Amine-catalyzed coupling of aldehydes and ketenes derived from Fischer carbene complexes: Formation of β-lactones and enol ethers

Article abstract of DOI:10.1021/jo034476n

Aldehydes react with ketenes generated from photolysis of Fischer chromium carbene complexes to generate either β-lactones or enol ethers resulting from decarboxylation of β-lactones. The reaction is catalyzed by tertiary amines and can occur with diaster

Full text of DOI:10.1021/jo034476n

However, ketenes are usually not reactive enough to trap
efficiently most carbonyl compounds. Hence, a Lewis acid
catalyst is typically required to enhance the reactivity
of the carbonyl partner.6 Intriguingly, Lewis base cata-
lysts can also be utilized and, in pioneering work by
Wynberg,7 the cinchona alkaloid quinine and its pseu-
doenantiomer quinidine catalyzed enantioselective reac-
tions between ketene and aldehydes or ketones. This
method utilizes the enhanced nucleophilicity of the
enolate-like ketene-amine adduct to promote reaction
with the carbonyl partner (eq 2). However, additional
activation of the carbonyl reactant was found to be
extremely important since multiple electron withdrawing
groups, as in trichloroacetaldehyde, were required at the
R-position for productive reaction. Romo and co-workers
recently investigated intramolecular ketene-aldehyde
cyclizations in the hope that minimization of the entropic
barriers might aid the reaction.8 Using cinchona alkaloid
derived amines, chiral â-lactones were prepared with
enantiomeric excesses near or above 90%, but with
modest yields. In general, limitations on the carbonyl
reaction partner make this Lewis base methodology less
attractive than the Lewis acid-catalyzed versions.
Am in e-Ca ta lyzed Cou p lin g of Ald eh yd es
a n d Keten es Der ived fr om F isch er Ca r ben e
Com p lexes: F or m a tion of â-La cton es a n d
En ol Eth er s
Craig A. Merlic* and Brandon C. Doroh
Department of Chemistry and Biochemistry, University of
California, Los Angeles, 607 Charles Young Drive,
Los Angeles, California 90095-1569
merlic@chem.ucla.edu
Received April 15, 2003
Abstr a ct: Aldehydes react with ketenes generated from
photolysis of Fischer chromium carbene complexes to gener-
ate either â-lactones or enol ethers resulting from decar-
boxylation of â-lactones. The reaction is catalyzed by tertiary
amines and can occur with diastereoselectivity greater than
20:1 with DMAP as the catalyst.
Considerable effort has recently been directed toward
the diastereo- and enantioselective synthesis of â-lactones
(2-oxetanones).1,2 This is due in part to their presence in
biologically active natural products,3 but more impor-
tantly, the 2-oxetanone moiety displays a rich pattern of
reactivity. Apart from more common ester-type reactivity,
these species undergo several interesting reactions that
are atypical of esters and lactones.1 Also, â-lactones con-
tain a “masked aldol” connectivity. Therefore, diastereo-
selective â-lactone synthesis effectively constitutes a
means for producing either syn or anti aldol products
while at the same time obviating the necessity for enolate
or enol ether formation that is required in many current
selective aldol reactions.4 Despite their synthetic utility,
few effective diastereoselective methods exist for forma-
tion of â-lactones, and even fewer of these are enantiose-
lective.1b,2
The most versatile route to â-lactones is the [2 + 2]
cycloaddition between ketenes and carbonyl compounds,
which has been known for more than 90 years (eq 1).5
Fischer chromium carbene complexes are extremely
useful in organic synthesis owing to their versatile
reactivity.9 A particularly useful reactivity pattern is
direct insertion of a chromium-bound CO ligand into the
carbene moiety yielding a chromium-bound ketene spe-
cies (eq 3).10 Hegedus employed this reactivity advanta-
geously in â-lactam syntheses via the photoreaction of
Fischer chromium carbene complexes and imines.11 He-
gedus also reported that â-lactones can be prepared via
photolysis of Fischer carbene complexes and aldehydes
in the presence of a Lewis acid catalyst.12 The yields
obtained in the intermolecular reactions were low (<30%),
* Corresponding author.
(1) For reviews, see: (a) Pommier, A.; Pons, J .-M. Synthesis 1993,
441. (b) Yang, H.; Romo, D. Tetrahedron 1999, 55, 6403. (c) Orr, R.
K.; Calter, M. A. Tetrahedron 2003, 59, 3545.
(2) For recent syntheses of â-lactones, see: (a) Bodkin, J . A.;
Humphries, E. J .; McLeod, M. D. Tetrahedron Lett. 2003, 44, 2869.
(b) Castle, K.; Hau, C.-S.; Sweeney, J . B.; Tindall, C. Org. Lett. 2003,
5, 757. (c) Molnar, F.; Luinstra, G. A.; Allmendinger, M.; Rieger, B.
Chem.sEur. J . 2003, 9, 1273. (d) Schneider, C. Angew. Chem., Int.
Ed. 2002, 41, 744. (e) Cortez, G. S.; Oh, S. H.; Romo, D. Synthesis,
2001, 11, 1731. (f) Cortez, G. S.; Tennyson, R. L.; Romo, D. J . Am.
Chem. Soc 2001, 123, 7945. (g) Doyle, M. P.; May, E. J . Synlett, 2001,
967. (h) Evans, D. A.; J aney, J . M. Org. Lett. 2001, 3, 2125. (i) Nelson,
S. G.; Wan, Z. Org. Lett. 2000, 2, 1883. (j) Nelson, S. G.; Peelen, T. J .;
Wan, Z. J . J . Am. Chem. Soc. 1999, 121, 9742. (k) Caldwell, J . J .; Kerr,
W. J .; McKendry, S. Tetrahedron Lett. 1999, 40, 3485. (l) Yang, H. W.;
Romo, D. Tetrahedron Lett. 1998, 39, 2877. (m) Yang, H. W.; Romo,
D. J . Org. Chem. 1998, 63, 1344.
(3) Lowe, C.; Vederas, J . C. Org. Prep. Proced. Int. 1995, 27, 305.
(4) For selected reviews, see: (a) Machajewski, T. D.; Wong, C.-H.
Angew. Chem., Int. Ed. 2000, 39, 1352. (b) Nelson, S. G. Tetrahedron:
Asymmetry 1998, 9, 357. (c) Franklin, A. S.; Paterson, I. Cont. Org.
Synth. 1994, 1, 317. (d) Heathcock, C. H. In Comprehensive Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, 1991;
Vol. 2, Chapter 1.6, pp 181-238. (e) Kim, B. M.; Williams, S. F.;
Masamune, S. In Comprehensive Organic Synthesis, Trost, B. M.,
Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 2, Chapter 1.7, pp
239-276. (f) Paterson, I. In Comprehensive Organic Synthesis; Trost,
B. M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 2, Chapter 1.9,
pp 301-320.
(5) Staudinger, H.; Bereza, S. Ann. 1911, 380, 243.
(6) (a) Kung, F. E. U.S. Patent 2 356 459, 1944. (b) Zaugg, H. E.
Org. React. 1954, 8, 314.
(7) Wynberg, H.; Staring, E. G. J . J . Org. Chem. 1985, 50, 1977.
(8) Cortez, G. S.; Tennyson, R. L.; Romo, D. J . Am. Chem. Soc. 2001,
123, 7945.
(9) For reviews, see: (a) de Meijere, A.; Schirmer, H.; Duetsch, M.
Angew. Chem., Int. Ed. 2000, 39, 3964. (b) Herndon, J . W. Coord.
Chem. Rev. 2000, 206, 237. (c) Do¨rwald, F. Z. Metal Carbenes in
Organic Synthesis; Wiley-VCH: New York, 1999. (d) Do¨tz, K. H.;
Tomuschat, P. Chem. Soc. Rev. 1999, 28, 187-198.
(10) Hegedus, L. S.; de Weck, G.; D’Andrea, S. J . Am. Chem. Soc.
1988, 110, 2122.
(11) Hegedus, L. S.; McGuire, M. A.; Schultze, L. M.; Yijun, C.;
Anderson, O. P. J . Am. Chem. Soc. 1984, 106, 2680.
(12) Colson, P.-J .; Hegedus, L. S. J . Org. Chem. 1994, 59, 4972.
10.1021/jo034476n CCC: $25.00 © 2003 American Chemical Society
Published on Web 06/26/2003
6056
J . Org. Chem. 2003, 68, 6056-6059

TABLE 1. Effica cy of Lew is Ba se Ca ta lysts
TABLE 2. Resu lts of Keten e-Ald eh yd e Cou p lin g
Lewis base
yield (%)
dr (3a /3a ′)
none
PBu3
0
0
cinchonidine
pyridine
DBU
DABCO
DMAP
0
<5
15
40
53
1/1
1.6/1
2/1
yield of 3 (%) yield of 4 (%)
aldehyde
product
(syn/anti)
(E/Z)
15/1
propionaldehyde
acetaldehyde
n-butyraldehyde
isobutyraldehyde
benzaldehyde
3a /3a ′
3b/b′
3c/c′
3d /d ′
3e/e′
4f/f ′
4g/g′
4h /h ′
0
53 (15/1)
55 (8/1)
0
0
43 (23/1)
35 (4/1)
0
0
and the diastereoselectivity was less than 3:1 in all cases.
We reinvestigated this coupling of aldehydes and photo-
generated ketenes and report herein our results of an
amine base-catalyzed version of this reaction that can
be highly stereoselective.
33 (15/1)
0
0
0
<5
0
0
2-furfural
52 (3/1)
61 (3/1)
43 (1/1)
0
p-anisaldehyde
3-methyl-2-butenal
methyl 4-formylbenzoate
bromal (Br3CCHO)
0
0
4f and 4f ′ favoring the Z isomer as determined by a
NOESY spectrum. The 3:1 mixture of olefin isomers was
at first alarming; however, the enol ether product, like
stilbene,14 is likely to isomerize under photochemical
conditions leading to a mixture of olefin isomers that is
not necessarily representative of the initial â-lactone
isomeric ratio. Reaction with p-anisaldehyde again yielded
the olefin product resulting from decarboxylation. As with
the furfural reaction, the ratio of olefin isomers 4g and
4g′ was 3:1 in favor of the Z olefin and the yield exceeded
60%. 3-Methyl-2-butenal gave similar results.
It was surprising that decarboxylation took place with
electron-rich aryl or unsaturated aldehydes. Substitution
at the C3 position in â-lactones is extremely important
to the rate of thermal decarboxylation.15 Substituents
that donate electron density increase the rate of decar-
boxylation, since they are better able to stabilize emerg-
ing carbocation character at C3. Nevertheless, similar
compounds are known to be stable at room temperature,
so two other factors can be considered.16 First, chromium-
(III) salts are Lewis acidic, which could increase the rate
of decarboxylation if some decomposition occurred during
the reaction. Second, it is also possible that the decar-
boxylation is facilitated by the photolytic conditions with
these unsaturated systems.
We began by exploring the reaction of phenyl-methoxy
Fischer chromium carbene complex 1 and propionalde-
hyde (Table 1). Photolysis of 1 in THF with propional-
dehyde under CO pressure did not yield the desired â-
lactone. In the presence of catalytic DABCO (12%), the
results proved encouraging as the reaction produced the
desired â-lactone in 40% yield, albeit as a 2:1 mixture of
diastereomers. Screening of other potential bases initially
failed to discover another that was as effective. Especially
disappointing was that cinchonidine was ineffective, since
cinchona alkaloids catalyze coupling with highly acti-
vated aldehydes and often with excellent enantioselec-
tivity.7,13 Finally, we found that DMAP catalyzed the
reaction, yielding 3a /a ′ in 53% yield. Even more impres-
sive was that DMAP afforded excellent selectivity, yield-
ing a 15:1 diastereomeric ratio. THF and toluene were
both effective solvents as the yield and selectivity were
not appreciably different between the two. However,
reactions in either acetonitrile or methylene chloride were
not productive.
The DMAP-catalyzed reaction of 1 with benzaldehyde
also yielded excellent 15:1 diastereoselectivity (Table 2).
Acetaldehyde, n-butyraldehyde, and isobutyraldehyde all
reacted diastereoselectively with complex 1 to yield
â-lactones 3b/b′, 3c/c′, and 3d /d ′. The yields were con-
sistently modest while the diastereoselectivities were
good to excellent. The reaction with isobutyraldehyde also
produced some of the dimerized ketene (vide infra).
Since typical Wynberg lactonization7 requires electron-
deficient aldehydes, we sought to determine if such al-
dehydes were better substrates. However, bromal and p-
chlorobenzaldehyde were ineffective, and methyl 4-formyl-
benzoate gave a low yield. Perplexed by this failure of
electrophilic aldehydes to participate, more electron-rich
aldehydes were examined. Surprisingly, a coupling reac-
tion proceeded smoothly with 2-furfural, but product
characterization revealed it to be the olefin resulting from
decarboxylation. The olefin was a 3:1 mixture of isomers
A competing reaction is dimerization of the chromium-
bound ketene in the absence of aldehyde (eq 6). The yield
is modest (19%), but the product is a single olefin isomer.
The low yield is expected due to the low concentration of
photogenerated ketene complex. This result is interesting
in that Hegedus reported that chromium carbene-derived
ketenes do not dimerize under photolytic conditions, even
in the presence of a strong Lewis acid catalyst.10
Understanding the diastereoselectivity required estab-
lishing the relative stereochemistry at C2 and C3.
NOESY spectroscopy experiments showed only correla-
tions between the phenyl and methoxy protons and
(13) Calter, M. A. J . Org. Chem. 1996, 61, 8006.
J . Org. Chem, Vol. 68, No. 15, 2003 6057

TABLE 3. 1H NMR Sh ifts for â-La cton es
â- δ OMe δ H-C3
boxylation. Satisfyingly, the resulting enol ethers (4a and
4a ′) had the same 19:1 ratio of alkene isomers indicating
that no isomerization took place during the conversion
(eq 7). Compounds 4a and 4a ′ are known in the litera-
ture,19 and identification of the major isomer 4a was
confirmed by comparing the chemical shifts of the vinyl
protons.20 That 4a was indeed the Z-enol ether was
satisfying, as it confirmed the analysis of the â-lactone
NMR data. The stereochemistries of the remaining C3
alkyl substituted â-lactones were assigned in analogy to
3a and 3a ′ due to the similarities in chemical shifts
described above.
(maj,min)
lactones (maj,min) ∆δ (maj,min) ∆δ
other H
∆δ
3a /a ′ 3.37, 3.33, +0.04 4.60, 4.36, C4 methylene (1.28, 1.18)a
+0.24
3b/b′ 3.36, 3.35, +0.01 4.87, 4.63, C4 methyl
+0.24
1.99, -0.76b
1.04, 1.18,
-0.14
3c/c′
3.38, 3.35, +0.03 4.70, 4.45,
+0.25
3d /d ′ 3.45, 3.37, +0.08 4.27, 3.99, C4 methine
1.42, 2.34,
-0.92
+0.28
3e/e′ 3.44, 3.19, +0.25 5.76, 5.53,
+0.23
a
b
Shifts of diastereotopic protons. Calculated from the mean
value of the diastereotopic protons.
between the methine proton and the ethyl substituent.
There is a literature report17 that the relative stereo-
chemistry of 3-methoxy-4-phenyl â-lactams can be elu-
cidated by observing the chemical shift of the methoxy
singlets. When the two groups are syn, the methoxy
singlet is shifted upfield to ∼3.0 ppm (instead of the usual
∼3.3 ppm) due to the methoxy group residing in the
phenyl ring shielding cone. The methoxy singlet of the
major diastereomer in benzaldehyde adduct 3e occurs at
1
3.44 ppm in the H NMR spectrum, while the analogous
signal in the minor diastereomer occurs at 3.19 ppm,
suggesting that the two phenyl groups are syn in the
major isomer. Since the â-lactones described herein have
a phenyl substituent at the C2 position, it is logical that
a similar result might be seen for alkyl substituents syn
at C3. Table 3 lists the proton chemical shift values and
differences between diagnostic protons for several â-lac-
tones and the stereochemical assignments. The data
suggest that the methoxy singlet is an excellent marker
for determining the relative stereochemistry of â-lactones
formed from aromatic aldehydes. However, for those
derived from aliphatic aldehydes the methine proton at
C3 has greater utility. In all cases, the methine proton
is shifted downfield by approximately 0.25 ppm in the
major isomer. Also useful are the signals for protons at
C4 (originating from the aldehyde R-position). Though
the magnitudes of the shift differences for these protons
are not as consistent as those of the C3 methine protons,
in all cases the C4 protons are shifted upfield for the
major isomer. Further support for a syn relationship
between the phenyl and alkyl substituents is the obser-
vation that the diastereotopic methylene protons in major
diastereomer 3a and the diastereotopic methyl groups
in major diastereomer 3d show quite different chemical
shifts, whereas in the minor diastereomers they appear
as essentially one signal.
Having defined the reaction to be diastereoselective for
carbon substituents cis on the â-lactone ring and most
efficient with aliphatic aldehydes, we turned to the
question of diastereoselection with a chiral aldehyde.
Reaction with (R)-glyceraldehyde acetonide led to a 1.3:1
mixture of cis diastereomers 7a /b in a 65% yield (eq 8).
Surpassing all the previous examples, there was less than
1% of the trans diastereomers detected by 1H NMR using
the trends delineated above. The identity of the major
diastereomer 7a was established by conversion to γ-lac-
tone 8 via cleavage of the acetonide group in acidic
methanol and trans-lactonization promoted by silica gel
(eq 9). The stereochemistry of 8 was established by
analysis of coupling constants and a NOESY spectrum.
Aldol-type additions to (R)-glyceraldehyde often occur
with high levels of substrate control,21 so the lack of
Felkin-Anh diastereoselectivity for this â-lactone forma-
tion is unusual, but the level of cis diastereoselectivity
was remarkable nonetheless.
Conclusive chemical evidence was sought to confirm
these spectral assignments. The stereochemistry of 3a
was thus investigated by examining the geometry of the
enol ether resulting from thermal decarboxylation.18
Passing a 19:1 mixture of 3a and 3a ′ (in benzene) through
a quartz tube packed with glass beads at 420 °C under
reduced pressure (100 mmHg) gave quantitative decar-
(14) Fleming, G. R. Annu. Rev. Phys. Chem. 1986, 37, 81.
(15) Krabbenhoft, H. O. J . Org. Chem. 1978, 43, 1305.
(16) Mulzer, J .; Zippel, M. J . Chem. Soc., Chem. Commun. 1981,
891.
(17) Bose, A. K.; Lal, B.; Dayal, B.; Manhas, M. S. Tetrahedron Lett.
1974, 15, 2633.
Definitive explanations for the reaction stereoselectiv-
ity and high reactivity with electron-rich aldehydes re-
(19) Casey, C. P.; Bertz, S. H.; Burkhardt, T. J . Tetrahedron Lett.
1973, 14, 1421.
(18) Adam, W.; Baeza, J .; Liu, J .-C. J . Am. Chem. Soc. 1972, 94,
2000.
(20) Fahey, R. C.; Schubert, C. J . Am. Chem. Soc. 1965, 87, 5172.
(21) J urczak, J .; Pikul, S.; Bauer, T. Tetrahedron 1986, 42, 447.
6058 J . Org. Chem., Vol. 68, No. 15, 2003

2972, 2835, 1822, 1495, 1450, 1068, 858 cm-1. MS (CI): 162.1
(M - CO2)+, 147.1 (45), 129.1 (15), 121.1 (41), 111.1 (27), 105.0
(100). HRMS: mass calcd for C11H14O (M - CO2)+ 162.1045,
SCHEME 1
1
found 162.1049. Partial data for minor isomer. H NMR (minor
diastereomer, 400 MHz): δ 7.42-7.33 (m, 5H), 4.36 (t, 1H, J )
6.7 Hz), 3.33 (s, 3H), 2.0-1.97 (m, 2H), 1.06 (t, 3H, J ) 7.4 Hz).
2-Eth yl-1-m eth oxy-1-p h en yleth en e (4a , 4a ′). A quartz
tube, packed with glass beads and equipped with a cold trap of
dry ice/acetone and then a vacuum pump, was heated to 420 °C
under reduced pressure (100 mmHg). 4-Ethyl-3-methoxy-3-
phenyloxetan-2-one 19:1 (3a /3a ′) (50 mg) in benzene (1 mL) was
injected through a septa into tube, and product was collected in
the trap. After removal of the solvent under reduced pressure
2-ethyl-1-methoxy-1-phenylethene was isolated in 99% yield in
a 19:1 mixture of olefin isomers. Assignment of the major isomer
was made by correlation of the spectral data with the reported
literature values.18,19
(Z)- a n d (E)-1-Meth oxy-1-p h en yl-2-(2-fu r yl)eth en e (4f/
4f′). 2-Furfuraldehyde (1.32 mL, 16.0 mmol), DMAP (60 mg, 0.5
mmol), and pentacarbonyl[(methoxy)(phenyl)methylene]chrom-
ium(0) (1.25 g, 4.10 mmol) in THF (35 mL) were photolyzed
following the general procedure. The crude reaction mixture was
then concentrated and chromatographed on silica gel (99:1
hexanes/EtOAc) to yield 282 mg (52%) of 1-methoxy-1-phenyl-
2-(2-furyl)ethene as a colorless oil as a 2.6:1 mixture of Z/E
isomers, as evidenced by methoxy signals at 3.81 and 3.69 ppm
in the 1H NMR, inseparable by chromatography. 1H NMR: (Z
isomer) δ 7.55-7.49 (m, 2H), 7.41-7.29 (m, 4H), 6.72 (d, 1H, J
) 3.4 Hz), 6.45 (ddd, 1H, J ) 3.33, 1.83, 0.57 Hz), 6.20 (s, 1H),
3.69 (s, 3H); (E isomer) 7.41-7.29 (m, 5H), 7.17 (d, 1H, J ) 1.82
Hz), 6.16 (dd, 1 H, J ) 3.33, 1.84 Hz), 5.70 (s, 1H), 5.52 (d, 1H,
J ) 3.33 Hz), 3.79 (s, 3 H). 13C NMR (mixture, 125 MHz): δ
157.6, 154.5, 151.7, 151.3, 140.6, 140.0, 136.2, 135.1, 128.8, 128.7,
128.5, 128.3, 128.2, 126.2, 111.7, 110.9, 108.5, 105.2, 102.9, 92.0,
57.7, 55.6. IR (neat): mixture 3057, 2936, 2837, 1639, 1487,
1236, 1088, 1012, 773 cm-1. MS (mixture, EI): 200 (M+, 15),
161 (100), 134.1 (36), 115 (27), 105.0 (85). HRMS: mass calcd
for C12H12O2 200.0837, found 200.0834.
main elusive. The diastereoselectivity is dependent on
the detailed geometry of reaction with the aldehyde. How-
ever, the reactivity patterns of the aldehyde partner sug-
gest there are additional aspects of the reaction beyond
the simplistic mechanism outlined in eq 2. One possible
rationale that contains several unique aspects is illus-
trated in Scheme 1. First, the chromium tetracarbonyl
moiety may remain coordinated to the ketene unit after
addition of DMAP. Second, addition leads to the adduct
with the methoxy group cis to the ammonium group due
to favorable electronic interactions. Third, only electron-
rich aldehydes displace solvent to coordinate to the weak-
ly Lewis acidic chromium. Fourth, only aldehydes within
the coordination sphere of the chromium react with the
bound enolate that is expected to be otherwise fairly
unreactive. Finally, reaction via a geometry that places
the aldehyde substituent in a sterically favorable position
leads to the correct final observed stereochemistry.
In conclusion, we have discovered an amine-catalyzed
reaction between in situ formed chromium-bound ketenes
and aldehydes. This reaction is particularly interesting
since it can be highly diastereoselective and electron-rich
aldehydes are required as the reaction partner. This
heretofore unreported reactivity suggests that this reac-
tion may be viewed as a complementary extension to the
Wynberg â-lactone synthesis.
(E)-3-Meth oxy-4-[(m eth oxy)ben zyliden e]-3-ph en yloxetan -
2-on e (5). DMAP (47 mg, 0.39 mmol) and pentacarbonyl-
[(methoxy)(phenyl) methylene]chromium(0) (1.0 g, 3.2 mmol) in
THF (35 mL) were photolyzed following the general procedure.
The crude reaction mixture was then concentrated and chro-
matographed on silica gel (99:1 hexanes/EtOAc) to yield 90 mg
(19%) of the ketene dimer as a colorless oil and a single olefin
isomer. 1H NMR (400 MHz): δ 7.73-7.70 (m, 4 H), 7.49-7.42
(m, 6H), 3.74 (s, 3H), 3.58 (s, 3H). 13C NMR (100 MHz): δ 168.2,
138.8, 137.9, 133.5, 131.1, 129.6, 128.8, 128.8, 128.5, 126.9, 126.2,
96.9, 59.3, 54.8. IR (neat): 3061, 2936, 2833, 1875, 1495, 1450,
1188, 993, 696 cm-1. MS (EI): 296 (M+, 5), 282.1 (61), 269.1
(100), 265.1 (90), 253.1 (84), 236.1 (44), 225 (70), 197 (71), 161
(39). HRMS (EI): mass calcd for C18H16O4 296.1049, found
296.1048.
Exp er im en ta l Section
cis- a n d tr a n s-4-Eth yl-3-m eth oxy-3-p h en yloxeta n -2-on e
(3a , 3a ′). Gen er a l P h otolysis P r oced u r e. Propionaldehyde
(1.2 mL, 16.0 mmol) and DMAP (60 mg, 0.5 mmol) were added
to a solution of pentacarbonyl[(methoxy)(phenyl) methylene]-
chromium(0) (1.25 g, 4.10 mmol) in THF (35 mL) in a Pyrex
pressure tube. The tube was pressurized and purged with carbon
monoxide (5 cycles) and then pressurized (30 psi). The tube was
irradiated with a 450 W medium-pressure Hg lamp water cooled
in a quartz immersion well for 20 h. The crude reaction mixture
was then concentrated and chromatographed on silica gel (99:1
hexanes/EtOAc) to yield 435 mg (53%) of 4-ethyl-3-methoxy-3-
phenyloxetan-2-one as a colorless oil as a 15:1 mixture of
diastereomers as evidenced by methoxy resonances at 3.37 and
3.33 ppm in the 1H NMR spectrum. 1H NMR (major diastere-
omer, 400 MHz): δ 7.42-7.33 (m, 5H), 4.62 (dd, 1H, J ) 9.4,
4.7 Hz), 3.37 (s, 3H), 1.29-1.17 (m, 2H), 0.84 (t, 3H, J ) 7.4
Hz). 13C NMR (major diastereomer, 100 MHz): δ 169.0, 131.5,
129.5, 128.8, 127.5, 94.4, 86.3, 54.0, 24.4, 9.3. IR (neat): 3063,
Ack n ow led gm en t. This work relied upon instru-
mentation supported by the National Science Founda-
tion under Grant Nos. CHE-9974928 and CHE-9808175.
We thank Carina Cannizzaro for helpful mechanistic
discussions.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and spectral data for compounds 3b/b′, 3c/c′, 3d /d ′,
3e/e′, 4g/g′, 4h /h ′, 7a /b, and 8 and copies of 1H NMR spectra
for compounds 3-5, 7, and 8. This material is available free
of charge via the Internet at http://pubs.acs.org.
J O034476N
J . Org. Chem, Vol. 68, No. 15, 2003 6059