Downfield chemical shifts at α-protons and carbons of β-propiothiolactones

Article abstract of DOI:10.1002/1097-458X(200006)38:6<468::AID-MRC675>3.0.CO;2-F

Both the α-protons and carbons of β-propiothiolactones exhibit atypical downfield chemical shifts. The α-protons of β-propiothiolactones with no heteroatom at the α-position appear at 3.53-5.35ppm, whereas the α-carbons appear at 56.9-86.2 ppm. The major cause of the unexpected deshielding effect was rationalized by assuming a through-space interaction between the occupied orbital of the α-carbon and the vacant orbital of sulfur. Copyright

Full text of DOI:10.1002/1097-458X(200006)38:6<468::AID-MRC675>3.0.CO;2-F

MAGNETIC RESONANCE IN CHEMISTRY
Magn. Reson. Chem. 2000; 38: 468–471
Note
Downfield chemical shifts at a-protons and carbons of
b-propiothiolactones
1
2
2
1
∗
Hee Bong Lee, Hyung-Yeon Park, Bon-Su Lee and Young Gyu Kim
1
School of Chemical Engineering, College of Engineering, Seoul National University, Seoul 151-742, Korea
Department of Chemistry, Inha University, Inchon 402-751, Korea
2
Received 13 August 1999; revised 4 February 2000; accepted 7 February 2000
ABSTRACT: Both the ˛-protons and carbons of ˇ-propiothiolactones exhibit atypical downfield chemical shifts. The
˛-protons of ˇ-propiothiolactones with no heteroatom at the ˛-position appear at 3.53–5.35 ppm, whereas the ˛-
carbons appear at 56.9–86.2 ppm. The major cause of the unexpected deshielding effect was rationalized by assuming
a through-space interaction between the occupied orbital of the ˛-carbon and the vacant orbital of sulfur. Copyright
2000 John Wiley & Sons, Ltd.
1
KEYWORDS: NMR; H NMR; 13C NMR; 2-D INADEQUATE; ˇ-propiothiolactones; deshielding effect; through-
space interaction
A literature survey revealed only five papers which
reported on the 13C chemical shifts of ˇ-propiothiolactones
having no heteroatom at the ˛-position (Scheme 3).4
Moreover, the assignments of the chemical shifts for
the ˛-carbons are not consistent with each other. The
anomalous deshielding effect at the ˛-carbons in certain
four-membered rings was reported several years ago and
explained as an inherent property of the carbon orbitals
that contribute to the paramagnetic shielding, ꢀp.5 How-
ever, there has been no study that has particularly focused
on ˇ-propiothiolactones.
INTRODUCTION
As part of a research program directed at the efficient
preparation of captopril, the first angiotensin-converting
enzyme-inhibiting antihypertensive,1 we recently became
interested in the synthesis of ˇ-propiothiolactone 1
(Scheme 1).
Scheme 1. Structure of 1 and captopril.
Comparable yields of 1 have been obtained according to
procedures outlined in the literature.2 Its 13C NMR spec-
trum shows a surprising downfield shift (65.7 ppm) for
the ˛-carbon. In comparison with other structurally related
heterocyclic four-membered ring compounds (Scheme 2),3
this chemical shift is unusually large. In contrast, the ˛-
carbons to the carbonyl group of ˇ-lactone and ˇ-lactam
are relatively shielded compared with that of cyclobu-
tanone.
Scheme 3. 13C NMR chemical shifts (ppm) of ˇ-propiothio-
lactones reported in the literature.
Scheme 2. 13C chemical shifts (ppm) of related four-mem-
bered rings.
It was therefore considered necessary to establish the
structure of 1 unambiguously and then examine NMR
spectral properties of the related ˇ-propiothiolactones.
Here we report a systematic study of the chemical shifts
of several ˇ-propiothiolactones and propose an origin for
the atypical deshielding effect at their ˛-carbons.
* Correspondence to: Y. G. Kim, School of Chemical Engineering,
College of Engineering, Seoul National University, Seoul 151-742,
Korea; e-mail: ygkim@plaza.snu.ac.kr
Contract/grant sponsor: Development Fund for Industrial Basic Tech-
nology.
Copyright 2000 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2000; 38: 468–471

DOWNFIELD CHEMICAL SHIFTS IN ˇ-PROPIOTHIOLACTONES
469
show chemical shifts larger than 65 and 72 ppm, respec-
tively. The largest 13C chemical shift is observed with
13 that has two deshielding substituents. In view of
these results, the 13C chemical shifts reported by earlier
workers4b,c should be interchanged between the ˛- and
ˇ-carbons. Indeed, ˇ, ˇ-dimethyl-ˇ-propiothiolactone (14)
prepared by us has chemical shifts of 67.3 and 40.7 ppm
for its ˛- and ˇ-carbon, respectively, which was sup-
ported by an ab initio calculation (see Scheme 3 and
below).
The unexpected deshielding at the ˛-protons and ˛-
carbons of ˇ-propiothiolactones might be rationalized
by two main factors. One is the less efficient overlap
between the C O ꢁ and S 3p orbitals, resulting in a
stronger deshielding effect by the carbonyl group. The
other involves a through-space interaction between the
occupied orbital of the ˛-carbon and the vacant orbital
of sulfur. For example, a fairly extensive overlap between
the back lobe of the ˛-C bonding orbital and the empty
S d orbital can be assumed to be due to the almost pla-
nar tetragonal geometry of the ˇ-propiothiolactone rings
(we thank a referee for suggesting the back-lobe over-
lap mechanism as an alternative to our rationale of the
through-space interaction between the occupied pseudo-
ꢁ-orbital of ˛-CH2 and the vacant orbital of S7).
An ab initio calculation of the isotropic shielding con-
stants (ꢀ) was carried out to test the latter proposition
using the individual gauge atoms in molecules (IGAIM)
method8 with the HF/6-31CCGŁŁ basis set implemented
in Gaussian 94 (Revision A.1) program (Gaussian, Pitts-
burgh, PA, USA) (Table 2). The geometries used for
the IGAIM calculations were fully optimized at the
B3LYP/6-31CCGŁŁ level in order to consider the electron-
correlation effect. In fact, a significant difference in the
shielding constants (ꢀ) of a saturated system is observed
in the contribution from the direction perpendicular to the
molecular plane of thietane (ꢀzz). The difference becomes
much larger in the unsaturated system, thiolactone. The
nearly planar geometry of ˇ-propiothiolactone makes the
orbitals overlap more effectively, resulting in a greater
deshielding effect.
RESULTS AND DISCUSSION
First, the structure of 1 was confirmed by both spectral
and chemical means. A 2-D INADEQUATE experiment
together with the DEPT and proton non-decoupling spec-
tra established that the tertiary carbon was connected to
the other three carbons. The IR spectrum and the exact
mass data indicated a four-membered ring structure.
The ring-opening reactions of 1 in acidic aqueous solu-
tion or with benzylamine gave back the starting mercapto
acid or its N-benzylamide, respectively. A product iden-
tical with 1 was also produced by treatment of 4-methyl-
1,2-dithiolan-3-one, obtained via an independent route,6
with triphenylphosphine, as shown in Scheme 4.
Scheme 4. Structural correlation studies.
Various ˇ-propiothiolactones were then synthesized to
identify their spectral characteristics. The substituted ˇ-
propiothiolactones were prepared using various adapta-
tions of methods reported in the literature (Scheme 5). As
a result, ˇ-propiothiolactones 3–7 and 14 were prepared
by the cyclization of the corresponding 3-mercapto acids
in 17–42% yield, while 2 and 8–13 were obtained by
the substitution reactions of 6 or 1 with the corresponding
electrophiles in 30–55% yield. All ˇ-propiothiolactones
prepared in the present study showed satisfactory spectro-
scopic (NMR, Mass, IR) data.
In conclusion, a unique deshielding effect has been
observed at the ˛-protons and carbons of ˇ-propiothiolac-
tones that can be mainly attributed to the through-space
interaction between the occupied orbital of the ˛-carbon
and the vacant orbital of sulfur.
EXPERIMENTAL
Scheme 5. Structure of the ˇ-propiothiolactones pre-
pared in the present study.
NMR measurements
All the NMR spectra in CDCl3 were recorded on a JEOL
JNM-LA 300 or 400 spectrometer. In all cases, a 5 mm
o.d. tube was used at room temperature. All proton and
carbon spectra, except for the 2-D INADEQUATE spec-
trum, were recorded at 300.40 and 75.45 MHz, respec-
tively. The chemical shifts were referenced from TMS at
The spectral data for the ˇ-propiothiolactones are
summarized in Table 1. These data clearly exhibit a
downfield resonance trend for the ˛-protons and car-
1
bons. All of the H chemical shifts at the ˛-position of
ˇ-propiothiolactones are larger than 3.5 ppm, and the ˛-
protons of 6 are deshielded further by ca 1 ppm than those
of cyclobutanone. Mono- and disubstituted ˛-carbons
1
0.00 ppm for the H spectra and from the central solvent
peak at 77.0 ppm for the 13C spectra.
Copyright 2000 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2000; 38: 468–471

470
H. B. LEE ET AL.
Table 1. 1H NMR and 13C NMR data for the ˇ-propiothiolactones in CDCl3 (υ, ppm)
Atom
1H H-3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
4.25 5.02 4.16
4.47
5.35
4.02 3.53
4.11
3.71
2.67 3.19 2.75
3.18 3.50 3.13
2.81
3.03
3.02
3.13
3.20
3.46
2.70
2.89
1.37
2.62
3.00
1.41
2.84
3.62
1.70
3.38
4.04
H-4
3.05
3.74
2.83
1.39
2.79
1.37
1.81
1.68a
H-10
2.70
3.13
2.73
3.19
H-100
1.71
1.01
1.81a
H-20
1.01
H-200
CO2Me
3.81
3.81
3.78
7.34
7.18
¾7.35 ¾7.42
7.29
7.16
¾7.34
7.18
Ph
¾7.34 ¾7.45
13C C-2
195.0 184.9 195.1 194.4 192.5 191.1 190.2
65.7 72.3 71.8
23.8 18.0 21.4
199.0 199.0
198.7 190.0
198.4
79.6
24.0
187.5 189.7
C-3
C-4
70.7
21.4
36.3
74.3
23.7
56.9 62.6
15.0 27.9
23.5a
72.1
31.2
76.3
28.6
21.5
29.5
75.7
28.0
22.5
42.1
78.8
26.9
20.8
86.2
27.5
67.3
40.7
C-10
C-100
C-20
C-200
15.8
24.0
23.5
41.9
31.6a
10.2
8.4
53.0
165.2
53.1
168.6
53.6
167.5
125.5
CO2Me
Ph
126.8 126.6
¾136.6 ¾135.4
126.9
¾135.8
127.0
¾135.5 ¾135.5
a 4-CH3.
Table 2. Calculated values of components for isotropic
shielding constants (ꢀ) and chemical shifts (υ) for the
four-membered rings
The 2-D INADEQUATE experiment was recorded
on a JEOL JNM-LA 400 spectrometer operating at
100.40 MHz (500 mg mlꢀ1 solution). The spectral widths
were 20 and 40 kHz in the F2 and F1 dimensions, respec-
tively. The matrix used 2K ð 128 data points, and 16
scans were accumulated for each increment. The repetition
time was 37.4 s and the 13C–13C coupling constant was set
to 50 Hz.
Compound
υtheo
υexp
(ꢀxx C ꢀyy)/3
ꢀzz/3
X
Y
(ppm)
(ppm)
(ppm)a
(ppm)b
CH2
NH
O
H2
H2
H2
H2
O
O
O
O
125.5
125.7
126.4
126.9
116.9
120.1
121.5
119.5
53.6
53.4
52.5
47.8
39.4
42.2
41.5
30.0
24.3
24.3
24.5
28.7
47.1
41.1
40.4
53.9
22.4
19.0
22.9
28.1
47.8
38.3
39.1
56.9c
Materials
Preparation of b-propiothiolactones by the cyclization reactions.
To a magnetically stirred solution of substituted acrylic acid (10 mmol)
in THF (15 ml) was added dropwise thiolacetic acid (13 mmol) in THF
(15 ml), and the reaction mixture was heated at 50 °C for 18 h. The
solvent and excess thiolacetic acid were evaporated, then dry-column
flash chromatography of the residue with n-hexane and EtOAc afforded
substituted S-acetyl-3-thiopropionic acid.
S
CH2
NH
O
S
To a stirred solution of substituted S-acetyl-3-thiopropionic acid
(5 mmol) in water (10 ml) was added concentrated H2SO4 (50 mg,
0.5 mmol), and the reaction mixture was heated under reflux for 5 h. The
resulting mixture was extracted with CH2Cl2 (3 ð 20 ml) and the com-
bined organic layers were dried (MgSO4), filtered and concentrated in
vacuo to give crude substituted 3-mercaptopropionic acid. To a magnet-
ically stirred solution of the crude substituted 3-mercaptopropionic acid
and triethylamine (5.5 mmol) in diethyl ether (10 ml) at 5 °C was added
dropwise methyl chloroformate (5.5 mmol) in diethyl ether (10 ml).
After the dropping was complete, the reaction mixture was warmed
to room temperature and the stirring continued for 3 h. The resulting
mixture was washed with 0.5 M HCl and the separated organic layer
was dried (MgSO4), filtered and concentrated. Dry-column flash chro-
matography with n-hexane and EtOAc afforded ˇ-propiothiolactone.
a υtheo D 203.4 (estimated value of ꢀTMS/ ꢀ ꢀ, [ꢀ D .ꢀxx C ꢀyy C ꢀzz//3].
b Literature value3.
c This work.
1
The H NMR spectra were taken with a spectral width
°
of 6 kHz and a 30 pulse angle, 32K data points and
a repetition time of 1.5 s. For the 13C NMR spectra, a
°
spectral width of 22.7 kHz, a pulse angle of 37.5 , 32K
data points and a repetition time of 1.0 s were used.
Copyright 2000 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2000; 38: 468–471

DOWNFIELD CHEMICAL SHIFTS IN ˇ-PROPIOTHIOLACTONES
471
Cushman DW. US Pat. 4 046 889, 1977; Chem. Abstr. 1978; 88:
7376c.
Preparation of b-propiothiolactones by the substitution reac-
tions. To a magnetically stirred solution of lithium bis(trimethylsilyl)-
amide (LHMDS) (4.0 ml of a 1.0 M solution in THF, 4.0 mmol) in THF
at ꢀ78 °C in an acetone–dry-ice bath was added ˇ-propiothiolactone
(3.0 mmol) in THF (5.0 ml). After stirring for 30 min, an electrophile
(5.0 mmol) (MeI, EtI, BnBr or ClCO2Me) in THF (5.0 ml) at ꢀ78 °C
was added dropwise via a syringe over 2 min. The reaction mixture was
stirred at the same temperature for 1 h and then allowed to warm to 0 °C.
After addition of saturated aqueous NH4Cl solution, the resulting mix-
ture was extracted with diethyl ether (2 ð 10 ml). The combined organic
layers were dried (MgSO4), filtered, and concentrated. Dry-column flash
chromatography with n-hexane and EtOAc afforded ˛-substituted ˇ-
propiothiolactone.
3. (a) Kalinowski H-O, Berger S, Braun S. Carbon-13 NMR Spec-
troscopy. Wiley: Chichester, 1988; 92–467; (b) Pretsch E, Clerc T,
Seibl J, Simon W. Tables of Spectral Data for Structure Determi-
nation of Organic Compounds (2nd edn). Springer: Berlin, 1989;
C173–C197; (c) Pouchert CJ, Behnke J. The Aldrich library of 13C
and 1H NMR Spectra (1st edn), vol. 1. Aldrich: Milwaukee, WI,
1993; 1285.
4. (a) Barbaro G, Battaglia A, Giorgianni P. J. Org. Chem. 1987; 52:
3289; (b) Kohn H, Charumilind P, Gopichand Y. J. Org. Chem.
1978; 43: 4961; (c) Pattenden G, Shuker AJ. J. Chem. Soc., Perkin
Trans. 1 1992; 1215; (d) Sakamoto M, Takahashi M, Yoshiaki M,
Fujita T, Watanabe S, Aoyama H. J. Chem. Soc., Perkin Trans. 1
1994; 2983; (e) Adam W, Peters E-M, Peters K, Rebollo H, Rosen-
thal RJ, von Schnering HG. Chem. Ber. 1984; 117: 2393; (f) For
other references related to ˇ-propiothiolactones, see Block E, de
Wang M. In Comprehensive Heterocyclic Chemistry II, vol. 1B.
Katritzky AR, Rees CW, Scriven EFV (eds). Pergamon Press:
Oxford, 1996; 773–821; Block E. In Comprehensive Heterocyclic
Chemistry, vol. 7. Katritzky AR, Rees CW (eds). Pergamon Press:
Oxford, 1984; 403–447.
Acknowledgments
This work was financially supported by the Development Fund for
Industrial Basic Technology, administered by the Ministry of Trade,
Industry and Energy (MOTIE, ’95–’97) of Korea. We are also grateful
to Ms Jong Ok Park for the 2-D INADEQUATE spectrum.
5. Lambert JB, Wharry SM, Block E, Bazzi AA. J. Org. Chem. 1983;
48: 3982.
6. Lee HB, Kim YG. J. Ind. Eng. Chem. 1998; 4: 127.
7. (a) Grutzner JB, Jautelat M, Dence JB, Smith RA, Roberts JD. J.
Am. Chem. Soc. 1970; 92: 7107; (b) Barfield M. J. Chem. Phys.
1964; 41: 3825; (c) Meinwald J, Lewis A. J. Am. Chem. Soc. 1961;
83: 2769.
REFERENCES
1. Redshaw S. In Medicinal Chemistry, (2nd edn). Ganellin CR,
Roberts SM (eds). Academic Press: London, 1993; 163–185.
2. (a) Lin’kova MG, Knunyants IL. Izv. Akad. Nauk SSSR, Ser. Khim.
1968; 1889; Chem. Abstr. 1969; 70: 3975c; (b) Ondetti MA,
8. Keith TA, Bader RFW. Chem. Phys. Lett. 1993; 210: 223.
Copyright 2000 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2000; 38: 468–471