Regio- and Stereoselectivity of Water Elimination as a Function of Ring Size

Article abstract of DOI:10.1021/jo970989g

The elimination of water from tertiary alcohols was studied for carbocycles of ring size 5-16. The resulting olefins 3-5 were discriminated by NMR spectroscopy. The relative amount of their formation reflects the steric idiosyncrasies of the respective ri

Full text of DOI:10.1021/jo970989g

8058
J . Org. Chem. 1997, 62, 8058-8062
Regio- a n d Ster eoselectivity of Wa ter Elim in a tion a s a F u n ction of
Rin g Size
Bjo¨rn Greve and Peter Imming*
Institut fu¨r Pharmazeutische Chemie der Philipps-Universita¨t, Marbacher Weg 6,
35032 Marburg, Germany
Received J une 3, 1997X
The elimination of water from tertiary alcohols was studied for carbocycles of ring size 5-16. The
resulting olefins 3-5 were discriminated by NMR spectroscopy. The relative amount of their
formation reflects the steric idiosyncrasies of the respective ring systems. The behavior of the
medium-sized rings yielded experimental proof to the I-strain concept.
Sch em e 1. P r ep a r a tion of th e Hom ologou s
Mon osu bstitu ted Cycloa lk en es 3-5 fr om th e
Ter tia r y Alcoh ols 1 (n ) 4-15)
In tr od u ction
We recently developed a new class of potent cyclooxy-
genase and lipoxygenase inhibitors1 as potential dual-
action antirheumatics.2 Their preparation involved the
synthesis of a homologous series of monosubstituted
cycloalkenes 3-5 from the tertiary alcohols 1 (Scheme
1). We herein report the influence of ring size on the
position and stereochemistry of the double bond in 3-5.
This is the first study on elimination regioselectivity
covering normal, medium, and large rings.
Resu lts a n d Discu ssion
Mech a n ism of th e Rea ction a n d P r od u ct Sta bili-
ties. The elimination of water from the cycloalkanols 1
was performed in refluxing toluene in a Dean-Stark
apparatus, adding a catalytic amount of 4-toluenesulfonic
acid. The reaction was run under identical conditions
for all ring systems investigated (amount of substrate,
time, etc.). After a short workup (see Experimental
Section), the raw product was isolated in a yield of
carbon, as can be visualized by models. The endocyclic
olefins 4 and 5 lack the advantages 3 has, but with some
ring systems (vide infra), the positioning of two sp2
centers within the ring reduces steric interactions. As
expected, the formation of the trans6 olefins 5 was not
observed up to a ring size of eight.
1
approximately 90% and examined by H and 13C NMR
Assign m en t of NMR sign a ls. The discrimination of
the isomeric olefins 3-5 rests on the following data. Only
the endocyclic alkenes 4 and 5 have a proton attached to
an sp2 center, giving rise to a triplet at approximately
5.1 ppm. The assignment of these signals to 4 or 5 by
the increment method7 was not possible since the chemi-
cal shift difference was small, preventing a decision.
However, the presence or absence of one or two olefinic
proton resonances showed the formation of 4 and 5.
spectroscopy.
Following the literature,3 we assume that the reaction
proceeds by an E1 mechanism. Consequently, the ab-
straction of a proton from the intermediate carbenium
ion 2 will determine which of the regioisomers 3-5 will
be formed. Statistics would predict a 4:1 ratio in favor
of the endocyclic isomers, but in such cases the direction
the new double bond takes is determined almost entirely
by the relative stabilities of the possible olefins.4 The
following factors influence the relative stabilities of the
cycloalkenes 3-5. In the semicyclic5 isomers 3, the
double bond has gone to the most highly substituted
carbon according to Zaitsev’s rule, and it is also in
conjugation with the ester carbonyl group. These ad-
vantages are counterbalanced, for some ring sizes, by
steric interactions of the methyl and ester groups with
the ring methylene groups attached to the ring sp2
The identification of 3 was straightforwardly possible
by 13C NMR. Due to the conjugation of double bond and
ester group, the carbonyl and R-carbon atoms were
shifted to high field (to approximately 170 and 120 ppm)
and the â-carbon to low field (to approximately 150 ppm)
as compared to 4 and 5 with isolated double bonds.
According to the literature on 13C NMR spectroscopy,8
sp2 centers of cis-configured alkenes usually show a small
high field shift as compared to their trans isomers. In
the cycloalkenes studied by us, we did not find this trend,
but one of the sp2 signals of one of the isomers was at
higher and the other signal at lower field than the signals
X Abstract published in Advance ACS Abstracts, November 1, 1997.
(1) Greve, B.; Imming, P.; Laufer, S. Angew. Chem. 1996, 108, 1312-
1314 (Angew. Chem., Int. Ed. Engl. 1996, 35, 1221-1223).
(2) Laufer, S. A.; Augustin, J .; Dannhardt, G.; Kiefer, W. J . Med.
Chem. 1994, 37, 1894-1897.
(3) Kno¨zinger, H. In The Chemistry of the Hydroxyl Group; Patai,
S., Ed.; Interscience Publishers: New York, 1971; pp 641-718.
(4) (a) Hu¨ckel, W.; Hanack, M. Angew. Chem. 1967, 79, 555-565
(Angew. Chem., Int. Ed. Engl. 1967, 6, 534-544). (b) March, J .
Advanced Organic Chemistry; Wiley: New York, 1992; p 998.
(5) We use the term “semicyclic” to denote a double bond shared by
a ring and an exocyclic atom, as distinguished from a fully “exocyclic”
double bond.
(6) We use the more familiar cis/trans nomenclature to designate
the stereochemistry of the (unsubstituted) rings themselves. Trans
corresponds to Z with our 1-substituted cycloalkenes.
(7) Friedrich, E. C.; Runkle, K. G. J . Chem. Educ. 1984, 61, 830-
831.
(8) Kalinowsky, H.-O.; Berger, S.; Braun, S. 13C NMR-Spektroskopie,
Thieme: Stuttgart, 1984; pp 118-119.
S0022-3263(97)00989-4 CCC: $14.00 © 1997 American Chemical Society

Regio- and Stereoselectivity of Water Elimination
J . Org. Chem., Vol. 62, No. 23, 1997 8059
Ta ble 1. Am ou n t of Isom er s a s a F u n ction of Rin g size
systems studied. The ratios were determined on the
basis of the integrals of the olefinic, methine, and ester
methylene proton signals and the relative height of the
olefinic carbon signals. With the exception of the cyclo-
hexenyl derivative that gave rise to 100% endocyclic
isomer 4, we observed a steady decline of the proportion
of the semicyclic isomers 3 on going from the 5- to the
10-membered system. The latter formed the endocyclic
E-isomer 4 exclusively. The proportion of 3 then rises
steadily, at the 15-membered system reaching the value
of approximately 30% it had with the 5-membered ring.
ring size
semi (3)
endo E (4)
endo Z (5)
5
6
7
34
0
20
8
66
100
80
0
0
0
8
92
0
9
2
96
2
10
11
12
13
14*
15
16
0
1
1
4
15
28
7
100
89
86
68
60
0
10
13
28
25
18
20
54
73
As expected, for the 5-7-membered systems we did not
detect any trans isomers 5 (Z for our system). This was
first observed in 2% amount for the cyclononenyl deriva-
tive, but was not detectable in the next higher homolog,
then steadily rising to a proportion of 20-25% for the
large ring systems. The endocyclic E isomer 4 always
was the main product, in some cases being formed with
high regio- and stereoselectivity. Data for the cyclotet-
radecenyl system were not determined experimentally,
but calculated by extrapolation. The calculated numbers
of course have to be considered with care since the
relation of the isomers may not change linearly, but
follow an oscillating behavior (cf. the deviation from the
trend we found for the 16-membered ring, Figure 1). Such
a behavior has long been known e.g. for the melting
points of homologous fatty acids and the yield of ketones
from the cyclization of dinitriles.11
* Data for cyclotetradecenyl system not determined experimen-
tally, but calculated by extrapolation.
Discu ssion of th e Da ta a n d Rela tion to th e Lit-
er a tu r e. The reluctance of six-membered rings to ac-
commodate semicyclic double bonds can be demonstrated
by several literature examples, e.g.: elimination reac-
tions;12 the fact that the enol content of cyclohexanone
is about 25 times that of cyclopentanone and cyclohep-
tanone;13 and the finding that in the cycloalkanone series,
the reactivity of cyclohexanone is surpassed by cyclopro-
panone and cyclobutanone only.14 This tendency has
been attributed to unfavorable interactions of especially
ring hydrogen atoms with the carbonyl oxygen and also
to the ecliptic positioning of protons in the ring.15
F igu r e 1. Ratio of isomeric cycloalkenes 3 (semicyclic), 4
(endocyclic, E), and 5 (endocyclic, Z) formed during the
elimination reaction.
of the isomer. The cycloalkenes studied also did not show
the larger shift difference of the sp2 signals reported9 for
cyclic trans in comparison to cis olefins.
We found the methine carbon signals to be an unam-
biguous probe of stereochemistry. In trans alkenes, the
signals of the R-carbon atoms are shifted to lower field
relative to their cis isomers10 (R-C in 6-Z and 6-E). Due
We explain as follows the low or vanishing proportion
of semicyclic olefins 3 in the medium-ring derivatives.
Medium rings are strained because of transannular
interactions (Prelog strain) and too large bond angles
(Baeyer strain).16 Thus it is advantageous for medium-
sized rings to have sp2 centers in the ring, relieving the
strain because they have one substituent less and can
accommodate larger bond angles. Since the 8-12-
membered rings predominantly formed the endocyclic
olefins 4 and 5, the introduction of two sp2 centers in the
ring was more favorable than the conjugation with an
ester group. (The latter would have been regarded more
favorable at first glance, assuming Zaitsev’s rule and
normal double-bond preferences17 to direct product for-
mation.) Typical medium-ring strain effects vanishing
to the large number of signals, we could not identify the
relative positions of the R-ring carbon signals. But the
R-carbon of the propionate side chain resonated well off
the ring atoms, allowing assignments and quantitative
comparison of the 6-Z (R3C, approximately 40 ppm) and
6-E (R3C, approximately 45 ppm) isomers. Further
corroboration stems from comparison with the NMR data
of the 5-7-membered cycloalkene products that cannot
form 5 with a trans double bond.
(11) Ziegler, K.; Hechelhammer, W. Liebigs Ann. Chem. 1937, 528,
114-142.
(12) Askani, R. Methods in Organic Chemistry (Houben/ Weyl);
Thieme: Stuttgart, 1972; Vol. 5/1b, pp 63-70.
(13) (a) Toullec, J . Tetrahedron Lett. 1984, 25, 4401-4402. (b) Keeffe,
J . R.; Kresge, A. J .; Schepp, N. P. J . Am. Chem. Soc. 1990, 112, 4862-
4868.
Ra tio of isom er s. The Table 1 and Figure 1 sum-
marize the relative ratio of the isomers 3-5 for the ring
(14) Prelog, V.; Kobelt, M. Helv. Chim. Acta 1949, 32, 1187-1192.
(15) For a discussion, see Allinger, N. L.; Tribble, M. T.; Miller, M.
A. Tetrahedron 1972, 28, 1173-1190.
(16) Eliel, E. H.; Wilen, S. H.; Mander, L. N. Stereochemistry of
Organic Compounds; Wiley: New York, 1994; pp 762-771.
(17) Hine, J .; Skoglund, M. J . J . Org. Chem. 1982, 47, 4766-4770.
(9) Ibid. pp 267-269.
(10) Ibid. p 119.

8060 J . Org. Chem., Vol. 62, No. 23, 1997
Greve and Imming
Sch em e 2. Th e Str a in En er gies of th e La cton es 7
Wer e Ca lcu la ted on th e Ba sis of Th is Isod esm ic
Rea ction 21
with increasing ring size, the proportion of the semicyclic
conjugated Zaitsev product 3 again rose for the large
rings.
This was not taken into account in a paper18 that dealt
with the elimination of water from ethyl 1-hydroxy-1-
cycloalkyl acetates (ring size 5-10) under acid catalysis.
The authors, without commenting on the possibility of
E/Z-isomers, noted an increase of the amount of the endo-
(cyclic) olefin, corresponding to 3. They did not offer an
explanation for this, but predicted that “in the case of a
C12 or larger ring, dehydration will afford just one
product, the endo one”. Our study proves this prediction
not to be true since one has to take into account the
different factors operating with normal, medium, and
large rings.
Our data can also be interpreted in terms of Brown’s
internal (I-)strain concept, furnishing experimental proof
for a concept that was supported theoretically by force
field analyses some years ago.19 In the original render-
ing, “I-strain is that change in internal strain of a ring
compound which results from a change in the coordina-
tion number (and the preferred bond angle) of a ring atom
involved in the reaction”.20 Two ring atoms were involved
in the reaction leading to the endocyclic olefins 4 and 5.
Both changed from coordination number 4 to 3, and from
a preferred bond angle of approximately 109° to 120°. The
outcome of the elimination verified a reduction in I-strain
by this change for the medium-sized rings.
The product ratios highlight the sensitivity of the
quantitative outcome of a reaction to the stereochemical
demands of the ring involved. These data complement
well the findings of a recent thermodynamic study on the
strain energies of lactones 7 as a function of ring size.21
The strain energies were calculated from the isodesmic
reaction shown in Scheme 2, the enthalpies of formation
being derived from the enthalpies of reduction of 7. The
shape of the curve (Figure 2, based on the authors’ data)
reflects the same trend we extracted from the product
pattern changes in the elimination reactions. They noted
a relatively high strain energy of the six-membered
lactone, accompanied by a second strain maximum in the
region of the medium rings with the eight-membered ring
as the extreme. This also exemplifies that in the case of
carbocycles with a semicyclic double bond, the six-
membered ring is more akin to its medium-sized conge-
ners.22 In contrast, the well-known relative stabilities
of saturated cycloalkanes, based on combustion enthal-
F igu r e 2. Strain energies of lactones 7 as a function of ring
size (from data in lit.,21 Table 5).
pies and comparison with open-chain alkanes,23 show
strain maxima for the small and medium rings, and
minimal strain for cyclohexane.
Numerically, the differences in our study were more
pronounced than in the lactone series and provided more
insight into the reactivity changes with increasing ring
size.
Con clu sion
The elimination study presented furnishes an instance
of a reaction that reflects and illustrates the character-
istics of normal (5-7), medium (8-11), and large rings
(12-16) on both regio- and stereoselectivity. The trends
observed should be of use in the design of preparations
involving transformations at carbocycles of various size.
Exp er im en ta l Section
High-resolution masses were determined by peak matching
with tris(nonafluorobutyl)amine at a resolution of 0.2 mmu
or better. The tertiary alcohols 1 were prepared from the
respective cycloalkanones as described previously.1
Gen er a l P r oced u r e for th e Elim in a tion of Wa ter fr om
th e Ter tia r y Alk a n ols 1. Alkanol 1 (7 mmol) was dissolved
in 70 mL of toluene. After the addition of 4-toluenesulfonic
acid monohydrate (25 wt % of 1), the mixture was refluxed in
a Dean-Stark apparatus for 24 h. The solvent was partially
distilled off (40 mL), and the cooled solution was poured into
100 mL of 10% sodium hydrogen carbonate and 50 mL of
diethyl ether. The aqueous phase was extracted twice with
diethyl ether (60 mL each). The ether extracts were dried over
magnesium sulfate and filtered. The solvent was removed
under reduced pressure, and the residue was taken to NMR
analysis for the determination of the isomer ratio, followed
by flash chromatography or vacuum distillation to isolate the
main product.
(E)-Eth yl 2-(Cycloh exadec-1-en yl)pr opion ate (4a). Eth-
yl 2-(1-hydroxycyclohexadecyl)propionate (1a) (1.5 g, 4.4 mmol),
after flash chromatography (hexane/diethyl ether 4:1, Rf )
0.52), furnished 1.0 g (73%) of 4a as a colorless oil: 1H NMR
(CDCl3) δ 5.31 (t, J ) 8 Hz, 1H), [5.21 (t, J ) 8 Hz, 1H)], 4.17-
4.02 (m, 2H), [3.44 (q, J ) 7 Hz, 1H)], 3.02 (q, J ) 7 Hz, 1H,),
2.09-1.95 (m, 4H), 1.46-1.15 (m, 30H), (recognizable signals
(18) Screttas, S. G.; Smonou, I. C. J . Org. Chem. 1988, 53, 893-
894. See also Screttas, S. G.; Smonou, I. C. J . Organomet. Chem. 1988,
342, 143-152.
(19) (a) Schneider, H.-J .; Thomas, F. J . Am. Chem. Soc. 1980, 102,
1424-1425. (b) Schneider, H.-J .; Schmidt, G.; Thomas, F. J . Am. Chem.
Soc. 1983, 105, 3556-3563.
(20) Brown, H. C.; Gerstein, M. J . Am. Chem. Soc. 1950, 72, 2926-
2933. See also Eliel, E. H.; Wilen, S. H.; Mander, L. N. Stereochemistry
of Organic Compounds; Wiley: New York, 1994; pp 769-771.
(21) Wiberg, K. B.; Waldron, R. F. J . Am. Chem. Soc. 1991, 113,
7697-7705 and work cited therein.
(22) Cf. also Thalhammer, F.; Wallfahrer, U.; Sauer, F. Tetrahedron
Lett. 1990, 47, 6851-6854.
(23) (a) Kaarsemaker, S.; Coops, J . Recl. Trav. Chim. Pays-Bas 1952,
71, 261-276. (b) Dunitz, J . D.; Prelog, V. Angew. Chem. 1960, 23, 896-
902. See also Eliel, E. H.; Wilen, S. H.; Mander, L. N. Stereochemistry
of Organic Compounds; Wiley: New York, 1994; pp 675-678.

Regio- and Stereoselectivity of Water Elimination
J . Org. Chem., Vol. 62, No. 23, 1997 8061
of the Z-isomer 5a in square brackets); 13C NMR (CDCl3): δ
175.3, [174.9], [150.4], 138.4, [136.7], [128.4], 127.2, [122.9],
[60.3], 60.2, [60.0], 45.5, [40.8], [34.2], [33.4], [33.3], 30.0, [29.8],
[29.1], 28.6, [28.2], 28.11, 28.08, 27.94, 27.87, 27.7, [27.6],
27.39, [27.36], [27.33], [27.28], [27.2], [27.14], 27.07, 27.04,
[26.97], 26.9, [26.75], [26.69], [26.65], 26.4, 26.3, [26.04],
[25.98], [25.6], 25.4, 16.9, [15.5], 14.2, (recognizable signals of
the semicyclic and Z-isomers 3a and 5a in square brackets);
IR (neat) 2928 (vs), 2857 (s), 1735 (s), 1461 (m), 1180 (m); MS
(EI) m/z (rel int) 323 (M + 1, 6), 322 (M+, 26), 102 (100), 97
(35), 95 (31), 83 (43), 81 (35), 69 (35), 67 (39), 55 (47), 41 (38);
HRMS (EI) calcd for C21H38O2 322.2872, found 322.2866.
[23.51], 22.9, [22.6], 17.0, [15.4], [14.7], 14.2, [14.1] (recogniz-
able signals of the semicyclic and Z-isomers 3 and 5a in square
brackets); IR (neat) 2931 (vs), 2860 (s), 1735 (vs), 1468 (m),
1446 (m), 1373 (m), 1180 (s), 1096 (m), 916 (w); MS (EI) m/z
(rel int) 252 (M + 1, 3), 237 (3), 168 (19), 125 (26), 111 (32),
102 (92), 98 (28), 97 (26), 95 (35), 92 (65), 91 (100), 69 (40), 67
(39), 58 (46), 55 (86), 43 (72), 41 (61), 39 (37); HRMS (EI) calcd
for C16H28O2 252.2089, found 252.2087.
(E)-Eth yl 2-(Cyclod ec-1-en yl)p r op ion a te (4g). Ethyl
2-(1-hydroxycyclodecyl)propionate (1g) (1.8 g, 7.0 mmol), after
distillation at 1.9 × 10-2 mmHg, furnished 1.3 g (79%) of 4g
as a colorless oil; bp, 84 °C (1.9 × 10-2 mmHg); 1H NMR
(CDCl3) δ 5.34 (t, J ) 7 Hz, 1H), 4.17-4.08 (m, 2H), 3.06 (q, J
) 7 Hz, 1H), 2.29-2.26 (m, 4H), 1.57-1.51 (m, 4H), 1.40-
1.21 (m, 14H).; 13C NMR (CDCl3) δ 175.4, 137.7, 127.4, 60.3,
43.9, 27.9, 27.4, 26.8, 26.7, 25.2, 24.6, 20.9, 20.7, 16.8, 14.2;
IR (neat) 3510 (vs), 2920 (vs), 1705 (vs), 1470 (vs), 770 (vs);
MS (EI) m/z (rel int) 238 (M+, 0.4), 223 (2), 102 (100), 95 (25),
81 (30), 67 (25), 55 (31), 41 (36); HRMS (EI) calcd for C15H26O2
238.1933, found 238.1935.
(E)-Eth yl 2-(Cyclon on -1-en yl)-p r op ion a te (4h ). Accord-
ing to the general procedure, ethyl 2-(1-hydroxycyclononyl)-
propionate (1h ) (4.5 g, 19 mmol), after distillation at 2 × 10-2
mmHg, yielded 4h (3.1 g, 14 mmol, 74%) as a colorless oil; bp,
74-76 °C (2 × 10-2 mmHg); 1H NMR (CDCl3) δ 5.43 (t, J ) 8
Hz), 4.16-4.08 (m, 2H), 3.07 (q, J ) 7 Hz, 1H), 2.21-2.12 (m,
4H), 1.51-1.45 (m, 10H), 1.62-1.22 (m, 6H); 13C NMR (CDCl3)
δ 175.2, [170.4], [152.2], 139.2, 127.2, [123.7], 60.3, [60.0], 46.9,
[35.5], [35.1], 29.1, [28.6], [28.3], [27.1], 26.8, 26.7, 26.6, [26.4],
26.2, [25.8], 25.5, 24.0, 16.6, [14.3], 14.2 (recognizable signals
of the semicyclic isomer 3 in square brackets); IR (neat) 2977
(s), 2929 (vs), 2862 (s), 1734 (vs), 1451 (m), 1180 (s); MS (EI)
m/z (rel int) 224 (M+, 1), 209 (3), 102 (100), 95 (33), 81 (40), 67
(32), 55 (27), 41 (32); HRMS (EI) calcd for C14H24O2 224.1776,
found 224.1765.
(E)-Eth yl 2-(Cyclopen tadec-1-en yl)pr opion ate (4b). Eth-
yl 2-(1-hydroxycyclopentadecyl)propionate (1b ) (3.8 g, 12
mmol), after distillation at 1.8 × 10-2 mmHg, furnished 2.5 g
(67%) of 4b as a colorless oil; bp 138-142 °C (1.8 × 10-2
1
mmHg); H NMR (CDCl3) δ 5.31 (t, J ) 7 Hz, 1H), [5.22 (t, J
) 7 Hz, 1H)], 4.20-4.06 (m, 2H), 3.05 (q, J ) 7 Hz, 1H), 2.17-
2.00 (m, 8H), 1.37-1.20 (m, 24H), (recognizable signals of the
Z-isomer 5b in square brackets); 13C NMR (CDCl3) δ 175.1,
[174.7], [169.8], [150.9], 138.3, [136.5], [127.7], 127.3, [122.7],
60.1, [59.8], 45.8, [40.7], [34.3], [33.5], 29.5, [28.9], 28.0, [27.8],
[27.6], 27.4, 27.3, [27.24], 27.17, [27.00], 26.96, [26.93], [26.87],
[26.81], 26.76, [26.7], 26.59, 26.56, [26.5], [26.4], 26.22, 26.17,
26.1, [26.0], 25.8, [25.7], [25.6], 16.6, [14.14], [14.08], 14.0
(recognizable signals of the semicyclic and Z-isomer 3b and
5b in square brackets); IR (neat) 2931 (vs), 2858 (vs), 1779
(m), 1735 (vs), 1459 (s), 1181 (s). MS (EI) m/z (rel int) 309 (M
+ 1, 6), 308 (M+, 28), 293 (9), 102 (100), 97 (35), 83 (43), 69
(38), 55 (42), 41 (48). C20H36O2 (308.50) calcd C 77.87, H 11.76,
found C 78.00, H 11.51.
(E)-Eth yl 2-(Cyclotr id ec-1-en yl)p r op ion a te (4d ). Ethyl
2-(1-hydroxycyclotridecyl)propionate (1d ) (1.7 g, 5.7 mmol)
yielded 4d (1.2 g, 75%) as a colorless oil. 1H NMR (CDCl3) δ
5.22-5.13 (m, 1H), 4.12-3.97 (m, 2H), [3.43 (q, J ) 7 Hz, 1H)],
2.98 (q, J ) 7 Hz, 1H), 2.04-1.93 (m, 4H), 1.44-1.12 (m, 24H)
(recognizable signals of the Z-isomer 5a in square brackets);
13C NMR (CDCl3) δ 175.1, [174.6], [169.8], [150.2], 138.7,
[136.2], [128.9], 127.7, [123.0], [60.2], 60.1, 45.5, [41.9], [31.2],
28.2, [28.0], 27.6, [27.1], [26.9], 26.7, 26.6, [26.4], 26.2 (double
intensity), 26.1, [25.9], [25.84], 25.78, 25.72, [25.68], [25.6],
25.4, 24.81, [24.77], [24.7], [24.6], [24.4], [24.3], [23.2], 16.6,
[15.1], 14.0 (recognizable signals of the semicyclic and Z-
isomers 3 and 5a in square brackets); IR (neat) 2927 (vs), 2855
(vs), 1734 (vs), 1461 (s), 1179 (vs), 1097 (s).
(E)-Eth yl 2-(Cyclooct-1-en yl)p r op ion a te (4i). According
to the general procedure, ethyl 2-(1-hydroxycyclooctyl)propi-
onate (1i) (22.8 g, 99 mmol), after distillation at 1.8 × 10-2
mmHg, yielded 4i (15.1 g, 72 mmol, 73%) as a colorless oil;
1
bp, 70 °C (1.8 × 10-2 mmHg); H NMR (CDCl3) δ 5.52 (t, J )
8 Hz, 1H), 4.11 (q, J ) 7 Hz, 2H), 3.09 (q, J ) 7 Hz, 1H), 2.23-
2.09 (m, 4H), 1.56-1.42 (m, 6H), 1.26-1.22 (m, 8H); 13C NMR
(CDCl3) δ 175.1, [151.5], 139.3, 126.6, [123.9], 60.3, 47.1, 29.6,
29.2, 27.8, 26.4, 26.3, 26.2, 16.4, 14.2 (recognizable signals of
the semicyclic isomer 3 in square brackets); IR (neat) 2980
(s), 2927 (vs), 2854 (s), 1733 (vs), 1650 (w), 1469 (m), 1183 (s);
HRMS (EI) calcd for C13H22O2 210.1620, found 210.1614.
(E)-Eth yl 2-(Cyclod od ec-1-en yl)p r op ion a te (4e). Ethyl
2-(1-hydroxycyclododecyl)propionate (1e) (2.0 g, 7.0 mmol),
after flash chromatography (hexane/diethyl ether 4:1, Rf )
(E)-Eth yl 2-(Cycloh ep t-1-en yl)p r op ion a te (4j). Ethyl
2-(1-hydroxycycloheptyl)propionate (1j) (21.2 g, 99 mmol), after
flash chromatography (pentane/diethyl ether 4:1, Rf ) 0.52),
furnished 13.9 g (72%) of 4j as a colorless oil; 1H NMR (CDCl3)
δ 5.71 (t, J ) 7 Hz, 1H), 4.19-4.08 (m, 2H), 3.07 (q, J ) 7 Hz,
1H), 2.14-2.09 (m, 3H), 1.49-1.43 (m, 3H), 1.31-1.21 (m,
10H); 13C NMR (CDCl3) δ 174.7, [170.1], [150.5], 143.1, 128.6,
[122.6], 60.2, [60.1], 48.7, [33.3], [33.2], 32.6, 30.3, [29.0], [28.6],
28.3, [27.7], 26.9, 26.8, [26.1], 15.8, [15.7], 14.2, [14.1]; (signals
of the semicyclic isomer 3 in square brackets); IR (neat) 2923
(vs), 2855 (vs), 1712 (vs), 1623 (s), 1433 (vs), 1276 (vs), 1226
(vs), 1190 (vs), 1113 (vs), 1044 (s), 773 (s); MS (EI) m/z (rel
int) 196 (M+, 3), 182 (97), 151 (100), 150 (52), 125 (48), 123
(24), 122 (38), 101 (27), 96 (28), 95 (36), 93 (33), 88 (25), 83
(50), 82 (30), 81 (90), 79 (40), 77 (24), 67 (82), 55 (78), 54 (20),
53 (22), 43 (28), 41 (43), 39 (34); HRMS (EI) calcd for C12H20O2
196.1463, found 196.1484.
1
0.52), furnished 1.2 g (64%) of 4e as a colorless oil; H NMR
(CDCl3) δ [5.45 (t, J ) 8 Hz, 1H], 5.21 (t, J ) 8 Hz, 1H), 4.08-
3.96 (m, 2H), [3.47 (q, J ) 7 Hz, 1H)], 2.98 (q, J ) 7 Hz, 1H,),
2.13-1.91 (m, 4H), 1.48-1.10 (m, 22H) (recognizable signals
of the Z-isomer 5a in square brackets); 13C NMR (CDCl3) δ
175.1, [174.5], 138.0, [135.2], [129.2], 127.3, [60.1], 59.9, 43.8,
[40.2], [31.4], [27.3], 26.8, 26.3, [26.2], [26.1], 25.1, 24.7, 24.6,
24.5, 24.2, 23.8, [23.6], [23.1], 22.5, 22.1, 16.6, [15.3], 13.9,
[13.8] (recognizable signals of the Z-isomer 5a in square
brackets); IR (neat) 2928 (vs), 2863 (vs), 1735 (vs), 1712 (s),
1600 (w), 1469 (s), 1446 (s), 1180 (vs), 1097 (s), 805 (m), 726
(m); MS (EI) m/z (rel int) 267 (M + 1, 3), 266 (M+, 16), 251 (7),
200 (18), 155 (26), 109 (32), 102 (100), 95 (56), 92 (27), 91 (79),
83 (47), 81 (49), 79 (33), 69 (38), 67 (54), 65 (31), 55 (72), 41
(50); HRMS (EI) calcd for C17H30O2 266.2246, found 266.2231.
(E)-Eth yl 2-(Cycloh ex-1-en yl)p r op ion a te (4k ). Ethyl
2-(1-hydroxycyclohexyl)propionate (1k ) (3.4 g, 17 mmol), after
flash chromatography (pentane/diethyl ether 4:1, Rf ) 0.52),
furnished 2.4 g (76%) of 4k as a colorless oil; 1H NMR (CDCl3)
δ 5.59-5.55 (m, 1H), 4.21-4.10 (m, 2H), 3.02 (q, J ) 7 Hz,
1H), 2.04-1.86 (m, 4H), 1.65-1.52 (m, 4H), 1.33-1.21 (m, 6H);
13C NMR (CDCl3) δ 174.9, 136.6, 123.4, 60.3, 47.1, 26.3, 25.3,
22.9, 22.3, 15.7, 14.3. The compound was mentioned in the
literature24 with no characterization and analytical data.
(E)-Eth yl 2-(Cyclou n d ec-1-en yl)p r op ion a te (4f). Ethyl
2-(1-hydroxycycloundecyl)propionate (1f) (2.0 g, 7.0 mmol),
after distillation at 2.5 × 10-2 mmHg, furnished 1.2 g (64%)
of 4f as a colorless oil; bp, 100 °C (2.5 × 10-2 mmHg); 1H NMR
(CDCl3) δ [5.49 (t, J ) 4 Hz, 1H], 5.24 (t, J ) 7 Hz, 1H), 4.17-
4.07 (m, 2H), [3.67 (q, J ) 7 Hz, 1H], 3.10 (q, J ) 7 Hz, 1H),
2.27-218 (m, 4H), 1.53-1.17 (m, 20H) (recognizable signals
of the Z-isomer 5a in square brackets); 13C NMR (CDCl3) δ
175.4, [174.7], 138.4, [135.7], [127.9], 128.3, [60.4], 60.3, 43.6,
[41.1], [31.1], 29.8, [28.1], 27.9, 27.3, [26.93], 26.88, [26.8],
[26.7], [26.0], 25.8, [25.4], 25.2, [24.9], 24.7, 24.6, [23.66],
(24) Patel, Sh. K.; Paterson, I. Tetrahedron Lett. 1983, 24, 1315.

8062 J . Org. Chem., Vol. 62, No. 23, 1997
Greve and Imming
(E)-Eth yl 2-(Cyclop en t-1-en yl)p r op ion a te (4l). Ethyl
2-(1-hydroxycyclopentyl)propionate (1l) (3.3 g, 18 mmol) fur-
nished 2.6 g (88%) of 4l as a pale yellow oil; 1H NMR (CDCl3)
δ 5.52-5.51 (m, 1H), 4.19-4.10 (m, 2H), 3.23 (q, J ) 7 Hz,
1H), 2.36-2.26 (m, 2H), 1.90-1.82 (m, 2H), 1.30-1.23 (m, 8H);
13C NMR (CDCl3) δ 174.5, [168.4], [160.2], 142.7, 125.3, [118.6],
60.4, [59.8], 41.7, [34.2], [34.1], 33.3, 32.3, [27.2], [25.6], 23.3,
[21.5], 15.9, [14.5], 14.2 (signals of the semicyclic isomer 3l in
square brackets).
Su p p or tin g In for m a tion Ava ila ble: 13C NMR spectra of
compounds 4a ,d -j,l (9 pages). This material is contained in
libraries on microfiche, immediately follows this article in the
microfilm version of the journal, and can be ordered from the
ACS; see any current masthead page for ordering information.
J O970989G