecular
rate-limiting step, solely involving zinc reagent 4. In the
same manner, the rates of decomposition of 4 were
monitored at 35, 45, and 55 °C. All runs were monitored
for a minimum of two half-lives (and indeed in the
majority of the kinetic runs, to more than 90% decom-
position of the reagent) and the quality of fit to the first-
2
order rate law (as measured by the value of R ) did not
depend on the extent of reaction monitored. The zinc
reagent was also prepared in DMF-d
7
by the same
procedure. A plot of ln(%A) against time generated a
2
straight line (R ) 0.9989), again indicating that in DMF
the reaction was unimolecular, involving solely zinc
reagent 4. This implies that the formation of zinc salts
during the reaction plays no part in the elimination
process. The rate constant for the reaction at 25 °C in
Our search for a suitable solvent in which to carry out
cross-coupling of 4 with acid chlorides began with further
NMR studies. We required a solvent system that would
both stabilize the organozinc reagents toward â-elimina-
tion, while also being tolerant of reactive electrophiles.
Our previous work had shown that the structurally
related serine-derived zinc reagent 1 (also a â-amidozinc
-
4
-1
DMF (k ) 0.26 × 10
s ) is approximately four times
smaller than the corresponding rate constant in THF (k
-
4
-1
)
1.02 × 10
s ). Although â-elimination of 4 is not
totally suppressed in DMF, it is significantly slowed,
thereby allowing subsequent cross-coupling reactions to
have greater opportunity to proceed. The decomposition
of 4 in DMF was monitored in the same manner at 40,
reagent) was significantly less prone to â-elimination in
THF than reagents 4 and 5.1
1,12
This gave us the
and study
opportunity to prepare the reagent 1 in THF-d
8
5
5, and 70 °C.
the change in chemical shift of the coordinating carbonyl
groups upon addition of DMF to the solution (Table 2).
We hoped to establish from this study how many equiva-
lents of DMF would be required to disrupt intramolecular
coordination in 1 and therefore deduce how many equiva-
lents of DMF would be expected to stabilize the related
reagent 4 (and 5) in THF.
Discu ssion of th e Activa tion P a r a m eter s. The
activation parameters for the elimination reaction were
determined in the usual way from the Arrhenius equa-
tion,15 in both solvents (Table 1). The fact that ∆S is
negative for a first-order reaction in both solvents sug-
gests that the elimination is a syn-process (Figure 1); if
it were anti we might expect an increase, rather than a
decrease in entropy because there are three developing
‡
We interpret a downfield shift of the carbonyl group
resonance, as shown by a positive value of ∆δ, as
(
15) Maskill, H. The physical basis of organic chemistry; Oxford
(16) Majid, T. N.; Knochel, P. Tetrahedron Lett. 1990, 31, 4413-
4416.
University Press: Oxford, 1985.

Decomposition of â-Amidozinc Reagents
J . Org. Chem., Vol. 65, No. 22, 2000 7419
Ta ble 2. Va r ia tion of ∆δ of th e 13C Sh ifts of th e
Ca r bon yl Gr ou p s w ith Nu m ber of Equ iva len ts of DMF
Ad d ed to a THF -d 8 Solu tion of 1
Sch em e 2
DMF (equiv)
0
∆δcarbamatea
∆δestera
∆δDMF amideb
+2.711
+2.180
+1.785
+1.045
+0.526
+5.347
+5.586
+5.786
+6.112
+6.383
-
0
1
2
4
.5
+4.55
+4.35
+3.70
+2.50
a
∆
δ was calculated as the difference in chemical shift relative
to the δcarbonyl of the precursor iodide in THF-d8 (i.e., ∆δ ) [δ(R-ZnI)
b
-
δ(R-I)]). These shifts are relative to δcarbonyl of a reference sample
of 2.0 equiv of DMF in THF-d8.
Sch em e 3
Sch em e 4
F igu r e 3. Addition of DMF to zinc reagent 1 in THF.
Following these results, the aspartic acid derived zinc
reagent 4 was generated in THF-d containing 2 equiv
8
of DMF, and then transferred into an NMR tube. The
1
initial H NMR spectrum indicated clean formation of the
zinc reagent without significant evidence for decomposi-
tion. However, a 1H NMR spectrum, taken after a
13
C
NMR spectrum had been run, indicated that decomposi-
tion was starting to occur. The rate of decomposition was
measured at 25 °C, which allowed us to establish that
the reaction was first order, with the rate constant (k )
-
4
-1
0
.6 × 10
s ) intermediate between the values in neat
DMF and neat THF.
P a lla d iu m -Ca ta lyzed Cr oss-Cou p lin g Rea ction s
w ith Acid Ch lor id es. Given that THF is problematic
when used as a solvent for cross-coupling reactions
1
7,18
between organozinc reagents and acid chlorides,
we
identified toluene as a suitable inert bulk solvent, and
dimethylacetamide (DMA) as the dipolar aprotic cosol-
vent. Thus, zinc reagent 4 was prepared in toluene/DMA
and then cross-coupled with iodobenzene (Scheme 2) to
give â-homophenylalanine derivative 7 in identical yield
to that obtained in neat DMF12 and much better than
that obtained when using THF alone. We have found that
activation of zinc using only chlorotrimethylsilane, and
the use of three equivalents of zinc, are perfectly satisfac-
tory for the formation of 4.
F igu r e 4. Convergence of the diastereotopic methylene
protons on addition of DMF to a solution of 1 in THF-d .
8
implying coordination between the carbonyl group and
zinc. As DMF is added, coordination between the zinc
and carbamate becomes progressively weaker (Figure 3).
Furthermore, with less electron density being donated
from the carbamate group, the ester group is seen to
coordinate more strongly. The carbonyl group of solvent
DMF is also observed to shift, the magnitude decreasing
as more equivalents of DMF are added. Presumably, this
chemical shift reflects an average for both the coordinat-
ing and noncoordinating solvent molecules. Interestingly,
The coupling reactions of this reagent and the homolo-
gous reagent 5 with acid chlorides (catalyzed by pal-
ladium acetate and triphenylphosphine) was then inves-
tigated. Use of a small range of acid chlorides allowed
the preparation of the ketones 8-16 in moderate yields,
with the exception of hexanoyl chloride which was a poor
substrate (Schemes 3 and 4, Table 3).
1
as more equivalents of DMF were added, the H spectra
showed the diastereotopic CH -ZnI protons converging,
2
becoming more like the spectrum of the same reagent in
neat DMF (Figure 4). These observations show that it is
possible to achieve disruption of carbamate coordination
with relatively low concentrations of DMF.
(
(
17) Bhar, S.; Ranu, B. C. J . Org. Chem. 1995, 60, 745-747.
18) Fraser, J . L.; J ackson, R. F. W.; Porter, B. Synlett 1995, 819-
820.

7
420 J . Org. Chem., Vol. 65, No. 22, 2000
Dexter et al.
Ta ble 3. Yield s of P a lla d iu m -Ca ta lyzed Cr oss-cou p lin g
Rea ction s betw een 4 a n d 5 w ith Acyl Ch lor id es in
Tolu en e w ith 2.0 Equ iv of DMA
mmol), and triphenylphosphine (0.0264 g, 0.010 mmol) were
added successively to the reaction mixture, which was stirred
at room temperature for 3 h. The mixture was partitioned
between saturated aqueous ammonium chloride (30 mL) and
ethyl acetate (50 mL), and then the mixture was filtered. The
organic layer was separated, washed with brine (20 mL), dried,
filtered, and evaporated to dryness. Flash chromatography
over silica gel eluting with an appropriate petroleum ether-
ethyl acetate gradient furnished the protected 5-oxo-â-amino
acids 8-12 and the 6-oxo-γ-amino acids 13-16.
zinc reagent
R
coupled product
% yield
4
4
4
4
4
5
5
5
5
Ph
8
9
59
46
49
20
51
51
48
52
45
CH2dCH
AcOCH2
CH3(CH2)4
2-furyl
10
11
12
13
14
15
16
Ph
Meth yl 3(R)-[(N-ter t-Bu toxyca r bon yl)a m in o]-5-oxo-5-
p h en ylp en ta n oa te (8). Treatment with benzoyl chloride
yielded 8 (0.142 g, 59%), isolated as a pale brown solid, mp
CH2dCH
AcOCH2
2-furyl
6
1-64 °C; found C, 63.41; H, 7.14; N, 4.18, C17
5
H23NO requires
-
1
C, 63.54; H, 7.21; N, 4.36%; νmax (KBr disk)/cm 3373, 2983,
Exp er im en ta l Section
DMF and toluene were distilled from calcium hydride and
stored over 4 Å molecular sieves. DMA was obtained from
Aldrich in Sure-Seal bottles and used as supplied. Organic
extracts were dried over magnesium sulfate and filtered, and
the solvent was then removed using a rotary evaporator.
F or m a tion of Am in o Acid Der ived Or ga n ozin c Re-
a gen ts 1, 4, a n d 5 in DMF . Gen er a l P r oced u r e. Zinc dust
1
6
6
4
H
740, 1679, 1529, and 1160; δ 1.41 (9 H, s), 2.72 (1 H, dd, J
.4 and 16.2), 2.82 (1 H, dd, J 5.2 and 16.2), 3.29 (1 H, dd, J
.6 and 17.4), 3.41 (1 H, dd, J 4.0 and 17.4), 3.66 (3 H, s), 4.43-
.48 (1 H, m), 5.49 (1 H, d, J 7), 7.44 (2 H, m), 7.53-7.59 (1 H,
m), and 7.94-7.97 (2 H, m); δ
7
C
28.30, 37.80, 41.64, 44.34, 51.66,
9.44, 128.07, 128.63, 133.39, 136.63, 153.04, and 172.06; m/z
+
+
(EI) 265 (9%, M - C
4
+
H
8
), 220 (19), 105 (100, PhCO ), 77 (28)
and 57 (78); found M 321.1564, C17 requires 321.1576;
H
23NO
5
(
0.284 g, 4.5 mmol) was weighed into a 50 mL round-bottom
flask with sidearm which was flushed with nitrogen. Dry DMF
0.5 mL) and 1,2-dibromoethane (19 µL, 0.225 mmol) were
26
[R]
D
2 2
-14.1 (c 1.085 in CH Cl ).
Meth yl 3(R)-[(N-ter t-Bu toxycar bon yl)am in o]-5-oxoh ept-
(
6
-en oa te (9). Treatment with acryloyl chloride yielded 9 (0.092
added, and the mixture was stirred vigorously. The mixture
was gently heated using a heat gun, so that the evolution of
ethene was observed, before being allowed to attain room
temperature. Chlorotrimethylsilane (6 µL, 0.046 mmol) was
added to the mixture which was stirred for a further 30 min.
A solution of amino acid derived iodide (0.75 mmol) in DMF
was transferred under nitrogen via syringe to the reaction
mixture at room temperature. The reaction was judged to be
complete by TLC analysis (petroleum ether-ethyl acetate, 2:1)
after approximately 15 min (The starting iodide is UV active
while the products are not).
-1
g, 46%), isolated as a pale brown oil. νmax (cap. film)/cm 3366,
2
H
978, 1736, 1713, 1367, and 1168; δ 1.43 (9 H, s), 2.65 (1 H,
dd, J 6.4 and 16.2), 2.74 (1 H, dd, J 5.1 and 16.2), 2.91 (1 H,
dd, J 4.9 and 17.0), 3.04 (1 H, dd, J 6.7 and 17.0), 3.68 (3 H,
s), 4.28-4.36 (1 H, m), 5.36 (1 H, d, J 7.6), 5.89 (1 H, dd, J 1.2
and 10.2), 6.26 (1 H, dd, J 1.2 and 17.7), and 6.33 (1 H, dd, J
1
1
0.2 and 17.7); δ
29.19, 136.54, 155.05, 172.00, and 198; m/z (EI) 215 (3%, M
), 184 (40), 170 (10), 154 (20), 102 (30), and 57 (100);
C
28.35, 37.83, 42.57, 44.17, 51.67, 79.55,
+
-
4 8
C H
+
26
found MH 272.1492, C13
H
22NO
5
requires 272.1498. [R]
D
+3.2
2 2
(c 0.25 in CH Cl ).
NMR Exp er im en ts: Stu d ies of th e Asp a r tic Acid
Meth yl 3(R)-[(N-ter t-Bu toxycar bon yl)am in o]-6-acetoxy-
-oxoh exa n oa te (10). Treatment with acetoxyacetyl chloride
Der ived Or ga n ozin c Rea gen t in both THF -d
. Organozinc reagent 4 was prepared from iodide 6 using
the general procedure described above, but using DMF-d and
THF-d in place of the nondeuterated solvents. When no
8
a n d DMF -
5
d
7
yielded 10 (0.116 g, 49%), isolated as a pale brown oil. νmax
3
7
371, 2978, 1734, 1368, 1236, and 1169; δ 1.45 (9 H, s), 2.17
H
8
(
3 H, s), 2.64-2.88, 2.64 (1 H, dd, J 6.3 and 16.2), 2.71 (1 H,
starting material remained (as judged by TLC), the excess zinc
dust was allowed to settle. The supernatant was transferred
via syringe into a nitrogen-filled NMR tube fitted with a
Young’s tap. Generally, a small amount of excess zinc was
unavoidably transferred into the NMR tube, although this did
not appear to affect the quality of the spectra. References used
dd, J 5.4 and 16.2), 2.77 (1 H, dd, J 6.6 and 17.0), 2.88 (1 H,
dd, J 5.1 and 17.0), 3.68 (3 H, s), 4.29-4.33 (1 H, m), 4.64 (2
H, d, J 1.3), and 5.34 (1 H, br d, J 8.2); δ 20.42, 28.32, 31.70,
C
4
2
+
2.10, 43.74, 68.16, 79.72, 155.01, 170.20, and 202.56. m/z (EI)
2
6
01 (13%), 188 (32), 157 (24), 102 (65), and 57 (100). [R]
4.4 (c 0.5 in CH Cl ).
Meth yl 3(R)-[(N-ter t-Bu toxyca r bon yl)a m in o]-5-oxod e-
ca n oa te (11). Treatment with hexanoyl chloride yielded 11
0.045 g, 20%), isolated as a pale brown oil. νmax (cap. film)/
D
2
2
for the deuterated solvents; DMF-d
THF-d (δ 1.80, δ 26.70).
7
(δ
H
C
2.90, δ 161.70) and
8
H
C
Mea su r em en t of th e Ra tes of Decom p osition . The NMR
tube containing zinc reagent 4 was placed in the spectrometer
and allowed to equilibrate at the temperature of the experi-
ment. The proportions of zinc reagent to elimination product
(
-
1
H
cm 3371, 2956, 1714, 1517, 1366, 1249, and 1097; δ 0.89 (3
H, t, J 7), 1.21-1.34 (4 H, m), 1.42 (9 H, s), 1.55 (2 H, quint,
J 7), 2.32 (2 H, t, J 7), 2.62 (1 H, dd, J 6.6 and 16), 2.68-2.72
were measured by comparison of the integrals of the CH
protons at approximately 0.3 ppm and the CH CO Me protons
2
ZnI
(2 H, m), and 2.83 (1 H, dd, J 4.7 and 17.3), 3.67 (3 H, s), 4.22-
2
2
4
2
1
.30 (1 H, m), and 5.34 (1 H, d, J 7.6); δ
C
13.84, 22.37, 22.54,
at approximately 3 ppm, since these signals did not overlap
with any others. Spectra were recorded at intervals of ap-
proximately 15 min for temperatures below 50 °C, and every
8.29, 31.23, 37.93, 43.19, 43.99, 45.53, 51. 65, 79.44, 155.00,
+
71.92, and 209.69; m/z (EI) 260 (42%, MH - C
4
H
8
), 216 (37,
+
MH - C
4
H
8
- CO
MH 316.2128, C16
.15 in CH Cl ).
Meth yl 3(R)-[(N-ter t-Bu toxyca r bon yl)a m in o]-5-oxo-5-
2
), 116 (66), 102 (56), and 57 (100); found
5
min for temperatures above 50 °C, until a minimum of two
+
26
H
30NO
5
requires 316.2151. [R]
D
+10.0 (c
half-lives had elapsed, and in most cases for a substantially
longer period.
2
2
2
P r ep a r a tion of 5 Oxo-â-a m in o Acid Der iva tives a n d
-Oxo-γ-a m in o Acid Der iva tives. Gen er a l P r oced u r e.
6
(2′-fu r yl)p en ta n oa te (12). Treatment with 2-furoyl chloride
yielded 12 (0.118 g, 51%), isolated as a pale brown oil. νmax
(KBr disc/cm ) 3365, 2978, 1735, and 1712; δ 1.35 (9H, s),
H
2.63 (1H, dd, J 6.4 and 16.2), 2.72 (1H, dd, J 4.8 and 16.5),
3.05 (1H, dd, J 6.4 and 16.5), 3.20 (1H, dd, J 4.6 and 16.2),
3.61 (3H, s), 4.32-4.39 (1H, m) 5.40 (1H, brd, J 8.0), 6.48 (1H,
dd, J 1.8, 3.5), 7.20 (1H, d, J 3.4), and 7.53-7.55 (1H, m); δ
28.18, 37.87, 41.59, 44.42, 51.58, 79.37, 112.27. 117.86. 146.70,
Zinc dust (0.142 g, 2.25 mmol) was weighed into a 50 mL
round-bottom flask with sidearm which was flushed with
nitrogen. Dry toluene (0.5 mL) and chlorotrimethylsilane (6µL,
-
1
0
.046 mmol) were added to the zinc dust and stirred for 30
min at room temperature. Iodide 4 or 5 (0.75 mmol) was
dissolved in dry toluene (0.5 mL) and DMA (0.2 mL) under
nitrogen. The iodide solution was transferred via syringe to
the zinc mixture, and stirring was continued at room temper-
ature. TLC (petroleum ether-ethyl acetate 2:1) showed com-
plete consumption of the starting material within 15 min. The
electrophile (1.0 mmol), palladium acetate (0.0112 g, 0.050
C
+
152.36, 154.92, 171.78, and 187.00; m/z (EI) 255 (61%, M
C H
4 8
-
), 238 (14), 211 (100) 195 (20), 116 (12), 95 (26), and 57
+
(57); found M - C
4
H
8
255.0743 C11
Cl ).
6
H13NO requires 255.0747;
[R]24
-10.2 (c 0.500 in CH
D
2
2

Decomposition of â-Amidozinc Reagents
J . Org. Chem., Vol. 65, No. 22, 2000 7421
20.55, 28.44, 29.46, 31.03, 43.65, 47.04,
Meth yl 4(S)-[(N-ter t-Bu toxyca r bon yl)a m in o]-6-oxo-6-
p h en ylh ep ta n oa te (13). Treatment with benzoyl chloride
yielded 13 (0.128 g, 51%), isolated as a pale brown solid, mp
(1H, brd, J 8.0); δ
C
51.85, 68.34, 79.69, 155.51, 170.32, 173.76, and 202.81; m/z
+
(EI) 331 (25%, M ), 288 (44), 230 (47) 215 (17), and 57 (100);
+
26
6
7-68 °C; found C, 64.4; H, 7.9; N, 4.1, C18
H
25NO
5
requires
C, 64.5; H, 7.5; N, 4.2%; νmax (KBr disc/cm ) 3361, 2982, 1733,
694, and 1526; δ
found M 331.1644, C15
(c 0.820 in CH Cl ).
H
25NO
7
requires 331.1631; [R]
D
-12.9
-1
2
2
1
H
1.41 (9H, s), 1.89-2.04 (2H, m), 2.43 (2H,
Meth yl 4(S)-[(N-ter t-Bu toxyca r bon yl)a m in o]-6-oxo-6-
(2′-fu r yl)h exa n oa te (16). Treatment with 2-furoyl chloride
yielded 16 (0.111 g, 45%), isolated as a pale brown oil. ν
dt, J 2.4 and 7.5), 3.15 (1H, dd, J 6.1 and 17.0), 3.34 (1H, dd,
J 3.7 and 17.1), 3.66 (3H, s), 4.03-4.11 (1H, m), 5.16-5.19
max
(1H, brd, J 8.2), 7.45-7.49 (2H, m), 7.56-7.59 (1H, m), and
(KBr disc/cm-1) 3362, 2977, 1735, 1711, and 1169; δ 1.35 (9H,
H
7
7
1
.93-7.96 (2H, m); δ
C
28.32, 29.38, 31.13, 42.83, 47.48, 51.66,
s), 1.83-1.87 (2H, m), 2.36 (2H, t, J 7.5), 2.93 (1H, dd, J 5.5
and 16.0), 3.10 (1H, dd, J 4.0 and 16.0), 3.60 (3H, s), 3.96-
4.01 (1H, m), 5.03 (1H, brd, J 8.2), 6.47 (1H, dd, J 1.5 and
9.29, 128.05, 128.66, 133.35, 136.83, 155.46, 173.79, and
+
4 8
99.73; m/z (EI) 279 (7%, M - C H ), 262 (6), 234 (34), 105
+
(
100), 77 (22), and 57 (79); found M - C
4
H
8
279.114, C14
H
17
-
3.4), 7.17 (1H, d, J 3.4), and 7.52-7.54 (1H, m); δ 28.29, 29.49,
C
2
6
NO
5
,requires 279.1107; [R]
D
-18.0 (c 0.850 in CH Cl ).
2
2
31.02, 42.82, 47.61, 51.64, 79.29, 112.35. 117.71. 146.64,
Meth yl 4(S)-[(N-ter t-Bu toxyca r bon yl)a m in o]-6-oxooct-
-en oa te (14). Treatment with acryloyl chloride yielded 14
152.62, 155.37, 173.69, and 187.47; m/z (EI) 326 (72%, MH
+
),
7
269 (10), 252 (9) 224 (63), 116 (33), 95 (42), and 57 (100); found
-
1
+
25
(
3
0.102 g, 48%), isolated as a brown oil. νmax (KBr disc/cm
)
MH 326.1603, C H NO requires 326.1595; [R]
-12.3 (c
1
6
24
6
D
364, 2977, 1737, 1711, and 1171; δ
H
1.44 (9H, s), 1.75-1.86
2 2
0.600 in CH Cl ).
(
1
(
2H, m), 2.33 (2H, dt, J 1.2 and 7.5), 2.71 (1H, dd, J 5.8 and
6.8), 2.85 (1H, dd, J 4.0 and 16.6), 3.61 (3H, s), 3.86-3.91
1H, m), 4.95 (1H, brd, J 8.2), 5.81 (1H, dd, J 0.9 and 10.4),
Ack n ow led gm en t. We thank the EPSRC (student-
ships to C.S.D. and C.H.), Merck Sharp and Dohme
CASE award to C.S.D), and SmithKline Beecham
(CASE award to C.H.) for support. We also thank Dr.
M.N.S. Hill for recording NMR spectra. We are grateful
to Dr. H. Maskill for valuable discussions concerning
the kinetic analysis.
6
1
1
.17 (1H, dd, J 1.0 and 17.4), and 6.27 (1H, dd, J 10.4 and
7.7); δ 28.42, 29.51, 31.12, 44.00, 47.36, 51.76, 79.43, 129.04,
(
C
36.70, 155.51, 173.83, and 199.25; m/z (EI) 229 (18%, M+
-
-
+
C
C
4
H
H
8
), 212 (44), 185 (100), 168 (48) and 154 (22); found M
2
2
4
8
229.0958, C10
Cl ).
H
15NO
5
requires 229.0950; [R]
D
-15.8 (c
1
.750 in CH
2
2
Meth yl 4(S)-[(N-ter t-Bu toxycar bon yl)am in o]-7-acetoxy-
-oxoh ep ta n oa te (15). Treatment with acetoxyacetyl chloride
6
Su p p or tin g In for m a tion Ava ila ble: Arrhenius plots for
yielded 15 (0.130 g, 52%), isolated as a pale brown solid, mp
3-84 °C. Found: C, 54.72; H, 7.78; N, 4.13:
requires C, 54.38; H, 7.55; N, 4.23); νmax (KBr disc/cm ) 3346,
984, 1731, 1370, 1228, and 1170; δ 1.35 (9H, s), 1.77-1.84
2H, m), 2.10 (3H, s), 2.32 (2H, t, J 7.4), 2.55-2.66 (2H, m),
.61 (3H, s), 3.84-3.91 (1H, m), 4.57 (2H, d, J 1.8), and 4.90
1
13
the decomposition of 4 in both THF and DMF, and H and
C
8
15 7
C H25NO
NMR spectra for all coupled products (8-16). This informa-
tion is also available free of charge via the Internet at
http://pubs.acs.org.
-
1
2
(
H
3
J O000558P